From 7385db5be4d9d41edeb0ef73a177cb17b7c74cd1 Mon Sep 17 00:00:00 2001 From: Divya Date: Wed, 11 Oct 2023 11:22:22 -0400 Subject: [PATCH 01/22] Updated formatting for ml-frameworks --- frameworks.qmd | 1816 +++++++++++++++++++++++++++++-- images_ml_frameworks/image1.png | Bin 0 -> 34481 bytes images_ml_frameworks/image2.png | Bin 0 -> 71931 bytes images_ml_frameworks/image3.png | Bin 0 -> 44837 bytes images_ml_frameworks/image4.png | Bin 0 -> 50250 bytes images_ml_frameworks/image5.png | Bin 0 -> 34613 bytes images_ml_frameworks/image6.png | Bin 0 -> 186596 bytes images_ml_frameworks/image7.png | Bin 0 -> 100282 bytes images_ml_frameworks/image8.png | Bin 0 -> 129032 bytes 9 files changed, 1742 insertions(+), 74 deletions(-) create mode 100644 images_ml_frameworks/image1.png create mode 100644 images_ml_frameworks/image2.png create mode 100644 images_ml_frameworks/image3.png create mode 100644 images_ml_frameworks/image4.png create mode 100644 images_ml_frameworks/image5.png create mode 100644 images_ml_frameworks/image6.png create mode 100644 images_ml_frameworks/image7.png create mode 100644 images_ml_frameworks/image8.png diff --git a/frameworks.qmd b/frameworks.qmd index 85f2ab9b..271c3fd7 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -1,119 +1,1787 @@ # AI Frameworks -::: {.callout-tip collapse="true"} -## Learning Objectives +Learning Objectives -* coming soon. +- The evolution, core components, and advanced features of ML + > frameworks -::: +- How frameworks specialize for cloud, edge, and tinyML environments + +- Challenges of embedded ML and how frameworks optimize models + +- Criteria for selecting the right framework based on models, + > hardware, software factors + +- How to match framework capabilities to the constraints and + > requirements of a project + +- Ongoing innovations in frameworks for next-generation machine + > learning ## Introduction -Explanation: Discuss what ML frameworks are and why they are important. Also, elaborate on the aspects involved in understanding how an ML framework is developed and deployed. +Machine learning frameworks provide the tools and infrastructure to +efficiently build, train, and deploy machine learning models. In this +chapter, we will explore the evolution and key capabilities of major +frameworks like TensorFlow, PyTorch, and specialized frameworks for +embedded devices. We will dive into the components like computational +graphs, optimization algorithms, hardware acceleration, and more that +enable developers to quickly construct performant models. Understanding +these frameworks is essential to leverage the power of deep learning +across the spectrum from cloud to edge devices. + +ML frameworks handle much of the complexity of model development through +high-level APIs and domain-specific languages that allow practitioners +to quickly construct models by combining pre-made components and +abstractions. For example, frameworks like TensorFlow and PyTorch +provide Python APIs to define neural network architectures using layers, +optimizers, datasets, and more. This enables rapid iteration compared to +coding every model detail from scratch. + +A key capability offered by frameworks is distributed training engines +that can scale model training across clusters of GPUs and TPUs. This +makes it feasible to train state-of-the-art models with billions or +trillions of parameters on vast datasets. Frameworks also integrate with +specialized hardware like NVIDIA GPUs to further accelerate training via +optimizations like parallelization and efficient matrix operations. + +In addition, frameworks simplify deploying finished models into +production through tools like TensorFlow Serving for scalable model +serving and TensorFlow Lite for optimization on mobile and edge devices. +Other valuable capabilities include visualization, model optimization +techniques like quantization and pruning, and monitoring metrics during +training. + +Leading open source frameworks like TensorFlow, PyTorch, and MXNet power +much of AI research and development today. Commercial offerings like +Amazon SageMaker and Microsoft Azure Machine Learning integrate these +open source frameworks with proprietary capabilities and enterprise +tools. + +Machine learning engineers and practitioners leverage these robust +frameworks to focus on high-value tasks like model architecture, feature +engineering, and hyperparameter tuning instead of infrastructure. The +goal is to efficiently build and deploy performant models that solve +real-world problems. + +In this chapter, we will explore today\'s leading cloud frameworks and +how they have adapted models and tools specifically for embedded and +edge deployment. We will compare programming models, supported hardware, +optimization capabilities, and more to fully understand how frameworks +enable scalable machine learning from the cloud to the edge. + +## Framework Evolution + +Machine learning frameworks have evolved significantly over time to meet +the diverse needs of machine learning practitioners and advancements in +AI techniques. A few decades ago, building and training machine learning +models required extensive low-level coding and infrastructure. Machine +learning frameworks have evolved considerably over the past decade to +meet the expanding needs of practitioners and rapid advances in deep +learning techniques. Early neural network research was constrained by +insufficient data and compute power. Building and training machine +learning models required extensive low-level coding and infrastructure. +But the release of large datasets like ImageNet in 2009 and advancements +in parallel GPU computing unlocked the potential for far deeper neural +networks. + +The first ML frameworks, Theano (2007) and Caffe (2014), were developed +by academic institutions (Montreal Institute for Learning Algorithms, +Berkeley Vision and Learning Center). Amid a growing interest in deep +learning due to state-of-the-art performance of AlexNet (2012) on the +ImageNet dataset, private companies and individuals began developing ML +frameworks, resulting in frameworks such as Keras by Google researcher +François Chollet (2015), Chainer by Preferred Networks (2015), +TensorFlow by Google (2015), CNTK by Microsoft (2016), and PyTorch by +Facebook (2016). + +Many of these ML frameworks can be divided into categories, namely +high-level vs. low-level frameworks and static vs. dynamic computational +graph frameworks. High-level frameworks provide a higher level of +abstraction than low-level frameworks. That is, high-level frameworks +have pre-built functions and modules for common ML tasks, such as +creating, training, and evaluating common ML models as well as +preprocessing data, engineering features, and visualizing data, which +low-level frameworks do not have. Thus, high-level frameworks may be +easier to use, but are not as customizable as low-level frameworks (i.e. +users of low-level frameworks can define custom layers, loss functions, +optimization algorithms, etc.). Examples of high-level frameworks +include TensorFlow/Keras and PyTorch. Examples of low-level ML +frameworks include TensorFlow with low-level APIs, Theano, Caffe, +Chainer, and CNTK. + +Frameworks like Theano and Caffe used static computational graphs which +required rigidly defining the full model architecture upfront. Static +graphs require upfront declaration and limit flexibility. Dynamic graphs +construct on-the-fly for more iterative development. But around 2016, +frameworks began adopting dynamic graphs like PyTorch and TensorFlow 2.0 +which can construct graphs on-the-fly. This provides greater flexibility +for model development. We will discuss these concepts and details later +on in Section XXX. + +The development of these frameworks facilitated an explosion in model +size and complexity over time---from early multilayer perceptrons and +convolutional networks to modern transformers with billions or trillions +of parameters. In 2017, ResNet models achieved record ImageNet accuracy +with over 150 layers and 25 million parameters. Then in 2020, the GPT-3 +language model pushed parameters to an astonishing 175 billion using +model parallelism in frameworks to train across thousands of GPUs and +TPUs. + +Each generation of frameworks unlocked new capabilities that powered +advancement: + +- Theano and TensorFlow (2015) introduced computational graphs and + > automatic differentiation to simplify model building. + +- CNTK (2016) pioneered efficient distributed training by combining + > model and data parallelism. + +- PyTorch (2016) provided imperative programming and dynamic graphs + > for flexible experimentation. + +- TensorFlow 2.0 (2019) made eager execution default for intuitiveness + > and debugging. + +- TensorFlow Graphics (2020) added 3D data structures to handle point + > clouds and meshes. + +In recent years, there has been a convergence on the frameworks. +TensorFlow and PyTorch have become the overwhelmingly dominant ML +frameworks, representing more than 95% of ML frameworks used in research +and production. Keras was integrated into TensorFlow in 2019; Preferred +Networks transitioned Chainer to PyTorch in 2019; and Microsoft stopped +actively developing CNTK in 2022 in favor of supporting PyTorch on +Windows. + +![](images_ml_frameworks/image6.png){width="3.821385608048994in" +height="2.5558081802274715in"} + +Popularity of ML frameworks in the United States as measured by Google +web searches + +However, a one-size-fits-all approach does not work well across the +spectrum from cloud to tiny edge devices. Different frameworks represent +various philosophies around graph execution, declarative versus +imperative APIs, and more. Declarative defines what the program should +do while imperative focuses on how it should do it step-by-step. For +instance, TensorFlow uses graph execution and declarative-style modeling +while PyTorch adopts eager execution and imperative modeling for more +Pythonic flexibility. Each approach carries tradeoffs that we will +discuss later in the Basic Components section. + +Today\'s advanced frameworks enable practitioners to develop and deploy +increasingly complex models - a key driver of innovation in the AI +field. But they continue to evolve and expand their capabilities for the +next generation of machine learning. To understand how these systems +continue to evolve, we will dive deeper into TensorFlow as an example of +how the framework grew in complexity over time. + +## DeepDive into TensorFlow + +TensorFlow was developed by the Google Brain team and was released as an +open-source software library on November 9, 2015. It was designed for +numerical computation using data flow graphs and has since become +popular for a wide range of machine learning and deep learning +applications. + +TensorFlow is both a training and inference framework and provides +built-in functionality to handle everything from model creation and +training, to deployment. Since its initial development, the TensorFlow +ecosystem has grown to include many different "varieties" of TensorFlow +that are each intended to allow users to support ML on different +platforms. In this section, we will mainly discuss only the core +package. + +### TF Ecosystem + +1. [[TensorFlow + > Core]{.underline}](https://www.tensorflow.org/tutorials): primary + > package that most developers engage with. It provides a + > comprehensive, flexible platform for defining, training, and + > deploying machine learning models. It includes tf.keras as its + > high-level API. + +2. [[TensorFlow Lite]{.underline}](https://www.tensorflow.org/lite): + > designed for deploying lightweight models on mobile, embedded, and + > edge devices. It offers tools to convert TensorFlow models to a + > more compact format suitable for limited-resource devices and + > provides optimized pre-trained models for mobile. + +3. [[TensorFlow.js]{.underline}](https://www.tensorflow.org/js): + > JavaScript library that allows training and deployment of machine + > learning models directly in the browser or on Node.js. It also + > provides tools for porting pre-trained TensorFlow models to the + > browser-friendly format. + +4. [[TensorFlow on Edge Devices + > (Coral)]{.underline}](https://developers.googleblog.com/2019/03/introducing-coral-our-platform-for.html): + > platform of hardware components and software tools from Google + > that allows the execution of TensorFlow models on edge devices, + > leveraging Edge TPUs for acceleration. + +5. [[TensorFlow Federated + > (TFF)]{.underline}](https://www.tensorflow.org/federated): + > framework for machine learning and other computations on + > decentralized data. TFF facilitates federated learning, allowing + > model training across many devices without centralizing the data. + +6. [[TensorFlow + > Graphics]{.underline}](https://www.tensorflow.org/graphics): + > library for using TensorFlow to carry out graphics-related tasks, + > including 3D shapes and point clouds processing, using deep + > learning. + +7. [[TensorFlow Hub]{.underline}](https://www.tensorflow.org/hub): + > repository of reusable machine learning model components to allow + > developers to reuse pre-trained model components, facilitating + > transfer learning and model composition + +8. [[TensorFlow + > Serving]{.underline}](https://www.tensorflow.org/tfx/guide/serving): + > framework designed for serving and deploying machine learning + > models for inference in production environments. It provides tools + > for versioning and dynamically updating deployed models without + > service interruption. + +9. [[TensorFlow Extended + > (TFX)]{.underline}](https://www.tensorflow.org/tfx): end-to-end + > platform designed to deploy and manage machine learning pipelines + > in production settings. TFX encompasses components for data + > validation, preprocessing, model training, validation, and + > serving. + +TensorFlow was developed to address the limitations of DistBelief---the +framework in use at Google from 2011 to 2015---by providing flexibility +along three axes: 1) defining new layers, 2) refining training +algorithms, and 3) defining new training algorithms. To understand what +limitations in DistBelief led to the development of TensorFlow, we will +first give a brief overview of the Parameter Server Architecture that +DistBelief employed. + +The Parameter Server (PS) architecture is a popular design for +distributing the training of machine learning models, especially deep +neural networks, across multiple machines. The fundamental idea is to +separate the storage and management of model parameters from the +computation used to update these parameters: + +**Storage**: The storage and management of model parameters were handled +by the stateful parameter server processes. Given the large scale of +models and the distributed nature of the system, these parameters were +sharded across multiple parameter servers. Each server maintained a +portion of the model parameters, making it \"stateful\" as it had to +maintain and manage this state across the training process. + +**Computation**: The worker processes, which could be run in parallel, +were stateless and purely computational, processing data and computing +gradients without maintaining any state or long-term memory. + +DistBelief and its architecture defined above were crucial in enabling +distributed deep learning at Google but also introduced limitations that +motivated the development of TensorFlow: + +### Static Computation Graph + +In the parameter server architecture, model parameters are distributed +across various parameter servers. Since DistBelief was primarily +designed for the neural network paradigm, parameters corresponded to a +fixed structure of the neural network. If the computation graph were +dynamic, the distribution and coordination of parameters would become +significantly more complicated. For example, a change in the graph might +require the initialization of new parameters or the removal of existing +ones, complicating the management and synchronization tasks of the +parameter servers. This made it harder to implement models outside the +neural framework or models that required dynamic computation graphs. + +TensorFlow was designed to be a more general computation framework where +the computation is expressed as a data flow graph. This allows for a +wider variety of machine learning models and algorithms outside of just +neural networks, and provides flexibility in refining models. + +### Usability & Deployment + +The parameter server model involves a clear delineation of roles (worker +nodes and parameter servers), and is optimized for data center +deployments which might not be optimal for all use cases. For instance, +on edge devices or in other non-data center environments, this division +introduces overheads or complexities. + +TensorFlow was built to run on multiple platforms, from mobile devices +and edge devices, to cloud infrastructure. It also aimed to provide ease +of use between local and distributed training, and to be more +lightweight, and developer friendly. + +### Architecture Design + +Rather than using the parameter server architecture, TensorFlow instead +deploys tasks across a cluster. These tasks are named processes that can +communicate over a network, and each can execute TensorFlow\'s core +construct: the dataflow graph, and interface with various computing +devices (like CPUs or GPUs). This graph is a directed representation +where nodes symbolize computational operations, and edges depict the +tensors (data) flowing between these operations. + +Despite the absence of traditional parameter servers, some tasks, called +"PS tasks", still perform the role of storing and managing parameters, +reminiscent of parameter servers in other systems. The remaining tasks, +which usually handle computation, data processing, and gradient +calculations, are referred to as \"worker tasks.\" TensorFlow\'s PS +tasks can execute any computation representable by the dataflow graph, +meaning they aren\'t just limited to parameter storage, and the +computation can be distributed. This capability makes them significantly +more versatile and gives users the power to program the PS tasks using +the standard TensorFlow interface, the same one they\'d use to define +their models. As mentioned above, dataflow graphs' structure also makes +it inherently good for parallelism allowing for processing of large +datasets. + +### Built-in Functionality & Keras + +TensorFlow includes libraries to help users develop and deploy more +use-case specific models, and since this framework is open-source, this +list continues to grow. These libraries address the entire ML +development life-cycle: data preparation, model building, deployment, as +well as responsible AI. + +Additionally, one of TensorFlow's biggest advantages is its integration +with Keras. Keras is another ML framework that was built to be extremely +user-friendly and as a result has a high level of abstraction. We will +cover Keras in more depth later in this chapter, but when discussing its +integration with TensorFlow, the most important thing to note is that it +was originally built to be backend agnostic. This means users could +abstract away these complexities, offering a cleaner, more intuitive way +to define and train models without worrying about compatibility issues +with different backends. TensorFlow users had some complaints about the +usability and readability of TensorFlow's API, so as TF gained +prominence it integrated Keras as its high-level API. This integration +offered major benefits to TensorFlow users since it introduced more +intuitive readability, and portability of models while still taking +advantage of powerful backend features, Google support, and +infrastructure to deploy models on various platforms. + +### Limitations and Challenges + +TensorFlow is one of the most popular deep learning frameworks but does +have criticisms and weaknesses-- mostly focusing on usability, and +resource usage. The rapid pace of updates through its support from +Google, while advantageous, has sometimes led to issues of backward +compatibility, deprecated functions, and shifting documentation. +Additionally, even with the Keras implementation, the syntax and +learning curve of TensorFlow can be difficult for new users. One major +critique of TensorFlow is its high overhead and memory consumption due +to the range of built in libraries and support. Some of these concerns +can be addressed by using pared down versions, but can still be limiting +in resource-constrained environments. + +### PyTorch vs. TensorFlow + +PyTorch and TensorFlow have established themselves as frontrunners in +the industry. Both frameworks offer robust functionalities, but they +differ in terms of their design philosophies, ease of use, ecosystem, +and deployment capabilities. + +**Design Philosophy and Programming Paradigm:** PyTorch uses a dynamic +computational graph, termed as eager execution. This makes it intuitive +and facilitates debugging since operations are executed immediately and +can be inspected on-the-fly. In comparison, earlier versions of +TensorFlow were centered around a static computational graph, which +required the graph\'s complete definition before execution. However, +TensorFlow 2.0 introduced eager execution by default, making it more +aligned with PyTorch in this regard. PyTorch\'s dynamic nature and +Python based approach has enabled its simplicity and flexibility, +particularly for rapid prototyping. TensorFlow\'s static graph approach +in its earlier versions had a steeper learning curve; the introduction +of TensorFlow 2.0, with its Keras integration as the high-level API, has +significantly simplified the development process. + +**Deployment:** PyTorch is heavily favored in research environments, +deploying PyTorch models in production settings was traditionally +challenging. However, with the introduction of TorchScript and the +TorchServe tool, deployment has become more feasible. One of +TensorFlow\'s strengths lies in its scalability and deployment +capabilities, especially on embedded and mobile platforms with +TensorFlow Lite. TensorFlow Serving and TensorFlow.js further facilitate +deployment in various environments, thus giving it a broader reach in +the ecosystem. + +**Performance:** Both frameworks offer efficient hardware acceleration +for their operations. However, TensorFlow has a slightly more robust +optimization workflow, such as the XLA (Accelerated Linear Algebra) +compiler, which can further boost performance. Its static computational +graph, in the early versions, was also advantageous for certain +optimizations. + +**Ecosystem:** PyTorch has a growing ecosystem with tools like +TorchServe for serving models and libraries like TorchVision, TorchText, +and TorchAudio for specific domains. As we mentioned earlier, TensorFlow +has a broad and mature ecosystem. TensorFlow Extended (TFX) provides an +end-to-end platform for deploying production machine learning pipelines. +Other tools and libraries include TensorFlow Lite, TensorFlow.js, +TensorFlow Hub, and TensorFlow Serving. + +Here's a summarizing comparative analysis: + +| Feature/Aspect | PyTorch | TensorFlow | +|-----------------------------|------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------| +| Design Philosophy | Dynamic computational graph (eager execution) | Static computational graph (early versions); Eager execution in TensorFlow 2.0 | +| Deployment | Traditionally challenging; Improved with TorchScript & TorchServe | Scalable, especially on embedded platforms with TensorFlow Lite | +| Performance & Optimization | Efficient GPU acceleration | Robust optimization with XLA compiler | +| Ecosystem | TorchServe, TorchVision, TorchText, TorchAudio | TensorFlow Extended (TFX), TensorFlow Lite, TensorFlow.js, TensorFlow Hub, TensorFlow Serving | +| Ease of Use | Preferred for its Pythonic approach and rapid prototyping | Initially steep learning curve; Simplified with Keras in TensorFlow 2.0 | + + +## Basic Framework Components + +### Tensor data structures + +To understand tensors, let us start from the familiar concepts in linear +algebra. Vectors can be represented as a stack of numbers in a +1-dimensional array. Matrices follow the same idea, and one can think of +them as many vectors being stacked on each other, making it 2 +dimensional. Higher dimensional tensors work the same way. A +3-dimensional tensor is simply a set of matrices stacked on top of each +other in another direction. The figure below demonstrates this step. +Therefore, vectors and matrices can be considered special cases of +tensors, with 1D and 2D dimensions respectively. + +![](images_ml_frameworks/image2.png){width="3.9791666666666665in" +height="1.9672287839020122in"} + +Source: +[[https://medium.com/mlait/tensors-representation-of-data-in-neural-networks-bbe8a711b93b]{.underline}](https://medium.com/mlait/tensors-representation-of-data-in-neural-networks-bbe8a711b93b) + +Defining formally, in machine learning, tensors are a multi-dimensional +array of numbers. The number of dimensions defines the rank of the +tensor. As a generalization of linear algebra, the study of tensors is +called multilinear algebra. There are noticeable similarities between +matrices and higher ranked tensors. First, it is possible to extend the +definitions given in linear algebra to tensors, such as with +eigenvalues, eigenvectors, and rank (in the linear algebra sense) . +Furthermore, with the way that we have defined tensors, it is possible +to turn higher dimensional tensors into matrices. This turns out to be +very critical in practice, as multiplication of abstract representations +of higher dimensional tensors are often completed by first converting +them into matrices for multiplication. + +(APMTH 210 Lecture Notes, CS242 Lecture Notes) + +Tensors offer a flexible data structure with its ability to represent +data in higher dimensions. For example, to represent color image data, +for each of the pixel values (in 2 dimensions), one needs the color +values for red, green and blue. With tensors, it is easy to contain +image data in a single 3-dimensional tensor with each of the numbers +within it representing a certain color value in the certain location of +the image. Extending even further, if we wanted to store a series of +images, we can simply extend the dimensions such that the new dimension +(to create a 4-dimensional tensor) represents the different images that +we have. This is exactly what the famous MNIST dataset does +(https://www.tensorflow.org/datasets/community_catalog/huggingface/mnist), +loading a single 4-dimensional tensor when one calls to load the +dataset, allowing a compact representation of all the data in one place. + +### Computational graphs + +#### Graph Definition + +Computational graphs are a key component of deep learning frameworks +like TensorFlow and PyTorch. They allow us to express complex neural +network architectures in a way that can be efficiently executed and +differentiated. A computational graph consists of a directed acyclic +graph (DAG) where each node represents an operation or variable, and +edges represent data dependencies between them. + +For example, a node might represent a matrix multiplication operation, +taking two input matrices (or tensors) and producing an output matrix +(or tensor). To visualize this, consider the simple example below. The +directed acyclic graph above computes \$z = x \\times y\$, where each of +the variables are just numbers. + +![](images_ml_frameworks/image1.png){width="1.9947922134733158in" +height="1.6791174540682414in"} + +https://cedar.buffalo.edu/\~srihari/CSE676/6.5.1%20Computational%20Graphs.pdf + +Underneath the hood, the computational graphs represent abstractions for +common layers like convolutional, pooling, recurrent, and dense layers, +with data including activations, weights, biases, are represented in +tensors. Convolutional layers form the backbone of CNN models for +computer vision. They detect spatial patterns in input data through +learned filters. Recurrent layers like LSTMs and GRUs enable processing +sequential data for tasks like language translation. Attention layers +are used in transformers to draw global context from the entire input. + +Broadly speaking, layers are higher level abstractions that define +computations on top of those tensors. For example, a Dense layer +performs a matrix multiplication and addition between input/weight/bias +tensors. Note that a layer operates on tensors as inputs and outputs and +the layer itself is not a tensor. Some key differences: + +- Layers contain states like weights and biases. Tensors are + > stateless, just holding data. + +- Layers can modify internal state during training. Tensors are + > immutable/read-only. + +- Layers are higher level abstractions. Tensors are lower level, + > directly representing data and math operations. + +- Layers define fixed computation patterns. Tensors flow between + > layers during execution. + +- Layers are used indirectly when building models. Tensors flow + > between layers during execution. + +So while tensors are a core data structure that layers consume and +produce, layers have additional functionality for defining parameterized +operations and training. While a layer configures tensor operations +under the hood, the layer itself remains distinct from the tensor +objects. The layer abstraction makes building and training neural +networks much more intuitive. This sort of abstraction enables +developers to build models by stacking these layers together, without +having to implement the layer logic themselves. For example, calling +tf.keras.layers.Conv2D in TensorFlow creates a convolutional layer. The +framework handles computing the convolutions, managing parameters, etc. +This simplifies model development, allowing developers to focus on +architecture rather than low-level implementations. Layer abstractions +utilize highly optimized implementations for performance. They also +enable portability, as the same architecture can run on different +hardware backends like GPUs and TPUs. + +In addition, computational graphs include activation functions like +ReLU, sigmoid, and tanh that are essential to neural networks and many +frameworks provide these as standard abstractions. These functions +introduce non-linearities that enable models to approximate complex +functions. Frameworks provide these as simple, pre-defined operations +that can be used when constructing models. For example, tf.nn.relu in +TensorFlow. This abstraction enables flexibility, as developers can +easily swap activation functions for tuning performance. Pre-defined +activations are also optimized by the framework for faster execution. + +In recent years, models like ResNets and MobileNets have emerged as +popular architectures, with current frameworks pre-packaging these as +computational graphs. Rather than worrying about the fine details, +developers can utilize them as a starting point, customizing as needed +by substituting layers. This simplifies and speeds up model development, +avoiding reinventing architectures from scratch. Pre-defined models +include well-tested, optimized implementations that ensure good +performance. Their modular design also enables transferring learned +features to new tasks via transfer learning. In essence, these +pre-defined architectures provide high-performance building blocks to +quickly create robust models. + +These layer abstractions, activation functions, and predefined +architectures provided by the frameworks are what constitute a +computational graph. When a user defines a layer in a framework (e.g. +tf.keras.layers.Dense()), the framework is configuring computational +graph nodes and edges to represent that layer. The layer parameters like +weights and biases become variables in the graph. The layer computations +become operation nodes (such as the x and y in the figure above). When +you call an activation function like tf.nn.relu(), the framework adds a +ReLU operation node to the graph. Predefined architectures are just +pre-configured subgraphs that can be inserted into your model\'s graph. +Thus, model definition via high-level abstractions creates a +computational graph. The layers, activations, and architectures we use +become graph nodes and edges. + +When we define a neural network architecture in a framework, we are +implicitly constructing a computational graph. The framework uses this +graph to determine operations to run during training and inference. +Computational graphs bring several advantages over raw code and that's +one of the core functionalities that is offered by a good ML framework: + +- Explicit representation of data flow and operations + +- Ability to optimize graph before execution + +- Automatic differentiation for training + +- Language agnosticism - graph can be translated to run on GPUs, TPUs, + > etc + +- Portability - graph can be serialized, saved, and restored later + +Computational graphs are the fundamental building blocks of ML +frameworks. Model definition via high-level abstractions creates a +computational graph. The layers, activations, and architectures we use +become graph nodes and edges. The framework compilers and optimizers +operate on this graph to generate executable code. Essentially, the +abstractions provide a developer-friendly API for building computational +graphs. Under the hood, it\'s still graphs all the way down! So while +you may not directly manipulate graphs as a framework user, they enable +your high-level model specifications to be efficiently executed. The +abstractions simplify model-building while computational graphs make it +possible. + +#### Static vs. Dynamic Graphs + +Deep learning frameworks have traditionally followed one of two +approaches for expressing computational graphs. + +**Static graphs (declare-then-execute):** With this model, the entire +computational graph must be defined upfront before it can be run. All +operations and data dependencies must be specified during the +declaration phase. TensorFlow originally followed this static approach - +models were defined in a separate context, then a session was created to +run them. The benefit of static graphs is they allow more aggressive +optimization, since the framework can see the full graph. But it also +tends to be less flexible for research and interactivity. Changes to the +graph require re-declaring the full model. + +For example: + +x = tf.placeholder(tf.float32) + +y = tf.matmul(x, weights) + biases + +The model is defined separately from execution, like building a +blueprint. For TensorFlow 1.x, this is done using tf.Graph(). All ops +and variables must be declared upfront. Subsequently, the graph is +compiled and optimized before running. Execution is done later by +feeding in tensor values. + +**Dynamic graphs (define-by-run):** In contrast to declare (all) first +and then execute, the graph is built dynamically as execution happens. +There is no separate declaration phase - operations execute immediately +as they are defined. This style is more imperative and flexible, +facilitating experimentation. + +PyTorch uses dynamic graphs, building the graph on-the-fly as execution +happens. For example, consider the following code snippet, where the +graph is built as the execution is taking place: + +x = torch.randn(4,784) + +y = torch.matmul(x, weights) + biases + +In the above example, there are no separate compile/build/run phases. +Ops define and execute immediately. With dynamic graphs, definition is +intertwined with execution. This provides a more intuitive, interactive +workflow. But the downside is less potential for optimizations, since +the framework only sees the graph as it is built. + +Recently, however, the distinction has blurred as frameworks adopt both +modes. TensorFlow 2.0 defaults to dynamic graph mode, while still +letting users work with static graphs when needed. Dynamic declaration +makes frameworks easier to use, while static models provide optimization +benefits. The ideal framework offers both options. + +Static graph declaration provides optimization opportunities but less +interactivity. While dynamic execution offers flexibility and ease of +use, it may have performance overhead. Here is a table comparing the +pros and cons of static vs dynamic execution graphs: + +| Execution Graph | Pros | Cons | +|-------------------------------|------------------------------------------------------------------------------------------------|-----------------------------------------------------------------| +| Static (Declare-then-execute) | - Enable graph optimizations by seeing full model ahead of time
- Can export and deploy frozen graphs
classification. -- High-level vs. low-level frameworks -- Static vs. dynamic computation graph frameworks -- Plot showing number of different frameworks and shrinking +- Precision - Of positive predictions, how many were actually + > positive. Useful for imbalanced datasets. -## Types of AI Frameworks +- Recall - Of actual positives, how many did we predict correctly. + > Measures completeness. -- Cloud-based AI frameworks -- Edge AI frameworks -- TinyML frameworks +- F1-score - Harmonic mean of precision and recall. Combines both + > metrics. -## Popular AI Frameworks +- AUC-ROC - Area under ROC curve. Used for classification threshold + > analysis. -Explanation: Discuss the most common types of ML frameworks available and provide a high-level overview, so that we can set into motion what makes embedded ML frameworks unique. +- MAP - Mean Average Precision. Evaluates ranked predictions in + > retrieval/detection. -- TensorFlow, PyTorch, Keras, ONNX Runtime, Scikit-learn -- Key Features and Advantages -- API and Programming Paradigms -- Table comparing the different frameworks +- Confusion Matrix - Matrix that shows the true positives, true + > negatives, false positives, and false negatives. Provides a more + > detailed view of classification performance. -## Basic Components +These metrics quantify model performance on validation data for +comparison. -- Computational graphs -- Tensor data structures -- Distributed training -- Model optimizations -- Code generation -- Differentiable programming -- Hardware acceleration support (GPUs, TPUs) +##### Visualization + +Visualization tools provide insight into models: + +- Loss curves - Plot training and validation loss over time to spot + > overfitting. + +![](images_ml_frameworks/image7.png){width="3.5052088801399823in" +height="2.3043503937007874in"} + +Example of a loss curve from TensorBoard + +Source: +[[https://www.tensorflow.org/tensorboard/scalars_and_keras]{.underline}](https://www.tensorflow.org/tensorboard/scalars_and_keras) + +- Activation grids - Illustrate features learned by convolutional + > filters. + +- Projection - Reduce dimensionality for intuitive visualization. + +- Precision-recall curves - Assess classification tradeoffs. + +Tools like TensorBoard (TensorFlow) and TensorWatch (PyTorch) enable +real-time metrics and visualization during training. + +### Differentiable programming + +With the machine learning training methods such as backpropagation +relying on the change in the loss function with respect to the change in +weights (which essentially is the definition of derivatives), the +ability to quickly and efficiently train large machine learning models +rely on the computer's ability to take derivatives. This makes +differentiable programming one of the most important elements of a +machine learning framework. + +There are primarily four methods that we can use to make computers take +derivatives. First, we can manually figure out the derivatives by hand +and input them to the computer. One can see that this would quickly +become a nightmare with many layers of neural networks, if we had to +compute all the derivatives in the backpropagation steps by hand. +Another method is symbolic differentiation using computer algebra +systems such as Mathematica, but this can introduce a layer of +inefficiency, as there needs to be a level of abstraction to take +derivatives. Numerical derivatives, the practice of approximating +gradients using finite difference methods, suffer from many problems +including high computational costs, and larger grid size can lead to a +significant amount of errors. This leads to automatic differentiation, +which exploits the primitive functions that computers use to represent +operations to obtain an exact derivative. With automatic +differentiation, computational complexity of computing the gradient is +proportional to computing the function itself. Intricacies of automatic +differentiation are not dealt with by end users now, but resources to +learn more can be found widely, such as from +[[here]{.underline}](https://www.cs.toronto.edu/~rgrosse/courses/csc321_2018/slides/lec10.pdf). +Automatic differentiation and differentiable programming today is +ubiquitous and is done efficiently and automatically by modern machine +learning frameworks. + +### Hardware Acceleration + +The trend to continuously train and deploy larger machine learning +models has essentially made hardware acceleration support a necessity +for machine learning platforms. Deep layers of neural networks require +many matrix multiplications, which attracts hardware that can compute +matrix operations fast and in parallel. In this landscape, two types of +hardware architectures, the [[GPU and +TPU]{.underline}](https://cloud.google.com/tpu/docs/intro-to-tpu), have +emerged as leading choices for training machine learning models. + +The use of hardware accelerators began with +[[AlexNet]{.underline}](https://proceedings.neurips.cc/paper_files/paper/2012/file/c399862d3b9d6b76c8436e924a68c45b-Paper.pdf), +which paved the way for future works to utilize GPUs as hardware +accelerators for training computer vision models. GPUs, or Graphics +Processing Units, + +excel in handling a large number of computations at once, making them +ideal for the matrix operations that are central to neural network +training. Their architecture, designed for rendering graphics, turns out +to be perfect for the kind of mathematical operations required in +machine learning. While they are very useful for machine learning tasks +and have been implemented in many hardware platforms, GPU's are still +general purpose in that they can be used for other applications. + +On the other hand, [[Tensor Processing +Units]{.underline}](https://cloud.google.com/tpu/docs/intro-to-tpu) +(TPU) are hardware units designed specifically for neural networks. They +focus on the multiply and accumulate (MAC) operation, and their hardware +essentially consists of a large hardware matrix that contains elements +efficiently computing the MAC operation. This, called the [[systolic +array +architecture]{.underline}](https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1653825), +was pioneered in 1979 by HT Kung and Charles E. Leiserson, but has +proven to be a useful structure to efficiently compute matrix products +and other operations within neural networks (such as convolutions). + +While TPU's can drastically reduce training times, it also has +disadvantages. For example, many operations within the machine learning +frameworks (primarily TensorFlow here since the TPU directly integrates +with it) are not supported with the TPU's. It also cannot support custom +custom operations from the machine learning frameworks, and the network +design must closely align to the hardware capabilities. + +Today, NVIDIA GPUs dominate training, aided by software libraries like +[[CUDA]{.underline}](https://developer.nvidia.com/cuda-toolkit), +[[cuDNN]{.underline}](https://developer.nvidia.com/cudnn), and +[[TensorRT.]{.underline}](https://developer.nvidia.com/tensorrt#:~:text=NVIDIA%20TensorRT%2DLLM%20is%20an,knowledge%20of%20C%2B%2B%20or%20CUDA.) +Frameworks also tend to include optimizations to maximize performance on +these hardware types, like pruning unimportant connections and fusing +layers. Combining these techniques with hardware acceleration provides +greater efficiency. For inference, hardware is increasingly moving +towards optimized ASICs and SoCs. Google\'s TPUs accelerate models in +data centers. Apple, Qualcomm, and others now produce AI-focused mobile +chips. The NVIDIA Jetson family targets autonomous robots. ## Advanced Features -- AutoML, No-Code/Low-Code ML -- Transfer learning -- Federated learning -- Model conversion -- Distributed training -- End-to-End ML Platforms +### Distributed training + +As machine learning models have become larger over the years, it has +become essential for large models to utilize multiple computing nodes in +the training process. This process, called distributed learning, has +allowed for higher training capabilities, but has also imposed +challenges in implementation. + +We can consider three different ways to spread the work of training +machine learning models to multiple computing nodes. Input data +partitioning, referring to multiple processors running the same model on +different input partitions. This is the easiest to implement that is +available for many machine learning frameworks. The more challenging +distribution of work comes with model parallelism, which refers to +multiple computing nodes working on different parts of the model, and +pipelined model parallelism, which refers to multiple computing nodes +working on different layers of the model on the same input. The latter +two mentioned here are active research areas. + +ML frameworks that support distributed learning include TensorFlow +(through its +[[tf.distribute]{.underline}](https://www.tensorflow.org/api_docs/python/tf/distribute) +module), PyTorch (through its +[[torch.nn.DataParallel]{.underline}](https://pytorch.org/docs/stable/generated/torch.nn.DataParallel.html) +and +[[torch.nn.DistributedDataParallel]{.underline}](https://pytorch.org/docs/stable/generated/torch.nn.parallel.DistributedDataParallel.html) +modules), and MXNet (through its +[[gluon]{.underline}](https://mxnet.apache.org/versions/1.9.1/api/python/docs/api/gluon/index.html) +API). + +### Model Conversion + +Machine learning models have various methods to be represented in order +to be used within different frameworks and for different device types. +For example, a model can be converted to be compatible with inference +frameworks within the mobile device. The default format for TensorFlow +models is checkpoint files containing weights and architectures, which +are needed in case we have to retrain the models. But for mobile +deployment, models are typically converted to TensorFlow Lite format. +TensorFlow Lite uses a compact flatbuffer representation and +optimizations for fast inference on mobile hardware, discarding all the +unnecessary baggage associated with training metadata such as checkpoint +file structures. + +The default format for TensorFlow models is checkpoint files containing +weights and architectures. For mobile deployment, models are typically +converted to TensorFlow Lite format. TensorFlow Lite uses a compact +flatbuffer representation and optimizations for fast inference on mobile +hardware. + +Model optimizations like quantization (see Optimizations chapter) can +further optimize models for target architectures like mobile. This +reduces precision of weights and activations to uint8 or int8 for a +smaller footprint and faster execution with supported hardware +accelerators. For post-training quantization, TensorFlow\'s converter +handles analysis and conversion automatically. + +Frameworks like TensorFlow simplify deploying trained models to mobile +and embedded IoT devices through easy conversion APIs for TFLite format +and quantization. Ready-to-use conversion enables high performance +inference on mobile without manual optimization burden. Besides TFLite, +other common targets include TensorFlow.js for web deployment, +TensorFlow Serving for cloud services, and TensorFlow Hub for transfer +learning. TensorFlow\'s conversion utilities handle these scenarios to +streamline end-to-end workflows. + +More information about model conversion in TensorFlow is linked +[[here]{.underline}](https://www.tensorflow.org/lite/models/convert). + +### AutoML, No-Code/Low-Code ML + +In many cases, machine learning can have a relatively high barrier of +entry compared to other fields. To successfully train and deploy models, +one needs to have a critical understanding of a variety of disciplines, +from data science (data processing, data cleaning), model structures +(hyperparameter tuning, neural network architecture), hardware +(acceleration, parallel processing), and more depending on the problem +at hand. The complexity of these problems have led to the introduction +to frameworks such as AutoML, which aims to make "Machine learning +available for non-Machine Learning exports" and to "automate research in +machine learning". They have constructed AutoWEKA, which aids in the +complex process of hyperparameter selection, as well as Auto-sklearn and +Auto-pytorch, an extension of AutoWEKA into the popular sklearn and +PyTorch Libraries. + +While these works of automating parts of machine learning tasks are +underway, others have focused on constructing machine learning models +easier by deploying no-code/low code machine learning, utilizing a drag +and drop interface with an easy to navigate user interface. Companies +such as Apple, Google, and Amazon have already created these easy to use +platforms to allow users to construct machine learning models that can +integrate to their ecosystem. + +These steps to remove barrier to entry continue to democratize machine +learning and make it easier to access for beginners and simplify +workflow for experts. + +### Advanced Learning Methods + +##### Transfer Learning + +Transfer learning is the practice of using knowledge gained from a +pretrained model to train and improve performance of a model that is for +a different task. For example, datasets that have been trained on +ImageNet datasets such as MobileNet and ResNet can help classify other +image datasets. To do so, one may freeze the pretrained model, utilizing +it as a feature extractor to train a much smaller model that is built on +top of the feature extraction. One can also fine tune the entire model +to fit the new task. Transfer learning has a series of challenges, in +that the modified model may not be able to conduct its original tasks +after transfer learning. Papers such as [["Learning without +Forgetting"]{.underline}](https://browse.arxiv.org/pdf/1606.09282.pdf) +paper aims to address these challenges and have been implemented in +modern machine learning platforms. + +##### Federated Learning -## Embedded AI Constraints +Consider the problem of labeling items that are present in a photo from +personal devices. One may consider moving the image data from the +devices to a central server, where a single model will train Using these +image data provided by the devices. However, this presents many +potential challenges. First, with many devices one needs a massive +network infrastructure to move and store data from these devices to a +central location. With the number of devices that are present today this +is often not feasible, and very costly. Furthermore, there are privacy +challenges associated with moving personal data, such as Photos central +servers. -Explanation: Describe the constraints of embedded systems, referring to the previous chapters, and remind readers about the challenges and why we need to consider creating lean and efficient solutions. +[[Federated learning]{.underline}](https://arxiv.org/abs/1602.05629) is +a form of distributed computing that resolves these issues by +distributing the models into personal devices for them to be trained on +device. At the beginning, a base global model is trained on a central +server to be distributed to all devices. Using this base model, the +devices individually compute the gradients and send them back to the +central hub. Intuitively this is the transfer of model parameters +instead of the data itself. This innovative approach allows the model to +be trained with many different datasets (which, in our example, would be +the set of images that are on personal devices), without the need to +transfer a large amount of potentially sensitive data. However, +federated learning also comes with a series of challenges. -### Hardware +Training in machine learning usually assumes the data to be identically +and independently distributed. However, the nature of federated learning +makes it so that this is not necessarily the case. In our example, +devices may have similar photos from being at the same location, and the +types of the cameras that each of the devices have may be different. +There are many different techniques to compensate for this, such as +adding a proximal term to achieve a balance between the local and global +model, and adding a frozen [[global hypersphere +classifier]{.underline}](https://arxiv.org/abs/2207.09413). -- Memory Usage -- Processing Power -- Energy Efficiency -- Storage Limitations -- Hardware Diversity +## Framework Specialization -### Software +Thus far, we have talked about ML frameworks generally. However, +typically frameworks are optimized based on the target environment\'s +computational capabilities and application requirements, ranging from +the cloud to the edge to tiny devices. Choosing the right framework is +crucial based on the target environment for deployment. This section +provides an overview of the major types of AI frameworks tailored for +cloud, edge, and tinyML environments to help understand the similarities +and differences between these different ecosystems. -- Library Dependency -- Lack of OS +### Cloud + +Cloud-based AI frameworks assume access to ample computational power, +memory, and storage resources in the cloud. They generally support both +training and inference. Cloud-based AI frameworks are suited for +applications where data can be sent to the cloud for processing, such as +cloud-based AI services, large-scale data analytics, and web +applications. Popular cloud AI frameworks include the ones we mentioned +earlier such as TensorFlow, PyTorch, MXNet, Keras, and others. These +frameworks utilize technologies like GPUs, TPUs, distributed training, +and AutoML to deliver scalable AI. Concepts like model serving, MLOps, +and AIOps relate to the operationalization of AI in the cloud. Cloud AI +powers services like Google Cloud AI and enables transfer learning using +pre-trained models. + +### Edge + +Edge AI frameworks are tailored for deploying AI models on edge devices, +such as IoT devices, smartphones, and edge servers. Edge AI frameworks +are optimized for devices with moderate computational resources, +offering a balance between power and performance. Edge AI frameworks are +ideal for applications requiring real-time or near-real-time processing, +including robotics, autonomous vehicles, and smart devices. Key edge AI +frameworks include TensorFlow Lite, PyTorch Mobile, CoreML, and others. +They employ optimizations like model compression, quantization, and +efficient neural network architectures. Hardware support includes CPUs, +GPUs, NPUs and accelerators like the Edge TPU. Edge AI enables use cases +like mobile vision, speech recognition, and real-time anomaly detection. + +### Embedded + +TinyML frameworks are specialized for deploying AI models on extremely +resource-constrained devices, specifically microcontrollers and sensors +within the IoT ecosystem. TinyML frameworks are designed for devices +with severely limited resources, emphasizing minimal memory and power +consumption. TinyML frameworks are specialized for use cases on +resource-constrained IoT devices for applications such as predictive +maintenance, gesture recognition, and environmental monitoring. Major +tinyML frameworks include TensorFlow Lite Micro, uTensor, and ARM NN. +They optimize complex models to fit within kilobytes of memory through +techniques like quantization-aware training and reduced precision. +TinyML allows intelligent sensing across battery-powered devices, +enabling collaborative learning via federated learning. The choice of +framework involves balancing model performance and computational +constraints of the target platform, whether cloud, edge or tinyML. Here +is a summary table comparing the major AI frameworks across cloud, edge, +and tinyML environments: + + +| Framework Type | Examples | Key Technologies | Use Cases | +|----------------|-----------------------------------|-------------------------------------------------------------------------|------------------------------------------------------| +| Cloud AI | TensorFlow, PyTorch, MXNet, Keras | GPUs, TPUs, distributed training, AutoML, MLOps | Cloud services, web apps, big data analytics | +| Edge AI | TensorFlow Lite, PyTorch Mobile, Core ML | Model optimization, compression, quantization, efficient NN architectures | Mobile apps, robots, autonomous systems, real-time processing | +| TinyML | TensorFlow Lite Micro, uTensor, ARM NN | Quantization-aware training, reduced precision, neural architecture search | IoT sensors, wearables, predictive maintenance, gesture recognition | + + +**Key differences:** + +- Cloud AI leverages massive computational power for complex models + > using GPUs/TPUs and distributed training + +- Edge AI optimizes models to run locally on resource-constrained edge + > devices. + +- TinyML fits models into extremely low memory and compute + > environments like microcontrollers ## Embedded AI Frameworks -Explanation: Now, discuss specifically about the unique embedded AI frameworks that are available and why they are special, etc. +### Resource Constraints + +Embedded systems face severe resource constraints that pose unique +challenges for deploying machine learning models compared to traditional +computing platforms. For example, microcontroller units (MCUs) commonly +used in IoT devices often have: + +- **RAM** in the range of tens of kilobytes to a few megabytes. The + > popular ESP8266 MCU has around 80KB RAM available to developers. + > This contrasts with 8GB or more on typical laptops and desktops + > today. + +- **Flash storage** ranging from hundreds of kilobytes to a few + > megabytes. The Arduino Uno microcontroller provides just 32KB of + > storage for code. Standard computers today have disk storage in + > the order of terabytes. + +- **Processing power** from just a few MHz to approximately 200MHz. + > The ESP8266 operates at 80MHz. This is several orders of magnitude + > slower than multi-GHz multi-core CPUs in servers and high-end + > laptops. + +These tight constraints make training machine learning models directly +on microcontrollers infeasible in most cases. The limited RAM precludes +handling large datasets for training. Energy usage for training would +also quickly deplete battery-powered devices. Instead, models are +trained on resource-rich systems and deployed on microcontrollers for +optimized inference. But even inference poses challenges: + +1. **Model Size:** AI models are too large to fit on embedded and IoT + > devices. This necessitates the need for model compression + > techniques, such as quantization, pruning, and knowledge + > distillation. Additionally, as we will see in the Embedded AI + > Frameworks section, many of the frameworks used by developers for + > AI development have large amounts of overhead, and built in + > libraries that embedded systems can't support. + +2. **Complexity of Tasks:** With only tens of KBs to a few MBs of RAM, + > IoT devices and embedded systems are constrained in the complexity + > of tasks they can handle. Tasks that require large datasets or + > sophisticated algorithms-- for example LLMs-- which would run + > smoothly on traditional computing platforms, might be infeasible + > on embedded systems without compression or other optimization + > techniques due to memory limitations. + +3. **Data Storage and Processing:** Embedded systems often process data + > in real-time and might not store large amounts of data locally. + > Conversely, traditional computing systems can hold and process + > large datasets in memory, enabling faster data operations and + > analysis as well as real-time updates. + +4. **Security and Privacy:** Limited memory also restricts the + > complexity of security algorithms and protocols, data encryption, + > reverse engineering protections, and more that can be implemented + > on the device. This can potentially make some IoT devices more + > vulnerable to attacks. + +Consequently, specialized software optimizations and ML frameworks +tailored for microcontrollers are necessary to work within these tight +resource bounds. Clever optimization techniques like quantization, +pruning and knowledge distillation compress models to fit within limited +memory (see Optimizations section). Learnings from neural architecture +search help guide model designs. + +Hardware improvements like dedicated ML accelerators on microcontrollers +also help alleviate constraints. For instance, Qualcomm\'s Hexagon DSP +provides acceleration for TensorFlow Lite models on Snapdragon mobile +chips. Google\'s Edge TPU packs ML performance into a tiny ASIC for edge +devices. ARM Ethos-U55 offers efficient inference on Cortex-M class +microcontrollers. These customized ML chips unlock advanced capabilities +for resource-constrained applications. + +Generally, due to the limited processing power, it\'s almost always +infeasible to train AI models on IoT or embedded systems. Instead, +models are trained on powerful traditional computers (often with GPUs) +and then deployed on the embedded device for inference. TinyML +specifically deals with this, ensuring models are lightweight enough for +real-time inference on these constrained devices. + +### Frameworks & Libraries + +Embedded AI frameworks are software tools and libraries designed to +enable artificial intelligence (AI) and machine learning (ML) +capabilities on embedded systems. These frameworks are essential for +bringing AI to IoT (Internet of Things) devices, robotics, and other +edge computing platforms and they are designed to work where +computational resources, memory, and power consumption are limited. + +#### Challenges + +While embedded systems present an enormous opportunity for deploying +machine learning to enable intelligent capabilities at the edge, these +resource-constrained environments also pose significant challenges. +Unlike typical cloud or desktop environments rich with computational +resources, embedded devices introduce severe constraints around memory, +processing power, energy efficiency, and specialized hardware. As a +result, existing machine learning techniques and frameworks designed for +server clusters with abundant resources do not directly translate to +embedded systems. This section uncovers some of the challenges and +opportunities for embedded systems and ML frameworks. + +##### Fragmented Ecosystem + +The lack of a unified ML framework led to a highly fragmented ecosystem. +Engineers at companies like STMicroelectronics, NXP Semiconductors, and +Renesas had to develop custom solutions tailored to their specific +microcontroller and DSP architectures. These ad-hoc frameworks required +extensive manual optimization for each low-level hardware platform. This +made porting models extremely difficult, requiring redevelopment for new +Arm, RISC-V or proprietary architectures. + +##### Disparate Hardware Needs + +Without a shared framework, there was no standard way to assess +hardware\'s capabilities. Vendors like Intel, Qualcomm and NVIDIA +created integrated solutions blending model, software and hardware +improvements. This made it hard to discern the sources of performance +gains - whether new chip designs like Intel\'s low-power x86 cores or +software optimizations were responsible. A standard framework was needed +so vendors could evaluate their hardware\'s capabilities in a fair, +reproducible way. + +##### Lack of Portability + +Adapting models trained in common frameworks like TensorFlow or PyTorch +to run efficiently on microcontrollers was very challenging without +standardized tools. It required time-consuming manual translation of +models to run on specialized DSPs from companies like CEVA or low-power +Arm M-series cores. There were no turnkey tools enabling portable +deployment across different architectures. + +##### Incomplete Infrastructure + +The infrastructure to support key model development workflows was +lacking. There was minimal support for compression techniques to fit +large models within constrained memory budgets. Tools for quantization +to lower precision for faster inference were missing. Standardized APIs +for integration into applications were incomplete. Essential +functionality like on-device debugging, metrics, and performance +profiling was absent. These gaps increased the cost and difficulty of +embedded ML development. -- TensorFlow Lite -- ONNX Runtime -- MicroPython -- CMSIS-NN -- Edge Impulse -- Others (briefly mention some less common but significant frameworks) +##### No Standard Benchmark + +Without unified benchmarks, there was no standard way to assess and +compare the capabilities of different hardware platforms from vendors +like NVIDIA, Arm and Ambiq Micro. Existing evaluations relied on +proprietary benchmarks tailored to showcased strengths of particular +chips. This made it impossible to objectively measure hardware +improvements in a fair, neutral manner. + +##### Minimal Real-World Testing + +Much of the benchmarks relied on synthetic data. Rigorously testing +models on real-world embedded applications was difficult without +standardized datasets and benchmarks. This raised questions on how +performance claims would translate to real-world usage. More extensive +testing was needed to validate chips in actual use cases. + +The lack of shared frameworks and infrastructure slowed TinyML adoption, +hampering the integration of ML into embedded products. Recent +standardized frameworks have begun addressing these issues through +improved portability, performance profiling, and benchmarking support. +But ongoing innovation is still needed to enable seamless, +cost-effective deployment of AI to edge devices. + +Summary + +The absence of standardized frameworks, benchmarks, and infrastructure +for embedded ML has traditionally hampered adoption. However, recent +progress has been made in developing shared frameworks like TensorFlow +Lite Micro and benchmark suites like MLPerf Tiny that aim to accelerate +the proliferation of TinyML solutions. But overcoming the fragmentation +and difficulty of embedded deployment remains an ongoing process. + +#### Examples + +Machine learning deployment on microcontrollers and other embedded +devices often requires specially optimized software libraries and +frameworks to work within the tight constraints of memory, compute, and +power. Several options exist for performing inference on such +resource-limited hardware, each with their own approach to optimizing +model execution. This section will explore the key characteristics and +design principles behind TFLite Micro, TinyEngine, and CMSIS-NN, +providing insight into how each framework tackles the complex problem of +high-accuracy yet efficient neural network execution on +microcontrollers. They showcase different approaches for implementing +efficient TinyML frameworks. + +The table summarizes the key differences and similarities between these +three specialized machine learning inference frameworks for embedded +systems and microcontrollers. + +| Framework | TensorFlow Lite Micro | TinyEngine | CMSIS-NN | +|------------------------|:----------------------------:|:--------------------------------------:|:--------------------------------------:| +| **Approach** | Interpreter-based | Static compilation | Optimized neural network kernels | +| **Hardware Focus** | General embedded devices | Microcontrollers | ARM Cortex-M processors | +| **Arithmetic Support** | Floating point | Floating point, fixed point | Floating point, fixed point | +| **Model Support** | General neural network models| Models co-designed with TinyNAS | Common neural network layer types | +| **Code Footprint** | Larger due to inclusion of interpreter and ops | Small, includes only ops needed for model | Lightweight by design | +| **Latency** | Higher due to interpretation overhead | Very low due to compiled model | Low latency focus | +| **Memory Management** | Dynamically managed by interpreter | Model-level optimization | Tools for efficient allocation | +| **Optimization Approach** | Some code generation features | Specialized kernels, operator fusion | Architecture-specific assembly optimizations | +| **Key Benefits** | Flexibility, portability, ease of updating models | Maximizes performance, optimized memory usage | Hardware acceleration, standardized API, portability | + + +In the following sections, we will dive into understanding each of these +in greater detail. + +##### TFLM (Interpreter) + +TensorFlow Lite Micro (TFLM) is a machine learning inference framework +designed for embedded devices with limited resources. It uses an +interpreter to load and execute machine learning models, which provides +flexibility and ease of updating models in the field. + +Traditional interpreters often have significant branching overhead, +which can reduce performance. However, machine learning model +interpretation benefits from the efficiency of long-running kernels, +where each kernel runtime is relatively large and helps mitigate +interpreter overhead. + +An alternative to an interpreter-based inference engine is to generate +native code from a model during export. This can improve performance, +but it sacrifices portability and flexibility, as the generated code +needs recompilation for each target platform and must be replaced +entirely to modify a model. + +TFLM strikes a balance between the simplicity of code compilation and +the flexibility of an interpreter-based approach by incorporating +certain code-generation features. For example, the library can be +constructed solely from source files, offering much of the compilation +simplicity associated with code generation while retaining the benefits +of an interpreter-based model execution framework. + +An interpreter-based approach offers several benefits over code +generation for machine learning inference on embedded devices: + +- Flexibility: Models can be updated in the field without recompiling + > the entire application. + +- Portability: The interpreter can be used to execute models on + > different target platforms without porting the code. + +- Memory efficiency: The interpreter can share code across multiple + > models, reducing memory usage. + +- Ease of development: Interpreters are easier to develop and maintain + > than code generators. + +TensorFlow Lite Micro is a powerful and flexible framework for machine +learning inference on embedded devices. Its interpreter-based approach +offers several benefits over code generation, including flexibility, +portability, memory efficiency, and ease of development. + +##### TinyEngine (Compiler-based) + +TinyEngine is an ML inference framework designed specifically for +resource-constrained microcontrollers. It employs several optimizations +to enable high-accuracy neural network execution within the tight +constraints of memory, compute, and storage on microcontrollers. + +While inference frameworks like TFLite Micro use interpreters to execute +the neural network graph dynamically at runtime, this adds significant +overhead in terms of memory usage to store metadata, interpretation +latency, and lack of optimizations, although TFLite argues that the +overhead is small. TinyEngine eliminates this overhead by employing a +code generation approach. During compilation, it analyzes the network +graph and generates specialized code to execute just that model. This +code is natively compiled into the application binary, avoiding runtime +interpretation costs. + +Conventional ML frameworks schedule memory per layer, trying to minimize +usage for each layer separately. TinyEngine does model-level scheduling +instead, analyzing memory usage across layers. It allocates a common +buffer size based on the max memory needs of all layers. This buffer is +then shared efficiently across layers to increase data reuse. + +TinyEngine also specializes the kernels for each layer through +techniques like tiling, unrolling, and fusing operators. For example, it +will generate unrolled compute kernels with the exact number of loops +needed for a 3x3 or 5x5 convolution. These specialized kernels extract +maximum performance from the microcontroller hardware. It uses depthwise +convolutions that are optimized to minimize memory allocations by +computing each channel\'s output in-place over the input channel data. +This technique exploits the channel-separable nature of depthwise +convolutions to reduce peak memory size. + +Similar to TFLite Micro, the compiled TinyEngine binary only includes +ops needed for a specific model rather than all possible operations. +This results in a very small binary footprint, keeping code size low for +memory-constrained devices. + +One difference between TFLite Micro and TinyEngine is that the latter is +co-designed with "TinyNAS," an architecture search method for +microcontroller models, similar to differential NAS for +microcontrollers. The efficiency of TinyEngine allows exploring larger +and more accurate models through NAS. It also provides feedback to +TinyNAS on which models can fit within the hardware constraints. + +Through all these various custom techniques like static compilation, +model-based scheduling, specialized kernels, and co-design with NAS, +TinyEngine enables high-accuracy deep learning inference within the +tight resource constraints of microcontrollers. + +##### CMSIS-NN (Library) + +CMSIS-NN, standing for Cortex Microcontroller Software Interface +Standard for Neural Networks, is a software library devised by ARM. It +offers a standardized interface for deploying neural network inference +on microcontrollers and embedded systems, with a particular focus on +optimization for ARM Cortex-M processors. + +**Neural Network Kernels:** CMSIS-NN is equipped with highly efficient +kernels that handle fundamental neural network operations such as +convolution, pooling, fully connected layers, and activation functions. +It caters to a broad range of neural network models by supporting both +floating-point and fixed-point arithmetic. The latter is especially +beneficial for resource-constrained devices as it curtails memory and +computational requirements (Quantization). + +**Hardware Acceleration:** CMSIS-NN harnesses the power of Single +Instruction, Multiple Data (SIMD) instructions available on many +Cortex-M processors. This allows for parallel processing of multiple +data elements within a single instruction, thereby boosting +computational efficiency. Certain Cortex-M processors feature Digital +Signal Processing (DSP) extensions that CMSIS-NN can exploit for +accelerated neural network execution. The library also incorporates +assembly-level optimizations tailored to specific microcontroller +architectures to further enhance performance. + +**Standardized API:** CMSIS-NN offers a consistent and abstracted API +that protects developers from the complexities of low-level hardware +details. This makes the integration of neural network models into +applications simpler. It may also encompass tools or utilities for +converting popular neural network model formats into a format that is +compatible with CMSIS-NN. + +**Memory Management:** CMSIS-NN provides functions for efficient memory +allocation and management, which is vital in embedded systems where +memory resources are scarce. It ensures optimal memory usage during +inference and in some instances, allows for in-place operations to +further decrease memory overhead. + +**Portability**: CMSIS-NN is designed with portability in mind across +various Cortex-M processors. This enables developers to write code that +can operate on different microcontrollers without significant +modifications. + +**Low Latency:** CMSIS-NN minimizes inference latency, making it an +ideal choice for real-time applications where swift decision-making is +paramount. + +**Energy Efficiency:** The library is designed with a focus on energy +efficiency, making it suitable for battery-powered and +energy-constrained devices. ## Choosing the Right Framework -- Factors to consider: ease of use, community support, performance, scalability, etc. -- Integration with data engineering tools -- Integration with model optimization tools +Choosing the right machine learning framework for a given application +requires carefully evaluating models, hardware, and software +considerations. By analyzing these three aspects - models, hardware, and +software - ML engineers can select the optimal framework and customize +as needed for efficient and performant on-device ML applications. The +goal is to balance model complexity, hardware limitations, and software +integration to design a tailored ML pipeline for embedded and edge +devices. + +#### ![](images_ml_frameworks/image4.png){width="6.5in" height="3.1666666666666665in"}Model + +TensorFlow supports significantly more ops than TensorFlow Lite and +TensorFlow Lite Micro as it is typically used for research or cloud +deployment, which require a large number of and more flexibility with +operators (ops),. TensorFlow Lite supports select ops for on-device +training, whereas TensorFlow Micro does not. TensorFlow Lite also +supports dynamic shapes and quantization aware training, but TensorFlow +Micro does not. In contrast, TensorFlow Lite and TensorFlow Micro offer +native quantization tooling and support, where quantization refers to +the process of transforming an ML program into an approximated +representation with available lower precision operations. + +![](images_ml_frameworks/image5.png){width="6.5in" +height="2.0833333333333335in"} + +#### Software + +TensorFlow Lite Micro does not have OS support, while TensorFlow and +TensorFlow Lite do, in order to reduce memory overhead, make startup +times faster, and consume less energy. TensorFlow Lite Micro can be used +in conjunction with real-time operating systems (RTOS) like FreeRTOS, +Zephyr, and Mbed OS. TensorFlow Lite and TensorFlow Lite Micro support +model memory mapping, allowing models to be directly accessed from flash +storage rather than loaded into RAM, whereas TensorFlow does not. +TensorFlow and TensorFlow Lite support accelerator delegation to +schedule code to different accelerators, whereas TensorFlow Lite Micro +does not, as embedded systems tend not to have a rich array of +specialized accelerators. + +#### Hardware![](images_ml_frameworks/image3.png){width="6.5in" height="2.2083333333333335in"} + +TensorFlow Lite and TensorFlow Lite Micro have significantly smaller +base binary sizes and base memory footprints compared to TensorFlow. For +example, a typical TensorFlow Lite Micro binary is less than 200KB, +whereas TensorFlow is much larger. This is due to the +resource-constrained environments of embedded systems. TensorFlow +provides support for x86, TPUs, and GPUs like NVIDIA, AMD, and Intel. +TensorFlow Lite provides support for Arm Cortex A and x86 processors +commonly used in mobile and tablets. The latter is stripped out of all +the training logic that is not necessary for ondevice deployment. +TensorFlow Lite Micro provides support for microcontroller-focused Arm +Cortex M cores like M0, M3, M4, and M7, as well as DSPs like Hexagon and +SHARC and MCUs like STM32, NXP Kinetis, Microchip AVR. + +Selecting the appropriate AI framework is essential to ensure that +embedded systems can efficiently execute AI models. There are key +factors to consider when choosing a machine learning framework, with a +focus on ease of use, community support, performance, scalability, +integration with data engineering tools, and integration with model +optimization tools. By understanding these factors, you can make +informed decisions and maximize the potential of your machine learning +initiatives. + +#### Other Factors + +When evaluating AI frameworks for embedded systems, several other key +factors beyond models, hardware, and software should be considered. + +##### Performance + +Performance is critical in embedded systems where computational +resources are limited. Evaluate the framework\'s ability to optimize +model inference for embedded hardware. Factors such as model +quantization and hardware acceleration support play a crucial role in +achieving efficient inference. + +##### Scalability + +Scalability is essential when considering the potential growth of an +embedded AI project. The framework should support the deployment of +models on a variety of embedded devices, from microcontrollers to more +powerful processors. It should also handle both small-scale and +large-scale deployments seamlessly. -## Framework Comparison +##### Integration with Data Engineering Tools -Explanation: Provide a high-level comparison of the different frameworks based on class slides, etc. +Data engineering tools are essential for data preprocessing and pipeline +management. An ideal AI framework for embedded systems should seamlessly +integrate with these tools, allowing for efficient data ingestion, +transformation, and model training. -- Table of differences and similarities +##### Integration with Model Optimization Tools -## Trends in ML Frameworks +Model optimization is crucial to ensure that AI models are well-suited +for embedded deployment. Evaluate whether the framework integrates with +model optimization tools, such as TensorFlow Lite Converter or ONNX +Runtime, to facilitate model quantization and size reduction. -Explanation: Discuss where these ML frameworks are heading in the future. Perhaps consider discussing ML for ML frameworks? +##### Ease of Use -- Framework Developments on the Horizon -- Anticipated Innovations in the Field +The ease of use of an AI framework significantly impacts development +efficiency. A framework with a user-friendly interface and clear +documentation reduces the learning curve for developers. Consideration +should be given to whether the framework supports high-level APIs, +allowing developers to focus on model design rather than low-level +implementation details. This factor is incredibly important for embedded +systems, which have less features that typical developers might be +accustomed to. -## Challenges and Limitations +##### Community Support -Explanation: None of the frameworks are perfect, so it is important to understand their limitations and challenges. +Community support plays another essential factor. Frameworks with active +and engaged communities often have well-maintained codebases, receive +regular updates, and provide valuable forums for problem-solving. As a +result, community support plays into Ease of Use as well because it +ensures that developers have access to a wealth of resources, including +tutorials and example projects. Community support provides some +assurance that the framework will continue to be supported for future +updates. There are only a handful of frameworks that cater to TinyML +needs. Of that, TensorFlow Lite Micro is the most popular and has the +most community support. -- Model compatibility and interoperability issues -- Scalability and performance challenges -- Addressing the evolving needs of AI developers +## + +## Future Trends in ML Frameworks + +#### Decomposition + +Currently, the ML system stack consists of four abstractions, namely (1) +computational graphs, (2) tensor programs, (3) libraries and runtimes, +and (4) hardware +primitives.![](images_ml_frameworks/image8.png){width="2.557292213473316in" +height="2.9092125984251966in"} + +This has led to vertical (i.e. between abstraction levels) and +horizontal (i.e. library-driven vs. compilation-driven approaches to +tensor computation) boundaries, which hinder innovation for ML. Future +work in ML frameworks can look toward breaking these boundaries. In +December 2021, Apache TVM Unity was proposed, which aimed to facilitate +interactions between the different abstraction levels (as well as the +people behind them, such as ML scientists, ML engineers, and hardware +engineers) and co-optimize decisions in all four abstraction levels.[^1] + +#### High-Performance Compilers & Libraries + +As ML frameworks further develop, high-performance compilers and +libraries will continue to emerge. Some current examples include +[[TensorFlow +XLA]{.underline}](https://www.tensorflow.org/xla/architecture) and +Nvidia's +[[CUTLASS]{.underline}](https://developer.nvidia.com/blog/cutlass-linear-algebra-cuda/), +which accelerate linear algebra operations in computational graphs, and +Nvidia's +[[TensorRT]{.underline}](https://developer.nvidia.com/tensorrt), which +accelerates and optimizes inference. + +#### ML for ML Frameworks + +We can also use ML to improve ML frameworks in the future. Some current +uses of ML for ML frameworks include: + +- hyperparameter optimization using techniques such as Bayesian + > optimization, random search, and grid search + +- neural architecture search (NAS) to automatically search for optimal + > network architectures + +- AutoML, which as described in the Advanced Features section, + > automates the ML pipeline. ## Conclusion -- Summary of Key Takeaways -- Recommendations for Further Learning \ No newline at end of file +In summary, selecting the optimal framework requires thoroughly +evaluating options against criteria like usability, community support, +performance, hardware compatibility, and model conversion abilities. +There is no universal best solution, as the right framework depends on +the specific constraints and use case. + +For extremely resource constrained microcontroller-based platforms, +TensorFlow Lite Micro currently provides a strong starting point. Its +comprehensive optimization tooling like quantization mapping and kernel +optimizations enables high performance on devices like Arm Cortex-M and +RISC-V processors. The active developer community ensures accessible +technical support. Seamless integration with TensorFlow for training and +converting models makes the workflow cohesive. + +For platforms with more capable CPUs like Cortex-A, TensorFlow Lite for +Microcontrollers expand possibilities. They provide greater flexibility +for custom and advanced models beyond the core operators in TFLite +Micro. However, this comes at the cost of a larger memory footprint. +These frameworks are ideal for automotive systems, drones, and more +powerful edge devices that can benefit from greater model +sophistication. + +Frameworks specifically built for specialized hardware like CMSIS-NN on +Cortex-M processors can further maximize performance, but sacrifice +portability. Integrated frameworks from processor vendors tailor the +stack to their architectures. This can unlock the full potential of +their chips but lock you into their ecosystem. + +Ultimately, choosing the right framework involves finding the best match +between its capabilities and the requirements of the target platform. +This requires balancing tradeoffs between performance needs, hardware +constraints, model complexity, and other factors. Thoroughly assessing +intended models, use cases, and evaluating options against key metrics +will guide developers towards picking the ideal framework for their +embedded ML application. + +For Future (Federated Learning) + +[^1]: [[https://tvm.apache.org/2021/12/15/tvm-unity]{.underline}](https://tvm.apache.org/2021/12/15/tvm-unity) diff --git a/images_ml_frameworks/image1.png b/images_ml_frameworks/image1.png new file mode 100644 index 0000000000000000000000000000000000000000..afb2412121e41d550872c13d0d0502ca4e63f026 GIT binary patch literal 34481 zcmeFY_dnHPA3t7Fj+J9Y5e_1Il|6DCWUpf+Te4TO53+LX5RsL=LmAoGE2Ck{9?9OD z&-L!U@6Y#d_&y$=A4>Jean0B3xvo%kRe559I|Ns*Tp?C?jMBVv1qb}Px(B-f{^j6B z>)k6?UR_Z@NuxcC)>857Rf0z@WTdtQS!3U_So4NwBdtC>V0asbvLGm+KoIjY$H3Tf zo^e{FSt!4K^4x;OWvT31S<2@oAFB9k(g^&EA1x)pH|w$pYQbY_;Nn2`sRDGlE>6JAa#jZu(cg7 zE;sf&AEKuc$7>;w4o)OPVi#Toy@k&OeNaRVSuOXca$|Hp8d&cx^?gnfupIy5Ja4!6 z=er>0$cg5G)f=DvA>H8|C92h51#$HD6IGQa zZX1Y4a?lxl=V4E}G(?8VX#NS+pJe>50l*Dib(Exv;tD$*Ue z^T2Ax*!z5cqy((saLH34zcb5ppQQmIvo=^VzeQMIl0avX4z}RKI#?@(NNB2wFmW}H zclqL6`W0R6-@0O$)>liLs`^vK>aKY0eV1SF;V7-x9gv#V%u!sF?|I7>O}m`p5|!#c z<(VMq>zM#C_VQ`k~FQ zF8yi;`)hbN_X>w9tbf{V&$U;%t&cO&aUg;oD?_FG5|--__2DmSACLa$+Sq6N(K6HF zCOc_e@u#@i6FNdF49{V5RM{#`o((cr0SfZxK* zz$eRr4da0nkeK4ayy%Ag;io#e%D3ylS1#Ew3b%3VykqB%+m@p> z%W12W1-7LP!#hiTw#Pe*pOYPGSGu;_$kp^p^ec-M;#lNy)tzSIGJ86iVWGXnD7w$}-j>ct3Bo@}2ea}xkD~aY$UK7(y7JocQ zhtEA@Lfi`y##t9*7xMSS{H}GaGHWL&-PzmtJz+EQaUCbm>U*YP|pDcnKA1&<%& zBck9(9|G{j+d9EUTY_1CI@w>Huy|*sQ(gXUY;% zSyIn>-rOlhr;;I4ppC-1?l^r!CKdV%P?wR{d)~9}yL$7az_mxktvN-#0m%hC6{MV&!Z{CnGs^aHv!__Nh zwn%4LGiBX}v~@jCxj_+epzCx;5faCV*w>ld;TA#Oix>%HyJ$F^`Yg6t|8r-ddmb4! zA~O-)EnTQnpXS(j+(Ep8o)@M^!kopy_K{lRYTpln4^fXhqcf&F=821*R&T^8-YmaX zv1JuX$~RbMDsu|1orM0YPK|cOs5f_mg;T^mKZVQSJla;E0RxWI8wv^`y<4*tmaJ{pWh*w7u3GrMoyXha^=U9C|Z$Xul>J9l}|oB z3$>s8!nacpD*d9*S10Rt65`e0FMzd9e-$^%M1?Yyh!ZUq z`!r+k!d`nIgoU1&%u<=J6D{BPEYTY)BBads<8Nyjo!YV-%y|tuA6bJPnY9%*L+r*= z6CVx59G6M+6>`buFDrIyU#^&G%`S^kD*ZvxkPL0;O6-3LFds0?4m3>825 zc_~iEvYUxWbYiv*a~1aJf~RvzvakpAP}pnkRK~7`(IQYk{>}9j>?^Eq@9q=z?3YqU zGl2>G7K$X}-_j)Ns*Al2)>b6xJ$)`z%R=DSM5jK166B2|TIA9%JWmAms}iFIFG0(G z+KGD4&*_fsj4;w^XjDwN8OZ=)RH~+m9xvR+VTUEz4Sw+NOBNa^(w>{f&m?WUyzq{e zGHKZcM|eN*tOP3E?6(({@DWhmBz4=Ew6-1jSozVgZakc)@sGmqwb#E6 zPTg;#ST%v`9{FxvV$Jb*Zu{l~s0M+e-yU>9u5%AtzwwkN!M7pkfneR*XUK$&)Pt^G zZ-cz}4m#2XW0)WXA6{V^Eq%oXJRLXTlj`eRY(@gO>>Vx`tJRUxcZp<-Q8C~|FATn( zlT#)9+b1+#ady00C-VNX-R@$4#CT$|#>J50e*EbX3}flh%c7iU0uV|$=*rzbmO*X_ zTu+BNd@8hq;oG6mAVLnvxpJ%QC%n9lbKV*K74p(ae)(sxxo>etHkCTXTyyU9V10F& z@~4PWRCm(f$(ql^bmGE02qab*D*kYp;84_tq|eDwny^mO8-sq8w0b7ZT5A>S*^N-? z_UK)4fEASlSoo|v)`52*RjgV!S1$67 zg@9#$1O|&GJF`$;@!9PYlDzKO_i*IPalO-Q%fUjNTHy{WDg`=P*x8OK(pt0ju#MlQ z;`F5LMqko>PZYdoRTrbwlc(qqOhJEzxqy?A(l5m zElqrE9^MaNzEY}aCEf?g%d_1L6{a&ZjFkT0uz6$1KRu~=VKIHx9e`Btr_pXdi=TWx zljA|e3lHk#*q_WtOL-B-D>C$Els2B&crPVcihqf7l|W*T9zpy5$!QizGMjxzwnb3L zMO;+EYj1EtkF44A0(AAsKWY!(7_fPX2 zV6_8qH^&nB%*AyTnI&li?aC~A6Fj}+wZ$ur92$?GgEst8mk=COe`u|O63Dt2(V5gu zelz}fYn3BLy9D@J7L40twXQ`ahIPY6e&?=2XXFynxYy&Q+ZkY_QqV*h6Nk;U@#8$z zB=+B`h4cAFcPF{>CJx4bOnt!nFti|rwp6D?zh*X_8aH-vtRX~AQSu?ej2=3fxGc+& zG(!4$s;QG!QEvbqOx^Sb7cf1XbA3qZjt39#KFM9tU%U0HQs*zJG|@R8JX;EEYKquV zd1!s4i2c-MBRGT1ZIJ*^%%g?xf;2MrPZa>2TmVFmIl%tWnL4N z*L&`sFq-3PXURp@5A}=$!^{5_mjJk`Tdl;wsCfdD{LS8UsixnqBZeE@8JhQE>CE7u zS}nZxn<5$a9xBw{HNA2#g^=|0-+Jk(v`7lEv|=S5p(IZJWHr`%to+D{ndXkY(xQRC zT&$HB6@Oo&rSFnN>Z}ZEa^NtwU4>30>_vF3>*{l`h5LU3!KE9J-`rkNFCkMPsWDuT zL!^wldTi7z*A7}3MuY7vhkiqt0Se^6*>pJN{ri_P{8{9khjs^RV^!eLcR7-XDAR9L zMRF{S7S_bs^`s05T8@=gg_P2cnui=XR#$S2O!O?;GSMeyGh_qexqR(PJ2BVpF-e$q^sYKLoA6C~mWB7STGDW0tj_Rc zT>e~c;w@#eefOtyyWSt6f?mH~((v24Es!&_wP}=n3OT%nqtig>S?ha4}B-*vG**6iAtJNx;foo2u3@_fMI%j^wfMlNu9)&TcG zVZRb;e}1l#KFV|@c7<1`6(0VTSb{~kto7zbiV68?gJ~aq?NHBK4$Jml?Q%tn%}+ZP zN)x|Hw9EMqrJhCqc1eF|B`|1o`O<4afSD~j*<6?;mkq!dYh5GHoK8(>|NnCca@Eo| z=2iel+>d7-5Z{fcehd3M+2HNIP#qF0)8zQ+T5{dfQukH;@*>CbJY#D|Vkx?3=Tdgh z?k0;nkQn&+Thy(|i80B@o|;Um37XCqP;#WW2d(KE`&eif>Fgf{pNG`TLkC9^IyhI{ zs!RA<;@sbbVLvgDYfSUE9#&fBuIk7=zXNwuYqn5I6)Rp@ELXG`h}#`*vV0wV?e zvI5YECmWMBT!3_?F#&x+i)0?99Gvb`<=Dyo^kHML&q;^*!C;9q$RIMrp43RYXbIMPg!^Er9rx#(ufoNk739EpAqzne)kR?9-Tv z&~IjN5@{KrE?Nbz>Y#J#ZH~?#jz!%CqZQ9nuyC{ zk45#gZyl%=&X6#LF#DbY&8 zp@E;DRWtj^qyP3Nf4#rI>#8CB?8VYIYoO?CA4~zxU9V4RgoO#4cng+P+@;YS+lj2R z`|0Rc(SCD%+AQS2zKp&z`ZIm5YW0ee;8-D7U4*{c3%Dle@_Tv&Q*kZ9giiq@!gxjN z05AwKBbp+UPw~&JJO(0&JOe)KPG_!M~b>g>n*q%?AkSUe-(z{%7sd;jK zvj2M&1H_-FKxS(o2)LapTf?|5jf!EpgrpJB&p82bPg{3gEvX*=^F7OO>l2d(jC2Aj zK}fkMT5HFq^D23UnQkD#c^5m+cL+Mqp-L4)+bjXy;5IP!n7!@|L>39DFqBK~q23}> z5S9e07Kn(6`QD6^7zuB;CVweFZXCRpmZG|!483U!zJd-A4o&|8Ajc$=b|wlxS_0-y z5&U2IC5Y4D%b@%QJ!VUq1D*e7x-DtWd?D>2zefRcHrYE!@W>E1Zp64P#_Ntl;+Il5 z;0`o95NMqLM;1Og&&xH%%ts<0L0X@$OlOgR>tFQkRPvawc^#g)I9&@1Pb^a{o_v1p z`tuUN(Mhp;{~PmU{(ry+PmhHer+gGG9j=J)nTPSy%mbr2xUy;#jZnYj@dCLqew?VQ z4J5=Ao3_5LvK!;SeG!ah`ZsMXv%{YSd5ObG7heHZBC>&Bb(@X0OmY!rzw6v>Wfd4- zVej=80k|;L!xRwEt?;6u)@{9zlLs+J1ht08Zwk-HEYruHIyNk3q<~gdnU^BaF!I?_ zt!egW02wP@p?8W)p-;s=yi7LyRwW_#m#^Fndzz&;4^~IiR(6nWx1mB$-LV#O^X^~o zDX6iZPG4Rumo(lWzH@{g18OxBJ!2*=Le`B!&X0v<%=z)cRmNMbmr&RIY&j5xNrW?Y zK*+F7{jMo)6M(IfUIUR0Cd~nEm*oe75vCiI<24^VsA_MA8#6Tn7bk6bMo?ZCXl3_Og2pez#a^hagk1m9qix-iKBAxWg^HB9L5LWKi=-R|Tw`B{*}@mDWEyB7b=J zw3WAtW}~xv28(R@;vBLw<+%;YK91&r#59sC;VlIUNUD!J1iE4+Z60poWYBdg%x_u`TpjDf0SW03(+JIc4nOh=KWXs(V_s z68;>s1CH9AY1V5^|B)>Cki4cHEe05|iZnelWy*WtTexxHT0l|F)y=8UJKV%w0aTaA zoev1s+J~|+1MQL$`nl*+*i@TTj>2O^5ano zX?_B6rR3KxE(2DW9#Urj)z+J+viJF|y7Vn+Ju{qpN1JzzZ=rtf;m_-XoLu-F$FnY^Ue^*>zhJz&y#1)e$IK z2u??QGVi?kwA?J%=WM$(ts@mMC8`@#22TDE^D#2a*g;x^5$M$G^1#x+%Op1JCYoaj)8^Njb>ZfQy+;ZPVqyQ_ZAa+J z$pdGjryUJi-C8JXT3OM@+MXg!%M|C=)n{>TSn{^|-#gbBx_pR2GWHI-TKEP|d&`V2hr0 zRpZ_H$>B{2&Pp1aRU z7S90sE>U{N?KWXwU6W4`bR!^)KpVj263|YEOO2(dXxH8=Ma={<7QWKvUTbIiG1Ecg zxW}t$M6#}w0N)h_Hu0hnTiY$|`|AVbNLn7~*sv)tT#skNVYBReZvpQ5b}^eyW(s6| z0?m_Ys6dIQ;TLe67UCSFE{?Kj$_uOz9je*}IDW=0Bhv51E?9{60qWWkoEp+KUS|xe zT1>J`ThXpISyTcpw>`Mh7bk%uAg9q82zIYU8pXhpFL?cGT4Fz+Dc<)hSoR>;I$!Mf zCpr8t79O7h?@6tDs_~I{OpL5W=wu$$SC^vL}T)A+j981S;f=+opkY-QLs=DXK%q5;cM9!{=k?BtnG5m8~iZ%BI3;7t& zXpM^{WQd2it0W87IK8D;ggG-pLx_S?epp1;sCnhzp(1ak$Sh#J`l?I(ooc9~ z=TueoC^Jd=s@@o~Nld5Yv1A@C@f={igC?p?aAZcwtv>BZ?xgpdoWks? zT-Fp6f&fDQr5}o#99#Tanv($0778@L{|t@s`?z@_@zvG`#_5tilTT5~kER=aes`%n zz%jc)E~L%Ca-Oe=aDeNf3!qvXUmS#|=Q~FNnL%8S3pL9V#(`V=5ZV=5T#B?@fR{b@ zLkodW>S^!`jt)I!<8{|6Z@DI0L|pnKUn6U>C}^t?G-xSqfK9lIquk)qkM_g%AH?8s zi8iW&PUKwt$rL9#jEYN4A8T0WR#|S|Ii9T$Hz_828nvZfsM^NBa0b^|BWvT3rDiRM zVWk0GhqH!3bxPG_DOWM@N=C!$X4*{+(2{({iDOVooLg$94D!k`II8gkE+PS8n=+GS zHaWl(c2IBH>ESR$U+=v4&R|ZyMM}v--6u0lHaPTB>c}N6Bt4ko1&U%_s_G3hn>@4fH@a2AFjL@Tn*g>`siOo1`I*hxeNFCx=UvG3Y$69HQhVl^Z?N|a*M}k zZLS1!ykmqv?|Qu)h8qEFj$uC*-{-dngW;Xw&`uDv1S%4-Eh>M8M9q(N8=_Qqnxz+S zL{qHuWfEmxjcK1Ir>%eo_6$kg#gMGm1Vx9le0sGhe_vCBN&Kyj>DD;x_F_{3S_3~z zqI;U-zm?4Sh7kkqwROIPV={q9cZA8EoA(Tz7#@s%mVZhtrSU-hn1Ef57Tqiim%Dt4 zTT%;s=&!hCko5>`Qq82^(!X9B87dV!(EUEMZv^N#(MB_;q+X!+a)QH4Y6-6P$sj*r z?igIT4WF4pgBBJIBw)I+5-LHsNxhE#>o%^z{%}m>f-z~qqvXwCDx<9)8Z!{FiMkHZ z=^Tt*1B;tG3MYmU#&$Ec%R2Ga=mLBHkoCI}Aq~1L--&cZJx}#+`Wg6$X6ZGYi!vZA zd5Yr%5YQ*Uk!Ljj(1f&7B=(ibB4OP!{3iMtqpYkGgLqe5%fekbCcz=0aqq+l1$ove z#LmMTc=jsFe<<0}TW+j3_8?9|yxuvXN%0>SqwOj>&dS*bDBx-7C=MQ3*OEJzCbi@jEKhu09-|#>o1>?R7mR(hVhqTjX$u16i^Jb zsI6~SQLpad&s6L}z&~9xoPs01dHc$OVDRlAW>P%mYKNi5ApMUG^<#NznHo=cO3fUX9Gu z&TzInQEuun885VIUs`(u1KnbLF*5kmiD|+8HE&s|QNuX#d7&ykX%n>k zKth`Q_GA{BlMS#6H#Kj3p3=Jw$IKbMn`asoU+)Oo{fSTEeeKK=1==3R;nQx`9&^)MFl}MP}=Bf+~vc+ zmuh(0oc!;wL?2>Xr%9h%weq>j*O8M_l)PyX2zFolTU1&FvOv=|!^PGB^Ei`xm652! zNR6R^n&k(Tbv#=eEk6L(dJW>v)t8)17ng3lzung>M}S=3@Eqs~TQ4_x$)8!1UiBYX z(Nj4DrOFhYn;1|JkVXk0HQZ*NJa8>ANN6qLWW|^y;8H?0l+N$WIh`{rUY_!=Ufz$3 zl}A{MfwbRRgH4wgr|KU=6kyxL;Z!^aVRv8el9OuErWp4Na1B0nsA-o`#k)ys1?0q7 zOQ3l{xez!q7_0beFhiz9KALXhRQib5(5+tXZ`NJq{1}|5?nX{G+4!XrB?<8U@oE(NyG-uvS1XQ+~Nn2EBXH2uSvGaPMo>K=NVk8mK@>n0+A- z0KVM!*lNa;u#4*o=t{;^;330Y`GI#^0N^SMOVBUi=#-duMy37o)@n&HZAO;aH?^DBAmL%<3#_H(}`<;D{s|sajz{VwunZ9Q=6IBBwCi)-_(LOEn$1)MWe+2UP zNMW7Hdudy+1Gr{6lj7BDVuUfFx+W6Qxgc%818N=&0ZTS{NOW9d<5)KFM-}m-m7RadG^@v|JNw1i?^m1zFlpXr7SDfHUY01~ z10Y89*(y>6++p#O)Zt0LoMgP^uw8!=0_V_jdKpiU-}G7h1{Bu?4GL4Cc|t@cWQbA+ zCX24$b6Y4Ui3)mAs9BNSm2^_NB8j>YouPu&!`{m;gA`c+_x4IxHevVssmMK3ND4L9 zxKTmQMnjF{|H-eIt0OXoMnL2={cl!no4-fbJD&9FHAqVn$`M$hs)>%?=k?rKC4v?4!rYYklb9wzXacQn;+JbK`4^s(u}OK8G{4kjx5*z_x_LqMmkM)`j$ z@3p3z_Kkqoj|&Wl&(FQGA@GlZOe=gTjAV{t{0l9wmL8&bYTRqKYP_YNcbF3xyu-gL zH!toTw`4OBEB^S&<>8 z6m*-@uqR+wX`^=X#q#9?keb{zW-RP|V0U>r?Q>{ote%39dCg?@vg-|ELms>q0JqL8 z$VA_6s@^lr5@oTtl=WY%D3Sr%!1s7j=hHI< zY@7meZqB$5f^}w4NoF#6 zWmNMbrvVg62a3*B6lD&`)k!wUKYR{UQ29owt`Bnqj}_>J98z>`^EG3vR0-dlzbwZs zbpcLGT4nIef*)zb+3l%MUDVf%q>OF9X9m;sC3&Xq~R+v5S< zb$&(nuP>VqevmdQkxAjd8mf`XPTAqw$Q^_=hVHZoI$?BO&O3}9pYndkX?H*i3QAF& z1el}Rnk?uM_XD)qVX(%_X_Oo_K7W;meSHT*%Cva@ggRMT95aGaisjp- z6uG-WQv&~;Wmm?XcHAjEdd!SA*|))G@}%yslHT9NRFtFkoGy6N=IPD%Z&I@L3pA&C zl<4&WXF&~`2w%jMc`O}h9Ke(7dFuJj&@D+f8~gAN@bi+pDd{}T7<57-zXnAw6Ol{h z?m|Xd5K|p-U~v2Yx?wA^FYrR>iWQRauLg&NbX_A_(#lhf{1*k%Mj?1tr9&Bv|n>E=Y2E`8a_W@ObkJy+z}N0fin-JhP}x43rp zMgPfoSGE;R*OmvK_ov)G<^|yyNUMZp^tHh*p&nrx1KIVZ?#*Z43gdukSc?r){;TSH zb{Y07_j;B0qQ~mNb93**zFP;wR4#U-B(`?Tn|?aEH_sb~s!CJsT}~cWA5PoZsG%L7 zOLUw)3tVRHYokSm?WzuiCRhVg%4bL!n0S-#I?^P6%#uF6Oc((!4A;NrBdxvwg~qR@ zpyl*Lm7hl6t3%N-`;qRmVwu@h=Jj=}2HxvbJOyyvas=E$iC@!KGJS(_p9ho zui?(X6Mur=_6zl{RT#@3jjHwP70OSm>{g7}#tYU2o>e0{cC$*}lOo%=p~L$IVhb*j zP`n?qfUEZifE*U*700S}aM$Vb)`O%q;g^HwYTngcmBisqmHvI~qqaN`Kk*ZvjFvqi zycwQD93rbDds?R8%CYlU-u8Tq;?#8AZKCn;)!oN)3L+>nXBat)@)iNbPj47o&>Fd# z_-u1(c~zD=Yh&NEML^}^yzaxG974LYHweB_NwMI%tWF3JF%BSh!zVhx= znGr-yH9pCaQ{&Qm&vQ=Fm%~vVA(Xj5jLZN;TSJcxu1tpY0b}v7Xp}3huTY zAbSbGe~Xqeas3uZB54Kr3Fnx^pRGr9fT)o`FbP-4S>KB~If=Fx#Q1a6(x;w!RUE-} zGfH36j&Iu6FHZX|*w^EQEgiTD9ryZKX&c=&Q*z&^<~ogDLaFHs4*)4}e@1KE3x!aR zZ#j~B{*BskTLH}LKM4=mbau%|xw*U}6CzXCY8x^q4%eNpkvx%b zto1Y2Y}lVRk0`Yg;0?He#J^syauO%7`9N$e-E)W+y{?Twq5mxyRM82s05~qhlSK5x z*xc4)YVVTEEi)nil1B@BoHZidR!fd!Q{R}`#QvQDPc2G-IkllxHxodJ35XsN_sB7` zU-=B5+k`$1$BMLkDVSXz$p0lkFY|3>*?3h&XFc&u5s}$qa9w2Y^wvbpENk+%g86B_ zp;FPs#fe5=S-zdo=FJZxiN*si zAP+SL^hoP_N(ms^Tj+^cZwf;u%D)H;ICY!ASk#Jv8b#V>yReo zXmqH!-GnA8fZGgY{5PI;B~FA(f~5s3X0nh*;|AQ2VxS>yj7VDr$2ToMj;v4(^!?O# zdedvDJCE|Pq%tyGZlddXo}fHCS-%+D4rEAaLr!-kqzdK%BM%6VTu>^L8e$dCZgMf7 z33DPkyaSVxDw$Dq4{V3O)w*wPa4|QsV$Ndh8vRzuCh9J{SI08_T`>&>4QI1Nr#3x` z6Wga5xZO+1_7jD-1+(O@+5W63m)JuQf@DExnE}(Hb5oROr4rMMreosO7Wm=JZ!Q zSqdBuZ>w!mCF%Azl>Eg}88(zBh%1OT4G1gD6_BEcV3j#`9DTFgEl6r=->NqF|QDUjGE zsMPZZCv#`*R)i(qV8X<$-FFx~yKVcpMZAD2oWZVuj2~A$19=^ab*B1V98bks%*7rm zwdi6;|Lr`%)P0>s@r$E*2TrN=il39-GD^{tjQnBf-;a$Ld`XppvVP2EHMj#y#G4jJ zyD*fTwRV&Y*#_ZKW`O`f($KMr&BxsxH0>v(%1ef|FU!HZuM^^@NOn&UWfjJqs)M)# z#{sb7KDqRZbo$u0A%IaYx0wv*gUr2dB8N|C!7h3)P#vJBQ+o%4ATabyP4YOK<{dEi z?jG~TM3@9@0y?!FDlzVb&V_5EZ;>{KA>oW}kQ$j(?tSRcgs;``1pvARuC|LRU?q+L zr+(E_wqhT(V*#EiT+l&iPp8_!P$`Bn7(IA`Uc*#4D&pD6dqkjUZaUMKC_41rjP<(@YHJ~11!Xie+!YXQxEJE z9T0#hl_UMDS7stjBZ>KP4ruPwNvr(5Se|+*TJ$X-L|a3_TpNmbHvU}j?1<(VjzB#I z=hs=y=_g?!toZc9oXp-8(|5{GoYnCeS?>!UZO+&>e-yh(H?X^fX?CIK5cdUv$Ae#5 zYBj83HWZA1)4Vsf6(q>|Wj(%f1h1TZm#5zkRbtnIjKtGitJxPNv?R^0cMOV}A9g}t zK>OVO?bqzugIt}_Ulg@?jQFDJnPx&Vvu%)v#7g~VC|df!R!II*0P9Pan)W@K#WSWj z148+!w9@N8{xdU8fu&H!dMb)3IHhw}T-=x;KZ-Ifa7IqwYt44PBfK*gxF@~>ZQe{R z*EXe!OsU}>;FZhSdD~_O^EALK%d@GS*G9+&ssS)Ab>q@2`&xQCSM2m~b92ND>{bKu zw}<1OER+UH2|f;iP=+V&wEzt8SuVt}Tx~=D-#D+Pe>q<0%X4DC8u7t}d_K2@7)wJe zln!zZW66$K<>Vy( zvVv&!P#F7Fd{ca9fU4?kV(uy3YMO1_?x67{N1?X$Na;MHNxxP+gCa71YWO<$Zq9;J z@tS0bc&TES_vR*?k=zO_!Ku#dK4oEc^(>?PE3VYF@Q8Q<$QK^XalS3-DU2m?X z9H+tH$Ug#jw-t7LctZ+`4>;?chQGDgr~tdgdblK69s z-An6t6c?)NtY#aQL^E2(^3>>U!BdsrnBBf9x;CMATl$OGdL5B~)dxERBJdZ@=D_0J z3_S+-D^Fsli)k3BgODn8`K{u5<**#q5Agan=E(>+2grQv1UG>U_*T!&^lzl>lMftf z7HPwZ$L>q^pQ%- zgU#Ai?n^K+s|Iz`5Bd+?Hw+xBCw&foCrJ7E`iV=vhhEhIT3;X8+NUg3vE_f%o*blG z2sS83m=1o$6g`X$dg9Qv4OZo-(0P-N68BV=XOfK?FFf( z3@i<@42rNG7s;y++&3ntxaGR;VZ;aCf$(3GYE4wlhKIOz3BUzo(Cd(ql@HQ^1_rZD zWXLn%G-DOunjQ*YM7HryK}cms5Kkx!kj`UPX9q=CDgZ0TkzBLV?XQyiz}wFmA$|v| z=Ez0hZ#yB{7POtRAmj4lfVR1LclU7XaQkrQaM#6&JV!IQDL9OLh&^JlV&8){bg2UG z9p{tG$$U6DZt;%fTs*RV_u_>L!%L6pz$|c$eC?rEg4}lblGjJ97hL1F9(`l_THtXV znW|$1GTrA*n{}I#rLQbN93VYzRcuG(i)zhMfexbTNV4trh)}!$#;YQa6zGspk>I=b z`biskV;p4G$SMGHGkP+jYuofJg~=*4p4)I-g2f{#-=Hs%Z=Qz|hp2`_r`Ty%xRKp~ z@yyX_YZ;TesdOal5H{Z#eYm{Hg1R0~fGfoApnpar{ShwI2f4)U+GPs|z;AvT)9#{v zRd{NJI|c(|5h_sRYY%x>04y%wyMts)qS=)l`~KM30Wt#GkA#H^3t^;N-4CVXW7%1r zyL_w$(UJ-9a{26?t-x%hqD8_g1z(8Jn@(PJ=`_mTQfo0Ghsn`G-BO2l1}{QOnF!h_ zN7!1zNUofDBhQ>w1WcM;OKr8@bSkmNN+jYYi_r2s@0KbbPNPgDHl1S#DU$4Bo-oZV2|)%(%HhQ z|CWC1l$D9oM*NE4ioHng^y`~73>fE|`n$MhHNo5o;Oh?JY9uH%}ReWX}dCgyz@crT47-Yp&GrVY8v-{MV zuqg8FZ`X6(P@*H&3{?R~B;#+;=z@qJQ(1kT16`Fp4|_u>>iviw^IG=|bj=52g_yHZ z`7kpgILoBTm)WiEFMv|}k|R;uD@kexSoU!^Fq04X98tf3;&IN>%o%BQIBiq;3;&P+64Lh4+A>iFIIG)iUZyU?d2_6{?__;Z>RlJV6 zmCO)J)qEO_UGmBLt^Dl@rMkE#_Vh!g{C?;vMW9Ie4fJccfdQ*@KRNP!k1bptPRmNB z1aITS6|oVQ3JBBTAGsY6mTnG2WZzUxM2>7<6GvfPbCW<}9rz)o;LZK8BzT7B7`^Ku z`8cI(fEA?&piKtj(MvUMk|ZJMF_7`bWFnUH1CH#&u&S}o1CpyxL@KzfCY-g9gHE4k zL}Eo18NQfqgTk=2yXCUI`_O*;7svDZD3e0=Ew%492pa6B4DO9gHD=?0!}sjOQ(|rW zwr2Sc7!v{sn9YI+C~?A8lSuF?y5XuGg{l>QDY%-OKF?gcVh^6u4xVJ~^ghBhOCR~# zBzDoKQOWo;@-Of7fGaoR-rQAuHd2BBwL^>|YXw;=H_kOF{PeKnAOm0!Qt6=6w`MEGvBwJwtl#3&pRbFCBrUMvEyMJ%KWuR$g~AF9W&p6LfsoV zBs1-7lK!a0a-{&~lKSIg_K(-X3)754Xp|pFR$*}7Y_`sK&d06VlFy+)b?r`rK=j_7 z5+3A@f9Dh=|qnd&T-^u3Uq_p&pn(|F>Lw`xvMZ1yO zGZoU${c1Vws1)+rvvTy~+n1vZyTvsZnm{+gmL}PDeU2?(GC1wcn7wp`3la*+P?1Q|hoVLH!SG+vG;Qe*gvN$4fdu(zKuOBgs-G{75=0^w(9t1qh7 z61$BO;60~r>RccJqhQM&s#kC3L}g)&x$qs%qvwAD71_0gV~4FRV#R_JGsC`Bu$yUV z&@m2!%g}x6j1tByxx!D@&uXya_C@;pos39%N`kADZB%V!(ioh8u;d3qR&CE(3~e#v zmShoXr$K>X*&zzbQ%tw6#`zBH-!7zR`G3B)nTgH4r%o%!MLVTCipdRCwDFx>EB ze}VE;mu<&JbhO>_n4RR#gAZMz)UDxX&c) z`eY}0TRCO*38XF#!)Wn*hk61uaC$=NtSP$qH=-#wD^WMK%(SB*G~DCpdYJVm#-%Fi(W%up^tP;9LsY-&_V)s-tqo22co zOw+GTglh+kP@!11Hsf>z`0!J*7;dNHPXPEn=5Ap2P2P%t)C~;f69ey)-QIY+yW$!Z z*Yi=%E{q#jw!e#6?6T&khO

nFFr$L;|fwDRbo`KK|7 zqMloHPLkHk8h5i(4?$x8OZgow5wipWWq{ew&f=`12DpYfa^(a4da1vM?tZ9L5PR>} ziF%~|Bv*DC0Q&R;!kyniFb=R0PhN%KwsLp8h$yU(PgFO5M@24$&J2j>ZaI6d`ZRP# zL*HjIWz0Of4C`-lu|Ttve?sDG34m9qYA~~U{XI2;Z%Ul5gp8hN;#-| zygWh;uzL5?;QyTbH70_67omrndnfR%9jOHNK09lLHsAh8fmSC6Cv`7r`2q%E(;0Dl zK9eEhSUvRWvVRW+m$xJlsV*`nN$~#mJ%eiKiNE)$aMO-TC@(+(s%*_zDf>byW6by9 zM`gEy1u3<>gQ(0NQTc@=3kn^+GxF+gCtr?MJ4=xgx0uW}yAsVl zXD@Pc!aoOk z*r}^yEvuv)S_b?XRs81n`1)!`io7ZUsUhwG$y)HK&i~WiTYg3PeqrARN+?|_8NK1FeNJ|Sy*K^MIx9&gTd2z3G`)c0E6=z)M-upN{ zJO73?Pl%RAc%TrZAXI_tHDh&+-w?KyCr?gXO1QrP_3~w8$$QNv=P1Zqo}cENoMEQe zT&M`C>T2`*wJw4?5w4tx`#q5UCbH%NY3{qUzdC2`_y&*Vq0#T5*-rWile(_?UhrWo zJ-b^xn)~Tz2Ij`gduS@fFun@D&Dn(u>#0Kw274Yd1;DLe?Y3%ft~nIRK+l;sk)j|bJK;k;pN@S z+D(Xy+E3(LYZj|AQ6OkHGl4k6E6tF3OT=ehp%?4^ENOi$c|BDz2_`YdkSoh){0pZX zWrvT@0LWOWd1DWb_VlS~6S0jrH+Awd&9?wb<=tjXCjR5$mwC$y-#@rD#S#(R;fJ;$ zvs!gZI8Q#M-Ea%k52{x=yWwV~22!A~(CAmBD`Ln(_9&W1h+}iVzY@x*3 zZYN8(TO2|L4wOn!>m}L^Epf5$ zzZ82A+1LF!iJ$HsQy0v{7-JM`_PhxCSBDxv!Q7neYp*GatwP&{*NGFTRHk%eCbM6) z87NEu=>Xm}33lJVL%`&5P2}4AM&{2E{lD9D#JOJjg{<+S02|K3#2hx*+P*MQhg%$t zMQ3;CM-G3-zZhZ_ouJIs>@X(3BLlT9jUG2KQTmkKuHRQw#tA$4UK|VxFWHUlM7%-B zhh!ezHNC&%h*|`ne95P+Xf}e%QVgmkf7)1+CV}p|1}h!ERlJQyKJ=#_tcutD8SW$3 zT#1!3-Ihj@8#jvexGN|c{V*o}cgB%2#oXXt(#RqIh=>$ul1^|m5ASckY>ZowkJ12D z@Mg?3kW?d90SSd*?}})B!ag!EV#cFbS)~Ci!KXOJWKjd1|FANok|ObFu{$v5o`G+B zW&fq=Etcfs9yb-1@C_lHfJs0V52S`k#7qg^>f_w6Co7a9eKd-~JG^;!fMVbt65(eV zFBAjZ;V$5_k&Pc3jr6s5UI5G5HX+`dO!dc|kJN2GhKhy6pmxNP5bO)wD1dJ|zY#Ho z_1k8}W|rWDi^s!HxSd7cfJ*bimS!%@0TjqJhLYbc6_h)P(P9sC??dxkFK>T-M6p&L z=FO(~eh)6}loo6ZOw(?&cuFD2Onefc#dFOyKN>p||EDBCL-fQ^=n!>F z2P`HvjdKc3Kfrs>g9UFNb=RgMSai6+o*U(82#mtDWCjQZ^5q$ zI&ohK`v}1Z{b3dE$(Ma|S^#_E-jBAP(=9qBI#tm1@H_KXx+zb zHAzlTwexHC>zUb-z#-Q+kHOdM6G(SPVdK)Gmnn4-Jf>C76tl z@}oM|&e$J@R&5iBr31JNNXz`SLVw z?(k5w1rCk_{|K)Q^R9`7ltRl5_GatUDV#%LvA$pYkbZ5zC7RPNofjdK zW5C|)4hxBp6uhDJ5YSyHuGw?F1v1P^kn`L6^+&*9`Y z`(m<(0`B2HUtK*}cv$YK*^xFuW~nw`g|k0wD5 z^Y0iMLArQ?HE32^7brQ&$sEx?XWvGHDX`sAy>O){Xsr0Q24GJoBx=pfAy^%+7k*fp zfwZb4L~K|CC=rmwRt!HGnD0mBWo-6bq!*T0v6)upvg%kOtAMfroEq>j1^XtL0T(13 zb(by3&&Q<$j?P?B&%6CJjuU0;%g$us8IUqwRGBYhuhJz-iRgD1XMF*?h)w+#9}_@u zAaPmbBf{wfPd#<=CB-Y`f)QjgrvP)a5lckz^mFb&$p6M$g37cQioWU5kyEeFWQ`NjnV%*?sSqq2qGuEnC8ct`4pq77yV>jEN zT8-FhSzo-L#!FzGMyN%|?=!^rE-GH7daFk03)U z1XUC_Eln_3$W%+V@F-O!Dx73BVs%1rAI}3TbS)%@_Lr1lD(Tq!=Z|z>8{#DHg^1Ua z&xwO-?@QuqnO(NIUm5nT9jHw6*0uVRW@kG0*d^D02)Vute-Qa{!z)*DUlBNOl9Ism zLvO~m2J;$gF5Q^j@@r(r&4^RKoJ2e^RnD(7!z;D_a|b>nK1Wx`y1G9>1FNtU=YH(p z&JR}z>+@-++Wi@)T8E7f3(cMh`+WLm*UMzJ-ZMm6)egC(I(1{ywo69tXK7!AWu_BuN7)7){^3u znJY2N<(hPr;;D}0mW7#b%*z&SR<~5j|3=g2YE4dgcIRsfJx70(J0;bR#aqje#{~IA zQm&BtjAUZKg8LBp8D7X22}Qn0pU94xpD%b}!6eS0ChK7~2U65jZ36Hcn*gi>fty0F zWTp)0QcrlA^)s)#DWRpqrPgt0TWw>dp70-q6Q*;t}6&%M3r~rO4Cf}%&YT@%CMQ3G{STyp?;_;N6zfh z#(gq0VdY3B}4umfRJDmt7AyvRNahYblx#wp;l!MsvaZ1}06VHInyw#%Z5$6lD(V(*QVN3u!GU zRs^!Ke??hfd$mFCfREcnL*V25YiBq2uH+Z75x4@)75q+Vp4W0JyIQ-d1FFB}1{{@y z-|Dbd>$rQ8L(hZpR2f}t-o~AOc^iB3sA(tq88!lqS-_+Ejve`A3_i*zh0lTIX1RBu zK>v>`F*E;yI6d$sqWdrc7E;;ODW|#DZP@q!nZNvH&0*~m7-d4 zr~J(nX!KlEfA6Iwg4RLgA;I|@jF2VUHkP?+v-52_6jgp0HHt22-l<}agaIg-gFrjkZHiN-@A{gH`G+im zv(C%2wdz|2Q{|EEvO$dAp#6nFYllNzk%2ACoKbac^^R!@W_euiGKZ9>->E7CfC4 zN!Z|j-E#iq^196tIMt?YD*ocVV1jDh+bCIknu1!=)1tw2YH@GoulK@yV?MMlX@6iA zbtJAbyPNz+)OT7jL4)Yd2HMRBD_`XqvLY${1<7Z+L^0@X7WZbU_W5Oez3=0vYONKq zNrY@donBVYwz^eToR+6fDW)p?#E5t6`%<=K7UEpB;E;CuW$&qbi`^UPs7CnGi42sa zb~m$hA%p)I`w_(dH?3`We<2vAVdNb2(Ss+78A(g9In^yp%aR#7E_=2&Mx>jO^VYI&LoWv&i-ug^3craVSt%Fa ztBi%UyL=3Xrj*{>e+_dze$~X{t}1=h_GtQ32W+=( zWm0}yj1pX38KIuW0i?CFc#5c-fYS1~)HKSn#PypalVjz9>c$VYgS~|&mt?ZVsa^){ zj4>Ozj4@{V`JtUwj`p^AZEZk%p|nE23J8uOoCG)(x!{vynPgSN@v=J>X~B9{V&%Uq z=tr7Iz6@P=brUOKxN%JM+T0{w0g8H_@aHj2A8 zdoIsEDQCZOE%xV&r574_HRmU3NUR5ZJOkPQp<^h+l){PJQku9l#N%`Hi7kv_nP0J8 zw+<*Dx?&^w>OJdkTgw`j(x|F+;?Ev6ZR9qbt!|-&+}SLcH5-v!3F&K0WY`kxer^%C znq$~K);ZS11jH0%x$*(`MQ4VIO$OWt*c zPEu*M3EkwBDGS2N9qXX-aZ}G+XkTZ^+`wb2vq3$|47R?p$z=p@vN%JilKVr{TFkoJ zV!a>H4eSdWW_i_Gdg&%C9*whSd&{Q90o;{hWWlp0#5vU`92>RsR9p*YCou^L@=EUi z8xOzt0*cw?0G@GA^J{zzNW9#{;QbPRcxv`5!jttMBMMFBbv)yBzQ{T?vG!Wl8k%D6 z9@$7jg}N=A#iE|3&s2ZD#k#(&`%#;>a65fFr5Rrg^Csb3nJgzUc*%&qNu7jDbR&r4 ziK}ga=tHh$6iO(Q!ZkEsXKM>I0M(O)O4ng2ztx9(0T~N?I4>Rbr&e}9fBWI}x=z4z z|A})uX2e2RcmG_mleA`(c7o=&NqR<={$^+U>Ep9gTU2XRc1vMY7zSj(*U|a4Lt36> zOb>zzjQS!-G6y{+#EK5oXtGDDUEUTt?M{19A%O3Gy^hu!%cfG~`kPjUYJ^s|^!RMI zTKtb^WyRLFjqySrF^xmaQ0^Z(UP^>bwJ>=SeyhZx7eb?gw(32aTo%R=KT?fjHq);N z89n-GU8{kS@a%6Vc9Wfc9Vv8BB0ECdbTH8XNN4PYCC8z~juKyb*NDr0tD}qwFWfIE zE}ZQ0?EZt9SSn}bEZOTZxnES$mb$G4>>u2yX-(JAGo8-4%$YQNn$KB6weD+Or_I5U z(#}WPN(hzE3IRh;6L|4jk^vGb8XtK$*55!2nxi(0$qbu;Og5ciSylKoxogzrZR}0o zOPg#n_gS0iDRc#I4O)sh>YDjGm{tswx6BQXXpR-d-Fxq+p(e&owe;Z28p_Bpzu~^3 zW;=i7_a<1avd2MT#TQKh=VS0|Srp+wG&D2}Dg4McG|JoKA^)-&X(VFuevaqRjjO81 zo48*iZprwcs&O8(L7~;xcj;U1qf1G1a~x)EC+X*Uau$|%$}gK*jq!^SHZ`{#f=l7~ zgg05;Tj@V~=NwXfB-A6qPZTbb;9tq0&h=RHq;V$+Z8tQ*BlFi)MgOXYi0TZ2hqO+8 zWWzNqrN5olNQyLRoI}z|Y2u~>XjI^`3`L7`KU=o>0yS8G@;(J-!*?mSH0U^Q{Bzk%tWGvh@4>k zym`geNA<(g+kI84g{Ef3llqy~^Q~jkWr_L&AEvA3vEwT4ull^z=cXr0S28ibmE&7D zCF@Fb_w>vIWMB!<4aiA=*mM|O8&u&ez-H<3xDr0$v*vRG5^G_9h!pjr-8R4Gl4)5? ze0W|hymUR68nJ4<)VdZQWBmzOXh8=ksav(jG>-!B(xnk=iY3cfWv zimw%-?Zn#CvJ&qRG;|_wZuk1{fk>Y5lL(*>ilCP zSN6E6A1RSw`nbs2AbPjf`wF+oVMFSzR&9x+E}&k&sT;`PvCp;)`AER)`ZtpVzr$_) zB#QR*&-aLV09Q(NCw_OrxeMROVgAwAlB{{F6|9xS$Fgw$Zr55wuh^zE!)?-r-abRc zR$x#we8Hv{e}XbE#!dInzL*NjZ3~cw@p6mfD&DvN>kSA#y8fHpHs9FT*pSg-9nV-| z%*SzXjMwSoFOE(rkAtmAV$Y4pHhMT85ao#I8dG+^ zy&!*K){~;0iA(iimc{#0n%E{6W)0+{!*2APOPee>U&O|)eBR`OT!YJPFtBmj0BNgJ zb_S*}C%V^FLauk^RQyo7IY)0;)z~3pr5jtyevFcZGoL_}H@3GGHZx?%9SN+wo*VHp zMG8R{9)S42-IHkSpIcxDfDZoRuAduw5KTg9KrKoDl|Hs7VpRWq{5p7i&gKa^cwh3# zoPFq8f&oYt1x>1a5_`9&oS2jSu@ML04Q&lyri&?auyfT@;oYpXKl9?!2Prtmj3<;k zy+U`Dj+G&wE;K~Bz*0c`B@3?n)5mMz;)VpAO9^&Oj(r#ZUft+AfD|wJ+WY2nZZ1!RhS4WN7V_jam{2!G{h2lXIgTUhV}Cgz|nxoqSku53uu# z$3++~{!W2_JK=mq3~!U(Gvrs)RVT&W?~e3($`mf^;Nfb>cN~D2_Pd^mv| zS3}&G_V!y0ZsY^C=jDsA5h_sgAVAo02Jr5)#~TAW^7K*nd&?~gpcJ%L^hP?CZiV?S zvfv4iqW1|VM2e1;DC?)a& zD+Q?Gzh5`^W__R0aY=)Q8zK4lE{D3rn)F2mxi{bJO+4ys5eLVf>X~TUW3G5nU!VCY z(F_O6YC4XJ>;S)OmEQf;%t0!5(fOb85uV^1e^NRwHLKd$&nx;I^5~LlMHsk2$Za*C zC$(Pj-*(m_dD9(IlXNEF8TM;rj51E>f`KmCF8W@C=P*DB(A94KcAg>Yi_Q2nf7AV2 z&6E$vJllY6%O<@l6F+{r&Nc&ss^fuMjB+IV*`N|!8N5ercPxK3u^do!J0^=3=lz-3 z6Y)8G^G(p(sU{cO8HHB^x*^JU?;V(Xj|S7AGrVutlPlgMDQu_btL|byyp?t7Jl-oS zHi!$+Z_=kBaMOy_%AI*VDZT%C4i^j>m3@IY&6y-<%#`V*IhKk#;sW}7kVh|qy0F7Q zVfCrxT^~S7xy+D!8;vf?K1Ovus+UuioCf<@+$qOhYI(14_ov`1mV(9apF*Wtz^>fT zO}Y{kB4z#sYz%H0lrR*^AR>a$YFqdB{w!j+plO3MBF2)55}7(ayx=Dj#SdxfLGz^@-lNSlF6%dZ*EqiB3u@W3;ST>⁣;1Qxf=uWVz^uq%)KuS>?v20 zjCE~dsfA-mAiIJ!zsp9UpDzBcFdHP;Y$AyU1J_wi{BE?XyA>7qx~qG_!BlAbI!Nal zpbsS|^(1~u|AvKvwdFxB=M}$pf1i!gZ_cWVAagNCJrEHQ{G#0+e#a5 z-9i8&Tm#4Nf@HibBjC}~JN;{@VAyOStN;x0=e%^>J1M`v;C#p)k6rgDI`^P?*#q{a>v_<*A#I~{%p+0H`$3)daMVjkw7Oxee z3&4Ewi1YKM%G3@F>5PCt7)XC2A;X<|7FL0QpqD|~HEGRzPZ0jB`9gi)BAK5qZy*K= zsr*-20N_8!OP`!ijJhUnLv6#^d%|foJDTS8mBa}wv?i<@_)V-3JtZi};Bw7+Mlap~ zvSuS(@zufYflHesdM@W83Dmhn&KdimS1Pc`S%NHDvsNNnuER3KKv0`7?Bw#O_tjW6 z+Ex(O&H8>MZMOnVJ@ufD_S-N-cRSD82e#osrGv-XXkHE%JP}IWIFzrzr_#WWjE-9@ zS!5U_LN-JqS;T@I!IVNec*`Op_+IFd)^KJX;)dV{;9Yy*FA|DQ0^iblfyXm6`ktp* zqQ@sdL#}-(0HK9Xf~we|LoN^3w@2u>q{H4hh6y4=7QMp6j>D7RUp#V2{#&$J82A)9 zb_T>!kLYPG>J=IPt;5 zQyThQ@xE*xW<1VDmwbCJBMQWkSit+<@CJUNeskDzuDg{v`ZFelIV7n= zVJpBM@*FnK3>ib(ago~RR65uf&bU-gi)ib``Z4rJD(`h(3@Y^`Xp?u*t+|2E1}~($ z?dBeM&Qn=G?uMVNb>_fN>_l;M2PI)&0u{MFu1l<5v4z*Gz(hgtl>Pp9PyazIIr}0% z?gQn-C_k7kw&XZzU;AozmXcqh9bxB$j^&Gq0f=I(C%C24w4`n$845n6e71wpasOPfFSWWJYc56~mmLJR< z;ql3~x}^1Pp= z(w_AtgocDT9%N_>;97(4#G~#l*-z+lw{QGJ7 z?7$v2zmddhL)z3?U#ec&fzHHBLyT=Cp+X(r4=D3G>_Tg@eS&U+v4>O@pS@PNkFw07 z2Y9hCaQz>cRCghuDFvVpPy<_hAl?`#aM6AApuxUWq}~TWh)J4^wIZ-Fp!s_#ftDTi zeUtwZ8Tz@kDez1TDB0>jo0x-oG8naT4DP&9fV?5+T3>z9@zLrL7K33h4h3Z#ucS0e z(QsmZAbqmaYS}1}&q!fGc++9o2+!+Hi(Z9N&36e+y#^cF$Ua1XHwvYdI#Cwo0NXF6 z{*2V+9l#vvP$U6E>!`n+hZBKWSLOQ^ZCs4?BeCtCU2ioa2Nb-OVF2yaiepe&EK6ra z$C3mqY^m~1y?ZF_8a2#s~UgLaK0$P}olGz>ks6bwX5{h7@!-aew_ zUG_!{3s?@Sz_(Qc8Mh!f+)Vys?`&`cyptEJ0rnosr_MKlC`1sOs?VDrNN;x3+8m%7 z9V+z?;Hq{{g2%%VEUN(LV3&mhlP0ZPbGhJApiY?ZcLy|HW{bLlP}%x+YlvC@Rz90N zFW@QHIJXuRUUt&>`4S=P{_avr%Z{RgAhzZruxxuf;t(e;tLqOW|2A5t0A@HtQ> zd@^%_E6!NJ{p@q@#pf`yhY6AW6n07GeLgsWVPQ^YNA_iq6M{PbXfHtu*!X(IQNv0- z!KPckwvo17g1H7eXy1^)@xidcwzr+X+v3e2H#kZsG?yFciy8pKjwos1gCdUtp3bC3 z;4E*G`bdKv_)!(iMC}PVn;!pLx|d!;UL#Rd;*CV8sVTi!2#OCkpo-dXIF^AbE^iS1 z_b1Pqj6ztCb{9mHG4iWif{myO2-&nVI7vG?_~lzb_mdd7`zHOq%nQ^GHy#C~i(igl zHJVfi-BSUn%*928Y7x<=(mS8rUZ1->8%L5|9NyN`JC;1Ze2<2v{VM$UwZRWo;d@4+ zJWwNS%+c=On_I%)=?a5=9P12B6B3X69a-RJaiO1@R2=9v?}(#nlD?X}n(2@_2(0Wo zg526~vG_LYlRBn@9}x#Pv%UJmPHgc+c!Ms5yvK(KwT5lw52Yw=IQ=7Km=pmKGPLVp zP8SRAG(gfSrEVnubO)_G4NJ^yK!BpEHqXPEJk}MG7txlTFwvwImuwG9w4xIS9Eu`WveZ*|2 zPBd2d6=spIA*B34M^e^fKB(@iE999JogU?hPP2XCpI8~*ch6o*mZ4RcqXc?F@M` zksiC>9YuJ5Zlkp!?Q7b*vcMPCV~PtGe;W9B^X0*FLQ7@U)mX|mzfxdmC%kw&YmT|r zkL2EFF~eqCb5Tzqcv}pIA&7|cpo{OAeX;aPi#^}&KEhcpp`!0A`yjRvh?q99RKCTH z4x({_4zQ@R9}xk_o1x!dG=pf6$!yZ8aLHFM!BOK;MHZq%UUtA!Tyugky=X-V46tET z@UOpaDHR$(UH9Oel2cSt0R0TkW4~x87_4|5VBYSsv&GR(*c}Clz!a7l##s@JF873D z9|s}%)H5G^dSfDoPpB}Li6N59w5V8X@zKwEnvK&}SOrjc;Fy;QEKq}tR=ebDGdRm+p?8_WeW_N&X>D8|opM{jWpeCe%g~$8d53Y4QODUO2DDWUyv+ulY~fOsdGs(2G!u zDPg*DK#eP-rxAY*xv;mxAxFd8EF`Aj;h{E;aq0FjScacepE1)Pyvj1gc!^fL=6fuN z#3s&ghcpBa^t2HOk8&myH3QN_{Wo9%JP1=apcAD7Iwl2w+lv6IPTBMnsG9;@3Vbsi zW{;{xJ_LSrkUeH;@*<_+Q#?epAzD7vUte7~TmqOcRH65}$7Hod$^XwtKN1SP2$N=^ zT!tQ|u;PS233+=cut@n4#czaUB_zf#MFN|1ll~Eg%TCn*v*QyKG<=wW@_Ce=|E`oF z*NE9Ne8O|xEDC2BLA>(0!x2rmOK|v=Kh{mym@EY`k}^<+NGZ>xIS(0<;z3m7JVCmH z7(A*NxGkKA0z-fztD$;?r@0Pbc$kmkI7aLHCVIcwOaw9<7XYl0Ig&};J6fTSoA+~x zQu_nXH`d?XXoM8=D^7Ol=7{&MxHCZ>(LfW>u&({ee3ffOE|T?-C3?U#giCG4WeVS^tzCob3x9^3!Z(On}dTUn1;dIYiWXK&m$HfBomLY+@)5MN50Zfsx=SpU;k< zLjhU>ZfoAp2kkJ|gvgg!LYql#iBvaCtP~^MYL~AN=O?jSzhB|*Y}%r31+|f(ke?7l ziI+iAB7mE?(fR3Jg~6g?7okc}y%msbIpsfhJ(mg02H8J4PFHTUhazGmOei?4Jp5?H z;HBRJQ(wcgEQMq_x=2Cp&$&HM^bL$wj)_;E9}6On+b}CPmR7HWwcGC9zh4*w*DY80 zDn8;3uWaEHK@`rT#L0mt(KlQd8au4i?Zj4=7-;2DM#@a}2<`?iFicmt?eU=g+{1r# z;(t$jGvJZ^+NVKmmUyOrg!echQ3O7FJgmfkhEC=S4fp#Dnfq}U=s*!`H^3it~mjkYVL(fPGzxj2f-=0aG z7V;4TYSY?tka9yvgy~&*auz@n^Ed5$b$3f}S+5zZJRlID72l5?28U#b9T3?(E>j1z zM(du?mwL}4L>&g6+S^~KKB`ZETiYwzi;0K6SlOg6qv2H=#MjilaO>n3^EeUMB+3rK z{F>ni0MGWTXh#_?>8qas_HY6O8I|SaBaXPHgJLkJbwS(?qq1YizO^r)#Y-wjH~+?- zbgJSaC*J5Kw9{vtpHK`$wF-0)Zoeh=EP`mgK0ite(EiegStxjkAL>qo=e;+GBqWl4 zJ<;J_wDZ%_Ki2gSI5Yx7an#8FP3)OjsDE_E$n{?V=h?aHf~S2>xc-k3-5ITlH61@? zyKmb~D2dT~14(IM-fmk-ZFW(1>h_V~g?IE2=xcx3&PBK`M|l0CIv2RZ!uf>O01kP) zs3H8YFLiA4eIu)$MH~${dcYJ(Am!mF>iwG~gn*e4B18lS44L{=oWijXMv)0fw%lmz4KsJS_?WjpG+ch&+Sh9&xI&;1 z`Cg8&cZd~{=r^dONvB+@c}n6KTYlmRe__Lz9lEvZs_^ocTlk>TA}hlMjw2d$8MJw8 z0vo&-hFH*84KBT^N{}^kcVPwt8RX}$}?e-GIA9P8>=llJH@AYQ48XnPcWJdPO-3<Wq*m2mC=aY#x=~QWaSVVh94!jlSW`~5DDA==UYah{pKf~^KICc}us=Ykxb5RV zNq+5d?Le|n8yA+$Ii7l2F$ zpI6+kJ~guH^(8<1{M}$W_Z}7^J6IF5P0W(R__nV~$$9Q7iOTxlvk}dapyK@@f7Ivn zV#8pCz>&RoMA0ONTjXZ@>s~E{s$wrpbUH7jga~3A81j@y8_?Fu)!r(Rs1Mby?XPf2 z?}tJVpxgNkask3Vn2wW63suf87tH-iZ>mnlyysaaBIj9Kmp5lma*eZ+zBLS^IUqM3 z$}qQ#5E@;Mrx>W%M?VI$a0TquXt?a`cwAaLS(Vex%0}+%JNef39*x%g_+NZtVqRI! z#yNT z_=(E)hOw4hj|J4pHDO-^y09tx(&#(rR(MoBGq-0_NtBRk6NHFUio?%vC#$h2eY^0J z_OSMf>U{R+dBnOWcqQx5|1R+Lk7RXy*qZfB!6((cF)e4MwC(uF;;iUn!AW)|DAgRE z;F6VlZAr!Xrfl`gn4X2MO525c^y>YzZE42MPEVriZFFc%)Cm(Nzxh5h^aw za#khkM$G<*>s3$FWpkz!pK^x<{zWcAw1+5vooD;USTsh)?@gwlzzq9XsQ>!4xZDmy z$O3fv)y>w1>Q3YJezN$txj%_ZLAs?rjV4^YQ~XhbeEA>EzzP6&+%nM z&O=LqH1uBKdPv}QuMf*{zdDt4216LEKx98ZN)%Vf6huveu%)wFwXTdZA_B8dw(QO;dF>SO>J#g^{($0M4U%it8+{d$sCx3No!k}ke3*nYV?pB<&*UKR9Xn{V(-Xu8?J%wSCe$Cc*ELRroDMlmnsop&c~yer=_ z9!C`u@5Df1bymyTVVh~?A-|>C(gMQ&3=80_ARVFqT`y-{x5@gn-g)|b<-Rt#T#iS+ ziZ=|MeeYkdpY=?$v-4W+%p9Ow{n(;Ctb2z#;*s>Rbt3#A%|t+24;TV62o*WD%4>`k zrWb}UK3{W=xv<=Ssk)rXeX1FzMm!sJa#jj@qu=~*rHnh4Q`MT7s-=9agR6MDg{u6C zs=IC1fH=8YplS)NS~ML8)M*t4@0V@~t?jL?nQ4(4=PUi-UY=i69qf;H=SK<@*g6rl zvvZx+WBGlxg5foezY(I>=2nlj=37b#%LV&(&UH-k9fw=-!6Ab9zn$Q(=PQSFLdEMm zuFAx8!hKXB*HI~z-A_j6)N{1?BYd)R^l$fnZsJbYUl6M9wD&i_K7GU79_yaQ_Uh}| z6Eq6Jx@1vp(-cO0I^wOBLsO_ib6iYwCLBDrpW!YP~-n8s6%< zjq!1M{Uk5dZufFAV*)t&Tmij96Zb>)B5S^l+q{d#-OM5L`+7{TopM#%NnQ8d+wXdJ zDBu~9ZD7E5dN#Vs(PX-C^kk?+d*4R2xkLY&pJPGJ*i&adCr#V3*Lk?6mhhnF+f;fD zhU}Hr2!zVIrwM;=C|sv>Yhb@*WLI9}_S?IC@4LEhQw3A)$(&z0r+f0}!a4LdB}^zr zTAb$+i6RzA;ts2)tFaamooL(69Km;l^nc$GxgF*KL4gt7!i!V^Cmp5~7yhMb+Q=8_ z0eg%Ca(5lA7oYR3HCnn1m-TF(#g|mI*!P`XZYvM;@mS}FlULm?2-c2-_bhBRlt;y! zun5Pe{fjb5OEqQczga5MzHe?J&ak1NxfK2Ue;zk7D#q(qmD2P=ps8%>N#|4vV%~_$ z((HzNWN~1)Hnm1nn%}+)>tNZm5^A{r-MVSd+%UO2KSxW>G?*rX&%i#z)09@8M#IBl zGXWN?)>p$35U;TqAh18DxwKy?eljqCJUrI+v-1SxyKp@gT(K0@%FJqg*qt5?~r z>uzjqVu4Hk4$LLE_86jDV9zO<+Gk%XI#b{BDk?gK<2*rj-K2!SPGa9%FLT)+ZN!}} z(Z$|fwhS&6O#T!gg7alSO2}Z_o~_ny4`+8WGY&JQZSIXV-_E*AL!RG85jHV0TS6L`UC_ zFTxOjh{BQW7JjK!RaY@`&~%TFb70=akbRWUP;E5X2(F5al^OW;@lyj1B|CDr{{K59 bh`dKtE=}EUeCt#80Q{2_lM^k2>HGeF4`WDg*+-P>`2?4}rk>Kp-$J$cW%K{LZXy;2Vmgyp9V5f`$kE z3j_I_O$>ohLlmSX)IGBgay)$1HSV6pJdSccD=5nc@!(1#qr5MIuMMOo{BYQuPSA$j zhFl?rK#+-N!Ud+8T_3T(mAmd{f4 zY&){rn8@{lTJE;+CP%Oyb-oHCK5Cfc7x3*(mgKpdR^Wer{_n45Da#I2|L^~kg);g4 z@4={HzY)Lur2OwQp^yG9QBo%deVPA0)eigrKlMM0@&EcV-4`;8f4s+*?B@7I5FmTI z+`WEUGMDo%bzuHyw zxncb|A^V;&wGmb0im?mrC%n?jv&IP7Le;$eXEzz;)qs|+`#&No9exBpKWEnG8|_+H z$(H7ye{1JIJn>{3Sp|Gi@fT5_D}1o$`3`%73WMJF{bNuu>v}o9%3C{;(aYH?rB2`l6y6 z&g7H2j1X?p?L?mcYZV?_<`k?&**MXCqp-ETU%Txi)pGvTCVEW-#Nn#<+=uMXSv4%>SIXgb903Mq zk2xm9UWz6<;&71<4svH#%Y;mBCjre3?iA_X|dARzeQ+kd2KG zuJQ+vUVe6c@UP@#Ty+hN9$P-T7cUTWE3Z7tT73Q57URu^;wA89K3G`LatYzu7n6Ax zQu5C5W{L%fuC1*(&Kh{o2=EXmPw=IsF5@aBCuU{EG}z3p{n2xfS?=^(J@w3=+NX}P zD;KfjBRe}khk+Qhx=N_1Oi=05y|Gypz>E!tAr73WDPgJZkEeL;c(fomJu^e|`ZZb8 zf-_ohNJx(}i@5(IVwz@k*rN4N6yD2~Q%`I%vY4V&dih}Zbn_YB=h^&WB>N^Z2?>dS z;cuj2XFif?B0n9%>$1whw$z}1Q_6{U+yB$Ep2Xp}Lj5a;pQyXK{^ZpxBwv?&<%?$D zataBzh4Iqtc{{3~AcIo&jhxmlR@l10^q)Vbf41vu{TGGgDY1^4*D-%R`8KwmPA*vV zSLoE&xX!&_5Ll7@mBoMTQPJ{FRTXD2mISVRCVF?O*z886?+gFl>i5^2?g!cx25p48 zIfQ&!1mnWz2gin8&u-tnS;q+T25wH*o$6NTJMWKP`#xUI?ku+PM;0HVWWQ9-?N3n_ zMS(!q+52+u5@|N_x309Rx(L7Kqa}gpy)6SODk^m(?E)K7K|w*S)4HrxgKn2%snNPO z+ZpXxu1HMR(_wY9z8{#|FY)(pl6CWJ(%|0}PEs=pAU6t;xe8Il@fUypXm$MU!2(-S z{krJ<>$cOuU(Om9w7BqrFH#YXtt%J2sz0Sw`_DEESFlhn z%*Q*i>xwOd!`C)?d`Fpp6ebg9xStYPP_QP;EbqdZ)z#F{@zv7(>-zbdfTs%i-JpcDZW#3YTWlHRn80~JXqs4uT6s@L2YfS5T68wNy(UUB>JzC{tTRE7xlPH|kLw@Xci)eiL8hO9w*B#aHM)#69)ZC@s> zj4`=pnNe|rdW^eW#ziCCCl4ozm5qciCZrty<@i5JM>T!43vehMtK3D>x_==1)7G%u z@t)Lv0XaBuEa&xl4-%EMr?$XR^CRZP;XDy7yI~rdWnXKl-`jO|V*VxtXV$bf@i*(o zudwf#nlQF+nJcBF24o@OQ-$qf`9v)=3Lp1)r9*)kB`8qZ7E8<_#K z?1?4eK3Pu?TPuEk2xaejz(ov!0bkx7ltMX2WqAz_darYoGDAiofy2Y{EtJC5Rg0`% zayL~UJFg2KG;!&F<-8_R7QLk>WYG^i-yXkjaprp`^zDJn>d4lf^_$9*Uj9$YP5y|P zy1(B);?b-2=Szf~^y4Xur_w)MOeuF0R7m}l9{DI(!@{gQ%mNqEL;W&-ML=*c(2m85 zuVDw_{h%+=k8l1%y8_LN_6_I*BICQRWoW8>6mFZF%-zgwy>H&fhsWqp!^CZSaoGBa z_@!yT_$E<{ji4JSK=JZfwl8qo3sUEnx-QUQ4^qwK@TnfF)9j*$$y}YecB3uHQR{t` zY#wvErb2S}d3zZ;Z$+2nTUb;YX+?V+`5yZ?=O4QQ^2yWkb7yt|%Z*ky2cKKR9YlCy z5eeV2wz}o3O0STf1h!5D2U zNJy7Q8ihf|XFORTGyL;saF6jxy}JN;ysVqC4k7u_Es8<5(veMeI$ZJQw9PBCXmg!r z{VL6dw=Q<23=$s$xe!0sA24ci477qViOFd89J=3Mnk|dIQb^7_t8O3XKxa_SjRjGB z6>O)hv^qR8jiXQw%yv$=OYqb#{P;Nxt6R|BN2!@a zKRq=A%1iUiKF{Ny`E|06zU&;YZupscT)*K(36f1PHP2px;^Ww=qj>pj!I){!(PEo~ zFeN%E53&bl8(uQK#0o;u*RM2;jL0Bn6)={$SgWnSk-&!Ux>V9BPi+93_Sf#oCgFEy z=iyI|SfN4Kf#dl@cMeinzdw+CkU2n>1A)?|hNOVHz{Xu`2jh$(YtjC?4Qbit)}Frn z*d|jJb$N#AAO24VggAh7whrd&u_CajWK~qK5i!XlF)4&r<{i3ZoSbkP9hT#t9`Bo4 zStOA?`LS{gi~iq$lo=}|C?S!|hLgXk6f+X6fBct>h4=Wt;J%?ct)df(ZNkdHqj(Cy~MTktM_$X<4>I`){xzxdP1sgA`;$hItdo3({Vg>1T8k|5hlO{+ktpkTxFP{lE*TE+u zEHu-Vv%1s`-R)Hu6kohB+i?DJ8gIk?bD^gOT{}E%YS)%WPKEIqW5Q6OW$T75GZY(& zv6AU&n3zcY?k*%eKM2&-*VlSqT32Y-N@`q@wkq<3)p?=?L$XE@Bl=bi)I~T*%r|9D z&-t@P_cr#mst`*`t9ruFh-y5J4YeyZ>0i8nf>0C`6i7a5Z8YTTLc)x-M!n}NSg9X# zxa4B~2o%L34Cy)T0K?=H})CZhQD|-n_}5+D%MP4{vT3 z03*c7jz7nXOcUL5Zc)+Gi>ayM(CPjInaL`d=4ROovD^sp^v zHV{qlx7G}4$^YICL_Yb)gC^%^)qkqRDFh2rVe~t!I z6zhq?R_>TlN+zJl-BwjIfK$N7Atl_h)Nm8Z!A5}Tp`j_`Fm$SgML<9h@P81Hl$1 zFafVKdPym%q}0?~8Z^*VLrt=O-d!)3=)LN)r^XHO7`yQB$LHpxtrhK+5W= z%`PjtTZMsw?m9FgM`3m>mJwPS8W^YA+MVQ2yR+2)Av{WJjlOho;-p!3? zs9bnC30_t`xxp)GS^<|W^q)U}CT3(HNh0Io;o*R~sLG&?ua1=2g?0R#k)1u_hn2MY zJ06HmkC68YBNH>T2@mOHxfVMyI--M>1gmGR+=mbJAP&+?yuVMKv6hjQMfASd3*s=y z6wYl&jhWi*FI6j6QCAN=J9F+I8Y+~Ez})#;OGjHsr?Dxcdcn#|_}9*zI2MsuQ8rR? z#A%-|3j?G(*}~K4Itn~m=K8OeZ=JS&k!);jfqU(`{|BR#LJcPm8;c__-0A^eR3RCx zBV30eR{0E5f|ElVsyz%ZFYjLXKpF%8B#CQ{-XS9!Q}0|dEBY%z)BWM?$wlIa(mwkeE>5$i>>!oGj! zX@7pYzq#&u#(0~9AUs~a34#d{-b*Fvp?Hd!yOE=Bb>bbMY}0P{MZnSX4ja97>9@`0*TTSf+cd1 zNP3>X2J`0aTh6l|N!SRgieI0Xe2z^H|1~%Y95}Y>6Q52XzsnQt7}z#&JfqL0x`c(S z#LVT<)!S%wjVEkU4p`P(m|Pz%u6R*m7E2M9(b7U+T#m{QKj3aE51Wz%v*!g5vjF2` z@RZZvCQGWV(X0p&pR1e2E(!Lb#08(@4$rMmme^EO$6Dc<)dL4u7SQ74<~;8?HsEPL zPrJ)qMD*##EL->!*5tl@*|bA{WW>qE)Hu(-Fb4$`%}y;nJv7Y`(}$Z=Miv(7o>Exf zvmY;sU%MX=9R4<9jc%hNHXgd7$Q1G6J3Bj5(J(+z#m2_Pjh>iLS@1m{^*kLQdfYo| z>j6AZjY%KkyIf%s6BCmgrvcPZIqATexCNqSp40B{7bopxl#=;Vd-*fgoW|XdSc=!0kE|+rj{*V$#L#0-$ZiW+HZs1iGy!-EdeXvolsabuXm)daY|sS9`;>+uT{BV`5fa zg;rc!x;fAbXbUI4z9ot!7sQovph8VFD&FArcvi|#)A;PBOM!?^9C*4uNCS?W*noQZ ztm)`y-b|j4xS$BaPleNyw9oC66# zqIimq^A!#qC27zQRY3dG5)V3b)>xu95D;r_Zq^efEA^Tz=Al)Y=lLi<4lZdh3Du~0 zo~YkveL6VK-5CY#dJ7C|Yn$noz;yxXP`Q=(*GJtulZ8o%sSrZv`;V8p>KgjJW(gwU zt*t`H!4L^-Y;yA0uWvKfmYJV1f{0ce1x^-R#!W7ECbcenuaA~ubm}cvZ)(>|%FCtf z>{u@lMNIz<7_W2(WeT}j$C#_@S9EXZ1;odZ@vkl1x|EeSXqL;=2oj6YL)u#1}Zk1rA7X_j-Snul63x z-$py;u{^PaPu!NYuU?fJMwtZaa9f%H!mp(M4-F4JK3F@+cWuO}>FH}AylQt?@LnCx zM?hDuFB}UlDblC{df$!rra#tbvv44M0--EHDMzSB*HQ)$hi~4zJz1K*uJJmz;B{+Q z`;8d)5DVBNeKsPHvQMjLpTkD8_~^K}a4lRov^AFy>@buv&@#%|ea1ihzG>;*+}>Ot z0OtjyLO^5yp@$k!xV#O#fa)a%q-r$(GK(a57Y4=8@PKI3h}e0$=}Yhge4{FebcH>U z-=}#%t=eN+n_p9dJu{lH&uGRI2w5u!6netX)T>-4u%d?Y7IqvQ{@?76L1%bn}wSX6BEx1I2E8k!|SAvi@ zERVJgA8bOL8T?MW@x!MD3yln8u1^!3U+I4T{td7}2p}KK zRtv%eye}9H{ckrdMqJt!Yji6e1#%*%tv+PDt0c@|H%ey>;N{i+ZiU0}Y{rG|4Ods+ zCQ&_0CniP(DA#57CrZG<-2g_0)91?8_i12z!ddr&r8%9DkZ!{oxsWe1dz=r7OJ{aS zY%F@0kfRXhP(1m-4GNy4@UcMfXpj7xHN)cTp4vajg8ichXAtp31vK>$<5 zuy-O$lz4Qpva=7JkBcRQAYpae@)erBkt_Q4t#{XT4bZ~FVxJP4mc7V)t`2$t$rlaK z_~rE*a*dn@%+k`*vd!s3cRE14LxrGtlZZ61q6?;EGP%rA{pU{6q*bNccHDPn*^vNrvK-6qnX9*~b(vKn7xg7x zA4(W<85i@Asr`$gs;}jOwQYK~@k79Og9D6BU!NTD115x~?rU|?1`D6UEZXyDDm!0jw8QxDIZ8+=Kj522Ds>z8_V!K^UOji0X_T+H3gF=3g#e0?j)@5s>^~zL+u)aw zh(vHlqN7pYWpYSVYPo9?0)@e8>wu&}w*gh8`nXir)o-Y)fW!Cs@w%j}H8fYmhxYaB zwgHzd{ES3F+pbwg{qV^JoG4Lbn-4&QwJZ%eGUcH<(k^}p?Ma>Nm zECSJljcd~f6LiG7A+hdTx;9yvw z+nDr)V+@NWB>;hB4Is0$fRd!~(1tlX>xa5Ud$_*gs?WNWgdilGg$K#n2LMVLY@6A? z`>7S@R%}Yj_=5^tcqF7gi~27W73hFr_dKuUueCg;eCOxDWNKzMa{^PTG|x1`8g`~9 z1sH1HAt36pg#$n<2MG4Ni=&WVzuw6Qhl2d{;e&;7cOcBocz}N|AQ4Z7=@P^(6V*ET zJtDuinhipfMK1B}yB^&^jotouHQ%i6hSGOT86P1R3t>kFG(Qjy%2t4M!g)O+c(YgD z;@Gm-{M>ne`ipyXp$dcujY7K7=CEw);VPF=+fmG&TCx`Uw2xPf;_`B;WNPXy{X8 zhzXpJA-mRXE<%`=G2^BOZ`oJ-ql8`eQscWG4;_d-M_41Hq_gM*1C(Sk zo+|>4Y!;)RHI|?KVuyULj(Px4o&5Ti;oG}U!C;eiT%iZ+aK33BN_zK9Ctm70)fwQ zyXN)gnTWWxH4~(p92YtqGk_;|ri!r)N$+vJuMSY+*NyAvYTOTXeNTFjN-8UREc5)r z{U7gWSb!E40;8)LbGbij2Es&SR8$XGGgeC+46F>Z)&={6xq77I&TF=Ka=~E!Wp6wk zgSLWa41ON#?evqyUn%e?dX2U{t8@`K1SDY~6_GYAx|4zQ-2G8-TE7aqTBmcCd8>bG z3q1qw?xD1kh2}RSaw_UJ+W-I{YUU3;uRd8i-5)mf8FS>9m7&TW|Gl8_0e84>b0lN( zMIoAO0%hF6fh`cBK_x#^4<%d5H7kRa$_Gwx`c4n($7wk^aaLDX0}=mrd5DQVoJf%i zI6^@5lkoKwemW!4Avq4{!yxARfDBb??B*Lyf)zF4)97_TReZEI-{c@CAprw;q|*s0 ztey{iyscf~J8|9{+nZ(!H>UyttHSTq)J|u+o(SL&(Z%1t$Ay=0?|T3hPA#~|_yk1I zGg+7A_Dqf)MGY1cTFbm+zyrnbW=ml!>h-dR7^Uxt1n;D#uC8tjQVw(UQY-KN$sn0S z_fZX0Oiy7@l6-!8oI%fh@!|!v#)MWvO-tTrkZ#12GiasgbaP_TKR9UG8;U|K>MQWe zLNI*Bq8^GwGw)~g$pw9hrT~Ukz9(keRP%(?oWGq`S z^KAT;b(V@u!kJOX3L&!Jx7+tUat z1>GT@-_;Ic&NnM2lE^fgF%5HdGKT8U8jgbcMGkqL^xq87y|ybokZD3ar4#Nx&F?aV zVmn_rvh6NazE;>t6d!CCkbT5VLZ+c?5O{B>sT95_7~4PCpX?L9EVXC7k9@p_LKBLg z%S5m9Nr%4Dj8sA44Z~EiqSij`S;0dSeTJa-wAJq~en`$zV;DtireP1FG|(XyN!?>^ zv>xma*DjdtIUqFU|0z6Cr@0*}3P_Yf!o<cKdt7EWM>Z0S4hT9k>6Iz`2p}L-rA|x6E2M!gECDmXYGcv$8!|mnJ(+g zy}(FK%j=U&JTU1F*XL8R{RKC6R`na$fN8)S~ zVj=|*6%S4?po7uR@S}Jos8(==MT^5_vH*`uhWA$J`>{A>`!h_7KKsvQ&6qUkG?G&@ zrA-Uwla@NXQ-U90TiU$ti{&n`lZbx24e*JAPWIw(Dnp;Cy23#`DiM$&bgDS2MV8mO zdkbGKw33KsI61CITAi6xr;=t z=jC$fNxmIf4^Yj6IbMsJF?~@e7stRkc_@EvzxZP>PwQTUeff+%)sVVr!4(a9vJo$+ z#F^MMr>|Y25;I=5zr)4DRnWI3Cvo1#ey^ePBH*aBnwci_R~~;J^oT&mkoWYMa@PW8 z-!!Gz`T3kc=XU*#+mqw#MgJ~#K{ZLlMXpF`IoTAj2r2X9um89oFDPR+sHiP&FIq^QZi zdCvviE0za$lSipbQ6uv)@Ma(SL+GYE#SX?ytn+LH*#ChDfK0Q@ewXXZd5_lRUph8} zOVf~H^P?l6H3NF&z)dZTcloN=YVJk{PSJ z-mQI^XN3V8E4thybc*4d#4#xZq^pj-L_QJ`D1xj1j-r??;;mg_bPNIX_R53x`gFRA z{u}#i=jiCHMw-S({FkN_a4((DhlDKqRurYS`lYl9xQ&esYAz<+uV26ZZTI4qHni0^X~Feb zulezbzY?frNJXNTZKj`qRAa0oc-mVwQ?9C}7m~VxA4|&n#+{em+1o-Ddxl~vvzhqe z&>div<miNeOy&Dyj88?qD#Xl$T! zl(j7+(@R!Q0X==BmMW2Ztd@nj&Dbk&`n!7j?N-*#zXs;u;1G3>Pf#HBYmOTbg+t}l zozd>KhtowkN)c1+tk-Z@B7v5=|H2NmEDto*78- z@~yLoriE7xT{O%~HkunCT!0mb?<-PBIXUUWk}96!-6t#>T<;uAXij#&w%-##=RzutD0!P48kO>L?HidxF!i0Kmn}}7y*bxs z4*v)EIrI5@mpECtrYQ`UNRkfk%V2s1WK7JS;Uv1G+$7A1(J!J@;c482m?T{A42394 z5qKG8W$pYg+g{o)))4@u*K_Af!~{szCv24MdPi*ll5lDyh+GjPE^%Q8Gny{8x?P?p zUwUocGqb@HXv zrH(xQ`Q9^@{I)2Yq=sN)a}$agCU!49_ezVQU|k%Se#U^XKdUH;2*@r8P(BK9azKRY z#}a$Uw^&inMJb=8rWJ;T)|faX6ez$jVUwewWV*IyQeJNYcrYkr?qf?J5T^u7V{~3x zoZEXMhM~1s?}nJOOi~)95M_w=78_c!12JY3OZVzb?ewr7h(Jk`NyNO~_;;84NF0|x zWMdf$b=}AJeM|ujIZ?xciiQUFa{ZK&AV{INLfdYsP@^rt+mq=I;{J)3hTi96>2Zc-@|^ZNph4h<&*i%&BQWcGS1)L8cH88 zbxbxn7y&OF5L~pcLX9_X-+wB6`{vEb$dwWH2h3hT-(!}lxwX|m#VD@ zU;>k)%ysUfMgnp1;;W&*-s@9P8SItXH)>&F*xTEe8|#UX%N;mc9NsbIs4Rqy*v$5r zh~b#Zvp0X6o#@`Y9q-w&o1a!H#a;;htnmrE;A$T{H{auUsb*S-Iy!Es{KtphOinYX zoYVDGGBsjTx8pFHHKoAaYz28_UOdV$It%4MBNt$_6!{lJe^tD@^`V4K`B9X6U5M;CYZM^T+t&x)+;OEF zrpqM@AHnWqr3V=kBb^ATdTlGqU6_VOBHqj_^X`Tp_F~UXgjg0XYcV|F!Hf$~8EhG& zKR)Q&{ES9gv@8-r&+V7zVamow8AaBg+AF z0LPXsu0Itzc#i$s7vmBU{g~EkaGTmlYwMbu<>KE%oi}9U<&lB<4}maiR&;;XumKXT z;6S^oQC{QXHcQ=B)#`PN2j`}NFjPAA5g>Zqc-+0T=#vC~B#c2=O!Vg&79|BsNdu!( zqN1O0)OOryVMHevAG&CdRJ@+gy;Ea0ggHH`yNwL>h{nIcfnF8mW_o5u>E1pO7G@r>AATUvih_0s(SgeYicJ zux=pcG!4y$UD}RS7(h^qA{!eUL!^OkX=xGgIMOe?3k?k|nNxkC5Ug3QabGkXV;!!w zIrnNvq<>&&6_hCO$6K}jkKVr?oiPEkXJu7#z;+;V2ExXzzC^s8#-Co0+IrqwN|;L0 z9(5i2p9ohEx^SYyDWp#OmoZ|3;1x3P9$HwApvG#=(S?ei{l}cU)Zx=-;!$7cdyWB`2he^>hw^k2<1196(_3tZ-zJB`a z8e1Pv34V@Ra9fwzsHms}LfN+}iq}qi8?V$2TH(b-UG?y#|JpVNW~mi0t^p==l=fUmf)Q8Pz)-X*6DnP;ko3hi+1P#Rb*|}}exPb*)D8+0N;gH}&AyN5a2Oi)qy+2C; zF6Um;QkjG!5snYT*j?~OLOEm4&8Sf;Q0p^rpBj4zK)ob^)65H<)kb|lmxuBpV8!}Y zdO~LI7_TLelM?t(Dbdb$=o0==;r)iBSBlwx`PPbvbWPR26_-`(Hlj%`=n9{~wCSJp z=~HQLbqQlrQ0fQPm)BlP8UYqXP+hTDjo-OJmlbf((Mm)n%S(QL+dp!x7YX}w&_mFM~(m@18zFtl>!A;ka(0NfmHRJ3FT|7Y&2Z9rjYGzz|Qe$ID<7DITRMyb_xQV$G6$Igdu7w1p=7814^tjW{^PqNU zqTW)usJOT=;Q2{jNlDh!nf9Fp-hxu~Wda0{`?4}J2(gH8Rc%K?f^nVHcZ*GR`HUa` zn5~%t-%TXg7ga+;%9=O8Sn%jiC2!oC{Fn7BCLjxlMZD3KvWepWxo5J_>;!ZSxdcj) zi9$In*Ta7V;0{laT?VO*@du4;26GNaQp|f814ac1J!J3_fg_utj0!8-gr%h)n*mO8 zYT2q&9>_<$|hHoIc2w!ffLAAbGAP+BiZRZg4c&c~feF9=Ik5Crd z#I>R<9*}1C@Yw?)mk7E|c;j^yOgmza7gGYx8^~wp%Ete`U)Q?sa)C=$hAO45p89XA zbLcBiM6ztBTCoyH^Y51(GHbQHYIO<0ogV?qhL35wjd!bB_Jxqea*_ zq9vgr@bv)lcL)&eD!RI`O|x*_R*7j}B|;EsK-^+$dTTdg1UKXqh$vv-NE$=P+6!nz z=&Knf5*VrCCuL{Hg0d0_DzYg!?rkFZTok0Z(TR!JmjYqo=kNO5kufmR#-}f)9i(9{ z50Al$5#k`W&BQ}7D&U`>QWm^~3SI(``YI}_hKF~4mxv1G__ECdS`#X&L5&7bwhj0j zk_fm$FoX5FKl9pvM9K*+s#i#nL~_+PMrLLyd3i>VCUD3}4~h?7-8@m?lZRehxRQ%} zjscu{WUy_A`ncoe$KrBTZD~Vka0)LF;ka5qln@4VUDoyZ&NtX-WrqA(@$y)QRy4kmH|a*k$9$&S^=#R%}&shV;59nX8<;X&Z5FI6ljO7}rLODs!`DTrj% zYcvLh!|?cc#K3#BrM6Q`uFr{|K1JCtw)U8K+&DD9-;6&qf(K17v*>QLGm!m1m3^C^ zMF5YV8+=+=?k6*4Ea5!^ez}8q56HG*|1kuZd0}^nGTwHTim&h>h&L>~3J$lXL4%xH ztY$qV`~X*SaG}Koc))_)_Ge8lKhHpwwqU7XPRAAThuc*#x^lhrJ3Y`T0@#0IblLv? z{t_h4`}E51LA1ZPDG{ThqW(7c5V3iA^pX`U$f+HAeT_+XZp@)Z-dXMv+xqn>1c<6@ z-9c~yo;%c6hs#JnoJd^3hPuH4&BY#$P6FGZx?J`y8!@COtH^?>j9rTk09K0EKXtH4 z!om9XtoB8m?*3AKBs4MtQA{{F^yI|xM0k0l>ewO3RKfo!Y`ey8p3io^6Jx5lD*?#7 zJVIouLD+Ne?}5&_4!BFZMqf!}U=*8?>Wvx4@3Z168_K=$AHOydC$yg*8uAIVoY!bj z08T}L_z-Gj7%XDVgxR?=eG(9xHnA7!e_BKY-1n5x;xHyD_gf*fx zG&GFDexwFm+;H*5vYwYHO9O>AvsE{1G3+N#_s8oQM*clf4|oFZ55(!uHRRxBK%g1| z4S95J=e4#m(UjB<(#;0;{R0D~!Cl9HH7mPCfmAC|Xwq}u-uej`KWEZ}XlW&@S`)aY z|2g@A_L*4HOIX0>!-M^_mb-m7=zjPQC{KizZ^ujzTN}y<4z$)A-W>hygiB#il-zB3 z4**V)#~V4IlK-;c298DGx~R3i<=XkKJL5KM-N5IZp@GZK+uzi6t81igt^tRg#B*m4 z)&knT0#Za2FgJAXI@17Z#rt9h;3i~+WXH1&g@c6#G{C%SO`VoEZL_`vy+eHwSaL^` z`nHR$1h-zdJJY3j5aaW$QK(4|Fx(ub{m7T?nqT8b<5B=~4pJ)yG4ASq{O&vvQ13yK z0!2&c$fhLp>Nwsv#yRADJX}!G&2qYO?5|j!5S^FR+Lyyg@=a~Y&l~9#oBBnowcOA{$|lkEnTuh2@BM6-q(<`&sF5F zxPkECd%frZ;5#-hc`(pXXc-t}Mcp?ayr4za_83KEy#01l60}_-01tGU>Erpn^an92 zFW`v;R9H3tDxe+!c*8*8(mz%D*1rx@oQ5^ehm-}5ML4+jKxtQT`_xMMz@VKcMsj?s zEqpxqx0c;xIFY6VP~-wM`>sO3_0`<*ajvhgZzjig0JeuCwl$JpUqDaNVH$8%o)@)p zoEMGrju(G950*N}fJd+wsDB{LhF5pqp_2V^=pJG3ibaweW)lV-58%tbNw4u;ZR}^J zHk!_=4FA`kmhMjmfIQY-p>NP#MetL}dkjMmYNrGN7Mz+eU^xYoUH#CbP_bM?Dfaw8 zC49Yb?zx$!Wz0jWr%!(k91cy({$y@{gg~V0+4zxSPQ%fU-opx7{_?+lLj=hOkC072 zaw@EtEDiK=2~)zx3^f?SCUmy>7ny)eYcU?Ak!iW!j zj~_LP6+gU=-cO!14*lVyS>({t37ZZ(QhO#CvUJl_u6mWNRJ?~suygGXTNvJAO2nC> zpaipAZG;JI4%4RYZ}Xo>fD;0~!a>rsVEI~4azkdOV7b;z?$>8tG#MG0zjfwl0R1Fo zWJLV_{hnOF6B2<*)eG{($!>98fq`us;kqaRk$k$jM$<|9-V0<{PYKO}Mf+-A>DpXR zGt@<;Yl2a=H~MQ})=u>NM4X(p0C4c+daj|&ENYa`^qSVL0Ez%qU^MPW3v+vCS-~D7 zf7*b*-=jDtRr|*s99BIWS1QFP+VxV&PdV=N3NfMngF@rIKWkdB_=XXuj zF9Z_TiY}A~;=U&lp^O*S@RuTJ;b z3kjgBFNU=kjShVM`KCg@1sBw~2*vAAg-E;74eG|2F_Y6R7p@^k%Q0NuxgXW}y_?>TMF%&aoxOBop7TmTCpt@V~ zYlaZ{OQ%+VmwS_6yydVQBR_!+P@IpKY|#Xk%d>Kj=@$hzUCA}84O#W@&}D_McFfA! z%+x^CVlimN-z3=7?(!Gm;^KZ7qp(((P$disZ;S8zm{3?y&|Sv@>=?jL(CTV;U|6^o z)b4Y9UQIo0pB==4t<@?})=LQo9}#Liu_6J`8a8}WAZUTCFzTX2BVv^{mqA$aT*sh) z`LfVsc?lU6^-LFWk@3~5q+=cH_Nl%7pS!L?(B>GhD1hUEF|)ex7AYt!)cDf&C2J(D zpnnrZ)RJNE(%bcTl48-v9SNds%>wNbg5C$e`|H&#hb{?eY4|}!;@JW>2Vhq#X_yAi zA{3}XoBTd`di=Y$U?VM`DB*a=2{{ygtIVxyd+nJCHr8@@#IgEDZ_W|{t_Fi=XQ|^g zrT>iya9>g^2Rw^}V~~Y|{HLREuIig6`iFF+m>nqpKvfseam3x++$?KQo~xmj7M=Qa z=sNQH7TvM54mX%O(1b^82K4Fl7<2HH_TR?um)DXrrxQi=1p#57*Iu)H_9qI@=cs!P z_#2xbXg9~%L*KcukjGJM>@W0gx;&WApzrgz^Bxnb>c!{01`;Hd@!q|fUi&ft$2%C_ z8+pHw3X|r$R(%T@-J%;Bi;Ccbsl}^j3k_U(w3U?9`<-nIy=uSimk-(3oH@ZEVUh>_ zumZhQ!a>`Ho>i*937`cD(o!tI7$6vl8j~-iK}%|BfV zV*a$O{6wXlF7svxKU+ux(k2z3vt`h3bbp`zm462mG*yjX&pIcq`hkN&MO`}-;Gx*Z zdF_SA;(SVard*Z6K+C1$%R~z@39WF_9q=he{(FwS(;0Y-q(U24|;_a z17Xc7eUxyM2~cd@Y@{dyht;&8c9Iu3XD0xMtbwvu&S24@%S--?J_fXF+L}~%iqByQ zT>?9C=@kXkQtW7T&Pfgo=b&oJNSQG$vX2{;g_9hv=Ln2{%TD@IhO9q%1 z;RSTi21J zaDY=~gApqjIy3AGe zZJ8521n_Ctl8z@6642HFK0eL3%d=(?YcciEhr_N1*oZ#NxUDYk6#dyM0~}J)a9~`g z0bVq}yQ?IAk9Nt-V}_9GOK}BcSJo8)rO|j0Ni<+|qRvq2;r;DwUPqr7NcoC4%)cc z6%cUg_kwWFB!ktJWvGlZ9`f1Mpgnzhdb)envHJs$3E=mf33VjOJgQ%)m@zatEK}B? zFB7it# zrM0gO8a+U_3@}JbX=tL{;%XaVR z`S3w<6q!Rir}1Kymu--iih$P|J+$i`nCO80spkV%X_Rg1#R;G`p{<#t7Es^C%s`hQ z0Q|yBwlP5!av+TXM>=Tqb~-iHc|?lsnmH>=ZPR=x`sz$J*exwB19ePe<4^{TX7QVV zGJ*r?Q&#L{%X_GkD>Nc(9WarZ{0@|0ni|(eUonNpLuTq2Mf9S9eT6caKPcMao^SK$ z4}O`c#6!S*@{9H_VJ)jZaQ4j-qaTDEmV^ZkFdcO?6esn`1+04)JuxM6BLc2v>gs-T< z)bf?ln#ko9uri6EquBD2R|*zwvo2$o_vPY zO4aw{W!^g%7jCdOn|jN4z;Ia^{RhTJwKvw5k4&+t@(v`7sQs$8c)%P6rl(sd37r=4 zz{obdr{F`G`Qu=nd96j4F^cia$}wV22{FB^IlJwsHO5ylUqCbGrGo6#)6&y-bV)N8 zI{U%RqzI~2(C+c*{`Tr%SdeXDuytJb(>n$cKBlyv`8YIY?{r0rZOn+oZht5v#!+Hg z0u$qQ^Ll(?Q*j2UfYQnAo59CY2%P`o-r3n9zP!ENDNOiQ*$}s#8xUV!-GWHWNCID^ zrkJ8bGpp}{5up3@lEFa(ekTorUDWJLNHDfp3m|)A>{5 z6`x}``v6RcoNd#c-oSS2ewEqVYr?Y6gRk&XgGEU26xJV`x*l0OIz+xVHl7}NborF0 zUiJA&sE=3!ze?&)b70_tU5!^xu(fe8ySvzh2wzUC!M!>g6%Q$VU*>qXTg>gSwq z=zBY2wAJZz6eBwq?px_?PC)hCKXkjv9`J*|>vq&iL`qR7yx)q=w?SeM&`0CS0e5Ou zop+>1E&GcHEVu7x1c$Dp0B`Q) zhb?~2Dh0Alg%f>NEuB`^SDjQG9!Cpy3^58kO+4$(Y@m@t0QlKgH1Efo!}FFd)T%nJ z{^1xc=Tphy-5lE1SjyTzXpU$o30~qWi9NaoHzR&t7XTl*Q<+SA5E~}PvH>=E@Y=8Q z`OyYYVhaDPuz~~i+2r{0%r{4vZ>Nhz2bh_SzR)ZuDw&1cXp@`I`xRZcaZ89F@?{F! zp#lpTU;H}sLq44T4|{L;=tAx%(CoTz#beW{HSrK$5O8B*W0{^Ri(R9>FaoRt94_Mg`=QO$hn zd&&_{7PJ;+Y;tJ=WMug(;dU`{(*JOOf4wsGfodQ{oq3ndXW>R;dL#M7nG7KRtS>(t zZ*9eD>MuWEI4mqk2xDXgZO5@WudOBh*^a-RqGVZiAYQqfwjXC&?vDE|>)G-^xU%KT z^GH;3Uj55;i6@a`JUm) z&r~7Zj4`mpoSyJf4xgqwz$j1VEALeeaYh27#urqQ$J1}zudDYX4$<1`k6x_Xl;Cz8 z`(a~acRcGo-1;33ztiPrrAJos$sIttr^29dhLx1ipmN`84Ri09JO+$pQAE~0o)6UR zW@4T7_bb+(wj;4^MdgKa{W;eS==u`Mc@F3`@yOIXv$!~zFTE9j^7f*zEVkBP z9mMgx*fu_tE^`?=&H^H9`+=q2iA3An+J&V#C`^{z2}GK`fXfA@SFHl(eRGHBh0X18 zV?vh^d-)>e#$woML|BMZodt*KiZhfhNK_% zmY@9l6G>#R03^|)4JBu-`ap>1UivjR1R7XB^3?(=bpL7|p6^fMFOly|9@$DDg59%9t zxK4y;BuGvu5{Jge7u$ibCZ5O572K~ZUHz+>v^MGO8})Y zK0n_Ng#C%lEx+-I<& z*FU?ywwGP4CL~|2vN_My(JM$si9kLx%@l=EUA?*9)z$ljyolu@`c}4Xupx5NE!?7Yv6PqbVX$fFk zUP1TL`EsnAtNaeP;l2|a$S~S;bo^Rp>s^nB)v4uXPwd=}YiQQv!enC;3mpe&4qQ7) z)~M4e!)!Zvt~aeGO7)l6+?O?k>rXpPsXqcSfys8Sbi~@N7}eUHlm@ozSslQ1XR8LI zemjly=W9zjWY9rol|E67c-r^EkVvazQgQkeQF9IDz)hP z6noKQkg^7xTOuCdluig<#T--LHjg|%9F45JJgr>JoID-a)Z@~T<8CV;HE+O*qrWv8 zm7$4U&p08P&C|ZRXK7A3aXE2vUdKpXy~9e`InaE;0pRkc>|A;P23@ZI6DsE^Z}Vj~ zExq-$SU}e=FxQwZ+wG|59pHYt^kUV!J<+&&gIShaUmizBjDG*lQKPBU8_);JJ4PlZ z>*mkf=C_+sQy;)_R<%yD&wI*jh}BBQ0?)nW$7+`sM`34CqZ4 zsq<^DgiIxidMsPjiRCM0$mC~b4iPiQPz}a0dA|CDP?txYGx8tw;w5F@0OynBvTZRX zueo8J`hbVV^@#q8Y_EB_w*vy-_8S*wThk06KE4>{3I!anCx9qo9THSsI28{VyScv3 zA;zixRi2(*Qxhwa4a`XN``y1~b$N3&F*_T_-2U|2X7Tlpm06y{YsnlW(uFsl{XE07 z(rRO`Zh_GWl0eSz%7P2!zK0LxIb$^~Hr1_EW;UDunXQ&ySs9rb%AN47v@s$?7~gZ6 z#@60Gx2raiSGzXtD?D`n7K8Wmua7_Vn#k!zySK(gX}drew!ar4UH}1=b4hb)Np+X@ zUAp8ek{1G5 zmqZ3N{wm`7hK4%pzT`z4YdOV|@^X3_yA?^A*qw?2n%5Bh5zqwE-K)$C-bWkZr@GCD zqkM%zz)GBY(>lAq24iF|r(b7YZXM0)wlL4{IGax<7G^{Zgveg*OJDlOvqkUDf_e0x z4>V$FDBkQI9Aws%#`-w(%fWu#KoOab{>k3#=;So9QyTTnlO4+EzE_C*wVYpjM#f?V zG#6goT1202hGXj1hpFGuZnd|Btan+FwLsA_!wTmNIXoRdP1xTvQi9!o<>}Z}B|(+- zpr~>@zcv#EB$EAsx|mwcc1r&7q2+}GfbM^*S$Y0;tY$@)jGru5C=cWrACcYxBZ}IA znOZ$SI3eKV#No|HBwi-Vfuo!+M}GN9kgEa?X2t^uhGJjitH%wVXPv0@YnGPn6sO~y z?CfYjW9^d`Hy_bbM^;>S&Hvo4`R*N@gkGyp!{FMEssZFjHLE@n&g)wpHJ72OS6Gm= zfpt1>x{_A^CWZ6E!pgRN%LD$+iW^V1jQ_9lpMf(T5N2v>O8^OIwOqGjIU&xo23Q1x zxypfEm6(B)-jYY=;{44} zJ;<6$bqd_6xiTemw)=9F(TL`b6N$4xQo*`<1*bLZ$_M3{UI&Sn#Y*j;q>wc-HPv}d zS<^gG*C%Bq^;Vv)#Wj6Vs7selAXMC2aygQTsGw>S|D3;DR%{v<2{WfD_AS1mv`tZ& zp^Sarsh?Xj$5!uI3$q-ob-Zfhmy{$KM8 z3+2-l0O9s?zlr^5;&P|v^OolMo}^m>KPH|BiYYsE_$BX15?AC#C`SrgVuA z<8)Y8ZQ+nb3c0MW#a^8#VDVc`xizz8w4WRA@T>>v2N&Fy?P`QTK1RuP!^091oM)X? zNk5Nlmc8gnkEBz963fug5J1xn1u#107HjRv65c7`$sl^xZteam3My_r`iUHgt~he- z5J=2+j6N#cEjmr(DIZPksYg}pW0)t}8UTdJt;}R}M4m}k$seDb?Mz&G)CcY(V0DKF z8(E^60JMHyZ57b|6@>}Oa+R+z>$AjDQB{5Qxyk^4PT`<9;;N<3u@Z>dyn_ zYuR;Pdb&ICR8~NLVPbk(FkemtcTzT2Y^!!7bfaU(EiTmMPzQ(y)eN*$ephUp0&y~) z%cUyY?W*S)r+H%Lob~qbE33Tmtd&KECOVU;ts4m5B>aRfDopm4`O2j_ZPoP$=!Lxb6Yj%vo5g|_IRsuPQytvV6~f8Z0sByud8HR{R<f~~y$fc+-0Z+W16)Ca30mOv4sg~>|8|xt7 z4m8c)HD7!QYisL|i}Sl@XsQ-UHUuWdxOTWR!(PDi3?^^|Vo0)vhuk)q&sBJfk@5p( zi7eHbLGylH4-atpqYD!QgV>C0=e@)eU*$?Ra)^6;@iDo~MXd+QX1>rKXm=Pc?E~Ns zw&&Z8Ir|m&4`lqB#^q93>G|$DA>2MQ&D)UMX32D^pF-ZDjswza_uisxJ54(RZJPO= z01Z>Q%5ba*3~ovQeKpAKB0fa1-)8OHxQuM-6Y1*pJO$#r~(Yl;t~Y4nqH*_XAX z9{CHi;_w*=T=8Uc#h22n1A86lr>x{t1?Ncm5Wq%7iXzcq;*HU-Xv7E|bm|UhV{$29 z%log#hsZ)xhknR1^>;B${k~ib9tt+979F`5kD3rBt@vHnzduzt0XqvYLNmt#YA^XK za`J(&hhn;N&V;>mrS#S_r)iQj;^%ZbNsTB@J%r1xWJiYTBQWMdTqVqYJWF_QGWP3e zVUwZ5oV(cHzJdjDT;^=93sX~5X>Dz|GU;tdI@b2Q?T>63ir>f$$E}V%L=X0=5n>s_ zu&u3VFte~s%+Dv7ousEmhEI=7!t%U&@r6Xvv3Owg1bar3 z?w;87cPW%{+s}}_JE41Lk`cuQH2becQW5lY*Z@%C^CzUP@5=hdskuPmCirRoEy%Jj zR2|{^cPjA^_yFhL>0$5SK=kGYs3Bztnn#$@yY~NbyG&rL6H8FxR%TS7_|fB)XqkBv zcevxAO2W0hNrP$aD1^!#@S2GMhn*XY#L4@E^qoTw^DjXU;j z&>w8tWSThVk_<9;J~x{|xxy)%ghG0ls}9EoCuLxY3Hqn&~#j*y47YJd;JpwWr#7BO;0@yfD1ta&i2NcnwRn=7j@dQL%J7f{oEgJ&(;(OJo za578g#x8ishn3yig0I`Kzn@WB8VUdkvbTgIpbsGr7Nd8zUe*JnH^&c5BpmCB3Y5`p z2aOR628-YYi;7}^y#a)4gn((6u9o-x-V>pDR(Qv@pMr-kPQK`!;$PvmY!C`h>oTt;hnSV!3Oo2wN{Cp%*bP+uPj&8`M^MYea@{;{RgXTGb{Wb_^kGqfjV01*!N^G*Z zdMsr|@PhsKi^c;^cZWr4;aZMDdu#8s#Y`4H{UW5a+KaFKg0mSYMV>zqI;ZlZu|>Uh z24Z4j)G03*7hRWQ)|MAWhx_JdaHrN5%a5d;29Nh$m!mhDukUJZ=8DvFSrZPrs#9cj zCC3xE0Dk6 zqr-!_;aZ(kWU+HM_Nyg8|3DmJyB+xOy$lZOUy5h~o5~$`UbHb)}(TBw~C2d)Ha&Rko zFrZ@R-5+vjy4n2W!@U9|k}qW@a|{|U_bjLXbIr9B9ibE12PTEkCKlJ~?tP|{;1l)z zNWp8}8f??hU4cX4<*DVRbUS^KNhL{dyn`OxK7r*_r+qfkb>_L0GuNSphr8<>7rGc? zrw!A_VaA4mi%_ z6Rk@d_BB`dwOa^`2lN0YxmW^u|8bil>6a@p1Qp#HTych=wdT}bS)%SQ;cWtp3J59! zy6bsdA-)_rzQeyjG!I8g^B4#!cHK(B$CDBpViB&d7g}D{NJvPCW1*!y$T$rC!Z7{u zu+Qvf2lFskk0IJ$>x;SD`#JhdF0{~{kJbLpA=n2;Qi(=Pkpof`@@34FOp}v~V3Rfb z_u=Z&g#7QnVDN%`Y_WFz{d#8ybQvyhu62TlX2OU8{>9aY9|GMGg4Y6RpqJSH#R7DP z?eF{f^;+1tpoepPCwBVs@5lCMTS_lTrnK144_Rmi?)@Cb$3uU=*PltEcmJQy{QIY$ z_`sR^_ca;c|NAjuOaAYM{J&cC|62{|U6`QdeLd8i@mx>P`%z0>4~axh-QA&v95Q|O-{p%U zUJH=EffVR>D*k_bB=+eCT?!gYL-mqJZEKx@iYE&-ydodClen z{+z8{C+^-N%X^Ah3~U1?`lU3@`mf(T-x>7!gZGY#xQNJF3fN?QwOK8&3L}rrw5j64 zM&Gh_Ye55GW%|7F$x%O9>+jKy+P;tGBdPD z`6|r1wX%(b^dws|gx$!Q?kg z+GUS>yG=TOUxMp~IyuLgP7dri8fgN9G{-xV`12jFRZwIIY;CvR>_8m;^SV(muDZrX z!&_##n{Bg*OAU_c)wgHd2%l%-))YgRx<&|}QM{)S^-3#zHS>d*a9MORjp}_yL5rhW zi1ORpIj3kRjeqBYC(c~n-D3oopM9*Jmte$5wg}eAG!K^fdzf&0ubc5Al(Nmz-V~{& zTln3E!Q^NSE!;(9p}~$zw%~clKvgF_*pN z%s8Lr(pc7jRm|x`5RC2yDQ5>Dnltc}Yr(q62c$`D^ODNsf{W@=0o@qLLFfl0eK*74=Y)mg?WCf#t`Qfp# z)a+*;(3VnRE`olWV?e~)3%V>b$i}%7K!8sP@^!#YfoO??ZbI0``o_zZXji=PNA@^=otm1q3T<@G4_{GbxSUwGCw8p4; z3*k_P(H_mkk~Ii(dJssGIZfSI1kzBaV2~Ih-~~bUsu<%z8Bf}$;=HSTT_}4wolHD; zw9|H_aC{Z%UFA$2qVv8SRPoyek+h6-(BOGE5g~sMbh0vCK2Ka$Dac?ly{~=BbFA22 zk_Sp8jC3^_%JDkk_*-*cxALFHsz{hhvWvZv&o|7#6?1D!6MeclrwRz+0!WVxnUSlw ztfsXJFgH{g(g!rDT;Y71&%$aa{yeS}B1-}u*`d61qSb^mmn-EXNb=0taPLIa{?cs)!YUEMWJj&2@nAX~Qxldr4&O~i z=;h$)!8Hmx3Mq%UFQ;jl8xVZ^#TSUhyKH|}5R)DH60YicVJ*RP5oBYU7E&u+kwf~; z0IW*1GRvEKH7B#T!(-kQ5=&`JzU7kCw_5we8-?5R3!Z*V7ZWLjNtcfwgJx>^iv-Cg z3(Hf$@9g$=7MiqQPrG+P!a_Hb1asD}dXT9bOb~O~X++Jci;%#1I7m7@bgCSKYt>T~ zNTuAAq8l10h;fRqq39ix!ugalsEZVm*(a<+Td(8|f}UwZXbXhXR)=O_4$83n^sPXL zDL&dSM*+_AvwJ$}obMq;x8T8VBIR%R$viG(x+H~n?QUv)-HjuM#Mha$s-!jrv#p`H z*@&lh!>hcscQ`1A6S9|$rhRY+gvC7^rOy-tllqeZ?7kY~YnNL_68NSRjZjn)fuoZ7=BW(DD%#FA~~lR}b-5J}rv-%P%*fr~dp zEQS&Tm8RQlXe}b`<+f+P-09tuvWd;xW8yYKyO?>8%I%ST&b-)?!0v4~f)&5RO?E(O zpm)HmNKfAf5jfnV6@9a{+-+lh5)qCdRZ$#?Ye6W%-na-An1+NL{`b@*L>!eSC1#Q#|c|xi%qB@kp z<;B4{<|=_$35HS3(hYQc3oWQ9bNQByJDo<6u!0YhwR5smX9vaaOJ?=rVTXQKp zcm1I)!Y{~hIV@7W(PgN*9-l*I@qOYsX*20Y@0ji}(sYnZm~{F1yK8Er|MonWMcFnz z*+>}W8oCrnfYfn{=Z3#Z;Cu2b8)s&jS~9C_{2g+1-l;ZLQc)A`k)D{1JLtYZjx@@v zBZheZ(@4g*r9;8>pw(S1!)Xl(&Kh8PKU0#@OvZ}M6(x65IE8p{gpQ?=$!%rgnE;&zn!VD$FwkkrK1$$QW?8B9=irA1V9_S$aUar!-7(eNQBFZ4w zu#!bPxurmTEqUykKa=E{IZp#lr6i+RwVWzmYbXq|sLZsWlf_}sJ{2fcXbHVSUR1T( zz32HvSVN@}N(Fb%N~;0^p@v0Bg~LKO#9y$1D8XU9-EHxf>-c3N(*~~`ZQx$feJJc( z?cT@TbYy9=fDJe=jTm6c9U+D3VkcB-YENYomkC8IJwIakx6#CZ7Q80?>YUFrl9VCF zpN8VPu<2@thCQSsSuORe&Ikl{qhXAX3t5{~Yc6Kzt|nDoTsuL?OpusV7OMCrL8>S? zBIdov89#x`{>$w~!4w??!-JAlx;>_mssFSVBcd^Tx^qehPpQcag{Se%k$GsqY9yJ~ zL)GsC;#=auAEiUC`wnH=j5fB6=4Y0T^Goki_+#GuSj30~cABd@i`kf`l^;w)E4BTwBU(4Yb5pdp z3JQe%L@NKs^HTo>o4KS){03*F)f)hP~g_AJs$+K*h0mlvTn~ zW`Bez=6j*O^FsH>oQ@d~jYq&WZ5 zJBD-;@N7#vQlx>pij-EHuA=JLZ~z z*6V4qt-K#J98eWX-D9fT>@6}r>tvSKhK#*kRM#z2!uU{2L_(WIrbC3$&6M8$USFG( zNe^$F&&BL=WBx0Bbh#WH?XOq_fwmOCUL}zj-TeeI22xmYYZm>^za#cA)3>;<&(L+B z!!WTk@~d(snBGT!rK?=((j;n%UarHY5m`B#+pv%$q{D@l%2(c4OU6kpw`$vmwEtMi zA|u~}QoBCzuyu=qPAX{T6yXbsvRaueEa)s6^Vg}pdC&~S3(J3cu^G@zsY-EqZ=Y@r`nSzbH0Tm~_rC%nPRC`z2x zLUu^$_%;PmbG%aAP2L+XGe;^6Q1?=XBQs0GT+Ex~lgEIvuQV0XQD5R3G&Iem@Y;!O zenVU8S&iANo6ZE&+`{jx08+sd3SZe3z$YOM$wI*#nRnNsIYPzOZxF<)VMS-rb;-N+ z(Kp@;fN5*LZF-MQV(`H7{(bm7$UHUy^gH!%t9tA+P_dQjU?|qZM%x47Mk`wIW6WSf z!mJ(Y9w{{+m!qR{9KnY(eWbZTsf=FTIpu$*yrK}`KV_ZR>F+9XYHg~-&CLtac*SZc zwGpPM4sXWjpK_8brtKT=n$=W#UI$}@o}*mmv39tjAYxG^pShk#zUSFUyIAl1C)pt6 z--54-rH>3tW?@lCa7&<779SSd54d_JO>Ob!)b1y;33Y0e`cxEt679K2KHQj2n*t;UiI0ca_EHo$pX5?m z!NIZux`5B;vEhfGZ90-R9;=Z#5L<>300#!G=^ zzJ>h3r;jcMyJg*P)b!K>dD~eIVWdB5B%2pwZi|=_5)zqGHz#m+a-)(b(9{AXz z{BRaN5-2u7ZI!%Jx!L*dpWIa17)PG{L-HfxaVub~5vk9(W&g5p4>G&XyEmW~q-z@v zf$1~X!{!bd-KIIYZ&^-!(E;2CIM^Cm`d^g^BLG7PLtFf#v@O4aoWl83Vl)5BM}D(Z zg?sV$NL9Bq-F$sW>+8c64W06|uw+F6xh1Nw{nc!|vivZ#oHt+4xUa!*hb*xM{z4}a zbJf~b2AOj;4XPCz76_in$Ym2=b<&t4E=g25g|Rb^k#~b|v{_s$UJwQM<+nik3K(b7 zzYYO$OrQgS7mk=#l>B-k7m0XU*Ox*$1O46+qs;mJ_fZ+h$=%%tGT@(8s*@p`%CK-w zkZdD79Sdkt(xB;G>X!*C0+)58?coFoMJewf3^+$St@<1Y;$^oxy>L38XRxQq3~9V6 z{^z2;0yd8@!xnE|tzO&cxa5P)mxtCKR)_Ngx@i*R_RG7Em*U7k*#RhGgjRO{N=Ao+ z`3^>uro84a>?!DXq-oj~p)4K35HtTgcsEx68TW$l4xio^kTt*1`EPI7Aj^!G zjrA*KWR3j({aD3|$8UwblJTA2*x(kkBvc!>zkPBu2iMm$fOImB%raa?HL1oIC&emb zbBBzxDCj=I&#pV1xq>Q)i-LdO`(Qe=wi>T0CvJ(GR^GL)+gsmm^23w!?@Hy%_3APp zUXr5!^DPxZQ#s!D<2WNkW`R8c?Km_KrGXVWFV;~IzH6QcyYd}1Dy{eP<*}yEFaTKd$p0U#@SpDX)ig^p6D`Nn1{rR zXInqKZDVZ_#rFSK^8xZ>yKG|=($mjxsJ!C;Mo(R-U27_qH|KX2%?zave1vZgqr+#C zvZx{R>eIqooz1IPp#g*8%8sMuIZrnUwN=Qnok29cl@&4ePdpSP}9<1@S-~0 zWuV#FMfjWr-Gm?rI7mOg!#^<&3Cg^2!3Vo87a}5z41=6nsENrRvF&SS<#_!6=k^L~Mb~d^%z4$nYHTo{KM# z9}FMW6=tcX(A*dSDJ@FRgv1ZkA)|YUf+?f)mjv9G3D8MKf&{gU2Hpcba!$e(l-sQd zd+USDn~8HD>>}nSxjX^;0HKj$|K<@T{v4cluAh3^=&>oAJ{>!qvlWyUZCo+tgc%wB zOqF$C?o9u8c)L4$BPlamW^1#}eI8BR5uucO*7RQFxo?)fkxN*YGvZP(ae(-eSRVp@ zd7oh^Dtt!8vJkHl965twNa^WURJiEOlx(ds+N_LkW15T5r7Ey&=>aNhN#w$!N_jF& zP;52f3#reaW`GIYm7y*ei3}t`8yj8e?$}WRKvo!|w%ge-Epz1(L#2wyrK)wTd#@0S zfRCSy5&oaj=|MnJxD0OSNpPL+37bCJX{#XegD*rg-h6mRnw6DnSr) zBz?r9OVpxUr3H9iD~^CaSC_%vMd7G7KuL$@a6&cU(3 z+bJr$ilrdZ6h7qq21yl~Y|3^RmGVkVHnJSN)2zB4VQTP+GiICqLeFUW;=Vp=&V?4M zExI;_s&_Vr@Emga4?R+xGz+WHWjO|1nh(o0w;P9V2viaN5F)-yb2pu)&6i6rfA1Jr zW=0$t${iH^{SywuB?rmJI7u)8TPa%h{<#w01z- z{Z#hatiV7Qe^<4IICKY_P}JcgZ>)6RZe?iXsYc-85Uf#*ucBrI+fVOhmsXaFeK(Ie zWibIU?@#iEdO5g?xuVLR?d5#KEt7T18#`DoAqi~jR3S3L zf@yrn+vlA{mwL#MSdu2@8iGvibgaxTN-Bv&u!PHOvC|Y9mOB1jR3}BFgQn`DrkrrY zR=*-pOb&_OKh=e0pLg$O9oj$S9Y(}7m%H}0flZjA<(}<2^pWtHz`>;MNx1+2*d6Y% zu=g5q@;o zL_^b=PD`+^cLyBW`(H&&gQM&yMn#b^eo^+~VqlI-eDF04Hre%h8$zNSfJdk@12x53T{vBcJV>Z^gjHg8W&N>4i5n=2zA)3}KY+Vx)Wd;GW?T$~PLQ{vr7 zkFJDM$;Mow+Or{=>M3kGx{S;WQp(IFvH5~A%`9{L+W82A!bPu_wm6*xvf6pcm&pA? z&gUw;{wVeCK1v%{8iQC5LjBT7_pQgV?a)xq2+TvHq}gMsh{NAZkD?yku!3_ zpI6^-=NLn(BA@AlB_s$1bBM|waq2l9wx=pno zKYbF{{j5|Yys7vgfAN{u{=;q;@u?D0U;X@M4!>&*m(T?34 z<1MAQFlTvzvr|Y}yrhyaNLgMUav)+&_biwA$nPcX5%YaR;4IHmNZkJ>#ZjANN*R zP@Wd7Ta>OwbvB^62!&5=MPVC&JCN7*YtT8poMrFUUai#o^o2uo~2#jF0JbH4Q%+{9=^0 zz=BSvH)@6JG^Q~fLgPR7^C#}S9Z<4>J}rcq z&36jsUNF}>zlCrQmj0$#(I-^r3O;BU?FKD;XR{sGp$t0ML8;5Ng}rX@m#s>05Gbwl z?&kBRXil{22Apn{LeAlz|F*=WOMcq(bLTr|^W*hi9n{=fHw?Fa)VEH3Q5ERqsQ1(z zZ=E`xWV%HPZ4FWJa~x0{NCr0Sx!#KZVb)iNryR>0QgHu!NL>W+V*c7mR5d1NM5ZpA zINA79;Q~dwJz*i9e6|u(EM?Bcr@EO6@usSehqxHm!)7V$kbpFW{h)p@p2{R} zTJ_n34*slypc{1m@X)tqwQ)FitAar+Cy`#%8U9W4L>| z5jgl@ORvXW=Wwxg%9Nma-i&Cuhz2I=*y$jl!KsfbhE&wN#fD#8i#!4$v5E~bjp4Oh ztq`k%)o#;~ZVOqcVOuRTdB-gA7~%(6oB8K3|0~Wg7v&4zOGWB-ndyLgrI3Wj;h++w zNOJ`Zm2dRd|6&1X?aRyS0)1aj5Ko2(ZeE#pO z`q=9WLfiCYPaI3??+-pM@T*u+U(S7)8#S|S83lRNcK`bM6I*nu^z`=(4nflck)e6U zChOZX$3;3kef5$<`u|m=ETdJ#7=Jh`A z1HF^t*hU-%3CzD0xaaPYu5l~HjyP^+-x3wxgz)GhUQK@|oT3D8rb~H=Cu*;cM`*3!BehX6{%Ch|Nfq`t4#^)m>hy3!voFc+#fpSavt4+5z>tFUk z@TPJZ+M+x~>Y~QR7`P^HQU>&++mg!uol^imLa&1&!P%_Az}+-QxjczS%c9cjAYF#_ zR$9)mqqc+qMo+egA1pKp#q+mFDAeDWJt&MH?4vinA9ak)MLIni%)LOM;O9DB<_hp+*g(@52-0IEBa)*{DWlGrq+@KW|wNFfp&15l_lr!E^^XeGZ;LOPuH~Ie; z|M>k}s?^F%qY79VY(3G2+t!{&-9UM^9R>|5WO^{X2U@xRL?dlD|Tr7BdP4~a6>O{U1alU7|e+^3tqkTnKpsNgV{oSau21KR$BW*5G|rt zv>b5w<-hM#K*01uM8Ga-lSe|myPvmfP4~~*}nhJ zY9okJJWkMfz`A0K!-S-(jQXQ_p+P*A!9(WsZ%OY1JetCS&f3mu>PgnZ29x^yG<7g; zQ9nMks^lnG*B%5uw#36-)Lqc(2h6ymP(_P)1CUYlw}`i-Z~j~Ohti#jOYiUZg?KUN zX#sOMa|^i2pTX$fmasGI;CD2)1hZdjk&Hc@kc7vh>9Ds@;$nt2^b(ZR?JP8@Ejp86 z?iL?-`fcn#plrgM@RkteQnXj8Bn4jywI>^Y5T z3egERAI3Teem%b-gnpCHw|i7ug+$UVy>(y1R~8e0c!&Gch}}pewZl>k0CF1rHXF}d zew(gk4$aWL`{wm{o?)Syxy^2B$~NeWUQ@zSF{{`b5E(Oq<#z1~ zY-FKu1t{?F;es%@-k>UVSU0-|xKl5gP}*2;-86^naL83hIqc-SwPo;Qh7B0$ntzcl zt4UNcw!|~Y^&*kf2fxE8p13IDFI?y}efdM27@%%~>NL0h>Fyk| z@-=SaCuGi_!La-8o*F%Cf@fnzxtG>XN!5lT>Vdp;Zm>i2xUH>XH|*VBT9?Inx47SCkB~DnRA(6?fV(+j~ zhdu1HqYr;%V=qCbmN^UXUU@N67+%s?>?ar{!$C0WB$xK*wPI-Q*V8rkUbrMW3&jW} zj9yNRmt{%q;Q-g6@$^F_YLXBel{Cp`c>bgyQLR?X?kN>%c>L7ZX3ndm&-X9S!b6HUbk-~BY1I9?<+aiNdT$J~>DWJ8h-4M%rKPS% z*KlXd-6DfjMhxCY;bpl>@{qh2FqevnxK>V}IF~5i%9MMf!5iq0mU@d`k$l0B>HUp| zn$nKSqu3grE~143xGOwSbD2ZdFpHkmodPD4Yx~j7QWx&Q!|8OcmP%R<4kGHT;k~rp zsyUW9ZSrIHVElVg_IFNy6|~ei z(Ku&JRaa#Gt<8^?m*Ug#)C9rRTwCManUL-bZf_fg-L`)Y+GzaW1hE8?Kc|+4d*{OK z?t$<7byhi3V+o(Iil?w!|DN#Ai;uJZc=^(%p-KM0i(af=`HR>oifM}>YQQzMS^jnF z;LG);RWhqGJ_V{qr*Bd&tJjD9sD6_VMiP6^Nj7|UxQ;;e_T6@E0^3kVE03w@!0Xg9==Hl|Yq|@w zh2_Hrq>0Tw2Z`D%1Taph!9l|q)lidzN!6UytymwZ5w0CQQ9?ib*ntLKBma7d%TGGU z7P z9&b48I8+uif7l>p05K%*Xx=idUGSySOEzL1TbQ)|jc0K*REVpH)mtb33PimQjVOz( z(v7W)K6>JRf_3yVO_a~dI1n)ABomj_;m};4nk_DXf22B4SWKZ$)8*QA6&Ad9j0|D&D1o>8BR^SSZ*<^~$7XImYl(^6;Dkb_|G4Luf0yG9B`Jaek;B{EPIP}-CHk$4 zQ;!E10MlOWhYsJX!e3kt{uW4Pl=Kwy)~QM0edwVrJs!bQ=DAQ}b!L}cqa0^6p47h$ z`QX5ILIg+x zqFir8U@7n40kRY}&sHx$w)1~rM`29_#Xwxj8fJXfK^O3g-Q)8Sn_#33? zl+mz^<}JZ3MUoy=BHE&#Y!80dv~A-1Thb)xyehEN!+YQ%+}qaQCd~f|Kv7`|Ig8tY z?c}0VQKrHq#4Y`S;N=?*6{TPzb8blfR#%jOh8_9TWZ`KO^x@yVQhQWz^>hSv{wRba zf?-5xfVn1lyCMxAlJQj&5Tjr>#?fo37*KK1u!qm!2!=vnYM@%B4jXtxis)X~K5vh` z%g&7i5Dhs4ND~pDJ)5Z#vSF5y>oW?;tW(Ys4I=-*VRt3Pj9cdWh=k6ul`f z;F0*J`9+xPpV5jfY!j+Pq!VU>Zd{2~_wkTS-EB;LB$r*0Oy1PtArruV_l*+MX(h*` z9&BsGBXj}#HXJwj!kq)JMZ3Oz)_5piidk6wgX(RCFmj>LmHLf6Osf=*BD?+bAP%e^ z?CHgZn}-Ns4nU^#pmzEux+U57u&!g7($bc4aNf-5AD#CrZFhNPvFB~;m;?R^5&g)^ z`l9`w5)n_yytO_Fd}g?;bPh#a;d6uK(vKne9IJl)K&hb)%giF?iBjE%2Kc$>TXO=*a<;BKX z{GQ~nI<_enE~z{yZ~v51?*h9Xl$WWL3B1|eSbVnRT@VvFRZ8L61xXc}d{$rBRwNtJ z?LtP)oZ+D0+dYfXcG#G-jl47^J;1mbAzIbp?TH}@v8@Rx>3s9qpe!WFmk1>hxta{& zZadahBoLwPj_PJ67=S;b3?1s6|d#d zN&gAFW5k81IL}v98NZzwih|_H zgYWN0G5_(C0c=UQ3jYFF0@HNtG)zxxiqe{^Lg##U|Rt6W-kGt!W}V_t7>pO-RBT7aaKU&V?0qx!u`Sc6a`I)%41Ukj`1BeHjB`)=J^y z@NDQOo@g)We;gz5@CW0sKMYi6Vkh(MS5DQ}_*1T-omm8LzWJCD^c#Pa+Wzp-T96rL zrLCPpg*|Pa-#9VlwF;>neoW}Wg~;pui;oCKg3Q0egCn`4UeJ~F&N&2JN^8-S{9uX6 zNHV)i`{$Mf+<3(q(>lZzF1|y+xCV$Qb{yj6Hq)_#N2ZCL!!(11?Hoqggdi(0IgS+QHJM$(^5O!*g7ExG@ zE32wI!jkz1?DukPGu(HsCn-cI;pl`p;UQfJwD7Q0h58>e(xS9V+3YGS9k~)`x`-8f zH2+zF7bp6iOhGWu!)ml;|7_GaY5iVMkig6->9P%$Q7Qam(k7!%Q#cgzn3bgZDdfW2MVY0kA?4Q1nGG5fk#C55<}%-Xw>Jr$u|{FKu}H+ke7HOo58Fq<~2&ze1i z{!30OzEnFOjSLqoWU3Gx28ITUUZ58n2tpX&TlJ1u>a=@6INVkQPsDliC6HS;7>mAbA=hCu+LjO|E_l^7>;< zWJYFtbBl>T_pcM^NaoPKfMqRSEWEi|S890-D^KQm6e`GrQx1R&80+vQYqL&nxq)=rh?e~uif(xn zZ{!gpT403L8M2cH7&AO56*;;BDm4wzhiH%cSL6QB!FtSOx~lp5;0}D&=*Y-!V!47V zx$d}7sP-~pB&vI2uUU5a79;n1=i=S#e@!D$vGhAaFTI5{vOvKm9UT3>7W`;Z=D;up2*ZY%oXR2bKE z+`MNPr^p`77d2QR!s;_!{hV$OKL1+|9c+1XzN$B>s|&A-UmD7)l(DL!r9X5Oqh8Vl z_l_3^^WtLPQBaq;sZ+uLppj}koXfWQ$#LVDN4{1lBwp+}2Rt1T-+x`E zhgd}V1-IX~qDL7Se=zxr6vvJc#l8njK3QU8E>_BbwOG~|F=2KsHG&d8PBr~zH&;%% znUKHX(*^u59K2uNI=@52lCj^F-MrST_({{gu8+Fx{sAWd+m=$b@}Bm4dH>h{*}D?P z6pj{Zp1*L|tE{;k6+aKY9?|DSBV(?TZ45Y8v6u+HQE%(k+izTu;rg}g+A{9tOujIG z)G3v9lTloCA9>#MoG`aXNp4JC9F5&9i#v|aL}%`acVL|i2DJ_yUWrwlkNi*SzY&vX zEqHmLUJy%$)wx-McK5LIsUl zysR$tOf)k3C9ij8D-NPdxKBnkJ6=BRwZ3pdsSz(o& zo^>>X5keN!gx#+Zi3&%J%+zUw%&RP{?7O392{+}sk zTy@#D+QE41ysIU!cE9oETP70s5{M=?TosE#vE6_!|%vkp-gx}iBkke;Ye98weVk~ARG%MY3cCFhW*nh zo^anW@vE%kCFRyxdi&>>WFPsl*zQ*0yu!hZ-{v43#HNqX>PDGb9$| zRmD|`xM=a+7FE|2=p&fa9_<$p~;4SZI;Q!_GCZl=4*hpiDJ zeYljJYYV;57w+htkm@J#icXH{%|LCAs&kZ%ywPn3!WiP@Ugn1h><_pZb z^_kzeBqOfqR~m@#K4jnT=S{G?)FS0oz{1tc&MO&Lu9L6#Cz;Q1HKnTXQ>z{dX!9FC zV5J4EwH?#k9?gynK4{4o8NTG;CP$Pm;FjZ06-XaSr!;&^1{QQbQ@$ zi@~M)gZmU3ZxE=N<9_n9M(xe(9d~OSGkh8NaKY8AF;-%aEh&zsrO^vR(@c(dZ=@?Y z29&I?wOO<5JU=GTQ#+M*W#O-})PfZh75lgcJcRxy(;}_bX{g246cRuFDpdL0*f>=~ zGc#stMjnmtRoj`%?TOEa@5wFVX8H^-@1)|oB|>)8Ln#s-iH9yUO_6!_@9GVSJ%1dm zA2VHvK~QFU7_Xtm3~B4Qx#{j&Tg9`D{un|Ox3lN2<-=`%nrc_SPZtskn48nF{xqNc zZ$aJ!q7<^v2FJ9Fh_7VP*~=rW<)tFGEAUy8p`pN?FFwna;Pxgx#6Wa0>aE?d{;nKc zdX9u0HfWHZl}|IX>ap1Gf}*|rEv;-go!N5e4(s&Bf}~`@$AbZ3-gm8|Z`*e9Y39y( z=djV~L6gCDTGr6&_9S;~a#G0N-hSmVDC66I`s9ZXHf^_D-xvPWu>Ol+B{?!vIX~3ESu9@v;6OWNvMUJ z5XbgS5=cVnCU}a2S}W#T8>A9CSI=H@-LjbMNb@brWi295&gsl57h`_A{I0bKlJCF}4q@Fsbp+;5=TX$9xec zzbAjO^X{aomlmFmoJg*T{Yk&1c}tCT&KV0Op6+*j_vpSKe_hEey_k0R~>Kaz?xN0IoptdbwU z?nwx_$%OI!G?Dl>wd&zJ(txdcV!SB-Rg;o+3N8k+s(U5%GXZ;#{c7J2pnwC&21bDj&h?48eL7S zqZ;qKhbc*|SF0F`3I`?}Xzk|2@N3eV{PJkD*QkzP;XGD-WwZJb$qB19|8HTDQ{>n) z{f5oYQLw}A;bBN<2Pe)Q-r*8kAqE_=kcc6(DPre1Ag*S7WGx!b`^hqPr_uD#rbQ{2 z({)Dmve0+lhB=L)Wc{_}xkc)`Z9yup29tXy;xAZJCu$y`0{mZoaSJzMmKGMLg%3uo zO&s)jduXUn&wj{Fap(jSgO!EFFOJc3);9mXS^AtO#Jo1tzd8=aw-i;e%m1G8zh?oA zKM*-c`nV#fO;D{e$-c_NQpjzi5?HOZe-`K=`MCulH=B2DhQ~SK+s^a=(#Y$C(f$w(yrTD8QULwLdJ^Z@REBkI2T?b9B zK2l$|l=3D^Nq5>uNn7g`d#F)G%!+2W>T}|eHCIvzMb%879bSGsJj&vEx;XlUR?k2t z2JBS|+r#vQby^s)+V7>Pr|-Qq>HhhEul98%)D)8;iaARZKjVf7q%v+dLr|QBVsT^L z)Cv&_GyYN)6w6xDd~rkldDhoF)2YwLnL>KtA%nN$tR;ydj9e6cSdw*W;Q5*~>IZCE z&3AL$>SPvd2qEVpC9G!a2M`1bC4>cE_$pdcciD7ix>(KB`2FV z3YQv1e_>U48aJ+_GlZ))eKEHoK3TAXS3=~h8KOXfjvV3}=JVs4$T0Ixfr#rnuPj5W z>==)@oGRlSI%fkOO3MJ^P~5nyc>QtjL_*bqkOIeNd|?@*F6t1$?aW-=bcR@VQNiDzGx{~^gV=hiVZHE0j^_^ahkvN(we8rlUj)A^B%~mq7`0hEyY_!~^YE@{2LVcbE z5cfuB0)76Zfc3}d3tw}efQy2Ver9ElGo3-TEJ9s-R&By z^(*Oh5m#cLQ`iq-N6*&|E$q9dEn3~V$oT^mKM<756SpSM4=wqka|tZ)RhtmvOk%6j zx9D`tyY&0XF+9Pp4}dZ_kaz6yeL<0M?H>PxR?2b6lu1Uc2~8*Lmt;H3jyuXJ)&(`e zJ|sP!ObrY+?aYwzH43Ix;iR#dxTqR!X~U+6thEN0K>3*lamIaie0h-+x@qlvhZk0( z?3Cwf2{l-r+~Kf2Us6=2UtivXr{AitsX2M8T9XYi3u7?D&SQ`YEw+gNo)I{DFd0g6 zr=B#13f`m>QC7u`s@3itk2dV3TO4pv-=D8@KY=v_dX|(cb51_mL&hmtWPIv< z{SX%oM)di*gR>*y2;%{du~*7*@1kb&#!x9d$tH122C* z>iqA~`HL6fRcFc!HK$^=9~h?Aq@;ey>Jh&fH_IR{;G?s>zIwUMrTIWnzeNPN6K;A< z`~E7~8=*8pX@SyFIQlH+nYLcjMs3FBsKdD@r@{7|==;JSzG?Dnc+2`%v8bf0aw$Bg z-;0q)eIKGrg!)CQjE55ad>yel>@AO97pF}7bJFo{Uz^csacU*ZSpM5nmzqpK#)I?d zM;DX`u2N*jW2PY3w7OlFyt2fWca4F=CnfMMFTF;2IV7_<<6tZLs+B=~&7_RGK3l>JI^Qp|)$WM+*PbLf6!e&#|a zp|7S8G~f?Psp@z;B91H{IOV0-`D9zP>S9|3))v7>D>$uJAC=rX9s03wo|m z9=aU-yiS!_>y(_ZW?W5hjCZ2fr9N}?U?zHmeM*gwRRRI!O$2f zH`rD7)^CjE^d(bDx@E)9^Dfez>izVBQ1xYHRh>;)=!3t1G4r#a$}nw5lvT#w5qip6 zPhPZ`u*zJ2i~a6&{>N_8ykjq_W8O=|^#ZK!5_jRKs3rv9%(K*fx(e6UvHkZObGg4$qhd&2eSy#06*Hp3_pu*RdM8toJ6*GpN&Z;2_V zuPQ?AQ{JObTK6b7s`12;ePkvaC@j%YDH%lEWFFZ7Z%rMeh2IWqD=bLFG&esJ{~E-b z+uv1ThW+MNf}KLo4sSHSjvq_qneJC-5)(gpn2v74riO3ogNqFKL?RgQDK`9F{c<>& z65sN12#3FomiWr*6Or#4nBUi+`&7}r#urn7o1*CXDrdSfDm~&b&t`K5S5WB zc2}>e_K~JLuAlj!)wDZx z$(B($Cu(<$VqUQ2$j88)+cBj`g#`uKqk^FEoS3Pxv#%v?rGZ}@%$U6y@X>;2P(v(~ zO&7|B2{P{v;>`Zmcw*%01!VVRHi&}RFAtIN$(e0qB;J~c44;ktK3OnZs-ClbQ4K}E z($ZgV{;mz@i%uKv@M8Z|kOS%RmP$X<*bzAsRX_Ku_|MF_5y%3JsoAFj+?;Ah0|k|I z#oxUu0ih>(Cnln6Yr;W3Uux;zk-uww2mp z(N#6`@ydns0NBwRW;7bm)wk zb5$VuSQ#s4mw)kTo|5Yd>eFGA@{-e@LnCb*#R^dJ;a1>^mGdOIMcjX9SWo@J`o z>kWg&qf-6ZXl|FjL?w$s*GG1nlJ`TE+lAoawT$_~L{#Yci*luqIIr@)*Gy_(nw90n zx)v&mUqz@8`>iM=k5b$(SKvIu)GFdv^*i(1p6F+62Ird6fh{>vj}yqJ2lZ-Y1tGta z`nA44<9YL%RSYRK0!4IBjgCCGiqfyNu^2D6F65E6Un0M#E*}dy>`{O`E0ZWIJY~Wr z^h*5SDl!91gnR7z6|U$n+?t^fZ|+{4$r1PcLO!@!t;gB9!u4GGf7RxT@|vHKm({`l zYlyAeQe`(3SLBtAskvo5MN>UD#e`-eQwI+6pv{fO#2t^jxkF3r!u-R(!MzSRrg1y_ z7xgQGr^3}PYLECdyd&;&!R@iJ|2(k)T_yDMILA-1M`MrX94mU(=^r#5OjD%>XnU<2 z1e9toc$FQ@%VHl~m3Qa$aLNYKl#O>?G&Ju%U_1;XyHP3Ptl!8n$ew(vz&&RGmpeAI zL0k7FdWhzSiftdhh2&&OrVTlGw|o2bL&*OwU!bVVx&v*ySD3Z6I9Hm{3{6V3}f z;Ql?Lc=5(1^}pYKQI7aurSjs<^ymMUEdp_2E znBmmP)Z7L`ATYJ2$yQuC*)c|r=Tcpu^KK$|wBagILM)%*T%#qg|Hy;GLCEvNozX~? z))C6i5N7i$fk5nuq%fUwhcUI%FqOTn75b{iYb)LpFcOnyD9#nd%!Om;Q$GIXwR;6<0=~d*LY$zAiYE z?g<$-X64Ss*Nr+$VMgxk4{V)@m8;7y6UB=0wIVHH&s}KqQS->Pyyq@k?}^!Kh#AmV zpUGCCjT+ARBe43MJr%R%&_;uXBp=1cOMx1J)2;NxQ9uU$*u8b>u7n$X^Smk(rs^5-j4#xEt6i_=yjQ`SjfZr^OyuN_QFb@p0wrZ4VaUvzy1bu)?}|3$p_n%2!Oe_o$ILLx4I2@Z+G6Qsi~jNVM$gMDDhkV>H=xl&R%-Vq%a2czBW}(oq+#(m?I=4$P+6>b}>upHf)i?N2>X6g}%+RV7~4RB*jxNI*jc%V1t=< zjFsIT-eFGUN^duv4MSFaUm{9Iu!gStzDtn=!{vKZ9GCk5MZMLk4EM^-^ZGGB-DNi$ zjFf)@w<|*(i^uKs5g?0r?+$BAnSZ+coF-tVRjUf*TfBVkU&QqTK#oXP7%<`eMYg&7 z0gr^^unxk0P&E-{zBJ)Y^0{7NUHa-VNUpr|8VSO8CEg4X<%t!=h=GHSMTR0apva?N zjJi?Xox!Z5!cSVEfBvlH=cejFxfQJsmsu0L8w2Ow2`{YxJ(meETv{GdEHlUYzRvXr zZxeTI_W{t%FV~~`faaERu1sD;5m-j&zxZH#V8EPnf%3NV-P&?6MiUAsV$|#X8S4^OLNL*H zgbk)zV{+dgYjWJ}SLPQKh#gig`d1-fF|)Bk;l4SNNz;d0kO;oii4ksMwoS3NT0|K! zM%PQTM~=HRv+60Om|$|E9&G?@gYq&4fnBXUUe=oNnw)7EpHUGXc0hL*80aPdDjEPT zo>4JZHMSBnjD)^i)77vmbj~&aH)uYd;v%9EOw!*86(v`5&V~>oO{@JyoV4z^n5o>{ z8N)(G0PzV5{e+?^aMi6~lguiR7LMw8QnxRzcT|#XFkb31T(x|6(`HZYXtZn{0OX6F z^{x;ONGQ`RVAMcKDu98)m7C%~fJ@ut^)8$H%9O=N6Vj}BFjM+AiU4^AhYo3=g_N{iAj$`+J>W@h)RwK<@7~n7w_q z-Iu`F*SRQ$m8uSN`nWj~2?Crc6fd5Ex!}$lks@`gK7HzTC&KFXn~Hjsb0l=k3VmnX z6>|krS0Ps?^dlz(Af7=fOEcd+i35miUOXpudIiJu#2sCjtB%&TYnV#E@?`oV8{74; z`~%yP9p#gp4s5RP zF|1iJZ)3&EZ}Nt2mX16Ej_#mnt2Ioc<8f`sym)|j8iZYx#GXEWZ&I9OWy{5ay&>8} zJTe5Xrys6Mvc9e)DNCR$p6o>fk?OfPFCcl)k$T=UvaQ%*en@tDrt3&D;MjWg9gGSe zK5RT1hH_E@jyz}Sds-VqRA3Ry-Y;-2dw~0#vBw9S7XRG^f5EI74GIa^)Zv($)|l@Q z4=_eCSN)c7Wo9h0F#O9B3b!cPEp#yGzWNgh6N=Ql`E%v@+1{Q-XDDjOAiuQqL#&i6TUq!)%}Tg9Lw$_%+4cz-%8Fp; zL|?!?!yX>JuL%HfI_0w#ErChF!}_TU4m>Wj1v;9coG6?)QF(4DVxk}MQ~(u$1MCGV z^p*i-h7EeF_=YKeMuwm8LDWSZ0Ykw_-FdP zx~DV13Gn;cwAX+{6Egrro~mieAZ%zv*!4XLn;CJ>Zx!109Q9Df7_sb)J2GC4bt!(+ z4riZmT!SOK8KVgK6$+_ey^6}3U4UBCf2g^VvZDS-csb`vYSDb5wB^aF`Q^{dE46C# z9tBQ62nR?FmrItKQ2tM<^CiB9`-NgUc$@5wH+=7hJ#ofEpLI&s1G4${Zr!Dz8(%R? zadEM^Z7$9GOv^;6gHR_8ZIN)udwjcWfuOs?#t1-3h&!lRC>gjDU$EO0t>sB^WhpQ% zK)UzRF^!CTCsJ{3}wBY{{*PiR&KD?)6KMQG8a9i?=# z{^S}727X(g%Zi+gC3D=LM0o0fDfp8!GnF7Ze6m=qEBP=&06amju(FD|`WR5j04ea7 zRWCduI>Btd_j&hT=|5Pg{qmlQg_#zy{8y~!)zbi20%gvjbMAa5aoKqz1-M-#PxnHu zC4{UVg(d~F7L5mhqvP7UY16mZx1HvmiqB+FW;FH|hILj8?(%8*3a@cOsxRKd!a6+3 zq9F*-+Ftiq|6(;Ku-zK`qtd~4bwlx-*zVQT)D*ZVI}bQkR8EL9B?=C$R$B9w!DQUG z8gYoMSY)5xtVv-pbR0Z%_&NZD9Y&9QE#l>k zu!+ZP^y-^=H8!L>EVHHt$CNI7`#7TK@ejaBbQ*)X^L(UF9{{^34NA34GOxjh;H%d)rOAo$mBSZi78dKXA-1!An$q_P9GZ9YGqWH@ zHoCYt2yV}LueeqPUjo3L(oD~SUDT!8VJmHk2F z1|lH&A-eMB;mDKp^3`AT*m7oI>_80I?qG?YFrV&PVL}Pn z&?UZ6?Xo}UHXhXI-Vr;j0x;ZH7wz{iZwJILV|PT9jjOFcKLXS>_yocq#{8RJyAzg> z-ggljP!Qm3nKpRcJ#V&tXPNT04nKJn!hM*h(G-3x`|p|Pn;ySuyR zF8GqAmzPH-!m{)c<&(Xj#0j2ltI+!MtiNT%AjzEK?5A+V?jaXrtmw(&*@!$fc79mw2H>W>%LnD~wiq$(bVIlSPtAJ< zLh~v{L4VZ9m~9p>AaREa0S^MaAqFy4zbp{eT!5XkM+ zefZ0-(cna-tZt0}k^+(4DnEOwUaXNsLHLUo(#5wUPq@w&qZdjP(x7qHauF4Y}n5pMu}3DR=+*jYWJJ^(a* zM=j2*l;9!Qe__+5={u64gcJ8PvKL_Sbt)y+h0GNw*DV$ZJ9A zD-PG%Fj!~80F4B8^Q9FCO`i+`7(tjMT@u%qTXcE|yFXQK-8d#C83t9X70aw|!9q9uw zTsA;ZQ+|)u4+xc-kGHrM4Z9RI_hRhZM@zLUn*g7(4v?>Cz?TUx&@R3oN<-Iw2NUiZ6M$Ff`&k8zd)hiURE5LTu?#n;u;)9_^hK4!oM7)|}+Q<*IKoHoaipAuB@M1u!4Km<-Y1)xHps<7ViBf;mZ6W_{C% z023#_q<9O!dsYF?aSI@V!24SLEH3nqU0wGFmYxtGIkACmbiC$Vva_RAulpIsHuuZl z-{0Y^i740n2gn@M$yTW&Oje`m8@wkg49%1@h5&_jzrFh;LyMJPT)YV?H zGO;raR3Dy?XKA5x+7b)*^4Y*g0UGEzG~Rtkv-`xCbkFp0txwT(DL=;X6?`A>Tjvwg zoDlkkKwyMcXr*?ExoI8*UQlvsam_sS9BE12DY91<6IrY%qr-MN$WTzO^iN?FTt++J z-=c=&ks?HWahgc-zQjG+l|yX0pxtvDe@F7)vjCepNfzr9(tMIW&2GHXoGCd9l-9s` zPyZH{+Cb3QYyhwI{dQIV=hDpS>FKghzx_48U(%#HZyD3qtqS;Cd{hv#4Pl^r9G+4O zIE$-V2U%`wkZB~VK0CpTLL98gG^7%b$AuYJq6LO+Gs<+h`~iKT@&R4DgM+?F5o~XM zZdyOIi2y+w>|4qv*UAmL=gojd6UbD3uH$MGlZ!&Y z3`yP3F`%5FJv7w_nvyynG}c+N!7k+aC$!kJE?=8AD5VHwJ81K!uX>9eu=aS`0};6;5k9z|Xh= zyG}k=N`E3U>6v-Ymu$IwA@lEarOl@V6FHKQWZPlU`GK&ou!BpaKj0F<=+tixc9sO}uY4S4#R#O;_r6_n2adhK%*Kj(2r41@Pk>+@;y zNsv?J5UaQcE?ZI6^nnYuFPRqT^UJPeN#gz+rni;JCUr1wnU!snv^R_3Idrv>pxNMVouI!KV(srNI{ zZl@<7V#LT<7p(+Ym+a&}B}g=yCrcSZw|uI4s_$*%QqXqj{Y9zgrVk2(R;lHF7g%NS zdKIbZq98^3Zj>>P`Gb$ODoZSEY?;+tez+Q|z!ukzJ3KqHM5FPVr;JYPxnJdk;hqT} zw%l1jIqzFTV63edC+mtEAa|vRgG2r98pYWnQiYWRz1wgTQ&T{t)B}0uex&GWZ9Rm3 ztIhhk+#qgrr2l3RZf(n217Y2x{~pKYhCI8;xa=(1{^+SY)dz-_x;ondaaZ$veTzE2 zSaE%g`=b!hilJ{>eRjEUSY(Z4c)RWFPx_z^Zei6@*59Fh-Q$G!zSQ#}Bzqw4 zAABF+(|)pA{R|*b`F6peJ(tX3>jb{%1PNm^GB(x*9jH)Zt~l>lRSj88tsY_rJy5lX zZozt_LffH*r7ekoS{MSQ4oiwy^xuzceW1FHC4MbNWnyCTq8}C@OF6g>>)rwSmjRu2 zM_ai9wJcXU>pd3@^_IBEd4Hrx4gC!Wc zzd6INzxkT#b}AZvy%Sg_5^w z9s5G)=iOr5D#>M`n~eqFo0L#ubdp&U5<~0#r_!cVWL*cmXlB}VOoN{4QR=7g@R5b+ zJ+a^x!{W`qg(U_%h=R_b#DWF{5N$)3O_*8B`fjv9(@cVM3vb!;$^CNK)zV?v*wAfd z8`^URMsB1muyExb+$HdfmBj54*NQ$~w?Bpo&_sV>5Qb_^1H*+CyG!=U&@TBizM`&=7t{GLi#!ku`i7miIIWI0QL|xyNyumnf1bTVVc{>H#nqLG`Evs^kyIHP0W->vgObZrN02twr zuLT}2f&JD*Tl|)R(Yg*?OLjf1C$imVROi}Mlj4F@Wr?lpdU}TTJU-&hNYqm#h~PHv z`djIk_Hf8hyg+HBEVi<0v(s>^Q=kR~R05DAP;UV%2LX|*cS^owtAM8TS%YSb{r+tE zqDRj`6KKtV%FnK$xxPFB3j1c%?Un%$`b<}zLA+>qh|bCcM2Qi2gH`f6~~1kL*Aq0GbcbCjb6uUIf683{93PkhLZiwRJB46DPG+M36xH{ z;XMA|Eb9FLS&2iH)W$iaVZ{wH1l70fz#HnX*F&PmFT)|dZ^v}9DYK8ckDe+Wm)B$` zcj2BS7NX2JMPGQodhv#gO!CF67m;s++t+^E1fq?tP&t%)A#`yyQs!-^AYil8Q0wqx zl*_^+@MryaP3DgRJI*%8%hk$9!9;~943J7`+b4|g8<{G(j;^>YJ%<>-Iy8PS{ZO%R z1A4gq^VOyafL=+=X-CG(n>y>nqw@(g^nL!20q3qzbwyh(zOVrR@ZIl5dtZKbc6Yy) z6Ld0dHbAzUFmx8wWxzrnqsT^sg0Mk#wE6h_(6IOC8@q#fY|+T@eJu>TcVtB9nX}j? z1*Wgyy^$~}j*mNDo`B}WuT4sv9!C3GzE{Uds+yoQ+St3$JOe+Rs5F$mz1$0-3>Nw# zeA-QDxq9NH0#?*~Vm>i|ipS~Kf<jXH2;x_lWZY9TKV`rWmeetro}uw zcpVy-u*8IA6LDqAd39c6X9pCfNPVcFWdiWlZqWFJ^+mKAVwpJ|_k0=v&4nSZsYUK4 z?GAr{qbQ!MP*OUdQ}O-`Ge=Qrw=3sxINz5in<13Snn#PtT&|#(?+w z7*50WfW^Hg>TS8}l|}p7N6>|XlP%3Wn{s z+_VC{sPS@n5vZ(Tb*uE|u-)?NeJQt;C|{J+AE&*enz{->(G6ph@MiqW@j)uNSJg1J z3!L5bwD8%%>lqsE$!;_x%C%eVjJ|>APY0)U=@zWNQco?iya`UTXZWZ1 z7STAG*Mn*Bm<&+giG=C3J7fN0azx*2eVn&@&m7a}&-b(!2NZj|M9kav^ z$&pg;Rd4xSD*U=eeruZ~6coTegJmNAgf}^{Lh8ED*gr5Zl!w&j4flMu$`}*c>Ol%Z z0x|X%qJpt8mFwj`qoyel?q9jZ$~!HXk1&Y;lH4RGY!vm0upprd1bpW#>CaN|_bq8#G&qITJa z#`?q>gh^t9cYoTO{fabOQx_K(zxZnuH{x^|WTvNo<}>i+nK8Pi5WU^!6=aALdU(<( zZP_OL6AbE)#Lz)ha9sAq#f1|jOPPnf3WK*k^zk83p7#E}nOe1pu$UNfety25!=Z$n z&1>E?+<*>I9i6u%5!XwFoCeyO$hB30*CU*R}%$oxAebzB zwP`%Bdb7psO;X761~`+mU}h`Lms@_7Ntp3_^6tSy;P@FD;5%HnOcnc3bv+r|bm4iXgEGU;xK!Z07G`PbRHH+SUq8<;0A)!*7*gnHRMZEZG zx5T2dR2I&M3OD$xdiMNr3A8JPgoStgF6}c$IqO-TqhnP9nQ3@cLC-xb zMmU%0SwM$3(|Ayhv4XAd9PK|re5Ca{ z836?c1}vt8jm;<184pol3(8d}a}=pQ!utBz9_j&TjZ5K_fd*7x^_{cx z)m@u4u=R@=s7WGvm@h+T@J0|sqki$ab3Waz zc#wcn1(=JP3e&r}?n+8>azD_e`FM2WJm{`NZ%dNa(fatnaM*mLfrv)*qhn3)aIs#Z zK`Uzc*Fc6adwE)neF}5O+Q&aKG7$wvE{A1c(G+dog1QYV*#Jiw4$9rW-(>JyJE&dR ziu-1vy>S}(o}9eg{m!A~>Cx5yCFyW5HqGF3f7*eI<+_H4sjqy&p`pxZP5xN8xO;qo z4RW3!jM@A_z_kL$*k@@^_qlV4JS=%OW5T%Qw%bw$e4I&&h02k!vBLVov{V{n2Q86V%XGiG?&=KJzz zTAK9UbTNnB^((^Gxn2-?oWZUR-qE~xR$a{!mi!!1F*f!`hqi4cE<9}MOi1puuEV~9 zV(f0Ml;Ok7UIK?LMQO{eNhQ+|fllSz3qOG$d%WL-gnqCDUa@!P>7L!Vg>6w2=o#qS zY{n|v+!5EFbz(|e0qOeDdR>V3>E5;k{54n(DIpk3372UG4PWEo^|{&!=%UZ-@l_Jl1v%JCifCMF8&e5)$HVZypbkxTJ`=?-HB21 z0%Tq%pmy@|ipco*(m*VLNgI*Y(|f;W^((j^`H^+aQr-2Tx^=+OZkuCDW@O< z(4l~--2qOQ+S9$^hK3hjav#HhBn_Swcr!UUc?_Bu z!a5Kg{!x)bFs5oUVtB0WJ0z#3W?G@`L4(gQ*%@)$&4J#s`2r_XB!X3b1C?2^1N&gT zB*D=doIltCjRAHdAcGP-=i>ix7XqG_KzyjU`;DkL;Ov+@A>+y9fsPg4Sm=E$P24(e8Y5O0!@z_az;kRjSW=RP3Z|{P_I!k(Bw80 z1S|CgDJUvn=!T4;)ikug0(VU33mn<*u&7N!Ovo`tC1U@XFJHQzpZh3WYq3-ZG8zGo zYiOj_m-9vWZ{md;Lhgg%lrkC`QqIEl&b*oM-Rm>&hMH`*mU?dAdRU^qtiTWkU9xtI zN+zdcEgfB*NIj&|o{$VmyZ#*^J3D63DnCD5O3IoRJwF1kqpgep^)+{XKruc({tFn- z!`)TS_E}_fw4j6pszzzN3}q*X&t~Y)S&@}!sHtTXW5F{3k-uUBX~h&2m7(Hh9kPy8 z*zhSWHC4*Qgu1%A+U{_1t|-a)S~V21wYfR=rWpMRsaxYS2`HDgIX5ec%8o$=6=>@^-QIgfnyCij9p0!eN4{ z5>uN+0vsUB&CTgeqe;h`e~K*mI-grT!_`%7EC}3O|!oZ^x#n@;Bo9o@vE zza}R=AMjqh4SGE_<$`kM@oIMY3{o0)mjn=pB9z^KRNg@LB@+c8e z@x0!P#l*))CnkznSd?(Q7KjnLz1$6%*~@L+g%{?PR!TKYaKUQj51APQZZbx+M$f;ygXxhKh8R?1a0qNX(Rc#}JO3 zpI39a*cXnEZWwJ;wiFW+yS@(@p`xXI&&C!{XxW6J8V611TwH|wuV!04`I=oW-nzEI zz{3L#2Qp@bBt?-w^cNTyUV;u{8!6e_w?07UynXv_uqbr7*%jij2N}54zj{|RJyOoB z@owOgLoPAM+Wc65O(n4r;s=Dm4Nggv`C<4xt|Admu%WiRqpGIk@GRN&zZH*IkTbTI zH(2 zx^sb)fOLs;zhmy_zK`Sm_WT3S`mlBHO{_KNbzNhO^Zbo5X;b4UPuq8k5DO&DerEROa@)pQE}%#J^+Mq4_WtFH5@`d^IxYt(u1q_e-w+AG05!j@9n z?l2+tk3wg2)L}LHSx0!wmSXd4pZtyi$@Xni2Ujw9wQ;t26H1OQRTUq8 z)hj(|YvS2w|Np(tU+T3c!wUM}S4#>}*H~HcVMMVl&HeLsKD1Q(4|vT*$~1HQYTP{| zBa1CLxK>Qc7@jek_3PRbJu0QQi`QcZtV~Qyw%fkUDqF*6?Zg%IPP4IMQ_Rt9Zw`Dd z@rj43@WSto?PLBKE3ZeY%5PP4->=rh=Mso~;U5bhNi`4*)P2s!+4^IyN={pQUDfBT zF~l~bo_(1xee7nx6hTf;uUu)d5HZ)mj!05!^-@F7%v#%Kx!~Ysc6NV4%3i~VSbV9y zf>}8UIgy{NQ(x`rN3@bBGPt=^2ssJOzO4_vnuuOo0QwVqm+oCDyPyGiZtmQ{2?YQE z0pd$V!!>qq-l7;u4=pb1+<*5JzwP@!i!LM+cp;L66?`e(Q2llKYxLR`spz{8rT?Ud zbI_N$FsWwU(5@iG3ZIFEEt$>?3V0YuH_1|5fhr~O)Fx)1${@_Fxbgs7dAzq4MvI#{ z1eft)(^%hAzawq%`EN5S{#e^1p`nS)%{}y$2u`C-d-nR9)Q^;uoZzF#NdMc}dGqESow#_?W9FALTUl@fzQm+kcNa~Q zTGl0$yZCxK7=KsNy*`%s-_AG`nWgj@?ReOP)PJNpE{RyVUzLirAeEhkv>%I>^w_CQ z%H?1&Yg1RL)7XXgR^)C?3h(-_xdZw4SYp(Ka6F?AgyO|DoZYQQ_8Kv;#Wksk@<5H3Dnzn2nlE^M)2@!54y5$m!v z9FpnGFouuRTwhLmBtyhfscT~(U&icRGI2qb#>F|+;Tn}I{|LYRbY1-Esqd-3WJyWM z`b1Tvtq1=-w+$3y$MERrT(Kz5l?M;yPf(qiS34dV-s$ zi|;J%XQnU_;Sdp1G}Ug^b(Px;c_R8YxD;?N#_R>4gfC^k_;Y5f8tcC*!>!4~Zn8#v#prGy_AZ`iMmiRfG%1CmH0QBBDj$O2pwwzyhUE(Z0dr{*+Wgzm zr7BMod2ZleU1Q*ObG9D6KH3pZi@WIeJiDEIC9{;>tZcj+J1lCD6z}EkjzIQP ze)F-^>IzdQduwmbx7g<9=3o4uV&Pl~dHH}&^E!MG4Ms+NOqH352TjK((~ogUv4W&% z18H#%&;02#=!05$^aHqS*u&FbGb#;_4nwWb?U!YegiCqo=a1o%!t-ZpOx{BOxt5H> zW}drjoSO7S-}{}%%;qduWRlmK$7vO#^=q@3@zNgSB=&?t)k~=b7Yzfl?oX}6=^Q?g zd4W)uaFV5dJ>AA2<9Ixf^Vuq;WrQ~QP|)a$sVgDUTG}l)H!yd1$Zf_#oow6d+UB{@ zUn#Y2v-AfZDG%@!sEeno`;7IERUDd;;IH;Ql_q*2M1_&5sTR%XE5QjvVHHqN2y)#a zC{>Zu+;jelql=YH8M!i9A-ZAdK=W)g&)cqEh#|>3H*>2eN_9sBKWXGWg?=F~4ds|c zC z&Z4+vWXN4?b?q1qqZ%%gA6OzDoA#rI2ayyReXnhbMXrGw5Cef|BX0Wbd^%;6C#JFyZ-c*cMw&+b?z`?@I z&PBnk`10swFr9J0nVZDft|s{9PSxq;fa5?YXjrKz1Hoah^~uNBSm1`@{+`!v?EoP# zzwN1@sAzI}xQ*XU@tPU6xYjsP9xv-y*R(lRON#Hp(Hwqd_{lGFiAD;(#O;9d zgfJ9%*RHMfr(H%+9$*s&2_6OnD(EtvgzxYCvMWN0dUy{C3JT8#kCu3?72bIjsB^1l zOUJ~d6qy4V4kbMUieB7j{|_(-nVL(gHS~{{0zVzBi~Xu}=z32l;|VuMuhCx|`I6Av z&uNKp^!P`}?!RM3WrnCgDB-6j<1lOwCAj_9yB_W^l1?9P&)*h~1mE81p}IL#g6#UG zFFmHcuwkM|clRG~A~T*uKKB%(Hk~uY82e_z;8=Hb^-AqEBH1Y9+MXXYLWXTS>L8q? z`hf1ttJHhX_R;lNq6L&Gc)y=>sVA@eFAETGe)P?d!)1n(g=Mw^ZEUu?JfIN$v(9Tr z=*4dgv$l5@li!?b=KD=#2&j#0b6ZLltPebzxry+Wz3og)=ny++WYgByn8rb4;&0Us zQ-21!1I(rjW<20EyHg}nG}XjodXq$Vz>wbovBb1$Vr3u$`@=moCJv5XSHA4rPJox% z!I88OS%>C|=Ha|7%+xD*d9AQ2!ur2!OE&{1)WekK}4k zf}Y0ywDR*^rC1^yEQHH~iW9rIG~EBnV7I8%v|m2wTx4+Hk{N)m=9ZQ`|H`9MN3tYf zW#_m5$FI1JXhRfyA0PjwYheyd4HMWk@P*OwF>h$`)>+1yL4%NnIv2cvwDi*GeVdK( zdLKc~#(zAJG?@QskFrkKlt!2^&;4;Lg-4Il!SkmQdJ*W=*OyUT{Qk4F@M7~~Goi#E zSV!+Z-D8 z`6E7GN~sA&fMi5&F0;U^BvJQL-y`SNNE)?AiKIVl{k$U>A8G06IB!&-dEn%L!rK~M zcn9k8y5m(g`>C23z=T%^r1a+`FU|?KIQ>#e2YJZY^`ziXaIZ}QWA-ijX+Tn;QO#hr zk`<%AX*2rvw|}`g>ysrPKB$V|%Bt(tTk>DK#m>mUAR~DA!~0j;?vpHDpM8e*(g!E| z8@N|Pki`kgUSOe7ZS?1lZ<&$7H`@nn=kL(-3kVzntPx*EV0Cr%*oHgirKv85j&_>J zP<3yHQr%rY;;n`g4)}k&Le*jV?T#K}xL0?(1AQ*MI1i>hI8pgTNKL2#=Rx{x)DQvW=7K|1^%k8%D7BvF?Qn4`I-XVp%Sr5~%}k207}XYiNpvO;K#8n>SytBp6hOe?agL za%V6UGT#cHG3Xxrs&@s3^Z+x;)G|TuJT?E@$`0`Af+OB6H0|HKeT6^MlnjS!uh7I zk=xB$k1eJO2Ug^=0$t2*c=(k5Gcla`IFuh_Vz~EmbEqU-KY_pgy8dK;u))s<6&|h+ z9PyJ%dlNWUEX}tNY7@j)2=39KL3?=#Blnq2!ni4WNNyJ8K3gn$c$`;M?b`WueX{z) z*eAWY`T35%z9mrUtw8GdgWoxDnfm+f@af-Rv8Se|2Uoi+&Fvh)L5hDzWZ3@QFxTy# zxh1C`vitrl?paCq6RWuFM%j<|w%VEr#g)OG2xfe&VfX_qZVv#DrQ6=!SsBY=K&69y zfXMMkA4!~Y0~VHex47T2>olClpd!jl7mGEVAKS%p_@w}{u7B`$_!T_o&2k6A0e_wI62Gohq1s81Lbn+D0AbGYX_y8+finFNTR8q7>YOdNz>vRLSh zLGW-(mH4Pwi$88p%EdxUfuK_>JDpz;JNo(IEd_6@avoB!eFG{MAwE7qD-FiapFjVe z9n6$HpAta=bhM^x$Z4%g;!z{pc`qEipg?e>Gb2PU2D0JN&y6lIv$D?($kTw@@+csi zjRb+ni|UT?E>7Dd(5f#5ms>hn-OXZ&SC1v z>7UIg%nRZ7jJW(?(wXr%>vu^w>kcsmRchbZ71p-y{wp%BPXK?18!k7(EHTQe;kwt_ z>D7Dl11UxBcd~co88nL-`fj%%VZxu9<#=ZbxHB2?`|iI3u)iIz<{>&YE-tP)VR!3_E6MHKU7$j)po?gb z66<xL#G4J|k{WO072>%1fp4I;P_*)85IkiZ#*)_MPE+8|y^ziE!(A@@AHG z_x&7M5a1c2^v}CLoxlV|Fj;$41lXzlwT_Pcsdf3EfiU1*OV(#(WJH_=T7??AC)!3W zU@-Op6#8sc1jD!py)I$9-BsLBozesZkCPCU*v>9*jk94~$YhO*9hf#Mpp$Na1C+`s zl+vfv=zsPJmRZ`fc0;jS#-4QkjgfHpV1oMx=%kc;Y}6i=PkiyT5G5NRB$VCVq;Mz)g**$ z-pX$NVvn5q+Yi525Ln+U-FwZxeP42 zxq`h101>^Px9*=EH? zIl)!J=U~HPsW%A&s(yMD{Q$Q&%7B@Oo`dckvSI*pB-7T;$tj!#wcUKVUknbnAVx)S znt$hPzmHD%(RjYci6#9EvZ6q06+$dbX4DVygGm{~2UJk`@}pvv4a)h_nG^W^5(S={ zBHfRk^8|122v+(pQoLBPh;!cmR^|nfFeg7BD>XCosyZp~E0hohYHsL`kn_0WFv!0{ zEZkPjHEDHH%uDsPrjYbtpp4QWZ57)a*8B{i8{?1TOfL`w)j;h~?8I>g7K&o*c5Sx7 z8)>fGn9wg@cH40TQ|IkFib?;84G?>nQG?d?ZZO9!_cZ`~3lIvlh zjOQ0!9(lH_pWy`ztn$ucoVShg@lovj=Er($ncD`JXp%9w!TI`~vwRia5&L!h#M(584vsXN>+r z%O1no>b`%QG?;*%fOjJUjs%a08xMm=sfs1!+&IDCpI#j65=l#OO@xIfC_o%u885pv z<9|{x9ivVH2_d*7H*V8}XWDxEJu|{{Y9<&(B~vvjlsitd-D~kT^-MiVg3TiFugYnO}UN>+422LN-)r%{%va2+h9xbVg1RuzSs#kOK5YW`i z%j?V6uTjwn!Hmmp`L_k1!!6^r>?ZH*?0g*=S>(tOvia~f^g7z13tEg8T4!IR=F-)8 zY^iE!Xk=vS6D(}&MYNV2?wFWcV<3W>6xa7q$A+~1l4{5S6W@I}JoMb(m}qPG;fLh% zu%s9j)BR3d;U_$|bmbGZlqV57S#NX(gy6*{FkGO30xGYZrJB?o`W%C%NH3_iwFkJD z?nMrQHGLOnXKQpNthnXtw=x=vjB4GFUFc4mdy^wz`alY%8KLB3XN)3r7c`x{99nru zgjJ>=H$eI|H8n7hGjL)V>9&nLHQ8dy<({`^^%Gd~Qg$*HKiaAr4) zWT-k2ml+5CcxcD;@QZa!YR^ zHqNM>CB_;1zo{Z?Y$>(q0u2yh4stZJ2FpCu>Wn$DA6+N9bqjRi*Cxb@QQg;-=;>iS zx0fnIwt7}=PEjJw*ZTM^`{kkq>s))vBrQT33HhU~P>vL#DouO*y1iY2SG*`q zxBvAVoZ~&+JAyY~*Vx%VXRfOlKQG}n(1lqHkI^edbo`r?h65bF5;W0_ng#})AY=pd z?@%rXMKSzcP1(q`H~ZiXYA`VujM_2^)X>V$+3c-J0WK?AD4p1mFiaH`?oAFq*p5#Usk3?oyGn?T}$7y&96Es7qt*ZuA7?av7wx#Rr`L|9Pb?41o2 ztytWcBWznI3Vri|3G%S;Tbzr~9`SoNj3(3XWFLV80not!2Mcm*rxGWS`DeD!OXr~v zAo9fTV2HXdBbR(YtH*V)+R$59EklN3H8Rq^Sf~L9ZSjIaA^PRaT2^T~^2Xgae!W7k zjhv{q^OUBeQ{7hvU%^UAx1j`c?}d=PTx?>qCkN;kNEaxQ@{!e-y#A^bhlz|vKn_c1 z7ng0&?59S`9VU9&m3V>)dll}bNxg>6hIvU6;HLped85ByW+%~^HM#$1aTdL!xj}C> z?<%)!guf)geao@nmaIDPPas0iqPSvcswN^lu`+UPHa_I$mfN616CLa0kw|$N`+6~G z9>)Gh796$fJV+C6J$aoPDqZqW9~pXvJ)MmxVsBDUNRpwH`3PpMrUZsd;fFMZ z<&EhO?{fVpoZ~go3>MTN^9a-LYyF(HG)Pcc+lz+Da*mw(KOm*OT5d@z+IoMh% z)c1%F&7j3O7D^GJXJAkXn}pQ{m3UcouuKP4@eHYP}%5VZY;$w$9)VVG54vbg=^x2UXSZ`8!| z+K4u4zmL)(E}-~%)*EX^a-9Q82o%`woue1k>D?Nnpe?hHWW6%*@#!)hSs#I*2g2{- z-$&M5*bhnJyZ#XLm_aG_@eW5bMNVI+vk+A~3Y;&}b=Lk{vUM3@X-Lq^g9BG~b zB@%>D$Gd%@ZMY_OAJiL)YpcVs0afFjaK47gNaU7Z+4?ENQK zhY26tlqJR*t9KEkDLVWF@)iXa4%&}w-cT;WxLHIzT#|!e$UY*hQ9aSF{S$ZdHQE9{ zpXlglz4L8n$gCb~R}2)pR86Sp>&I84$+vb+&a7?c6L9rodxBs}P?!=62Wu;2F%opi z(lXAT%TZz9_J6G%?oX)L|0;je-K`9smSe*a({{mJPxbp&C+Gx5@bl%-f<$2#{MDrl zfscKeZuiN^cc-+J%J$IB<|q-EngIt8LHCY=?>-lqmXCD~hPNcoh z&H9GDc3`C+t7wz-1ByzP<8vR<+F<}OL9=OLLus9Ii;W4e2?#xkhyn7EZ-m9?^ zGwftaLa12t^VykH6WZB|>VClP0BLH1RtjBsZySVXbv?aoD^pJ~oM5SkUpCDR%dLrn zq`p(-rCMbyQ?Rt2N`}Wa^Xz9wu6osbm-jw*6ywsnW=H!2x|g6y`EHqI&kv)8t6bN# zE&v}xvXFmM9y3AN*9tQf6>4UWDCHiEYr1yLW6#bI*5Q~~Dw2gcT8x^bYPjC=lhewzJ-5!{H-Sp4iZM50fL#lvJJUJs- zb`%H03Uu;AL$yc;-5hk_#b1b-30Osu;sT%rh+!wl)!|ukkz#Akd(8@T%l~YIN&rK= z_;x@8EFZ}$pKEKHv)yPGX4C^HYrw9MP?|SC(AxmAGAy6`6 zHy~rnjO1u9O&2VDf&R8P`g{0xP;PGeE51@}OBb|}$4+FxX$WYRJl6ey$Iy?JT3ln( zE%;N)5;suq;~}MMTPH#DysEaN!?nn#qV|`e6m;}GMS7JGG+hr=)Qw*i88tNt3te$T zS&mOtY1~9?A6xu=-!$pemISJ-BuIS-p_ACt3Yi9?LoX%yP(y?Ct9a^cLY0HWr3!0d ze1_?z9rd-G5kVo9gJe)F%f5L+Gy+!!(OCfd#R0YnoH7Jf8e12z?5*p{7tP%pCc)!( ze2A%4n+FygxP6Eo2jc1-s|ew6;%0e>eSHx1kgOX1YrSx#2?!?;*&Qm`7A0TUV<6kbbv(S^#U_=}}gm3jb zm}x^QdnnCtUp+uXUzxD$+b}JFNIh=68?6`;?8qyiCH|Hlu%Nq!fm%UALH zH4{Rlfu-70c!hJxF;&0iJ8DxO9&eZ!gOoc)`QQnj8aKuh6N81~cEHY2>x-vr>~$A9 z?o%Oq27ja&90yQd_e!o|U@q3!4D%QdUFhiRzhP&C^76q0%>R7+8UNb|(GJ0wwdvV= z_(lZlhv^>zyfx>@LV^0`BhqG^CKEQus_vM*z5Oq8+f$5PulW1qi8wW3P~Z&TUkE$h zJU7DUZ<)0w_gFGe>43ifv!#Xj@5*%(wSr`(_^9|MIJw|5PPH8^rSt*O3e@`-D3;VD zPd@UhN`?%niW2l6jOMPg@)|YJr9FEOF&aEYIFGYFJL5s&IO%pU<-0$=1XfqbzkkjR zdPi5{;Q<1m*(c;i#U{5x4{0^@?!d%^YwaB!O|vf^8ya?pbxAPR1_*!$mz8YHzWVOR zOC4Ox@*0~WjPR#n{8~vh1W$SEyd`SfHy#xw*ArPp6;l>E?oo~k0gAmcl6MRCvo8oH zq@<+pKYeO1zn#SmQ?crmtc8{JSys6zabYwnib(!cXwDLTudVaUsjYn|$1q0Yn zBWty*_z{4?!JR`|NxS!t zGIF73M35I89j_H3nV`e3@>ysqUWDHKwxzwaSNu?Fo%8H-{%mP2@z^pREqv0I77|AO zk@t1&RzqTt@MYYq!Ko|P@fsylyDW5KSvUzKtG3HPE6V_Ay$R-vJp#+?q2h2@e1?45 zo$J_${Gq6*2p4tPX4S$1CY8X`^=iWFOuUA_*%gmF9M(xd(z!BG#p&ea6v%O{m0MQ* zi&}~VBh%gJUji{;&Vh#$os!ZABCf2VAr&GHAqyY$M!0MM4S|fs2q@9UL{%9G;NV^e z!|n9?+k{QwH)>s8Y?+zwB}UD|%e&&PAxJ7~Q%T(!trngVLW^U0(3p@wnV$ZiRjxju z{&K`SE0p1fw3ad{gZ$@E+U34gm>h(`}uZ^aho|+ z8^ZM{wq0Na90#fqMCHye)W`C@H2fYl^&AR>sKi92uy=R5NqZMXf(@u7$yJ86w7=H; zQn9pT@;lzmvHNz{q-t&4*uN-+THO0Re6^DkH(1e0;|lGcO6-Jku1J4h+a4@awrgeQN^)%6gvzCLY%4NxEuACZ^~{h7|rh^ka4PCh)(%*)1BGyYXPvxY8`S0_c3CfLzoY`z4xu!?2MXcc$bQR zK{(cziiXcgmY=9XT3vC5#^X`Mb$8tQs0 zeyX5P(wOyNVppscCnpnaw%wLl%@ISer==ftJLco*nA7jDnz^XMdh?O@N=jr;uCqn# z+_SbZDYH#I7dwr;ijM5)*ebbXAft5Q%eOwZ@3Q5fsed7E(fo_|zQ2-xH|{Tz&Vmu4 z_Vlkdy(az3GddwR!kP41r?YWFi!}JDOa9CayRA|o zvhrc-vHt!t^YOW=pYJO&Vx}Z>#SD(+yUZwjsEmH8DR~~StKu8{Gk)T1HZWNJQTc4T zbbRtCe^HvO@#g)QQp=I{C#Tb=IzDqHV><VsXEBmH?Y)5D3gIEi5{mOJ9vWR{@0(DZ*8y`ND z3&gN}T=D1U+$U3VTfn0_Nc>YSvahP!{a5ya+fEPa3^3Pcqj9#tahjMo{dgw3pOAJp zW-&uphl?CkH`p#mM@OV&WOFZsipT9GkNlOIF6MV!GH%Mds1e*fKKZdVUTJZgIug5| zp}BFlPCngaxCK>rT5a8NVuPPH4Tm%k7~RX$YS0w}As7U|VmP(QYN@aNpI6tIXeN25 z3;eQ@6}(8|j%z?ThDb}%aM>_xkOi>&7nDwN)%fs*)<+dm;De<83NSyNLPJGnQq1qq zt1fcaN`y)dD|{ZgIM=$Z=`@^diZUvS)a=ncfB+nVO~`cb-V$~|B^5pyStEhYapS9& zH$gG2`tr?}auYJsfHH+m3m=V$@yyZov8=W>Z>i~*Vd=BhoUxk^f*xngD@DJg#Rcga zqrEggZE%+Y$z|^`5A%?0k8&=cMv=qwJ6M|eZdap3aa(=D%8l7gBGdP zVHQGEt&7EJISF2lSkf58m<@p}b^>7pjp!MBzs&r0TB&{-MpC@l7ed`9EM286k=Mg6 zs|-Qe-ClZ5-KZv+$&A8D$H0WM3^ND=m7Y3ATK=Mb*4j|58IA3G2>>s63})z;mhD&S zyjIf_A{4_~1kxnk;*TmjhzdQJ$b0W5PKBc9Wz|Xj9Eq<42hk>ELJL;OxP3(LJvJVL zZ~in62|sMs)rQAIZ52i=B=au{JZ7ekVPXNTo08HHpO*GnQ;C+gY-I=5SySxAKX&%> zmwayBdNO(U6HGgrQ*LyP6{#)@h-`2}A-gfTd}PzFU@=R84>f z21pNt9ngkAiv*m4`Tl(q0gEG(=ib_f+LG-K8a9ZnrA5yZfW?AouMgyuWy2v1!4e${tMq5u* z-iIG=Q+a#sJpBwRcihIOf|GM|a|&u|S)d`(e&0;09S1@lkP?7AM+=RVXo+aq*usa0 zwT9m3F&9}gra&K{YLdtRIQ$RQhY230LOKSS6}aEhAjg6)`Y@WYw5&`~LBZP9VYK=8 z;rIFNzW`JqNB{)racEtBC<2%>K(B;%k)LlT0_G60n29ei$8@jb0{({yNVK~j5g zYG;9mQ&YPKs2(CrI?hPsva+rCbh3b?3~oE36M>!twGab*1rTuXhU7u4gY=uLei_8o z&(a1Kf!5w{n3CE%bD#rP&VngsAbY4?ls&y!{MP)3o*4j_gB2?OFgOy*=p z1rM!=jXvE-xt=xDku2`R%*r|kCBnJ%OW>&PAR3J-MG^)`A4nAfOB$4)2O1hAaGw!T z?Z=M?q4?CwhQ4Bti&7oOORpsM$U(|6z!w2JW0@UV^(sNc+;s@Qz`lK`Th* z^C?trF65L$0tNR$+l99XN0?1#65JKIX~-`E8jRO<`LR@FGypIq4if@EmjDn#2TK{w zmcAuB(!EL{3Fxhc6bxC#a3!FekfN0WpBu@dw_9!jbXwNRAuJpZo&;#PeUK|qfLjJv zP=-_%E=dz4hQRcG$;<>0m6?;%YGiffBb6A^{sxrz`c!QkeA@I}R&8x<==}l{jD>{- ziVSr_LuK&FpmPd7yS+eSd?7X7E5wRxZ*Qk%;JpkNKwqB%glZGzVRBPbQw(Od`({V~ z77)D>uy#hpQjGz?_32{gdwfHVmXbv>_|fGykzDHQSDH{dYCjM<0Ik9ZPtL4!h+HoZoZ%Wo7F&E>2TR{X$X zE4M?xgFZn7I+@szXq{P4A*^ji>U}WK{(v$Ga$@Dcj0dbKSRXbq+46UgLKCiTM^DdJ z-7RCR%ii@CIezRA+aL(gmHaaSxD=kEvamiW!XT9nLIKg6wVImz@9`w%_mQ~00Mzv z2n97W`H$ySAL8QVjg6_nQwP%YqX5fWQ6DNuDTbfk8q}03!^WrO;CKgK?hfVKW768@ zcVJ^ci)dL}ii(Ctw0`4PW;{pIsobvWaS<_!FonWguW%SVfHaj{mLB@{fe|8_G<1ueCvzcj@fp`cAd*U) zoN6Ya@6lcWLc&F3y^jNG&R{~we^0qD$mGSgH%^Fp{VXBx;xFY1!GW6$X`C=(EOks+y5E z_n)rUJ)f%i3Lnj&0PzKL>g$av9daHsXY1s6RlI^kL36G5_jEne)s?&=FP6zFl6ZK4Y%6*qe>t3n32+HHEiu)f^^ILw$T=6+!L8!W4^XFPKqBs zd@#`0S7#)NvPoE7T|JgmVI;uE_eA%PS$8f7&}&TP+^3?Z9tGj)$DZ;Dj-u)3} z+}5)6#63OTD?qA5e-2Xh@u}@xD1AQDP`~E+O8y9r47Y{On{KYI%rTET_*52lyr#Z6 z$y6l0d!K33S-FvP$)`-v)7h#kr`o%6as=JfB(0*NV&mkbE)y17{%;3mX=&+d?y0x! z=hP8fSy^eGkSQQ2SYNzJo%p-Ze|LAcz0%=kxU6=bHh-HbUV=2VIuuEVI!qX|`5eT1 zO+G_cIygLko(uO=ns-?|h44Js<)KW)s_XJc#yp)57UOtWmvgo4#5g6u_+)JMGPWH_ zIG}p+d;iK?8FpMKJPu(nhYy^P zm{>0<`23~-zD$($!`!YlYMEqsoAs)5+Rd1j(Yj zp`j4k%>4klt?}GWNASr#!u`DJYWWy_JQqT~zbKyO}lu5N5 z*GD=dxHc0L6URXGQhSNR?`+vfM{1rsLNBDiRLV4%FwkbvC4Cp_7T~2#bGnn#i@x4| z8-u^nU}oYjNuM12OlB$#fy5EmHONj&laZN@?Tx#B=ij@=7rItz$zm%G^KMmb)dz5e z`S}HHyllpEZcho*8admF(%c-Sx!F2-v-yQkKFp+zd2~1H)tT`nD^ms{98Ut@w@FJ` zh8$bRlP?=?^-a@~;$OAUE-<1YTvw)qA<|*ze)CTEV=Zi)oi$)EkvDw$7<(lVd3vlb z6(jsfk^$#YPs_Glgab?t+F)3M{K!ikwqA2NW>SqDt--xD^v&=VEoM?aZtgT#z-Jx) zL>L%>3F3;iI^S3KYWx!}6pwt3VY$HLEb4iVQPUR_Gv*VOi;Ih2YFu#suCM1;Ih&D( zqFU0GhqOu_55s>D?1usFu%TgF=YC|J4urOZqQyd3^6$0F%8QFrBO(Z_YM0M-a?YRq zEib8F>G`Mx4;$$Ayg0{ihA$z-d*yog*QBSf&+fUU>dwv%{3w{YrdhH2Sw2>Q(Of^) zCFOGO8Mn&NM5}&|R^25JJG(sHX8lPH?~0_^7h|G}W)KQ>HSwx8JEpv?CMxMwh~3=W zY@Ric!7R$0ly2~~{0^<_S(y{8zJ*ch}U_!<*OwQrw<;{4^`12ta28P;iPMZDwLet55AKhCVhW#M; zX|o7h%?Lk1eZHSuVW7hE1yXs~-bM|+`gV49c%i|zisjh`1*_n&AA|Gjw$yt^HbcE- z-l^!4ybz^MzHla7Rik=u?ZM3;z4=t}m`5y83PP7pok#38ODbzJIU7$kVlBK>#D2^c zyAb_LHpBRP@)v8Ez+;F78Q~8vXdT z78bH*W@a_c3#grZ*=9;O?tqlc&(kgPzHCFGnhhhR`gQ^S^ZamS^K5!@Qj0ESrlQ`# z`h@}yu&uwd_rfe2s!{EFf z#0mmhj0(oD*woG6qJo*Y7Z?~v~`|bzD zz5RXL-JjVtyi+1Bi*KP}>;RsNFf=!p?p*y@PF`L+Z`eFk+3^*Z_0@oR z(bk<_zqpY#HwQ;YpR`;d$4A_@F9f!}2 z_Uhk=UZJL{gM$e26IYP}ur63WbO<$d7UpGS$gH^jy7nwDFApV`A>zH84Kl+5Kbht>7qP0u@lim_UW-Y{f#usGq6b=#U2hs-8P$mzj~s{y>@ zQ2e{mxGf2$#+TFm-uRUkOqs-3kzxd5zqPG?6oqzKwh^(eiLlroZ4t9Mo0cg!qF_I3 ztg~)2Ivw5H^rJTZvPJN&)u-;j|AM&BZhjNDtF*3pG^J(l?>Q#M@9pz{{67B0g3F5d z?U%&jCFhI=zG=H|Nw3xAVPG7d5$z{ys_}~UJX<;Xmp&DFYr}hcmvU=6#i=pHa^c+) zF-D+OSTQyRyd-wxwpwv*4gb!R3;L(d zhhvwKf2W~ezXX2-5)$tt|N1=)2UKzljNdXZ|KGR&-!9!TVTfhKqsK4B1d&UoD69Ig J@WHd-{|~Zw7pMRL literal 0 HcmV?d00001 diff --git a/images_ml_frameworks/image3.png b/images_ml_frameworks/image3.png new file mode 100644 index 0000000000000000000000000000000000000000..91043427998c9dda87adce2e8a3bece134f88b7d GIT binary patch literal 44837 zcmeFYXE>Z+_dlveiyBd)2hm#+iQc1^=$!~6AsBTqdI_S7UK4~Mh&rRU2tt(Tj56A& zqZ^EJ?&SMCzvua#|C{sXT-SMVytrrFbKU!1d+oJ8d+oJ8EBd*%3MuhDVk|5yQZ-d2 zJuEC-6c!e?5#cT14tlva0ShY_OHE1Mz{hGQ7vIOCKeeIz+l*{;2q_oC?2=1OXQtgF ziVVAq%+A4o|Gn*R=&KPJXvmQIkSTt1rCwRS)dV%K;jF(SHoR4 z&ep9x2FwYC-ktrUITOFhv>skV7;TOeuba-i4j~aMt)ytJWs^lQZZvuUQv&x#w+gpt zsFD_*)mlG)&h6{#%lk!6j*@TL60_UPY|RtY9_ALcfk<83Y zV$Y_Y_k`S=9OwlhwpHoIz^f)Z2ZwU)r)SQFvQDnO+<(U^0zJhgiq)L9E(jm@d{|qv z`ZJYJy~T<0`9!Bu)S?91cJUFLXl}Xj7*nNs+DbcJ=h)0c!(FIjU9OBqRAfQ+-p|h( z>8rE#PU1#x#${dMs`V?{?B{6I8Wcw)xC+D+hEGEs3(dM#; z`dHL7x_A^Xaj_G#vdWbA6GiWkFu6!*D0?L@+7;hsdq9DTj*fOeKeV6Fjn_q+Z2Zr_ zj}h^{&t$BN!^iD@k5%5Ng-^5Y%Ysok z_gcpav8p-YnLLV7S}$KFb8v8|nwe#(r%z=*G~BE;gB2bnbmb;KU%KNkl%vb!vbniQ zjVcBvPi4WY14BaEREbJAi#}VX!eC;e#GsWQaTydSiM0lw?We?Gg}1D;!}+pRlO@6t z5JN3Dk%(rq-8^78ilsXxI{8MMhZ6elx~|gx4hO7MG_&#V-%?#n`Wk=!u@r8_2fkbU z`5pi~pq)!hiR$Am1knN`hu)5Wh!S8k88NN%-iTm_V`gd{0J|Zgpg?6T7fcm@1aWaR zjCL{2Jfuebd;D^qA;ofTKCQz^baKH!31`sMYaUJbuZplTI%2TiEAAgJZvmeV&Vr)o zx#8?|fDSt&x?1n)nXsNWb7>5X({ebiw(4bkGj^w)OwQTVQD1na(zmF3%B)y`6CueH zSFq#Bqbb1};V{|%YsV?ik1Bv0u4XbqpBG?wNdC(sufF-SuuCm`@Yj9=yKY36bm+h4 zT~0X^R>ta5zN!h)s(H7-XNnuNec7TxEVz>dKX+9-{j#WW3X@eE%0U(_jTkg$=C4Q- zlp1iuqlI`CA6OC(y-nK@Ay-9ZxgkyvlWuaZUK5qU`R%)mj zP5Ea&Bry|vdBxnn zg(N)1Zs(yy5jDp|U{qCL3Ep%N=&ELu%3<16a@Y4=&HRM>)Tr-@F2&S2v#jDC+meA8 z^EGHCh{X%azOa`Df95J#@FIzNO_$IBpBlx1hs-nu`PZ)wI&DbV-W2&y-bH<2<_{dw zhUpwW@lNYnmEQ3z$N7)`r8{1v2y%53ktvH|-{f%(&HRYKbi0Ih z4fh8UIDxa$$?X;8)Ryh6bn^QaVFuiQSVh9;uvNj=JTr7GenuhAN>gnEr3krx$JpeY&26`|zz&tcT|po;geP<)>rx&{=a<#Wsqk(Rh+eR0{N{>h|2uhT2#{IuZkk>?E>VzqL-gQ z2`>N`!J{Qczt97DL+^~YQ}Dtu_xpmziSDDw?xK)jV2KWb+`})rlOG+iXv-Z$QXyoH z5&#DZ;R#431cWL1Cg;9hM|LnLNrQ@SUl;V7hLGwZp2Mcz9yVwu4{G1=Wm*Or${=LA z59S(!k_t2kq^DT#l0-o805s!(J%?OS(| zpo!3`3r$br6AEg7QJ_dzj;+EB#HdlC*)0Nh6vH7#!;7k@WU}N(<=#FBnNyd3OS8Fy z6{F-wCpUmGc6gBLRX>NR2>(GSKu%$vD35T=7h2BpcJ?NRC&{6O<*ToTNO^v?@y@K| z5$3K01~zlTS-@>j+VZ56f&mrIpy3xP9qk;A9w^{Jbz-!7cvtJY!f=R&G0Z{7=NYlS zj`(%UJkc==DIeV9!PU&cF9E%jr2FS{O_E(#H{9?-K(D_ZxE|s4Ixry?x(|AZs?=?M z>o#Vlb2i49J)=hXGBkfdzOjKjBIz$P*pK(tq97?B|9dSnKuYfEXt&Ta9=(*br6CQW zpV~r}FdbjHL-28W!FF^{@0x=+a*pF%vRT9CC)vM5Tp>im8=W~J(|3}CUT3a2^~@>V zwVI|ih3R-IbMYK3s+sGPMWjzZYaV1-zR7zx+^>LW)nA_tTm^1Vw&S=Zmak2{9UU9KwV_oEr>oYHxw%+dxkOLQZ*5gw z?WX^mA8b9F3CC1C^LTKh=APcp9>D{VLb9deMo5a_Hy3*%0tXyQUQ|LJskCsxC1@pQ z3a}#}TD{0CfiRHJLMuL`QT91p%_Tp~;NE|xXw?c6K*(&m^ygx}Q2(ep&51N117Izo zh7jCZ=8uA_6bYw#*cB2*NFF>lLIgS)o`;T+v?5tkZ~b)=q1>Y?SwY%`KxInwalZ7) z5}Ypq=DeJb$u5!M6+;*AAugQ~e7Du=!@64ibRQsOP~_Ue2fZ&Tg)uffQRbWCjAG>y%pRWZ9*=+6B8Y&T6z~H8wH21!-dMisytSGl zD}e8i4jEG9B_IYbtCV&HEC(b*D-u>KoY)#k@`^zTz+V1c@l6-gTxUf`yije9_pm-p zh5j!JT1fzVixIGUS*57KY$ahUM-z_F2%_xI8(TcXzN%|-2<=O5Em-;P5>~tgiPU0+SEh|F(pIBMt)tuly9YYF#W{b zTVpHLlZSh?AUT!vUyJje3Ux4itu@qoMy})g=DX}o@ORlGS{+}v`Ner(^e8n7d|@^& zVlzq9;q={i*PV))7!BB11cqw3y66y)El`OHWg0)6NAI3+j2h9j@++yAI`j|lJxu*^ z*kvwYztmUizP9Bxe6HF9h%A+IAO0b7p0j(lm7mn?GzH0IT`|?t++tB)Ux#|AI_bQ? z8F86!$otp&T0MWV!-rds6D_jE}k)QXE-FntS z<~iye%o{Uvf?a*j_vf?SQEQPg!<$iy+qGP!-Yji9ZgJ-Bm0BxXAt7AiE&d#s;v^13m=FSQk^M zsrh8qXH>ngUe#Pz&C?2=Qpmc{p}c$>uSzdgC8W7(ZEITI#d0BJURgsIIlBAMSnzh~ zBY~R{E9A-b#L*2a=X02`(9Z|(YS#IC|78^#5^z3&Idn1;glwRH3~6_sB~bHb&gum7 zP5@FA0R2?+V#aS+T&+j`pUF2v7B~~6%+2E&W!c@3RQ9{?m>g3Aw&SlH-wzaR#uT|_ zFI=|&!B8LsYW37aSkZVyU>I~H_Vy@6U*j7Es=Yr%L%Y>uF$@PJs&0V~oPzX_@ zCSnS5wv!)SdG+7(OgDxlbkT@{CEMHA2_u~IY}M9)CBEeTsrdn(YRQy;m8${+S6(Hm z$5>Z;m*1wBww(21;1`f2O+1g8>YL^{T8n_1t0!RFa&o0FHm2eO(7F`W5jw`e>9D1d zEV@em4iu#Npzf^v@~YJA;oF_tnNXh!^@P2rZBaQju-@Fujy#Y8Ikv-CTPQLSoWiGgAL0nQq8?wg3aq(g}kXZ z%L8uMbS0oDA^lU+?W@4&07XLwVn@P#Nne3tkP1PV_`RxxvBC@0+}^9~lbP(PFHPtp z_pYI6Eu40tY7Fmk4ezo zJe+z|(}Lu<3!Hookd{16X&kisx&;Ut%_RB70i;EpkVn^GB| z!q+L|57!h2ZG=-D2L+Wia9Fo2duZ!X{tInIA|&;uJsK*}FR}#SV_XTI;m2g6yz!o= z@YcD{8<(cgR+_P5P|#D+84nhoU53UJ`hj)G)`zSX#AJ5Y)sL1pF21vUf$D@ekN6a? zNU^IVaUZ4)8txusK3^9j&|BSGt$*uvwPrH=lZT2qjDD;>$4ZWvwa$c8Sj+qv)=f-p|5s@~P)lld!u>zC=`< zGD`^-J{U|28CuMR#3fE_L-~aLrd&5T`hOvBDrXT)tdag)CpGMoUlI3w$7a4_xLXZAJ1Da1jTQc9-Ew@6p0d5_Y+emEmht8v?Ao@w2S%8`M{Klo+Be`0=|IKklU6VHB^AVRW{$;HmD z|H!o=@rfnDz$e?1i7wB6TA=|%VTt|B)~`dedGDX#-4#^;ZvI9GNbLy7Y{m%w>^F6( z@8aw47F#k_C(nt{J6h$^QymAhmBX1krS1DewJxo#OgY*Py}bMMF%GmhflgMC*$r%z z8=bDuc|Z!}a<`2}A7)KIIp5vQebEyTHqzVeQJI^B&kBIsBf(KFqBtziPOB zPu$R|azgo(QH(d&FYY#&OivbJZTBN$sotdbVO_a!K@1sly0IfPm94N? zt=UD<{@^|7@6iLxK@Zn4FT#L^6}0v7n>m#qI+onDLZ5wSKfM1ZxsX1TOAOxmRJpqaV^a@*N!h?aQrx`ut%ewk zv0}|&OBC619gkxF&L$Rhand}s+PK4jU!Xw?5fX?5oCkTyN9QH$>Yq<<8ZXo zSZ|~Ggw1Ea)+MF`Ae>M=25cLwZZoPCEt}7g-LU%s*l-3SiVR9(Lj~>udYAmP@yc!X zaubFp?zjAvgQG;9S#lQ0l>9lp3FS)OMe#7*V=r?bNXm*&owAu`v7Fl_M3cD6t`xV< zy%U%^K6ggA2&$IM(%da;g?j~Dq>Q>S-R2v}?p;JV@x*|7Th0qSI7#Ma-a#~7^oKnd zc|w5W`3A6~sqO=kahUuIRT}jXWP&H#U-9an6h+z|&%)z(y~8Ct7IsvHrn=})t^Bfz zJm22GX>rrK+NDJ4*pR^Px^FU^$CbJsHA~~BVzt@9wzz6ch6ZTBuk$;*RwJx5AI$_a$ZLS_4Nq;;%xMi zjt)Y(=3V^!SYt)<5GOgS?ellPIx7vY;D4mII%oaNxcQ~=9KY?OU|P6#piR>1>N+OW zVD2*X{=}{Jxn3ghDqcbnlZUp$+PRp?(i5 z9qsQ!W$&?W7Kbiq)!C!?vSSLG95 zPE29((j3|@YiRB$in^~gcS?$_Kvea~5E+uJ*UzWRB6(K`S4Gp1ecPMqr*@ur)pzOZ zA1}>G(kjvaQh^#z`sd-q=-&n1Sk~XxW+p8Oqn&*6+OvCHGqc{}a7HEhCP%l3IK4W^>B^JB$al zIOo0Rk+<0sXqo$;>CA5vmF+-|9wakC{ zkW;$4^ij&9HH~c^6UWo$+7fSG^-RCro?JBR(65#3P`hx53M|;~L8kYF8_vfPrHMB{ zM=9gEQ`a)_?L*zxW7R9t8jmA`;Mmc(o~oJbgJ(>I{SwYnV{i7L#E+F}9@V&yzs?lM zB6f}YhPXW45FLs;dU;HRRF)3)K3hOtsZ3mDEWrc2n0ia6UeC+wZMmTmr)*=tApFnf zTtzQbyDV8qKtrtrI(>HrHorrh8zZnXh zJXI1_`VwGaSnaWAUF% z-G9dSM0K{F!TlFyOQvbhrcaOb)>CPD56b8AZU;K^Lqc@=p#_A<{VD>npaZXwACMOM z;1y(Q`O_ked|k(J*Rjb~l+$RDwF<9Y+8wNVC2V~6@fgC2x;f#(j%|cR5NG0 zqtmM`53*`$+XZ6Z>t8g78%Lz8GgH|EgT8$8k57wIG=2l&hA*5FPfS)~_moMh@_SGo z`(eJ{=jd_QQAHrSUCwE){P}&f>uO2&relEyRkAFoiec_H=w+24%!ck&DR`Tv>zt$nOdL%GqAIT4>k~-kSb4d{>DZ-s#;|cg$iGcQcBMDr^l@M$}S)Wz?7;$9646R@&sgmO!jI zj=bDIc;uMEm}_*=6dk{axD;C)HAl`jA`|_JNtqsXd+c3YYa!y1!ts)*Gi($#8mGOd zmXu!NRir3t?duWac)qJ=O|FKjQPpspHjqSE^DTtZcsojCW^Oa{Upe(tbsH7 z$?QQByH;ihXc4^e<$}3p8w&+Kc_&boE2yYae0APQ=zodtN)TYA?M8sRoT=}yqIfC! zRA`Y$vLtY}*AU-FiyCF~uS~I3b)6d?cIduWws)hw&sn|(jN2)<#+1ZJ{Qir4;;xe~ z<2?h$hWj8vwrA2A=~XlLYjeE`67r0ga9@YXF_NP#Aqjn#_S<&-Wm=ApDY3%?H|{TK zH~E-0?7)Q{A*)hSw;Lp|e8ao<7{%`@y1~6tBABT5h-P5>WL6_A zP1-OLPcLos04rbFT-p%JW7VzF3b&W9I_SYdHa4{yj2+e+aGwurk-hnpDM_BaWZq0} z&+?=-LOXP9$oJr>vIB-ZpYasK@%lLoQXPRYMIKF+P7$7*Lk?3PIIjAP9+IkOyyT$I zwL8lxu-&Ytz1PP+g#0$V6s?}JY#RE;UTDgZeYj;()64&pqI%4hSTDSp zRZl-4Y#qu<8GPcG&m%mRFV0%3Oq#X4ciN;8fg`?T_1iKAsnQ6Gl^&=$&rO#Htv+JJ zG;#>1WlF=yMGxHwxe@W&!A3H=N-olyD0cs`)_|%bDQ4u^;k6={{c%wvZw&S#lS+KP z;fdraqjy`$`p@s1Bom%3HnznY?ir;r6sxdy`|_tgTOI4W%@aq{)F^ZB9o_4=MM^$? ztLh((Q!DX0`RVlx(0;@UCpvCLJ171FzJ@W6O_gm10F^)Ut|Hg`6>7dN)+icP%|t0PYSI};#+7%!1e7-F%G(rYp5)8;3XHKTq2g+{<}Nj?VEUW zbeW1HBH+<`6A#$L4QD0g@k42%$fB_8=r{4G={SSGsR&;gg^>+;WtE@T2(?Y9GY7P_ z5kI7ewH+r#M|!lZN8QEa7Q5E8DBy#mnIt~x1!(O{yhX$hMFG~WT4>a2fQ8F;0A#z? z#JOOm4XDrhHO0?y`bMf-$5Iy)&qSng!QcsPZ$Pd>sNJ zL|T1VcY=}Y^d_rm*M;Pc!Of#++s_nkZhBjn!sA0c*&n^9otW~*K1zx z8hMfG!wMD2*>`JcE4{m`6$Xk*Bm)378DbYYoF`EAMSh*1rbQSTjR8-n+}wMOKmj~ z@!Eo(+#TRa7x-q>#vNV=9{p2~9Ebh;$h^^mV0VA=+jKI|EUVB<(Oa!1ZLi7?HQ>!x zGN}*X2aHTIF2#rRo-pZcsrC~xvOYY=k(%_2z|O@D?ruo%>8Wop%_;*FWsn)@-AZ$e z2*)hnb&~FtZ`mHE#gh=;W*blPvQ4v8)zr+Nzz-%G6CeAOeb2le`9iS^oi5a`7b zKASd;7p^SF0mVAXZb#YRZBlSjteWf)ZMG2(9h|=djZUc~CJae`kMm)laci8-oMVBBjLFMiJ*5D!B*h_iM_Ayw@Jy9hB1_?T= ziZLUv$(g-7+zB9+686don3gN8<{i295bY{bAQNu|

B?CYa`xUlA!isA(Fnt6pPZ zV#_U-!Sp>Jr8nIdjq*4n`7RkBIXzW4YjSmErkn-@4#LWN4E3m8?bfbGo2ACtV)@lW zy!u)sV5#(D$R{LhyWaea^P-MGQzS(09a)X{mgruk z#CPU^PoM1UW_-s5?W#*@hSez{8@jGj{lVKdR`KP zi{2G-F?ZsukG{-f4Bgv>kg0uW2rx9;KF~*w6XGeCN5}P4?=Yra@u+g~i5gxpPgf>5c$|7aJ^$&ZNOHMnYB@X%1#Cvcjc7 zcXLbYG3=siB!IfKC4h^TEW&z`Df33Tx88w%Q2qq+40vG~IwT>XOH8H$s&T!l%%Nby z*HJD30h;)}^9R|#!y%?f7}zsh@}Z+JtI)LEC*!meT^gW#rWCs*`c#9M)gl{S2UT8D^4iK3Q{z$<7mN11z(0uPS4>z`hZN5rlPqz}( zi(^EmHsuTqO-Y$Ehvz{J#5VD`c34X-r#Ui%Cq1gG!Ib*rIbKCN%U5zG_&8*5dn3jb ziw>)-+7^%`9=rcuW?+z>wUN$*Y`18YABgZyR`?5pRvU73;4Ju)|A!nHEcotxGB>#)yi#Z(4vuRxoFR!8-Eca-MQ2 zOPG(k*$a;0R7|9SU5qIzFmgt5!?=yPUPIb{mN*bADOTWi$)H)84BtIMBk5&kZP>~+ zn;&#frYMhRV?%*SqF~B-q2i(4{^jE;>k>6%QTy95@Mlyg`^*)x&g-+7xbbl^OLAE` z6^knx=?QeV95VCW5rUrVQ~XXN?Sf0G#MM&a^y!4omo(t*%%WP8F0-W)SX$#QR5CXn z`X<)e8pcOM2av-ajiE;ktC@RT@B_(p=p6DkI52oL2x4blYqK5Cdv4+QUm1s&f5KNW zoJ1R1+6;VRpUB=rP*|1KhMlpCP~-YUhWYIgS%O)(bbGR*BfR%EWbF-#Kq zFhX)Yk^M{LYe6kox|f?o6r~-N+~AAw`L`tzl6Y9*lIt5QEfn|YGL_y-8SO@7D1au* zj#xLvK{b31O~0DTdARpG8)J;hdANGIu3wdQG3A2^mpbD=n(Pr8 zYT||n=R@r(vhwuW?R7+#<2;5lnib-9+AYd`l&l;^fu*pmN2W90bVaey94XwB+7fw& zUv+o4`UOp8nD)c7maO~2UY~ds*AoH)f9(>2qd4tuHhfP_HW}FCGuFe}N7X{{o1eSn z^F;yCE|6wd{&A4god!}BQIVcZkW)5&WmGBIE1d&UQKru$=({zV`-aZmbt&qVC-G8a zm_t#Z%zaMgT>X8auRpA;5M-c0&4ZC#7a!&293StO;mk2r%dU2JG=VLD*lzbmjApxg zdJ1ce66It4O(r5hQ|zrx1vH?W2DI+Lkpp5``@oy9SAT5oF}-^f2I4u$aBxk3_P^Xu zL4PeKTyQWAId!YfqxjDEJHzetRxm}A6@Q8W3YZw#+=Sd)_I)|I>bEB} zqKl03#7Ye^k=m(|FDkha@dmy{B4q}?Twg{?4}B~ts7!R}JdQy586`1EzF))R9$VVc zvDQOt+0%rL>3CYie(3NUr~ved7y&pUC}V)E6I-^&AA~o5CRNx5aPe4a+G0}?cTMOa zeM(&41WvJ|oZb>e4?PTKWkD(>ChRW#v(* zEB0cCleKZ}967PVt3k|_k@yd_gd=CDQ4|gts4(AN3YW7=#ak2$+cZqK%5JlJ_Ss0p z2IR6TAW%J4hRl3XeD);8E;StmmzOMNM(_I&tR-(=q_eL9yFGd6=HXhyT>O^FOpCF|wT75db*1Lczcp$EO^v~!a{c7`~oQ<$)` zhD6DFown?NaEN33*qqT>cRGYjnym0~5#f?F-JSP!93feoG2?JIygfQZ*9ME8LG?bv zI{vcR2BBPboH_anD{ z``LJ#?W*SW`bMKf5gxWIjz{pO?x8$_^5MPoS0tO3oX0>_#M4SV1wuGBQvVx z81(H(A1#Y3Ys&wzNsH)Dh7jamG>~0&MTvzIP z94B>J;HqA?S$)b>5%Eb_gDOg(Figj?!SqiuXJPB-3CGxQcI8`1?_%0TwxSKIMV1!| z6oXRzT&=5`{?sji>Hs=)sL%r0_|l4@z3_-2lV^i}%xhyq{zr74*{7c)^eNUYIIV)+ zM^wZ2CQR*GA4J)C`XlH|q2`bG@R1J(_H%gJih~HKBpUWD9#@57CY+kioHL_rK@}r4 zU5Qg`1Va(%?psw%=r`VN2#bwGMU^ODOMa>tvQAJ{$^B007xn(lEssYs9nuZ&{#Qn_ z`eS6^)uc2plSR5ca=qNjD^`^~mpW#bRe+{&_e<~V^R!%o`3aX3w+zMRH<-YaU=>iX zt_y!WerH16xpOm_+V37dm!&XDTtff1&SBdxbQ!)DWTyu3?UJG0c4Dc18UNPMGtmv~ zLb9p{ivK$b<^(N9={O2WVPDZHG6)NxR=2(gL!?Hhhqm!05g-t!T z{8PR2`BPfq|IbeTf0lRt#^{o4qd?{E{}AeL9R%R;{`u|2Kb1!!B0-==qeT9Po(I$| z**Zi2!@>pF#}p_xH@81U972F6h#cla30dFN?~6SlRqOJVpwwsj9`rBW_BB6$`g_aNzrg46X<0kNFHw zqF-N6?+<+)AlYjmDoKG{pe!4Qk43)Zg!iW869IVAMNE=pVtTrexOii7YU(awDt~`p zUkZaDeYk<`ds*b!VA$YSet=y6jzjXk97%8)b7NR0lS@QAXjq6l#iG}>*26sqGL&}v zLZDQaE^yob6oFe>nm&G|4{lGopGm{>-VY`m#?7%bUyx{ zHVgocM@YtPWLUpT;t=~s-^aos6eKJqS~`j!ZPG{=EL~zL^v-Z4NMhu}>BvEnQDTLJ z1{37oo^{=wdZy~7j1ln1Vlt(xS9HkM{!rzJ{V}2$#oY@g{w|5SyC@0+14>idIHMZo z|DF>i?d6b-;;8z$(=aDO489jfbN7+TJmu07$sKI`HYrL2xw@r4od!5EM_3`U5;YW; zPNU2uI9RXVV+kyXsE0s`7n8FHWsc5vx(Sv@_KD7J`4^Q*&b@dQ$PjK*;^GQPw`zqNbOD zlh`shj%++!QBjZ@0wpXvWtL`+krf*b)&__#qG%XdEt`F75taMla{tF&>?&?RWn z-)&f$2lJrl!5xD_P|1IF9SOYqE6V@esiS-HN7w(;rZoC; znj8M+zq*cMZ2n&#ph2A$wLR&Y@^rCOgOG?qD+_9Ljk(6K^!)-avWMeEffq4Ift!Fu zeHFrJ=Z$s%hIZRi?EwU#ApZ+Wx2}O(7eaqp7IG=5y#9ca;O$G`a%}yFiY&>-vGoT{ zE$GgG%ire#_);DJ1G6yUec&<@^ba~FegWuGsvw&s9iSWFhKlM>{{@?Fc|K9 z?>C%(sMmKC3dsI;NQB4&@4VLd(>lNvt^^Q@fY;#$0hg=vKV*4r!f1e+3CQ;U3;6#) zgF;;O+oH*BR|Mz3OP$Dlo4o^uMdXVDN-FcF2riBpxZ|=K_Uop`D;7p_Bb}SI_UDk>_tlzGRj#ZA=zIzs`MAtA}ov0rk&LRyGl} zN!RI*s+#JD@0P&H@_;;l`npt|a zlYk{T|7yqK@Mn3|&*|!wqL1JD#=Q3v*$$Hu{FM$akA@*zR~x_U%ORQ-v-OHwc|OZ( z(4ln=>x<;^O8rP|f2MA3j?)pVkCm$aI)GJn#cA5BaM4bb>DQeb+M`>^4Sh z+&2&J>X>IA(NYyz-&`N|u_yEm2VHykZtp_vx5nQjTR-1;LvoB(Kn__-d13Wlom2(s zJ71p#-IOe|9*N1y9Ayso;gVlkwrXSiffkCJA5ZoVCM)LpdSXiRBtXsATfLo%k*SOG zki~#J3DQdj=j#m=Bjhs!x_+dh86jbj^Te(&cun{H|3eSmod5$Dl@IaOK?CkoC( zeB;T9fpp)mVu88U3bPaViudjea`LlvGo_w!u4%1H7_wD_hrfvjmw+5R2i=|)EebrW zqtEj&;^Kyv--->sSt5`HvDC0&rhDOwECI*5cQ3Pdf5~CS@~`Wtu1>|ZgdMN*Z_e{k zB0xrdB|(<$WNQNKdphHg3Q}Rh3B<^#zU}OgZQ{46n?50ra%*O0pKS-EwP?qD=aNBO zUtjG7$=cNozC!1<@(QCRPg!n!o=w%f*B1dP@Z;kIf#_mtEK9Uwz$~ zZsaQha(rWv*b)mP*pW>Q&U3y)YP`3I(e5va(JhxarZAtjHYa%Q@@)GWQ zw1#fKY2|et|KipsA~ABkOxyk?aM=~p^1x}blJpIWHp~@X*YClv=M;3cZC!pr|H#32 zOKqp4Lsr7h9{GiznkfU<9LBzO7l+u-v#i085_Zx%C*$Qc^tbS=&Bl0cfnb zdV1QyeQ|TOsl(RhLV}h&G5$lNcAje}kN5N38P0C+(65SUwj>kY)snsWpsy!&Qgrzo zmVQSc=3w^bCtti5;GdYQT`;eEWP8!sSLwG`BQv&sjvxx* z=;e-2>}*~pcM!*!Vwunv)sJ0fe|x-beWxXDKS4>1-Uu6YFB-j&{*t7PIYs!{YOkF^9q1 zr6gP(N%X-lSPEUtp)N-`kKk_iJQe&8Kja$}qP&?(#Wvcjdn)H4tBW^Ri>aCS0pNEz14eS-M zvFdKJuF?&AVc`iWU$6xy1?Ibp7*?gNe{0nOUDZlQb9(ynPrEVr&l z5fQXj&qEpXRfM~4Eec0SKAdU0n5Zh%hR#Dy{o)Bs3=WU-Z(97f^z(^LU%20dgl+jR zUamhiY={Y9G^z2-bPV&YTSvdk$9k%Q>+maYMcf#+qDjGiHP?QXO?At!Bg{bRcySu% znv`X%?WzQV(hUTV18zSW^vvCX52G#dfAE`(jB>hlxj)8nssR7_m_9 zs+KU(KjxsYMZ<4KHQFdF3x7+l+qBogrXxuj-Y8u|8WM}CnZqgaV2_wG*9uNO^16`RSSYV5i7S2+Gs)&WS{*tZGC%agLg|a? zKttJAC-edvnt?3#FdI1w^DQrlyHz!VT8aYJrMf4t6*C@2Lo$#7I2&rB8LF8l_3X+e z#;w`8+^E9-Ao+8fv&J2ipUch{-9l-vd!CFRe}xm>k@n({5Z+H9TPJMF2Tkegqna=- z!_^N=f~j_uW9c^2Y(7b*)V~><99vkOTfA;?1n>Ctr-+#c7uud}TMIGoc7r1V;z7ML zFN!Em>@5jD!07??cOQGnBCME$PP%DOFks)gq&0k(bfvm^niMW6!FG)fx-p)ZYyO$| zKBTKP2gtHJ&$o@?hiXZ(D_8p+Zy}i@D@GG%LnV&0_GZ9xaTl{VG=Kt@f*@K{-o$ zl5Hw&PIG$VK~CYxzA`>6yXf9qg%_uyNroYsk)XLwfb={x?^os<|HMS6QwqIg!#;oX z)ujH1KrhlI@~GW>`z+cwFvW09Ap~$6av2qqPWnPyI?mp(N$u*f5buP%o_xikBb(zP z*8|wi_1=w9p>xL6`xH|~T5TA3cT&rlr?6963Jv!tBV)&GoI{7JfEFn~x-%6PS%i$= zF`eP*k8D@J+3e$;35|NE$*+y(??(Mov@#z~j4j}iXn$f-KJv`u={Oj?UK0u`z5a1& zhVi)*z=#*Rc{7k!l}ICCR`wOk6HCrux;OcK*fqo48PQ))X#n;9JU(NXc7w&ypo+%#Jn&KiPnfAjN!Os~YYr#T72%vzD5zJCm6_ zH%S>10mXZhMtyX^m$Z`Bd=MIUjCh3T#@@sb0oE|2V1A$ zy!zg^QXn-|&g>y-M(ENW+(C^xV<5M^lzyWZZH+5Xyw`p+ytM~ZSH%*Sjm?OqA|Mmp z9kLgIe$oWYOCJt;KjNAUNRqPiOQA z9K3jSSZ;l#E^sR%QeeHohBXvmDNF_L-&WBXb_r&}8 z^zR32%g=_dfNEelGq7{?>cF<|B75T)_TIRLjtbX}F*(T%+e>>q85&zCs%4iNsnnQC z7SM3U{u4_MT0Sqg=~mA3IAe0Ski`ku;SPz|z20uW=G&0~8lXSk7zvh7ebt>|FHU&z zN$u-hkK<`KW9~VorF;JXmm&!`I38m6+r~!flGM7cRzj-|^fZ7@>SSEsv4ZhT7>cf5x!r>a(#DQlT;c z><@@_7WFzIdjiC zhX@)4@{qrOaD^l;)4Y@Ks_5@}x3KI|OU3K!xmhDtoZ(ETo{ozyZ)TvPvu2tD#+9qj zsWltub~cIzvQcY&o*ns_=DajqX5#W^zU#a8*S=f+ySncPrc)BQ&Y7}s;8GY3EHW-L zwzEVVB+I9(_X`GQ&$Keen;;Zt7#?^7F#D#Tr;vUwln~6q^Yt2&oKbI~Wz*+5Pl-Yc z)=W3(iNGW4N!&x*IDf8e1GEY1*LSlRrS+8F+lqoI;O98jlr`BPDg8+sAW;;*^2}pD zYKFn5le(ksiY*|2U;5e)HMSIIRp_u*xe@X?YO_=cPP=dBr<4K43}7h9S$P_tmF2RmGgwr)uiAca)eqrP7I zNH^`7aAs=t^Zl$kC}=o~%sbdBy!@(HhzT3G{5IVx1p#b zKt0dY=8t+lKJoGHo6o|K;Q_1i)%x*?JB%CH+I06_*Aszq_ZYE|56f=Wl#pMWUFy== zUW_A)C!jo1(s6X4t&~<`aTO8Tnbv#!`KjdYTsf07*IkR1ane?L+k%DbT*TRR69}*? zIMqv|@&o2o+rB7W^I!fO7C$hO`O2n>i?`J7U_UHmK}Ux%wEpCrVF9aewm-kIvLf$^ zHl~`=%IF4$71_rucqeJ+m_e68djVO?7n-=`8PD|oD3#v$KXk<}lD@w@O$=(PX9aDj zmv%YY9hCtRVJFjMZ(UZdnH@l#6}U-DYFb<&h1-ZWoF>GxpjgSH{cOHl+e+yrrn70< zoWzmK`WRbR`p-rqgyV4S@lvO}gWNq+04`ix?4t2Jx%^5#Z$tRbm-jjxn%wQ5oW%ZR zyMDjvesnQ6n06-CjsCVb_Gbn?< z98zkm8vPc&pO}7DXY=vE>Lw{MEMdY^Sm;@RrCAgO)c23FYM}S*_#RaBM%WAdClESK z3>bg)LF3UbXv+{v?Q_Eb+s8Bg%;c)M<8^X}91CzYvy_U7rFUVn5ndISQ5Ig7FVuRw z?YruhL4)- zkbb-rzNaPJhRK#9UGD0t_om#wnAaD7v>!Q!{r;gjoXQ|NTY6-5I!8Nx%i8b4d*Mra z%_vO{;@r+Z%88{_?F@GAs(^1ycUL-Xawz@rthaDXl`YU-5)&L~`QuGR@;y^z|q z96`M>q!r^i!to%6`hI)=;m5j#4KHLE;AXxF5|9Qk%{`S<#NU@LZH(9_f89gu&z9we zfVTC&F-#N}5k?{!&5F==P-@}+`|WqHTMKT7;B~k@j^`82C->4eAeH*7Uz7B!ScVm< zKQVj7dJN7`<^~t!9F*D6cZ-dlg@p2EdUJd=1+{cyo)=@_Vp;eIwsq8|$h2!fJIiIj zDQXayhD@ zb2}f;O}SNeN)2{JuP4WV>8=vHwwXyPD!_j3xeM51y1B|M%)lyRxSd##r4{h8DX5&o zG_ol_TvkZQkawupJ+FYam2(W8KS2&hj$0g8=QQmy$YW$VyI1n;eJkv^sOO|Fn0!!n zW{XLwWnV6#|AUhdJ1mxTE0Qs$id_QtEcaea)A-8jcMzH$td)0^&cP-rxiD*;CC!e! zFtw8pHdgB?#PR~Rj~ljtaRm0Gr!LYQ#Ce^_tQ+qSM)s+|4WS}hr>DpJ%Ax7{lo|q@yr^>XK`2l4(|H!fLVlKM+v~oYt%;MewarzJhOehY z!N!LC#F|s2%g9`qhgDuEUFuwb^O5I*7+G?OOqrMBz=n)<-D0q%10DJ4HskVB<{Ysg zY5)a{J~-k##V)A2eCna;u~Vz5VJjf5S@~Zvh5x$mFmzJ~oqZiUpzzvDMAq)(x6ON< zlK0`*vr{*Z(1~(*u&Kuh0SJ)?xStlnR@%!m#XD?}YV26c>e-BK@A>vN9#7K2z&ntG zpG@0gwX57Fn>BuUE6jV{)IJL`tM>fe*_*UPDA@h(mAxVnhf~CK%mC7ucv#1v}^ z7=*pLsjSj2^Wb;qHEmjeY;AUyT3~bGLBLHX`iae3J=Ez4D8@wGw4LM9`Y*l~^!M=Q z%K|6J-;S%{bzIsyx=Hd}w{;_*MJzu*jrnmetz#4(WMFxLA-3uCK1V?*v+k$)&Mv2;Ac1MxUf z@H~Vsd<^}kEX4X*FY7BQnW2G9BoDp;754vza`FFbKYx1*{vTJ*#MY2%K*=VpKy-ji z2t=Dl2NZmaNC%@x=uIZ((<1#q#=CN;`Q59a8K*s}_o7Wv+5cE>-Q>7^v<R93{d#%KrDHrYk>6ZZd@8q%lsJljLD2C(a zaQY{wM^@w`UnH$n6?&D0|Mw@8e7XHJaY@1bb##3CW#i(Dr0Qtr>9aB303cBb%(+K3{X|3030am07@)f{`Xp zPx@ZZ_7_gxG{8Rg7x;oX3G~o{gP#@a1nC>JNWal)r#w~>rJqoU-Jt1ceORGIeW)f- zNxsf3D!R@o$$Y`SkJHOj(zX;KJfOuGIXzk2%4}v~?a8d|ZQ|WeByWtwy;-b$N@M8| zH^{-)w9-b@8TkNrgms*Ft_kXbf5_n`qJWo-N-RZp$}pYkkiu4_$A3*%qv@U0s%juF z3ivc1M842x>&zknji%E?HeJW=($}rB&Z&uH}Q3##Wx}G5^iQ1-GI<29)qGS zb=IJ6EP=5Be##@{kF9?6m0_0JH<#)wy$OG&@CCW{N0hmEWGso77(ofCyOu?W&OjY~ z9$?=6RnBRsy4$|j?5T1i;gCLdNv6q-C7h9)km~&1NnVk^5h2485 zNcPOH3N8;bJ0D#gLJJU|amu*Pq?`tal1(f7(~r{&SPM2tOVDXE|8&%4! zqJAB4sw3PXv;v7ubQ`XoB-iOxHutN6hbRtbQ+@=HN5>vt9cKF%=hevpAkQ@T-w?NW zp?T#WW`?3y--|CpBJg%-?;}V*;(l+6aqL~A8v(1MCaD^5SHly}m~6i!-|E;YPqDv! zET~MK9_I_Dporg32pbg5PCoM-aj(`Y7bXYR)V?$QNfL+PmA0Cz(I7-hqtsR=D$!Kz zX(9`cly@h)pfcyi<9D^Lh5d7eG2~kaadWu+J>@D*wPCia$5o#GrKLbeN>;R9>KTWP zyJeo=gdY<9Lrq6MR%GQc&C!1I5ak@ZAJG&Ct35Ekp^7l#(b|(*lx`Y34|%{pg=a4w zOfLin_;089fCrz{H=?x3vykJo69`9L#;;KGz&hFg+ z!$yP=(yND3(8RDOZh1Ixe(=7KJqx+H$9qogoy%^i37sRRxQ60wT`PBM8*CU6qK192j<2-)m-;eQa-)x zd)semul?icTGcyn%03#}rGIkM)MCngmgY-xTxZ6um>RDY@<`%bX=GXdMa~qpx zvkw)!440`h#eQW=ZOnm{MszIWqZsoVyVQ4tB()`fGY2Me_fJX8cW^`1w~4uWU`edlnu}#749jIXx#_8q2u` zT3Z@}<57@XAo&r3I4!~ZfaGkiqOi=&_KFRPo*j0XS7k0c>68EeQq1HmvL)5$n<&fadBq50s^IB$>=U$amE(6s|BD=q|PZwyV7F(Z+ zqpZ>@VOzuz5*pjh%NxWfq+7nFeyNz}pYk+6+^LnaT@BZr|&*wt6e zM+zr<#=R$zL@!yCMP_Jy7ryeMFoyvX^{UqT&HWpAHoa5v$Je_Ki832Xy^gg92jTO@ zQdUr_i%eRrp^xQ{TX`SoN2yFWZ8o|MA1BKJ;f|f{(A1#H?(s&+mk?d07`ZWdK&s8% z9L-LQPfrftDj>>&``#MS6tZo}F*1YWSk2q*a(V(}+uY699iXccIk7<3$DKE^JIdJgM>m=#>ldVo|9`T4}f2-5_;QNyGAw@JkX&t^h z`>OR)HyEP*$mw04Cm{JJ+4-d~KtV6OpZ8%ALK#JO0Mu}M5hx^7F4eV`$ssvj|9PD% z^xex~%I#&)H-w^R!`9m36^a==MHo*Z9LHa9te?Fu@dBWA?!*FV7r^bO9!$}ttQUw-e;H|CW@k7^?QGm-86*gw3;5mdu*e?@^r3~puaze@ zPF0al&nu^vwXq%3PJ z4oy&B>K(h^!ID@caNgr6VVK^~xIe}+m+u_T`v^J&=)%onrK9s!*?;kykKZwy zj?7?-j|2iRPR$-#zZT!OkrOTd!}s$^6H@lZNcWoFUBCBU>{mHvMmhzjPu>1iN~Yf| z6Xlb2O$@J1BFF0~(eUXpgWX2mE&oJ-RB23_;YH3GWib$>H=F8DZHkoWqEN1k- z7vgba8QH!CI_3B*8Z*8L_wZ4_FCjlfZZMoY^DXbBfWk&q-_1KV^3*qZ`ybC})<$mY zJd0sK9)WhrLmt!tuaAXeckX$}!FqC-_OJku9@JMkMjUrF*wsFEEDe_slt6A2@aRzP z_wH6saE;i79B2#yl>*~?rJvAec0>eFG#_X9qqwH-XDJxXPEsbwszR(ONdy1IQuH7k zTxb-qJkB06n~giu@wv)R^6 zhL#^P*L_JqEgcASM{VUS+(!0DKpOS);tYw|kc96g5c2 zs|(7<4U+}~DHiVl!+8v{$jG{7eKYgn;pnUOt8Lmz-Z|VXd}G^(4-KD<9~(QtstE>` z_94=zD-%XdR2*EzKd|yQPq?yZSP!O8XJBHK5Z072Ox+p()3#IA{_{!Se!nS%JA~9T zDbWo?<7It~sRpOzKZ|x@W21RyLzl zdKk{fe34n2PWwZ8PdlKIdgyWA7P}}$uVozuFWnV)$j-QXP4;Jo-L*^azVyip+4I}m z<$pHic*}O;oY1X$zgwqTkLqw8t2p=kX@C6biX(P(_If$LLnjsGC^NNd)L?E{LVW1ME>r z@tw#?1LGwF2-Uoll|84boWJ0e7wR72(5zI1R#cL>923>RalMO_gZ`1@zgWc|w^w`) zpL>#lRd7B$TRTB;;kyx+Q%wB$E+KFAJN9#&gTPzJ?w=EkG}>@&k4{zF$-GGup^1n*1p5u)5XGhCij$t?$ifmPoB@|&)g$$y?NR@fr>vp?+ z$>D!GR^j%Kr-MW|@AV}O1xniY$6uRvXd^LDvn_b5HUq2IkKd5p8UCCxRWq|kT1p=% zV?MQmXOEa)Y!2V}eY`f4YqsFPJS#y!aUzU_0MY8F1DCj-p96%li{@Gh+fn--wxQZu|4a8^QB0dg(nK=KjddS8cDC&B;rmHRIgw+Ji)%9pASpCGcaz-~A z9euvEWxW=;J2Pv-; z-ovl@B-Ra6=xI$36vbP`x&-Q@mCC-Tp4l-;4yRHGp%cFsI6P=iMn9XB=m&nfUYwjL zc)f$#-N7GGiA+~uCbSL5S8cz!|JYhWA$OIb0hI?re%wshGCXuGCq<@FOI~Pua;)w3 zy$x&!t)$Tn<56nBK@b+6(a1L)>#Hs9;w6$`Zl}#J$y5mmUS;MGKL0HeoNxZ34%|e< zZc^vv3~lH2+n3Min7??XI{aAf@05;D&J3FrthK&^?cSkidrnpiKv6%HbRNuDgPESO z#||7}k_e&9DYLt^8I5*GA+ni>@p_%L9pz<-{L48_dfH;E6XlPQew@kA1v$M&(9_}Q z4+0nX%pv53rLoZ`nPPeEzGChIK8s%q>uQP{&CD*gnDy1V4}=Cr5tx!^R9aRC;FJwZ z4IwOM1-3ieUXj1>g*MC+m2XNTj&8olR=zhDvQ+Yh3$cL|W4MAAEKk69$qz3*E;)^OL@pkOD^LVQ;4`Jm&D#uJxZNPX4 z;jvTXlSqD=7sNydOXJ5p9-CPtrpF+u)fjeon02On`OvOIXRa5N@6L#KbQG+aFHn^! zNsTm;y%g6`@GPS9X+t9vJy+I(e|&G0%EBv_v@ZJQ`pT8_O?54DUksLJ)N7)-!+JVG z@jZ9W+_sr39wmC#d}pbB$(r2BJL6%+;FP<&awo@Gc@a@f&xJ{QmGoCedG7wT!RR;s zqFfUCJ}aS~yw-58w4Z-e!v?1ZSZ3QVo}TGZY9=?eHJEbf^A5*%oWn~;eNpBl4#hy7 zk<`ch@EKtPA4yV@W^W9N`}t^Ccq28n`7e0e4_@o#4jyUcD_2a12&=)HyzS7AQ^coP z(oki;`HU}Iski89X_4w3gmG&ziQ%``#0f}}x`GXvGPjNrS?$O;&4d+ToZLxuYde;` z`e2nMaLiN*3paeRrtjM-BaWjp5NT)Nim=UbJ_8m*2YZh_Sl-IKy39(KKO8KmVHZly$I|J$q*Y zd+3x1Tl{S8o7(%isz0cf&J=l)kyANaczq zQ4eWVYw%46&sn6spuZl9499)`aDd(DMM$+rt7nLhJu8<#cFi1_^5$GIqc}=mdEr8! zG>tsQOT^0-Xe9^C2CeP9@z9dRv&~&Yr}^9Fd=dQDm7}cLCrPS)tZU@NzSvRUn=kM^ z7WB@GaoTKhwdU`X71(wl!Dwg7SHW^lun)0jL z?g=jran+mE!mm%Wv>A#GP@1iIZUE6=BE%l|o&9lgqO7txx5icbifZKE`J@}?Bp@#$ zEY$Hy98dU z#gVya2V<>f&%O4oXbfIIlcXYDaoJSwDOb3L+H)kD(|Y~}k(%w> zwcU%z5JR&IxTUw)&%*lZMyw`cgWgmvl8o~DWSf}hA(pq3Qlg6OV zySmuPFjqe{jeFv$g zGWAW@M+xVw9N;PB4?@Ib<)?ZIW|%hd|uv(3t#w zWsPAvYseX*rS>s!Rb84inDk-t+5)MatCI^F#>dhN$-Gc$pXsbJlOV?>xpVlSWhr=i zle@!I;eIYX-lA0+Pu5s*20_w&s)q}wGP3G~5(my+vJAQctBz4=K99DVr*vp?433!vWlj?^6e5`3HS1 zh>)p1q=Nod!vnIO<65GP_=MV4>?MY_pJxn7ZIZjo&ua&uT>R_xwey| zU^&w)3-dxMyxQo!K$h*UIIWAHXTG77e!17}IGlwadL*YJgQBE|CXX8J_KvI2zi z7XrrxpxEZNlEL#gYtg03WCcq)CC0xhy#Fc`^r%8cqVR zV^otqLiBt5UKw}=!pJ3Uc&*xnn}I9gqU)lvx#JZB4tAc4-q99hopSq?eW#$Ub%~Ao z6)Dr5n?2M7K{bU^jT;Egc6?8^8m2l+-oe#^H1d7+E6V5~RF}sJ?lk9)$xYa@&0)U{ z3GIHY0wV7d-jn;$zoZut;1x4jSxiq~n>FwZ#&3RltnypqX5F|Fii0Bb(!sco%Dv^2 zgsmjc4)U3GjL0cW%JD7B2sxq~mvYV4y_?rayW8dnf9$Bd7F@tWMlY?+_w}o!_K@9N z#^F-ztDuPg@$K^FPcDpmzITIowj$xIeQNIQ^YiH@HU2;p5_j@6;cC9G#P;$BJO*9v zy*Ye?qPqS^GWW3%rjF`3najr%ttdGMM;Hap4q>!%Ce*=qQRI|ND+eqPr}5t1w1Qgd zk9k?H>6z0rRv_{JW2vW`cCX^~-Ro07{YXr%|76?1Jh|O~n|Uvb45LYwDS;tGa!TdC z7$UrT>ce@9I+?~(TbA`?oooT!Q^Z(5yFb*KSX6CD#3ySoYySccY~q<6aP0;oz$;Fo zzwSkx8NWef>XT!#6|+^K!HPC@%sZ%Aa?Ik5vk}qT7{iCxFAgzfuOxIOW7i@8_PNj5 zu*O$#8h?Nz@^YfOGi%S5B0PzBx99)!jsAI|P@~b@oOGRiDN!+e+YR#~Xfz_pEh(qoBK^ziM1hF~ ztv@&HzB*edUcX_@l*S+H$-ab-o0XhV;F_42E6G?>xO@p`4O!Egz5yQTf81P`Vy}q( zIEfij@dNAdM-6b|ONEtue=CdH)fy|JcR3}Rccy-wOQPW#XA2#TocC-fB%;h89z60T z+nvbMX{8q=&@F}gUA{UgGyT>FCj5W#79#RNreN$w`-@40^WCboXJQ_oa2=;Qs0?f# z8{Xw0JY|Q}+_x{fL5gP2qtlT`RpR6}o1K4jJW|RJqa<0b2gfse7Su(nG_DsZ&eU5s z?JhZQ54aVa%us*8Ll+2v~LmKzF1|2x&QX)6;>hvp{?OJ7)TNeu*l4Qv6 zXOS`DB>Nb7d^0W8glmPh{ouOL89Tz^Z{`Hwb`iG8Nk#?u;g$ZU`~OK0&(@}$7)NTt zalky5$H_k5c03lnXC&^&u{HOV>vmhMLC5^P{14>1KzypFOpNxe9;~6@32M`MW)12S z>Ejp08xwVz0>96p5fO$-iY;>`SWbfT%0D8Y=H;dkaBA@gWd#CrrIVEvY?V zymU3S(lT2HClbEpZKB0WeR@k#JMsl~^;kDTF!JtkZ$41-eUE220FThNHFy=( z@Wrej{D=J?*FN77WJ2rz{lCt<>hcn6b#lZ`x|kP4Jj>YS{AUOd`=}OGWCQ{Z8yE#sXUej#z^!l#PWQhGc;hfp?6f2%aQo9u+gwz_sV!%IN z4)Lg6LWD9?(KIH}Yd^5(8Np6oi#-hVpO76*(@4wGjoVMy>pH-D{&5AL>+dV~Wa5$4 z;iib<|L=Wfuzrq5+W)WKR*;+%tgS{B7KaKW!5G(nBqoBQ|< z<9dZl{C?c(N*^ju=U*Ra`^Fk?e)V6;RC19u-LpV`248tXp90;C6`6i#XgNBHWMO~${L+y7W z_V4AZcY?S?cZe(1ngYV6yYJ7`A}C?<4vSUoO7D(hCTnTjs9}wKrDEC-ZWC?K{l)Gb zVdK)rZ=44#9&SFKJ&ODH6* zmeuRgDma9fVgCJJ1~PSbkWG4TyvmLW9L1bc9{WBknccA8HR@ZCm{qHVptXtkjvXd` z?_8f&Y6E_Zk!#zFA~Z1mk&ZkZG4<+BnD2vbx9dF2^toIY=9jQ0}+IuCAI=VF6i8OxxC$kNPFa-?%G#_u#Rz) z>QWOXE@F`I+G-skH{E&0cP-Icb2S(Z%P&Y)9}AY%fh+c0z)5!f!LI%a*5H*1zSzg0 z>|ZAx5=<-dZosfNp0W(PkGj*~1MqU&BPh0kZ@;{e6YOl_EhP2oA?ipE5!hkrd-tXA zARfB~e6k^5bxh?~%Nm$MZ z&si!1?sWR(H4i(Bi6&HHUt$>JouXk?-+7kBn^kUP7NRgmF!7+lO5_kMkL@k9DXy2h zYz{mBnXiZ=bOlOTiAbVR85-o7JQfVao693n!DDQ?`G$39LYv4m?aWI%)utIztDIXg{E5Qb>^^*-HoF9|R)E zi~jccQpJ|F6Jp_6iV@FzRaW~yx%nZPCGOC+3}~m-_D1a2m7F*8BKFGWe{)Obl&jBW z*yqck_XFhJch*dwoELV1jky$R({_?+^A@87892ndTVB|-ab%nT+OX@yf0WYJAqR5r zl(&=S%_o)H)ac2-skm^;;Na#Sx@PM`O!hrh?-@c)U=VkfX~<%C_x|A(8=r))Q3#^8 zxG7PM{0J&0!k+0A{&%b-wpQL6JnAjj3mb7le@Qo1d%Z=|v-q z*{aNZe53CJH7>E5IY{;L%)SdZc?J4P+dcoRhSOqR8)s0t_jt+6iJcAPdG7S{v8R^t zF~@7%!xZZCyRQ@ZzSqWVtH)ZZK#-Fir-qFWT3&zh-hCd;4ww0`riw|EA?qpr)gIYhKrpcK?x|XYb%QipFNEV*2dG;W9tUsd)!Wv^(sOy`cTts0IM( z%s%FVoL{A|_?2em)p}e%ySv#li9(6ius6XKoT^BuLz1cr!oU~47JxAW>O5a=x+_7| zfks%5IF(a8BJACMst+hN&G2zRE@yJB95%X)Pqu$X%m}S|(F(&Z$&$t!ta+`^7fW*g z%;dH7GYqwkpwvbPG)&N!-Ou>d9ic!Q4dz*dkSdaA-wI@7dY(<}jA0PM7+JG5@Ov@T zydf`fvEr2)r&^U-|JMw}@Pp`V7+!6P&$z$MX;T*8bD_<*Auq5@R+!FK#b)|xjo!tp zvTZamoPZN2k^4_6%<+fkN14|wecSc9>#}|!=GRHl4eXV!5v(%cOKbiQ_Mfxvs0zE; zZ_W!J&{)Z`W}bATpWhmV+7vwTf9QX2APIE;e$DUEe)#W~9W0h*f7!C^K9n8w%PaB) zh3%14Fu20c&UbmlyXTD$B`Jq!J(dC2Ot6j!V^@4(Z&5pS5ZxC8*|#+zH>_B?RTUZf#=CistH*{Q0KH&+y6@CMTeXhfY#_2n>}Z&v5KO)u zb<#zPdR}YwW4ksu_8?9I_^-|@XfC4U&HdAX)i<03_xFtr+&yZYEgwWj#P3>T-3$8ID8 zd#4n{4-&xoMkq+kG%PPXa$32OzIZ>HmwZPpWZQc-iu=mrY)UaF@i0oKgz$qXedcTW zfr`)IMTsu2YEq}SP~rY`yo|n}JyM8c+Ey(Cew_>xv6x*6x`v3^u6b##_o`AC-Qis^ zLJ9@!%GU-q@;mWRcHz*rI(!V;1^x4-k|Y7|t<}s&uG^ez@G-6$NUGiQ;?B8yw8Pum zsV^a6Ta~844#~cut#c7Kip@o2zFot5LoSrtKY0E_0kr7qA)B7>cwmazVzyn*hAhYj zWHaBhhM(^(TOB44k6B_lT~Dnt6qqWAqZwmHhP@-;DsvS^<x`s=+C94|X|${&y9jssDVneq*H6i!MvX)0{6Ge9 zGLO%1_eSJ}JC<-c+ zF}|KPRPdhu2=BUmfr)$SU+djkfLsheZ?*u%1Ygy8cjSol)ZII`9ya>|Sk+WXj87i* z{$gZaCxcV;gmTQPhYKiwOjS(x$_vLdDWP_dLMHFmvH^A8URg(#Q@QbEZRbwU?@5AbK2i?)aG>XY!^>UE4o7RO7Mss0raCuGjf z+FJIDIP70R&zS?Sb%XUH#|?VZw1E0&(J;|Oxv;29XX2YAai9)X+l>;+Ngx{=&yR*~ zFb8g)wwaC$+PMAdVlO0MGp%Hua@!(2s5|r{bD-V&CCWxRV)?+40?8eMltsaU#qn0R zbbe{!YMU~F`u=jPYU!1;)7ihMPkfsHB55i~uDgu(|DN!GhW0J61lG8)Bu9qY)j!1R zYwd!yGLdc1YoYFHaI6ANS~nY$W54@yM=aq!JBp0P6x7^itqbLU9dqyTnCK^F{qP@bgt}ufmfs}H~Goam1Uu_$Fc^u|BluP@Z zYEinE3>sicR`1F1ez;TMvft5uPhZWf01t%Uq*Qc3xWd*|C871r1EpVQ4PGl0A^bZA z;YlMNk_4+OtyvKVV43fbIcnq7Cuab2n0lkIb<<1*dDNXnHBxF)Cl?dZ`zqTfg5*25 zT&Ls@s{X(V6Jr^?6$}|heUT`w&`tYbK+n!GmXa+FMLkV>af^;@a)M%|uDXj_4mD+T z%MBOzd|~0(l8ZNz81zj0zqgb*G>e8fm@9F;xpS8_NhStrY9$;R2F&blCV9OOK@fuy054mbX zzdxcUc@W`4kD#2;L|}apYj^)}nCXQlct0?{t@WG~^=4fQQ}8@^-kxN{^9H>X$pTql zh}cw&FzWxP2}GW?!fe75nUJ1^#l)08*b;4q7ZF=e8Gcg zy!P9e)_XpA_QHkiai%EGDvOJdW1wep56-1EB|(P|W#N(5zv{3kk7*7o z^iV=K`eHoN*Teh&V8?f%@v01*QCCK)c_A4M)L$Cv zb;w0==Ybn!+@?iCXBqsQp?1EYBL%TnU8ZTh+hKxxj|X6uYh}ZG_L0da$08_ejpe;d zxB%=5k&oi7*H)7hd=Sm+TWU{yIq>=V1vfqVTenpp!X;i$~MBK z=+0jJdae0|`N`C4XOG?U*qRZc-7rk&c6FFSfYDqt2?(bD(z%}rf6!^YR*_5AYCFq|7Tc4RJ zf!&#yo!gqfDC9dw)x@QcH4M^QqzGp67#Fc69b!!aC4p5wn`6(rmBly#mYGgsUEt@= zce-k8Gd~frYS(gK5Kh$1B)OETnRom#pOrp-t^2!XQfuXi+U{(*-7Fp|wt`Bbdv2X9 zc@6?MM!7ay78*xW0Ct>XFeZ)7T)yEqa#eASfO>a}1_4u*8bZDgyf`PQ2d9+%kdZ;C zP7q_rnPlu(IyhEl@?2!=hp^IMNZ4xS;}>)=R8kJ^>6_J9xt<#XB~R9O+mEN-sy#XX zJHfJqP8=g1P=kYGb&Xh7Yd0@1_sTZ2bDdmkWe@-2{9U&Fr#DK7FgEE=R*Tzn{R~PukMv1KY=Q}44{KT9V_ z*$3p^5~M$I=Umai{tY$}*>iiqtm$ULdl0*~JYI{A(>!MD+wgl@twk*Ybc^Z_@{?wE zx{VC(58ldBFlh*Zj2e`j2xK^8k$j5-%{@#%8P62-dgWu?Ql=Ms=d-(f69(eP-BsQ_ zV2C=4pHRIz6^3b;nB6;1uWV&x`~VqH%@pZFOsB9Tv_p4@Q@#P6Qg!A9#wVgKkRSb` z)_nSsN`@h$;)r7aqc@>ek56Kv@PGACxgXz>Cqr?$ZS}5(z;SEA!s)=qaC`*PYBBM4BgBMMx#~*P|+w31CP<6uoG%o?0|KUbzgrbn_z8l zksnEkJ)iWePTIU_H#ZouGb>}cE0%8c4$dHaS?|czIp6{g%b&P-0mrOVK#bzj*BPAU zA~uDL^BE{dp`OqIdhL^Mt+0L<8=s%=;Abf1PFSpr``sX?Spd}iDPOrz!oqJDifHX4 zfr}ww`O=+h7V!2rl20Qx?>~*=z^E_wQhHbF1vvdnG9U90Te_}V?-ak^sd!s+CSt)M zT`Y-TArxNJ;IcM#N2t_&@KaqCf6eWLdG@7uy1q<-QAAmk;7n9h_h(GLmqc2Qo9xX? zKUkP$^2=FIFbD4ZHVB_yBy_1QLmzfL1tB{t28?m2!=$9Tab>$uzYm|CM5;`zC(Wy^ z1{-`U)Mi>=wKe z{t%zUa?$E!ko;BKJ4sOjp36CbErYKIku8%JHfCRKH)iJxlBMU>VrD0$<*!KY~9W^nNEd zJap3Qibt^vmpzVg8~94}2WJZ^Rs%1snVogvA-^Wor#Qb>=42SVaCXuN7n4Ving@>F zqRltY_~AD;sqw&dcdfWgyt+3c+L?KNGt^|*^~PV90ILYBeP6`QIEclUv&>bI6nha$ z1_ct?Q!b^K>62B2a=2D(pBk%d*NRkTsEchUH_@oaxu&m2&v6o22fbx0Def=5oO*lC zrfra$R>Sn~WH{R#Ef8Gycq6&H!6p9!n88|aH~sjkpb4J|ce%S|%G^?owi0M}fQuR2 z8|bHxn2rsG@;x=ntb6i$Qq-M&?woE@SWnAdR8vIgX9&3tT}P99W>=aI_w%vzZE_ge zrx)3Y*->rx=h<8fT2ermI$ckYG?{Q1=4pSt7|aX6W1#?-C-%J294B$7MRz0N+H zSB0%UZXa7W$AUB7RxM5SclwtIz418M#8O(97A8CI!ALjyy8k?We63WzBN%bdVJv50 zJh*x1hs<2mUW$DbklC)6GmlZ`y1JFvxk{tv{z@0`8qH5oMqWVHyAJdN3F6`L zdZoozQnEiQjl?h(hmad^i|(34}$K%pF^?Q?(YQhH-0gpL@afW-&Dc&-cQ+>LU+5v%cWZ}SJJ7Zsf2ZkndL_~%) zn*eSPM_I^%uU})f#Z-Uv4mxdSUGwqbmbZn=iq+w$(7GBIQO(_p1gZyX4GSX`-Ek%D z9qU}Uc_UYlP)4W?G7^()Y6m;H_J;|LhW)%Fi0WO<@kVFdcCGcN(#j+Ik#mT^!#&>p9sIM9@1vpSIzIdXWw$E^@<>_eBB@;OI#2MyXWKH7>7cRk1H%UP~Zv0Ui9OtTj zGp?@>4@Wi79&Zr`FekfCtuS%Ys4g$PYFGSC6SuI_Rt@%q(srfC@6rT!QMD3pTK%~+<7}626QUV*zLMWHabrl!G9YJ+hvd%jfm#B z4=R127k)p7GrPOaXc3JQ*CI=i^%Lozsc(I+ z`x^v07kjbWwj`}F%T=XI;=DDkW4tVMHdj7Pn0AG!O%B6~-^QTlW?0HXDRBDR$9THs!0eF71J) zPF#k?4`ck*=h6~8LrhIhv`=g{wHUtzMh@xaP=YD0N$vk=@4Um={NKJ`RjW3&w;FBj zJ%YBXRl8GZ?CnbkN~l;BtBRIdrIc26sJ*ucshO%RAwt_I64VG{+?W2I=eeKzcmMYs z$MfIqpB$1a<2tYFv(ESHeSU;QuR4PZ;Ybmce%LAeUAS?{ReE?TDDjzkgWzS6@&YM$XM0zG3?R<*0wq>mVYugkaiB`ix}M*E zniu|DDxaa$=mj;7QjT8j{7tSrk|~?q<+2easxF?_UY=j zv0l1d%y`H8?!!0I3`d%MT~f@&UlUrH#;P6*mM=YY{B_8NVcI#W0$;IVNR(SnQet0O zS{sWv;l0Q9<~Oc_+O3eycPf)g4JANMNg81myV!y+QvrK`g@(K{Eq`No&v-1OrfXR! z?_~v!(jt7IUu0s+H*vlkqrGVUz^bgJquFwaPHw9VMP)jaACw##cmkzGT;7I@#cyX8 zOH(U2ioJm`MdFhsF9IZ)PQw!$Kp$6MZ1n!Ovf>LKH-x^NG726~MDk?Nh{&y5-^jl} z9MH;<2~th!zj=Cm%;-`ICwb2WH`KmXrXBckv0J1w`cwGt9b|!wRKSEzRN1XoBre`ciVR)2MYf z^#eP`ed-fvBjW%PYeYZo437Omv`N`%cz7wnlnEyE^;aHT*<9VW__KPWak@{glSi?X z?_CzImKS>ki^*X${EQwU?BeoWX)69@P*2qalw;$_N$V+j&MW7;9?7i*QEVyGjfXr; z&l22KyNX(&LD;-=u3GZ7{UrtlHhoyHeFK>}>#I@_MYs-=_oD)9S|;~_&$s_ZKtzeV ze~*nh?P9#eEx}fHZyoyYUj5WroAt-ehCSuRj%+>k9og_uy>I@xp>98C`=`R9*RpD- zV4ciKL7E9IQ6FpX9rGBfivUuW(fCJq!tkdYX;(AurNl7DZ?p?(ndal))W~9O3vubPy4mA-Q6Za#TLkWrW3&(?Fwt;M zL1VpPmE?_u2>eK#S2GdYmVNrq^+xUx7nPCFZN#<&oMA6u5#_X*%g8MtC8B#?g>r+eItOSCB6q}^z(Lr zY&i0~=)6m_!=&2Cd%-PI)Ll6HUlf4&79X8goFEz|F9?dfQ*4>|99RQf^J|}GefQmN zq041D0KGHY1c{z_KQyi*nXK9P{0=es_OQhb=l+%uQdSVFQ6)*5)Z)vfgn83k<(HYs zMTYidK4@O#-;K}UwVT}Wq{v?FeI~HdZ?OH*s)t)3z6nqL9O?E=Mscc@=c#i!5~0K* zWT1C{i_6R(dxbOB?4ZTbF%_!yz{47MpvTZj9ny3gtJyBhJRfRPvgmNA(CRe4tWln- zkL>oTlTuJw_rz*CfYJ?k&3?GMEMo&M} z$|@f!`GQFRTx3Q4^6{L)NEBF5CTbT><(M-nytuRX$NBVAN(xjFY4|lPBSf!od;M`| zDvx)EaVDyuO$A)`N8gj;3x4@479SR9sjTQ+*6pvzy-9#EuImu<%>(0t4TQ~F;2HB@ z0J=iZ8K5fy=W>R?_0luo}b2UvqOSdxN(#&*Qpq;X0QU2U>Yo(SG17RTdBU z@xD5zIOxn}=F`oJ<@2`0!yH0iXT6iAKdOJY!u%0#Bje9?{h_S_6HHw6)qUS^{GRXT zo-e*VmReT*)Lf)ch`#cuDv^HKX2|0hD$@e(YKZvNeFSvCEHWdj!f4^3XL<6j=_$JV zFPru@^^F}x01M5)ZzsGn+>d1U%X)aZM(N(V{MEAbTjk&C(}uPqseUWo;;p*{$ZKed z%X~PxzG(A0eMZnu+YRf`eJv(dQrkOBg52>=GaY3-Ifq6&K;rc{J{HR@o~L+ zTnH0^!mIQgo@5+jB4mg}y2en{^P?4O&6o>mWdfKYm@4iv9eA?0xqKJ@(akPoVphPc z)Vh0Ti{u|Ay8)-qNZ#T*=HJ%~HfM_`ZeG3nEp_cxWqI96%N8Fy)QazVO!U*|B3-i3 zLADuJ*ugqJ!iK+o#8-L;)%qXHrJks;!j2X zV))9O)hBwKoS_utmk8ej)$8Cpy0lfzgIKwO$>{Lna#-O)d5=MBJ zcPP3PvaC&{CMoY56cW{ooJ4P$--!$UCcLJFCv0sR7TtjR+gekwel7a9{XhRqWQnR~ z^{2Uf*Fc%@m14WpMBD0UzBFwlx!%&r8#c~MxY8`h}-o zNH+S+4_f7DXxY;#RfjyWxp0*}gRLbzDn#FO=Y^s{mGa5dvbFYVhCd4CNv-_b!Ao#A*Yfp5Q{0s z5miPRLo}G-ccR=Y=1?ANNj#G@nbrTfI$?M%x6OL>EVrn?GQGSc+TZ{cuqu902NuMQ09Kx)3aqdtwiR)5*Rk~^+}S%bMviSyqMf9r6vFbn z1RgGSI=V6Na;{fikh=6x?c^77k?BPhYn`P9w4idH!MT5+4Ro$6^`7>3Zg!OAK#5xU z&~tiK$Zt=7#d{i8f)x`D!^GB~7qMdbUzO%5;c`DHr{w{@Rq0grF4ta)(Q~N3E*Dwj z)alu|=KuvOx_{bhN$NnPJ^|1;0{Z5oJNyMe5;O1*-EEQcDPGiJV&F`5@#Wx|XI?)) zX(Y6ZvxzPvsBZxc2!=)wx8uSAq^_879&L1jc(?7(#9VW<#J&Xzyj~Y+d1NK51v^6S zM$WXgE%><~G58(ETa^3ueNW&fE`4t+m5Vz1#wSw)6x)O(lAArVf_m7of*p+u%U%FH zFmjm8YYc*s3?AX~M#`^YDxCGtx#Q>Sb(-@)R<}XPX62-yxEl?8ob6}a<<#& zbQo=p*BcTLjJZXT7Mz;nXpvblg!egAiQSt1<}JRgNwtF|J~KDzPi8*CyUD&?%br!4 zJrkxrs~shbF!6juNMfkmd7~87%OAZ|o&v)|m?UJqx=OXR7FsH+#fB^pVe#vT_($@f zL}h)o430I`(NRZg#!H7tZ}rq?ZpDH8)?N~J#6-boEqKU?hOOV&$vDN^#!noPDQCkN z#)Y`Z3WJ?_eyMmC9F8Lx5Ux@OpjX+`Xt<{aXJ*xcLWcAD(=zLT{R3`Y)v9mWT(iii4=AyZJ-N!=vxpokQ1ecq)hskv~(S9Ts}TGtLJ-fL1Vt|@j~B0 z?nTZ1lr}eiK!?v9o1yx*5A3+5V&opQZm#~wa}qZBDlmDIdDam0 z{P~M|JvE6_MbL(KPrboSS%!BJ)T6@+ZL3rL8~21@*s9}ct9tT&zE6i{cRc% zXY~q-KXFS4`1fmTN21f`>al|xt6V?bi zryLSYVODCM^hXwX4#Jc=e=rasYy@(*+nI_6CK%U(#@M0E!>YjhwOijjr=kuoEbe(> z&M>XX{X{Y9hkZy5YHcia7J7qbZ(}1wOaQGD;14HG_)`b3l*0x<9cO7ta*sMNSJgGF zzhJz})nH-y3aSS=$4v%|K+tX3g3w}WUcd#FpNTGk;!{)EIQc}o^4S~`4jvkUP&jTN zNli~^`g->9PfN#iKq$i8WIZ`1-f1Qoz508vBykXQ7N9e!>4vVeKIoNvqd&?X3H4jh z+fGD;Gd}^>cBBZBan?l7_SforJE*Fn&Vm5SU?@*(c$&~`2j;Pw3n)aNct-W-DyH3Q z)L_yKEES|n zyfFy2Z$chE)tp4h5g4Iod>B<$#%f;qp*em0iTP?ule1zut>VLa@)hw=qRv-#4pB3k zCv_4M&b#^M1~&g9gv7}t4Pq+_sn)gFv14oWUqhjJqp?Or4zfwAY4=>4^8)s9(kWlf zwtMVtnP*6Ctoda5$0&+HK<|=5G8HI*u@;icSOHdShb9DJ^fC5mj7YbOM8TQHLa^-x zoI=YHU&n3_O2f3|A~Iq(k+b|+xbIMyzZ4G45f#T{b6ewp2ZjJ&I1>@0?}0<1ni5z< zI43Kxl^1UpvLCm;^%KbcK(QY=)TRW+QRnj#(7`26b7!!vx!>9k%^ZWvuK|_#-}{)dPot^)%P!ae$$D9fW)0ni#Pc|n z1N<`JQhwUnmGy4#&*eSxuCEehJHsj@Y3l2sXg!1lz?OOneNOhXNgV*}_r!j{N3E_1 zZ0yDn&QADycCV|0hIM2k?WYUU@@ahw_o?Ff4F$kumiQVf+WhgH;fwaOd=0|0Of{*AqQ3PS5TtB#Z5|?KZ{)+ zY3FO8<4sxusWTWELDne{J_X3)1Q-h#!rC>zQav88T}Xh-0+>oy_E0 zA6@e}9Tf+hV#wV4voJ)JW5QB>^AI>)f@W`>d$g+Vp-mw>VC_B@ouE=;0adrP#{Ts< zk(=xT6kV3NRc6r+IeOZKz%*XNtZ&V1Z#3n2`wAfp$Ns(RJ4YB+)GviE3YE_S8;>mn zS|bDE*fYe0>Yole{I<3(hf=O8r}68kw!`dtx;0BU&cc>vODjOcrI7{{7RA}I)?mU- zKCzm+m%a{pJpCEx`x&is*8RO_Mc9m8&`@2HE^S(Zv1o+4E_JoW9&AnQRw!KIWTNmQ z9aUtqguD0+aW%&tFK~mYN-~vQ`1I9O&|3e;l`vPw1}%`uGRIy?L4z>CQWEs}n)clr z&)VEA8>sKNtNmM}on4NwwJRLi_+`jiufuObzDN!pFtf^Mtw6Ec|1YFwpDh`(u?U zRgvBA8^3gr9;n2y2K`d_+BB@@Gigj2Y{k4r^Ub&Wz^XR0&f@~8^}1XmU+f#Y!f2*0 zsU*!DJeE=A*hMhxk_lz1SRIoY5Aau-1%MbH-d^2_S2A#UdEk3EiT?&FrzOWu#<7ql zdNE3!*<4HnZ7uNcJ+!8EV*`wWg97!ALX2UvY6B#}XoeVR?nfs3bG!?obT(7~{wEZq z^(D|T^PztZaygF3Osc(^`0Guo?F9hJZ9`yM=M4mNVxLv$Efcp)?}D(FU;jZ!W87V1 zT-cK2(aZ`%zIJIlP?I7c%r%ahqvC2G&Q;O2@j1Rqfdghad;dXQ(tuxEaP>g7gQ?~= zP$C+Cd_<7{gWBW{5^XLA)?rR!Q~>^MD^DE{s-8ktKB3) zkHe3)<#cMoC&}tlYO{FR_NlM|PI6;hf+OR63rxu&9W=^UP2B^)BZkEFb>`s&p9O1C zgYI7Oi{k6Lwh{cWwZ*li4H1L&Vh=2^K=wg7p!_);`PwLKXVv#U-vrQQRE7;qgn~i!+0HEL1r;U=`{q58g4_0^A~?NGqGnW1t=k3b>`Igm)-XgH=j}ahX1~nL-}b zYG)@)v|8#*5**y2haNKhbI~71J)~xK=S9ir7nK#dKYPy9V^(!Gi?erD2RJbYg_$AN zF^Ba%!{AhC!bZmXAB@%xwD@fik6Eorrp0vwTTXU+LW*eHzAh#3SR)SLpdTguK<;#^ zRYX393WXiQ=|;W?crqGy>!)(96yE0w1-J+rBS8I{>T1>vq_x{vgvQd? zPu+&6c3xtEBK=QB|f zjxmb3o3s{&Gotucg?kjgJMNSq3seBTsO4(p1h5#EZ24Eqd$R988tO;c^kY^-<;dC;NSFDf~Atoffw8z94zi?H>Dz(c$_3RbJ#j&5G0X*+&VYiuoO; zV@iJ&GvJW_C*$FN`{MsE-pE`&vQ@c_KyVA0UcxATNef!J(!mI8M<6(+3CBkr$B`|^ zN3i1mV{t4yw(h^Ah;vuXHJ8cgbNl;{>CZ*}zyBYvG-(|8*GqqouVh*AoQG}0Y{lG5EJAhqa@ zGg0?`-~H`6u%NVQBkpY+!H-<>SF;NFmuo~Vx z_O5Ue-mYGFHtmCKm~uwm+iswki2updS!!ML***qZG0R{`}JOI)8f({PTI>NqL?PDe&)W^3fF_i2u9+yujAt z|6Bm-@2?(=UW`95_p{#rnQTt1UFdWsm;H91{#R9Y?b1=0lF-?0MzlBcWv?IJsLZJ5 zZ5i|4HwyEP)r^2=`x7=9MB0V@bABh#=kA(JovAG-mr1%dsdIi+a|-b_85IY{+G$Sb zg*t?pT0rK&lb=s|dolfEi%n{nyu7^P-@UUV!9>N(C0|2I_`;;6%};~NT!=a8NVclP zo-{XzU5;KhV&-;LU+QB}#I2OUF)mLW-p|Fazsi_Xm@KOlC{#m3@D36vSp*2mh6gTZ zq%XVwvJ;?__dS|ejSoPR2_d`4$5(D#{-7hPRKaTQrsnMIth*}jp~m!H=fmX@HZ-8y zasTT@Qln}fR|`_wm%IfH`cjD);<{6uUO_myP*vGC=&0<*da~Ko+pFJa!dlz@o=86bh5{}BDIz&e-MEqWiHH7fKzJ6u)$B((u5#*4Y6%?vFqyI-nCX% zRve7YoKW3QqJr+z$Euj8YCPtRw^LG5RO;n2?r5VPciq7TLJ02};sCXBK?8EfG5m}p zbg0CueHjm!5cq7j>U6YYD@bThJ>~=%N79CID3*iJ~zA;WV>Mb-sNT0LRoy{0PkBvN*f!OGtNw=z8#&*0LxoE62ZmOU<7~Yy%4%Rz8*_ z(VkJRghnhcTm5Y6n2jJ^P>& zWb?#x)YJ68nyRS0e2cC*0YwN-;WIIz0kU5MHE6}%d&sv-OG}xUTyjdIz=z7qKoGGn znOT3%B*omU>1X$bnlvW;YVNTxp9S zQ30AiJOZB_;X6;F7)^^s0@{fjBop^WniW|{W}XXcHJsGLIvf*x;3IlMfX@vB3XE=N z0&pd(EHuChsR?lK-Be~Er0(@`Vghy;_`moI39L2+RkJ~<&ii7SWhGlcfuO0@(_h7}S9_{JJ`?%&VW0Xp+ChWZc zA4;0v`qtyI{{!tL6y85T(%{SI=cV{bt!kU!53AlEUZcVDdI57|K>LS7F zmA{ez`;}bj%1B~!x>kI0`liFXL{U;8JfVIVwR2Wv6nig&V1J4a0$VOiNuW8$QLBqC;aC*ecCZm_LH`>j*;`xr@O7u_%iHEYL`o=)rv>P+cikWVPEBCl+Zm z*+WU0wTlx7h5}kvHcGvpl;ioY{FKiRAj;QwrXBsz5V3ua?=|oQ`Yo|YC)j!El7brp zJ!x)aP10z5a=IBDKRdHGksiQd1R(mNE7Hpnt~e+5&6Df1w`SdyYv7RPxtqfT-qph* z>|ZrfQxX&weXMGfbIwxO?e~Ba$aL8&u~I*cwEE#I?@Ur9VETm?+&pIuhS*p9NEM^1 zu*A7fEsm;;07pjJnMbFQrj9d}rY!q}CvuYl7k_<2vl{e^apC;Dkn+a!3{{|dP7xSiCIh}3(8W6uUZDl z+cTR^$(qpImp?Jl5Vq>i-~u+~*!xrY!~mOg!G)0B?9i-TtUwqPe#j>7oto|ET%7Gz zkDNqBa3;<5dh0qTc=!~_bdmpSQ`iD6jSKMG^IPB__8C~VbhtDr_iL9o%`My{(y>VZ z=cQIOL{Dd*N^g=(eA%u}XX??L!?R*EDEvZz=Z=1~Ch&O$4!oHaFsa_Sq~Ox@yf=t( zQSlsNabV8EwQkB4?#W^Tyg3g!rq1qp4`?=YM$^-<0dV`i)V=p)n`5_S0nZNXBlCpK z7Zofniy1hmh?E#5vpt?ZA*Bh~e{T+9-8XD}v)AQC1j=Xgt1LD{sGSux8fWwgkIH3VOaCi_Dk%GGDp|Udk zDZU~za+!_3a_DGZKs%BGV3XtwG__=v-q3#D7%+Mdi#Fvc8aac?cISw`@x|0@iW@lz zfloro2So&`?gt#5s^Z1Ie8+A0psb&iAh6?Qw)`kA#(*?#Lzi?=tulY4q~rc2d6k=9 z<&`Wez5(fXu&I6(fSl0VFNYcep+y>zUy-1i7{b7TM-5b!%1YXW>E_V*IsxyrlA zevM9s109ryu~zqs8AZhvl%A=mJ^vZ7;h5y{D;soOrB{7IKo<@+FhaW_*gJ7&1bgN9 z1$U)kDuA$Hvh(Nel(bAsPT(Pg?hO*N*w}d>g_Kgjc;ihXQm+nG6z00Fh@oHyUY#UZ zsr&Zq4Ftw}Y{zo!^2B}|o#?Dfov%UXH2z0<@R1MlP{b^n#n(9&A1}*J)!EgjVTlB! zfRcZ@M{?BS&B<3kOJ27BmG$Wib!u@gp9zdOKX#|6a!{f0?;N+MBM!x9OVZ1H!#Nnpve*_F?03UukM|EY4Y8qP1ZwJyU~3xbd3+;b25NlO!_70E-p+< zlkR@G81a_s^-hAq(c~ZqPY!*+Q>paBY;r2z42#|OAyVKLbS%ih)Ak3oOaGDtD8Xrd8?E2a zofM#G+-Imhm8YBH{!SF!zgF6|a((b_%Vt{Q`xLerK0x ze`(bkFf93NjFk*{S%;5?c=-bQ`xG<}Qxk_9F8_yE_du~QE|t&jc&nIQXQtdX`(iff zN)NMS`|)~V>2XJ-qi|fK9J`7Fd4OX#$&L9DLE)1n7}m1gipw8`X*kU3_&euQ z#2y#rw~gqz-&G-Bql7)udp=NN?O|Bs9+`7r+x@pu5d!!JzM8i6dyVqkX3QA)?OiX0 z-dS}=`w=%h&DosvF_794HWE9g-gZbxr24CV5mZDUFS63xAp=c1fD7;##i6w~)kH4p zpI<+Nz=S{LACht|FC31n5{pc?xDcG?u+`e#I$w{Rjt!qD0|+ky4h?HJUd0+BTN$Y!BEO&6SBr|!*pOzcpmvLmGWO^WdAk2`6~RB)faEZ& zw9V=SsxHPQc@`6QG=tv#ts9r^CgOm3xl7-DDJDa$A1sy!L&PHWx&k>(jB=9=>>7=_ z4;B|mhRf5`B}2Q9W-zRWLkx*sHbc}>YLw^$zC@?0C>>QQccK1ud9=`G3Mxh2=SAGo z58vh`RGgj$^#B{l4^Y24^+1qAVtium^@(u#8(vA@>BU#JbCJoR)#YzTIO6Qr2c+k; zIY6w(qijFg<3$M!nI>|^y~VJJc1}R|Ht3((ynb=UH;nFW#l@&@NBjW8e%w9F&imsJ zU%L#HbOnL<1K$c|`FmkXdw@1eAZ@2)rT(7wBUa(3P4e6ZMbHaFt?t)Azhj5I&~ejsiE`=F zDf(aImT!y*ceJTJ<~c!Qy2(XIz%5TvEVPX2O7`MhcjGhg*WHUIsN7hl-(n1G%@{9s zfB<%=7FUhd=2qXli!FDpdpY&4Mr=~@Ox?D4_vKdGyw*|DR*0z+K zYsid~T$tb&7Tww9X8h4D=Ssk*MWxRxdlL*EDsr-eqF%qY*aUp1)h2xAKfgV1a_M87 zD0hpIcbsMAmioo)7piuUs@R49cP}NN@RQVw=9U;eIKlb)K5ggZNp>+1+mN5+hl_LN zRb;tq*o9q(|A`_b&BZ+~N2L=fGyA~A=?myXvKMcb?ylLV$ZpM%`>3AK z>%8;40GD3qep1mS2ToE^Le+YCT#Ptg$-2qkwX^TcBOwdhAFJMVRdZ?$`t*BFJVCp@ z484!N1?@NNHLuI`jJx#WJ;g@)h1sJ|dmrX2+en{W6Dn@Q# zpgECzx?VeOtI11gFfMAo!j|ux1I(K+0ML66mp2`B`@zEo!e4-q49rQijr!vxz0O_{ zLrRiQNO(u|&RMW{Jl7LS|DfsLzMiyam6y-HJ4`?t0*nXNkj<}d7LmBqr5eKsH@QUE#id+($@0}3 zN?f2}lc7Z*fLvzpEs3dnaNd62DbBAFLD{?X$EAnp(vkulBg7p)5@yb$I}!~etV73w>E^4Ngug{yV5WDV6>< zlmgTifB+QPhIxMSAj3xaDi_N(PD+gv4iw|rVXwJ7_a7H!Z_ne#_yY1PfkCBE2*7o< z(E}+EM;tl3k5q2gB>4J!B}=u`y8E2df~E>aYuQ_%^@O|u^(fl|=IA%!E>0E|pdG{6 zX-VZ;r(gpRTs9vz1*GfZ?{vFeUWah6uikCC2gqy>XgJ-eFKsPJdvElFPd5V!|0a7w z>44ajaRp2}2+RIthich;kmf~|!BmTu2N|lPNz_NXA5Z~A+%+#sggf1MvM5%3fOBdN zPsij*h;LGxF5McScD;|#~WfwZaFU3c{C^aQ#Ho1)Fccl4a z0kSP@jLg^Qnyuza(BmG>I1{J?-@yCc2y_wuhY21kLi<8!j@PB^L+z! zc+J$4ncLOR&=W*!yqq~!s##cbK9ALSjXL_m`gDr(WFat0P3zTVqq$D_dsJs;oBPCtv%`WvHnj30j^w9perW=^lT;a6#q<%8mZu_gA zfa3slN|RM#Pn8+32|0IUl4AEF*@pF2=$12M(1Z2fQZX6vBieY5joB|5Ck3C$9^N)@ zFm&7AncZou_8S-h(Z+>grFTO8+zyFX)=|6me#6R9-l-?v+wxzz^3Y?i*ATH>k{sUH z>*3Gb#bdUt$Pq`1@848^?G+t5cH{f8@{{@|QyA!$AY&sM1N|PmM;AoT-r{dp|2&pM z96vIj*(>MV%WfNo?-=}V-$aW0QwQ&-pB zQmdcnKZB!;+#eLT5OIZn3WMdCAq3c4rYLQU#Zg{-k&elaRV`Ax;^c-VU0d{vB8F@*~q_8+}QUU2r>8Gvb5W~I{1vc zwsVQsm&1~+@AVAEqk>X{hzLol0psiJY_s@i$O>yLF^7CH zpD*niUVoJOEy$utKMQ}bb_h796q}I5h)SlIv;81nykKdQ7w4YnFwv+HRnKc)W=LIe zraKb2gSG%es_-Gq$r|)#9sA8}>rdnKsu~JR>PFQ4LNj7As6V30q)yfQISLs?3xuO1 zhwYfRYVqTxg4sJ%k#}CZBq5(dEROxJffjBc+pN9-wJv$ifsNO2Y;Dpeaqpo?j`9Y5 zSMp&Eb9-Ns*nRQt7m+Qv*&`cO`NvgH49i!=Krlf_1JkrvE>|s%+e<&ZGWW_!xnn_>(`rToFKkzVfS9wp>yl z_G%2&Nq;D!F~G)FVtdN#zDfJ#)y||Nw!9M*K|TVs0Vfs4*{#>RR7;YzeI8JVMSJ?C zvl#!$yCw9Dxzdq=ymYCRV-L6jd>RXxdC&Xt^L5f~WXBhgk@OjN_h`4x>a#IFGwXj5 zeMNW@R(#Ysr`47LBPohN;8@F^uTYcw*erADSFy(H{x2n>9hgNi8EWOwggg0ye##$l zicIS6sT&eFt*)_%0g6lNoo}P2tXoHC@}_5qRH`QLDboJ}GyCla4jTflFV&4?WbH4W zDyv8pAF+gr1bQ`4PRQ6iZQl-lPo&a*Ub_7rjA-1FI#HEHQ{_4lXqZK=Hj8b)1WS?4 zT0YwXlVb0jwdmTXAJfWF&H8Na4z16oAwS-dg=9W*!T~V##a$VgcpGMPjXdZr(ANC2 zztcon_jlyx`9_h!EQbkyzPZ(3Vl7O@l1U#Fzm*gqac(4+J5~_siWF9`72f{5UNl#) zil#Vo9+k9Oa7JluZ&^VQ_+C9hy{m7&>B)IS^7`joOt-?jggtf;i=|P?TQrKnxRB`t zFl~yt^K?ix0;f>yK&O6ZMRDCcn^LX-Q3A*SO`;42S1+|D-DrY;on00H*PY25U9-W8 zk{NnhP#_3(n%!S#XQ~3sPW6PjxrgvgY7eOjr5G;E<}ck(T*Ij+T&t{^K(m4&t}-8D zh`LKqa-`p{fyuS^XwKdDU9~lA478yN8eMjw7(OOlq(HruD6LZ#Nc@?{c#cuR z5`JoUe!`>W|K@@`yGnz?2PgLRXF9tAGW-m^~M!a=zx3R0w<{o#I zR2i;@I1nh=hyy0RM~Q)MG*x+XXgL63!_#+rju#9Nzv0E7qYd=sz9O^h#r7rsT-e(* zt8_0TUW?)+-3o+9vP3>8h;tA>R=MazH6&|LPAW24t#QBiFE*=-oEw;b7(KQ&PAlSU zg4U;DnD#QqEVo#UW38ku1NYM)uSuo)N6xB}AY4>j4n=l4I+huHXfpj}yAfT&*?W9# zl#%yu1?;|XvtJT6hZ<0pRDO!wS?2Uv=+?k|7 zt!)jNh`DX zc-B-WIRpliZ`^=IHr*WeN$c>0>Tn`}X|>amU6#9EcimO@W5fg3P~h;byVl{ZX;hSS zVPk6ttql3(j+5Ys3u$8rg3RHoqBjA3ak^KA0Yt!T$tMiS)h1C+FaUAeHT~|g`h~(a zzaTMdR8SS=J#HU+4uG(mrwX?j*W9RiVP~;UpB|V0p{f&iBYh_|I1_CyPh?mKbBastFzwv(&AO{ z;a_=*b)+!>pO4lOac#@#HTob%xF(X7#ox8wmZcNzhHB>5^btEL$C&P|LB8M}i*c)w z7Whkkh3+^lL@#DYu%Nz;?MC$Ft;-|Rd|Eop?2|{XKugP;&V3e9Rz2?ZW7i+(dy7-| zhdH1Mmm`;kbUDJd723J=`J)bcfx-&3{qMhW;0F1z*PNO`fC-G5vuN#O3Hu)02j(*zRy!+WIa^aLx&yvn=$M=2AQy;`}^kX6U^y5MMsN+OoZwqD^{@Xl(8zcj2 z8UG;UX^RET2OW=%^GGCL5a|iEuMQ62%_FG&dSD_97KU8K=X$OPj~bAArTIup?bV{? zu+PQt>8#!w@ zp~}q97<|GsP3(kdm0J>-{_HSpi#F1%H-2Gi+EiGWvy4&)v?9IxqJefq00^Zm(X8e4 zc0S4=guY@gU$mkBHb71QOv>|+jpR+wrB8~Q9ZLv#1mt&S+C{y^uzfLpsy=w8S?3~g z^?_`jfVz0J0#S_yu|K||tWcu&1^f##h-`+-vg7tz+XPqbA+R@2glZXMc0jZF!VZaXKE$Bz}Wd`H;oaV^r@35N@-S=kK4@q z-!Da9>jM2qoUJthwK^H9*%!KWgc^UE>rv0w_CUS$Pm)RV%gecKtOHkV_X=DpC1#dkkYgJ|vKxv) z1Pwr~#8hHLEcjq)sRzd?_v{~&0O&EPk8yyOSLq^Tv6dbX)TuX?F%D;;h7NLstxqx= zrNn+Ag-_(8cx0=&)BwaBBN^rK=Kxl#N(eY|9fC)Nj;|ceJxY zs~R_p~|vv@s^2nZc7SnT zZqlG4{SV6l#If--G^lqU_(VnyRu2)Mdo~$Dk&i(<3}(^0`qa=ikjk9=}oEw~8i{fDfogg+za)NAN!GS>8d9_QmTa_a$8}IN1}JG^1m=Z&vHN({UdXJ_ zbiq7}&6@59A7IQ$2VCXbuViV7b>5587;7M?2&=&x?6MQgDu(^vSj* zNV!Xh{BpwBwQK)P9j^132j5sk(}f!fkmG!VLH&7CEp8#qpKsyz_kGIF^Ey2Vkku|d z6xZ760T07YK+m;Dm7&2ofE-JEnrYY|RbBD^-1FWae&cs}zU#&%YvH*{@#Vd*M6sVz zeaq>cudZG!v!uuVX!gShJp{;<3y${nR;YYi7EYw{s#^H!L`P^ZP%AEO$tHJrF^=SM z{&c#1Mn#jd{V_nm-m81I&-kG*RFsa6i4yCWw~rC_C~9Fc2$w=ATx}Ce z=w-}YOa+bmXCXYJa%6r{e%}}p62u6lz>M?m6&OsJ*|Jk%1}unVhDT&D6`n3kdGqki zQjj0ZlgHL^lYmk98z(S-Od}NpFo@gE`xCBZR6?Dz4>p7Vu#C@lOWN*A1&oECd%qo{ z4}IfXj7E|-c2`h)lu*3<<|SA_5t(ScXAWVJbqRn?fU!pr_C+4yw;dFl57D>*RM6Qd*d*+WeYVvL47e zkbyZ1LL`)^Q@9XDPbFB~FehtkWHC@2j|qTO$>(2OJL0dj7sc^1WJ@CV<2r*fQHP9d z8>dVnoQEFsYOQ`WpZU4LHK_`^UUk8^5F zpt_2F)XRXq8EfUaI2wS7W1tB63cZRcVD=47^TB)2qILZ+K`+H@ZP=9q@C1EMwSScq z^T+|uh4>Wq$`=a;a%kHzHBLV{E6|LD=`!o!54FS>~E;6ybf!e0U2dU4yWhz0af| z6p*8I6bedb;}4V92LuiXjX%I*^7 zRZO`LYTZs#tD?r|UeD)O_9YI(GE)YmB*yOOQ~1#wA8@xUi0;@lCncqSjHak;7xqOz zj%nx!J`T7SRr>uq`{zUHd5vCIi*mnNF^qh#jSZNREbLDlwfV8tyh!(ruee@k~{q*e~8Rif8-&wU#j^nU{IxS zPA+BHK*=~-)y#jwRLRzM0Gwo>Q`ef>6TDny5;#uX949eBXyO3Cqn{X~LM7q6{v_(g ze&|-x7I`L@k4^R1==|%Z#~}WYv*v{*2|E%}Heg=zk$k{NZ4_8@;(7f&V0qkWZa>w2 zyi_i!$jCTCY8;mw_hA{f6o{qVdOkNS_pz^H;ja&|B!Rc=I9QNcN>S17J z1zU3huxf*P-v;lfL>o*}Levsr0fmD@l|I1u?Sfzn_KIf|936>)#9M`xwfHsUVc!t{ zV4a)7%3+}<2xh<_SAr{p{q$5Ff_c4Y__=J>)f6?|2!`@Gc$FwU*l1_)i4)$h>)ryv zb2EWpoQYqSPbju1*a~FAHOD+6Y&=I7sJAnzV3F-c zWNlb<3M2FPzW7tsu{_aMj&>#m5}Vt)%*SF;6tLLuv_7t+{aP~Lo7a?G^lZ7?pxK;E zo}+Y7)SqvBUTrLijjK|*U>+6$YZ4Dt091^El7eCvQEr{2!bz5_f{lm0pUx#AuA#BB zF_*FYD&dxE9E=r|W389--Uh)i5hJ7LJcAVNYUwmd-!b{2TMPcFk7V?~#?1dx>-gOS zX^oTU+aO!LFQWbU&~^!PFD(n75&~c6C*N>(U)E%3Ps`$r+nDh0Dev$>M2wjcib99l=IFe(bvI4tDQt`L#)aW%ft^ zQvUx?#`*nfU_FZv0DEh1Rwg6_7MGT8{=zf@MBLU%(SWeO-KgMGv$IokKoK5-8REf# zD=37#h(8oK`$P(K_1`SU^OGGw{nPTST@bBgVv?FF>hhf&=#Pj&g#cO1fhOQR)uK<) zhzKFQY2p-4P&kaRQHYC?g#)0K^wzy;dd|(wB?V;vcGrS0?lij|E}|1L;8ko70mirq zfjx<1H6=F;nGkyAo{kR9MrcN^TE<7Pop+&`J=P)l%^MC19+TLDg4@^tSh^x4{vZK{ z**lgjMAFSQUZ?TzXmD@Gn&CQQ=05(6Z2XBr!DFV|A`iAVOU8djLv#t`JjXB}@z3|Y zyIi$>ET`lKKX({R5wx9DXF((G%@yC7kdUyW^WX624bi12pJTFP7Zo+Mp0c*F$!~mW zeePgwoPrI2vl8eoF)Fa;mT)XF^%KTW>d3@`H*YA>R&0nsU_Iwna1o=|#{V)Lzw=up zg&yp^A~CZJ0AIz_A&yM}0+|>YWvsbGz*hV#%-omrbj8{==68rox0qTufc_`t&isN0=Top;cOL{0oBg$v-?!aIIr{*hAbXX%7n z{S;BCUnGfNLBSyrE{gM8!<;f0Skl!e3op)I_L4vuZhwdFw1T^Mv-a_TXc#Q?Tlg>! z)#<5w7^=Tc%BM^67F6w*L~H%vD8Rjdx!svCf@a4BlAS0e!YzyHMxEzjUm_mHM1+OM&HYJn?(>&x9}{?B6q<76fM~9+c1Jy*Rp^(ccn~eah{xTJ9M(N4fRe3iK4Ei{~+R zj1Oeob{IJQjZ^+C@J{p|;HtK>6xoaAAC8q<)Qkboc2`Jc0xpwcVUsosBy1;zS7}a7 z1hRP~aobg|t^dw*kVAH|^2OP~$$)a{C)eS3vgU{vU-+wrd&$?%>4*Lv+aP;<)}BmA z+=lA%d-Om_`1q1od*=NMu+`yTgE#|ufxoZ*&*7XxYRdl&E-*;J0k|F%#vR*^u!c~! zDgH3xyuMtFQO~gZsOO@$&yIwx#p6dAeP1U-qJVm?8 z-%`WtDd)zK(y0stR;sKn#g(sLoV2!fm@Xi^G3Kc0)|aY$lo*$Xg@RdD?J9czltUyy zYD8z-3fXwcnx_aFj2-!b`CbOC&wDuwDu`HyM!W>Z)>^vAC(y|gJPvSYYROD9+KA`Z|eDD z+m06e;`G=9&q1Cycslk#Ng$VFx7V85uzs5~2VEQ!P;XvmX3$*Xwc)OjHmEte6$g%V zz?TJiZ5!8>rroC;;?v`ParX6eC5a>Jbw*O8EI;L>>Y7r*-@Fh+0FmUpN5cbrh|g#J z0Y8GSR8xOjKVp#IYaN%2*6vaC_fWqtsn!<#{X{p)b0T3=Rd}3~sw&EBT6_4f_v*=1 zh@#FD9H@1p(J)-)G1Z+|ijiL6WY(iWCw%-KC49=U$+Yq6u$W7Wms_$`hv2`hXqV-c z9~3PaWaHuGrIX`tc{<;EmcH~r)v!vj3j-T?{9KGt=2`Uudd=0@#?Sr*=A+%EUH0Xv zxvR=W1N+of3GIp-wTU4-w`*2Es;AYiU;lV3bUciL3(J!_FfYrzCT9Qb9OfrEA^<+i zQ4_-&??PxX1|Ag;20mR<;ywk-ih+s2Oyea+_fc9MD0+ z&B8iu`jj_!NlL$-^$I;1)>Vi}v#6S?*E{7-G60EC3;bYoJbk*C0hb#@=i~c*7=kFZ zu3Oal!o1fn)za>o(t9myRfINQ9?$V|8`wBU98r72ru*8rb zKol$TO@{dNzwEqS77G-fN_OypHtqbEpiP(`QJD)guQ%+Fy0b{~T*6%U{xh>iwYLJ2 z1v6`hG+WN7$g`7pYp%^Dmn`y*pLZPdNQ$Nk6_L&k1T9Tv_t^>YVb9jaQk^x$IDj`O zzQ-*X47zDAJ9R!y;#|+TjhS{irn+q?o`G@e_ZQ<)R#iI@Ptx2Cr|I?qzTYKhL zd&$m2_k|~lXBMIV@q$;UeF3>+24yJ4*C*}Lx31ksxo;!J%e^9fkv?-e)?#?4wHrk` z#ZBik$gWVlt3DNCum9y0??4d=snZfd7|J-nB|3f_!VTBD0~bDDZ~H^+FyK5DUR3_n zJQhnc9$^~1F@KfX4juy=hD0}?>Hr(j^$fAJwbK?fQa}almICJ;QJL{6y{ea@#dUKY zZq+`&rWCOHEarbaYcsxMAbE7+NDo2&wgi~(fx8u+&#}y`f68$Wt8y~qcNa^ z(=VfZpZnwz7KZ|QnSMyy-1RQO`FMhcRBrSk10SrQ$Q%gVcdXx4AmcmO9#XV!T{xL= z;T$Xu2ynZ=kUj}ShO7K3)|!oNH|W~> z&7d>9NJCDiJFaYc9qq=WZ+b@Ejk8Al!t5J#;@98+A%a+KD`ArL+4T_JM=ay*ZVi~S>!W!+M7Tl0m0x&K(1C7Y@SlfL4t;zH= z#YfTJr8r8<&A`{&7f{TCB}m4_)o~LqyLx(Q%sCn%!S!k1`uJz9EJ=r8r={=t&T6Ml z$#EUQ|8}S*G#|+TK8jh>ZB!ZO^JvrI3(x+n2j*1#wgrAyq4$k9Ygw_zW7;!xb5Cju z^wZEE8ppe>O&6L>w$?(<#z!5*zs@p$Hg5XT^`sx9niIhWw-t!=02Ka6Tiamb>%GV= z3wr+r401l%v^QDp!=jW9&33PjEa?DyuC+$C2(_EzWDY>Q54Q`WrT2RTkN7PeOCN(<67jBD;t6<;NF>#$0kF8m|F<#~7=V@Po?a?A~j;P0J;A_1#ny{C3)k4R8t)Z076%+xq4&38XaJ9czA)Kkd(a7*}A_ zAhu~@1@$$^iN08*Xe#*z+}6M2e}UR}sBCp(51@d}qYhIewDTF4o8&@gVel3zxW$4- z{jkb7_`E2g#Sjt^XcY}!#Wt}Yn9aIEk ztHNaaU5AMk(H!;5?+;xP!O13Y!yjwE0TAja(ZY`OjqVChTY>@^{D$Q}G6gwId3dhJ zc7@%jkPMChqgTGOJ<-v&5Ap>X;V!#+-q<)+n_qf@wB$i-URWVmE@LUpync&znJn@i zAQZp`#uN;(0M1n$g|l~6wlw-HCimw^A@f?x&lVPGBWx)CL8(-{uQbOi^B z5r0u^uOBA z$d8jx)7~_Z$K4O+k2x9s1-7%v$UgdD5^#CAmymHWtfRn0a8_*C9}oc!HZW!Z&m9t9 z10{YoEk}&`V+yJXD}|&6hjXXjgzf>rhOj+;ok#YAZ?uv7rfI2A_^(XBg6#zs05Icn z_-t%dNixBQFVJk|uaBl2t5EzUAM93)neCUqr~owYI-KEZQ04l92jiX-1oe$Lugb`P z2nuF{etWV1Ay-9K*nWL>w&32oxeJBYRgTvYIi~;3^%kH41#giiD9xt#dOqhEaBi%3RX2C39<^wRdi_3(LBB=1^7dq{ zL!Ie$bBo+w8q!yBr2X@~smiV4^fvD*z@EzUm-E|IvFyE@e1H`T*1X?0wu1*ls0%j! z`t&aQ@sGE*XwKmtJ36%OV)WZbN>;MhLO5r7M&zOpNbJB3On0i9u1(d`_cR``ye0r$J;zfZko4gu*ER(lKHUC{q1FWD zA%R^02$PkT|BFyeIH)~zVjc-Ha-pgl?sO7A*#@hXar*edoT!>YECd1$Ac5ccphx4% zR%t);X^3?c%Vnz)KUqpG04={3-z6}=6|h{nZCJ(g_B@mZ z^Rns;Tn;Q9f~EuS3a-VI(w=~VDa}u3qtJ5OA*ll2weW#T!R^ViJ{PH*mHs@ATxAe4g#{@iFNE=wqf_(Qp>S)K1N^)pckXfBFAtY zV+Z>+{|AVsOUfNB3erTX8n{2qK@6jm*Xg;{V@%-Q0=c#Yn+sAIqs~-#NjCYlg|f7|C@Ku|$);hMG0D-!UM?%OJ+zI8HbZi{l=)wx>mB!`Sb-tq?O!e?4lkoA8~mOEFiHKb7FcNw zh^qge0r79Q!Av6+pse)|J@Xgjn*85^+#5eK*Adodu1)z5H)Me>WIL=$_iD6o%Ck$% zeBUDcwe_&t6J&;efa+pbEVM`i~d!~ zil$kg`6l&qFnh@_-6*P42chNM=U32rh3^n|^5=6n5xDnc5ODxlmla_JQ2-YNv$t>iH88ZZD_%Y-?EuV-S4VaGd+1LwCL#6TtMv zseNk+%!m|GsUW$z6cWd>w;2AMwz1n?OQ%zH8Vqim1B%xU(l##8n!yTVVYiE#)%4?O zjQ=8sgi)ovbqrhm>8SBT>s9UEZsEpVywe4OfHiS;A{VL~h?u8^ncJ$Vl7k&#bD0GcGZap8H3z*MbrHr8c*Q-jDki~5ySlC!!yyiA3 zAID$ee>sn#K&r{Uvc~V$+iO}k@&{D(D7u;_$g}<3L$Acc{VK;M|8*hl>jy4OE!V4b8I`%y7}{e>LlWzE$cr z<$jaZ!~daDA8Q+6SRJhKuY2yVa7(NvqTe<+7*M=F?5j}XaE`A*r+xY#hfF)Y&%C_G z`vWu<8ksGXQSF2`dr_CX;%$<13@e1D^J&iY&_o_37fH4Uf8T5Jr^3p{Q?lJ z+^yZX`*cvZV(g0!`=8A8cj6$(*}dwxH{l}S_gzgx65-xITkAHEE}(t-kMlsJkd8ti z-o=Ymv;I>zMLd-n_tTJ{&u)%) zxx+|j*-3!0r$jVtx50Q zphSR{`+-3LF|M$~EYs6pROwhmMLnw@_y}*6~hoC%{X^; zR7KYQXh~u&N=#;4s70&UPHf zdEQo)F`ndQU%Du(?4SJE@zFB7BKaRF@4upcGTCs@KK1;Bm!9y6H`2`eiI*mKiStN) z-$reVafsf4YdGu(|MG`iWjNY#hw4Wak*XXaz+ApO0fRq+)w2Vys;R2SJ`_X`*jM> zdw1gn%<4V$y2?|-33BRh6#j=4#*$OewEDZ+-mE_cUUuJ+Ck);SJ!<2^mtG94A}@`Ayq`S?mMO$E*Sz)@9WyiHszLu!9xF%ZJO9)f1- zb@b!^s003;cD>b|kXvvCj=M2z;Je(+#vS zLyk|k0uToyouVEQ(=aDB>g21@9y2}WWtidRhx?M3Ii#NDtO!#HJbgM4?}^E=o+z8= zuVPWWP|c;>@+rZibHvcDrF;4N@^9I$Cp&;i0cb>^a(GLiqH7uuD`-iijwMkyg%DIBM#`vUs;_f`dlYp zvVonUYsfhV5f%}C=>bo2D?CjsY>3BD~njtWMjNH=xa+#e54j5?P?!?$_!0ES_2fx{r0&11VVw-b`? zG7Fm-1{9o~phdO5@}1vP4mrq?;s_u*3=_{by_-MZd7e@{w{VVxG(|XpYMcdYOxYgXVFEbFldU$O5QV=Dlb{7|T!xfkrbiQ%PP}$j=p-JX`PmWE$y=DFMJ^Ziq z)`yRdBnuoT=atGL__E2Z>RFo1bJdvZTSZURUG;C1%HECiY}?kIs|f&8?GT{qu_AA2V9!0rjR_wXUgkK2_C{--i_jy;*{jKWzy;kCFhUsMI7E!s#T;+YFbjW9wsQUCCO3vJ zm*x!)i=}dKXkRTaZX(bWP649Xi9E2@WP@k@ty>u%m)!ZjTaGYQy_!>dZtAz`-dR%X zCG8_?+x<@R-D;VqF!QLaB<7LJnBf<=@$H?vT2IzPqYMk)AfBbmJw;V+E!i<|Db`FC z_uVy2kNpnx`A8-(FO#43}Bvmi5bSKIKYnewE#DNF?kwpAnjO zw=>V~IZajT*|lzZwS-0`=~Hk27H+=Re6lROJP>dkRSCd|>k5*XHNseC%pUZk{RaCJ zk1uxi0Y}+*W~AFJnd+Xj=VpoO_}T0kAz(LL{zP$Xw!%BJK({8~ zKxXKlsRz>Q6cM8>A{qyiAzR#82kOs>9|Cyc!(CbQvFojsLjT<-lwz`!kdFl!ej4}# z4BD57!^~PtJLu4m({ai?9t+YoMjiYdgK`3m9!`Kz9gr8|xFp);VrvOh`O_BnL?A1U zgtXOt6-tIi{G~FC(mh_sZ=j%uObi|~Hh*?rfkNEhJ-2~8doZ|9Ut*!T$7Rs{tBNXd_ z{w9@*Y;W!ujSYzmHxI4EzzE4g^w$a*B-unh%`%nlPMu2BmyE0z!HN6RXos@*;_I)3 z>g2+}#D4FlTv90cFmD&H%uKNJ)gU;babX^UPWMK6LU*Dp3VnYbon03H!!7NNt;R5U zHs}46J>$BWDAjmfA>;^%7B2e%4FxYM&>>?;_wn za?>SZ*Qi%lSKZSH*#W5Ba5Ia67_U9J3j{XIWZ7Wy&mAuCtsm2hbO6IwM655oFNemU zv%w_K$UE?ofE*Z4=A&uJ!OpH0&{F&RLd$~-ip@~NtXSwB#WYSyw*-8^?=NtfL-gDA z%C&>{^*%Nh`c7-LkM*OU#G76U(H$_&1*1a6Ub2F8uP^ zkAACntEzxj)!h&gbAU}-U}-T7E);(?s14Q>Qf)ifl+GgE$&F>y6D6I^a(zI6R)Xfp zNn~}Ysz8I^J;6^J-~DPrPW!>qb7fy7UjgmC?=_ zIF&bSKw3}7CfO;6wx1C%VkHIXAGdOT(r(oWo1-P+Vfv#y7;~R@CQfFzkiUzTjGj`P>*hG#xOd2EokdFY zhfT=|d}|N z{($!61zwmzeYIKs5b2Wv^TrKupz(XFvwe%R+hmlq-&Gi&-BytO!?tPWt+1W3?Oh9C z_8X73#RoIzVTh(;Rj+LVX=dj&S}LB^tjB7Int`EoVSn@0xjC2j?jB4erBU~z=d$GT z?dD|X*a22|o3+a`74YMK)yPa*Is@KRT~tTqL<*+a5r~e=A@`@n2JspH2OI2y->l{* zIc349ftzo})}ke3=sY3OC5y*F*8-?^@j#au&(x&>5t+SrGwn}4%O0-2*_d!G=FIRt zYNjpW(S6lNmmtZUmo8&nEM+)B+G&itx?F+oZ7e=ew3DaP1^tD|Z0qWh%+1ZS*bO%% zo~L`bb!i2P**eF$xor<`@@c3O=>~oLb?R$_mc&Y%*|Ob23JQ|8%Ig%96wV~C0J)02 z329G!U0i_ZZG|MOp1_J}xzh`%pN9Y~zp>~{0zzOO9jVKeAl0kp@4;pZiy$)}wYN#> zTa^7KBbLC^NKdBd9z1?(IY_ew&g)G^2| z1|JM|^A9@koSkEgsOu<^-u$6l6{_!lC<*13<=B_@SheM9YkMgl|IuVsj#BS}c5Hu; zT)s4EGUr?=-QdMtLiAr8{UX$Kd#sS*Sv~IeD<5nXQh)MO-qSG$DjNCYUs_}6m&Pe~ zLdd`2GO4U(EODexC#PIzWf<-}T`qUT6Of2TkemMnJWKwCITj|%-Bi|WY0>_BWD!>- z!t~X=A)jCU%_T$?6r!aAZ8vHnHoiXim4T+opp&?8QjKX7(+;y?G+#Up0Lal|EA=jK zBqiLmrhMuizZqwc*!RWl=$CblUkljZOW^5vk*Ac;v>w+>Y?j8Yh4ThS1XN*@4)(a` z|4A}{v1;@i;?Bc2HX4AGeQG(`V(XrO255B%hA{J*o3mV-s0FhCf|yF_vE%^6fIBq^ zI(6|z%lITE6>b{)enOj#su1U^Niyei^5pir)$af?5t4UgHOke+Lu*Sy{sMHFy%nUu z=EK!*W%RfA_0Xu9r4k}_NMOu+WmU?(zB;tA@Lu_J@NvtUOl<8ci<4?kIw7sKC797k zQg2b(uUozcd0RpgaN+t<^q zpV!FK7L$nKI?M@cS*$N8ePVh+-qWpvxonqNAL=7^rn1$JK4#BB>E`CL+t%4j^s=Uo zyX1*QvB%igAQ7Xh%jfu*`T_FMetRcwsCLx#ISNiA#fct1^@Hu3z z6zMN8FFU`3V6!MXQky`tD<0#4K&$52G?r6FKi)hBd)f7p-gR90H24tQzq4pRD=umQ zQefp=S>Un7fsbxfanhBEIUxOd9)ZV7%jEP^ zPtVmm2+|xIp}AD&OGx+DujOR;c@XAfc15m4ugjS-_l`u^8r`OYg1<8Q;#bvx}_milF#W+)#dwgpgl1OE@AlSt$*TJ4&UZ+Y!-|A#J<1h-Q-ljAuuW+yLiQ8EiZFg@s_8+NJgx* zH@IJU5b^sd&{u_0jOnW6_NwuZ0g~=#3bp3*#xQ)cMcfcB=bhIL!;+Up z{d>P3P5@;&zxx0)xgW9NLDdx++J$5b{MWBvV=ah8?IxN%Z8qh;di2w~Y7$_G;A>rQHb7 z#qzp-^ZwoL{I-hIvzo&@ue=Nl@J$s?H4@x6XyYPsxcP%4Ck_JjU=b0KIq#?Wl}RnB zd4d$EF5bEoA9RbHltk%DXRkssTGTS*NTXQ;h_8cED(6Cuqsu~`4M z_p~2!uR%)}yyd=GdpKU|K3;<~sjupY;JM3y{ z;r1HMw5amy8&E^y1Rh#g9u_=LHvSwp`vt_E@oO(|IWU@CwlrVMm*B_6td+?adGZ1b zzzj!K?X`OS2+5UID@yADxf=d!WC_Mu)v--3wzMP)rl;K>pTV#L&3mQq_O2L&ZiEcT-2B^u=6;c30Bgq5rV@yHx zCfPMabcRKGU(?ce#}n2rqprzEP^Cm7Ho1xcRQcW6Dj~{qKV{35wyJci=(ZrZaMp7G z(jm&h2wGsJ!r$~fOmR2&*@@ohOm%_(S{!2v>8O?3ZpxM<2ZiQfI+@W!G^QO)Q_l5H z)%IiSDNtD!$A+EGYzxzg7xkeXd9Yc2THE}bitk-?BpVKMZT?_0-y#G+AD&1))2FPt|UZ<_pMKajEtgGXxtPjkQ#F?RN5K@<= z;@E>d8dfLw0tdO!`s>EB+UuST@)Hm1$+88d4RV#}_Nu{;bF*eG8GmV<(R$oTX*PlA zdqhXR?erAF4VE4krHfXcCJ#^H2qkAVhB5w5+PqaO9yj~wBu!I(F4WL=8=x7x5a%L8U-mfew4#2Z|rQsLx_p77gy?&HCIW6uS$Z(_Zgm2)K z&R@RBa@Wk2&BLa{)4@ztCh89h>lN4!iyTo>e#BURPHh@e1!pRd^`j6_7JQfKj=*O= z3e{()o)G85hlwh zs5_s5`+2LQdRR{g>eqLLba+Q;>ZC~CjG-#Exp;L8&?jZvek8vnsbz0;d&yYl^T>^Y z;u-!QoIbeYZ!jSx*nypntYzzm5GsiIP1V@mgGuk%*SA-GC<=;Q-S8W6qFLUsHBf>8 zNvRM|lHM#rwxLiFnP2*X^x3X(w)|z+$`kEV*j5iUgJ14f4l20u+PuC} zU+;^Gr>6i-x?++Dqkg`5@76{KZCKgZvk#76I$!fntL&+q8AWP?m4d7)Q$O@6_RUuY zD8KepUU#~Pj(MBWoUYPHAUqeJW;ZOTy3qu{@?qedO-1U^n!y8~yId-DMR0Rh3)qLN z1)taCiZaJaS*7db9csA8I)^apz6$Qor!EwrpMGqCgDd*;B9vvhR^GUZuF zm}#UFUa8e34ijV3nr~)Oj=l%Qvym!%z03om3Y~|EWVFsNS{N1`yY-BS991!b+AFVD ze82rPi7~sO)o_hYs#2?O!jl2ij{90sQII4g-1xVzv=TRy^ELNb;T~JPSt$-zt+S> zZ(~O^qBtKdU7N&Uc%En;)uJ^^uRFlR4LZUrt%42TEgf}isl?VPaW5iAsN>QSk6a!{ zbN1t|cUwkH?H@l`DsxBDq&j+}j0FuITGQ?1YDwmeHfNVNOQPQ2$6_vBgMXVfYYbCL zPuwc`V*wFz+rYYf=?IV>0V5Oackwmv-@8lw1N;=udmq1}i6IxT@dj8*{AUDWWo8Tpk%GDRQ-jhR}KBgXcFP(4z z0gU3W0LJ zJ`oRFEk2xfd>9PkUV4UEnMIlZbi2ZcDlTp`piAPb+xdOnAl4Eb*-<1_Zm>E_zgm@u zDKH_2Gs% zkro$aW1?#2*PoT_dXLd^_{AGp0RO!H*{U|~J0n1yRn?ql7E1I|_ZhQyK;6ZOzpNdM zo7nkL-$An4L|K`{~0%w0rfq8AVnWJp=%#zIElL10N1M23nk8UU)Lf? z1#@`+5lyprZ;i64kN@Md_&iT;tNZ5&*uhCL1O@H~%sgP@sH@ z3W&Zq%>rQ}tngqsnl?m@)S-j!~ z*Rxt*7-%W+?DphU!R2AvxH$W;LQqntK)|@$0Ki8e-^IYObk4S7n%T$HYU3I!Moqhk zW!APSUO9!g%pBr4W!q{)BO%3ok!GE|m!CjHsn5pHU&78DI^6#J&{IyoGDe{j(U?|L zyzn*hPx-(-te$NZbppyJXD{8o|632EYA6c^vVMwjA1BPZ{T5V4eY6iK6!{usVB2i| zKy&2#;kTxvhJxtpSMh9cIYl+w&f_VAcLCw5WamK}1PXZg8Kf;t$DZ}V?h&OD?97eS zK>7x}PVm}$hbd=1B?W1v!qt4J;i^v^zZ|aps&Z!ahe|vPfEU;(n^B!wgbL}+DutAWi+;S~Lc(z%UlixmYEpyuo7L?;b8XXwz0b zzJf8LYE;_5qM@g;CD(-uOtG%SW^#(L>tzE9wcX_vLP0l(1P2w>2V|FR{};Iu8UpAO;WKX#$A*ow$MS12{*Axaam zhFCYvIJSqp>;>z{9Z{2Iqk47-+{e_3R`R%dxbz+2h-GSJmu>b))p$qzu7&h-+kKPR z=Ki~iJdK?!jbR2mXIPg2w(6FItGQ!t=|;5WyD9Lj0_ronaGDQ{US{|gw!ala6$Nf6 zLZlG;*FWD$H|K>_Leda1Ekq++F zSXzCgKO3K|9JPb`-Y4g6pl&4eefhb7_#`=%Sujd};u-q;z|$L1jfjV{CO6_bA1Zt9 zzD;cgO<(-SI)L%bb4hRAxqCktbPy^B3XNX50~kInt>?=%RCUkiie$qHLk3Cre0X26 zlf!advBbAws$O3MPhzEfXR+fMp9dMn)lW2$<6$n(uqX_E0Oajci0e*l2Z?0A8XH{$F!F3Y`L+9q*^^S&P?+fS| z!&1STm$W79+EudbPLd9`V&)Ln?gq&$15H>qmKP&AA6EUlk`W4behN|_5?hn$DTs?3 zWAe5)GT6g4pcoY)%v!1D>%hwUimDUc#@LOgw%4qGCxu%jEWC0$sO87WLFmHlMQy8Q zb8cVYJgkn;sfr3>5t*A9L||XQpTCdRy_AsiF*dHelJ>Op8)U||`j}=gUE#kkU1=WU zwy7T>gw-o-HjJ%muIX&={(ywe^&rG;jOG{D2`l2wFQk%;PnPtuXAMge5s1vH*4dOc z$09NV%^%tEr#MzjIq=DgF?Lwyn0b}Knx-`N8lKY^SFStTFC8F`d$HQ&beq?bo5PgO z$+y@lK`J}b$+QWVg2@)t_WT1Vv0>YLc}a~N3z<2jVe2(HkxEmrA~^r)P1O!^vC`|* ztzuIr6swxtprtV2WzYfU-uy6?%(7!Wp+4h)Pu~n8y-}u2g+R{C?ybyTZ=L89omrcj z73y@j{nE*8V#NL8avscL>~y~(vTx0%QB&$sXZ0$~t@W~sNV48N$_BB{SPmohf;y5& zR_z}UKg|zU*qkbLcuSM=A4EKVG&e|r_Shl*dT6!N z@%Rd)@Lf)Nm)kK0`@3{Y`_xsc>$*+abyTcWae$qFZ!>YKIn>q8RUz)-|FR2P@}Vx(j*kl}}0r7^?R8c{!a18U8 zT$kE|YJ4N6bl)iJ3`k#p{whU>u&wm2vkn9Eq3wfX2|wwe#_OMXbDD}HF2EaL!nD(8 ziHAGXnEN%6Fyp>w_Ky0K5}WWpZdHz#sO7jon?+`*c0aEu;SHF zE@I?ANOtyI{djJnSnK_DW+kW3mD6NYpjsXf=!lF2^qSMpN0F*I_Hfml%UH2;e789J z8L*px8Jo;)@VjLLqp?cUoq9jos@>HwMB^qL3n=XDY@gN~KS;i zEniQqJhpPg;j#F|dr~V!x3!p^+?HW084EAD{+T|G?BAwzpfn?9Tn;VHuhRNM4cYM$nDDs zs%r5f;=a2HWrDV-8lFjqt;V*r1D2uS^9&UTy7VCKPR`a5#)(o@%?S3(@DjsiIc2fF z6@^Pc%so0Cq4xXZoH$$6%7I2ZFit!sV~{5B-zj(`d9a3w5Dqy0ojtod9@=j)@RJ?V z!`Fea^_0cN4T;$gfZ-u9sQ?rLPr{#411OiqA78<(xj&XTlSDt`=dfa+$yJRL+UJW@ zgzOmlN3z5(mO8{&FvKN2K$u5WBh@dwO}>SW6Z2SWTqh2bQd=~ z=mmVxE@a15L!md6*^C{3xHXiUKOK2Fh5K{PmW~dY#>c^aGP`EA6epq*W_7DIkfiHE z_7L$);cCG?9h;oY3Dh+6M$i&&s{iKQOV@R!&?vV*Sv557{04;?Ewwz!;%EKu%yxus znE%_~&{zEGlIP?vFmf^-A4?9tiday>oSU-m5&sS58V6Ge*&oGPQ|rI38!`Gg?Hzs{ z@V_IG4U~0@I0UrRNXNYPz^z3dXp0cT*35W;sYH?GCO^l?YImo^XZ`o)o?(JsLNtt? z41CQoa9!STXjrH29x@F1{a1;Y-=uF_aV?_QjOyTz`@o=)0rybr>0*wb5-iN^(aag~ zFnsjF@{vHDg347fBEZVVwH6sYC#!;>Er@w5>@Y((t@Oo07Xr41ZXseODuhI!n}vo+ zsu^{Q{2P+bkx~A6Us<98quZz^4JGFUEI;5?Q$^t8Pq;L5Ag616PB0Q=VFmUU&J#{gXI@Bj}i6K30VfDrJP48RNbG zWDCyk_hi*20pFMhy78}=AW}kwG6!P~)=e#J6BYVD8FGxs@s0rbui>A4u+PfZt4|?H zN-fsz=(lmQ#K%R52SCBU;MjjV=`En%6WFg}W~>`tZ_vG_-z|i-;9A|&b#3n&vHg>rouD^pAk}m z&V_pOr(;(&_r(oQQ51Ku4sYl|Zpt9gHYI6GU3OwoFrC~j$LD$4Z{QnG`btgm^k%T< zc{p_nU%NR^8LRNJ1tAmGC`&hLa-8tGG*Pm#+EAJ?<&Oz%BhXiUX&`pUKCRRUX+;PV z*pQS{MLx6}uI1@9&Th#h{6T-uJo7(XuzZ<4af;`7;Rla>nYnzs#*pfgiLEN->z&V9 zq5L~3%rOZtY7Ey05{>4;tzh=T%30|Q^cLh$o_^Ct>3w{Nt+M#7#xJ^b_+QH? z4!>MK^h-z6sC-!jrbWkYnEf(z>spgz*9*nc>cN?rFBT>7N;F*8{6Hg-R({cLZj2^y z(K_c)3R7A{=`ZKey4opGzvSb%vm+f>=P|0IrMt8JHRF=Fce!T;Qz4#9_z*MYu{PR- zoehP`Fi!gY_>6oKg+ry+@x(9r^-2$EGSG6_4)h=z*eR zq3}mSAt0MCY4olVXcdBTp?4w8LY(xNG`4&KKQEuGM|ba0VBQb zkd~pww+q`lXswEePwsWBbtIV!?ThjME52u0ss9@9(@1Q$=}k*)#O(@Xo54uW)iO$% z)u1RN5hs^-)3^5HvnZL10F*0W?QKz0wR51MHNlFDV0_dZisKKr=olHjTOcdf$vN&3 zZMc^=12Yp}gB(r?<0hY!NTjY4-VwUk7luBDVQ=?kKe%N(1!*q>hol+ocR}imI?d{3}jG_juNyVo8#s5P~ZT)Hz9^2edUI? zqM%Fi?M)vwR6zGq`GYt#@GX?Q$VdYE$~>q~L&fKiq5XaUAx3KIai`LcQD zGpg#{tF2!dqOTn(ZwyPWwf@w#QHQ@DE$OGGaJW3p-@N)yIJm)%-i`vby*H-Vqdsa+ji}*!u**pciVX7FzY_QZ@dZuGd zi%KxJb!My1=*vzVwM{I)&s0Ii(lXNYplObQJ>;g5608FdJnGpdDW@?FB`7$?e$yQ+ zk&R{78%?Uh5VH~_OrKbkZGTJwda2(T+Ja9^P)zzCZ?5Z+r)od`xc^&pyt=}xf&7re zYk~(?b-TqIN6fz_EqRRUyf3S|+9VM4xT^r-mgXrJr8y}3QnS8Pc|l3a9IegLtShwP zkH5E1OklNKEHSnj`%=*AwH)EiG}Ic3x;6T(H7x@?nxuN&_AXhKgLgbm?6o6}A$efw zd9@GNG00>R8KXf_H7sZKs>wQ0X{F>jPnf-4KSavBFg6fx;C~PMAi=#*)P!VES%uN? zU03VdP%*9<{gMM8F&Ow^;y(qq-S(4meI zjYW*54ivSfEN}WaklpVC`5KtMAl=K4#eG!gv`sYfZKZl>F~>z*ElkUf+_hd;YSi4t zR9ku+n)%u_f9T(JC{|gVAD%DpaHy3C(-vi)?!_s*cB-b(9_y*mcD=<%fTxQ?C+ujJ z`PNyc+Dwn3YsZ&DdF-RHjd1#>r!Ni4B|58_DD^;4)LQ zu483SqC8XL1%{_(Zl#%KMm7~c=UHO9fw=a*7|ywz(78ro$n~b8Pwqu!U+V0YJ1*Vx zT3fPkQGj#Eb7$dbY+gIF*XvRA>;H}TaRbNqO%7Y(P%{XT4d6v74P3-d_9@?p?zgfe zo+pxZGk%3kQL#F!P0;Z#xogLvIYOg|?)V?mQ07~e-?~_yW|*^f@-B1#XH-=nXp31* z7Xrh(Afkrs2=A@7wgGAka=_H3(bKLZo;M51xb=kadc#sJwzhTIRQV&h?qL-^`@PBz z|Fg&<+|@g24A2aTos#W}qBDc;HDqvvE{~KSWak)1UliWrM;EBw2?J^!C>pWb-=}!e z=K|GRuTd($0grSET?)YxN?*H=zm;6#t5Y-4k*r5|U#wo#|9nSQ)%d7Bd+>N%oo2Sf z1F#GcFozN2srbpa&q|rQnz*g3A86SR32xmr;MiP9ZBNLlkN zFfI;W-W5Q~LoQqLu@}SXYcTc0t!fU3I0r<%5WA@Xd)deTd;z}EJza#O{M0Y_*Oi}2 z*;1(K*KDRA@4K@z(7T_fe#CRgE3YZ}Ay=8;HO+!l7t<+>@nulxi**};l3PHD?ob%` zu~QxRM)iTYMKnBQb^aERST;xj^VYqICcF)U_HKV{EHO%l!8l{Agw^SlE2H^|b`G!_ zYDe0_vV|a~9wF%NXETj1s=T`sECvWI!j!VwmJ7<+2!SB@qYlH*PJrAyQMYi+dmn%W z>}l7~4||)$9IfTd&r4DR&r9pBOCQxjt=7lTLyuzCjec4>RLq7-Sm#nZ-7Ik}$dt<- zMRxx;cNw|gI_kvb?o?ZFB3G5}vHJ@y7iep(aB4tU?6j>?SuEZ~helqrGkX7{kD7g~ z-aH1W{5%c%itf$XF5=9DRk(8ZwKVy}qo+AHkv6>Yx?SdwKCwKx!vDH&rNjLilXjJb z_I>d`+BgI3!4)*#0m<$VmwQ29LP$<^aX~xa_uV+7LjsU0`3YJ|9=xu{RtqdAW6bjl z$>8lZ)f$3;{p;yGp0KF5{QrDjFS2g4!a6!q^-(GKQ7Du8c zKs5AMqvLbUR5~2Z2ovN_|C#2QEqDpJj1Pe?t+Ec^(A4Ha(2|3W(wpxPBS#rx(%lSJ z;|L2fM2TBwC+L7vBiFRy@k$N&*cWGj2%Hzkw#H4>DB>O1;=Arn9bvL5H;$7Ly;oVB(EbWs@t-@|+yyPVKiv+#&s!}$t zE>j?~HRH~jCx8k*rg-vLQUCe9S$Qbu(Cfg*0Z!H|E=9aJju#qP85|{MrnJR^GfX0? zrZ->#9g&GDsnbYL_?k~!&J)~e0VIa@UumGNd=IQf(BoSthD1f+gnuDd8(bw2E!$r! zuJdYW;qFuP)fMfUDvH=NE1K+4^G*#h7BN2$dU}rsM&Oj-0)Ar8b^S~t2Y|D^8#x71 zQ%=rjUX#{Ig9RcTM?a>Vn{$1(b#DdobHQ5TVn60O(;2w-zfmjIGPYjy5W8s7b?!;z z&viGXp%;Wr~ zraL{Bi1~y<)HxtOd|cU>jjtrr2BxqP*U+~Ve`egR6w9ebO1n;!ZJ_06W&fn!=51%7 z_7fVKc=)}){nGT}eg;nNTAxj9L-le+gaSjlI)vru%rB4fnNB`dUbzpB16%Zoke8lM$H+~ z(kSStN`B`D_(j=!lfWlDwk#PrzrTcH2hupb`>K!^TwD1cVx;!DEBrem zfzJ}8j2J{5m}kPZc{Of8AhsNFzVHI;-5`UlOX|em~nQ+~uXpJ9nHNw|4WRn2^XBzCT0p4=4 z*9$Q4%dFnj%WU7T9PoOoPC=1v;58-_F&icKAaw2NyNIryk49%bi317!xi^BZFf1Hq zgX?nG7vK5AbiVBspy>_Ubr|UFGH=jrfQKKwN4+-2`Uwq#wZRUiIDGnYnwZ z%04@hVp%KybWnYvlv}` zf;(cMICo2V^?@OlH2p68)Ks5a9d@}V=2;DSNEI1SpbGa|tj6+MD-O_UTvW$g@g@C-V&MiUL{>}a3S~ME zu6>a=*_T+&3Ra#4dN4b@R2aV5QUJ|$#~KBQZI8R}xIrhaC!w3&SwDl7vDwwbk61bW zjqMveOYPyN!+mha5FP3yj<(j&#pb{a~=4z2trqU7w+P<{MTql6}?lZ{yA8QXk zpo^^mU*SI~c&`$9MZK#HPx<$1LCydEE!!vqm|=KMz_m38nZK7ovHiCa;Jo|;ZPU<_ z2U;XIgP{M8njq*;;7`)*F}Dc+>jFV7!V>f#s3no~=r36az%X$Ga3{t8`wxPe=M+Fk zp~p0H%>Ulv95Xu=s97X{a4BLDw- z<^P*b(%}ust1~j%{;*6Lei4tO2;FAaxu1yTJq7rCUoq*$=m!;z2Q$)y-K6F;Y$_Kx8z*;%fD9ni zICgP(_tMHYXnF06Arm?kkwIib?3It2J>^qH=A$>x2S`-!kDJ=apB-;b4qnxAbdS9+ z&&IXl;_eCrbta%=>Fs&m+4-so{}(eKcv+brB(pvEemFV#E|a{$gQgecfkDAxAzSy- zvkXd%j~%^E2IuqKJ=EwyteKQtei8&=(BzlKMUJg*=<#ohFL>AyP;6c8j8cipytf}^ zli~YF#&+VzVFOyW;|y5EfK8kX-gu40YN5R%a%amQEnVT=h%xIZC%eFI%1UQOMjH}f zk~%Bgc%pIeIdem|7w(e*0or(aIRjqntogq`ULyk*Ka{IB@H)f5wS|Q8ODsdLQRZI( zrOZ;D0t9#CnyCq}OOsfc_?c%h_m&>y$KmEvjSo$-2fp5yB$ICDIwr<*W8wR6L~n*M z?#=TzfrfH)m=T0W>E;De1uOUUqQwe2|2eXbHidO=eb=9%%wsYh`WHcN_p1N6{T@7v zvN*Y6deoh3IU`f`c4k7k_fD##uJ49jK{_WMKxP&lcbE7ai220ZU_f0`@l>es9$~{q zo(=DJDA(Y@?ja{Z-(=?iJzV$HwmpYQGp2o!)05{&Q#~<^?fZGVel`S9`3a_fK3<~) z_PUeHbt=(|c=EZw?`?mI*+LV6iQX4Wr~`0}nJFCUts<16P+c3ot12D`&u@3W*YSeMu@lf$?5+ z9Av|4GCvi1gV4&{q-_LZfet?jCEMwoow>Ka1~12^0=2`~BnSiAgSBQ7W+<(+D?RQ(8yAloH2G zqq|ee`DMKOJjKa7Gof#vXBw8kwF-}3{Ru#Yd{CK{L~9C~sXYURlkI-BdAfKodDbF^ zB)Hi*CW5lmgV>zwZkwKretOrd_w>tK?-_{8qum78<#IAgW@Isx5moQ8nK+pnCv4*G z1goIr^p{{`lUxeta@FqxZ#Ax} zRiimgS*O&wK@Ww1TA7D{-txcu=Ln!(>oh2NtO1EW&Ei`C#|0=;%|t+g#4HCS>aD@& zYK1TTJQ?P74+(?eatVzY0j_13yO5V z15o!Cp;nc;wobiT79s(ulzk%sN8th8O}ISaMEv?s|NYBA&DAFa>OO8JhNV7j28Fc; z=Ehz(j!t@D3jt2%rF-dvafy_j@`Br;ge4!-*-;uHelweBz&xOt{_ATCKeF5Gy`?3r zH;!_&=C=>d^g+p@4w}=Hw60wz_4b`{obL8RcZU?R57$H&TB8cWckt{bZ>0q6*fhS{vC~-B`2U z$Iz|1KS5;cD1dAoafJK;Q+N7k_kncwPF@*fdE`}4X&rujuxX^G*-T2UcP8LSnWaF) z+Tn9sPPIU3!}o_XKCk;`rDKl&!eFx8e+F$o2d{YdT9`uwdxbgn-ryQqPH+@^HNM8x z{=UgZ8f5B(|2igfa?nn^EIKqE5-FpIwMH)u z{02m+7i-E#-uCkE8;S0uxkp*y8co6XKrHyp;jI@8u-W@ z2K*RlOUdCM#@Le6gZu}-1ZNrG#ISN4!%D1?9TP7k~7uAS@PV)$u3>r7(g1rvk|HvZAry1Xd=#+CeWCiIksm*K-9ae@4=EOkEng3W}A;aj`Yx2*k;V@|%2ft=}R> zZCB*D;`V@1tUr5JlUaUiH61oF1<^`zu}5^~z*cIGh=&36oh*dIRUN(N;NvyA6_>>v zw3|w^om{5xO8Y0Wvz8R}t3DfWb;kYuhLq!?94Km9*^UJ78_way+wKdPJ>9w4-vjkF zeY@Rz^69sbaQa#N45XsDX{TmTI9yk9!#*x*lMnIIiJGhCk>u<_faLG>W@6`;hrxCa z-X@s~+-3p+3ZiFTdxC=V*d9A1%h=7bYFKHlkVMgr*f;bwsigLZio4H$slr`%6K7rp zd|UFE`e6);uL$V2R2xl0H`$6=w0@`(z|Eo-4UUc~49S)J%=mh$YIf{wp|Tffof#Ca z1--(SgWgSX>^<#TYy8@hLn-`0(mpO%`O7^sCS4hQWp3Ff21Sr@z$F=H?LM;i)r6sS zz`*(kIL*sMrhxgl%3y=?%yD*dEJJA%kX_a1H)epiyA-j0g--?;-P>Zs!(8iBPfJLH z@y8>XL=QX1o^v%+bmLBEss?1xcTV9UEm^%Slr5ZNT>2Z4(Jg0WRbR7hkjyT$3cQ3Wu1xIsddkr;5)c1A^eTDkTh61~i9;a96( zFd{qI`M}~7(50l_CCRclvl5WRr2XnhHU=59V8bCN@$-3mG>jhvIxj)M(T3mPee6!K zo7`jCzW8w$QpJ~8@lrnRIDHVy{f(ssGzIXCt@PUmQ2OnqVk8Qg1}_aJB5DTpJc454 zC<0X&8E78ZSm@$lO@URvccOF267KWNzZqNqa`<^%v~Zz1rQvjZgyPag2CrxAee^US zKuN@T7f=gXimGN-p>XwAEoyi~=N8r~l-ia3GE_-OUY%wg=e|=2-JJ0}l%3p=OnL*} zC|4`oWRrh0>JD0$q;k8nM9-TbVwPQa7j#K>r+sjyV1OYH`9aij{#>wpG^xa}JYfsV z@$m1xD}P+{&T(w^$|`^Zum+c=%o?m#Rib#qAES*{%TfOv@7`7e?8l*9-dRaSztHYJ zK$BfJm8!PiT=NY$mwI&Z7Fam-8l3^@;ED=pRh^+_Ss;@SWw0$Mt63tV{PE}!ArHd|JFm2AeHhLAL|Ixzo$#{Xg z{O;;vwygzwj=mOXP?Ka05RBIljY0H1pKIhU-qc_y%QT=8l{I4I7cyqwTS9SMRlEIC z>M$}#mI<`}fo><+$2<6Dtc=vCS%f!>2(JFyG{dmF7%3G9L)j+)9~bmr8cOjBrEuVS zWpI?(=JfF~p8@Urm2dD%yh}cSxXfPrIcVPu+QB$)L&My(x?q=mdalY^^L$M!iz^e+ zJdPT#?t3)#|Frkr(Qt2V+do3oC{ZK2B!nQ2=ru$nBnSr4%Sp5dqn8XK5uzq15snZo zLbTCIgb@joqLxGqIp=xKbIy65zus@Hue|HEEX!KM%-;LA%Y9$hbzl3T zqK0rj_Sk?<53h=ZSF~cwdyo$B`nT@gjlYqnALCLj>Hhn{f0E$v*^qiLHj3} z%6F_V#e)}z^HXasRg@h=9}88}kx2UB@_R^jn;W`oiygYT`%u11MnJrseUPTcIcVnn zNo$?JnY9CuQSp+o7w*@iid4h7VARceeYZKiDs)xdzlG$WUP0$WJPeg{YRjv}f_oyG z;d&0P3=mwJIvRs+IATTCv2okf2ro2^(^WllIhxw5#KQDBizA}K&p~#7W_ZEcx88hN z7;kAEDXLw4%kcx=8MbsqEE1Y}AW1h(Qk##it!dRsPI*}ZqvK*uphfZS3+%QjHL)i> zi*&>r$eWDDJ{2=W^IGrsrzwxIzdy|4jO|{(bXVfX~ z&u3N6o*4q^*$20O^}aJL*r=z>Ez+Mr#>v}cjQMXxvgfru4|o4Oy`wU7;eOhXoL8bG zmG;g8{7ci5XLo4vOAOF@+}`J^uF{aqdSxfjV;#LC*kw+b+L7wZ=Yr|I@CD zxQH%0@*v&{GI`ok?Tg9?U5=Q4w-6urn9b7857F@=N_1j%ePUOz5a-DWPG^%jVYiy~ zLDWHjUHkeZaA5k>RUwws627Dc1RVK;$E|D3Q9t;xu$ZvbM%ikqu@GG4t$ge$i8~do z+)nMSeb%{t*`Ce!-B^m|>mI)ELah2Zv9z@wBrwT*kl^Pa%zykn)+y(_U(e>9U>emg zge1v3Ai(ZqBhLg`_aqm0NF_LkEc?0%S$F@!A#>-}q)9*oj3W{K`P~oq^vG^WzfiBu z!sF1T5BRJne%^>pn#kaj7h&(TZcj~_e6-J5T@PElMm6=kLOLwrB#4JC_y+lk1lEFI zS>F2M-b0F38uK^2g1AD+e`50f%eN5x8vV*vp>638Iq?Q;L5u+Y;8?PGJqM3p5J(>9ep)^#- zgVKKBKK;V;)pQmTS0zBL2{di_A?V+eFmP_ts|Grw2D9Rui*YMs*kWv*%*r(?ZR#r@PG>V7349&Z^pH`ie2AN@Nl8|5 z{Oemk*PJNoW}$+IK4esW675Bw&HA#Xks2T)t)K!elWoF7xFLFwE%|>AbU$93TyMg)7b)yg^@FcmQyM{+E+Q zuOA|16}?)q~n! zI;cL{S)$;l>2^^0hIw?!_mA;)_28EBb%xRs^uCyWaxo5R8Cez=O08{sBVtazYe2bV zlLpGMqn>FrXsw)+`Xu*8JMax8Jkxa{$$XJsJelvA2I+HdukKS*VNQl@od-C%%jcj3 zUJb>S^K8Xsl5L`{4mSjF6_mL5?5oAb{Ig<-G}Mu-;E*Zp0VS0nHO@o;UnW*;f-z#r z%;^y8e<$H!{%q-#(BZmJJa)ElZ;DRKc4Nyq#9iJ07$sEo+}Nc^;YF#K@pAUY*&OZ0 zLLen#>d?nm=0zbC38yNhj(Jb)sTpFA%@-R33>W}U*;PFrWxl&Lu(2tTPa`SFRoce_ zBOQ)w4B|)SMDT=jv)9lzivmG=-Mwi^*`BLfjwZpZ#}zR18lR~nJp>D{5A1H*735y$ zI)1bI*{8({Y!ydL@;MKg3xJ9cejZWC)`rfqGGFpM#^5>f=o-s4$;gX?;7K8#JSo_7 z_A<2SS~qqIS5$1DoWI0mVvVOEv1m#s4>q;r|DiwrxB0XGn^PRm1(CF- zrtWNi&10nGxb6wt8{u^Q35L53Q6I%K((L@kj}5;q$h@)Sd5Qu!-f4!TT{*fml^Q^$ zM-@(f;pe=zjC27^fdc?%a|!yBecn^~R}O3S)3PW()7I0X$_Cp2D`NVUC1;sB43-tX;TnTU^EqNk&kU=wRPxIvxn;$-Qb(KZ^ z@WAo(1}{!W-^>b6Tn;XiBHOPtF75Viy+|5<(w>2Eha*^J?^X9t0$Ad&9BFMCb!0bV zplVqa0MehxI_z9?T;;M%A@8#Cns0TVkFZM9XSt}~o?V3JZ(L9v9QQGU@4IH2amx`K)ek<;tZCqO z4R#ySZ+*Aq9<)qp%B_Crnq|XvDedaBlZKb#{(f+k%F{sb!SO9j!YCVcq%wsm@;aIR zcp<}PGt|YY>!Jk+=~H1^Q$SBK(gT;S4l1LvaZI`H`B*Mwi<`On0Z(>)`%8I z!NHSFP2YHXJwx5&np_|z<9i&*o^Fc;8PWwNt9?5lH1&ikCQvjH`AoKpTrVP zer5Z^ACMnBRUjkKQafyM7nu{*Fy`|0ci1sB>_nA`3vfJ(gF?NQ&G_!yEhOBWtBwbB z59j`Ifte;<7B$hAh_7QSJ?K~TIe1Xo!+e9mth3?#Xq`SXy<&HB1(u;tyG~T`7~5x6 ze_m?6uR{WWs4IXU8C#k0yaS5e+uLfUmmXv;#EK=eU@5U}kD_O1e(Sn;qH?~Q_pyEZ zfLV=U!sUCwJMYlZEndrI7s0{NsopTpucN?3+^t;eL02wVmUg}h3lFHBdWx?6YSxA1 zJSLN_Sf!j5jv;K}H&&T%95vS5^bf>uDn05trBMbhckth7?H&RcBJjF0-*CRt(266KY)qi|WXlulKJ{c2V$JcDYA- zYTW$tK0f172w;3|;bYymk-vB^u-y}RSkvZSrq3@|{xhyJSL0xXf@Z544Q!Sk)~e0( z;KeQ{eziC3WpbWXMjI6Z4~(pqP47m8J&qUNg<$1GkNIzu<{zYCV^~QGch&IbTdqr zJ*QC=0)c&K3&MZht@4Xqgx15}bZ_+5PQ=jq=B&)6Oq9rUREE?5Bp+&z`6BmMR1G8=5PRg(@6Za#( z(QV~YQEtfD{vK}5^d&!BAg}un(g|$Xj=oPdJGtmKM@Q@;DUK`>%4f%6T6745`u|U%6vraEj6y~O+486b&W?iP#5(|?f#(X!8_J8Aqw33Li zw2|iL9_N{z5`MjK=h)x(M(`GkgNH+=7qwN;B)@aeYom1b7`+0Pp7->S?;>_B;QP2G zXH<<{_}oD2wv%{Gi74_|C;?Z)8z)%~kGphbH+qS!aP?z`A@D$=V-Ymv&r=WDT>nt{w8@8-JO zXnR%7AhuN7z>uB-$rZ`P+r@jKcEzT$=KJGqjF0I`pQ`>USH$kP@bp~1lj~64)ljMi-biJq zuWq5hTw+W+-#h5e>T8e92=TfpF0Le)r}GFSJHvH{3ta_Y55R3)6>y18vq1suW4KPy zYnlf+zpQGZ+4tkufLbWKKZ!U5-ga(NYr8ut*FlZV+^cl(>DJGmskN;i=m6GTRI|Xt z>NUD`V+NQ1Jw6ajmn(l(X@Hgsx4jxMWjo)-zh5u$%Ay480%L6^n^FmPR=nVF#aBdj zr$fTkSnW_d=k3^@{EAvhr0UUca>=`={KvrYnlogx)NTm}IJ-j6u@M=g24*YNgf6Tm zxYPqpFm{@&P-f$4?%SF90_=9HUq;o8-}{bN7|c(>7y0Y}?~>=m-iYHzO2#Mj!QQ$b z*Iurvr_Ar~wuuQA2r+J(DR{w&IG(angQs9WN|#6Zi`w+gHC?H%|BbQpENjSF1i4=% z=bzMPHm*#N4%c2E>lpvJQa78I{D9R<<u zD*T3z2)+5;!>gV6t6!WPqZBH%W8qW`ysbpVX;*i-LF%z!VT=d!YhdT%F_~k&}7La*H8GxRHxqo0OVmwp0;k^uk5e7 z?_Mn3MlV!Lv(i#S^!d7WVZfIS*25wM8#`=%;Hkg@M0Pzx>Q(a-uT_G2E@ijT;7+| zY+iraF%F{Fh9Ed7!E+@9qt&Q^8MNzLt$9ln8O96rveIWYx@`etqphat*g zMy1LE@>@T?e@ZhB|6%6TFjg&;&heKHLhL-_9B_#S{+w!y$QCjT>f+Cpq?!UmoGx#~ zO0)ai5P9%2)Ex*{ysYyuW8D8*J22FVauGniIY2HRff0<+belmaSB}*S`gy1MD((O!LFk({#V@6{%HHbQ2>W@Ma?mVok#GZ&;=4 z0kj&NPI3jAa0?JQzR)5~ZwVXExm}0Y#EegX@TSL9zuDxQ&e9$e4a8Of3Cp}3H9fCD zOO?Iz@xI$VpeN#(pPql|ZEs!8W?jAwlq76z6a%dMa5wujR3hsJ&(eK>Cx_8Sc3yIh z@;K_IMJy^ESjzDsc;ppx}=KPz-Y?EzMK%$2)svcrzA4-B`^7y)K^JW z*Ur0Kp#(;Rr58(SCbmL zogc!5h&xh=r||?2fTKH6H*?$EdD;jrz+h%Q4evX0vdHLy?JD{HPme6)H9d)u+MWH|c1*LWefI8+d{-H&k zVgf5{Bs(her5EYziUuLXIB-j2>qT4b8R#5fiB=+yu(UUkYLK^tCjI-?kTlWT(?|J) z`02G01k_~O(Lf2Zt2esZfNHD^-eVB&Pk^aT?C{v8Ar5!7=vBTlQX%To$J` zt1QmJepPSec4JOM0Z>|p5j+KeX##;TW%Li5Hahi2Wg1qFk6h03HRu=uR17+Tm?w0M zl}|jG_Xejrg7j{oio$@AD8hymcG~XXDeY~Y4P$n-xUrqR1udcZg=0z+R`T=xX6Tqe z723LqP#cjIjkPJ=VTx>t+?J?T!;1_lsRjLv8D1KYeW-|k<=^r|B*zY?gA34hY2MqU z-QaQl1L%m91<~Z3h{s%bUZUc+SJu6irMFU`wXVq>BR92;_Q&~t@qV4Za+0|!(?}Pw z_m4~`Hz9VTRnCwOJ_&h{tBZTUm4vHzfYp6j`&kQ_%K7&eT&4I_)ZQ)SR8T{{&ZR48BB>VT_fkv3Eq5~MUhAQ2`9mH ziEtB4KT-j-{OL`GvM*1LY>j)P${DSqEocDNCV#9Jp~`rgh6r?a z#h+V0K6gF$(P%fN_)&}shs~V|z3ql4O_~>uwmW2)bqBFk79O~fw-F;`gkrVmY^rsVuBxXVY0AE~zxUbVlP~psSfDfTpd1wu2NOPsBra;|qr{+G+IOYmu z$Pl~1m@9aQ*r?O=IB~lWen`kCK<)@H;Vt^F@$x93E{=ppMz8 z;it#<-79{gYyz522T{MdRAhrO1td#H!1Z)rd)(0N=_1cEDFiDn=qF!deJ{LQ9^eX4!PyBak0#m?V6}!NQcf+sTTM4VxcPUdu*!nIfz$pei}!2<^W;C(!z-`Z3m%$BqI366C>p!oif;La)kiPbkX4<(+Ww5s~L{d0y%i(;7^BPpD1w+sHHvCTxYZ z%1Te3>vbB@3*3J!2Fw7pHkZ=4ug$uitxpfp^pN}-MSZ%m7VI4V6Tg%TB#{dAC#MeX#Wxx&BTn9Ix9QAK~Ou#QTA;i9NPF9TT-aXHHlC@Jpr z%oJeMXPYugDI#QWB+?cr2#oe%0Tue32PqmUD)T6)r}|gDhH$}p~@hq*Z?5%-q8qaW2APH3Ns?KGs;PYEj5{=BnNL*w0X#+gmBu+&5TI+bFr(y2FAWS2`Kk`(*`Cj*MB8QTl>Rv`lo& z>P`L`ITA5%7aX2T-y>G|0Ern2F*f^lAx1O>w@us?2wTEYaFa3>QZ#_!c%o|=IJ#K& z*M+#|qKD!7{2~6h519;qhdmN0>8_bvA~9ZmN#!eEgCYikX_g|l*nzC>IClwRiIxXe z^uYHgGj{Dx6it+WN&Qw69pl^@(tEFz*fgTkS&)<&F{qK>gn9DG>nx;`wAwu_W`K2` ztsf13EZ`oo!kgu{+MjF3QawW4{|&~tscFAM*0DHVVoudkM&ON!KO4Y$^0~`;%slzrx>zCtKgyr z;BTzA_wC;TZ|`LL>8d`NBk-Q*hBBV)#=zMh&;;CiCu(dy^%g7&2uJ4=7`C> z3THl83@qEJBP`&yl2As_$5uV**zo=2(hy`n$Sz5LW_CRl*vy5DfdEtjc z)-R>cY`e=(1dgeOj=g7u+zmcXaTuK=`wL*omdF?xSG7h(25JmSk-GWjvokSUjzbls z31%44IJp!Jz|pU7x-!g8q-2w66w5%qSU=9IFyEi+Q2YZU$OP4m%c1-;7JO;F)< zq~#D3w`AHJD<3)F9NDxOG4(Mfp0vF?V8n)l%YlHd+y2khRf%UHW~~tOo*=*=q$F{D z%9We14SahvHQ%8!P?Ztl9WtBbG#(W~sH&OUC zp$1qVKa47pTi;T)l+Ezb}83-)Ywo*Wsb)2 zdDv)r_66R<$rT|?Htk&&>}?p{px=^Fo6G4X8iA_?9cgx|M!-x#t~SLn<6Cz_*VB9# z@}`gp(QvVxRNFKOk^D z=mvh(JN3ai89M-Zf;)oL_6p98U3;#7blE_eLNe)PG2*Y9l!tTS&B8Jaatdjso`_eI z?;=3bkuAVJgC%N=g&jW7Lxt;?$Z|u~-#^KrS;|6KLBJ$K&I(!G$=z>rD{g97ZK^x7 zB#hADczDTX()*gNrH*rCjZ~tj!0504*4$WX?f25-4_KwY+-^^QW*70PL%|01b-K=D zO+e!=0;t>r`^xtLd-Q1Fhrh$_=2_^5XCwSD;a@{l0nFIm)Dh! zsh5?pDDtxahjSEN$H`F&$2ar*LPo(?`~F*YloFrYXK}oWgla48juSr02A=k)L$o7 z&J6?5^0~bfH^h0uXEZ$vOHZRGDe(|x@AUC$K?0ckRyQ}Sb)2s)?Ztx;e2zWJ3Tjnq zUMpW_=K_wlGKtIA{*LGZ#!ow}+rsNxs%7Uv>2Gd!N}%3r+_0W zL*QmB;aq6xJ4F?wwTk7_57kKxp6-a%--7ovSAg@ z=4aaMTb^nfyK{G>C0pbbsI^ZvaD48khd9o>AUFI6y-$e{g{-(f1on3i8x56RgX`ra zONTIx643&*5-1@e@^O3ptBEo{ybDA?PtMyW;p`&&G^I6zKT z3G~4LfoOi|GBiRBtUCXnuILp38y&C(_;1q4z-6zXpApG?pyXH?L^0Tm)81=sZ33b+% literal 0 HcmV?d00001 diff --git a/images_ml_frameworks/image5.png b/images_ml_frameworks/image5.png new file mode 100644 index 0000000000000000000000000000000000000000..644fed120c040a89a095048839be1d23dc882c6e GIT binary patch literal 34613 zcmeFY^;=Y5`!5bMGzdudASzwb&43buNFySll!$|y}swV&QG&vUwf~$_8qVLUa$LJ;aZw1q(lrvSXfx3YN}6ku&{7e zu&}Td2=IVAqA%*&u&@HK)Sf`}JWO{|@jo(k)t03e`!M%9H)+;ieMl-dxodSPrRF#2 zH(AHq^GxiaQEWo8aQkuebjh`!gvYHKIP7reG*pd(}9?r19ErxMqRA~yZ& za>?`4|45GSLmm$3if&1zs;Wwdo0~fafq329+e={UwlIaiAJpuQE;45uy@UHY^zO!H zeX!m+pAS|GR{Aiu--GiKnx(Vn?T*A{`nM>LRx-AMKyyDvHT3kiolKCYv)>m)!xpt- z*k_U%;nD3;GAk(skM&glp1JyY!uFY95#ge=>%&7LE!YmHhv(vZ1?6jq7s?3RFe&+8r4$ zO^j)?!5leX?`c9D@CY{4_HQKajm|dezT^~lnAOU9jgN_9egHZvb&*BcJ<4f}OG-k+ zIX=T~v$~$!)Qi;37onUkc-iE*(Jg$FWhQYc;*Q+fWsGf%&1|J&{~FiRcgT z17c^9t-pg2_}qW^@Tq9d>iW7aoKo?A$_3?$RST=C;y3BWM$g!9--=%CI?lH+Q$Szm zw;1p_0Tvua8+LGf{7LjSh^w2?t{|MNdwd5?wGs@%N>|m?H26EiXXtnGdzD(i7k`}X z9ylR%COOCeb3AZv7o;R`CK1g9t`7rWyl(i&rT}Lrj{N&{j22wniM-$qjp?8lD}@5w z4B-(2+>ZzVcWF_|VsL=ZjESOI>2?5K4t^p-wep;5-Ad0WnI~6TMRoPWl1o}bm-{i9 zQ}}n~5a{^_up*b~6_iTzwpP!QDKYkhFV|yJ zScd^ka6B`0+ zo#j;rwZK*uTP@CRLN)^k1Qq1xgN^vmN%1@OZ zX0Fx>^)H6?E?1`GfUaVFjK$F-K=>Eg%N*zRA&=Kr9 z_t3}?E^y{D?(K$h%=v^-DD2jJYw?O5kR(AR5s`BQr2Lgr;L&LM8tAY!?#JRS`{!(? z&MOHmxS8Ar#fp4VR-=5<(m71f?2w0RM*1bUqX|vT+2tVQ2Ou1w!frX9hucl+BP;Tt zGy4E=>w5LC2tPpWSvgCinqpa^f>bL{bUkICe@Uv1`9atk^J8ROZkuwYdOyuq<4e-5 zQSWIKygA{@-d^?JigMsuwbF6(YCc+WWOEO>ik_vv$YubY2QEN;0@=H1A@TBS@$!2h z%D$YcgYf3_usO!dHFY=?0t7XPK9`(w!P13b#@4X8(E9`aXJ5J8- z6v=m2EpZ%H-n0_;uacbc(bE)nvv2chAb#bG-iz5fRaf;oF*NM%eMFz5=1%~lV*AO> z*}a=1vm16a>Tk41auqX6f0fOE>-m%!yozui0Ax$RD6!}y%~fg1gW%FYtv ze_`OgKZ;vi;QNaIR*f`Qt4BBCgc$+VN;U;TeiMWrjYo8Nd>sh&*%M&KzM2nY5%Yi*t1vh}*ba{*duzNgkh2nrdYrWi%Mq*`|$b zbufcf_B+^Y?QC~zUVXK8`Tv}Dm=J4^@Gw(TQ1C>JD|~PJ!~8Pn&Wkb5?uaDWhgK@q zYEwN=b*j!?xlGSP-$T+^wh0cM=g(C=GQf^&3h?qBia&6%#rnazQyAH;Tn#uK!^0J| zy?VyUFhT-pxf%M8G`OB8vx2*cLvLR`6FZGtqApb=KAehv){|!dW5vLie%(&8cSoNC zwnqHl$`}Vc#5Dd#B`nz#rQQ3&F|PSneQp)!i!W<3%n~6ozUeCPTm9Z4mEyGh-tGKc zUfV48ezVwE{Z=Q?xdG9kS1=UuH5=sVZV0Cx@HXOh=9joZ3=uA*lW=uM&|VE541fxG z8}3_qhe|=OZgt>)eGib^nI~*W3hI-*9DQUO^QGjsbA650GqK|WZFtfiOAacZy0CIk z`ix7Z?e?4O3MuHnmV19WKK=v!Drh|-=dpMPGseUy-mDpHcHF8{=U<|S>MfZrFBz6Z zv2aNX>T#FtoiJ2pySUT*irdlV%eUM{i5I;?<@@PD!OTR5Yow1{ zixG)iVi{cTsR2M)`rMi@x;wYWr+-%Wey6NomLuWU`TtnUVcyd@&^fZln&w95Ssmgb zwPCDJ7uyLCWQ0C9>NBw=-> zUcp!!3A`{x%i|S_b9VRmjz~47M$U+9=P`fdGO0g>CmA5OiYf1!8XYwqM=)|C-H$N| zF>9o_p1-p+p+18an(+OIq6NQIvm6=MVzfk#9{|bi1txH5$Tifg5LFke@9zYXw6S#n z5HCDhioiLigjV3jBmvx7G3ZaC;|Q!wYNVH9iPkaC%Q2TKp0>NqM9JrbHQvAECjY0I zX1qJIW;Pgp9xcpu z_*}{o5GG;pn2d=J(<*&e4tGg@#Zvb9T*j>7{Ofy}`qENOimm-iyTg)t5yUOa@CzWY zgR;dSWr?zVf7Z|3%k*;4iM`Md^FpfEfIfW2An8wqUYGafT)ov)5B1~ieJ1@7QxAJ%fzFH5 zhXeRTfW4&RN=~qK$t6$+3U|eN0SOK$n;(jprC8bgYG|eY0r4S9JwCF8iKr6qZ=@~= z0=L#8%OB)hb$^SJwj9i=trA{@ugrEQ9TI7FzMKa%@Jce#WftkqEapzg8KFM&u$pWr z!7z1UsyV%l2qdvtP(NHJ5d4O$&<964rkKq~I!sm){+xWb0va{IE)G&$|wd>%? zhxsx_@>f!OP@il^Go9Oi%vjAt_e%D|7y>9Gmq5eQPUp z>le+qOmSm(|LrHggmu#HEzcy4$y2R7pc5rL>{p9^Hg7Q0y}f(U+oIHEwG)q9<)-_+ zqI=$Km$z8_L()?ovEEAEb*rgwbY1?m zy?v!#JqNl>+y?1?Rr&Rk`4?~WzSaX1NnYRgU$2?oitJ82TrR_BxWOwg_0=7Za(;3Y z^lE6=cXd^?TQGivoawhZb%mz8zF+Ypr~Oj=OPU~A zGZQR(ZXzNI?@~_JE+#O0^Lm~`x_Nx9$9Xk#VPfQfI|WTuV=69g%Yq+lN?TTy%piWdYuuCjiw{8Rles9L#Bnq6Of&U) z%$WMuDvL?Kah|k>+X9C7Rpy&`n`3qlZXHet01KdO=HER(!A|e!&RPLpN>hKoUhdCK7Mm+N#A4>{4B;FgLEH1ZN6>V zCbHLscYwPERvz}jYErk%ED17fuFNoA(?UWlk4Xes{7$~s7SS@h$y1w8w20!FC3n);e+O6t$SRQd(zyv-oG`1m-P zIsCEhn$D)MW!5KFLCpze$E2x2>N*me zLP~IX{-dlLu)j}TuEtlS#>5k}mhF~B9`8pV_t?YbyRym!hqG5RH_NdV0BmlN=rjfm z)TOy{7)v>9ZcS`EMq-qW`bh2P9`mM$)Hv90EKXP{xskcoJ2q83T$rtmVoPsmZK7pe zVx0j`AZ_>8orj7ai`;()d9CHdVJ>@e;}qnxU7RAj?xpNm4o-KmOhY$;ciud7SHH7I z(H87F?d6DGpGg9RLJ{GMvzK2Y2z9E;s;LI9ve~;?ksRIg@mZ3su63ib4w{V zzG_GSu@#rj9L|3|;W9n=7Qe^Bp6lHyaqGxtB3%;;$+qsUWjufEGs`6E;B&o}TE*v3 zzU9j(<(Bx&xK&?kL;evVG2nX=j_Rueyg+83&YW#33mLJ zk&oECl(Y0HSN9FTOg z4>4NRo}^j{ts=WD!;c#n9~U%6y8GW_B078&d8Am>-?{@j(}aJ#{ytl}FrQNi03xRE zCWmsgsL$M}Sx9q~-==5p=L5-4EAbD^5WK{RPt}_X-rQ@pYju?Wizq$UADc6@ULjZ> zBJp0tQ`*Fi2Rlp|miP7i?x~U#Gc~pUt#V%G{?DUoQHYugefk7J1K4ZXqRtraKJoAK z8>q=fmGxqK^54^+yy|t+ z6Ht7vNfz7h2|h$dr^kq>hdU4`RENqWgC}hdAC;HeZE7AHyfn($v*7&5i@tdv0nim9 zOjIi?evQ8Qs*g0xcyHfLsmH=n6S3Lz3h5_k%V!_>U)mVbOE~}xR^sI;`ywFxSpgMT zdP0)%Nuru?qG~!@+(Rkrq+@p<+U0I<(zGY(r8b=1YEhr2N5@l#NrXRpcS)gvECXf8 zd->U(JbC)AHLFgR>15aK0L!Pak0u6jU~%t;5s3w)=Ke&8SQ%F5J-g{?$Gwu?5n@uM z_swZ1uLa!k1Q%1pBGF>eQQ^}y5f`U!MbM ztFjheD~~mNg-KjD-OTbRwKw}J=`X4~&WJ_i?A*HG82f1_l7v-W#j4?|`nlQWtap6B z3Bt6@wC`k`V-d(|#=g*035FeI8%FaSc%AjeZ0*9@K7*Bg2{eug?4(_7HNp7la*7O^M@R3V4H3$0oYNHj<%=a z=xDX&5Q;sv0$uQWu-BTQzaH zhoeORpF|y}Ico`@155zJkSO0F;pmbMELZRsqK@Jq8)u7GrbQxyX&b)%I43Wu#bhfn zla@KlILTu7Q-6pR%xb zES@`7Yf2C4*la|J^$o1{4R@er@ymG1MdnL!q*J^#!>#RsnX#XE9-lvtHI7}M2z9Zd zvAke-thx)#I{NYN)U1Ho|KQ?EHVO>)Mb>r*nKCVnGqJJ|@!o!}?Wt_3H)Ys;({|6s za;PN(lgU_2zFwNKU2c4&#lARfSA>+SI=Y&kBIfmwyjsXi*{L|O~uHu6D zrP!(r@M>)dLT9}F7I_&$IUDD^E`!=-c#XeT>hA3l0MK*I0FVX01lGVtijV7g?Xu?c zjrUch+8Un`YrOt;z4}e*oq4|~+zHyQl@INlyxa-3DQ>-Wsh49Pwo$$vYJ(>-YJdEB z3pP3L%FxTm^UH`I#Ja`Et;Q9YMtnKUH8eO?v2y$48;zV5!htZC%2>MdD)Y~?1} zXV6u!V70eY``CXIuhETt-kn4T*tTt_n}n;47EW^7&RK;aCd$&UR?+sWn@LhdLmB8{ zF6W2qA&#w`SP=F^fy?FU(j2RN$Y?HR0}C9m&+J9QBw)DZX}u;)J8F4V^Q(ws);Pp| zU3WC$U6+l9POZalO*3R35$?Hktm|r1Ey45mDmAIU1c+H9Qo(t1@| zDs!&r?YZ^q?xA%TcC5~jz;!4a+B%SgS-5zs=|FbRm-cz{UWXpKOMd1wu3qYH{Fu5? ziv_$10x`W+khxEwOtu#;+;gZLial!420EdqN{BCNqeTwAd-vEo_`v`8X6txYTJzex z`hGS?MMlNT69>;{ zIx|i?0ceZkfJKiprVnlaCN-jH7m`Dx)8o{HlPWs`RhBReV-$e^x1XL?Qvr^GSM4_Taxdx03+Si3z{0PtX^@R(AS5h2@{ZpSrRPD z%g?_ub(m6L!Z7O(Nj^EQ5)~R9B-n(z1UA_t21Ixa(kq z&VA*~=;4++^TEwdUuv$&m>`fa$GYr@al0rJA1>zy;qRD2{Mb&aj+w=tik@XImK)Dk z7Gb1>FTd4n$*49R)9+p^CHD~P2Q7}<>C2j@)IIDz8pO=pOIeV9@f^x@-~0LTUCFhu zFX6(8+!P^~qeG1_nRWNOt>WZj>>T|GWrOEVtr!-WJO*AwR~0~g%U?4?F0DfYoexZL z_{y-+KU`?}dwPVqV%;QXp1%p*`00}3dlK@xrzb&ooxh0X!`LL6Q--H-)9C&D@Acmg z*1i1NFirYe1!+fdHoAJ{Vhq{Gp2Ezn9S2T`-x!!Tcr(~5|9M(%zx(dR#d8P@BX}yk zeTH3zJ$a|~S2HQSsI`(v*=p7i%h-@O!w|ofWifhP_Juu^sg-Ne#qn91i-X7~Sd8y`t`7jatwc}11NdJYS5>77W91`t0ZROwL{l-bZ z1AFN!G41a=SIs#!{Znpg*V}XDLT&C@f8iG(N<21KK^FvN%co4A>KAkUxR#4{KGv1K ztREoUi!NArg}HBnOv{Ht)#(k#h~0tBpwZ@-w4am_q2CxJAS&s!y(8apcTHld!9IHH zTFE@aERP_ky38~w7)QnGe7dwslz+st(xzEp;4O#5kF)6<3!Bx~&P_X!FLbcQpkaaR z`Na#X8$An~4(%XGXbG>4S){Z{t7p>BU#xl%BB*)GY;nWwnd{0*FY`Wxdwy4@d_%v( zMb3MXfx2I5Sn&z_1`3I(AX-w9_~GJu>_#uoL_p3@C{klL zuJ-UWGZDW<$)`ahx&k#x9N*VJj{Q7;_F>5cx@ht@D^2sSEq*_&s1!%jR|u1DTTk{ANALPUxkK?7gQYyCNJmz}bRDc_a}4(N;}+{r zVEajHrXmY^F&wF%{PGR@u5!1>D;l_}6Q8+-v5AvUH?#~?Hv3qC{pQ(DHtCrx?U zSJ&V%-^j@D@7h%;Z8=yaBM?cldzl>MT_FNs^3Fq# zP0uIU_=TkJjZ11kU?Eb+k>#&hxA8h=+;@|@n4Kt@tBn&EE|?aB?9^mFY7i#Tnxgr$ zIRv!6^4m9)>aKH<9=ewmgS%|P=yW$SoUllV8f>s8kpNfOq$zBLGT`jKPKGkAt4wqO zq*?Jg%@96F1aunyyT|#&%TNEJ;Rx04Ql!%vs;RxiJjAutCJ-ut#uj<9Jat}u#PYc< zB$@TGb)Hvw*&J0qOxA9Z;XLlB<=c`>kO%FF+hEkP$_cB0(_jmEiN5R#pFT}jJA1y# zVD4+3OJLF{!*`qk=$${lZ4_PWdUg1ZEe#~m3#5ah*%0iR18G5rt^2MmfWnHsE3Le< z4%J^bW8}skbsw;#0wlsAy-3dWfT#a9H7UIDcgrQg$t85?9ae{|f-l-BfM&XxlB}R# z#<7uR3+Q?H)PHy0k9_sw9J-HZEeVpSvh24{5-#UFaCEQ(__cqEyDX-Fc&QoyhL<`R zTmW;uv;Ilt6aVWL?eIT^E(QfPH5R0mjkE;shrs~>L<0S?-q8h}1RBAUOxE^!V8@plkh^cf2t6e;h|>>e`(Z4Ag^f0rlxP{%M*0(*_v;2LAloJR|r>hU=RBAenasV2dN z(+3T1T&GCqvw*ZQ(l~C#vG;I$sIx8{!zlM0cZXc$f@xp&J1x@g%tRp@%WDf9JHI8O zdUBY#NJ5HVUb9!qu!z7A%0z-yEbb4E7f!0u=4`X;D&#j#li)V8>caDF)*6r_-Jn2^ zCp3u+Ewxtl>~NRy>gdROsSsSVppB4Lg?J19FpoLnsHtH`8me9)m6fb|q{UFo9yuo> zF3mT$`Y4iA@2k}-_%bpF{h39nY+ZR|0r@%eWgfVznYeJtS^N{W$j`-Zg4IaO z%UAC@%$ApmtKN!m41G6LoGQWBfXKPr=*$Z_uFUBz+#1)RV`ngrQ6iba;yTO78HRiR zg>kxY_NR<|>jZZMHK~_}$@S3cQf#T-(FU8jHUh78Wbrmk(qvo;84I?L4Z?6iCmXAs zw_Tb>1)Z^M846g;_3QBW`tvn>rnn$gc2To2yR7=+Blee*P$neZM(w2QFiv0^!FM!a zB#?8GP(}&?Ax?*qtBNUG5iCptVO}QE@bkWc7L1oFryzTs6e1Zc z^s`1zKE(3s_d&`u$kU_?j&TP`J~kJe&`Yw=*wYoLGwjQO>@2urSs;XBm>Ea}Kj0d1 zZCC|wC^+U|m=@lEBtduT*alFbw+Nx1ZyPef7z-k2k((lnN`xD?3o&??(XaZ`I8B#L zS+O(8*I}iCQx=J`G;acD1dmV;Lz=yao2#~PCabOy58cw|Muv)uP|uq(-_9cCo#p&k z6A0{R!;NWKQ=1&#lPG7V|jQ~vmR z8CXQpwg6=!y@Qgp{NL1_P0fnb_ONBwr;3w}asr)z*R8&!+2!Q2IGOHkKmtBKjc(-1 zs|AO0s=7emhY^8z2}lIjNNnpjm=y$rScE_d+v4a?#?9DX{ctz&cPk`@aEm35W1nrQ zJg^EG9?Tvp6aZWWVR`L1y+$WXM&>>pT5Pb57)$<`Bedj~A#^7Yl$feW1lHNIaNje8 z_;#hSkG*vJkB$Ma$ocTD;fLjWYcJ8Cm3JNNC|9~S6r}jBsM;TOx+@0vJKfp)V{$oG zxp*Y?kf{agEE2XEe*y?a-N4=;%F>P*JM(GsmhGp(=D`EsE1I@^FhBSaSZv>WxN=f; zJUa|8*c{3TI6SMoZ{=1XW^Yn2;~)K=VgD3J+%m3t8|P&pLY~zWDiaUxI#KH1lm4WL z)Ey4^{8&V=>ldpIJ)ejFw{^!k&kBa7&pnIp@2$AfE8S(gnWU92Vxwtd@(aMLbJOk^ zM|S}2Y9Y&u%mC5*p`I4CQ^s(#ko^)Rl8og`2a$8?yf;<39`OE;uy!#I=sYC_dhX@W z!Y~u*H&pI-!$*ghZ!P*3`Cg z0gV?G9vrS68a?(+clDJZ3qi%=KZ`P1^`X}erGN!W2k$;n;oSKI;ta7+_*g#GphQY( zJL%i7mDyRDR0`9UJ-B^mG#*q|ZOWnyw>Vu+V!Bl%V}8`GC?(Y^qz9U(J`ozB8Vu?b zI72)eH$#}(E9Jp3XWuR&dU6C=xigpNeL2!v+w(nrn$!^ym=Y_<)z1#%d(-l zL-4IA=l-Dpk_ze15BKN}wV+J&<{qqIZzb{AOhW{cJ%sJ_0M#z7)5;q24W`U-Mr+OC z{}OWdqYp{#aFnfw%2Kk2dzK1wT;h_j*Db9?OQ?I*ssB#Ixgla5i;TV7$xQGz$UR-l z1Gm}~MWUYibx4&HBFC=UE9EzNh>oJbu&@R{d)_7ChOP?<2H2^$)Kn{r4+X7|I)vK1 z(ZMVa5E~Q)OHVT=={x0;GZDYGGjSzB!Z3UalEp9A6}-exMD7rQh*B&l$2=TBtmH!% zbW&E#iW0A(OkCTl9C}>aniUzk69-5zxnKu>XUn3L&|MnY&18Cp&8lt!g)h4bDxf#K z{5rOkF9a!useszSO9YhZCbV*i{NBE9WWBOr)H?lYEuk|gf8VMKe1r;njB6JwSPw*5YUbe z#+)$#ryqU)x5xoS8N{k`79zz!FR4aan z>JkgQkCB93!h$`Xr*Q0$`77U0SaHG=z@6S9+Cd+Zm$8JuWaBfS*)^_0-q0mqVqaP` z{N8tR8Ji_LLp;rGHUwI2x`aX^zdn4qi*>|C!Xm5wJdA+#w<69?eMYaq0P&0Ntrr0)@&TKpIEm;RN#5jbLnw)IRHT(JWgg(Oyk&jRFJz zs0y+$``zmwm6fn6&ku9t>&!Mq(X704)CzEOR^3l;Si|0uf1q(06K8BZCK)*MLp515 z6JVd^CvlsVeZq7-h*E7=af98+CI8v0Tb`$@hF!i#Da`;LCv3m>4A>XvSy~#FdXk#G zM7)iAAK4CnATsgc6Uu4Bq-F68=dmulC)YbeNP}>RHtJ1-AMa*(CiNd$Ql*Wr6Qoie zdx>Q*?fa~&8}uq-G^ZUdAqk`(eU1gUx0TqbQ7Rttp5@FM>Ge8R-`-5P`gFRG9H1^x zl7eoGlkZT*=rennqnBIc^qk$G{WKi&DjkiAKc#58c?30G3e0`=%gM~_MHr5k6x!BD z{!EMk{Snool&^aG=C|JReLrKwI(9qgXyq&MZ1^#BTo_lfg~tow-ieO8R_oH6mYkp= zt4r@e1*urh40=d3S6-V@?Wr|~29 z%7PH1Rgonbo_rFCjil(WMJ{-bbc#XLsE9Ssk4Ueh6IXBwW7Rcq|`nio?eiFDQopgHiG<=cqe7rE{kKd0eSejd8*O4Db zK)_qGQMClq{3JI>u~(JLfCTFVOV>r|+06#7I0l(B@j&+eCjB^}vz<DwBN>VTbvG2g#4 zfx9VEbpCc-#{yl~tN-r0)-;DFCO(psl+@AE%7EH}#$&+UBi>r3z@%YGBpd}Q0_RbT z;`cS-Nl8gS%k~m5RBY^hBO{~O^z;`1iy!`BOC8Rxx__epG{yJ0+`$2zdhv2!spzPP2OdtAJ{&LGuHDyrF-8GhVgkvBzYE;2_8nr|{^ zn_P{nQWs3|Pow63;|Vj4ax+0p4NTb{!scnHX{9G$KfpaU9xV~Pn(#bb#F3Po921NK zn(Uuym^iej2#@+L6ZW0X=|oT88=)h@@F*!UpJL^3?`|tBHl_&H5b4%w=^%HMz_E)3K;`O$W_F)gxPP z{~o5{6+Gl*J-?X^F%8fvag`3`^#qnJI!AdJC2g+Ni1FH=7MR2 z^Kan$Ixb}iZJ&9hyPxs-V9RCx^2j+arCB|*Z?E;5uLZ1O?BIeIWn$VI1xha+_(z(5 zNaVe&1ES5b<--W*2y6%2$QHKeJ1bSQ>}2fe{gQk znGMb#dG|(KxNQ;}hi!m*7e9uDY1omluv0lb7qB82sUKtiKcgeR-asA%|DmMwC{KZ} zMe?m79TdL_*8%4#{{MbZNwx+YJ=17x>R(LC_Vd1!`*z_KaGYp9uGz6wiJ&s9e*Sz? z_SSdOP#M62jn?2l5vTJX83FemApZ+JpRr}xP8Dw-{xtm?Pu~E24dmW$6kdb@oDX3o zE%1cupT8h!N;gdTR^ZqScma6gp4eZ^q{TZ9{zH77MDG3f5JoWR{~PZAbJ-yRH>Axg zB3Yy%LcNPoiT1v(3PW<&b6&DuCr1e?clc&%+{>Ld!bVyS^OU1e*H+V!Q&qNo22)Iw zJil!Q5=?#ZgRdC9+OHX;d}?^j8kOI=@5DC})2Ov#0}a66?oh474Ak09TCJCv#KPel zJV&v~4Tj?bJ&kKT?9VCjj$=r#lg2>Tll%^}v)Xb*hlhv0tB*Kdq?YN*z7+u2?w>6< zp?4gv5V#S(u5F2>l@6&J&xKkJRWsD}USJr?B?P=3J$2vg568qN_>QlOeEV)EZuxV= zspVS3>y$oY%M*hedY3gpY**_1w343aup1A%&9LQ=+H?LA#WBJte%>o8WL#e)Wp=fc zw6az@)8afT@@x`LO@)(ph+1?+Zu-4WG~Iqaj($_afSQm;CZb5mL@>MuKv1 zS46FES6Pb})k@S*O6hbVYX3`bvs~zbjwi-`d^^hYBihLe!~D0=;OtaLO;fGZ{<$zS*X0>P>Si>h^4d=ibI5%3|-VixgJbct? zr~(|tI0m+WU|15PTR?xxuTX_C$?~fwmd1ZpPwdIeVVBE{YE8jN zq||+I=__A*oiyb_+g%{EY~_3W2x!Il)c5E`>gLQOK*#Sce6>l_7Qmp8muI~pB+Ny< zH^=$%m%XaLL$Hy|z#0UBHDuzRPqv(jdfr~nmorR}J9=$n1Feu`KnFI320R`wt467iWqFyQ30E#)u50_0?wd{t{aQa z=ZDNQKzGhAm`B=J?!4zkuZr9ug{1opwE0@+!s>iOMZTyZ`P;?7p+o1pj3m4hbUXIt-^+|YaxNoDYC*vI~X z=EBk9b-vfU%u?($!Jw6YOx@tkxRYD>h7zzs&Jmf1dhMkqSM_EU+u>dbCSR^1ArOA@gTL zCr_^vf3*y|a35a>@^o(=*yq=Wa_Qz3(O#kbFy8Jnzh4-hAV`_Tcr*Azta;5Ub^LBG zgI%4ba{h3pe-tcPWkf#YzesErPv*O!q}lH~YgqpX;&-E1 zRbxM+*>Zc0?pv3;^*j*(wM@NKx&Lo*C+E&+s0(p;6sy6NBF&x+?O#)fs(wCAqwW7D zZH9y+Pq2UT2iLo^_MiG48SEbn9fV)bnp*0@y>W|@jw3Z}O@i8R=e$q!H|J2kqECN4 zZ23C(o8QE(FRAP$^>Oo(_r#-Kp}UA#)rXig4!Q=@BN^+U_Pgu!6Ei{+j=L98qx{_n z?B|3;&fW?JT|<7S!Y8q2O&=PiDaPtsukE?`c*qK7YVE}J@#HS2hI(z}F_IXUR=;cC zuA7oC!L=!NFf$X1#6T{%woiH}<*t(n^apstxNTc3s(M zt8rs8W2Lm+_mX~6fr2eZr3f0OlF4>gul189!~3aLgl6jSWmyt$pu?T6-`2@+8K@1) z=g2Ear?A%|t6okTCw+x7wRc7uxs0FOXJ)0W*nE6E>^)=!)xb5u9Dj9bIn+`x-@Y~| zGo{sMI^@dtnmZyLMmGAM!h%R!Yy{{Z`Pb;J|j|>p!;pPi}sf}d4 zNUNtjGG24Xbv<$U1s2Hhs@(FM&nr}BXy$Mx02Eakt}#H1xY<<0dso5D7}GpZ{J zx3v*~$c_}+IOGq)Uv#fVZh3}52@kvc4v1^*2fsJASOnn`(-b+AXdfiTbO%8Zfr7Y= zn%XMQYBb9yC=sP5th*BckJ72Ny_?;ssTCLe)%<%6KU&;8Y#M*L_e3nR>vRuCyI6N9 z4#wm%l<>_!R+9vYLZRQD?xjF|F8U?++>S5GUsY@JU9Zly@`G!ck==CaFlDUQ0t|e| z2LmGhy6|kFcA2}6Uj4S(NIgyK`ibzH#c}lHgfYu+?uAw!qMQY-uO7HY55#-;ri0%i zPwYQKXXS>6b8#+s42#$6u~O=PHY_o;QnxEa({DY<*Atb=6lF;tVT3uY&_!>_a)fLA zOFHgX!Gv4FI^)RA$_G`Mx6@ArA$L0?nTzY>&sXC1#LB^;^ZsWYtOhQM9?^EtND-0V zO%45fjQ}f04iNXMswFJ*fg;YlrMdEU5LnvY<)%I~?3W*-n8Rtp=lcS;dDHlcFC&ye zqwS-0Y)!|0d0<1^6XlLTJS1jj*=@Wrl&KHq`FkL+(H_V(9DH@*FTTqRxXU0}Z%z`w zW^l~Q(0Z*RT@heLHoYNVK|xbiCA@y8z2tyF`xb)fG@{Pi?EES0X&T!3n*985u(EI@!bY zyUx5FnFXxLWMI)~$3*v?dk>KpclUS2-EZ!@%g^#SHkN9^XVeM;Zlc8JKLs70?n;k^ zxmR%<$Jf7~dv&wjdecX`Cv@KbcTBl3FX{3N5|{S2@|!Ikc}?u=e??yzyWN9;Cm}r z>Jgr|f^8Z`lW`@P`7$rdT?GXVOjwD9eZ2dWVp5XOd!a(Zs6&eEuEteUNLgURE|5<% zkijZvn&c8Sk0wMhLN;#Vzoc7^Zi3VtjC={4&Bfj>ysF@DtcT%Y@Zz>NFJ5jvPg;}` zq-SQ}7nDqGtYL0@;O8Z8uBl@qOts>6k2|AFooOaJtLgHzkyS8IZ+*~2Z;MG)>a}b= z&OaES=6fLkYqY~ScrJx=j8%fkiT0>2wqiO%fn-UP(0RDhfpjV>05Pznu-SQ)y3d?x46d%*=HF8bkQBx3@Nhg{d`MHw@^FNaO?aKCwZHLWa0Oh1*24&Lt{&)^kuh4W&vF>d? z;dZ6^&C2|A7Pn#uhh;)!% z0z^SX6afXL6ObxZq_+e?1nE6=R6;L-KoV+rC->g_Is10+@9*`T>u}`@39QUD>lnZP z7-Jd6P!{q>(JV^sGSawa;C{8__$wsLfPd9v&FR zb@}mZiWAXF#Do)brRp6SSxbdUl6E~m5c{o&yj}Yt+;|o8K9h0JTqN&@&KELgWyR8~ zVxR1mJh)1~oNOcPQRt@dq}OT(oU-|qha1rdd}^pV%@(XK?=_UxK>0xTw#q55`+Igu zn-?7VK87V_aJ>2D*JCX|=f1%i{g*kXZb?<+8zNHzr9EAKdSveYPl{ONNlv+bQ?489 z51q?prP*)%{D>KsN8O{mllGot@yv2LtNGlc?$^d!# zszYm4nXMPe-hGl&wcg23BMh`3sQ`1K_Jl`PmW`#yAYoTfuIO`8hzKS<oJ(wL^N6Etm?-efkWiOjR1ys)neXMbFL{+`?lW8#(_Al%&7`cq+;cg&pK##3{ z0>XLgu7g)=6f#fzG7Z$l)!eyo<>hO`Hy%?SXYYxnyNyjmAAS^?wa#~bIVG?MVP{+r zg1)6csM-n1h@+>E_~J;oc&(KZ!%=mF+W#MS`d<&(E}6v4&&{>J&1|LTW^u<4I?SRE z$t3h_18M5=bF&uoBKo=6U;OYBe)`F{8Uvm*6=X1_@%k9|w~JH$^Bg6D|5HV(oTf6I z1*>}x$TbOH^j|iJ&AtcN@$js*=$Yg%`v1zl5oKHeSaMes5KSO`(1U3S03~GrfR<#P zmkmd%aT5(7pTwr8Y5)9h@E0+kkF=Bqv2;K=kND_O<%hqDd|3bgez~suF;$M^cQm3? ze@b|LvxjkrPS%o6jpX{GpP}Tk*0wmruKx`7SsQCMZS6QZjfoDUISnQHJd%$_v?9hU zG;_KVRt3vA{tjl7DxkR0Tts76Uj1I>*@>wDzVLTZR4Zk0gO4SrKFRN$Vn0!j+_*Mv zK<9^3V?47^#QLe+Vfx;yS2`0;T(SZ6B~e{R=|=zFIcqiCPQy3%mK?@>m-uXFnU4{! zv~is(fr<)UHA!Tc3P=K0WFSH^naUqYes=N%L%h)D5CD3A;Ace{?5*x=FvYHSwt_+b zyaFw7<48l4Zk3MwGH#Q^p(^c|P(^xp_D6iNg>D_ss+&#m~w6vf=}QZP#^a7?<$du_Nlo%r)sII%hDv;*c<(~aDy0B8af zk!fq5sN@-yA{S)hik;Kec>KXRdR*8rr)w+Q#zs{_cb)n}v01iKiCKk7#NkAJplpJd zdX7GFcDPmDDq`E#gv5T)a$&ap2Y=IvqtmPY%;A{n7jw5UBi^k z1*gSPyQf=RQbzBh*bV2~QSUPOJ~#6?*6Mdwt}d}p&HO|y$)$VO>7(S5fOlxKlNmtV z7;35TTRvU_#SLcHsk^;;i;Xk-f%Z2qw&;2ro7suKrK-DD9mQS~4A z-~!^TuV7rUw^V;CC5k&0y?+El5&N=!9opJN+U@zRPJCWXTr8~WVUEQ9^;7J!g`-$zkHVYvda;S3_{Q;TSjTn zyVZaiR9$(U5fNsnc4d0$X*ihF)3_)(AD0fi zVdxX{%Ek5Gy<<{O7VyR2c|so3&|I|Z75k+)^=2XvDG+`>+=?_z-D;5XPCo4B(xD%! z>tKoUzuax){3Jr`$AJQTO0|mfyU28{WqPGr0y}X}Xm1R18Y)vT#D)Y7C z=l*GIUibrZOCRjQ#?wvJK$eM5XP;6u)IY9aHH=YvzH~TmQFrEck5IF3=;vW%1+-PM zAtvySOEcjTA*f^6OxANP@W$n)sRP={nHTNv8hwdVHd{(F_xutLZUCEz#(J#Y=dR;ck@r^C5?)13BX1QXbYzScX>#$3@av|{E%Szn-GspT=SSu?pzjDZ zF($=THixr`*RQCP0c_w z(=^1k+Ri>u{E)_mUpF}K=(c6o1n*99V>j3nJ61Df9^gW`=>y>I+MgmXz*0+}K4kIV zk(u=PFMV)Kh5Bty%h_*6qz6I94VPQ>s|jMK6pSD{FWw|it~z1djGFo$k(+p3Wxq?> zZ~w_#G(rFHuuj`Sh{v$%8Gy_Ph54tsCZG-#$sPbABT z!7fr8Gv33WyN$BZ1FoG->3lZ~8>~N42OGK|=9f^X_xg^fup92Y$5ZjC8Ge1^nSAd1 z8&%&wQ#BYwKR7T_+`P~@UGF>HC>3laQt@JA8dAmuY-CatiBqDm_6*gurCaL+Re=1( zSM?Z*XT=>yAqv;N{@d|La;2h@tyIxkb~!z=a6zM1Mrpd}HQ<~C89Fs_Cx>5Mjw-lN%DjJ|bOUU?_g#Cx(v%3aQ10_}GoP6g?T^`#WQ z-aEeGT3q_3d{=r^evi2Mn;Cn)uB1i!BEu8q%GOnnWg$z{tyR`rF`1~*WM}J~U2s!v4!}9R>Kk zV^PwQmuZm+P`JY%2bJN!7;2?{-&?@zRn3()apD;K4VX=JR(Eb(2(a}|ywdjeYH^h& zp>S!I6f!#Q&9Sv>lMLRxh?{Ixw`ICc(!?Oe;SU-gxh7Ec)ES5Yb!WWu#{!Gc{iUc} zQNfa!Q>g`fQVhIp;;-wcqBNdiUpw3?Uy7vA1Idb|b5qNk!z!Y6zT*cgw4kI8u!xuB zABd^A5_f5rv%{eno3Z~g9M4&>Kc9;a$bM&AJ|rzFYaNBlX_XF-s}FJ60j}xTi~;0g@i*eKWWXvAt(cPIw5}<&acn!hE?r zm4FBP9~%SEyd*iyL(@zwId8@EBv$oKMxC?8d6>rLb{lovUOeZ1n=2#`_``rQen|cZ zNpRRnyf?HsOcHG4_S1V<_%#9UHhG(h3EQ}`GnBxTNzH`aWD)VH;$2fb{UlA~j-r#^ zJ=%n?lM@i8;lxCWZ8qi3U?_;$VnQzliCwOfRnO4Gk-I%p=RbgW7|}MR(Hf`tOQNFF zrmi0N!9}^D%fS3o`cmVwT8sbs{CTfVlH*w>Qhu$h#&>pyv0lVwwq4Or^e6Ce86Ze4 ztV-E!(42ytZ{o4U{Zg^E3?S&Z9X===s7aq7udF7RY7!!uIo% zRI_om#i2Qp1pk9Nn98Y6!sxY31*U-^bpaJu5lRv%ve zzD*SjLWSQM-nCr?&hMTxS&{nzbcO4&@@ajQD2+2X-{#t_g#B)LHDypp$z%VU_K^gd zYY$x8k_l|he3~KQw~q-#k6Vx7NR*x8h3&MP#&K!5O<)Bv-<&7&mjX9l!|=ksw~|D* zrrTbSbsR%N_NU&6QG@jNPe~S?3F$&{z!Uy)x^arF=$Tf?4Gdvf4U`v%oFDFwrGDnR zwh0AWW2yh3>2Ql#T+d;8rjZGdWi zpd(4~A7f+nTZQq$dK#P6j#6ZH>*1BQ2yj+~3X6-Q*6+IWBRf}GL& z1HdbY@89N}JiweMt5m~rH*T?P5FOBfd%>4)$8|yG;g|c-@V50k{a@Z7 zj(?%0X8Pm+&d%s_?ry2yDH^_;$vF=uY+MsTKR)49*Z_`A`cmMs)95m?Mjb3MFqYY4 zN(ABKy@^A3HNS!N4pL9HMq#)UuSTU70_-9P19 zAe!Ac(M>JlXmbQwAWq`T9wJA zUbrc>%9iiDQ4BHC^7iXi-p-LWK%hbhxSH)t1muX!H1 zkL5qs!%WB9M&i0Iz;(FYYMiJKKWwB`N>@Ux*l@)*zzgCZ^1MAs832EZAg|ad8wOW; zje7WA$^*LNxFhOxnJcDv(&)lA_(%|WFfl*TbFfRTfBV(FR5Ymp_eo-VqY6b39Yg5QXHq8$gNnqM_lUKKlI~Z>2XgdG$IoN|E;sFBFYmn@x&M<^`L?ZLuD<7T zb9Ct{kfjmtkO)JU=B(_$Bth}|7B~lwMEcwmq#f0 zUJp$F3iNnYqHqpU3kwnt95ky*@cGeYMsa=s0buO~@!E&quBSt*j>NE;k>0X{2!lXX z5j&%j`=@ndl!>olc`lxFodgaxA+yi7-(PO&HrKcW0eX2~rl!HH&CglW2lN8X9vNJq zrLzM_vRAM8r}G^rV45Y#Q^b|Bo_w1OOt<;rR$>{-g~tQ5llCC~D8J7^)#`U+Fy)Ip z=Zgb#ipBJIN6r{MQ5MUT{0&OX5 zI%#}wYjK%oucFj%<7@?Fj?I7RQ|`nnY_~r?4pL%{R=IVnBXFMYX!7eXpsD^zKx!<( zPWbb7|HErE>}BO2ec77WyepuZx7h!Sj`9_-cvb2xxB8tNwzw*{=?sm|A88;-?WDb# z=mpOk3qQ$Du)U;R%k?zm5rCBkOy;Q2>k)hLxf2Ve<&gKeJ8RazQh|X%PGf;9IcV2; zZh5i?J0Go-zx2>S>PWLX(tPrsa=x(|;bb0~tTM3tD(<^FC2XLu@m!AO;==zCMEJL0 z$H5RC#i>n*I5J_)rpX;J@|UG6m@6xD)l!j=*6aArLZN9Cbg!McVz3=So@WiT{fESs zaeUWvUNmo2dABb*u;aWVNMSQRvt?Hp_jDH0K%j^UooQ+1-v;u7+F=FLuIMpPg_BBm zeaN@bDzVP$ZKrv)SpccPU(XTLe((v=nJ-oh^scvZ5cWiVa2eZMdd?OYy;t)+X`!zU z4KH+6+}&>12DMqnZ-RvX!s*jROZK`d#qoD{=48;H9tU0{)dgJB)DsHDlMiUOqjZku z?U7VO8&EW3`XFeCbx_l7B8$&L{i+~eeA9VDsm2U)XJXm$CF`M6LOsvF%i^$xf2}S0 zgLn#13HAnNtzqc?=p0wDLDK`6Vr&rYDeA##4&*|C&GBQnplWD$J)!Aju@2G=P!JS; z@1dtIp>|2ow1n_Jb__(UyarbE`qn@X6*KlmZUtga68V-z_$CIl%~S~v1?dy5RKT6#j*ceyZKHHGoXVn_mINTFqNp`Dias^-6aR>%GLE$pla`^ zfK3dP$JEkpt;K$&@}Hfo8gl+zYY}?mn9X#3(Jqe%I3xaI=W3dtoqt`9Ryg}JGvk9s z)AQR4dXmc@?TWmYr5#!9+=Bj!On6S#p`i9>qIJOTMqA!D5di}4kKfsXkGku_P_=x( zH+|fdh%2D|O!WSI?ZuV0WKd71i#1ZlKuasmxB!zU?)WvBN$XMbQb> zxG>&H`}xwsjU!fkhHt0)q4f#;V7V9Ra9P@!F%OgB85cLnMhbDC16-`myWub2fZNbH z=9Q$&cjsyE{X=56d|UoZ=XkeqlO!YZR}$^P~YFJ3iz9(9kiQYrnJiUFB-$w}52rtWm9Yds@TJu`l4@>DeVJ{wtO-Lcy z#KXB&N|=|f3!i52;CSdmZXzfa=4WbC)9>jeSMkLj4|p*dK-6A3cNY8lgOE-?ZzljS zb09_lDS4*vlTLKX9?DoHhH?Z%`98H=KvVdr#IRYBn_7M_#BRs#RUZEo@rC=c$45RT z3HSs3Nk8vU_P}FNwJpSS}hqq-PIWe)z``wf9clA$@zfiyH6U*l20%a7@VzQrY z$IG+>xK%jebPMt?wz!)$ckJ?usNDCeBcrA%KQkP#Hj2=BP}=wS`8}j;|57}s zCK&dx4dK&ToKkO1ho@)Ct=@p>>nnlyl;QZr{sz-`If`SqUh~t2KrxPiTCtxG#}!HQ z7COfjd|CnLMjME-$z9SOuR65KNvl7mkCaW0qYtsrY@o(b(7Qz}Nc{>BwWyCPH`R)9 znpMtn>e;KNkF_Yn@Y6t9sbrs8p6>;{M`o~MrSgwpDiJ|%;omS(FE zGw?w>pp>en-+DtHP;7K_?G!OumA)9b0L7%GoHg1*52g zww+AB;rJ`8!hT33cDhn+o(J|@tXnV#it8(Id&k0HxOxmiz|(u&G=(^XU3%AwVdxv? z*G+nt#Vn^51P&tYL}*(DS^@kjugEn-kAGT!276Z-zPA79th)U*WlN$7mC7kA5Is)e zlW}-_gj&UwBBF1-=^XF=8qJ$`_3$+(Y^Ce;TkLY=MzyOrb&Y_skHV&o)RXnkJjzI@ z+h}djJL28!7+n0T_2Qt7hdos$+sRUHvjg5H0eSJF67@unR+Z`B>8=mmYA#g?u3K6d zh|KjEDQVUuw;me>eP2_kTp1V!S0KoLS`NpXXa0QhpVMz^QAU9ORt-^PSwMsXoz1)v zBhojP>5EROZDGM~h}-?{9QBm_It+TDR5(ehH>h~J`Uj2~)3fbY`6js%JJI^<@~n26 z$*ea_k~49JpRd0)ujQ|eO%@ZIz^Uj{udI1s??05)lp{R;PHuHbGIJ^$LQ!8iY4UhShKjEZQ@;IYaa%}+ZgY!+u&5(CBLmrt|~kT@>) z6*C`JD{^?6xG#C-I75Brqp53NqTxzB#F>Wa9jclbj7qI*B(c9kPYFdVI`PLoc3O_6 z@Hh5Tr;frMcZrhotl)lC3&w86*ZCDxa(LoF|8d_z1$5Q2tB6V^rZ>!vAyZKh?$UzI zCgw@S;>;gUKMmONI5?ssHuw+%k!Q)%kQ; z)_T!XFuL#HUu$y(uU23v8qXOcLd*OqZUc~YY~}6{;VU%PWoY=R zJ#xXJ1(}36)_ATYnhqQaG)Jp2JL=qyvQV4*J=)K-BFhv$UuT*Bu^Yq{cF2J#!uiJr zO(6i%jsLb~|1W$8FZa=R`dIN{<5$@&w%c=ZUTbS>ZW9WIsPjPal&vx8kmius8D8uw zC^08v4PCJd?uiNlC;yl|-rqn_l>|h7&VI+8)Ejwk4W-c~;XMtCMkyqfE6=Y zozhWHRkUBZJiAzons>Z1&r9S8VZa3=h{aEkZ_EI-7EogUdh`_|0I{|`TCzC3cCqe941BGrpKte$B?h}UZ)B9%a zC&(jyfjW1~KImx6wV#V$o@waauLi-@dqW3h^WAlpb);HgUF`k{1-0#Vp-zsHmIl)| zz4eV!yjar^UrQX1j0EqMckrUp8LlmtZ|FEtSk2Oa`mzs4>=&Si>i2uY7M+L-lRZTb z%DOq^AN)!~{Q4-U9Nbi7aZtWIcDar>9T-{9}-o#M(#sJa4$8yaTfyoapJ-e$X0;FjnT;2>U!tTQND>};EPX5Q3v#E#- z7f&a|4WMTfB8Hq0hsaJHru$(oeYHXQJ4;kGR0}OSqL0VUn1WIiG>jLMYGfk1E^9{H zw!bcVAA;|nU0$`U*}Zjlerhk-uQXAFtC#BA;G*O^e8E*}&syC`J|5N4fu2Tqwyq=V$g<)EpPtfRF8Su8|Sy8(evB7Lm)E!5!l~RQpyykgQ z=Rd}9?1V?ez-|4HVp$LlrK|T8(Yx&h!Cbrjw$XJ5v=H7|U>!eslT@jDWRA+6THOz0 z4XjmP{;H-qk=gN~?#N)hWU@C>M_7p+tSNbD2ooXO-eJqtH;wpxL#U0`^u(m;Rccf} zfsv;gX;vyR@4vA3+}OV+zaSVT9cI#?HXcT`GeXPRxgm9WzI&c#Xh88Kmn`2WCITaZ>3xIrx;gTmC>Wa;4x_q?7f4lm8gPx8r7-KXN*-*r?0JYV@B7xVNKIq{+Q37 zdIqr1`Ak*GL=JgLLeYOLdq=^x*{huG+Ssc_(@($fxBxm)Y#yy7yR|Z@d?sV8j(x11 zzm0dT`9$$^-&)}e#m;sJ)A8S>vx4s`b9CVX@M&9PQYtv;o2opk(p>jLV-~jC6{s3nEHk*v(2i_2Y^A-Yn8cnq#WP32dM zyr?g{hP}J=g7wI*(}n|ktp2K{q4_0s9#LR?ei}g8NcSfAGWT3=3uokd&t#!@2UOoz z{&~|ekgMe^R?x(RT6^GDU~SZ2=69z}2eLjG`4(y_we7_bu^t0s1l5d}yxx`AlR$fd zXL7Q1>!&Br{{8#}XAz%@h&bo#+kF*#-s1a>&$YFaOnL{r8QvOzEzz8r+F}j2C>QoJ z`cm?B|NDH(KFrb7J1^{{`4>0HK4%zamZ)|@V{;xC^Z(N{PioV`oG~aFkBOu28(zK} zvpku*>Aioy4a1`?2YWt!LC@ZZRjWAe5BCsr1~>`bL!64Ps&f-dC)s3pZLQA~NtO-> z=6eWY{C{S)M)izJrW(XBq{51&Y7%mLQzdKUtw;a-a$DqoMn#$k7b9c*gErK8ERbaxb6fF8VfN&8;G{R0E zlaXOu&;in&5$yqn<<)qW($(&pHuHhb2p`R&_GiGU@>UmW6May-BBFrkvkczI?qoR4 ze^uXKg%yKw^l6J{C@0>E&Oi4ks~gO6S+~3b&w1NwVY{_yje%vcFv zbibthy!ZgSTh#D-_X3wT3vHudC6smJ%wvy33_~un0W6B(uD?Dvs&jV|N zJAt#3xOW@tzIx0co>`7hNLTDzNFYcdk_}NLSA7f*N90o1bQ@x_Ztlu$ytM%n#AIiN zJ~B|NeA*M1nWb+@hX-P_T`?O{csj>2mmLxkI7qyoeIlfOD^UjV?68?U+7m2|Qiw>Gk_ID2(_J=RuybHi}~ewKZb$htIxK*|6L6c;sNbRZ9K!j+?o{0 zA7Ok9Ta4cO<`yUss+zPPAU)8-WVT|LBi>;=!?W9e1DDve6;W++r9 zStPd=mE)>e!?_P9~SPWjPTP_RqEy|2!69c+=Um ze4Uv3<<3|^=ABu>xzPrd=A6X(B>RcNY~>}F!30{1X|iqbtfFO3ai4kwaWt3zr1d#| ztKOdvmD(&o{;!7#PB;`=^>u@Ze}YzN{TQj(x1nQos!T>(IS`u%2QYV4_R0F~9L|Xe z0D6xdu;RhQx_KIVNXGeA)0|w}JairXVw{t5BW8LZ+Quy^Jnr=baHmap282`Y-MxoV zl;zo(pPGvR4m;Hx9=(qEiNMDq`5b&F+=FI|>`ETU+Rkck_BKmSfN3p>4krL=8#5lLQ zx;4}|hg=|!A`HAeFxiAd;}GH;+wX&x&qv&~n2!vg0??3{O?@(Rs+{ zAr67qwT{5%>LCQ^sR%05|wgG;T8;eaKE45_}Z3{A6dOVB` z>RMGDfWCB7NkxQ)k$qtRaj*>I2ehO(_^{g-HPHH8tBUP%w3J@^;`3ZD&NvoeBOE!} z$!cot_F))Q`4`t~*4hHAJ?$wj% zV>9S$tRL&Ss30H=oZTZ{B7Fr6@wlZgZzx{l%|%lkxsh0)z`OJ|1_7-|n}&2kNz$Po z+eyff0&6wangm;w#85n#Pj!MgE}-Q^{hPnrNnpbNn&PS<7I{bqNVUFf?^5M+U!k^& zY;`1f>gwB1Y?(HLMy~tY^uWvh_bURP1RzBbh8=n@$E2 zP%PLQ95;V33r>^-69lI}P&Kva#7oYmlR>`VM>!RdBq@ibP4BEx!~qF0Plvy(*>PZ( zS{_r(x@n-ipCOu?+1KxX`#$h?((R#bE_dy3lz8gCCj)(p-;+`2hM?1{rHe`(MaE^M zv6&Ad1*3r#m?Ahymvc~=alz$oC%HhYOD_0|rG6eqMU}^F{?P|u&mZJzLjqrNZ!lMB z*0ep80nUz5*=w$d-XGbRNv!>?eltutf8B;_&yNpsvBU>6*Id@EM(p6S*an1dzUxe6H;9 zMdmv%R899K%+7bqI&a@;kPi^Fi#{0dk;FXxYj%v$dAPT7pt}UyZa#5B9}veU!6T6u zFPA$>SRm})7he}yB&JKL#55;2y@I;+*S|kvBxUd(l&;b_R;n7ke70VO8r0&*`*GJg z;S1+qs%l>TVJyw~(}vs=jaX@qp7kvW-33Cu-m6IjbFW^(FZX*zkuvsAkkL(PB9Lql zWx)a*6c24o4fbmCS_y_sRwhe@#c=hiCtGV~HCU&6{p-7jKk9iP?Yh%?uxjoqG1C(Z z(R;fm*&5g;KNOAcTR|=PG+aHSMa<_oVW?IF_wxbJkTEScPLn@=lu&*UU4|*43k-pblZjKl?l7L4WmH>Tg!8aC{>l zSUzFtNB$^(KX^j1%)dYoNF9dK>j233QVLS9f8>p>rKE+_ZL&Ui@)wL?74;T2uj+|o zJ<_T)-uO2vmRe`a-*NB9U5@P_#yLRh_AW|y#eqj>q{)u2D*NT(FZR?a?h)(!8CdFq zDcdD%oz7xw?JN&RIqLQ+W^;B&@XW5;ZQ^SHL15z==2051JFy1-If@Dch%{NJsG~{^ z@}MAuIdPjwdo-uZyl#gfzB*_RvO9m<9UnK2vdK-B3wlvkr;?dvk;x|<2xpUV`LPoQ z!Ss1*O?v{1Wn$^t@l>um`ea*|8z6|Xv*C-(nSpxojJb5Eg*6{d3a0ue+nd2sg|@c5bX&##1dj0U{Rxpvd`F3x8;xcTP| zxc#&ejk76-C}N!;^+$MZlf_7Hn88klf`VpUTV2ISq%*A!I`9GI*@i*64QVrY6qy0P zTSGGu1Rnk}U75XAOdqu^kS!+54QhKH2%_EEFlDb>>SHJ{!3Vc06IeNh@ZG;aA;#00Vpt<`4Nzeti7zQdn?k41;H2)I1>X`>zTwrMIe} zt^dJT(%h(}^^W9Iya%hn{Jfg$)`sZwkFQNN&4xsH+6>YHMCR+VmRq&r^7Vz<2$B7y z^Dj;FXO+-^>!M*wiW%z~RQ~LNM+i)E#~MEv{!SuV7}1{##5=&d66M4}u-r!9)(rey zk~$p|y9~pho6Ga4(Gw8MJKtu8mVv>@>kpO!W$lj^#mUD?zZ)GKe2J>h>3*H9tnkhS z&0(e!$s(=Eh#87jTCeb#=hd#~URaO-yZ1PcpiKmf0awLG>Z zWd9j`>^rYiw^X0m;iWZ@2Xf?uGCkvBur9jM+5~Gdwu1{Qo)Tce^q~mb7gg;ZJ8}8f z--2nyw2}(778VbdfG)Q&YXd6S(RtiNLTtTE{8IRPo~nv0pIzg&Ortv+c?;5W^O7u9 z@N5IIx77tAz?M|lSb!py2sTIq<-N{9dB0e3?_C4wncwd$^Pc6etFAPEXeu%<9Uxu0 zoV6B3%x}$)`}<`leAx4Ddre(q;>)b9h;`sKzw@10NchxLj=--IEID$(hyQUV_w3IT zbX8uXGqNJ~Geyi#Z z#XN97h@AXoEIRn<4K&(xxI(UkB<-~=Q1xUZQ@eJm;bUv=^8G(F+~80kz}rNvvpmFA zhXjU|o5e8Pw2G#-y;#95vrrv1kv}l+*s9f=KUp2cN8^}(`3`$K>sc&0P2Q&~QA;Zm z{^15>*6DKF&D}u4`u>)rk{tk~c3p|c8d?rR7Ci4tMG{hk>2e<@!#o%r4h#!&CJmB$ z6IuM9$BKFg@kqlK{i)t^1;K81x*15eM*`SQxBIEGUs_Ek4RrTh+zLQYK`U5 zIz&$Wd1>qN=#{i$&uuM#zIHfZFk0(CkXP(cggu$z(`~iqCABRH; zt=E1cRTFCMj^ z|M`rufL1NW=6^10rg%Pt{h!waS6=%1_pc-#9!t71$$mGP5=8-iv~TLG7pq!@{D0H@ BElU6Z literal 0 HcmV?d00001 diff --git a/images_ml_frameworks/image6.png b/images_ml_frameworks/image6.png new file mode 100644 index 0000000000000000000000000000000000000000..39feb8f8175d9a86285cbf7763c84211bc52d7f5 GIT binary patch literal 186596 zcmb@tcT`i|yDhBJY=DZ0^j8s45TjIS$*VM}kuD(cTIjt82#JV@bOGrl(mN=4%0wf_Jx$$@J8RI+S-gEAE&bWWU+GFpPU7od``OG<=s3(TnoM#2jo;Y!W zQ|GaU@re_xwkJ-U+B(h7JTjf?+jio_>k~Q}4^1JqTRD*ki|%m-#pHxubkEiPV|^-n(+S@Rnxg$i?-oTEAcOR1$3`W5!`9 zBP7t~-iP1BAbWha@5%rYf4sxDC|vAUaGqE7q>B0H?4PD#rFYg)suypkoCJ~^E4(C6 z0Ld%)M|&EDrtD9%wiyR3*Oe6p^=3oahvVb)%9IJjXJUk< z7%76ZAk_1j?syntwOH2n(GQo^D|b{>E^{PT0D6P3TF7vrutQv7+$i6J<<|hQ#8*O4 zbdK;TQRwPM|4BQo>pvclVJa%DK=ROd8Vf2`Bgva^91w6CO30}Ilw6ROuhsVds^72Z zFr6{5ulZ5GD^{6#?Gy(&{v4`?9lGhF{jq~X6v}VFirQj<_Of9uSWx~AV!F6+ZdB~e zBhTNFvbGwl(Lc{987Mo|ly`!!~KM%n|(=BDO7%-2N{N+%v@*5+v048fdUY4y)Mu>UPZ~TIm3oc!oCC=bmM824=ca zLScouEH@5Tu!1CsqtH{J2A`P~vp1qC6lY)Ed=$q?1VAS9R@*z~x1r6h;CeL!bUiqf zv-C7eLmG5QO2%Q|DCMB|WVouV`LqpCmZdfGkRhiLmsg}I`cR9wlE5Gae zp2AV};isVzYd^TzV8I84VfaAP{PSUWRy9gq2e>9zj&K~W`|%nf=aU5JGI@PGnH_~S zijrYP@$CPOJOyeG9_b*dNE*HC2z--Edr|@f@5 zHjs2D*bPj=&FtNPu%g__KAm7a0?*8-8o5#Rj7XTLAnF-m^b+mcp&vI&8vI7v+r4Q^ z&}$j0L)9G z*4v`K$Ky54Z!Mc!owCcq&78v4c7k8Fyr0$8!@2T6t#Q^0Wam44iPp|%pzd!x&q4dy zL2=91eWawrY77YOVV+aHotX69eD|u)3LuG;fuUAQ!f|8j)_pq}L1-^?lfX_SsjcWa zgyWl|I!H5CwY|_EZUkY)i)5I!-eEc*gA=uM^Ay(8^@j*iysk5FRs%) z&pHi7SBZxVqzwS4Xs>YMZQ+oYSEa=6*IC(jTYBeYCC{CL61qmt>Zsq68ajoQ4$YM& zbaLxS-Qditd~gohv>(*i!2!Bc^+E1!?cyE6+{o(GDN)+z`bQ%d`l9PP0`(A%%#GyV zCQz0q&zTxW<`gG2?T|;JXpmq-Ny2fpuIVZ4!Hp5`4sPjjX)?@0TrvzFd};jwE;4oF z3KU&&DWxM&Ey_=bpsweC4uv&~(#5sqYAAD&^NL#8u=sKfae@u_y721dC2PI7+SZH0 z(|x5M7V0H}8rV>qVF2l(*t7iX*t?xgFUyz{5sHdF8eQwc&4`l&?(mYUIZ*W!-1|abeYGKmt&Qbcb@F{_LV>ZmG!O0fS(Nyn zwO&=vhKhF%6HO6t$-qfi5TyTX!_Gz01b_W=w9>IB{WQ#C#h^b=8GFEAt(qxawH|hY zC>4@k(QD6{skzrRUvLtA9&5V)yfQm?w?F-(d6EO`^J>AF1@=Nx z{c%q_ken)Z4lGFEG2wqCvvU9zhAA&~zxj2kX(MIz@S{+*V4S|0Gb_sah!xdX$jsf6 z0vY8ar6Y*mlju4z8F@8dDR zOXkm&XzFJ!LFq8WLXc0}pT&0Rzj{zP3d%4Rv%}|pNQ|O-+=-*)0xH^`y_c3QP<=a$P3-o`TDek#7FBQ0@j=TVT0+O!Ek-1^{+a)=W?t zVg*bEtfbYXFn69NHEGd)K4Sl5-xVed%}q^Mj-f2O+B3HpcW{62UN>#8p_Yvf7sF=m zUScy?&@p@@T(I{laOsd0X7w7MprVz=L&m za@k3gG-31<6h3>u^#(!R{*L^fAHc&j;ZoV(dl&vaxaCH*l>S!{@+u)YjL*!Sw3XNw z`1^xNnU__~5exPp!W1RO=3_%B%A-eJd*BUL!pI+aE`f7ZGlCDdbkN7!o`Z-@ zxV+t28hP<7Ge=)jXc=NmL~<((nqpygb?gt(G;<1*{9D z9tKg4kLL8%!tooDm1DoN3Gn4&Xl8naTmuyyDaF zF}M5^5zR`NKmUZmdvgL=5zgS&{YAa}U+7ePGlLyUv{@gXgK~qiiFql!MXa&X){RdX zCyxjJ$u$W*MGnxoz36!6KQ1xUPSP41Pnk_Lj?Om3$$11wtC3&GA1|-_v_9A8w^2cU ztQIzMfK`jyXf^5IDxlhx78-eu|3m@=*SzdqXX7G^#s2ju-YHvMdvO^_uZjWrA3N>p zJUQo-sa&ngIWx{1X`@b#>mG)&u~B5v1!gC&g#6wwJ5g=w&(6qN(;(uT zb2w~+e~PCOI`n##hiS+|V2&Go0@wBx`%sp!*O@(gSWe~X2+Z_jPW7J_N0zje##7r$ zv%0c;lO~KZGrz8MpH>Hr+XuoXTJX^7Xc1UZv`9#MG&|bL0G?&62^X<8oa6!&01Nux z&~$@WiNWQrHz+%6w-NwF<>=+rwCD1XvH zuH4m!U(o=?#G>C{xqG@#^IU$bW_{48gCsyIA!PGoEg4w0@jFjiwxk^lr00o6uJr@Z zh2nZ?Pjqz7$ji&G&pseX6Y2`f!Kct!)lFjN+#ZSFTWxIkD>@j-$M1?ft`qktJOjoB zUd5gC(BY1lkCE7s=@+_F2!mgZ?ye)GNYS@-S6{)_+7u@zCbS|BU&ZI9iKf+}8xPH0 zz|%(vwAvbIXm>Z|4$Rlr_v(8|zvO#9y~0gdj_BV>G25T}7&mEGy5dh6m1X_9*f@u> zhiTJNQ&Zm}%6(}QYbq}y?O_9o;eM(r(P$~r4Ss;lLY?ruxPCx4< zHxyCX9$8NelYx~*eVGA^7F+r4A+Wc?(XFp|ZAe;n8*0fMwzl_J01}z#m%UuU z5Zqq^WWwLUke`1ET&A5DlIs|NivDC=6S(K~DT*_gyih9i*b_y$xjj`EX~>E#Wxjk2 zZ)$7Qz@{>M?&!e5u0Jn5u~9HEC7>Y{k)25WD-3X8`@*{pyUMr5>u*PF{@ho3c+E=6B3~h-J`8 z>)dcMOI+~cneuZRmozb$kT=hqkEreZ>$9J|#SYraM01GYqCZI}l3ICHm2ajO1JGVb zg3WyN$KAMdTAKeg>3v##8EdcnSk#zN)3 zL2L0{6+ixtW%P=|-F(QaxX$qWRHv>ZPQPB^z@I^@IgThk2Hp`=fTk@2>Y)@LJShP! zL5ppg_o?6?6Hp1Lq|6ZmfB*1ax{Ng^+q5LXy0IxvxMe;xSiAnpAiz07ZTiFQTT)!hZgf-SLKnUofGKz6*vuBF+19OWU;GaBc|1 z+X~!&(M;hYkiyW4@a7Hf{T~vDr=gKPJLvYmMF%PA!3P`mDDFQgE@*;-P7jovlA3IO z*B3$`dJLV-{o8Nwg#B9;=?IgPgwC{GQ~m`*y#><~Ok}QVrqw)Z{DAiall=jwFhk?A zAzKHdk&Hw-tt17@zW*A2S20$y^HnkB<)aXMg|LC;!abr>r8f0_Wl4ASRj9 zfzCyERA)_wsP_?1Dg=c7%70t;IQXDd!FjaZaaxfG zXy2EaZjsIGxe~?UxtLAoR?`4IP!URh&p$ps?#wVSSkY9fjzK66{Q)!K^gzq@$VK4Y zHd^gHMF=|;wTLh((2DrcIk-D7cDE~(9>&66jsbxa)DF{w=y&MM@GATA<-UHrGKyB) z&xy7}!<@AjNtkf0SqX0a$Gd--V@APxR}~OTOFePP22PI&Y747C293CmqVD+S!pWbX zM|*BM{sq3UA^3g!k2U?MiHbp1o8dQzG(tOdE#y6LnskWueSG7MgdQ(T+2ZBI&b z6LF61t~(R3J*z`pg(vMF@96=Ct8LoeAm$v*HGCX9L+3;A?KfAmMYGjChk}#2M3voO z^T9-SqD(&xn<4gg$vnaf7r`NxPY+7fLuXNkKdvH5oJWf@I`)T@xz4sC|l(yCR zwl0i34m5^FkA6fT5C^fNV8s$$on8G;`0SwK7W``XGiw^30ETt(;ah7DRD^=(tbRkM z-7CA$h6s9OJ3_d=X99TC5c4ItIQk{-n zC7|x`zX0u5p%ww`h^Zk$+oc%o<`(*!a~QDl0*vh2KvDg)KOM{O@Y#cUxV@r;SzJB4 zF$AQ#DN`%!*0fi1_-3)2ruB{q8L8x*HvC8eW3~6@s){2l@-St~1YJ038!|YC2(SVR zs0I5GROSqLJME72((cNKQ0^hnk=pS>_pz?yVMSX4x+wURj*;d*%{C0f$8Ye_MxDuP z!#sQmOV?Gzo5HgN{;<}*SqxWlkr`X?#+*Q1njZH4vfNsV%XDi~ia=&T)p=zck@~iU zGePESo?+rD#_+6^`L>Pbxb89@NA8t>)D|5icyV2K6 z5F$t%JjGmN(yTKkWL8_Xqmy@;mv^{7AaGQ{l3ncb&T}5*oh-UyTt_vsU%J5?jGJ4rOyA zosX&9_YU~7;mAI%dV?L*P#!T@R30H}G(HkxiuRV47gf`bFn4W9O2E!P+UDdaYk+b# z2w6Usj$AR`Oazw}`6)F_@|xtuHJN;0;HU|p>diQ9S36iCTN1B$ILTq`pv`gd<_%e` zHPP1Lh%4o;r)ScOwCcVVX*uA!>&8ms5EX^*T`b^+W--A0jR;Yb!BvFO(xLR2xrsVG zvFzSTi$Q^nO5kY6!|(ynw`yy&LMZk#oR^L644yZET&OH7ODvbHD6Om?7>IV}V^KgZ zYdAaK76w9pB27#=NWCF(x?dtf{Dw zeMaZ7&VBN8dfII#aPG+eMVV23w3tgTZ6j0CwLp5v5r=O$15h~-zDBfL^&W?zwFF~WQJNnF_=1W znBP3*T@4k?Q?p?I*qrgqm;8O<6?=EHH~3~t-q%JvlBmEbsPNV8^$iNN{Gu-a^?RiB zddhFTgZ5pmE+C#p960e|NcxGpyL3*-PRk0#vt{{8BwP*CxIQ25vE2MKCve^wh@Wq- z@m*Z#(0K4y7u>t~tco3qMs)HjRB@kezF&5q1#l9t(n%`JM}Bo#O~X^R!_v6N&OCkY zA+xmn#;Hfo^d-A3q){K7YkA;f>=;9;y^l@9t>(u7xadB$#n-hK(a1$1Tb0c6znic_ zS>haMOw4TK4-?j&=G~=%P(tN3(tCLe)N0ee#29y!;P)#`2}i<4ObjP=4tnuF7%_xt zd@JL_IT-UNH5ZHIUTMjv2)W8qo3fw4Lc*-x@EvFJ>aG2%zzm~$>4uu{O>Iqhd*UI` zPeu5u?AtyEwAO*0`=`co9J;Do|J0Ja#S>SGvWsNKL(Z(oCv%x9r2*pxri|q(Q}pgN zmucelhSa+SkHi>(p{gG)*n|h}XC9q$pB{Z-=q_--Vp=F76i-Z%bFz{*$gu;9R(pJE zN|XOd4s>;*Z3#>V-VtK8vR9aFzb^z?DAH10>JFw)ywo0et+&2^yWaCVizyUqVCv^d zc*cUy>q_oU9~NxLlG zEB+1sD`>`~=dZtG_yEP42$*hnx8~T-Ul&y2o#&yK`S`NNS&@M&*$KenAjCkyvQ%o; zV5pQNQthK4!LWRXW1iAcr zdF;yy%WFpUCGX9vt?Zk9R|kv0;UP24&UKcFUtA6(gLhh1oz{N#j;h(d=T>!Az@t^O z;8V5Ba-Q?5zAgkB8_{_%&2>QX6bv+wL_8)wp8WovdhB~XV`Df9R}zAJ+F4)khKXyL z3%aE0BRZBw9VA1u{T_7uR8uxoHhAwSL+Ggv$Qrx#vrs2F>14E5^7Q4)xh|Ivg%K{V zh!*f&wv2Z9H;2eOI^r4v3U2A^k!il)8lcafLSKxGSrl7M-Dbo9%q3_P#6Vz$wC^vk z9dwACinsx;-L6>;0A2un49Fr=8ooXaE;kx;47QG%oyfDWvvYJH+)E3rORc4ZbAVp# z!Dz{%D*O&L8LvZEG91(86Ye7zZ^(2K$`gohnum@q*ug^|WGzQa!Q3@%%q%9;AD+qm z!9xdji7cLMC4{^HmD&dqXhVUT%sY?buuX?ESwxMvE|GYV80NAEA5tBdkO7$7;lRCL_ifE5D66EK zkdI@*&=P#BI`u-QtVc`9^McOQalYYrTfWhud33Q&X!R%I$>ZooJKrx+ z_z*I%-11Ah)n+QOyyV zdh?$i`L-UP2sGkB)`os5$e{qCGlUCR>P~WC`t>TRxoNK2ABP9QW94Xm%j z)Hplh@Xfork|9)i7a$VBNSSS|aP!{YnQI$>WvCKrENe3sW?tPw0 z_BlTCmm|hytf=|??1*$G6ofmkCyAR{nL8XNOf@@;8Dx_ygRsjA&-N<)f#OlzXDQ5L z`3)l%q_w@h4irNSP~VEcY4-(W9nQOBzot2*j$P+g5IgH(P`$1f#LP5ihh zj|8~u%G-Q-*U1r+C4>wfe4jM@>G;Dnb*ICvseL*XA6X(`DQgR1H*od>KtJ?J zGt*0=d3EF{`%Ai?%`IzD5+Qs+HBa~KjX-9`d)3iYiTs+K{tYlvgIq)` zci7FR!^KIMaM(imEB8r)6{Sw#W=q$OCD7mA?RX-hug-L!HgCt&A`NzkOgM*^{|cRN zyW#Re>p4X!=t3BWksFdn@&(ErEMxiCt za!x9j#pi;ucP|aaFD7T}n?e~~`~YYIU*R!Z&zG0l`m^Ve`f&Mb?{6ktOZZa^dwyLw z9cd(o?P7t$|5^}L{S2BSkn&^h7SB(zAvodiH+ZR;&?Ca?SR{-=#b#jzF?)*)_0QwO zMCuI$sOq!T)h_IODu@Nz8h!%HK|ok`M3b zVz|AORU^VUIYSGgMd0n3?@+e9v~S|QU-u3$8v`rIVlzD+;VfW8^v=4TO0dlfb}Pt< zmG9r~`}u9u)x&@PH67{-nBMKR`+2x|w;(heU$P;7P8de)Y%xVI*p}NKxyp7Jx0Nf( zL*?HJ1R56hR27tY*xQv`?n#wf`6!Kt2Az}V1)md$WZ|uHXFGZka_;yzLbui9mC}KL ztszOxP6ym>tv7mQV$_6DsnGk}PO8H^rII~r&WyIB*P zkjzE5Zp#ZRzcN_V&OQH7rJ9rDS+rNAI8#nVZSQZRBpC>3x!o zi?mcs?k%$kJc0xly61J_K=j`0i7FfI?d_d1qqsPO714(roWbc@v8Tam0t(P(rPC1- zX9>9p99%-@=QCWK4>rc-0=BL)xyQ(G(wi^)J)1RIq=Y7H(9HR1!t0**3temBN}|OT zz&BYLq#saX!si54+G&IId%(4jdP{jiXQU{ryQAPuwA9{TXBq0vmW`LyxEU|W(Y#WF zW#y{PvMysha>wKA`wQ7h`yPa$Bh#M&YoGJvRGS^?{8zkErR^LXkObU0G@L7F%iH&@ z0-o(?ArjAl4gO&Ys7H@x+~R5XfN+Vp#1B}_4vioG%A4@LTQ6mxBc?kcE#Z5S))(7!nQr90 zUZ|}?SIL?m*f8@9JXkAh;PCw7uAdap;XyTAiT~4Ik(gegUTCtQ_j#$PHs}tPP7B1( z8wAdIr&$8O(wdKhKu3v`u0HtHPz3zJl6ZE}XyO)@k>-lh&xOw;{dlYqWZ5`G1jx1E z7O4i#sd&fm5OKzj!9vQq3Q(-=Hh_F8mPNpY1Qq0R;<`dkdJmjOLc8#yF|ri;OFGBy z)-9E5KVL`EN(6+XNRTry+ng)h$Yv;AU5G&+4NEJ%`{UZt4>0^iE;e_!*17BE0qsX% zd1Od*3jgWN=NR0^Qw8i@9v!>yWjl+jp8prG$|Y!t{*BN)hABOVN2)Y|A5!+I?64!| zLJ7LAGE5}hSVuAbb^$0sx@;wO|TNfKlvZ6|^Xz|8ZaFOVz2(iu1S4hIrO_m>YIqB~*GFTQ zEr{eW%Z1dX0_T2dXg;LRfnokSCjz-GPep$Ix>QKbkP%H9{aR@qtS<|EkngBTuBlv6! zcti!Ui(sV?=~i9|$~~a+`-k5{AM# zxH7|T$wK5-)qO!ZMiV3u)54U!@r~(ODyk-o8yDQ`9rSxqJ{kXUMX=(pkiY{2%8!-S zPQCT8$2h_zwl{6~#+x~IXI>_Ez z4It&sd)VFYm(EDwU_4jwinBujj+QkYym|)jrt<>^+JBEgf<7{cX^Wu&i)lxJ4rukm z5@`^_c{rADXc_$^DI0e*wb=a*L#-Ygo5=7F1O4X7ZVfD<5cxFJ3Ky4kRqIyY33_}# zrmYTg_Sc~4%FbDLbj}?G=E@s8=E7$V-G9jBKzj5_H3MJ`IKQ=JC~-X}i0)?3poea3 ztRkcd!c1R5^7}v4n;;9ed_A@!kSR-nuUknZF>w86!>@Ie3^Gmh9O6UR;X7$@a!~WC zDbeLHKFPH(J}H*AzK?+h<%OT~jyBh+Sc*v{ij;_}1NMsmi?Ed*yr&rXlyx$BQNVw|x*Im&{& zU*u0XxK!LKI2zn6-CKf>gn1|YLC7-wrH8ZQk&>&&vfsTOtKW{G898kM8EUrdtaujJ z$-OeIo!pqXIL`Wi7QW&_pP%3ihFlngk!&RAt@Yh!;ro=v#)eFWp{SBiVaC+g{y2ee z(VV%Coe~8oCV%hn!M?|}#>iE#g8%P@aTfLPtK5}@he9+gRVV@N`s}VP@@)@DKDQ@U z*p>Ya1T)+^ZIvmzDpvP==a)op=EJM=FMHVd22S+5)}S3y?{w?MNmi6Qg%At}>yCot z`(oEe%L-rF)n?q4-F)MSQ}c?Bi5cvRe2@!`$Kt4>Oh6Q*_TY})7VB~Qawp3fz=6+Q> zH%2WjvtY2&jeZbXt%&LBkEy58p8goPMGZ-k^PX`~V>DLHBwAq!!Gwyw@MC=p&|7rAqpJP!)Pe7Xmi+}*E&&vU$c_*dw;x%aL`xmtoHC?c3% zl2jAZa5T-z^vk4j!UXy?fB)}>Mu}9=%C4gPj;d$g?zwqGLGh+?vyTg(kX4#daNXylAL~2Bs2PRt<;A{%2 zilcQ!PzeFg0+o2@rXu4F;9=x7`@Y)8(_n5o%Ww; zsBdD}_<4OjU2fBG2Lhuc%zv5X&T)Lpl!<>Lxzp8blfu&OVYYT<;9(gTyF#i7H|tf; z-5kPA;m60#V=xzqc$#=mB*PlK!*p#}gnEs1|E=qRIPu#yKjM@pt3 zm}Dao8pQ#gJvLBJ#=`A_ciCi{iSr<=EVb_A){ORO30RTZrJr@^0y&%5QdWOE1`&mw z(u%#B;ZZF(OYCIf7>GWRd$}_*h~R*vL8)(GdI`evSsR&gufZK8keNr3IEiAW59#gt z_iocE&%T5N>y@-VFW))Sg>(KST!a82T<&}bUX!WN15s0L{s$39X z>D8k*Z!9+TvK&iWQ7p#VF^(oz0B2GMR@VJN41Rq7NH|keP5HJ>Ulc3xL?3g>ZpXvH zG`_eT{#ByrE0i5DCmIegPV%h$ZlEx!M&;nblHAH>MS^-_D0lrxoXEF9$P9BPz}()w zB;Uxy?2gDn_e&~XNkv)(l(xSRx5k)yG>7feG#gk8rOs<`+(Hn$GPAKk47~CG(?83O z_2xO2*mZ=94DI7O-@%d3G&cj}P4p_phen_731*$Ef8l!0Ms`1)ACf_aNxRR@Jc?^; zGTJ>pcA_?<84ar`8h7_D*v8tjF{&`(JZ)4PUK%ZoOHCyj!XCI}(N(%atSAl81h~v9 z3>E6aDpx{1jkY&Xrfp2}!OmBrVSK09Vr7yIW3GGLlQR0T<-o7`YtHti&A**h4D3uj zQ~4uU01M^e;~{fle6}uQ_2{?p?aEVk;-(LoAx`fVEffO)4Kzvf8Y*ub=eqTZJzArq zMUZjv`ZA)t2f;1zWv_1X(Y><2 znRJ~)kP(B<;zSTjNQZwDQFgn^DX2W)_BD%szUW#4*L`Z4>cJRf+QRq+eIe7ai92-5 zJ9zJh#Gp*xZ_zsGft`Fc2c-+#xdd1N6p^LiQFxmv)fV)|D%3H#_&VsL-Pc?Qmv1x5 zcVx3R&5iXsrfTiasmVV-rFLbO4^Xt&^@waqZjVyjQuEn=Z>hVibb7|P7YnzRX{*CB z-2UZ(W{MYETqEn+=8pv)S$dm4TU~k8^nDdXJkO1M8v^;(tJ8Z1sgn?t2kf1f8((I2 zTR=RyL?BUIkc4leMP&OF$9g3F$kjQb8-09;PCrC0xEdKH>fc+G59)fy8K}kE-`}4^ z%~d`~8CYfX+j*Qzn}x0p)Wc&ii|bZ2j=OV0&41(R#JNJo;^Ja+SXoT}HFXi$gVmfs z)FL(z8(O>rr>}l39qPLPEhAy!0D&90lTu2jFE$1d2Q7ab4`wdsKDq2VdUHOjP$hJ= z4SQ~9SKHyAH;yMMWTZKve$-Gl6uAErl+-*%#DKK>Q}2qmNyEWp5+huce0ZvlLGM0` zmRgWi!Sd=YthCN`nZ#8#B5%qHb!=t~L3A$fQ5dDnMb;IT_StZw1axfIg&XvD-wnKj zMDOf_>z;p^No{}kPK(GFs#CuuRlPnyY-3uz7nKJ|{~xFKe{L23`T+00w^qUHH^`K4 zf}jx@2S#Rm+=hPfc0QFU4tHGcgL995wHZ*x+7{&2e>p7mxNu&kr-7?ICgmA8C0F7J$_PQ88^Ge>z1mJvkl{WE|pt zv0Vl|k^RDNwE1l2p0}6XkB*prtljh#HTCA+7^U%B-!0AdG?|x+87zjSy;LcqoZv>P zIx?kzp8j}{jh4z0?dB~2=gx70!-=xs-A67;&4}p@mw`(+ju*;a7d?WM4|kk&_<19H zcaiJ<+YscPdv&G|mYJ7DNN>-~;d%2hE2C2-64f8-KT)fjY%iN;Y)E<<~2^oyQ%Y5!EgBi$YH zD{b0RZ-Rs4dlCe}dZ~Ac16DPPpDwh&t8=%FPZf|(I23g+DK758<@ncRP-LEeWhILA zv8dBTdi3&d8e~6~GOv7}de8ruinn++*@9;@b>oYZl(*zkkp8`ChQBwdv`7a%#!`U@p|PADrsJZ}W z8(7+5(>wUlx?_6IaL}uqO^DqmrtI-^+=4L-1=ln$o^|0-siOW~vaa6LV5(FSAv!iG zrL3q;ox-KpqE_YB|4DlUBLbRRCG*lqZSMGvXx%fdK>bQNz4!5EQyEjVqvBK2&n7jv z^~>^^=D2~I?RsdJ{NjmI^1H?b3&X|*D0I=tr0u7}E3RcmG-RiWy0yP}%6VFa(t#66 zcyjd1Iw13H&s_4rZLM+D^x6ecQ(TeO?w9VvP`2G-o8&G7NI{X-_5l-G*)JH20bAA+ z*Ue0!V{?TH6Bfv^H5*$OQ$q`GDR7!V#-P9FV@IUW%tLAIg!Gq%f%6x*kXx-x8f@>o zUD^Kjx@nt(xI8buxUoDMXKSslv3G79fnNHC(7tBn^3kojZDMMvn>G)@*)PzXaQ3Vz zsjafM#`sT2Rk9g#Ndz<-Lw`3>HPtN|oS!r#jI>%PW_c`?YP+uy%g( zf=L|Eu17Dfe2_@w#(Z8Ni2&Ks7-6C7BjPI`abI5cWKuKtz7qGB)9j|zB$ZUTvEJ7~ z6o2tcCpRf7BfEsEA}!0g_gwDwrr*7$labRK0w5vC8M{Vq4O!bo8ad*~jfsmonLSqP zJ3Hn-{qGV5>sc%Sg~Xn`A^?k33;UPIIFqR5L*>{{en=wwcy zadO5GuYydkjy7eXL+_A?$ztgIWVAeoF=_NS+$4A~<})0n2waaW@s5ywRQ0SSX-EaH z+O>MDZm4qX*SeO5z86xS9_sz(foPWy*OO)%r=X$+!WDA)KxvisQ~1+!d)kW!19i}W z-EYH0`09Y{`l8<5JXy>GwdjMwERR;5D$iJ65C{Plh0~!o+aLRUV|L~-9W&n~&UGWU zF|b4UegRJQp0*1Kx2)lqf$8=T0O-^uB=2mIFcX8vN1B)kHClDH|CQ}uC^4O-L4OL6 ziNq4WGM7v)6E!Ab;QuUP;k8=>zAurz=ZM=yTHe#~IV$UW{dWz3jL~pB3q0X#L7O_g zD9zS|emxah!|dRBu&rb3-g1%Ysr$G79o)_Px-qlhIX9!r zGhdM>sFjw{Zp*G;RS2(k`lJ>)Q>J7Ua9Oe;Xlu#IOi+6Ir$(2f`LI9uBL|3AZZ~}T zlMOk4Fl#fF_yc$h@txCiVtl5Xq66eT_tFn9)3l{LA6wsNs7oI8BL-GlsUoPz@-%4~ zCFw{|UaNm#WY%W&h9H?(_S2YOAui;0y^-F!kL*1Y=7iPA)Xc+noro0~tH#^&&zff4 zs;SOiXx@gMt;9@RN<4z(iAl*~ZoAIL$gx|@;!Ow=G=CmhKRIbMcLZ=+9b~5QHjmbk z0UzgSYYjG6OM?&_X?c&C7YbY+lG6s;aqBK;D6Ts2ug0j33N2x#-k$jvG0k5PnCsC5!}FFbFCEhm3m#{nhI($W=+8OCv$&J-@Zli1rS? za}yH|vSE1;&nOCq9qh{8ApT3-H+R={ku>M;FaIb)UJfXlDi3qPe5TR1pGs2BJ4@vV z-Y$6bE|f-iuwL5p@YZOnvKrqO*V|^pp8EJm#{MYJfDV8*(hNt;F4=#k@Iir9H}zH? zIM+?;g3EcbHftfemob;$gDV%(NZT`CiSx)aPV*m9?dF$G#>fsm$f<7KlMFt} zaGW(VI184K=HRe}QBq1J4t_=j{U%(i!=L58et=s_40%_lQ{B3uTld^WwZF$KO!p0d zy#26ER9k|-8>AZZPFp5OPJWZPx>OtFpsVWT%Uq;o7!Zv9tG*kaj<`x;J>G|RSzg`6 zw5;a3D)>ipwAPX}u=5SdYTSjZvL3@WcPX4#$A0Ro4Pf%}DecUwM+Y%re4ROs!h?9!5!m#7~}RImXd zS&E$cr@^&;8smJjylUO~bcgHAXP{E}}y7>B4YTHtbxC6+x}?!>ey2z^nBh;; zeVd1c_|)7V+7O;0-`O402(h$T_yQ*sUwNCngo;SqT2I!+K4Dp6Dkl4Q27T@r| z=I1NT&O0To<YWRwK2LfLS|oB9TMbT}lXcK931b;CiF<%&imLU{?z44{ z=MlyICEGns0}}GWDggs`@)iSX2GcKMqeJ|GCM*W}7Ici=#te;E@-lC1r8UZwMdiX= ze&2J;rW#Ic4z8 z-&2Z8q3EFl2CbAK|DXj<8%Ix1jE((lJ=h%l@`^ShV&MRKhrqBHAc0<TEW);V5sM2^HQa1e&P`~%2X$`_v0vn?6h{# zo=Fhd%RwDD+f{4gZ5frh#e+2^5{{E2YRMIAt%=Bs?n+Gg-a}!zb|81&_19GjOPx!sKb;EsrC zXvSPyfCE52->awpD}|YRF%bTR^T@Yn`K_Cv2BXDj7cPzFM}cDg3AvzHC30yg&<5F?791|&$+n>nFx{4SD%$`MbV=pZ*$K#%|nj;1%PhNL59FaW#IT9qd(JbuEX5Oce22?i-7}2?}BCT)!m2G zkEUH)k_{iHF+o4g$a7+(ylQZ4 z%&q;26?oLjQpdi|u3!;8=azcO9q%7Cl|Hd`{5AZ_fwp;N-M3Z!}3$+ zuEGt%Hu{{Ke?)9P6JCIX)t)QUN@87ZdzjUM%60*^rEg5qA~H#6&9SeH$LaGr?xA55 zTW-Z1c9r?_1znCCvgHG+qXz~31F!p~m-^Sui39cyEk-g+-OQ9wzz+#x`ae>3Vu3?A zOY6*1mmkn+#u_%9icSw;$9(x?AK?R8;QO7mYxJBQ2_{fGP`RyY!gLUpZ*@ptyKhk@ zdspU`vSF}V1G5GsyC_WBG{0s)FYs7QJGRi6m~h3%+TX zsWDoDc5xU=2=S@3jNX4rklh?q986{34J902RduzWZQgaj70Fwj7gGx9Ld)pJsuUEtBJ|FCR#UMofpSY_R_m$okBkx5_c+Os!e^1S8 z4I}WSrsq&-$GTK6vFo>3Y<@>Fn-sK9F5Z@N`&UOYs-x|{*Sp-=BxDJhevL@)yO88v z!;M=0{y%to52&VsZU0|HK$?h1lcpll2?B!Dh$u)0=~AT%2uPEj2neWv^xnG^L3(e} zYovD)dWR5N0wMVy@9XdP?tAy%_tt+c7BGi!%FI49duH$NXMT;XgF{T;QkiTT&e|S~ zh`v78VOKTOcZz3Qc>IX@-PYsmSnr5c*{@{hJ<2-A$3E2}XQ9NiEv<1yJAU}DZ1sD4 z<1VW=CrV$d^;kCnE9AJquLrH+U-$PEd@kUsBt4XQs55KL%Xn4mm<43gE(u7($Caa6 z$1!ugS-9{G5QR1L5d-oaT#>F!6Ha$ts=m-F%yT2E7Grgmp`mjEy&BWVEc-gQ`GX@| zZ-HbOeWKWm$LNvG<^!BV7l=x_hJK^hC5w?Z=IY7 z+o~Hxjd|F=FlaMX08GAsFo$Gpkp%y&>kOClH4xSIUE#aiBrlM0IB>Ttm2AW%>-i#F zfnjxmrb&HeQ!~eQAigY`AiR#Uj1SRHYVxep@C_0Lg>S$Hygda2ZBAgm7Ific=PLwM z>3t})MMMh!?LUE6;l{kWw`J84%Kq{EhVn3(BRTov2->?%0dIzqA6r9T$<&;WwENy4(YPZaB9(ofBhHo-Xl*^5YCBtI*qCl- zZ?D3Zm)A4cHlz${dETKwQP2}`u_l8~uV0n6^t4z^-rim{@r4}#x^)P1m{KjWN|7*0 zyd2QK7XaC_q3CBl|2MjIcb-G9zMC0Ws=R$@{3>dq7L6C5I)Q%V%w$ZB^%UQ+?OT6% z?(LK^^Ieoxy?0-{#NO$<*hRy5se#%=D80r>cD?oyra-&ZOS(FqCGUetezrvURfi$% zw-V3x)^>o1d4cuHLVO4L_q=yHzdcB+kG?3d9`;O*%C|SueP4ac>o7g`Oj%59%!VoD z?OKhs@NZz{S~%RRU~&+$1<2=W_Q-+x{15488T+r9vA*)AY`A0|rjzK4)AfgT;vYV5 zPnKnEb6}0wv`$8Y{P7&s%<~E9j+VKsf>D>4qv+&`V!g2<9p0fy+Vxef9YAG6(`2*} z3Tnv&^7NC&6e_*fYfdOBt<~e=c=di4Wcb+;OVRNgy%nupBM4(xi1^g=!Z2-jt`?SR z*LnhPZeELp!xOz+CQ0}qyf6HV0StDYpV9hR zKEFuhyatVvHnH^C;vA)wSl0B%C&X|U=l&B$004p{{O<}GH)RftwOmTVTc#K7zBS9} z8oVU_YJT|q+l!r)@Az>Z;YypfDxO(Cjdu@tIsWCcoSPUw;9jWFoM{I)ni82RE^joH z7s+X@_LOpCZb>`d)@ufMK`X@+dI)kFUj>9A2l|iD;?93O3>RwM)ZlGy$o%Gb$57T! zUhnrW_i(qWinQd{ffI2r>_I z(D8o-Dkxc5P0*b$6x>f&UI8!@Kg z9n_rZ+fwQ213UU3#K+^tFt|MVmPG(eKy$XBf9FWrRM#U*syaQA8hzM>T3|p6MK2h0 zssG@#1ej0dz5viA%+iCtUu%by5dcZ>!NdJQ>@y2NsB!OSe>Cj_9u|m!mmmJyQ=B>+ z8ovHgb6}rCFH@QYsDL-s)ni9~sKU zh|>>F9`S!}ZA8%*@=qLva;+$MxUr9Xcp21G?jL6#0JN@xH4BV61oL|4DjuF>I`zED z56T2YfE47cwg1%Ur=FJ)e0ePv{afjZw8xz(6UVrIVjgaX#|3xmC1hGa@VGyu1w0c5 zP-JnvB~y2~wapO6*~nDO`EOtZEkXcWQ$U2q)mgYHA_|DnHErtT$V=y*&FD6DN~FW( z1_BI1dS>EyFy!&^`l$Eo)`B3OonIf|fPnitb{y@rZu9>; zY2hCtC;)_pzw&d!tTW#rw=cK;C zn%c;7d^q-fcj%@OwZy#wMUp+8$7scC*v;s%MNWW&1K)UCXorlsjvnCv)arjwWw4P~ z69`UfWR9|$PsSghSw>FYf1_0YFLd62Ka9-uuu)u9_>MK8t}SRj;j{g#qFNHuZmqvZ zgX^?Xkg~IW!dj-zvs!?OUB8e|;9L8M5;^>7ujn&jmplvedn6Y)eYJGcm3;t&i*+FCm-{Y^o9AdGH zKec#*Xp?N!uJ~qbQT?N!F+_o$`W-YxfusK>TJ3iX+xGZIfv|~CW!1c6wOe>tt_=ED zWFadc-|`!MEOfP{qK)4F-QQhuL1DI7LArwc{By1rjvheH_dLnTT0llTg0rT;2k8I^ zDz3ssed`eg9Oy55Tdcl@z(?9@DNO~?wxkeqp9T_mJZC#Q(xD@P9P&CKV>*4my>?9z z$yK@+??+gQWyJgpFyOS}6YKE4fs9cG4gho@USxX&A2N*$NblM!Ij#9`j;R~BBpu!l z{I#i}cor3fG*>je?wc_t%dU)2;6SS-z5ItkWH_e7|2L8E4;`+%t=3M0{jVmY0gd8B z;ai+9d=*{}pt!!gy(~tXBf7Rl9{Gt;&m1=QlI8vFq%lN6xzTeV5d>LH3V?9wC7>0TLO|M6gOA!~BhHQ0~20S{q=%mw^EYppZP;=wJ=qR3d;7B?Z8O z*y8^wrcVbDp^gBo(7&FbfVee9Y5V7xHfoZ^R3eH^9i~Y1*GXa|*x#=f0uDTMvTvbF zIitQ2Lf#fSwx2cfXp;I)#k}kx09R?(A};zEGwSS<41#vZPQ~xcF3~%6DUGI2*qV2}> z+V3v!#yXekSF{Pz^-j7{x?f~dAwVAQ+`@kIed}`R$g2aJaXlOJp=MW8$s?xq>eRi_;O+T;uuXqRnWSJg7{lGFqgpk=qd+E}zFyEM6cxcsjbfAdFmdJjO z+^41FEh29J*Cih^WMf|gSNiNErROV}!Ul?bewrigk4`fPpOsfdH?UGu01NJajrjU+ z5V$11j9(u7g^2}tH2(#TEA?5`crWV&sW3l0(bh(2j9ycSQg|D{=UTBLBn8kdcaT4d zSv3D(0<1)+(G^rk7J6S(O7!H!H?BYIMWN)))+n}NPwF(@XPm1dTba%OF94Ska*F`m zuv&HY$F)5Ze_eD3``{|XtZLfH>30Z znQHqtx}bj}JXcMp#img60pUQ&P+`_jp;|9a)@!~ky>GL3{~6&B$VNZIUC**Q{y#ge zET|YD{9m>jJj(PzDevcf;UUtG2f~+sD*^XEJEmj+?pBcye(jY&hGP&>7n%Qwc8iz`{`;Z1UDXQ} zPHk~ub~atXo!qcBYke32#HBCpugmQIOX*$VfS9m}r3M2F&z6kPEfc;ia{mKYiMN)d z*zeci+L%tnp+TF_b^l>W~|+&>KBaDS%>H+h64;JR^A z|A5j0LJj@2hxrE&?@PAO8)U`)bq4N#@=TeqBJrWUKdP>^h_=@P{BAZ9utj%hP4%x? zY^t`98!_xndlxc|X9MMyf{{}a%rsasd8F>}$r5PaW-Ey}kMor0l+J&((^W zW2fo+^fZlG19z<7@BF_z;3M-hJpZvA7*s?vt7aXUzQ>>#?T!LgTkz zmZOexP_71_PzB;^3567oiC&dpvEQ*qbXiziHH;OHoT~qAUND#kobZARnydv$>etY^ z+tm+?_~4Wf^MQl(MZD!j?i%aj$1DB0d?Bbt3$^~}ezwzoSSGf=3|k&4hjBHyvF6c} z8p{y}-i^zU#l)FnEbfJ%HY%9=559?WE8{knVF5eDmb7(kY3BKaKgVD+dBPxeElo&z zsgpd$_4OLn_p-$M|8RNMCIXyd7b^;)AtI0@+|`;$JSr?Hys{SIBS8I(TLTxV2PrJV z3dKF9&BQvxIG%A4^|PV+K{|1!`4#zBW5N6N&#&h_#mJdp@`_Lc+LPKJQ9aTL*RyVO zL3?>ywSR2XYK0p;WJd~!ytF1I;0CP|`FV4VYvLjYZ(V(5GRO-a=#|U*h6=HazZr=# zH^CUlvy!>Tf*(|V`WQ=p)DIIvX?)MY-b>=RmfMu52Z@ggdB(Cv)yH@5?pjw9sxnOK z_BI#vpHCf^VYQm2p@5MSQojDu>k)nzsSrUg|LdMw_2*w(RxdrZ3YD#cuC1v@aBxA> zqctF{1Gxf8w6|ix((##CFQsRgOM#dMOWl`$j5h$a!A-h$)~K=_ZEz1TFALKCjDiQT z@B^X$a1zf+^uQw%OTQ$lUUC;j0%^4U@7NbYY?1V<(N|+lF)1QeQG+C_MD@8>*7R@> zRSg~-5qQZ8a6iSwwJK`S_4z(EiBu-jfUG%<{%!vTb-|4zpK*2`HH&n7b$GRT!N*eW zy5J57;eqo%$gs`TfJABIuJyyd0rmxaX{J|~V*4{zg~M<6gQEL0o?>P+F>zp&VTW?o zrFQHiDb@#^@y{^!E<_6uIm{S)AZlurtl=^kRiHq;R@O)Bc0mc$7e&8m4DaSCQ-204 zV{Gd~z2AFcK&1(BK~+mL=bX5p-j*7;WFy&HIGPkQpcSt4gZ}xo=ma&sex=J|`{8Vj z*Egwwn$wqw<8F-|7qiji;s>T&2OfMG{jsJ>+bNY?v2@R`t({56t6TMhI8KHPZYz=B z4?tb83={~hB}AXE^O5BdQOja3zpZH{p_CQKTe0V*rcb4bn10l*aJxKb{rjS317ra+d zLw4(niX0{`!$!D41?W4}8%ur!Jik8%SkA^wm;epX(T)k&7vGKQ6#=ZyOEhQTT0gvv zjRIuw$f^FL3Rw0RbJQ0LugP>&ftNo}l#jKJGku6U1v*l4%X3npY>{HDP*iOJ*4E_i zn0h32p21qemnP{o^`omvkbp3*Asiw>)pOe$UXHxM*OJKrp9;+HqG7miFh4I&(cIZ*J>e$STY(|K~T#e0ITg++#@K>xa zno}*+^7VvYA{Yv}om|jk+ItJIyH{9oB-j`I+o*BU%5kwNDK74sLH!mRex}l9sJN_2 z3C@eVuo}x(RZP2}+D~z3`ym`iLY0mB_DxOL4DmGo%?KM#!qQ6pJKnvwKem-y8L%?t zEG*v-?2G#6y7^dDgJ0@>p!_?1%d5Bb8%$P%-k6TQNmIPHc7kPUX|XWDjzz3{ke4R4 z`t0#Iw3;HDy_P!IiYgk_!|*#DrYQjVPiO9>u-z(G<2_Rrg$QN``DZ3;3PNckIcdY+ zN7je;<{~;XGVh{(aZcc`5sl;^O`AOZOUE|1C4%h3*p0y{YW+p_bMDo=<7{HqMvvQU zle$9b&qBd^kcLZGT*eLj(9q6c94Y&t2Ncb=e(Z!V7C)N9>*nD{j2OZ5^Yh11WwGOh zG$$eWwZj$rgF|Idv!(S=x|ME;z9WFa7~A2nO~50kLRdmBN+yw0($Ly1S3V}<*%gpn zzt7e?+0`I1p#2n13^w1oC+^MFlE#_8h!LBH;(BKAB*sib%_|l|X_t8G{auWEwX|s- zeqvrybXcmcqFjH&83{`j&pE}NonGQc8r*A|x1C0FzldB(14I3DY$L!!0~HmU&Znjr zpTU}(l)+fll_9o-%SOy47M=`E!o!E~2B###Rx1P6bG0d-WRRX1uk*!bpYFe; zV0y=i@?<{szUjnqK1Csou4JJxG2DqM{86jvxE0R&Tn9jcmnOHSL4ro7BXNegZ$24# z4*42w1%HkB0)^u)^E;ZiY;}6nuyvVijC_0jdsMg22n($S92Z(>=RRzbxBkm%k9LtA z4oQ#_ck{}rQktn7UPex}hdXfh6Q#E{^V5$Wge-B$reFm&MC1LGE4dilaUBm%urG_VF8f_IU$iUmX?n5sv@*4Nck9;UtB*y)8=Oa#nFYx`v zWyxbUYMRXoOYm0wQ-->#BA9g}7Kdqc-L}!Kc9~M&yrH*#S_&1c{eYKts6Eq8QHIZ4@(T4nJ##+l(8}BYOPl{etk7|)CBB5? zM_0OXzJV>F**k0%U^$#ak7yED0B^eP;Rt=-m~S z62qrpbj-Wz4)Wkbnuo{d*d@kBA9a;WFA8iMmWcc+9!SC%40rUL%XL2lH_^yvDLI{2 z*{)BQuSA5K#tPc<^!>#XdvzxhTLFx875%0>UJoEUhxhNm7-pZo%1byma8AKi)+|yG z$G0Xez96P!#@Eb3NaezW+pt4bMepuV287^Em8l1~mlx)?Qq0qVb@Ksg$vl zt+c#(|7Fh(^kDJwcP9gQVsnV&a#=l^@DKU3c_)r(dn9W28iF z!o+nPYcjE8Nt`ABS{-M9;J zpu?{f4tpFaQfvbDl^lDf)~eP`+qy7wNj0%P22~Gcu>a^kf5&P+J&9({=Q-pZ0O#Xg z#CyOEUJ&zBJoVU{Wof~eM?p?@O`2UEi+~?_HVslS?;+2K6yr>`ZMyo|9!!JIcPWox zf^}`Z%M|>rDQ78ml^FO?;JtOR;VXm~r@>z=PEw+f!uOXUJFA*&t$5uZMw)&g*1Uu# zwT=s-1#0O+PJ?MXO)3(Db58s|C0JJE>a&X&Q@OKCmVwTK!WwYg(#maj)wG^(^NtlABl;fMPnKjJe#dfq z?_%75_x138zvLtEHK4~6SpehqQv{o){TZZ8(nX;GtbOBy(1Sr{oA;QwIkUIUCIUX$ z1i`2W5fa~1vLdjy5(oHGhq+&O(Z7O1lfGS(G7%&a@@|s40}dIQy6t`Z^5NQ*B^bHVW29t9|Cm32BV~41W3h#rn|FU{bM+gX^oUr0tprK z%UISRwmQEZ(dy6x_=!lkfZZpFi8lZ>CRYrgSbsfPUW@1l-5`i=y$*D^CB{C3#V@7T zHp~mluqr&{(twNeq7lZSW6v}Bc~1US#t8>mGG+?uH>Jkz5*%A+OxjKB7`DW;DmZ-i zAw?D&_VF@BJF(A!edxczXp zBrTXEJGob-r+UKp0LMF^-Gqm+apH1g5xC?oKU*pTkDx2TH=2F~=_!*MUH+(DB!l(t z8DgrMZ+Wj!`Eb?-C%k>Sw{rs3@Dq3<)SwcH5}Wj>gTg88?5@D1jtS(8e#|BIC{IiW zUSNIhZzO!Dt_?3~OuCB#H#Lkf>ZJNxMVwg6o_|X*zuj#1e3sG1ay|j@3y*)`=K_}U zUKH0H2b0ny^@JjN&OpEEAx~x4RU;St#xLOXjvuF>o_JZvGYIGlvt6eHJK#Hhh)Cq0 z|Nd+X(|T`(tjP$nkJ)fH2@U(_c?PHx*Vht&~ zstAg9Gw}*9m6!7uwvXubaHe~S0(dxIx!nZMvcf!L#6w6%eS}SsKVCC(#42F%0?*a^ z7azAZR$&}yOZfy+pmFo0vgPMRfy*F|RwCNnQ(m7P#ds%AD1AO}rWj!&gSNJHyJc<3 zwSQl>lscdk9(;PFEXGgjAfh3Hs-45X>Dc?|FtYsC({>vi|EAMu>kG|8jm(C2IpUN; zt%77g-IlK`B7_s82Z;h0^TKThj*)M$4~YPUPdl({$r_RAZ*6|U4qeF)Z@;d!+>=S+ zl;7PdwMfDIw}Lx< zYc7!Q9KO1;gD+iZT*$1AiRUJYZ#khe#?QD~hhmbW@Y`{`$JnLHN{gOv&qLaZygCN% zAJk6{ob-Go$x!e;C8oS%DM4~0^GYJ1bFZm?e;6>O^`H?@ z;_$ZX$&eWQc^4~_{!z?chDI2ZueF3EiFQW&BkQGrpmi(}jCzR@gsf^Sl0VyDcx;b9 zCh%)T2uBdy86}}=*t&vRed4_ZrJ1eAU(@^S$)DtGrFZ-`ivd)1DHqA%H7<^HsxVNh z?lAqd+4{I?`UxF<-%1|qZ-o+uPS2!TGnqalb|_ssC0%$vR1rOl^Rzg;2tlCLGb9}7 z_>A5H-~m}|9iG9=Lc)Yamy#jjUe$q+@yHSeC3r*M_zlxM<8>0-4Kt2mH!yFn{ zJ1$6^4xO+UTU6_&k9XG{IkGe$4Vf`Emm%Gad*Y+B*&DP(bQmXrx44P2sK z$(;6g?EZ9#?lTtXo1!SfsEK&dSO&vs{pRp*PavK}FC*x7x^4w-TQlV|>&(ILPlk@k zr6Ep6NpKReO1B~sy0@i8u0S2m?cbP)I_!tvsv%wJ8RG%pVwhjzKaI{)j;_U5)thA< zciS&w?}Ot5nMHlSkJ(#c8Jg{J(V^n55~;hyJK zh1Gx%gu(iw>qb%!d2=4ftu0GC+Oui3!0B{vP5~^!{gG$oELGTgSP%}yH|u=3?76~V z^#p&=av@N)mJ&QTWm@5>$Y+TKU_8VVoGYptKSN%DuLteGG)>KgAXJ`3`Z=Luxjzf# zr&AR|LFC?xS2{+;lABEI(CnkQj<=t8YR&EmH;o_Te`j?pVUFYj!;dZ*q$m@o^HF9~r{#kY@Ryk5alulsIRTgqT0(p{kM$ z{zK8niH<9Fi}UqaDiEAtuRlS87?s?OuGgDSGp8o3g^1O%)XuR^3wGgkq;{pP3VvU> zP4Kr%pBh0+OrEFnw@)uq2$*z_0{6LQ`N^5dtNAzzEGvgUTv#oGj;y>mhC^*Eev+>w zH{ozCgC4<*bMpOA^zVCT3vqC0FZlekA1foi-W8tYywV+XVCsc`vpU!8<59Bnk)U>} z!_GXXh`=3Tfki`|w__h?uSg@0-R1#xbbp+R=l|R$ZCtINFY|OCh;>A?LSf z%KU!3x|nLcal#@0XML2Zn70)?UMwQ|Tf;mS5BmsKdF+CrDCA0i?=omcZZ~mDRv1st zhTD|VGt+J_dLjD0^zqNwa1it&Ij-Z$O2M6b_~j6>cR7BD17-YW3aTq~@8b{o@N1z7 zeD{nQ?g27Za#sLA>3VswvwpH?aPc3kfV z=T|M#-s5A``-{ysCZ2~c&Sn4P{-b#6-7xnjx(o%cA>|r9nIK*KW4k+xbQLpKBxP6M|<1bLXsZ)xnl1SolNtk0*#{@-**;UR(wZr2w6EU2E_VM%FwBL4H4+3 z%3~C`v&I4cz_ueBbM6z7z@a?KegDZa$yVleh|Fw;sC$DrYOglGv2iZ7_5HVVwZP=< z26J1BeI;m_gcwD-rqfaUP`Rv!Q7gm`z4ZCQdgnFKu1}Hd(akz&4$fNaiP~_A)pMwe zrYoNPJ{WAm2p(1CK7Vm9U-|fL`5fr2o#x=2t>X0x^5%gRh2tdfQ^4dLm9)Yw@nTN1 z%;``*Co+UX6I(2xfN#vA+fYE5NI_RjAmS?tig=7+6M85tgN;?j^DXd1^yxlLmt9Yr z8ZUyyebGOYj{9>3>oW6xz7S@JyhxOPg16NwNTMnxAPFEf{m^cZ)nLA+bW#7E_#uNv z8rUO!(TM#!Z_|+PB|5(@M9KuQw~QI#^vqjweUIwoKHB#=n1Q|PiV*Updv0T_9K_uE z^Tp~ESa#^72laboZjB9ilH>R7r@UmjOie}swaN_(Ge^9x2k1JPqm8exEL}e*1=`Qr zO)!`|$6so*Gey>(51X9Vh`&c>UIL?tAA`|Kz07t?rS$LaEnyq3_mW*-lEmPt;(o#V zU@4NzqYz3wi8;T8qwbm8=ZpP2rha99sJ?>y;`0(W<7QpI1c!StqIsw%|FTc2#|L=J zrAT!rlr7@Nu=7}c`0A328J+;u^l6H=A338mBG>(^exiN}!*o-IDPgl-%6HQqHJ^P{ zitez<8QL+0T1b!XWE65-O-ZV7_FW$dD10GvU{QZJ((!uJ03ou%VLLy8Ciw%}YSaRe zDeIPoxOdKV*|Xt;MBFYRgBV9!pll%?rJp4HxN@CGXZlZNLY z?cFoQ0?Zf*^gd(yC0m*-5y(nFN2pV;@16K$7&J7ih;Z}-5asCQu2}1Sw$?h>C zB3MDyuf<2LT~|3?-8yPM#2@`Ss6_MM%M{?}9Li^|=a$$0=9cJUN%^SNa1*Lg1v7nS zjIZP-BPiTIKiPTd)b??BTOs5q+YGDENC*x<=A0Clod$VdfppHXb?{e0I?i(`+9J5I zzw01%Jv4mOL0rd=yxExLITrtn}+>gHl+qLJ{|1ga&5gd@5dpoPrL`KF+VS3jd<; zZ51XG?_Rwq$ba;U9Q1Q5P}gF)lXL>HkUn+pzF)~Vd>NY7_-k%`!SUlZ#-HSjEm5(o z>DgN14TQ4e2RMaOGwi}XHAm!d4py(wMJuy_7QMHg1Uf(1wVWn~@qm^A@x=OsjA9huS2~hy+W;vUQ zq%XTN1O3wUTxoQN3VXN*){mgKRw+;Cw-q#Y{B&5%5+tPk@3G&;-6LmJTExYl>eIyW4M*lEOAzmO=f7VDg>tj_8$3iUrp!cIokXOBaAlaE^IWgVA<*U-j`mCZx^g?~rN4 z9yQ4#-?9Srq9w%V8ew@8en;j!a=F#_WFjF83oQA!-@O_^NY#fw>|vWDP$bEBO{L_J zS-e&QfRnx5TggB6kOy3U=7dB5dwq02&Rfx7E_T>_|20q8=5#j(ijA@JY^a&F^1f&tSrcToUx1^e6@%R~u({2AY zeMIf&vqy@q&gqu*uTtm=pckeN^YLb8&cqDhu9pl}(IUdVzrViJG4xv_QMx=Uk1FX= zUc{#?i1i8v*Pb3T>%tC7(AndIz% zL_NR$XnOAL$a-H4C!ttA;UVsV05fjuo3vkr_l$QIK)NY^HE@9~O6L+feVr2+eswmq z-*y2ca#^}ZKAk5T-C7+Be}#(_a0Bhx%4S^|2Dv*OKo;jo$3oHCIi`;S!yZ?C+LC`` zU~58U0)lZm_=p@D9xp-}teZJC!?a$(p{RF7ZUw3-CIjm58GaS&uFBN`ZzCzj9owg4 zcMXnYFnc3TiX7E6zDg8w&GmWZkb9pq)lbcP>u53gzLg<=$Ujd+&T;KOD8{EE&8xYL@n*Zj|$GJAz$D< zR4#nZMSB^6*!$_0<+ZlK#dSLqE2TNAgY(n|s!*g1%S;q>HJui>v8{ghg+WddA92HN ze+PZA_X=kRxmd-u4y7>ot#A5vvy4_|VVJy>ub*h6AedQNyU^bxkcbNTS^S%c2ukeT zGpBy#OE3g`t^WIpD&73h4tlR5EKOwkT2nAv$qV~;Li=fu5E+i0& ziY?6FN%jE2Fyeymj|4VEIDl|-Pwf3;wz&;cU)FQD-;uo0ES;0C4+Hy3?SSH(y#@4r zxA<~8+)1mz++ZaZx~tDLVD?40$rHCV(ME^8zP&l2jMXolKI#bUA#}=>Zl&5JGwZYO zjTbOphVXqzPB8I(6dTB9B7}a_h)Ro-HR*wW0v-ZNiov!cUCyU~mHM|zQ7MksV06ij zs|Bp}LPB-OiHOvtE?^SwG+ye|daaL*d=y7jkO)p&eSD3b!n{R4aPGa(hrWx06|(n` zCg%`CnSH#E$O0D%ln9R4Z#}yng+*cg1{ocn#z|w>Q`ozyRH&GzK}!5m_)gN5FVMbd z2EywcG@Pnc=`!T&T}oFMLwDF~KG}bAVC$w+age~COsx;IFijD#f_}tHx9&}>_>Vum za^>R{#pklxW(E7_E0j<2iA#)`!i70{afY2crq3GBz{eJ_qv6lX&HCA5$`Wh635p$I z#-v}QtWIU%Hn@lH!x@aG%IvgD4=|kVm4z`sM6?hY!W-@lde=4jf$fQUmnLF6YFu`7 zYcJlpC6~2)YM+m*8G&BY@3}Os6AZVx76ItRL!ap8u@ZMEy0aj$Ud*yae7+I7u#7aE zHVphmD&VB12-f|oG%Bl;E$?xj!!i7!_aN{NcvwF_G~pXb;fUA@5OU0wHI|g#GEH5| z17SH3_;2{(i^v`%si*HHN$<}EQdoyn%|yiTN>B^1Ww^>PUhQNd?7*xkjO!4V{5S;y zRq^J4&<`9b42t2Upj+u1$*f;3zTRpbpt}H?6$q~P+y;Ws0(n?Pe-wTm)3@@1>_xWT zr*-4|NUz?er!@EjI+IYX@J^Re-9;1R3^8d(r*(#c7%jPew&8@{@E}dcLv9d0#i_U4o~_$A=zI zv9wEUDPV7(Jxn|7xF8%8=zaQ%jb0~PSWi&{AxPV zYzo9T@*^is0KPm?X9TVJ6ReGRQgY8eI4rdV|CvxHWc1{~+QfRUW?hR!>ShYD{Ex@D zqki<k%%MM?BXi*9H_T3t!fzV(Gj(utgHM=K97A)yH-*;GuE7fTbWQ z63Lq@9Hk;!Q}{15B_zvN9}$faZeMQ#f?+Y2!sblvN+3kL-6P@RgA^pxJpU5>$aS4h z*kV%UT%f0l#dBH_XDH@0awene6u87HKh!>fr+w{Deo}N=-ayelQOycE#fbL=e05FD zw~bA+)DBqcI$Ew=+DQjPTee=!>UjQk!q2>2IafaZaOq=D)&(^>C9#)E+9BMcySnpr zE%ygjY&|0to3U=bwY4#XhZw3nwhIKqjuU}7=_eRi_g0CV>3PIKESgUx*M@-D9?yrk z@Sv6^w`CWZ=xvsY(u#Tl&d*rFLzfSKCEpQV=ddF*X~l^PT;dLu^)7T`l7A=Q5;Qz6 z=YS=s9295Il_Yu%+^hTqB?=(BsXv36(9#4^fPm%SF-X%Ej{;_(j>4;!F(Tv&_X7lubo2ZjQk4Jm2V)P0L;I$uaml_BQM6 ze1sWz^8qYko8+a~k8_K;b|o&Zm0)^OlxKT{fwxbxDPC?O!m_iI0RegIg>wc)2s3Sk z_@t^44kk;73@7R8zTE_WthL_ML*H|Hm#IF-Msc-v!UWpsd3d=2-k)v8t)!Ooqyg~x z<+3J1_0O!37q76$m56GrJvA(+L+$WtgOmFh&rNjuSGL}^irPIMt)2O~KCJ==U;53q zD~K{E3pnz)`hoo>lu?t7z}ex7vAuQ{g8+n#no$?B6ed_R@7*<>Ls>%J5yqc`XmGfP znmU2^$p30X4bD`GI1rQ`+6+T%hE1P1IWBac>^0-{NqeECYN=0Tg*dW!M?a1KPmjpl|M5Dm=Ih5kQ+Gblh4!e7@s>bnv5u73-~lZ$Ey z>1Jmgyy`hu=0=v9CgHN@TCCabj@u?gg%Db&0 zmyk+qNb5IN5R+|xp){w8%@LO|o|bpCkbBPv-*OMQJ!!7GeP0H()R8Wq7LAX_I}6c# z{NY>j^EZJ2%)6QK@JZ!~Y-tG93|=#b8rdRTiax!5sepjYL-7}c{t2euwiMFj(`Fyb zF{$a##JkAKOl8-IPW#NKGzk;k$ zMKkxaoxPZW-h#aGd=nmQx%I$kHY1d;_qTp;GPN8IC{EqmNZ42$hq4vlhib9f`yFC7Rtfys@)EyQwx0X%)Dm8tbzxh$R*HAg zv_i;JZIsWa+JYrYAIv$t&MPp!p{orBJCK(i9B7pbG|miJ_gx{mwHzyQ!iEgotc`D3 zn=pUBpw_D5#<&6N%1g3%zi?WHpLPc7)py|ZREQ0TZw%wRHWoYDqL~%iR+PT2>AiV7 zQ_1Ady;qIGh7`3waznDy{VcbV7}_a=IXj#XPc}xrYsH`Z3=$KAu7+_=W=8X|sRk#y z1+k}`Dyz>moqgLXo!tg{XsmGM`*tkXt^%G-%J*gfSL@lLkM`;5+yDincZ0Vq){N=6 z-Fe2Ad%ZWUlc+gXV)e2JMvJQ%HK)nU)@^PlQ0ivc&#Rp zDUJJZ!Dil6RkD;Q=%a3Amll&8UoA#?al&AMqN8TNrhg1#x>4do-n2B#Oz=?oa+l0B5rJFx> zH>2U(F7^VsU`F~2BCo2rY!66i6TVEfcEH)ksebVsx`nbzUJME+Im)+rgSg`D_(-+> zm`{ZP@8V0>dD7Y!uoKJxBErQL>qeQ}S>Z@OAb$+u=bJ{TZBbKpvXGYfMf_RCamf($?V=u)@B@TvKJh`OI!A?cEo= zDIEkRrQf6wYATe>XeE>rIw2>X&v%m0N-@JOW_6!p`jzJ%O;1D!n_y1_E6wZXqm{kF z%R`X}6i3=TGCJeOE;x*oH@@trQ>ujITiR=;yM&#;>=ADSulCVV+b@6c4D2HGzW>_> z`GDF9F8<49RPeVpcwfmcv=_r&=g2G8)FiXvZs*Y*_5>|($bK_#1HYc)6m71;gbJ=M zqhiY!i+p)7edzUDR-^A;;9eCSMe?5C+az@`=7Hqgt=8MLUl}b!hmf-y_!THIVL#*k z-f|ePB@mTKdP2nA?cM&N)DGC9S8>tG>-~bl9=<5@BR#+Q%R;rCfScTvqN-&iI~Fvp}C$*rAw<+jg%oQqE7jD2bn^lyek|jzRC2WsX^WQX6HT}rs0o4_BqG~L~rjp$djUF=*I zqYi5f%&$Yyizeo8We}-^@*lD*qMHf#ay11QIzgf?;X$INH&Sp zc^K{@p^(&i<5lB0l$+yjuk3D4-kaZkmYEesYi3?N3 z?w!J0>JKc7vsQm9&&?8y0Maq;-nli-+LL%Af`?tH;+1`#H_e-#v=I`PYqiXW)^2#= zo!~8Uem{fLFj=c@vYswgpXr?nx6dP<+?LsVpgW7q%yUp&lr6razlyzWRNK*T_vi{_ z&jiZ!xoV8hB=tt(QYnRaKDIMY#1)sPoNx=g%!b~QONEa7df{e#xZ<)IndQc|6foqE z7TUz;52Xy63H_=XoG2FfF5G^!j1`uDXD5ojf000B@@6xObgJ$#R=6jQ^G}wIZ{`QTo<=mB;>Qhyngj(B zI_*48p^W*zUUZP>)4SjJ3BxuXbX!I8-^Z^F=>4);Q9qvHy4Q-^J8ca^ohUfYjSLO` zT&~^d(9SXQ&3Yo}cpjAzy?Q(n{7&+PG?Y|7ub((Y;mW&~`)1L1_2nv?H;xjcsD7eu zRGP1marR;MDiIbf3k}I!#yE1E_WkQus==rm-Zt0Qw76 zk*}w8gx@0WE}NF$HwH#r^DB?91P0ORa{nS@vfa57&k^k3=0mEIB9j(3yMtd#5an9? zyu`kV7W$zSw}`*}>}ExhIq?ks0Z2FuKg%=*tWPkRVWaq&RC!sP3Bwi|7GE}t-k;r2 zr%rI{j!o?WX6_HV{61w-o*^hzPV4}fr$ol^I-=o{{xig?O|e>*hWLNfKE z5hsvhZHKnv?Ol`AIE^oTCj;o@vZqM? z_JB=$z-yd%nc+blcg*F+KtJ_P?;TEwLDbrSG2_(F=_aoj%=#R=rUtF~`Hk&nLGM_v z_izZ@6~YkG_i~0HX9KH2HC+oPo{EE|^(d~-1TZ|u`XZ#az)WZyzcJK)vnF^1*gWzH zk!iB=cZx#!%aZ~Xf4E}wd^T)df{;ArL#Jp?`bR_$Q8~;+5 z%dY-$blj!sMFCY9wEfSrbM9^zuxAe*nj_y%6WQ~=^2a9|0+4&}mGl~uRJ+)>(>qR- zAQo%%&n3kqJsH<^$-p?;cB_mH7a(dk^L#8PS!3B|Mq~wbS&wtqz=?2j$AqEgv?ohY z36(%guBp353-xies#*=wI`Wxpj~<^IOsC#`UR6@KIUhQJ*!%Ai+L-bFIe z)h|D?IeIyb7XEw{pYI!m^o+vlKXi#qi4K_H`MEsQ5po*xD1V1p%Ck3&IY>-9?YKK; z?1A$=JGH?U>F{9svf#`HnQsG481lEImQEk*y-&zgsZiitFy@#>Z{QRzkYNw^RLzh{ z%O6Dtml@S786maXGihaA#CL_p_%CvXynqCOMH(rSpAK{Tl%PFRytCuhf>=-w`8aMh zzO&oQEYpsVPS0wBlVIx2>PvgB_UJJrmU{uu20^zs_f>Dyairkh6ui6lGHc49(wov= zabNtE;ur8AnNI?`u@@cfU4v|*Y_pd(ZW0@OZ0+_&)hcrN7I^AQaKh#x3W%-o7~_Cx zXT3dr-wv8T``QMLUQf|RD`q7eu_g*YoRj50Z;+l}ud^>8oo?^nHztsj8@MzU4E61K_zH(|!f&$wkG1!XYwBC}zm=+@BBE3Q1(8muA~m82QUs;<&=C-j z-V+c3k*0J+iS!QAL~7_Ds1Xp5novTd2SQ08B+vGH&OOgP_w;*zf9Q*_l1=trYtNoF zGoSZJSG^#Fg`3l8_fl9O%tIMRSWnDL%%x9<^W8_wJ)fDEXAjV{Y+yd1w#m0av^~!# zp`tkITvhgBecU#$vUNw2&u&e&dK0qmZm>s)pagMxD4pJ;=lN@sf|WyhKt|>D1h|lS z4#SB}^$M>cQd7F}TE(2UNAKdcpVi*`@{4zRQv~s~*F2-zQu1!}&u|nq{kiMi^qbO# zIIE|7uB^v;`8GiBb`+0xhUj?;1t?5cjTqWSP=JDwON>LD==C}Y#d!^3pbTI%{BAeZ zEymCQpul721Lb^aY0mj;9&x$*0Oq1~A-Q^c!~&*=s6xJw)ex=bg%gYRo*|H{97mUFdO zkif<#829GNDJYgY!Y)S%>d>l0c0BTD-kh#vX(DyCJjbW>YnP|cy2g`o+`he?Sc&-o z6pTqj&zxN`CUydM0toGH8e=6-E!D$|<`&6;4W)~HtYSDHe~aT1X}Gdp@@YqhL^1fs zdZ%IjW=Do@$W!ypD+qk2`{#G^tJ~yJl@)_^%1_C0Ty5a?MPnxeQ=8W-2Q3usB_k^&)GGrO{1Jj)B_IyQ~ zz^3nR6a#9D^M?Juw}OsOtxl8&e}%`eaqW{AkLc&ZBzh)8X7aXb?0KDjGJ-AT^3@Z0 z_2?ckFsAUDTQ0?rD^w3 z27d(ltL+jGiAJM)fm((Pe$w|pt~{_RmTpZwO3Kg%dZ#fU!?8T`*3kE!nS*iR z`{C}-Hd1aHkS{qxWBscTJ0h^3BH+j!5yVdaoy1_W=qHcc#t{rNLQV=N6pxi2Dd44~ z(o$i^69Hf1V28sbk6~GSfOOSJ^oeFr7=b#i(1;rhD`n^8DZCu%z2B9js*>(eZ@{^p zbm8ptBHedRH!CbWxm9%JPlA;@K@Z;K0|gwo{Gh?5t}=C_o#|Ej?Jrr<-JF^fd-|%C zKtPS-v0^rD{IrOJhWEZ;kY9&Q1b$QPLo54|{KcvV6&FopS3eACvT_KhC1rf;TGd!- zu*r}-5qZjzA{X}5Dv0;8|HjO{a=E~RFxc1a<&D8K5mmj98?@uy?;ZP0z|VZR-uH*r z_y}0RNP+d-_g40*O93tAN*jfRxVM$S1Ki+?8kt*UT+(mdj9bI89DWu-Nz%HRJSduT z1py8mVxx1xlWOE$<#&S+LC!3F7lW7&(xjNBk- z*=r5|*wrI=-qr!)R1sDFAfTubdc3)zj5L`BKfEw^)ZdpyZxy=7i{8d-pEQoGrAD9k zxfZvbjMwuWmC4Vr8~>E;+EWzcB)lJj!DMXDV$Ml$IsIVBUzt=qnw@;#Pn^&C^u6+m zngs(&JJO|4;cU3%yD=CAkFGp&G2%Po$`TcL@Zm}^dV8Ur0tAp!Dg8XK>*RTlRer)Y zdncM)hpdP(W{M)?eZM-&Btx!S)|yI28FZexXX<*GG(?^yd5}H7Zk3-~`J8sPFY8KD z2An}+gKduY#gST{D6pc-FFy!&yjV%~qsJ2-71gzRk<5|StPBj4X@FoQgHEFDwS4W z^TrOUEhH)Zhnmeu$R!Qz$_^j^{DcJtpHYF0H#!&t&q}P@EUj$YyR4)L=)GA3clZde zD(gbezO*^fD|7ubyiA3itoELCjZ#klKqPk$C< z4j(|Rj7<^9H?e&tfP`yZC$|!|(Cj5-Pe{A(B$V-$VOcuSwxd(ah!r$%b{|<`1sJ`x zkKLdM3c|UZ(mPQFqom+0dT#2+UeJoziD})@HUh-WeTKx?S=Lj!*q!+a!-d zd{JiaPH{H`{-0@et4X4Kj8O`MJjrHvZV$!&aiJCIF$O6ls6bR@m?s- zLFFoFfS(pnAUw}}LDas%qdr{VtGA|S z{#NUo6pLtS$u_R<(&DIY+kay<^^r^I2Khf136q?$AU#>(2I~t<_pKFb1V#3$tNy!M zw}_;#k9vDAxBe=xAOckrDr^81D)sz-EE=Ay8=zdNoyjly0WG!DG*&NVi2-!ANQ|IY znuBAi>u*p2-ZJr`$^U7Ha)k&eESN`f-(OK{GmwptOZiyv4@JK)9YJu+PJiT121YXv z0#su^gr3JF(PKbEDQbxKBox4WnJ~<{U6faYEu2s|3)PYqXq(O4!x}y zDyd_d(>`i95;0~WfXhn@ykvj$#t51p;02xb6i@Sz!ZeutyuykjeLh}pG( zB)sL)pB2Flj)SOFeqenXy$E6t9tOUPijSka{@w*X~Dgu+Y z0s(0my?uTzOvXieArrE9vvmQWf?wBG0OaN8^{Yr2-?sa}tudM53q-cA6j%Zw$l%w2 z=Ti$>+5J?iK2v5dYnGE-BUx&e?C}1jtmE&xKt`Pl*SN*Y9q8k$HNU%V%iBUv9&pat3k0!)k-F##>nhrkH{e?=&TPsEsf z^%1nutpxJyx2KgMC13k@-b6kP5HNe@UHR3Hll@|IOCJ(vYs@NXFiE9- z%@3l0G%zqE2-A@rcNaE<20SqqQ6<`!)p!`>{wvYKvz>lJ``My906xR1;bM*v8OX=b z^OlRb3rc?cP!r6_^$5CNAoISx=Z|Y{?}s^v2~O*yKBk=KfL$MC3!&} z8j!tj7^L!I^q$j(F#uJ3&))20NN43;5`yuH{<08+&9MGRMZEodsqj9L*;W;`Fp2%^ zs@Fl#Mg0Z$6B^a#r^5tjVDu7CN`c;! z3UxqfY^f-&G_X9~Kx;T~uMDpqi-#z>liQT;GDgsMB z_mJ$Z5c8ylUEhM3@&q{BQF$Pr^=b+*In=;P`gYCYN%Rk6@^#MlfJHk$X(wIqaYtbC z1RzZJ`6Jf?b0ymZmU?#3N855-GczgIJx{1LEN+hmZUAMe*E&FC|0FPLg;xN3@qhS5 zgmWaNNvCu|oo}gBASlZ#Xe);~_y3CJ`CGUTk`0(q!-@FEukwp-{}KE8R}h}R3F!PS z%jfZdzr2e{@z*BSH{L?|v7h?bIUt4U{pH8IG;AfF1(zyQnai>8@A zvD=#XhoJ|vcniYBhjK}qnJKkPF!8^;RlbbDt{gh0f{TsF;fF7Ur2f~h!}bXPr^p2_ z_2|L}=?)d61bO)2n=^mjV^CojgD@FogP`BL7@!!zlJ~hju$2Jpw-G=oaudEyT+Gpa z{_spHcqt8120WS_AJ78!$mcIRfU#hx;@E(r6xG(9$Umcaz8C-dsGYy>zW=+FALM!1 zZlLY;D+`8kf^V&(pUb8FcM1a~w)9h`*YB znMerL?oBQTFaAL>{~x=LYSM!@wfC2)cYv=L4w$r*znlT^Fh&*pWz(WWO)8y_(tUvq zaf!Y>UGV+DOu-J@ytZVv7KR459u}A#!N9UU6sO9p9f)z}=dqgyp5vK}=I}Rn-E_?h z1qKkIlTyDSB^*C=vZThwm=Dj9;LyWTc6p9HjgkHCJ)Vo?e2t_x*COe%=DRgAt6U`@ zH2G^HV9-^^!hZ8f2_TbNu*V_-wsI!7C(lc- zq?zbAI~U9D`1%_jxPQvpo;vHZlH+g!&so0Z0!f>962RPv9sH>oXh0dGYhxSLJIi`H^Dgz?yyOpDC!F zC)r⁣R<<%k}zhX#Ju`=!u(txVl}`al00jDFIbtrz?S$kA}jIj|xIZm%rw2M8=%& zdDhjm&k58V^*x|#uI zhGqlhe@y`g9u>c~X%0~WN>mglawxYF+YnbGkD`(7}W@a{o-l8yFSc;Upn67L{J(%XOkZ+KSsl@97Eb?f?GQTdxmFc-H{En-9zbi z*?uG+RHnB3c{YjgsrlP*FJ#RaN1S}lQ%XqGe?pKVJttvG)eFZNc&R>X3$QpAlE-8L zN>rs9Kn?0`bp zUi80!rS~7(KR%ejjo3GP&dHRURNq`1uXM}tQ1dqaNr%wE{-#YmcaEnl^v3xrd+9wr z4ixgV`}tztcLOL~?=9e_mAcjpJyr<1ML0ruW3=!<-@7*!vyERWEgHX;rU-nQY4U$H zRW^u%9&HqlP%`6d^(kv_0EX2WrlRccc;av{9!+jDMenkR)klBcUhHQjdius;6;-!L zCjp2Mfdn)+6MfR3v=I1qGhr>|VxdWw#ja%}L{)sWXG574=Ot0ptH1o@)28%XA#osUQ41mAmo2(2u;C}2u-~jIuJ-~7*k=(N z=+5KXpGuo#tl-!}pcILLMf^P*?_xjVO9B6eKsOJ7`x*`)+jd6}W4Zgrlc6vBkr06t z)0$A)Fa(g|H9j&92q2JwoJoH#V|N^E{D7`oyOgCrWesufe;XC(*Xu?qS-*XCo)&X; z^itR2?zmR=6S^lTxRP!U_MC5LURuAfr<6rJ$~BN$BRxpSZn8%z~) zvXLc3PpcF`cB8Y$-aF&os}_Kd8zyv|i{Itpjt?d(?qriLGV<&jW?ME0Xp~ZvjoNa! zO&X&q^T+g?C&lmw(MJcB^tWTdr@F3TiceIv?V=WctKU>HLoA(UckjfcJ!U?==S*^;0E8IxbN4MADkf{}H3ILx)s>Z6 z#!Wy{9?@G&G`O*Yt0;_M(qn+`OfBuSZYurs{a$m{Ew4NvCUhA&=?g`Ka z(c#-$dHdyW?=(Dd^o-n3sGl9~CnVVJ08FFojJ|kDYDN0Y&bu*hR;nM!0azBv+1XML zZ-1j5RoSPgcK4Q`y)lz7ugbWoyaB~=oXqj^QYrJJ#^+_^xi-ifc4;S?t!G|97vHF{ z+Xv`toO)_1FxcU!`Req~Rr7~e5?yVSyt;a{Zcge{EomfPkp%L1jK>1@W-;%g=d1B@ zr=AZ@>2y?qt27x5;>%R5nwO~dUy}Rn^>B%*SIynN-?!K*RfvobK8#aUrmMCuBa55Y z3-;;Rl0NF{1fZJT=fVQgf-q?&g8O|8`49;_>z85|5F__YF*zTWxv=ay3f*4ly~H-9 zC)bH59Ip8c2+DRSYjetZ)MG^nj6CG%)o5&;@V(pS&FLw{ne%+=-^18s9^0TbELfD6 zvvHw%h5U$)&@EvJ8!YL^#G%;$J z!Jt!g1Q3uEbrM#xDmD>vt@p{V3ha}bjKZp(CU_~+pW`sJn%2d;`c+FDpEX$2LX!xV z9Up!%Oql2*ZPc^$&(D(p83yerhGrQC||6`$SKRM>-Yw!l6 zSN9z_fPI2e_V1{N33STQ{qpsTS->s}jCAtN0zxWq|KH13ttgk2lm=$AuC<%myc2rEIzQs+u6-Z=HN*_}tgMjE*8hvuat&a?eos6431 zl6HBxbxzR#X@;MSL%-e!Rlj+r51uHK+&S5;?S&5^2Of&I%5DML-PF{R6R=O#=e|MC z?2T4(JV{`4N#x$6eh|0GDOakD16h2GH$L33<-jJd)&6jMYi)z6z122w+o&yQaiL#L=AMC6TTJh&KZNO>ud%O~zC#$drZ) zHzl`a9c|M*G#9Lh(@@StKB$*LFxJApujv2XLs9&*JP~&6^ zTuCl;TCvb}Bih!#`q}0jfIIB0qqN1GKCMgER%a}C(! z^s)h+T(p|q>o=4Rv6iQVYwZKPxyijoA);9&f-=tbQ!0*iBB-_JD9Nb0=s|Y6GKH+r ziHBtm1%BY)oh4_)9b@Ozj z1kYmc+DONbbKxLJJ+%euy754kQjpYkcN`V9b!(dX?eEp?hcDxKA7`YOlILR{vYvcI zJmKiozZL%e%Z;<4**7*PA)(2_q*D2H$rlKcg3KH^H|xdtJUn>dfQ2w;q+fUq6v-OVt#X*j{Co~oSz%9^ZmGLX;VvxKoR9@qU^JK{&Q&C}ry zc;+k`Jy3&H;LYLpWoKKn+bu_?p5}ao&rG^+^6bel`L`cN}&Wd^Jg;0S6yU3ei#~zG+T#ja4hR1*9ZEIqv{}{#2MiD?2Z)>h zJg)y7o($2EP44|H^uf1nfT6p+pj#p3&S1#;$gsoh?efdEm-#Xny{z8m0vhL^?^}#x z&Pu>ySJZhvIq@gOq)r^bjb+Q|u2$;=NA0jl_Lt4T4pvAg z6Wwip`IVBlfi#wV9Wv>*FHg3K@yPtNYzpNc5DR{M_4$hOceCSmt4aD#L8Ib)ah)1S)dCZhDtounL|D}nDoTE_3uUBDqIVitmIK{4*nGs$#iPxgLk3`^>=3&1%Y5mNIUgJIe6*u4o$?Pdf#IJ(|6CWh9)pl z==}|z(Q({}l&82bU=Y_3P8+Vx>bEBNP~5h{##rD|H~|Pn-(yJNc9bwoVAasH&Z(hI zy##h$HxmCT1nYr|?%b^n37EfpnfO_;aDnHmUF!*n8G_rEvbD|9wn)P~{@}BKgg*9I zqr=+rZm8VAkl*|aZ8Q0?J*YQOvl`Z6Pasz^Z^fS{|_vuXpsiQfpR&GUYs>`ggcRHpv>@+7&USkic(!N$`j3 zlpN)d$oU%yJ|p%3BG#qNSXg(upfXOG-M36i6mzAp@5%JXgm1LbqU@m0`0nq&;1uaE zk0&yYlH4;bnB16p17G`zp_OmO$HU(_ZF2g=BAx@a5OmeNZZtKbrv#OKe@PsYkbpMR zDpopQCSB}6J+;!}W}QBpCVG3sMLpq48gAMsL1E<(_&*q6JGNn_T(Zvbgz)8jKGf4T z1gf7h1RFj*CBJpg7uI{WHWHh0#JIb6C~%KflkRt@CS&9TCc_<2Z5fgEV|VJ5GaP`N zgb7*ALD^buk^L^`{jyzsu!sjbriLau_1DTmf|$y8@g;xZN2VSe5dlIS9?i|K6 z5a$^#ST=IlwC;|u4c-mgPUe@vu_-!P8pEHh`u`NVbn0fTok$(^`>pC}Hc$=>xtXjH zo~z1Fmst|9PN=!6r`x3EYqcR2gHX}gc2*)8^0nFmPPf}DJh}Q9t>^K6PSxPz>WRB4 zvbTNTfK({F^KgkdU0ZgkCxQ>MUJr)#OgxOcu3oZ0+Ixb536LtaY?NCi75(f8%8 zY+R0;j-$5T4Jyfa5}qeR@n*$n5Jekz@|EP}FL(DT+0}?#a^zNrs$xUzWr44wzBO6_ zntR9dLq~Z)g}-kW}+X z9up4{G{Bkw3y`rp-r}}m1u-Z5n}c#S7|9O5v!|MS%jB?go7AY!VrB>LSqcQU z1V%DczNr85g+YX~0SP)<)}G&frrLME^?|>B1@9ugwgyoqr;LEvzIj5J&cqrv+Lz?g zK_hRMO5?a#OK&8=KYk^gZe{ z%qt>O`}<{_e{->u!FIh9eTkdtvD1` zeDs`Vslsn4X67(4e8w#dN?h3W2A~3uBTg<2riGTh7SkRTqHMGTbno@AsvEyubr8hy zoo66;QS4#vW^XpHd7j_mqJo~FJH6y}EcadJTCw=;=#2DpeJ|50L@O2wVwqon-|eLB zAhV>i13%n32ih};>PSj^mwd-w+4?EvGfw{(SP9ryVGQ|q6YLrcF)aaGJ9aO>j;yX1 zuqmoss7H`5p&K0SFLrX$sa89Gd*d=H?evDd&gs=zZpp_J!G#zkVjVV67n!vE%qT$# z3_fuxa1BnuW4|%nfPd7tq$IAkelULpnyy)tR?vH5TGVu#X$OzxOVnxA;9h#C|9O79 zi&#_kdXMBX{L*0GXF_vglT6)|8l=_wBns~8kRabm`Bh9=*G$SNA{W037DD{MlFa^t z3_3-+|F2W%?{i@8{m$Y~tE_qlIE|lS8Q{h}_U#hX@dwY%rVs0pq}!TL&Vj@@I()7| zY3RaD&-`2zWGD++___W4>VhlbmiUB(@-wC=Hny?w7a^@S!HJ#>s{Ydvo-Et|5J;;$ zg%02P;{f*UiQ@Mh8KEjRgH^D^9D9nyKzfG_$Ii8}6H}ZS9}=#pbXAzvysUihU;*|T z;W+ia4kF4gN|Y?A+h|f1irfqIqP8|k|DjPt58zFG`PrU7OT6y=%8n=BX}*g#uOR*+ zm|e}1=R4{Kv}fqU=wmzh1?PUS;)~0MF$R~D5ai$ZqmygfBY6=pF3J_wCw{{0w7kh| zYO(V@UDZ;uwj_cb1t8bBnWmCVS2-~FyR9ttbHQ$FCDSfYzd0HgnL+X>azUWulC(~FBqv)I-{5^6oMJRvRo z5Iw(l555zi!RN6 zQawCX7RSc9D4=`am+8V7^*}!WNMicm2_s>sNgtFgPFDf+lP7hF05GrHtNWX^> zcP{&&JD_zG^vN?NXKV(OfpHQVn_OJKBfQ7<4*)=>>PMqF?Pz8JTXg4-@^tyY zpCPTLHZRXinVv|>;2=*(j+K;YhdCa=?dorlN4&9fI4V6*-0tBhe&aCda^ZEL3{sW7 z)#5|hW#bw4o0o=e^&7?BQcr?L>^dWJ-1isFw;HEa6u3)1hB@)1z!e3Ga>sVzS2qE+ zr*CqkPJgjani(oTr?_ z3yLUvjUc1@eWgJ|RiHH(E-&|{0UEC!<|JvoKpa1Se>XJ6ACCfQ{vwXGo0o+ViZ{#C z-jgK9OdJ1ok;=Ro1_bC0uGUVCL|?RL;4E((X6j2yOCap>uJtaM$lC!*uTytBeIgTzGo7EC|+ffL4w-t z@5(L5hWS78(LJ##^I|IL=HhwFhy6gg`#4}uDL1f>s1xZK*LOzvuMSkYmv@KaRL}m< zE}7T3z^BuRJ2t;ki$_RHzt5Qo4Me1O(^b{7huNC zvMOAAzIXgi?k#Ah_t z{7jj8<%l64gJtn+K8=S*-<5a1%B8Bqd1ess*(~5-!^-_Xc!Q38K0V91k#H*rBWmV5-Z?+U* ziauK!QH(0$n4dG6fQ_$Dj$ujH!aNh$6{D9-#mVL-8E%62cJtVU;`EbPdjda~D=JOx z(XVRj{DQS5?;PE2)3T>N$N#sWNuaU|QPcdFHx!AHCB_CY(gk^eU7sZ*#Y&8{Y1;5S z&n6hyO+L83W_YVjm;Nb_303sm=+(Ng1}<5ZQ!~F%=Nv43t;REp33Z3VI^1z+{meDC zfhUu#S#&Nh%)Hjp87{59xt7Gr`XiYR7yslSr%k4T#*gUTA_r|%rRE&?4O$er{_7ew z%`=$YYPcs0fF2T}edm-dK}Ec7+}=sNIw006vwBD&vO*qGa)&Ys>9(dC%1b@lt6D1n z1bv>+rBD9KbdnI!54)1zJnc4Qin&LDv~_%t%h{C&@SDm)v>m9Gsug)r0e1&ZV~Nn` zTtBLV`*me^yib2-d@FxU`5A%-DVF2s3}A)_TOF&Lw{}-s#BX$eeSpgTIrMqs){1{^ zd?z*c2Yc%6OPn9V60Ut*jOdIGOpJU*Gw@>ap|Iu`;nNRKr}ZV5$rz3W9jz0y_Ej`A zjIeK$+fh+-$2v#E*6Q2QQ*&ET#^Cl4GLZfg|HU7r5a7pfDi(tnF&H6VH9a^0BgP$Z zCKKb$NZ2(M&%;io@g`|yOUS|C1(SKaNy7e!Z{XOkK3o}7g26HJ6YdU@$E(|Cvs`_9 z3t@e0B}&>_#XAx@CRosdg}hL7^0*GUcQI;{c^qn6LUG=Rpy$~=hOT~F()24J=f-gF z9DB4AmpxooWV9_pM;}E=U23hzN4M(^j4`xmEwH9`P4t^8tPOOvY|msR`67r7M+Jf3 zM@GB@Mjy-@EILAofP5(gBZ@qL+kj{fz)um-v68rIj$V01?&+MxP54^&47MA*z~Z&J zK0acYWAHib4VxzO2&2=@PmB5p}n3(ZasLvecp-{YB4k2na)-L!ze%dq<5_*FHkT;su;2-TT7g$3rDP(D-y6iiw}B!fD?*IYB%Io^t?HlHQMX17d0u!M_!cAPY%v{wr;CKp~456`1%v`)-m(K4Vwm=)vY_V-`GfRJoc+OW;@zl)zZwB z2q_;Hm+X0t9tV8cy88O^rfkoL`yrzVa(gfyMrpDh$<81UssLA^Bpu;rQLx4%;jnbY zhCp=ZuK5MjyV_KUd}}V{hza4rBA~n-Na$%h@*!w9rIayLbI9kcuUz`ZbtbN>C1>iM z)Q(&OA)iDZ#Wog5|G6!8Gq9Mn75MnjA`GESB@86*hQAqt_QzdnqqgIe>zih$jV45z zH~6(fLbu_u%Q4H*wCVLnQHW8*^PHoc1>tUi;tRJOO(K#AOA z)>E3}QshSN{5tT0y!CF89D|U^|j4Y$BS60+_THz(XYLJw8_+n>w-l>H&nE_@kHtIrAEn6ab+@o11sRwek7b+y+*+= zv|Fmd;hA$jig)$(4)!Dh`t*}6QUkw(96t4aPdbv0a)`P54S7>?ws>F`OhGwUb8KS+ zyh*+nsx88U7Zc=KWGTqZX}{VZvoK27{6Puj0QqjE3wv;gXI&(5HY4X2mJ!2y3mn*; z5&Zzv@x0HlEmVQE6vy)k!t;Esb`$y7JbV_&ynZWu`BX)Zfaf94rC|@0XP-@pZE9?rc$JQSH;DD`7c{=jN`xAgf%opzI5`UrwWfa~O*e9BIJQ>LIw&c&fW-gYGg~ zL_FGg`Z1$Q+9R7&p{}M4i}RkPs1q4h!_iSZ4sE?613s3bs5YskZ=t7SR(5(>SbZ48 zHz$-Q$hFzWli5N0!P_Za^UZRvkm5edO>3VwqdB`6uR)FnGais(N;@woCkNcWg1w&c zq;mSxJm;Chn?PRV1T9Z)E^g~w3gFpHscD?`Z-Kn>0$*z8IS=G9S;%@y&~;+Z7v^8zx3X)I-Uf)rh&ULuvRoBxN|;W8I_e zc>L2TCIXz&RXMvtKE6+`S?nj->7!?Uv0k5HM6FytMY%gW8xY{;=Arw>WErtL5@#E8 zpAysU9bQI0Lbx9#enWUL!6>s-Z!CV6YEsTO@hoGzH{TPwi&`K7;-C9x+U9;)-#fGI z30laE}&p5}F0+-am_TewI1w9_G zNd9^p@}+utx2xQD<$yB4qzo1c;~EglOf_*yCZJ#kE?x&NnB4=^_N?7ku}yd{XE#_; z4n@+WOdgO9Mmfh(R^sN5&QCn3=9t9KQpinuo2u#zAzcB-}o+2lDw1BiA zvff}r3>Tv*)D@H>+Ivp1*uJWsHJ|?Mk zjiNFexTqnGI_@C-v`->hU0gYUq2>bnv<(SwcQr`mZe|O_*{-NVD^D$@o!^W!zJ)N{Al4OtXxcGnGu3FIS2~^T(|4mf8sDBwdC`%? zM|kojx%~nmUE1a%6TyoHo?tez^$S~&f9G7agXEasE3LN1;95mi0>-uPZ2M8}KwH&3 z8SYrAjjgWQBdIoQemU1{BlR|Tw$z}w`S|GI$i!=gU5d3ET<2R9QU{)R$GV6(<)8$! z*uWwW{MR#K=hSxMSnbcXTRicpPHi&joWUkVPK63JGCZb`<&Mj>Hr=FkyLV>ok{98; zaQ5PA>#r@+x9M;o3{mWza?ASG=eAX!&bFI3)yY78Q46rZmhL?O`^1yp>u-rRcCUOr zJ}QSl2r~6Y{bCevB6APpoB^biV);2d=QDQ<&JV$sw(8bS`RmGr`9&BmO1-Kj1 zbSj_E^5L>*RS5EVw9uwEk~mE#wg_Z}(0p7J>xfMD_yaY?**G0R*%URj8*`{IE_OBm zcu-?m`&?hZkJqv9rj-}{K^0Xik;zTsIdJ>S(1+5Rzx(eNQ^9|tGfbI>buLC!KfslC z2;3y_kE_Z2!& z+di-pwbL65-;U+|Twm*1(?i36gV3n(h$Wr*{}0lE&Br{7V1%-1-siZ}@QIlaTCXIG z>ty6Rt$DYEB>*sN=^C_TbzEafUc`gArAldt!m!@7&&!TRE29z`m4oZa@)<&?cb9ke zs$2A)qAoC=)}@t1`okjk+R=Lwzq?m++f z|2hPkUb}%7t2-bc_Y)Js+i&(IrsLF*fxg+A!_;*$mwsEn1a>j*N7D z{%hSCiYTtGwRWk7z1JGeA)8J82P3E9BAX-3DOkC`ROo_Zn+9db|Ou1u^j+TD!G6P(*LJ7;&&FxyZ!2Z&~UtHEOC81 zu;JF->|;?z6Y4rK9bR37b{-S(F1?5=weqBH?|TL}f)5mGbrk!zog;8CK#jwAGlWmV zr}l25-9!myd$E>(`QT{33E*z+k4R80ou5g|U;&wx8L<#BT70~M)M#g(P_r}b^@io5 zxS=$4wTE;aNEJfurBIhrV>5JyNWzQvXT#xPI7cDEh<;@)a6@r0sj{B{AEAdv6u_;n zm$+!}TReVSdiEN-Tv!A)Qj(tZMELIgz6U&^3j>{*)X@`XczvV5l91efrci#(X@ydJ#7l`H#zsFdrSBO^pQ2wx?CJ0A-d^*%!GXgFy%<&*?_xEz^m`J}wSjo%# z_lF;FqEr^|eanER>;8I?n{!nDNXTFWIdYmyl17D4sg;%d$G|mQr>D_i>TaFy9gVZV ziZ9nMDu5RLL5 zvBK{D;Q1b15Gk~p0w9=TlQjeJKy^YdH3Z96+5{;T4=Ads|E`p^g0z{fxUr`AynmJ1Y|pY>NTwjPi#kZ zp!&uDa52=-CQAAK4iQKVUYtnIBgNVdEFTl$GO~orjCpCgn!{{1;Z;VGg`Bb7IDTx3 zmfehRi5A*?0p8z{@jAHWDxcf))($<^1w0!;O5c;!!6D%lD`DkMC^t1keRbnmd5^YuP&hd6i*1=xiu^X(NMFt@;21$@iK{QUr)^&tFx2{8SJyQ9ali zmnBax@!3cAdyO~DB`wB90fSMTlipiv_p!pv5FM$fEEeb3`^2oSq=OzFY|b+Dam0x4QSpSXx|oTII1RnI2iW!>qqU~SU(#jc&X7c2>-o3#yyuBuxWW9zi4;6$vV3{K)WNv*iNk@~u8M#!Fld%~7GWjur#R{cGl zGd)(g!rY?>WEe110MOnc9L@j7p;;lc=j$dm(9Nck`7+uVVI5%}>5FJ)&QNLbLT_}V zB?vbf#Sm&2>DgCkc%GEL2F&);2)!q`ub)x}BT(LLz-)`-V$Zd+XXwl0EUH4s?iWpp zU1;If>I`jg9q4}g`4-`$6(cYeE#|Voq^HJ5dQR_MAfW5Yji}q>)&9xwC=i|+HhT9^ zGt@KPsdnV0ATc)0k{+-&^n@?8sKIc(Dw#asc0&(6X8Ak68&4ZxRJ2y}xyN?r$zV89 z_#8AboClOZOb4uJ=+Lmf3(kjz6zKA$@%JGt64SLJ;e6 z$5%kjQ#-J+>Y$H2=0A#lr1#nnJ^G=797ESwtr>%hqv&ejI?{bQ>&`Dfp z#&(`o;A5dbmPI;_cGSMDmEB5d;Jo9*cU(}I>Oj)X@CYu2gqjRMn6__~5F&b1zS@_oSfw}J9e zAh?ud$Z*O?aS_mrOIVYHyy~92nCQup-vLFI6DX;Te?Oo|*S=A5hn)&en{ap)H#8&2vVl|s;hUHun>43{`O!sogHXG z9V0+ge*b#t1`_1l`pSkq)_sP})A=RvKi13uBzQGj1qk)^uKjd7T*GOV%Y+^}M9-TS zPRXZ2-Jj{vs?>CXYHs9ZU8jO%1Ev(PDB2aVk8i|wOxz$CqM45_68@nx%<|tHz-9N2&8eWHEzc>k5_YS8x&Jyi54Qz(ZB6+8s3tzDY@>i?td z+XI>2|Nl!#QYk`|luIg?<(49LPNJg7Ex9hag^=r9W=kdIk`O|y$o*F2&gOoNMJ|aM zHkSJ?E*rD`-l}u`oO3?se9rgx-`e~AUfA|}J)V!-+nO}BT|I&i0Kla5bfZhHCuv?= zkC?>g7^X)poebM?(02FZg-o4K@lZv@f((Dkk{{fgfCW`=M{S&s=GKyVB z51)o7p=zoB$G|g?^jM6;W;bo{>+3TYNgYoF*k19JIklIoNi+E27>HmYn84+$svM5pKD_*>cS?|mGV<0(NcEV13)l}t`*uz>5 zwDCW~s*J}}dVLuB$TasYjh?!FOKnSORgEnX?tQ{AQbOU0@>+qTw9BJ0dx43C1Kns0 zCDl2gaz@!8&qo!4^t6C6hVh1UbCGyN^Sa2}1Ni?He2!A1wlhK_v8p#y>@!bt?jW9gi+%?*OoM z!3Mb41+W47j-Lmg<~ZCWi1Nw$^>M@Ixw8F_vX9HsdOobz5>db*gTwRpK)bO2^3eP~ zQdz4-xIohIUH!y#55A+Zk%FAR!j-@jQNOWgDB}%IaKJm3fGMJUgF&&k(4MGutJs(# z1b`g|7p{|ikS1e-auWYI%m2udJN*XAOo?D!ZhH_mR1g9nZp(ZjlAXAAb6we-KQM<4 z_T_AbXr=ze;i)E*h%Y(1S>h;V9R*WI&1Z~8?Sd^Lu)!^`$$M^gwUQ=6`oMHSvQy;x zgXd&;!angL_C1ft>qyt+h z37;;9;{nDxw=-rZ)HXBulQZ}B1aX@Ze7xneiOlHV!%1~oj@$mFeWo1hyE^QmlhjYR%myKsy~&P!48Y2p)VAVh zzd@56kmj%xW3SThoYD-kDUa$?z$!T z{JD<(7qD{=04kpaxvAz{eOT-8W19M{Rk~e^PKkC{SznK!;Bj?Y{nPY^0CxH>a^_zs zoiW6C1*?m1!W&PyhL&&I9px--bP1$StJDz0|K;<9V@Cy_ECqYBO3sJk{p zlhU66O8$~LcSdt`sNjSV01HCb54R(AsGeG zZ`X5VXr;M3Vrz*h=jAyVH?m8Ax%kC#e2htilgoLH9PM)tbtL`L{|W58r)Q%%>d6gr zJZ{XhL7Fxw!E{yZ_n%83w5zP#+Hxd5Moxx=&wACk6%479+*liRqIbZAKLQ^XA#LD> z?0hiMjc>DfcJ09T6RWFf(|54aNZibEC718z08%8pZHU z6s^;37^F%7ccB6sckzjvD}A@|0hQ(0Ke`lw@Q5~niu&g!u_8ypFe|7X%hY7)~`fr z|5o~PuGTfE7-0h69})X}30CqezVaX9%6~|f-WH;pztDm9USH3D{Eftfk+L)G-sb&# zNXpVC#0Z1rzHp?S4%wDwmg?dw`6rZT(MX^%+ZX#3t!#7H71!Ehody*3qT2AA5(5C;IJ;Thv#v~jG)5}o z*aqz;pUD0P&v{Ew8lDJrTkL(fCrSPW>OaU%A;!VYU@IZ2%*IQ^rKR_b61UUHJIuNfB6|a=3qJ-WKFTJ2jg4iI_{H_X}75 zJ7ro(Gyl}v-U#rYZ|MY6?SDvn{s~YmRD_?gDC<+JLH0Vwo1R$epXSk*;)EX@vWu$) zUGsGPKX{xG>jnr~6V`pV#Xbhxdamx_W2^$~N7%hIOFQT~4D8i?kr8f;Z zO@)W=6Mb?7DFcwt;X!)x4clS%fWY_U7QiOk?}SAQYK}V@sh~c4qjs`d$qrZPR+YvwHP~A%Z zro#gpl+wQ-J^xgoM~uTzq;a=Lt6M%l^*X5;=&}Aqmh5h0;SH#Yles%&Nx_`zyfoXb zz{FAq`U8%X3oKo|f8Fom-u?@-&EdwbL5gr0U0|EJv6AKcwnh~>=cBw+Ip)$9H(k0^Uq0h)62ep!!<2bie4CY@m zu(C8Pe?l1fX3T+;+Qmi%^{**%Vw-VcI3K<23mXb-3!n&#IZzIhR0VQy(y zGBQuqJE%G@pjWD1nGe{ah>V*Zv$MC+EjIA^Fc9jPJE|=V#2g`t1k93MeK3bGfs+Xi zu=h)<@bL$gnA0VnW*7`3b$jtYr*nS~yW347Y!k6;Z)aZZdvB8a@a)d`pIrq~*$GLi zqtKMImEO4!n(tV!z&*4J&>>Ws=)lJg1d>t~4h4GU;%d%$`;)y^Xf}~+!OwVG^crtH z_nf7X;kbNtbal1S(WMf%g%S9?K89FC|6CAahCwR5!m*_PE^~aVn6zHcW!c9W+qh!y zcvLF?Q8e;;oA>B_P{3Wxu=hBcKAD6q#;6U|EcKgXgSn)z6iN|f(_p2y^MJXIs@3U< z_v4P}yW{*$vUz#OK06np#s#eG9ab7Iet7>r9MbEzzN9PK^#bzz42)*U*2l1J%Gmk5 z@IU5<4SdiVT_tYU)lza_`e-XOpYSxc|+u)K2Tm>CHCaD zFB2wKx41-p7#56&PP36G}CRToDLS#^efMr8k zog1(Q{V!gRf(zUOF5Y@9H{U<@c@XT{R~)JvHF;N!#d6e&i zvNMM+51WdZoB34TSkx6#St%K?;!<^7eweJ7BiFZ04Q%*7&qn;o&VM=)F|RTv(kEy+ z-K8+5&T-CN!wzX!wzeeKvEU4Rd7OaNxU*)UOtLA32$XkJ$}C4s2M5Q)J5%0~M3qMA zE10>3$;nAt;W}j?2!6I#_(qBQJac_rT^xRQ*+VJ#N22Nh7)kjaYZ_P|i>sh(?mUbh zXL$_wmHk)6Hf_4Q>C*Whjo{PlOmW)Un%oKzv!1vWy_oo(dLQW@#LXlzfGSq+sx?*; z@D3L0JrJ2}g$T4ECfi)+Q>rQ}5bL_iP~1voMaaU;Bz(yYJ6w=_pn0K<^+r#KgkvpT zR$jz_UB8)IuvZz39R0;u->l)22V&2qnl-bHnr8_AVpRzXA{nQVa-tx<`oWfEs3Xg6I^xHQ#6VC?umCX41urmz$CzTK>yhX+5&D)R z^UR4oOo=EDj(c;=;5)VJI@Ob766G)gt%jHpe!mozYY$}nv zbB%S^{B+S%dF=yjCclzWoV97WQn^eF1kT?7L5pIXeFT`vqPtqR)7se}Tc?X?u6wOl zX6Ieq-Meoc5u3qe0>kcGdJAg3Z@$*s1j;cIP$I(J%>JB?UJJ7j(V=c!`>2x`R4L2B zDd70^QK(-JAN;vLPJ^~69UU5EOGQ@{t= zw%D3je5OkN;$4EAMKIqCk>!VEq)EaczI*#Cw8jQ(zlBwlMG)SdbG`}2s#Yhq%6UZ8 z7qs0w}*h)^6>L-O+UqRiOHsm7peqneJAV|#ALJNf|zzFM>jyq<0 z%-2W}nBC5OH23Oz^m*g7goHuh+y3H2;H#z4H8M{MdVu8?uo|jANata2xm8fF-_GZN zr~+Os7cG>3-?3hjHou`IM|N*Ou4I+xujqs@5~evg9#-y}vzH+NNhDX94I`c|w+=+= zCaVn2eR|t~0d7{On9xl}Y_2eQ>NERkB>AjgPm9ICh#5%aQ&oWd!_G@e(P({)8|P12 zTIFfxE59}Y^ND$1g+`yRPc?1(-8 z?qf~s1A||rxA<0wx|D#`ZPZ$Pn~mxrYT-7u3g>tnR$hT>Q08KOy2TDN5_q)UWdOd| z?SfsN8>$2v)!q)RP*C&POqf6b<8e!sW|53S>!e{mk?Q<4R&gqE!D(^r@Z+D#ub7IPkJa>UbJ&MQonELI6)e=O>d5Ggufqcen< znK>0IqF6!X77{@h;WdZMjtmtXkM1i;V#k%1*tV3~D$(9|Rz}kt;>YW`0J^amVx>N$ z#$r&JA1qd;qlPvYdlI``SgZzSn`<81WPwr71@{< znNebnE-^0Q$N!EN>5igyy|9{=eE}ub^x99+=h z%BrIR_3!diw41cC@c6;`i_DyEUwV^Nv6qVxz-fK+^d;_wpT^sj4#l9I6s_Roc4^B4 z#=kQ3jWm(5CJ{Yjl|}G_uoHrZ9wJDXLOU)QPv&ibvMtSXz7>^8=&AcRwK9{=w~99j#i_2I6a*hew`WD#-JUJhjK;jNpGNtfxe9XeUCK?->K1>hKuIy`Hab*|rj>`-k1HD>0$Gxzl) zZb1;7f$eUJh};ShE=Wp0`r(IaO=J9dSIB|>+hqo_=kHBLq`j$-}$!vF2G^{m?+KcfF)T`^q%}C`0@a-5;{HGJ&?Q=ISEKl zdx4Zo{1#kbU#mz#RV3hpzw-ICdgp!6E-~QTy7+WFzDgLRu+GAsmVj1o2k5&>MR+Bp zXN;yZ#Ka=B{pVNTw2nh7Pp|nqWNJRot?J{wfDd1q;D|kR7Wn_(TmqJ=v@geguo`pf z6$fH4TxUi8B=_=LGLE@OZFmw-Gfv|K!Xltwhi33Zz?nqO)#3$=dXUS%%hr@nE{lAz zhhpM@Q5(CBYCOE?!X%3f3qvbDSc1m9(>x|S`mpo?NPl1qkiv0+FAX*?GWtdiw1QNv zwm;Aw&N+@>ezrgU!hfUl1ZNkG)5OT0J@`qT?{&hoP<5C-ZU z4QTs=_|F%4LP()dXQ)ZUiO-whYBw*ExYyU$+5|!FL@qVzL?vmpdkV)4S9I@D6qZx8 z!xV~t4;S^uaDkW)v%oKRUE(n8l^lqi=MhnE{twHg@)le&`DTrit)vJ>;wDA zD=osHOkg6|#GDpXnf?1H9|b{E#%V9f zh)ASjHc>B8k<~g?GS9tmb{bJbEC@cmysp}_89DQE+gISSk!|W%+0Z!YXQoPZVXPs` zg<;cTViV${V=H3AiX)z-qYAdEpLT&J848 z(A~3Ac&uu=Nkq zSwfB65T#UAtLBakl&*K1?^|MvAK@+Wbl);KMuR6U_aKePt+ovm_b`? zTYztFx2jvG>gBGvB)-AJXB_q*?Knt|b5+P1ee|_wJ{66SF7+?FM{1ZtE3)=qUfwxz zav%DraX0>LXBUuq*r z3GxJ7t{dLxgqtXHUv~tvzP^J-_l8W)ewe6d(J;DA!m^G7I&ZMnRmk;{^#+#@tE7^R zix^Td3thhMtg5%=zM-#AvlvN*bj_V*&!hMuaiy%%eaD(fjMOnCVy^db^`?D*m^wc* zhpsL#r$w6|7ImmOA3%TX@>)+W>><&)jn=G=jth4?hyzuB_-`>kX2uwRo~j%iDudI= z;qh`*K6XI$bnEsy@5oCwF)?8v(KT!1f*_kosdm10*66w4epFEbUH9r7m zEZPP z7TLiCvNZ`azu`x8W9mQ|BdS!j)h#}Fw$3(06$pbig7>^m0Xy_GU+py@4rcpwB8yrE#i*OfU7cyd*xR7eJ0e)bl|vwh zs&ar^(0u*{y%5mSN_e?6Sq50eS?!Zm0HGljO21<%#~L094)uv z7JSJL)#VM|Eg0gJ+W4C9hM%NI=*rS@76aHOnmdSB3NE9*UvUQCqLvw9i0Rv)6ymr~)j?{Z zKt=F#=Qe>ST`vwW3xFUdFaW22cg{1@Dco-gh1_M-%A`P*G>kkXes@t#rK%4`7&U1? zm&tC42r8~ZVpQE2Y?m!8{8m1S%y)I7o(b62kH0s@*3A^o??%in+$wuzgpaSaO+jyM z#(+X*N-Hy0V*Kaib<#)7EJd9p>LP8b-t*9!oVr~8^Kp+o9wXAj1;P`O0ImwSXjCBg zOj+zfF`RILxli@$VHz!?;3RYSZC`cmxBz7SSe4bVXm~OAA&3;OkWg@;;v~`J+lBaX zOs6mj`6`|ZRKWZIhP_n1gEhulsN*Dha5-`<)M`TMI+2h}B(wG%kQ!=&`#W~-@EUcBTQSJ`peKl5@qlemF=$rI@Zo;#a6hs-wAbG08823LTQrbOWB_} z_fRdMSCDL+7D34UmC;N|g3-r^Hx_&K_33z3mIElcQ`LM#17-gV4lHjWaP$lN6PFU7 zyrbWL*Np*R{N_W8S|Y9m$J283+h4#JO<*B97Q@Gw##n|VQ+*#q6shfCD^3doqtP#v z@#x$8n(-_il!YSe)e61vU^-UTar5MGQsKm*jpT1AGN1D-xVZWK-;eShs2c#bbRiMs z%I2SozCLzA%&!2@vQQ`@60>HJz*tZ#*#}Vw1?uujo&qdMOVtyf6{f3W1I*v)8&;f##kjllKzHox8OhpY>+QA(CVF;OB=1lZ(rX4`1fQtB@mZ8IHD)cWmsV_V)HbgWY;wx))yC0IMm|sXMRd7= zQIYDQOl} z`V{|~QvajJDc4XqvOMr-HG9)@euD_FY~C0c-GEbH_(@Y(*d!+56%{tVfNZi7Ex>v9b>L^e>Dobw4|oR+=y zAX#+GJEh!Ne08cXZG`Umw&M$MpXa_V6_1XH+hY=ex~sDb$Ymvm;R0b~7!lnFxq!M; z+zs&ib22s*;O$5}EBm#b=^{tN)6m=f_CVO@c#HG2!tc2A->CNlFzDJ)V83k`F#wmvqDq1Py& z#%JeoYio-fn=^p|*@g8xt6P0y4Z~Dj-YWZ#z-|YWI7APF9#d0ZYp{DD9o;?<@x-U_km(VHfe?Uu-ni}x zFD_8mzZqD$0b*wUQyGd!z-(OVEWEouu)9^}<_GqK+{)O75bd=`|Sp}J42JyFk;KaCZInmDM8OlDW_!bu56(GNj8}fy8|{K1d>Sy zLQ-{^Vl>5TK!w%!J}H3)ZinjkJ{Kq(Xc=UcknWuJb7fSw7pXxIRE$kw7^jgbVJ7xd zpR<6MQ08S#G)TDzHOW*zNR}Oo6=POLtu4(mxl%7Yoeh*@j9;0ghjD!Co;t(StR5Z1Yl_Z_>OyjP}?wX10+;!GA zfXr%l-Cu^EnB!*bb(0ovP*NYDA(772f1|W2r?X`R_)aKTJ$bgV!o|N-{90Zm3qm3K zT9e7q+ylZO^wxAgI})_U4^`EV*4>A`Nq{KraHu>Dj1?7Pu(>a|mGWY_{&+xa5I5=P zg^>c_V|FW-&)_Gp=C0er1l=>khLz5(7Y8`lNU}@VCjkarptnHN4b57;17_wWkS0Wl zg~F;rlgXA{Tp+u;EH_r0$Wz%gYD_7ALT08Sn6RBBs8aLfl}TSPMZ|+4FH|84aV=qT$oVcKOW8#H|&) zL)^sTyaC-b$M!7dyJXR6OPjib{TP-Bqmd#IG3c!X^|}`VW(87V`~AiyRxr=BE(x@O zgqU<}W*#-L5hP6yj-Qw~D-n3Jau&w9oT$2o#zVi79leBLW*D`tQvO*y4SfiE*Xs{B-@fB&9>l?B z-&ax{)rXvTIPW6Yr4j(R{eF=J#TyNiDP1_G^0L1^aQntE0dh>wy1U#u{lG0F`-%n&)>-$pU>gHNfQ63I)`Nt z!%ZE26gd_xU_mJm5Rh`HhB1}LLM%pJPjF8vTmft+KWQY&PrzlYAC#*CPNAI^`ymWs zb=F>=O41hZkMja$G}x{^B>u!mto5!6hAQmQ)Z18YqHR(A~U zv05Lbtdm?gQ7|WflH2cjYMvfZ_q^EspPHcSbRkMAJYY#RSJb}m)e)z-5Hg_ZSsOm5 z6W#ul^a4eu3tG(*aNFDq8(Oai#8rq0ed>6(=H{Y@6FL!V^mHKHm+m?=uvnl2%+LEdl{?IFy|d~#c#K*(wab^?O|8n8AL#E$% z139RYG_RVVx3HAVBBG0qUQlO|ePyrYk%|2leKO!01x22oy&4BdYmmtOiEQe?ZOD|* zuWxp_sPa5jy#nqPPb-%zs01ukRAY7lTK!feCrthxT5yBgoiP?OJW%T(2W0kuvNmP# zuf!T3b_%N{QkTFU@7%Vzif&>7bLWy92%OoUoLDI8s}JKpK%rw^%v1&|IAez-O)8ZI zu2t5Am|Taa(VO-G4OH$n^bNjx!$>^c4W;gmfA?V=8gC(j@h2Aw2P+s_FbLcWZR=VS zkXdS#NiYv?of39(eGNv`SkNg}V|^KL!-r|P2iZ?`<1BcP8O8phr9l|IkN@{_j*tLr zSUYlwQ5^6j-XwgyaAlaU%ild@Zc@Q&oGly)&gmbcvk0no){juzyh*eZpm* z_PWaJaG-ovT{6&mepl)By}Lho`4P+hBIp`NtfbM|53T!%=?Wk-fQhsy^aa;O&8;9LEb4j)@-m9^kYM2CDoQJ9$4~7 z7N8yxbjz~Cy!>Y<@e!-|SG>+|&3iRTGx9~TA4Epn^AO;PJ)>TSNmP!mVigr_^Nx+q z5)tHkYCzD!2ExN)$FrN}hi7%hb+p{a?#VVk9eDa=WJxo3s`lx?BMka1^`QBF?d2#v z#B<{m*l2*ed-KFcfu{xYx(t%)s~hLxbM9Z;`98n?`=)+@g%4v56UJ#4buDHaU3yNH zDY&1sBcNli$=uXjs!(WRhkF0ln!b-q44-e>QQk`pNfW?Z20rUsqzZjCkbK>2X2eqf z_v5&_=I{QUNmfynWj-7H29CbAu5hQzTdfo8#&vA>mu?KuoAD|@kId8@DP zuba7FO97kS0M10~79CSxK@cEPf_0fA^w{~&wDeSNwL7Upp0^}W+WQ3LhG))0WxNpU zX}nkmM9HUMVUV?QcoNa+5F&q8T2O0-&RC1A3>IUC>Q}Y^bi=cmto|5qXf3GOpF&&) z^x3e?Dup4Yn(VvR>;h#r1R++_VY=z${q^B`B6!5|1mb%7Ai!Ql)2+KDq_+yMw}Yc| z;8@%-x8-|0!{=P-ts0q?Ri#MID(mrk9Rc&(Rc+Ks_wnnDo4^-Ca@|?y_4*HR zot%1KX8ed1IZS%qsxX51fd+vdoI^3bB95e)mx-+h%cCvxgw^&5ZS z!sVDl>w0<$UW?=1Lhlo-$9dgsJt|coS!{8maG&RcMF?t&4|(w5vrD@b4s2s|<_W-* z?5wBs-5G7o=(QI>39E_Lp)j>RhOOLFNCeDm6d&7K7Eryo96E0-vb{P=*5c`{padD& zBWcm{@<*tIEw`%;=&0*2;U#w#%vTatU~d$6@JJ_jENY<>!5*a}<~MB%;v^Ra?guN% zA@}*7H9QDDP9kRj#n3JveN42t_uMJD>B4zg+L24~y!wd?@DI98pGCp>C-~@-4JL#% z4+6)WQC|LKxU+Lu){gNYdh)hZ>Wn~jbf16)ka99MW&0jVLB~D=t7PEotJmB_c2o~*L|*qnVoxWUtDV- z7V{uzt-ay?_zw+lT#7S=64KI_*j=QSSEL@$FOPZZ2lz9P2k)AmR>yU=PF)T3Eej=U z9Nky3x|4ohz2ugD02%fP&(5sq9vTu;8S%e@;Yq+t-0<%&h`towT;Usob#7{c;MzHC zy+8Ax;yZvC2-SqMT~x;(?50kfm_63aMSgzj#mfrc0lfwrfi~{2&zZ8!H(7zxGZf&rueSSH9{qwz6vLT`LJ+SPxq&wtOwr(z?LSEGm&Y>V`cLVb8%oK0d15(@ zj-zBUzMjEy8jYN2o z9!4OzFo?>)aG$7~4e@O)l*-E=kV+R&CC@rS(zV{4-y0`XqPCW$Qqp=*B%hbBLGnr3 z)`tm8sXQlRCgOz#mxA@7Pc;LfkFi)}nUUaRSChfQk69o8#9BT1+5LY@Bb@F2}~# z!>h9-G_hOEJMbn{H&2e~?nch$^e{0@&8t(iWVTRppPp0@wpV-pdIk=VP6 z0_riy&X%1DlFX+(?={25o*$uzK?1{x9xCK<`)Z>Y5rcabx8x2goMt{0Dzlyq4o`e5 zXUss#EUlamY792_c-_HOJZ=E>o$SO)4U>DyaZ{%d8JBo)f)iJjf6K()&=$Mo(v z2K*M+%Y$dykw^;7J`Fo2#_^`is3V5E)i!EK`nqMB!=lG~&VA!+58KczMc`*NLN1&d zxN-oSr`MiY6-;1g3`)Fy*W5a=;#Kxk6c=l-mZ7ADLIH>N(F>k^R?e5BEU95-UJo|A zxr{xIccR=6g$3$CxsFpTkC(Wa+3?ZdR1_qUZ+^~c;m*-!M3Jk@@mrqHrSfo;#Cx;Z z?4Vus(Frl?X%d$svAf(Cb1mF=wmEFZIy;i=ed&4UQbWJwW6ZPu=X+R~?y7p2C#1Rx zp-pc(_WjlOGMbx@`t4K`w@P$np~c^_b@)v_B1c)f)9X|%DD^4Fyz=)uyS&m+;j7a# zFkrYKV2$`B3aL$qkPc8MEpyb$wDdw=4KPW3moX3Wc)t6NaR+VZ7i{nw*->#1BuX>h zZSQ;jx#rOiq9W>9eJ3h3fhF?-+-ap%xy&}cce8=nd4k@(LS3bMYF6(J(%wXyH@>D_ z`?wEnOlaUJF_S2J)Qa3HVqaamv1aHV3)u=hbBUavafFF^l4a;`Bq}bpTPbmZ+v2gt z{m@b9u5g`)mQD|N_#bi?Tf163OHWR}@x^X!+tu*~%Z0f^&!4)vw#jC1Y2|M(i62Ru z?28C|mvqgJ_E_GMH(3-n9Bpa9aYF`Jfm6t-r+`-Jg)p|Y^6lM}Pl5F>_H1q2AJe+u zP+jk7aXQy>cvFJ1)!UA+%hDeT5d-|n;@9pXoh!baFeEY6 z*lDN?q+7huluC@72lp&C!AK)D3HVZ6mAf4Z!|W%Sgu+$??}Rb#1fMZ5Bm^|yMH>*R zFn7&&pfy2CPMBKv*Y6g3soA37t2zfi9iw0!YF)NbvEc$Gy;+}OIZouY3zYG=Ca0L| zu|KqiCOd>=h9i!s7Q_Ht_u-4zhS5bjvo$A0nQO^;<<{0A1y(OShR{c`5J zf&*3T^tv6MuaCAYh;+>m_7HI|ZTtxetTTBhYFFkj=N##lp)~MKUcAxOC76##MLD0t_zA%L}_GOnN1*taV z13nrzD%uKC=sjyl&{>Ei65QW+TqPcLCCM*X?U<6WEc3?3Z$CkEZPrN@hzCmxxV_>x ziG&%vEZ7X(L2c*bva33TVgmxNMv*8k@L{7lu;H3s5|}keduAAs5sk6fvhhdcaeEcl z>X0s2kBMN96WV)40$H(5k&j^)z|kAd`|ihCWy#@a;XG|pWNW9pb#vH=kqJo2-#=gF z_{6PO>lvD+D8ECH8U%i@U+GgvDS8%QR6q^2MnSLxEiWeAf=R#kIYeyqC z&i;sj22>wTt4i}2tz(R_?cPiedI%H4Va0H9TN*B?U#mUxvb81nsT-jxfH357gO)u? zHH`S755I;#kCNed7zxX}tMP+EGB`ycQ)mo;oWKF2IBuOg^6ReXefS}z)3UBG6Nx>& z`khaIXnh2FvinO)?AZOwGo^+n%{7ZZjK~%S+VIHu>El#RFj0bWGY{x#!B=lw3!jeE z)0p0~KK3GN=E?X<@Q+!SX?J>}RihqL`>JO5oGAj`NvFMj&M+VkaLdoe{7-uT!QHUDX4V5P)1X6aO?|rj*g)eYtuV0&g+x0ug*zq|!5>!~ShzN^@=a&hOIIJj?Hhlf%0=#%v1)VEuB z*v`FMda`z4__J8D!g$R>XL!69UqfcJPLd)o)#T8@-a|pUo7cTP(d8r3WYbLny|z16 zcE3e6OGJFRda{8Sv*rG{O<0Ng+GBlJdm>rVfm9mPyIq#i{?=(qw1I8jR`Q{Q0gS3UDLhKE=^|@Ff1l;QZ zc|rl}9{3dObtjUd*>XukM5!5QHWhUVK5Z@Xoctrz;rgQ#kjdbchat{SuB%>?7(rJH z2!i{(NP`Aq6ZioCQ>UiSdoYWhWu~odEKBb+)t!!OMJ>dNf+_jIQosuS;Zzf>q{1~q z8K0i`^-l&o#ZS*l-yhp75|cF3|0rh3$Ji>zRGtqgc1MDs{ zY`EV0bU$W$p8hIE_9}DWR4goOD=yB9;pRG5+;Yi=sZ~*UnM^J``&##=!CHP+g5=40 zbQxZPH_M3jY2f6|^!C_%c`mi@L1)oDg=YXED$6c+2s4r1T*~L1qkjtkQR( z^GD`~$3iu4=SwK!;t96oy#GjEkFCng{>F(=E$&zl&40#wDDxxc!8hFKJDrK{3u|%WtK)c3M-!XDX zFe5Y^_t{|Yx1X(v+-Hzx)?Gw)jt_q-@P7HZm~s~T60Gzb(up7`TWNAGfX-Olv)=r+ z^#9TJ-eFBd``)hwMMR9!ixBBbl`bem=}mf*E*+#vkrs%GNbk~{fOG``=_Me&iAt}b zcL)h3354Vw+o^#K==Y5`g?_Y$qW+pT1S()`KpWk;GbcdF$w$EfyEyaqPeG6!w zoD$Ni6|b>T5vX`a5~;x!U1M5{3br`Tn%Lccq0ddQ(T~k}upE-op!c#dmg-2uf3g>p z(y$S>nKvJg)c?u`F%LN))PSg&He^2K|6|oV1XMPir9CW@>jId1Z{6z9zGM=^}Bu6&yPrplMr!)-ZfULT(eg6|#<}#!{V1 zv;y2VB+*NCWrH7@Gh>$_GdEQ}Js=hweUl$(R`dESx+I5d@IHL#ED}R>&gIs>exUNt z+hQqwLDK_xh1ZF*_1`@j)WwUZI9zL$5XsShDv&eYTYBt{6H1bFd(7e0Pa~F3RjY#0 zRx5*K-JzYzGa?5i)eoSX6@YXCg`c-G53TgFLag_uL|uP&3M?Qc8oV>)gHO3L92J>_ z?@WqCz(bvkbUWZ5B%eJjrV8|`(BN%!EnI)75LDlPCk+VYX@52hhw58(DE`-mldF@a zh5}++slJ|bB@3E9`1s-L;So@{V!U~Qa*(UDDZf~0&OiDT@ZdeSJ2|jIsfhjQ9a1KxEbb*bCL(C4 z5Hex?va#<5-Otd|VBRK@)l;^sCblu^txAqa#C>#!{#x&Q>4k@z>5ay(P{9g3zuzRU zwas&14j6m&JMdB6*85Nu<>G*~9Kpn=YRf5<#9t^dJ$TR1952TP#~p9GJa!$ADE<0PC07sQWt z48NcIHk}=eU+Qkx_J|jy zwHpRBkA_nn>*a5u&4!MaT8+slM)^{Xo>*P5^L*m*L?E)gT2$+jti`vB$UyMZC0jI< zk@(9sb`^b8jrZ*C*Pj zI#w`+v7JS57qIcga6gqJ0mFO5XWe@a&_~FWUx1?F^{z2dNaJqu6{jdM-s19`l%+dVLNF8si z!C=8-nM}%+371!tLAd@TH}$;Ts?urNTTcXu-#4v+RyVjVTjY$0Jl0~f+Pf1!U>YC9Zm&6Ioq`%W84Y%QpcM`+ z3I|e_uF}hLY@AIS6vLphT^u6!eZ0AWUf3`T?^h}zJ3I1dguB$dZxV=w(zC=APsSW5 zl+up|j3$_`@iqEiRBc7bEvLfH4^@EG6wcqNOmQAO)ih(zNuJa)h)2p)hR~O=qFq`j zfl1x#T>gj|I?ICle(QmJ>sJd3piSAe+$2^*%qpueCU%Cm9p+;QH#PTa(YTQssA6t} z;N3I5neivxR=B_CQ9+UW)LUtq9>|0YkE>Y%4e0C6_t*6uw|^z@=JcT~u$nm&@JZ|= z($PM}{T-gxD2>lHHab@^zMGIO(Y#wZ-EiOGH@23xCmv4F4Z%d-t?$LRz}+&F50<*X z$pN9Qz>PkoHM2mBjj`nz7jyLhON>|u%r!FrMV6MjpLokt-c>8hRm`JOo11m;%WW~H z&)x*Vjb2XF(BB~-!I=A1B@Bn!4+_8};z0v(cl+`vdAMW}O|`wl^f~Q)@m!9cxXRO8 z7YDi)Ov2(B`@}D@6h=g2?`0_-4w_U3JwC+hxZ=Hg{-MnKDfkv;)4e;EHA4-pY)v zmoW&Q=;b>u1OU5+iTfmlUu`HU&)BRyoeiu1BOgPxe1yu?wHD8&4L4ER~_n;Jsig8}-wq zTaaqfZJC_ee>_{ep|xCMaoJa@^_)O?PLV@2vslK8)IeyO;A=UPh8s$pyaHz@;xezv zpmw0~q{`Fk)hN*Lbc+YW-XcoGjW~Clt5)!l*-A_;N*P}p7!yS2zx(0};`}SRH-f2p zvSnK{Q19*bm$85iSX-%a&GJC%C-}K{yqM~_1#%zl7tVtoBCX!a zceWe1|}>N&UYsRYGulME-{9ys1);8O1DCLX3p zJ3aQRiUX56Pm_ySe6WNwRSd0<1PKz0BzlRt&pFMpc8xCxq#^b0_`AyG9H^P2M?Gd- z5jn}l_rU*g?_|?Jn7T%62@hT$kPn`6pe>JtEQ?A>WcStA26SMz9)z5yP75GrkaiJpJ{LbaN!@#jQ= z56|lxug<~O5mUbfW6ceK&DA{v{nU{qvY`duyc6FJcEZf|I;?_BI4#E8xSIX_w8Au( zVIE(zQL@;X)vGQ3uOm%$KA&)*hy?zuK6ZJz*L?9X))~N(`o1BRFHX?ACQGd zRlRO%xk%d0I`2qEPqi0c0LhVtn<9LKUVD~Swj65i;ZSLFS+(eQ>6cpVBcZ0tz*RHP zfFvI0_tAUB_ZI?cRnUmpKSc?b{K1r|(HFztVqTy2@C9dM^OQI_w3Ze=Tw!=smBexJ z+cn~oVAY^Wj(4-YhI4{mZC9�N?ni?{y9G#^9lrO5AQyC^rpp@1P?bdIz3{v&Ooo zC#b)R!-7vp+yr(G$55=I74IDR4#cm z8uTihsIw&TTHK!o$AhAzM@R_C+O9~aR`hBXUx5x7tepCM?lr77i(!7m6~hcEMTTr| z2pN6|Se&kX#;GG!z45U3ow2d8&C-==pWWH}ulHv4>t9osXO*^E7uCBj%CbR>$g#;^ z$(BIlms)Lk(Ll+1Y?9ltuaH{w$I>2&^m1%88zh%@pWh@Ni3pilUM1l${{nQB%Hw@1 zeuclO8~SoXOl|o|KJG;YGdX5mqD_Yx5`O{H@oeFd5LkS$YvARet#i*ISKX(g@ndW_ z`6SHlMd);-LfI^qwe}Xc$89^E*qt2Y_Btw@_|ujPwR8Y%I^Bj}Jejde(#peI1lnjA zdKmja!@Ar~+EnybyDbuaVKixQWHoODl>tv~ET9=H@_5?Z|HnZKc7W-uGejoaues+e zP=^OjZA-v(-Gx~qG`8h%ml1Bd;fxv1iz#)@*UX%P&F+;2noko?GC^dW^0?fk&tlQ| zeVzPY8y6Eleh+0CoRURxQWQ};P^5llLqI<@x&Qo(G9tgBEw^z%Aeo=?YZkc zf=P1mNUr-eM@5j-07dQs10TCec1Fu6UzT8ilRE<4W$rHa4CvZHht%k68IF zUK?`WQF2~1cL&5=m$%k+hD*Sr^}TVDMvfGJ3Vj2o^5Spq4iN72zn(F$UiZDGp+IrG z1J3>Tdg;$mq0;z(1pY5tf?oiZ!H|gw?4uU_B^0Jv6+*KWA|{4%le*YPBGDmdPA;;9 z@t7h{ULwBCcurV8W`4`jR>$3;#F~MwRI?6wMW)0>HORu#Uw-VS-4b*xb37cjeUxx# z^2S4IO-9HH=;Sc8-4egYzzn98;X*xK25Xy4sHB~iW%#>LD$RT2<36iKAiz(1qXQiA zAK=HuIlj4g4ilqXb0ggYOpb0b-S}9$o3pc4IGyzJl>`bWP?99uxtZOxT6t=^bcvGM zQ(xHnI|@35w?Ub^!2CslLZb?b7}DB!a{J~p6H}cldlTWw5PW~_Z|v8=Y~GG49ua>| zAQmUv@~tp5)~?whzxhh$S4S(o^!*0}<#?oPJh(O>qzXtwA6vDU&7Zsn4+l9J!+6Vlt0&H(i;y`7EHF(Y^ZQ@?m{$2^x$p_TVN_S@nkJkB#b zWed$qL)=CjHLUWY%PFrPuiX7%(QG1^D0G^;5oEz)c-qc^G6cUP2?cgDkJ)zH8#t~H zv>g-%OO9CQX4qOW^W1{H9QJ(>^r(ph_0&gTgp7_7R%kK&<0-U3@A^=x5UOJy5PrKo z{g@zg_GskB!2Z&E!$SNF(sC-m|H$3wS zte2nAuIj?}a;mP8aM&_Q21LleTVBDdcDr34=>r1J)b!bp?+sH51yj6>run8N6bno+ zQ`_?d`2DqsiJXM+qysv}kHIc_b#kO2KwHIZZi`see&SbuDPKe=_5vdLcah;oOiAS5 z4^dEDd-rvt;?$nl_n?60w#{q1yU-GCF|6VsJDMGA$PmL|#|eWz<|g8C0nK5hsOVIs zaJ+r%UX2P#bLWTMsXR^$e`WZaFT2&~PioEYvE+;I*``24To(m6m~Rokx@_6xXgpjp z&c3&jO;-LW`Gjezid%hQK=`Dy3-NW&4pk-}whZ~Mx?6@?Q1vdQiYC5$PI!;?D52gN zU<(|@-hz4BYs{X$uzwJwOd`2@wp{!B3SOb<5H(_C+gY(U0!-m1;zR7SXAa%uJHb{M zxit31iE^)>SIqo2LG5Q+A7YrRWD3w~;Y4(Y&sCiDSJ!c=-9#|`t>}YM^9RCj^Qi8- z*R-I&)6c?3q7{~UL~K{Y++ye_8{=<5wSzd9K~Xh0~2kDcuai1)zI126v{X z{x>2mUXvW!+>6WSI4Iz8UC@)~uCDEfR7L_Bj%OO^EA`iNcJe^D8@E#AQA zh|F%uYiO7{DPbu)R_D^_n=BYU3em=!A2?? zx6*C!pK6pypW_8s^RO_)7fM;pFCmz&9`Og#go9ylbFuxi>c1$THms>9($wj5jET4i=N-3BIxbQ?x@aw2M;y~68#;YW{@>cqRrWEz{8$Wu%&7feH)fJFi{Rbn*5x7rb< zjyPevebtjK9!%Wxv=hAE z^=6y9F^kLm$#BpZRTMEU)A6%cm3LGZ((Zz7C;9g8rc`vN+NL|a4^SK@h1@>$i~jg*{iIH z*TZ1zqzo5Pu;VjM$&Ah&7#r2i5cEMCnNWWiqLV}>HprtZ(qb36gxv<2CbSB|<`|=h ziDoI%gm94|DV01ckxE3>Lv|eOAdVGA_&?@=4IUH) z#e(DZhocXMfBL9UD2qNxSZa!9fuoL6TrT}M`N=yA!MzcQSV^ITWIm=|38fE19QKRk z;o<}2M=sYz76z5VIRRoGox4 zrCi7}ehGq)7|?y8QFonkq*~RR7rw-x<-MPK1#1t-g{Zt%cc20sFt4ayOsPl|Vs!Ri zB9~Gu_#jYi0SqY@pO-eE&)1Xb#Sawp;RoJ+)HgdDP&^~QPE_lhY#+bNc-?2UjA%PT zLbpYJbhRs3ZQNTD%??k%j}&Q!1Ki@PH9ii2-ptv<0zls?cBIqIq5dur5E^*Xsdj`* z3WKkyC!T79R7jL>@%8quJfQ?r-r}0NOLbLB8#U^r-25KLJuE>6br$~U!uEC_b|TJU z&e(N2BlxJv&dkEMmN@sTH0LE6)#RU#b+?aD# z9}-#{?Zyb=nNL1jeLzbOOmhN)^IzMshR&g#*NcW9HgXf~wcxhG!SSMbP5(q58yhBOHzPuxlU#Z?#zi*{-2(2@o8tp3Rq z%C{1fH^Dk9hM^#0+%!AJb&9Jh%k`N<3I15UX|mS4Y_i(kO$E=9)Q|Z$CRw8MKckVO zl+#xF|6nG`d?XR&$J!0^&c0<-7csCr@dVorI-VZEvi!J;x^Tv^^}ZVE$q`*Sac-(> zb|S`2@}%O~6#Jw~2Ub8PbN6KeMZeG3i|>PPr58HUjh|Ez?tls=bKqO+7FMVE^Y4LEJ z!doqak()Zz+SiCnEfRglefy-bEPB?oOK>mb(6lkuEgdnJ|6J;KK91d6$;^KD5krs= zD%j{ibiZP8ck-R$zMH|uE5Ceg7DC=~=_91<85=mG1h&G?>mnL~k3gABPuu)hX?uIQ z*Hj8#0k3%QlePg9S>Yw4@NwWTm8c*iu1oAyGMIMIy97S=ef#W0m|@bi?gR>p>n#F^ z#lKymk=O>55N=%)@WcXJAzMi|txt$Q9vfPd9oJ4(^cBNgaE&S0-Fxd}tZ9X^J_ zH|n>3Hfh027A=;&pKYXC3K*Nis3S!HYEQRG_a6+Vd{8C+>=ZM=Da+5#-`jsBBlrjP z&urHt*t;*H@ak&ay@RULfSSaBO1#H86$zNVPClv+==eWaU+qaLV$2=#&QihwBxtH6 zD)b}p1pxPXp1Qq!ET|j)HwlXe7|`%(|9Wt11zdd^-VidUY5!AeYq653U++(Tf5 za2Sa{g^jE!rcciHLvTl5t%b?h?BBn6cB~)%p@mM7^769>bYGGCr*YNKsfqk)dfNGJ zk3{Z75d{&4e)#BH)koT7ckX383xi888a)hBeyBoowJ9KNI*z`_O?+Ame>*UuW>5Hj zWkm)9&nX5>(;?Kcv))m~f_&(hW%Gtpo_Yc^MSo+4DS*Iwfns+Y zN43oToITov1CVV>+#g1E3&hBrYy{L$bvI+29)})}-@Ii~>q^t)gDg~|OE9B672hJBX(;aXy=$!WR$?VZ}+)J=>=eEcU1%iYu-bkF@9LLF46OPtj5 zz!S0%C9ebW;&|-meYMs7JVoJwur&$UwEq z^gSZO@GAQ+y2_DqOyGRD55 zcgR6xsZhZcEWx*dIq_M$X6Laft#c1-7mWAtD>C_6-Ns+zxUCR=mFymb@aq)jqFWL9 z>>ySw);Ohl*Sa0MUSZH<$;3Y|{mH&El~ZL)SS52>Ex#Q*BwR7@p!(h|c=Iiy6*os- zLcY4|?RIP+3Yk;#1BsXz>v3y#ZwGd0uh^D9*qxx#RF>Zv4+ni8Eq1SgA1kfv0idfY zzM}_39`HqE$--}w3{j~bglrVc+wQ4l+B!gGGl6*Bw%Da?9hA(Zh&aGnSWw}|?>xjZaE`-k| z8H4-yZv}_g%735pe}>-PCgP4-oQ~g{HRc7)4g~6b{((11wlULLbg71@7#x;mzDrs} zoN5=np$c?fdU7y7I}TnMi1yZI-a}L8$~GjR4w9Scndq$ctT2!31hhr`qTG<5vQctN za7Q7H=4-Desx-3h^eMxrhTaAa?CTYmtsl?Z7p>uAwdo_7So1w)owOC~wzO3e* zGo~cQnFX+Q+DFSI2ao2V(T!X@;clHKXsbi3d_(&JmDc}LM0ay3Haa?deqKiC@jGIL zNw1F+R>3F*r0BV zwszlJQy$`;ZJhEem5YgY>;I&XqvygOcU#PDUaem9xABfmzhXsoqZGv9p2-D+eS z0uA_GuuQ>^cYjkxSnU1O!jjwdrjn3l_jz_aVzwd?esqO`Zco&8^Po0e@{o{ykrpyP z)3slh^`Ss5DH?2JV#~S`*#l&XZ{COvkjN~b5HVkyVc>H&BYc(QS0+1&GIJh1E!!Xo zn?L;wFe1|(zO=)k#-|V>U(fsJ{3=^>%^ISYWB02gE%QyZaQL{S{@Q-oUQACroC|jm z8Y2DYgjSaz#r!5hkzIpUfsJg_R) z1LRv_7#du4Yz`BUTK9JDR=f0eHZZ9}Q)iNW5N>~EC;ty?Ie;OpRCbYzct}RS0-;_= zMyX!HGWCetYX4B=eH7;temMgG=1ma85&7RTx!T~bZby~bjrFFnNL@so4j;vj-Oqy- zxZ@QM&%myXHFzH%id3!xR@VRv2vs=zrMs#Bj@z46RW7@I?bT4oQ4X=tXY8S9kdekw z$}^`D$`}j&XToZ+qm$b&XS?B_-e2pRql-p}y$6a!&L41=X-hL=x>4%YVr*>jnfhA$ zrnIL#c_ytqw&9pJ4`tUfYrj|H@A${t|Fyai6-tTe=YY_gIn*$O_vPK7$>4(98nkTq zpSDP+*l%1*q0~#;rDcX2XQ85r024?qmhj_!Rmb+bRfgYE+W%l+fua(iSkF~qTRt9K zs{V`YK42K2TbgLTQZ3pUIdMo#B#D?u=yKvlm|-i`Ki&g}S6l!mGxiO6uSTUT>8S*9 zF2Py&Q{o}`BUGPU{-RLy5n#;fvt8iUoj2%tqT7C7zF+XACV69Po1cSz(L++Ev8z~L zQ+MvT49K2z*rd9Xg0b{~TCX5#I zMX0lZQ@*9@0>%Cw%%QXMX!r`=_P(YPFt7GR93r8kI#Fy0UZ9T4gxnA)$oS+EfphYl zV{|*Gm(GU?XyQj}s|oQ$L%4uQ;BD@0zzO=~gyv=2RYEZ_I%fbUx4FCP z)bP7RJX`%_j?w6P*D{1vqH!p4ed6d^Q?5W~*)rmN!oyFYmTOvf2?xxQuLc61(?Wtm zKfqT4JTK|nE^D3=p6Z?vDhvoAjf`pEl7Ldm&)v;06=pnRjlmsYFjI9ODD$&p<0|oE z7g*9qwJLpgo_}I1`sLFgycsWT*~*PY4D{$eS|B<{AO{3?WXBs}IZ*Ge)^z=n+o=dm-Rp` zx^!T@>kl?%q23Bf+G0%UiMb9-oQxw>k!9(8wjX*Tyn5=%z7((7)YV(R3=6XP_4A7m z8O_67`}UZiO*tGFFk<1opWkbkI)4!va01?MM&*^qRmQHOBt`dbOJl^QnMk`}pPmk} zq}Fu~?ZtD;IObIEI)F0YWt&E726&*5yirsU2vb*B+D+72vW&DbFisPraH7|cIU&Tl zXJ@TZJYJ{19FKGSqgJv>YqS-dARv5OSZQTQJ&6+g6?T#}U4124VzYAELP1G4-#iQ1 zQlKfOL-;V)w8=YMLEYMnx)ra(6d&{$|DHjxd9c8EBrvns=g&D)IjEqdPInXot#G2j zEeg{y3z>~RLbsIa1EsX4UY%ar$h3X{1l_VM(dHVBl0QFIW9gi=1PgfnzCEJQ!j332 z1`-7T6Po`-g!1p{;gBh;rs;7Ct}QkQTM1&?)98d*zsyn1+j##*Mc7$}EFi-xtVU_B zl^waK!%h7buFkbF#aIR^Nn`YYQ+G-LATS?jN|Re4#=?Uet1eF9c!BI0_aa>;&aAoP zO*CV}2IiK>V^Y}84w=2!62}LK+C6?FSi>D8s5_SG!3Sc0x49IbHV!qT^T|_e9?pM7 zY8aSr%Sxkbjl?9`W{&Z!NiLh<`aLA(Mxu1N`{4@{yPdo#>h`TxDJ| z=okS@bXkbRojqy=XjIJvfDg`_ucF?-TQAOZs-Z8~S<|n)h*W7w*tt8hcFKVPst65z zZ1S$;aXh+*J_{)RK!b~~*z1)imsfeM1!TqHS4GT$^yFXA(DdB9{3jSZ z&mcGkNcwam0xsHleJT)c#4lZy3VwI+qT_jrk$ZS>ozFXI_MqdSBKr4fY$0&-scYeM zfwH%!qh6mnVj;(Vs5TxW*JbR_oL z6Wxt+oDPfQimvocaZH5ANLWlOddZZZ7)aDWZgkv&-cM}gxsK>Dgj!!Y8!oakY>mUeppp{(-Dn21!!Sv zoW?o`d`rxu|kO{NCmIv4>#PLwAAK{W81fiJEdXKplCS;{pW`6-RpEy9TnC#hT(bxB(!e zrniW*@cmV%7$`Ir>g)+X(zLkmogV>&xL|*bMbJBMF8jR1x2(Z@){yt#B(dGq^ZY-z z#-uLrhV&7eh&*+=!f3#9=X*w7Dfo=%={G2%?LA%$Qm-2@O3(B3LO@396oV&XAR;@r zIrGdarGG=*(ri;-Ns#DWZ5&~PMJwCZ%e{NSI;jU%UZ9sTXg~!QPesc^D$z>=*8T?M z;lwPnv!!rBFE%oM@^{#}Y_!Dqm9J7e^ULz6rh!0v_ql@?2C)V$*fZlJ1l%IS4DU&b znHlmPTB~vQ7Qp$$A<%~%9F^dwAWzSt&GvgF!9Pc<_Oe7A(O zCN^z1uY6J0+jk>jxhHQ;3qKlaT>iRMO2fF$_nLux&U~-m;|sUHYcmi6&s^8F{e30 zc+XQz#57}dLwTCEu#y+Q#rv$U<;~doXa2>BA&1((EQ1T4c&=~AN2oTZQbj5XT72v! z>>@S@0Gy1;ABetgh@5!I=lQwS0HyV#YhQ`+rzTI(MA#LV(Qcsh9?4i~Cy?V({<2HT zGB`+n?Xqm_FnKOFD6&CbMt~bcg}cLqGw3_VX?uHCS6;B?ShFK|8uCA;q22-e1&l44 z{JX;*gaE3_l>P-7zrw?V@?+$aA?V$xk5Q9BS96jnH;jHAkNBria=6+ED-88(4ZPI5 z)E&jmz*X|8!h$a}F>$o|tCZ*lYSoNORFAS}a_A2y*Kt@^7bFb>hd*^;Q(uOc-7>sK z$e>Gme&5Ki^5Bb1OK}+X6+kNAlVPSyN$j`>3DO+oM88VA>FJ8Vc!V#7*spKPhpWE# zsxV%Av=y@p3;d}Tgyp`|p3QI99jWrMm_&VzVp4=lP+*=YA{mq~nWc;gC81cWlR1i8 zt&^_Mfc)su7?6R$pGJYwIX#UzyOE;ONQC3-^OCIpKj`9t#j-Oa;gitjw5Rd6l>|x? zg8@X@ud?pND5s01WYnDw4pw)MC$SY&WLl)B;MQM7S7^SYI^%jIxJepTLejeB5R((U+IoE* zbX11Rs|U8;Y18mMD(2Vf3yNCBPj!Hxh5XIt?@Jje1l>LJ(4>pX|I4 z^}^2X6YyK^dRl(vIci#B4?8WrZh3rdx<}rXC(_=(xbo-sSk4{;Mf~Q`AvQzGgQM^e zo3h97OZxZSD_IQ*)=0>GfMK%f5;R;u4w~_QiweJ~xh#B+KuZVB_b^ZfMiomN!ToXe zKD*T3HQufz@#Zq4O^cqZPq4wE=V~zn3d38gc>|f2->>J`eK$32KH!#8c}L0hV4#aj z^OqPMd0OnR*Rb|QStML_7X3>~sXB?`0!k9BnB>>S={Z;XVvlOHzt17Rrg zcb4%=6g>OB>MRp=TWl{s0`*8NIx7d1`YRbAyD^3Fdo(zLG+;Y(mi9`32TLPMoi)Uq zCVEs1GXU7_pFLAQz;)p%{=>oE3Z)!5rFsMxwXxXOZjks8<~&dL#}C4WA)dB_NKCE9 z!;Lh0f(N~!ct9fG+;TGSXkCmSv_|53U~LrrH!AbV4i%uxF(JZ_Xm)uSm~Lm1Qjy9@ z#sTQ_`9|#BViAIQ*F8fo1Oa?ZgtQdpYN(z%9w-2XfykzR?#&ut*A=555_@pMB`?Gq zcTZop94!WaJ_wK@DsuvFTXGJa!zj8$-}vemHfEeU#$aOfvr_t2KE$h*t&Rk7PkE$N z3#s}W5RZ6}aUKtM(66z65E3POI?_nvwvhY=4V1(G2ZIavUVz-)e~afT5TaJ=YHRy| zOb)X{G;C7;JtIH~Q@2w8i{Jgv=U#`TFr~lrp^xvCh9FWkBR7FJ1a2W|Vq z2(Oqwh@fjLRo4c!l?myEg-Z~Fwrx!VW_ivE)tmc@Spi1t5a zF+aQCX`hB31Pzj*I)n=iC5|GB;5$!SV_=7Q*(k~&z1vQ}^XV+5nmn!d}NnFzgZn9}^>d)oa; zZI41k$&_&9A~O*t!}!(YoCq%N#Zvu)dQQIxg`n8L*Cx2>uZoqip(OPELP5;swbj3Y zq;8AU!9ES{Gd(Vw4LrvV6W@OkoAOI1|5M=`#sOMFeGAf-Qk{Du_~Na2 zQ*%*U&n2zpVj~}@ve#h;rSMNoDQ72gs&MfPa4j~!B7j0;x`l_1Er#SA53geOQmNrS zXX`HnAdce0DTFoV&Y%gO1potFg08xA5;MBmkY5cp+g}v|!u0(dIR2l!w zn-W#zM0=pW#QC7r&FX!BZ@>eK_#CZs9<~>QRQ|tyWayO$XihCqR2^GpdS-}M{Pz(* zc2OReCjct}?e-+9LB8J@R^{XkpzQQQwTOCN8<)wx^Uc64H=lrn zH?B~Lm-ZdSO*88xTW&660Y6Brc(@4(G`>Wi0jemj7-9$ohoijL%&FbT>cF%buS5WUYD0CQ77TiDA% z(A%3Q0fw45STRDAuqUejHr!4GMWBM}>U4%v%JA|OhfPy_(!v34)F(smCI=w%noOP(6~T2> z-dSOJo_^A}|0Dc@pVxQkw@SF)+ADFU+N*%|OGljZA2F&XjmjC^xb-rIgOlQtCfx&- zT&&q#U7C&%3esF!R*W@d`7@l|Mm$pT@0bN3ubYUe(}1hW?-WrJpFa1?<+ORCFvNWx ztg9q^7MpaA@aoCpA>xFzAtu`&#{lEHAs?Dk$p@tFZ;3Cn=Ef$I@mdGDnl%(@Q;juH>m9*DH(4cHi^w01M30^qh@Apwc7qP z!?z(Q*JS!Tv_i_|Dt%pRMNS7cH^8&V*p=GA@qgoK%YtDi85BIIJp1M+RhPx#%;gnA zW^ob*xAB02jTh0yeChL%W^4pLY(|&tJadgMY3U6vj1|v+dX|s zBIe(<=P&&n7Zu`p=1UI)a-4O&pqUog=RlG)4Ep6i{!I@9q-1`Gn3oQm^DE$_v^n07 z1puIW3nEcy*IeJJ=8a1XCmyf|UlOzz8ql7f+tpx74~|ewn93NNz=`6>n*Mg-`E4xX}Kk4i-UoDcOKW z7EXE;t+x+=*h!B*pnvqd4-6jegTM$Zf_Y9*_BCx1u((qVWPUNUfpFU`r7Mnw^=vIs zKw{vs*?xz`&+cCkY=woTon0}7HSzmRPWHonBEEdMcDeFy+Qq(L(MoPUwjirtm_PGG zcrE4`J#ji;jtWf#M1KQ`UVKzUvjNKyYu4JF*CI)GbFHwj5s{>Y_r>2au;I|SLo1M; z0UlskkuzKoh&k9Dya*BZ2Pj8Vf^-BPT3rV7TIO1@Ec?B}{SJ^nEz^vIx<;Xj9swmJ zAM{$EHGi{v?D1p}k*M_A%5LIkcO(?9(TSJ-mk8khtz@D70}G&D(2#d}X2!)tS+~pQ z!*|r>|4!Rb7vMrN@drNWPO=%=KYFMvRu}vm*wcyu`Op)gn3M)*sa(yT`|m*2k}G_% zSNwCRM5|?^i@HCmrD@G%*yWo_GdgT$2c>AeA?JL*$Juv@R+9nssPf5~lPX@=FNqSe zV?QGHP==M5#T-ZCsTS#gw+vV!7gO7?>?nrj{FxZ6y1!36teMUzI!152M-mTdnk$9K zM=J_DkvAMsSN=X}%XM7Q_*VtB=+Rfh?%<%*Rw7?I zWYjhZ`@eZpM)Ok5(n`hT6DwL<5*_@SXPg%2sDJ)`Z)F1~u|ph+tXyo-CjTv2S)#bV zKFlXpDxQ8+%L(}u+zI`nc+{ww7ZGm7~(bc_=Dq=8dy4C|?nl9`DA-OrE2%)D}f1NI(I*wxb z1FM!xYK2ZUP&XJRxvwrEykFb=z!2v#HVxWoC|%te1Vn6zHt5XoH4hX6;piu#w&$C+JB>s!kvs-3ieU@EAAB<_wJSg_?T zJ8gcTeddN|5-t4C$~@%GVkRxohr7=)D*wWTHy2dopX$Akv)pEh6Yl?<7(~p)<`PgB zlfE=giv64ynuQ(3pZq3t2~9Tj)Rc5aGR`!^s+}FHP$N>!v6xmQ9dUl1n}C^V{eb`n zE;UpH$jd3kG^yz)pX)Q8?+g_=3~2EDgggRg-4#`e0u{fXOdjbFp2DU)Ygsx%`ghGd zI$3ozP)b*tyBnqb7~$x@AA1nqKr*<$d<^WW3b=Ay(L6My|5Z5-EE(DW&TZ(nZCU&IC9iIEpqQUSMI>toJ& z^_?NPcY`EaW{t5sWPMueC8bzWWDfePP0xGr#RH!lCXHw~tP@v=s>@`oMM9_{lkKFV zl5q!i&Gdxfd1Qi|0KuG}RRoEU0r?Z%$IKN~Siev%6}jOI!LKWk;H?bJbSiHL#2c=5 zLj|S0Y*unz8tjq97fNrS8uDEl%A;E9STaYKtP~bhq&XmqfIugR)AJU$WojSv#gMsN zO5dOd6g!IE{C}^LNV2y^iiNdzmwAX790mJ141%L5Vu85V%?Qv@38B}cAI_pT?XP@66h(^%X zbim-tOFejnM`s1kL>$g%Heau~fvP^*rBwqC}zId=-9*v|}Wvmm0M#3^Ahu(kdv zzeEPLg}Nu64Xw6n(J@nbC3*qz<<~M=G5;}Mp-2B*Lh#?P9yKn07}!4cKeg#yviRv6 z57a;af-vi+!SXW987E3IReu!opLY&N#{vtQYl*fupsv_;061gjRc(XN21o5$4nkF+ zot!E5m774)wKxbPo!zQB?XApi>tg|iw<|Xp_#Qnh&sN}myYnSRwxI=P`X0O54KQ?^ zoiG_Wi7IatvgyGOd?FTL&BPZeBf!&AC#F?f-Cgccb*^&w+~2QIopf{ar#{uzhc~Ft zq97(#$xOF_%H*4uCUcQstX)OMXRLHmpP?8`OI5p zxtclIHP$1+lt!`Tg-VGjKQ*02`#UVWOug65iY->GMUDhhV9^s|xz+%O=Xa*f_wyDm zUUOH2Y%Q(NQFQP$Ti0#{6c6n*&SelIY5!p$oX`)idM*7GLVLgyh+HopbK}2+ z6`>JvbJ&Xd!G4csd}D)z#HJ7|lyp?5e)Oset3SKS{pwAU#AR{mH^4s`2I$2i6ygIV zT^Ep8d_z|Kr`5Oa!HeqD(_a5zQqnwF1f*R97rXZ|KEnuyq4>?s1>+TIPO6IKtd#P+ zVe0=rRmRis4I9~3Itl0Cc=v>GTye3a;Kk429a{v;#~NZK>eVabt1%@sN4#DkDPHdk zyb>bFDwIX2J2m74_SX15j-lVPmc;MMg7Y)KM)6Wk?&-L1EY^Mq&eEbD8amt3Tj)Vg zAJ)(GJr`t$M6w-FnL#XPS?#4@`Sw(*;9P6$z%HsduH&cj9I%I!=&?dYJ*LGMet5Ip zh=4tQd+JWgD^#|_4^(!iFaM-$$!<=Ka4%=L9qb@N^9pH80SmCYNkmrSQCdaCcAZb9-7u&@un4tXn zAKga&c8M77JCMa*-fK){b8XC@gC56`O;&faOSU>HjrZnk^(GtAz%g}O7a&h+$BBV} zn_~Gq9$T-xO&t>b66xn+7Sn<3c~4kpJEAs1iqVF0tg8H8q!^N`<~c)ADj;n*wjV?8 zkRU(i%|A=jeXjdlWtg0aOc~PQQ@~}pnZZdeW7LhVi+e{szBPaI@JuUm%p^3~<#|vv zVjBj&B!XdDk3Y{j&@U#P1mNw({LdU zD@cFZ);IfdWK8n=*F{-@7L(mSXuHH~dq-B(k&bti^9G5tp7BKh;ihLs?R*QQBFOTe z<_5CPY=PnY6Od8aCN#pRg}>)5B!BCri-Ne0B&=w)9dnE`%p`S4o3k`jQ(n~gF;!Ft z%*WGtGaqY7KB|BF{5mS}I*KFi_`Vze2pZN}Ai1*A9BwDV7P+1Qpu-2?Ar&Q&z6)>{ z(UPICm%`*;nNJsvEh(S&Hy0B@!)bu7YY zLIO4nUs)U<6WTnFW#0=rno!zJPyQMXrxiPI?oT%!!Cw@-oe4?eW?MZLDj--7>6lVV zAfO|vBn>OSBlJWip-d7M0`T`*q~#r!UsaN8 ziaGLT=!Zfpi%+CRbuN4#5r~?<`h? z@!ftPBR5^jS?aTmi@*dT7|`BxEQOZH4wsk!t02H1@z4gX-E`-BitDGuLhvIGB=3zG z7@sDqqU*GH{KKiv$bufF04HA!WmSZv^`Az z3Vv@o90~clvIwcDG(k&0%5HN1wQ81^C1*E5+={Cwkv^FV3&h{-!ec2$2*)co96HaP zJvzGb^bU3D_sS3TS0Kz?-%^AW>##N1=mkg?{%Cj(#5*mrI?@!0`~y*i*(%}|x<#~l z_I~bzq|JKIY(0HXT!>_MxbgOkP3Q}XUh*ea zo%yxE`9NQh=YlPLz<}ve1f=1)l=EckkHz`+3>*2t{DjhfiynI55O#Kr)$kMCM<%4$ z(4_n(B_3Ahx2e(%xB&b2srRgqDVN|kGn{yor=Lh)gZraiP=Iw;?tV@Ts2$H1U3lm9 zFJy<;9K)T5=j&IJaX2{o zY!0^3U%<+B%^M6DUEwGD zyiETTq81ke3z7)NBykCWv4iD+3QgQLGE0F6kiYyl)f5f`M#WfHF;hTyC!|tnF~85- zvHlpFI7U*APj#w^wSBuy{dg=SomOC&y7oG+OG1_^TeUl8BTNe2k`?Uy_2~>Lh|hl) z_p`VDRkHK}BVXjiaIu}krNY%8sDc$-V+9Y3gn00EEbjGrNxUNMk;_lt-PY;4Y&>Eu z2$Yw6$#b~9aI15!W%rk--8VRvh0^-sQL*H=F;`gmjH@{16AEmP9_M!%X;BL7U01J)`b)bqm z_n)Nv|06{x8cnNmweU39=fj##U`T=Qzlun7Bn0giaHjRvxNF~uMMVRmhQbi4mx=?Q zjtyD8F*L~6qLTiJ!KKY@$j0X)ODw}(4~`(j{1Ji{5vbqtkoGO~#<|`C64p+s46@v} zPh*pJzDaPqe-9GIoH4Avc%MZT6A&wC&h{}M1QKY+14NeI*#4hiyp8)4%mKNvD8DZB zA&9u^em8D{Mt%CAulH)ef%^8ovi~1!<1sy}tHwuSOOI)u%v8tw06|5k|L{n_1~Y^I zK_3BZ&OZ=SMlK)|nsq*j%jQAfqd)A@jXxW|{|jqMOe5;#Xv&*EP8ND?7gb;T7^Uz= zZp7Rmm1}}ExAPF0UowmK31^{{A2e4Q4D37W8naAHG$_> z0cpNtZdK+Z@yni{JK%)uS^_ABe`887ZMOl%;eYE7=HI2F(09~+uX0}AQY%x+`#t!B z0u`0+xS9`+@ZkH|!V@jLmmZ%X*L6PSBR;>H6a>6&80WjQ)YjatUAi_IMk<1<6clr^ zpm;ccss~fh>tz_Q3`VWJO2A+~ZUP1DRVu`kvgiO#Zj{ zFCA3l+j!j!ha1?OvZ=DKo;t9i8}i?_h@Qti0jmAqsVa!PLc>A5R^oMcKnh2QL9Opi zw$)ig#8IhI%c%NaUY)c+Ez<@r=CS;QCc(EVX3i87WHmktO&hE_`eEW=-HlSBlpWo7 zMbqjGy^`8GhUue9RN86O*_QIW`GTmoJORC39e99nipNU^QGoF8y!)?zP$4~!g1}u5 zvRUWCa}OaS1D99pf@;J23bv<2LR=kgTNg{cC8lyc3`p!WSC&+GlV3 zhZLH7qc{u*oEmyIR2N@%_}Ex`Ma;?#GLETQda?*9t5u8n48NMkFKCSRJq=a~Pf=gU ze{N5-(%yEOZ57BD&jr4Pc7K}M(IE(BJLd`4BsM*6=gxqibdAOV}fYzb!ZfDMx zlirFwwlV-Uy*?$hsl&;@`tNne^NBp4*grCX`4Ru<1NwMQt!|~qULGVjlE3!!C6%Zv zOU>yl{m8kXeVi)9lA1F$M~$ibA(3R)CZ=At6^f(28&E+D9UKEZUtng}9`5)jZt0gP z16tgJA-(3p5VM~GDZ4TL_i#3w-gi!vZ_74%3o?JkH%c;L1MOklwC<3QBv%$jpJJAe z6)#wCah9(2mT( zdTFHU!Rv(k+LFnTDWV#D1y)&8#J2W|`QTzQ=(M7e#G4ES(M1+=b((_4yr$F`kGv;L-jsO$f_kgf=m#%1yHnI!Qni9|O>X zJ3sKZ1Ic1jwkTwQX7)GZi`&yqdXV`IDL%7|z(UgO$bNv}Jg})5Qxh9UHo`mMt6=4l z1tzv%X}p7v-?A~DdWPA;?f)<>dEqeKD7YBci$QPKPMPkvly5Ki(&uRM9L+l(qUWbJ zW2|hy=FZ1{`06tOxwp~3OtAvHXTubUZ(=IH;8QT8-NwdQ%OlICKYuJD= zd!i1_iW}$4{c41e{STO7CSv$GU5&q&68omYD6Vh$lD6N^QP>$x0Gl z;peXLk%Caku=a@2;g{z~Npw)be@dXoc2*01>090=N3xgZ4{#8~Y?@KGS$RhnSoo-a zp96U}NI5K>-d2=P2Eq7n8MlZ>K0HegDG!cXAOOR!R3P z46?AQItvo+P|wnDb#%Cif*B@n4b!KOGOM-6-&!N3&&S}iO&QQIV!`#U!M2=@sW&s9 zDY+8oOM5Iu>_0AC=aGP^ucT%M>`pJy#)SjbS2V~vG2%1oBR^-ar_l$rGeLSS&_ceC zQ!|~&I=`MEJyyy{&-uEFkLJC3aSM7vb%ecbhdaNtAkIr8Ps|~+-+zrfldpse`*7Pl zNG5ayP=fI}rZq6R_K1n;J`pkl`t8pP6!mW_muvj_8=I;woTW5r<^A6tx%U+3=C)l8 z%00E|cYi>iQ-8+fG(SPxB{tP?iwVZ`xd}x1n)08clYo}b)oBD^*1Rw3{ziv=a5Ff@ zBVw|iMPhkvDVm*)qQ@O?D=WAAcB@yb5_q!_TzBaM+XuF2A=LVudgRql;&(0FgnCra zftuEJhllFZO9<%w_=ySwIxrjCl+xg+HzO{Nu)m|CUGBya#>w5nSzhs)arEi3z6d?s z(!5K8Aogafw>S{b4@?2l<|G}yO^(2|p7aB*Zj7!d=5=8s`5eCs%}qIV!Ws6j+ZCwR zKr?)~B}ngjVJR+=mKk2FJ)#-256>BnHcZn)2!-{Atp+plT}@S&NbUCst|yg#_F$~$#84^WM5cE!0bu~T9^+)D>7$IxYx+%V7HaHeN}D|ygYC7 zK{kmeh|JhK>K2eNCQ6pW50}_n+MXXJI8OUT>xTir$_x3GE||fua@3}D5i2Unx*s;{ z4xQo3868?EytnP`;nZFnAM7d=jNy?`p;dPA&8z*sQqD$_y!XYW|19j#=_bL-LuD$B zjkGuRY^GyAtoEKG_m)j0d(BXo;|@v3N(P?W6w`Il+tXlX!Ypxj770%B z{}%cJ=VuLXDPVk-LAobG*eL(Glg{(IH_KPaDD?bgzfZ{i%iAQkgBM%d5M89L2r(?M zwi4BV&Wmb5XD2qGQ@|31PNk$<^7Q>Nmt{)h4G*iQKE~2_guJzf|GwE4iIt7o+}(fL zMOj7ic4VnuNJ#DQyC`j*&7+Gt1EhP9x8?F*sZRMFRu`At{SH;Wb(@;D%f2|M9`$Yd z4C~pAp(}%T9ANHK?iaJ(oM9tvwRKKZHu;JX|DR7>c@K?KJ>eU8unTpII}d&3rO<`06maboWnbvoSQ-(JS}`1CXcCUW?S|BGSRefkd%B zz8&4;iod8HM0jZpt*^O&FO;Gqk!kDc#n zss%gGyFZ#X8qBl(L6G9bFI2@(Y2rUwE6KnOhbprFRP!CCaSs(}ywLq{l6c*={%=Y; zW22%YLVQrSzzrc-=b6;d8e2~-*$ou`Qw?SWrR&S{t69e>rv9xcUbI#utuvAcvg$-SJR#ByG}qbd&V;e|L?vD~x2 zK{5o_r0W-5kT#!A`C`jAbwm#e=tbJNhZ+#aczIhH>DF5+SexI^ra&D&NWUw?*=5FZ zAnu%n6#9-VuY|Sr{$&DqKxLvy_2%PxC zcdQINCoGd9>|+j0{XSw|<`%5jRcXXaTlqB^&1fanyCzWuq<pFr(l#cj^;Y^0%-%oH$yi*Os_W)%APnsi`dFV+WriWyHq%8J5)4bd;^zwDeu8y>? z0GW*4&gYbU6AK$tZNH{nZkjzNf1|myR6~Nl*6OnFuj>8~Y44*`SNDlAyStcoAXIQT z@wIvs$p^bA#b;V66NTVKxo$o~#rq%GXz=P?hC!$Q-f+t3_~+GjvNbj>MT6~EuI35* z+em8YzT`VghkeATWbeC>$N<;5d^}H^0Z;Q4r`t>nsl|;=~4%1tDCAq$xg-KsPVoEcp4(*-{5-6z9m zU?bP($LUd;9jSB(KNv6x%A2dAz$o9Im;bthvjgT<-*g0|pUuR?VCrnfzt_~ODb%l! zcnQz;gWRu-iDF?UnGm=MLJ8jb9n&Uyc2rDFTr%Yo_pMx4r61MTHkUxshT=o5=Ps!M zk*=(vg3DekgvxqwI?|?O*UNS6{qo)ZN~+JRr$>H*yb`cllZwD?Szg1NDBS)aF*~-J z3V)t@N?x86iDaAc6S88T4#-%-*F5dzW|0p0f{PI28FgBT)s%_vtf!Xy)=B#R+761^ zBn2@XH;!Bh93!?oLlx>po3Z0zp|;bv{&H0$WAavo;G=?M3JDkiTV5S}$TQ){0ffqo+J zJqP&KX?8RFpmTFTqs{GqKgx#@^`$;HZ-TlWp( zZ*MfrH5pzGhaL_%wrjAjtbYjh5iA=Y$*_gs*9l5AlP{qM`WpoqN_UM1Bihypy&3o>u>Nz?U*o#0FgA z?Ux4T&29F{;m*@V`HnyGi;AKPp;gg+>GJF`-r8JV^PLwf9TuMWG}Jyf)NIYU^q*(y zd&v@C{6ZQ)E>1Ut7$EAx>F1=3Q zfF`Yo4=2gH`dN(HU6w7~d+PW|iQBere`KQ0zKvLM&7UH<7`ZU3pu9<$X0mgm+h>`K zy~R)r%HgXedc#G7Fr!*6&FT)9XdUG z+V->QBW6Dogqvpk!}pgv*(@kP#NMk!!_flVSQqCUKzV30Vk5M-H*WCT^mgc?+LG@y z_EWX4A>psOv^?E2${O4>BPi_#TNB>8<>s!*}gX zTarM+E%T3~3L(KKmFlYvvPdkKNx=I3eIGqEV=n)O&xdC%bO;U@I{)GizgYFCm-ip5 zsWFNO3tOrq#clNU^_|?@J`|eOB^8>~Mep+*32t`Awiet9SPtQo^l^lT^G6N$YT10&>#u47ep6Ao__8WiAss%!VRX^#<2RpOA6;X?I`a02{GA64@7{)wU*4eC@|z?CP?Uo#&#sMI)s)Oz}=;+SHz> z;^E%N@Uma7^ObolXXxyHWyZZFVQPA*HV-7XNvp?DJxXOd-uaj7s9fM2L+P-n@{I1| z@k^wn*l!p%(m~?4%OtP%{N)rntQ5;`4IQ+RWtfFDY?-Sal-Ddae9t|v;MG| z>F{42JqNpveGX3BNn%FDbtWCKfiiMY97!)o-28%x7-1#Sa6pCuZF6JQbD@9a*kzIF zMd^M*<29|?l!aMHkxdA^qLG`;n=0shFW66Z6Q_=z!t0tH{hpJ=Iqy_Ew8X!QiA~Mp zt$(`y@^&!e7+bl9AB+vYGcr;Zh*)rcKEEVEUwy|%ZsOd-^v%11NdoU{p$*1g500ds z;G$Iry_;-}`V9`7drB!NXnHQ3y7|$bwz>5#ae>re>eQiAf3@;CFWD?~FCZG3*Tp(| zAPChR0T<1>3fAdMB(mjt&xz!h#Rhci`XJcC+xfwyBf}%tizL15hix{YXh5-%FnnzYA zHSOzMm;9+ED&t|jwGYr6>U6z8ov4KSHz!%05yW}2J zXn2ZJPRF&%mSVH&2Tj7))V` z8VQd?5z`691*pa%xaZZ)Pi0pc^9Rt>A`SS+CQIqi%TjZPOACgO?dZ~Zkv2{9K~(c} zL!i<9Sp}a7jJMx$ABST6#{9ayem424e*SorCzH{x^{3B@t?D^cgYNwLaizSDPxwL^ z9rNviD zYCQ9Hraps$eLXtI_}xy0hI+gr!1i;k*5jYfyc$F`DpaD9>gJcH8-Ez<)Ng7P_x%ov zcvVBsJoB}x1F_pYVM!K2`Hvis>!#;=%Pw~NP=uXRj=sib9t?HMHfH)U9gnuxC2 z^#bas!_I?q!Ct3*2A<-WNIkn}nagd(hj#%h;U({SE zQS5XX)Hw7TuS02nHXM978|wD(z=e=X4Nf3 zHEyL)>m#AzMV}9IZgWxg`W)RUi6Ve(5@#K0k7+qUIAyD*mI}Ib`}qb2C676~c)3id z{5;V}VNiE{0y^>ux3I$BXT%71ZS7axlIM$dML9>claAXQvxJLeA6dTf>xs7nOSs^| zA`gR7;^V>2XEPC=GbVQYviO^XT#XdfcXZG+daRsCjlY_$q^L~GOzmA9zwuGf0R2>W zqSN$ia>4iS3EsZ@TY!gfR9Z(_nSP?=1-*#qP3uq)*yp)pR1RSNBOm2Au`ekQUgTap zHPXf@u`)a|f=LRo@}$Ciw&sO?vnqa~?xp&`TNJB_`%%kR6+ObxjG^*WyCnCH9LQ2k+J z6`}(l`4u~%V|5RH+dbM^&Ocgq9V={L^K&8HAijCSWvGWWfK{KkQR z5qg{sd$UVrjCd$n?IM)_ZNAz2{tC$tBre)=2&xKB+@JaK){hp7izpnK=n%`1yi0zMqu#njWX~n>I zNU}iQ?+v23OaPOSiwP1m!qbjHa@lwEd^fbqU2Ci{W$cyNZEFwMqplY^sh+s|auy^w z`B>g{R>~)DCxO)eXi>yu?gEY}#&pFB>xu-_ufdIFZG1*450nu+@ZLyv$VLQGU}V6< zv7=$EDQ6Hp$s51$`#ek}{}Rb-964cCHaQ;bYLiHZ(meDC)JQT3tlx46#u!nVC)#cA zCV6WD0t1cJ)tQw3P63J_9=l?-a)N3U1$qj3NCpOH?CKMUW>I05$}Ta_hZL88N8^^H z;FGExmH~G`Q#{u{<(N%$vc57UC3pV0ziVMXA>Bfl8t&e7jzfWINo}wgVtEBiY^l0^ zhioDoFsckSR0*QQOcjp`Jny%zIY0$|gw;BkHQbE_Id77Pd2{`vHWIGg-W1}W+R8&I z?JZC@3LxJZvWD*Bk-#>Xe=DMm$=Ur5e^jV+N?u)Nrr{8?BD0}sx>&9k#jtcVZKkS}Az>AX=9cn5TNFy2Rs!AtKDjmPBO>9USyEu3G?i$XDbg07^ zrEW!PEA!)?w~gH|>n&o%UMJ9_P|WeEZkOyBk3*Oq$G&^=*sYo~YVJPIfH_K@QQF)u z5=igCAC1@;*Ei77;Z|I(sY8wiffWNTSv3<5&!o6D$GkM-tX6Z)XsvlU-HLb z4Fk}{|0-v2Rjw9nr}r&As!8azyHR!F+4LK%#;v|6%cIbx`}JTdITgYbfc$*Tpt?9- z46zsF)WIX<@%|zv|D)8vfUY|W)?_dZ%tE#ZB&~V0Q79nlRlG>uv9nXF!!_kdF&b=F zKckZ=f76yDF{D3n{!S-?><407Yn0pPFZVPj{$so$9XRj71K5ey2nA2e2tJ zpwm5L4Fm}O-D`KGsTMn69TFzXf5(`t1hTia5ccs@X96PeNJZ=RoZBflz-Q#s|L}Qo z(~?lJ$;4mH-tM{D$_F#G7eeW>1$@1mzi;|c7oT2nKkSa~@BjYC`4ITt=wI(prv6R4 z3T&xNR;m{NUZA)CsnIhsdF-IWAadg8^tAnD3(oZ2>2HF@%1R%nUaKqm{nRrGupaJp zqtp;=#O*1;5<2EpU6lR?`2^NCnA@KRcTwnIF>@1UlPB94D5NuIv{YpB9*6>FM=eS| zFDaZ(?i{ZSkgAgHKR7>;g{D(`ssJ8WIvRuTZtuSf6KPF6B*j~ga3rsxZhRZK6(uq9 zdgH6Qd+x=%<}S9V`UdXeRvzgW$;iIq0);3nXPYVdBpxRwf(YIS9xc#{8QuQJ%5~ZgR?06ow%}vUF8e!rWiIg%Z_L2ET+1vA} zT!;~K+FtDQD=@C$qFGhC;{Mnlj)bZoqFfX1mWyPaFYnh^8>zt?4p^8{ zENE#2GFLABiZ*1qaf73u)Nyc-gUdl5p6meCadpkqx@}Acx{8@cs4W8Sq3H7zmsj|= zzkaa}&_b$S+%hBell+=b>}?es4U}r7k%^1zEO+cS>GmA;@V{4f{bH~(9nr(A`|YNl zzeqG0Xn)cmL|UTgrOk$7z`KMAln&6_nsbh0NgBlv{t*RUGEClH+wxldSE9qNg0;`g z>2A~1;wE-Gu)_^{Xu~P==Yd)LL|rJGtAs7TyiZBGyiaL*oljZ1{Ol{rc@2=U<1^h(wlq+F{C+L(hMX&;I)Ljg0IzwErLE0Zdx;K1r?~I&E1$w8kkR08abp z$#k|Y`XOxOOAUjnQnY~_Q@NPHRHJ~XX;kW?}7^b|UP8Kz8L!0GUX5x=}%8;6nD zs$4b>3N|4hSLpmnijVB?;#@a)tr4jLyeUJXp{hQJ;3MST4dze`XCA#;Fn$8YPn@_@ zq}ma24i=&L@i8kidJ~yl<+nS-y`=yD!-g+&9%>_uMRR#ltqQ^R5_d z>BfR-@(myDaHn)NP|2FsbbM#%jvBr|2ZFfZdlQTsd5jxgW#s@~!iOoQQFyo^cg?R} z6$cf`$2O2>1?s67G5hA#wcmxhbs05k_W^;9OMrNeY1IX^$yt*4TW~Lxp2zN%*18h#=VWGT3K7OVrCXfa(HQ>b{^c!*t9l0VytYX zZ0%|h^`$^kj!TDmd2)Si&Aayfl8C>)svYHtd1907AaUv>I;Jm8!meJAE1F$=>eO6q zny`q~f_B5)^8WS7kPa=fnWDmsCN?{VMR|GQjpest{KaP#skM9ec`@`> z0R+a|e$N7(;JTVP!>zoZFlcKX6?j9VK;v#zR*SY&v)jk8+b3JuCgh-jbK(V1AOfTC zEj7CO(jknA1{qB=mfx6)_~E`e-B`Ms8}nY_htsUI`Bl^bb?`gv?$u8P6f!+rsa@-e zD*j;bRNpbG!(Df-D3DmVL>vd!G?M1Ds*#pBkuZkg-tKO)?}NugAFq-ouz;bHVtaPK zTxWK_Qm6SFX>Msy>5?6{EiFU)bCIApe`NpDbXZO%>4mitRh_wnDamxz+DZ9*be+A> z9_pdka;Hdm^jk)GVm;|YX@UPoii*F!Mr#yiKZ$E>s_smSMLoS=#lI*pz*iz87Cj(y z65Y*+Hj#l)sMAWO!g^(%p}N=Ol^S>S?1yZx$RnurMFRZz*Vzx%rfRKQt|S zaJ$B9Gu5ENZyp`%T zPgcmC7}uxAJ_XRV0m^Uw?V(5(H;)<7OR-D#o2P>nh9LS=7;rZ6{_}o|N2N>1iV`3k z>a?d0dASsRF7nDk42*Eb5qnVg(X5q6C`~Z^)(W}Y{6PDNa-y+jaB5iOkQk^s^6K%_ z@1i_UzP92ZJK<>BPIub?&0ni@`!3FgE^u~m%I^r`>BIDT`z;e8y)@2y->ckn7slk~ zRT7ro_45RsJ>^^ttBRbOxa`LhTQ*WvL_GRuPr9fv;d23cDs1rXIkebRsIz-y(d?;( z?JU*nxD)f2xE}oj*~$AIs(e2KOWa|J{zIh!P+ZuZV9@JPmJ28b%W^;OuiWwq7hJIB zkrNK99(|$d1{SIs&SuWfl?47F?qpDRFFMmf`TvW*(*Q67OT$cVlA@wwL1AI~ZjC&T zO=ndCJ>(vkq}s0dU-u9G1L|o3)SM;^qDR!bh18@rVU9 zAtB1v=$oY)Ry;r61)I6Txv%n_)AFfhou$l#iVJz2H(sCx`rSym;}gmMf;9Wji@%yS zlLy}f>-hqa1doJHv+z%ii-oCU7AC2KIT@VsXAoVUs5}&82Ks0Kw7l6A-(DPN-V*C5 zxJ_Ofso%Q|cyHQR()z9=xfGAa?$=Wx_zUF?RzSkH0W8AaqDrgS*7wM?jVjq-5M3O2le>Igk;O*NsN5e>kj+oIU_~neUy>9-{JcF=+3{XL1HW5 zZ~cD<0M8bbL;}(Jju^1v-n4lGEO-uiX*=sN;p(?1=T2%gPosmb)WOJ7vNs5Q(){80 z?u9D`u{&qAgI!TV3}}=`q7Zb`K~JJGzW)uVj%{PKG(^n2{?U(&n?H7EMGOj_?K@=y z#2{h&9^CMMJp%T@WlwlmVq zeh1!;hRMAhz&Y^Oo;k;VNW&GjFI=`k9qD{6kLk)azT!e$S48P&r+ZJOknrb7l-;t} zQ%Zr?NaNO;I=)_<>p2|Yn{L`L+*GuQfYZ)+o;2(tcarQ>1H^nOGXps2A|8gYQ zsy%WqyaaRA5HSjG_DO72mQ(w(z3JIr z_TsZ?2W{g{Dg9+Ry`80Or<34rBzU9ppeqp-vj(cY2;>@KuWI-gPI!%kd^RZNvHb)d z?0O9dx}x;{bL165EQZ!at4E#ZQF?PW;)eFk>7PU?72bx!|xDfHwX);wLbk zIz^<00QyZ{`T|TANnDML*Zy8Ktu&erVOq@pE5#4gO`8<_hnwWGRQnQYLN2+1ymJ45 zCsOQUroxsP`SOEclw{*<2gWVwy(u*lxffM;R3&3jAU zVLFLfh>mx}D1G(g;YWN+zSVs{{#xUFsohR**z!%Kd>ZtnPy{jlpQd*6(GJQ%Goygh ztJiN89mdehI9$` z)NCZJKtkpa=jt>T>w>5HLTHNY>vvsmQIQ7US-fA^$8MKf5w6jO!5(Tlv=-9k2mbZc zA^;3^Z>r+?e?=-|XQ8!TkDaeE_g^~wu}<(H=7uO6-9$f+ys@Ln75r5lJ43JSZ$Y&>OmI<@11%|zZgavPJZ>LuLfa2Y%S3HV7H^i`IHceRHi*vV^N5!{Jti(Tc zz83{CK?8s%Big+`3QB7`96xbA9qgMz)SvlFu(Piqr|eQ{L!G(8vmQoU(+6>uOxdu0 zyZoji6<=URk=N$3gYo$dU_3qJE%iB1_(;!!;RL0%6G46M#KV^yIIRTjKb7UVPp#;{ zw-62gMSL!EP0=>))o=3K1t`D~*uSXL8SA@O!W1R=GSYKBABwPg%fPT+ zVH{XNe?#u_3pbNK#)gKeQXlA3p`;$_dHFikl9lL8?$tEmd}?rSbUpSb&kdnDYn z3q^Q=r$e2y@yi6bzzmG=IW*bS+*(dk&KG%8{B&jZ9pnw0ZipuaTNt|2a`Z0#mCh;L z{xKt)$VeRyr3G7s{)AqXGrn8Nk~5-t*6Dkgyqzavb2-JjRT=Zp2OLZ zg@Mx7Q`eJR@0UzKt$R6?t1t;CJT{axCYDlnX%qLIr`PS@mAE~>8JNAZb6Y@qx>$_~gJOtFA`w;kSpN z%8>pyjm~R(%h&S$vU3w4`fg6X@A=Rl9(XGnwLxOvC|K;dINzESHcS|+T}^98i9xV{ zN8Op4h@`dhTv;Z}lxzL8|3WvfY3&iX+>@Vl)vzJaYALdBtCz()0 zs)N3ZV-i3eg@dbY^oa1FvP#=goM8ut8O@6#;!A^YLt=;?;bwl@4E6n0G`*;i7fbJj zV?aP8=d zjvDN`ou7eFVXT(n5&X@`eVui5;S02MYV~M@@M2`5>@mJvUk4SNS+!6 z{Ci3aO2>Gu5W)L8l%kuJRb6lz9{cG2j9D1t;INIBY+~WHSJ&1PymlQD&En`GSN`Wy zJSe<_UN;be^$7>j2Ixb(D;!4>qIBFJODb`@IKolnviXr=- z&O1bu8ehG+*;mkg4$h>eH(dr(s?;`#>{$>Q@L%x+o1AB@5%b!fL@_+x4h^WizNpgA zVE6PL>@I}Ju)f#|-peU+tGvB5m9HXztJ*NQF1Qd685+!Nht@pk_49W;D~}kV>5epH zcH}h0-3Y7m874KDl^BYGSc!gDkBG&oZnnMHoRNG^L5!&Vuk?oj_iEdRj_`1*1q6I# z+UYCp>ney^kut-S1=qd}^cUsGU6G$CZO9{0CKS-;W+Z+yKr zR(+mx=xROsbcD!%AmeLNX9`Dt94Hr}wI|nic5TaGU2GVElSZWYQYgTS< z>?h{ZOtt8lNaXnV^QsFA6`YyG)OSs?Wq%N@^T6CrSI0Lrk}7G=(dFvwY_+|}R+iP1 zRJypRn)xj4R4CyZOVZt@b&lc(uj{!^LbIX_MUq=dS@m4UF#KD zj3w#g9rHr#8C=8&`wMq|$G-^%ZXfi5GgbZ%;@&(Q>c0Q`FOo`{Pv<3-pFwhJkVboGs=rrrPIm| z2VR3#*dSwHz1n>}cqc)vTqs-i@HJ;nQwlJn;-s$V`*movD^O?pqiioSv@caxRhI4# zdUupB6RXJmqwZ<+U8hxlI1c)*7SfVZ3GFGKlNSP=bEL~d@kwrQZ?FST+BF=aD`DX| z2L9|ch1Y^Q_0oagk$d7H&wiS8{Cl3*Q&3rV#j#w=N0Gz3hPK#Y^(afETZ2!#^+0gL zLHwQac$<+0^Zx*#xkyeeI1#+QTW5bN`<-eeQj0D8;3yz~U5O^;mXOev2)9yUM#Hso zM@yh^tAXS7jGX#}xJzsmb8{Pos_Qf#+j8cItLGVi_gmkCtN#a#eHgw4>^Y9*Et?k@ zT77yMaK?_2`XC6E?L1ZJf8uB_0y{fL(d7(EFS$E&FV z;3#hlz3NHm;#M@24B<6eI^O}lS%TyTy|meG13MHNVhu#ir~Lv`8>m{`)Bw5;u#yj- z=OTuKZdESR2d%i*UD&vObZ=r*+dZMYmX5Bba>op%XH1mjZV12XwX6lEqm&JdP!-cA?5(`9z{!e z?37r#_MLG0UB0@x@%f=kMkkFZL$JNi}QAngj56sqbWh> zv=V@|?Z2$_Y3jj&mUJm%LPACP+RLDk0cVVPuU34#b=x0oTRtkiynd^j1|kp+1Zpz$ zb#)^VR}(ZM#b%X0ATvT3T)SxH=lg_G!gUiFGPmC{av51#vW&&g)n{u06#1P_LVY$c zXz$+a7VP;5>LJWfVu5QH(jzyULk1O=rx718r7!sGUwQDhu=jaCULg%>keKPj@82Nn z`zhr$R>}`cfq@i5uc*($9lZD~)`59p{bGe?Ft*JDcE9mp05i-OsWo0*$kxvy-qF)n zc(7|@u|JiiT03MFlg1Mp1zHlpnC4oWG_6N&Dqgs9mm?BUSOQSVUl#K@Ar3c%yd`C~ zeg??EU7Q}(IzAw!K5avZUx{acdq1S_ePUL1;Jx=FC44>J)-Qb7c^lWJAFNJH7?)sd`G-J2qw$p4Dh(s=6SNlTZi-qW5@vi!DbyD)0QY6xwrL9 z8<^_!33c_tG2yMsjB@S-u8q;BEm|q^OqKWgF04n7J4|$VZH$LZJe|nchs#b&F#jRW zffiqj%a3gWeor-6es?$&3N9!qva#t;vA(GtU(UcLQvUYsrLjBj|4{NnE}j2L%d5=< zVd*v@n?7Ap7z}#AZp|*!AMdbpxOu(aXNAW85tB*D4Lo++i95~_w_mztMxE}-2gPFWM>jKg z*YCHb3Vd)DPICO|Atxn zVl_9U%_k5S$>6Louy?ffl$$D8kd$Z53bu|wj|`WI+YFRCC-6Mi?w9XMBkeMCMdZti zK~Ie<(}$gj;hMstO-PafPHN$G#J)pBf10r8Q5@|JS&C$|+?)K&J-4n89{EkKOLvI% zVW@%7C23Gev)ax8#f-=&+74O{niWBXkIHbvjY*g zMEv_0fMtBjOzJ4GC0rtL0D6d?ymFm*nw+z&`3X;F{XaDs?l~i+comx$#r!S35m2TyAO--zypMc9tB@3>T6txu zi76#|SYYU5m7_k^UTywdj@O^LlS(*kY5U?dIVAK-ntIq#Lu{wE?+ar=ke%RM^=omu zsIYPP$h3HyQ(`~Pmq;nWvH#Y(?9y=`Y>~XONxABED&f}In;RaZ!|$s7UFmw?5jv@x zew#BNVf*l^M@z-Vi8^nwZ1-H*$3Cx>6jikk$;@wah-fYGUZpjD{Odb0j23E418ul| zO!{ZGhKl5E77LFI4tDjbyMLBmL#{|y82r)m?bdaQB$;uk6 zVK-xOvdaEk!OOq=)qzn1;xwFk9E82R5N93vy_*fn&oZ8T@H{|{a`~CUrDoBhk30>( zmkji`>cFR=-F7kAjfUK$Zi_8?9nbFBuQo}KY(md_FjN?k;s*@V8w`myC-QR?^T+Mz zyHAZEE>~^Gc?Q7OR@=+SNrvfLB$qFS<9@NAenaYk4F8aCgMZjZEh+C#2}TCs96D zL(7~n+{X1qsKn?&QjNatOZXAT@DB%Y-{-=-uki0>QKn@>I`UN^C zO&tU}8%_O{cZVc{rL%0f?_!n=>e)6ZGHx9yH19JAw;aALQ!?kP$z>(hc*3=K5x-x&tjHvQ>C0tJ{puOq!Tr&G6&=xqzc-#CL>mg;w*P)3l= z+)0f#``6fc{Z>LuJE>w}8>abHZ)T|tSMQaWx1eZY3|hbAaa5)@{{D?u+4<$bi3s@j zVg)G<9-;Alv6HDcf*$8RgyY7}0hxrjD<4XFLS;>O)35RN@6$ro!(!J+zT?O$r+~C* zflr)Q;l#2aJ2Jaz^(Q%VDFp&^6quX!i(QLM6s;Wcp} zSoDKKg#qL#Q14d!uv0`%c60*uzUW(7k$K1WwL*y8S&u4s(_9a>`7g+G8Gf9b|J%u8 zps1uv+KZii*%~JuoLj)JCvKAJb&3mSVxRQbNpmTtk-+p%&tl?R3qT_RAqfEKXP&rm z`7~1IreUT>Cc!>Gpfh~Fy1KF=d(nC%LX`m`Ey611>Uo5lsWGG&_5$^VB@(^L55C&SE+^4t4G{k%IjQ086?lFSy>PIMu ze_%EdkR!(uwX5Q*esHkYMnG*Nicvb}QSL-9NlJH|h6|UUOT!(+o&xt_d>C&bW_p?_ zKwBaYN9QR)a1Z}jTt|c9=!wf*Y#l@7lJ3;N1Z5Gs^rwEa26_)S;}Esj!8C@(mai=d zbbA;NZYgO+W6+4A%CFkN81a2D%w-Su#%Y32Y>xZJnR>%T*N8^e&@^379(X}I+MGI zlgCV{M*JOX!(vlcySDAPj#p&$K#)kWz5p_qj&hd-;9r}luq`+9t11Nk^9=h6M!|?m z^YWK!JNK=og9xy!^PL0KEl0)I1p4cfk8f1@Qb^zaosW2HgdRNFi(q#Brsx}FhM+)i zmqe*aS4?Xx2`S#W={C`-8D~#TONmc^(tX;$PEFC_USCjc3v+_XUOVgUel=^X_N~Xa zUW|}z7eX;e#~@BQVM%6j^LY8gvr&MAw?;-o0!-s4VpRGXVROg47e3)W?G^P7+i!ev zNSS^(uCqt5tv};w4m$G?T#TBIe1}ey=i-+5_0jT!^-##xx*7{1cyfkS0ej<~rp5H~%J zU&P5l2ab0!$`@J{YJA0^&rVeId2LK~)%iasf|P)3R!E@BpD#n6G zojq^zpfeO2^9X` zesI6Zo7qYKufP(hz&?L5KfY0KM((Z8e>@#lwetlomH$su~@JzyBxX8^(yyM2#zwM@asMWk(G;U1O(TMF-AS%nw0B zw1DR1@JB`wEZBF)9xXlM1XpK!ZR_Hq?4;g>M+>&u?6%0A(HJTz%vA z#qOIZ-+kC|pBvs5k+3utk&uHOO_rngw7NRRJy)5Bi};Jw*r-IF9{I8+bTmiN;;7Ih z9Xa*AWFkVMaT%DI>WM3?5xIO9Qs`)R%OHeQriGl1pJhHf#yP|ktf-|aMNlxL^=;k# zU_AjKVj_{gI_>mW?}~&uNxB_%(!?Y3<>G6>__uqT+oa#ApmeHH5A-CG^n?ziY$SVZIT6H%b~>9i zj=FvR;jv9lEcBf%nY_xE!b6t-SSc)^w-Jab+XflXX)#?k^1degRQ(8_nETC~S$R13 za9KYZRN+2Nb`e)h!CZ@PnU{s@dN2kA2rv>4k~8ci*!py^C8j5V?8az&HS{^8QHs73 z+zB!BP3nU_36^7R#%%y=+#FfYHWXUQ>|~8U1?hJ>2X6C~>irn(woEmOno#Mf5w^&? z*zdt~f!!#n=D>#xw!g36M|&MS*Z0%VX1)Yf)$u9HQnBC1HB#$Zksk5dS05dx>8CTg zi4u|Lf#*Lx($m0gJBlW~`Xx z#$}%uVnBeF0fKLQ_zbvuDG;O)dACNm-{(EsWROXQ|HVzzo0i%_fU5KK-ZmC)owX6C zq$#>INX$9$lU$@C>8IvFN^gI8#0f%gAIBg(GpxEyVWozFt|8mstj6R=s+>Yfwe_?v zkKU+zu?J#BIkwtGth_m57k!bm1e9WGBMjvcLaBbRcS_rY3Bk73dypi#*u(BPz%_z~ zq$V-I4UVUXSy@&x*o8IpQboNr7xBw-zAkyX!!c`V3wNH}SY+=9+mvBK>r5_f&3M77 zh*D%97j^KqqUBrqvAEq@Y?~?alvb)zg7SySU;-d(60sNCmm8SC_-A?^WxedBY_F_La@%b<$V_ zD$o82DyM=abI%49lYD!gv=Q>I0p2aua z)7Dkrko!N^qbh$a+?`=>-jjdi+%rd0kftYvbxvZ4qE7~`;Glv}AhGx_z%qAQJ4tQg zmO0{=HJ-}`UH~lY+lk6LTU+&E&MNRGw zXkugkpYsvJjsV0FgfI(5eZDMY)D*?5@K)nmx8QBp!)K#W%v*a;-y<5m6SRJE9uoe^ zS@@kI0Xa}NU(~3hw<_@prCqiTv@`&10@k+T2}kq`UENOPVIjZ*1J0elz<=;n`}Q_) zsSo9mDFWhaxh?1AU{w~9<^ac@_bFxT@o96o?2`7~*}TUu3anGl8EARB=O(7t4ueo5 zGUc63kuC>$MBRt77*8B57$_94cVNmbzIKKZrI)h6ziAbE-^sYfMS0H)%E51-(*=Pz zh!^>Hbb0{+K)faiTZ8olqra^z>}^h?s=W)0gz^4|qk~v-eak*F_Y;Q2rHQEKK`C8+ z=M~emB{aK!BBxX6+9ty!wLNDg>o4ztz?q4+6%S!ImM^gc%too?&J$YQzR6Ik)RP7z z#n6aC2z3^3Y284wg*HgkM#Y=RZT6j3Ox;`9`+TR`YsSPl>wgolz{W)b-4sv;x?6C% z#+Fa(>7qu2&t0?8a@OsZt4cf}rQFwKZCZ|PDgSvlen42$;VpSBY%DvH%-g#_36j!r zB|CH=YF$r&{eCMY`iw5fb=6ev7N|45b|4){HcpNs{NOu!BluRt*f6~_))=dw^{yjZ zV3+UC!pWs`(A(Fe`(uO(i7#yn-(eY} zKQG!-EgL)O>3e@IDT;B62vlJ;owcwx!x;Z$KT<*P)iuyC35X}3Mh9%u?QI^ejn=Qd z{Ao<;AN{t8piEswN}p@<4kkkw@!I6xyeLU^$m7-AaDo6lK|RT!PmlhWhn^<6C6P!J+?on?%5k|0l}P zKA&LtX$eMczQ==x%HJ=|zqKm{1 zH5Y1hpvce-SCzB9&zLII>-KQgL3w_gkD3=$EIPVUr7P^CF@3$$KHzcRAdUDh(MtXj z?*?JTB!V~3q!mGAlIrb6`bs}W7J#0)c@oqlYnd=9+W9J_nN&NsPXOBD3z@J62~djs zCz-Ln8(mdeQdL#3N=m)DNe#fK24d*(jZKa&)1fwSyc=B8q$-xaMbf(H`F+zDu|$;6 zk$Dbo{jMUzEUXz8)>v8?SLx}W>Habr;jRqo3Pnn%tQt7Z#i7=~nA`mqIKkxP26bDx zVK+=&oVWNjk({Mhbc(6FxBbq+v7e)>#t zxJD%BT+o_&WSin0|}v`T+R}`2on-Z_uOZB2Ph*3F!^fHZrF3{rr(6 z!_C2&=v`7LA0oFTeHgLzwo)S39X*ml@&tFA1?`+X@&wEMcdf)cv-64ocDHzZQ7qC$Cu6$F>Hi0FsobZSowJzi5O9+ zEy)HvNOZ{LtxLKyNwgSwdZaCY_{y=7oVs;^Gy2~S0Oq$czsiEaaDeN{{$cc zoyWg-&>y?dT2VS0^-#lGKvKh|6@722yQIf+KE|WrbA7x-sdt3cg)blHIxp8OqRyjo zl82GvbT)CH`(rxZq(`nMZrfx0rrlmWN5nc?D_qVpT3Vf=3bI?EuQhWmztTO@L+!{^ zbqZy*bFA)|)_qxZ**}hi_?;`M!?8w#u!!}Z*j1aY_Bi=zwJNN)Zkok8(~8Zm0LeH7 z>H!L;gZ`UO1|QVD5t9F%waOR@q23RDS70F`-C{48Vv_DSQep37QnY1P%PrbIFr-U5 z`oeu(T_T;6+V9@TiXl^sJsWxoASnc0SQZdW`M-!V0BXc0bV#<7ydqF{j$LFI(CNS0 z`ghI%A^i8tw`(?+k6%3?D=f{PJVQwmiog1f{;Rfh7k$~s1^1Imo-@lOt>E?b=atU# zHZp%TH8asdtI7Rgbs#V({nhRZ#(O4qG@U_XKUe;oEJ?9sh^;LB@Op+Yq5e zQWgOF)n}4dNKQ1E&S{xrariHn2A9YS(5`~H?QlW=oz3g3v$xxk(>J(9_i614pbI;zy>r4~ zyXS`@iS;V`SgPUG^p7+>@#PCcZ?Afh)YIqcF_6;8%Bkx_!qD5neKMA@b;%!C*WR2u z_2TaI$%b)#Pp#~RVqiHDmBiY)%t-@Xpv4q%0{^~i>VVm{Y5=cV2Vr@X- z8}H8rrMiva#3n!CRL^I1TR@GgbneQKrb>&do|l_HIW7`Vs}R4(GTu85>@s>YlV2aS zCuq^iYp1<>@(l*RfyGVg&JnJn)&_6-I57FJr>9w5mV}{sN(8eeNq?aKzdsHPt~+>V z>p%%f{N>wPD;1b(Ui&#t1;Z5zAz#FGTIvTFIM?E@1~lYD2W}+_C$MDuaHjAK9zmxv zN`xdSM1-@IJJDU$qzX3hiajmJSfq_O;8j_pZr&!&VJ*H2Z*S&+ zJhlMd8<-e?dIwBlJTegc$rSpc1+6o+Js-^HhL8*1nXcC40?=gsxJKvM9fqEy`-~7+ zcK2g==;&>~(xN_rr3=9Y(bN4WZApm{C=_x8+e1jMcXTXzZOkzi31LAV-+GL&Hti1x z5I=0*8p|kg^w9g;lUrs+Jv!IvU)(=7eUDx-g8ElJ_z;i9 z0iKn_kR2{7h4SoUz7y(L$PYeb(h6D|P`vKaMzHZ}j=r8m)SmQdtUF-1fWfTWWn#qD%dEwb;zw-^Sn9!BvU8pI=F z2e^5!D0uqzyo&^1FVb6??fE6}5<#7WRCg80UW4V8aia8u|Nj;5KAA;cSbL$VMTEMI zdN&w#v!ve!6uFe^8r@aot!=WSR>!%PxSDs|W8Y0Am_eVvwtGTM*TZT999P z2YP=id8w~yYuY`o4ul1|!(L+f+<13`hTGX2>PB55=^O=}x2c^`a60vSq2j&H`r#ajC(yrkIl+HbHpf{G&H#rQ>h>dyd1r%_U@NP2nE| zoDLa#9;OlKoI*=gnE|Yoc6m&BOfx1`P15d28jRXB=^w~jk_Ofj{}y6z80uFVcT&bx z1I=(9-0r%h)@*Yc`*M(0ltBfmbh^Y*;se9qg_2N7Ebm z28UHCdHK~;f?SvO5t%OD8w_zElxQxNnXf9s7<*z#@_?n6Kg=Y)pRW_OGZsx!+W&rf zZ^z;@-ABxtv%w#H{r0Oxa@1Vk{BA;2d;yXA7=e76dQKE;HLG24cQ0&RJG}zCEkE!$ zDJ#JtQSMT+XUM|aj66U6ZWf5aQ$mlRNdpe-6V26AryAmXGxY%b`TuLXZ(;=a&2m1y z*#6zU3ZLb!hpqD<6KNTgC0c6j>*O=@m)tYKuglZQV@HK|`&1pi9N1k$X?JJ^?!{Ir zVaKuF1+u_{FNS=ejhSCyCmC#8ooAiNWCr>vX#Uy{xEnc?*YDAh=Cq=c44>x0`}7KC z5~DbZuLR{)HA;U+^39h?clcN@@1325@ejZQK3g@s2D`w@q>1)u!e2q&yR8KOwVRH( z-{~uw@F2jt-z7|!m(==;eG5>}>EVvuxT}%2T$0F|zg4~ele_nS^TI@ATZwX~hCn*O z;&$UZdD24%`{;D0KJLFL*a7u9zx3;*L>mtp!dx#sjdqD5--+4BmX&T zv}(xu%Fst_W`_$oS9HEr6LjV_L;dNz&>c0un|!tFDjm;l_Z>Ob6#YKP3GgDTv%OCUi3z!HT$Vd2m=g&-u!t#phOEAM?B$sQ$W zAI#>5oz$^!Q+U4?b*vx0^_?r^cAle{l@}A52Ys=?H^dpVX^tST|GZILTSJlW=n>d! zU7uQCY3~~i-Rx7Ww}pBMw7E7HB`cR&R8~CoaN2xbw+}|$Y&j+CTZuWO1a2$1f6sKh zzCWqxg&l}Pu)OT!kkq68uSsUW47;sdcI#hO;nafSqG63lJ;p%|Ru0{;Lz9O!@=rJMt;7-C0EGbj1QM__ser zLUE>`M!;fUtIu9L&o2O7{F7|M+a&8~AOFnD$nMnAVu<^x5aL6^lqij>L%#{FQib^@MXm&f#UTCP1HsZ)V`{<|@0}V6Ne2~Hh;K#Fsyn`x0vfT03 zxz~qDMV=`xPC((dnzgi-c3_Eip!P)3LoGl_cLt{W?{C#Rm+Cna#m!X`{%BbEHpc!w z=L;YQ0#y874H3e?ToqM~ybN~h6`S#6BM7t53odCSi!$`tEG)92HB2^)h+#Yoho5RXpfzpzL3K2_Fm(+>PEfpB}8du z;=G}P4_CZ2Ti>;P$I`W@y(IqhDwDGsc_k1%*uwhz8VGW)z=(C|$4H!)$rSLud13(V zqnK5a?0)pZVfn+y^dgzthCn(ICoCU*VNk}Z#k-T*x94Tq!4REEEovnD*kt8O;Om+aa6GSy4-)*K!ulm2258@0y=4ls(&ybs>EB^K| z{*BoF*GuzlfSA2h?yrDG)bCdkBmmF_ST^ysS@~ARE-*J>=pBe6w5UXiEDT>EITzZt zhH{hEh($xY&=hSmmZ$`##e(kyJAUW8aP0*FgWaa0W%=C09ijRUGuw5OiBI{hW19vM zNBM9g>3HruOEH9xw4+%lk)u(3#P@UO_c(4E-6pc(lB)Jnz4DL1gt>s1|emCNn{wXnn~9yn+f(Imy>sHD`FSo_|c=$7{0&Oya7wm*h3!4ZvIQgp!MY@vV0^QllZ7 zsvtBvM=`tm)~<9`gE7bjKXyZ@@E}BWaOLaz+^m`0%bA}G2{;OPq}Drh^%y7EOmU5` z`G!0=(X&7*QFN;(@_xZIzrCapaCEXO^h>=dkXcZ$>J&nz?CW^pHb~-UDtJk~{d=^4z8bk*a?ym=}HkQq+?OpoG$^Z8KeFY9fE%>I-_C#&~5j>VnrS=_*MTf~p$6yw~%%v{HM_(o0 z5ei>9(R2V< zuu_aD!y6bwXZUEq*?1)aE=Q$u%ELc;2s~(81^Zwq6^5nvwX!AspIFyG+_T_+Nt{8^ zK&2l**tNWU43#?mL&IV}gQD1Ks5Mxtoqd9DbwpX`qJ1Yrx_%Q@@b~v_RArAO7m>Cp>~OShHNS@vn`JgV5WF7D54 zZUMz3tr#Hz*=z4eLNmRup54`Gprx2|BK-#DG)~{q&xh|lh=DvYDLQC^Z+RSXO-(-QL?-Du04$XI+4~vWQ6V24>D+KEmMXYn`>*_EQd?RwY z6P-+rLyz;H5r!l_?%Iyy`0Mpk)!nM!7Dy4iTA zh8nEHPmL?dX+c5YJe+bg)<28)=O~ntDlvx5k=G!hhuIb?iFW5PQ9chMS^m1Lp0Agy z9Toz<)8Z|1Z};9=D5+&hOz)c!srlC{xVLu6_;j7wF1FoMFxPD{kHQJj z&tiz8q$Y*4nBT^ge?>MB%2#EHm*OGPvN{QCkUGLt6FT6{5ej=?(??CFPT8mi{SE^E z`CKQd%iG$28f^DPkdR;u3YBK}E$7y1Q_rTk>@etdG-jD15)YW`q@27PJ5HMz9ME~U ze#$)Q>D+V65qX;YO+Dn`Tg^!_%d30eU$g$wz35Z_{inS-HFToV<{91LS#VMr3Lkix z%@1g|Z)X14`ladkuN;RAU?bPBXd04rTevY+S)|xR9^&|rRO7fV0(;&bV&3x(VwfbE zj*=P9HG+>nQl$FZ2b8ue%zPHIq1QVr{p=>Z*XEL+BnN_=$+-B{7;-=KdB}8GN6OKe zc$VSi!;*>aVn!F8sf1KZaCaYVU4gEaHf{nd->rx~x;A=ri5s&w3n8mSB9GLCA-q+Q z#p1{Ub@7!~VVT;BF(jW3ym)jH69+dc4Nd*YZ*%&>ytABT+(C1I!s@=r&Q8$*!MM*- zM+%DLjPpElOkejX`l(7JN{6pijW_k2FyHYyNmd~u1jyjz`*+zl=2Ve2S0bv+2rWdc${mmH|AG+4q3A>colHC)6H3 z&=c#`51)DWfYeegVr|G*TH~~1)Nm$>Jt>Ep@*jt?YVWc6yyr6yn|JTNJMmgWtPsO% zbV~JnIMz_`4c}Xf{U8vA4V*F+RC)IG>!uLJx9NVtBi>=4r4h7zM8evDc8BMSDDpP0 zgZGW#xHkO*gQB~-AJzGkYGLk$Q4H7%b-R{dRHl8d&L+z)Y&SJl)630+<+uHJ-jEE7 zGj9*|+N}o>W8wEHQ|LlnVllYr9u&GH&%`YXkheJDz6Od1TLFoPq4Sb#xh7*_^o%O_ ze&*d*De`W8_PVf2y5?u0(j%0{i|d;EM!QYVr%1-YS+~-_Z0J7wsR8Yz0dMT3quOQW zC+1)AQ}-~Dv3Sed)lUDbZ-(5kM312+jgK&c7VjmPv{p-W-ejF`2;J81T%BuMxQJlX zK-g^bep&bG`iTgUB~|G1G^^vDyrdQMTkU)Q;?yHdIvfnUfbny@fzfgulsbG3A&jy9 zqLw|z(IG)JucD^Sx~e{Ify~@a+clq+K400ihFA{U#`>Qm?+qbK;6w+IS9Q0&OlV2< z%*auCI>9722k4Os7pE|iO9!oDWTG>Iz)`_?)K~Il*ZqDzGeq~^QOP%zEh_FLkH=sp zVLhW>(CkBv|GVPt*a`f#{b!NuC8kA)4wV*{tIX{_Y0Do?? zO9pI6h*kZ()Cb*+VALt()bOJc!dwm~pi_DrZsg8MRf$hM7cf3_#dZ?=`|;n?F&PD`?6rv*>AIChO-1gfuJ{yP8Vuj}Y~;bE-|v@bAqvWgl?E44OZnlJ z8?t89_u7i(xR6LD9W_$kSr-#%=?{8*l+&%Mac|wX1d=pfu zzSPq1F$alP^ZAuKzwNy>-jJsu`r8j3xiZkN$&(Hbn|dJ9ycZmWKWX!n32Z=?oQJu| zsLF_pJQXj-&pD8yK1}*{bva$4)BCTYWSKesq43*|@nP}LHF_Dq{H20?vz>-o-%ptY zYwNVAzI8YYLAqt|k6wetxV^tSe9Oi#({Jnjf1dQXB|X_HoAc^$thOnKZ9-)I&A#8) z8Z)1sY4`-F!WVB1ovON%n!fZ%qcqEvyuTuFSv4l*)rbk<0Z4rAR0QYX16nOw(3IuG zPllc1E&JCg-`+y~`#&im%^$KwnI=k_+{-z_QP=tsVJs(ZrRR(zwN^>!+Sbqq&s%wK zluVrCdvy&(?dMXO8a?;uXSk|f-)_3YQOd#^+z~@ko=d)7p3JyKRzk^$n^MyX5K-M< zOG9wfz;t+eBSjaAU@b0y;4z3!je*2Gl>zvq>Z{=`U8O1*wc;W0^RiP4CT>ij9xL8= z81QZo*MzM#4^&+mqB?|g5Z=&XPCgQMzT+6>`uvmEr|rvdc?QV0#KjO$yUbl?MNJoE zOSH{wsKgLVf#hMxS2Rbmm1_anX8X=6iC&0EN`)qE*K;=Ksk9S@yvp~bQKHNUH7Wa> zqoGO3)9*l%DR%9uY9FxU5q1)YgXcMB*N5ew*S~%Zx_CeNUV83g)T*Sr4rMkO9vd}( zcUN1FB|X4_=XI1K!R3tXdBq*+Dk@7l{Bsw2C5SDp+e`cAQceynZ!auq^9ThG2z-+I zjQDGh>ruP^eCuMLyg;nF!hJ|E*xaq%{|rPShZ%`CQzxkf;*;YrLj|f0-?PHk@-=B> zbDo`tXf%Z4hum+^ia;o)YSBKne0=6&XO>g!L}7)*iJadF3-s*EqGI=OZ#QyK*AdY{1Hf%ospLB8S>yC{jwx_&Zm4|<7jZNz%_XM@+# zJmL-N!kN#zk38g@o*Gp+Rxt?k7!yw{*nGbLd`c))2tj3=T_chxB?#1Y3VE_n)~DId z?Pzyg?8zhsV|5lmTCf>i!yFz$xvJB+guSaqK|iSEO7!qvmDKK2f7407t7P>#1&hNJ z&OV5klnup9cI-~!8nuX1Vq%)D+1Z1-4GgXrq%s~Vhb)?1!#B9_s*y~y|Kh(gTVH{X zb-&opcQ(C8FZa}Sm^Ulj?Lv$1b)!O{E@!)E-@AlWJq>u~p$IG{_ttE^J0r<2|D&-f zI>IMUt8yMo!4Uf2$u59tZ;>HNiIDaR`hoUIhdb1XG}Q^km-MeQ?CV0gIZAt}i+0lkHo zk4zX{gOCO!v*mHP0W~%-a=~uz0Mz)r#E_~`{U%R$2ZvhN_13euPd;$N%w&)xbiXn` z8QBH3_B4_M3)p6^~36Gob1KsvK}3`-M`KmY#~)xyG|KYR!j*O z7wBIH6RDk4P6^#G3J!d7(a1A!R}zAv-&= zEGc&bPaK_K47yZsQSC&r8%2^BLt-HjGWezH8ZU?`sZlC`7K~gLXNDLX|aEI zO!ThJ+m%pgIQl}ap~=*%1UJ8hq@6%(jZTlk^jEGh3#D2=V*2P4CsA_q;pTU~yFY^} z%vJf%n%2DT{pd*y^7_PinA<-3;-?Da3vp_%z71sOhZAb z`x2L(M4_W5BsHl=Bv{h9>dzwUQ-v_`xva<9lB-K$fiB_uEz`!p6a@OO!$RciTrX7` zzK@wjezv%Bd@FBOaFY2!H@)#>zp_&>19y2zkgzOjb@qXz2~FHb)>fN;ZWB32Of-AD zPWwImAS0TBnx0d!9HsZc4N;epW4zv`A65&x^MU_*ZS6Q08iEl*Q~9C`uad{ol=5_+ zIJnMCPgwq>i<%LopZvA|P!bpi8YS2IR3a&k(wG$d(WR1`&ak_BBR3VYHX9J&u7x#a zCs4{!z}~?jCEC#sJY3#_W8I@J>v{=sl8F%5gl84Z!hFhYXsye$olKmbvY@Hrd*la)-uM5E`*kyBHH(LM!UUb5Sm)tqPdEBM-s%yB3#r5q{eKEx6iM z3tgQ&sw%HB1C6j#aJWCzH9vY|;&<^z4bu&J{Y34{8&e{R33$(@zn~S;!gV@~kBv|h z7+yv>kH@bop@r8F%ZZ2yxF{M!M7V4~Bb}>Y#ycyGn__J2>Qo9J)wh{6g|am@^EwQD zp~u*XSZf?Gn^8twVC%@amz!mV$hX+vP^b`ozFyr}O|?#K6y{0%H2h0vU$1>Hy*obj ze#&hr&Jc`WOeaSIvDG~zc);d8-$(oqx5^EU*@;WG1|(R#4D(q?-;0+qOvuL{t#yOU zP0^Z?oVg7(P70*RO%z`3nr@4>JA2~G{P}^t_;tO^z3%X}ks#0^H8YglsxIHc`s-HX z#*D;+8;Qzz!Mtf;=F^VfrTOlo${CPG@rr&N%5-U@7|jy6day#R^@CpuRq(h^%NMH| zcLdpSSCH)73WWv;3$$)YhC*S#8ob%%@ke+0{9!1Ak|#s-zfu$KqDIh&`~d$JA{x1tGVMFVAdQ$15LT z%WOCFpzt3Oxzwvcr04lh4a9RFkU*UcqLOVGI5n>xpMUx4h69 z9(5lTwY=;bQW}NLezH_Ivcofr5xv7{5dWP=MxA_49IzN;BYpq&P#;KK+v!)ZfJOrb zQ??Fm{Z(1)%OgnCcK4#;5CXEHyuEaKC%3cv2poXc3L@}Hk|jQmvT?x%mwqwH#l%z@ z=d3{#iewGe9E*j(=G1L!l^_|luIkZP(OSIyW?vQT|6?z-mg4iRThm!qK@fMK?Qry? z@}qf0u{&kp6LsN;X?y>0Itgh?(GZgBdvs4>_zW~&9dsH(2IKlKor`G!m9)?U zqsS02!`FB)MpN>aeZx*WlZf(GCezIjuz3i`!-3GJ$Jf}QVFWyUkm#Xwt>8-2by;Dr zfK&A!Hm{s%+9?Q?c1M-V(Uv`Rg(aOTQAp~~xLxoccJ zVKwFpXuf+-`#yUbEDrw=92l*GyP6wLFOAUTwGA+@Q+orohP-gQ37u}_umeb$RYib^ zPaPAx?c8B^-ZUD4D-qT#u0~O$CwJcNBSwXpi@b5Gh9Dy4t0`hCq?A2mn@J0bC`)7;5wa&cVReEDfhJ#O48mn z6LQwIi(Nn>f!ID(rIdc?Rz&P!Rv>ZM4vSi{ewb?FXXB(5VV&!2$WNzE0>67y2tQKBXQejkI8p5i>hD9 z;;al*l7u7g`1C{BZ3f?K8V9dHU7yP#;hn4wKH)AdjRdHPd53oaEu(lr>K-*HF1o8g zrYda0u;2v8=*~eZe{ufzi8pdWiQwpEZjMFt#ucfZ9ox2qaYoM0;K?IFd}?Nw8d$GZ zX;3tD;e1n3-I@<5`0Cl-bua%V2GNo3K!@${Rc`r{T??y>eeaCA;Vwfz1)CuI-G4IZ znDO@`ZD=}ckDOZ$MzjBLx?(Sy^5l5RgpIC>fZ58!;+#6k#!3zNL`_%Ia?e5CJ2q}%o9-+?z;55d-K zUUK5>S*y7sA8k)B;XV(0%3s7Qgr0~6KO?$*bE`m`q;FV7ELkg;{?6JzP_z**H4+Cq zz_ErHXgDfGdTac#bgfXP^-Hr2gthe7j_*eb9h+NL|QK^(oM)-=-P(%eit6LRORG#hvg&r zjNCnFcntB!NuyiZPdgY;#tcm5iG9X<6ufCr3K`)+1|_g9l%qacZXy4eBoert|z zMy+-jOvtb;{8+lC7Jhm4Dhg9Bf{vn5kRGedmQ8B5WPI2Qg(>EYwv1!JII+@8eA4kR zza>^lvm*ObuFDvoE>M%oF7WaUA_2pMc6!biGqPK5War7Z;Jazj{J!s2?05w39tzXzIl+K5W3L-^vX3EPJ)H*L9+F?u>k!ru(uKwIV!NaZPr+`l`oy-j?x% z2P%SN==pHG^NQ${6e|un1ML;0`O{Dfi=mEnCFe{UJSWle_HVmO2!+EBb9di&VoM8; zMfc+S3bfVU-Jf>xp#BRR_vW}PkAe9A)ex(v)qkq=i`g}`DpW3-A%T?j1ONc zUg26dpWq&sRTOt!Y>nVEIdJ%Bt?sj~Q9z~0!3sP~1qOLZK|=r8SR#$RsEYskt^M$c zAD+-~8jZaCEUv^FllK%gs!{a{68n!m2j39l?oMVXu3ysiwQ0wBZB4Jw`1Xn@%YJe2 zwz|I^7Bk=(0hY4x+Ul%NbC#OBC-s%_q;zH%?|R1TRoTI}MX~P3U=cy<9QfDoS|D3Hm8(CxX9=A!2<3xwtM`V*|8DY`aD@a zfoujz**%y%EVJ=)V+9?#x5VVshpl{r>7uwRNqErayiR<+^w3CMA7$^gl^=I783Vi+ zL@O1@-pM>uqL8!4Moab(W_)xlCRU|8Bza5~k)+IPt2 z1wW9N^_;qr#POARph1=NxN~h`K`LZxJCR2?$wROC+!OY+i|A7#4S$dGPL!&kY4mPW zspPG}`QOC3*+;XRSy^1n>jl9#zHlnb!t-HJnV4mhm$(P#pE#Bbvwo^ykE$Z&T;9&J z>K_j+`C-VMTPOJ7Huye!cED4AlA<2$G;e{t-cqykivQ-zrMOJ^R5XX~dGdil>TP(Z z)FRO}&cHB!#0+H4W6ZE@Xq#rm<}kd1wN4MI4hqXY+#W~xWIyA8I2J9USAUR9>b4ov z$3;gNt&NN||HV&*Q%se^uCH{H=)VgYDVrMFnsA8%X-4zhz|0 zEVN0KESyb1)oc;d^4waNw_6^JvCg!;WGM3~qWiZC89vIoIEGoiAhk(JIcSY``md@w4l02!vKUPMk}nV3czE| zBCZhpi-~beu)_6*$5LjxtO&C+kglg}V1YR3zrLf-Z^1n#<%dHLy&hGKi||*Oif!hL z-n8jNTkA{JT$%PGeaJsZD->i*_C6bCb(R!<0Huxwl7>^5TI}}{jUP^a?G*m*O`Pg) zW2%DT&39hQ-1)t^{j`_f+pC@ftHHovOBchME3mN8d z_?)eQGu5%dJkQ@ahTt=7O?b$-1do;aZRE}`C>NvC0XM1ys6AQSrDlVEvJ?s%-Z=H+ zGNZGbuvW-%=3QT!aRKJ)<|ZrCean0S8NYjAQRFL?n$U(iwG&`Y)^0w-5v*p;;~(U)kr81aD7G#4Hd-%j zZ%u$2dU3k?6&qABucD>%1dj(B;Lg5Y-ao4(B62qeG`yl`Rp=ZNKKOYJo4vHA%YiUr z;%Cut@j{Gf*nj4h4=Cv*C6B<>1M{DTc_WQTkJ(tDhlrMRe&wOq|zr=~w zY10#Y-9lZtVOR7LYUEvn@NoBY@r=-RRT7d74p-t zaPicE1^@B!wd$t_^rdDE_s{X_GttWX#8=Z`MX&;X?X9^5c!HL2c~XpH!7cs$?gb-C zTrN3Bpj!T!`gL;&GZHJ(Gj>D%!e?|9{i4t8)A?Aw{dS4NW5jZXSDBW#%h%TUw?_>J z6B)hR3}F8Jx^5tm`y#O5bk=w=4k0~wELTRfN}{Y(Xmw`52ix2(zdlzKwssP&S-3~J zyt8Rl=D;B@woqbtt*i?tFHyU@U~=X=&xIY|xDp@*OnGCl-{kC$C6vLB#A35Y7osD9 zfuJo*IvnZO({M`4h+_1hrzIqhSpZIGym+p4%C4ru0GM0@&zIB9e{N_sa(8r0vN@~p z%Q-d-q{p6Hx!3ERK6jH_w19IzwdP!$q4jT@g?_`fww}O;BwN3bw|!`<*UJP7nstnX zvuCU5H7Z^`Kh}Tndhvi1u>uNh)u#7*u^>UJe;oZ5X`Yq(-605tX5t-eg^oRuLS@zh zHKa7O%lGDj7pEx-dwZKhx>$ddkk`fAbCj9ArDBuYdN5la`18s^78$SiYKv*Dc+)~g zmKExpa{BycGEe}$)lc`aV+cZqzdbVN`;FP`9mYUsFr*MXD}+uZGrRd*m_e2sS@nY` zc9Q|29)toe9VZ-WmlMH9GHnWxOgJ-rQT@|XPrvSTVK!*WyWY|0rmy_e_>+(r;`BpO zjpDJBrXc<69^M)Iz^Pt;ldhKy+zy;AbGOhIl9lv5xrV38abrbTDg#Lw zk>pGE;+!|9E(15t z&H7XR!~jPVRl`u`5Dy?%Bc}Gc>H9Ub$@GJ@@SivpKyjv!fIYj(~$bZaUL{vTIhbI@9hr_zTMU%tV$JnLKB*UA@Jz z=Ha3ihj;?x!5xxyX%0x$kfQ1g_&reI9AdPKmRBvvI(76lWicMx%zW^;`4QY}4gBZv zQAS9(g6pt7fx#G*L^PQ&3?01H#)%_A6!;c5Uf#juYFQd(#b4?R1IX!vSD<9{@vm=t z@f_fbI-@6gveA7}Y&G$HVj`12-h_cKn3G}42l!0h>^VLV)^AEv#^Hkvz6S*oZ96!8 zL0ZQQE$>s$2+UkF3;&(2NXYW!pF#z_{q>FLWDxSo zp-07!yY{IBdp#c?pVUz8T2kLf9Ax@G{8=}l_tR|GoFm{@1nj3s!qrGzRH^*tw4@%h zX)7JZ44p>_1Bbgny??xl-uH2wPVKt2t$F{>VjGlh#yc_1ROxdKj98JKv-Sy0d2R*Y zGPpcTxcFFnqjj>1a&J^82K6#-2q`>;YKz^TAm$MfuCUC}vU;=lFDux$C@iDh9+q7o z`7B{+4UbCUN-x|jK*4)~8*RcEW@W-CU{b-RrVhUx_y9rQUfLqMKUPLp(4CP;7`>^` zfg{c67OYURf3+Mcvb3cV9)BqWe7smI0)CKcUrh}H-Uin&j4(94p_L`GR&GMv{bzh>A0qGl0NDLcP)RrI4h}us}!w173ba1u<@z=9_)3CU&GqY1%_P#pmT zBUvW5lF#knGox6GdD^s8yMLL_5cFXf!(DqyW{;k@t^{T=4aFoEd~~>EG<)C&p*@Y*dI{E~||q>dhnT&OdBy=G;mdUD5YWHWjFt`(MV5Z^c$ERS5!G zk!yyrnBJ0JZ|nBQEvwmCZrE|8%y>bH@?lCZji$)JmD`^{VRSi;=O&BHPf?JahI|$* zMn*Ep4g>VZDph$;1L(F=8+*3Ez?LrL7m>+Dn3@rO>r&Rv-FRb;NiO>6Xh3gP@#6*% zVxxn61Zk05{K)UF zP(OkoDruZwEt*h?i2&l1yt2Q6q=%kI*5X1IMwnf<%~Ez92l_)7ds$`O-`bs7_jagI z_*N|8Tp_>RO>j>x$SNZRdQeUlfIH?`C}-$h^%})2LEFWU7cxox=(MW{?V#wxig*yN5MN#e&cW zW!HCT5pmrwf+N{>NC9rF*aw;?a>9EwBp-<1EHJcS_}4$bE0G1s`UBAcK%v%I#$kgK z!e0Wm&K{l+5#CC`F$iuepP$tu8Uii|z* zkG{r+kh!d|BjEBdS^&qlil4R@*5@7-&u)UUxUeY4O!avl?YhE+JsioIw}1cs(hQ&_ z1mb;~nJujn#pJ#wr9i$z*9o_Z9W{@X3^j&l&(q1)>YqGaCzG_lMaXZx`kb>PtruLo zkMm)(8Ixyqjw1`o4PBFW@0nhYEnIm{1A9!a63a`h81BJ1{EXfLPxj!+CS=`NMew(p z#RzWoWDRIlj?pgE9b!hU4xAkrUb6dnpY_R<*CT(+v+5>{?Nu``D?aesBHv#JT~}`0 z1Mt8TL+_#^9GW0?;TLPxLCKNBa*NHb z(?P(UL@eGz66ceAUft^F`L(;VQkUeK%Llzy_??XuC_>3ehMeZ@m_Bi@pRzN_rBpRs zPhpr1l8*3Q8WVrrYoC+WPmnJthgkV&b}x@(&(g~G5xq`d48srH`h^{M=Ui7Far?^U zwHK*QUt>sqn;HO9WUIDzA|7+g(C7k&nJVRCD5TGPFFvF97{M~lh~a8F`(9$HtXyRP zhKu-YKfxdqD)#Icp^M97j%BH>zHv5tjxl=rb4$R_Ykzte(`=Q=C3`{N(puxg4*(;} zn-3yxvzM`0#9bq>!$_6bfHlowk;>yUG1my9zK-tuaseqvx#ox3Kom*o_#*P-zjl|T zQVd!e?mdg|Db9o`RyBtUt^eXP*#aV|6Sbp{r3I?*7g2SJ7@CNqZkA3Q@Anie1$v}-gpUDu_E@iK`C<}@N}Sg_10&slrum~=D>EwQ&_dlN|{ zcz3#@3o|j}VIh7MUP@A{(5&UtrOPB*1jLjfA|__Z*;tCqa_p6R(5;z{(jMBPtY}+{ zHRnfeZOTI=55^HItso3UZeLciTj$Of<{dxgWVVEZEY(2Ldw~$``!_G;)buacOu`=8DL`}Zsl&D*7pBCAhVwuA&X0){x zIpf<0Jp$RR9QU=%fqR^IbnMc-i$yu&R3;FO|7MUa=X2MzD{}P%F*(Wk*|9#!V|z~} z9<^v#KYx+=Jw|zZ{%{A*m$#Ptq?oK!CWAk2_&aW6YUz}k^w1RW76@5h%(0LgXG}?2 zKNujAN|GIBBBC4bV%-joe-CHhB;%CgTOm$LJR8bIALeWYeB+pRfDUKcA%p<7ttFq0 zkIff>Xf?b2!gs~|$nBr(J_*L5 zMn9WFJ)GFz?SxOtJ;iPm3pc|U-{1oDHg3_tKB4v0u`q+@`8d0x`k09&gT?$4Aa#pyWlMt;_PpxQS7i*FH z|BkV7HDeZBDJLxv%p3>95w4ru3kk}*yrJ7jBX)L>^b zFZh-UZfV@Wcs?zZNmT^(_MV>EM1N0i*_g|^I(6dSr0>{aqq_%MW=ngUK&ZES9d?Ri zt|RVdKbx_AeAioK997|I03O|XhT)HaT=X;BakR+d^4V@u8;r0634#jJf}q%TtvB=^26gN;UNUnM;eRPIxqRVr=(^_x)#Sz z4nkm@(K>H~W^c|rm^Xc3AV%)ZOb{10w_Zk+AyqUDk)xwkTwHA9;HKOiq()_(27p9cNsmbEbGuuMr$p@+rmBRZ-}dBiHuj!gyZmQU zHvk9e)*i@0khom5y_U_A^f2k*7&BUB6z}>kW#yWeCx2ENVUnngOO2T;pKO;!m!Jrb zJ7GOO-c|#vFJERVHzhwy8CpaHu;#{cVAXL*acU|kMfK2CvA1R*TmZ*{V9qMu zD1HDuPAVvJryUXbW?10f(w;k`@^prn>JhQG0}kH3?M4FIFO(p`lC^CjNz~uZv+@=$ zct?Czx-+Mf022a@?+ku7xKL(Yj$+^p#!o0qEh4P~fi`9;cW+5q8Q%eD@7?S%WZY?l z)=v*0Sn&(H_Vj70XWS75(Il1Akhq*DY^tQyo79<$?>1GjL=GOgbZKP8bNiw1moIV~ z8VmpnmtwcR)qbd_*4Had<9?ujHx>SX8eGCOQd~yIP$^SJk8|PAvvZa$T zfWyfBk%ie&{6ddNt4WQRsesGnb(U2_v!W>3gpXS~e8m+77uL0s;k29$+@xz^EO=wg zB~>igqWjC9j6sv$(g1SARL#i0fqokA_Uak7is30VwKsP?#8IWXBH4cIUL^#4d&%2v zCYU{)^d)%WsuvZ*&4gLo!4G>mP+rx3DuHl(tL(pb@k6JMNB^DeMP#WQCy9s2d9l$M z$(eb6!F0sg7GGEJO=x{i?a}z&U=X=$>6C@znreAi!2`y&PTJt0v^@5RUbAPq0kK;) zjqxer@3nd*4wt~T?WL!D2rDso*oRM_by~^+_~hq_M5TvPrx8&Ghmoy4Zly08z0SO{ z9xr>mGG1M1H<~K#F!IVz5QW&wxhrm*wwmIgU6^6C=Fb6nlGxG8nz-c|-R>v?#v>^9 zT7lCtRg4B!X@Y;0O^Fy9V6vouYsHW0qk#}`3;PCE0iXeljo~!ux);jTO_QDEh=p#mka2kByIjamX_!InB9wD1Z8fg z3)(p^#6{f86SW#HXM}8?Z&d-_cgpoOEb6W&wOk{d(Vk2#jwZd z15S-GOJ&)L-apDM)E|&ZwV-jNk=}pdY~6_~>HW>FA;0_6iK*{6esu;hiUq?H>fRrB z8?sj3BdtSSzwqq^7>Bt5fyF2Cz34}@4eN0q}8)8MzQ9(wy2 zQ?>3|u6#d?0!w}i^CveJD$E_aD)M?tId5yiCF44nF-iNL7s>Xsq7=5}PZ?@)$Ie$X zoVM#@(%+C%H}#lVbT_vc9y2h?yv$|Qoxsrv=fNqG+5#rorit04WU;qbf?J|5ZOyY8%Hk;p3+_}5)<`fhUWd&<6Ab~+HRfrMDi&& zVIOoY(hk@X-&l06FQ(w;06!|@pAT3~Pnc|-aaD~7a`H{AIJx#=alWcFGPK@l`d+LLck65M)BBuV;Z!YLAjP4(LBA zhwmsQ*7~@4JC?tO*IV^K`7K{vD~F!}XJKzmLsDj|5|IUHMwcup36`D<5*xkN_8%QWX?5}(R-bX=-fmX)Ebo@%vl;3qzUt`69wizDA zcyd1*KHO_TFNs|nlIfs!*mmJ(i}u6 zRZ6@Um#lpbEFhs%Z*G|L?e$Z7T|OBPRl~*B0t=fr~5H&3({9`1g+;M z8v?Be_Y8G>$a16?nV^NqpwFCVRdQT-`V0q04BW!vq>DXj3d2U(g+TBMJ=UjohE81C zn5w>ZcqD z^6$ZucI(yP;VN;{X@60^wHnJC0eac93z%*mBo$!g$Tt$rDB0^{>Y0rU)axFI{#q)J8C+_Ow+WeCk}ciq)L{JjIo$W6pJ$>BKl1o%=2}KWM)ece{E1e);_B zQ7pM_B$qt<%9IgdMd#q1*8|%v*_?+Cf!Z7$9rZ>H%kyfzcN5BjXGi5G%HX6~H^c;f zuKHbdt7i1%xtP!Db_X8Dz2Mg6BfWGQzod1LYs`=zVlcVVe|WdY^Aw2C=;9{Xkfl#x zwa`<4V5EgMIb32vV=1-XWVLx?BXPYyWiumc1dHxemswm`eI_aCw(4bFkiN4*yLuSL z!ZKA>`2p~R0^K2looLLH9J|b^on3M787deiL~4_0B=PrK^!YAOU``t54iyw&kHbk? z6WB2=a2}67Oa-#oJHqBLo;Pd>L)^YhHP~=Rm@|o9J@BLDF-=^F!)H~`5$Sg6yL^Wc z5WT^58Zopm-cg4m1w)a-OVYer)#%|u4WA*24H9~}m`Cn_4uvkWwV`2Xa)x;p^NzMM}YVyE4C-xxlM>rKRbK7j_*h2jbkqvD%^3<~c3V`7u$^^I!K z!nRNeUJdp*Vc~%%JQ=!+cQah_+5%^)YAY3m4fur)Ju#;@;ZVb0K1^i1Dn0RK=u%fb zJK{dv{EZ0@np}Q$a?1Y}M6p1e73dl;4nSQ~)M}Z8%y#t6NVH$r)s&|^2>O6!%{)d-d z$R~mzZIgeXtA(C4zBHSDR&acuP~ZukqrXc;;v(QQ)10i!wB2cjNRTGH=6iCt3dGrL z@d%Xv;UTus;Tob63rfAAV13F>%H{wxl4u^q{}I@U9sH8`AxE2=p1RRE=HC6a>UH`D zZl51+cWP_Rg$9ajYbuPxawi$fAJPhhYg5!-x@@(ww%MBS_WWKl?HXTdRUO3={ek0n zr^#pS6J6A_2=D=IJ;Aw>VpB|F; z*0;B~6&1zp1jA^HU4s8Y!)QQqfc@>A)bfsCPw)F|GNu2MfZ`8O{#clG)@O?lXen;Y zLuiE@yL4nzMMb51PfS>tJDCYq{&^22@Gpca{V~NjY;Vl%*MrR>fK8BPLuG#FYF|5_ zRLu8yjO(UT{yFXgp5(mtI?;eSwd!`gsnMr5FMTm{ljeG&Em_pEHCD5i%zGia9@qNb z?Y6N@?rB(_7NeBQlUADhdzZR%TCBD$Sr3)2Yly)! zR0n~Z@GmH;p?9RvaQKr8Pa=FIc;Wxzpq#>DvGNjDBb&Zw zq2u#gDqC(9IhLLFW28N?PF!o>hl&g-0^#rJnPJE)btoH32d_L@D|!@Co~GO7SK%-a zVwh*K`dY^3{*8=oH-zmWDRbd9ez&hz7oj))30|>bX7^ujd?vqglrQ8mK}P{_>-b!j zFKF|a<2kNUuU8VDj$w=6NYM8J7P8u7TaS6yzV(u|EQh582b4WHA|K9qsI z*yD?Nh3I3hKS-ZDfH?L&M~FXg5&1P0w6MN~I1fT5R}s*vA~x`B_PdCv?smMJ>D68O zl5e+p{a?mu>gA}by)OlG!^8zD{e|B9X|$T$eX0R7e;t&aqs>r+fiJTK`yZX{-P*3m zQSxHt<9@x)XWYdl?BF{Nu<)~7cg;wW_|^Irb;!cF!VhdEBi)q%7`B)^c^2VuA;358 z5Lx;IHGre1ySdyKlt~-7f28#3MvIoP44To*#PaYo^4i@D7ShQtU`HHKG?cik47{v(FzLfPZkZyd@5C;GDGYv#oc@!aO_9!I8%y zZ7<>b8I498LE6R9AR*f zpjXRhMR`jj3*G0c)B4yR;yZzwH`NAM_4V)yHpKJF2U}2WW@E<9Fvp2W<&tlzDiv>Y zl=tRg-d5jq?)`~MU04hp0C?2G+h{t23;XeJH?WDlno?5+R_e+vJ+(!zruX8oN$0dX z;zJYOfzfZfubr!~Fim1gV$r_C1-Bk8=9Xm&kM52uF(yNvMsGp8v!n^;YHrSd=$~9k z#;7AkN(of!x|M}OWI`BJ=WSpfN_9@8Q;S7AJQF7{NMysB902_o!yphU@t(6}wf(3x z_SY{ni6)+rKPr*Bu~*o;TN3(S4!qJYe@RJP z48WisFvP)GIr~IZ$zggQ5_-q)rlksB`He_ui*#WjJJfKWYXa!egX94>1>y>q=qR@B zJEJ^ZACd6yI8})%olTBiW#ILS^EuglQfQ#4K9D_QSkx052pSr1JLgSu!5A!L@qgy_ zyIra}>cTzyzCItZ7C;`Jnvl}5-$M!S(?yA&F;=_4npF2fhWDPcK5@C!%&NWoJgzi; z3&=J-v9BFeLyi1sE;!!*>&HZrjN`<&owePN&DqV;3iKOk6C+_jD*zs;+niR@%~-!j zbhgJwhJ=-UFAAihn{=`u+FPy#| zeqa1ZTGpj*-tGw-@9XfqmTe$K-+BmBT-?ZF(iQ{){>+{M`;@ z5zVJOrh0pNK8t4LSnn>G)Wb!GLV=_PSTk&T)HrN1D9(($U&f=9Fa!`xE(#lXac#dJ zYnTELlfrhh(d0w`jx(DSNJL`Q?qznnxNLkGaeIB0WJ;hH;(x1D4-_(<#DjIkFt-YK zP-rdJeC0Hwt%fRyvB(mddQqoH`A;ER>sd^3W0=Q<_7ux(T~S0SFh1XkLWyI8ta)pn zatT~{|C8+(r+NJq-p4)A3-pAq{`?NxUXou0$Yy2rXB9rRfu&DQD_7pBMnBc(&fYLGap^kO`jD!Uzp1A-}Z;3V=yMyufXLa!=3yezW%o>pktL z(;KE2)2o+F&_iwO+H@51|0SCmR`49kHMicPVvd575~IO~MAOfgbp??GAGJMgy4d;9 zyE7EU@?`!NpQQaYwl;kzY!&CrmS!~gg(oTr1Degc!t=P;bsi}Ma5)XFY$GYeeudNP z0FwihT-{b0J%dP4xkCSNFc*!Z{~}-e_fzxwQ@^)_e6|4^Rd#}^b2!E_AmU8Pe}r)A-^?^JzoEr>Q_Xhx z^d~Qxmk%XYvjt)5-ky^9y>~Z3`_l_OoC_4qlp2% z!$qpk&#a?{7zEC(@FC4;`KzrLkQAWeDE%;U%=RW;xHgB z_Oh7#AOoC1ecJ{6;FqdZ-C7fM=WSFr8);n6`UCMF{Afk34 z2&zY#r*Qucr@NcKY!VqTz6BNfxN>8T4{qnwu(P-Rc#>J!^_yT@;!C?@-fuXsXp5(* z^Kio{&50TuKhpGBRzM}AmIiUm)h9B)VF5z|+15c$hqGP8q4QnsQ}@`P9;5|wUfCB< zhq(CRV=e2QKpv~Cyfl}7P$s{`6p{)|fC75^vIDZ~silc6LKDd2VWY0kXJ;+D=Nzr7 z4W|YC8H;5Piv^^{9vNgYCK_%Q3{TQ!{Gc*N2D2Ne3)f3w5wD%njOFx_&$~$pg*63) zhjF%Rz9Xvw3FIpT@FrXln{Ds?hdRl*OZs^ikTa)8Pl;wb-OaxH@YB)lhwI{XIHgFj z2d~18mwnq}FYvvb19@v!p6DiX;Nk0LR+~FKo6L^L{C2JVx_1IQs`E*t8G*HPy=QH8 zL3e-FjR5KunX$p(sJ|Xi-5m_W&R5NpLjQZkE$(gwFClOVO#NlYc4615Bi z)mAvHQ%g<}rpx|e+awi_#L1_fTf^*Zbzi_&{`y9yHs>fSFk)p?)N5a^#BV5$WGZU+ zFlE=%L=&J;*;F8}7`|9bF0ViFkgpKxn91o{$?Uo^*mpyR0douwJ1-or`Hs|A1fdC@ zri=IPBCcO9S%T5v;9B&z;EC)AxKRny;Il7}AGYb{qDD`{fPU z3f5LfqsBw2a-u*&o>kGGv!k^EZj)`#DB5ofW{gaUHV)GQuoJ-8L~!5P3Dg}CvNT(| zIw#mYSM3x`H*#USR`~h?vHPPgG~)>H<{7;|N^bJf_+H-F2)9mLP=0Q)4*GNx5gWXS zjC}{zu5{fz8nD(wc1Zd^K5%xdEoAz}T_Jx1${qh}V_X}CAjRX+j$!P*S}R${g)9N` zJNi@yqn=6SiTn=T@l`{?xoPY%+U_5NB=(uz2{-#A-&|lmABvUfsjw7_&c<+MX~k0K zM8_-GL*yr6^U#>TzDop&Nq4tZnbDuqb!tzkH~#^b4Epw}OTCf!-t{Z#v<16!o?c(1 zMT;i--8h%9U?S|9-Onb*6*HA0(w(F6HZPa!S`QmiOVaf5?3-r*LiPhNX&GB=*v+zw z8)JF%D25m(j$gG&$9b%4hk01UWecqPa}6*q5tcx9#{mbT1{G=^j5o)Fu|1(7|3Edu zqYpso=dhHnmL9F}Y`sxQ=8yf-R;(kieV?eoeo?Da$rI5BTsc?ju z(ZYC$-6T*!atq3oX)PS(I8aCDW&<1u6LXY3!a&Psb-aNR5gA1&&@Ioe184EC-(3CU zh(j2fY!saYSb|D76Q^SZ6_*R-F5Me+>Dw95(-hwDgvQ+_NUY&~nilT$~ zq}ib267**PKoDwpX0v)+Ck>6@ekqy|AppM$fD3Uzs%ZL|7(=g@L$`+E^?T`|_T_Mo zBJHr-^n#SG$xh2nq z4E%(}k>7|bakq~|&AXf*S;Tx<7+yAeiEc`U8hp`__Rt&`9>-(st_gjqaNk=j?WF5FE9Vq%@N7{=J#nFUo#XU1S5 z@80WU1B7#r9{X(u&?dFWQ*ZWb6CP?p`NkaCG$adC-iSB zpx-IwizM|vFS?ua@LwiD3g57TM#Gt%=ZMIa%i&4!UBSt-ituhW+%IC*aM zjXqogxP_NZ?EGV+5{s*Szrdp=wvWrB9AfF9W-#LSbIH=So3Yh<{MgD4mSW6)&!+LrJRTT_|>p(iQM-Qx-N{OcStHxUVbYmPFhNi3xpGn0=yKvJ7xGx zzMm2h61uf^+IKEsQ?>OF{;d-x*?8`NrWsVoD!`;>@JGEy2t8jCy#>5n|Fv~wDf?5Zyr6y2IGBxH(MHmc=R7C z*HkRrH?DK=lce8XkuNEFK;#}r60+tlC8+?KzgO;4z<4Iv=3U;Df|f98cC3z`_}7Hz zMn!T-^!7u=Gu+mtQC~3+yQc@Wh4s+MYa!@?@sk7mb!G_rRA7}-JaC69qYotR^iEmj z#nY|SNDm*88aO!jNu@!z(9M;ps{HmKU_-#R zv4`9JAE{z`q-J}E!*_gSY_77+#?XpMWaBR3!v=81=BdhnI}KS63HR8*=>cH-O2CZo zOUvfA+b@8NY2L8bB0XtfU9e~- z(Cg{7LqO5fYU+`asBihNH~(JTv8Ul$^~^(w9+Hsr#BnVZyx@8{@&NR`!H;k`^y zSMQpc!{v{}{gXV0%VUGsYhGbdsED>dHZ_BRDDkO)%p5ywwC35J9hx5$S$0){LLIm$F6 z@ss;HvwR3pHx97ne|=0MK=hd|xb7|tq-as@C;e9pHySlWC%f4XnTA_u!RLkZHs9gx z{}`K;&VsA9zKMMmI9@X(oMC^bC6hJ}Qrym_{BRx^i}Zk1>mvkStQqWLqW*Zu`~;%Q zQ=(4t*12ArH`u({q3tZ^nmK&qmRYC`Vs(7{#WsNV3Ht=z3>??oH&YuR2&7?(O+X>1zXkOk=QSTy5bN_45hzQrh8P3Tp( z+&!Z9Ur!gHNj@aOlUc!G%7pTi8K;)tG-6V=xxeEf{@3nTzSk9e7I~R@<#+joU6cJV{_1N8smXOiRUk-pzK4)BlNXU@d} z^4RQTn2}x~^YmKtWQ4`qYSK&c<^C`9Ky@Cn%3!{f?^*fWW)^*yOS5&dS%BmvNBYVn znWvYnQT!871)aM^b*z&-RgDkh=U&G}tMgFS_AOK1Icld5oBJ336`Hpg{n_6SpnLu) zmvc9=YKQ$uP&$6s;;gB&9UJDZup|2TcarGL`CIFaM%{py#y2J*OM@szX*1aROLN%^ z$CJx3d`qZn(_t^1PtpEJvio}SBa*cSU|6s(B6ogp#Cce2`@GNtJekBnwohSMfGG(l zM1dVV@^!Hh`WmpKrp2MJm)X{o>1{-k!V*-J;xdT#;Cuhg z!sHkLdzY0@0XdH8*1OjT+W(u#8#g=KD(Vit)(88!-Z`A~_)ha<#iKj=oQ}s>9#1w$ zyjf>&TqvFk%qLS9i6SS@AfF$Ay`zWp4Y!pcAw#D(ZWtw>EAl_L2gb%Pfd$~f=MYca znsu5Uki?S5Ly6bRywLyC`onITB=2nr^nLL{nJ);C$60+r_@Nd>?Nk%}q;ey`y<}NV zU?Re;+KHy}Az_o1w<@rfp+Kd;0)?1%CoDVm9BQ9+YxaIAlell>0nsg@Wu1mJ^HSg# zVCl6sG5>rhG*_G!0eq(g`D_y-9dRb~U=0&3p{gZG6VN4NW$=JTs&txLb)8S2>T>zE z=n%uk%>Tw!Q}g$ARR=;f5~blMTSgQH5bfm2;N$7@q6!!d#<7RKfazoWcaF(i`-(o*<|B%Ar$jPj%@4Kk9 zHJTu)n`iYI4Ox90+hlMNf7N^(kY(Y)x6biXNk@g-f{r)HhYa4e#IN#Pn3WUo_x?ATjoPx}dq z3b8Pc2P3`vw;ZHCGx z<-W2i*Ga*DS#N|B??v%zxwF_PD=)j?I`-*5CTPA%DJCPpoWl@ z)uJoQYvss=<-9D; z)GAApAEllF&9Cc6ZpxjEKKv}o)v)fx;QPLiIloil^|&y&lV_UH9?JfOPT&&o^S{|m zH1rae|N2dE4Aq64TFb2fEGc3dWg7lr5fMoYK+XMag=r^c#lbjMXcZ&9oS@~rZ+70V z9p={&wrCtOrTF>8(9o${UU7Wax~z0f8PBnAp-{=R9n=nHnKq~2DLNR5huNJm6aARr(er1wOrA{~?t5m68j5Jc&OB7`34Nbf}mC7~pw z?6ZB}?|k=n?-}11_xzRYGP37dJ8Q1B=JU*FKB6a>{#C62V^M+sNMSsCH8c8|UQfQq zkcaS_RkO|GCX~K0L-92JN)}BA2p@)y0hCkMEHcm&-hUm-ihaJHt#CWe7=G*)v>U6jPy75M}Th(%s# zd(R|oXj}Rr8$_XdAA_9diFFJC#6*!ef@}s5Ti$s(gKwtT-8CHL}Ads z9b>>K<=?C8xjarM&~xK6gnCOe#aKQ}{02gGJPo>ML>mnJK3D<_`VI8Vij;Bjw<`7c z5r1_iHoJFT2v}*_%zwJo{Ld+G?>)EqNGGT~(;c_Bjg=<7@qEp3g9~nl+-n@@mvQQ( z$k|1jQ&Pv{9EgH>V)uDgJ`o#Np%&(zMP7#AngGu}5zYyaEr~sNE z`2-}V-$IMc{UhjXeg%ExYyvd>E9FqO^b){5PJIM!S4qUXW=rGeQ2Z>`T(FtXEt`8qZ!>O=|gj zP8^#cNOgQ4geMkjT5QI3cTLnOvbtFSRayV&0vM(4YW%wXWt4f1vBGCVF5~17Rr6}j zOc%$Cj56At#P%mkC+o!e67F8Ju6i@o-4!1}155F%*5S0PF~Ap>z1~Ew z{-62_Eb2e(6dnGmE)>|t&AFU}lz~YPCZ>^`__Dt^<9|S#|01n7c73HjO_D0Ks4ClI z0Zxyw+Vl5{VQMON+o*Z9EPW3Ot6So<9VNOrYfo6%WIPc7=P7Q$QJ~n+Tirz97@QFP zVx`!Cc~IHqM`^MRUw~-Z|JyE!l>d+Zfv@XsYUE^JdL2@OZp(f7jao=OtZe^biciyb zPv_(x8w&F$8?2;1f55yze_(g5bg4Z-a$aX)ySr&|MXT4RJYlohk!#D=3it-Xj3xaC z7}m$M;MOl%=|)1w?*Py5@4>iP#^yFQ9vpuENFacqpZsGWFmXS}{x34gOs(ugY^7F$ z?s)%m+CR{vSC*)#FAi!r-10X&zq+LraN2E3ZTgq4!XZZ8U$KEAs z(BA(LacKK7Fu=eSa81Ap)N=T~;P6-8fvmaP`kKu_PPkJk<3<7Qbhh@>C$nTiBm=0q zaa-}oMVozQam;&gilYr?*cM_3Gnf;1M1qhcu3@*W&bry{d`SRb=;3Dl352Z7b9v5{ z#qn^@gAy#_d&&4g1sE>70B#BYw?2YN4491-28!IT{KHOndS<{JJG9q{AQbF@?b{CV zH|g&O;cMw#nTp9^hjvOjU;LNNT>M0G#H08(+fY*pLa3r;$1H+jI=I2V$3%jl0}_qy z5Daye{Qbb0N}!-t#Hfid`hXxkcpu(rB2XB(uj%ymHTxame+3BSlmI(|%NuG61OC-a znES4j*dKvxm96-fTq`HB$C@^7InE;+Cq>mWhR$exBUr{*TsosKSVeP1jq~j96oKu8 z(<+(cfF=yF!1AA}2B=H_gPwuh83eY_$P4n$7dVI>kAC!zxWS9~bm1TR5hMs;JJ};2 z{wKkIOT&~GB<5GhBUW!kH?7%iLxr0O*9o28Zp3RkV0au^2&seP-HM1&YzdOz5}zj5Uc;`B+#eNZ2|D^mZPZ!NVL1RpQ* z2|EV28Q|~#gfh{5GGACa?Y47sCA-6S>PIO^)G+KUwjw)rxgh@!;F5oR`=9{yQkKM_ zq%5)iOi<5`M@5&m{ow=3?rkXLDQQZ166mOy-d4N&vB2p^3ssnZH4;w(yu52+5l6^# z14?HCKs^M@lXI~IVgC2vZY}3M7L=hBumS#WA;|{bre~vl0nbKX6i>>LZCS) z43loscFt+x+g}}l`)~t;8s{@Eji8de5QsQ^vrVoNS^zdIc>;>JshcT9iCLEMAlzpl ze>|>b+f89>d^{Bzz`FDy>m$HnUJ!{ZYMyb~p{1w^)B@ZA>%4|K$$lJtVZ=C>OQI}r zt1(cqMi7kM1*vR+EWR+}KNf&6=b(vzMNZ?HqydFM+klmjTMouHB;3LjWcg-KM@VsY zvY6zFrB#k8Vu#;M+O8mPkDKKbL#@#Xdn=*W%$nsBkoNLN$}QmH`r5CslTtyKV|n}U zoRsQBPug4(x`6_W;_IlOR;85^--2x{%fGqu5h%H4q_Z6;SAzeBuZ>d#=lpK5wTD-M z)p4zoW*W)oPnoBIHi^t;Zzw-Tsajk?9*BMgzF_!?p*1k%G*{H$ z13$%|a*lfwBCNl`fe}49@WSiT|#=r7@-0@Fd zgQnTH{ZBCHEAl3ctN57ea?q~4E8_x!hg1F~=M)lC!C@Vq#fvqjpZolZqF6$M2-C-)YNfS?CY`X77Z>7rJ433Z!-FaJ%i3ZgTjWgg{%^iYHw@$L%QfD~^tJC# z#DcV%tGZRqu{?txP`8$XB;C(DP4H+V!6d9d$BYJd@q`B?A8h&mfS&%3b!_ODfj@Qm zXJ!y(aKcw2*^T0S4&)>FEBs;4pv693XBkPiTYOJBeh(S^N+x=i7zQ~86THU%1yHLc7;NP?Ti@3oN8L)#bkcQMc+Ga{ z1u;3>bomtI>IvI19~#)(Acq1Uv0Km~!Ll{RGUsEYuvcA-&Bi#?;Qu1MKJihrm6ZRr zBUmZlAMaP`xm%QRtcNV4>6O2L?ZGDMZ+gD*r%b0&-_@f&wYjdTiaX~5mYMZ&P+fO#Qhm&ZshPSrWgSER<)!sS zLRj^yD{7(ZKYwm-{*h&vB;1Nu=Xx{+w8(Z{aU^;Uv8LZCBEL~Y7|Sul8uLx=S{O;Ue{%~M_- ze(}_lnVGA%n00KgRfeCg^gI>QL?cJT?6^f+dFf}fyfQ3QxF%Kj+n?Fl?GCTz^^!i- zKp9j|{pcg}BK_x8INs^ldz{8?+!B4?A{(MR^R-<+Gby-d5$7Lm{j9Ds4R#LuA+`(<{EPhM;0ZW5gbauZ$xJDJ|HOxe|K3@0<5{k$0WSe zSC-w}cG-3w^K0SI!4SUnQSQAuS~ZC*H*$YqkN@3Cp*-j-+}r`#=5w-+!~J`EFH@z$ z`#(R@Iel~v5fHF+jYp{G`_h}G_L)%oSGR?-htJ?l*viY)-P&u1&gdmQW#_!h*ePXH zZdvvk@#(9^8!73moZ&NGr7ktm-QTNc4vW^wM1y>?C(Q66)rU}tN&F1AiUmaXx6W|G5Ozi9WF1EV-aLzvT&b}~qk#_BCijiq0dYzy% zSmo$7344=vuLpUa_L!BC7x64$<0RtZUn3?hTlmzhoWFa099`}5{&>A`hYe1;3Z4p~u6atjNh%#24<|I8LYmIQq>Psy=pv*Cj!r7= zXO7dwDG0uu#3ZlX@BWEIB&cdVcnggcMO_-vATBjW{Cik;pWS)nYssJ;&Y|QC*xvM{ zwfiwRhf3|mbczJu0CBz~(0T73KKx#~NQkr1cZoX3LX_VtD)3`_+6Dfa+>u_g%e+x^ zP=ki0`IJ)xyS>GVGFu-8-x+x4TN+D2;ZY$2@Rzof2dJ5e8kOSIf-8ucsS0urvZJ%x zAc{?KM5N-*1ED;m04AAv1@nw&HPO3K%7)Io@rAXM@0rMxLs4lyJ9<>|g4Jvntq{9e zjwDxboc8oHHFdyMsw-gVXKEy!IDFy~tVv{U-9}Nek>(Rn@}%7OrRWOVu|LhHNtSvTj)>PQ8B32?fjSexVdJ5 zsFw}Y@yAxXl+QW9z^>adz{Ne?i$iz9?sSzc7;I(+P#qrYE6%M4Ppj5EtKX|fs#E;G zQ&5*2&StXA5m|01vQph^VE^?nS0u2zpfRu{2y@P}ts=x;JR<^4eCwmq4j58sK8m== zrN;ldDf;FU=NFzMT9ka!#KLVqZqI`x|J16%s>s$)2aR0D%R);X7E*7~4wH#2s1Am3 z*~Xg_-9AJO=6s6sZJ0>mGm(^c!&QT=q!k-Zu(TA1-1PZ83AyPmMm4WM{qcfFB%SqT zBIGvl-CW08XtyD?;TZcRa)mCvUH|yMmi8`Pm=}1+pkA`=j z@SE)Lzrtt~p4mPJmr3aR8L;^08YYT%Fnmq|<`_*EXshdPl20GSUYNd4RnD~-uYak| z9e0CWA;Pov_cX-)CR19Ze*?$PI}YU$!w!23Q()^I z)T|~ocn~jjX?0CWr!>C|u5dLaW>+-Odv<}YZDaz_Iet@)48;GUF>j*hB4q!HIaIv; zzaPmU&&fVv8_n7MK}JG~iG10wl5+V5g86J$^+K<uBm> zEO^Z)^pjq2`sJOotT?~!c!<@U!3Qo?SO(p39x_g%HMXO3;WzWp!fg)w{g;9sY&#cD z+0u_6;X5WmLWn0Ooe*lu-qg)}vbHYq9rV7}Ungdt+UFrW&U0VYO1XOudnoqzR1d~0 z%kvMjsuhmPmX=Z8jh@)X+FNMMj3Bvfx9tY~ z;Lz>bS}1E~a=vux!t6>->QZNP$n;{$(AZ5D`WqqcY%Z^DR&vL6pY;{*t3egBew||W z47x9)ana57J~M}dbx0-S$j=w}TB+&U`YAUAUevwMj-5pGD0J+zGo-4Pkhyy>i#>ey z@t5N>=>)2Ce3a|l9fPQZ^BFYf8OlFk`yON0M+o91mg3E0lJPF7C$5r<(} zIQ8AHHPUYk@Ou9~t9n99m(>?ql0fSqg$#HoA}?B`I7xU8+GrV)Ky2w-m?`7 z$E^)^FI4OWbT28nx_{G;7%5ko2!00*D(s4fh`KE$2~b}1rTnCmvblHtrzPROSx^?E zycbN6d}@fxa#>2?=u=y8)h@>$;x7TV z(7JMqS;bjPBnGS2ZPSReHK%=j5?#p@ftPFjm@X7=TGSxM`1O-+>$cFqD$`i znGIf+`H@r;8ymcn?y)~Xi1MJQb36@u94FiLGcmxkJbmGzFt>&*Dqrm%kt9oDw zYxi9u&^i|S%4j=$P1uMj^*YBdQL1%7aYc&(UAx?!>T;7@?%lv*{m_`;a46@KX*O1< ziAYRCG=e>B**=3yh!Zw+?LNHQi8R_d%Ew}9IV)u*F?=Xspmt!X zGt}Y%8YNUiVfFR8MI8S6cF~aPN9bOPq=Tfh1SWJZ{I|{EcZA48XLF|r)&1)dU_qZ5 zZ~kpYxJ=USXw1av204q@QT(|@7XluJ$N7ifTi?Hv#bv#X~O0ImpAWVA(4@%E!P5K5V%&ok)Yy5%|=@dMxU zEc|5%4Kwf+4F7jSs+^!`4PVovEd*uXm-U-~C1Vn;gYh`H#Al8_mV5F}&=LYj1ZlYTN-$ROWdT5fkvRQ%(?rNLe=| znak+?hXwPTX8J)dqc~f4`DIjE65oE*W$y8?HUB=d;Ghxzr-bQ>*NEMH&oA2BO51Mm z?czmY%z+H2>D@drb_F|n$+6f-u1Cfiv1RR?j$26rr8~FtL}Su*_D%NMlgO*_kk<1W zRqN_moYnz685T87hu-%+g$JO~m&Krr?~uLxZ=rgks5tx2-ARjsyHiS5hZ(Xqj|v8G zXcO^wo66vWD{**3Vx9Q2MdBP^y*bz&?L9h?A55RE1^l)>jCvF=4pr6edcPQb`Oy^p z>I$y!qvrlVI@o;GCu=-69dUKVc+=vA$cZy+4(|yjs1mEv@1C zxJGtJd{!N*#1Pn6Dd^=JMjr`7P33W^n-3`Z zyWW58+TEV7>D6w?iQtc|3N^-`uefz0&^T=gJ}(@uz3a875l&7|bBxfAml)`}v$AD8 z^6?xc6RyraZbR_TJ|ufSvj`~u$nySPV%qu!KfSbeR?qLuP3@@qq6wn)Gwe6zD~J6Jom+kz?D6O3w#_ z9HOXB@uVBr7NXN!-eMGvyqYnqWP)wso*Wqil{l;tj0clCE!Z?IWlHwpNBXINnHoaj^BB@Y~^B4g? z_+UpX#%nHco{n#}x5J>%rJY?kk!|i!J)qfIKl>04tZxkQ0`sEH#wN4W*N23s{12Vf z|FLxzAj7?8debFY9fgt54=UJ1Lu!t~8YYVK(`fJg+bszvNG9@yA>ySs=4V%rB+F*+ zKcg3&4ewxNTNLBVH4q9F2&bpN?RA#@XEB3EM5T;@bG-Q9PGsyDjWrgCLjsY(4XS-Y z^S;XN_SZ|;_KMvT(iJ9mj%Jhrcm+7SLu}0-(e&^gS@1^~zb9W)VH_WtqTRqe z=*fLX1_@fM$Kno*n4v_9g=BhLV*2tt>@3t zMBE2TY{{J6-eH}cnYnS~ebQWK^w_%bDx+q>N#Fy|XK!8#^#}Z?0IaASb@ysHbY(5z z1z}-N#&AYa@kq94dONvfKDFcMr1Hegeb>xO9X z^z165*eAzJD4C5Jd5w5vcdXNj!v&7ZJtBzz`u%rEK|4bc{QTb2M7Ej|BQ^qy2qNy^ zNB?^%6~A;ZJNSVagtio1Vi;t2pGlpPG-En&42YAxOW6G_Xb@H3{q55YZ{JPi5HL6U zim}vZ-z$()leHvzw7aGRF2ig%BgJ~R;0_Pe$GY)rCo?Q+eB+=gL89IF3zNeW2_WnW zyb4w`nZn1r5~rDi*ltAr_TFkTnw(EWKUvkX;*MHYvlJ?)V_?^t|r9@sxT!aRvw>I`qmn>BKx`EEOwgSENt9 zQ7pH1WKPFl(|VT2UNP=Yve$i9g+axEQ|_GoKfl+5j1m+<@TMBVQN14}otNN`3S`}t zHuUj2A_&g3_vAp`XZ!@@)0IZ)>F2Qh1$*gn`U_z;t(5l-{Obr$k@fV(R#F z<$_oVe>Ym$##$xIybb?$*D$nJ51;V2Na5G<-c3T(_e*hf+Dawa@Aw1}QuVHQ@8MeQ zR|`$wXx4;rw$$_m3?u^2VDN4a7wxyNWWeqqm)&>RPWluPQ%n8xt9TQ|(1S!@7s^Ji z2K-jDmb8>)w9RpabEGLWsVzJ4LY^YY@^9@t6$jXL44yJ_6vyV@`yk#=YirMIBV(bUNJ}RjV#o|3}1;CaV+;jaJvWN?|ImnBpB09vztK~b&)^;Ko(;*ku4z^SM+M8Uk|gyX|c zY7uIqy+M3YXk9@JqesrI0N>5W8-T}-yGy`@ylagKCsB&03HkEDjFK*~(B7jwzQ!#P zLb${I?|WA=QrtkZ6SExH%$i&mVSdz##_vOM(~TTw=rH9uV_ne6LlPTQ=2u(+rIogX z!M2v#^G|>DAm;X_9>pKN&exRBI86=Gra43HP+84fGu3A#Ua(WXis=$ObD6;M-_7u+ zK{yBGpIY2t9sTOcb>&!H`QoqpaZoDol|vuu2IT{Gf?k+I5{5P<@tAm9Kek`raCh z82M88-5mBUj98#rN|BsLb3;CXmvm1e_#eXadh*Hq%1UHR+d9A1{D!>>ok}HS+5l5C z%Pn>DI%MmkLzDEEwHW-{_w|KIN9W=cXwwX&f$H_reW1o|Eqc}bN!e?Tb5C|5lR}LW zn$wjv^XTVmGM;%n?xeHX)<1aQPZLKo89mY^wnLJNdJ9GaD^pKS*=QwZ@^IO_ujzwnRO6MkEqK5mt zOAD2VbScn=p6*ftv^?i_P*-C6_>&rF{KZft9UEn$9C#vvdhM48B;f<<`sg7%%QZ&% z(2H;cu)&IAQxzeP?HYfEUa_ZR314n8o^~PIiP)S?u~yY?yNS`M`opw}Sm2ce@;I&M zd2HLGl&q{pM7z3Lm~o4$kc|>8_=lNlx;wFBSUwh^U2@_DaM08Bz*29lCvg1z%6J2X zT!aYlO?;_&TuD&chd_Jh|Y3s|t8N$P!Q9!1X1hqNnB=ee2zphupnwchO_J zRihl%Qb+U#4f$FdxCy`TS5a}0h}F3{PM9$~P*)Gjry8^J;o(amaYePLLmhJ##0I(j zL@KWyFOTA6E%lqpZT~)SJ7C8yfjS6C?$xeoIgLxaP+s0KwHm!S7o6*xf-+g748VAu zO-h=s@+UR=SR0u-!z_JSBd5b&zewLbSD>9y8fg%>Nhcg5Ehu00@hs<>Qu`0BX}cy~ z&#>v})Rp{&gOm!pWmTJ6n@El)%n-JVwxKiOs24NUhCxRhBO5{o$mS@GMUf!F#_XA8fggb$vK-!puYa?rjr~dmO=oQOWJ7_lRsd z$f@U`bG=VRT(3CakiuZ~M14_*XS^C|DtJ+h^Y@iEcXGO&MnA;+jef8zc$cts@qXE4 z;&Dq*V83xg${?!aN_?2jJ@zNYr{XO;E^Jtb6q8z~cMR4(OwJ#KFPMJ9BRNxfg~Ron ztS;!UZhkLi>o2-l_I!B&+W6Rxv~lDd4qeqb6@QR(K?5?R@jyCe93l2!?J^f8V7Va; z{hwlIcZbZWa=ATE_M+eQ1!0@Y?wej>5_$>Mx>j?Qf4YFoc&%H_`N2W@#WXwnw+}q0 z`N-XO>+E#*ylbWE@D`QXiAZ@EP9w!|d)(8J$?kKw!PiZu_#5YmZ|)6FJX?EVgDbW` zqfw3w1~IcT(M6qsu&YaMuFeRx2cK>Vt&+zJOx|b^-yw^<-qt>-k^xVG{=@dq^2V^| zXQN;D+m+2ZJ{fcSKPL5ZI++QJ{d4z@CY;W{wf9zr;zvlR=V2+`@5c4+7TAxa8!Dzz z)Nv$vJkesZw@Nx5jKN4`no4!{;}DZGhjM8gl*^m-%9Y38oMPN z&}c?2@LIG`n9~WE2;zx=Td@@yPkH_+hypY_L9Y4}LWQS}ZGY8AIA613TKUMmwZbEGOiu$&vOvk=0|BRdaNz0=! zZrJDFy+4keKerX5GVg_!K!X-6OXrqUc?@fUX`;7lte){g!uVz-aG$keO|c$FWN^)^ zLw8!8p@Q9-K(3pl&c$OX?qM_NbycA#b|@d+;_jQ@QFS&qc)Dg%M(mRIcxAg)WhM$U zri85y&bdTi8{Dy`=e)VLtSvKesPoglvlj(j^EjrohsZTsEF24jbK^6d){gOHIBq4z z9nmm9Dgs3l2#kDQypVLs-Y=(9JLV*>11UVa`Z^T|cG_>usXzI^K(v-m8tVCHqx>Mx z_jT_?SRyp-e6L(*&g`@t0L~vh-z#NS5_nKG#Kd8O%fAc%7RA1{(za5~Y|QQKGof~$ z(!MY9UpIj@lkNC^?nW_H*~OL$>}e`|M@@mVqwFi37!+Goe~~#xSA(e1mZ)U)RkRmeI53$h++=99Z@*yz&k=$D1wx^Yv)vRfloDTyMqOQXRz>_yhFjlg5I^ zWqGvOB;@o0+2No5i^3`5_>|gIt?FBl%L!qC67B)m?OjzesXt-Sx=ZRy;Mf$;9=Etv zh`ef((DhRqTDrr#WI!)MloH@nc*}10B@c6iZ#fM~&wLhny=dpQM$E@?oeAp*?P!kH zUtV(3T>)xbTA>Rx@sE-V&6ad|>_`vogq8L5+lU6qOrNw@(vut2R}6+vX2QebMD%W; z;+7B8-jn6(w)19I`!Z#(RYY_zCWaLr(;i=t`5;$JmPr}+Puy)cOle30ZeI|jb_Jge%Y!Vf(d2jgmIBa# zrjUqh{6$zP=9sMz!7hEDteh4+G~k7~*}(-oD$b(8Ssx%&IFk@-2vTeWl!1ry{2DS& z?CG>BWZ>6gH2*zY9aX@>UGc4+8cfRnC$v1Ncl5_NoIAa7KqmB@$`=QJ^}cgc+DE~8 z(=LdoNCdM+lwYdin**=F#>3;KaoVFGr+VtwCRc>uEv zXX%5RF=T(09Zu?#0Vq>cH1G8!jZ@Qawc#uV`O`_CBN*UI!;z$w=I{S{&O7v-vDL z(I8jZ>FfkIBzHBPAB2ZexN8sguMShkJE4H+!%w`1Ma?kQp5vqU5o1|1fT= zO}Ndg8Pk0DG4Oe-?V!ec#=J9oX9zKW?9<{-3~+`2@h8(NLiw85r+do~(rJ{6eDC)C zN9h4wVy!_+S3^QcC;!5D+#mhqz9R*CH(Q8nEj`*&^*KR*8saG^u3-O289_L^j$^lz zgHPSrQ*kPDv%UB75Fj$4k&wepB)GY6Be&CvsZP&ci|y$QmMvABZGMX0e4Nlz!m&g| ze{9klM=E|W?4pC%oPW4*ywqzM9O@d;OL{v8U+2`aweTG*ZoA<2P z4+VL~GA|jGia8b`BuN4S-^Ii{B9S$Ao z;sNy_Q^OXJ2@b~Fx?e;{nJG)$^38bYLB;iNd?z;qR-JbMLavh+xYglI@sJ7 zW=@F}MO-)!sd~f>$*$RD-l0D=u)bJ%y7xo9%>IWd#SY!jO2{UH2lXP`O8h$1sy;e8 zdIP|4TK$=V3I7=(LSpoR8GlX^qNo}8DE9bs?0iD~F;COl8^spFO;vLOaSa8~E1ru} z{7IOD_MKbmnN{aivoZdNJ|mE{%%|Q;$I$US4j@B*Cv1wKsO0Y)t3O{r%HT^@lGJ0` zvl$m^se;he@7>bW!}MX93h^LAqBNRzn=4(?(#|}0De-)tA8g2ztTzr*zy8(G=(Zdu zmD?B&;8GVNpZb0g(s?LnDkQIyLTB<-2m`K^#;j8q7^|$r*UBEpK7F1g)N|m){tzDj zD4vi5F=bW0xXjBHuat-6Eq`cpePR2fr?Ex$k4{#ZUjeqXvTXtF@qDJs@e=bImhDRM{aTHSuEwQLcxj&n2S;!OA1|cgh%V#X+Vs~U2E2o~KNHlHIXNqL zp3kQ=`E^dhp0`l3d5T?zH8p$J#}hKH*!Jh90-W}2W>Tlr)keqqbiMKJ8c4$SN>yUu z6+*h6V!mk8=b>FXQi>?iBg8dbFX06~^HBW~qjYo1fFE=|&LyujB(Y*j`xgF`@`TNM9ZJUi!u7ZVX>3 zd+WoMv1X@HBGqa4$u-Qo#DylA_yM4d(&dec&JIWis8T=7PU*=lChbLX7(O6p2t?7z zFWC2a<8AmCunuRBfcBD+q>n1k6l&*$qSIJ8JePfI#pi!LIzd;IzQqIhT zU2pBElTgz|XJ5)~hbn?n{UAKSeHV;a83lI!I-9~T(c|#@hz1(RZ}a+>u8S)%1^Gge zPN^z;^q`Z~A>%^l6_3w0_FZPSsIh=wr$#xn=JUfK+%~T-3C-r7Y>$)saZ`iVZPt2; zPD=mih17`IjNc#@4-y{`H8@g&ncm??ypG-PrDb$LP~+w36$OK5+@b1_FfCKHrDzrT z&~C0X&=c-HZ@SN@2)v}N%MD;cr`sofquet7n6K(PQCgM&EEf+yKWcIJHq|emI*G-d zGFMn)9E`#V>MDR0jtjg+EVtBzV&_m@;iwC#9k6qa=Z8-hGBkE!m+E6!70{8xho}SAC(8;dWsff{N zL#jbAsc#&9979~$5tjlmFxLx|;@ENQ`OrT6-bgA`7DAq*?W-(98vmYZ_g@D*UhHT! z|M`)WGrV=WsH^HX@o(EW+-GFa%|!G_#nfTfcpn7y%fl40SNsbg?ekFw;-+hvMnVtJ zKXQtHd|RD}cKA%W&*!_rj2ea7CU_M4K)>w&n8cw=v5ta#PjCjS^I`^3b8{F*mOC}S zT~>^;YE~H13Yu#h>P^qJ;YQdCSyi1%+-8vjayxw)OmWI%F3))_=R_D8dBi<<(i)oA zH+`J;5J|bSdwIacGB6f**tC3H(`zY!EkClh5QPDQ!)H9#k7M9P^@0`nHAq@ZVKw8- zsxMYm(Q16aUjP7>xRu~m+>aTECXnSlqJg6)wQ4I0pGbnxS(GJ>WHOl%u01jLJB4Uy zF%+OPqD$RWc#EkLehC#F`k-3H>V}-c9pl^Mucw*V+ZZmhOCVBsIA`h({j-65Rius? zz0;)@21U$1dn@wGcvK)$vA@Ue({LTRwCj10f%7bDFDrQ^c9UZ8b+rrkY<#arz+-)b zsn&O|or;nDu$!;dg+I&2Zj(Z^{GvtMob;1uF}?fYt1|N)I(g>`ifWe>Kc+g-V4Apg z`>%ugV#T$3@8c|iu=lk_j9=HLGksu>Mf3Xb&ppTod)1k7)iLv*8fuox564_I_}V{4 zAJSHr(YgcJU4eLQW+haJQuMWRYjgefQOa#@8d8>QD~ov`6f_dyW2K;~C;`i*WMfVXj|7_j1i%wfpJm(&aHv0P*Y; z^mYkCyCW9XKpyK_p47n}J|)k!5fforGi-zh(Lzo3kY0@VE)2`ZcmyhY6$=-oN-sYK z+?t6Uh^N-@7_9%XMs1|n<&Me1B}5+khDb(5x0@>?Ym1S*DW zcQp2=GBtxSeT7vHHn_K1H^OKm-yMUGgUfd3)-ZT-Jxw@vKG5qUVz|W-Q4lc9ByA6%JE{&Be z?SoQ)r*m)5^B|%8_iHf9CMRApOH(o}`l!iUwPay6}5r;Pes{(Cb>4 z0XJKyw^}2wpN7Rr=6gR=eDXVAgUrieDa6i&@*mY|Ahc(Ayr>yik2t||Ls?7BNv)aX zF`DtcYs_4FhmJB>BE5b__FZT{M=)vXOiyq@?3&gz7p#%NueyXO+d^?U{#4ljVq*N# zDA)qc^b3E8Y>#==$nf>+xNh-S9x|t$Pv+G%Ez03ahko1!QLBA38Qj_BAlqZLB#U%T zkkVD4mHLSl&SUPN=gm8fPn7_Gv{2)aq!oqm zJc3~DP7uQ6*{{+Yuw(e>Y70)piOim0ilZGp97i-qX3@3Zi2>HfhC$d5WC_C4ekA8^ z>;9VEfwgt@Z>~-NAtPUiO+{e~;U3wr%aGDXi|f%2PxBD2V}xhAK+RZc^6Fz$>KHu- zouh(#FG-ahMm}Iu&Cg*9Q+*Z|*M9*N{xWO%bu#=?LO)(Zr~`}FP_#)c@%GK*WZ77Y0V?-0hqD(nk>2?j=q*Y3GGCR++k4y@q>rI$sdzt(v`kNWWxX1N&0II+Zo5L*bo~cH$%9FA4V=)D* z8^7oJ2^7}G_ZA&R`m0gtd%O5KRHduN>WaKhfBz|xa0HBsID7dT7!;$2N2W4XSe{MJ@hu`*{XCuDz16;RFEep_5k6tnQ4fIbbb8= z8&#Pjw{cuxh4%(|yd@wgP*&+gwrC@qx9-S}N4o_U|2Vkcncec2=A6*;tG7vW2v5ir zO>>W=5*L)*taw}9@tu9g7)^0B+rBsT2>v}UYR+1n4WtU_(JA4^%jKR}*D`ZhwpD&2 zj}c_*H|Oy9?hKYLmR=)oYWwNog869IwJ3Cn%T}PuM`5}d&-S2@b%kuU^rnQPLL9)? z;Q}@W!{@4pnH!`64;w4?2d`TJ>kf7p8oy`Kmuq5Bqp{7=l=lzk;OgY3RO(IjS>T&Z zA?pjmF4L1J-Y7OY3B0`y7Bk5(@`$dwm64F|#w`fPx!SGeZyX&R!Tw+r$W%E#T^d0u zf0oRYE8>E`7y{%hpGqEGRaj=v{%zJ6A2n@Dj%oebKZrXY_`|Fq=E$znN_;k{Es8VQ zg7Yc)Yk}EpcyQ%?jRzS8k%2#cc6(P6|C^rO9RTuDMeRr|aZBVP#i#!52QsW{rdzH8 zr3zoB_EzV6pNbt{)-j!DWMA1qT61F=|DjU9T${fg=C%;!v=6?+cwt$pe|ocf5JGi$ z|IyPOgFHoGFFbRVhzgx0{|UxqBXWshM3Oxu8wI0Q1=dfpy5>F}-NTlYqx;(8hx_^t zDzczdJ!Iw+Hz;E1N{~@2yE%sIhnrvcQ`gc>#=WL3N?&+BkCiCMkC@UkL>TQdC?er~rV=6t>iYH*ZjWlS}%CHArl3@a9U&&Bp|Z<5K#yHy$_F zq*mw2wwsg%S7p8oEs&0L>{c(Ydtmk$#$Nc^0j>JYX*}pjninNO@eR3=vKLa{Z;9}X zML{7bM|{PWw8t(s!+kh=+QY#2_aSBzXbvPsLS-t*t480f5036*Y7o_A5gh>Og5tUP zf)a^Sc}A|gGTgRtq_Lp(r12OR3_ z@K17KeGab%YUyF0*eZsIiN2JPd0?){lT5wIgv1PaZ4zhcCWBB%vcn|G4QePCj>Nav z1kG}y5&@D!z4}HynV~@wrIuU%5V<-IS%ls%cYsAjR@ViU9wRC3h(yScjpIeGKZ{lq z$$YRMHgXdhT&jw-M!)A2n<{^M*qL}9Az`63?a##1kCq)f06tX1YbtQ|2cyXWc7R~> z^A=gHGt&)1aPu%0S_V2tNnsV}4NBBFP408*QQn-p6_w9jgyDO1u0i^hE}W*_Q?~7M5Qdc*G)Rk+t;%h1`D)owEtrQ9Wu5 zJX*u|H2^Z>jp4sM0L~>TlSSY%@NFXDOG4nzqp7K83ETN2EM#wqdh6(X>&_0}#0!AO z0~r5PnJ;wRL0MuOf!$rAf@GC_IZ1D#1@i_2#=?G(ru+qW6Dpa!O_)uB|y$L08 z3Kd5l-;2(lWf=i5^RZ($>f$7E(fOK^x!10AoYHTjQz$T%U;={#G3gI7M8*d;nd8OW zUo+9k{LYa~DynCHvOKI1CuaHUbX{crZ9BUQEGv_vZ$5Z-=(p*d&2-gg>nbas;Vra3 z5At2#R3Z`XMv*T022+=oOo}GCRA#tw2zb48EmthEMf}^4sp4`(^(Bc|ArytV2X?>m z&BKal0two%9`pY2aA@4Js2}lZ6Vq8e#kR2jB(rZ5lUO{XQvEg|c~V*v&K*>vduw)m)7zF!|M(JtokT~l7mxaG}H)WN1d9VVh_3&<1=uFh^C4pC=^ zt$~PAYqTq%i&wBap?JlSSHbWV0O|}!$$y_q+ngmL>EJ%As+Yi|is=XB^-H9RMcTA* ze=mY3{f}VE9trsl9emzLx>_Jmy6AXbZSla|{D+txCf zd<}*>yi@N%ZeQf{k3@R-t3ZXGC$Pf+6kWDaY1;Yv_WvX0V!QGVm~zqn>zcZaJLk|1 z&4Kkh$Irs7mhpiwRqu?e{%>Rpzs3*kT3z53FZL8HM6s_0zv4dbb2+P-&HMnn24-%2 zq>}N7=Olw{{nZYSlV5)g#kZ^-REGSt%<@T9&Oz&Z%-a}2l%B^xNX4t&RpfiX2^{%Fb!gHY`xAEDWf_jZE|zyQvt&4Fkr z;Sl56R8DE~2x#v`CA>j_B>gg#HG~YYeOwheX9J&nQq9-?!un;VI2=KruzI|71RY|8 zyqoR4nyV{6f}sL|vt|Ug5L@bJ;sU*l5jl_X82)+i@W@{QXJAKCLNs8Jj_@qi8a;R) zv1<3}W?!2}g<@=T-K(Lczo?TAdDr+AP2n(kH!21{^D*@TdO;^KC3^K6zI4sY7uMR; z;qi9cpR0SzZoImGZ3EpmjK_w?EdMGB;ZoP}RnyO}822#ILalflQQ2y`>j2+WBpoF9 z%fcrva&>$GW!JDZ$2)gEF_tS11;Fm)2?(+rQH}n7F7NTTCcf;TJ(p(s*ihuEo1^y8 z>@p_)#!7snb=a$pjjB3JsTlV!=rtt$Sq_7&fU2Zd9gJLyrEDu)KJ(ov?A8;1w5CN^ zqB+02BaHH2Cg~(wj7z>-1I#VCSxpccFBCVA%67ps+8VZb|)o1sY|rw^o6!lt3o@%>Y4h5y7HZ+QF2I z28qvEj-s@KlMCx?Gs=6{QpsKgdVSvZt%ZaCRoZ#SHL<3BUqPjVolZmqM37LV zmmEPUf>NX^5R@W4^ePDgQlu$WKzb1brAZYC(u;HfsiB2l1B3)p-qAgKcK6w{`|PtX zAO2vHnPHN-%Qe^a{r&FG{Y=W5)mw<^M)K8$#ZCC~E+!|k;sRm85yEg%HXM)T%Ro5W zj0Yc(_w)+;!d_2WkwVv)uG}@wNp@(}x@XbaD)~1}3|wFp+`GT`#r&x$Oj~9<7A$X^@31ij7 zsPP@>+maS9-;FWRwSB#`WT>>E`6Eg3ntklU_sh)Grc|AEXTE^7zSDdFO6lxHs_s=l zWxS&YIls*QfpLnf0>=+Kz4ZX&^@Z3?p51{RV=kzZEv7Sw-nzJI2u5FgnT)d%>X?Y( zeLE&kpfp5L?5&e8aDn~4VSD8$554j06A)uypG5J+s3AtkQt}(EF%r!2I>w?g@#5+p z`A1?!+z@={=>zuh(K@WwElAbiI|LfBgBwHcr;j&0i}68&4Lo0>dg?|+v6^w>yO*0=A}H1vJk`UH&|h*9MdJ$rM#@ZTwS#+`mw?r<=PrJP!z5t{95qUfoP zgYwwp3_lDkk4Gh*&3)Zvq!=m7CGyH>tjb2F($-#cuANyhgDvug*1q{_t$_aZ{i$n@ zv(oJr-iDzvnuW~}7Zraa@Kigc^uvh-z#Tz}!YgUls~%4;?utBNV%TOR6oFeqPxZ$z zC7j0IzT-m9G_Q;1PdF4rVkUk00uXHwSZ0?3GuiUv>?-{)(ZnHy+(RZNRT5(y5j&MM zXcUl(R3bm?WiC6?a)206VhEmASLD7^lM%qQMIhvkeceNNPxH^=*9~DG77k}r@|M9N z=UvjLR?c&0kclV&lHQQE$N9q_9?NVWBgfeflsLV{@P;I+I0SLy@LN+8m-zC}JCU)r z`t(?}B_<|s2awx{By2%G9+5dPq~Y*Grs%RAfP*i7@FBUg{Ly3XwTBLOM$?k9^2pA2JG%87ufq!-!L=r2EiNmmvPzs@Se9%?J-&RgOoF*XldTp1gM^t?GMZ%7- z+>1T?{9;Ahn5GWU9pD-5i_5@6(s3*q!bSEP-t!XI>_2>OI7z=Z5wk(d^%0jurjs{2 z^KCw3zU|ot!EE#+RuGxs5rme8y(g_ zALy)?L3damY!e5H;W&W`+PQ6F)6e9`w>2;+AoAktl0wN>U-lvX7HkZwe!DS<^Oo0_ zJq&d4?^hOHlW}p+ucg>67XevhL(}9=)PrLvejAzV>i%w|3MgdIfnEo`$eZuLt2Qo% zQuU>PwceAFrXUASkK?hLDqmpWEt&U)E_x6K)`G=0_rS+P?t4S~DC>SQbNai7j^JS| zh#DE?9F5-;2TK6-bOt_jZ>2ta%H|l|Hb13&h+Z!s@dBwxMKkRoh6_j-O=o$22msQGl$!stCJGGS$zYuqGeCCYKHY zF<#sO83ly4L9Ix5pPyx^|F~bh`&(d;1~HId7&baq72~*x5RK21WEd!Jk2I**{m+Ul z?zOqogEOt05j@l1cg>xp{HFH(-Z4(UJ-5shSUR5SdE{Km#8|jpYx9tFM+F^EelfR= zB_jmkC{p8lGkU3g%a;MA2OX)1gDXY{(-5yBU!vJT_U^)+F^IH{8&mF%h>Pv<*>WaM zBP^aWj}v~pQ_ghj*!)9x#-!Jl}i>x{j-ytp_GdKpLLp$zCI)$ODI=&4E zYDr>BuC(+xRp2^23kHpF!yorH4dIO*p&;U}-YopFsa9?8+qZSN5B zODUB}_+7tMh2ify0n}~XI@waWE5G!D=)nh?6JoF3C-6p}(OU;yX3 z0ZF=h9g(67XYJ9B<8JHjS4JbZ)wr^+~0MaG;<8N5>=s|EcHSzubVtRdjh!mQ$F_GKvnL{4uxuX6Q}Qn?gX` zJ7$d9Gufh(e#id*rj__e(ACFV0m_Wu@C)f0!P>O(klY05hta)B;92#} zId5jRI~8F8@yx@4iVUMozAOSfN!|2T&$YkO?@9np5?5!Fs187nHMRP6Nm=9=wrNMr0kn$Z7Lg3+(J}SZj=BE%0DPkJ=l$b~p-6sEb#B3a8eo6J% zH~1>A+k$n&E-%1{rt+Q!x`>qRAE5eQaPi-d3IGJJarxh3R|DjEk}hTvx>30T6~DvT z|K_l!uv_&5CwJW=?2M{@LdLmg<;G0-`|S)#qPry~wcY#s*p!(Wu8QrPm~ukbQjAVk z2|=h{4hNg5vvQiLvD%ud<*Afw24xG$Wi}xeRD!r&A_+WRk+t5t*)lF`xwY=Ax_2)y zsbSpcnV3>5p7p1=E%znM>DQZ*1d~eMKD;zGVsAK~)f(V@Ji58&RjVHI(7WUf)npXA zzPDhm?H7;XsYZXfqReneY&3}=rG&q$hre(gLYDx=1HSKwBL3jLQ0tF2nP#eGk4vu@Mf||bPN4Aw| zS&FRY5-;HlxJkdn@y8W!2SV_4`8_iwEWI zt3!HOG#8p6!a!pX)4Ys8kSXfS6lah~K?02(?#V@2cU-CQ7tfE$684F;3eN?tql3uU z%_nfu_`Tezc=}l3gFM+5D2OP^{&JmZE4$P{*E<1wu*tfHy1D&< zUI!8ZzyH*eRMD^b{+Gs+NiOk$)jK2-nZb4WSaAk~al6<_d$svGJ`YGthCsd@U2`E5)$zrA?qW^ zN_#VY+sTJB{z>ZkzZXB{6+~$%d2L+n^l`k_jlZV?KZ`V&zOuth{CKdwpnWEk zE(3aNyZ$)ZxZaXx`nOR-St{DZCpEJ=oFivSUIMFkqyQf#xow25{-A^r z23ByXph591xhj}cwbS$mM=~&s^e%lc?MWOzl607|d5|t|D-11+_M4Wi@Tox9K5&_D zb&*YfNynU$rvVj*A&HW;JSveyg2pf0I}IH=cbX}QxZ~|o{s6OFO87!`W}~IHE^!VW zXiwQOZm6)~G|lCiQm@B0+G02R1T{9MJl1z$|4myLjWAKr`+AyGvD{xF|Ijo^?gN!| zhH>Ll9}kXQGR3Ec9bsiIu40awXlM#UNlN1y8Hghz-%?U-Jl6yRa{w`;(u)LRLP>&= z7}_*;`ERO|B>AmDE(p{5FZvxAyZ7{>&l{>6s1*K;fFN4v>n;g@|5j2=4mpB^5UIBW z*liXNLLFt%w;^+87&QB1V0!~<)9^85LlG8VkKuakg#CB& zqc7O|iQ-EB72_4phFv&h-NP)?LH^?V5FIdrs?PP`(a0f%VF@b-hr5LFRENvub-~)fz=Y4d$r68~rf>{k<_v&pzVWJ}^~pFmoVE`$s2UWHbZxuW;@0<$_`UHC zR?q5LZ$7iil(e(*c#TWWT#NXQ(#pqX5zkXz^*8qIYSib^=h9{T^>Oz+l$HH!Zns!0 zA4$+pt4$)d2r}X^jo6&6^fhVga}c5LA{1GK`MVLutP!`z_;xI$X+qx4F42&_HUupw zk}Xb_gTLbq)$6zm55`kqorQUw(CN>hg>k@m==_Ym9xDk27ICBax|Bm_31!z zbTmkA>y||!k7~4bvd`WUuhcfMFmo(NI~z^G&9G0f{X>Fs3v@BU#KbW&csGirNji&4d@pSiR5k$2~J*nH=B^d z5f(FD&0B~I=EVX0+pAhy581ZqRuFM>R__SMhub-d2GieZze~Bbgi2%e{KemupnDRT z@lXVVN__g7f9GMZ%D2W>blVV(Bw;#Vjw(>t;ivfZ#lpD7n+^UQo^HF@M8};h4gxSp z3u3A#No9dR$jFg~E@8FPD=_1*apLVhm37Y8^mO>0{44@1B$BZPPror^ap}gaN^nwL ziz?$Y0$gYju-kZd?diw|{rJN&_P{`~nl}s96qH?!Ut_+N+TY?UOVI$A`zMe(vWTNQ zL+g<@{H^znQ@hIjj)t|=7lvhy>-Uhs#I@1s|EbHb0lci5A4*Z0&*R7`OJeKqh{E~2Ub~KceZ5l~M zZayzU)@F|$>m`^Pnx-!Ir3YEhBnF;RejzPfR~u1^&r*qnw2mu=*I_N@C0M3Q$7PIN z;h!&Ul9T)(t-ID=r zVTR{lZ-y5ygdV(tXiAl0F=HVEal!{uJ0qKJP$Njuso{7=ZizOejT(TXJ=SN$D8HO3 z0Mxhx$d-<}gdW4UoL@W#JFaJ@VVUCIyE;IQuq#^rH81@ZlLkK!Q>74UL>YR#z%CR= z2uzJB>7cKO8E~f`sbxEB*3C6pZLt=jxQCrAgPqYj9y2qQmZ79PN>POb!b z-9J8=!k9gzumw3WnyG(0DMOZgH?}hi@xk&6_trT6lRVx6#cVd90_kE?I<~EpDm!z;68R-1Fe2 zA++zKy z6@bO!j+wE;HAeQ{;rOHy6Um+=@cZ=>lY%8RJ_bj{RB_L>UmymvU*-mJJ303UbA3{M z=Otfn-MhwP!7dmd!mx>S-+elpq>MUI_zQqPO*-u~G#8{yWjY#$-`}^LSM5-4-dpg! zE2H?_apVSl+?qHQDr&O!AbBN!oq-Km-$`{MpKZDXwE!cD>9?_JCF*Tq|1E78Iae-0Q!cJqxzae*bb_g5b-Mrjbs?NTGUEC(U$ zt2v}w279}|`L8_7+^%-Oi|vh0RO*FXzUYiT_w<2$2{fiSEBO9a(DzV~H)T6+5~?zi zsKuxaxd?Jk{!#p>A!PzlYG)q~2ympSybo<}5rsc;SX^zamr>@f_RJoCO%jLN{G506n0NHmKx4wnimQOHgXSd(DR`K>R zdP8Pi|Ami-h=7nnK+*acH==XX&OD+bYGga8)^E0!0><-=6usWUPFKDl-n)6I9OX=8 zV1_G)3zg&(InqIKNApLA7Jdy|16kbj&_;UUe6%b<=+Nk8SCGN=51%04z>8mdxIVeN z5TEqr>gJgQqsRwiW;t#yi)uu}*y0qO&rl^y(&BmJbc_@)P4~8!0PyW4Ag?ENV&ahk z>OkD7;R4=0){hbi4lS`e1*4;1P@Os6hDj!s7{!?Ed(098NmL;XJNF!sKSu8g66iT| zW6BGd)0=jRZ<7k+-aq9I-S|Pf*%A66{?M&n^Ukq?<;-JyR5s~-#_-UxMf5!mV7avG zyCB8qVH4Z$VK$+l?fcg&BKk9oZ6LpKffI!3!Mtz*yHHRk#tsMAa%}6m2SVv3V#m9< zOT8O9=~^0dTiD{1x)YEbQcE6ETBnLX3aETwUw`Ds!}3EQB3Z(6cQeZ+oD25JR0Hf; zv+DgE%eoZ2%H1Q~e*GyUW6j)?a-Co`Af%f4A-Ydp<9WcA@bx}%=_ZR=_a@SXSckVk zsk&2|A>j?ikEd%!>&w~`Ha&!Wo5DCL^f@Qf6?QVu6|8Zwm4_rrcl3m@auC8Mm|B{! zjw7POyB3qT<8V$h^~r0kCg4>VH&c~Bi4dl_6uhe0v<>CAYeSw5Nm>J*D-0x>VU>%O zywql{wBEFAGp*->mM*om_vh zByG0ACjU(kn|@cmf{nqD_eqz?Q3wQKq>k%SSoywqKBz-GuFZ zvw%9J*b(@0s0(EUN9 z?KRN1TZgHUb@C=1bEq916ckn;PHWG{3cGysX6JD;@%6RGiSM!am1W+ z;3bSY%BFcyr;0pe#bUN`uv`)vZi8%{gu!UdZW{2GZ%kHa0QdNF++rdwsgcBHs!GMk zL0>!5a2TKZSPio<7hQGm<%w?|*00!%riF*kZ_;oPtS&|k59fokM``L-TJ zUX-8r=V`pR%#Eqx?NIjRXEBsBUO^Gt|=@Kc^4LOLH09TmkB zMi=htrHg)kk*gMgF|NW;sml>kQHU6)&!Z=vhF&_ILtVx7%wZ*B<~o;1n7MEgV3Dz= zLdC@RW=n3n@jgDCJn%a0@oAMe79BCE7do^C#=IQ(qe0>wG5V_))H1GUF~_%@howUq zB`8F4Y4Apuy;cUezto)<2`tB+r0{B=te(Fi$#N9v`5aTWzfs1*CVOYILz@TDJ>z%i zZ#O(qJvw2LzYxhj;YRORpd=<4g&`!pzG$nX*t2y70@pI(-%+D62T~kq`BqIR|7BCx ztYE2#wD1P7pSf^umWuz<%9|Wk2pj>+{NgI{Kz_aaF>t@zr1!u1*56Xbj8%em#+@13 zb{Fj1^rd8+q~p&4P2n4UbRjQt3<|xsyc9=CMZxmk=}eN&_H_pMA?m-ba>{5u*2oj(u5A+?_f z0aEE$zu!2wLAT=HzmJjM)1aQ|S3q<~tB zko~!0+wA+3#1TQj`)AF{*t@m!O$m~zpZ0CvQ)Nw3vB0YCbv)9 zKI;2W>c8s<<;$48TQBHom40f15Gg%-3KFtEHzoA=$kzb$%m~Rv zPF3*C%VQ&zRiQ!wK}6E%&4Yvtb*p^fF&|-2!mxK$R|{Wjv?d6N>(*pKBVl+^uLw}A zf7eap4VTYraRWPa(Qqw7h9!h0x+q{rN&(HsRb(oF5! zh7!%8sPl+N>d6@``SxN8EQqr;zu(6^k`W$s@73tV;&atex$^cJK>q&=A(U~Pozkgb z$UmG?qh*k8e`CDMva(=fWpE+>aI~(v6BCvCVlkATy205{Y9a>mm3~&nnR1?4&t&8} zr`@82jqTbRw}|_WcIRGght|8yB)|GnjK0GeNb~cXEN^LWDL>JQl zX4&8Ad>z|tii%vn_yPl8eF#05S->LMX25r-iKnQHcuA^$_|wyZQRypqx7&UFIy_G7 z0Qe7Bz9eD7*jy(6P95+I-UqsFo_w~7W`I=L_Omwqs11%NmLW*dn)1d}rQj#(HQXkZ z^mC4Kf2pvt^Jmw>Lz0Y61$;eavMti0A-97btOkmFK}=PawHA($fYY1?U&R|`B{ezT zF>W0qYG`U^MuNwY$9(bc7fyTEJe;4J{7_1XK+4t13p>yCOgVYTu_~R*b zOj&#R{p9ZLX6wrH@sM`5o>9v!RCP?EZ#;T!Ulu^b1hvYFG&s|#Jt8kaHmVKCR& zcvOj#?J{4MgWK%v$yoK42pE|v6_QDsvY`jp6WJwwVv_bL(|bOtw%|Lye3QjYZ0B&# zkxZ{GRv{GdZIE)pGrtqxo) zx%>NN2FsF~wRxzn7WF!Iib__Ld9yRPjbtJ?=EG58(olZNh-qno)Aslj@g z3I4tt5u(KltT#w+fP_|*`2H;Vf=c4yj9-7TQR9@bv1yk%Hx`(X6r5yi2c=o$o#mY? zTY9$F^ox2;Qo=6AZS?WB&|$JVCSxvHG{w~IHW@*u2{v6bHyrub4&v5j_1Zp}G(J6I zHpdtW2@=bV4@V#GHwAFvxP!hvyS1E=qu2pBCzULUsif$!30_}%?#U&nGrkw&$0f=H${eKtXx zPEjEE>P8t6X68Cdeb!K~6hpwv3b#5b)2%-F*>fHsXZ)q4Voi`Pz3_nbCoNe~67`vo ziue$NWrk9>`X5;;l^qP9u?B$ zz5?+93s|4zal6U9)a27UwPV72YX@L_oosw>w`KQYnFpI-b7Jwx+j(VyAsoY(5iE9# zL#(9 z6rnO%yJ|m0Cf*g+w=<&nMZoe!gDw=#wMXwU#eg8K^mZ(dPE&I9wXB!iw7RQmW2*ow z1quq+<8z%25S~PI)sNw|HJ=h;Ki1~X2GH_woAW!t!!b+#VF79Ty`l;%}hE-ztHd-a0mvcl`gODHs9n&t+70w5I*xc zoomD=aFBj^nzFK~$f^x8Z?jYP?bY z6cgtcS6)S2PDofd)}lZMS|X9i5FukGPsVcYtgPi&G42uc%I!vO$R-;5SA4^udpe!N z9F0euUV)3ME=dM+cT;i(j7{CKE741Pcfc+=5Mnhjs91dBwNRRq5aj-hT#u)df!;y0 z$ED?qwk+}Gc&N!KL0>#-t;Su|!{&(i0&uN=ltJeEn^V0gnmN$-e{8y^AgKJq%KQbs zwN*P2eJgwqZb;pB%2+KqvY!%U(Kr7=E?UXauH{<$a?`l{2lixk)edZ6tmV1(vuzTs`Y$ zE5z4ah9EbZe|yyUG;cGtmIPjl*h3n<^W3v~N;xLBCMTS8q|Ai6>0E}Jt2Yqz2_PI{{`4bJ zfiBO%k)p8y2Jr`kM<2wW3K1^4x^P()>ZYAbn}zsci;XK?^}TD%Tkf>L8fzs@U0;s` zbd5foQ?U}Mwy!`I~T4N?@lWZg(w zvCewJnu)`YnzFSW?;Pzu1mVZ72ovLoDUnOWAT2Y?z3@|qYS8HGnzteTWwLu~cUJ?I zPI6a=Kt8JCmFu+|fb#N1z_^b~;Md3fn;x2u4^Q$MN-#Z{Phv| zW5`)YlxjMoHhd8k8WG`+Mso3x4K|(88Zv(Fet?gp$Suklj>bMYFO_ejA0Mw+OZj0n zx`OMuE_~hA8GTE@`o&A6;CHXI)C)61kmg&?X=i@A@h?r``-9i5IhDo9men9-m!0E| zP^(v6v!VPCOcXL3uuO)WIc^zg5zxH$uNm12@J0Tg)6N375HM~vA05=uDcII3@kYWT;Lvk98 z*Yt1dv|MSvxVf{3mCE{c(s`#Pz)4o%;+sV3kuPeqd#`OzqQSh6EbnFpvpw#dc}|NVZF z&t+_>isJ0}%8N%0!j3Mozo~rxbH3>2(grHXeJMCdoO`LymYo;m?1B&y<5B+Q0I!2y)`nk3G~xR* zs(UiMB7SmCa6bFoKTPBQ&d$C&Lp32|0lowzRoQwa)K(~dVaQC3h^P(?uKp)(sUBBz zJ*U3{PTlFsqd(~X#Q**O4=?nr#s|&iHBvxC>4&KY>vTO_o0jur4|su)%(3&1*j4R6 zH&WOAY5Rr>(Y(e(@rRu`99_+Qp)2gSG>)I#)-Nl#zn)@?krXPHmzbc^sjM|5Ha9e* zz^{bw>CzMp!n_wyPreq9YA?4-K)(BSQ5*pLV&m17@Qy-707m(L+GR@&F8RIS6PpgV z>iv+59Vv_H%Kuk>uz+B48R$m$0HjVCI0s-6zBvOJcDtT`CDe^o@8zkdIJu~9%m_Tv zJ?PHv24TDVf8`2AWlzWv zC$AXAwo;JI<#VqwPah4uznv9{3H^NwZ>)6YrC2v8+xX6J-T%ThIETQUa`#kj{aXWg z;_LtX;B79_%5JlobZgdS9qD>{8rY40$pW52_gkC?$~DQ~?^b9A0$?lPyV(IWuWhkE zAjEh`%s*T4)=c7h-0C|qVw|s^Eyhx)?2KjpC&b@>2?_>aywV^Nf6fZR(ON^vDzs>3 z9kf;^K}Zk&i&(`Jdjj=_YiYK*+VFGwq*LAfO2Bp}i~;=e=%G%#h^QK)&`Zo%E^VDf z2>x-A~oI)S|#LE7STjxI*bN=)P{TEP|rxfO&n5gdlhcfp}cz;nwz^o^q zkkhLFJ0bJmy~T7DvghrhL`&O!`+LdnIa;@#fl0mlS zS++ML`mfdga0R3Q?%}_K4!Zl}{*vN)Z)#w8`|2CppP-T@;7k#Zd;$6Mvi{%M;r&TL z`llNoUZI)^aNGdwq((_^7`ST+3KbAqW2tuC-cWdt?z9-yQ2fSHLY0=MI(BN*Myo-k2JO8sZ z39Qm0h|m#!gBCzM)MfnBB4r6PF#Yl$$3|l`!h_-k5i(|AU7#lbZfNMF^Hc(cB-eVg zTtauohK4NAwitCk0eF0a8n6V;zb<%ngV%vMypWK`fqv#E{AV?ts^wVvpScEb-btpb zfVJ(pglF`QR{#f|=ZPN)x3dXOW7>_Ues(`FudC%a5DRo~dorLkfK;R*N%yMipY3Ei zi@4sueBMBr?Nn(?uegHB4*z=P6SF&O0uW{mfAllmw`yK7{xgV2nMay&>*fT5uO_wY&M$dh9jvF%a_Z(Sd=-V>Y+x54Eh zuhVwth&hK zd0_SZv3y7qCsAC$Up*TsNj$l=lA*q;5xY%@NmNy$9${{eEa BAWi@P literal 0 HcmV?d00001 diff --git a/images_ml_frameworks/image7.png b/images_ml_frameworks/image7.png new file mode 100644 index 0000000000000000000000000000000000000000..b8a50fecd3923224736155510f85aba98a586db1 GIT binary patch literal 100282 zcmcG$cRXBQ+vptxQDP)|lp#g$M2Rv)M506wqKw{4^fru6B1n{|A<;t+C5SSkPof49 z-7ur~Q3r#w^Zf4job#OLxj*Op^ZvQ7wfF3`)?VvcSNUG^O6Qr{b;?_mSFT*S{`AQs zy(?FUyRKXz`hATIkSMq)MqjxScID|KWrF~V&DImRS$Ep;B|&08D^DD4E509+o4N86 zCQ)@)+di4XHkq3CW}DR2e)mWs5oU?`gaNYYfbI#)=QdJ(XT%nZA?dazE`3wSvoF4E z!8UT5d_CNb%kKH2pEqrgteg6-&jk5D_1f?`{9JY^CPns~Wm#;D8;y#OX>U%<3^5UG zq6VS-Vc(tSsUxv+iVDpCUZl+m_wk{RiHZ5x+{}6rD|)Np=c9OPxN5HirENx}UjYPV zw2=i|lRc?;fUpof^M-w^{JCpn8}_>lyFps}e`!W$oD|p;{2CqYB(~N-);MVfSu+|z zADjO0ey1v>5sB5XLp4nKvbZIJ4e!Y!0(6$}dXu;cKl8`gST)0^i6R^ZvIv93cUAqg z-fInf+FJav*-^vrx|1i721V2CXbT!?Sp-%fNeYY_7s~Z%#!mMwa03Hh8eYcYo4Qln z{b9kfGm;FU{{Bp$xFz14W6rV9%t}7m+1<3W%p+upJNO`QVwC|q`VR&_Qs64KhzCSX zW9it9kVS4}Gpj!L4-DjL_wHGRo;%A*Xu%{dBpw?Kx^X=HA9vQj=KIg0%kjaIREmAY z$=kDK7BF1UlY(E&7gfi7Rm(z6dDuStSv2lm`5Z6su*bpwMF>fQiNqSOix+m%dXMOB zr}r>3v!QXm!5@K3@s}~}ICf|QIi(nn5Q)RTZzv{5HfvZIFsU+}Fxd>dOOamG`;#Q+ z-(p8MKW?5}#4Gdd&6Pez9ajQvvY|S!Tcpc|ajhpHx*Z0N3|r#CFP`{i$~#42W?3$sM9{EL}~_*C%)Bb6PgB8+CHP}99qImGi{ z=0aYC`C;x_)dm@at`}^8Ru~21a1%H6iN~_uLyc1}{BO;-_8sDUe-3H#o92FfT`gmL z7edrrg(Wva9rwN8W{1A-ZoiF6r;>uAK$VF*=|Zz|2*LPB)kF~{S41M%47;>%&MkOd zYEmIT@6p$Vn)km16NYf-u}>xy*r2UtUNF?LdSaEN3<68O^!3c{Us3{67yb-y&m|OP|}vCZWFXq#s&$f zn_`1HKYg57l?qIINIV}f`U)vj`itauSb=+}yn4bmho?`i$j>p6soWeR#vxq~QTj>V zpSXd1Y@ZkG^CmuA=;n?8USJI)a`FNm=$`EX53#p=GW}(?>Tw0k3sv%u#P2sV|u^+5xT_6Q^&+c}M&AYdP zf%xvP%^jP(!^Njj=mmV@P*w6%5l70}0$>)WoNIt#oEVC)k!o4_rDDsCOtszvnhdGV z-TINQKhXpG4VwBYG8Z%c?j2VT$2~+@0IKS2l@U-;t+u>{%F;Z2pnCKD7Yuc}sT^r@J&QDw zR@v?nOZd65D0NlTF_;~ijFkfxQbJ9!2>j(s+6NCFd@L)Yd>yax@#9C@N(M8VabjLU z!LSII&E8&PWX}|(i7D3!eT~#nwBql$vAc*{vi4KoW3+nXXI;tgxe!O7U4vZ^v-5a7~KXQ z(cxiW*C>N9quJ@fX??+-?YVDmZbm)2O=MJQ7E|jm)bUxrq_ArAB|jt&YNC+mdNOCo zzfxO2y2+2o*s2@%n{wfSuW!)j$a<3ro3~tDvMc)48)2REi6%{~{&ym}7-8DLYgzS0 zQz<_J8pw|1PxT^guK!lQX5MCie@&f!kOY0VkjxGvbmf=VlBjEFEMU5+4h&6AsTxk* zPZaMK?EK!n;~vFf+8gzHjEXVxYMa{F{nA&-!)aCh`_*o{M{dlvJGrH-&Deu7?G8g9F=CR z#3kB!;od8Q)k=rNW{o~Kgw0y6r3;(chIgoz*<1<65tGs6hYnZmi=PH<&83&>=Zzc^ zPWA;~Dqc?)Hn(mMV0iNtsS%E0M^+BMvnnk^P~?SN$Zirgz41Bbkm;Yjc{@=tf7qi-B3XuKCu@6mx}!jP^rcUic9pjtzzsHhe?LH=#8(*!KL zU#Gp?Z6hj0m=w*v<3IqjRc#5z=zOJbi9Wry=I3G_{9?V^OS^@DFb&bs(<5e@pn9gO zd*cl?C|D?E1#M5{r>{@P_8=fbv7_U3BI9__44WNtHAV-_HS1nMRX|CzHI0kR*ZoyE zUEzyL=RfWCO2VG07JPi#V7iegVt5e0lzV`qdv!I(s$Fz&^ZEF1C665;M&`qs@`LUR zrFgF-k5L%xOsFwBM92K#-g-;x_wTPFb|f60btbS%PoNt(-x^Tm=OS<$Ow{;94p~az z(_BX^_v8A!VWz(jq31!Ad!8`ATno@jIjJs)`jg_rCiC*0>1MD+XduN|kZhx{74-64 z9M($=d+2m*_8m644`+EFVAK?AZd7G){qXQGd|!s3nFQV4SYMxy1l4!Oiy*EgDV#FP zU*K(aPgy-dQ2Ep~jOMr4p5VzZ*#>V)aBJb` zsS9>2k>m5Gse))%8fv`feX_M?!~`kT1iH5*FZ7m0@Ya3&d?(0iJiU2-f5^W>&Sfr~ zbbqCDb26#py^!fVzS>4dSC{CVgyb>2sRvsUOpsGNy=Fg{K_Ov37?pW3PTf>#-u(PE z1@Q;@z~r+b>*Im>aTAF^@Q~t};LMk@+elk{CI@ofXtssIW&F!0X-R8BJd;RF%WU(Z z>66dPre7>{Ul=$%LF!hWiL+|PtJ>hNEQ zR@qWnnTwtY>HyacG8Sc+`H1GHx}4XadUV0Z+hXzKf+ka$`RUH1fAty*yBx1DCl?V4 zq&V81U1>FGn;GSj@jjw%*&mb>br@vh6{)(yI#6dUNrQI$r+*!XH^{myIOSK?WJTc% zFylr`F49XmzfGOIB%`i6CReB^oM_+tiby}4h$9exgYs@(h#!Z6;k}4p`Dh%2oh9L6 zhDgg{YsFW~Vd8+)*6K`!V5u+0)xw-is3A|VK2a$CVqMxi3 zC|7sOhXO|yDt#j+mLE0#Iq&!*Bel4#O9vTcUFnHiJhW zYBff)6=%!~ijj3OEG;=<$$dLj#;)SQA$277gIUV8&FIkDY&~-4jg5(Kj8oIC-Ql;8 zmw$wM@2f!x9np>%!r{t?Ee?6nWslN!`5X;xz zQeFLS&rQtt)0^`S>~k@^n7XMN>zg+LmjU+_izVlsP5RGT{kCqU)(Ffc`I+P8PtVAi zn-g4 z80&;Fcr&yHVB(?dm5VK7C6$8DL#+&o)>puMD5I)mHb|j8?n515SE+ju-ez_Ci>q7q zMm119@}>iw+D_sde>IoaQr)_B>tk_o%*FG04FAkSA(n-wW-CK^cb0u$+_AaPVyM9# zVp19Q&QW0jI1L${?Ijyz(|kB zAnjR=ju7|FNjGg~_$!amX0Ybs<#M)%<;L00s2mAQ?d8&xAtT$7&c}G7{iVM2h57c# zR1L4qiR0aFCTLsmPOI8U3!chA-@G$Ic21V80Y9v3V!{IRKL|eRt0$Rr2!RuDsIPEz zC?oIljtUF2??3N%5kID8t1@dW-LW(s7J5wn{IC~fhwX_l>RlaWY ziL9MHL}$0iH~OsIo{dGyKV;uxKM-RHh1PpaKgOPK)(%aGjIl6o{`qx$!@sUkA%Hc@ z^uAxeNOxKfzOQuo!=$a#w~Qy)H&Ef9=25;VyT4a}r(D?N>2sfsl%X~aNrxSb?^s($O+T471G-l^iAl=|GFvBue4M|R;R znuQd*OTryZDix*;jPH4g#C~zzbr{SkN~JJ!HAbyd#YHh8Aq^sL*CJZ zt*J}INlRwK)#F+yf%sxC9r=sDi-}1n zqIsu@b-VeJ=(Ke4IXG{jD=G1&q2X8CpEnyX!&yW<=Dff9td;NfOANK`z%PS)pSF)R zo7vD`#mK!=vxjI>Zw8<2+3oIjf;ve+GeZ&m<~kw!+@xLE^{(NN3tZCfXx;@qhczM| zYDP`P)Oroxa-6hV0=dk*Igg8e)Ia|?0X#N2d1H~b7KaT$87sX(_RE>H1&y0ub|a)~ zt=7t`Is~d(h>wRdl+opP)Do4v=`MM%vjhh3Ia8&!&V~ z$ase-6G*<~#a`}-elhzVfzX!OW|07YTp5|06ISRvI468}_5TSRS?Mmtu7cSE!qet| zU7@8cKGr__+CIN@V*`{vcUD$J*uU@DPBVn?_)(kia| zMs~q}YLZu`gKAF7Yi94cobGO^Nso4qyz;2=b*6OSLPCxdIWnf6c`xz)p^XBkuZ?o_U(I7roD-XMbkA81ddVu zWHIII=~lzS*$%QKsl488Wg0#&H@L%01&Nqyagqq;Kt{takKqoR>@OErt<>>zgSnRq z$AlZXp{IODNp&pl=p)9Nggd=lia&`ldFK(KnK()MoP@daE%%JR+)FAEc@Q{smng?p zv9au6)^}&58NxQbCQ&j;)$Yd3!4i`u@Vuj+rFMR+!KgF*{LI8Mm<@5dE^xZKSkW+- zuhoBccD69{egD8ZqTFY^wK;L_8k+;cWxTRL(LWSN8gjXdD!sU?f3^=wnMnYAZo)hU z@7JuCk;(?)wM~Ya=>7!#k_mZzksHT2b4urbn^H3HqHX?)R_eU9JpD!cv41A7pf%j6 zKO_!&3&Qw2w`Y|Oi`nRwG>L`_3_)c;AI#Z*KM<$z zG3KQCM_haZgi@SZq2}t~%tv{x&mHmML`JQc^>Lx>hBoBIhBn`BH_S5Ub9YsXLbKHc zmB_C@tZIpDMLDI{S*6-O_|H-J&-yae)z=SG3nC)BC+CgQg`-)HAy=l2=Z$@9UtcI+ zj>BiQ!okHR*A!2!F()Mb$1IEnn0yk#V|cX-)4*xvi*K)`?lBE9I7A(9O@opYj+oDD zFqA(AvZdQU_|H%rU#w|wl8>to5qT%eO-(HvIyD?>%l zFS9KuSWTd%eU5&{6UF{D*gv->&oVHOrKMp>0S~=A&`u>D86bAl1O(>r%m>MY&5UoY)q!Iv3|#O?m~)o2!6H*PTQx<#+#1}Bb? zA_#jLy&#K0*yWB0{3#!i+g61l7!0uvU)UdlfT~*d(DcTDHx!9HBXXT-46l+pjFPRB zxpRgFd@DD@i1ypUzo+Y*=m;Cya7rlrgyFaJ?=v*0-?QfmJ4Dgh***XI2-^Hon6R9P zZVQ25)cQwI?=bDZ&kI`TYdoF{=HqEDS0q{~Rg$@m7M>GER7YCz6ol=X@xUAwu+_V*PIb)J*3RS;DLnK)U{ahd$ZDI z@Ob2HlEUZgP?3ytS;UKRTfBg{lJgj@HEd!!w()Z&_uH4^OjRZ#VB)6M53Q-oBjnKzs_J$OGIk(gJ8T=c&iV zvd3yyO3W$Q_R5LP%SPuze#$tpirep=v}N6QS%6>WU6PQ~uzg$$;>wEV;P{kjIHN$U z&n@LXa}8Aek&fi~%$PFArM~vir%d~q3)O+nZ!iye5>T9W$B<)80E4)X(ASNm<(c#&Ib%!^?*|FYZ8v;WsC1>zG-SsE-`t_+s%7 zv@=EB-#8F>>lSXml()SK8R#-l*JOg>->I$c?5Kg^kAe?E`!9N|XoN4$O}~5Hlb6|0 zR;PgBx>Zy4&CJZGTZ2v?CBsZrai*14a0NoH9?bd zTheG4orNVLyTLWztm2!?c$@ac63DZ=t*^LLCRiaU~dHNk?gHU}A&6NC-od#*su$R=fY)k;2I@mmR@ zDmSW#;mLy$+xKU*=U%cskY*!+Hy;pbZLd_xPr(YIUWQGxxUHp^DL}w<&vD-qHj8=M zP?7}wyd@rZ%C=8{-*R+Wt8VTZL^kx|`^wsDUqZIr+3=*O%OA}bnz3#}!EQqf)~814 zBepllBrI0eu`)2jCWUjXQPFg55?TkU?cw#lrR8dF0EQwUIi&TQGXtAkXg^XwXZ@t` zEzTy0pR~)1elYm$x&bw3Q^|n1_=ATMwF2)&VlPi1mrSIdTW%Gp*!aucGu+g;wyb`m zCAEW)QR^WiVciH>%z`;ev}9yzp1Ek3dO=$s(JI$6ICq^x-iJ2ypa9&Sg%6wv z+STwJneyP}O{Xj_8dl_&!9n3rQPfcAKNS`e za7!>@_Hy69j+Q3LVRL;uEIC<^S4hAn{ERaMB(~bB`-0dzH}tihtmuS?lY2h>4On=a zK&Z&Bd@wAsdLv$BmFeY;V4F{sUeT$SgzLizTyFFD#H!;3y!sfwOI1abN-WK=EJbo{qr?&{y9^jin6Q z=MxtTew3%rk*$=%7A;DFmH4Eu^_7ViRv1keb903OclU|LT*k54lj+N-jZN@Ig#~ri znIjR)U@-=DH~4t-6KO!Yh`DY1Y>H@7t43F}1KyLa(!$kNL{M!K^5yIT25D(rRc+Mu ziac#2(97SZ^-pbMyDNXj1`+1BA-d=O{b*gFO6lTrillebQ?}U_N)l_|PH(f>*L-vi zLtn!hwfe2S`OxB@F+1X1LS-A9>k+;?2o0)h(r#0l0->Ia%t%Pdvpg##Z&iEM#ko;k z$gMf_$do$D1T(pv0gCxP(1@w4{`{`hC6_7*awN0$H5}Z!PZ&3*m*#(O(i!&1gqI7#f40Q|UUF0&=#O_i+LQRo@c6SmPN{(EwJm zQ}Ww84L^HOhrBXtclU%;d#@Hd(=#%(mu7Gsk+Y-ms{5iG&$_LnV|s#0D=TBN$L{Nd z)Mg=WzPfrI0&Q|-Ib%KN@1|PS*AGegHvVgok&KYBEn|Z4*cA*F!-Qt{U~SfW>)obT z%M0s#^a#p~N%L8p^0D*rADUN1B|hQsHE)*hp6IH-Pl}hr(FGFDQJur0AIPytL}>0_ zvb$ohTskPlz%s_D!tce17#MexM4l9~D1^1c1IOF*>+{gI2jRjyL5g6N7Phb;d@_dz*~d z<~Y*t!AFa_oP_Q{5l0Y44IGjhirdY86KgrsXhFTxiX_bqUFTCTkqKD6qBZP+K2mI{ zv!x*IEivYN$}{FRxyU}-(yNhqttSSIp7Gvk^>Ho!@pl8im1#P02=LkW+ObTmwwls& z|EUr01ZfvCt=D<`+Q#DBx=@C+2Zs=6OVA;u5TR3XH|kbUdP|^$TXPIrP2QDc|10zb z;r4lCr~aFl8&Wy!s7O^&AyS9Pq!2juA4K!}o+R0sM!W0jLiZK!L*S}551yv7qgzyu zy_9p>OO?8mWJ+>+F{oCk`b)<3-CfMa4cZY!+#0LKvNf8T>b z5KkdI$c#n1=FoOn8Xs%(LGhjbfO%d{rc_BYbf7-d?&jgvF^xPv4t zLV3vpnKGu!CVlDIQrqoznrm@b_<>oiGpM#}qe3F^Pl zgMv<+tP@LP^}J;wj4{VriH#h5$pMv#mF+4yX=f`SSkj<*uF_jSdr^-oCzAvFXX5wD ztLCpW(juGP*@bTAyrIsq`nn(eu>UF5PD!M!F{hkMkEV>@X4Ic7HZadY||uj zrPY%}hzW~Kr>FuR|5WvBsqkz5J&~ULS)W&HxN@Va_V!3}%auwI-hfUgI5nAP-Xl%A zHoD1rL_=DHOb$g)&dnJSrejbKJGZb!qmpcOBT_6ORsk&E!(t?e37iNJlq^<+HJ_Pg z-KeOn_>}Zorvof4hYT&quz&99$8x;bZO-!&YCY5B8sNx$N`C6JFjdr-?zB{fTy0~8 zkY({a%Plml&Zvz)y9JR*2j}B}$WqX_gXMPZ&@XvGf<3<2q7$ z%WbO*#RAQl?b*3LcoQ4XP#p9FmUB<|yDl*~4M(8?C7Vbh3Df+7pC?8w>b#*=nx%1e zFmy9LWXONJYv;s529ApWz<^Js;LW@6i_$jqk-LtbK>;>++)+9khz+RDv5u|c=k8Ap zaHBx#d#&`eEBmUxKB8&{C=Rf9xL8PKS=sBO`EG4Zv038F=TT3W{SSXMYdXfSC(oC7 z5n==r@cxwksX_Y*(vBO?H4DBrsyg$-pnS}iYI1&UZ%o>U2V(Lc9O0S23hk3%0uO>k zbBp`%q~IZ+Ht8_@E2*Avyy3%XVm~tJHFYkC!Iv9Mxi9qo7Xnf`n|Pv*A86B`icyP! zL3}=?zN#>_x`;U0B|BR}v6Fq94-hmSrDS_ z1ufoxqxyZm@J6vF>?t3ejLMt5)6;?E@=x@*J1_3MXzRV4$mZ&)VwY1ajNVC~90Kv6 zUxpWP^`Ii#eV|4>VAv`SOKRsu*q|~U z&s{7(CRwz+`~+UsW1TGytNI72K{XgrtNk}fjU5On{%$^g{n>shh1Gb`zSwLfzpg5g zP^dMpB88#d3YI_UK@*EUcV)aex41;8i>h}Q{nMGJfaCTu25*0;H8!+r%zGa;x;XYw zj#2zX-p-nFt5wcrIZ|{(`RKk2WAk8P^>AU{A23;>#BglLg730mi$CMmtOk*1zItnW z`cI?Oa2%E4f@0Usm+0?9@||eT{w!r48f@{E`U|}Q>FdhTdfWu()!AZ$96kW!x#}xu zMcGl32cr~tUKy6aDwVYpaK9+4ir&X4{+FGrYwBPeN|Gk$Q?W>0^~SZSuh%bY5u=$8 zAicw`!$OH3sccY*s6JezT6RCBr5$%bZdcQVO# zG4~JV$aZE6GEC6M1Mzx6lMrV|GNe`ZU5_I2KXRz4y3fo7mbU-Cw7yrj73>AM)x2@x}ETe#x|M; zG9WV;ymR*(9$&T?%7zv(ZSseiFp|f!L4_w^?2yI>2xz^}<#<1;?Fz`T!^C>VGymJS zv}Y`CZf*;9P3m^ebcz)|ZS>1@`mG6PIC6c%4gb?cMa?hpZaQb=L`BP5#YuvDPP1~-{A9|Z)+!+ z>tf+=e4je=(QhXC{TVid)g**u>8^A`UEFrsS$^tg9MyeEsb87b0zOi@>xIQudPB&x5XlJx~=o=jJlhMd>1a+Y9)0!?c7X@|uOUb~)sPYHZ#@Ac*G~ zTV&5;#WQd*ZG@kp9X8l-1AktS?dwn>S_%7Nw3`?E)?$1SU*9Vu{S-iF*73b#>Hk&W z(4i3S`HtEBB=BfPmg9@HPWBFIUFcVVb@?BMB(~@sNh-hSElD?QjGMaBhar;9+LEDP zM;~52B7PwEK?f0pe?e;|T* z7P79PxraB&yJ`XorXsB+9I<@6B_kPg&HVqF{_=;Rt$Ts3P%fkjn=!l=iD_yIqSWzh z9`a&jn^!hmTCy_`7IA|O)m&ZG8EB!x!WdG$BR^~cYU#nRcP7|){O|q*`;i*e_=KZ)U16k zuC<&Z?RDY~J2g!nP=p zx@ij-yq9bXD@wXLaYumg*f1s@+qbmq&W7%KfT-GKbz*&V$L?YGI%&Kz@qEGht>2G+ zuJ)#5$&+eSCDzGM#~X=N%>WPl?^r2-+~hTtEO39qg+|B9&Z=X!;ZqcbF0)xB+}f4q-4X@;(^Qdbdcmnr2`4!#?Lj1&_P>cK&6Vzq3S9{!gNQ1OOfRf8tb}8PQMsfVFxT z1^R3P;}QBlviiMO(K$dS_tOi_yEW6fFD+{mgy4Uj0kEnB@sp3$n znNHa@2C-6b!}?}hKsoo_*Oz)fV4uDv{C7ur4W$0fM%p!N0GuYmOv27)($#NrJ9!D8 zC^D`&X)Kj(4m_p>Mx3wOHyJ;`Y1KbVNuNoW0?=Y6RKs*yM^JK%mU->Q_9w_bfR$E*XgKSX#1qwd6lIx9>=SQZ2Z zQs%KcI(GRtpXK()&ISfAObi4W!wYCak!OO0i)ruO?eBV}?vBwe-G(~JrLoIWNBJ`H z9Xp4Jh(SVZ;Ym}fX`#A?DLKr4k8S377ZzZ7^f17oQ|;HEVFp$-cR}E5!i}R4Apw!t z^{iWrSKfU6$b5T8x+(8AxM)Uv@cgm&``zZI=1(!_)LFN(7oP73&32~`hrGsL2w`VD zBpq&m(QHS@*Ofxd74DYZTxX@zetgWgn9ZOMJh(?rM*gv`j{a8mt+f?SVo<@EB(Bp| z(m@uC{wOUesT2~TaBz6IHsG|fA)Zx2k*CMqe`Xu+Qyzrt-r!1jXoE-2i6FBYRzoL$ ziU@Xu;b{-%MlN6WbTS$3GtLeTqL>EtLArIt?ml5J*{; zD$b>>HOv?;ons&BSfG=*>&g9w7*ou>nsRB5>XN<{?G>j@~FD9cd#>ngZ~M9v?k-T=2w(UWiI7m=-ml7R-&x6GTHYBBfuE>vJs^7R`txh>0$8 zAisX4r2)6IVCLHX2|q|7oHJy=k(H0`4bs{uwU9rvR1&lYhSLm;#)`<3h$_y*KwYAX z55rUboRJo+aHF5L%H;A!-8X*?j$z1vz@0_@!Q3b zrVY_bSHs7Zm0SUVGu{d)v)~HGt1IO^X`>y~?4@y9p8k(pfe%LZ@eOI$^#s6Ocmv}C zbWlxpQl8l>7EI_Roztg-{p&ea862>jU)LDNRCT+-6V-q(PYj=KCVHxyVDh6REJKgG z1(Ku7UJ~xXTQ=MQ)QnKaImHBk*pQP~Sk(iR2+b{=oCqLln|A%z%oM509qX8he2Tpk zLvK?#m_J5-<~W^c{z9#r+B1EdaTZ69Upu|%>j-OXX^Dss1Bxu1b!uie1j;kA`J3~uz+9MnQ^gI0^7Ka-n(ED%*`Awdkd z?mjg;FGA^B8){~P>}cmxrE)*5Dg79Z{v@A1(8It+69zl4FVqak=eaR;YnZr2Z13nU zS;i9Z+aUJ8a&-x)S%>*;xv*ieTyZY@I6k|MAudG?*{=J(TaqefdL|$tJsMAb{DRQO zH^^vWjQ2}q2G@IiZtMB!rTc|(K<*Nlj^7(H2S>m6uD>!>`B0OhEzvbLM=iEYJDu}n zZQ-o<7e2)9?T~le8DDH;SyXmC|hMA6lVSx;OH;wy{8vpi;mb z=JL1pgecD|PGFM=2i_a}MeXSuQvP0iVJqw{F~aLU{TrxOrg4`;9FV398g|c;fg(=z zsP$LQh^nPk`a6iTO+V7_YXPM$4T4MYsjrro9|H*$PYqTGt;aRrK5Ip?E)tTE0~Au+ z>S?A@(Yv8;)`zoRQD~1sP2K_f?h&*B{*$|G4`3PG!c%+p!#drv_JFCZikWj$@6H6O zTQqcr3=vN#sMq!}d9C!x`P6~*s~iE)#*W`tjV z*s%V`2TDT4ngJ4Bp8LK_x51F_x7xxQKjr?BiDSVu;%mgkLQOK4v8;5|aoM+6Ep-&b z__;%iQ4n*y!mPOtnz!D3Bk4oZ)@zjH(wUm{#^4JpvlHX0c?gKBwGtSx>WDC1=6qik zH5fjLuXj1wy*@xY-gycPVRl<$x7p)Fv8%D~Dmu0?+&(TcB+gJPM#QBBH~vSeT}2yl z_3}6|cK%%nG$y8pu}EQ-fNRR)hsW8+^?hn;dS9ZsDocm;xo`MScYw^7j=8>8`TAi< zTtO=TUm-hC!xbLC?ikb93pfbspBMX#ph4#P^?y}sYe<}DIt4*?;!KkX4Y=2-TLg}y zrYMQ7>%K?k+#>_vs`9mtV-jnV7Tqo7YL+@c{pmYt?wA`^?ux}`Mb&rg@ax)RAp+M9+W>8q9T9He^m*fpoLznHn3j+@{dR2&8y(Z}Cq%fR9a|!t zzDTWq-#}AJmCtUP-mM;OSpY`X>5CA1VNilE%mY`Fe_kZ~Z+)Yzq++4kWTX3?AU!k||f*r27IEErQlVUuW4WL2uTXlN0AJ4zd0 z7iR2r7+=q`lGg40o0VEsO1+m|G1mYETE1oiOR-0o(=8+^M>q6!Ne2e=*;%`&-szX| zV%VMv=6L_1g@#ANWdZiCu2k3=Y(4S_|1$|$Ul~q}ffA0mG7^Bw zB8GWI1Y)yZaL|&287S5cMYR(1)^OTPio(5Z^6vLJj}(%B{l+fSXk!R4!&nL%&j}b^v^$ptWxwUz*}p zBj6HInEAeuR_|Bn(b>)jD@0$^-$Sv$XJ_q)ksf!49OKT$T@aGl&nSca-jqS!u3SwD z)Y$k+EC^hk<5(o10Gs6@B+_0mLk-gNfK)UrCW9DbqbkU$2-0|FtprOU#w1@Y);tJ^ zU)SpzcYAFhr=4lsL2TX6WkH;fs=@|_8m^;~(BnXyS$7H0RloPOfT+Y|-%7LQ?;7A} zn)cIreu@w8*eu;#wz+`8;b63%(83a}fVxPYK!0K{J0ed-laEeij`a|a_gmrsRD!dk z;{oGX=4Uf|tv|m{$}nJL*-J$Z%Y2T0t{1-_mL3SFHCf=+mlqA)_ng4i>X~n8mO04_ zXv;qDk~qcOJ;7rhPgkChd;e}fd&MoniMXPXxX*$a1&(O)fr{U>8DB~v%? z8=`g?3e8MBl|11$+=Y*BZAHJkyu^k816$AdUDxN{4QAVg4N2~)YN_$3vCaDrD|Ru{ z2J?(Pb@SXk9-a6fI_?R-}gJ9fF`J25v0)LrY}kY9~44* zv*e-uwAlAhAi32la*`E<5T4BufK``P-f)O3N`2Lcf7m3Ojwe_qa@tv6bKq{Rm%;PS z=9I&z9o^L7#purCgX zF;{c=X!+=FOjl|}fUo7fL5Fh(30vhx1_Ruk1xQ6Vp9vypjMDiJTXD)4#nv^CvD<9l`H9>Q-En1`+)QD!SL+&W6YX zoUcHj>@aN$Hlv`-)^9cICp197Nt)x$>s5UbZxU@IkAGhWr%DReTC%<~kJ0eMErm_+ z1J!nWsI-OrJ>0C~AZ=b7oApfWgsH*tZZ&aZ@D6DJ3kl{nqBL1PhQ*L6UwO5Un+huFlzUXQ!|;&`+Kwlsgm+uHn0Qp z%MCgCrc7YjWDcc$$9dbmhe!dKOxJZ86*DK($8c$rR- z^BEkU=E1jk7b4S?>Cw#SRlwhAw+EDQK}-}wI~)o%OO0DC#hzS&8JI_A@v@ko^a_P? zeE=Zp>Hrj@2>N#y9uMA9d7~zced;T{u@J&Y3wb2bC2xmmJG?r?_3U?Gv*w!&^OG!Q zs1Vt4S!GL%Y!^te+aV5I&(`0brid^DVq^CM0c~y+PVW6Rj2Z3ogXiFhdzR|>;>liX znTtCPSMU_Em1S(eBmxE-j?Z9|o-}J3EKpb1^hFg3KtC^(S1JoJokYs0eEDGD;Ikxf z2>Z^h{@4gnla3x8-P#i8VxM*84S{{P1={vu@V_N{2RGD8limVxv~MLv7J|pNB6;?J zL0)W__bH_0oyT|MZHg6aJgau*F5{Tt{kHmEMiILRkuo`Bv+E8fO0a?m(yjpjm900= zet>@Kp1uyisJjz0MY}u5U}lccb-P9reYN-@42Im-%AoyI2+z+FRbM-Xx{rO@nZM3Q ze((SRK85C!##UV|t#5X^>>k?Ea6-`0{&_0J3j>}l@f&oIrd(5)UHU8K8M}v@p>!bh z7b+Wf+?r)#(2QOFphjj~I{0_AXV9M5k`Uky%-7Tjf2omoQ|aB;t@v93XD5(XqRB>S zA7`_y@x01WGPwg!SY3|t7bS-@1)AR576-OYQAzxQv7JJWfVz^E)BTw*E>`fIh>*Be zi^BY8^OV@WTONdgimk7(F~&y!Ki;C)Vbzz*9fqs6NP;I$_qVlVNlcQ5Bw!DXiHh^D z(2}e3Kt{27!Z3oP-qo(&;eig+>fLVy-YT<&q(t0bKknv~V%KAy&nvNjNoAc*VP>m- zM!TWbTaqavK!;6Fd#dQ({8*x^XLta&tDDIN;b%b{uG%Mu7FZnHNo6p`_v~I?`(J^> z(J-iv!4rrowTo}w8z`P5{a92%^1SMjb{lR>Brh7w#U>X>^kQn-w?Qe0=U-KE>Nvb- z_3dKKd0{%|-bZnI{Ymx4sYL)}RS#HDd|I&P?tH;zz*Blv#>x3S^)bS!npX{&ZF`TH z35`tJXP&z%X*MuVABiQ}+YQpfdp!ti*{Yl_2989O> ze0J(^k;)~zXGwcmV$#jj-f`vSR`ev+>~HNub$%1P-vG5{E+4BWsenbos<6xnrzXLK zhfY~jLfSjF5ST@Qn`3NC*!(8(7Y4D-n$$IahK)}`$6|ZrF4j6w24OKVL8Tt0SilX~ zKVbliOI+BVDECIqS!YmtITy-X8BPw0?0LndL2{Bu26(ir2>9}QE3{C+Y+Nx^rV$&U z<%wXyWJ{`iacc(tY~j)GnLk7x$ask^suOC@3sOfLB2P1dPEp(tVBsBcarnm=^Qsb_ z3iRQv!PY?4vLWBHwz0_%qa^oA43THXNL$|9nw%sc(?}a15|@x(B)N|=1Y#}AbpEHM zRUSHkyNro-&cY3_SVI}hx*y$c-l8Cvnr`g;Tb}bmvk<8ls;v9F-i0xv`%%_i$bW={ z?e~Bj=no)37D!JE?;*y-B36iSH4@_sjZC!AW?fZ2x=1~m>S)fNe= zU+rzgEB_DX-aD$PHQpB8VnL)uL5h?Z-E^c%$Dl@00i_E_ldd97I)oqsNe~c}4hkZ@ zgGh%U5PDM(=`Hl0Py+;bAMSnbz2}@e-W_ke@y2+6G5i9{m9^IQD|61@T*f6c7kpjr zOhcyNAckIfgwp8fAZIX{f`#TSvU4~i-MOWI(B~^;F985bq^Xgdb~q^P6B)c$E;Ic5 z*@uZg8MWrMhnnY{wR6bgkJ*AYx0u|KY3hB*j&%U^*{i8$Rv9|1`UD**u?8SQm`^f> zlcb3L$*_Hq?;o`LCmlU+e7u&u>5Rpo(W%@4yv=2U%a_3ImPIIfksT+k0ZOe%Sm zGJ+*Y?c)zbeZ$@2_-Pns{{oG2?k4GYtfvhH3b>KWZgCpJy`5z-f$5CnC>Kd%;ZI~< z!;DtdRpk|dGIHSQK*xD8`>OO+1ShNS;uNbS9;n?JjdJZ@k8XV;5*km9xD&4^ZX8Gr zxaCnjDlB+&fd!%mJSnuA)lIIgutC;7-*t^dgq(0nNJM;u<8oa>d;_$29fzgpu=h${ z>+8ysnKY$0JZ~Dc@4!k?>k%d>i8i;x`FZ{6FO@DE&DtWQgS)N|*&tJ;wiZR}c5Qdz zAh4DhM}Q3cUQ~6}x{~z2$Kx&80Z#H9t?%31WHl0De&`{gc>Qpme^tV7IRDdYhxP6H zv>~s71}Xwi43p!LpQ_m0KGgs+i}RhS5XN>-gvojhO&2B*0@`wLkW{sg@}P3L3JR8= zF^yl^iP#CoNfD-tR&{hm17rNWaM4TXk5l^q<$2gch$55YVSKyS5hKl1gfmWBTq3hQ zR!_k2#e?;zE5=h?wOSowNl7|-x8?AWm4tE2P5TyN?5QMENg1%|Sp}GY21fbMe@+a= z39}fZ@6jQ7E=$`-1hKFmzKyhDZV{tiQT8?HnmZxBis-QU5OUbl(SOd`cZ%p#t1h4Q z=!wo8HJ)0u{^OedO=JZ?1}}onD7o(GuvDIy8E&%RJ{La-v|lQzxxrpasS{yxBKfe- z{vtMhsl04;d-yL?{)b3y;ILyYpmaa(**EC>cx%*UFlA62+pW#J_w2-Z99%#WFHWqW z3xCpi{_wTC7FW934}$VcXr-3ezUqm8QWU6LPwc=}*TO!nwTr2X!WpsG4%&Ekr}X|8 zh62V)1}#>)8XMn}TZ+hEI!)@-&ScP&CEN9Hg9aw(G#(>T_c5jaY5PSWfqz|o=|US_ zQ9tJBv&!7SGT=2@fAsvO$$5s>?}BY>@rH0{cbDlr$Rb$W3;}J#330EcmV4}WaxEFg zyx9iFYtN7J?RL`Rp;lr;Foj)aNbyZNMp$Nc55w$_NCs4F-80Pc9OK?TM26%qmx@zl zZ4b?wx~KD?qJC|4mGsX3u`_mxV3?h{(DYv!lV?X_R14e7Gpt@!YMH~53tbCS!)e0} zyX0Vyf4zaQ9I}?T^oA!!X!6iMW@2G2pTthcxd=3F`1Mj-feB39%crUv{SJ-`ftAOJk zp*rJ=Ce1FNI$40pdr4u!N-gd?yEzNQp&~>Q9|by;MP3WE{xb<=U-lE1k`mXt^c^J( zCJ{-q>TL;MR&t?QsmU{I0e^50@`B?u0Z=!m3Z=^lERy>tH zoe^r&lLqeOs_Ux@|2QVeQSY7(40OCBoZbGH>GQ->P+yFo0ETS`Hp|yHvm3l9qTPnK zelIcE4cT}-Fps)`G;X^#bTC>|Q)qJ}9$Pj~U+&#hTxhi~Qmz?)mJ@6{tLy6T!@L=L zAGjWVk;hh#&!gTVBmbdbi}9>8RC}`*C8G>P$*%fQQ+59z*k7^V3hC_}$8o+2_Zo{Qus`$<%Ef zC|O%35o7mqz7i&bb3B7rZH^1;mE-Ol+>9Hp=9*kdurV+QeNjUn$#g;Lp_t{egt#XT zQh>ScwYY*i?aj#LKWB1`7<3N9F@zo!Y;f4Nj_*-mAzYYYHnVgE(ebILmC=s5?L52- zudkjR9OQ8Q+ziB2F6+@LfKlePX@YbmbG?4z$K_MSE+nE)=tAlUVW+U4XT-H}#Oew+ zd`MCbR4<2hoYfB$V~?GpW>C*V{9rp|7qRB^;L%)TqB>sXFqKmK7+IN$q%r(ey^8TwNGlG*?eYS?hOp^zt<5rU>)azGzuxTsTmk z4tB{cv5EUV!NXYPv)q=3PMHAWkq=I0X} zZ(n>P0vKE&)_)Enb7nZ-^SDHUZh#R#PC^~bn?d~f8;WJ%;f)(U0Gc_o!Qw56Ut5c$ zGH(@E)1D62ny%3dog&_4IPSDEWXYOsJaE+YASHk05v%&~N1FTx$b?s+YikmK`zo(G zp{Iy6b_z4qlap06pJwGhX92ifA*BX`rTtw zt)^HqdC%@a9Py6&m2XmCPJ38PuRNd=YX>n9D*qC!M_S{`1FG)0O+ZI;ml0qWm#ICe1J#OQlqOI)nr`u5mD?4aT)EzYB6`v%P;FNGkLJY=KI8UzQQ=aDU{| zV&o@eJgmvC-x1B`jJ11|x=Hl0dPc!Xv>YROcKw1mpk&4AAnsb-og7K8XRUHrvA1`*D!r{L?ocMFTBj))ZhZ#@^=Vt3KQ*Jrfb0!-;^h9lof zT9wX6e@@InzL)qZ1oUNb4w1m}ruTl= zt4Yhq=%>Z*iZCUN?^t84#;>nGRDIE0K#eEd%)D6%D?u7t2ivzk!X+KoGg~}mwH|`I zw~v=h0E^>ua1wll|05?-BpZ9x|H{E5MXgy`~EtEr{DV41PIIK#+WGJ zan*JmIvxuq(|!&7Q5F;eXC!>=T}{`Ed{-ZI2JXfB#&@DEc3I>LpO632-6PBGDvm2B*7n zT%&5ii_A%z8!1r=6z(F%qE{}?NMzNkVD;d|NKh~w08}!a>d0sc%rak40>~xr+lcOq zWNiWJ-78jK7;LZ(q=?1(NCuo%+jnx5)+wmcWG@KR|JkNaz@6-&#ibN`)splue%I|` zfYOgTlrg;N#|5d<;o9fumVtd=wl`Gw) z*ADUIyi~Oj9F-2@)tl~RsFwm}WEWBo>B*2)dO`}&osRCy| zZWcPwJl?s<$p;Rqe(xU@rH3isfAZ&h|A|sxPbG^aV`uIN881vO!6)4Vh|T0Ll@z2} z91#%wz;WUP=_0`SUSVlf#H_wF8UpMcXRn<)1mrEr6JG4XzP6H-?J>T|M!hk=-no2! zxY>=gHwA)QQ_v=MbKn2}Q6dd-U8n(r;S*sfcly;<1x9tw%2WF_ptH#qx1zEU4NbhA zjZR6nY!35fZWqd#3@sU+lyO(0ab~S%Xx`{fc&8%eip=)datTRVLx7}X)^_R){zUf< zsJMT*q-wREK(PXR`9zVs)~8=A)TYp{TCL|nVLlT{;{YYZlC^a6rY|eSf@m}BPk9|F zyQ|WeGj|25ed5xhxYZZGMvjA9Cj}1v;Mq7pwC8%$|1AWF4(b~VVPAXj&?&xI7kAfN zhTaOm8SED8{^>g81a@`Q_!;ze}Z`+1h3d8hzV`FB?0kt24T|xP0A~ zo?o8-%sQ29CZ+dV8a3qh&_E|9ouuj-SFk@HO(KKmHU?3x;2>(0{XF2!*s$ZbEfz)3 zT*$G`s$VtE5aw?FzDt<>jx)_u|+>5t!oP$C!6||=Q7Q3uc z1)C)QW530SxnOqlk@a9`NAeDjsdjgNT7q++C`+FdfvzZ0;ykPAeVMrnNzjh-scLkS z9jW~+U;`vLM}?hmGI-7oKzqJ3>-}kDprhwqf>-h+fM{WCfw%!C9M>54@5uZ7xk{+Q zmH7Vb&!Jy7GdHDq-9OU|yCh_}02F?z&hM72_qK5L!S-+=xslb3L})AowsHJILK|h5 zi%3qQKS#$=v<}rIdBvl8+Cs-VntQ0b!$Zh>6dPuI#Jn zvs~}_y?!M{gv(~VD>W`Ab`tQ-(8l8|vPU4RCn}z(a-X|pW_*=-he7iu4qkor30#ssgHPu7Q0qHM-y8-AtjsW z3nQxhf6fBQYQ7N8-mr4{o1;3zp8Pr0EbjC$c3HxR7f8zxbc#v`(@gqB=BKO6wr=Pj zt$KOPt@IH|&Cp|azDR|@$^mGpvoIDFc)Hy7cdj3>7^-gy<(Sx#8G9{SOZo8VWRBY3 zPmjiY3JE1qg`Vx!1AB*$&wbMqHa+9Lv*ItlUC$|nlDk~m#x{&^b`T&_oUF<-k;pOt zDf6wX*RR%iPk8w-MQ+m9x#!hfM^LCg!p6M0w~d7Tf=|zA=sY=j@-XqU@?1JyOlugF z)kWiGb@XYxyV8knVg0OA_{Z!%>WKxi=<&=pN18d%Q(jcWctl1WfTEK8hfD1sIiXLK z;TlpBo`X%fYZS+!q-VWrk#ZGaO@P^sicU#ru>WOM{&(}vu!pATG_&vLafPh>6Yg=} z?;e%EHoUznCmXmFo~{f@5eMgz5Z>Mj zoy66#8jeKdx5)j{K2pGk5}*;bOCBzY8ua zh(uBv!mzo=Jsoc*RPvSsdI5Y#qqDW$3bc#_I&oU8=*7o3&Dd$NHHs2dyrAf;tH~BB z80Eh7=BpqaK}qmVd|h!J@lrkRKDDU`pwr?Nq3YQ=-@zo{O~u@0iSvx6I6c?SYoOC= zEi+~cUL?#+r7XsKQ_ch8R`4V2v0Gdr;IX8Tz_D}YiZ8?lTN1GdTsPAaMf;Dn=X%@g z-uj$)_qgzP_R^96&}&hifcSUf*Wbfl{dMm?aFZPB)dFJi^rq&KF{}3Ix=Poj&zVH| zJUBNeX8@Qj{Jw_TQ5eA4Re9y;$+|8wza0E}TH0>pEDsOQra0#r-B`Dx&AzlqdnW>( zMixbp?c6lyCbbaxfqzXMle*%Gf|~71F0O8x_8Fpfl#7%2B)vhuWR$V)LvhI0EF+|_HcU#9c^tp1D$379z$yr z4}d`H4NKnxAiBbb<CU4jz51b8yhf(aX1S^rapA+meS?)$RHmew=z4;9Qn9-P1CoH zJR(y;8aHU$@^i{q+fl%dpO+a{vu2tBX&VyzD%UGzwP}nk`d$`nMd%l z2RqG*05o2kUw%DQDG#P%(mZ(VQQ1vlOQkQ*1s_>1kYX&i-u8TvN9=|E%e2*#CI4e? zV6dlWhzb}A$aXH5mX`L&u<4pSW^ku$GYnKeXQE?Z&WriO{;37Nr!XKF=n6OPxY;fIH1-@e_v z^8R(tSFL#FporsDKKS9mH4%|?U&gj>*uN4s@|hYi*E)8&XGyCAnFeexr!*&&FH7Y& zehDY{(qUUQ^AltQKE!)!Zm{ZB68z)o>^2k4QL2)(Sg1GM9IH9XTCY$c6q0L_OH=mmner~jf7AcvCCZ;w{6Ccl))N1;fXyk*!&b& z(-k_bnPfU_y2U}%o1rdZsE@F&;VSC4Z{I$1WAt|!`H~(PRrIK|N6YQn&eEptYKd8e z&Om$3zDmXDM&0)Juq|DawgVN6ebnuuwktA zYe1IbEq)pI_(TkCdvdC93B|J6w2hMG?g)hGK=KBw*~5HJUm6m1M&;?q7|C; zs55%WnT2OnRP6Cx$DJl^E!j()>(pZnkI>(8jh*U!Wbs7p>yId`7kyhhZ?k;{vK#XUWXWEd2P+G8%-qzA4!QZOp#wd~$l|>3JP}=~`EFE$9xY)`{`Lu$6R5cS*OBcGZx?lSz zIfO4as80~4m!yB{iL8_69BVSMhc_n1$IEwLnS$B95r7m*#lrE2HzH&7G`Ne_IGB`l zr!oTDn6O7pQ7pdkl6ai3PH@m2H^c9j23|Tj=(`(5KMoE8ia>waXkh~?+u!+VN;_yODr@%K#yfA>YE3>=ojs;gv*UpO2H*PgvUf|a zfTm5kR`#^3)x6xc+aMPB;}vv`zn1J5d^TKiST_;+=3hRTSL|>l33;-%d;_+(x4d-n zUpDyRYf<;}zyC3w3^i-((r|RyC>+4Z!v6;oTyWyQFu?(!TmI~W1$B|ry1%(~lhP+L zAyu9(gfw><%EtWfvF;MLv__`-RpSU#t1@KC@2)zBgS4oehsRed@)UUN)uY!=hxtw7T3wTv)&GwKc}u>sV8E59-Yc-blh*dgLIfW;YR>rjFrXJg(ZsxwM;`LtqPkeI}wBsl9E2^}b9UamN zuw)dfA_H+3YP-3XA^SpkdjKUq#XCi-UFHuzy-v~)HnJf=J5H3|)m)fz2J=@@2grZq zRuI6Gtt`v}YP%QXpi~(r68#7nk=HxTgi+VFUXX?TII1d%9#q&GQfVKell^K3WzABGiZ;NUN z{*8jJB)kk`(~eIFcPo=b{$o`D+ooaL;O}9-}v|LyEiE_ zegnIiE+NqIj>%>-#mn4%ROY%16Y6(>`1^#@wQ5*C`m?#KQ3RB_8QUMrB5-d{;J zMKt(s%%nIC9EC(@d|Mu_g%Y35#F(PkS*|@f)vhP(FBBu%t0u}F;uIIL~TY}gK_Mm=myI&-?b}X1Vu)%o6)l#r z`1LVR99VcH0bHA(GSn73S$d>CrLX~5pX2}y_;g5DWV|pzEpe@qF6`_fWc8JnCQAn- z%qvu!tQ}?i~cy>OXp;&WVh9*p3QN6|BsE~e__(8=*Q@sh{~h$ zqv6Nbq}JLV(3_%Kyl+wxLdWRYxwR)a{=9Q0j1A1s{I}sdEqpWU{SK?e<@wj?XyxyD zPb;guo{d-i)I20f99lF6*io#`9}sdm65;*7$CQJ=Iu5ga+^YL_jv2S#$h1Vq-*zg3!kxA|)yUqDmky8A$GnwFos4{H zeShq;j_fphN6b~osK4bJeqh8j{FZ5FO!Wd^k%dCRl%RXFfSb{W^81lIZ@>Qiw$$sY zKv3%a!jq|jLVvxP8M$HOW}UG@b;^ypWwoETNWs|}&K5+)c*dryPJ>Iz^0w;h?kLE3 z-6Ge+4`PS?$dIn$ELOnHUc<28zzb zNSVd9Ksi%7igiJ)E}2{zFjKt26ruBbM9Q?~nf*1mGxpj>+r#zd0v2spiYORR{35M- z?*cN!^KSDAt}`N^x((JfIE&vtI&;7(l2XvNSL7mzR%5%EcR6m0-Ct8S7n#-_^S+S`u3R+T>TjA94~B1V35W_FHWl{1$b?3ZOj_oN@W( zWU{3EWd;1Rh43(mJ&BfPb54ttM1-Q<v9oyq<4sM3qye};x zCNNTL2jM7Sr)|YRVvdaBGB-14XM-vcq9ZRJE-!yHtuzRQxiyxJe0UP-Koj=^1KtNz z$>2Z;B~*1P?_3p{NF|`mLo#M@%ojXa}}pOC_f-Z zs^a*4GYdV@mkn%$6E4`nRWD9TVS47)(TsohWUZnN*>Jp8Q(lKpHA&0_r_r%Ar+}YI z(XV7@bgP`nxnQ|HY8{~k=O%D%smgTs*4maK(`WONvMPXK1VgdfM9|&{5#pm|!+#4^ z+&`{D(W*Fm;B&AUMDrNJtbgyHyXDM=P-Af7e#_iu#dUTzfJCG(;fFTZw!$8dS4@_; zsK=bRQ}(Td?Ta7GC;TCN6{1Z?BHY#{e_foMime-gxU4KN!goU{vV zy>k+=cliLR*dDoacteA(XyvnSJyKdMmGKECOkqLGjN!T0CnNtzm~IotDEsvB5DT@E zPX8%fk1;jW7memevu!}iG+Xdljpvr(wP(AxLK;Nfx4-M!?N7K@RZW^=_979kO`EPn zR{_NBrj2l#j_ai4YQ!?O+sZ|zrFo=*O+s9dq3_nThsEBRU4E*Jv4@8etpy9!}|xaa`zgulF_Gb0#lgOf|fC$lWQ+=`x)S zlvB3z#KviUp!|50A3}8EL$5)Ek~Q9umHG70WJSiJ4{n{X@$8WGK*UGUn-yB#78-%IhTHyX! zeI!6$y;xS??oiEG@t)=tZ{WkT{HYbz?CnyQE=^gqoSN-na0YsSpFh}Ynzu1|8yb*O(sjT%cj6SAv;#Dzm8g*{$|UtYjpfeGNzBJ z9y2H1cjoDqaa2aw5}goNWx-3WW@?}If9ZUAZrN__%-?ri)5D=U*&Du2$`3#OHLjtP z5vUGhLYXnJAQb6+?vBG&8950UCya*@S}o-E&vjDT5Bc2^6_cY<<3t2yqL(-d9i(5=~<&0+a$5pi1%hT7N3*ILHEX#fQ1!NGV*4 z#H0c~cJy^Cy=lzS*3t`JC>H^))P8E#-z4-B*H2whXkW)x$`ik^q9Qhg;(JCF8IHH{ z<+tFqZ#g;o7O-Ij2uE4zca4+{n#p7qnug2f+1Uv!?^u7TfvHWB!+FnN0}ik#sSMnFQvkl*A>dEA~glP;;~#HiUU=ySxTOVSa6#U!J-kooTP^(yh)tqX{{1| zL(FQaikHl}lC4z`(z`cm6x&bH#!akLejpe+&>?CBCba3q(~Wz%2())S(tO{zVST@P zCb3A=_2!?J{N1j{O+VN_j&l|TA&cU&4qrF6 z>Nyh)pGwj)OO7)yot07*NW!S8sTcQj!mNWxCyU-`6fe&b)3mT(`5y@!@OEdS_<8S1q`ONh@&YkRG zAe~)>><2#KB--sb<(j%hv?LgSrhex=xap3X>m&~w+MRZ$j&tC%r$%txw z;i8_-2F3OO`vgwt@#^42Fv2wu`D_7L5>8$lBv_z~XCWK)oc6ocsk`ejW z)b$(n>4ZNdy&je3pd{LtxHw;yLAse0=e=L6DW_j&BS^boX%87lUwkuH|89gxVq^zk z??MYqRNPDD;>0Wo@2HdBtR0Y9JQk#8A;E0s;EwbERArSO-r9)K`vK!7_U49A42=ZaAajlL9ofXma>8`{kky9Yt!W*Do+7nCi6&0&G{T(HI=mOnUX zvw(?8JY(NABAYn|Y4D}&^sd&z?EC`_x8Qj`SnX9E;nXdwsPWK1IeB>sGHsjmb(v*r zgTq+q(T(fZ=P_z8zo@wSG+GV#!5s*se~{GwI8rV!RchRs70j0zxi5U?Ah1vz4;eE zZ4{XnJ=4VTkLGgNvg+~_9tL`F8|~Gk3$J^p2{Uz0v}a?&lwJz2JYhvXQow6FzA$5V z1A&k`d)S!@_3gRR$O!7x*daPl?UW~0x33{acWI9ix*BL>>On11v&i7FzMz+EyLO&G zs;!G`_9D_WnTyaXrjnMRrserqt#t0F$uit31a*_pSIwP#H#?bC5)m}GE5)hAkAd5$W{ zdJ-tgX>O3K{pr?o$!C4c1HFCm@T%-wqBuHMH(MLQYeZ_$5rU38&tz(iau0m-l-%?d z_3{sHFPt+!KgJc0-rw1-yO^-kKyS!cXFLA!GNybUoS46971jP!%>i+d1NuF!e}pEC zGJAReH)YGuqWZz^E3#7M=>aoB(IZ|(rZ-u>k=GOT?Q28QlqWW-r*oC;W)-ZCS&Xdb zwE8*s#zC;K`%+RPWG5(`M)1vacsDpl-7~w2mgZFMSXTYCN->@~z5Mt>)QsSpuq5Jh zh9XNr5#UBtJKL?f2HtZgAfSNE)oDKG9KeZt5bMzIMYa{=|El(_u4VdqNMv-fZThLB zkJV>u;j(A-Uf#x#v7q7wmkp+X_xs%dY;LR6HyfH{lwLzkoLLpur)hs2My*2l+6(gc z$Wb#FWp6lMYHEr@a3Xc^Y2%nI_;KAQh*XU9a~jD;2n)Z^u%I~JbyL~vo(rz$)UTAo zFo;{fcWMSw0&T5IvM}1?vaF7cq!zQNl7?kQs>^y~a;2hr6-&A$Sy!G%DkdzM2b{@l ziNJ3TJEr?(42E|elSj)DBl!yWRdQ&4(pNPyDAOIr=&lAnXA====pTW>@mvH1aZiRX zxys>`NJmG**QCB=O-c1}Dt%w}{D`GCF!&i|AnJ8{{Nd?v~sy zdvw)w8YazY-YZXoiga1}w!Wp}O40aa|A|>itID{2b({vC(rPGL4aC-ddGPG0~UhBOC5cMw>w$VlUI9M*A^DgE5i!H7xP?r z8v5<(?*$P12a+rEMd;#)?$}nCh&Kc2GWf|kxR`o?6-zWbU}#CBJZZx4{0vU6bmy= zB_fP|^yIe!R~i0Xt|)tWQ#WgHFxaPEVX()M!bt*Pkqd`~rH2Q-JyIU!8qZ%Qks@-C zxCv;sB%%Vm^&@mcx5D&zxROOolh&EYy>&%rFS4x$H8q})pf@gK0K&; z&f!_@n#&`qv`fd2mpJ|#;$sq8 zQ^}IRGy-FRF=5Q6WcWVM{Sy0FBSsx{ZWty#X9P`R?n#!}@K)3*BjxQ(V*1Tn@WDUw zIkQz0ySZo9zAnyfaG9&^HlA~ycWwKUG5!LvwGhx)TcRA!b`O3Yq)0^0vjvNV3M{Q0!{-hSHfV4 zBrbxFV?zvPQYr)nmm*-)<^#nRr>N`suJZo4qprg*qj(02tY#OfU9~Tt^2?Y=@pVX3 z@ykDs&b3CQ4M-SBw4b|A*A6Z@T!cp-S{m@C`(OsHjayoUb(>>VnMH#P935@G#uKJ< z>ftq=I!|`fw9fDCzp3Y}f26I2{GCx2r=kEY<|Ec~{{Ci5vyqdfOO`A}xZZBM^UD9| z7d=PhH{p%?(koST5iTSv%z&W9s>iObRqP)UTmk=y4KdQh&}T+Rl99Yi&J{aZtDtyF z4_O37m2N>^E-%u6zRfOBOf90RU&&`%apjz;YeNmOH~|_M>fk$3qyG#hWYjbgy5X<7 zbW<1oKsHIRkNztgL3i!RGDLxp^Qso5+c`DGa55d&(9;=-84yp(@?5fcA>wQ28t8`I zC`Tw-3XF0}Ys8poahPrvsUBR>rfB^v(1~dlW-8aNosD#;X!%=-f3%|~KaftxD)v{m z!}gx-%rOJr0^ZUnDolUjuiBjSI<8uY^b2h)wHU#^WJJe!QN>Kc6Dag3MG<^Mi8@fM zw&TYJB{D}XT9?oMfoRpGCpR8$x!wFl@_4r;tY?t-Pv zUF7>?vojz3`4LcLbHC2F$+cq?neG z-$A0}h-59D!>8ES?YMz8pB4Y3;wp|xb@S*T10~BZPB>W2=qqsdIJr`*KCD4%w-{p+ zk{FW|tWiJam9oJBOGuy|{eGEi{sN4t()bPX!+NQR zF}Y+uYLl46LVc31;XYV>2{GXH*yfNyZY0WB2df9l@q)`70F{`GhK_7B{hqdHF)Ty~ z*I7N+*0=H;LPyS}lTkJPF>33$@8Ln`mXj

m8l;r#Fb$GtQ7JT6{=LTJq2m{&LU2&;EBbo-d9tK9~ z%AsX7KKu)4{rubgxD*&^iv8OJ6{-7}oMYQ@9G^c%V+LaXMxkQAw#*su62k>%V82z{ z_&Q?xLblpL^VOUGlc7kna*+N{+x247P*o;k({HaPe9){^{~P+~p8lfozyJD0MOhuQ zpATP$EFHYxxowRZG}okCYEClkKvZYYf+#P{BvcsYh7}^R$m0`03+0YI2oFkYdnJtI zEpB*8b4MJnP>U8SsAWWXwO=I;4MT@4uv-r{yyF z<>m7q9Ww9krX$qQ>os!s-~Z|qk@7~VhLJmcFQKZ$z0u_3av;XuttF-^@OQ(`9vW22 z4a^85GZ+&s4nf(4mveFvK%a|z!PaH`(MlW-l#g06Rlk;Z^JE+A$0Oq z`@3HT(2hBOSc#aPiYbq*3|ySH7%Djt0^{Zi)OEqS-mIrgXcc#}C9OchZkY|aW4Sg$ z*(F;{5twqFRAy~54NLm|t5!X{gJfGX!u6y}CV<2uYsPdlkWNCM?T_msy3;Lld}b0# zb>`r22d|=RyFfeu&6ro%Ps{MB_v-<=X`jZInju29vOGMUd{Tud%1)*f^uj|oUGXbvyVPW&@mY5V+2p7Q~sVQL!UyL6!& zh8uUs(i=RD=IxQ+^-3^4KJs&V(5fin=OLmSZ07F&Zi+W=t2RN^0(m?27CMq+XFw>( z2MtySq61{mb^rF++}bFzBAQJpzqyvW;;F{JCn@lzz_h}o7S$&Vx2~LDr|P+}z58nvVQ{t%90TIs}$@ z1rrNm(2N{DMmsvFMi>n?h3L94)7QnR;h|-%drlbguu}O?uqZo|@B(?jCL_#gZh?_q zj?ZyIeS4W=nT81y%|wDt;MW89HbPb%F9mnIWBGveN)}_l+on}uANn@Np-o!lvDdCu zo5W96=Gngzt!_Vlh2$OPH&~bpW+NzeOC)deb(WbrmoSW*RK{@&M__q}Q%h76l|Km> z9;;fPjlX-||4R{{PYx{r1@Ba{yjoS08+NB#%7BXR|44!t6+LBZ8* zz-_+7r_{P`t2wlrlVkfHv-n|FPr3HwgrFNxK_k_Fe+s2%edu!*`*rY)p~> zK%)(~h^sqHW4lE{(H*2lhN2w@m~1uSU4fM4 z2avCV!RY8*)FM$VMx+>=jOJC-iqv~$OL`&+9X3&md;I9w+j{uR7H$aiT`uJIK=oGH zIe^2QN=pbGq-STLCx%uC1Ez_V8rcYwMz`Jxrt`krCV9^CI`u&lQeaLJfKia*Pk&qB z0(OT0hj9@;yPvA!3aYuwmgY2Yb~x?M@`7JZDQCdF8yJs>IS6Z)7lV@(4-3w$phZ8k zoa+(^5D#LOR9u8{T96GyVYqd+-o4%9zyCimD)*_e#H|N|Ft_hE5;qs^l*v4`T3l$g zACt$^l&{RXSWo^_$|->oMiJlm#Tgg=bL0tf7GgLXl`tJJuPlNGDrb?Rb`d(vPgF_(c;CgP@wB5z&C5gS;zm*i_Y05@6 z49Q7C_I`$|2V5cVX0i$cJEQY0X4~J|9sqc4NNN(2l7@h{&9`3$1$c-T!!dMrXgu81 z)EV|dY!85>udp}n(FH^37;OXPz?)kcZNcgRDD~B`o@|K1A zY6vhEtpS8kT&P3yi!c+EG098;Z$~{@^y%JYu)Pa*?(=O=tSdvz7=O@-t7asn91hevXA1K3K{?QdP*gmnYNa|gD4F0F+D-!zt(b zrsKGkL!1*s52#l?PSp@ux+%sNf;morn5gc!u-|l1H4KJbV(XzEoStG3 zZ-j`7Tn`k@;fB9s@1VccpZG|C(McnAiU|Z^cg9r+^vykY04EjozQnKz^*#R&F6Hu4 zJ9z82jVY2fzh=Q~z;U06xOLdz_?R=7scF$J1etKjSP2-dm?71KI$#{UE&YacRsl+L ztg~W*eR71H{tpuPxj;EFlUK4B1^mSfWR66es5VHfk7HAee#`oOAkhB&9uB{+lyuJ7 zh2`3rGxbIx%fX0Z$G2le3?&HanU+f-D&e(#Ag0Q_xI*N{VsW?ND(qTT)jvO7fODIeVHVRv!)XkfHpNMe;eu14tIVXq5hu z^;(a{)-CKP#^qJsViDJ8@I_|v>%L_m@89aFy4Rny@GZg zqT193=5&8fD8(aOsun3$Cb1U3qA{`TJgtl?(`~@d+&njGk&_hqD4&iLrQ$jQf^nH( z3UU%MN2`$^2^&arZeV@V`(TjJ?OC82>)O~MF^LXa=``}qLu$tQyOJEzZ^zq4OPD;d z_z`{sesOmGX`o!dKkUaK`9;It?!3!PYENg1N_5bgHk(SSsUwqfiF*Q-UsH zii#yQ1OWCfp-mh;^XA8FJWp6N8^OkeENSH3izu9Y$M@Ox$m|u?m^8Jw5fcbw@;}c| zAZ=aAOz+E7RO_pxpJn_4F>He0m$Mc>+DU82Xc`vBq|Ppv^1lUKgT+`m3Y1!H4H1a8b0sgIxfHlmr1dV z)A;rXmDke`G6*w*5)PGcI@JYmre9!S6>+i8Ah6p48zszyGa{x${+YS-r9Ryu0qYTX z#nW9TV2gZjRqu`!uR0OIffMOp>hDv7%j10j3)Mkgb4g16OO6RVJ8tP-2^G}BLLBXJ z1sZ8D%`D~Qk3?ORlg{ov4>hg524lj_KM2@X-u!DRgwv#fO1p~>Q|?p6V_C0Pk359f zUy$?T9|i55e%5M7oTOMu04O*L9n$GcADINmRNk9B^9sw@&~)G<_C^8!s^!i}3XvOY z1lr2(8VcaD9K7op;|iJQ?qj=|*Y@smo8u56yYQ=+=5`r827M_l0(|jUH+t|T&;!Q{ep9#x^A&Sjp6ktI*J(JPJc8zNn>;*8=e z5YQ34Bd0%!P-iIWv5dQEcaB$!n;!5Ql4-C6N}@@k)tf*yqMPvpQl1Ptd5oDTfM)B> zH{c{3IP;Jte9xOKv7nys%&h>yS&T!tn?Tw1?lcUq9nkKnGb&6i{l6w$sy|rO zUsS5*F+_G0v~!x9#5P_;9%TlTJ z8rHH~N@x#f4tad9#%C_V1-VAU4PkVw9R4N7gWMq;DJGg3n{fi?B``WTOXmVyu4EiW zX?7IxoCNp7b@h2DXSDE#nP8L3!omEc+Uhq9q}NGq^t8?cvsNl^3dX?SfNFCXW*sO8 zaDVLw>vnmSJzdA$4V(Fml_x#O8eP3;G8>{pRR&;eG023kEzSKrs7!mmW{ARc`-Pfo zEio!G_tf8fHN2xlv&)ri4$BkGq1X#|{QNmNCscCGnPeu7&+j=cCMVxgvNUJ<7}vx_ z&>c|HOZRszM^K*Zh{@s0>=jOlt|*=MVX~?54OgVUkaQmz964%!uv9B6Nu5x-98d*C zvE&aCFIR7jDVaSLYsLf!27JDe{greutp6wdheja2DKc*kHg$L z`=jo{QLl7wV`P;c*l>~=iiB{v*14vSi+9OQr??1Nx4GMNvbxHg@c=wB3=BK%&r7Sg z^5mx+bpK*7%S>=jXJwI_rY_ZzDR!Bt0GjZBe)Mx-1bw4HwNa^-ijVCD)@GJ8`pCYC;>Ox@AOb!XqTKbmBoM6&ye@; zvHp=QP?(V;<>RJnFbV3f9GwibS~&re1^Aq~Hcb$oHAV*|#X5o7gaLrS%!rP!03IGu z0B@DrNYf{msDk%X2Nx3q=RomB>YInu?x?dMa`pL|n;kSIe>G&V5$bw%D&rXX1f*NR z;3n50XE!1@(%ubF5{5{5oebt#{$MyOQrpkxH=rlV)dSR?T3?AlrhiY&muLZ-m#G|4 z{avsxP=E%!Xf$9;Nbfw75}@vk<&>ycG} zO7Xv?)Q(T@v_rMv&#gcr5Q?qt*sveT@-A5Q5cP&?D~#+9FY;ENFWL?3UUwr*@uZ&J zQ$WHXJ}N3b*&+$G3psjCTn`BaR3!cBbflY(nO)YY{3fW+-~<<@0n_T4=ypx5m6+qA@fnLrba|5q6h*~B3-2S77~?`Cs$MrG4@{LoH4$?)K|)zw>;&(uIqQJ-}1nA**?mV8Ridn z;3>i85GtH8WyYS^F(ObWE#nC4GNojN#^nb*u2|2cxWKUcTJ@SEObtC?cF(9$KB|=V zsGiRu(lJl)$*2zsb@UI?m4nxNU4=w+->!8PtEO-9-My}`F|6ZOyYn9p&n>hr>4> z#+si$9f{%K!AWwLzkm?sKml^hd@+OXZKr_Q%B3|R1v7XCQLgJK*~$d! z&grdFALaiYIR>ldZXap+3Yy&@i-WU+1=()sw9{6!KVg2AS5{t+Uoe-OtM{ER%8 zvbA9Uxv#rrnXXbCb7SR6J>~gx*8Px`2T9=)??xwK?hQ0ou60xvjIr*;8RRpjy~URd)6cA1^BUG60}2_v-sD93@mdKR-$oHoux`NA%6q=8e` z&t|89od->!IA)GnOxS>blrUSEZBs%b{kW`N(ahBUeRkTj|GF>e#nI6yzRC=~LgW!$ zQ-hwSe}%|;M;SavH(ToS**7U-6Njg^xju)t+7n^XUa*3<5of7qi{DV;{Rei*7z3n^ zf#m3&r<@Tc$}7yJjv0?qQe6dFOBMB-12%|FZv!>J0P?w~{#Rf2sW@4m!c)6o@A4-< zw65Qtb}}_-o>_~`$MSZAabe~6?hQdP-KQ3I$CCYs#-#l)NB@!41D@rbop6gwLPx9J z(qQ_{znU{ZS=-<3gL;6*ya(O-J)ep#rsL9L$uVIec#;-ofs8aJ*4^z=vl)1X85A5_4hnmY-+B0rkei-xZ|AQ)lh$Q%Ly;#9 z2bV;)MEAxAC=~8QMfbD&TlwfZ*glboRJvW}Iv^_RxyrqAZZH7S5V*nq)N{^(wEVN@ z8Fy$)yEE>^$a#5A*7!LRak>CKf!ew2W1{{p2zt}bc7i4axlUJku)j-qxxMAN_NBO( zr8`+k-Bo|KYhrAKkw=@?tO~aN_c^?o>VY04iZc@=jXN;~Ou0O=1~uy$t*oZqG4B|Jy)W;IZm?vQfaifiD@7!9XO z?34$VAwB9cj~JYtf5=Td2wvC3#4^nKx`toON-%!@hZ4<8OgOks@i70hMX0Kd5V&7Y z3q>B;o;Wx{?eRK3y+^2+*n>cLNj?@)H@6W z=V(XW=aLBB6>GHwf;i9B3kJWz*t6oZfZxi2UU4s0?;0;cISaYd=VI2j81LB(daYKyRo> z28Hq?=2%r#Q&t{8^%RGd{aTi42G%q42eE&6%vzA<%?#@Uy?mmqLQUsTuL2jUjn-L- zBO&X=?TI5i*<44h;8K1KL<_ce-O{!;EW~qdSY>;yYMYl^>3JLXBN=hr*uH>FVMl6CtYuHn}|LYJ-~TIBg;4;1}1lEVy?T?hG0jLs^6EmsSV z*|n7?$0>KVO6Hi+ySwwujxV%o22(u+x&@lPqQK(16>j8?i52EYjj3^_%<+{3YUr~{ zp~uwt5ZA^OJGpVEdpKTOt zfGSFr6sSJXHLJ14KC$3qflfWj$$7r>3Hs_<4_*kT4 z(_iZd!0hDt)$`L|e+>ezU|w)7V!Rm&75*OTkT;u*eWE?CoyvDK_B}hOCw^O@=_xo#x@CU+rJUn4nH7@6VXK zBz_lv?E3SwCIP*-cEAWl5|}{OT@G$#B$hj3W3Go@A*n}&k7Z?L+_h{{ZdQF0avpKL zJvl8bj1*|EZ$l?_=_vynug0VwC4TFq`HUEMFjV*T?&%7=$RFLj;INaTPPxA2<&)}b zZ*QhfRvo`qRMbU^GBkQc={z=Cu_PPJ=KwoK>&oIZg)(cf2%Isj3xOT)-fXG;6TfGl9{tiyPToHv2PcJ+njK`MEbRfF) zLBW4+4xgI8c1}Y>qqjNC_MiJ?d>+6<3W*O7+}o}pMrF!So8R5E+VNu9MZ1WX#}0n} zHzhF3WB>j3M;Rq06^RBd*w*#63$)v<@6rgh$G0C1256%9MkXeAmhv-X&D^rTC;P4% zxL}EH>+Nbz9C`K~R$k<&m>~#JaFw}OKwCt#&tdam2XR0T9tft%7TAu2k3ZZ_<>xSoL5Wm#&7FLOC`aKIjb5wg0}yrFtt zK?T@WdGnoC)ipjyQA3dhicUtD}>fsCj6=E8iBwE zUUP?pXf`)HTI`8c?y!@h=a>%$u5e8{LXr4H=-du*?BF$vz}|6@VJ}+h!>3CsQy86q z9r5>%J?7XcN0qg}6GI8}ts&9U>@3vYsFrhXqu!F&FbTPvT(ILAv`@_vt5z#!WPJB> z>!F-H#C@g2&l60P-h6eG(Nao6gVxyl4cMhy{BGvIKa^$lF;-kZPf>-18@wC(*ijHp zDx;T{c(P-yV-+TG4#tNl+QKpjpZ4qKrrWr5q_rZ zN4jkNXz51z8X9s{Qr0^NujiPTwL?_OZrQ^imfg?Rrh z;SJvsjz< z3o<-gcU$Dn(AWIfs@WaTnr!D<^*`9g$kYit?(^JFT-^TMICnQ0T>cgT{>qwQ-ZzMr zF^+-1uRqC=^8iAln8AUj`;HGLYCb^=6Mh`!<>3reUW+++J8 z>ACGZ59Rp4qxR4D&j){y3dz?b(M8)*uYcWbx>5i4iC0$-Y#1LOgtgh-6au@7PV5Uk z%Ryo7H9w#q?V7#CMUjQ@FuV0id}iZu=`;bIu+cA! zD?EI+Io|W#djd-02-Nq$<`e~ANO!Wrl3!9DR-Ta8oRhV>A#m410}&xfp^%?X3-;u9R7wuJJ`FfbU$Zo&E0deP2DX)2-182O4R z0ejk3zVV?1|92Ff!*Qi815w}Qc|cR`N*0?U>y7(gYVkE)?Bukj)P|196rn;Un#XpI zuCX(jRNxa3KjfWy{SnuW5wm6g2IU26qPm-tm`Sa&}@=N2xft_MJ zJTjtgt=a8`X3G29Urp?4?VY^907>Lm?Y{x7!Rgl^FZ2@FSvf;L`*7jVl9)ls{L#xP z`lYd_IsN#0RM@^5$Y&GvqGJ}?%-g>Tpf^6vH^8n_P8|%?kZ&M2r&P_j*HryBf2c;v z9h18E^tAMHErAA1DCm2VP$|JmoI+#TImx;J=T3b|_9R#CFmtu0i_Qq&y&I(Mnte@I z=Y@{OzOI##(npD`ZxO~Ug&nc#5qSfT4SpX%t&H?4Dd&+r_BqBwOT5Z1&m-vZ=Mboo zH?RSD@?f{R%(UoES7fMhlCno^k`EcAzTa;o3* zF3~@AqF5tyA(WW?Iuzd%N@2x6janQepZL$XAu<qUZEyyH*^gs2WYV8{EsnVPfx z@Wj>f4G3wuxMy;*&Y#dVU`&{a(5i19kk+n*>=9B$mcVjXt)c|mPyAhj`lf3yVKGHb zRG8Da!h64EJeuLF);D+=LC-2XFV}QDn3V%j%Y!p5v#FS4c2t^?Yn&FBxBVRV<)8TD zhv8C7U-nVd#M+cXOfipHh<&r&C7LUI)j4+tdbn{nIiNpJyS4_=t^b{aBnQg%^8Kb? zEt@=4>s!Vyi*&xs?`e?JsCew$PF06Ho5C;}S>meSbDWP|X<|Q2XQIGXp0KY3YI5=K z@x>so*(*;O8j|}ij=5fSKksJyp=>TsDx*Bc4P zKw`V}!kGji3P8~xmq8Po1LEp5VPs+-;!p;>OO!XllkH#fJ{mP6kkxIlSZlSHAVaE0&3R9uGDFC%&IQgyA8Hn_H#I;TtEQjMyAF^8M3$>Ut)9zC}=XJBE z1LevP(2M^P)r?(>;KtQoui6i2i&f^Y<~TGQTvT?5JjbpsE_VjRu40w%RC}AZ&)*k9 zL^=Q^D*c07@qeZ@d*1MvQ9Yo_rq!Zp6)HfVHVut1{FJ)r&s@Q@Fas(KTi!#|044og znrm=j#7JrBao*+OZ?5diBc)+xubne;ii}d>qR0I>TCU$yXe>+cn5y?4L_e*uK44uY zEeey`cW+^iP22wd>wZiLO|1JdPhSb?@B?vYQ?HZqB}qvYS=g5De2@B6IA-4I?Zu8A z35$;nzT+;7GUFL0n0#0nIEx|znR*JvpajLpr0%BOdP2C<*%ZhfbL~z(4_&aO&_KqO zeTYPmMq}-q**0I2k*&N9;~!pGni?PsNo`FVPG!NW>2{g6LBv*A8GRiy@6k`@=(rCO zq42ByS33V!A440swIJ)f9PX%lhg{Fs&Hlki2wtE*7bN!eb3{>7N*ae58WXGh&7-5` zk5wKhH4Apum6K#1@pau$=M>A8c$S$Gre4o}GvwqQ>w)m?um^eJmo|QwipCfOVSUJg zZnwl0kCTI>jz!Pr71nc)C^%tab-z9jKxj%=4;P@6^Q2_X0SCy*QwuR@J|O|!UXQlg_yAnc#n?U7t% zIHs7nWYbb5P&cS~P#z@mrlmq7Fk%AZ!+*LRgdxtNsx|B&Rmu$g55L1u%)j~_ywl|= zG7tZEf}=odH!7EVjypjL`Y3-5X=j$6um^Xy?ye8C#4|$)=GY&6C|*yjLZ~-{$Vc*o zP4JaGOQJksEkU_pf%TQ-;Z+aRpyg3@R?i%@7oewS@#&Nn=0^)w*MbI5!qg0K)2b09 z&j+y(?n8C9SpYK_#@9XTqVq8LzobUZI%(fFX%}1^dj4x3o99E$w913_pGJnfZQ{Va zMm_*W^IwgO>tx-T!OZ)r)CQaZ8YF51POcxvbEI6n80j_SQn+z{R;Y1Ox(zbC?B2|Y zuS@qKg*9*uK9-9?Wf~HV<{RoF79Xbt?FM~LDKB@|^?o_`ndX1)cz|zQ0dnHm zN1U`#3*MmH%F5;jtS-Av`WTlV0fEhe1`R?3=k2%Mprbo3C z7NjT%)04(TO=@B^a^1VNantRk2NvFWcAA!&VHLus!Huq7>k~UjO}t`pEb2naYv70g zWs~oi{&2eVk;=l(fm5(wBIc9}78iRwKz*LEm~Hk^Yd{$z&tY#)-_7Vrd_{-%hj$vXk8lBhb{wh2csC`lyn*6GzTJGq8gWM_f-~=u0 z`rIDptM8G)=gAL)Kz0-dEVqC3Nc&o=PazUaalM+?-uOEj0dMpepEOJC5+iZlJx<9- zU7mc=;@Ho;YrmrriuaXcK$i7WOc;b{R(KnOb@^W-rvK`Az|jG%gR6R|FCK};95(E< z?O=HSAZ10Uq@Vd%^)tG)o=GB*5VvQS< zAj4_-ZbH~H?imf@u@aJEYHl-7{*+~fIpyVsx~+(XS@bNQ=((6UIAu#{m=9$Im2cdK z6e`AS@yCmON!|iP2C#RnWQsBJRxm;@O%w3GDZQ$>d8DHA#%Pe~(P|Kka%k3`9Ui>; z=tlct$p$9Xpe8e&<)Z}X)VRY_tOdH9qK>LUW5Y%2v6+srTObAdbzMEZCEfz{tT1}es#Dn($p#-4 zxW%8z`vR`~=&6<-=M0E_(~a_(JO~b60LB}m8>V~hx`#|^ZnHKWqv7O4bXQ8+xTzpW zZhDaPvqD+%r{s_k;uCpWgn_5M06t>lngEqdl>?p(MIP>KglKm{H zxevL1GDfJCv3kDdFjBx-{D|(vPT?qN;pI

27Y^WZFg1J!OrpznAUPsSF|0j{yhe zq5t`53`mRip_Qo#eUPtO;-fRGlm1g?h4doWN?)@t`+Ef?TcbSl+r=Z z)FaF>i{<5@>^s57kMSCE#j1;}oNr7+0J zX87wJQs$lCS`Z1~0p!4i4fugxl~k((M7aYVVl!}PiGLv=!ZNd=z>_s<@d5_-25Xl; z!u|LBL`ggEqY^?0anv9x2`I|Mbi*xvf67ygER(wB;|on!5Z{Sw#jPz5UQjLvY0N00 z;iu^@U->()XNy^^|4ePdDZhz-%D-2UV@DKPiFe8DzJ@6MVVF$>N@u3dz#RjAQfm2e z;O-%>pWZY!dJz-GSO>r^Q+Y~NgnkmIv`K=NV^*_3pxb;`$ySP-=b4g~`B%bu6FM(uhdQ#XB3t&On4u1}*ja&VdFE?u_D%l43D4`1nQ_8z8?}*bIY6RMy?tO4*fz_x5BwkM=Rq_eVghV z$=wFrVyK9DUk6xG|5 z%Wt;^IK*5?`2^yrKf78s@0gpXMZ?{5U^b|&-OUY^))PG05k)`|RGqL6&bNpYG>@iB z-fL!@sGNH{gnMKfdaqMbVhW>CHcPu*S$KoYvCg*4tnOqmeD1M9atYroIrA_OswSKg zsPUB+JQZZ^Lyti{(7Ro`#X9TMWDuD-wXanUodiRNIR%Ee`CsPPrMZr5&pm-%TL19Z zs>xg9eDPDl_tvMMpaY9Ep0+j$=H#?XG`@WQw>Hrd#{Zq_Sj0;=UnZ`n=gYMsR4O0W1uA01-n2XsIUJpi*#XCzS%)@>>t=+rBOb5_NN>OF!|XJnj?IueHEV zNO9B08x&Z*_FCUR|6-f;-1dNDA~E`FVU_HX3Y~1hX8N#wiWx6X$yh(u-#Sj8{uaL? zhE(5onxR{ycB#b`WG|NT90sL*_}P>rsCmHEoqS)ASxR~d^=<_GS#xwUs>rK|*qTBzibSws4$c`EOotaBupa%oXrN*-Qq1AxuNA7N zH}wdaa#;f}pwCcfuI$_B*kwQs^jq9olZa<vqUUow02tFeqWqwNv%tjDBGQ~p6nA|3sB%io~Ch4 z+jJIgKMBj*Dz4|g$&1TYXf%=_E21|kYx3WYhkXSPI9^RE$lj@>DD#_uM&Ufk?N|3qp$2Qo(`jq_#zhdprB~i z4Fo`t%OFnsP<+GbX#%MU)^FF}T_o*90h*Ow7tZ_-7f@2PG|S-J(mWCA1AVu%UxImS zpnaPMcVjjwp-cSvt%RzpAeuQeL+g#5OQ>hknb-k!FEGuF@;Ly=o=q&mz<&zZFXIe9 zCgE<`oD>y3H|0Nm#sw=j@>Nf^@#SH5wCJsw8HAVsZ2d3cNR-w6!ka_-CDPP<=nWwC zaj`)P3*p=zUH!7KCnXPy`WnC|tA^X<|7DY+Bnp~wfNQe)3J{v2*A;i1RQQT&VMloXfwYWY>q>sisirIX2FAAGh?E(dJ79!Zn41W)1vgG+Vvh@aXro?FY{YcSw|!RN z7yk-glKkh8!tBeWrHAq4PCi~h{K!qAT<*wMv6;*F$rpZj7%KfQK~hbhF3;LhpfwkPaW!ulM=V2EuZpm0Nwu63rgXupmuvvFn+WfH8{}iG{C3xi2n43#BH+JUp zH{6=e&ohgr)2R4OmmpnQH&hH_3ZKf3sagT24iGWYSu?$Y?7tryEn5+%w7Oy?1cZ`hQ8b%zP~h=u95Yvg#YdH(s*AInCne+kQRdZ=8*%Uu zZ|u%(0@HaGNL1DQqi30U|{?(DSYV+nS9aSD+V$zw)~NoYBD=vatA*BMP<0ve)_{jUPLC2EzRdxl zKx|)Eg3!t^;2q7yeDJJ376^9K;5|R>@B#2kz}>St8V&riNdRDoU|bT@Nb%p^-xs3b zx)BX;OBEq0_TxvVS-h<&WxvCyQBoFr|Mw2*3sb>PATY(J1VU77f1Z#PQGq)7wJ zLr|b}Z7eudM<{q43x9N$Qb0Z0BL+Znk8*&t5##tjLUX2mw(ffy0UC9dVm64hQR~dR z5Q%1Hv78s^%)Dbt9PrAo46Hr(9H4Qp3`?Ldp^nN&dZ5aTs@+Bb??Kq#1Ev4d=Qnzl zqCy&VeYJ)p*p#iZGkJFtYlODqEjV0w#=r7b3sqG~SKqCWev_UMW8)Kp)P2{AnG?po@+z&3_3h6K;DEZ(UIu3I_or3*xr zDknO^63hN>2su9{}+fS>no$Mpo4jBJU@3wz*6VS8yNmlon ziTRm!W<-__kUufhXfQj(x%$Yg8s{rxc^PEtjiktH`-$7-0S0=VA}kK1!SPNo`+|In zKKp^yRRG)+nohq#!#Fez*5?rLz77Vwpr%vb=1b?Kgx%M>e{~G`xJ~lR44(;j(vAnx z6FT>m%N#mxdgXMuVqF1V#}m6Wov+&LvKM5Qb;2*hl`tcfUze^oLAMS2qdJH#Q!U}Y z(!_NBydhv`GTf1E1W?XrAqY(>o0ArCuYzBWdIXDs9_CIl`OyyDh41T%Zq_3csMJpo z@yni0PXNAlxN;0NyxJ85P_K_mT9&-OH0o_73Y^RIV{Gs1DJd7 z;hh3Hj;scxZ&#o1d67OJY>HC=yBO`&pX4+FIAF*E!{S$sq&FYszYx3`3PI>Eh8`7J z)Cl$z#I+AE4ES%$tK`j)r7hoz0rV9>waGknX&9LG*(;B4Q^l0^fNcf&gI?MlRk!{Q z)f=vRnU5#U?L?ZHwPt|ZKIkLALwHaA#g;qn^0?`xQjj}kS9;w$j^%xt^sd2H|7knG z9+x_n0_8?6wg+0Tc~O8lJt5Z{TPWUAImaCH7yuhczf`=)Phpf_uGqDu2 z7dS|wW_s^&*bdIgHEJnu$WsW{hPOpcs#HJ&fKPA(R*+-% z4RFrFn$dK>K0m!U`-2(KG#{NW0wA|p1AhJ16OIE>|L7`5Q^2^e65qt9ERO(@?Ry|d zIJwL~*_*#GTDBz){3fTxc)5Z$ECKeq=BrFo1qwia2q>HVKSKjkeP7mI_(HuQf1EM; zH+0u-3E;EeI*xJ#`9vh;v*QV)#}W)a#F9T#GpK4*@Sjh7lmkRUJp_uS232FjAkJ9< zfL~-PXiXRQ*OgXls)oZk^>gXVS;nF#-C9Kf%=v8^%;e)>i3E zH~Q5YCVAa=lg*>1sAzGI%s!>Xb#VG%f(d;z#@LtL?qHFt>$i>(ayI+r))GWQ>=f%= zM{G=f$MO$DP^jf&0fy(o5(>_F{d*^G)s-qx!(tD<^0?eP);g}&K4LnJdXx+4oS7HS zbZsp0^8Y-d$_pP2f)Kv~1qPr9I!;78;HhG*@(C{a&|38|Ia$PQOxb^gR4@B1yc(<8 z`2(8-9ccbSfX1Tl8Ab`ZZ5zLu5hgb=UbR{kL4k618dsWbW2d*01gl+H*}o`XHKB};FFzOV;<p_7}6G`W>T}G zpn&x=_UhZ4ueRoZ!b9lcHzvY;rJ7Tc68xPHS%k(L?w0SuQjDT{fp*cyNek+8Z^XWE z0#mY`aP7B-d zMXhU5MH<)rsql&Yg2J!1#swfJ)bm%{MiXGhkw}!YGjD_+fHB&&SY^=F=d6Pb>=?{u zgelC=?`-$Ejyj{h%5UqIH_G?*#!q`-M@n2UW$hR4yRMK{U!b4E)0Kh}JyWSqND{H6Ugz`h!!%Y&UTD@^0t zI#Ee4^lCpDv)G-SO2BTYf9YM3Yv7P}E-zz9*A4l3t?u2FP?_cKnX=^#Kc}@--qP>#q{Lfr*8Cd?{<*u0Y0y}%Q{|#*i^QT$C^uKC5;!^L_ z|5e)oE~@&-6NvB6>vy8QoC1NU47tYlS!G|9xUix&L{(5q+Tn}IK7@#w-|KOvMF9WYsE|2Quy4asOzX~5ywP8NUS6&oy z=>XS}stpe=j0V=FzyUqCcRuVL|b~Fyw8*vROpOnriHRPFKnA; zhy7g>Rtglbgl)Rq`SS z#cM;k%y-UN)*Hq4rem1?lP({!O7o$6035JC$Fu<=gg2^Qdy&Ta|*JsthFz1 zmcn7C(zYH44h~HLxbo?^Op@cTG2%))FA##;=jiuS<)tD!DkJ)35mCEaUrx`NA7%L_ zPP_I{>{hx~*65PVe${vHC{5)<6iSvjEhh(i>hl2TUE9oeRN}mc<~e*Upy35a-FU;N ztXOdZ*x}HavQsUf2=+s%%E=v&Iow;E2X8pDbw`TG-LjoY!#z2eN)wnQ>rMEt?a#B3 zw$^0o^OR@>qYQ2|HI+`B4+<(f`G2S)EU#GNYN@ebtK5qoQlhw?EW+{_bLo1OYN!{uMg-KSNlzy{`yls z6PiEKVKH{c@g56QxSw@B!YwmTp8<3hl>@m1gfQ%G1%CVLvuW*}yv`DStq*j&o6?U! zI8Mzy(j%7+X|P53!DfyHNUMOZ1ZfG(pl!Za#FnT?17K7(pY5ja07W~wmb9OQ*g0lI z@$7}s@@bCccv%5%oDb*UU;6FJ=Gel{+?@z84Z9T^31dC)rF_VvDt)%n8N+7;_%lswcCvl;!XW` z+-9F=NRK)Br;yn?GOcDd=O=-4Fe&F z{3w8k=Hz3!`naax`rRWy%c1$3=xIpd1)N!k1iA$u?fBdeNeNqLHeayH(ecDH#F=tC zgFYy0Al4=^3L0dXAtkuut51ooDx9nt^x%;G3a~yY-2T;^aws1^=G(N(&Q|L@YWQctF6&F4XWwNa5COyY?hQ^it*#cE}+{skeqjpFryx^JJv zsovm}6#3&R_Ny<{^D22zC;<9eojB0!kt--*hyoo{j6o>(3CZh7c)BsyzaO#m1lYc| zo&dp1047oZ?8r6HFTw*nc0L>?OI>ws8&BHhM3XWm=;L^-f>o%a=6I1mH;xg&+Q#=1 zKwjHC@JsEu=pnr{2!`hWU076k3v5sKbBskl*@fTG7W{+M| z!pR)ikXD+PIdLFAUjdB2e27)HCrlGdT<`nxG{{;AJ3Sy8s!^r_rUs!6@;hoNNMSf%;1xCm>f9S#JMm-lfTrud7s3Cvh?yYL#3b7 zk=19g6R?d$96K;i6iT65Z}U*}FXtY3IYNy*x{MEncno^))wa5B_ltLmhykD9J*Qnk zKY=f+%&tw1`G!K%8;Xr3;dEcj@ZcXafC~mza@?O!ke0OD$sF(yh(r(Eu2SSmXMQ&5 zC^N?*<)?8U4@`}!ln389CO~k!=)pD3%g;-drap9)tp1Z{>_ZFIRLERii9cy&bFsgT zDIv72@5>E;bAS-%WC4cwC*QYZam#XdM z9A>_C;lx>=d$-oEo#0sPy_bj!2*+Js)m$yPtoL1^#O}+;68iUW#sEw#|=9+wAHhUgvC754x!+QF>ycl&LNcF+3@z8vnE<&_R|xfnDI6Zmi>% z+e>Q<68@x83+9?*LUk8t++U=hC92pwxuN^X-7QXhnmO%e;x{CAZ#u?zFRH#phANMl z4sd&3s3F}q$e}-rzGh8kHs?0OOHK()d>8525q|ARW}aU0p?FTR(0tcy`DD|t^KP9m zD_k=mhKLFd`{78KIIiYmfWa2t@AW+MpJe17;Yl}UdtEmHC z{Uf&RQ)+1`d-cJ-FNqZ4PiPZJd2kUlUzXSjbe~3)yU~GGghwV&8`qG8siUO%NmZx9 zYTnNo87rtbjb@T*Fg)`MeRoQ}KmDD9rY03%QU0+ncotU?TXLnFZGZm)_$}%+C=+?M z>hGyHk!+q=+g?npA0h`B3udo>?SnY8J(iM-4bH!%`Ro+0LO;y_Ck^rVT=`^YBX?Ih zXLa6vu`xlD5J%3&18=fDC?HjE-De^Wrn5)ujj z=7HVf6x=F~d6OUGa4vrlX9yZ*1>NL+`!td+MnS1WtwRr-*B2jhSyu1F#N1{45#ja| zVrwLg@`sh#DUQ3$EjfZLW)%a-y>IF5Q%jFzenwcGtx%J(A>uO5SYK6wvap4`eOt!_ z+dop_xhBC|VLFf&R__Y($1Pzzn1^Xy?;7eD6voEJ2hFpFZ}X*t+}uyS>FP`E`blRVu0Z>uo+;sol)Kl4}cW&A4&ec?bd2dFRh> z*K+|iNC#s&a65+XI@U0H2;%(QKFQ;??X5Fjo99OOP?L7-iTwV9eRs$S5hFVlJ~Wff z3cN25%Gd9EJ?jY;uPL}2^cg|P&4HNrF0( zn=*&Vw#~kp)o^!{4SJ!w$X!5+J>SUTVLzr1#frc8(9_7<1KOb0H7^+Iy* zH#A%#jCoEpBzcb0ef#!J{Du-E{J1}%Kmz|-AmxM1P<8du!&DAPfrv(*w{Nv7`!y5CFE*7!4#iVj)ia0kOVC(3^yie@i=pFC3u6*b8P+Sq; zq3ts>Wi8&`*!!XwZWk!C{<=pr>P;mT-&s5 z!3TdcLm#ubytWhAA-Q4jxmz~@T~c3|I*t&>Mwu!;yp&#ZVTx5wJ3Gn4vidAR%=pO*YD>p)>Jn=)1A zw1133O|FVnlM=b6ItPA@o}z{AoH7}$fyy&Burgh{5(52H<%@VB!069$Fh6zedv&#m ztPRIE1n9$Bx!x{ZVBN#pAkO)5D97NWbARWBX4ytNfqMp8biY%=#EdgHSSQUZwc2$A zc(SGJ5V7JRldDl8*IOQm$9L8hnO@01PvZ>l5m!WiX4!YkJnEo0eoQqj;t!H*R(S8E z5It*fJh-x|Q*Bjiwcli@HB=v(C&LwOJk6Tvc;OBEr+dFsHotjR1EWCW4$KjlECO!Ki z^2+4lXjuyBZxV+%Uz}2yh}JeXKw2@|;%C2*M)KqHN+P9W2I_+suQR6OdU|^5x|0;z z`j3uX%F)UqsqbK5lrvzF=qDe*x764cF|$wDs;LgTd)aKDW_Ns-K-uVl8`^S-h=?4y z)qBLo(GhgfA)nwaWi)$!_HvD~##o3bi`pGw!LglmJN5CQkObzk7|DTEO+g0+vuxo@HhV2s_PtXy{$={~oY4DM^@t~L60#Ch zpk_6s_={P6r1LR!-m!$fGTZH|U;6g5moDv@$-ZAI1=Yr+m%gjjv?jhQy3;-&jQ5d- zmS(c3L(G4B+?TztSu9tIv7-JqOIb0vM6h>EMg1Ox$(w^Uj<7@+IyuCC*@3e$1sOs^ z1mmq@Yu8NmemJM)Ery=Up5wzi8^o>mOJkr1qM9>S(|>TC4?KB47u@F!+Wz|Ka{A&I z1AR$lx641DF^AO&2i-(AX|YeW!?y^R&Nt$6yxUz8s-G7xF)^jxq)4wYPn1r}Suu+_ z$8EZ?*knt0SifHiHmu?Cus`ug#@faQM*e6gxtqQ8C>u;HD_<6z>P1cw-ggx>m z+uF*#k-X`QO*5U|GiLGVf>Wo*dlZYSM?qsmx0Zk__=56hw6El+ev@WR(CLKfdAMDW zaf8AbYJ*i*5IWr&aZ?24(qNmX5d*osn)T)3?c>J7wF>*ooeZPy17n2tj{M#m=OLK{ z(A&{tXx4%`qwn6ZuG3ct26Dh4Rz-Pl3aGvk@|RoiN+e=Sf#U%xQvnT(jQC8CF7dxw zF95A#?0V_Nt{tFJw!XW>92(FzpUeTpIQxWfI^s#SOLQDiE^EtaZ@WaLOgj^;r)XCZ zbXWT*x49Kpmo2b*4t^%5u8Ka(tT|=fo%r(Q%bU0`PBeBW1OLGf#Tm~Q!5seR>-zebx>=G;FpqF%%DV=8lvqMduthx^z({K)p4&9azD z^xkWQ1tT+vgHQa6H!A8J+D|L{md~KA9B2TmLfLEUo1T!-`Jj97sg+sSeh+lYe|th| zu?Ww18T5T@e0KYQ(7u%{i7~X5p$^FAl2dPQ?5cLkWhVJbLvJC1 zOX-wXU1Y9bzcuz<=81L82b2&U&rkYZ7_Wb-)4AdDdcyE`W+T_K zI&rL6z3&#^Ja5Qf7A%yhd2H4GyQwqH%{!*tj@=Swp-}0iTyoIRc5sxz1inO?G&!9= z`P+B97UAf9$^^a3l3CGPE789dv%Bj^BoZF@s?xgsU(CH{P?KHQE~>8zBB2OKhk#0x z-g^)bP(YBbQbLiUNbeGYNC^mtNbe$21OXB05Ly7~Akw6Us`OA3>RIvKXP@uvJ^ROQ zGv5!#A;W|xd7ib_eP8WnBDsB&2A2Wf*fY7(7-)H}wzd|c`k8@b>)@9?TG2jb#HIg{ zI*3@HqJ${>m`FD$VL9!U2tV_qR`X%kTA*Su4sXUyEyV`K-ObyZ_wAJGN>TsbmJ7ib z25b-yI=(;h<3{=sU*x96d~YTBoSXOT6;+Rls;WUsT;^0dm8fv!9P#7_ z!os$Hx5=@*zH7xe8hNiyQCVl(9g296_0g~4^}KP0+xNTHSwME7Uq^Aj&5xe~^*;RoLQYQFMjYndtFu#WaZ)(D< z@mq1i2b|o!wicuzO7h}bF4JI2kIOW2r@P@)b9VlSru}Wl+vH%?;9ZSAU9=IrWY2^q zuo>^u#Cpy*$x9S{*|%<5m$)DCAw2C_*_rb0bMB)TguxByptwJ7(QSV={$mV2A`W-t zS+cx8!eAsc43j!(|7IMJoJdx&7*FUYhf(z>`(@?HRh( zyXvh{X;wnl1O)}Ra2-&Cdh1EQ?&4zAE8`2~_lhxWAr`yz=&&Wt3R#> z76@ex*Ema+EN|w;4Z*8edidb;+=OZa@7*y`--D@xS@1Lm(>A7mykNQ6cm}Hn8CI_7 z4Fx7@hn~ItT@`pB^yG*2oh!SWQ((k9;&j~Jb!qphj9pJKYH6Ul9pb}yjy?3uw+BdS z#&)WoQP&mEbQ;>T7j{>DxnBZZY&VH7)obmzbsTI1Elf+veGw%l8mh;=^}t+#!b-gB z&M!Pnqdgy+lZUJ(_>wbWHa*oXUEMBy#WzfHG(NmmcS<0r$q-i6d@V;U@T*Q#tt^`X zlqbP@Bfl>wx!9bC0hWYoNhC)*c{?~uS9#9n9O76y8N*BKEu*G2z2kL-dFBmqoZt1U zubrC~sNjh+t4T&Wu+xk+@f>YSh2Is3oE-_L#F$&lhk6N38Z%?bxFZpE4}EF=E@>P+ zVZ>;%&h-& zB|)Gi3Z;C7RhIYs33FnscWKz8uRm)(qT4$!J_x?jrV{g*2XoOyCAC*hw`R^Lgd28g zK<3@A%P*u3KV;E~Wg_P8@rY_nCpGp#=6=yGJ@t&%{DM5tvAckb!H{TN!W*yHnX}HN z(2>`Te2dS%$>u99ei+j=*X0^-cUov63u9&SS&tYVsIkR_l7&RZmOZ zlAZXMo6~#`5fb5|AQ9#6M_yXkj*$^}7Q{MRQ#|P5qV`P%ert-7sgNt1c!zJc;!MAA zYO4ElZ7-@_ao4|9RJKkYE3_5#EQ!&HUZ;t8OU2F1fRC(ueyqH@z8=(`#|_F3_C9#E zB_rE~FTPLpW^Lfds^L{P2Ea7Xi}QzK(XStvFidb-Tr1D+{MB`6UPpG#O!*J5i7dj* z$M>#e_#l0xy@$xR)X*^B*WRz~DI%rsli(nTakJv z7&5OpAx@r9A81%sJ*D0p+Y$TbwU2L465caXvf8n8&~bZJ2Mm(#LGxVozG));JnNjojkehM}iJZ z6IW&`Oq=jJd9{q?x`-wIsax^7LbTq4Uf*tA&DV5_bxi*XM@Y;LjJ^xCFd$SHsi&>{ zu#7T~Lbu`79RpgXAaEg5A=+;6c*Klu8dsJ|eb)%gbiE1jvR}Oz%s3$;ZZ46MtLl}# z2sXR8k1~cgkp9~(cy^cgtkRrc*GCIL7N00!g22Cn`io9% zirz{V7F$(ck>ebdG%B+1_%*weax1?}4tG|WbLWY8dAQ-)v^uLk5?|`>IQTPs=UuMz zo5vBd3+QQ_-m^$ES_txoPR~`3`%`ONGY#Ai5mBYx5Nm!>R3oAyPmBO!(*#Zv+(uIw z1gs_4Bn-yJ*~!)Dr^r=zC=aC#lj_O>c@15v&qIk4W75 zB*OEt6P3!oyQALK`_#-BMIuv4^R4LVr@Lt1&IC_K_a{|9LWdZ1>-MSCuo26UV0`I8 z>wd?+-l1Jxx!s}uHXClobu{%)arbjdin;TQoWaRcSl2iPMA^r*Aw9{RaUhcEY1Wr6 z+J0ABlwh=uP{0S1RtrF_IlaoKx@&nA;Ux*zZo&?!O+Tdp)I*6NcPTbzQ>%v~Qv_D9 z!{Yi3E{v~V8@aOJ@hNRb_tlSs?`PNXx*;l5&H>7cK4G=IY7$zwLTJC%jLpt*aS1e^ z67?uPq7_9op0e_w$bp)teBfn-EF{=hDkIb|kqI+XK|zG=u>B|&E8PGnD_dsRlQ(Jy z-+5B{4Quz7-B%Z+$~O06W)DoJ$xM#-psuwg5zr5Me*o;T+5D`0JgP=jL7gPX%jJhK z+R?EIuZ5F|%I*%SQC*XU8&eZ}m#?(SV<|;5-M)B-pUmyt>NxaENQ;4L4{&6rMk&7* z8%{Yv%>7$P3k$Yj7A`VjirQyn8VO@Uy9E!>8MTi#1kCB;qa}scMy>7%XEDQB7{SY{T5dyY6>|8BpozByf3BB{5=H$;JW0gci_DTYT0_S1vGc2wm9CFY@y9)>Fpj zP~q;u)|h8HNYQDp7w7Fu?uN)4O0VLnc?M976mz~;G2fSXZROR zA&r!(cJUFaD94LQWRM1_%r4~<{nyr4WAeW~q1NpeZcah#`7W9&>L8Z?gZEj9|tq za`J=n^8;QL)A>4nmmB$kfv`|>)nMPrZO%TK=t~t|XbW(4Mhsa5YAPAjif6{fgv3o= z@QW)86Q~ty<1tBTzgFRa@1N=so`P=UoW`X_Wh2J{tep*0^jGPSAa!@Ni=-?NEUl~$ z+99p3`)jfVSnqJbwjTgU>Uj9t_Dfo(X$bjA&epzPD@V!G5P99oR?}==fpHQy9KuVD zz;)nldWOnElA6p^Qy~2bPFn%J!qhxamy89b9j%S^c5tsefDTOFFkd{pVUgiRjgRE^ zx(w#E^5^~iS6RC5o6i;j=nFix8|&9*?z?YtY!w41Z(LkFM`*vq5@}Y_r2O4uyng!E zOV;sZ-5mRIG4K3MuESkZT(iaw%{unJ2v4ck4Cj27DvLP^2NTj%GF;~{_wK9nKbqcJ zgl*g~!>_JAFgFKJMz}m?jIsN@GJsZnC8Fgqj$z967rg+K^TvvL=E2Z8pW`3RkU0rE z6OTA@cnR%-aZFd^=_XYMm^sI~I+cFN?6UNd$AhFK7c z6;&!Y>KNtsMbvv-i@cwTS}`;cs=zp|BJ@s6;qmfN+T`#E+0XR27NhNWl}AF!wq858 zk`b>}wS{OwBfaw8toj*0LPK9I7J=}{`Qiv_kf&C6dOTCA@5*DCzS|AhUZ?a5umRIN zq{YK^LC7*jB)o=~R&86);`7;2R$x%w00(+bYKY?v*?N8btP99N28l7|s!3$x7L_rG zqMJ8DRnFlf>A*_07B^rUT_S~Be0^~FauyS&zB_~oHO2!op(8)PAP@G3yZH}a(J2vB zngvK>V@l4Ik$I+s>(vM7O%ppMYr+rUYy87SEu2JB^DE`vX&3)P#FaX3R-k(kK8SU^ zg5&E7$@o;#uG6FJ)KsVa(duSGXu4lPiIW2ztaLMoB}SrouOk&3ns>fBUjO9y3htf| zh5MfRNZ{Nk`IptsQE+Z~`_51GbN-y7a!j56c)duv#YeM`x1yA>y7z zwN7!1V!DY(;0VWbXnVz#k;xe*-L0OA-G2jmb%t)AmXeg>e;(H?ps$n{@6g_Dk*d)# zq9z#U$z7>4CLwO^Wx|KlS@J&glprhtV(knO5087$7%oIB&~>p^dpDHleO?WfvnXVa zKd=O)(dX5S_=jRH8pno>EMwq^uy9but-3s?Y(_?@J!i!9ulY0dWx9xoHxgQXZY^s0 z!18!_k|M5lqsE@@1pfsOpay|od1;68l-GpewwGf15#ezQ!V!AJL9uT&I~hWq+%pQF zE%L;X{oeGT@fbU~>DOle_qAo4-HV{~kXlRaI94g7JN>ZnAVbjhI=Th%M-xg{C9tPb zb-f{8Xt-^e!n!!VMyp$xH#hQohwf3eDuO-$l(l661#fUXgB%u5P%G+#wk^Ay8!V|W zZ}c|gDI;PbbDwj=lQ;y*&xe^_Esy z7z4eDC5PP84QlDu1e7bPdO{*xx}8Bdqs-bQ|A{%D!ML>`MVRvOoFlY(v$Xa?=C3oM zborTomgviVjMfOSWw}(9CMaKriVss*SyQ%h^j*>8KF03Yw3Rf^%@^OJ_wJQ5To#5l zzBUi;=y%iO{s3k?-$@zJwk?uCloc6Vxvmr{kI~PQv%O`SyS(%7)$eW3SV#5Ww9N&wBv`+-Cl-bOLJo);ZxjADQO+I;5qS z#H`p0(=sbwyj>b{9H@8~%H7J#{0vfZG5wyy=dp@$30+cebL>q9?ZcOCIX?kTey*rw$$XHzJ+z&%w1ew zy+(KXOHMqz)AsuHHcn%1KE;{)$_u8Ty?@Ks#@Xkef8W{W*Oe?2*^&ODbOfYM9mJro z*ahMX+<{{l0#B<=xk(k@UHOxwLOsXpSV>p6=AO`mKFd$$=vyym#)rvxYGg3({?lIt zE0l`|3oa7p$wsLnN+@YPWWaY;vg1Hqno1(NL2VDJ;x>u!v?HsdPg2#Vcg2rSlBDo3 z*5Io!)~KGi|MC|URyU!nVHdIxq-S1x8DCBYPqqwt`FdGuxGF4)2YIQoZc2CBntnci z!}opnMfDVFb8a(3Yvd_&bYVPzWe>Hz-YLwjI!Y8;!#2gPM#zzraZnMSMXR63mjCPp zy~`glE3fU~&BB+s;-9?>s_75+GD-_n2!+bpY=p0Jcd18vgmO=fV>5A)1hJkMw)^LZ zo?>I2DL_fOR90L?57IBSl-$>n$9OnU3#XCbG{dYZI!tJ(!Cv_*QQ3FFkaW`Z8<}&F zQ_w-(hk?KdHkDhcG250W;vVE%JA-g6mB|^~sdvYXtspB_o970_3bW*1Y9-JczL57+ zxbK%gXDbU_P*{>r(m#8N1Nt&vZJjg{@a)*OpW7ZKwx_%py6NT?S2%L}(xbMQ+qef) z#0+p=TE3iNh~*#J8KGYf0{?=0m%`h)Ew4-UA;JTe%I%N!Gl}Pd8TSHxt|n)adJ8$d z(fYB_5b7bmwX@R^_LO|J11i{zN{&>G0>g^?P%>|XX!-fT`f3(-y71-9?DB@p2-xe3 zw~!7O59!-3a)si#V(m_n$L6eUNRP3)KCPRjmd><}tFrmQ9SAL!M@m{fSPZ8SP}6?8 zb9}$33lCt2jP4L)t|ga;M&B6>q%~5M*SM+hEcyvMgTw=!pR#PbazV85{XZ=o8M;xs zqM}!6&AAbCb%ibX70NO}3L^ux4=<+3y;nR}z0=;@8!G&C{UmK$zHU3ze5gsn=;QtE zsXUt>-!zT(vun$}9@wa-%pM6oaj&GpaWG@Tg3^1QSb190>wt17(u)&+grm2k_MuQc zZw|6K1_QohYc*LVL^I;wgvCu@&o#4^gR1kMRDq_A9A(Y$!_6;9DR>rnA5-Q z33i_4*2a(WFOGfv<#0s4z{sVD#>8?25lQh#*^1jVOSC^pz*Hf^ao&oj@>^uVlf;XA zR(RJBznuW^tqM;f)olQK%JwYW)_w+Fzn7krJBd%lrcDA-5VR)vrZ?zifWQ zftVT~r#kyVG75Ekh3Y?OQp=f6A33z{if)v>R(^^Oy1e#cMG`+W!~SwTW<4}Bi1Xo_ zRY-bt5#uS)z1;x4Qe7AcF7*j@5hd2#QsU+o9EV=Ht& zS|C>=0oRF2_6}dveXiY=OtV9N?++(#Fp@*tYP7E7E;C3ZO7iaP(h zm$g_02~!RT-(!&@eVT;-1E0X^qCn_%hv{sR-r##7%z^vKjhwix4UMx`X`3$~r|5M) z8Sgn(ZQb6(b`3tokv6%^tY%(X)eGRz@wz;~0|U~De>$ps!g=zmTlr6Mxvy_;KSlWt zi-g;Nf>5D{ORG3GcbE5_u-T9D##IKP?jHM;mBa6-WHXaB<*v2Y@$6}{<;&Z}fK5y6 z4vDLMlwuzH!-gF3?*~LtA(}ai!*F+^<4<3Om@M4=4($v~UJYC7;~ewJV=FnX$z+!t zyyvW-6Ps09Z=u|!(qox*uoEvYt}q+v~TOGKwXXYNaJq4o>zjD+{_l(J0mQyZFAUTtgFj@Mxs^5pJFC7v&WR=JqaD zY<-ZsuS$(iu4OXkj*l{)rGi!@?EP1wJ~hYJj!petwQ3X)h*(cjiZZwMW>}5+=Upf{ zj$p!Q$+WvOttlTO0_EW3U;U&3lsVS$KlVRZ9oh&E;)?T6NsMUmbd1+>o5e_Icw zH!zIdSjEK}_oskzw|a_yeJei3Z2{<$lxnb&CV5(<*AIRU^L_{ta-!CTZBWNi;Z`&s z;p?SZJ497uMSf<;ch_jk%x^uIx(m$c;sXuf6zpW3;8af{^FZNq+L8E-u-SJVjaIqD_vzH6=O+g{V>2yN!1f7e&4M)W?JN?d7|T;1?x5)Kb4ut(uNYC^foG z;JZ>M)kNd*QJXj&y)C=D?A)&J$y$VVoh-^Pv#4*#AI4aS0PP>#9Us$d79YG?Bw`9hh3JGR}lBNxC zY`k)$lx#vx>9ulLu5OS`*^2BuiTG%|fJm7sCw$DS&nYDo!KUu$-n z$S*Lov*&4h?PtqZV3bFJeri#?uzHHC8F&02#ksnBPfOfkY9Kjnm~Y6xCn;CLF+-ReBxGOqQ z>N`VeTg~F>!#;qnt9!bIKjIN4ltd>I>l;R=Jh?2qW=`%oT&T8JqVn1ec+tykr(hW= zvQOGq(#_=p@s%UC`d{#F^8sWzB?ZJr%qBpVsIt`QX-@qtTxeDpnJAKJnqzQP0dylW zouF+FFuP4;YdaYDrTR2-_*-(FGZQF>b1CM1`b5IzA;HyP&u@(o_!w17Ajs@eMyWdT zLI6J9A`VupTq_6i1DfiYh8;(SL2%V7o+b9HS*}XsX+tUEa%0W06u!UPIaGW(4bx6> zx0&zS3p7{KGt=KX zNi`LL%r%SJQCCqmmoxZdg&nBjYr*q5;%|R2V&srzfNz5PE}lEy*fzyVpLvj{e{$T; za4Bj!6a-X3R|uaHokly+?~%{LOUdTVnG5e>uPPE6;;tX^A$I67w1 zTSoh#7|NP%EypRF`Ky4}yKK;V3)5Vi7rj^PoLH@G9?ID>Hhz-)#i<@FV-0pEWd#XM z65OW^Ue@2?O?w)tv!~)2Tm3tC@3pOJIZ(=aIZn{h z5g#?ui@^Im4yBi;ECZz#4hHy9L3pg>&k6%CP!NF12Ji@n)T?r__1KA9Lx)ZYu`WDI zKq-pljSMXME)V+@waR03St;S}AZ6J#)L4zjBGpU-sM{Qf>S5rgCKl^Og+HW!XPy zmPa;6*s7!%l1;W_hFyw~He?qjQM1*gsOOzT8DRZH$UUc66Ov0`OICXq=>v77sA4aW z<9qHzZH5zHvW78ZI&X3>CBJmNGcOjj{kETYotHwg9++{5*j2nwap6Y*4sc#xh*q>x zbxj=p$2c3|p7JHKB*!R1#t0B*^YOu5Rc73YprM5ABot9>i?T4QrxeJW<}+f(GNqIbfDHHXA^V04a4+s z0KN55So`~mI3|qKyoaln6S-}X|1pB<^AoX21wZ^8m&uRf&e$nUhFWx=0t3sh+0Z1V zkQoakze3enfJ_WYSquXESiHe2r8iqy!4gQ|Fs!wHDfP3FTvO?93R6JL@FQ!=#M(y* zNeJ`e&(D^Ls3=t#$V75PoALACMeK4`>PrLhHFDeQ%5~^W(q~mK{zNM9t`^>om$qwu z_7ZrM|M)KiUOnbf$tJBrzu%Sz^loRIUA_K^uf-vI+Omt+iHw(w!joEU);Y0ZQ6@Vk z0lVbxJ6Z~t8bEHB6B|9fi4_M;aAFDX9rgV8UY;bC8O|t2$5cxbbLps(o4ilb1S&`t z*AIoIJd;UqPFrh{dCCIcO<}?TRK|`da~8H6!A-T&e;8&eBrT6~eOcQFY+HbzDovO_ z9ANATcff|PY|r``4AR8ccaJ&)m4S43MsDP*uVqW2CvpOvw)d15d;lWX)x2E6MD$vJarANx_YSwGgKb%O%*y&Xh9by1HC4;yT<1O}^)iXdCPUN3ztke!?8zsh?V1%Olcm`v zd~nDg1SoOef;ZmbIo_;uh+AA)ft6fITt@s#DcUJ+;i6p^Pvw6v@4j{+GJ4*~Nnbda zBsEqI>k1h?RxvQTG9wl4SElYplC6_{?cGpT$;jfk+NIDK7)HP#<2M-3K6Afq8z*iO z0_aWDid`FSX@ELWB;Gh*&9tJ5ICeq=JDre~ZWER$D>Jd1*m4#slR#D7k~4 z6~n^H_G;d&TET0h-HS9HIaxywt^zkus2mm(5tkZe{Iwq29Wbd;y)`4b;uO|U0zh^n*)+&spb z&YBUV$nauz>6hKuC{5UaS^E74w~?P4Yl>N%>pd8bajA7y{pBt#J;z)-QlAxmupnWN z6oD-S;4eb74%G4rb@?C&=zo{rW7=15dCK9}51#DEMJ?3IWM0zkdNbyQXZG9#ez1J` z)>6SVY&d#Sql6EW59DWgwai?iq*o~BfCYiAaMetw)!mtGj=w0c`%XMyPPO4;kXbI@ z>@up;SA6OHVegU8vN^IAesGw$*q$_;Sot1@R$=?4*I$xCC#amVEgqVQDPaoY-Y%>Z zZxQ&_?1Dj^#n}itQa_>7nq9;rcdrNdk?h)gf19l1In#GV!5I4pS=OWo$cA>R6LL!r z{06~+nrfHoL7rR>Gk;tAAtVQz^M56i5L*a>sV9PW!)^reagP z{|91QTS+x6d)$*c$P+6Wg{Aw}CN2`!+1%nzV{HJ;IJBXe&3cTYQQe&hH|wY6?u@T& zxw+qwEU^V-pl01^Sp*+`d*X)cQGoyKexSm6c%Tm*x3z&^9-kl_&RLQrHcTN3os*pZ zHO;vnA>4Jg7kLW`^9BDYuh#O+6#NppB_ zmb!T1f7*1pd$a+>L>yndir$uF2wIT;l=f&$GAZaoyinRQI9b6VN0~Jz#%aAse*Y3_ zQ|bRp<6WUwx|6apz)=Y+BXpZoh{8@az@h3TZdV8@DE-m3LuMw-I{#A#bDdUgIZ~|% ziJm@TWsQE3kV6g2C)RJgC8Ncz2H^(NGdxBDu}8s-iD@l9 z*a*^X#1D!n*jV#W5j1nn%KE@({=d%)iugpfR>W}CSePbSB;r3Z88iSRREeT;KFoZ1 z|81$?k0`h0`7{=f6cJF+_CHo!Lo&@bdxsAVRK|miRshq|hAWp- z%WAjV3H!P3*zo^Ed|{0V$6tUa!lDFN8qVvCO13mm!AM6Zzp=6VAA0@%6LUGBeu1I<6rz@L8?g1? zi*Yc_oBF?KT1$s%T%xkTN9W2z-njV0tWuK)c*oTq4(_Cga5~h(l#jaz=}WOLYAH>s zdEjRqzD?2pOY|YWS|E2ZxU$?`QpdoYD!+c}ZBp|} z*f}W6#7;cSoaHujF81m})!>IDzCAvG)I3iNE7nU+m!h!FB6LNaUnm8}VC@IteD%Sh zSoHA%`PbLy7w5BU&&wI20HdQ`BADkS8KQo>nyp~Fm_vyORyG;-XoM%+{6-?fN-BKf6d90#3z`!2t$*s}u(!I-7jeSTC zEJ^40rS?_1JIuH}5#||0&hWMAS<24+t)J6A-<{^#MyXU(wZ1!dm;Try*w;{+>pTyB zIL_%^SNY8gi3Ar>$PPniKL(sWB>r$*t8J7>M0$bfsX4b+jyLKeU3XqPCiv(4SF<(W zYq3G1*E)kd-3P%LplmI9YR28k8gzmg_j9hrL!Tl4-a44_Af0wUi%$hhWYWi&YeEE{j{PhP64C zA0t-#W5MKkqF(a8Sh$*B^GrBEIUsYQ^r(ma%N?dj(L(RjU(D#nYG~n3D*UeU9MDYW z4G@i)ph6&G968!wzBrn(f)1e`|MXUL7I3<5^xO(b>by3Vt8XQ&^RM4m&la)Gna5^L zpk;{@%flS2=)bs&P}ro;`{HPi9l$atXDoV$_r9ra26r{F`yw+jW1v1&Tg7DNvIO8& zHwR?c9;@yN3$*TwbK;Av=#*UL?vSq9(de)ax|cN8`BP6H&uZb<&!$PFr)$Vzgt@m< zOQcYKyd3g`2{=b2X4j}qOSoaU@m&3@GUx_X1=%(>3tWt_(ZzSV@|YrEl`j_?###wI zdPby3V-eVQnJ4Cru8-_s-G-wYK#5puO{M?W!F(}GG06p(XFy+_{?Bg!O&HC#aniWGOJtm95qzmxK%D9g`%cAhw2h03T9llxdpLF~S-bD8@ zd2FFp!c0JIK-w;B3yM-wr(i|`Z4n%frv5;Mn*|BC-lz&qL~GGXFt3167Jqc@ur+LM*j zL%ETtWJGvM_npo6<*Eu#88L8{MJIa|*HPhh*Do$~Es=c(8aw(3|fC}zbyTu`Iqk=ClTG>jg8U&_0M zv2Y4iNF4PYB7=3$Vhb1|ybF-Xt~aU*Z8RRP9x3J$7qmM14Q^fvZTv*85p1s8=!TJT?upT;#$lW5&C%uMoFQ1d%yUP_;H5k}2Nx$)P!Q5x)spA2 z5v=_a=IvcoET;qTOfN0jIY_p5aN}-1e|>s>D^VQjad`M*d{VG|a@|QpQPBzUqj$_h z(Bg4OW+uCXlT&DWn7YF=hgV_kB-_hT!osS~f<;^RNNxbwg}PO|Gh?R|ZUVen<0j@= zG>9|Y(Vf3GVjQP8-XXO&cjUcNEYWohDOs|-3TSL@&Y%5e+=oDP9y#59jy%lGPe_W& zzwe+$utIQ4#Y4%n-xS-sFq&7o?S8@-h)=BZZnYhqq@Y@#rixbgbGt{`xWopJePjzz z!seg7k5Z>~(Y$Fr=#@i5zAUaK%aoPGKL@|yFayF!zoEtHug7mbXg^?`b71_jL~5+e z{#Q0X$728#ldJji+(1n_tR_$WldP)!FRTOAfLe>MJ$_|Y^X-Ci#uo{M+O@6@nQb6+ z$eP?r@P(*rT}eaIXNDRaqSkYYRWo82u^yh3T{n$^f%8AaID9V&*AZr zOwo5{+Zf=lCS2;+P7&)_Pmg_OkXltII10%G9Tocm+DzZMd)141Y1rCEGxDk#q&%yl zH)LsPdHw0&+v&aUI@Q=Pmf0@;(vYH(`d66;>k}GTQE|ThDs!eSY}VVH7Utu-hZ&E{ z)MLZ?9mNr>;&f!HQ#Ix3-67wbo|=3O!`7#Uw&&jTEPi4s6sM!UxkSN`%7pFN%W|7?8QQddw6*)QHfA*&qpyEW3&^4e!s|V9jg`6J9A|~%GgHR^| zgvfSIUxbcst^3ThIF>e0!V#yJSWAmTOVOvPkh#sqqw^$D@H-g;4^G-eknk4-F8p@4 zFh$PpO~f{iXKP>HrCE9gVRPKX*zT&qgVhm>DWg4Mkak^J=0FAi+8J_r;oh1^& zr>IU8<2-e_wwD;7P>mrMWYR%5<@=fmlOlOHuUtd{c46w|#(<)T9pcsf+uqRuFPpPn zs?=t?vR1HNK!SPHOqM*dzI}Jw28dKd@yDyGuH?*^2~a>dBY4wtfSB32Wvn&+Qy%zr)obW=_eEjbLmJ#V=cRW3UL%Y>7zVHalzfGH+z+~T zxfY5p8EJYXY`5{!zUdATYtTSQ^Yt3_ft_hGTh8`H4B>11(iI{vDEFom8}?ea^5J;y z;19V|;r{N-VNQ1?iolHZ*NP{*LtA)S-KN>ub0?|?d;85ODDrYC`ta8Ai?qy0^k1~lL&KF6DklBs5?mqX^hKci<(jH7b`0_-T9veh%^tA}Cj zKqh48Os4;IwY60aF(^$^mE{mrL}?y!pn7#l3j6hr(x{Sr)j9h<=~l&wte{nw88YDp zBz4Q6TXu{w*PMrL<(Kpd8jq;obJXQJ7Te}5f|o*KRRXL$WvB;C`|)x zd>(iq+@90ur_jmNd7f@L6?cEM#Lc%K(Wsm2Yzdv^_R)X z$zSqp4~F}mhBC|^Hd?qaR_?gYHl6ofE1RMtv+fje#kmShzn(6eSN5F^A{|QyyWHRn zjoGu~z_SjKEB){MF)J%qm{z_je-QAQ((ZA42Qn6pzLQHX>YIP_M3h<21H);!>!&pP zMOCk{4#w(d0DvX4NBIglF=E2>6eQ~|H(=dy=4Yh3fX%9h_H}9`6x(Y#btEOOj$yH2hk+ewiea zlTRl1gB)IEOtOzik*ht$I}?9vZ>#NR@6|-H^^mW6^OlYauPJF^^BPQhT9@Nl%D^ECw0^+`#MOfyZW^<>ZkO4)bf(mNtsI*KE9-H zQv~B?oz~Uy`wIT#le`J3Mhv?o5+;sAR!^ZmHkwf!lvly;q-m&UgiE- zM$g?lR!^QY3keEM%=bn0HcjUt@k7}pzLjVy9d#b2s((geaJJBZ6-fZzP@MYqPPyWL zKTS^0+>0(XRYI<4z#Bj=(BX{*=^3GH`H!OqW&PNQLr(5&~-Sxh$P%4HnnUZ)AE&(;au#*xYS2GTOJ(A)WL9`+>d6pFWJ$@;<#O5sZd3kjX zc3y6EUhpM~{l#T4di@gU)XF9vPgr&Y%g@egkXE!R5f*5xaW2vib}#Sll-Jhd=OPVb zvhQsjLJy`$e+vcGQ6#b>@m9Y$b-0cHh)kx4+ka<7E~VI8ZcaCBVdiz4DxD@~=YyfG zyFVqKI7y#-wEBb^s^}hn3>%z!$zHxB4$I5uWBY2<&i3&1MXIAm`}def{*=gi#nR^6 zGc&zgqTlFhQ)hLKJ{jaYNVi8Ti9kUUL-&&$n}?j-eEPP7Snq++Px>G$a6#o*6GnoW2-+YIwUWVat1#l*@&!H zo0*!TxNzaZuE(;->}_*U{{bcBY0*5o+a!dOLlR?RDE&h<4*9C`jyf0 zv$hl0nSe78G{sB)#Oc{`AIj?J!UP|!K6hO97suGUA>O~YjM+9GDDC|;AXZLIq2vs| zjW20FO4Pe5HqAR=Qj=58_QKzPJO@O2YCuB+R?E8azd4nj)@2RM`0eiP4T!&47tbn| z9j}HRF6fNf!M)He$NQDu4$q(8%&9IfA4-^!lDgOKdZN$#IC?d61J=sO9#T2LA;A(e z(s%x{m&d9YyON|&a|PI~UW0SE44#_ZErRyHWqG1F)v(oOqSUglanQ0^T7AJ=>EsIH zVE*b1zIG3g&BryHzkk1XY$R;oC^#0H_hD9*4O3zBT$i4C5f5Suz@0eD%oI8_Z#uwR zyqE#00;w@EV9$StH*ZJndL!Usr&6<|-i`k{4+ck?W>9b5=)8b$r@;&N=CRt{NE%?v z?mNAiL{|4(?r}o60nv=C<0pRn%_>Srpr>`8w5nBHSJ7s4Ua&hXjAX3Ao(h3mWSG2c z|3kPDPqR@y@9psH+4U@?=gJ3R0saR7dO*_ntrk}~Br%<2eg4@sX5aCKZL4X2{cZ0e z@3iBsj8}{Juj|rLb?wl=W6}E4_B7Y(C6ONChn&9_)XLcA<~w3ovQq8;#8LH?*IUDE z#(h5z>|9xYV(Nu&V(jSdc3TnthV^|0w;HueIdq>Vi0!ql0!>Gy)!A(&vx2(YB~Tk1 zo08+QMI#5ej%tn--GOjP%Wlhc?S7%#fxeJbxmG%)({se**(_*g$m%m@01lmJm!G(l zK|3V!6mhN$zxY@w%ud#On{~}^-wO!GHbWYO1Bi$9VY|Iek8BUPE6Rx0rUEEl9NYzG z{vjewD{tvLK(~3=)7N|0M||XiVygbwFE4S(v4#dUpLaJ0F%$id@<7-J^z0N%8Xp^L z^-BQU17dPZ;ENZp&1yaj^3KZgHxhCm2i{7U`^-0<=T7v1zc?D21R3Y1d6SaxVheXJ zCZX4X=MT}q11iNxRi;?)!Nc~xHFWmgkzOlrYdER@!A9-_OWw(vHq_!!Bz6$&Mw~Lp zHJKWX_oDLBN*bIa8#T-9V&9aG*2Hb5JFt^A<=@P+`ab@q2YCv0vp7ZF0R=|Af0P!} zPxmCEJ6ip!4UPvjNN1y^r(fD=#55nL)0U6DtB87#pIkexhut%!Y4lH}WF9Z}%w9FU zIfF-1VzcF<#PovQ!t-+}-_VOW4pgOPKleIm-^tacr0Ik}V(!O^PO>t;`ctD;F3mC{ z-*{1ewRScvr}xYJbDnhOw9u4`@}kb4SXVpV+Fig}`EE9iZq)&B@d%1MPYIP5jP@8c zCLB|*(zG5~eY{5ReP8isXE!6w=YGyYUUBk+&32WxTa>8yviJ&=PH^v)gOJA0$+Hfo-kS*WF3xKU+; z!7azdX`SrKRG2e^Z{+zC=e|sN4Rp(9{SGCr6?-|<_ULHljB<~J+~<#o+9G+s5|wb{ zE={i!^w-emMp{~rbcK6a$e+WXejM=591vaYs2*Ng)b9(y0+EeT*`x0sIY>dQwdWO5 z9tTGaO28yZ1v9K1$S7KN*JwOgq$SbsLWD`$vCw>ZY^DL**1Z`2NU~cYWARwIiA9>n;L7)t#_KY^Aji)4& z0h@|n{12Rge2rnUaUCh+x6M!+5V1iC3MFx_6^CcfKaRbRYZqP?Zan^nYJE%@wJ`7= zy*%wtp?diL@b;cjO}1UPsEX2iks2bPAVs=T4Js(8ND+`uq>1$2B!CDKI?_9c2uSZ8 zfI&8n9CCAHT$xhUOay0!W#JR|h5s2#xXqjFA%#HL$`j1C|#R_}svxC8;Vx$yOSZfh zt;c_qiiR~IVo}?GQQC!C=O_H}!)paILg+0+zz>Kr&@~QtK-NnL}Ap7m|yI0B-g@qSHyoi|*|f;jX;r9wby>}L3~4tJ6M|Do^gDU(HhXfOIky!5h6Z(@ z0PhOfj?K9&HU7Ru5Qy_`)I`N1U)T5U)UG|miRiA!mu=-33oe1zE^oXNt!J)7`ECWtVz>1uQ!+7O!6hkWVI{nE*}2 zGSG`MVd&TGU8Y~5&~BSB@2(8!O^^e6765c`a{t4L)Hh4NB&k&_MrK$-H#en4%5DmJ zO2rrsTIJflp*S?z32Fy92MKKXeTFl7toj|u`_`NtuWK|q_tv-&JdZjP#9C>Te`?F` z6!mjL#GcJ7N~;xR2wFFd^!fz^07d%X594D)|JMIoi~+tM7xn)E@Y_oG;>@#$D}bQw zqW-_F;pB&n+wrL0B?17)l2Sj=FNaC_bg-89R|YHz(Si2>luBZ995L1=i<@~D?H2r{ zH9GU&aBo(>11pw<@t)%8m zTF2!75gcJitNK%VxA1?z^2!2wpIU}9`H!OygYf_`Ts9DrUyD_>6U_>)Pm-=>f8r`@ z4r`CXKEEin9C~1^f0gu$L@Lz4%}Zve(0iuqLM4#z3b}ues_5h>}=dTHQn`uJ26{x^3CPIGx5Q(3lY+I&cmjif+6g0K{4J77XfETw)F!g zr=K_K{g`vhu=7a`aii@@fYv#vD1+}Uy-jA%hZ%Wj1#sHMkwEn@*=D&20g$Rv2;D(z zFFes0?}Hi{SwWCh6v_G>dJ7mTX*VG;u<3IXuds1@z0$jZU{yB&rf`UGhS^{~XW5a< z{@SNmp+Z0K?Sq{?$i5w?d2V|m(MJUOHrc$)_QuY z0#bdT$#c^{5xW>VmOp;7xhToB%A-+-o*uimAKIT+{B;P0p|%^A1VrC{(JHci42UeAdIAJMC=Blltc?CpkkiR1i-u3+sfqSiMijZ9hF^Cb`P z5)Jgv3Z(O6`!MbJ_&DGz(yClKoGiUS^)s+hB0FR_Ng7!rXJGWS1Ham8POqA4^OlR> zMpW>uV#C56^9c9ZoE{o-61W@UOe!E`aY{w?Tnjx$>+KwvP7FIg5MLf!>;)b*Z>--YjD6g7r=EqkH(Y`W*sXlS;B)i2WZEO%1!LTCnpGBh|u=i-jGmT z7AALtJIfzSRs}y+n}`o=t?P*sHkAhf0%1qEG}vvaxC{e&%JO(0I~3hn4iIQkqm=Pt zU){IqpRRwtlO1$;;k)BTuW9|-D}Z6f?+g7_Z08Srl=+Au$+SD}FK_EKLy*#>bmusn zOm7|ctfL{AGVWDnpx_ITezxW< z@u7wK)N<)Bo$=;VKMD5fkK$gbC3K{uh%jI2v>}z)lkE7Sj_=-xu;6(Qvw_P#5&byz zM*w;|b?wIfPnH|^f&dwE*Mdvr*iF{w7Ye@rPdcXpI*DC+_p=(eKoOhhGp106xwiDs zi8bk{o0B)PCAbGDHUY7G@(4Zk6H6~E+Q?=Lw)PDH*4ML5PK{@8r4)Tcuiot%6;Dzt zx}oD%$GWX*{fiynw9c%EpNa&0G-X;3Ckg=_Lf+q_2+@?FVK!YrxB3 z$rz~O%VXTTkZs1BfNow?ahUL9yvH;l&&p|qz%TZd^5BhE2DIS3XX^PAKLUYyQH4n^%toRG$~fZ}n%g8r?Z4zIf5VohPP(#@ zQmlcdV^?5y6KiM8-08aQa!GRMnwlG)P@8s;hlzEz@`;M~AfELrEd3$N4pmkSw2pq| zOh-`P{xk%Xg{lJfLWM(g0l-YIp48PlmIJsbgu;+Yif8+JuQ2#Qxa`xjc(!X~-uzTW zi7@Rc{>h@^SCA+6TmJ{l4m3~#!Lk6THNgHhy6{H)k4{E*q#*f0^n>U6Y{vlYmDQnc zPc(R+`g3Dr^QMEsCWC0{{y6EPaRuA4Q#_E`;D>F5(8p>015oI|Pg+-M@31Za z$)E2yi--PC`lsF>nt;8>leF(y$DU9^dgOzBpt$1A{~!I+x|e*D*U}8h|7we%)hvID zIv_*=X)2uBDfgItd%ea!vD5c9NAPeq`Csbj4htBIo8g;Hqr9ILvf0C(?$IQeLS-6E z3STtF>o%ohO{BH4QsDHw0(nLIsU{vbMAsQJFV-PeD8`Xh<fi9dJSU7C&wKuQBDUIF455(TBb4h@qGGF03B?>0}UOW})DO7DLJXui3tJb;o? zck#P5S1d_kUd?C2e{wn*E8(sRgBIPl9-w8DEaM zWEi{$sL3=I53mIVr&?*Gn-~WtbV)HM;G5I=eQL`&1Guw4{0?z;I6yW_5xfjZonpVn9M)B$@KQ6UMX!00D>7doP15f@zERmkxX|e^4ka@?Lymq+T78h$_$KNpaVWNT=`2g^d!VVmK)yQ!_E&lmr2zRaos|5b&3UzP*Zu!3=LOFp z{r?7?lx)trq!{4$o!&4B*lqi#4q(A&&mBnEflgj!&n~pa znX<@h-eGr=!rE>1*wGQnrS54X#YiGe2_?Z|YaOs!n_s;gs9fv@IOOF$bO7u^7Lpa$ zS#=-a8*$-r5Pp7(F9RW(Fnv8X=sZt-elFl=pj+YBp^@b@|6h8eK%D>^eMF@_M`}St z?Nh`*LaL&Eh^}`%tsQ_VSSZ~OlxQJcQn}8NGCE?<)+6=%ZofR-a4TH_{`oF*Xw%1A zex-NZ<1;(Dtw3N8x-L*V;Pps%w<7~&zAv_Z(u7@;Z<7W*85qYaMvqp`o%ddxuQZn3 z-#bwn8f@`hA3~;WxF`5+EFi0xdcEQ~1!hW`K$xw9~HFt zyXi&?%2M%Cqi(bk<;o%j>Nor7@|tV|A=ou83t|(u5}AM3=WkmfTuro0`qb!{Tk`=5 z$-X+^_dc-*vysGYszD zNmo~SxY9?XfD~{EX}r2Y?`_-EtbTT5n_i}xR>EmG2((-tZacw4`qp-P3AF59sysJp z?-k|=3y=L~RbGJCsmj?C%-nabm13bqH3zwQ$ z_VzVgjTO%P9h}FO39?tpb8nNrsT5lEm`PwL-iQ)7_;Wt{eJkBT4rWpDqSBO5~8cZb=kU zO0zI;6QMx-pRx?!#cwKUCP@WiZ5%ZP%gg$DuR3zO8h6s0XkA|tyBnY}^ef7qjXOr~ zbogDoo7Y_@c(^kc?HO0zfbbYsdsRA<=3kdKgtu}_3Ui=SAz$#9lnkay zU9{M!ah`igG-oi~AF!W1PAuE>3W0<^Q6&(Id|+sn`Z8dzdoNIwg$a6JcQPDgVdCX@ zPaE;P`$w-HKeF0>MjlDA@A-Rtl~d_Wf5@%~bnK^9HPUK@yMapsH*yVv>LIL7RsR8fN6cKt~4c zKG08=ke0pTM|Va(kPW@2bRK(xuOCJ`O>YYF;Dm9dEA;2yW%M3qmDywu96B#Evs8xj#QdnD^jY$_3p1O)y3s;L zFtIH8#4h^f6bWU5tWHcU&_Dy*$Y~*q*BYM(c7e=zeEHVk|5lBI2Pm#nk=M?j_8J3=j~ zhqJuCgDy+^dInAFQuu@U*sjw+C~c0@%$2)!G~$Y-_op6E4(G&PJku{c!+wssc_j@o zgCFlld_w7`y?t{rK6Gr>(*GtMY;%L4l?q$nc>}7t;Ir%~y^GzGhDeqAAMEF12!eIA zu7y*n$|b@suFQj6O8k9#BE*i4+}GQ0+<_MxrW!Ad#>WUUc}c94*<#-&NEhp;#2$2$ zThCIZKv_y9%|%gG3O2EO$&Rt>D@wWuTGtUC+6{WtP)=s{(l)~imqn)h&&2c~V@tu! zoj)tMFfrYQ_su7mfDBOWS`@AAt~;8qT$PQmp*;tysA1yapYIADrZ= zzdLF1bLxl#4=mErDUtxMMEm!>vu#`a4zA3ltYAUpif3(o@X^Icdib4yMdUNW?SVcwS5UNk^ZjIU$tX&odAS$2)0+p(xq)( z{=8eoVUWi(^IFiWM%jS!*&HzhEAsoT$_@)n<|fZ^ zZ^^6fn_lNN6X&IiNQmV_!f3IZY8v?F>z%!l2Q)H^V|`3BiQ(j--8mLm+6jstT33P= zrD#>2tq0=_X%GG#GHEj7b2EbF-HGaJw=0GBwGiq9lROh0<=f!UHX2YRaquqU_-f2` zF$Gtag?1kjMC-(r!WqBZEYH@$vO6lQzXC?>jJ{$lU^Q(S&fCOQpuS5i0C$5p$%pF7 z*mHWXtg49{QuHZ$y#E;9$Zcd?8ZGD1OC5MtFkVZ~LJ@d-Dm8dVH-#P@WiP~wm`f#5 zxbz}g_7Ru5_Oq}(pyQMdY`V^LO1jiru@tqL?LsJ`~CWo1e+D>z~zah?C+lF_2~fPCk%Ij zvc_Zna8EVkZYb`Lm0$+YidVgfzn}Un)z!Hav|e$l_pAyev3%bm+xG>xwldQNE0jOb z0wI$H#)UUK@!!53wy0E7g~ZZTgu%C(ecz?RC+u8Vj)ZTyfR?@9juohH z@{Zel(qz)T+46v}F2uByNb%A?8L!PdeQkK7MExz@c-&UaV8i2TTnMUAeZ;N#v9KNK zGhR@f;N8gc*@K8r*Q-t`6JmAvpV5IS1B}L#sG*^?VY?gaD}B)w%7caKS2^8B`a^i} z!Hx2rY{R!S9!`&34IH6ua7WCHbx+=eKn;84yLVf49B}z6>2D4T`Imx=Nf#Ip{@(0F zI}|(qfX>=w65) zNx>!woUI(ib|#?OD2%R9OuFMsQ@<93$frrTi&ZclFI{G18_8~&&alby$4vbdqMZpcO{O6$+()I}vkxIzn z=Z#yGrq;adkQw*sGY^8DC2ia-=r`U$+kx8^brRCd;2h^cZJ~SXS`dcM7cA}&M6YCBUVjdH(M zG4;!z9TToZbMw zwn_#jlkS9XEq~j;qQJSg{&I7#nRA3W#qd4RrLcLilU?8SI$p&!jTVx=+QVf*>l*7e zd6yMDd4oIEDr-`ELXbP{eNCy3$%GX3nv8SDbg0+2w_72cS^naw=^q%@WhC!YBrtIC8XiYwzc?i3)p^Lo`nk0TWS8npvOOYG zooD_!q?eQQiLXFTuOZwtBu{Hr=4x-T-J*vn$=DHHS9I?8K&|7vJ@T3Xpk)l(j>~Fc zm%go`SxI3kK52K}<@g;*qD=kw@nPe+?3+rHyC|}N`?#_{e}ZccUbyTVfyU-K_q5$$ z+$i%CUqwy)US@h#JL?b zK2Dlsaz?s`IoHFDWtf^ohhN5Q2R5f~A6GD^q|t*5lVuCkuP!_DO3Lt)sl0K(N^=-2 z0uwaEV@krBS1Xwy8RivU_P}v|#{}+knRq4$+%MMWRagz|ciBXmn67DKzET~ztCI$W z_2xcTjCyQ7d%`%cNPuu0I1NnqEVv8Z6J9vM4d&hmR5TRJIA_g}r? z+irDk?%x^rsV$cOyPskAp?c6e&wO&(KroH`_bvA3 zI`pLJm`FC8nj53ZF>wwa*Ef3YsYli7vqVZfEwslJ@+D!Qs=|x0szL^>SN>RmHlXGL z1b7W*d)~U0#2QiuQG(NbQ@QU_nkQZAol>s7j%qkXNS%05iTujel4GdDcd_E)7J zFmCyGPm_ENlDhMFZ|n0sK}vAr`;R@)9)v@zOWZ#Ab1C98p_>d7YM=p z-0S#7h0#AWZQOexhigi_4URRb2W&EqYEVmJ`>jYNxYb2kC5j)=Fbf*0E7MXC{~}tm zlo;+M@o1;ovJ1GUMt^l_@PjJP?5Il3K`VDzoxg8I z(ug=Lmb?-<|6MF^CaZl-le^@7Va%&erk&Npo7}xB_^UNopHBhtSyL*vV^nEtsP1jc zo+f#wp1a)$+fzb`Eh{254$5_0Hr8L_*9sW%N71)1&YVZZD=CU|BfcPVa=YcjxbJ)-qQC%H8r!$0A$CMqBW!dP zHTbJ-Z4$tAJk_xF>d;YKDCEFr4)v#;GB^TFV%w?b(iHb0!=rTt4- zb)YF#G`l9OsaqZVQ9#1XdMbP2fR6|$J@H4AlU-s^Q?ho0ju`!kj&{-Wt7qpaN*~T~ zzu+#M++qRbN`6=p{4~{G*LlFh5T%u+nHyuVb#oi{{sx-i*4DaRGlXJuuWy_gINdpF zkM6#v`4zZQ_n{@DIK8ZFvDh8lo)1n6?}MiJ=47j{uM3;fhv>@i0_zUf(=6Pv^OL&7 zl3}giX6No5ll4w96<5~|!vdL>M*mENyeaeV?yo>b%RPG@eDG|3&9mwL^^Yo$z8?cEmaqkj_@>A9*~2wF4d#Yvp}(3j^0X_ zmD*2S@p7?Bv`Q_{rviuf1ABhWZ5(L=o|l$9$LOAArFPEKZ1OGg{VPz0FE5hmwzW-A zbp7uT8-9q5jhORURkIMy_UUWm2ho5FDko{3=MNwY`PF78=j{j_22@P`Q0 zao$0}I8H9V@}Wu_ouGna5gS3jR*&DUW3rAupmAqLaH>l+tfgi+Co?_hXB9Iwb<87U z8?bbP7j{SifE_>A=YopuXBUl3g{hn$y_kP-omG{l;kj`M@U!lEdqF>?Qdff+PUrBR z*FMEoALBXKws?+6Nl1cC$!q9uc6`8wIeurrv~sD$wDPd#4yLDt)11&BYu=r)y?Ee# z?>kjR06D76CkF(^U1~X+*Xyv}JR@RNq4k1&e*x?~mmqk;iFx}v z@UQK4n>B9#nSDzyhR}oSyyr^-W*<+^&lnZ)M{6*qPwsbLN};Rr0WosE|iCs}vXpoAcE+F(2cmN=;jeES7+^bd^q)oHegtK;o~p zmHYZ6=ps>{Z{_e9dMaypZ0B8XxM~mKr;O|s(|MASBJh2-!x9g&P}7G$X#ehQ{yHTK zRul=wJ~Sf>ZM-g7(f)kB?^yX&Io>g=&1KC9WK!W6GI7ZL?OXB3{Yd>>BBO$+8H1}= zrnTztD#@v)BXYFHJ8$~pyw$g#^!ry@4y*H4$!pA8b@4R`Wwm>vKL53cA948sM2vVNQLmE zP3|qQ7;mU$^NxLN84AHktD6p;@Q@8Llm8t^^iHEsK^K3sFMYN9DDQvz25c^ zqTx5XS7}o<4M^Z(>fXaQlAnugGo!wb{S{Gvp@F-TEx^AUJS`ybICD+&8RxOlS;8+m zy18G5HCQ*D(NC29>eb>b_^*`^#I;&O=u6_yDvCxk9kLfgL*nNbKBB6JZv^5R5v)~fBb)6oYq4X3e#I9I!;NOFm2o$xuI9Ibk9UA~Bq z&~QE^e23Q6jo4kEQz6OVFmJ-B@JJ-l7E&f*R`1P&Eu8MFC`Zty8_y|>KW%rwj>hep zhQSazrZrQ|)2D62qDmU|-nsj`SzGvtp?>_1}%~(rg=2tW2@V6an9;#VdVxIY7>gO(#5)opj$ZgoK41$eHF#UAzkq2 z;~%A(RlM-^njYcgXQe*h{)`D`N}9*gt4=0MARGZsP?d!{(9VuyYikQ%-yv&J4)ysa zi7Je!&ZA5bt9Np;1})RFvDfLx#`SY60h`yGf)U37IXCl+LZ$NDS9xS^^#CjI=?J(I zBzqb7Fv1gO<&u}pg)UoOS#h)2-34_hd+xU;a-nticV$?HdTG=Lqvfn(*uxV|^W*B7 z9;VZpu?Xb+vqpc*>Fz?t7(Ql!&lOQlhbcB;ttZENxJ<9w-jp0vP2=vX2PcWah`y!e z<=4}V(}QVUf9PZzt_~zdOEW(H4rYr7Ef2pwNEH{OK+Ie5Y2O6kZ0_L0d;TG^a)@MU zrpG_%fGAm5&ov9!5QDvI+14a~$ev-yQ2fPKGP9 z)kEE#;AKAKosl3uB!FFg#lNg?O2x}sTj*v;N`&(zJTzm3B|3_}f_R15S4=(vnpB=} z@qfU*bb~3BB>+2lsRo-XD$i?+R??NZcl{vJ!Kv;59s_c#^#x`u2;~dwb9E^2kO{sXw}KPIvmr9qiCsgL*cCj!uFW zRNxh0o~8Tcpt&drzg0UT`|EEbG3xZ zJ}NfR(Yv6m!J0Hi^5eh;5NK#|g%G~&VO}(YY`+Q)yUGtNg6YOR57g+oQ0g`@npU$} zzdKiP@GZM-0*R>|z{sRLhNJ)+p~{FI;H;~aR^TpXB22HNvP5Bsbjc0?00Pb<44fk> zim2*Wq0~?NRD!; zPE#jtJdg?r)QD_u{fdYXM?4i^f*DL=m~k`*Z4tPo^BMZssAd( zs(+FvyUM;#6sj*eQz4uJGx3U!rB5i!NLb2qn-0*Ghiyy@nRq=^C{vO8)}3oeRO%TW zt{#0R9c;waPZF?}^z8Qd1W!HA@;^flPOxaPa%I>)v(q3$x#W^de*YzC*c+`#NGc%q8?10kQs4CAdS++0UC}#3jUd z<|71#$?gXZ-@*u7{^HnCy zM1}#dH+-iogK2ep`WbGirj>K1+TOIBVcB6{+o(S5x&S!Ew4_@O2~3W`APde<)&XpF z#K=F1AXhA5Dy3DI0bXyM=Bw%Xo2)S(Z!srndI&g6qHrSa}eu1*gCFy8xDgR)WFpja93;&d1?)#zCHY zqM7N{LmNgaC)k{13AlkVT-NcyZ08$%SF36-|ovP%_ zas$24oaP;tmw4^0Rd*2vy!60Qj8;PKj&+(+VMItQUykg@Z)IiknUZnzN-uqT3i*)F zXaVb|?%i@3t{Icvm|~3|aoggqceH^!*JwS-B@Pd_4IZ06Gz%dzV+?+w^oBAz(Z6-QEOFuQ1I;a#xCBokd8GFGh%fb|M`xz{bXcPa;NOgM;093d+c8%PX6JWftdi*bq z%6FstMRG1E@+E*xND5B}j^|T}`er{j^vA^@Zk%h`tEoREEyod|Aey8EJDM0$bkY5y zUjA$}#Vp!T2MBFKC5t2G0G(w5mlZr)o)uFK!{*AHc-?6eWJ+5RQ~&rQ=i+IfzPm$@ zG5}fVOfRBTKMt~jmS0uTR^d`|cJ@cLb8$9r-T;kN|GU}Xvq9=|$)JJqETHiC`R77a zP9BYas6?hfM*XyJ6*Q`0Yx_?qu#XejB4R3@gel0^!S&-;j6=Vy;uw~-CwTsRQW#(A zGO7Otu+3@-&MLmC4kyuXV+a<7%^VPTs-~&-N3~rQq~&E8!+QBCp)*B?zq2gk>BSHO z>}5)qMWoNo?=nE7N!9QO!*fb5Xy3M=#Xa@EHp~hzfN+DTn{mh0gAKg32swP?f&jW` zn4kr^oYqH;32*1Q;75B>=*Bc=stQSX zhRlQlX^hucNygtCD5(6|O|j*+&EifIjEvb;KNy_a1L!pxroxIL;; zdl%dPDo}6ie-jM02G~wKG0_O#sN*o0JI>N5OZg@SoUjUn=EX&=O3xZO;4wn*^(8sO zdJ`#HQn3@YKDAFd#!Cizk1ax~cRvnPP`*kNPqC@as_FF}s8_{d!9q~D>?S?jJuG2K zH7z2pBH7fwr!1q;=iur73&Cn1PXhC8Ae``c6?3kmx2*_|n(W)PoCs1Ncc*`!R9_z%L1%;7&PYN432rOem%1#t5ZU zmq5P`q?`{xEZ4I*d+MZPL+NKnt$CIK((SV>LH5z$27PSIYDk4W4@^V{?LwlFrI`C7 zE$p>bkh9SXD(xWL|3vzU>^o9v$6??fqUC;Qcv4`88Vh)r2cqRFNk+!P_>c+V`&mLS z10uv)UJ2h@SIZaC_~w93^63auybZgUOUv%e6%D_vCwzIW!o5ldR{d2MEmync=*A%# zF{qtw#mduOT}=55DD9S86QhvMiWjr+P} zRclWVB3JXGc-<2PrmuDO!u65Lo~Bm)A)+t+L-EBwxYnL}UB-3;EiPgGA@)fW%ter3 zSo&vxQtOw|Pa`i>|Fl(ujAiT-b%8*w9|6WPx;fT}JMmcpFo^Ys@y9Ra-A~!oOsN2I0|BywCoCD>y zAPayu8pPzOxFEG;VJ*so4b&4L0AUYb49gV<+E!=HH5&ps>5mj%S&Q2a2*l1z>hWVE zp%WPJHR9sR`zOH6%B5%a+Gb_mg!0tXn&Tq)h#;{Idd?E{tIs}w3oU@WF#*r+u_(l3C-=B$2WKjD4`F1@ z3nUauVAj)0)bxM|fA-pFrYePuE@J6sr)P^SeEO4(IiBLQ#pEORYqT*LqsR~E5& zG0Q6J6-lP$6~Zh)htLNy#WhLD-H^15Ed2TX*7x3+A3reSe~*root?%09v$HBiWl-1 zuCD@BcDY@EP9f;mqN-5P;XqkN);7Js_t@c}FDoUUVQb_P{3_Ys2_%GP1`xg+6>vuR6=TPvE)`ZHxm`h;xWacXxx# z`s;IfQ>R?X!NQA=UF!#;e{(Z(Vu%4)sKNgk8tldG<7I(om6j$NIQMTN`yW-Fd^HB5 zGq6hF%i!laxDyKg{mSYhrnj*1XryY+?I!{x^fxOjTOlUZ$L0L_8 ztE)Ga7&KpKVk4I^ zk#*DN77?-r$sUS@oCcSQ^>1^+q}Gb1 zba`;yPwr{tUcdG4uj`-T>C=V-`<0 z9{oe^LQYmRr?_fwEYqw1!OaiZPH`GDf*F}<&;=G>qvCRM9jX9}C|#?$KP>4LUllB( z0thAaU!%`XVKWbnjfLl0r+Z>DToUHH>-rnLYKvu$Ey5s@!=10p?JaE;eeRd>wnwiu zxP@YD9ckDG$p(N`v!_X;OLnXqU_U@rTvEccb}^G6do`w-_T9C#4Jpb>buy*@$tMF$ z{qO6(diCI7yo5NJ@E`cnS@MmgG6{5Izw;y2$k=WN*1i|_rnky6 zH>VmL%-5@O{DUDbP-UFwSMQaU;@$|c#P&cmWU9_y#$mP!M9Zz1Dhr)*pA0ZZ-EU7M zBqt_*NBEPmov1AnO&Wm30h~6eGr%jsZ}HNJXkzky>&<_mS}giJR~f!^>4Xcg+@Qzu0tsH^;ZRlG-Y@CjW6DkyPC^xw-Rs2~QCb-6@aO>)Bd6el$vFbC_;S3fw{ zg&U;&khT|h2)UQ?D3Ch?II6l@Jj30`_)S==@oc@Pumc)F;`L`vFve8DoGg}VHYTej z(UGiXx3d)_sfpjV_P(~}kx}|QX_{xNQ+lGo68>8WeC9R5u+TsA9{xi!`JXfatP}~F zZCH!C_=6-+;Dox~89nUmfmKVt7@-26AL8W}y?I$-^REVBA~jiA31i;8=Ksi+V7`!< z(3~zWB7iv${I!^c>KUeupP?5&0gvoccRlTz#-HOQ|d3eyLxv;goT7Zf8kXW z5*CI39w;bdbld?W*FIC9N&CrRnoP6>iDk#gj(w}Pq=)z4Dnn5+unR6mNzD&ojsqHW z!BBKNa*Hcmo9YMcCLA1Gl|X}gxMt-412ih4L50=!Q7OB-geNciaPzK0Ej>Qd3)#*$ zlm1`+<<m{c`XU=Y?6`$E^o06j;(ZG=Jt$B|Fa|q*1q;Fcn0=U(N4|$IVk1&Z zYMSA>wpZdtMy_!F06-Omr1!-b0Pvn7n!1bG>9vK^AdmZ?+SeVS=-MpuUmlcU8!ZXQ zxF5Ko3kWV-%;KvN%3C0uFK#14kkj_2A{41sm%XRKsLG3y(Yg0a%?A+{XQv!2@rcW` z?b3lo#W=^ye+u9hGVEjrTE| zQ@8(mc#|mr;+HbxhEzup288+I&q{~feVc+pa|ak#v%ErXQ56FR2PZz}0m&)J``Pg? zZ>GxrAXjgqpS0xXU*q6(dSI!;^!H%lLwM2C?^%A=`(^UkeWXR~n?nWJ0buCPi-16= z(%6LAMvLH*e6pmCh1&=Jd;TM1bJWp0HJzy3!a{gpsMDP&u`lL^|2aro+kF86e)G3) z>6@FIKSyT8l1@E#d}jm$zxE1sW}ovbfuEmpa&X`e*YR@dnyB{#M*}vfN>}xNV6k7n z048*BA7tDzjpjZtLj$DF$~yn%-X8q-VrFnBP7_EVYHREFQMtLf3{F+JUwEHq6JOi5 zt`EHB{O8E+_7V&n?%aoDAE}0_PSrKQ&og=))1vo90u!g54m?V-*56^|qXUBY0_b|C zJzLrKBsP zW&3>)5%FQSX8)+$gra*jyi&xVq0yW&7v^N77lv4{^ZD!F5JqK0ThnG&SD@Evi&UEj z!6&!<-GLCS@le@kIb2||gABYH=sm|Hp7gAW7I(p7JhZVPbC{{6$R=z^VMdhcx^y(} zce7gW{n=WYjUZ!b`Ct(6^&g0ib>U16EO@c|>~?1~1V6tQKvsd+=g3s_eKw7K!r)!% z2l(3no=@EG3$tMy2;x3fW5B8W)Z(}1(Iw$&FB>BcDmc2&iC!~XkzyM{V7})#hx9J) zD(wF09PbaFnVnr~*+T>ia0kZw(;juXqwbTT15Rkrq+X$(q)|ew-}S3^hij$NX<|$L zC@_Tun9R|>v%{RO8kFL|upC{7(sWayl{3ZmrK6$@CdG4y*?N{_G3c-o9B>R{yBFBP zzO}vm+UU<#^Cd()02WH21OB;|F-l!$Z}84+4Fc$DYbqG-pg~K|@K#$RdUKP_q{;0X z;&@9qAmH6@oxVA4eQ0OK|0zs#(8kAyytu-2MPb+p??BRCu;TqKHa?x zYI)U1DbbJ2%v>R(<$iy(;0K17HpG~|udhP=X2M&|3&0w{#Lm$fYw7d~xTtl@>ATIB zqIjQ&cFy*}N9aKC*=*psGiJ&%wqvOaen}OiQ$!()L;xP+e;M252mu2}7y>KwI}UZ9 z&hKz%slDJep&R45_T{wMt)n-_z*3nH&4JIlyH~04r@>{1&1ui5cJ=O< z988DORQk_);W1rwOQ%N*!GNa=E<-incmLmQ^z}l&9D-#p*-w-PdG09Icy3=mlsOB_ z3B=T}b909x&LFb>>(s5pUjsz>ppATI1c7t`=6>ruASyPtt8|CiI6X49-}*%RDGnYQ znp;pnim=R@J(_A?lX-=>47f5Nar_>fLE@Qf=eI<`~%ByclO3M8B|4y!HrOkpv2k~4y0 z+pv&bm9|G`1s3U|XbS<%@L;v-6G~vn(|?a59EY`ez2gCrdr679f6hhRu~7;Z$brQ} zT?z^ceO0NQbxHb}p~&3U@1?*%$lWkd^owX>N@6VJIPCmXKtO=-c7KL$ zqO)bMol3vBd}^F>TJb-yG`4Ufj>C_GmRyd=8TI3BDAC#_Gk;3ZJ@y|6F3psQ}}#~Pv6d}=v zvqSg1Xq4!LUzIK*d@Mf%8^Y;8p}=eRP+6Hz$$Ofpb%CH8S)C|f+}_(CG!>9@+M1(} zLTeou>WJ3ya6Dsc^mh%rBVy}@WiM6WOJDCHj&Btu8?4#k;ZJ3M$lLi$LOcJ6`c3u4 z2vfn70Z7~E>{=%n1PZdBEM96ELSq@&xlEA^?-K+xIzwqV7qiCqFa$r~{)e>R-zVpi zhmp@H$m59!W`#a~%o!Ob13fnk<6uR`CMMtclH9H?wAncWF5<$vdN3+LD@uJndWX2G zoE^w4Eb<|likNk4P+L-7!HvYVHH#MQLpcvm&&~~~!w1djq5d;}G6Oi!_?^xcm*h4U zkk5t1W{#pKdjr4p#`OWlt&k&zc7d08P}3`ant;<1D7J`M*7WW5H=lE>7`*n5gS?qF zGUbcLDS@y5dpTa-m6rmR!6avZdEHfEv;hUWiq`qX(A@xBj0~?VWBbl;>(&~-^}p=E z0ny%bAX1mV(q9>KIwQc2r{DPy)>7~jt;ETaYnY+80ArfhhrtqqaRFEYL+jE)l|R`w zTgCfdzqGj1bS`^sN1fqJ0(j*+U%nVZ9)*H}<-LCC&~afEg~t*#cTqBsGv+4u4UIGJ?nsIS za>VK3tkyfcIM`ZHOl(8P_*I435cu%X!NFmtmlscGoF%c(M;+{&CZJ?#^aV!bTnXC) zpW{{WUEs6t3Q!=0uh7Le?7q$lid!;o^!Ux#!a$@sPctaa%L5(mq3Hsf_8^ci{M)rG zy{OddQ)f|vBt%XbCrrCsjO71l?8@Vz-2Q*}R!Xv^?3xftmSh_l=Gqfc*|If}HM?Ye zGDR9&l(oXBD@&;?*@{MEU(#gDZZMdH7*CkV;CIx0eeeB!?|1W>KX{#IJm+(sb3W(u zeBSTR`%D~SH91Gj*pKr4s-pQQYWor78Rz!27sHq@h4Qr%q({tvFG>i7PpNX)b8Ww^ z@#XGZYOL?9yG<3#oLB@pu*FzXZrh8Jl6&a-8`xEtPwq&Vv$I5YMfufc4x}G*2cl$i z0+wf#?~J;3S!%7B(#*KF197X2A<+HnP_}|KbW;2ImPUYe*L3K;@GTD@+c zExe=A!W#a2E4)VP)N64dL+o?Gq?oxpcbgW;sX3;KyHHkC#2NTXhlV;c`VImu`5n~G5e0+={*}eHhI+f$}5!WtphLn_t^*0 z7>%WSm?+=DP7Q2TI?lThW}kYre%t?jZEtrwfm3R%2)lj^2+wIOE`ngo=WEsWQ9A}p z=FwvmRe5AHzYjl6jYTawqIZ0C8_5AF=+5yw@)D>vOc=`c2)}B^mNGX|uilW`FUrCE zGJ7SYna_t;{$fpw*Zrt0dz8<0(g*9$iK@rQ#S6OvpGS4eAZD6BEOzD?FjDv|`g4UV z(aE4F5#+TgvWrxOM?5@myahUz&^<1asbU?I$0i$t;L4@B+X}O!YIu06MquOEQ z&9xfw8&{c5^D|5KvVK4MSN@IOA9Auz?$hS?qRl;<(6sgEnb$jY<|Mh*E$KW8CgLZT zyN@ERz{y~nYt?kBgvUF~d5%{Ji*Hv)c>;}JB?U!SdJVli`HP%vnos(}qO=o;+A6D}i#KKG7HS+D{YQVPpL+WwaAA}^hPZdFb9J1CV;VT*sob+aak46dn;A~!FK7!IlQ?| z*jeQB*%_EG$G8lu(5AH6(`p6<9JGpiClr5``T-W=feJVf^tz1n^PgryS4UR+?@%lW zCMD0O2S42V6uIVSz z(#KwJezmV=$XlZ~g2AcH6KwCwyUrmKtKNK+3ENl6Dl_1ow^9N6_ebj|#M zF>H~fl+M2=Hd{~#xAMV+aI0^y+{3zg&I72LPhqjG&1_xdjTP1YS|9Xgrf-&oovZ&m z4zZN2*h_MJ5-2@!kFu4FSH{*e9zt_Z$@m=&<$N4^pSVD zaf{<>b}sCyFNo}Bi&i|s%1tKUIxmAABf-y~_5)+1M9p^P?$MNvfvdp&dfB$`r9B0H zU6RbC(7;3a95foTYg0K+e!g*3DIwE~T*8hITd_u7osKs-zQ37j*k$L~VzU;KG!`w& zt+^qy`7RQiaIuP`*ZM&hygdW0}ybrH4kW3|{CcOtIZH%HWN-yF~& zpAb)4Y15rq3b6}Q{++cfySdy-{NXl)s@A}(%3C8TZsH9xMLA|D^VKhfMa2p8rcAQt zv**wER(KE98DRY;={--)peuYYcZgZKKDf!oC%2TLml>HzR<_QMBT)tmxz9XWN4CF#_jl^{jvA!q4U|)40odjG8X_b^%iLjQ9-GR>+1>y1ywSEXN80) ze4)&z`=sV6-*0ONW?5EkBkZTC>i*{9&${GS{%Z|g2D4Fmp;|GDFZk-b32IwRgjb+lB$1O>+gjYGNrxF)`I7-P10b9Ux|yXNnS z-O|1WJq20Ht!v6#T~q82uiV{YtiPy->pxfIofO1dF~Zv#7cBp#s^i``!v;m72iFxk zNFi&6$Yt*05RJhHl}Gk2)YMz4y?f!ApO=5?Eyw!^P(c^6v!OXRe$jnQ{VP)zdu3xB zcW?r2NLc$S7+BLz-Azd=t5TLq*heuZ($sKrXqg{ya47z-DVW``W>=*q!u+4dVfU z(N>U)nCh+?4W0E04q^Cj+~EG2veG0}1iK!5V{H(s1>sIzR2f>}YBh>ZRE6lO;ZfIF zkXRuf0>%|fR<$Cu2JL7tVUZiL2wagttq#AMcxChIHdelP?^jD4S>WvwV=?#CMeX@? zu@F|8U)v76nI#}<8JnJg?>zMV@~L@lS{N(An9eBbl~$z}4G#`~;z$ZjvSxk#?4GDR>fdr27ZE-CVAR#_Q3eec=FkHgS2ry!8weCEU17Eef0aln*f4np})gl$bnbi;*7I@^E zCDWQSGE7rF*PilzKnM$Si8#LHle_Y<3ni%SX__7t%@^OYorHPDY@In8SZ73RXxM?E zoqmY;0pl;y>Td4Z6T9>04BROEK_3Ep|?ra)e^u@D+my^h}upNOBxb)jLaFLHN@Rb*3VU53c8 za$(ldDF)v3%K=eY$F&fGcg4B}=GEYX^B_b2yR{9{ne8aEbA1a&?YxGktZLYxJ_7efcP$N;+&_?IHl$VX^%K7&-W z&-CNZ}MPw16kL} z+n6KzegA~J(npxqSRt)Py|Go)(agIF3FuZgSaKzA2iD*bw!^S9*Ze+PEn^Ou`+u*E zLy-?$%RrHjpJYa%(9pH25Q4gvuvaowNFP^wmx?RmPz9h#bdJ{&RR4LAZ;83sJI+g_ znn{|UC{;?0$@yKaLc%zVaG=CqWUEaAz`&^N1{C>(ZaV-V(%llgXwW_wYUh$AaboPj z)s3*Br{x{4p)T?R-k=%>9*VE;DeQJ{42e^RWaVeDj4x^*AG=GP48 zSU7*8bW!{KIK#!Q4Nm9%TM0`mY#*bx$9;);=o=(%UfP|8BB$QZxeLN~ZIY_Cgu2(? zE^w;&0z``+xUk>8^AS|!k}kgroX^`DwLui7-OvE<0}O{#quPMw`5kb1X22vv|2+(+ zp^=9xy)C=!EP9rwBdVapK$df!d6K03bHFl&%`mKED;{klx?z)fda`NlR_WUu!Ewe{WUBq;8c!dRYSL-^LW0?Va#Q=t3MBq? z^4~Y57!d&3;*F!4s5N}Hv*0{H(%vgx|J^9~zd&pDNA1Bp66-MzY&f4HGR1J=&tusI z)>q+TX7Sm;RQ*Ad9qwxHMcUvI!--Rq43{>2K~@rBz;J3{Dx*LmJ{5%q^uQM7kB6VV6)f-ga^ON8YDj;Ip>^u9 zO%L2#NcC8Pz(hE1mQFPjN~P`+jk_W$gH(F1$~5C_oG?j)sKO zpV`Xfc(okDywDM5O}pIsuYsa4P>jFbK!9X*ZePh3G(qay7dUo=!jL{KmImOVd4^3L zhp;mWqi2{Adbe!bwq5pyx?1K9Tkwbm#;E^hj&bRKfHA?bSvVdLgy)V$2kWU(k1Vx*~h z`_IIB3jmBi{DaX9#y=7MGtEwhDgXG|Pld+)nXKD~w zaWxK!ZP+v?vV0&Aoe? zQ6cf>q^QBm`i3dSlAck93aL`I^QRt^94TGiX0R-T;1%=oC*cHkmPEyIZ}|>~yF%gbQ&@+r`Qs~}O6aEWn^R8NbHvF`j5F{>QmhI5z#lQ# zp0+!b=%K)`%ZG{Fk<>6vtmrj0nTlu})iui}AlgtN>oXTk)u|@uY=iK0@^`QXZNd8( zi0{FieY^~??FwBcJ79`_gVeQWVMx&{hAsr&C1x0bw?0uBE`$4iyn%JW^SfWj@A`ue z>Qb&c{TEFg6ecc=(Cr;J84_-n)DBO@9VPWm;1|k(1`ZyP4zM+bV;fu^9P!VLO#_c7 z0Jvm)4K8a1a|7@&g9%=&PD2`Gkwg{12@oohLx6{M)3cgd4Z_UE#a+90>EbRdM8ye~ z{gq7l4t)7#J$&a0wb~zr3G0>B-P#&`YymsrWe$=N8L9KVN!VstMU7oRZ~BO7sRqUx zg$|B6x{$Or1FQF$A0C2J8{x?G`%7}*w({3BvrlNX&+xt|5I7POpqtktv*FWsOxvbm z1%^gE*TM)c<%+przzyGuZLttZeH|6GR1Zsavo=^GyXVgfU6cNqIS3vP!He^NgNZGm z1APx$0sA5@gIdkg z&WQEj=0dbisC zh#=m>vN~m38*j{XKQBp2fmFG-^Dp6tOYiOy+4U@Uzh)s71&&>~Z>nP}%gp27paVhX z3ye3y92_BpgMN(AaQ;w2topN6$R1n-oO~6ef!>=u^(VP2#+T^qq5bNUCdo>^#6K*U zAn*`4Hcm%8P^%?SakPS^H^T?tYU>6sb+qZG!m^ue%{8aRt zQ9G{mh0YI$Vmo>!8A@M(^8iaQ+$~P8!R5*lB)AOONzLaAMxhzdri}P?72{&A52g;aqi=n9~uG+DG4k%^T0c_(zn26 zOM{^C-!(~eWaII-dvs$>)A@ykBd`Jj0uFB}toJh?IzCVwcDOCVAA5OEUX=)J9eP{D zO!2)zzsqrv6dr@I!Xb8qwdAA&GWU;i<*mKo$Hx_iR#s-GP$ zFgEhjjj@Np5JP9V@q{&+G}eBwd{U;8SkQ#;G~{abtJ+S=pF^eGDQ`u4m% zTP&wPJ}yJXU}#Age#Bb!x{Sv0k0b*ePGA3$Q^4L0#ByCFve|=cdK>sL)HBh2seLKp F{{YT}rV;=E literal 0 HcmV?d00001 diff --git a/images_ml_frameworks/image8.png b/images_ml_frameworks/image8.png new file mode 100644 index 0000000000000000000000000000000000000000..e193c05fa5164f9fda02c4920a6793205601c6fe GIT binary patch literal 129032 zcmeFZcTkhv*ES0B2o?kzRf+|qgA_r76dOucklqOpktP^|lz=>nN>`+al%O=}gbtwy zQlv;PflyQoH34Y}5X!f2)ZhEQXXecLX1?>s`Qtd_43oLYTGuN3T08iT&aESd zxDV0M(H&90eM66q?w@8ly8X@v7{M=n+pmAo(LJG4zj5{6V@uN5!IG22q@AUB5>00K zd}#bnJKYW$={ID%DA{-|ik6H+5%UjE`h&NlZ(cpOPwiiB_Eb)}d|tsr`$V7Dr4(_d zp81FCpS$$O?)`K3Anv8+58Cw5qs=LDW_DF}RSqtvs;=uY=ENg610y5Pt2Ntq+9$S& zQl0c*8sOj9)Z!}B?>|))QaBm*{!}&k@c;ky{|y;Dfho~QkHD_X=)vUwxubOwPIpI$ zmj~=#ith~msH*hcFnZM_8?=Zn!_jmh&nZROM&2&g{-a91; zEVa<%{4iAe3Et^_Us#$rpG2fHgrNHBIiP{Bx43$Lm^7ywe$UC!eHVGOM+^40?O#b~ z=HY$PM(ju~athprZWfx1FwFS=d0rm-t8rjXPnu`niXE;BLv>>>JXwW4cam3J?hQ*s zQA#^_S#=WG5wqVyO>xGUQ`HB|^N>2zev|Hc9iAkWh)!1U{#G~EjC<462AP~Y`!_TKPgjdkN(20A_k^(;7&4nC|dmf=M*_;Hks6qN2N#O>YX5Po( zTZ2qFrtqzMwzr%9BzO?^G=kRKz2q=jp`V!(7g0ceFO`$wm7>puf;bk*21c$;+$Gqh zkd=P@u6;mNKOy}%ZnZ?ZM}=v7CYnj{dNBZ~d3D`~&d8TSdPJS5>ya=i>i-(QV&=8f z6j!k{+4b<8TG&Gd#;q@FvwbbwUAtSlXXVLH=o#8pBKTFtEs#A;OqMH~%T5JWb@k-& z?HdCrH<*;&k=8)VpV;;+S@1}d{#c(h-?qTZg?GCv&;bp<{FA>L^Vqh|tn@)?!o_)R z1@|8U3Xgw9G7t@mhh>Dc-p)95CXx;Ft8>gdX6b~(Sh7v+cXtrbVv{2R->J8Fm)_4} zYvHHOEjDC0K8wK7-wC7t5}%pu)0}6Cj>K)wnYfbhedZIjXoi~xxM5#zMO1Er^yHj} zBZ8R3k}Qg(<4~J!-Ba+CH#2ylBj~(KumDbpmWKPoY-*voAp(7A>5~sr%z(n*$GH$t z75fjfSE|1`UwJO_ip-rG#H1R8Vk#*kfxY4u|5&Wgw{(Uawb^-KFUYgJ41(zeCvHu% zrV1K!O4K*p_cnNK_pSt~#ex(AYj?q-X*vpS{uOREy*zXK2O$ z$k0xha`^g=IMtan&n{^`<65_kD)-y)@aj^cRvbAcA3YpE=puQ^hif~n;HxRz6jRE( z@aoK-)Ru}}>h4r!3m@%~wKba>lVRyhH4xa#40x7Li~c6JP00a$(|?0{mb1^c$~n%i z^UN3mTQhBD`Q$ZT*{1zY?(&*o6v}Y0!hLd-7-zM>E@ep?EU*f&epNJaQH(ND#%vZV z(HEOUah5l`g_9atCVDY48ihYzZp{QZaP)f1oiz=;Y)AFj67d{^w~nu&MGGKve=k}a z@iZvtFbD30&L!zPd^fSj=fY$=A3mDuaGxK{Q~U)B=vl}nVyV`PP2+EB%H?hj|N1Tz zEmc}_iaM_OT5$SVRr5Ta(K#1CK!EVnXi$z8jo|D2M4b1$?)qd)e7G7hX z-&A+QD0-ElH7x@9z|cBO&d56rIu8M@R+UcExw3Ra!I$D(Zr{#MPCCS4#$?7} zCS;*&PiRhluraN&$jm$bja#U?oFA~e6L|IuBJQ;1~PfMrj3b$O9QRGgTb~qgLTnKiu0ao$t%kO&MJ}jGfrsp-QsqMt5jBo8jfT zkK}r}=hZwjiDk#Qs6IbXM)xD(LtqsH|;J`!{E2tJ-7MI5LR4c_SWo0J!^1Zy#SQg__ z{4rsbP*%IO?C3h(sTBAxGZ?O?WHn?m&I%giLz|4m=LQYzM~u^DLpSw|CL``mUb-11 zu&U$UKVRP(`*=5#F(e_R^^FifkTS=`>&KhRl3cEj<+G`5w2HV+HZw6WD1$L?^x295 zhup+j1#N|Kz}6(jyv~1npM5yjB>+KX&heBhX`T485|?bkFb{gQXJOFH?J)B?c=c5T zyJhyEoN35_enilAuHog3AY)^C_O+FgRUBRSELh-;qNl$~9F5d0VcwKs1Ww79a=-jK zcbJPlp{%MV9L-MYerkcE@V#~A2gB+0Mu~^T+SsHU_jj$0!)@$M3=xrzS&%6UZzla- z+cold%}_YMit?QfP8|Q>kEYPlt}|RMW$yhoEjlXbL_*ct(kE5m+Ub@`4jfKRf6H!t zS7qp0m@V@c1z@@K2?}nWpPwILjh1%!`Q6$*F0KSi z71l>H?_X36)$Ob;Ebc5ULUKCc=F^ujS~;D}VG?^Y#^&ya5bVc&;&apLGgO9@J}9gy zwQbVyYCKl2N!_H7=cl?b1?VG=?yl8K&C7`Qm@aB9Rf#&KPOQI~{sP<8sC;U>nj~uu_FYHkz?o-9ubIC6Xe*vA zuyVMlut+Nu>kwzRX+aMunra+wrPJU&B4qf;gck|_zQsH04IXg%coh1u++~v_>DaW>eb>Zu=|VTXk8kqso4Em|%`sAoj#J!OKNBcp-}v%aqKM%q zJi-B! zXL^_zwRPuzf>|FplcpYA6|E=+Q)chc$0 z4jp(^bs5KAfd$2ZdTiIAGJJcqC4$%2x^6x6&zfZpHF4e%2b=ODK^d)PW5XF}2vk-h59r}pN}{C9aat`#=^DO>AJQvUB-=Bs4w z+oC_sbJT+Xjc5NWvBkqxE)YunSuJdm1CH}jIc5sH`VH>DNdnad7BTZX5&zWnnDm&nCb=G#TG~XPc|HXK1&8=D)Jq-CX`Otg0~j@F`!; zd|PUyvZ;Yw5^BWJl|~`MO!Al4CnIqHc-CsCYtgi-wfUiWI|~ki5K^UbYq{N1X3vuJ zZh;xwf;1G?m>TK&=mo$3V}oI7zx9W41H`9~raQ@K3*7KdXOg04lzDTTbX~Vtj0Q}7 zsEKlrpsC<1&(8UzY24@*@GKVJe@c`RGO*8F2Z54TJ8czydsLAW~bSV%)f)c+(Pn~j`D_`k!Z{)fa&*d9JmPz z$N(()#&0#yG?!*ao%63d+Gnff*2wRv?++AN8Pd4NB&-Xu{h#e9c*k35pDn~Wy zXmYw{++4w4y``L;dHmOEs-q7O?;Wcul6>L^|KHe68bz)rZeMI2UcG=c9e zC;TspGdvqJb6%SpfCk)gBADphY|ASjH(lEy@QO8U`CZnRf2 zHCKy}Tionyg+(_{D3|G8g!7+;WH$S}U$nWQO&YAT2Fz(6ZaVp*LD2JfMX!9m1liy^ z<`C@Wqd^md;7;xqb*v-PafH^regEsbi!d%oUrYR{saO=*blcNF-eu(8+OWJ>Sr>kuMwkIn$x!Q_j@*t{SRD!$?cYf5pg6f|zmwX89XY>&SgXJBXI04y!vs?I;qUcg58eQpfF6_@8V z;$W%ah(+I;z_9uMHVlg*;bu`H7zU~xsa{QTZ;m$*YzBLr;PV+Np&cULuJ!i9nO~dg zM|ffV!`hV3%W(|+%dJ9~QAL+x*IKJ!vSJzQ#^NGS+WtQa)cRw#hwg(|Ip1xf)09lkZAtVh82j@*9)T$2X5!k{hkkI*@u{m-6z$Q&Lg-3 zObq4OL@55-RLsrQ<870XD&vARLPNlQ@2>f)Mexk#*X{U}+ct$PpE$cM+tjh2!#qW* zCmOH(Q)+0`qg~?9erLS~0Z%$lpI(O)kdLu3S8N(*sOYB;{(Lv<-q0oe3&N6vt&T8g_R&0hUX=Bqq*-&e`h{^O!y7UTG~;xcEeKdO#nd zcKJxN(LPy1dukaR!48k{=#Xj!GT#E_n?qBbibnenMH`+p$g{MJ_5M}QGzPqc&&A_n z*^JDb-8hL}^J`qOiVGE~epzW2I*tVhM3ljT!C=Jw;QNM9ocP;gzI?_8j{$3M1mpTa z<2sC#Ox%lMzr~L(DNZO=P`oMGT+{)uvag5GKoKNlZh3Bfb@NaC`}$uU@6h%4&@gc9 z$@W$C^x%q?(~OhM8(~S;Qx}?jiB~2TOp9bDet5HI?XO<`ed#6dC03vh*}b+qTN^&& zt{PP}ob8*(+pe@wM0Quo(5|4E^Jkk5g5+{9WC_eQ#ujB4USPfQ>j__ILX@jiGi>{~ zhDw?4tbdsYQR5G!JbU&h{#6B)6$=AH9?$-}3k3n&InX<(m}{opX>4q`O0-@v!pMyZ zA_%FWyI6D^MIs={<<#$KuJ`MQ831=9&}S>24+TDRkeB}Q?BI!#8Di%L8ck-nXZi`AdX^44c=smzr*4}U0L&3NsNQ&Q)T{Vabz4WCuc337ms5G3un zvVKqs7JdbfI_3dg0m_9s_xxfZy=uZfq7{TKt$$BK4Ki%Ump!d)hM zx4({ywJNi!Y!&+MtYG+|@9vj+97w40%&EPR{j-$QX$>_)k<+V|11ls33nf1U^}=l5 zR!vL2QnYaMf!VF~0L#xHAUzHI_+hM1=ux3~xU+WedSN!>J-5 z^axyT^{U^`!NYlg+-G8ToKWt}zcSG!Q9kZ3lXno@yE<{ZOi)38z>u*KXW zmfZ2>X12SZmdeU&Bnn-CyE~J_zuP4SKs^9LV{^}euqJOuqXsVYpI?nJZ*WR#d|s_^UW*9V(a>UKXHr;fXrA8oz>V1mmZ8Dcat5{ z9y{8NyMA#gA7jU#R}PWB22y9`pSm?0D{k`Pjhs?hO(y`j4TG<%#C7XtSfScY6 z_Z%~IlCtYueZYpPc;U9**u|;rZgedG^K}2F`RI7~sGqW1lQFIg(pTHCejRqOPigq2 z-46XQ`VA28o^0&7#cyGthv&MPr&ZJRR>f-k+0D$J*$3N_Bxv8@6{ok}mV5x_a$y zQ*pkFHPlZWtrnZ}9iV!zTgk-tIHkH1#8r|LOhvCxW^ooXRfD=6+I6hT(-mZOE;u(# z5`DS*B&b-_f-pFvHnN33pn41OB^EKi11C-%M6VZTxzn!n58$qJf$Q=X!j> zctY279;ed~s?gkTsfqYP>)Fz7MClK^krKy+70hn6Z9GYwn!h3gYpxnV!_KOyxBZ>IaS*;a`@P+jB;^ zNWY)w7R)9#4ZU#}J-FV-1RGhQJvU07Cw#}|Ejf&ny@%T^r&wS!Da%L>3*>8GP<{ik zl*fEtImPv90h;U-IJAQP#=w8Tx!fSf(4fY0kWdk|yFIb%`ILzjwY$}|%Tu+0V$ax) zFSZ}`Tlt!$nL4pUDG#KYNqa4fW;>Gk10s|t#?-;`+JSBG8hU_MxcyDjYEy5+c~&jC zbE`#Vr-%1Zhct1>XVG_J-g2`yj&X1}cKf+Pbv=6W)WkzXqPtq9us~YRIt>cn_ALGW zRV;8)N{k8@cZ!_QHoY^n&NqcKaxja$iJ8qk{ywT!Q7dL=HS<%P z?8H2jZgJReS7uL7t$2$zk5uxCW41{Mh6017V$62#m9HOehK`zzFUNjb{EUxV0BRv9 z>N-;Cta+}<_}t1Swho>f*1a)_m&a=M^j*OY9a1AG8#c_$3obQq)OLq+U(DWu*<4=h z*JoHsvmqv6a?9KW)Ad@>Bsl-!Y8@Ui%ju5zR4I=5)Cid9FK)JN8dXXc9;agV{bOCE zEyiSoa7)~*OuQ^5%r>ZajQX%WR${a0nLtFHE_t*PR1LmdhES*^#kklZ9D;;^VMnH) zKi%S$(LRf)L>c>#ohnfZEv--fTv_C-|E;3nr1})AWBGvpBV>yvC5zii+yf+Mhem5-J~M z)*?@%V(|ex(*cZA1utSWZGy}#L!}jqNp-ulI);_ZwWLMA=}#P#n$6|FMm9Lg!2%!u zz@3zvztW{*#li9J^|@Q%#TFbWoa1G22ZKZEq(IGS4;r4apFAb++zo%I4?&MJVT0fS z5^R|I<(*;M=31>1*xr)3BV*VL{D@)h*FG zeTH!3Z0jc6n2ip^*)GYV|175zA8!QHt#P(+Vhcoxq->9+uUu@5250qjdR6puFuzF)Nyql>!&6&XM%U6`tn>#CQC`DfMr^ZhGg+`nRXQ z(Ljjn)tedfhxPjn2z@)da|1c;3jdVtmP*~;StPhWat|fLt0rDIKN_}=+h6OqF%GJu zYdSbIjneHaxgi=iCa7%*5YCNBGtWcxa4Fxro=$tSJ-=a>AK4a0^_Tu{w#I){XDQf7H)iwvb>gN~h3n>p(XziB->abm5oUG$*(sQ`Z9 zke8vz=YWK;>9*9{ouWy`pJ^zPS_5hYzOD6{nOp9o`Js|?J}Uv&xpmFY*ix@9uZ%7F zD_jU)KNT3WU}F^_Y8cJ^S$8#bJrtms$CX6M~tpI@p`vd19p!~uLO z*^0KODH69A0|KnM*S4lr>bCyTyw=+|@d66j_ptZyp%AQX%g)8!#7CV6{qmI+^>Z!K z-v-%=)9@zeH`Ch@_RS2eSMFSlR3d6;>aR?96}-zdT)q=Vv{5wfjn+40#Ft^)6ANl6 zrl4}%5>t0!9ytdv0aR-eQIdq-+e>2!JU#tt$D!bm5GewyUBfI{)vYdkjd&&BGFqEatsdn)D_ZzM z&0@HlAAGTd&D3@f7cQaR)mHq1a}^Yh{j!R%MKd59cw}95;jY3@UDL4PH)_BunJIS~ zzjx7XFUia~FCs*Kf6^wabPSm7?m(VZoh3p9QUC9LqV2uurC}9^V^QaKc0geLS05jW z*drQiXJR$dm+O$!)zxS>ZfC1ItD1U>4e8Ot)6Y)m9Psq_Te18aA;uWMy@+219VGtJ!)7)4hpTHf@Chowws& zZ6cX7;(`6e;bI$&LdwV9S`MRoQJ@rG9IKM)=D(5a^1Y7W8?OgYjQ?ktd2IexNqenICh$McCjc+2hn!2w`RJY zH4HO$>cPs?SD2YW?KatsX=-IyMReVJ{rih{gnW|!<}9&&YJ4S`x~e?8(r&fmF;HTY zNZQ#;>%RKpwK&0Pn8m!dCC=<6E7T_pij8W{Oi9^tpGi!|#P)zv83_?BzxVRXa8idXI z$e$I5X`ds7KD9VLoP|Ye4aN6kOo|=$*XC}JI-XhJP_xd&Q4od(&Klt3nhG1yv(Bi! zl>^|<#Yr>1{z6#Sci*v~;e~j&y1yKiW{1dX) z6tQw|`R5zm*u~4SQ@Nkyu(BPwegLh4YqSOqWqd3t<_z`4e*y<#ImI`c7wR%rXj?)p zhR2PZ@}&^TRJe_+ny=wGEQYIA=?Hn=h6ng_=lf6Ewd|np!Tkjcc5}BHH?O znqU$}bW}9{(I$dpvW_B+R?1!|=#pIOhn$;FD2Jxf%E)#v5@vUhg3aY;X(4rP|5<~@^B){1QL(CGlTwNgR4t$3Io^xo?NwsP zHdfYA<_-EKc;%^pg)-KtO^*@B%>A{`vS#}-<1o$~#JIXGduzUQ-L}J;ilNvmE=-!{ zmkqp4ZKA4|V+xo%76~{nw}Dl%d0G1SkESS;CC=||=8{Q#UkF~lbbI;{hGzaDlYh`Q zj(juSq6rU=RAPVIN$riRm0Nwjad)FVu>RI4YNX23Y2XT2m{R{~NlK&)3?r4zSxmK! ztC(9|d;HbOa#9C2n^vaG^$sLIJOA4Ze^$~CK z@taA_)K0J3;s=|#aH5*7uC8}oxBXCiaxP~KcpdfJC{M_7#6L>Dluy-~;Wq0b6wh>U;T9vp5L}ol*_B7q_vVEN2oV{)7GBy$|{)@Yo ziQljAf6V(}IV^k;@=OD(Cd~^m>lq3)B844lyvRH2H3V_8XfsJ91eR1V$iVQyRv~9- zXvwjM%_wXPq+bSoFNpcLg+m?FIR;@SB&A`cZ^&AoVXk=UL_nS7^H=YdsAvABn{Q=^ z)!9)E$s_%3cf$6=5oaK;=!Fa8x!gOPF63C`v-ZHNiOr~@C-qib-5vby$hP)!VYmy( zKB)lzb%o9xv{iUBBxU;5&dT3u>_*jpEsvCZU;PC2#ufB&Yeiu0nUAvs$khSB=`r-J zC+}`eXYSx+)%nn?cd3-}I;FyDQ&I+PV!c^mRLXxT^c^enM+2GC9L{3Ro1QQX+cU?N zY@-^{Uohn{-N_JX#T*GqlzIP#AJDL-WSGz9Y-ZeHt)x1>7d##>Cq-ffw70gQV12J- zkYDvc%|Kn+?#odqXE_P~Zc=D{b)XFEds?tnw(H8IM!McW^<1x}l*@cg`(7a=j7t`I z{P+plht0iYPB3S9>T5!#8$ed!=H&`rad+aqDB2kl5-H!5m921Mwk8QmP?VZ91s$hL z|1BC6^grL4Uf}qMgl%DrO(oTufneKftUpV?9U;3xfZ`n=>&+0;xPUUSKq?<~TKuav zaq^HhoK^bQ38Szei@}`M51cuT6eZuNS@#xRBY)exT0`#gQIF8tD`#FWQ9&8~cY(sV zWF{8czC=5sM$Qi@MQ7FVuUO48Q9?#$W{b;k(M_SO5VWfh#?K?=z+pJz{cw1W8Lw|7 zRICV++CSSjqu)DcFh8~OpM~{rOk)a)pE*o6hX;D2-5!5Fg^VrxV@v?wPV54w6r--b zEVa8VwQ!w(^A&A}T4$wvZU!JQ#EN?4(>%N|h8q*JTnhQ*;fHE=%S*@!SPHdb?$BRa z@TxNtTGTd|b_m#QNl$ih4#q`V^^%<{H>MI8R%i{BylOT$H;x!McK+K7m96)HeK5}c zm~09vA3W&G*UnBZD*km#?u#`}4C(4o^bUqgai=bxUL3>|VTLaa%5+C-DoD990lq9C zp!P$(9ebK}PqecBDU;Fmxc}fBzBZjk7%6q~lmYv@ufQtbwBA*Q&!D?3m9u>k#MUwT zhW_`zACHxWH~%xWh9-(x^_YxQIcwH5g@SmbZi^_zl(E9=zy2LKU&9A8er!_St(v28 zwX5mlj(J~(9R12H8amTf^yl9*E_!$QBA3TO5(cP<-6s385U0$XPtBAli_xOayYiP# z{(D}k>FxF~_S8~WlH$slWV#{w#W^DIE%Ub}`%2P(~9ANA4Gg#$Dk9p05!zwK@D0LpGjt!~W@XHCsR=YB3j&cOZf?9u4 zo+xFyVL>x}Wuu{O8A4UJCEF~7C5;COn}EWWa_ zy|vWRzNkK(^U+hpepDF)D#1Vkl9fNBK9yt}-o4ifsBj!_b+|DZ$}Wh&Zq+=h|z<1IZ_}KSB+3A+tfOXQ}+eh zukb*1yzRbYpSU|!IJr~$PG*E7wXFoDW+Dcs#n$nX z#O0$T%UEYP#3n%OxO|UFTY_6HJNu)xt@^)Cn^}YrJr#|kqV*|_jl}l2`CK9%L*w6` zxfST$zZkIVn^cYP^m@$wv!{eM`6j7>vhcF~8y_@gE*R<8X| zs$-TWbY<68R|UAgn31BGl}?cW2`leM9Ad~Spr-bQ!PY5SB=<#)fU*8{sVa>yy=T_dC=O@V1D~+0{0_HtUviz6E8S z4g=W+M4D-{l5jnK*uci|mb3#__BL=9N*nz7WY-#%{cn5i6#;a2;E}{_7lZZ?HZde68&7-_ z<7npXc7l-U?7icwy;-8Y98f7xPt-!e=hbYj($+2xl-F&$B;}0=`{JmnI;oax*+!Kj`WY1$ zod%;PipjLn1+k3P;IJ6{33<x6o;Zop<=pp>C9j#(^J zkKjs|UqC*L+wl9NB_kf9d0l=}vXXO*n@}LRcXU#>yS>Pi;nuG>z7(;nR4<2OkJm{M z6zgwTCKY$ZE;^<64S4`|*1a0Fb6~S~Z>O%<)k|8G$+my!%?GE;aj>}(6Rc*#LG=>> zyMSr2S4=so_;92~eXF6NE>lBxQcBiUb^IV!D%$6=jrnVXU>AOg*oGSx;|9SNw)*k% zbBrTjUy}8AXeDULIm`P3oVY8Rb@xPYB%#p8G4YW^kV_h@lw@5wDk78Gipl`yjKG_c+jAE*kf+aAm68(M}`_`!}Es5!Rq z`8r$Is^LvSQO^zXKmt=NyB=k3n>Z*0Xp~=y8KBA=D+q8_Gz<19eRo%J2Y*u#Hd=Lc zvP31$Yrjit`H|6RhE!60m?6#|6ZsLZcM@I>n4TJ>f`_g_)PAGhd9Pii-5zFsZ zf5#4vHoIui`Z_(ph#SXB2>r0hv)UwaVVqY)ZLO}G@*-E@EEQ;Du%wbJ$hX5!N`-fr1#j0Nc7Gi;y05k| zI<3r6VS_d%o!@!E=X2kB$CR|=0O%7$DyPC?mFGP2H4#sYwpwrC*w=TfCt(AD4Hv^Y zLPazt%X1<)ks6B;i~w1(Ssca8D{cGC7s#6Tt%3;7u7BsQ@SI52v&NnFOo|NH>%ttS z>b;<>Nz$yV)d?E=+b+Sdxs_2coJO7E3@=7tFZ7E;EONb?>_AX2sf1=l&IR{BY69s!^+x{ znG2xv7NYH-QyDQa7kmWK`!H`jss&J`R=W zDl3igFGG|05r9(eyGfJ^m<3%EZ?o;b`jWa@K(f&;h{;=7Uz%*6cfvf~F5f1x2CQG- zh$88%7vETZZ3^ff%utIJ`6f8g>uwAlpg!T2v3toEDH~Z={ecbc!+uW*+kSgX^Yksy zput`FY@qh-O#J<}+JDM755EEO@V!5vL^|Mjt2b8PBqOQ)Jm7+k{-rc@f4Euxk4AJ! zoIzT6U9PKmvp5JA&;#>!?yd1-tTP}{lCW}x7dRqb4x9X=?wXtg&Ukva=VHxVds+|7 zzAY(Ed@x-ESd(%JYwU2t=S^&I!w@jnVdz8}qS+WAq_}V=QvkFWYr`0QsIHXzyrEN2spx8Jjw3t2D?la&jbzuFv-3i`9tZrI$YnI%sIosg({{vG+A|4d#|+F>)*xGz7}y7H*@8nmY`EGMuIP%w=NsMS%s-IAU}yr&qa z4LHS>V0|GF`L57+!Q_On6===LX_?ac&@NJxt0x+hZRypWa+N4#ly8h}kg%xg2jqml zBb*mf9&m&7!wrj8EAM%C(FrMY4JP?t>(vH^nQlxBC_No2)*N-^b(AghM~Gcvjd}tD zMXH=b;b*Qtmm{9vGEF>b!rtTS0mv9TpFY`RbPSbG&F2<(X1_xb+hSlQ1Y24NS$2n3 zjLznw9X;c%`Coet2(UoA1tpqGP6E>}f)1@mfjk+Sx*!^3%k=Zfy4tbwoZDFHeBG|N z2dGZ0)+Q&h?Zw?`3WgCIx_;HGy_%j;Wla>ev4G!f)85x6I8>fl^<(#s4qHldng17^ zPvq5#gw1`-wuJ*`wvV6dC5Ta2xh$VwZ5h>mjyy$+nY|UklU3(xwzYPlOe!d8<2S=C zN>CdT$OuGq2a|>9HOME>5ofl?XMDQ9NppZph_sP`+BmXLdvjiOwIlb+4JoFJaaR77 ze>ie_0*=y#InYW?0m#kJMJknj1vK08>~Ea14q<$$r$Tl|Oisd0zx4Ose|I-*17RAY zA-L^%dgsUjSv}VKCHdSA3X8r^D0BJC{yPTpB4eQ7EbiZfIQAb(hY}a}s)h9eP3fu8 z7$eJJnh$EX5FC8Y3)Fye9qax6k9TYObdz8PLXO_f6W!#*pHX*yvsx9wm+$JP-E$1Z z3PW93i~OpYfMNCDNUl8)$rX{V#XbxEp~u9Nvp=S5NrY!`HNCB!0wvuIdM7$u4d` zy9kQ4u+3NH`?C|PL!K?klCy43xUAY3OF)740Bt5B6GkM!yU2Ohe5eG~LOD35TT%X- zI4b|g0%lI5vQeO3o^vP?uzN?yc}podRV#GYtix#nY{mX1t{VMN1JfK{VUUhjPZazieuV~-B=!d zKC;3P)B#FZhm|ZdBLW6vAhpwpYR|Q*tDQr4C0^DSauf!1$4BXUB86XLwEGqyqTIi= z945pZ5Oa{+al=kjuz>IT+%SVqYqvkh=*MnNN^kChuRU{}DX z1tewL#jbG#QHdX-h=Epvp;nn-wudOgBnyuR?b0YNBA^t$PB`8}p&ye18MXk2r<>f~ z4o5+RUo1d=-uvVcnmI-poqgyPl;AS4G~?DPl0;Fv;u40r2SfqcnnD_`w6Twl(E|sj zc|HZGL9_B^Wm+Hulc5VM;no6fQTdO6gA9{T*;!=YWbY#4xlLRPQm7p$;12lt}#mw|3O+^z&1)tTKWx*KKVu5QX zI266N)3Hs4(;6W#{M|MYbrtmA)jZRa%FG4zReHl5sMV_%{D+OM&dTm<>M6AOoT#Mz z8i^vwmUWq`>~hIE%={sDr|#{oekmYqG3OXSesLC(ekV&9ygBSh^tm_ZwLPHL4!#!k z7mdnhw6L{D`~dRjr`k348WG+>3|h##C4b@Jgi+}ho)g0ApzSoP#l80ow#!h(jlcv> zL^J;I-I)QH;%k~pj;?4*OEfkt2mva&9fUK-q4r788k?CdjNPrXYdx+M`{v&k<-Wo$ zACOnWMD`e)t(W|z*$qR2*x+l?5G`LG`mV>A=c9dyW1y(;ztdQ9OfhAzkHyh84*%c! zSZv}tbmC=#XsvrVP$FytD8grehmQhyBWQAui{0SK&Fw28{+QoQA@-!xdwn#-R-oLA zP>9vysrL$xKw#ec=cnF^RK9S>$5cpY&wR8)*x{ub(jYP_;RB!&ROU88j8O5vopOUq z+O{$9DaBH$>`h{evMi0Xy21mZCPJxG+}}2Mr8_qwc-38012U{ZtTIsrp3)PQe6t{kTW+1kGd3~;O+BgD^2}i4K zRt_U7p>CW%?+X7h-G_jv`Uu$Mk!bN2Uav$#wRbV!ym|s06O5MM^d3f30NnWd&IQvj zgF_c2T1t-gwF{@VT>v@(G;tL4vK-&6p@+Jh=-P654l6zac;ydai%+$XO9<$OwO~7CJj>?9)Xc982lETr9prV2S^*reH zs8l(RW|05ZI;zZ7|I?efM+#l-x#cPUdw}9U(fny? zMPVz?fz>aXVq|~gVni3}N~Mgqv)QnGXjL2@$||UUXwwgvZe2Otc2L9e?gvOMwrC-< zg3+d7jtAKdeSj$*+-9Le!ayfv))<<~?X|UZeV?FC6X4#D0lpRwwLT=m!NbQw$8^pmXr&1aubg1$17W%mLx(eLgUVC-ziutVUhL72ghKz}4RdX9!*= z?E*qw4inCoxgbG%t%)skl;c}}*zzc+_1?5{PtrEJgaDI%KWJ)5IRlD^b&AC3yMZ3J z4i|wYZN5tvR8p1h4mA#rm0iud){*G9tUC%?tD&9ZctBkcqARI?cZ8KKIsXJ5Qj}Mv zX&7{}8B-z~*kV4k?*jgoo>W4v!ONg+t@gr?wpyBbpqc(g59qgr+EGE{?&mOr#8 zK>6U9=MXQaIcZ@0BOE^eVBUE?pK!hWYq-L=`H(e>>Pp^NMb3z4Ix1P71Pn*wjCTL-lKP&C5Q^x6ISiaExY#CwVle}UfL^tDL z`Z&w9Z(Y#%-kGU53!5e*eAfduTmK$5gFQ9i?P3@MymZw*= zIre2#g34_>+joCE3Lm*<#J7x^KYzjj>RB7ibs8;mQQY@5R{(j3Ddl2`o#A!$z1xpZbnEBg@cpQZoSdH=}!aZ!2=ldlvd znN-tcaelV=N5{gMGN+;OcTe*KkiVExHvWF$2Fz9CvsBfpmrWlqh$N^TyB&X(3TiIU zwFo%S(8bu6YMT3uTfsFE2im#DK7nr|pM$`P0)2IMg%?{tW7l+-Q z%(1xvYDG>#GO;;ANn%YfF3)-R6XG9#Yon+DzAF_R($F5HOfr9btA(i@fa5;8r)%wsVT-ie@K5Kr&5(w`ERu*WV0cQHy4=ZD0h95;=FIDq~8`pqo$H9#TNl4HuYHo-RMF1Z;xfiM=k zh8gRR?7ZfJ?+fTt(DKgXrjH>^(?%x7JgZfcfax%&PJX=rb3JmZ4qD5fD_GK)e#u2K zb0g`Ft1-#gAXQJ^Vls0_`qPn91@({3!st~MdfW}6D|jMKe4cqM8x~>qxw&8yCx;w8 z0`REy(3c&Swrw27##^@mN zz2f@>Zn+653=G%TsnUNXMCP!n(*5rqf)1wrBZ9i@10C|Hc(O z%Qu+bgAe>xjMKWvjwmD8g68#>1ouguq=AEOi1MIJr$ z|NOC&&-s29r!c&dV>p&11YTiEkv=7_vbS${=ilw0fEEAd|231PVN9{^Ps;UwZyA?C6(&S&VQnGT!u(d*Q#deZ+SFhpLlg!K0bsTrkgp*D!zI-DX6 zR2~+>!c&yNfbp4~-}gn~&u+dN!*^!3e|Z-XKraW%eX6&50&aCJZ zVFSr!(zhh_q!|Fgj4|xxmlM3I^j{Qetc)O4r^}vwETM}`JbW6kBA~0Zgu&+9(k*NM zf!2F{OQ!$hKCPesi@o;>XlmOQhE)VqnvLG9prUjL9eddbh>ap00U?wC(gR3WX)0Uk zK?Si;10q2}Xp*P|qNo&sKnNrvkPreS5XwJ;?o;kP=WxEa|KaAr2U%;(IY*zPtvkZ+ z){W%+R#Y#F$khg){giJ_aG`tZ{aANPHLRg*RV$CArCwRD2?BdymPPEgb(?+**ShI5 zxL3bMd0BpG+O#h2H=nJWFL1fCt{p_`z~HcZ8y?sFg|xZPNX1)o!0z4Jsu!~U)4*@D ze3bK}XU9%X3NxoNQJaptf8 zzp&x|;s4jIb2dN)UL$eQ1?D<|Stz?6{XYLX?ntL~x2%4zy^bsD;PKyE2cv7a6YOq3 z`^_i+_cB1v{~_%EnB@Niat?d)~wbZ7Lxa2YF89bi?Q+es?)i5l69?*^EOnTQtRS-o6ZAv_vOLC4q)GDS4-;$k1VuYgfJIs zX_hPb_2FqXuU>WIqLicZn1kgVkm}}6JO3OPt|hc53-ex{Q>K5c0V6T#NKbNlvL=+OCNo+eFJu`nk{y19|{ z-aUKv=x#k4pILL`POUdKg3_r}rs<-B(RB5$2yEKVFcNrglq1e*TCQ*~62lVCojiHc zs3pQ*X{i6$@BsC^VvmJekkabE$SwjW=#$euI5wa{}j3ZBlcqL9UW7H?)aZR`)$MYvo7D})_ zl->&9zf<5?atw7VXLHo9+hG5eGV0c81cw`CXp*np1~qA0(@cZb{+r{IZe1?s`=Yx| z+6E`I@v-x&DipfZgHl628Ijbaw1@fW;AMCMXXI*QZR|S>)vGSHFJ|BQj@S%EaG}o0 zxiH0z^ZR!hN35u9yi8WkyWx`=3H8TzCzUtc-~7gNGt>LMP`QYzx*iIjIW4HnkK$i)`DrZT_%CxKA&X(iQHOEufs$14nK~j7cbw2>p@y}q^iC;f5 zCH-oXuLBe=!@C06~ z+cv{MJ~9=&GMUO5u`=-8M#PBeRNaGvv))Ro5M#GrjH^+JrEcH0?S3^elxi11$=aLl z_yMs$h%suJ$uFOuzPTWK+v)*Ns|HOqMWj<>sjY-7fVz2p`7%k=Bh7u?8{EYnm^WDo z!M1g6Xt2IJHRn2I%RMg^j3wKm5i)ge-}%ZX5|2J!|f2F3>Fku>o9MPU|I8J!~Fdr2VI_3R@qJTCxm z%U|#=f(Z3D3TQzY6KoYWUW75Ha-XX^T`F$MGz<@#hI{EGQ14LNpt}Gl63RH35L4MCH!pQ^U^1ypfksQ`)X?L;>R9^oEuDy5wXH{S!#gE}o?uU_*{{J~h$EY+_cD zBoHbHNO?Tq(Kym_e|(J}$=fr!O7O!))vFuCtqv&FeTA0f}3ZkxeodEbB@KcdZ3J?GFx(@kX_0Hsnji*GD zlE5MdtGqWwRwdoO{hAx73r=v>$@vM!HcSXU(7En;%3x}-6EcEhQndh=dj$8J;D(9B zUoUm&TtC9=vL9yAI=lxQ`lFY0%-{sr*Q6J%f1Yyh1dQ+7casC4DgFEYlf(C3iq`TW zsNVq@IiIe}hIj+j@?DbEiA2@>epf@-8fdEZ%7s?8J*Rj((v+Hgi8hVMPc)`xBD*l<^U zeWdN-4x{f`F~|+J21rz9(%o!%%ecYh>PC4f?z5kJkYo_=gq}cxQlyR&o9esFg7fdc z-i7lJ#0k;_`SDw`Lt232$L^r`%Xn$ul3-#N{@gr8cJ`b5+_5uRbZGcaYpd<$uZyTX zi*6;n%b&J_XFWjc0%MI=7zM+cIsxd1)>9l(RJ0zFsW zFIMeYED9h&>HsrbFSTEa)&)s!eU&G84B`@pF>4xn+As z^J}MaTH^b`Ck5~N!67~i=os?!47hHCG^MCi@dw@l6= z8hcMl=HB5RRoHf)E6nfh-9Mu!&AwvdkzYhfkU&X;44{OHvyG#isJ+f$l*~_h(+_DXZS1)q3fvFg0L@*>~C9k&?LZ}8F?HGn{49~91` z-pz~!!2GX%1^M}=+Is<9;2#A`1iJXg<7S7ob^>V>?>82o{k*6nPMa5faK)slrvaw+ zXVa-4hfZ_xdT?hlINaB+pWoy&!HnNmO!bdoOqajHhzu8_>R^{D&)*G5+7(IZsIr-OF* z37qz)Lno`d{l)$UH8!yMukD3xr=_%Q8pGAr)aVGj(A!5YDU>}R!4X$ZP=}ww@s2gO zmalpdksx#SHB@pG413iTm=*^5!J;>M7MXF$`PVW33C;!8u8=L*<@ls0f)1bkj=hC3 zlq3Mpzf6M=t?uV8eFg1W0icyMc+qfiv->e()9&=ync@8uc62O$!D0SxFC(Kt-Esit zLQ$(uh;Vf1QJhP08N#9dl;v`Jm%w5i1OY-c`A`OeE4XS6}dW470Y{p zPdumi2isjt9G<0$XHkt7-Bid+EMu-}P;(xN_$;gZjw2@ApJ=c}ieHm!|hkhcqd@Q0e zW;riBGD#!m>AmEDEtX9vf8J7*77rha9G)YmAqV4B&>&-} z%kvS?*Lt>3A1AXTC0v$q4MizHuD%5b#Gn zqWG0#!+pgaRyqtDS|y`nkk|{VVs`{OuiRYJNJ_G8$^_E01d6|&FtHpB_l^<41HrKO z*S#x930aRnCb+n#>Oq)ntIRghyG@pmo?Xlk?zt)ur)NM(~^I4d&mQFrmRdGw>aX%ns z;Bq`!BSX8W1zB}EFQd2UFrFmPT0vJMt9N+g}38+^sdj>j*4uvbjFY!hv9YtF;#pX92j!8^jUCpV;nK z_qM1;GT%-mp58{Z2mAv%J(F%d_uW1NI)YY)R>?Fn-Tmab>xPn>Izb&j)Uv$cQ$6Sl zg2!ZhFYR-4%FVUo?O6K$?r07MRpSB_rkFUe^#od#p7Ktk@*@Mfb$S>P(mp>sr`!WqK3&Tn^j9~OKuuUH~$c4ZP$!MieG<& z+?qQY?|i~cSF6hOMb8ti)&NT3Hz*C-XvC6{`$pgbE&!h^ru;E#Q|V2{LX6sA&aFWm zq?1I$OFLhmWngX3nqQ`y9Q?w!G&^$s#Y16SN1dasoAgGv&?gcbAjG2N%%+Ts(_D6) zIoWSn7v{d4Wn!;$^-msl4D#MRe1;_qYr#i7Xm|%81J2ue>W1IvE+q*vB%hu#9=?M~ z0Nzp9)1Dl9)Pl#nqaYt*ru#e!ta3W7u7Zi&^l|#Q+D31KC~LJ7kU{Zn_k-R=R2K(R z@@v5zO$adLl;}?JC#|Qhr8x@vz5zwaZJ>WavM88NbOQr;#DIZG1Y@(ldR4V+pZf$- zg}>Q!i+A`)E$UVRVRrAM(X6uj>GEjAv@+L}yKL)!(D{wMT&`sqasU@4Jbtd&@mwkNlAX5SX<7WS6J zq>sbHCGM(NUp+u|MjzTs66&UT=EcKGavj21`Jxg}>OJ`os|T zD+;i?ffJh*-2k7uEuw51{cwUMBCBk)>LlC;mr9#h;FM#UNlz=`PMPJ65#Zk#HK3(J zfUjTO8$wQL2hC$%WndggM~pv?Up^vtbKcv66^7!?1ZVcjBGabne*p#l;Bobi4YqH0 zB+vT`b{O4-z3EH`PVkVDjI#DGjlisANrG__k?z15d}n=u6yn|5mo@ZKe|a&3ykilx z=#o>o_B)-X0#y0-#PE>c%-}$RA+6iRl8;!=**vG(hjDMUrG}hE6Ut*Ef6kJK!GYTI zu(8f(x%{K4EaO-0rfg|moU}%T@^5DV3k71(>C?HS1q^b3i;@J?>^Ftlw`|73cUv#C zlLWBDaC(c`s^4G9z31vPa|7Cm99qUkV@q2tv);l6uZ79)5-qsS0bjDBaxAcoAKO5BKqB4^4FncHz zc%M?PAFp&PxSyCE3#gC;nROg})5-|^1q{vTcy)f)H$j*!k_X(x{sHP0GO;th=A{;` z-VWZBsj^StgOI0a^1K|mrS}hWu;i_BtInv7XdrT`2IT;i1&2duykF|*Puv=Kgp-Y+ zUSK3D=g7l>cyg0pt8n-mi7wvp*uddxt`8{1dZdgtE%l>M0$^IJEc9q5KVVmp z<6%JpxbJkH-SONuD_8Ve9igcReDpvzss@A8wIDS{pzR~crRR@c;gvZv5z(qTwTKk9 z5&@)2?B<5KBw@Z|&3#{`02BNr0nX=Eborf6HD#u$NHFWO*rrr)@l|KX0=yI|TBAmI z1A6X9sdFZ3=ImWS3FvUifgA9}55|W?a1~K9XM*_<#$D@Ii8hrDyn~s2{X7{>^))7E z9q0mB_$qRkqY@Zg+)(bbZ@@kdDT%E*d!@eLy@M6p&&Y;+l25;I160iBit`HPzP$3y z^bWDn$#R#ifta{cmgC_ig-x5WovWy@7hi^fOSOlh`4Ow@3vywZz3vT5rh_Mh z)j%LXQ*H2~2kuNh+}YlNVW#lz)qW8Dycx_52@HoS5*l3o3JOfdg}aFZW#dk~4(7ynz~WX6^sMTH!8kmwR_ zk|11(s}Tn%5KFQT=@m2%xW0Y+5;EAA`%Qps(1muJS@(P|`eU74J+N zaM6Fgfu;)`2%koa{e!-ff%7)Ex*U`c!1aN6lV(mW!N5ILcTNS`@UoK7M44(4fq*hZ z3jF1Megr>41R;))M<^mx%A>*J#Y2dyu6jiR6v$+=AsX1yuVPIS4YyCN!u)nf!lXg> zr))AKryhhp&C$NVlAO+A&S+3>8_T-PeHV96mCdyc(ur8O$@DNRgL*YK`jOcTgDLnF zdTMdXV7`{SBLSKyG@4}*S5t#x1u^2CjAl1sk53DAoV*L8bpHI!HXng&cx`ooLk32^ zPjsh$7zYyr1R<~F4Z~hrS9^+5%}fnpXdNG!hI-IR1`c8D&h@^V>z$8G8G(=|7bUCG zug32X4YRNY0oEke5fF!bSSui&vp#kuJMJ96Ovr`p%Cq#rh9U@txx6R`V+l&*iF^&2lqcMDTBn5~YHjAE)dE`^L^HZn<6;6(wZ4bH$#eT(A*EgWJvoA`H7~dZ0hAmTXoN?VIhIIm!E%}7l|yBN<;IP zw%#kP*IjW+ly$?VNV|i!C@|*Vx@G=~DL(}lSz$a8?7gc?Usp~>JXHku0jCTAK*1e* zU;Pn4(cKaCG030)j2goMt$;_mexL3sua{JII(c>$bYb_}-u!&Z4Zx-31rtfON}i5# z_{2~(yqo^p z&repyPCM|%{ybk?EBCYmBJia z@Gae;$Fb+>#EA2Q7-(#1Wrge;oIxg`BV`q=JfO?%k^cy^kVOB?k&ncm&-H&M37>xH zwi3JCBk4veVxuPf2A4$6cCf#%+=_kRn`Tf!R%8D?lT>$v(ge>GXO7dMT<5VL1OR>8 zR_$N8V@cRQvDJM8g3-!@TKbb-%A-EIwxxY|Z0(QEbCihyg1OJV!4(z7Ow5ungd+>P zFx|lRUDS1C3?)m&O7SATjwf#bH5XZIoBia%zckTHm~;OTVZ;xunz#^mG$2=49&O)ty{a z>R0Ll3Ce%1HLScDey_9p-0@mxWW%fH7a-;zHkuf2{=0<*+c+ESS8fGGH|!bnu4M&n z9vc@;zGB|A>WjQM3_OBX+~+N@xv#D1baNV^b^XQC2vQqQ`+~ztaqo?)CzQTx2&Xp- zpL^Zv^3$~SX$IQ=knlwmbWNcdi-BU?Oa%x_Z(-f7#&VP%5Fds6X57QrH(Aa8-Mn^1 zpy^i`Gs7A9+ZHuBzTTl>3Wx8`)&`nu4)!}|_PW+EC@);fw5Yj|mj0VeTjofAM*GKy z@|l^@lZb52AL=5M7IdaONMo}loVj@+D{*@zaQ_AR!K(j6*9LmQzd;d;fyo)@k5}0T zDJ*XteNc}HdN4m=pOJaLfTqndoXQPl4W#LGzZ!1@`EiiDjwX{_4hw)>Nz<=H0yx*7 zt%e}|YIOHs(=t$=UT57NK6fgh(*u$tq-~_TgD<}MS&IG=<#+(4H#qBzdqa=KHKn(zu!fhEgdBt?TheJjR>xijk}6M#&A+w@L^?&r-$S{19N>lUu^u5q zi2sHM`~Yark6I#G^kvvdg^7bP8l`iFum`pCk3uPA^W_hrmrR6Los{xSjSm&6R_69D zzc7EH{~C5U^xrUfD=9)VhhILtYcVo^a~ACR`;|!MWexZ{S-1J=U|q|f7CZ#{j0XKI z@CA?EADlGYZS7zm@C)vqwcsY5Qb9YPPQ|pqLtA!d9aSW-B14L5Zw|M782@heE6tbR zy-%b|d^~xU(tUJQ*}O{(Z?iHe|K+3nb>aD&z8Rh}%5G{gKdfE17-XyP?nBkA`RS~2 zpODIeuIzDzg%r8pw;T((5=d8`gTU%Ww~5n^xG>Nl&)oVitW#18=f!tdO9Myk{(Xy- z>A^v_>;9J(w4g?#$uCN|DJ5<~j@k6qWoe$9hy_2V)pOg0NGJUZ0TN4OSv=$=)L$3Y zU-*Mr8(0;c(snvWNGroM3{{};og;{R`YW~*B4k&w93ndvHU$MG3E=1K=PQ{+4gBTx z{ZG%sN``nwlomB2$t=jvOW*8aiP+>cUV;q16nk&1q! z9AY=MFa-ti81QSZcRh^g{^j5OF9ijdGer|VQc8)!=iCjVjser(h(kCC7U88SfbRG z(cRBmE^ZHrr#Dk|rL<(M97;6G!95H7 zUh~s>kFN#E(6Aczve)T3Q$zmUM4Pm*TDQG%qJ~V0p71}g-B>TmSD zL4z(aq3)ynx!pH3VU9}-y(#G4Rq`3`VEH~-3*lWbb7Vf<5pVP9?*SwV%+_KV!i zGO*U;6B9xez%(P$q=_n&oDXK*qS^RY_Krm%uMmke>DHt zRDUi`TG#jUjVICFX$~Nv1b!ZN{oQLG!HQtefM0}D0i#xWaAZ_p7KnAhx;kIG{JfHu z6$?JnFvuZoV%DaTRiH^n%z^){*=wuEAfiJn3S>u>x@C53M6?3t(CdBfH?v&o}>`1rOFW@5o z1#s+0u^hg>cNu~)#G@B}9_p~n_8-ERfdLI&Z~F2x@Pj+}?9I_-i#FUV4=BJREw4$U zKWE?Pwo;&jI{TR<&(c2Vy{x64q}*c7$wt?-F;ytl4wrTAlAM+Z-sYl|a_h)F@2@{3 z(QS`pL+ls}(@^du6jVgShNFL3*#BS=uusM9)wni`607AsJaR_I+~a3-c+9gCNE(9F z>i&2VgWDK+)m{qZ3qm9;qRPrteO;Oj=0zbwdxYEciE8f*7Chq&L?B6af5 z<1BdGb9;3mJwL85`?ZOnI~KqH#>;;JI4)a@WBp|us_1HTxm4auqx1F0_PH;6BYO8L z$CtD=8jOX8e{UR5?#_Y!GTPbltgtni~X6Qniw)r zEA~s#)%umMsUMEp8q6=_6FqE{1|5O(y4X10QOzv?196!YPW_ssxW4>UyuD~p^3l9% zbrAboW@>j`IHaW9|A0i2e8>P}A;6T&KDW;@hY@QMV5elrv}u;eMS+aYBUWrZ>i#2I z+6$DIdF!tlrWwvVgCg2>=T08G?1LGEw(MRc%V~)jG2be`1etLlzq|XHYs#-dno(^J z+#_~3SOofGs!vDJqZY|p-l2b@#DQ*vDWJTKn!6U(krynBGp70R1U6Rh{)LG<9L!pn zxKFJpD96KZ+YBxcC1uB1o89*vfnbyKdPgg9hAhmx^9_NRP{*zZofQL-{H8SZ!P`}| z5xY55wBAue7zo3zB8h}q6G?0Xc2AXCo_(yk{L`+KsPMbY`;TOFPE5G#BL?^5Up*|1 zycu*cd{Xs-i0v;+)d@$j)>Q9Cq3%hOD5e#2?0LSwc4sAd{+Vxi4zECd%C*-a zy7X5#;jcVDW1y?GvO?LMV>$>G2nZYJylN8#f@q;WL!-W!Rrb>~}* zFUZ?!tkSE~0JXB_k-7JZtQ3fR_WaaqTK+d?hnRrm7Ho4uhZG3h0fzGF=Sf(@rilbG zMmQSsxn05yQZ2e?`viGiNIZef)xQwm9v$Ae7S}Wd5hH*W`|@@ftEkK4k@LjyF#NF@ zD(IK|>cT`M*0S%Nav?L4%=uX}JRH`H&B^aV(pE+U3%wI!kW_?NJH8b`sXlzZNpmULT2sou-`F0Tb)(6u%0a`+Sug zFrJ$k=bZ1r+W3X@=_RE#gGfodQONhRHF)mHD$8F^OfS<610Kv3#@tTb9Uq z=+AFyew5E*WMWoQ`&DDTI+I;k469`urK~Q0)Kp84@Ba;|HitHa=T|6UJ#3;31RD?I zLEp2bz1>4w zU~-wMa{3n(3r-VM>$PPbY{tClnZBv9)HVNsEEg8?vX`Pry)GBqrj(fHcSb!8`#t&} z7BXIQpCs}-hP`DUy#(s$YE6f2 z(76zg;^>8AvVPvbSXc`sza|R@AnZMtsYyKQV6B|N*?!-Glcvc!;~gW(s=}|GeO}P( z(9;4AQjTV_LD$Yq!tpt;+mHUk0*Y%ckYN_Oi+Ri=MNU6mBlTk<{Q*uLmbF(oc*OY3 zS#2&C*1tb!3t>|IJJv#2+Fp)B-u&a&+$$Pb z=S=VkHF>zeeoRWG*k1bdN7`|nE>E7och%g=_&s~R4Ixz5a&NrqmR{ib_rtx@FX6Mb z^8K+E^Fxnch$+}E?4lPba9vo}xhn#t+855eP%Km9a^yA-1KQ>j|MB*gJF{r668RzA zWhO!-_uc&IOUXBn-aNlD{$T2*V0X%#pM1sc=jS0xmR*!2z6-u=1(WWjw#eEx-Yl0x zG1T6VO%i-R??6F58T_QuI3534T`C%`m_|kIVea!u>yMhIB;KyBB@x z@(;R}svg1KeBN0AGIV?$!qS@8ZT#OzBwFdyG`sMj{*t~A^u>P zS#uRl$k~{N*qZzVfx;GsYY#Z1bVsI&@#g>{sA+i~Wj|h|pUW_I?$D*nY5XX!**D1d zmbMV9-gjr_&a_WXSy2UJ3{xX)dLr zzb`sx2s|JUd+#Tn&Jk`$(}wC6j-JdGXXX88!)z9|Y?NsF0e-fb+GY5;#O6v)+ZL!H zs5^X3oTP zhup~{x1Tv@ko^h#}4z61Sa z6$c@%Na_Y*&M8_wIR$bx=aN~S*8Jj#?*4voWmiz~rZ&=#ErxI{aPS&pGBzKzki-Op zs%T}MnW7t*YIw>f_ePuSf{fq5==3-5zlYLo7~g(&aDSF`1PH`=d@NX;`tf;3jGEwA z(eCWH`Fs^&Ri4E28cM_t5l<#|fMrlzrC2DCYoRld+^4^+y!g(5F;a;KUq1|khYIP( zT#rP*EA}jHs2`Q2JW0%$RyGle^Kj1#sG0w`vaF@iu(R*y8F%J`pwzI$r+Y4j{}%^< z^pA|-6ZcKjubf=Yd+u?%K1MB>Mdawrk9Gvkb!!{z;tZzB+f4k#GRF^Vcm*%s%JB@^ z5v3pYhibyxVViVk2O{itEx9Znoohj!?H>Hr>+(S7F>91-YiT7Z=%JTeTvN+mRTRly zL3W{RD<8QdjhEk zw;l1RV>KPYkuBGyl%KuZ|8Q0c6~{DRDj%;j?{>{ARlQ(o3ms16jNwti+f-ak`Hm+9 zhL2$vCw~Ah*H3rfUX>8f!b~#srq!1YsoJinqkxU@9FW=?bB7qbS#D)X^<=a}cG@9JU> z`EDBYkrGYz=2QiwtB4$wf_A|@5!a@RRzuhm0rYd2yzMK-Go5XdQMw9MzJ8Zf6+pyj zQ1U-}ls&a)lBZX;Gbf&oF6x^PoO$1Cjf7oVq9MM;y9u%0AK&$cO3c`oajgrA&tctV zUF^`b#prU7;-SnU*%wM>T7D-A9B#S|X%LhlD=cz=dfx_2Q~8fhVupitR{z)hnumxZ z3IN?bWX=M_OiS8ZTazsC*0!t&6u6R>9X8e7<-_Dn7faXDAKsOo=u`1|FP&ZJql1qCG1=iBQsd=1F1x0Q~yPWdudFEMp34+4mKc86h zE92g!jLrL-Hd^1r$plAa7`LcAU)15GWVU6LQ^s;UA7fCI!s6M6$OoY;yA{`?<|-lW zaG!x48hy21%L8B12?m1)0@Ct8kgP{0?A;Z&j}NQTr7cw=X{O9MeY#?_vhpAr_$2N3 z&p1G}iHooCZ;XtW=5GY_Z-%%oUQcA|$;VbvVH6yXh_afAgVp_sfX~AaJWz_z1^K8G z{%)HHoE1Hy$cOU#VDj3#U2e+SIhX1r6XYQzO#KlK#JZJ2K%fn?SsYHA-nEoN)*4swU1mml`Y2ib) zBb|-098*)w?O3Yv#kTJEo+nik$kQoN%K@fSoR7^?BSMi!*5&IjW;Y=?w%Lm^bnC@| z45~+{d)aCEp za$pxrjNRjJa5Pkfo~v;}%%Iz~FJ!prMXtxH^vK*K5vDW@68M#oIhvyuRO_G~pLf>5 zdNt4li=ro9mi%*B&23$-LsYaW;-zJ9y}0~wh~`05qGZg$NCVDyh3ET(&ipV8pUN_L zLT2nPuck{=u?_}G9+cB52&Ye*s6rbzHOvp3$_v#bu$nQI~Z-889*H z+6`%=arqgNn}mZnOL2I3wGM`spfy|Kw=n|~-Dc3hjNRteb%MZ5Q+JvDnT8)q_x*sU zUrg(M|C44TKEK?xH}t#mUW$qBR@M5c;kzBWutyTd{o%{IE}Ytg?b|iuF9bJHU2bw> z3`5vw@_3?6!^*-iT$&x)f~`;v?ytTV^{1p&V#^kz?!)~&!Yi9Dx(!9ZPvhvv*c+q~ z=G@NVwJGCGU}TA$WiNNRGw2dRl4G#vS0UeU2JKSGB!qie_$yKg^4N(Y-hrgqJr`p3=&daBUG3j8j3q)S>g+?3bt`6+LG=J>y^QIIVLZ zQB#K#m+sPhie<*Am7MC0UaIYYRdSJlgv$P0PHos5!Q=j5nmpbro*XY|>jR~}WQ~7* zPk4!TJnS~a^2;s#z*!Ghz`_yC+qx|m3hn+2Gh4IrkhA?z|M;`JpTESaZM0`=n+TT; zb=Rdq0{olreTuHSGd*WP@a+_oeW6mOJ9entx&zt>9jm*RmzA4mkrYj^9o(N&EsS;M z%U0>Ob_o@7pO8?SF2hGY7L8G}?Yvyv1o=s$FqU0h9U&|%&43fydAqyde*Pr@P!&#qk~GShCkn9Oo~S?%L^zz>nas!s#j*)8 zf6pEF)J81`u(I?U7H4@E!%v+!Z7zb{dfv|}-a-BOpkBWG_enoGzO|#O^C!c^yEl4a zI6Ixd&0M3e+wm!jl=L0q&V_)RL%j=&6+1zNkLrpL-R&-$8Al2N8oN7|+ zFnI{dC++-uPkex0p3&>Yt@F{zClac=37`c|xSuMNF}^(dt{Sq`;JCO{%NoxPT?w$G z08Bj;f=XGFpEOC##hTy}(olPA3y=5?JL#;7s{Ca z-!g&(y7yz$mIFXF+=ahY!&%Abvx4*yAmJ0|FsYXs_58;B8*`3=5G9=J9U1LF{^rV1 zraj3nFnkEF3g0VF7r{Bq9*yv8|KL8=7D?>6nMBI+_Vga)QDpek3Fdxo%6PZzm~;Pw zWKt*(q?QsP%Im^TFv9Gf89A6X-J;GTNuyUl!f&=JWk?7b3O)xnKp2DCNq=Ys? zAhWRX$>s1)l60?ahHNv4RreUoOkE@v1`L%*6BSe>!l`*84<6sE>Amsg$>Jwv9`{`{ z3pR5SY4Sr2_p6GZsCer5o!;o{;$$?~*j_QuXqQXgo!jr7w%P5IgYUTx;t5&!4CA}Q z#wbo=l15q$yn5z4v{*f@+Bvn{%ZjY8oNUt_LJXJUIi}*vGalhuVLdkurs8%W_eb16 zS72XI3AMv&_oWOUtP>QzAb|+SA8lvYo@$baQ6=)tI{8ivzE~An@IM`kBi`#AB+`TF z=iwUA1%?g}9-^GD*B72v4~l-wy67bVG&4-SWVOv)n+CTeAD$(eVAX6-Df>W!ELsmu6XDx(yayu&6!j1O2icp^j8I!R+ zamsU6>Y?>`C9KGuc+|1EuTbldL>-gnuAZh+e6{?{UgPJ}F-S=bad(QD{t})L$v=!7 z5Vl)+BIeFFJf9%EaR8+<_=!}LCGT54sCM$W-`dy6;Jz!6dJSsj-DeZ^Q1Y!j zRgZ2;Q_d0k^6n#^Uv07@X4&U>s~Pl)G^=zsnsOrC5kk7IVPCV{QzW=HZ(eXY(Abx) zy!)kS|Hki-tv%CZyHB}nKA*#gNH2TUN5RIs2O*E)x7 zS&MZ#rs+w%iJ^UgI#hNqJ?D55Q6L3$5sb%uBhTI#p1U{}ZNyyBjG(%L*j5BTH?yu|lVxp7^cp_hrY0{I;>AJI7%{nWW@SzWF~aMTC>pmu)E$nJYRMcr zC{#`lFd>e#n+U&HYEA)yWN2V$`q4-#6C4-(m!hp6nW48#VG_U%%q{ z_mzBs?ppEre2?^2rs7eUw@zHJwPzvqRiLUHKZb#y^?q=RLmU!!Tn^w=g~97w4f%d1 zLCxNNgg%B9Z864dk_xc(jQRAV@|=ib#(`mVxk#Cji141xJV`l+*)5n1Mp2#%s26x^ zpsga&QuCJ4FGf=#OO^3;zj9T@X0+9^G5dvhSlDyM;P3j=KNcfM=YS`25? zLc(~!UH}i3&tWNM{LNjSrfuFF>JT|x>i% z(GP<)y24V5%Om#Gs)!{TRo}WGIJoN@5U+Ckrc7Cs;7*ntiQ2^Ws&tbt^R0bQ^1;&J zqe-{!)=dm@QZ5LhBNXN{W}s!^d%md3gfFHne(H$kOQQzOGaY<}O!fI8Hla;D8QN2W z7kb5$4eM@%Qjk}V#Hz(9Eruj$)Wdm9gBOsYO;(?+&s*9tb$mN@JtdP#$@lI>`|fBp z%!EmHFUjV+mPId+=j#ICuFct}yF=_%gE|j>(Fl0{P-uq~V-;6O2EtX%?O$91 zebL5_46>aRBTRET;E=CS(%h9dkB;>F3a&@eHH@#LtB31fsF-a1uW+>Y^nBEvV*$M- z1^3rq3A>CO9QTEHhRP$Ka?-TM-wPZW(Fjf1aylzVV1*#gpXsX7&~oW!YB+s7tYw6o z&JY=sImNn%v^|z9%VVD1HHs{iN{%l%Ma()^w-k9i14rR?LGf=TED`=>b+p4oL*Tq?%{Dn5IN2k&LuA4=9{al;lPS>Ay?b72TKFo> zlgQel<={WR1hmdLLNbXRe78vQ%ErqQS0=?t;vGRJBof+8Pc$GaCHa5eg_>4sf4_Ny zrm|EeAFlQz(n;0->i{}5<1_VBS;j^ZHV$^OvZ<%6q5c3fr0bWrbA``^tuM0j z$z>&cz&;kQT$f_*DeT@m8vjxv#4yC{e$blv9TzGhl9~=sBEUN9(PeSv=tCkMr0~B? z<@7VHT}pM>ME&8!@aWMLeVR??MnkmSce!wf>~Jh-1;etWPV9*1!$a=U_NE7$@9Q^^ zYCpfg0BCTUay{*FV^;(;jZb-C=$O4#0Lay3&6mpiguiK8BEwriTPMva@m7jF)GmBU zCNhND2zfb8UZ1kE_}pPMQL!a>PT<{dCLz9XEU{SCxPeFMZJ|GvdFV%St8w`b$|3cB zmm&@12M>_#4pV3`+Bx0PkG%$g z)W%7@5b6I8L(G{@Jb&RsR;k^UwA1`9IxbLrPga(rXZvW1o`8cQSPnhPOL#(Kf0fX# z+ii8myJ;TSzSsV$a|`42Z0|SH4|{#;%LUD7;v>>!8u_q0_k2t2m+totbM~tDJ>F)! zn5%!X65!F0S;Dy(ntP-*t$Mm+4{<`jMI(*h_Tmp|jBihDCRqvG&1+0j%-=@z%D#q;JAF= zl5@uF3E16k^bJ%DvQ~G{F)6#n)p?Uwr~9kXdb3TcY`iS|CC{?YoRER#;AkU|KJ5~Z z>nXpbFR0YT8$It)>pu64Ke9Z0f{}s$S}bp}Y!pdu|F(Fq-y)66o4>Ku1TKYD6xFHy zGVC{hAflGa^RK%RS7eOQ0gFhwXqNZM{|a&W%gqrxd~@?Yl@x4i>QO2$Q?D1v%idhj zDE{Z}nGbdJ^TGV|a?!-`5^=_f9Xq#2z7y;p%4cOz>IkM268ba@-e^~Os8_sW$wUbI zzPLIE`Kd=-2!4Nq@&}(X+4ENUAx8tNY za&*jXp_G}9GXxjWs&?~^4+eZ0mcshv)awWv5uznX7VU7DR&3qs_1L3$^nj=1I>;`T zUFX%Hp~G)hSkES&nSJtU>LCP6<@Bb-hhO^^2ABXTP9%shg>#R@dt(}A&tAG2FkVUK zCTat`kT`MR@V)viE_bcRX-99T4fg%iqG71LBNLiL-{4r}udBJ|W#B`P$>m?FYhb~O z4$7y2DOlux>PrL#i025#&VR0mOTydCHBv#pA8K(!#wq`Gqjeb)w86OYD}#;&=^A~7 zCtlkK5zlbX(w!N1>?oLoC=hb4fY0+0mDMyFCXgSR_xyks4BKZ3@pq^cn%tS37y5>S z@DYp2ex|Z>P~@N?EL1dg00KyB9%Ym*PmyJJLF!~Xd|r-DMZ7$UXS610rMzIA=5_gW zyZb;;X&o8YmoI*^yh5GaUPIJ4vpANG$CS|cxPRt`y#Q&~u`awb9C@qngBm-yRFHd{ z|G$vzS3;Is7VkDK^|s}V~%tb{lac8Ka_qoK%kK9)*Zo>>?G01I_}vtvxsQA(>Mt`;uX z8!)-_k#K88^p zzl;t%>63>)w}{(Fq*`?58Xx{kq)idP4CIhCsa2(3{7%EcdmKdalz9+OeBOX=TRy&K zvp-_J(ib^w>TNm_iszX&d5>31ysNM*5^>+k`CfEfI%xut(s>&EeLK1dl-Oejp;d_g zg84p@riy1-@v{fJdVVPrJ|2rIFDXnnw za;bfvKDz59;otyvthjsWGS>hJvSwc!Bcl+=P_5W5d1T|eq(l*qN22tN{lX|bMi*2* z-CbJqw~0s0E5=|ZB;x^aO5xy@1>H(uv}`$E{2EPjlqbf#-yN~Q2OrO`vKe;9RIWba z)89Y7vHz}umPDiibhrmAOGq5x^Z>c8aaW*|d)cs*RP!w#(MxW@pFZzr_TBPefTqAV zabeiTW(eFX&R%SQG+&{4TuEmo(@~+7y?h0_70^7|9;&dR3t~&8E~e>q&Q8;u3OY;8 zk)<#6ib;ENpNKep^5m9IV5SJodiF#ps7{a?ndZqw8y!TN?|Hmh?AuzRg2^!MVwiex zj`=4~mZTsELfBT4(*v+mJElYb4!qRaFOSkcYf+HWd0LZ1;kq^M% zKb>lCuA!uz!(hWy>)_-XDc*3oPcva$&j|2OXNWBq3Z>}@p*^EBp^0}#UwFL8^jkad z7wg^>61T!=wtUOyJFamo^n>y7rz+ZW*`-u3-OcMxp2lWetHoKBx^fx9;@bkdxz;Qc zmY-$4E+QT23)r7@`n_=oRli1C^4}GGpdiWdfe#hgjyKYYvrDNvrC|vh~;B$$EcMZR|Czgv2YdiMu6ga(>{=z3|FHD-D1AXvR~e^eboB*8F~r z(_tcve&g@;oY79kw0_4G2|&1ybpGCYEDQ7LglVNne8hi#H$C9qmDq%T_*t{oc9n20 zPh#NN&2d-uMS@O5n5q=?0?RS*d5*dl%Qj*T=$^Ny+T$%lR^&$nyxqPIyX#uLrNP%? z6FAk5TDfe*9|IMLyMow$qW&Ip5nb1~-dZa<_~g$Am}wc2{oqXTEzVQ*fvT?aVw!nn z?$w$nK0t8n3eP0OCF4(jn_ZSYd%EBs;Qd4wzQ+NqUl%RBk1O`5|16(n(?5q5XI89y zBi(qCI9_GyGUTlP1D)kWwv|XkYV4m5w7HqW`T>9Zl|?$1|7cJHWml}|V?Bbm0MWz> z&z_a2z$WC?a^m40VNap*Q_$a~_8R3rwgVyOSXp)9QPubRv4aAaQPC<{ckB31x2PrE zJAEveT=Q#eTi}|B9Vqk#U{q~_pl?eB!Mh#>TPc0QH3lnbSh)3@zsetuaPRK3FT>Lm ztZAtD7`p{alFkKsLNe?A>HRBpl4rji>Tq~|y{LWS&9nLu!(H)!+CFmg)6Mj0k}J~~ z(pnCWEa=%|2A|IFkGJ%(TS+czsS;=B3#-#Hmf;G4_|Anw9PT}0)5)PHWPt?%$~HxV zrtie0aTl>xGfDB`?v`qZj06vNFm}J*feHtE?n=W~B@N82pS>Q|PISw+2_a1OP4Fms4|9nkdzO}O zosS5By{(tN)8)C~9fx>pcw1UuWg9(t_rg5guVId$Tu^rid2i%coeTGWfO?D%S-vn) zop$U^vH7Sg8c>Yq{DI_kgzO5i-NtPbbdfwsxCjr zwydI9%wR;TLoCTd$8$QUrv5o)$%o@obz*_tOq@EaaV1sl^U{*4=@+Yh8;zPh3U4*J75hIOMISe4?bL^ZCw0HPNkX zJLrDz0qSn$BFss3q16{b>B7GXBZR;4VhmiNHUJ3GcJ|FiZ|Fpvx3Yf3=K8YUgloqX z=TNqu+1MB6mX0hQ@pPt&mRZ-Y4~ja&T3h#;x!n?H{J)k@Ow?c!B|&}9oAo2b_M83p z;$JL(c=Cy+-kSE_6$Z6t&(3_Fw!ai0R|PVe7Rj*0*k1a6U|zR?(vn*xRsl~uhB>!! zysyUVs&=iX_#E!q)q|{dy@uI9)UID|3L9(E;e* z7WHlYnr<_B>Ev6U(rY`nvhIM}^;VX)%kWYe10E<{&v{n|8x2WG#ds1BvkEGQeOzex zSur!qbCv=4?A%fMf$v_k7J7%(Yl|<^h2cz_+AErg{vwRmg_G;Y-T*xqn;)e)VhOB? z!$lKO)eY@Zwr`nArFJ$W>$v-7o_|MN9+RuuAM1~;Y?ZyC!lb{y41n954#t5d4#1=!YY?CRTGMHI(hd2KVj<66 zi-qp|Z)HE0f0};o1MW;#1z0AP$Ai@qN>|qfO6!PRnC@3Mq zL^r7~Uk!=0Y75GHK;(+yp7Hpu&riA&Z2sI?fosik*A;%&rIo&!&nU}tu>W?K7ZH{R z{2&aJiLeg z&YA*p+$U{VU8YRLnP}BFdE$n!+&w$!b?d@>=L;ySRBEZ@kN6nSPiIlHgTdtPFD`VaI+mzL%UvuA$vdP)X<3dtKWHf&avz84l z@;Qq!{H~BXMQ{B8V78qP#(Gzg-H4yE&w@O)S!;MxdHl*gJ<6mlvFrKT`n%RAuPG2h(owewU;FHBYfD>?Un~u{5!UCY|oQ!);tl^`1XVNO|BHPYyFhiIir=g>jUMz z%-S;`Ze?MJaAutlP=y{5I@NKLUhVOx*$xxrS<8q)dAK933k}mZW18Z?iXej1TTkXW zWwVy3V2g$}d~boGkDJx&i)mKy7NAH$*B{GR-YqB9rJ`js5EH2fl^~l`OG4{tPUiI5 z6YAFIQ*bra7Jiirm}=`*-nSWVPd)qN{CV_nc1om~?70AtxTWeIZk`T zSROfk?>reR`<%=G}tNa zM1Yc4k@*Xk7d;HzPSBbYD9{=sZTE~)J(5VPY8I`0aeKm$bbn+om+PbnGSU%Ygv7FG+vv>ZFozN%GsDDXoL5xUu(n_Y`>-*kombKP+xXDuiHP0*5s+N z9kinpOaurU++O?*&_9ZyuK$3f>A(ZHc7OjdzTVQ@L5Ouc!+K)7hKfJXkeT_Ff3nu{ z>$mv3cfE-fw={7&%V~6Uys*(#AK5Xi$!-~-G?uD z`D`Za6nRrds&^-6n#VlkaLIe>`v#Oc+XM7-EA?M=2~-d2lXh=yJu>f zXm#e_X`NkIX`NkQPZv@OkB9iQ&E)g<}ta?@8Em@0k40lk2cmGXs)w002g zwB}X~wB}ZEe<)-C9I%NS*O=3^RoEG|a>oODPGj13?9~u|)9w3cVPA9H{UGh17w*obl$g4rxoY zHipk>J@cghbh`_By45a737AYSERKKbl0K{UX|_iVC0#TjG#r0TqoI~G>N!lCNYR`m6yQr7kgk+|`arDViEN=T`&m@cnDt15GjnYTiEq9P-)#`17 z?L@@YDdqV?wboW`zWe! z7bK;OjU-;!RsO9%^$U>WSF}g`W(%=Tq@$SwRmT*RH3Di&O}*TkcUozT4SWBC50#I* zip0KE@hxHKeN(S}S9!AeO7pwWFL19g-s1*COA!RS8#y(0Z^W z(z*^U(qa_n4yc2^YT6t_*NPRlqL3|XE+o~F@3Y^=Klis_7~5qg?nKOVlDUAX>|R;KJ)Mk!ctLNr< zN8`WFMOgVS*2tJtx*bMz4KAdhp?@%6-+V%A$qXvh+wCh{B_R?aHoN z8W%s==C$h*QeLxX;C(vMu`AN^;Ro!$&AKkAj$_q*jCD8YHXJuO=tJnk?!>N0P1|+p zlE2nNLU)MuXn7W)@m`sT-{o02{|#rz2}+SXyIJ`_9F*`-s^|OWm1p4=VTfGmH7H4Q zxvC_k77Uzm+*CN1!HHoXZT9(NM+-MOH59L%^^gQje183~o%T}jcq&_gW3)(EVY(Ur z#*wwX9H%6CV)6kfeBq zGc;i2?{Rg@T%K-ONqy&+PqUGIKRm3@_+{DOjUTqqcO;;84{pE4w~!xK~0 zYZS9)pCR)3wkDlj7hRY@F-geJKk^!>kHbB&NMMvxhTZwTz$c<_<=latm+HnXJa?ZM z40-|2D&gr1qttktQ2V6jEk8$?u%Qo*`a7zGrf$>rS%0FZt90G9P?cq;PrJPWH)P(e z%8nx26lc6@84Fkcq%wT2&HW>|e~BCw7izF&ynkR>gt+lmOat})aKGnhOF`Re0i*{q zgt=>?|NfHg+}+$S+Ya+los2EEZfA{Hs&CJWhRnO^kDSrtp%+FqJQ^)x z!QcMCC{2mzUBq-pyUY>9EEC9Y$`&J^jQ64t4+FeqI-6sP9>a6={m1eZx_nm>5JO4H zD`xUNJcP^VL+Y0F&NUssw8O-~E`2%_7Db-j7^@5+V;V18jO`rPo4G$xris%fa1;$) zL$H|KDYGkVXt2nFZ1&lrh#D41#XNCOh=Z@}_k6t;ydB{gt%aWvXe}x0mDqj9p4RX) zZ0Ci#es2YZ$7^^f0s(Slp8ACy(w?lo)o#01#}ve3Ety^1Y69+DU8u<}vRJ6xpm0&f zu~K3Gw%nf{D{S?kk5Nt}k)L~q0aZMp{41kk;m;eba7X1tA^03MGDz+`0!jD7fa6ubNv|U4gv1^IWx-is(YLo%F{@_PV zuxor9-Bfq~hp^sZdG<4ZrT01g`4QCvn}tGXqsc&HkK*Cnx2^$))t4jRm8LuWy$SIb za1-Shz0xHM>|XZgTp9pO#IY zkG_-Ms2rEmi^P>;K>avp6e1HmfkSN(hM=#f>+%d}PCBvRF2wF{^kSxFBKL7{B6J)*KQG<6M8fLz)r7^l(9aqVBY3W ze_lRDvEUhS{l%_@-B@Ad&@?;XxQoj)p!iKBW;$hE;O@LTPF@9G#M2PtoQy_$Gg z_!M{c{HpY_?wS}cT2*a10d#iwg?XA?17Ib6l-bJqAhm5r;YxkFURcvHgeB+v)xW=5 z#?q-(XiQ7B=arGYd^`JnJvP!rxV4<9G!XW;ryQA~Vk;mqynP7?G8sEc;o0|<9A~Cy zhC^sdI>XxE+hrB=d!(V(QoYLy^#^yFkzGl-&t=>31x&@83F3L>r$V&Bza&Ut)u zgT?;Mr8E&%YA(0Dvr|)v&;R+uxdmSA#bh*i1kyes{?7Y|WSl82r_xy@t2+HK2cF$m z#|~+o^|;@kZt*uN;u|BkL_N3DLqdJyP6LlCDV({LtIR=$5#+yfTAS@U{4&O$+sun| z*>?yy-re?SK-2rk;%>#S3Jaf!XV)4ACUpNdNAWCWW@W8qi#{P%<^-VU$6MqW$?m@P z?8O#HoMhKPLw5CeSo5@vElg7n(SJ=Z4wk(-crETKHJFOd8=W9=D}~Wssc?NH(1G8d zD3dEmOT(`>Oy`|xJ*@wS3#D0mnNg%-gr}$m$*dah2dJt0`uqVsIA2lUG)|Al17X&V zrVe&w7k9B)_%b3TTPW?52tb+DuJmirTm@Z*7AgdiDqtXwi|sugVo1fEfo0TOnhCKfkk!cqsJ}c2y+%p zO5fAdJlA*xYjadYsyYU|$#dgiGLRTnaV_4RveMo1_1)!``XH8;cv(0nHFr^Cnby-- z&N~!~CxX58JDXH#LrZ+;Oe%VmR(;oKa`igY9+y`TmGF8JabgYiKi%`}NbV2{#RwLY zG~3SW_!`j9>R=?jY5$d)j`A5hv(^5*;;FXL@8>J!2$jUpT=*qBgym`VcYOvycA z5qo+|${Rp1YIvfbQK;Z`^=gg$L|t;x|mt1d(G-FY8NLA9PT-d42J^&x(*NQXGsp6&MV32bxf~P zU7P2Nt45Gx=^n4*l> z^Rw)$%;X?SgSUu}*l)Q!{aD8k8t3(|LT8PKPFz%IV}Og~7~iVVHW(2R3*L?!#ll5b z7bg#nXA}2F7k77Rn1bGl$L5Mg?83p+gRrSy;{ELr0d5P%t?#W1^OJgvW~Eng)m6tZ z5ek|#NwP!Hd0tb@e_G%aU)CZ2GY(^&Thr5yt+H?#Du$0d0IkHn9m_6bv%fs>ExTZR zJWXRJ`F&QNebtJlxa+)rYm#+&{7+dMTS%e>|*ufJP3Hsg=R76%)^c4DKkSi{JtXF-ixgX z)9>?fJwDr}Veux(X$*L32WY0qA@p5`|GZq;eP-%P#LV@!)`Vkx`)S8BGXi=_FGdoz zZqQy$acqm@!6DWRT4SdR9k^w7n#gBJYl7R%w$3SL7L~>=f-w?2|AiDxH9h&SMsMY{ z=St#qCmz?|04ar(u^w!8$%d50ilmhET;EBqszJxkFHgmbbUNdAng;WQYB7HYad@3h z4=(Ow1~Ro(AO9=6XRaB~E<#bZE~Gn{`XIm`DXBV$7m#RJc1ecpY`#mIUEs>nOUhja z%#M@lIH15r7l+=F5sL>W5C}+ZC2Ix5b-(tC)W}>T~ zS$w3SJ@}q_mz^|{uPr09i|aqMZ25`>v5ANPdDQXzz(iHAG)+|P}Ft}DyrzYDnYmZ@WV}V zU3dJc|M;x$(0$;!zV67unE#JY9t%-i?g_T&hinQha)GQw%2_?T7X)SIchLD#;qgr( zRvJFhI=1;ihkbjtLKqJkC&Qy~4G^;sRYvM|<>w%SL45y~DQiUzuBEnz?ZssC@|!?E z`q7?NW#p0hD(t(q3;uNhjH3e|t{xBN!h|_~ax>O<%ci}GatcZ0d8k6Nar3K+4)LFY zlsRTRy9HH(l?<&;*25kG_0)^?GAq8&jE=K~d8|yE^LR|2Y7N%Ck0*&SaWIE4fM#kI z+d{vetN7w}jO%~iiK%E;fth6yg_&7>^Bholug1Ex;pvmpACkl7jW7+B8uw%c9G-z<^m63IJs8Vq2p4bbzT&U&>x)7KjRSKtfp1 z!I;~CA+P@;>;Ul4LC^%r^hN2W`{ED2YZGxt%`_Lc7TN?WyG5;2!<2Iu=9PIhwa2=l zJO#f~)8Otl+P>|>%qqmLe7rfP9}MQn`1t4LfSA@nHQl2jV5noF_$duD6Mb9JfW_L< zMQxzmj{EOKN#|2(5BvzlGtnEG$XpMn`+msmoX@eb!=Jfij)Tc!5j1w)v(BRr!8;KAlUv=i1CP+>m` zx=ru>&X}9^8DutorWUeEd*c-l*G?ZWJA{rL-M)vuh;C+4jEmgR^7LT8!b-=z_BODc8Qa(T$Sow=H^M)nAh1zSzLN&TB4W zOV?j3RWPShWU;9HfDOLYGs^cUs-lJ}u84OhM_pQw?7p z$o8Roxd!fDLc3u28PQTR$kxUeQl}9rrRx=g#8Ev zu6;X%2BQ|vZXGN7k7R(#YSKH#@iA%6=5VY;v?swv6gkjmoJ+tN=a#=Oct!4XfaA;R zT4WJ{an*_earfjz0;yp}b=ZXyUTjUqI@xqHwQ>daLy6!?ah$Ze*JXf z_H9NUg`Vu9R=J@3FVp?VJe_hUev!zEe4(zInd1C8B+fgIUGV=vjEDS&oIS=on`XXK z)c*bLgb;k%3^pK63VOnxX`t+4zWok|*v--rqJYuqdXrF{rqXSH=c~9~pG529rwsU> zjhVcp13B5ib&=jee77+aD#||`Ikuw1Wt;!ReI^I#R#3PTP^}{^M$_c_$xpdv4FNh& zW%I1X;=dETZj=Fj!HcpstMU*ez?e6jA zW-ujN{O!3;aDzrgi)b#sTEVP)>1t_jkjBD>l|9_91CBu}+C5z_PhAF^VBePcf3}Gc zBdV3CE3RnMI$!rtZK1JI0k+4Ln+^hKP&Mwoy1l#ir$B3prWt1oFGfW- z{5$zK-5Z=I>{G23HX0D;>#7={wz41d*sMkj-$Qb{dS&|PPNu{Vg2bAX?P&20;b~3sD-KN_~U&8iR z4;tH-rPjoEfI63lQ4dNx^oM&KLS!_##lw8Tu|HfB4TzO-eBWzdF?Z};WHEJ%948AO zUodUU!&`2;GtAVNXBp6@{H1jLav_mKvZe33ISQZFlGu18BmT(s;$FC)RR#5Y`@7DE z@W1*C#=kRx+3LA0r%6YrI;`be=e?@BR!KAHk#%l7Zm0E!(SHoZ3`Q+SK zcCXn$+S_7lqr;ZCrDqgtbuUVybc10O`$eD__pa_4;t_J;+8G+IQloUCt`TV zV@ESIBD_e|Sr*OFCN98Qzw+69uCPhi0=NW6H+{Cs{GE*ruD;=^76H?+pZ$vebypUW z8Pk;0AIRL#oGde#!^Cjq#@ZWto8-n^bS%)(^jKxp?_nzmvNMC^i;TJPaH98{)sLP# zeb+{9YOvdBs_()fk2KXdbb90Sj@jO#mTK~R+<}VhlP*1{YEkwkI@!q|YC&4>A995a zk9T%WyqoRC0|HmmlS0KPK8h`yGpS`3s}n|W)EzgIqX`EUmsjUB85_K0liWtY=v^Ea&G3!6+M9%Cb zenyD5ki>#ZQ6JW8mCSIChu>+1o6IJ9Ny&ogW>+hgGi`1Kj>;?liA7NH5m1!+ON))f6oBkDo`#*&SRv?yM)n#@&7c z$`vIuB*bw!IObW-jNXy}L?E-JqL z&-K9brSX^h+o?##vncbW;lXRsDy=c5N^S&|dEQt9Jcnbpj0I1i1joIQ3+QN!!IhLg z9>hl8olo#b*hJc7Nw3VToA?W1gXPXDekN?*np4sik$3+Ot1m3W8Pl5D8ZYsw#wjXa zVa~(&&g$v@5W^@54zm!+Fx0c_k~EV8tZKS8KMp20!ZyV#qqDrCgbg?i{2ShJcr!w? zB5HxD`%hoz4{6fShoaqvoyJV-ju-0}*QbVfTLytZDkZH*t}$K672VIl5OOdqoEgn& z7FsYtxO^o=*JD+2klUwriGG0@q7<(V)5tP!ii`mUvgOtD0pMxPO8vP79u=So-~Ww* zdanM-#O&Oeb3*q-OSoalx`rObHBM0B69Xjfl8Ni^?iWchWksm*#CHws=)GpXVk3^k znx0YlVZznz;Hmdx*^>6fG%dmqZFJg5yKd*qNU%s~hpCB|s2Ybz2v-P#}$pT9S7WmXch|?fIlR z+S_rs=^xIPQ-z@oH zCiF|`KmPmb(J#ShXQ`ytSTppkSnGfg?kwYGU*sKrs+S@)C){ogVs7YW5t>nesqD}& znH3G-P~kCR_mU=sk~~?CX1r<2t#($47TZwZBJZ@GMVLPx)pk@YX1aIpu3 z3*T_9J<{0?M)74TQ65b*IT9Uu8&9HRf=1ck>Td<$CL!umGQwg2VRE&1%#%4*Ra#+) zySpHr<`Q#Ks~04b%5@)Xd&RWl1z|u=C}p5Dsp$vD{XJw#m@c=4{nIjaMWV><70)@% zoLX(O=5@~_NBL{FTV-&C%Y7tR{fs>*nHPVkT{NS4_uw31v{~(E z9gL!Wc%gp})8)rY*8%-)3ox2 zKo47z>0d6RI&kyVixSjjIuet1cJFz3^nL=37tmCWk;7BfUMkZ3YNw@Y?V#dG1RP7o?!eEZ(PHbs` zgLb52Yi_$9>95VHf`j94Ly+oQ)7gz4pS>B`tK)(>QM_A2prO1u>LgBD&(1f1rTX)Od*MiyWW7TW&si}gjsJ+{?Su$Z9rJc)h~rdu&q7&i z>asM~$W1XXSmwrgFD+6l$rF^KMJ6E2q&=!NvwWkVu-UOFqw#whMFQML%lZZ2f3yx` zux7wDzRNg3*pJ}Dbj>zZa68g$)}v)Kh#)n{9RUrZak z!xyK7*oX!LK8YEbI>$p9FoC1>+hC1QRHy!`#-LaBcASex zlwlWGPyqIVbP7~!R+|0R7+9s-MggGJK|&$t#Tg`wFhOwjmuo~rjqc_q@YFBq}gSNiD0zp`iB z*{7s1kR?xHbAUwe_8$1Y;Fj);@A1Arl43>kY}@Kq_0;ecXxCQar7J11t2nC<>us9I zXs~jP?Rr9MT4&8*3s!yZWYyFnEe#7MV}Ebg`gfY}XL^lD!?AZ9n%`Hnn+DI%);piH z-flb3aK8mG6H$TopbkO3>b!i2P46vT5QGYYW_*rkbP}}^DLePEz1hHZ&|GPA;c}6a zm3iJn^hUu+Tjrl>O4|C3WH2r2u1zWXu9QD2O-g7?TjANt@)cG<6zY_O=_AS#)mF~DrG+vERCo;gf z6i_z(E#9pA?#02JDtqlP_T$g$)T9(WgqRzLzp02wf*_iwQ`0CMt7vKHLva>&n!GLU zIC2GCOgtK8L3Xye7E(Y2nU14*^o^Ac-xR7}hU=H$i#S!n_#bRYv z9HflDltv{c)ae9e`m>JzhGhW$bN(8$k+i(Mb6oGEQNugh{`5FP{@g(eR248hKgu^O zE0~+#);uH>I99-u%{VA+*6k6(?Z$L1=N^Y+WFO@uo!qzc5+MKdTs zuUxTK_U*L^d}1+SH!t3A@A9g5vwu)?uesa#tu^GVsW*wj(wr;rBI!BrVtU-|1Qmg=-jxXZ6`ZJk^*6p*ulVpb*E6Wh!o8T1Fpso(X7J7+m)jvf&#==l2MJjVjRC2uC1VKj?^I!n`W2P9qaFWBUa!a3R3^*r=&8+2tizTnUR6?R1i> z^(pRQVQv|@btU!rA-M0vc3I4`4(562F2@MUqh|0?VDD$YT7T53yskc=RmK3PgZoW@ ztR_uq(Eo!kps+z(#-ka^N)tj?Qgu^BmEDH6e|#M=yd}Piim#ZL9*VbR>O1THXs-g) z8vmD~-mM73XkL#lkn^T-!pJ}PM3x9woQ#+*3bi8Mdg;D^rAh9><_-z_FciyONT^L% z5WCm%iu~ewJO}YmRZ;lByvo@bdR0dBK;HUml>Mjaw{|sWY?O;&&d*{X8 zdotwP8Fv}r0TZvbDbXRdD;aA%u(MM>lUu{mgFsWHEPZJJ-2 z+-p7sRZBen&MZ#M!pc9XEbK@Sa%Sl}#B%X{*8R<*Tn}QgdysP>N+FUm-lE$ZdSS0! zrFT^26Mfl3@cr<>wVhH!YxcFIU2LQQSM{YY=VlV)1gT#sh6!5s)Xxaz7`4DKojp`L zp8({^Gz~A5f%7$2>fC!xOjk09fOmdzv2~WU9#dfvy|%a*j5>8pD^#OM)Rk1ab3m6} z4|pqyg@U{j3MqO<3@y}}QPdpTlD6;2?2o(X*eE~_*VIXp_sVHb7j#G5|3KugY=6as{2el{DWJxD$POac*bNbK9a5^^4+@DPfs*>f>d{ zWI-RUvjWvEX?x{$l8T^NPR1HJ|9^WF_ZZ2X!or32uB!7)s z*TgIqwEig0JiFg4T4fOdTm_&#`|4?;FZPSliEjh$O{DsuubdJNJQG~3StOUZW3vws zty=(b9~Y7QJ|`i5=B6EbE&8bSJGH zZ>IUl2jdf|xoGAPLDA6^4VdtyEQK`Q`3-G0>N`9UGSAWsO^j>G(n}RM8geS7C`pA2 z>XrMUHhx4b+=-qbH?TSTvW*qf&l*%G>o*a?tmIj*#0yde@bp>-+KaU-%{PaOzdv#3ZqyQ^5J8Kr}Rnr5W^W;k*NUlIZbYbn@y@={SlGbSs z&!w=UW8~qbS(YT?dkw*6{vpT>S9cV@LI`U{k!Nmv_-RLFh|CuuJ+$~ap9}h>9k*gi zd%PBgqQ+|~EFUDw1I>}*HsE7{R}%mGQqCqq5QfsSASjY%v~oYlV=T4o&f9ve6T_$q zBqU0(BS=Kc`jpeC9i+PZI=o@uz2vM<3)#61!J2bN7HwdRvFwA^y#>yN_=@q@R&ABc zId?2JgwkR$?=d!e$cB@LwoI4)A^0J9@I)M#$rJuw zLxAHlBZ8^||0>*Q2hzCL3uqFB_=QA88m03NoVtp$XDRiuDq%(4J^Uxz?>UcHbDuqf zl6a7^BNw7Bl!+qUs)LfAit+27^%d(#omfwDrd9hi>`FVOWVDYd$Pp}^InZsO^ujaB zLdPvFTG7JKl#aT%&nKnPKX4rbkwSFlWwJPRSew`{da*z&KbQek(NJXSV1r<6dhjZujMwIGLCSu%MeaTwk0lzOMBGTxACb?di|^MIr2%IO!x6 z^$ofdxGW8o3XG+_Pe{2);!QDOeOWdfo8iT;z?1;AlxV4`*hqpRhnrG^u)%LmU*L$l z*hn3?6i+-d%fu-Y#;$TdPk5B238??;j@$OiuPBTqV#%-skK*68)H-zyyjH&$A3@&J zEI(4&JBP<_lB&KQ+gbF*Q?ZPcuf{_28PGJ;9S-DS;zgQ4#J%gWh!}dlb83$@>jD|d zU>R-=$RrIc#ocA3CG4?L-P|i3Ne@yjYt!&tcrN|iL@&h8z70#OAqfAM5fG)z&*n zQXpnamm6G7m(j5INxcQ|kDx@UPS$gH_QJHH4a+F$ecyq;z z{#Xg`^%93H{DKKXi=arWB&upTT*JP&_1n~fSdf@6gIXLnW%C2&S`3b25xc(DWE8C< zj=Fe{Pb8?isDHzU(@SCO09sc%w84^J1hP)aNy=GTzX`ECTg}x~MUf#^2&ToL6easZ zthG6Ed`w4OhIk~Pa%|kb6rsc-H0IDr-Lpuepn{q;xaugO3MDl^6PMQ_&dasNjzUdH zY$;)Syq0U8PIf~od?ICwF!hu1bu-3LsJE9Xb7SLFnOJ_}_LzA1nN~E)LzR$*5%wdgq$O;7p=o$2dh*iszdgQnkH^~*Z(WET@ z+JO;DR9eOZQJ3bLb~v&XZxv(XGlMVvx$Uv()o< zj9QJX?Wu$5K`TW=d?8`6b9b%58$7-xZ~6Vuo{>HtQ331s1zw4|g%4uf%NM;~^ZETb zj2GA|f;lO5?vSCE|Fxvy*q5F|p z2F6hzO`S1ruDzzrIC=3yQ8)Shv?mf{c`hlE>un8%GNI6oGOJ?pvq^|^q=hZppCs9W zxj*jLiBA-5Tc!3Nov3mCt*h7C*TSIr9a1apX&Jw)OO)#YyF22Dl1ypK#ty z(AROKp*tT?A0u9#F5Lo9=j#Z*7=MPPee5M8BHxUqjpP%chIc8%^|G?K;dZt7BN7(p z{n#wmZb77t&Ogl}Ijr2$#oy#GG~yXMgm&mm%G=s_P;H#ekI|VT>_0LY>Cqs$Wm1PU z0#NhRP{!XC^yhltOjB*BP4~f5uHut$?;FP{g}NT2N3CyOZ`nUx5T2XP-C?GqyED^w z_WAzA1@^|%0@ch`{yuoLic`N+n-z^GQJoIqW8d$AL?4lQ^c- zYV9Qr9rR3zk*quE;IGMFqn-FXtOfOD;TXrJ;apcM#r2?(D4VVo5u;Vw7q-Tb>27%B zXoeQe2r>MX5t5FxYCL!Bt3KPEMZ|SM@`=j15C&9aaj5+M?8>=m2zH&zH=A;BFw8NT zEr%2%I6HD^Ff=E7^!$4{N{8awnOkR{iVYm=?Wkqt0zptj!!=#7iPEdVNO60o4OR(v z2IOlO*>6zTqWD0UPN?oj2c2cFb3^<60!|+gXwOZ&(cWQv_{q^k*9c9ST++ocg;r~X zH1lBiRwd@8fTbuKglph0?cDN-ag8Kq%LsW627W&y1;|_7iG4?3d7x=#x8yI(Zm(9E zUP|*Dp1CGm{&)u9@CeGShqwK*>Sry}R1IpOhIe$x?QbZbF2s{gB!D5 zbQY0@{jr{}o)QI1!FtyP_*;F8_hXBchlQ@>$h;t;UCEYPqJo{Id#Wj(@5v zK)evQzW`SZFoXIYduHq5eGZHdYJ@3DSkC66 zk5>dRR~7|eBfj>^VUlLzxw2_B6j{3wh_T z!Z>_F>%fZzsmnMo2iBo;pZkTn9W`ev{wy68R1jfCK3WnpbQyZnEZF;(Z^6fc8rJ(H z`8Qb$ifCM<6&@>Ov=B&0fHLdUa3snGO5Z|9)o?`@9m1Xu7V-6YcvIeTu+OJ63Vwg4 z@{iMZ3A-CX83qGBd+O<)2Y>jJ`?Eop0ve7KSE2tK|+R zL`kU1`W7hhk$kA5H^sC*??XyTB~_OXe{2lGTJAG@_|pBv!y?Q9aS98S5R+@C70_Dx ztxgMMkumGHE`GY;=)1o)IFbGtn2>X+kAZhV#|W?bBv-F&IA2jY_8ozo_q484rSg{M z`KL$xqH0{cK*^a-jrP6ZyvZkzPFfaMbNM}~9x*=0gx$QE_0m|##`egygf-%g@zojV zi{BMP>RvaHJi%>>G{Ze z!05--H8nZX#PEpRx^-7`-rb3d-nd}_d7}uM$}D=hVx|{^|drRa34mo zmvu!~$$Ep>vvdQ;<^*xF=jURq!l0$rx3ymop8}lfr(N}zuC-|7glg!AY>L$vZ>KOt zGaC!EUUO{W$@^W7$L|1CPF6g4TLKm|Hm!x&apIqbyzC4|iom-oPmM_;d2uJ|jSEVA zUWL0Q+wj9muRMEdR=O66y#@UATt3F!%EbsmugYPh>H2helV(+d#D(K0Hm_}z`e%Zk z&RG#8O}Q*?j7|<={{-A<{VA%HOo*=u&iQcHx1iE2d_VH$D7KL+FOg@kBzh%Gnxu%H zdHm7%F!J2e#D#&O=`1t7vOFhDoEXC2@>K&=@1GuddPVbQ|Miu**a58D@JJR({NcB) z+N5IWs^uy2Q7Sv-htY8~6-?Rzr`PW5)P8tmG5!?92*1iVqy|X7Y}SZWLF27ij2e8! zKGAhvo+(mCU0lP7)o~|0+Q-DBHlL+DuVTnQ@id@c%Wdk?W|7sq>qB!9nFBQ(_Jb88 zwGW0_v_eO-yIjLpxr;suCTH7-EE=WifV)^L^lF0RyV?qI)gwYIxqk)qOS3*h66_$KeZ(oO46+Sc>BzZ zK=y55m(?B$fpPy8Rze?gF_7{gWlghDXiSD?gIH-rFqU4Fe@v`eJfg9E^xyWt`@}nz zA_?2)HuU(-tvV`eK}c#V8~lBL1kL7_px^xs#aYcui;R^$&p5`43(PL3wqoBSZ!T4j zb~{A;HXf4);&wtF=mSb&g85z%#n2u!0Varmdb^{)pstMes<>zJW8$2-uioeLcw4&t zKHSt73m{6u_%3iXm47)m$LIvwelK|Pm^{myYQ0yw^a5h*U8k}2^7k4#LSPdicOj3C z`8T7nBK~)u^IEfS`6jZ3r24a}-uc1rwTBdn!nzy3g%)=h3lVx0mguSQjS+u#>)|qU z5+>^ZQFfhSO(xNL6+6wcNC!JAARyA)+6y8|@5RtVM|vQN0!mepULuQ%v>?*E5J?mQ z5d;E+jwVXzp$C#X39ff{?=I}kA0L14ec#NSIp@qd@4PR)W~8Q`PyfdB@3U`Q<{{ z7f}wb&*z>L`X6KY;QqQ8tQce4e($njquX5^dig6x;0rzUZpJ*nH8Eg_&hi?pDX!~9 z^;XIxJb6S#ZG=4*Sug!%vGAs5ok2N{6>uhP+W43>ehkj?=v8_-zB@D5@j%*2h8o*( z{c0Jz{;=_(Y$@djwMX|8_v7U3liasK@-lyc9v3Z`1spfZiRFQDFAg%CP+nMNxcgk`-6R@=Lo~n(^GO=X!LxL%X)Y zV(0Ygi%2hrx*g@C7D-_2(*9PVhYmUb>-T3w=P4>DeLSa#bc=T4tAo^!Rn>K@sV^K_ zq}+Yuh`q8vI7I#)R^X5dN(@OX8)hP)w6qGp=e+o7mHRVMXHnn6ziLM|45xYG^1&@A z@|W>>qtTAr$FbsRE^TK(Hd0Z{4~_HlgY}v{AVm7``Yf8!#cij#Zy3aKpxe@7ABx_I zQs`Qp*>k0G>tw#>vFNYgF@2>8>HAL^IGj?#1tfQXtpBB%_$-bT)9ht*-2KwR&PTZT z^Lbk#F9n!(JN{g#!q#Ru>_~4*qVCIW5MT7Rq2hekIexfMTYuGezw&MEwfx4}%!YF3-Es2sYc}_u7ou8gN)+4(l@ZnX+ zp@X9eJUN}-_pLt%az2;TbKy*K*;cJ6u^jL$6Z8sRZq@6O>pjOuSMINPD4Gom9*GJK z^exfWk=*uQ(@RDf)cng#O+j`)D@7_${U~zuBq}V`EKn|?jJV19ipNX zN$NP_F!P|n<{Q=g67J?T(E+ZYjveT>{{0W@5r@lSca>HQ9VC}Ej5RIM?yz7+q2B^< z=DWmuZ(|5|e+=h7Bc$AOx6?huqwfFtn8VoXi^P6o)$sm`i0!SUEQs;Z{G#j*i`=CD zc(A<`Nq;XeGxFu??kU)Fzr01aFIn9PxKo8-^^7@vkJ(TnBq2xRgtE3+AoYdrIN`Xu zVDvRGr07RyG|KNwUGZ14?yrwMQWJVlAJjWzB~YQ|$zzWz=p1zE&Zz?n(~uSY-=ArL z%H`LvQv6%y7*rTb@_Ff@luZjsW99z+?7bjciboHmHyW<`iRAy5|KGoUMNx2w)?`26 z++B)jcW8({?9!Y92ppbIExs-LD)nVl%4E-eIQ59@Vclj|F)QVhh6%>g0Z(7$03~j; zCn&8}RrCfEnvrFa-R;I;9eyeOEp{aF5#7%ox!@Ggxzh=gM^I zmN%CAwT=$~yWdokQzX~5w2M=Q%EKand48{(#G!xi%e(Q~=>dQHDTS)o!gdM? zZ!QrYH$vGidil6kHgR{v*_XBatojF{9X@cFzPfHA(5!Fb^fgDlMT4e#64N!9a!*Zv zrt`&gcm}C>seWdO^3>&K-d8YN%nH)wjoVn>_z3lsE#fOLtCP6k>Q;PJ`&IIW5uUo8 zICEcnqd(5_Gdge;UVS znxI`Z*>EJ4yVU%L5*2Jd-o}MMf(J@{c2ttRzA3*ESI)3QXMX22#3WBmv57@&&qr<5 zA6nXl3mE~fmB4c8BAi&Ui}N%I{O@RVq1J$gKnuNA(ZhZA?UP3%+?Oq-kiVzUKJKNo zz=_GwQ%moQmdsx>a>-J`3g=s5q^<)#$+8{v-X>6V4?B0ZGQZS(k}e$%pH!her(Tm^ z>A%MMOw|V$@KV+p{sm{mDB?|5ANa)u`xS@Lr7)@onj1E$2KGf)BIqpnRS81B#a7PH z)V+JiowufO6(`B0<5+EF+gu2_jXo2eL)h&zbn>)*EMHWEImh7rH~%}ST540d$Eu1a zN5j>my!)rJe5WROTRV))wDvY1HJ8%L`Sff=LA~~y^ofdPMoBwiU|`wsPQ&q*^MZWv zfy(~3+?&Z~4cuA{W15@sH=@yOkWh;+voXiM)a<(2n4dDhuF9`ShX>TCeXGw)Tv>8Q z)r_2!i(rWo9+Ev*%@MVIRL*v-F3pU&i1oTu3-kUT31@PRk(2?X&ZDpa)koX$BjyF- zE8ehyHh^(3R3q`Jz${11AFG8s8_%waO4V0@V^cIT+i=QyzE&}QZ&bJ2*~MVv)8h$) z0B+$((apjbe*_K50&wlYJAaiD{%zM4lk}z5Pg`3HGp8=0z%kajML}j&SWc)6x(#ts zzkOQGQ*DAl9nrSY+UO;DZLgrm@}IpaBUN@~ z3_5zz>&#Uwdf}JAUbB1KNA+@s3COP&xhb(C0;mp$K>$=&*$O}kFME@Y9MY0#OREm% zTc!nli>xXWA~Bf`d}FBJAZd%&R@ye<3UluqK6=LD}Bh|&u_RnU)rC5%d8&cQ}TNvGz-9=!Au;= zx&yeRx$<71xR$!V6f|~*YV-AT|KWXDF1z3d5agVz^|)>EE&$Jv0IBiEC)B7jF(P6_ zZS#*Zli~Rq_!3cb2D}(tl-+2RCW53_(;LcK!xtyVxv~~z&j%ehHbgWnmy1_b^nr@h z-3V@)5>&;@7q8AX%$XjopEYDqy#)dgHml%B6*a~A%OnR!r8^tvcML(C3lRQZ5oflZ z*M-~&`F7D2mqI7fKEmPo0L5d-T4$hK<={Q|`kUp){WJoNjXs{eI3s8rZB`&ggTty# zUOs@#?TxVHXjlOE)fAK#+8+^d(Sgp#BF%dj7t*X@*Gyiz#&)yfuu>{fp0(wGatI)~ zL4$#|4q>%u-=~l);$whve(35R>bA157}gH1rJ58pmu_n|o@P4*U?K}A^4_Y(&v_Vw zrufC%VkT(;!yaa5PF7ufs@61Ba#gOc2eM3}ryjjDqfa z83rbQ1yTQ!%d7Wlc7wt+}dD18$i39^J>U_G+e7dwyK; zV1X&OiCe7Rns1$Ez%p`7mbmbYMrkyOICChcozYdSxE2xLkgq~}m-}tue*W%<<>#cH zK?7<|e)SQ(m7r8ryN2hfWFCgs<76h*`1fww4QdqB##ioo{h~#?Ws7HrrF;PZW%00# z_W$QmHXq)c=qH>|D&d91o%;GAVL9r{6@ z{U0E87S*=%cS5x!7bD-uIb|6+TeeIogKIiqa(kE!m~|;*4JJpTmk|^uP;ue^4^#|| zPIQRsb-6J7en;q-Y&}eNRkTHw3A?oT|9pj4pT4=QtlRCPKsfZPaDBdmP;hxYWxXvKh(#2hYCa@84;*Bv^sl3j5My9djarna`!)D4&VSPe~8WLdbL z-hS9m`o+oW{||aKVR0GZ=YD(mi=C&POplmZQAFYH*pnx*?ASeNa%&$T{9iuc*1lIs zH!eq9PdSh$B9zqN`|uhU7jYX7#2WC+5@u}X4QPga25WXc+V>)ubh#OkX&afPCyw2z zVhp%^fHSCZ+Zc%qdEb@w>O4PP7vjyf_{?LEE~^A~XWAOmqLz7?+y>a!kkU<@eUr5g zxgGRK$$c1cqT_{*bF3blJc3970t0^S-2VYk{SJ_*Yv40hOVO!R%2k`hG}OQ_tH&=M z_eh~@4-x*ShpfWqCzQ>ivTyJ&DoN9uVMNbM1FeQBw|eEaY<(ZF$@|xPXXlS@33{k_ zZOhiP7xq{^avZ*}`|;78JdZD4(CK{eV9V3T0p>0PW31f^U#nK{D5pYxIGkasj~8z&K^6ph$732+#y{Pz7su4H)J#{)k+s zY+{Ta4rDN6LS5n?SNx4WT)^ydTLO1w^U(Gsgz#LoN2RB_^>$eS4*~zQHSl6vO_>lD zRSf9jnc4;OUOTh?#>3w*`1oF}ziBaqqsUVqD__pMWsP%<`n?mw z7bPX^&D=@f`OCI#I2Vgp!=m?W@*nAO*UBbaWKv$fo_i7kX#$5n#&4MexryTH#2Zd! zuy`V?P0x+~O-sPdN6^p=mPLRKp=c!afZ@A_He(*i^4hncln_YRpC3e@+2}#OLa;YY zu$AiXF|X&AI9p`>c;Ex}@^SL7U@2>p;WE@C!HJJ6C^KiU?)DIlzCiv>T_6;{=k}-A zZX7>#)RvUkA?odp{=L(N9kUzrG&-NTt7cV97QPPk0A&OQp{a>~<%9OWsR|5FDV`hj zg$#@#R;vnVm{VtxWk)q)$9r7IT-lb&0GcL0GNC&O7}7? zz@gF#C`Kbsg)fRLidXpZ+Cme``kOW~`gZcPnl9FIQWK#Q6DaTMHr|yLTRFrtWH3F= zq|ts-6Ai82aQ8(}L&O_l+D8XZ9_iCF><@u>JVE}-{QK_)CZrAL^>*J(?GmR0#v^C; zH`hnSYBUBaI#_mcRWRZDye-MyTtKFgnYQ*^GG*y~g-_XFqas zLnEv9Y$xiOdHNy~dr(~XQH<{ON)z`-C|H&F zS}hv2xVQ7QOP26Jgx37XoN{SIxdafwHKO6_ekzM>r88A z%O$~gl7Dh72jrl9MZs-fVRK)MmWgc?K%|?KYTug#euPn`S3lQ9G|AmCwj#FlpcA8w zZlDq@$9$9nRHx=`v*|qjsmqvK%q3C0TsuOyu#qrH$sU1i6zB97#{car`fRKb>sfI2l0U?;Tot|CW1EFRx`?J|` zg&Qw8O4v3*0=9Cm{So(6Zw_b8NV!hgzTx(KG0F4J#!d`hm8eCK9b=u}NkvonH$$>) z*Qm0uL}aKfJWiCYe473J?gmlb?pVrA30l8&v47{%gQot4SKG0n{Y&Rge%m?yXrbi{ zDW&N3{^Q~)rICfN+TQrM^{dZgG@EOw557VU_0={R-#5nx>rR>VKorCF^*&B(}6x zh(|PnHyYeN2EQx0!e+CM+31lCqAef0{v*@HzRc!LoAe)9`_hwzWrh7n;aLY>o0V-Q z&t~jz7c!&2hmfJ1ZwCrPGqQ;?xeN}i0`{)BuAKU(lCnI9h2+*;?L>2y3j=`Ts)0Po9ws^gz#H>rtDfWAYrdBfKb5K(FBprF=cZR}lQe~un*>0@ zOiVzLlJ-R>n6sAEc(wh`w<1SlgNFj1B!dvD{gE;0c|pj#5oYRns7*}R4|kJ-{h&}> z5eUF??px6F=KzpDw5vQc5k~L-LP8KKwH-up9I~w$g>|rjkVgzq-K%J`PfDfGMIWB~` z_QnfKNt48Hbi_-Ql`xq9|0Ttg3gJZDR`tA69+b>7&+owD*OURKDf*1C?}F9kZ#p-+ z^q2QD(eTg7;;raMpZ6!y6-W|U`>c3}d`-63&H<(( z?}(GSH$$B2WGxA^SV^X3*^B6}ESBp;8cHCg*EyZMwnQCE)D$v7=DZ(u8w^}X0$XRO z$0@Ar92qq!FwnkzDZ}LlX`v&FOZ?h4seG4lXQC|NTy{(Dc6`!02_;Tln>*2kW6@*u zL6^S3n*mASAk}VG#_r*)xct!b10hXP*oW|$F*B0FFpP&TE}65M68tAy+1B(s=WMX5 zugaRG1k{~%sm3TThw^3_N&kvZS?ASFDht=o2*VLbcV@`l^F$vud)LyYJz*)!PkIYJ z_22&#wKacOg%qLo@~85`RIsM9#W9V^YaJ=RWwdpceXM1m%1YJmIg{=fEW;B=5n;;f z#OmiYJ+aW6|89nsk;!GB*yQJ?+kdJq*bd62UGGRWEZe`%Nye3$k4cWM(&vm=t3^Cn zHl-Glw2L!2@frU%z#DJr+{eTuS2dR*qc4uF&FkLLyNsAvZmX3IhF9^gN6~_Sf#n=G z;`vG5B*pZ#%E7~S@?&o{Dfn_zt`2)+*~GtoEjl9EWm%9O+J2RpL4(4FoT}TW0GqM0 ztyzj^9+8>K1pi?5dWdpU_2xnIC~09>>CcHJF3t?qN^9HdYRP;TEw--^B8)UsVy_w zuP9?mJ^jxIx;Q6bVZ;|1aD<}5j|nQO*TC04&ekpe*ST@_;1-vXGn4Ct4GNyu5;@A; zzko zmx3%6wB&%{o)B5K3tFqr)Il78%0;vqW`Id3kt1$CzI`>_d^URS1zsEsDSaD#ra@;l zS(jCm2e2K-Egm2;=~DGF0dKsr3<364n)B5SKqz)7bt0ASo8gOs(lJ%@y5`H!$8#qr zujtEF3bTV5&G9upPZdTY3bEa`2va#gAJNLWe7^RZFi^^QW7f~OmHKg^rnJ{d2JyzcH=)NYy-c!fM>29%03;GBh)*RYb z-Oxnu+EsO;bf~b)4YhG>z)0^vYG7@>J(7CT- z)H^cuZ)-r{UcN~G$^OwV%lNYSq1DhNTSRh9_I)SuvHJg0n&{YB*M&eNSl7T7C$&UC zcz<$Zex6>2ULmL*&wgSUFO$!lT=*FI#vid znz<^46f z(cWb;mHl@3;^2L^%@#DxrSkpfhKLCnnlk3p9y}Hjtzt7j){QFHHa_DhS0enAtf*=d zOwUZ11Z;Sdw7s8x_ABE5iIxr+i?O=xnMs;(JD<@|Ghj`ht4Zv&via0Z2o_3TlB5h^ zQuO>kkCeHM$U_~PLRR0=ENb7tyEMlnqVIt~P&NDoA|zADJqzpS`9D>0Jzz_W{C|l~ z(HRF~2`JD!Zu>qDM+p)KXiHlt68VZ)0naZdic6Jlq{u&^yex6<68@Mr#EZr_vslCr*+8d4S_MVOO)C1lmiJ8(QN60z7- zP`tfEZTkJb6p^{CkX2ORZFm9D@(^Ow1bw%r;Z&?X9N7wq>V28|(>o^KFBqVCg)MzK zxbQOYIrzLUTrTwJ94wSJE$K1nAC((?k)gpLkfka62aT5eR2TExID9d?DVep{YaM@Y zG3fE$FBqxuOIlhysQ)VPA$aNwk*dULbF;aN6f_>*IXkKLj*aik>*ZBTZrdvXZ<)ge zVsYrRF4=(E{^R_k`xDX_A&*Ph9JYULUg&iP6-We3D1iu8DKm!>1y*C94{zoz1`KZS z-tSCiKQca_wDMY0ZcJjHpUO~;U^4q%(G>QBhl)H8nZ56*gd9@G&tAXAs9Z+QDyOPY&Wl1zT+PHEDFbRarK z|2m=?S7u*0|9zVD*>)VIcIOl%v&EH9`lTxP*HPXK^scUm_D$uOCuc;dw9JwR2G|7T z-z#rx&=yq z@UO1Yc>cLU=tI*z9XB&m`*AQTVE`GJ z*o*GId0Fmk&h)n#Jwplo9@9LPxUc!K;XZ6?AtsqXjF7%hs<8`{TG;B* zS5-hvT7qC>S<~|C;3K#x??pD%aO*c>%fFd_i@_g0)b=sZz4>`{>8-SYl5#-SuE6^L z`vf2N9vw<$Vhq;BCR$Do%MT5#a(%YsjNd^SZ!FZfS5v_8qh3{I3KGFzTnv>b-p4qU z?rcd>h$%c!e1V~FB5CHQ1UQeM?OboL=5v|fwXrINA_)i_Kjf6jnbTZ=@}CgoR9|Ar z?^}8C!g5k3CMv&gzfYrmW+Gr^D_wS3l9`>kqbV)GL+5rIi+ zZBrqDZQVemGt$ZK(%NVINf}}t16bbGTjMjC8;_l%ppz!OV?}l}-FIq`g%CVhUp<_? zcwQK0HA!h$UXar_50D}Kc556EUdz!*^c6$^jp#x{_xh3ZRtiF7W+Oe1;Jfq9)o2DN zKEQRZ>)&GF`G;`Cj~)BBToTayhT>aWXs4aB%OUeVl$fm<+V`|8>uW3k!#V@=GX8D3 zuC#M=8iNDXf3iAW`uXem1vq^{<=n|dLhY(C1!S2?YCYr-)mFLuNnzTM@L~o#^sjv1 zYt>N`^}ik@nA<1Hi!#=3Pn_u9=YX3WzL}L@r+OQR9w>8wHxEB;%9q3lc*Dla4G&+6 z7-T!1T|JlQqOnz1R-kgcgs?c$${qkvIzCOt{q@2w2B^RC%dZu+jhIp#CGWROMgP!_ z2R-Yea7?DAlw1oxbemq%!aLaFjgF)ApY~>o4yZll3oC{pe)yqY%|9*Llx#dfA`lhcv*69O)KSOp0sU$KvO<-EXa7yCR+C`h6fa z@?IE0@2Nv|>`gYox8?fo?2~Bw6yjFzW7#YDheXO+k-4~tA;abLi1W744;>TGAU~c# z%9%hYfWoHwu*0b>I-FG|Dg{nv_Ii`Oe0u^Kdw8%cGg6cv3Y{seoA)y;r|($E5%1uANu~G7sHXC zAtsb2bz*r}0p;XjtcK{ZBwN`#$+}9Vt^v1}L!8hh47G$R_(4f8p7TQbfB}A}IZT8u z!voa=DFj*TY}}kTe8HPIQ8`Jxs5OqPoX8G{i!a^%uK4X=q!QLKVX)ffiXbv$Yl&9GdLVnbT&s&H|y;5!gG}`tTB|~KGK0Vh-V2rt(BxGh~e%>{X zD{OUzq(-`!)cNe!TWfRno zy=y4>ipMxYD!TIFl^hwWUZYjzd=FGRUGKC2KVavT&|<~DJtwurInTKw761+#AFemQ zJEE}(j;G9Y1_*o-Cd!S^E1~>&&VZV{oQbtc{nM?Ghyvm?_a6I~Hmb0J4rDno zNv8}9F@!^~VzO{YBja^W7tg4`0AJeruZz8;k?2hGo(T7>j-B2femS`|ck;Bc`r__Y zJ-LWY=*^J&kuUcP&nv_{F_ zDgURGkde3@69-BJ-1!+?6KN`SY;3T&F7L|AozDEdxuWmg`O!O=8|qG-)VqSg5fJgO z$D0HOD-g>@pl5Lp_Z(i->1>}bM68N&`F*WB4SLQ0==%QY-NvO8lG6L{yky7$oH4sB z9hsuow*sRj==?T9A?s>4kER-ztKMv~=P4xUX{bkLaz^cpVi%U!^Tehsa4N4@PLu-( zMD~I(1=|JntDI>R)gFrHR5cFyXXkdOwT@?I+u!rd-TT~|;>K+nl00q9tRE%H`_}|X zahhE&T(Jgu3F@7(2)xKDBC*0+2EWz8Q#Rrt`Y8s#6|)WVFgp3+l??WPsV7G7KIg;f z_0G4dhQma)4)UOE8{5`|7|_ue9gukL_?hP(;yI6$%=fn!xBadET{_T=>i^>)BF2H9 zInY-unU&X^TsmH8|G6^&25$ie1j zR9ig(uX+>06m$ENNXwe%`O!*R{(=yp+_Q3imlzcOr)>YFg&1&zt$}zMm2Y`5K83^2 z#B<7oZbEOOx;iQI*TCv1&MJZPS*n*Kh~x~OVQsyp9)|3d&he0Ux!2i&J{}|ZF!dt% zf}!Ob0%s2Q878JEc31RPxj!{5g(~4UJFq+dp$w7}Fho>u-XtPyzb!@OT`!!z3afCi zr%p%1t0aC!!Q=D(B8!pnUvM1{x=q8I2OHJ1_KBlnt)lad?YxpzOt!z4$zmW)4(d-o zdU&Skgl4~ezeihqvzu+2^QoG}+p4<7Ri<9u)-u+Pf0k8iJ8pvC5ZmL&n%*UGaztCi z&}k%mF*Vu+7j(r&A|Hn!Tl z>$Xj1)KWCn|6a^!V>6*_xJk}ozm3Yq>!V-XcyYB9ndIOlx*tjifrLUnV016M;Jrcn zJYJlW5%tYmYkYq_r~}Q|XC3DIQB(S%M4A*#uGw(yA0~_L(-LWYL5|auBrJ>#Pf{J6 zI!h+IO_kzcFdz*#G?h~W?f0K~LVQ-ZoJ_%~4dj~FrfI4NR9G_Wcc4ggA9Dpb{=~A0 z4WWf!e4j)8Mz#3iZi)|+Y`>--OQhtx*-!q-q?$5@eEsPV*Vwsd|)A{>)L)AB(8QaiM6G1W>Rn+Kj)u5#)^n zVXkX-Emb~-R_1&Vj@Ktc*f9x=JQU@r0#OZU-fBaSsd0?@4o_O@!PnTJBY07nFH~FK zn0KlDsE%qfco(m*y0UKw6n0%w$EUPkmA0R1{OCuw+I~i0q)s&RMHJxq&j?_cIfNTt zJ->T-1gG4`5KXlie*Xw&(u)Svr7Iha>^C-EKndHxbs=VW_5WE=5W#z#ca|17#@Zy#<;tM+`b{ox^oQKKosGk zi+kvw@^2mMDon0jV7_tg)q@iqoiazfXBgBJBhi-X~ z$s|LY#<&h;EW1krOT|mA3r1hz`&FHy%D#%&z zuwu-{<=jYH`>g8f5f3L*K&P5ja@2N_9n$pQhE4Q~`x0>Yx z=bbu_?E}s`#c^KkdlPZ$bF|#*_7q+J%8^v>e*4Z`LsVwgIldA`*ROv5M}JFg5}=8% z@Vc6AxSj4EG}YD;xk?^x$Y>_^DV1+~=&-9HX)-QW+oy}aN0M7o#HSjB!dXTrJRCT; zkJ7JQtL1-pb&qX|PNl{k7e9g1Ru2uUE0*+|EBHFY=O9XNRJu-q7U#Y{nvW+~k*ph9 z24kNEHGXKn+QwJyb^^cYmQ!1tuFCQ%aFNA?6ZA6Y^x4hM8C9;+uZ18=&Xiq3v=G^~ z+b(&M7WMauG4b%qc`Wk&4TVMlU-StO;#jcwTAHUcP?eXqZv3^--`fKb`iS;ZM4?N_ zk-LPG@R0yca$2$`{2OkRiLAx^{U}NfBL&@eL8g~zOTBo!#2vGr_Rz@6xDuMYbawb8 z5*(1eMxQDALlVF~_VZ;nDdEaRGIjG?iBz|@-_G&=Ry*tMWhwW8F|}xqb!%&^x@QeMEm={csKFAWI)ODPd~@g$N1(1ZgHXBjhe!qibV%Sz5A#^<6(6 zjZa8+sVVnr8Ub3-4ws???2l=tIaX9uU@JVy*#k0OoiuK`Fg&UjLz?mh-~uP}Z``X8 z6$4>l0c3^S;0;k_v<>#=)p`1F=;Y)1#;W%UADb0h4P?lk8mgU`UMRcxrc~Vb;y!0N zHha$e)LIu_la5Y8$m*2|eQ5$~a$KpSX^Caj5&VfmTF(m6^cL9n&Zg!%QW^jyObt@c zYbu~&B?CK&U9`;c%A{N~1CFK);n~o@RyC$zMHK^OF&;$5ZIFKu7E-d+MQNe0?ipu2Yunior*N6RSbvRWzOtoh2^Q$`e-u6eW=N`8wh=fT(Lk}(oitiPre__%^e3MVoZ zBb-$^H)p}-aRJA-@5yOuXHe#Kx>T)<2z|-RV{BC}>F{J(8*CyZ9Lo8D{%Nai()|{N zx~heujK3gOP&3~`sQ^D>?`;C#VAFKfxMW9w-hZA{a-7??HgeopEms{+5Yh_4aQ}|o z)iG*%P{RD!X}lO&K}eg=wZX)k6a7h2|0x-iYe9!+{@tAI`nJ!poZ74V-OW>{?PB=l zL_vQ(dN%coBG4F5KkAaXrRo=p&dzxMD*Juv^Z@J{g;#hIXj-L*yv!If8v*(*zvSq0PCg>u#GmLBxb zX9iLSRN8WT3xX@J6)WjcF`qIM^Ilu9v3rCpFhC}t7#6vq`OW(KT3;W1VJA)~Nc4yf zDf*(w2iV=)>%|}YGF)TJFW9Z=U1iOpW#Kv~E?YH&F}T}3O=;sNbv#5iMsNhbXZ(1? z2ALbA(yzZi{?6)5ygbFuKpmO6r;aD0r|dzco2<#E5YhKOOGlI;j=mPh)lBZmv{A zZN~j$4l!H`AqF;KU{-l*&*q_%(*2hZHYGq&9V?io2{tWo)MgvnoQa;Vs^_@}_xcWF zZE(i?{H;&P?pq=0PK!+*ezV1{R6p7kZDVU{h!D9J+I)w0xCYuHamWtYuE8+3FHR^k>hhTj3O^MAXcriVz3ZLoPRQdH*FaG^z>um34Id?&hbcUZR`n(-taKH^@S1mjtI zP6#8_D6Wy~T)vz0%|0#Xd8jDOV4z{?63W;;U~U`4dRK#^sIk0W99Wx2q&0}~Gfu|J zpx2wmUfsl}ihGQFo(B{%p0foq7jnC@IW43*YwyTgC)_F8NI>58Ap^r8sX&FGKP~sV z4>i&CJ?MGp0V{WOcJS3^=Z~Ht+aOZuU4l889je|Q zr#(>s9{~zFDc76<1_=B%{(9F!QS~YuccpcQPK#_(+fy>zPPE^?u%o8sV;7~}QNvhQ zRwqnapG`aZC^6f5L@S`^JadBBJ;E?I*Q_C!VGXwHSVI{}lvPXQYz`NW`@A|*`Up=Y zw0Tyj>40N>tO=6H+VHexQpi38q@>(=#i95XxdC$Esp9Kedh#LT=jb1C^xF+bfYlf$ z21%*du3dU#dQMN9f4{35>|b_T+H-G_jY~kHN8~Hu$j%2z>30QBg`uSIBpA$+i6)d%Z%l0;{( zg+L79(c7m+vJ1a(&jK1g1cuUJ;}oDEH`_?1>Xn=ZBp})e#vjL@xbD~=KiQSiRviF2 z&)Om^A=nh^Z0>)@bou0GOQ_8>X1Ot0m{1sbNtyGN{j*l);cW#LHi9g>| zZ);wqcj8MW*~AmZdWoT(Eji5|X8M`b)~8Z7;82^TrrEZ1_CgXfMy$J-P-V{kTE|Cp zR{acP_v&;o3W=4ZN~b&uZYk`o)iEbhOL7;U`@_0 zCN;BLCj!l>BNf3}IQsn8?v;Xoar9^}38$Xm7k&SZEKTlY_ii%^i5#3$lho=p(QOL$ z61{Kif_$`^-fDNhnP5znoC0~}z@cU|BTh@NbCb09BG&?UtDcIFuy}dDUy#%peR;G= zI**awH*4d5TazIk(bK=47d4iAwS5W$V6TRwDjaRyCiF|)dPECs$<`=acdw@eeGO@- zmvf(-5YU}}VaEssiqijn1-2ghTI@#ZFL|NT{WD$P;!Iqs>%jZuft7!nIVVx{7-tm> zxfpue&YBesR4G!9LdE^zhHGBh?rE^*#gRl42N_K+~>3M^gTH zZ`_=VPURMe-C}7BzArwnr#v|Hc_(NMUTI}U+4c6A@9)5eZZ5KMHbo5-CUrg~N3x?o zJsmPIJ07xsuU{|>U19}%#Z9(6N01)?PQjg@vn_J_Lls4NpizvqF3E*wfTRf|8;`~U z6h4Hd8e>OOJ0`DIf_`<2tbJ2sdj<7vm&J9`_C5Bp43D0-jzI>3yY>6Sa?GC2vf3SVZPG0S3Jn?d8!>7jm5e%u07|Fa{=wB{Q9aT@$x_6$H$BOfydWV;eSsL+1 zbeykCxHA2g7I{)#(nh5)J-PB*tK#?ZhU%SY03BjRUg$c$Hi00TzvcYZMU4?~SnYft zXX1UU&fsD0?-^L;D#QgQ3ufxp@kDXfBdO+TG45aAa%&?vccM?3C}+=i7H7)pIOyvQ zmt;?j^EXqH&TLguBPBKnH3SR@i+?O7$M ziJ$1Pi5u&(B03vG(+FkndyD-pdDdQf8Rt@DBcML}NTCs9aXeUqhjmO8`S9G$A~!sT zbY=zJfP0<}e7kl)=sVoD&S(S|sFH0}JFlrntPLT&*zugKk3`j4h1)yH3%eo>p!9Lu7b)vW~I7_?KH>Y*+bpODr5GB?ebKKF>{3=z7sH z`F({pHlviQhWfAF?skNYtloa_Er=br5}r#FhDqr1-AaWdKs9yXfm8Bfxgk@WrKld& zJ=x-Y1&`F?9Cjp1!hMoW6vumyYS}P6vC(>-xYxkVJl*XgIzG>)=t~7vOS0sZk*k}q z>Opfch%qln@TG)UKH{T*h*YyE6vwv9X-CvsW>FOZuP7nx%^chlwkG`BmM3_3(j;hi05eOAMoYngIM0b+$gdzH1_{Znzz zF8oqQw$K*~kn&DDLO`liLe-!(jOC(;<*a_1Zi)fu(@4dB>s{zK;^8GOgKG1#4{Uc} zT(`Qakh~szs>$)nyP@R$=$RlR0qtBTwUAWHe<4x4eP!F5Sn=`+y={I87xM5d91LRRoO!{-}2I-u`izI z0ie5Izy%vHmO~eDbtyz_3RFJdiOV3b?A{wOv}s|n!C@F*y>9*mxpiRaLlOv z3ceWBLE{2k$o`B|CHXnS(8*rgr=fTYHp?x$IL#(%WD2ESH26x(Nxkfb+gQKC(HGbUq@rPBV>LDEC$tj0QGZ8(8Yk!gfHsceR9S&W>i7# zH#4B80iD4q_diI&@76gQzXS@=iKIiY_^W|6`*5Z|3W)24e;!k*{dmXYHU}TOI(g4% z*e<#_??8T>_FP{4=u7K(94SE?|kT-|8a*XUx6%m>Dr> zw5zN1QMyGvZsdFi;^>Ow^8+X@#U0J zj|@UfKg9}`h1KKH&PK$ldQR7^+tHl*t^EaleonP8f;MAzs-ABd+9biUnD|~iq=xAC z@qNgs>|)O(zC|;j-|qk^eRzC##89IkO->?9tj^$YwA?d*r>PY$yB!u^*JUMRvh(4A zi1?IgW;>Q~V#l#O?s)zprOokOO+d^H-@0i;*v#e71#7 z8p_>tkq5SO9S3Fv1cR{DWbN8&&2t6gl|5VWo2LV<^H}fgSRV2GQa5kS^~J|&a z)lg#F8+{gIAK9n*LnWuSWcJcUo1(BV3wYvvsg=q{q-^U{k>_lTk(cU*0M^7(3Mntw z_p4H=l?rvzep%?HNBk9Ex7EZ@X!F0b8Ufm7txh~Zg0Yrf9ya7(Je0UH6EPs`j$&De zX&UDG%J1bOS8E`3qC+A_L_icG`KnY4Jgj0>EuGj$O$tx$rz7dZEh+%1s z`}q&iHMj0i&)u_eXfMF-x@O}m^s7tyZwoSkahYhl9;gR$q?wI#By<>6m@0a@#!)8Y zDBiTw0?QRPgu4rGiX1pIgD4kDQ;Q6Ph@om5CwCt=C=lD_IY7yTu_H*d=OBN@xqYb6+Imr_=f61nfx1IPt*mPmFl5m2h31qcZOk|@0h0g@0f zQbR%w0Yb?42Cw&?Ta=F<`Ng|CJ3I62voo{1bx&mjWxtIm7y|wd+Jw%435ogy9$(bJ z?g9=S9Q3P_g-m^Eu$7SxnZY)<`-C0)y)bkHUz5;c{o7Fenj1l$2^n2%{d_ts>j1z8 zPH%?djG9#T*=pa@NILr2o8Hnf9v znNT?WAhEbRY^2YfYChFA6Pxn-OiB8O5rkMNl65E|FT#|df_Nb@Q1y@m4d|RYy_s0o z7?AOba9aArXF)Z|mb1E^M$f&qMp8G3Eaxp+6a!&J%SOb`R@>T)jadOHLvb#Ko5>gM z0xu%!+#B92F~D>_1TPu)wFHJAXo|Q$Ne+!5&pOCjZ5nkx6uG!GsqdT(nEtCZP2V%1 zKB!RwrtMKZlB))c9w76+Ag!1fx@v3rrqhe>g9P0&K8oBTI*(5~lpmB?lsPRZlu@4_ zO9m+Ut0kBIr1#%r6iW#o;V z9k`0TqzW;57~w^G64`$yGUyX|OZzHUn)0wUPnQ=sW^O7Lwz0|FG`aM&s9uBrXoYKe zY+}6U=WY0zd3ou5IZ~p|-=@m|_B6Po5V!!D1ON*Rc#y(Sq5Csn>qRtvNc30x0eiE| z5!tI(7V^zEpOW}Y!iZ!au%yQ}Z;U)@d>;tT{%qcr+W4umRJE!@|3 zTSVab;jX1sZM*}t0bfna{4xkqC(Lt8i)V~dhK5-s-~1baSEL3TWHscEdV!1k2VN4b zOdt1#Tb_JD3Q;Z0Cl9V%K)QU7!{DwYz_;VAQXP&e!92eTNIK*|D>CI=XjGu2edLC$+T%KaL0ZTD> zJ91WoQC?3{u`+K6Qr(7^_Jb|Myd}OiSN3Qnt9lGwGO3#$(9$U#H!C}J;`6qYnUpJA zajJ9a!UrPP4WE|iPU(uf#a5)N1UwYS;kAXf(?*=_cHrHQ2?wjZOPSs9QdyM#5pZkF zz?xR>a*SL5xnXI27e9t#{y?$h*_<7-C=(js?QP~*m;ZeE(CI)ALD#>}WgE;&QL!{K z)_9^a^husK_$>o@}xJG zj!1nqnFeAHFSq@E{Y?M+ubCDfLP;XoF6PG4kv@!G1H;!k02NaBVEa*85YS|!Lu=>`#~2)`jTCpZ#Mp!FD)HXZ1l%s;v}QU zcjZ~(qQnfD#3y!j!g`Rrl*i&Jt7WmLGw?8M)%H+iJ-8^f^An+N*|_= ztgAk^bj3S*=JXa-6W>rN(vzL|z?cHFc%>|JTFr>S31@}w$9#9MS!=1|M|b1XEl>Bu zR_&tq0je*)#7?vNjckjbyG*E(hUE6v${L+TkIdd3UA3Y>8#&=V{;IK7{yaCi3VN)N6K=btWm@$c@8qD$|xc9|qst zl`vFVv=O>R2F+|)QQeMySQ))RLh3JN$n+s44WAZ-1Wywz^+-orT@<*4x>nfhtk4~7 zps^KpBqA?0{`q`dJZ0oD7cjMIy&VI5m`~<&%C9FgD4Jnh$!g8M%t@kUr z7%dP?Ew(=%oTppYrB$xK)$K@NkyGuk@T>D$Wk+%kx4uxuJ4UI;W|Sd`A_AwPYHxtVF)ffXe$=G<9__49zAb79(cVyjEaMb zq4Im|UBKwoEKPy*y5FxABaZXca}`9@0hUqWJ!c2YkW~?pEPtNe!q0XWdDdE?~grY8hBqAK&?(J*X7kS=hA;J?$->{5@Byr=ysJ(7Uw>5zuMP`1%Yzus zzf(|Fuv9#Z(MJI(0oc11|8~&9g1*5acg_izRDSmDL%Z<{0CM)EMaFM)uVV{f_$FN> z07b%qVG697d(nhUi=-L;snJvNOtzv0z6~#7MqgPUDen--`5ot2NN%6?yXq^qJw-Wg z`4H|frdEZ#?IWE%_&xWBHmxnSROklMYmD=KxClMZx|P9KSI7YRwO!aCwNJU+^?1pk zx#PmCd12QRQ`2Oix&Vln)?8SZz`|dGkZ=z1i~1itJ=?TX28BDPLISD|w6LB&5C!5c zfZfViVfXu!G!Xcy`HSQWqfWq5W$W0&aCIz@UT`OgOI(xeaBn1ucc~#+pn0J9U<>!+ zH_GFI?EYDuUcjB-0PYN4T_=C#&M>&$;{#@6iy?P%jfDmb3yJ|H;||ltg*ANElfp;7 zq)qj_Cj|>Po$*)MQ!5W-L&LR44%;S==pEVid*JSRl>w!4?)v)&$`mU!_!jLyg_ulR zaJi&2_`=GtyX&ul+FkXDxr2TR3jzD~1Y1JQE?E2Opk` z33nAmd3Z23{YExCdS=$_>k%e1!?tNEe&kKrv6FN^iz*3`ru$=E(9wYY-SdLlmMTqk zN++!IcMEMJ)O`w)m;bcwN|7zXUiqPD)%L?pwPYQ$L8tPo+hdNDuM}3lsjc$l%G?g# z756(w?yE2DiO2HS6ZBIxrCqbUfsgDk|5hQx09fzGR(q2krPkT3TR`+OwRGzBMygc} z>G#FiycYnF*sYar%+oUUllTyPoAoq&QqT0LmI7p z@B`FXM;yNT!0>9N!B)>x6|x=XM$3bKz9;0aH&M~{WfhecX#=f`#rdJ$G?O8qyf2r6 z3s0p&3KK0bT~D=11@4>!cyYH)^;&l(JnRJf$>Hmt75|o;INq>q^)x{>*C79%XPKtO*9VA-{AbfV>J9u%d`V& z9Zrc1FrFaJtgUyjd%$7d`n;Q74L}?>j~~ZX{za~YSH-Al)>6mjR@Qjjw#Kt1;j^^ub?*@W*Fw*}=yDa4h|*23>7Y#?Sam z9dKR(0v3bVnHW0917RDH)k}2m3a#0gQgKlaCt16$G*ce-4;hJnh!fUxb<>N6NQa`}jtM%*6`4c)`jdZ< z7x|wjZ1WIx(krSa*!Oo2j2v=u@>Xs-l4ANdtt|cGmgRBPf}#`Nr%Pbw%CVl&l?U0T zaNh-PnI^_pbsXlKw2y6V(hN(`sWRY;viqbIKaL&ri_EEqT&_l{`jJ}TfT#T|6PfJ2 zQI%OJKBscXIooJ^Vy}XaLWD2!Z?gUJPRi|LiO~woIXN-T&)Tu-jxlqG z{b~10J#W%a9*owxCzb?=>ihjlxBo|p*!k+!!z53-i^P-EX16#2MJF(bWK6#0r(o-x zvZ~4$5jlz2KDJq_%jmH2%h$It&q=*`;c9izRZaXhw2B75yvBM-ykA+?`$ma`6*%5A zCJhdxTG=cQpmt}OnCU!=&d;S>^6TwDVQ+LBZD*u(0mobSS{)8KW0OFh@G4GU7uDw_#mP&)YwQRUr?dG^6ETQM@5Mk?w!J1`)8jPXFqv zb*WxnWC3gz3sjzO?RZk*v>iPWRLH5g1=;dxs;ohmD60@`g*H04k@nc5eqknoc2cW8E*^a7vx|?%EaPVOb6BQ^QV2A3|005)#G@Ek+XUXw4nsNegXf z-rP1A?Y*UyB<)GTd;_u1r`LZiz*^m31r;XhoIfi!&1c)t4 zd~?+40NysJp?@#wo3CA~hSP zt(79KJin+YXoF-ZP<+3sX>0%HRb0C(ZzZy(5^ViV(G~BXBjgg4mxt7nW8Qi@)>#cK z9{E1%2zu>*jk<^DFp1+0IVLk%2bw=4&?vSnw~6YI`32cK;87D|wTFmt^kF^*b}ni(8r| zUU}N25@?GGmoNp`meIie32e6iM(8@!*4Hl@j*2#BaHg2@i{TBy`Mq<4a?O|DUN=z@ zaFH8b1_6IUQtweYA?B&4sAvMb3q@>wwSC z@j4Li4=`js$KX0%;)Vf}f4#(|*yjEF5R8aBW#s~G{pUp~eM0?TnjRh;v?C_#o&XvM zd3G_N#=ZpIlu6pr?ybSV>fBiJ4@Hgdj@Or3nuL-!nFs~+*H;zufd z=7aKjw<=#~EGV`1L<;c7u#_|(+U)4|N)Ijen(o^d5bzYBQ_P>2LY z_tVV>LsSi47QK-ym6&F9v|!>OskEC8JiH1@u-0Pz@u1|8-Q;`50nV3ISF?*Z$?+nh7a%4 z($EYT`SOJA;}~}#>aQZ_#(;XDjgV)00Uy@VLPk?^O%$xJO)yNzXfnXkp~6%PVX7pY z)7n_@&e{^&53~ciw3ES0Z0|oE4^#bvI9;5Tyg={#ahZ9d5j-!H87hnV&B_;2iXsQ_hY z`cj_O(w`8n#W-sd(2By0BbB(8PT#y|M_HdeQ#f_LZ?sRK$W;SNUlsCvqL|B;hHnl)yuE%^0!_WmIitE&*F z%WlshY$op0GV0QCbNwGWc z<|_H#B`RLhx~A;+C>kXtQ#Cm-DyomcedwDQTd;resxt5R5(RIe#TIFvDsLodv9@x<} zROD<+yimy3Mo7CZ2eZ~(0#CNqp^D3z;Tt3;zT}0E-{|cjDYq$oSOQw`$-VN&`CD}U ztlD~Y^cSp!JB|W^gc}mM%fv(}xGS9FD;rE~!1v+Fc$-2#Y=pGyaVpo6Cj$vPa5ccE z%eYY98#9l1b6{uJaFJpes(X4Bc^D&WY8Q@ib2sjwTybzqqQzD7!tK>=B{kr?@n!^Z zKnPIe{Jk}uO==nLj^r@gnra2yJ|V%xa-n%fmsLcAjF_I-P4QKyb$k)?7d|;ZN zP}dumCYK2>4d`)szcz6{o;-Osd-)LH0pfliVg(L)Po_uX*dm(R9vaKMC{hzk}#_2MOi0>aO%(@iX_jy_tBwzUiU&0 z=TWV@Gx6RBeSzeecQ>)RZdwCp6j|+-JN;d%3k3pcC`+N~W`Y>Sgqj@$=f{d_ zZ1wFE(2E3dz1Ju9X%TsEHi-ojBRbc(6|eb`opMQOjZG< zD8CI|zt)N^oFYjcXc2RgB`>a&^|Avk`zhQJ`oO=gdjo6Pd2 zEqv?6_p4=bE!J9fZrQN_tZLjLU(qR$D!ozNF3UBjhEfLoumdqyHmrNJAlzxhal~mP zk7hJ%S*ob%u|i>f;`r8R@wI*rT{U+h9U{2aDtjugzo^SvdE_&fYCrc$8P+fpAvkF8 zq!D60ni3q48xSz5n|PAt+FQ}1)>|>QnDri2qS#eE!5_upk2t^iXLf>uK}j8JtmUxc z!p&ucY289JT1r(rvTW(`g|-)8Njmc7(h$q9DFHx7&{j^OGApoGk%S_t^;C>5X1+%i zD}M3dFVZ3Q(O6zQYVvDN!$}7upYTDZxLuonx zp^HB3Bs#f>O?97S4d`(g2@^dNq6{QXjeAMhBZb>Y=O)7HjD|3}Blc(_9d(3(gM$fLY zV{kH?6hNu0aSdBBSVOz&L+@f6$uD*L*e$t=G1hbg8?S0kcbtRZ?d#xPij@NZ3N8IA z1n$6)A4^v}xGLyA=4)5I_k8R80n0LRpO9FTXPGBjvCV_uc?g`dI6>X#udxO@ zcSIW}2~Z5}wT7dsq zpR#NI`KW+)cT)U$2XSwtF+kN1tDN_`;RlrLc}X(nl7#grpM|JK3+EUwp%N8%{e+-+ zu&eF2hkwH5;0AD*KaSsCo2)LCBgp>vKCN4?$3xo8bMMg+0?1C8^tg zy3K%=OF(cv$hj$2m{40zl;4~MpoQ(KexDyO#l%wOHr_B^#D{0o34apo32zBI_|juP zrfjpQ^thSM*M{BCwNYW>()U#^Xl3x~&yCTeO&o>3Wo*)ue&%@P1Wh z7>V-G+^6!fXE5V9))x)FI~c?w9Lkmn_=GH)X;1{hYE*$0P*3MdKH`5C(ms;FK@1`=PJZdc zYnhr3Ki9vVgxS(_u~zFltKV&=NtENUb3NsUwn3C-neQ++Y>Ta$LFsgkmq=Y zYK3r8**^ak`x4&I)uKbe{E5)y5`{EZKRXyS`58m*15f~0966u%^DU>#+R{zqx*x|D#-0I$W@ahy90PtE@k2+4EouPlZ2xiL=! zimpPj(WZdX);#50%kihBMArY>>_3>u3|GZ2REc>~jl+bFtV)-2W-Ogf`5Kk<9O%gJ ze|6+iUExeD4-UF>Y4E}7CMU#ZI-lU?nTrOH2z#WD?f+j9mVQSe{Dd5eKB7&J9+Xq$ zcXUqj?BuHeXY54Tg~{9!-hazi)qW_l*#^^xTv=d*2_0A!&cC3 zlsvK8!Q5U!uO11MxLX?ZFNHip*^ zvoQ;}8>$$J!o3key;==LMSZhv!qv`;c&2y(7GpQ~0C4#~$Xq2huV68qL$1iK(AJ#Z z!iH4OsWN?d(}i@c+4^da;AP%QA*Z--`2hzK5!BA@8ic@u1HM(4S8FfjvUsIDH(vh= zIj(q!TUjPfN^vEYxp7WTh=D+VzMB6S=`VCm!NlP2ntFCHQnT1849sXx5|rEzFxH7i zqV?#?3`VL#TVq+09bpbxaAs#i76aJ9(aRz}3tOR-H!lp~H?ywYeU$pRb#|`M+YSmj16k}evY2f!Qt5yI|`iZ5lu_KkO(Htkugry95t|Z$K8tpEX5_m(iYH1!_&A-i6UDcuG z5WCybtpfhil~CqDDrJxmF2Q9GQnh%5c#L?w1Q6r8?D=Oi<>|=o$cv7{9g{x-OR9T9 z6)}1fPt0S8TA*_N;O%jN+%T5NH_o{UUX*47Qp2wZ%qZz3cbP&Maq7VPe?` zOdX2bi*V(^4{04PrWMn^Kmjjq7WbT#CwEH63;X|vMX>@D=D#&$!e)-D?$_D^?9kG8 zR?kW?<#9@%x&0#Akl>RmA~9;OB`)hD)k{}C(W1q{XR61qe(wRU0k?t&!js`uaI(J~ zfMeuxfdUWt0!!1;?5+4G6?)n~g0ME#o1wWXJ1PM{9pWRzo*oWnWy%@ywt8V@R(IKx zs|m?Y@z22w#aWVDkYMH9oHr9^CwwQ$Cl)5oGJF~3)l;*c+eS8mKn=*HQGHl_mTXvx z#ecaZ9jI9b{F}VWJkx=JM=bSK^S*G04!j^wA_hH~hPY)PX2P`87b-(JXvquhO*d)I z@_+rWei#c3jfudyL+e22$n2D=#(Z(|C58Eu%~ng|KLYDZCq*qQ{MxMNYmc(xw~h;e zL7=l=ZY#<8q$1sYxk&3^&yU(f$(zY4u8B#iZYlQHHO;@P2)@jUpCz^sTJ&0lL7pnjV-M$ zH(LQt^01dvA!$_^Vk`3(W=`pco3idH%zv7>$ov>lAiaDJK(P0PE#+kHbb>sF$~&{K z*v;SSz3qK_{UCF=1Y8Sl14qD9;5c=_(+<1*lRf-rXqEV)rststb2K&Ca^VQ@uO1P+ zrETn94~RF9C79}|8wD;{VMOc2$+QQ)D{TCSkew~W7E1!a{2%HEPFbkVzj<7*l?<#2 zKis&G7>ZJ)E>?nP;45?>;FZ5L+ZpX$%?Yc9Lo^ZBxTrdW-)BUw-H6?Y!^qtc=aHa~ z6m989z>c@?{CFf|yP? zIxe3Ryfl`rYw3ZZ~}U!#}mg(xV3xanx~R{WCS57Y8LA}Z#ICsLj%Wr-k+ z_|2_io=^3H8rw7?xI$t-1$ag%@o3biqS=z!(%G_Ev_~5X08kHIh`-5^T}MUfz@fG8 z`J+zTrsLDjV4sQ|I`Wuf&Z?F5&MJrQw2!j{WDc46wB?iv$e2|Ohv=+ubCd-}g{(g< z3vW0-8;7=EdD7*Y%>lf5TWbqsG&=R{XK^+Ayd(f|lC*;S8SDsb;*q}#@P4{YaXlz4 z{vQDyYzaSw0_CN2OsWoVh^n|hv`wrGccb-yj9Z&gq^*nZ50E3CT_@Wt-gCxmOWhXy z6E8^+=+Q8Xq<3VJSH~4*q*9AJ#%Hp7+3}BfWFZobtd4pR6%h?Y=COZ#-%m3# zbAbaAvl|=5F8DkH@-+X18`?(l(zsaI;cL4i8LMLg72iH_RqULP(Sb_M8fzB35ex#}11 zm-pFgkn%P9pi-H{HVSYHf1*fKP7yF8@2B5;pr+duH-OSaekjEYn3p*4uEej{uw*y z?nZgLij~QdvX}xOg#mjxtMO=Y?RIg`2@e+LO-cY$t@-gKyWX=szCGnV3q5DYde(uA zl)uLnO5%@;0!F9jzdPyiKa*_)$LTK}>wJi>yO{BUCK$nyLJw+6FBo%!pVR?F>0P?;MsqGe(bQ&XSWn3R!AN@JHIr@5x80b>bm23lV z-6+jCO#S+y$%DnyD31;lFgs4)1GA3u`U;m6y5r}5Q`!5kMA@^iuk1E3C4%u2+jWEkAtB1YTq)&@j_A3POx1|AJBfVZgc1hpC3%bO444{5hV zX|6n6qJ;uWRQI(Z*v*GbB4sb=%Z7MqfKuFHTOo38Ix z?tq@Wm)q~(ik%)Lz^GDvug}O^61q!Wd&*pr%3Ohmz4X}Hhg9~f99K60wKd9>yDY5` zPGn#6V3R##5Rqk808_jt0trNMuvCsFyQ>hqN;{epU_ydf#;2PdIsqD!t9`uOpXJ1T>r!$2TEf6eUGRdC zz%qfiAW88iT4z5k%?G?G1|I~xYIR-2S+5sx&tBhY#!FC$&Ff!y%&7AI`9SA{`w5!k zzOze2vYuNNt)EhC8W7#kxav0T(Z(^=!YU>z_Uh4~Ht7^(_4^gGfvq$G2DnMLdu1oj zOnbj&0d)~cRr~$TY2KJNSIURWBGO;i$&s>LE!k|yN@H;QMbeEQic&$SX921@|0`8V zGtHBOBYMis>Ey7+a59b5?^TG?UJlPC*{%AW?a1qq_*#4m-iRO!%7GKOLX6<*o;O#R z1nm_RVxxVRgVyY;3r0hAC^1<1d68B*lF5mEO4U40b>UNVKHnwBClBOCrAD6w47}}o z0Q!F#_$|hC<_2c6pWpo+wWsKEsWR*sJgCYv;0aC^ECQASt7=mBsl=&dskeZPVpDB` ztThK-0m~zXUVxc{%I5;_Nwm~vL`4nuM5*?_%`LGii!wW5&w&vu9yXM@((GvVv~1`$ z&^h6MR6f`w{`<~#70ZoZl>#0--EmM=YZvq7D+OmHrz_EvoaLj{rgi6IEHVR`i!6@T zIUsgZ%u)<0?gJVv^OUdZ5%*|HSPCe%@g3a6ecM=&%3B`E9PH6orM$`gJwD%Z{}XG! zq1ei+mW}w&_(ps)-jJ{l6msgvDs`qtw8jk5@wDgKze-4Q+$XXl`Kq-pPFRmue-;Faw8iB8<@FR@Yg zqr##dN8gN$Dvhd$!rBOdM#J}JCy(PLd;&E`dvQS+M9B^eVyQEA|N1<<`$U{2tljvv zU1MC>o&6uZGx#>)LUvj9x&}aX=&zvKQo{NURmFTe>`qn_MeHC8 zEFRxqaNkA*G+Kh99WRF(ikb@f_xp{Ab1ysVS$otAhaGO@wDK0UF#rqEH$C|!AKh;O zF(I;LLg*&FfW!8d_b!YjtOprw`o*fMt7?U|Efv$w743d_Eg4&|(ldYwn0HD&ccoXf z*Q_^T>}+_C@_XlWq5u%4f20Vdc+l{~2Pv8X>g==2b7P%?w@sjD;T#5N|=04iquKt=4XNOR^ng>J1MeAziU#1aye zp>S8YrUy|OW3a#i_{+zSx>ALf?3{p9_9FmmWn8a$C+kcc>@Dq}3xF73@^w#&(E;@F zIP%SD^&OcO|8pk$pO}9zF`hng44{8^zn3Lp{}=rm5ReR<*sQW+|8cn*vuDf@_pO9pue$f*9$n_iY~Jhi6Mul$5Wn{ifzUJ1|(_DaO+FhkGYPE zChTEdN`x&BZ6}92{gSi2Fo?)ZOp3nKB$lnEF+lOrK6`k6k`*EM zXI-20dAa}u_PBt$voZ#5X6f1cdU&#|2=jtTDym~}{;&k)+_ZzbLR0REq+07>;{1cw z*`96D7@_X0g1Xq+p3}}!x6xi|kJ&b83S1v0Xh-`-Rs0W>{YQP-5hwKqsfd_-0ZsX;sZfN5 zcP8-I!`BO2vRq!-oD`E1lNVDJ(-yx8q{_|*Xl8ylwx6cS>TdgpWInt@wJz_)0xV=H zK>J@6GhQ83Ys3Ri(`_jE&F2>sYadFcq`q~opnTQ)UEzKr)?Ux#Z0`OYJleMQO`cpM z+b0^`D55G*&bNtw6z&CiHJHMJzHnE7?iBY*8L!;n2+EbG?RsKv z+wJ^sSS{sPoxMiI-ch2lo~p3c<*veR7oPy_!LU166S)3F-zaN+h=J(rx99#JW;1D0 z<=?49;r!08k}FdY0v8lHBR0}}#{X|x)LKi6YB2NmWsfRL)vDn~AN|o2eLcz@^vxjT zRPSSmPq5$Iid&Wy*1r_-yEnE)o3~tPBlDvVaXnz3E^^lFA-^)%_WhXZq5g_7Z?0zQ zj}osw#;kUEEw|m2PzkW~?*ww+^&NIJ=<4E@1>Fidm+d&Ymbx;8Tl2*3?X?%yJ|{*g z8&u|sr~W8G>l;he4r=Xrw)c1iaEH9EcKF-yB4G2v$LmhiMo`+#65h=wMo+)yYRO^X@Exqm_2)3lUA^&$rL#kS`T&oQ zjS_YlON)top9OIpU=FTu&aE}eC*{mbP*dr>qgBTqFf)RXz^f&$aeH@gEQ`?nLKeW+{bhJti!yB_(iXIq1C`YEIGB|4REGaxi|C5CT0?gXP3|uS zVj=(7Z+DhR-CDz3cPj;XEYiH#u-^1)qECs%e+|?))*2e@0&0^B=DiT3`&<;)^mv}6 z;%dQ|?Dk>N9B##aj}q~>v(Z3WiIS$7zE&B{U%Dpo)gfGhbXi3U&Z*igKiuYIseL5hJ8NteDpY>^*=*3mFGDZ)|y(U z#bH;s5!){&}%v4cOzKKrZOS>BvfS)_}O|Fi{7F~tIMz&qC#j4Ykxs88DyY zd>cBx-!YS~@VN`UrEJ1h5JO@3^D$;jjAk#lcQ*j9#|`0~*@iwDLzbr)8jL_j6=T^I zFlVnwz>tqKgw`@-UuWm6=9pl?As-PKTym1$lvKYN2oX@wLs2&zwdzC7Vv8bOZ}P}JiY3rjmsCuQUs?wwZCEi|mnObqhu;f?%r(1%RucA($Ls@m zK5}VV>9@ufe)O5tLdBm4l1L%2PKoRPAb&rrNssU$G{}YtU20|iW{Kehaj`g8yZbTB zy|;n>Vw>+5fw!$YOaMrL4wq{IQAhAN_!Pv_UGBnx|A;07`$*lj-0HRotbaJGDZ1L7 zm&Ga^n`F7lhN8TTH$zQSMCKAewq`bFw3V5rrxXWZqIXILzXlJzyh*O|o1FtD`XxWH zfXrSE!3N*H1_HeenSZ?2dH_71`8DPfQpN2-fKRIpmx9ah5{-;XEiG6{_-$vA;xz51 zw}QQA&G5sPy&%zOXsY$lO5iwqTs~df3BaxKciEN4^x{T!jUe&g6C2Dj~eGv$thT;EsqIUHB# ziEUn$zpQAn+HFsr5kb&)qN1V9y;P(^*D7a#9yiF*hA;F$V=4@_#%dv##7}@>U<6nQ zED4qeYl96neO3dXuYZkSQ6#j_Q{=6C=$@}gfUYixpA1FKjF;LrH4KNhhyak0Aqg4K z0#DOfgdIX+s;HIMOVWi^9gKdgrS=p5$_XI3_C4t}Q8lqVm%I)%x~VVb)Gvu-XE~Ro zhkHB&EX7M;1YvcUilW6uYOiNEgZgea-PEI`26H(qH`irtD4~K+jykjS=}GZ3B#*x7 zJ`d6~$**Uz$8W4+JrGie=8kn{WAZK0mvjE|&!t z+rOPUKibR5=wqqGI}>4U;}aJJCN+K&l{SGY*aLw&k&dV$wOR}wn*0^cB)^l?kA3+)@B1%lQuYFpTwYred5dxsAc-as@AmqO ztE!HINgi;yoiVy$_z6=C)aN>(Bu%DT`DSvO1_|8|TOPo!zewIJaq~}Bgocw)&JXBL84pNlF^|hJU>k6 z+~jiRgG4{C%q{qaTp}LMbaruQ!WvowTXU%)a07GBe}aKe*O)XeYtm1!TlgKNTf+@&L;#T6N|TS1k~ zFlDAQTWfMJ;&EU!# z8=*s8LVy{IC+Vmcg3_)MG*t(7R*$H$W0NG|kcIl~>|pr{JeO4wPlWu%bCBZYa%fOP zc<@vyZJg)Kqpo1N6|xSEB>ys;t1fy6=)>gK@!zgmDSBnZSk4qY13!4Fr`FeC^p2fom$%CI)apcK#Z`R=cnQ4 z5`sWWA^q9Fz*c^7ysUe>i(hY4s`2ual8B60M`VU;CdVn& z1%m!UphT%|{D!9y;czEhds&fD_N&jHLWm=v8Z*?v2a00S;?;lUGz>p)Sv|3WH3W2} zY0u9M{zO+5{=LJs0;%XRq|h1asMGWecgATu-hXk3W1Q59rj zCYE%DH0r-QGS*ak6F|bUy;ds~=TsQFvGcEfeiAV=sRH3VsMAXS#;L#Ylsf}ODO_4Xn3%!{#!1>v)l`h{Z05V&u)?>j~xCi<`^x~wwty%~KBeI$0&)}8nnxN0Z;2K0kNpK(-8 zLR4%tV3OYJ^BG@J@1# zKC&8hctmf9mstNBwR|_{-`l6B@E7~*AFb@?yxNAcGc4c+CZ7#JifKDr-?pUe12eY@ zOPUtIQVdnuK=$@AFjJgg%@p#q)bdkXRhz0pjtk#{)sEcksux_CHCtd^`DS?sk{<*g zo3WEKQ-70m)<_Uub<+${vT*xzxfRwv;I`kKs^I0{v~OHtwc~=AU2JW5t`BxnLX>=6 z$ka;`X*4fcuo5(z9!#8j07tf}Ldc_X*1LnVBC=0o z8oiw%vNAC%^Q=WN)@h+!t$z070DDfslDJ)EAw03{@QTkM))KxiZqmC>Gxc%lG%kDI zCvQ`1QvYJw)Chwt0*bWfyNvSBT}DqSKxerHNyW{?t+be3@um-Pjvf)E HM2QcoK6SP#4~o0cXDXgIRT3fNr3@{xu5q0|3NPu2r_eMehz8v@G@n z2eG3-m#8+DvVDMH|T#GcCm_XArNzio!9an^$-_%Qv;9!=#a zRtq1L=3|E*SF?Sp{@J4n!K*+0N1t{zX1%J|=)xdGua$X8@uIZAitCl-mWjdlZeNlQ z+7hxF=O37~N~wxA_KJXPwfVr6_zU2waA~hG%6!HR?TPAe)7#$eX6@pjoZ^|SlL-R~ zs%G+En*3Yu6*XoVRF{C8rkZaGEVx&1i*Kj8hbn^3NvzqA`yjqnAZ4*7w0Q|*uX~kE z0xr-677o-ctdR;~uH2{qC<1$2OssJ3hWWI*P>mrDvtY?ev{oOjf5o>jAaj1P0>&0# z1@lI`RVn;Dbh&xG>zN|mWS9-lqGm{%9UQ=XSs_M_qT6`OW2d7@#vGU6{CPj@axl4Z9X^r8j-Tl+YT)$A9LncxSAgW1F6$v_ZA*!mx+4*F9FO7U%&eSQTUV)d7 zKq$t%^&0}N_28RVBv3&;tl*_s^msaKG;xC{mQMhje@+16hO0IW+BPAsGcS!Hl8yko zEpyL9oq}CGkw^3kHv$`ZjWZ?1{UG34V@KZGG6*Hb;r6+=GG7+EpNi4%8|3A^>twvz zS_%!C8A5waU%;9>tLKlw_DA;4YdtyhZU6ew6|A1=N3>;Gzo98M@2TWl`i33;)mCB{ z=$8u(FMYgwf7@Sa=9ob`PypWuwOmF1W_dPC-I&Vq?%$oMJW zys&1B);57GmC0(w>IIHzcCg$;xg{6x5g;`h^Z=|rQ=kFj>HiW)} zh{yFJGeMWM*79{VdcC{Cd|jr_68O3ukP2^!fHh&Qk_2Dj zm`RG>wE3i6jH(K(Q9`$vaExTGC0eYz4=Rk=+riv?Dz^=py1%Fw>5^VVIOn~HC|98d zdeGkErh6n(F=i5xz81D>2qoVPi4xSOydOF#38efJOn&9|seqC-!#fL>eEcS#f8Mb@@8lQo^#*2JgREqv`K~>fo5#Iu z;FRA0yKZ~RWndjql|eLmgbYcZXy4*I{r+(2M9a|tcA^9{dT=vTsw=v#f0Ky`aOnWO z1S)g04ta^O6MwIv_5j!1bttg%K;(e{f@$7Fw3?<+Yv^JRxJ_oFoBb4U^*i){X@aNa zNzS3-4@PJq+qpJ?wiNz?ekPhJ;KiW~`B0S3Q!w6wC?o1@S|5rk8gA|2aQ0qjueH}+eXmoDMrN$R&xrZT_DorM!Wb6cpZMg}YBl zHfk!1ZD~`B%ekwCJO`qYnEou{2tpF{NhzqgGujRhvcnfL1UUV9wwh>F>`SqgMK#!UJI5w`@Bm44`T+1nSw$ORFZwFqC`K-wupJZU^yVMgisa1^IQI@j}N47 zk!sbd;NU-U@LZM8gwYZ7S&2i{v{i?3oR6)VEx5Ou@)A}cBGRlKN;9&cRIBcJm`^-2 z`ILj-rB?(zA#tl*h1@6|32VX8*CY`n?>V6c*Bx!U)j!ontIO9^0@yM$(lT2`X&n4_ z^pmGlg-)kC2d>q8d=etJgL{+FpGHzN%sau6^Nqq>Htaoym9%3OH528$nKilkuO{&1xGyMT6W*?PMcnA38i~J_vb9$ zk@EH|&bR^Y6G0)@kd>Z-)sIB#G3)o63%tDT$Y#J4>oFhXzUq1K;_9A`aoVp0;Qpv~ zgj_LlImgggk%^1e9%#lqOawn+s0*+jVs{NGkguDZI||n74gm7XdWn z_62hTEl*+OP5WyZ!r2Ptsp}DP8*ex0twe|ewwCj!S}lW{zhX%Vn7{)dU5MH#^3hZ} zGQ{$ybYFe-g0V*A76m1umO(sBbc0IY{m$fu_g#do%KqB#SFN?HDMS=bZ>~Pqy}DS< z%i_Igo_u*QwKDXROMqk(tmSvqn8GbMdAP{c`$p|-Yp+XLx%0r~G{aYukKV{8Z=gEB zloYs@iRM*uCP2rciDl&$f;|!hR||<5?cyS-d3KU<$**cce|3pVVFFC_Vd}Y>EE$BP zAS#wYtFL;{zDjW6!w(`Ukn7C&hU;uwyl)=$V+Qv1cP$GVe?=!?PZ(-C2cC=+*=Ije zcNsV!@1vR5y(BN|KXAdyO5|DlKG6+vmPjsmYy%d6<^6P^ZZJFE1xMttSf{aK_q)x36P(G6~hvVix-=GCY; z_knDBmX;>y;`Y&=Lo5nCYQA`}r&U;gGAeV8OYsBJuEFH@oj*078Bq(HY%5BT{9!Je zGscm#RwESD%=jR3Tzn5YMH3di7-`WTGXY6zfA9ropkUKpK~!j(mZ*-uuUYmSh(b*h ztA~4X*5xI1Uf-TYWtV#_*tW(;^~LE}<52Flm$;7KALOryyBNK9Nj*PSa>GDer(r`K zD?*(7TQcx-{|YKJVZ03nWD=D{sn`I^0GO$?{~hn_#5HJ&<-$3>f1v~LzrV-=CgAA6 z%L49^TD8p6;sdh^^rh6L4ih~0O!B@HCB1DY?_nY{<&;F*Y2+8VEIC=yEq5C7V2X{P zHH({JRwYK|sOf2G603On^j@B~>yZw^5qOW0)jMhP-Y*~7r)taK5* zcM`MI)xC3vXr(Ikqo<-efgooT5_mx6(=E)?R8hgSK;1P0WxLAW2a4BxQ(+R#t0~eS z(*JVQ8wj!_dX|?NBj&)oNj=uAS~~ z!Ois85M)`o60uihTV!>YqGR;WtX<}51)g&eqc!rrk?xJF)H{RHkz@_Rh08~`|2SrC z24wyins?9D@w>+hvTa1pB;l=Tm7GD$R_>EEXbdmC@Y#M-iSt7RU1vn8b-+yZ8DEV_ z&CpPEGYWNK+k}yQx<~q%Aq`!1Yg8KjLSYg4-BLuEjaVDUQXq_0@qnzfBv@q_{`vWLN2nGBB9 zNpL6&q?G(W)+TZ;Sj}Ab%DLYqb2EwmLbL07FfWw%^ADsq9V1h&jrX|tG^@KY*l*vNeNk_7@?)|KDnUsCDc-otl z9=J&Od8>Z{Y0Z(c=z@xhuFiCN*b|%#k|xA+WI7$T|45}Fi{+GL^tw%{$P_tcf6XLr z_heNOp#o+u;e>>bougk;uv2MteP>Bq#)rf60T)ZQx5t-Uah^d)wI!oqGB#}VIMsiBkdkl3>Xl6sr>3|Wn5Mwe2)1mWfb^HQLCtCgKqXe^v}3nV ztJK>g1qjht68>zwDmZXAZ_I8IO?D`n>f9(w&nn1Q+7_A6Dy%$-@eQ7xqESuhU1kz8 zy#=F_G$3$09i|rc@(f3QLyA%_Q9t(G)k@zZRY=VDVfAq7g*EGKa`X+jli9-qmAYFw zq^pZd#!pH0c5mHY1ukM0uMT=NQwx}~oqKOLK!Njb*$_iU30oIo32=%C_BjaHMBN2v zHq!S=-|ItC_zCM7kM-bVVRl`X6Wje&@|%c{v#&O?(KPbW--B!%&#<0aRs)AhZL z=lV4?WI^7PPpw_t!Zml!3qIl2v)^iqKf6rjPIvM(yR)J@r;2)K-aURdm;S2E?ZLnB z+NCNLfvf%fLX3#EGb&}Fqt)46*-Q|Xn!H$=UqR;{DT0ApVL6iAn;C(f9GmNOZbz|i zJ&Sk|jX@oHHUhN-1Fp04H(Y0KQha4p;}DD*zxj4Gt&5aS$4l?&-XwkNgpZZ|wG(}6l@Ha4Uwm;adWK$&t8b$F{qIka!KonK zT>x=|8{_usp_Rocq(R*mH2LV^FjwzT9jx=#NzJXInbT?KM_9h zP^bA7UECKH#bs@w=$X*vO+HP@04L%(hvzTcxZY&fyHWydLwM8SybVjv8;=LA6ap!||?L{ny3B>L|6%1geUGkq~0p7BKeX&5J6 z?9QE*`<_jzK-6(_$BLYM$GMw@FYBBz(MN`EMa==99-Ykq8(FBesO!Q%G$!U5PqEhl z<kM0kpkFdL*ahqp()9C$LP@y|TqvLX7n@pvj7Z8(Ltx?S z@MCVp|Keg-_T>XJ1^fNlt^^QS+pFV92MUXo>?~OCU3as{$ zu@`a4F2e@HAbQk}!kp0015Teu4zOr#^2W;ei?(1#MTT_OR8E9W^g>T zJJviC@U*2Q$>&F>I@W_%TpWbirI2_O!oi6gI)Uql}*+I8@X&%nzT z0|oexv82?&h4-CXfa!P@D|wb9>y`?1m*yqzv7vZZ^mlb-k3ARN%q6hMCTh)FaKX?5 zu@()C*-xj^n`U0Dg$pkUUJ*!n`PnLy%~KA?7Q8R>6pDzop*9o9AmZ3l_1g`nfZY4m zqH|^!BYGtB+h8W6<|H)_I?F^_Vt?q}*pXY-UuN zvDT{97++n|!Z@dsOx6bxQ|1uQH~w2D>53*H>xDK0>WSASj>a0X!~Biyi0 zxR+sxvy`b)UiXC&Bo4Gt6EIfK;ySBft_XZpXS|YVx?Z%QX5&z*XnP<{OH2>rTh&%# zt)ct)N6^*}m_*Uvh#|P)?ZpGF^5%h-x=e@-KRH(nP?RN|a!A{0!I4PO7PAl?A>TKz6zy1#W?_u$*! z5%s*A4 zW-Fmllbb+#V;k=ohK*!vx_Teea1&Je*>!0btIrRn;e?m1J`;}dk;nYCjpL^{274V+ z&C+Uyyz6T$Tix0tIGIQ-S4uA{JsMCZ(VD*{P_*{OtSO`oYU)NwA=plQtm;G`-Lr{l zR3Wxz{vvf*9a3K=WX-?LuZZh(4z>$bQVq@3oMk`#o7>)(f33&A(@8p9^nb z{Q7nxUc40VifGgb0;L$Y78)uGC10$GB0(}vBD78TgX)ntM5$4^d%m85RSTKL2%>kx zs$b;s-#=S1`Flqd){^-8OIT{DJ?mk=j2$Rc<1eRK|Kp6W*1(Ig(HZRY9o*zW+_vO>oyLdFLXMGa zquyC&DuHahJk%fM_TsaZvf_;Hn`7)Z!%)t5;EKUeU?LRKrhJdq2%9_#r9&szt*{6- z9EY4yZdYvWeFAR$G^=sSu%m*P^~fVPKOg;XQqza%#%-1T=m+(7dxn6LW}q?EKX%q- z7SS~%bwxGG@^2>R77?ha%R(oM| zyQ&aNYu6y>mg?6)`S4phQbf}K$$Yzlt9Wp!YUachaIx#NF39gYDabAJ92;a1!x_@a z`p_QDML@Rxx0)@_xTD@&L-z!K@b~cdMOAkHQSB)Iea}xaTM@b>psj2hW>-U7v$WFI zM0X23o5i>5t|uLqbidwN?wqYArB{js68k}`Z37x~!GO9yuS8p@Y_6og8dXWTIY3Md z_Q_sP{Wx_A81BDBLk+F|u(H0nyVTKc@}a;Irk}xD^>ss*V&zjvow5Dig6Hp6*J${0 z=PFYNRM64;hf`&Q(>W^W_D={+?1oXxmye>%@}4j@2Z1v4<@U@gQ0|CF5E~Ri|HfL^ zje1e$*J4=+@%QQ)Oq%H{YrFpEY@4a*5zG=f7UPCbbJ@TQEj$mC@Qr{t1avE20r;S9 zu3d5I*d3^huWWuzY{OTIj;&cqpF14@xQh4BNh_3FmPKRmC8cNgfa0u1w&$zJuhhf0 z(E9HpoFLqH1q;Hz%{7C%bvFR4zv^Ter^}Fkuf;|x;YBqR=Rre(<&M+FD^;Oh=GdW4 z&T)4TuP(SIM*EQdUPVtb)sUPYhYqO}{LPB+dO9!i#bOmdWp^%+op#%Or9vHS_jh>F z#hcKcM9Q}9`u3>>ea)Mj&*LfNv7SMkZ;#XT#SqUgo?j^j$Jr7E8tdjkEgo2?)yieg z4TB}FVdr@36A9xt7zSnLUm@i_qF`~kJ!7PHo5D)VywE7Au)*a?h?8>HE#t#6xu$== z{4i{7MeN7i0auEFD#%!|1b$;tz=Vb`EomlVhN2df-VSUlhS96d*4UiU#1U-(MC4$> z70;FKpzP{<)D#@DvG#O)ZtHNSf4fvuPJydii*+6XTE-*&m9HMCVWLeF{Yy9>WJ7jb z#quVi-p>E<5q@|RqW>7oq$@r z(#cj{S`u)7Kl`{)ck9?v@s`RfYt~+O4LMpKM{;89xYzml-k#>P&*1u2+@_y4ZE^7F zF3Dcqxx)vQj^C5b4zi1oU#r5caII$jCZVF?LSva(vXJ9z3 z%+_4udYXB(<1KhxjB=;s%+@6`4~^5&3#OJwa-IDu^vX*MZ# zM%>ImO~X(g9UhU2=qSDa(o=SZ9AsV^Qqo^|RX2UU1F&t=lTsL_I7w;RCdSn|f#gc< zyH(xF-%nMVTe~|f^Mw?CZ?syIA5(U?<4of$ET{$6*Zr%_M8c6&!$8m{y<(VF7k|jc zC{~zLtDaNWrH=L@Ik?S&+%RtSW4)$Jk%lofmeCa4u>F8jqOP!mz@I=ZX%MB`9~bu* zUeS%22aW%3H^up_v;=~@f3$FNYZbdSDCo0Bl zxsJAVX|9pu*~M{YjYfBKQel4wdlr~&QpyvyD=p-(ZSlYvKVERBj3+}V z-)3<@YciN$h+5h!RIqVlz+LLiyzBYeKqJ=e@?H^Gq4t|4%h|E_LDsCJ!bhvwq3KKJ zfkV^LH;Vg{jtsnn>-%*9I=_6sVWk7mdLAa{SCk~IOY?GK@9oH#rH32-FgbxzexK<5 zAt}OdsjUy6N1`P8a9^+y^NpI|PzUm|Bl2KO1?{$45A_vurz;V8$W=GJi z`B?@+X1=61m7VF%@b~EQ_ZH`uXS3~mbu1PKM>KsFL5mcT4|5-4OJYg-$f3UCCoHeb z4o}0}s~AD9bj-+hG(AHqYf$E)Y-XH;TcS3#p!{ZE6QAw8F~0EhBA3u?wUMhlK$(ZJ zKN|-oEHpf57hjl+SL_<`S4iUd-b=o|L}y3mkxP<%Qaqt%kX7}ivt33`D~ULjcJAsL zS`F5ho~))j`{|LFP#K|wSZwkvFlj3W(VE2RPvs>66mk+?5%&nm@0zX7cUL{Y<)55Rn zlxCc7$GB<6pmeVegpaoWn+^djnl>Fcp2bNlo^LMi?=RBirTM{T7pYh3?YzY&5FTK= z(MbW^T(3?v^M;qritj8PVa3xDB<)&CrBiR!=CG4}bvyS5_K!EzY>bZ(1W~jFRjF~Q z3v-$*Jyo15(qy77v8A=VIbdPp%D*{vZ+^C?ON?EwGb;NhDw1b79zweV-JjWp!W$0y8F=#uUr?nYp558J>05wFX1Bdx>uI z{?O)LWBIP2=g;P9BvGmsvU8Vbn|*_9kuwMc3^7lkC{dC>B&xe67BO@aaq2=hL8ojw z++03LC(*khDMlM8Sdj?+qJhUgnC!8+b$-w?61M)ckk`=j)?28{nHCE)a8u0)m}=w4 z$;c}?cPw4XP_h24za4yw{t9ZgP+$BBC1CRN{h;C2fL@M?anQis{vb_+tGsJUqqgtn zZV?-Ui4^>FS5mMx46Hz=C*7j^qCn1%JzcuI0?BWzpv~QwbfN1;&zB#QW0;ik^6>&K zNj_7hq^w^@gIDCo^pgh=HQEL|h)dI~WoXjhfJRHv)2G8rI&6%rYf63(P2$I9(eb%& ziRB9dMXeVfu}$Ui{W|JA)o!rP3Mb;=of>1ua-az@*zDSW(#WmrOSDTf%6ltak7<62 z^5M@~JXxAyw9|$(buV=-BQ?pJ(}eF8^f~ zcfTjOAJl1&$ZL-?WxH=pl+fEf{pbH1Hk74uX=8Ks}{sUswv=+6IOizSK z*L$?!*)-`L!Gj-Yitg@Ezrhldpn7{|0BT~ISB6ab@H%`6rPO(nB^8(cz{gizn<(9>8%r7eiaFD`(2dHfSf8psJ%LSE;RC46e0eoRvd zOlsXX(BZ{X=x;nHj-+$izT!(n6$)PjvnjS%gMof1j)(90RVcrJJIw&Z*69A*c?P@b z&LAh^nKI%*}CRUDeYh8Qg+tti%`^Da|wNfa`WoIUTBr{ z$>KbuT}rF}+nC^%LwG-Mm?pQ|;}F2l4EwjAiJ1H>Dx?Q>-7p!TLQcF*wGdXCR(@^p zilXRr0)qr3;@pq<6NU;T`GTqQ&Ft(?*yR%Pd|@jz;42S-yR-M)LBs>OW@b z0)Ozrzx+V~jJ>EMq^%?Sp$GOZzlt^B8T2eBEX?b=6K_X#tKiw6_%tg?qWz=Gtrw^5 zHpZI-fL6*Px7~*Uckmp)6#Z?mkt#LRW9_sqrKFrEqv_)Td?Ati!KArjh&?@>%AQq=3Fy{I&ll<=O(+P_`oKu|k*&D+)D|US4sI=B z548V#ylG_TL`N4tB_fa4aU2Y`C;AVBv^>cW$PGF6JYf+}s}<|)`_q9{eOR07Ne|63fWLTSvU(VSL*2Owk-3kPzW!J)#tCl9PxI`@1D9ZT` zRT;WqBsLz{#xn?2lEO`%@TM@1W z1h8?y_1)e+fV1ny!U;KN|KJtv9NEfI>&VWzK2f%vB(kv-V(5y@TaEE~`-rGxucZ79 z1u77Qe-8UiL!}VG+}ZKo!A&sW*U;qngZV^d8k(JC!<|}Yu~$cXpS#k%*G1QxMMnJ< zYv_CUDHJ0ZfZuI@Ek$J4>agWde=wva9c3G6!1C`itG0aM%BlyE0r>8s%Lxm90EK;2 zf$TBs#TLWvW8?JiCmeu!6e-ia(W?_qIM-5;yq~&uQjC5poIp;U20JKeg7i_hx~J|D z@MV?{&Qxb<1<;GPi*Vy9oyi=#MvKs>{nWWz!DI&&dTyG0IuDL^WEOP(0foaX+|-^~ zZ8^b9TN;R(o>|`Z^PlDexs8aK9-~F0aKh*bHH~MUmQFz_A+6>r)Z8FWsn^^ZiyH?Z znQO1BxX@e8b`P2xNZybu)gHJFy?QQm&$`!57Lij!$`DukMG%}f24n>A=ZIR8kiI$C zS&N#n_Nx)v^coAaDtUzgL)Sqg|IbRB#LTDLXtLL%}Dq=JUJjmnrt44k-PvX zmy;ny=c3PGm9_3WxE=5Q+?QVzXBP>`D?*Q#2H`CAWhxqp+Zd=pT!0;e zDpcLEI~#otXoH#dAb8Xs3>G17=6;V)Fh{91Yc3;V95i=0S#%>9VwoEq(6@^O5j2q_ z?ow@u-83<|orT*_>w!_c7%BF?=e5{H-R8DI2M*{U&$ETXLls`Q2|N*iz6*gihCnx; zmPAJ}0|nm6qR{uXEFVPUIK6Dpk?ysF8~#EhBhB-kSA})q>UN9AQ(}S0ZH1B-;~W{9 zHM-;XjpGXIk7aqEfMV-1BXx(F+86KHc{7UGpe>+Z#)c8)>bYCD$%ov4D#6N5ykA}x zfcMO6y$$@$$bPB|+cMBUyWbfh#56q}fR>jj9TN{UgJUQ&l48E`nN+)WH~~H>W-5MJ zISdO>)mx^!jPrS9JupH{!*iNe@1trkY1`3-{9yTynit@A)p6CWPa}SUCAlS=AS2cN zF1GYRA1GtWIHrHSgYk0&Ai}1!gOvat zc?)<%ywCyYCJ_y!Q`_r9=oUwF1=TYo`^Du?CA29pLKyO|?@n|O~0 zgO@QJc*G*Ugi)OPJu&(RjisAgPW7Sn3N)?OgcI&lH&~0iD~H)G6CSU8cuX4vS0~xi zb@Mr#_1n;RJvEJnrzhxX?(B!kQ~=N38?!b=9|&IZd`bc(R_>02n|HR4nuc=NgX-%wE?#MzZq4+2WK4SSloK=hNyw>LnaYwTMXwVqsdd|MljH_dB!2(9rOZ zWuS42g_9Rv)(I@XOQ`NUOImgKdf(V zXF`u_f#@sgMny+WwP?;S=24Iea+HTB742kzNe*C=#~I|O|0k1x9H`Uwy5PDRA1ziJ zapes1DOgy=oTDH`zkfbajdTLL$po{pzy(NpFZ=#SdNwj$0d8lI7xegFARca3ES%Op zG|&<`BvugoRLTFLS-O0r{khK&vfBPHj1`BUF=5+p6uSxDR1O;OLkAgc9GeidPdG5~lJ+3D`(h+x zXP8T-EI;isbHFfGQXRbbppg^j^`VK4n5BR&Q`Yq{Eg(etE*Qq@=^o+CmcQmdhZF?0 zHd4n7=sQj-6R)Ht9;mOTaZ$Y06g~#hEPGqP01>=$;Vr}5*R|>QXm=@PE(1tCNZurX$OK2fHMqtGhvTGa?BRO-YvIocp23BSi7}$;e<8s-`U(P z-xu092U0oU`SQt9n%3Ig0y~M7$Tn0f7J=W5cqmDKXp1o8LMbY2SsfLm3jhGWgqwWt^C=`KOSe&>a@+c zzPm7*?{0zx(0e-0_%#^D3IT82Hj7R015;1aJboSDh8B&hDt*L%j#rw7T0vwf9i9N8 z%BUxu-}mmPYFrQ>E?}Sf5j=>Dk@ja51lt4dSW8Q@D=vTa=a#^1u$#`OOSK<|zoEzd zYY#AV@n5fQLmND-f@I%?nB-MJINSUx>ji$@PULK}RH<&=()SRe)+fh|3WxpF;)k+< z;FrC+qkcs{5RE(iE`pyTSv$2>X@ohriPlCy`pi;{wK*RSH062ZQSh^(ImRyq*pbDI zz7S|<-E Date: Wed, 11 Oct 2023 11:26:08 -0400 Subject: [PATCH 02/22] Updating co-authors Co-authored-by: Henry Bae Co-authored-by: Sophia Cho --- frameworks.qmd | 2 -- 1 file changed, 2 deletions(-) diff --git a/frameworks.qmd b/frameworks.qmd index 271c3fd7..c8acb44d 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -1782,6 +1782,4 @@ intended models, use cases, and evaluating options against key metrics will guide developers towards picking the ideal framework for their embedded ML application. -For Future (Federated Learning) - [^1]: [[https://tvm.apache.org/2021/12/15/tvm-unity]{.underline}](https://tvm.apache.org/2021/12/15/tvm-unity) From d3dab82004d00375ffb20f79be93c5d49daf5da7 Mon Sep 17 00:00:00 2001 From: Divya Amirtharaj Date: Wed, 11 Oct 2023 11:32:21 -0400 Subject: [PATCH 03/22] Adding co-authors --- frameworks.qmd | 2 ++ 1 file changed, 2 insertions(+) diff --git a/frameworks.qmd b/frameworks.qmd index c8acb44d..7035db70 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -1782,4 +1782,6 @@ intended models, use cases, and evaluating options against key metrics will guide developers towards picking the ideal framework for their embedded ML application. + + [^1]: [[https://tvm.apache.org/2021/12/15/tvm-unity]{.underline}](https://tvm.apache.org/2021/12/15/tvm-unity) From f2edd8e64d4585f2a2a3789cbf89cbf8ce5398dd Mon Sep 17 00:00:00 2001 From: Divya Date: Wed, 11 Oct 2023 11:46:31 -0400 Subject: [PATCH 04/22] Updating co-authors Co-authored-by: Henry Bae Co-authored-by: Sophia Cho Co-authored-by: Matthew Steward Co-authored-by: Vijay Janapa Reddi Co-authored-by: Emeka Ezike --- frameworks.qmd | 2 -- 1 file changed, 2 deletions(-) diff --git a/frameworks.qmd b/frameworks.qmd index 7035db70..c8acb44d 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -1782,6 +1782,4 @@ intended models, use cases, and evaluating options against key metrics will guide developers towards picking the ideal framework for their embedded ML application. - - [^1]: [[https://tvm.apache.org/2021/12/15/tvm-unity]{.underline}](https://tvm.apache.org/2021/12/15/tvm-unity) From 0c40e99c0a271b4148c13ab2d66da6dd447037db Mon Sep 17 00:00:00 2001 From: Divya Date: Wed, 11 Oct 2023 11:46:31 -0400 Subject: [PATCH 05/22] Updating co-authors Co-authored-by: Henry Bae Co-authored-by: Sophia Cho Co-authored-by: Matthew Stewart Co-authored-by: Vijay Janapa Reddi Co-authored-by: Emeka Ezike --- frameworks.qmd | 2 -- 1 file changed, 2 deletions(-) diff --git a/frameworks.qmd b/frameworks.qmd index 7035db70..c8acb44d 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -1782,6 +1782,4 @@ intended models, use cases, and evaluating options against key metrics will guide developers towards picking the ideal framework for their embedded ML application. - - [^1]: [[https://tvm.apache.org/2021/12/15/tvm-unity]{.underline}](https://tvm.apache.org/2021/12/15/tvm-unity) From a809626d321b8914ab9c9f787b8906f6cd59d27e Mon Sep 17 00:00:00 2001 From: Henry Bae <69275685+BaeHenryS@users.noreply.github.com> Date: Thu, 12 Oct 2023 01:39:30 -0400 Subject: [PATCH 06/22] Test Commit --- frameworks.qmd | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/frameworks.qmd b/frameworks.qmd index c8acb44d..e9610efc 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -1,6 +1,6 @@ # AI Frameworks -Learning Objectives +Learning Objectives - The evolution, core components, and advanced features of ML > frameworks From 32b413cc27f8359e7d10a61fcc6c4410b27afa2f Mon Sep 17 00:00:00 2001 From: Divya Date: Mon, 16 Oct 2023 11:08:21 -0400 Subject: [PATCH 07/22] Added more intext citations --- frameworks.qmd | 18 +++++++++++++----- 1 file changed, 13 insertions(+), 5 deletions(-) diff --git a/frameworks.qmd b/frameworks.qmd index e9610efc..b24cf2e7 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -251,13 +251,16 @@ package. > validation, preprocessing, model training, validation, and > serving. -TensorFlow was developed to address the limitations of DistBelief---the +Source: +[[https://www.tensorflow.org/learn]{.underline}](https://www.tensorflow.org/learn) + +TensorFlow was developed to address the limitations of DistBelief[^2]---the framework in use at Google from 2011 to 2015---by providing flexibility along three axes: 1) defining new layers, 2) refining training algorithms, and 3) defining new training algorithms. To understand what limitations in DistBelief led to the development of TensorFlow, we will first give a brief overview of the Parameter Server Architecture that -DistBelief employed. +DistBelief employed[^3]. The Parameter Server (PS) architecture is a popular design for distributing the training of machine learning models, especially deep @@ -274,7 +277,7 @@ maintain and manage this state across the training process. **Computation**: The worker processes, which could be run in parallel, were stateless and purely computational, processing data and computing -gradients without maintaining any state or long-term memory. +gradients without maintaining any state or long-term memory[^4]. DistBelief and its architecture defined above were crucial in enabling distributed deep learning at Google but also introduced limitations that @@ -293,7 +296,7 @@ ones, complicating the management and synchronization tasks of the parameter servers. This made it harder to implement models outside the neural framework or models that required dynamic computation graphs. -TensorFlow was designed to be a more general computation framework where +TensorFlow was designed to be a more general computation framework[^2] where the computation is expressed as a data flow graph. This allows for a wider variety of machine learning models and algorithms outside of just neural networks, and provides flexibility in refining models. @@ -344,7 +347,7 @@ development life-cycle: data preparation, model building, deployment, as well as responsible AI. Additionally, one of TensorFlow's biggest advantages is its integration -with Keras. Keras is another ML framework that was built to be extremely +with Keras, though as we will cover in the next section, Pytorch recently also added a Keras integration. Keras is another ML framework that was built to be extremely user-friendly and as a result has a high level of abstraction. We will cover Keras in more depth later in this chapter, but when discussing its integration with TensorFlow, the most important thing to note is that it @@ -1783,3 +1786,8 @@ will guide developers towards picking the ideal framework for their embedded ML application. [^1]: [[https://tvm.apache.org/2021/12/15/tvm-unity]{.underline}](https://tvm.apache.org/2021/12/15/tvm-unity) +[^2]: [[https://arxiv.org/pdf/1603.04467.pdf]{.underline}](https://arxiv.org/pdf/1603.04467.pdf) +[^3]: [[https://storage.googleapis.com/pub-tools-public-publication-data/pdf/40565.pdf]{.underline}](https://storage.googleapis.com/pub-tools-public-publication-data/pdf/40565.pdf) +[^4]: [[https://proceedings.neurips.cc/paper_files/paper/2014/file/1ff1de774005f8da13f42943881c655f-Paper.pdf +]{.underline}](https://proceedings.neurips.cc/paper_files/paper/2014/file/1ff1de774005f8da13f42943881c655f-Paper.pdf +) From 4c9bec2a4641865a4b4ead0add1f2c343b4ab88c Mon Sep 17 00:00:00 2001 From: Henry Bae <69275685+BaeHenryS@users.noreply.github.com> Date: Mon, 23 Oct 2023 02:20:45 -0400 Subject: [PATCH 08/22] Framework Section Overhaul, Organization - Organized/Created Images - Addressed all comments on Basic ML Framework and Advanced ML Framework Section - Fixed External Image Issues - Fixed Section Headers - Addressed comments from @mpstewart1. --- frameworks.qmd | 143 +++++++++++++++++--------------- images_ml_frameworks/image1.png | Bin 34481 -> 43754 bytes images_ml_frameworks/image2.png | Bin 71931 -> 64462 bytes references.bib | 10 +++ 4 files changed, 86 insertions(+), 67 deletions(-) diff --git a/frameworks.qmd b/frameworks.qmd index b24cf2e7..10274413 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -117,7 +117,7 @@ construct on-the-fly for more iterative development. But around 2016, frameworks began adopting dynamic graphs like PyTorch and TensorFlow 2.0 which can construct graphs on-the-fly. This provides greater flexibility for model development. We will discuss these concepts and details later -on in Section XXX. +on in AI Training Section. The development of these frameworks facilitated an explosion in model size and complexity over time---from early multilayer perceptrons and @@ -450,9 +450,6 @@ tensors, with 1D and 2D dimensions respectively. ![](images_ml_frameworks/image2.png){width="3.9791666666666665in" height="1.9672287839020122in"} -Source: -[[https://medium.com/mlait/tensors-representation-of-data-in-neural-networks-bbe8a711b93b]{.underline}](https://medium.com/mlait/tensors-representation-of-data-in-neural-networks-bbe8a711b93b) - Defining formally, in machine learning, tensors are a multi-dimensional array of numbers. The number of dimensions defines the rank of the tensor. As a generalization of linear algebra, the study of tensors is @@ -466,8 +463,6 @@ very critical in practice, as multiplication of abstract representations of higher dimensional tensors are often completed by first converting them into matrices for multiplication. -(APMTH 210 Lecture Notes, CS242 Lecture Notes) - Tensors offer a flexible data structure with its ability to represent data in higher dimensions. For example, to represent color image data, for each of the pixel values (in 2 dimensions), one needs the color @@ -477,10 +472,9 @@ within it representing a certain color value in the certain location of the image. Extending even further, if we wanted to store a series of images, we can simply extend the dimensions such that the new dimension (to create a 4-dimensional tensor) represents the different images that -we have. This is exactly what the famous MNIST dataset does -(https://www.tensorflow.org/datasets/community_catalog/huggingface/mnist), +we have. This is exactly what the famous MNIST dataset does, loading a single 4-dimensional tensor when one calls to load the -dataset, allowing a compact representation of all the data in one place. +dataset, allowing a compact representation of all the data in one place. [^5] ### Computational graphs @@ -496,14 +490,12 @@ edges represent data dependencies between them. For example, a node might represent a matrix multiplication operation, taking two input matrices (or tensors) and producing an output matrix (or tensor). To visualize this, consider the simple example below. The -directed acyclic graph above computes \$z = x \\times y\$, where each of +directed acyclic graph above computes $z = x \times y$, where each of the variables are just numbers. ![](images_ml_frameworks/image1.png){width="1.9947922134733158in" height="1.6791174540682414in"} -https://cedar.buffalo.edu/\~srihari/CSE676/6.5.1%20Computational%20Graphs.pdf - Underneath the hood, the computational graphs represent abstractions for common layers like convolutional, pooling, recurrent, and dense layers, with data including activations, weights, biases, are represented in @@ -520,16 +512,16 @@ tensors. Note that a layer operates on tensors as inputs and outputs and the layer itself is not a tensor. Some key differences: - Layers contain states like weights and biases. Tensors are - > stateless, just holding data. + stateless, just holding data. - Layers can modify internal state during training. Tensors are - > immutable/read-only. + immutable/read-only. - Layers are higher level abstractions. Tensors are lower level, - > directly representing data and math operations. + directly representing data and math operations. - Layers define fixed computation patterns. Tensors flow between - > layers during execution. + layers during execution. - Layers are used indirectly when building models. Tensors flow > between layers during execution. @@ -632,9 +624,9 @@ graph require re-declaring the full model. For example: -x = tf.placeholder(tf.float32) +``x = tf.placeholder(tf.float32)`` -y = tf.matmul(x, weights) + biases +``y = tf.matmul(x, weights) + biases`` The model is defined separately from execution, like building a blueprint. For TensorFlow 1.x, this is done using tf.Graph(). All ops @@ -652,9 +644,9 @@ PyTorch uses dynamic graphs, building the graph on-the-fly as execution happens. For example, consider the following code snippet, where the graph is built as the execution is taking place: -x = torch.randn(4,784) +``x = torch.randn(4,784)`` -y = torch.matmul(x, weights) + biases +``y = torch.matmul(x, weights) + biases`` In the above example, there are no separate compile/build/run phases. Ops define and execute immediately. With dynamic graphs, definition is @@ -673,10 +665,10 @@ interactivity. While dynamic execution offers flexibility and ease of use, it may have performance overhead. Here is a table comparing the pros and cons of static vs dynamic execution graphs: -| Execution Graph | Pros | Cons | -|-------------------------------|------------------------------------------------------------------------------------------------|-----------------------------------------------------------------| -| Static (Declare-then-execute) | - Enable graph optimizations by seeing full model ahead of time
- Can export and deploy frozen graphs
Can export and deploy frozen graphs
Graph is packaged independently of code | Less flexible for research and iteration
Changes require rebuilding graph
Execution has separate compile and run phases | +| Dynamic (Define-by-run) | Intuitive imperative style like Python code
Interleave graph build with execution
Easy to modify graphs
Debugging seamlessly fits workflow | Harder to optimize without full graph
Possible slowdowns from graph building during execution
Can require more memory | ### Data Pipeline Tools @@ -687,7 +679,7 @@ overlooked as one of the core functionalities. Many modern AI frameworks provide specialized pipelines to ingest, process, and augment datasets for model training. -Data Loaders +#### Data Loaders At the core of these pipelines are data loaders, which handle reading examples from storage formats like CSV files or image folders. Reading @@ -724,7 +716,7 @@ processing. Prefetching, on the other hand, involves preloading subsequent batches, ensuring that the model never idles waiting for data. -Data Augmentation +### Data Augmentation Besides loading, data augmentation expands datasets synthetically. Augmentations apply random transformations like flipping, cropping, @@ -763,7 +755,17 @@ the computer the "objective" for it to aim to. Commonly used loss functions are Mean Squared Error (MSE) for regression tasks and Cross-Entropy Loss for classification tasks. -Once the loss is computed, we need methods to adjust the model\'s +To demonstrate some of the loss functions, imagine that you have a set of inputs and the corresponding outputs, $Y_n$ that denotes the output of $n$'th value. The inputs are fed into the model, and the model outputs a prediction, which we can call $\hat{Y_n}$. With the predicted value and the real value, we can for example use the MSE to calculate the loss function: + +$$MSE = \frac{1}{N}\sum_{n=1}^{N}(Y_n - \hat{Y_n})^2$$ + +If the problem is a classification problem, we do not want to use the MSE, since the distance between the predicted value and the real value does not have significant meaning. For example, if one wants to recognize handwritten models, while 9 is further away from 2, it does not mean that the model is more wrong by making the prediction. Therefore, we use the cross-entropy loss function, which is defined as: + +$$Cross-Entropy = -\sum_{n=1}^{N}Y_n\log(\hat{Y_n})$$ + + + +Once the loss like above is computed, we need methods to adjust the model\'s parameters to reduce this loss or error during the training process. To do so, current frameworks use a gradient based approach, where it computes how much changes tuning the weights in a certain way changes @@ -775,7 +777,7 @@ more details about this will come in the AI training sections. Modern frameworks come equipped with efficient implementations of several optimization algorithms, many of which are variants of gradient descent algorithms with stochastic methods and adaptive learning rates. More -information can be found in the AI training section. +information with clear examples can be found in the AI training section. Last but not least, overly complex models tend to overfit, meaning they perform well on the training data but fail to generalize to new, unseen @@ -784,6 +786,8 @@ employed to penalize model complexity and encourage it to learn simpler patterns. Dropout for instance randomly sets a fraction of input units to 0 at each update during training, which helps prevent overfitting. +However, there are cases where the problem is more complex than what the model can represent, and this may result in underfitting. Therefore, choosing the right model architecture is also a critical step in the training process. Further heuristics and techniques are discussed in the AI training section. + Frameworks also provide efficient implementations of gradient descent, Adagrad, Adadelta, and Adam. Adding regularization like dropout and L1/L2 penalties prevents overfitting during training. Batch @@ -833,7 +837,7 @@ processing methods, and utilities such as checkpointing and early stopping. These resources manage the complex aspects of performance, enabling practitioners to zero in on model development and training. As a result, developers experience both speed and ease when utilizing the -capabilities of neural networks. +capabilities of neural networks. [^6] ### Validation and Analysis @@ -881,10 +885,8 @@ Visualization tools provide insight into models: ![](images_ml_frameworks/image7.png){width="3.5052088801399823in" height="2.3043503937007874in"} -Example of a loss curve from TensorBoard +Example of a loss curve from TensorBoard [^7] -Source: -[[https://www.tensorflow.org/tensorboard/scalars_and_keras]{.underline}](https://www.tensorflow.org/tensorboard/scalars_and_keras) - Activation grids - Illustrate features learned by convolutional > filters. @@ -944,9 +946,7 @@ The use of hardware accelerators began with [[AlexNet]{.underline}](https://proceedings.neurips.cc/paper_files/paper/2012/file/c399862d3b9d6b76c8436e924a68c45b-Paper.pdf), which paved the way for future works to utilize GPUs as hardware accelerators for training computer vision models. GPUs, or Graphics -Processing Units, - -excel in handling a large number of computations at once, making them +Processing Units, excel in handling a large number of computations at once, making them ideal for the matrix operations that are central to neural network training. Their architecture, designed for rendering graphics, turns out to be perfect for the kind of mathematical operations required in @@ -1086,7 +1086,7 @@ workflow for experts. ### Advanced Learning Methods -##### Transfer Learning +#### Transfer Learning Transfer learning is the practice of using knowledge gained from a pretrained model to train and improve performance of a model that is for @@ -1102,7 +1102,7 @@ Forgetting"]{.underline}](https://browse.arxiv.org/pdf/1606.09282.pdf) paper aims to address these challenges and have been implemented in modern machine learning platforms. -##### Federated Learning +#### Federated Learning Consider the problem of labeling items that are present in a photo from personal devices. One may consider moving the image data from the @@ -1128,16 +1128,20 @@ the set of images that are on personal devices), without the need to transfer a large amount of potentially sensitive data. However, federated learning also comes with a series of challenges. -Training in machine learning usually assumes the data to be identically -and independently distributed. However, the nature of federated learning -makes it so that this is not necessarily the case. In our example, -devices may have similar photos from being at the same location, and the -types of the cameras that each of the devices have may be different. -There are many different techniques to compensate for this, such as + +In many real-world situations, data collected from devices may not come with suitable labels. This issue is compounded by the fact that users, who are often the primary source of data, can be unreliable. This unreliability means that even when data is labeled, there's no guarantee of its accuracy or relevance. Furthermore, each user's data is unique, resulting in a significant variance in the data generated by different users. This non-IID nature of data, coupled with the unbalanced data production where some users generate more data than others, can adversely impact the performance of the global model. Researchers have worked to compensate for this, such as by adding a proximal term to achieve a balance between the local and global model, and adding a frozen [[global hypersphere classifier]{.underline}](https://arxiv.org/abs/2207.09413). +There are additional challenges associated with federated learning. The number of mobile device owners can far exceed the average number of training samples on each device, leading to substantial communication overhead. This issue is particularly pronounced in the context of mobile networks, which are often used for such communication and can be unstable. This instability can result in delayed or failed transmission of model updates, thereby affecting the overall training process. + +The heterogeneity of device resources is another hurdle. Devices participating in Federated Learning can have varying computational powers and memory capacities. This diversity makes it challenging to design algorithms that are efficient across all devices. Privacy and security issues are not a guarantee for federated learning. Techniques such as inversion gradient attacks can be used to extract information about the training data from the model parameters. Despite these challenges, the large amount of potential benefits continue to make it a popular research area. Open source programs such as [[Flower]{.underline}](https://flower.dev/) have been developed to make it simpler to implement federated learning with a variety of machine learning frameworks. + + + + + ## Framework Specialization Thus far, we have talked about ML frameworks generally. However, @@ -1306,7 +1310,7 @@ bringing AI to IoT (Internet of Things) devices, robotics, and other edge computing platforms and they are designed to work where computational resources, memory, and power consumption are limited. -#### Challenges +### Challenges While embedded systems present an enormous opportunity for deploying machine learning to enable intelligent capabilities at the edge, these @@ -1319,7 +1323,7 @@ server clusters with abundant resources do not directly translate to embedded systems. This section uncovers some of the challenges and opportunities for embedded systems and ML frameworks. -##### Fragmented Ecosystem +#### Fragmented Ecosystem The lack of a unified ML framework led to a highly fragmented ecosystem. Engineers at companies like STMicroelectronics, NXP Semiconductors, and @@ -1329,7 +1333,7 @@ extensive manual optimization for each low-level hardware platform. This made porting models extremely difficult, requiring redevelopment for new Arm, RISC-V or proprietary architectures. -##### Disparate Hardware Needs +#### Disparate Hardware Needs Without a shared framework, there was no standard way to assess hardware\'s capabilities. Vendors like Intel, Qualcomm and NVIDIA @@ -1340,7 +1344,7 @@ software optimizations were responsible. A standard framework was needed so vendors could evaluate their hardware\'s capabilities in a fair, reproducible way. -##### Lack of Portability +#### Lack of Portability Adapting models trained in common frameworks like TensorFlow or PyTorch to run efficiently on microcontrollers was very challenging without @@ -1349,7 +1353,7 @@ models to run on specialized DSPs from companies like CEVA or low-power Arm M-series cores. There were no turnkey tools enabling portable deployment across different architectures. -##### Incomplete Infrastructure +#### Incomplete Infrastructure The infrastructure to support key model development workflows was lacking. There was minimal support for compression techniques to fit @@ -1360,7 +1364,7 @@ functionality like on-device debugging, metrics, and performance profiling was absent. These gaps increased the cost and difficulty of embedded ML development. -##### No Standard Benchmark +#### No Standard Benchmark Without unified benchmarks, there was no standard way to assess and compare the capabilities of different hardware platforms from vendors @@ -1369,7 +1373,7 @@ proprietary benchmarks tailored to showcased strengths of particular chips. This made it impossible to objectively measure hardware improvements in a fair, neutral manner. -##### Minimal Real-World Testing +#### Minimal Real-World Testing Much of the benchmarks relied on synthetic data. Rigorously testing models on real-world embedded applications was difficult without @@ -1457,16 +1461,16 @@ An interpreter-based approach offers several benefits over code generation for machine learning inference on embedded devices: - Flexibility: Models can be updated in the field without recompiling - > the entire application. + the entire application. - Portability: The interpreter can be used to execute models on - > different target platforms without porting the code. + different target platforms without porting the code. - Memory efficiency: The interpreter can share code across multiple - > models, reducing memory usage. + models, reducing memory usage. - Ease of development: Interpreters are easier to develop and maintain - > than code generators. + than code generators. TensorFlow Lite Micro is a powerful and flexible framework for machine learning inference on embedded devices. Its interpreter-based approach @@ -1586,7 +1590,7 @@ goal is to balance model complexity, hardware limitations, and software integration to design a tailored ML pipeline for embedded and edge devices. -#### ![](images_ml_frameworks/image4.png){width="6.5in" height="3.1666666666666665in"}Model +### ![](images_ml_frameworks/image4.png){width="6.5in" height="3.1666666666666665in"}Model TensorFlow supports significantly more ops than TensorFlow Lite and TensorFlow Lite Micro as it is typically used for research or cloud @@ -1602,7 +1606,7 @@ representation with available lower precision operations. ![](images_ml_frameworks/image5.png){width="6.5in" height="2.0833333333333335in"} -#### Software +### Software TensorFlow Lite Micro does not have OS support, while TensorFlow and TensorFlow Lite do, in order to reduce memory overhead, make startup @@ -1616,7 +1620,7 @@ schedule code to different accelerators, whereas TensorFlow Lite Micro does not, as embedded systems tend not to have a rich array of specialized accelerators. -#### Hardware![](images_ml_frameworks/image3.png){width="6.5in" height="2.2083333333333335in"} +### Hardware![](images_ml_frameworks/image3.png){width="6.5in" height="2.2083333333333335in"} TensorFlow Lite and TensorFlow Lite Micro have significantly smaller base binary sizes and base memory footprints compared to TensorFlow. For @@ -1640,12 +1644,12 @@ optimization tools. By understanding these factors, you can make informed decisions and maximize the potential of your machine learning initiatives. -#### Other Factors +### Other Factors When evaluating AI frameworks for embedded systems, several other key factors beyond models, hardware, and software should be considered. -##### Performance +#### Performance Performance is critical in embedded systems where computational resources are limited. Evaluate the framework\'s ability to optimize @@ -1653,7 +1657,7 @@ model inference for embedded hardware. Factors such as model quantization and hardware acceleration support play a crucial role in achieving efficient inference. -##### Scalability +#### Scalability Scalability is essential when considering the potential growth of an embedded AI project. The framework should support the deployment of @@ -1661,21 +1665,21 @@ models on a variety of embedded devices, from microcontrollers to more powerful processors. It should also handle both small-scale and large-scale deployments seamlessly. -##### Integration with Data Engineering Tools +#### Integration with Data Engineering Tools Data engineering tools are essential for data preprocessing and pipeline management. An ideal AI framework for embedded systems should seamlessly integrate with these tools, allowing for efficient data ingestion, transformation, and model training. -##### Integration with Model Optimization Tools +#### Integration with Model Optimization Tools Model optimization is crucial to ensure that AI models are well-suited for embedded deployment. Evaluate whether the framework integrates with model optimization tools, such as TensorFlow Lite Converter or ONNX Runtime, to facilitate model quantization and size reduction. -##### Ease of Use +#### Ease of Use The ease of use of an AI framework significantly impacts development efficiency. A framework with a user-friendly interface and clear @@ -1686,7 +1690,7 @@ implementation details. This factor is incredibly important for embedded systems, which have less features that typical developers might be accustomed to. -##### Community Support +#### Community Support Community support plays another essential factor. Frameworks with active and engaged communities often have well-maintained codebases, receive @@ -1703,7 +1707,7 @@ most community support. ## Future Trends in ML Frameworks -#### Decomposition +### Decomposition Currently, the ML system stack consists of four abstractions, namely (1) computational graphs, (2) tensor programs, (3) libraries and runtimes, @@ -1720,7 +1724,7 @@ interactions between the different abstraction levels (as well as the people behind them, such as ML scientists, ML engineers, and hardware engineers) and co-optimize decisions in all four abstraction levels.[^1] -#### High-Performance Compilers & Libraries +### High-Performance Compilers & Libraries As ML frameworks further develop, high-performance compilers and libraries will continue to emerge. Some current examples include @@ -1733,7 +1737,7 @@ Nvidia's [[TensorRT]{.underline}](https://developer.nvidia.com/tensorrt), which accelerates and optimizes inference. -#### ML for ML Frameworks +### ML for ML Frameworks We can also use ML to improve ML frameworks in the future. Some current uses of ML for ML frameworks include: @@ -1791,3 +1795,8 @@ embedded ML application. [^4]: [[https://proceedings.neurips.cc/paper_files/paper/2014/file/1ff1de774005f8da13f42943881c655f-Paper.pdf ]{.underline}](https://proceedings.neurips.cc/paper_files/paper/2014/file/1ff1de774005f8da13f42943881c655f-Paper.pdf ) +[^5]: [[https://www.tensorflow.org/datasets/catalog/mnist]{.underline}](https://www.tensorflow.org/datasets/catalog/mnist) + +[^6]: [[https://www.geeksforgeeks.org/batch-processing-operating-system/#]{.underline}](https://www.geeksforgeeks.org/batch-processing-operating-system/#) + +[^7]: [[https://www.tensorflow.org/tensorboard/scalars_and_keras]{.underline}](https://www.tensorflow.org/tensorboard/scalars_and_keras) \ No newline at end of file diff --git a/images_ml_frameworks/image1.png b/images_ml_frameworks/image1.png index afb2412121e41d550872c13d0d0502ca4e63f026..ba76db9c4d906e8c49a45e73cea25c70e0210196 100644 GIT binary patch literal 43754 zcmeFYWmJ?~*f2aZ4my+!AxICQpa>XrjsaM-N|#DWcgKK$iiBdIq@tiA4bqJwDuWWz zWiZkp-QRT|&*7ZsdDr{r`~A8Wvb}d*UHkH;mWCQV4F?Shg`z)m`jid|g(0I*JNj@t z;fr>dA`XS3Q?@;MQtQmglLReSCo5ZfOBCw#&HH-0^mUt9la21i#$IP463_lpkJUgO z;44J6MiLJ4QRC0vCZ%2$kEF>qIb?oPPlZLkIl9Y$#n0@}3*qt0wb>PC&?Zawr<<3W zleRY9MwG3|Nu^7>Q5=mvOlOo{Q4>ier#X088@TR2ekZ6%#S)B>b4Fj_sn)8Cj*7zM z(iH}YPiOBqHDKy;&u*h&tD04mlpD){Vxo0PPmMK0&7x5=2dhpQVo};;qw|OEzCZGs zs7ABnx_1?qwTR1oF6)D{JYnjW6?T2Qgld1GkSxK8THW{lSma8#R_Fk2UV%P=X-8`9 z^{(nv^9s~Ax*HBpqOg}sl|<%tB~4V@TJ65}+q*kdUzSZpo0fIY+21~KTP^PE(>k%@ zoHG{mB9dwxJmM+GKJ_+N%-(T5A6BB4$j-eZPpX0U6ediuL3p1UYx1)~wW#GVuJ~iA zyYdQx*|@0l7#p5^TmL1zC*{#Tnx7}=^l)-#E;OiL*XRAVQdbv?z2B@ly6k;lyin=f z2aC+*0)>}#Ienc02Bu6#&$r~%-d`NOGJiL*f#t>lyb){U-88=RlM@%>AL3u;vuoe2 zzsZw(=fk(3vqmSl#91P!4lLw-J7jB9@uBKq8C3c1?>?!na+|KiaP)X+7yGn`e18tvo zOy5&{61?jfffGkcyT!iu&eLCv^>!FxbkGSwY*^^`5SGdlT68Npp|5|B;qqB3-`-;M z8KQ2WA~5}^_j`2tp7&>?quzAvUz986%zxH<&sp4G+VTqw-XQ1lF+ab2bqCKQjvWsg z@RtuCN!)w3u%6x}sE;isq?4#`;=($mcsOQy@w(9U-E}AHBu@wpf0g-SKdPx6jZM;0 zzHmSC6R#V|=%hM>0&XdEOpCpS@*w9zhbFRbt$UDqIpKvtX3|M zXF@=ufnA2#>$->xmuB6{GS*+N3U3za)HptSX`N;hHT|}l(FQ9JvZFSlX5Vf8tIYeE z&QkBWo^?I+x~{Xh2jc+a%~Og`>A%oUy}fW*%!S5<&xPAXdR8VcbooY?8vh--)4K(g zb?9!Y9pt}%_iL=@-RL`Fab>4@+uJ+NcNlk^=xA+kxK_|{u-)?o_r>zpix(!38N}JE zEb=wRQvGQ9@%D%453AoQu52sER3a~*oJ|wdaXnBVoO!_Gp6_CkuOv#LMoZ zcn^E`61h^jpK|5yJs!RuMIQYgaUMdWm&SO$Ot#n0zg{TuUGa@vSR6I!T$5<<#`)cF zejmc;aZ-%ICr9~$vX&m}o3 zWf{MkaO=EYJkPNJa=T^zqm@q`pDtPDJi230^5~O&oI!cFwHxQ0;9U1edGoY^)**dX z16GkkhKGDb)kW0|HHuyrRXC`;xoq%LKPzAE%ECa;JGr;M$sVB#$2bExv(;~^XEmD` zJ$4A5#P(k6&GbAa|G*@F{jsn1wXg z%V_(?%Z-MOyEJ??v^5HnI1){F7jq>``H#*VtFM;zlcC5vxm10BI>|JAwD)=U&!Wsq zk-?(Yg;$)-qU{dhMrpkwUc0=Wtobe_@X*EF;h9q~RH*Xx^*O6Ju25cisPbsVs|v?l zcm2HAVXq4Yu+xNT_JN2>^`Z$*FIW&?jr>|Dix$eHZlbg=BWp?=dcP@r^A@w02 zLwZjZp1gbFYj_`nBt!G=Ec!|MIa~shr$Qp0hrLQ-eP?Ixv4TA@*ve%Yh*rtg+<4C`y!`XJ) zw#nA0ZB;emwyA8HNWVa(#K!90gd=h95;^T;q=F{yxNv;;{=R69KdN)o^Q!;w+nHBB z0gw7Vz4~OT6j6O^>jYj{pYIpzD)mF^6sr|W(T~0lCLdNRl*5pAg65sD+WEvEBc5@0-6mmOcf8HAz%|5w-tLx8ZxtVsIdHKvOq~-d3X5Z&`TC>7V z569QNaeiqzkvRG9y6yZia>VET~mvaVtJ zWA!A-g<6TlLiM&gqmSB*tn@#{%ct2{b$yyF5H9#_wV~AOvzXuQM0xhfHrYetvR#J$UU6hok7=sz-vi5oK|CGE34_Qd0i5=b>yBJIkVe`8F@po;?+r zdE`9DHpj~TpMEtw-N_WcV~eJPY^6Dt%638h=*%4#cd_sKICf{XX`zx{mOd%!WfZqM zqq=BwRr4+eS?(Ust_kO#jQs+UF*?y?#h#$oKlBG(nr4%kyM=g)Zrhc*xwbcDX1Qcd zX7Fn!DRTR+{$O8xbj5mu%%enc%N|u2_phgFp-H}hU)iDGH$`goG|YaQRyvEW8a6IG8<%!^Hu$UJK*`F)rOhn8;z*~Us$+9v!;amG z&3=29qNen=ES4IlG+M|43?aOQis!w(DU4fRA{)bEs@3fjxFC|&QR&zfwF3c(a6d2t?Z$G z%9157os#|5`&I7|-47WVDMEpK0R|f#Q+9L3@14BXXjj+vDICz86$!4rWo3z;@#RDn zUO?@g@mFQfQrJ2`^HibWAbwd>-$(TVizaGpXv@!-VZFQ*U6U+2vrM%h*^0jL93A_M zUQ4>TR?hj7ZlcJ~FkB0o^GdZ9>(!7T-Fwz#-fx?wRIV4jX57}q>(BaIyf>LvhK%Bt zrT!T!b#>H17~@b_GzW?bMrim^LUV49Rnfwz9e;htpimLEDD2;BG~hS#Cl-E?J%4`h zxEqez1^;4#AMaGmKdUk1)E)nfcl5y=>bUO7GiTtp?iE){OGh^wC-=z99$oN(+WEAB z8w$m`5BWi#(Gi@1^S|0&(0A8YKQDL1$wAEGs?%jlF>eQF#-VXMT zZgSp={C};GgE8`1oS*R55_daAetmT2}4{*4f?G$&rAZYjN4h!(EY|AGv7z&mTA~y>0)!$NdD*4xtF;FPTc&|6}KWX8!lii*A;#C!HMN zN_VAy+4c9e|DF7Mqk=es`F|+!hs=L{1v)FyD2Q+KOo^uN+4DSLBbV(d%?t1wgbewE zo`S!G|NKVAa?eJi)wJNZ^O;k}FL5a8J1;i*(ODe(7iIq}$$o*wL%bYkMl{X5w`v`&R19OdJWGcaLa4c^N15sP%l? z{z;tC$N_XM;W0ru*nEZ0euj2o%5M5&6(!lPrHV4u=<7cpF|D{ZQSj9%Ah7Y+ph3Zu ztWc8vGTXWUF&Kr$U~zZ?(OjJrhowjcD4hBClQz+LQI)X0=JN@nvm@yT>z@tCvXFwM zfA$BXNUWF;H(J4BIPC8u@QIbXkhA~%vk;F@G$NFSZSM#kno#^_DS?<87Ch94X%hVl zn!k2nt*8G64*91j7E570?Z~&igCOI_uzfh*hL0dqOmwCFcSmiv-`{Wm1VmbNZ7J1V zqkqA{kq37CdzT>&M;49kJF&e3Q0Msvay;Sxq5l84j)=zpUx9bP?&eY`^zJ2vrtIb| zot|lb$W^iw=J!~0s>0uI&cb`nl1*H3p!7Sv1tt1ti@4$_8=_=;FA$x1?)6^9;(idc zcKzrbTr)cN@e%8dhshJa9v(FBe6pI}ujeK4$^wz<)GW}SOCJKTI2Mt^?DzC&n-bpG zn@IE&uGBeyeL$P;d$C1*Q-1T;j*nsi8(pF2VlpYwyz*u9-oGDmdYKh?pNUPrHyaN) z_S53X{Rtd*P$W9+t6bx0VNKc(7f(Cx{9#Kfu?s(V3a))3uq%DH_-Rc`mpiZ_33Mmv zFvoQii3|Ic=Ho>zrRk?4tPf>Rop$^#D{@+ouSdh3^UM+gV-mnfZZg5*t{&3bH8MhW zbyJglkgAYLk5)Nx-1VIlK28r{A@n=$ws2Rqj4i%tX+t2p%$CH$tIE>)8EXkElkv6&-il zq|_bh{Rj}Xo0^k%gjC_T)@GxvdH%WiMtzY<2KZVo*$0 zERhk@AvTw7w{$3LVLj4QEWo0vRi(O+37=|or1U0=gaMLZ5*(YbuFN*cmTx}xnRq3j zd#2pHvQ_zs1QT?vZ2!Q^L==gNH1dV6{8NARzU!1J^{CUq#4@nVWtFw;djubgyvjG(PZ|?sMJT zSzbLH{=FH|nOxm?=`=xvWM$(iab@`{O-p@30?3LtP|Mr*2@?iug~K%;3*PB#=*@%c zk-*|k2@zz}I+>^#35V5FSJZ;Z$26Ha7JFvZcyUX->qFn3b+DyRR9~6SrHYRNX%YZw z5=c$Biy|#hvWUDGy!XFhDoFNG(`ziBkhUj;ortV?uGKJ#NI1{;p= z@|`te%V2yQ@tMae+R9tWcX`TUza-+~d|B{`ehrcmXv`#btl#QMX_DSqua##}_sX7M z0uE5)$OYFGG=qsb7!RB5{j<~@l{zs}%WBH&Q*TQHTHo1-9zKG`aKke0;~e2YX76&aX8p8d8(gUhZr^A4~)+9Iq`Wp-9KEW1o%HM_~64*!^Dh@7)j!NC#7% zBfE3Q%M^!eDkDPrM^@H#0xt$!y+ZlfTfZs4m|nqB#7xji1r0NA6Ma+@kMT@sY>&Az zwfa6>v-?IG?<+7Vm0(f^J|qFf2}2ARdE$A0oSGDqQ|fPO7GM=ypZ__b!-d4k-iaIU zKL4I=)|-}-&u27qx^ODEW(*TDHzt+LM@+p!@-6P>#d^{u3^{uyG21`BOO3~c6iB1d zm{44o1J8BYmuDvAggV6DG7)k?g8VpvQ#XS9M~=tGJ#NT$)0xycgvQWOJkin*wju!?t3x3UGfOR#&S<82Ip>8^wneAxqQ=V@Vku0PRu;Qr zDrwrwo1<;#KO}YxMcZBN(1_&QzP+Xdv*74<@hzy5Mc?I5$@hqr;}a#SIu5#rNSDE4 zTsCxgf(>VW5em{b{K@Un=&-=j^@wOLT(sSfR_$>@>_b39PdO``$bwlo66GMDO_LyH zwpT^{*6EWqE>RMz&(30=gW5j#94Mon7uMQk4Tzi~d)uUB-*IshjUs*nD8JEPP$G3< zx!P`I35<4pPjvHD)ICDdf$MeT9Fnj*ned6bLdf16jgm(lv!64bHYPf2RFFOh;S-1f z=j&HSy5!K!T47hKoMR={mlc4XYgRmItyJMgek9%X8z*>>AS2WxW669& zyXHvL-txOv>SWoBZt086?@hpNR@Q%2Gbe&Z5hp%n{Py`j_Lz5i_|=j87^o--S0*$J zC1ojzm#QzEOJxCBxTj$xL_%W|op?~Cu?Mh}6C4hYuuRPOL}t5qSLy~3^v1rV`D>Y5 z#F<~ir!5}M}hYv?9{kRY|l&CA66-ML%^6>-|(qPgcC`c!6E5I2qfLRO3qxC5X0F9&hA z#k4Lv54bL#tLK*9s-am^OYxRl?|zsU@V-g|R}`D5}{?MJfFVu`Qww}zWD=cbjp zrex=f_O#shKkh76y7pCrpaGzgt6tn9x?=Y2q$D{hNUs-Ed(GyXZj4OxQg5vU@t82i zXC?$XH)huLX0HrbF6l8$m8{JTdu=Z55XpWR2nTI&n{Qyh0qMVw-@JtrJOEHtEoKce z53K4Hi!F?UiDj#E+8X}yys_B~BCQ8A;@N9e&cDxTTr=IhH5tQ;n_io25icL9MXZGv z293!BGo0sfN)gMnM|k$5=D@<6X@;DDh3-ghdG{Y*W)_$HFJD^{>nS_Xx6159lTkVU z!?I;?Z8Rs$tm6CUI_)UShl<`vrgn)yBuD{*Z=zpB5T7ClE)D*ESocvni6c2+%4pbO zYsB39SEgpxuK|%3S;dO+jKM;0v$>`g-T;LO16+@$@?6%4Wwj=5d9j8X&p?gedlACV z`-4BQzZr#($fUURG!6fV(==3QS#Ot4G5hkVfGlIlCt=1R=KYraWAfUEJg4hHzo%lt zg{n3u^2d2zgNM+2Tyi3~#vAOc_oCJQ*!#qrwLV!plk@Q}27Na64KECC{I)DsT#2%- zSKjy;6Sga3LPNa#OQ;IEwwsfLBE^HY#g{!{#ZZIzeN6Cbi}80_7IE_J$dBx}6P*yW zz&w0RjmONhKePj4qYej!lfgAJaM{d=m0;|1;?0y-d4>X8Q-kYM&K(SuzHi%#H{My- z@diwYh-8=NmIV)0lto!XoNWq-4)oo|<^#xPHscH?E~d*(yn+3VBZ{}WIN7$$bGwJOX75DQ`7H)z`!r2& zXRYR1p!r&kag6WvQs;-($L7d%)yukrykwcldzR{(%arIaud3;|;p>`7MkB2wq@h05 zCPi?(pu`L>9mFu~wD-6YixW3mXL6iMBof^gZ9mGbXD|dQu1{32=`X))6H$KU*d4`l zeh)q|BfsQyaLo-M#Es?j{jqVxm$fT}x|*hawt>wn)}nS4ftVn-*@DM0ikqK+QH5;Y z%kLG3A*ip5D$k*oo~$a{pzVh`04W*hpKMNlE>(imyBI*|tD4Ss-c*sV_hB>9ZuqU~ zdj{thII&juJV)NT+c#x(#BDf5d_p2<^rNO&xvAodD!QNfoz>1;0VzHW6xkg7>LOL` zN-t}m0Fkg81T|XeMF{aXX4h^Wn)2*^W+{hmjyEQrmcPes3xdbD$Zuvky-af_zK#_< zc3LUY46iryw(nk8t!>`=6*a7$)1SXQLmpi5FIfvx)KdOQxegpbf8#^{ZK{Z9>~FlV zsLa-6^&83cUxTa3md)}m-%gcGN-J}2df#uk7Wn;5^Y8=S>Z_xY%4;ILv?&`;#ZMYw z9xeaeTvK9!+=ydwuvB`YFE8Mm0cZo^Fqlb&ic)s!t0Gz-ig#_+?90I^&k5u(2qn3< z=9>-(q;#ISbJ<23J%z!mpdR$O1pHnSZF@G;?2RSNe>nB4s%fME-8i!uh#&~Ih zChgYzbhXm@+bnBwll5<*jrHnVi&f7X0}Vh>awJ^BP$U;n1(%Ix4T#bELg#!a$vm4K zy<+qULq4zfWW<&E3|m8+n+-VY>DBVMLr+zQ5P%($@5jVLj#f zmFH_4`NlPBUqqK?ZlFjqAPO?8sWcb^Tzjo-r{>q?N4xDS|6?z z_d0Un!IkCunnS(Uw6S*~@rrYLLWRGfMp#nDuX8Fp2QJTwnDtUGPnj7m-~N`P)X}6b zyPIEz(sj{}Keh_oP#{BAFi`^I;harVHZLlDO84Q3R%Yj_KMsZ(yb-m@8&%8>n)XJ? zHrvghaM9_G0j(~MN;vvv%$kleb!0FX~ ztq_PNPdv=JQxl9?aEZEGb=;Xaz@N^P7->_W!gLxUHI00STl?`jMDD~)nk2XAV$CCZ zBmQom`n5Sp`9RwIHKzrvFQc<@8rxU?ogP(x; z?d)!BRr!4T)Y%ta37M_|bSC`jy8B6Wz=b*v$5L!K6NDcolqU^sDH*%To(XN!&-dWp zgTW>j6?|1Oze-vyQN0+h*}1or{pNWy-nQJ{#MAZn&ZI*8QCgF3bd>Z2J7!|tC>hSA z>6PqT?DJ2=U5>vG8EFx4!oyEA8VTRB3%+xFU*rqx9v_;BSEcFK5g@q&vaWtkKe0O* z|4LCPdE*K=ulzWf^gH(BEE=i8aLEf#qeG-&DnQpWDWX`@QnkGKrTbY3)Bn`n<_~J| zcD~r3H{K`WkssRYpz+3oMo23Wcw_ugG#-pN?2`V^t_w)%Aij*cg#hSzgCHOlVa`VW95+a-Zoid;-@ z#1ZS97@Oq>HAl+L9WmGuMtmYgdyWHAP(dx;Zkes8Ny8P9*bb~d(vWy z+pq1QX%Wgs80OXTtBa&F0y?*L|hv(HR_4k_S zAm5UIdguH{mb1Y%+h(tx+P-o*D>6JQs#6P|QCy6I%;Lf*m*o&tL#0=$igR?r7k$#EoH?`mq6x)JQw9C|nbDm!Qp zJanu&q~1LkvV6{D-0y+~$AYAhd*v@YG1zOs@CP$-D3Uf+q8HEo>W59^pN;9KjNQZ= zc0%fF4!4^-w*h%gQhf%xd1*DI&yHR@CIM`J>2?hkh{6IjEIx1Bmn} zCBunW^soYb$xdrtb66^2OxR&qTIbK#zd@3(b2@wepl*qo#e)))a98>4k*PKn3&`U|R$dH{< zdKrx(IRb$kxuQ-46JszQ-r4Chww#J5a@>B6@M7eW^E4ESL9L#FT3zjBQBPF>u(OkC zn_U)80&D4j$qe7^p~5PMw+ZYC7e6zKawqfomo_Cjo2mwbiwtr5En0}eSRx=T8rkr4 z4o*NN0%{F0<4!>S5j;MbLL493w;*JF;pB>H&2Az+*g#j<=PIgu7>&74ghEdYcmE5P z)s;?#7USmJEpjE48Yef-TU~%k<0(iW)cUxd0|THqks3znHXW}_^O2|*vVN`CvH|tS z{RA1xSM6~=NIhrz0@@N%T8yKnY*d%tsS5s^0iZ|+FwHZ@iVjOYG&HJgl#?TfZ5 zyza({#&mn`rXP~Y1}AhrxnHil}II6Q{#oTAeGJ?eQ+MqBAEB`J$hKrOa5cUn@7l`5lohE zUDAa78SU$nkU=#vrXOxH@>7F{5$c3s+ml(K#7VNJI?M1GhcbhJ%b%-|W_^^_~aSffKb_8w!CF zzjU(delvxcLBS`N5AHwdZwf^dM@*aNlT694@?VZnjUFWEZM_mdWV`d{nBkrWnL(RF zPIqd9V8iE|Q1PHnI}YVI^;DYA9o4-{{gNKGCwGm?32OFgXVbDgu%lLnfaP-xER4z2 z2G4Td{8DW8gL2!CUo9KuUiFWTRNJa4jS>4Qa#kyMzweliD{ty38Z$d4EUq*#TPHU6 zEpGE9QtwUzt|i5BKyi!-J4R5Pw0Gwn&pGKT>FKflbhTJl%hY{_7m%mh2NJx`fd_-l zXOnTB(So10bK|d1c+8n?>>27P%1Bu^lv(&RZQLX2IH0$&Ir34=6!*?azv2)5u7>K8IWFS-4sD4>0V z*yQ{49U7p`>C{q~`M5p93NuqNV?f^1=B~x1rjNHz6NS_Ic{>>4w3@>H5nmtAq(79# z=RjrB$>{;LV!~&ptsOD8N8`REf;#P0YkxulQUB; z0lGinfFJZaJjel1aqO&_Wr1NCaJZaR68Rbqd3;2H6aat?8JbXsw?&dWFwEUA1FFDG zV3+=s0tquu;juCcnMT7jI!|2%hS4yLK2EO&!|k^M6sZKhmaL`80mI?&V|Fio;t5*! zAh5YmF8TmP%7f`VdGkQTWPo(GI8p8cPWr`wljBiwXar58{mgwxg;j09|njSs~rX=_Vzh!P^xzG)4uC zgG%w>H>eaCLIQ2r=MR|t(jYodky}s0FdhWFMbLEzK>8H2(lm!CVDEO28%#U^;2lVM zqXE+ZUYboPKO$=&tO{%iY>2^D0{E1qa}Y_PgSB)fT5#4|Aj;dqo)bV6@CD@Dmk&+_ z6Dwf4qEee3uG%hsqe!~}{@q9D5nhpCNUBwb$DD3h+PzWii@~;n9JEcPj8WrHpNFvj zwZ#VzDIx?Hj^@kvL1qI0zW`d!6c}y*r8l}c9>Guq$h${S46!@xV0Ru}I&cqp_`*`S z%53Q2#RrOj4|{yZ_Y=*baOcC~8OH=V!3;XVtQz`1p1|-oAl{G86S)R8G$S~v_p9Mv zG=>V$qmt-mqK2#qrZdmMhG8y9N5z9^3Jmce$Xs4Q z7l6=Ku-}Tlhzoi1g)?eO-2SDL*p;rf}l3|!kqDKHsheO+E z4K` z@T{|S78!nnuYU&*U{c7hpF99Bqz#1Do_wQ^!M=m(#;M>HMEc;>SD`vd4=8Qk4>*-I zRUiS^7r^>U;R!y#2|Vh<&z5!&U+V!*on%8cG)5a}ru{}m8$h-JoQp5JBCe|pgu2{C zf(C}~z&bQ)NkfS97M8w!Gf)S15>RWeML(s#1&UH*43sdo zH$^)V6Q*c;hd0D8c4w$E430nd40 zI@$EabpUD{gtAj)1PK}Jft3TO19;kn$DwfnACFX@BAX)-i=6`Sr6lMflCl$S4HtU>)^+>I6n7PlvhKQY4v1n3 z(*px9Zo^dtKoo-$$%xKI0Q?d0;Xt378X!u|P&!ywdmzfhrV*HYTaW|WCCvpe9)CLF#6;M0_!LF~wOx=Q#A2G^1DLk#HC0Pz1Qxy3@?Qlf*M=xu0l zftxUo?Qg;~=y_us?R}65^GdkX_iY~n3O&$@vt|zQDgX9iXhHO0e5pDMg6j%6pkmcl z5-J+d)uT2*_=#8|P--7~E=?Gw+5rLVw7(-?W#Frf-7$o~P?E$Vj0QKVQS5aPOu})X zX>IxeB;FGT_6^X$^j`1?Jow%VF#M$tPGoPpGYs{CXQt;G-C%eF1Ub4~`4!l?W3ay< z+;BGp$)E@-MZK3HNQP;i@*N|{f6MaMuB>x^JkyK%6Je@Ar%F1ZcwfYaf>MlM}{phY^e!H#OI&>4UiN-z;@vr z7#BgHsn7uNC$wY~0?FOiM8Wnv0q~x@ZAOxyDWLw;p(w<7KLY&D^GhQtZ3y@Gd=&=v za=}ur@@oL}JqQRMw%U$^ouh&2TT;RyKxsF)$t}+Za$D>Ogb5V4xr2zLfKB|~iC;tRzlFTxv!=*Gbp`q)xS{ET z)sXW4+Vx=yGEP(f34aV;uX(*M#MlL zaV@=#1rTB$JS=)h?_8GG16}HmnMD+!-ytPQ_NS94a`ulPU4mAih6Pc5$B`OI8wMKM zeTt;iX}@6RuNnT;Wat6r;;*M)s3hGa`53Of!uR)uv8HAmC{+a_kAlFvv*`i4r-xv4)`(myh5_RXU#(f^IaTS4J4Q_v1O@+e0*EDy1 zXkeb)|J^Eh@c>srB?1v>liQ-V# zF|D7XFaI1Y`zpB5HC8Do4kgi^=V|Wp?7pAWVo`G+(~J^T$eBkTDAz$C*Zu6nu=z|j z;ZOw*cgmXi>tXdjDX)2W@Mnb`HmMn56jG`>1kvt519wqvJQ~x6eH?`Sr7$P)#3F|~ zXrAP}Rm&izVHjWmopg)4je?;}MK}!hQvp+UfEur#68yPBDd)))YjVV0DG;ALu!4Uc zSI+Nwak8MrI{5?h+qeHb2jtFismQYc&9}{>`4vp_=ZUyuQdZqdnPGpN;(y8!BG~IA zdj!9j;gi#x?qI0#U63JH94CrHPVPq<$*?wFeZX0kQgfpcC()v?mPj$ZK@YDs|JJQO zQaygsond{b_{{K~nZy$h+u`ND0)(l;)fAMFf;r(vd6q^etneq5+6kfNE}|3_xO!xt zG={u*Ya@E(;j@WqH>a-BD}EFEGLhJBJNGolk+RsTJzO}R_~0fjMr^{7E;1e~odTp3 z0nvD=n(;|!_|!-paA1Ah>+{_Gp=QtNmqyp%31!=w!dqR$$Tg;v)Gs0nHlacD_0y(_ zy+RJRzhanS>m5*Z#%C5F>eWQ#NyoI>>|l<~W?t$X`K7yh5*BwJYQH1M-Np~0pa!8QKSCQ@!~=^{2h3Qg=6ZT!q~y zc-{&|+n@Tg8x6a6c$)$}{kNOkx*VVp@uyygnv|z$l_K<&I(;~&4LLj&7&a=Y{=3w- z511GGoZpZc$&RsKz8594`h*6|9%#>GRv+ZzyD*!C+#L-K5zbcykWi0U3xZeMT+FO{LYS~gz#3oaooHymwLB#=b!k21igaHAdmmo zUcvusd*SDDXfGVN8J7f)m-$(`G9-Rb8~$F$FW7MC7F2lcb+(|}L*p-2{oTc<2hD|* zzpajxm{1P&*f3&mfw27bs=ZgU#t|BDUViH9^FW$NFMPNJ3F06f{VBQdKA7geQyjT{Z+2*T$pzIvl|jv zuFlhGny&=6GdJY<&e`*Pn%a-s5sRZ7?9!dB{{}uc)SbMtPVhDegAqc)f#}7 zAM88kK2k4FN%RR7je*Xs5VxmMx>67$K*wU9i0#)OiEfVrbR|iBaAcpneZb#k2(tgd zl>c*&r%AC8{NDmC=$Uvo_3UDt+n?a=EQBHFMhiWd_7YRC&o$>LQ08Z#i;KQ7#Xnp3 zi>P(AM54P6E=FPTe8%^5x%i>PYm|0%0+fJcr#B6`A%<{eAoo^{TMlz?%|s1%F>6LB zPOo@-Pb?QCEhlf?I#!oeE^a_aRxX}q98a({rT>RU#IN4 zPJuQ;`5;ct|Iw~(ihsYsu)}HV5%;hpC8g+f|3&YCg6Z$WUxY^FO^>iY9i0@}V z<&C15>?PjaEy_Vpt1e8mtj0+?_64Yy*e-RvmD7hra~3?yUlpsntZ<%Q^GbFYHzlQd zGs>dvk+{j%%w1dl78D<7bL~x{9F!PbWqx@5p_A{hx2j@SKuUn%dMkHCo!f~y{+r79 z*MsbVO#e2#6I@6-O1GxO6TPgu%5&$|^9x33TQb8&VuGghOBOnn_ays~HQY9;Lx%o( zbs`=D89J1twF{*1EH(LJ!kenGPj8bbP2ArNz2pD`aGvPLP zcHQK&!%B!-ak45riGBwM^o~9X5|t3*ncC88%h&jppX7-8_q@w_t?(wVXCoq7fg^CU zQ)y4K&n27Wb>1_7zpQS*Zal)>V@=F_b1~(X_6BM20h{vKY{#Ehy-h_Wf<_`^!i)kZ zB+zHpL{@`Wx+G5SiLt;>*z6-$vQ-v+{ zkNsM%&e^^zAKXlFEfhyno@Al5^{_O}>p?36n%6U`Y%_dXt$MEf-6MY+YA-SSoJ^ti zagQf4)3fUGN88B8qo#R3=U?`3Agx-xY*iX-P0)3^;KQZ|rd34Un2%_#0A5VN?-=uY zD3k;~NqSg0oR~}D^&e~>*!Ew8c zko@NTg9_sQ59y`QWF)QLJ-njrVHWtT z$7fUD`$wMj(a}KVj}Y;>NFGFE=70xty*~S(jz-c6ox5z49c`UB%S|7JS6bWRp`A8f zR;@Xvsl2amHO`49CviEnHX_q$ud&PtyrsVE&M<}#|B5B051Gl+(bl)desC~Vbfn&yA&QuRnolY2AUHETU?wBfHk^iK!6 z^i)C~~OccKr?vmLD9nfK3P!yq-D_4L!*n z?@_d9dyr+(YpFMq`|PEicQY6uFAdq`*)PgIg6rQ9`_PgUzgo8XjVedx%5c}aeZxN! z9z;;YFAjrUnPWC9o-;P0CKp{#fy9`9W($o%ZEbZN6%$9Hh$!U$o=*1r1+#SAU!JW$ ziNA_VduqXQgf<@Q>?8XjmP)dO(Lc|m{&qi?6{*Z)tlwCerf&m>T zY!f(cC!lSi7pu*anS0Zm|3zvDog6MiAuzx&GwapV=;+i-O6B01+OJpH*=0*^`-@{P zW@H+2H%mO>3r1l?S@bV%yh`@l3bziUPNB zfy={Bn{VA*e>vN>*)B2g3Q3zu6Xm?FQMy~c>J2`aUhg%u^dzpZnqf6#wK+$cHxUvg6nU&{eG4g=Lq2sPGf+`QhN%&oQ zSdrSatlmj~@4H;#g8xYEmajozm3x78QWUKqSxPnOh(N48nu&6?F=Bu(_8>ZrnNqNt zOlYg>((HL-n-tY_t0?D#L*%BOq2a{W0>6uI*3-V{P48>Tsi@MdN;=Tl)wb;T>N2mp zH7BR(1IqaHow9UN0R2d9OEg zuY4Rb!{SfhMcs6}`}DXO{&Z<@oXq&)^Xnw@5A|`crgR4rUpap&aN--{BZ!YwKcRm# z;QTm=3K9x)A1CvZW4|))ITM?bZW(+Yy<~qA1!dS+8MX1|FZjczHD?k%Qrl%Yd8f2? zo(^ymem%j0#$xaUm))p4CWPQ%yO|K+Xb?U%v2* z86cOjPAU~mLf785fsO|&MXur}k+`IKV3h&fVd;{01Oso#z^X(Kv5U7S6B`~CG_Q-o zv<{Fwj)Q$QjQ9b$f_tY|=V_nA0%KLW>9yS>H)>?*V52DUo3Ru-FTkNgl~_Q&XLs+Q z)wCzMd*pAqh^xfk<^m0Yb&0=bKFeZTMWNF(-B_UC<@VDhjc?~4sj~EN?LcD>?jg8j zUx;qu3nrey^skvb?vE}Y?{hIIXSC$_8lFrN0AXdKTnLLK^T#ToGmJwE$fJ=a7ZgYA z-7d6p?LuP>4-&ui=^R&zL19!7c>8Y@kX2s34u~sg{Y;|e?42!wkLlPf2+&_cFfI~7cg_SnE#tN?4_sat{k<~M3$ z9IK5S%L8sYn^~X47}g$6WTH=I(ns^~9{-P8bR8JFXWu^mI?hqCC6BqbRa@RVVVta_ z8|c@l{@~zYdr*i>0LM^3t(x(49%>)wf!(IG zYGQ@cnJlII)I#$U`vFWkb*xMRtM~_cb@_fg${DL??n4i^(~Os^$%|Z?V5T9!aN!yBmiVy0#2=*gFyqXWIUbXA$yYFx&N=-+Qd_<03!+UU8h5#%Mr+;0P~7XE^ajT8)#^BXG)qtFp&yqvL}^lhLJ*MFB3_|?L`Tf z17+KR2N_~l-2uZ3HAxENN1t=P!fUv79vf%;+hJPC?7EVJxQP&;FEbZcSvU&QOk2Wi zbC)tw`JungcI`&g*T4!U=Ohne0D4Snp!(HK5(EJHxRD~3|a*$ zw6PuYy^VtZrQFw4P}ST`!JOWg?V4iwHAd^PHa}MaUvTg&n7DTl@o0YYhbU_IpR{eSNL-h0nEpL5RVoO=i0a~z}gEnm-mqW3Zd9(z!o z(%GS0VzZ>6Bj~xi@9MWPU0C9VDa)E50-_7x$GR6^`UQJJh~nJ^SuK!Xa&8^R&!keBX!%VX`&lIB&FHl2EjdVn0?2`=aw%b94V5LoG2NPkD*O>OR zIvQwT?3}>L1CsGL%B`Gyk0`HpX<#FM=B02$V%W9Q0j>Eou!d(~=A5rxleC4p4N0&Z zzr(P5y2Fw4H(QHBf+8rilJQ|qG&pwc*`_S;b9O?topZf-zcs`OW0UVt~x#|NrE;LJNyB<^k$ zD1Cpx;l04~>C;G^Q-=UyX~-0TX42epd`$ol!JH`XSJ_C+IcE6xz>-LA2C1~=mtiJD z;dti@V`m0id!aYPPegW+r(2IR?`xh^U3FHjLYI2AZb}3-FNdQ|RzU)Zj^Tb6&ytkY ziXyaHZ+~28b~)~V8XTD?Y!XP$E3!y|gr@LD!<}^U{#*XnJ$Ua7xIf4FmPyfJa*2|$2+Cf&`<)W0(;)Zl`VzN#h5$6`=;9A?A3@@ zZ^5WWUN9<2R{0hIA)4e86BD8{9*X_Sc6cwpb{ifmkE|Pg@s(288Ef(}JG)5L`zJmc zTfn;k!}fXLF4I*ff0a+5oHu4$^UBAs+SIJyD%(Nser20kJvv-1F3^ztNm9Aq#DoCX z0_{d^n=>q5zGMLBSEvkBI`Of{0|2;>jV9{c?e>3_qnC$@5p&=T(KM&|?i-uS#fgaV zdL+`?{YSn&qAiNgZp`7wo#qf0`z2?*%-F*yz|opCdKrMY;Rexa%|^(ahRRNS#8S-`MsXdndgT4??)X+tgu@J z$lAfop&fPEt%tbn^7;g|l9AHqKbE&G@pG}$gaGK0%wIlFNFqSMF1>X@w|4h6p!*_~0YtoSMSt50;KpBRIq zA@F(w0EYkRv;vr{!Ij9Lilr@)bU?R*Mb@Q6=R&Q0(^5fw;>b#Rq&jxgmTSjvRVn9D zUQu~&U-5|5NZD2`n)7%W=fE=VU$LhEcxEv9)1M91g*@Eo;Q2h4gseMQ4@I^+yI~`f zqMBJ<1-TJ(e>*bo!7Er2=w>Tr+N>h=tml z78`$Ww$ylcjiY?*H7R*fc^o>=eCpfgLZ4FT4Fv))EI4tK1`DH=_ACCr@aA0V+l(Tn z!R9#s>ILHa;+BK#yos+;UBwDODnLlW31Wc|viBf_^>ik3VR)+|SgFYBGM`fO*+Rz` zr9Ch1(oE^@*=muMz#~+%c5}Q;)^$4y2nDfv+8y{aQ+i9t{*{OCZF`g$T8XJ;!p3-P zlymVD(~#lApoa#IAp}G`wG~81K+DVb3G|DVccoHepVE z!e_y(MQ>1LzedW<(6AEb#<9(hUi{kmo8RwNPt1uO50(Uje^wM}gTarDx)NoDUrfTB z*S|5yrhUI;3Q6m>dz1Vgg596_msgC6cqszbP%kIgV`n- zDh{w{#6oYzVgC^J3?z-L$^sgI^4%So?@7aqyRL2(m@*z~3pdb~Fry(5k-g))=Cq>M ztv1@UTkT|kx3+Dh-;I8iKcrD`BaB1&HP2MA20WLgZkX!*^&dyVPlYr2kG{0)` zs!oad{c(a))tl(csn6g%sTj^faqCgZ%BB43gB%@z`9RS3fS_Tjj=uOXK;Us93;_Qy z&>v?X)E)Cm#fNHW#ZB5VxcHRK##~p-8G4>w?P6dD0G+}BI>Ex0PZ5%|pBpT(i2Ob< z&{pTZFGKvjhzt_iEgbkE4i#iR6o3k?&;_WSQCDZMx4AHF(DD4(Dm;Nz5WhB&09KLM zaQJ&4vs~PT0`F!C+Pr{YNWswBib0rqu(Zgf9|cLdI>oo7Ud~SZY z{z^5X{oQqNsKR6r|L&rIia1oLIb$Fya(5)+vdT;wf5GM-El372UMHksIt{EevjGLs zEfQ;d`}S=j;WL~!>V`Q~@xgmkxH&Ws6e%x-{q34VKY^~mWed)ile8>*+Ny1AYwE?o+tZN4v zfM!ZY$(rL79>k%d&7A}5)#4>TO}Z!j9{F@EknS4rsf}Bfz&E1EMv+Iqg`a)@P}B3Q z(tF%R9nBUoT#~EUb?tddl(IZ~#!>BU1yNeOls z>y45&fAD@Yv$Yl#YhD>xt);DP#`ex`*dnV^d*c$`NRm~7kqCbt#Gfm@cQir_tT1US_O0)yXw^z{nJ2`Uw@hy!32A!T)*p{<-x4VnRl3s$T1b#H4) z3C-L)P%^5BRxded9D|oWyG$*+8K(E@r|(~x{9}p@636AiaJ6fO$lnfKlO^#_zP&j= zG6AyUHQV;pBo9Fjf#x z0K#^aZ3n%EH1Qb7La_s`p2>!n=2pDPb4K!m( zF2^r5v_=Vq68ipu6|k1nQ=-QT-wToimGY-50R9wVtAf{o=YzWK7Mdx->k6NIJ+&}g znq%hvo&gM8N+=chHa1*h!G(?yim~sK{qYx9{_$Pj!skB(7bfcKEhp;6L?j`)-B?nH z@im~0Z%`TL&Tp#wzkfG5Qux_Ouko^A_5^|HffcAV6atyI(|7m!0$0PW`{pNG0t#Y| zd;L)j{ori0sX1RWa3e;F7H>X&GS&kT8b0`)X?&ds!jKlCV$+TYi@r<@ixT>LiRBG6 zgQ^uAm5SkmXHMCG%P&Ey2~z#_?~g5YD4`#g=zSXtfzDpkskF<82d?<9jMqRWun&8zNI@rCfz}Xa(E8p9_C(^_H%7Bt&tl|9}s8YiOGnwWnvKRo}%** z!}W5g{cKD6oyflc+yb1%h8g%!)7f0ZdILWSM$Xez%xl5-(X=H3H_-H!R`9Y^3_rYS z${IMx7!bLct^Qs%=sHGG{1;DwFW&R2b=_(E>nXUt+mS$23V{;7L4ErFJjEmO7fnNp zYa*{m+Rs-kX6Kmnq$)WRe-|c!C@0tQCL;vkol`c`O>ZNsSE|q_BKddzbxqUg^J`MO zU!^~0Db?9z^AKe45SXh54E?wyUI`69usS@z2|xKla`X&^4^o~3)y<=nB~}K$P%TdU zj+Is3@<@4?2n-_KjU|RCBs-HHk1MJ9HYVk`d}V8Et2ax7?T@Jq;ImC8ms24iz}eqf z?c(WRlCW-Pibt`S;|TK#JyE6R&O!CS$7&4Z>Tt10yV;hx`8a zR<2o!1tG+~XGRBYFq)U0eN~rhJV#sT&ki2}P1QYbeJTj-@H91RR8>`#<%2n3R}AK{ zgn2TA5T~9Z24o)_O+LKz?r`bzxf3VYa2BkV|4TbN?KN%KDq6;7yrzcm*-+Eex1f!I z;#5(>tv)ORGEUO=>F0g9{c>Y`KywXX=TPGZVIC3z+KSb>{h1=PAet!;ML(Z6S5o8o zDAp5IJI`ne@m?=Spi!+K?x+1x(qNFci_0}q&;ZOWvl`*qgGKuJodWSpFbsW0bRn%& z6?QbOfHo>&ZW_zZ$tk8*h5jn(8_jXXr!M1SO;RD;AG-X*3li&lfAQWx)>+{`q7a>7C*o#ij`?I5E zHoFz=wQTP;T&(vH#}kgrVcg=mKLrirjq26Gh$qg|@rO782Sc z%;>ZPVApc6@CHaQW>3PJhTA|YTwmLgX|8GDQxv<;qzo;_p|U6mNeDxC5h1*S2+o>@ z5I%G~#Lc18LglXJ>6UN*QQ0VDSIpbA5imRNdT$oE7wo5u#D(7Pk})Wld3 zpPHy%0D^&51yd3#kR-mEI>UJVsSmT9+n@+1gt}Xp5WP&YkI4nj;To0Df31qyj zhyLw>yovLF_IeTQ)%a#B4D1!ALM&K06Jg2chkOaYkc=TLc|}rUnR(B`4`nB_)<7K9 z*+_pOV$)WTjbj~?F$Dh}bqj4)65KQau9{wmm2${FF*4X9 zfDui$o2NWIV~KLK3p&RSmq8e_MR5;JP?E2MCNf_?YmN&ZkbVw*Tg_tMetl?!n2J%j z@x=N4W3)4o>-KW1+zaESfxN`e3kf166fiWQc`Hq~^m}vY+aoVgS(nWgxv)%?7{S*0 zE4BD+=->4TSRql2p+}{FewC+i*KKzRlCxe8^bB3QGrn|^?}WjD>-+8( z62b*g0c3Ls>`W_&HqnQBFJTbk$_xDy>te*gmV=qG7? zAm=Ew9;;e5E~aLdoMcQkSTU^ksGY|k;liXPxuy-CgsBMelLVsK0sN%<&t9+MPl{1o z3RDw`Tz6KYZ+M?-o$ZL*IF%K%zf~j;86tukq#?u)2dbTHbgCWmUM_xbJ&8?V*&sC0 zJTE8}1WZm_-yvE6UUJ38(kMYIGQc5Ol`P*v+l=NJH;3S<9QVCV+mnbS>=OQfecps{p5zbUe-E18GU0a#M$17G zh~SQC7>UDya+?WD>}ZO@=|53tAiy$pw}5#7M`wfCtmS<;Y zTT;3f7}WlOOl~+hnx3C2_9sch=Vg6{O@HKJ*^8eyD?wTVP+7gwv#GRuOmy?KkIRoJ zV7eJ@J}=n_Ni6ZDyvx18^j^X{>89dR)~m})HxVpc6o&zkFmpyoh4X-M}xJuedhjwwmcaR`Cm-BSX4$W~2&uy|%OPP#v1d_# z_BRCv(lgLW`GEb6lvza*5}&%Qpx|u=^u<8v$MXkDBD%mU+-3_&#Yn9@($|N8OfIHR zUAp6sp8k8hGoDn6`1#e}jNasEesjIur83df-xIZ9UI|2Tv-at?vlb&|c}Ea@Kmn>2 zpuWL#?j;FGj51YYIm^dgo{tl3VQUOBvxY7wW2E3^(>8lf!p{c;PC}4>jQH=AVG-e6 zD^n1&VBvTMd<*r<92Zhu`mFyhAW=9f*S?UCowzV=ysBPLN#+9bC^T-S!iUbwqm_Li znq)4B!E9k=I@AgCp8vqIK&EZW0makiN6D0~TW__ZfVOWH%F__u?}t zcF;H!Lun0yQ#+92%;uT->J#lbU2|CcUUMS}d5FW#r-j}HAhj`WOo~spvp|j(zsJl3 zh(*)(cN`*So$X9127W)#>n=Kpra*C5%^5>c8zj1s{+(-5-o@o8%@F|nbeZYhxC{W-Np9zaJh$)3K|0CNu*{siFaAgC3qcM=|lHJ_!ehD-3q z=~%m*A2VS=c8;#Qh!wKhBg}^EtLLw67NG=TO#leip(|uB+(;;BAJ@8+QZR`NtZqMb ziPc|vzf~kn>GhNdKAyJK316AEcG>?D&9B`S#-Z@X)D#85BB9XJiVZSefE_boGymkU z^ayORlMzfHRNjZbExKBfijg}!J-Z1g*8PTfLH(+zBBnBsnv>iF_D1C!odC)z4Il|6 zmy^<6?Qumz$_ITvo9hZBBPCoy&5(V{4nq1>-%sL9c3}W0EgzsXAV_IXkk;nWNQ?)7 znVTQ;V@nvU){bP;;x&9b4I}?w#1f1fKyCOndm2DsJTAA^-~Kd&&lD)`31t<4%wZEN zMd@{yq-5Fb*fqfZAJ_2j3Fyr&Eu%hrpGN=j41HTgU^I$MK(|b_gra4_WoHG_C9@Iy zP)iFp@f06C{)-XBQ!XGi_!h)4M|n>uZ3*-8644DZbH*?fF>VCJObC!n{2^UZKnUmG zhVBFr^*?e3p`<0m%i4H#1@JdT5MmH7(nH9x62GOH zo={pJ4cP-2+N(t1Hro6D@eMTZfw7pEEMLTDf2$KoYVReiN8f=%w#<*|vLN#TuIvV! zb;bs`n+~3?e`gIe_>$&i`Z5iG79Q1i?BooB?M%t}N%BaKahFHzH0dx+BodyVf?&d@ z4qiWBWLE(eH)@94?#N*L?0IN%m zU#0-N!q*mRbG&#iA2#Vz9&idMb0lMK!pnZz?6EnyXvYAD@KtmF*^x|Htyo zK9P3#tr?rW#?SZDZv6>fN*;r}`J2x&o}9ra3dE-wuCuTyHt2bM)Xc(DMjOD_)^7<*QhITp!F#n-gwmlnqU~I)*mKk$ zJ^Kc>dA@J`2GFxtGvUfJsX>AfY09B^FI~HZEjKuo#Sz9$2;uE5;)Dp^qhRe@TT zo5A2!${&P$1H3$c=Ct?!dWO+F{2UVC>u1f!YEcXpvEkJYzlOE&B>q^HgZ+t6?}Y=Q zDp|Ub0ilL4Lpd4rUUimiL_+-!@m;;){N^PHW|H0@J^qd>?NUB1;2Ai z2{E!GxiRGypi2OT4k}Ea%v@MubwrVo1*jdpwB304w zHJ$k)Eq3q2N`^|zDTj;OpjvwYY)6;I#4o^8*Jx$55?^55$zFb5Ync~Qu*K&kN}@{F zopt7!P6Ub>4YoInSTq@y0x_va)bc}|IfT7$y&JLVs zW6jdDv)fLx15bR-_UmUNeB8_=p?*QgAYKAitC|;+V?j2^o{YE)&!4dYVS2Pu(1ky6 z@Cw8qvdjW_-u#O%IpT}Y8N6Eexx~VSwEI9g8-|VoMI%Kt>c9$6O2UAr`4h3lgR=~I z&G5gOLBSC%9dS;0SYXr*b^mW!5E1W!Nnzpv;Iv3j4DA=k3kYp zOsIN6bjw-*MTdPRxrHs}>2QI*2R$F0@TTnelE>R%4L8=1Q*Li=9h){&&Z^aw$DnurW3Ad8FR6*hrLgrLKd6!?qV6NUUERgWhAn`&&NgW-R(Q#wQWYvn9wGJQkAZ z10@A#K$rE}@gn=ODDxkGx;b2CwR8-4g`iDiDAh>hA&C6d6RsAUb#*c&8yF0r+krp6 zqCW}vEP9A5wwzBhvjxoOt;O%W_W3jd)Sgeg(!XZVFam|FUeJ>;RB4}O_QD0KBgDUiLEy&= z6+Y|Fw!}rEZdB^x`xUQCE|O(@j~M@W@Bj^=EIQXLZY#IAdI4ix8=M;xX#qMm?EvT2Yj*__WG4YY z)GT3Z86QY5<8w#ef#B~ngugl2t{HhAR|h!jTJR=J^A4tYlmge(6f$e-IH`jTX(o1nS?RB|n=s()oc1d>U2UD3H$<;WM4i@#22rig`(1goj{*%4&1GvsaciZntW&!3f3=79(_HFSwp@9PBZHn63wwjUi0m{||ho$c^y^BB!bMMBTT5 zM%TY(k$!Xm0-rf%ss1$tlz#Op?tzdu3s{x{A~-H1Yy` z$@p;+0~#j-?caDS|LFLy(l5Hs1*a`U42+l^N(5KWKxn}YFxD*6j!}DiE(?8Gp~u@~ z$H!fKRIK_E8b(KlzlYMDK}LZ$2QWGY=uazschOZy0k#~}lT9P%s92)^ptI$HcJ*d?3i>rnF(FdFweI<2+vD?0mWpUFEwn9beAD501_j zAhiLc{y&kLJr=C7HJH4%NO7`me}#XS1XdfL6XW2FeRmm2M$sopSSyy0_+&s>0NKY& z6>>?e?L<`g3iR36iVQ$hAexycR=afR5{MW4s>sW>LYAAm|{{A>0eC6IEkNO6Dlr|yy??Ut8vcPlXH#E;!0Ex`{wP)ic8_$N=p z{u!+H3>B5Wh_=z<_m9GWb;T$0y0yV7fVj3N5v*Po?kN&L#t9?dfc}z`D%0&ku-4b! z3O_(NzWdW3GZiF+`V)>@{m9ehbz#J7$1IsS6oo=qIHOLu8n?9@jNnGWiDv!lvl>SG z+rNa38VLSib>$m|HV#6H9+;F zi-CtM1Vy?sP^P_Cg_xKi;!=+R%dMJHveD#e2=#h!;aXJ?Op)l|u_UKCl{c&b$dAyw zNculuISX*@OL4%Cj0#|%Ml|#FDj$G?FE{&nl-Y&!Es~0e;`1^7P9%;gwkB}#OD1mz`39F-aG~2MF?$@LfmoN zO{|Qw-8dkfatu8Bvq*YDPa2^z3|(20tAegRumT;6!WO;UCw9yOoWi?FXl^VKpw7^8 z;`?Pzw4KG-&Ror4JA13hE|TRvXviWcdT;Q~D$hvoRBBLHX) z*QvP||AD{l*)h7u1MP|6vl$qJLvhFD$Dl7K{5S)p!Q*J-By*>#moGU(o6i|~EFiG8 zw}f*Htbyo3OYj-$~j0v$$z|1P^d*}ow;YZ0LZ6q5bkM}iHN3_CX}uQdR{ zUyBi4e&pw*xzb}aw~DUfm-FEv3|(JB5zG&dK8uBtyau5S)p(RFoV^f%zQ8DIe$Rk1 z)j+Nv{oy-1d`-E5hBg2Oi!_Ct?kl?H)HiW2SHJp^%hn?1I0Gfax5%3qIoc!=)VoLs zWd@UZag)FJk;tv=UpEAZM@!vCMw_6&oiiiUJ31X)PXLih!rkCLb=-IKw;^a$&?_@> zIU`18tB6lyyJ%_EtA~>&+H?H9I{IsgYWu`1;g5zs2$C-#xTfkt$ujX20)nvT^m|cJ z(O0)JsU>>+{{u`7@RUZ(onz&CNCaT= z0)%PfKSEzG0-%w^jV9Wl8hWty)eiK_p--xz=v0@5heDBfQ}X&iQlh7-^0maR{iIt) zZZ|oY8QOcy($gdZ{q%gd+$Q4%nh(&Pgv2JGK^N4`Fx`J>RGB3me2FZTrn~4mgso3l z__!*H?__xP?AgiI34wz=P10(8!hPd8XjbIGdJUN5Uai1gd>M0!t{ZzD!nIMp(J8Hg zr=b5;75ytDnFJzoEs^_xuJfUJ%$+=N+& z@Q6V}^NwBr`dIOZwpP;BglrAI584g~ADx$&oa&c#%FBO7kJ=f3MT5H!8KXr<-@=AP zigRwK{}vYDGRlG}ZS=3JwQI*#woklNw=1u$jx88vt+|y<2W<21_g&Rgl1+3oG)?cy z{Q>Tw1a^kI0p6@(JcH^FYJdA`kveH`>WBG8gI3miShp>Y*SYrEgKG4aydO>L&g#bf z&e8`a@{6{l%;sXVhBCKd`C&7w0w7PcYa>ycP z9&El>NsleKd*6gf6DW_`+uW3CyE@NNGt`(_RAj^ziRM-A-os*Pbt}<3FYRB~go4QE zO{aFONZb8*HL9DfcfjD+-6m`n_$r$#31l9bBE-g0W^W3GwAyK=46lFD94bFByc^$d zaW{Aigl)<5Zqd-##Oq0hpLEvm5;p#sn$=wNKBLpPQhuk#0%iXPJ@rl)I zhD;P{$QULdkuKT!Xh`K zKcBJZ(3#D-c{Nw$qz5ANrK*|-p$_WW{?}%P2)unyF__!c6$w1qR(_3Z6Z%qQ0dOu; zNi<^`{NwmUc>h+{tc#7yomd*f4iMQB6oKI^z@3NWj)-llhB_NiESJFPIc-=TxD z_tCfI@BFrR_ON^lt0Z#XIDf{z)u&zivgW5JX91vU(s1`xeK4~!ZHK>OdZgARwSdCW zrA7-iFg(DSx)Fi8!MlRCo~w0`4J~jhN6jC6LTbuHRcNf#D;_~5mq5`ryrqc=wP8Nyf44VG@S@rn5PVnrau}Q%wzS(G_9c zDfj}T3hZfQ8|gueRF~uuw#})DSLMPBBuBVROnZ|b3K-S5jnfTo=R>8H^Hc$WEG!|X z_1be^C7?H$u7POWM<~tn#!x~JQM>lKUkvW50W!qk&*efLKLW#>2qWJLKEqoaQ$^2> z!5Ku*-kr_CUkf+~bWm{RYxQ}u3gkltS$k4XLS(w_4jOyVw*%-lYSp3LnN-ivh83FI z1)gSgsfo~Sn7FvfY>vxp&LHuP)9*nLVjbZpag}z>?aVq?!(B2397QYs&Y~!WC=vH2 z9eTZo+gV-p%KU9YV6Q}1abc{cLvF?ybL%vS7}_Fd z75(1eu7*GPRHw<^IdF7_gn06hZmN^<^EXr!Fisy-yUB#Dgn7#Yr7TqaNl+)WhXSV;`;y>fTf|25lac8Cb{BPAD$4Y*fs z7%nQ27Cl;iCoz(VDbD5Al%-|~@ueIfKwhF#K9{DgN2yufU3aJnJ{rw3UZ(WF#r~LZ zHpfVIdYoxPYm|0!m%Hzr- zxU0D9!j=Bqojl4CKA)gwHn-m^G`w|=Ik1zRC3vyhK>DtLh*zzhDz<=*{_K*}%_&R$ z7JjzEjmzn-lT??YrFn3XT?cca#basX(AN8@HW_5s45L${R;=C?^s>~KY$rBC^TLBF zCRT(vI>nWP*?#^UmA{Za4!!w`>ih?=h~)UdvDp3Q*X`PYl##Y9T(;tZ=|3p~q&~mU zi>%OIJRN!Q{NBS5@IxqkVH}0KI<~H2ym}|T?_z1d{&IZnPgVQIwd<@Qd!H|^Lth2~ zWra7!v{8!~W(Zq;{3RM_|4e~FtvNx6L-vwGN6$OnpI4$RxjD{mrs#q5visIPm>5AR zGW8C1A7q@DA6(xmUhTHU4(Pi27I$1HWpDZ_6$6uYTH$6&(mQ9YAt_3faRzK})@15Y zyEe})u~=hHTXj=Cx*q<{ac)kEPmW(IN!s+qpHdQ^7P{ey%O>IeD)FRFSf=3&BWs}j zO``Bd>owvy?pv3Kn+udaml6l+4+pLRqoD(0L}E;nPs)~(T|cB&1@gg^6*Qyx*-sQK~(Q?*;lA-%1&Cj zc|78aC3tR?B8;bU?a>FjFJA$K_&Jp6&=v2X4*tZh$A)h1Wg}maf!#Kb@ZxutfMUsy z)52-2pWdss0ChW zJ#KIxz-7#>2-VO&rwz#UTYN`Az`=)q3jpScyGk?@#rDO0AQpEO^I7%Y&YgJrbHS3; zUxwl&xB46`=L9-~KTabK=HwFcJNtCGfT{$O8FPCZ)xGf<%J#;$46bCJ9&wtNbY9A< z{X(qqUVJKHhXhT+;L<+5{= zj`r>T^(@rd8SxvFvV?77yD9;2_7kV{Y&Pe^?>;w~HnCpjT9d7WMySquK#fEv1GrWl z-p6DM^T@MSyr?NcHkb;DyZy-L?l&ocNE8x^TIpA9y$QD()xMV#EkF9?SIZB9V7eE zhN|;>nQXY^$c6RYrntC*4>ln}WPI|%f%a|0@4{9hC3(4NY+ruGai-Qmsod>D?u;cv zKdEAG1MkM1u(|mbcUAV&-Detiz_s+)u>Eg{t}rjiu&jgk1ybqE6k)?}BsP;zVhboJ zo;50P+J;<|;q#vky22NJ8!2BQ|7_3r-6rid!^^Ta7Z%tyv|OG>e@~2%YzS^mX1;}FF?YK;W@J>!czWMIc3FxBQPgKIpmh^bYkERF8b zi*wKB_>gjc8{@E5w_T&_b)cK`A>mxKUQ;?(S~1jY@9|=z*M28X_TDXqARuyqKh!1iE}_SZ$hN|a)V_A@k7SzJ zQmt;$*)J~V7jt(8lHb{7nerF^$u#0fQWszu^y=j_qV7%SFpBHC;Sk-zi>5nM%puQH zcb;u9^nCg?a(gFIy6sN0Q<(cIa|DBdbyWFw-H2rkyp#_5NyT+xpwzp>st<19KFu<8 zxL0Q?Z-zeADb(-(J&~`zdV@#TuS1fHrqH2I3g<)hpl|E3MP#f^+3IMQ8~xfA3DX-R9B9 zUH=IElr~peFp7<;xrvPz`+53xqCDjVqH~2-byCGn47e}yn^6X1(yr$uemjJ^A%7MZ zIZbeMx$N95=*u>lm8X7D&{ivFxB0U}*0(8`AmsPE zc;ut<)a%YB{ruRz3E9=5TU`eS6VLb2*R6D22bR9r3yHQk9a_1MyXb0%Nse5?tu-ga z6;Noz`yW_o$8LYij5JW*L(4ygNP}C}YNhm5C1^d4{R#@n>D?MN(?ihLx38%%?lSTS z48E{#)T}baoNIP<8@10_#BR%Kt1c0{ITVVRP8I8rubsoS4>7lTe5&YL6s>W9rwbprp(GZVgFk#y1cD`Z8yj@x*x<*c~r;`-d!Eqr+7QzU2Y?kW|-MH z|0Fl1Lo<7Cw*{<7x_y=->%`^I11*n@mY_t3LsD6LS!kq8^+--E-CH3fvee|Nc&^Q{ z>nI@u(XW!*!oiD`pqAH-xR@jC8Qi_upt;+b%+p7QmbEhF2n}&_36GRex74H7L;{1D zk2(Wv2c^=PrK5YL#p%udq3AlK=!V+H(rJURF&mbsrRMF{{AygtQTW~*cvowDmXyvc zE{$zv&>mz(fg6oo9uEa2vv|mzyV7(q`KKrLL$Sz<9@(lF`$vcmvKV$+N}x+JO9$j% zOB$E$-DPuD4}M>$JX(*DS+djzb;gW^Nj2Wh-1 z2Akc6Gfv3C#i0=2FGMnLa|+3y>+%@wyKPO|d~aJ2OOOeOw9)4pahy4NH#9WAA9?C- zQs_(D3kckM7U7*0c?`;#9*tIW1(n3?>M7AuUzh-vYuD={Z7t}-g zwSHkLs8;*Mt4>3U>pd8}`iVfSi#U!Cs2!s0<*PS%B@zOhw^4;^; zvpNmoshK-*KQc2+4Le4~K0Hp32s8z#B>jK_Pu08rjRMw?Em*8NxQhuH)wVG;JXYOGCW7T_A z=V9dpC5x2{j+K-U*UxSUod#lNK|JAOF%XluVu@CtO#2ZJSerJWa2!fH&4FMr8^PzQC{b{Vtg7iiBo^TgOPk= z%q+hOz1iFCoqQ&Cv+e9oQiCN_!NKfV=!@?C`Cg{{y$TAZ&7aGTE@jSB%VI;ima}p# zLAeA@{gc!w_|h(VD-ASxY&8M&SNe2lbdtyO+tzZ^%^e}ChxOc@IG$WSz~NNB zOmCGiu9`a_*$i$ zu;rvHwZkxSvd|5K>MzC<9phc4($uPoqM-ekVo4zT@znzj@D8IJmzctr4oj! zgYZvD*{>cg%9pkD%44G%m;+XUIa2$4D*_w-MdXynNBhbEUMGNa4=v@>#Ab73^ z50)N|T^qjRPxa}EqzzhR@(+BqgwwZGae#Rf`*c2s_KG3U0iAfsoZ+ceS0wwcA(KHQ z@;8RZVyr6qV5SMaaCj)YP`*0M8)=#s{?)%S+Ef#1)vm49T3B}u?C?Pp#mKCPHd6V~ z4|)FMcbGI*twJc6o{hiAmM#Y~4I4p!Dm0r;=>A?-hnBfBF}`S8ue92_afRMSCsKD- zX#?oHm`&q_=re;dhLzvycB0#;yqxWlXsMF_ z1;|HELlEar_=shIszQu4q8$Y&OxM`!&r=?% zr#qZXu*)vEl*Q;@L4q$P8MPNxU#1^PTQ+pvxr&#xU>oEr%1{2(Q6K_h3i2d)Z!jt$DSXFJDKX( z^lv64Z9iJM zr(yKt5J!oNK^`S7SFw(TMCumE^Xt<8?)?1K4pVYI(MpHr=M$oX^FDk=dv029#y;{o zwa4pQSCy=o?HuZ`9<`$$jn-y*VvN^n!MK_kWI3TT}e@niMjniWb!RMSf~N zR=$=)qT`6ItM>gI65JbqRIk0W>^8qW@f*6~>Gc)si(KYwcJi<>tlvfI&eS#UmR1?h z9H1G(xjxrFA)A@y4??nk*hx6*fXuaBw-OO@v-&zaCj2{#$?dt03q_)e`NO3rgC>xeJgiE=)2AEWeK5I z1NFmu2x}#?-$}4}0uXvlHR^=h?BB|BOV5mK)@2(Og=(gL%K7POCQI&1T5(?g3Bv#i zbX4e7Mc>zzGF0Rl*D-#aX^J~#*im%K=0~E_jPAz3`CP*9=wx|Z(@<(@x9@PVkf=%5 zgMBtzU31-$fprCt2ebZurIrZT#Dm4b&{mw8!-kv5ip?Q6I%)W)PV>z^fUasIYwCE4 zGy}?59^EQm!MFKpI4g045AJe*S)H-;w|a|ALQc}$gZa-2$}{2Z?g_{(He|SyNPz_o z7!xM0kmoUBuH&#=6inWwVcquYzkTcZ{~zhon_%f}QpJ#d~vH&7g-0zlzO1 z7>)u8ZbI(i@{56b<=W&eBvA_SI~l|1H%@ze2`1axr&X$Vp2stdOR@hrDy*At^?a~n z)#uo*Ev_UT|0zP!fBOBKTB#WMu4{#z(DQ`Xk|WjKi>WXmrI#>Lq~Q2S1my8WYA&R? zESzoZ9zR>BjEnBb&-P!U^{%e7j_;AqA|FBiS0BG;`ipr(i&G)$je59b7N_dZm%=+r zic`!~g`}>SqAb!I>Mb-N976=!=d^bTX6lb=fY@A&A<)otWu-W{dVfaduKQ;M<7!tD$1Jov!;e-MJhgeOtd^W|zBy+k38W*+$p(tYaB6 z0KJO$)oK8QB&>`Nq&peHslJy;nN54ggb_XvdKU(J+)Wud$1j_c`)j^u=g{GBCp+lq zR(?gYfAN(Or(cma2lERxBNnsblS=mi50xrYqmv43DezvVHLCjHPNm;R4k|0X?-KsE z&>?X&z(bshem)<1=n?0A=M7!9tTw!vd&y!8ZoV#kQO@0Y;nbL{$sCr~q0{=jK@8`l z-Zt|C*UW>J&bYfcvJv)fPD1**oK|Bwo}}C@x$li{OO6_z#2&6LVQ>r+pyk>j7N|Mm z^(W^@pyq15gByPK7yL(hhv+v@A>W~{Y5PCR)FJ&d8yGwSD71-L9}Pes*gtK4O#<(8 z+xTLJiT<&D-YqHmgoJB+lmW_MF|#I{j{xNW9wDT%PxiL23y>UPzQQlff1@G&IQm8O zbbQ6$%G2O;H^QK?Abo0~Cfs=0W(dSz$CzXe?gdp^Fg+i)6+|w2R#y9oJ+bM5C7=`I zt*mLe*z#dumTdRx0c_}3jp+RT=CpfU&6+B~^ z+H6J+w!-k>(~AUfB6gJI-iEVbKxl2EZ9c{u4d6FKX`&v=?b=QLl32nBaGfJ@l-P0A z$)e025OdbS5#1s{tbo?!kTA3zh?V!_1glYc@hS)Z3e=my_Jc5cTx4;lX_G|2ecNpw z@E9Gf58w-E-k7d>Fi;=>EF%Z!^B>m@b(r?LPpeNcOWE#G zHR!}|eiK+`wcqBke6no_O5j=Uop%Cme%YIZ27T3;^HJhWTmk?0 z!|D_=TyrwvhS3E}5P07RGiK%tv=83BBPVKR3pgJ0{idNf^I9kz2hIi)7rS?6mS2-A zF5Wg=l9%^F;9}sM0F_w+{rR{n($j8d7Op$jI7VTabC?Oc+`j3|+06$J1N8jxg2(Y1 zU`Z&|4p{d8RIyAw*av0pU8~U3t2D9!dJUB`t(rt!S#zFe~7~^L#M!cZ~=ZdNG zh?E2`3WjeF+MkJvczgTTwA@8Ao2d|jW+81*DY##BU(hhS+^p`d!|sEAXZ$;c{E3}X zfC?N6XqFF2lLA~~Pk=&_ahW2R_Oi>ufdHsjOrr)2z_em^v^&-KM@5y>O}L)!UZk1> zTTj8=Frpxw41)l6^4aw#&m2KnT|ava9iCpGlhb`OW191wj!(*)WmDb4W*HP-m^ zh5I~HUyq9ryqw%GIbi^H39i!@^3UdwkbK4p0{Zvx^u2|(u`O@2HRI2B8I6zhrdMJx zxUALIZSQpG_KR+>F#I0nFHqcp9*q61YOQ5JZ(avR_Ef4y`(eWdLJ_O?{c501gedRz ztDAfRKzWp~%ops<_b*#ey+yIafIc-no{Mp=q%wxLxVsGEeT3G7HhoS9tdI&${@-S< z_n620owvCCOnBm}P4A09E&Pi~x0{&$6zx^;UKDNzJ@5~fSQh>EdJf#R1N61W*^F;T zfurlj?wY`p?RJ`|Z_9FP&|jR^AoRs7-iecoEwpxU}RuKjX~xF$Q~lT7U(i(xHSfb)ZlG zBp7^qFv;Y&9Fx_A1W1@vFY%ZKI?oYyTH!jM1Akwa%1sKInJ6&-Q6i(G#wPyDz?qCm zf{CbekGcX6POo6CV62&t^wI0ao!XU?;}=b5IHtxS0-S2pY7#yW zeI}eyVN-VmFb5+ z*Xz^wxH&)*z`5t_#2vH>r~Y2kTi4<}SM|_FNhVcLdBPETC=ED+t+3Ja5pXquu)I>S z^$GEyryIY*PagzcMS3QDsfOsiWA)Q!{#>;|IQueavt*xz)2%5b+9-)z#womK)gte1F%l`Mwh`6bI?wJVxZu`V* zYgh8>h@(-L;`o@_U?{9RSuih2qx|Gx_@8Ob{%v|Ty%0n zo#M{-haQ$VPk!NdbvbCJ!)mqA-}P#ztEJ-v*K$7jl(xv(^|`=`2;icM(wek=tNegN z3|yIU_M&;pn_sO+}EWsv24;4r}QavPZ-2o3jOJB;OI22XFbMg^8M%%=5@dW5T6^Z zoHw^@yOXNCpVOOre(~0voA#AWkA1P}@|}2_U3tKHueW=S?gy?j+z4DF6RNfnO=;S` zAXX`+r=S+q^P|@^MGE%K-~yeCC<8qg(M;&64)|Eav%n-`%6#ND#}VNEFqvWoiy9ly zLZa%*AN|=y?AMpvx;w|cVfXc&!b#tM*~$Oy04*K58LDz3PZD_2kAna(ZGy5kOA~0D z=dp=;W$30uqOTiJq6=;4o;7 z6GV^$w3ArE{TuM$NT|L9;GVJ3nd6`UJbeM!FS^MM zGpC{9A<+Id_8AXB>joi)v4D;^?NP1Zvb2*e)9u^u?^gjNT;a6^tA!I3es{3 zZNTlgP)7kPz|LZWzBjBea~uv709UFXWJean0B3xvo%kRe559I|Ns*Tp?C?jMBVv1qb}Px(B-f{^j6B z>)k6?UR_Z@NuxcC)>857Rf0z@WTdtQS!3U_So4NwBdtC>V0asbvLGm+KoIjY$H3Tf zo^e{FSt!4K^4x;OWvT31S<2@oAFB9k(g^&EA1x)pH|w$pYQbY_;Nn2`sRDGlE>6JAa#jZu(cg7 zE;sf&AEKuc$7>;w4o)OPVi#Toy@k&OeNaRVSuOXca$|Hp8d&cx^?gnfupIy5Ja4!6 z=er>0$cg5G)f=DvA>H8|C92h51#$HD6IGQa zZX1Y4a?lxl=V4E}G(?8VX#NS+pJe>50l*Dib(Exv;tD$*Ue z^T2Ax*!z5cqy((saLH34zcb5ppQQmIvo=^VzeQMIl0avX4z}RKI#?@(NNB2wFmW}H zclqL6`W0R6-@0O$)>liLs`^vK>aKY0eV1SF;V7-x9gv#V%u!sF?|I7>O}m`p5|!#c z<(VMq>zM#C_VQ`k~FQ zF8yi;`)hbN_X>w9tbf{V&$U;%t&cO&aUg;oD?_FG5|--__2DmSACLa$+Sq6N(K6HF zCOc_e@u#@i6FNdF49{V5RM{#`o((cr0SfZxK* zz$eRr4da0nkeK4ayy%Ag;io#e%D3ylS1#Ew3b%3VykqB%+m@p> z%W12W1-7LP!#hiTw#Pe*pOYPGSGu;_$kp^p^ec-M;#lNy)tzSIGJ86iVWGXnD7w$}-j>ct3Bo@}2ea}xkD~aY$UK7(y7JocQ zhtEA@Lfi`y##t9*7xMSS{H}GaGHWL&-PzmtJz+EQaUCbm>U*YP|pDcnKA1&<%& zBck9(9|G{j+d9EUTY_1CI@w>Huy|*sQ(gXUY;% zSyIn>-rOlhr;;I4ppC-1?l^r!CKdV%P?wR{d)~9}yL$7az_mxktvN-#0m%hC6{MV&!Z{CnGs^aHv!__Nh zwn%4LGiBX}v~@jCxj_+epzCx;5faCV*w>ld;TA#Oix>%HyJ$F^`Yg6t|8r-ddmb4! zA~O-)EnTQnpXS(j+(Ep8o)@M^!kopy_K{lRYTpln4^fXhqcf&F=821*R&T^8-YmaX zv1JuX$~RbMDsu|1orM0YPK|cOs5f_mg;T^mKZVQSJla;E0RxWI8wv^`y<4*tmaJ{pWh*w7u3GrMoyXha^=U9C|Z$Xul>J9l}|oB z3$>s8!nacpD*d9*S10Rt65`e0FMzd9e-$^%M1?Yyh!ZUq z`!r+k!d`nIgoU1&%u<=J6D{BPEYTY)BBads<8Nyjo!YV-%y|tuA6bJPnY9%*L+r*= z6CVx59G6M+6>`buFDrIyU#^&G%`S^kD*ZvxkPL0;O6-3LFds0?4m3>825 zc_~iEvYUxWbYiv*a~1aJf~RvzvakpAP}pnkRK~7`(IQYk{>}9j>?^Eq@9q=z?3YqU zGl2>G7K$X}-_j)Ns*Al2)>b6xJ$)`z%R=DSM5jK166B2|TIA9%JWmAms}iFIFG0(G z+KGD4&*_fsj4;w^XjDwN8OZ=)RH~+m9xvR+VTUEz4Sw+NOBNa^(w>{f&m?WUyzq{e zGHKZcM|eN*tOP3E?6(({@DWhmBz4=Ew6-1jSozVgZakc)@sGmqwb#E6 zPTg;#ST%v`9{FxvV$Jb*Zu{l~s0M+e-yU>9u5%AtzwwkN!M7pkfneR*XUK$&)Pt^G zZ-cz}4m#2XW0)WXA6{V^Eq%oXJRLXTlj`eRY(@gO>>Vx`tJRUxcZp<-Q8C~|FATn( zlT#)9+b1+#ady00C-VNX-R@$4#CT$|#>J50e*EbX3}flh%c7iU0uV|$=*rzbmO*X_ zTu+BNd@8hq;oG6mAVLnvxpJ%QC%n9lbKV*K74p(ae)(sxxo>etHkCTXTyyU9V10F& z@~4PWRCm(f$(ql^bmGE02qab*D*kYp;84_tq|eDwny^mO8-sq8w0b7ZT5A>S*^N-? z_UK)4fEASlSoo|v)`52*RjgV!S1$67 zg@9#$1O|&GJF`$;@!9PYlDzKO_i*IPalO-Q%fUjNTHy{WDg`=P*x8OK(pt0ju#MlQ z;`F5LMqko>PZYdoRTrbwlc(qqOhJEzxqy?A(l5m zElqrE9^MaNzEY}aCEf?g%d_1L6{a&ZjFkT0uz6$1KRu~=VKIHx9e`Btr_pXdi=TWx zljA|e3lHk#*q_WtOL-B-D>C$Els2B&crPVcihqf7l|W*T9zpy5$!QizGMjxzwnb3L zMO;+EYj1EtkF44A0(AAsKWY!(7_fPX2 zV6_8qH^&nB%*AyTnI&li?aC~A6Fj}+wZ$ur92$?GgEst8mk=COe`u|O63Dt2(V5gu zelz}fYn3BLy9D@J7L40twXQ`ahIPY6e&?=2XXFynxYy&Q+ZkY_QqV*h6Nk;U@#8$z zB=+B`h4cAFcPF{>CJx4bOnt!nFti|rwp6D?zh*X_8aH-vtRX~AQSu?ej2=3fxGc+& zG(!4$s;QG!QEvbqOx^Sb7cf1XbA3qZjt39#KFM9tU%U0HQs*zJG|@R8JX;EEYKquV zd1!s4i2c-MBRGT1ZIJ*^%%g?xf;2MrPZa>2TmVFmIl%tWnL4N z*L&`sFq-3PXURp@5A}=$!^{5_mjJk`Tdl;wsCfdD{LS8UsixnqBZeE@8JhQE>CE7u zS}nZxn<5$a9xBw{HNA2#g^=|0-+Jk(v`7lEv|=S5p(IZJWHr`%to+D{ndXkY(xQRC zT&$HB6@Oo&rSFnN>Z}ZEa^NtwU4>30>_vF3>*{l`h5LU3!KE9J-`rkNFCkMPsWDuT zL!^wldTi7z*A7}3MuY7vhkiqt0Se^6*>pJN{ri_P{8{9khjs^RV^!eLcR7-XDAR9L zMRF{S7S_bs^`s05T8@=gg_P2cnui=XR#$S2O!O?;GSMeyGh_qexqR(PJ2BVpF-e$q^sYKLoA6C~mWB7STGDW0tj_Rc zT>e~c;w@#eefOtyyWSt6f?mH~((v24Es!&_wP}=n3OT%nqtig>S?ha4}B-*vG**6iAtJNx;foo2u3@_fMI%j^wfMlNu9)&TcG zVZRb;e}1l#KFV|@c7<1`6(0VTSb{~kto7zbiV68?gJ~aq?NHBK4$Jml?Q%tn%}+ZP zN)x|Hw9EMqrJhCqc1eF|B`|1o`O<4afSD~j*<6?;mkq!dYh5GHoK8(>|NnCca@Eo| z=2iel+>d7-5Z{fcehd3M+2HNIP#qF0)8zQ+T5{dfQukH;@*>CbJY#D|Vkx?3=Tdgh z?k0;nkQn&+Thy(|i80B@o|;Um37XCqP;#WW2d(KE`&eif>Fgf{pNG`TLkC9^IyhI{ zs!RA<;@sbbVLvgDYfSUE9#&fBuIk7=zXNwuYqn5I6)Rp@ELXG`h}#`*vV0wV?e zvI5YECmWMBT!3_?F#&x+i)0?99Gvb`<=Dyo^kHML&q;^*!C;9q$RIMrp43RYXbIMPg!^Er9rx#(ufoNk739EpAqzne)kR?9-Tv z&~IjN5@{KrE?Nbz>Y#J#ZH~?#jz!%CqZQ9nuyC{ zk45#gZyl%=&X6#LF#DbY&8 zp@E;DRWtj^qyP3Nf4#rI>#8CB?8VYIYoO?CA4~zxU9V4RgoO#4cng+P+@;YS+lj2R z`|0Rc(SCD%+AQS2zKp&z`ZIm5YW0ee;8-D7U4*{c3%Dle@_Tv&Q*kZ9giiq@!gxjN z05AwKBbp+UPw~&JJO(0&JOe)KPG_!M~b>g>n*q%?AkSUe-(z{%7sd;jK zvj2M&1H_-FKxS(o2)LapTf?|5jf!EpgrpJB&p82bPg{3gEvX*=^F7OO>l2d(jC2Aj zK}fkMT5HFq^D23UnQkD#c^5m+cL+Mqp-L4)+bjXy;5IP!n7!@|L>39DFqBK~q23}> z5S9e07Kn(6`QD6^7zuB;CVweFZXCRpmZG|!483U!zJd-A4o&|8Ajc$=b|wlxS_0-y z5&U2IC5Y4D%b@%QJ!VUq1D*e7x-DtWd?D>2zefRcHrYE!@W>E1Zp64P#_Ntl;+Il5 z;0`o95NMqLM;1Og&&xH%%ts<0L0X@$OlOgR>tFQkRPvawc^#g)I9&@1Pb^a{o_v1p z`tuUN(Mhp;{~PmU{(ry+PmhHer+gGG9j=J)nTPSy%mbr2xUy;#jZnYj@dCLqew?VQ z4J5=Ao3_5LvK!;SeG!ah`ZsMXv%{YSd5ObG7heHZBC>&Bb(@X0OmY!rzw6v>Wfd4- zVej=80k|;L!xRwEt?;6u)@{9zlLs+J1ht08Zwk-HEYruHIyNk3q<~gdnU^BaF!I?_ zt!egW02wP@p?8W)p-;s=yi7LyRwW_#m#^Fndzz&;4^~IiR(6nWx1mB$-LV#O^X^~o zDX6iZPG4Rumo(lWzH@{g18OxBJ!2*=Le`B!&X0v<%=z)cRmNMbmr&RIY&j5xNrW?Y zK*+F7{jMo)6M(IfUIUR0Cd~nEm*oe75vCiI<24^VsA_MA8#6Tn7bk6bMo?ZCXl3_Og2pez#a^hagk1m9qix-iKBAxWg^HB9L5LWKi=-R|Tw`B{*}@mDWEyB7b=J zw3WAtW}~xv28(R@;vBLw<+%;YK91&r#59sC;VlIUNUD!J1iE4+Z60poWYBdg%x_u`TpjDf0SW03(+JIc4nOh=KWXs(V_s z68;>s1CH9AY1V5^|B)>Cki4cHEe05|iZnelWy*WtTexxHT0l|F)y=8UJKV%w0aTaA zoev1s+J~|+1MQL$`nl*+*i@TTj>2O^5ano zX?_B6rR3KxE(2DW9#Urj)z+J+viJF|y7Vn+Ju{qpN1JzzZ=rtf;m_-XoLu-F$FnY^Ue^*>zhJz&y#1)e$IK z2u??QGVi?kwA?J%=WM$(ts@mMC8`@#22TDE^D#2a*g;x^5$M$G^1#x+%Op1JCYoaj)8^Njb>ZfQy+;ZPVqyQ_ZAa+J z$pdGjryUJi-C8JXT3OM@+MXg!%M|C=)n{>TSn{^|-#gbBx_pR2GWHI-TKEP|d&`V2hr0 zRpZ_H$>B{2&Pp1aRU z7S90sE>U{N?KWXwU6W4`bR!^)KpVj263|YEOO2(dXxH8=Ma={<7QWKvUTbIiG1Ecg zxW}t$M6#}w0N)h_Hu0hnTiY$|`|AVbNLn7~*sv)tT#skNVYBReZvpQ5b}^eyW(s6| z0?m_Ys6dIQ;TLe67UCSFE{?Kj$_uOz9je*}IDW=0Bhv51E?9{60qWWkoEp+KUS|xe zT1>J`ThXpISyTcpw>`Mh7bk%uAg9q82zIYU8pXhpFL?cGT4Fz+Dc<)hSoR>;I$!Mf zCpr8t79O7h?@6tDs_~I{OpL5W=wu$$SC^vL}T)A+j981S;f=+opkY-QLs=DXK%q5;cM9!{=k?BtnG5m8~iZ%BI3;7t& zXpM^{WQd2it0W87IK8D;ggG-pLx_S?epp1;sCnhzp(1ak$Sh#J`l?I(ooc9~ z=TueoC^Jd=s@@o~Nld5Yv1A@C@f={igC?p?aAZcwtv>BZ?xgpdoWks? zT-Fp6f&fDQr5}o#99#Tanv($0778@L{|t@s`?z@_@zvG`#_5tilTT5~kER=aes`%n zz%jc)E~L%Ca-Oe=aDeNf3!qvXUmS#|=Q~FNnL%8S3pL9V#(`V=5ZV=5T#B?@fR{b@ zLkodW>S^!`jt)I!<8{|6Z@DI0L|pnKUn6U>C}^t?G-xSqfK9lIquk)qkM_g%AH?8s zi8iW&PUKwt$rL9#jEYN4A8T0WR#|S|Ii9T$Hz_828nvZfsM^NBa0b^|BWvT3rDiRM zVWk0GhqH!3bxPG_DOWM@N=C!$X4*{+(2{({iDOVooLg$94D!k`II8gkE+PS8n=+GS zHaWl(c2IBH>ESR$U+=v4&R|ZyMM}v--6u0lHaPTB>c}N6Bt4ko1&U%_s_G3hn>@4fH@a2AFjL@Tn*g>`siOo1`I*hxeNFCx=UvG3Y$69HQhVl^Z?N|a*M}k zZLS1!ykmqv?|Qu)h8qEFj$uC*-{-dngW;Xw&`uDv1S%4-Eh>M8M9q(N8=_Qqnxz+S zL{qHuWfEmxjcK1Ir>%eo_6$kg#gMGm1Vx9le0sGhe_vCBN&Kyj>DD;x_F_{3S_3~z zqI;U-zm?4Sh7kkqwROIPV={q9cZA8EoA(Tz7#@s%mVZhtrSU-hn1Ef57Tqiim%Dt4 zTT%;s=&!hCko5>`Qq82^(!X9B87dV!(EUEMZv^N#(MB_;q+X!+a)QH4Y6-6P$sj*r z?igIT4WF4pgBBJIBw)I+5-LHsNxhE#>o%^z{%}m>f-z~qqvXwCDx<9)8Z!{FiMkHZ z=^Tt*1B;tG3MYmU#&$Ec%R2Ga=mLBHkoCI}Aq~1L--&cZJx}#+`Wg6$X6ZGYi!vZA zd5Yr%5YQ*Uk!Ljj(1f&7B=(ibB4OP!{3iMtqpYkGgLqe5%fekbCcz=0aqq+l1$ove z#LmMTc=jsFe<<0}TW+j3_8?9|yxuvXN%0>SqwOj>&dS*bDBx-7C=MQ3*OEJzCbi@jEKhu09-|#>o1>?R7mR(hVhqTjX$u16i^Jb zsI6~SQLpad&s6L}z&~9xoPs01dHc$OVDRlAW>P%mYKNi5ApMUG^<#NznHo=cO3fUX9Gu z&TzInQEuun885VIUs`(u1KnbLF*5kmiD|+8HE&s|QNuX#d7&ykX%n>k zKth`Q_GA{BlMS#6H#Kj3p3=Jw$IKbMn`asoU+)Oo{fSTEeeKK=1==3R;nQx`9&^)MFl}MP}=Bf+~vc+ zmuh(0oc!;wL?2>Xr%9h%weq>j*O8M_l)PyX2zFolTU1&FvOv=|!^PGB^Ei`xm652! zNR6R^n&k(Tbv#=eEk6L(dJW>v)t8)17ng3lzung>M}S=3@Eqs~TQ4_x$)8!1UiBYX z(Nj4DrOFhYn;1|JkVXk0HQZ*NJa8>ANN6qLWW|^y;8H?0l+N$WIh`{rUY_!=Ufz$3 zl}A{MfwbRRgH4wgr|KU=6kyxL;Z!^aVRv8el9OuErWp4Na1B0nsA-o`#k)ys1?0q7 zOQ3l{xez!q7_0beFhiz9KALXhRQib5(5+tXZ`NJq{1}|5?nX{G+4!XrB?<8U@oE(NyG-uvS1XQ+~Nn2EBXH2uSvGaPMo>K=NVk8mK@>n0+A- z0KVM!*lNa;u#4*o=t{;^;330Y`GI#^0N^SMOVBUi=#-duMy37o)@n&HZAO;aH?^DBAmL%<3#_H(}`<;D{s|sajz{VwunZ9Q=6IBBwCi)-_(LOEn$1)MWe+2UP zNMW7Hdudy+1Gr{6lj7BDVuUfFx+W6Qxgc%818N=&0ZTS{NOW9d<5)KFM-}m-m7RadG^@v|JNw1i?^m1zFlpXr7SDfHUY01~ z10Y89*(y>6++p#O)Zt0LoMgP^uw8!=0_V_jdKpiU-}G7h1{Bu?4GL4Cc|t@cWQbA+ zCX24$b6Y4Ui3)mAs9BNSm2^_NB8j>YouPu&!`{m;gA`c+_x4IxHevVssmMK3ND4L9 zxKTmQMnjF{|H-eIt0OXoMnL2={cl!no4-fbJD&9FHAqVn$`M$hs)>%?=k?rKC4v?4!rYYklb9wzXacQn;+JbK`4^s(u}OK8G{4kjx5*z_x_LqMmkM)`j$ z@3p3z_Kkqoj|&Wl&(FQGA@GlZOe=gTjAV{t{0l9wmL8&bYTRqKYP_YNcbF3xyu-gL zH!toTw`4OBEB^S&<>8 z6m*-@uqR+wX`^=X#q#9?keb{zW-RP|V0U>r?Q>{ote%39dCg?@vg-|ELms>q0JqL8 z$VA_6s@^lr5@oTtl=WY%D3Sr%!1s7j=hHI< zY@7meZqB$5f^}w4NoF#6 zWmNMbrvVg62a3*B6lD&`)k!wUKYR{UQ29owt`Bnqj}_>J98z>`^EG3vR0-dlzbwZs zbpcLGT4nIef*)zb+3l%MUDVf%q>OF9X9m;sC3&Xq~R+v5S< zb$&(nuP>VqevmdQkxAjd8mf`XPTAqw$Q^_=hVHZoI$?BO&O3}9pYndkX?H*i3QAF& z1el}Rnk?uM_XD)qVX(%_X_Oo_K7W;meSHT*%Cva@ggRMT95aGaisjp- z6uG-WQv&~;Wmm?XcHAjEdd!SA*|))G@}%yslHT9NRFtFkoGy6N=IPD%Z&I@L3pA&C zl<4&WXF&~`2w%jMc`O}h9Ke(7dFuJj&@D+f8~gAN@bi+pDd{}T7<57-zXnAw6Ol{h z?m|Xd5K|p-U~v2Yx?wA^FYrR>iWQRauLg&NbX_A_(#lhf{1*k%Mj?1tr9&Bv|n>E=Y2E`8a_W@ObkJy+z}N0fin-JhP}x43rp zMgPfoSGE;R*OmvK_ov)G<^|yyNUMZp^tHh*p&nrx1KIVZ?#*Z43gdukSc?r){;TSH zb{Y07_j;B0qQ~mNb93**zFP;wR4#U-B(`?Tn|?aEH_sb~s!CJsT}~cWA5PoZsG%L7 zOLUw)3tVRHYokSm?WzuiCRhVg%4bL!n0S-#I?^P6%#uF6Oc((!4A;NrBdxvwg~qR@ zpyl*Lm7hl6t3%N-`;qRmVwu@h=Jj=}2HxvbJOyyvas=E$iC@!KGJS(_p9ho zui?(X6Mur=_6zl{RT#@3jjHwP70OSm>{g7}#tYU2o>e0{cC$*}lOo%=p~L$IVhb*j zP`n?qfUEZifE*U*700S}aM$Vb)`O%q;g^HwYTngcmBisqmHvI~qqaN`Kk*ZvjFvqi zycwQD93rbDds?R8%CYlU-u8Tq;?#8AZKCn;)!oN)3L+>nXBat)@)iNbPj47o&>Fd# z_-u1(c~zD=Yh&NEML^}^yzaxG974LYHweB_NwMI%tWF3JF%BSh!zVhx= znGr-yH9pCaQ{&Qm&vQ=Fm%~vVA(Xj5jLZN;TSJcxu1tpY0b}v7Xp}3huTY zAbSbGe~Xqeas3uZB54Kr3Fnx^pRGr9fT)o`FbP-4S>KB~If=Fx#Q1a6(x;w!RUE-} zGfH36j&Iu6FHZX|*w^EQEgiTD9ryZKX&c=&Q*z&^<~ogDLaFHs4*)4}e@1KE3x!aR zZ#j~B{*BskTLH}LKM4=mbau%|xw*U}6CzXCY8x^q4%eNpkvx%b zto1Y2Y}lVRk0`Yg;0?He#J^syauO%7`9N$e-E)W+y{?Twq5mxyRM82s05~qhlSK5x z*xc4)YVVTEEi)nil1B@BoHZidR!fd!Q{R}`#QvQDPc2G-IkllxHxodJ35XsN_sB7` zU-=B5+k`$1$BMLkDVSXz$p0lkFY|3>*?3h&XFc&u5s}$qa9w2Y^wvbpENk+%g86B_ zp;FPs#fe5=S-zdo=FJZxiN*si zAP+SL^hoP_N(ms^Tj+^cZwf;u%D)H;ICY!ASk#Jv8b#V>yReo zXmqH!-GnA8fZGgY{5PI;B~FA(f~5s3X0nh*;|AQ2VxS>yj7VDr$2ToMj;v4(^!?O# zdedvDJCE|Pq%tyGZlddXo}fHCS-%+D4rEAaLr!-kqzdK%BM%6VTu>^L8e$dCZgMf7 z33DPkyaSVxDw$Dq4{V3O)w*wPa4|QsV$Ndh8vRzuCh9J{SI08_T`>&>4QI1Nr#3x` z6Wga5xZO+1_7jD-1+(O@+5W63m)JuQf@DExnE}(Hb5oROr4rMMreosO7Wm=JZ!Q zSqdBuZ>w!mCF%Azl>Eg}88(zBh%1OT4G1gD6_BEcV3j#`9DTFgEl6r=->NqF|QDUjGE zsMPZZCv#`*R)i(qV8X<$-FFx~yKVcpMZAD2oWZVuj2~A$19=^ab*B1V98bks%*7rm zwdi6;|Lr`%)P0>s@r$E*2TrN=il39-GD^{tjQnBf-;a$Ld`XppvVP2EHMj#y#G4jJ zyD*fTwRV&Y*#_ZKW`O`f($KMr&BxsxH0>v(%1ef|FU!HZuM^^@NOn&UWfjJqs)M)# z#{sb7KDqRZbo$u0A%IaYx0wv*gUr2dB8N|C!7h3)P#vJBQ+o%4ATabyP4YOK<{dEi z?jG~TM3@9@0y?!FDlzVb&V_5EZ;>{KA>oW}kQ$j(?tSRcgs;``1pvARuC|LRU?q+L zr+(E_wqhT(V*#EiT+l&iPp8_!P$`Bn7(IA`Uc*#4D&pD6dqkjUZaUMKC_41rjP<(@YHJ~11!Xie+!YXQxEJE z9T0#hl_UMDS7stjBZ>KP4ruPwNvr(5Se|+*TJ$X-L|a3_TpNmbHvU}j?1<(VjzB#I z=hs=y=_g?!toZc9oXp-8(|5{GoYnCeS?>!UZO+&>e-yh(H?X^fX?CIK5cdUv$Ae#5 zYBj83HWZA1)4Vsf6(q>|Wj(%f1h1TZm#5zkRbtnIjKtGitJxPNv?R^0cMOV}A9g}t zK>OVO?bqzugIt}_Ulg@?jQFDJnPx&Vvu%)v#7g~VC|df!R!II*0P9Pan)W@K#WSWj z148+!w9@N8{xdU8fu&H!dMb)3IHhw}T-=x;KZ-Ifa7IqwYt44PBfK*gxF@~>ZQe{R z*EXe!OsU}>;FZhSdD~_O^EALK%d@GS*G9+&ssS)Ab>q@2`&xQCSM2m~b92ND>{bKu zw}<1OER+UH2|f;iP=+V&wEzt8SuVt}Tx~=D-#D+Pe>q<0%X4DC8u7t}d_K2@7)wJe zln!zZW66$K<>Vy( zvVv&!P#F7Fd{ca9fU4?kV(uy3YMO1_?x67{N1?X$Na;MHNxxP+gCa71YWO<$Zq9;J z@tS0bc&TES_vR*?k=zO_!Ku#dK4oEc^(>?PE3VYF@Q8Q<$QK^XalS3-DU2m?X z9H+tH$Ug#jw-t7LctZ+`4>;?chQGDgr~tdgdblK69s z-An6t6c?)NtY#aQL^E2(^3>>U!BdsrnBBf9x;CMATl$OGdL5B~)dxERBJdZ@=D_0J z3_S+-D^Fsli)k3BgODn8`K{u5<**#q5Agan=E(>+2grQv1UG>U_*T!&^lzl>lMftf z7HPwZ$L>q^pQ%- zgU#Ai?n^K+s|Iz`5Bd+?Hw+xBCw&foCrJ7E`iV=vhhEhIT3;X8+NUg3vE_f%o*blG z2sS83m=1o$6g`X$dg9Qv4OZo-(0P-N68BV=XOfK?FFf( z3@i<@42rNG7s;y++&3ntxaGR;VZ;aCf$(3GYE4wlhKIOz3BUzo(Cd(ql@HQ^1_rZD zWXLn%G-DOunjQ*YM7HryK}cms5Kkx!kj`UPX9q=CDgZ0TkzBLV?XQyiz}wFmA$|v| z=Ez0hZ#yB{7POtRAmj4lfVR1LclU7XaQkrQaM#6&JV!IQDL9OLh&^JlV&8){bg2UG z9p{tG$$U6DZt;%fTs*RV_u_>L!%L6pz$|c$eC?rEg4}lblGjJ97hL1F9(`l_THtXV znW|$1GTrA*n{}I#rLQbN93VYzRcuG(i)zhMfexbTNV4trh)}!$#;YQa6zGspk>I=b z`biskV;p4G$SMGHGkP+jYuofJg~=*4p4)I-g2f{#-=Hs%Z=Qz|hp2`_r`Ty%xRKp~ z@yyX_YZ;TesdOal5H{Z#eYm{Hg1R0~fGfoApnpar{ShwI2f4)U+GPs|z;AvT)9#{v zRd{NJI|c(|5h_sRYY%x>04y%wyMts)qS=)l`~KM30Wt#GkA#H^3t^;N-4CVXW7%1r zyL_w$(UJ-9a{26?t-x%hqD8_g1z(8Jn@(PJ=`_mTQfo0Ghsn`G-BO2l1}{QOnF!h_ zN7!1zNUofDBhQ>w1WcM;OKr8@bSkmNN+jYYi_r2s@0KbbPNPgDHl1S#DU$4Bo-oZV2|)%(%HhQ z|CWC1l$D9oM*NE4ioHng^y`~73>fE|`n$MhHNo5o;Oh?JY9uH%}ReWX}dCgyz@crT47-Yp&GrVY8v-{MV zuqg8FZ`X6(P@*H&3{?R~B;#+;=z@qJQ(1kT16`Fp4|_u>>iviw^IG=|bj=52g_yHZ z`7kpgILoBTm)WiEFMv|}k|R;uD@kexSoU!^Fq04X98tf3;&IN>%o%BQIBiq;3;&P+64Lh4+A>iFIIG)iUZyU?d2_6{?__;Z>RlJV6 zmCO)J)qEO_UGmBLt^Dl@rMkE#_Vh!g{C?;vMW9Ie4fJccfdQ*@KRNP!k1bptPRmNB z1aITS6|oVQ3JBBTAGsY6mTnG2WZzUxM2>7<6GvfPbCW<}9rz)o;LZK8BzT7B7`^Ku z`8cI(fEA?&piKtj(MvUMk|ZJMF_7`bWFnUH1CH#&u&S}o1CpyxL@KzfCY-g9gHE4k zL}Eo18NQfqgTk=2yXCUI`_O*;7svDZD3e0=Ew%492pa6B4DO9gHD=?0!}sjOQ(|rW zwr2Sc7!v{sn9YI+C~?A8lSuF?y5XuGg{l>QDY%-OKF?gcVh^6u4xVJ~^ghBhOCR~# zBzDoKQOWo;@-Of7fGaoR-rQAuHd2BBwL^>|YXw;=H_kOF{PeKnAOm0!Qt6=6w`MEGvBwJwtl#3&pRbFCBrUMvEyMJ%KWuR$g~AF9W&p6LfsoV zBs1-7lK!a0a-{&~lKSIg_K(-X3)754Xp|pFR$*}7Y_`sK&d06VlFy+)b?r`rK=j_7 z5+3A@f9Dh=|qnd&T-^u3Uq_p&pn(|F>Lw`xvMZ1yO zGZoU${c1Vws1)+rvvTy~+n1vZyTvsZnm{+gmL}PDeU2?(GC1wcn7wp`3la*+P?1Q|hoVLH!SG+vG;Qe*gvN$4fdu(zKuOBgs-G{75=0^w(9t1qh7 z61$BO;60~r>RccJqhQM&s#kC3L}g)&x$qs%qvwAD71_0gV~4FRV#R_JGsC`Bu$yUV z&@m2!%g}x6j1tByxx!D@&uXya_C@;pos39%N`kADZB%V!(ioh8u;d3qR&CE(3~e#v zmShoXr$K>X*&zzbQ%tw6#`zBH-!7zR`G3B)nTgH4r%o%!MLVTCipdRCwDFx>EB ze}VE;mu<&JbhO>_n4RR#gAZMz)UDxX&c) z`eY}0TRCO*38XF#!)Wn*hk61uaC$=NtSP$qH=-#wD^WMK%(SB*G~DCpdYJVm#-%Fi(W%up^tP;9LsY-&_V)s-tqo22co zOw+GTglh+kP@!11Hsf>z`0!J*7;dNHPXPEn=5Ap2P2P%t)C~;f69ey)-QIY+yW$!Z z*Yi=%E{q#jw!e#6?6T&khO

nFFr$L;|fwDRbo`KK|7 zqMloHPLkHk8h5i(4?$x8OZgow5wipWWq{ew&f=`12DpYfa^(a4da1vM?tZ9L5PR>} ziF%~|Bv*DC0Q&R;!kyniFb=R0PhN%KwsLp8h$yU(PgFO5M@24$&J2j>ZaI6d`ZRP# zL*HjIWz0Of4C`-lu|Ttve?sDG34m9qYA~~U{XI2;Z%Ul5gp8hN;#-| zygWh;uzL5?;QyTbH70_67omrndnfR%9jOHNK09lLHsAh8fmSC6Cv`7r`2q%E(;0Dl zK9eEhSUvRWvVRW+m$xJlsV*`nN$~#mJ%eiKiNE)$aMO-TC@(+(s%*_zDf>byW6by9 zM`gEy1u3<>gQ(0NQTc@=3kn^+GxF+gCtr?MJ4=xgx0uW}yAsVl zXD@Pc!aoOk z*r}^yEvuv)S_b?XRs81n`1)!`io7ZUsUhwG$y)HK&i~WiTYg3PeqrARN+?|_8NK1FeNJ|Sy*K^MIx9&gTd2z3G`)c0E6=z)M-upN{ zJO73?Pl%RAc%TrZAXI_tHDh&+-w?KyCr?gXO1QrP_3~w8$$QNv=P1Zqo}cENoMEQe zT&M`C>T2`*wJw4?5w4tx`#q5UCbH%NY3{qUzdC2`_y&*Vq0#T5*-rWile(_?UhrWo zJ-b^xn)~Tz2Ij`gduS@fFun@D&Dn(u>#0Kw274Yd1;DLe?Y3%ft~nIRK+l;sk)j|bJK;k;pN@S z+D(Xy+E3(LYZj|AQ6OkHGl4k6E6tF3OT=ehp%?4^ENOi$c|BDz2_`YdkSoh){0pZX zWrvT@0LWOWd1DWb_VlS~6S0jrH+Awd&9?wb<=tjXCjR5$mwC$y-#@rD#S#(R;fJ;$ zvs!gZI8Q#M-Ea%k52{x=yWwV~22!A~(CAmBD`Ln(_9&W1h+}iVzY@x*3 zZYN8(TO2|L4wOn!>m}L^Epf5$ zzZ82A+1LF!iJ$HsQy0v{7-JM`_PhxCSBDxv!Q7neYp*GatwP&{*NGFTRHk%eCbM6) z87NEu=>Xm}33lJVL%`&5P2}4AM&{2E{lD9D#JOJjg{<+S02|K3#2hx*+P*MQhg%$t zMQ3;CM-G3-zZhZ_ouJIs>@X(3BLlT9jUG2KQTmkKuHRQw#tA$4UK|VxFWHUlM7%-B zhh!ezHNC&%h*|`ne95P+Xf}e%QVgmkf7)1+CV}p|1}h!ERlJQyKJ=#_tcutD8SW$3 zT#1!3-Ihj@8#jvexGN|c{V*o}cgB%2#oXXt(#RqIh=>$ul1^|m5ASckY>ZowkJ12D z@Mg?3kW?d90SSd*?}})B!ag!EV#cFbS)~Ci!KXOJWKjd1|FANok|ObFu{$v5o`G+B zW&fq=Etcfs9yb-1@C_lHfJs0V52S`k#7qg^>f_w6Co7a9eKd-~JG^;!fMVbt65(eV zFBAjZ;V$5_k&Pc3jr6s5UI5G5HX+`dO!dc|kJN2GhKhy6pmxNP5bO)wD1dJ|zY#Ho z_1k8}W|rWDi^s!HxSd7cfJ*bimS!%@0TjqJhLYbc6_h)P(P9sC??dxkFK>T-M6p&L z=FO(~eh)6}loo6ZOw(?&cuFD2Onefc#dFOyKN>p||EDBCL-fQ^=n!>F z2P`HvjdKc3Kfrs>g9UFNb=RgMSai6+o*U(82#mtDWCjQZ^5q$ zI&ohK`v}1Z{b3dE$(Ma|S^#_E-jBAP(=9qBI#tm1@H_KXx+zb zHAzlTwexHC>zUb-z#-Q+kHOdM6G(SPVdK)Gmnn4-Jf>C76tl z@}oM|&e$J@R&5iBr31JNNXz`SLVw z?(k5w1rCk_{|K)Q^R9`7ltRl5_GatUDV#%LvA$pYkbZ5zC7RPNofjdK zW5C|)4hxBp6uhDJ5YSyHuGw?F1v1P^kn`L6^+&*9`Y z`(m<(0`B2HUtK*}cv$YK*^xFuW~nw`g|k0wD5 z^Y0iMLArQ?HE32^7brQ&$sEx?XWvGHDX`sAy>O){Xsr0Q24GJoBx=pfAy^%+7k*fp zfwZb4L~K|CC=rmwRt!HGnD0mBWo-6bq!*T0v6)upvg%kOtAMfroEq>j1^XtL0T(13 zb(by3&&Q<$j?P?B&%6CJjuU0;%g$us8IUqwRGBYhuhJz-iRgD1XMF*?h)w+#9}_@u zAaPmbBf{wfPd#<=CB-Y`f)QjgrvP)a5lckz^mFb&$p6M$g37cQioWU5kyEeFWQ`NjnV%*?sSqq2qGuEnC8ct`4pq77yV>jEN zT8-FhSzo-L#!FzGMyN%|?=!^rE-GH7daFk03)U z1XUC_Eln_3$W%+V@F-O!Dx73BVs%1rAI}3TbS)%@_Lr1lD(Tq!=Z|z>8{#DHg^1Ua z&xwO-?@QuqnO(NIUm5nT9jHw6*0uVRW@kG0*d^D02)Vute-Qa{!z)*DUlBNOl9Ism zLvO~m2J;$gF5Q^j@@r(r&4^RKoJ2e^RnD(7!z;D_a|b>nK1Wx`y1G9>1FNtU=YH(p z&JR}z>+@-++Wi@)T8E7f3(cMh`+WLm*UMzJ-ZMm6)egC(I(1{ywo69tXK7!AWu_BuN7)7){^3u znJY2N<(hPr;;D}0mW7#b%*z&SR<~5j|3=g2YE4dgcIRsfJx70(J0;bR#aqje#{~IA zQm&BtjAUZKg8LBp8D7X22}Qn0pU94xpD%b}!6eS0ChK7~2U65jZ36Hcn*gi>fty0F zWTp)0QcrlA^)s)#DWRpqrPgt0TWw>dp70-q6Q*;t}6&%M3r~rO4Cf}%&YT@%CMQ3G{STyp?;_;N6zfh z#(gq0VdY3B}4umfRJDmt7AyvRNahYblx#wp;l!MsvaZ1}06VHInyw#%Z5$6lD(V(*QVN3u!GU zRs^!Ke??hfd$mFCfREcnL*V25YiBq2uH+Z75x4@)75q+Vp4W0JyIQ-d1FFB}1{{@y z-|Dbd>$rQ8L(hZpR2f}t-o~AOc^iB3sA(tq88!lqS-_+Ejve`A3_i*zh0lTIX1RBu zK>v>`F*E;yI6d$sqWdrc7E;;ODW|#DZP@q!nZNvH&0*~m7-d4 zr~J(nX!KlEfA6Iwg4RLgA;I|@jF2VUHkP?+v-52_6jgp0HHt22-l<}agaIg-gFrjkZHiN-@A{gH`G+im zv(C%2wdz|2Q{|EEvO$dAp#6nFYllNzk%2ACoKbac^^R!@W_euiGKZ9>->E7CfC4 zN!Z|j-E#iq^196tIMt?YD*ocVV1jDh+bCIknu1!=)1tw2YH@GoulK@yV?MMlX@6iA zbtJAbyPNz+)OT7jL4)Yd2HMRBD_`XqvLY${1<7Z+L^0@X7WZbU_W5Oez3=0vYONKq zNrY@donBVYwz^eToR+6fDW)p?#E5t6`%<=K7UEpB;E;CuW$&qbi`^UPs7CnGi42sa zb~m$hA%p)I`w_(dH?3`We<2vAVdNb2(Ss+78A(g9In^yp%aR#7E_=2&Mx>jO^VYI&LoWv&i-ug^3craVSt%Fa ztBi%UyL=3Xrj*{>e+_dze$~X{t}1=h_GtQ32W+=( zWm0}yj1pX38KIuW0i?CFc#5c-fYS1~)HKSn#PypalVjz9>c$VYgS~|&mt?ZVsa^){ zj4>Ozj4@{V`JtUwj`p^AZEZk%p|nE23J8uOoCG)(x!{vynPgSN@v=J>X~B9{V&%Uq z=tr7Iz6@P=brUOKxN%JM+T0{w0g8H_@aHj2A8 zdoIsEDQCZOE%xV&r574_HRmU3NUR5ZJOkPQp<^h+l){PJQku9l#N%`Hi7kv_nP0J8 zw+<*Dx?&^w>OJdkTgw`j(x|F+;?Ev6ZR9qbt!|-&+}SLcH5-v!3F&K0WY`kxer^%C znq$~K);ZS11jH0%x$*(`MQ4VIO$OWt*c zPEu*M3EkwBDGS2N9qXX-aZ}G+XkTZ^+`wb2vq3$|47R?p$z=p@vN%JilKVr{TFkoJ zV!a>H4eSdWW_i_Gdg&%C9*whSd&{Q90o;{hWWlp0#5vU`92>RsR9p*YCou^L@=EUi z8xOzt0*cw?0G@GA^J{zzNW9#{;QbPRcxv`5!jttMBMMFBbv)yBzQ{T?vG!Wl8k%D6 z9@$7jg}N=A#iE|3&s2ZD#k#(&`%#;>a65fFr5Rrg^Csb3nJgzUc*%&qNu7jDbR&r4 ziK}ga=tHh$6iO(Q!ZkEsXKM>I0M(O)O4ng2ztx9(0T~N?I4>Rbr&e}9fBWI}x=z4z z|A})uX2e2RcmG_mleA`(c7o=&NqR<={$^+U>Ep9gTU2XRc1vMY7zSj(*U|a4Lt36> zOb>zzjQS!-G6y{+#EK5oXtGDDUEUTt?M{19A%O3Gy^hu!%cfG~`kPjUYJ^s|^!RMI zTKtb^WyRLFjqySrF^xmaQ0^Z(UP^>bwJ>=SeyhZx7eb?gw(32aTo%R=KT?fjHq);N z89n-GU8{kS@a%6Vc9Wfc9Vv8BB0ECdbTH8XNN4PYCC8z~juKyb*NDr0tD}qwFWfIE zE}ZQ0?EZt9SSn}bEZOTZxnES$mb$G4>>u2yX-(JAGo8-4%$YQNn$KB6weD+Or_I5U z(#}WPN(hzE3IRh;6L|4jk^vGb8XtK$*55!2nxi(0$qbu;Og5ciSylKoxogzrZR}0o zOPg#n_gS0iDRc#I4O)sh>YDjGm{tswx6BQXXpR-d-Fxq+p(e&owe;Z28p_Bpzu~^3 zW;=i7_a<1avd2MT#TQKh=VS0|Srp+wG&D2}Dg4McG|JoKA^)-&X(VFuevaqRjjO81 zo48*iZprwcs&O8(L7~;xcj;U1qf1G1a~x)EC+X*Uau$|%$}gK*jq!^SHZ`{#f=l7~ zgg05;Tj@V~=NwXfB-A6qPZTbb;9tq0&h=RHq;V$+Z8tQ*BlFi)MgOXYi0TZ2hqO+8 zWWzNqrN5olNQyLRoI}z|Y2u~>XjI^`3`L7`KU=o>0yS8G@;(J-!*?mSH0U^Q{Bzk%tWGvh@4>k zym`geNA<(g+kI84g{Ef3llqy~^Q~jkWr_L&AEvA3vEwT4ull^z=cXr0S28ibmE&7D zCF@Fb_w>vIWMB!<4aiA=*mM|O8&u&ez-H<3xDr0$v*vRG5^G_9h!pjr-8R4Gl4)5? ze0W|hymUR68nJ4<)VdZQWBmzOXh8=ksav(jG>-!B(xnk=iY3cfWv zimw%-?Zn#CvJ&qRG;|_wZuk1{fk>Y5lL(*>ilCP zSN6E6A1RSw`nbs2AbPjf`wF+oVMFSzR&9x+E}&k&sT;`PvCp;)`AER)`ZtpVzr$_) zB#QR*&-aLV09Q(NCw_OrxeMROVgAwAlB{{F6|9xS$Fgw$Zr55wuh^zE!)?-r-abRc zR$x#we8Hv{e}XbE#!dInzL*NjZ3~cw@p6mfD&DvN>kSA#y8fHpHs9FT*pSg-9nV-| z%*SzXjMwSoFOE(rkAtmAV$Y4pHhMT85ao#I8dG+^ zy&!*K){~;0iA(iimc{#0n%E{6W)0+{!*2APOPee>U&O|)eBR`OT!YJPFtBmj0BNgJ zb_S*}C%V^FLauk^RQyo7IY)0;)z~3pr5jtyevFcZGoL_}H@3GGHZx?%9SN+wo*VHp zMG8R{9)S42-IHkSpIcxDfDZoRuAduw5KTg9KrKoDl|Hs7VpRWq{5p7i&gKa^cwh3# zoPFq8f&oYt1x>1a5_`9&oS2jSu@ML04Q&lyri&?auyfT@;oYpXKl9?!2Prtmj3<;k zy+U`Dj+G&wE;K~Bz*0c`B@3?n)5mMz;)VpAO9^&Oj(r#ZUft+AfD|wJ+WY2nZZ1!RhS4WN7V_jam{2!G{h2lXIgTUhV}Cgz|nxoqSku53uu# z$3++~{!W2_JK=mq3~!U(Gvrs)RVT&W?~e3($`mf^;Nfb>cN~D2_Pd^mv| zS3}&G_V!y0ZsY^C=jDsA5h_sgAVAo02Jr5)#~TAW^7K*nd&?~gpcJ%L^hP?CZiV?S zvfv4iqW1|VM2e1;DC?)a& zD+Q?Gzh5`^W__R0aY=)Q8zK4lE{D3rn)F2mxi{bJO+4ys5eLVf>X~TUW3G5nU!VCY z(F_O6YC4XJ>;S)OmEQf;%t0!5(fOb85uV^1e^NRwHLKd$&nx;I^5~LlMHsk2$Za*C zC$(Pj-*(m_dD9(IlXNEF8TM;rj51E>f`KmCF8W@C=P*DB(A94KcAg>Yi_Q2nf7AV2 z&6E$vJllY6%O<@l6F+{r&Nc&ss^fuMjB+IV*`N|!8N5ercPxK3u^do!J0^=3=lz-3 z6Y)8G^G(p(sU{cO8HHB^x*^JU?;V(Xj|S7AGrVutlPlgMDQu_btL|byyp?t7Jl-oS zHi!$+Z_=kBaMOy_%AI*VDZT%C4i^j>m3@IY&6y-<%#`V*IhKk#;sW}7kVh|qy0F7Q zVfCrxT^~S7xy+D!8;vf?K1Ovus+UuioCf<@+$qOhYI(14_ov`1mV(9apF*Wtz^>fT zO}Y{kB4z#sYz%H0lrR*^AR>a$YFqdB{w!j+plO3MBF2)55}7(ayx=Dj#SdxfLGz^@-lNSlF6%dZ*EqiB3u@W3;ST>⁣;1Qxf=uWVz^uq%)KuS>?v20 zjCE~dsfA-mAiIJ!zsp9UpDzBcFdHP;Y$AyU1J_wi{BE?XyA>7qx~qG_!BlAbI!Nal zpbsS|^(1~u|AvKvwdFxB=M}$pf1i!gZ_cWVAagNCJrEHQ{G#0+e#a5 z-9i8&Tm#4Nf@HibBjC}~JN;{@VAyOStN;x0=e%^>J1M`v;C#p)k6rgDI`^P?*#q{a>v_<*A#I~{%p+0H`$3)daMVjkw7Oxee z3&4Ewi1YKM%G3@F>5PCt7)XC2A;X<|7FL0QpqD|~HEGRzPZ0jB`9gi)BAK5qZy*K= zsr*-20N_8!OP`!ijJhUnLv6#^d%|foJDTS8mBa}wv?i<@_)V-3JtZi};Bw7+Mlap~ zvSuS(@zufYflHesdM@W83Dmhn&KdimS1Pc`S%NHDvsNNnuER3KKv0`7?Bw#O_tjW6 z+Ex(O&H8>MZMOnVJ@ufD_S-N-cRSD82e#osrGv-XXkHE%JP}IWIFzrzr_#WWjE-9@ zS!5U_LN-JqS;T@I!IVNec*`Op_+IFd)^KJX;)dV{;9Yy*FA|DQ0^iblfyXm6`ktp* zqQ@sdL#}-(0HK9Xf~we|LoN^3w@2u>q{H4hh6y4=7QMp6j>D7RUp#V2{#&$J82A)9 zb_T>!kLYPG>J=IPt;5 zQyThQ@xE*xW<1VDmwbCJBMQWkSit+<@CJUNeskDzuDg{v`ZFelIV7n= zVJpBM@*FnK3>ib(ago~RR65uf&bU-gi)ib``Z4rJD(`h(3@Y^`Xp?u*t+|2E1}~($ z?dBeM&Qn=G?uMVNb>_fN>_l;M2PI)&0u{MFu1l<5v4z*Gz(hgtl>Pp9PyazIIr}0% z?gQn-C_k7kw&XZzU;AozmXcqh9bxB$j^&Gq0f=I(C%C24w4`n$845n6e71wpasOPfFSWWJYc56~mmLJR< z;ql3~x}^1Pp= z(w_AtgocDT9%N_>;97(4#G~#l*-z+lw{QGJ7 z?7$v2zmddhL)z3?U#ec&fzHHBLyT=Cp+X(r4=D3G>_Tg@eS&U+v4>O@pS@PNkFw07 z2Y9hCaQz>cRCghuDFvVpPy<_hAl?`#aM6AApuxUWq}~TWh)J4^wIZ-Fp!s_#ftDTi zeUtwZ8Tz@kDez1TDB0>jo0x-oG8naT4DP&9fV?5+T3>z9@zLrL7K33h4h3Z#ucS0e z(QsmZAbqmaYS}1}&q!fGc++9o2+!+Hi(Z9N&36e+y#^cF$Ua1XHwvYdI#Cwo0NXF6 z{*2V+9l#vvP$U6E>!`n+hZBKWSLOQ^ZCs4?BeCtCU2ioa2Nb-OVF2yaiepe&EK6ra z$C3mqY^m~1y?ZF_8a2#s~UgLaK0$P}olGz>ks6bwX5{h7@!-aew_ zUG_!{3s?@Sz_(Qc8Mh!f+)Vys?`&`cyptEJ0rnosr_MKlC`1sOs?VDrNN;x3+8m%7 z9V+z?;Hq{{g2%%VEUN(LV3&mhlP0ZPbGhJApiY?ZcLy|HW{bLlP}%x+YlvC@Rz90N zFW@QHIJXuRUUt&>`4S=P{_avr%Z{RgAhzZruxxuf;t(e;tLqOW|2A5t0A@HtQ> zd@^%_E6!NJ{p@q@#pf`yhY6AW6n07GeLgsWVPQ^YNA_iq6M{PbXfHtu*!X(IQNv0- z!KPckwvo17g1H7eXy1^)@xidcwzr+X+v3e2H#kZsG?yFciy8pKjwos1gCdUtp3bC3 z;4E*G`bdKv_)!(iMC}PVn;!pLx|d!;UL#Rd;*CV8sVTi!2#OCkpo-dXIF^AbE^iS1 z_b1Pqj6ztCb{9mHG4iWif{myO2-&nVI7vG?_~lzb_mdd7`zHOq%nQ^GHy#C~i(igl zHJVfi-BSUn%*928Y7x<=(mS8rUZ1->8%L5|9NyN`JC;1Ze2<2v{VM$UwZRWo;d@4+ zJWwNS%+c=On_I%)=?a5=9P12B6B3X69a-RJaiO1@R2=9v?}(#nlD?X}n(2@_2(0Wo zg526~vG_LYlRBn@9}x#Pv%UJmPHgc+c!Ms5yvK(KwT5lw52Yw=IQ=7Km=pmKGPLVp zP8SRAG(gfSrEVnubO)_G4NJ^yK!BpEHqXPEJk}MG7txlTFwvwImuwG9w4xIS9Eu`WveZ*|2 zPBd2d6=spIA*B34M^e^fKB(@iE999JogU?hPP2XCpI8~*ch6o*mZ4RcqXc?F@M` zksiC>9YuJ5Zlkp!?Q7b*vcMPCV~PtGe;W9B^X0*FLQ7@U)mX|mzfxdmC%kw&YmT|r zkL2EFF~eqCb5Tzqcv}pIA&7|cpo{OAeX;aPi#^}&KEhcpp`!0A`yjRvh?q99RKCTH z4x({_4zQ@R9}xk_o1x!dG=pf6$!yZ8aLHFM!BOK;MHZq%UUtA!Tyugky=X-V46tET z@UOpaDHR$(UH9Oel2cSt0R0TkW4~x87_4|5VBYSsv&GR(*c}Clz!a7l##s@JF873D z9|s}%)H5G^dSfDoPpB}Li6N59w5V8X@zKwEnvK&}SOrjc;Fy;QEKq}tR=ebDGdRm+p?8_WeW_N&X>D8|opM{jWpeCe%g~$8d53Y4QODUO2DDWUyv+ulY~fOsdGs(2G!u zDPg*DK#eP-rxAY*xv;mxAxFd8EF`Aj;h{E;aq0FjScacepE1)Pyvj1gc!^fL=6fuN z#3s&ghcpBa^t2HOk8&myH3QN_{Wo9%JP1=apcAD7Iwl2w+lv6IPTBMnsG9;@3Vbsi zW{;{xJ_LSrkUeH;@*<_+Q#?epAzD7vUte7~TmqOcRH65}$7Hod$^XwtKN1SP2$N=^ zT!tQ|u;PS233+=cut@n4#czaUB_zf#MFN|1ll~Eg%TCn*v*QyKG<=wW@_Ce=|E`oF z*NE9Ne8O|xEDC2BLA>(0!x2rmOK|v=Kh{mym@EY`k}^<+NGZ>xIS(0<;z3m7JVCmH z7(A*NxGkKA0z-fztD$;?r@0Pbc$kmkI7aLHCVIcwOaw9<7XYl0Ig&};J6fTSoA+~x zQu_nXH`d?XXoM8=D^7Ol=7{&MxHCZ>(LfW>u&({ee3ffOE|T?-C3?U#giCG4WeVS^tzCob3x9^3!Z(On}dTUn1;dIYiWXK&m$HfBomLY+@)5MN50Zfsx=SpU;k< zLjhU>ZfoAp2kkJ|gvgg!LYql#iBvaCtP~^MYL~AN=O?jSzhB|*Y}%r31+|f(ke?7l ziI+iAB7mE?(fR3Jg~6g?7okc}y%msbIpsfhJ(mg02H8J4PFHTUhazGmOei?4Jp5?H z;HBRJQ(wcgEQMq_x=2Cp&$&HM^bL$wj)_;E9}6On+b}CPmR7HWwcGC9zh4*w*DY80 zDn8;3uWaEHK@`rT#L0mt(KlQd8au4i?Zj4=7-;2DM#@a}2<`?iFicmt?eU=g+{1r# z;(t$jGvJZ^+NVKmmUyOrg!echQ3O7FJgmfkhEC=S4fp#Dnfq}U=s*!`H^3it~mjkYVL(fPGzxj2f-=0aG z7V;4TYSY?tka9yvgy~&*auz@n^Ed5$b$3f}S+5zZJRlID72l5?28U#b9T3?(E>j1z zM(du?mwL}4L>&g6+S^~KKB`ZETiYwzi;0K6SlOg6qv2H=#MjilaO>n3^EeUMB+3rK z{F>ni0MGWTXh#_?>8qas_HY6O8I|SaBaXPHgJLkJbwS(?qq1YizO^r)#Y-wjH~+?- zbgJSaC*J5Kw9{vtpHK`$wF-0)Zoeh=EP`mgK0ite(EiegStxjkAL>qo=e;+GBqWl4 zJ<;J_wDZ%_Ki2gSI5Yx7an#8FP3)OjsDE_E$n{?V=h?aHf~S2>xc-k3-5ITlH61@? zyKmb~D2dT~14(IM-fmk-ZFW(1>h_V~g?IE2=xcx3&PBK`M|l0CIv2RZ!uf>O01kP) zs3H8YFLiA4eIu)$MH~${dcYJ(Am!mF>iwG~gn*e4B18lS44L{=oWijXMv)0fw%lmz4KsJS_?WjpG+ch&+Sh9&xI&;1 z`Cg8&cZd~{=r^dONvB+@c}n6KTYlmRe__Lz9lEvZs_^ocTlk>TA}hlMjw2d$8MJw8 z0vo&-hFH*84KBT^N{}^kcVPwt8RX}$}?e-GIA9P8>=llJH@AYQ48XnPcWJdPO-3<Wq*m2mC=aY#x=~QWaSVVh94!jlSW`~5DDA==UYah{pKf~^KICc}us=Ykxb5RV zNq+5d?Le|n8yA+$Ii7l2F$ zpI6+kJ~guH^(8<1{M}$W_Z}7^J6IF5P0W(R__nV~$$9Q7iOTxlvk}dapyK@@f7Ivn zV#8pCz>&RoMA0ONTjXZ@>s~E{s$wrpbUH7jga~3A81j@y8_?Fu)!r(Rs1Mby?XPf2 z?}tJVpxgNkask3Vn2wW63suf87tH-iZ>mnlyysaaBIj9Kmp5lma*eZ+zBLS^IUqM3 z$}qQ#5E@;Mrx>W%M?VI$a0TquXt?a`cwAaLS(Vex%0}+%JNef39*x%g_+NZtVqRI! z#yNT z_=(E)hOw4hj|J4pHDO-^y09tx(&#(rR(MoBGq-0_NtBRk6NHFUio?%vC#$h2eY^0J z_OSMf>U{R+dBnOWcqQx5|1R+Lk7RXy*qZfB!6((cF)e4MwC(uF;;iUn!AW)|DAgRE z;F6VlZAr!Xrfl`gn4X2MO525c^y>YzZE42MPEVriZFFc%)Cm(Nzxh5h^aw za#khkM$G<*>s3$FWpkz!pK^x<{zWcAw1+5vooD;USTsh)?@gwlzzq9XsQ>!4xZDmy z$O3fv)y>w1>Q3YJezN$txj%_ZLAs?rjV4^YQ~XhbeEA>EzzP6&+%nM z&O=LqH1uBKdPv}QuMf*{zdDt4216LEKx98ZN)%Vf6huveu%)wFwXTdZA_B8dw(QO;dF>SO>J#g^{($0M4U%it8+{d$sCx3No!k}ke3*nYV?pB<&*UKR9Xn{V(-Xu8?J%wSCe$Cc*ELRroDMlmnsop&c~yer=_ z9!C`u@5Df1bymyTVVh~?A-|>C(gMQ&3=80_ARVFqT`y-{x5@gn-g)|b<-Rt#T#iS+ ziZ=|MeeYkdpY=?$v-4W+%p9Ow{n(;Ctb2z#;*s>Rbt3#A%|t+24;TV62o*WD%4>`k zrWb}UK3{W=xv<=Ssk)rXeX1FzMm!sJa#jj@qu=~*rHnh4Q`MT7s-=9agR6MDg{u6C zs=IC1fH=8YplS)NS~ML8)M*t4@0V@~t?jL?nQ4(4=PUi-UY=i69qf;H=SK<@*g6rl zvvZx+WBGlxg5foezY(I>=2nlj=37b#%LV&(&UH-k9fw=-!6Ab9zn$Q(=PQSFLdEMm zuFAx8!hKXB*HI~z-A_j6)N{1?BYd)R^l$fnZsJbYUl6M9wD&i_K7GU79_yaQ_Uh}| z6Eq6Jx@1vp(-cO0I^wOBLsO_ib6iYwCLBDrpW!YP~-n8s6%< zjq!1M{Uk5dZufFAV*)t&Tmij96Zb>)B5S^l+q{d#-OM5L`+7{TopM#%NnQ8d+wXdJ zDBu~9ZD7E5dN#Vs(PX-C^kk?+d*4R2xkLY&pJPGJ*i&adCr#V3*Lk?6mhhnF+f;fD zhU}Hr2!zVIrwM;=C|sv>Yhb@*WLI9}_S?IC@4LEhQw3A)$(&z0r+f0}!a4LdB}^zr zTAb$+i6RzA;ts2)tFaamooL(69Km;l^nc$GxgF*KL4gt7!i!V^Cmp5~7yhMb+Q=8_ z0eg%Ca(5lA7oYR3HCnn1m-TF(#g|mI*!P`XZYvM;@mS}FlULm?2-c2-_bhBRlt;y! zun5Pe{fjb5OEqQczga5MzHe?J&ak1NxfK2Ue;zk7D#q(qmD2P=ps8%>N#|4vV%~_$ z((HzNWN~1)Hnm1nn%}+)>tNZm5^A{r-MVSd+%UO2KSxW>G?*rX&%i#z)09@8M#IBl zGXWN?)>p$35U;TqAh18DxwKy?eljqCJUrI+v-1SxyKp@gT(K0@%FJqg*qt5?~r z>uzjqVu4Hk4$LLE_86jDV9zO<+Gk%XI#b{BDk?gK<2*rj-K2!SPGa9%FLT)+ZN!}} z(Z$|fwhS&6O#T!gg7alSO2}Z_o~_ny4`+8WGY&JQZSIXV-_E*AL!RG85jHV0TS6L`UC_ zFTxOjh{BQW7JjK!RaY@`&~%TFb70=akbRWUP;E5X2(F5al^OW;@lyj1B|CDr{{K59 bh`dKtE=}EUeCt#80Q{2_lM^k2>HGeF4`Gw0ejHPuxqNcWNA;o(u7I;o_E zhesrihevRhcn@5W>4?|I!y{L;IdMYs)QJ0!0o>Uama^6@^0aUS9#p*|gsO}-=&Mw(?LXnI0dnO?R&@{2ybkBML!|JbEB zStY0Njh60D)i2d2Y;C#>Dq1xp6fIHW?W^^oJ*ALNg2BEzxsQYKE$jWb8eVxq`fDf| zC;ap5<(id|5fP{y^85gisVoAe9%JWwwi|g{<%~kuoZD1*v}DdtlW&{g&EVrrA1YHa zK;xZjFgz=G_ub(Pj4CNXkY^dIm4NeoR;xoZ?7?c6_((7b)rzdp>*kQBS0^!ssgU1FJ2=ijOY>2dV0R94+90S@ygylCN;)gW2=tJh^9;*^69!5!*O7!p37ZCP+c>kIy&s@_4A_x`gb3HQL=dGc}iG>TZfsd46TGuNxbJQ zXP^XKI8hO%xu*$7Fm+G<(KVtcoGirHlv~Vu?>ru+sj@}!;|Cn)MF$5CT&FKRu1UU< z9hfmYN}NkyT78Sg>nq7yLQdLmRX(wo?s?)2k9d+Ze^M-&GyPuWIcIi%Y0Jm|+D1bT z7v1ldE)%fF?j!i^E%hbQ!w>hK&aa|y4(MWvy55G-Gje8}loySfS`6X~qO3eoDRG>y zuT|=k-H67yNOXdx;`#ewbsR2O!xL&$a>Pr4qngZZRTNmBNy?3!8NA_kf?U=MU9%aY z0iQQJ$^AX|9g@uzW8P22MReCbfc;cqstBM{_w@I zu8DQ!c3s|wyaB2N4W92WKjW88z41Q81kDvKFepo5{#vH^H9qH78NZ2%V?TW_g;Mr} ztmF_C-g@N1(h|pNgEA|DX*N^WrdqGGA>+Dh8iCWxnergSqF$l1u(D1Iu2`mIrV@fA8Z;M|co23k=hGVu{nuV`3%=zbIrjLrCojW$OfJn~ z?h{YBU2b2!%$&>E9@Na~bLYS}^i|3ci~|?dZHI3Xvn17;htdQ0rL+z^-3!kBbW)s# zEzmNB4oB-M{F>M(LMKJ`llMH@8^1Kf{jms6f!uU7hTSqm`jTLHV6~}@9^--Pr{)%V zggOx|X7$t3CU0b3u#fYozh#!9^9T}`dy+<;>~-z;a9uQIKODTiG;F_Xfpb@gB#|MP=x_q6X7 z-;*329UUC)8Fe1zUWgri-m;XIbRi&l^G>>2LF(1Ck@L)MG;S;lQVWa=6cH+UPYFMN zwpPwp&fnK@s>82?NG5PnY3MWIJoEf%Z~2WQ8^7<-uR0x=c~~b!Ddo6%d5v-eyIStk zf&;42snSMyT6w4DLM)RQPkdAsPs>0;2Bf2yOkgR28IQPdG%pm9ale3Mxvt38@;W7i1Ka*ek!hr2kznGgroZp{JuprrJBvEpS1M z#g8RR?WS60y^&#@{j~{n=atUq?n<)186|eh6rWj4D-HFF^!t57nW*_q)3I+k4Uf$C zn#C-M&g6`mNwPkRw5z>TYfww1?yY`KJuhM3Lu1N9*2JT}BhzA4<f4RdznocE6CC)J@;DE@Dq3KTfS#s+h8$uxt#ne<=s7OM1iRCMf2J<7wUH9#XS2mx-_MX^Z(d)avBJ zQ^NgJIbTJKhmN~VMIk&l>aL1Jl#;4S_(g?Thx`lqQm?u8ymd1Ft+p$y;AR4~4}aud zCKg@hCElL>{#Vvx(JGa%g4jr=Xai}P_gRz-!B!%GbtQu0f)%|J!HjE6Zs% zDZ)?oJ-wyH(sa`1y^Uehs!C|MvGi+!Zk|%{jn%#JhofsAve-%;4H&=Uyl=pBV9|>D zh}IGJ%f5Zp)9GG*v0ZiPb;b&z<+rwuQ}gR_4KuEiJRnK3T(J=P=>6NogHpL-Sw7BG z?o`#cG?@&gA1E5TZ?7qc9B?!=4)8y{If*K`9)A76iea=_rr273shh~-&zrT}Q0Vkp%x!lZ zdQxYTXe;M2*ri$?w4r5qTIBw3Ju#q4RldGu*8Xd#LPA?cUje#%2_hcC+5B6%x67y3 zvxUpEHVXpIwLV^~P7G@k;S#11DJl~;9a*$6_;~s^c>{&3UK^x?w{jUAbt4VEvMNh2 zRUP_ps6D-*<=k_YHoYwU+`*3>4-4831s?LyWxgPk=3TCyGda6jvEarME_u&!IsaV# z%;3`;wS|_}7KZqn@nMpR?RQJQ*EweNUU|9CL8y1tE#A{`XmNQWQ^Ht6LN?r8FiY9i zqM%#0$-}sVq2&3Ch8GS^4yE1SeJa@76Nh}n7L9wEin1*fZ3DXTpA%dlVkY`HdS~_h zLMeqbMMA{02sSkuHKF>ldLk-mwhs3%<4)gcx_QE)v?3eiI|4Gk>Ge9lpGl-^=VLDj zw=HtH+Wh``rgP>*Dz`?0Je&9GH|E7ybE}O8c7-9AtP!~}-zvgJvITo~d!m4)Lj#kP`mJCOw#d5?{LY%S1aQ>o0+YE$=>-K*`RY z*X8yjVH3q$Szo({mMl2P73{X&t$Ge>e@IPD;`8V7)8A;Bw4Ez_=jgFUwz{@o?x4nu zz_mBGEG_V-y;<<`&*SZ#_Elldl-oK;`dBXS5cRT#o|npbdJVkMuUkGBsn&~&@GBC9 zrk4p9BtGNc_#OZDFomXM;Tstz9qoq#--C%8@mb8vEg3Ig57546)xgoeSwwjCg2%MW znn=}YU$b{cQ;JYg+_KO+WvQlycL+WcwS4e?5~?KkgBncyt^4)2)uiBqTGr?&Z3 z3kwGqYe&~0pQ9vjfyC*gz6%~6<9_5X{wXcqX;{C_=DeP(p4wR%b4Pn&v&)W`EQCGn zosf0#!_v~yBBDn`jvNt! zJA_=k99+#jg&bUXb{6@uj*^9o`BfVyR~ttMPGnuPOO9@?^4#3WMnC`T&}rdm^XEJ*oHS zlZXG~(Otj#&qpt~SX@2fXb*e3D*P$0Kla`I%Re5J6G4u?ixoS3-o6TqRv?uV`6)C7 zQUl{eL?c*jlr+x6PtY>t5B^j5hkxfM^4XD4n*4hu9v%kol+v;Dp7`V4B>DR<_=#-# z`6MPYzrN!&Y^j+X>x^ku31Q%*yrv*A+DoLcbz1z2`1@YDh}GHFFon6q)g_z$SG0#K zeYg*Es-TVsU9)DX*%OOqiFACKN%TiK#g;*YNQF_5&|Jjt_S>N8mK~f4v2sHpO6)ZMgqo8EVek z8P_mk?XCRb|7D|{hp|tk|7o9Vm|PU9{<2|;=-(3ig8+2g;6LtydQVC$9&H~{`A;XO zrnZhF_-Eb-aN^Fslpy8#Cz63Nlt1YD+wFca=D*1D7h`@g<}ViW%as0$JNm_#UyS(~ zV|af>rT>!3{9?>6#_V4#F@7=T7i0d~p!n4#_$xnt;>5og^Zy!Sl0(sp@k=2Y zjqWp*AM>(IX3$!+U#8xbZ28^3`?~?+IFeUTVN+MxI@X0{PoP#v*F(t^XdH}lM0@Ep zr6Kklp>a~vb4)?UEv&L6K-PAS5FK}}h(d+a1|MtrdEy)^5}$Yvd>y=SiQ*h3KI;B8 z(Y(hwkl7mzi-VqPBK;P#X_*k@asFW^49<)zCbd7lv6Y{g~23TwS{TEwziu zY2o~`D2o_rx0*?_iP7H)0*S?o&E-F*kcz7wvQUIa#G!TBe(it^A&QNb z+S;A?v)MM&KEd)Tg>CO?-sxasV(rK1)vsX=pbSG|eK^)*d;ltFsV#Nr8DVW412(g&?h-ZDQa4T2PZK+s_TCo>b}P`6oSR7aK;miu6xp1s zMcRtk18_C|%Y`8&P7{1H%d%2!keL@TIfEtReKPT;;UtV$ghJYNL~dx1TMWfh|1xkj>Xu}B3Fu0vZP{Rp9tf+C*GWl z62`GUySIUG1rP2dh#RISc0CEl=QbnnH7m2JzLzY8locs+5Eq*;WsE~A|25b&@aWlkbSMuWB7H+LMkz;PaV zo*$m#uiL|QI~Y?)P_H^wEG*|cL`5V>>9ev$XF1_v$0&Tu_t@8>g2td!re1cPY)|nF z{TV$^_wUhWAM%Z=!n+sA#N4|T$6qedn2oCYluYBFK@2q+2*9ejokr6#P+K18*lR(< ziQ9{P5fV!X4gClkQ1BG0ba1;EdSDg3Wgzvtr94~24Ox3*@)wfBwg&XaQGdooik+hdmx-YZu_Jhr1dlk)QbJn~9L8+xd z{4SG;9VFj)IP2waB5)PE9?5h5y>0vI(#>!c=3W#t5BP-bA=W7!W1dyAdCpH zs{pQRxht>F>cPe-y|QmTx6UQ&pW`i`?s|Q{FXW5@&YOGg6HS|p{h)E*6PGQSv%MBm zGyye{aPW7^nB>?qsm-VUVtM&2Ao+uE=C)muAGSSmbmSa)Oe;8`5CW0#T;qptT zFCTlDZ)HCaG64aRQ?0MRch0mA5vJ>Suu21QVZTn9&=o#JXVOHWNe4KS> zv-u8jW0Lhz3FKmA%!oyt*rgN+l;Y;{clP7LkI0jMHb*58i-%<&C>?k$64Wn{TDU}O z?V!OIv2=@XVr#|Gm}7NKl8Rp3t!CV*W2Kj5;K12^FDv>->YkS`v>w|~9HYu6rsHee zyC~L!XuAWLTbImJW7G&4u~^g89a0kNcZSJLW|oiO_~z;$nc=`gAU$g3WNobF_%5Mz zW9ASeiYHM2Lnht67u{=KkVk@d0V6)FXtmcpe+*(7|48_%xz zy$UwDxH*&a?NR*mLX(xj#!&q=cI`|Id|N!ky3 zHijqq`ed`o$65s`Y;_gcC}TX_rM?9fPUiH)8*&Y6XNuku3p90$5l#Is8`VE5O%h*94Y=hGm6kaaQFBgDwq^~<&X$Fc$Z zxN+>tdEf8j7fb@Q`8PYK*Pn|#+}Pv@A&GvvP|+@{j6!>Zp|%SoxF@1t3vkwrXJx0* zfsc#_NeF->B1v{hq8pQ!^%qIFG=;dC^=Xu|QpyZI6m?nbI_8b4KgJqiv{KLZAXGMc zZR`15{EN=bzGjaNzXAs=_#`!Gj~smz)_}#7f;oC`L_lURNBEk}lJPoN;^3WK5m|=W zy2h9LOfs{D_ueP9=ne|;!rJ|~u0(aQ9v_G94^9)k<+fp|Vke}Ahb4njjMpAnK1|L7 ziB?QO40(#eM@Y9TF7tBZ9Fq0h==4$b zN(XKq*|#6KeO@*tvF>TaSXHC&T6E6%O!0H)T8$W1b=fUhvk&3=El0PF_&NxR#K@Hr zJxT+D8gMudoK)r1iP6773SLEbmshct#B~|p=|)acB=0pvvp4=4#>LK|bn{uihZxiP zbhO#~6OvSqrA7-l?BpM<)Ew)~Kq5T}sO16IPtw#kAaaE}Iw-*gXE-fm^xq*HnC#l% zFOl=$Gj2)84r+&@(d^2BB@@N8tPSI<-cplGo4&ZC0*L$hsbc-0?*sZQH}SAN^fy2N z9}oc`YWBA&(dnn}D8_WajKB%czX%Sz(lPcmx{Aaq;7B-&q#4frKh_4;4@j{D!`cjK zJavNk5OC_G!NK@x5<+VA7_yucQI;SKBb-vOuQ%hCnQvIdo7a~~lyJ2~B9Avz92C4~E*hwI` zcwjUQIK)plN-h?*4;^R0H@YE!%V_HzhO0LYL5B9hUKDi-_K9}Tpsg128XlaU+FCcM zdP*_2r)BPLv@mgK**II^OMS?`FxpTGs7++dapR;R`5F>ry9w%lf}|NZEkPn4I(l~9 z9#y|QsKN{P+5^cQ!xU#ZrNKL26*`rktn6WogqSmr$^%?3w7r_4t2~^#W%Ffff2wY$ zAfg{5@W@k-FDo?GkXp4TCTW1#nyR7>2E{5zpz7J&4mGTy>ZKY~5K?I2UfzymV$><% zd$je5yUK#E=R*4;ZOcW|Uf$GpCl&GQ4j{=}l(+mtj;}$xN7&r`(KW)Hb*#k+8c+$C z@}i!>;+d%WJFXdPYfuR|+My8};(;XQDFZ2&CPdo2MMhrzij*Z|7KfD?lY%Q_*`x7C zU9zWdxhSw$j=wf3s0X{;<#SXZ2(czSD1~%a)Bt<0vhJ}vcFgLZTbi#zZYvCwHuKct4VNx^))O z?v1@B>T+3q5#}!|^a!=c#KPvNts@KK6E+@V^70ZRRUq|y0)jeKsOpS<45=%TOSZvl zez`&iivs7GthLbhP43TH3yp`Trl>x$mp?Bk^1b^)1WEFc45PJJXKXCVh(<>w;E6$! zlvirbTt;B}f;ZDDQDq&2hP>we!(JU(W}BVm>`&{Ug5ie7H_*kcV{*k3+$xtL}M7qx@fPiu#9k*;c6ihq(8Vc62w|hY0 zY6-O<@qURUO%MOx`vi72k6{Pv6_@oquY9`xhTmLIb~8gG$dAfrVz0so4;ax&ZO95O zP0lNGzPu(X_VU#X@IwN7ud41b6D(=@y~|W)-J_{QDdA>n$98*vtW>DlE)1uMb5IJp7&E!ndPhkeG3o-c$t z6blNjkLac^7%j=jm8@o(WVMbNd@1oVU;gx2lxyuRI@emUU z?BIoyXkYjnc0fKKcJ$y~*cyF>k=nYo3$Kdq20sMJ8}|I*TeLXaMDqYk5sNqo8O;Bv zu0R2m<9B!|0iN8};aVSw62J>8*N>9pqvQo{yTs^=t3l@1%3}l$(rcJ{u;(}yQ%{r;)Yk6IU;0R>hrO@|z`@c%s61=2ON@UBl~0feNS&q+NxKUH zxkhpr=)G{i3=#khP--M%VU|dG@cHi)|L+Rx2S`}G?xqKeuB#`XGP86(4Uay3`7ecr zBZSb#q{)T|!;n~92!I8(j59s2wrPk4 z8d4pe-LII275eN;PJmFM@{q0*^hL)d#O$BoIFa05_J{}Z!z`frM>Dnf0k(s{r*4Ob z(D1X8lZO;-eB~e>EFerK4l1yWjx$PB;*l%CxCCjs0~Cd!gPc@t_^}Hojv1D+Vr2_P z2#yni{KGvU!Dx;!D7Jp;g%~d;*%)g~EJkLAwJe>>U^PnCKfQbg)@29fS;RyqaARp8 zN~scCNFo#rBQT>zVefk@h|_f?*bjUdPA{x6sDTRN|04gT8dU6s?ahPc4#S$^n#m6; z>#i#$OIP0f^t~Gnvd7&Mu0P?0F#L?=3YTU!gkj?Vq)fPmVF!`Otv)%8G6aOCusP=} zklH{tZxEwTOe!v`7*%ss2|K9Q>>C27jRnHR>>uWmxE=$qLwG)P+l*>JHY{yge9cav zN>wc5^Pt^E$xKysC^-9CGYdb+`F~(Ef9C@B217Wd*K-GWvtlpWTcS=d4IO75BWkpK z1Cs}DPJ;Wyo(CO#=B{Fd?gudr#IxK%gfL4-Z9Ui~#<27hVx8pLWCjWK4L;mQsDK1< z;gW4HCGTQfd}MkMDOQ0LDQwdKa18+_P{?t>W?=T<0T<6cPq3~5MGB8kb%g**J-h&3 z=q}jNvqS&V3yotiF+~KNj=L*o*8WMrHDUGyvuu{PH zo#YM=TLOpjTGCjk2TyXexV#|}54#HT6qyI``nqz$87Nv8BEVH+-@!xgzJ~DJ&H?`Q z4f01Ir@srtA7~;Crryg7sn}aHL@(K(_@q3*U`@H0W~-L0|DePf@-xjqIPxCX6SODK zgGj8+?UsosvD{MM^XjMTIo|C&W-YVTW=%D2W~1TlW~;($L4K48iqn$EQ2Qx(a@1Xsn$S82A2^v*yXe zhAqz8I_){#`ccIQ*EP_~8tJjhSxEjV}Y@C>txF>HDsR4^sicbK9 zS^}}TC=py&nMQ^OVcd%+5r_*zqh)BUTW6FY0%G5%1l5q_vsm|dw^&v7`(!bH%Y|$? z2*3}v17{yH>IiSBd8RK&`t{;oO_r(*qK43VGCP89ROwqs|fP(Gl&8e`KF#3uA`Df_~!-PEEfDT$Fz%?O*xLX)t(;{#Lp12noHP(5hh>rQ(Ww z-)*{Hovc;FLni_Six)p8I2{Ld*a=hn4uVN}84||v6pjuiqUs$+WsU-^v~<>dxQw}* zf>_H-bf=^vsQ20?7tIt|RYjG%JMbrd5jF<&jMS<$n?Y0k;gO7w3rywN3x@Gax>HHB z4UXP-vp=7-?dypzXPaDGDSZA|U1ln~YJ4LqyHyAnpWFsMs1{d!qTYY}BmA+9S}BqI1~R=cI`-K28` zNguR2_kzSpIrjcW7ga*XVN24)dAt%q9W?(J5H7=8g(T5YVTnN;FSy9WYX@6LG)Tl{g+@J*&Gr7= zJn{VLk2+wME>HM2=leUW)Oxq9ZI?E&QS>HiK7(uiUnRv=N%uOJ54C2JDsFa4p68kL z8@8>yX%%n?DHQbzT+fXqg;qUn)uot{8%q{-i3&8-aG1O}VY`iZAs9wMUW0%#n9NSA zO%3wR$kKS|Q*^q`Mx*N0ugEYWdcg#Wpr5E?gs|izBSOJ$IZ2m7Xs*LJr#U6PhR-6314G#6c8Ww;;9Z?@0Ks8QE9tDOxedCvz~(-^aw4s|=P-6eAYXBPJz#Mw1wYbP;UyPy8-jO1omOfh!wq{fRT zgbdX~j*2KAreT$2eYxtw!C1b^IEJWDFk}TmpY^O`g!N#CCA0M=)_!Ue;m0~NpKwio z6?za-?6BN)Csj#Vg+iHli`s7*(j;u!Km;7KF!Al{+PZD?S}HV{pH67-|p-vb>t z!Q$Re4D8^S*#qo2LUV)0+I;{qlLK^TWA#Z~d2l5&;5N^27)m{e(VOv@yJ6K{iR$R*J{v2NI) z=0$hP&z99?UR(r5tOfxD!wwy%r>UXj`*3CvNX49$)Df9U7Jl*`QQ3f45|x zVe-|ad~PylcBGg5(B&n+!C*g1SqVU-hg)^)Ch1xQ1H|PPriH**w;XV~BupGK0L00S zFWqe-^6NnWsH;qnd-NPzNEHOB6v;uWY=2N?Ea--*{qG@o`cVb_IR=S&-zqy@s-&FH0eer*PCq> z){C*<5c98mBD(Zy&On)hLypO>bhDrAM8MWy`Z8U^YFH~>LhAYLR>qIcj^bYp%e@Q? zt06w}`wg(Kv=u1o&_@`CeUq7o5KVeYr`~>w3^a*#G$@R3lkpGyRo9V8V>V%P@=RsW zsU`@1v-GbfMjL}wC(F_gZQ>r3@|1jgLYD4^K)>%01@<9oNs0siS`e8Q3jP&xf#Rlf zSWAzKsy^odz^WeC0RI|JD8pf0`@v>TcWaUXjt(7q_w47W`pNdl z!FQwBKrOmKq4=w7-P=D~KWk@Le4Qct%>!FCkXHyErL z&nF-;r2X3aA^N-9IhM`9s;I;I6I`3*`jp^re47;_A|&OnXIQ>pk^1^B4?!|d6*zEd zVELGYh!wZq)TZ|@Nqv0_AmNtJP7o=A^xS1Cc~u*dou@cL3Fv}c23nSi1#shKGbdG` z!c%&$0K%}ofvLmfagH@378`OFlX~Y~g&skD_0mN;#%?cVy}3F%f~|o{1g0K$8D3{g z?39=E6X+`cTHN9f0FPu{{1)~~_wiCxl%mrK&-KlukcGx9;<2_Y;?RfFtvS7YR^vHo zM2@9`U2Yq?=^KNrKE}_pr*M8%)@6@O!;(K8{(Z!g?crj$_Wz}nc=W3In{Zu{YCp_k#aZnPmFRWhaMiA8> zexwaDjtuSIFMy(42lX_5Ko~cZQTUe%re>dY>jq>$d1iE6!ftPy8%xyleKB42P-hWXC+u1`3Jjz%@OqPHzMLXRH3D@VATY5du=oPOPwn*3Y zY%M!|_F7VGe2j-DT#bAeK$2~yU$#T=l`zvq*3wcv+2B#oB1tm)h(OA)Yne)&jK#)vdzIwZTY_ zO4Ph*uN<_ygSVT?dTawYyo#p7Vz5Ly4(m18Xc zHs8s-JzHfqr0;?}na?HnY4ujSZ1%=+f%S_QCJg1C`E(F&Q6JFgEaByb+Ui>pZ2+X%d|}#R$PeI>ScZ z#xSVGuxxd8wbI$jzsvEl!_7-DxlseT8B5b?KH6izKm2xT(9wL*g0e+2@v*$#tCz8Z zzwP@CF&8JKR@foR_mWJh0n+*l%emTD?tf6uw=HVV#?e71K5e-oNykm1N`=sbBVEET z`8bi#=hP{&#Utn5>&BCAx&HRxBjRUEelyQ_I(pR+-BSbIGpubkd;q%&dkxTX;sn5Q zy=8@<$5Xx=YJv1z&~wxqacTt7O$Sw&pob?&W()%mSkt^^+2-fOCAV_MZs?YJe0^X4 zXz_)RJofh}A#x9+t@}Cs@7F>FvxdhI4<4Gdo>bYKu(q4s`FBR?3zyUbetLem41ve04Czm~P<}V>w*gf^8B>bZP^+wq+f=R9pX&W+~)S zI9Kp`O|s+b9S{;;3JlKyXTOvI^s=V#qKCn{=5EiX6smq=%@YYn5wr;B4qs9P*C+UW zV?RzBt4KfgmU6ab`#vELDVSCWu@PJ9xZ0Hb z-DGQRi^OBAvn}1ec-;ALFf{)WeFg);uem~6C@PneuOwa># zOSTI_bv=X=42wW&zC8M+2bdu!kphJX;{xR~Xpn~S915i-hONMXEFKxzA@gym2rPMb zyv`YrKxexpepJ0XpXMb+zlt{H(o)8JPJGI1gwz3q9kA%J2mQCNFO zlse4@GnO}*7G zl~)G-7tIIzjk0e}**|*^g&y5w)6=7c{zV3b+f4ZZc8-F4TE`p#jEVzz@@!r$0tRyb zJ1i<13X}sal+_?^3^4Sy8iIxIZY}%+CY=-xUME(EKL`&S0*7yM$N&&*bC+uW10250 z3w%G7e~bqFsUCDB{9TYirIY^-0c?Zdn^8(g3XZK0Bq_H~5Du&H?;y}J9XMv@`W_hd zBLWgu#Pwm%Y-p(bi}5^Pd)N)>-d@8*0SGpCB_9s#ziV*fpAKvQyA|ZH+yjIccs|%> z2U;sn_wPCnpFqt%y{9+{`$7UbXZH&BE&G@IrUA`qU3;M3Ac9p^jv~WgQ~&b3;=n8M zD+FL|DX>ZjiwN*J;NS6C0RbhDRtoj78LYBd&jd$3vCC5bL9u$>;EA@I6kwiFkTb*| z#z*Ky1{wczgzslhjHrJSWVB7ApVz`YTJ|B`80r1#najo9A4 z0cVcktdk$KBaMYjwaGde7WvPq3GD2i))^~bsW5MbozC4TK=`eX9P0E?Sw!ZiSfFk& zX5i1}LC38$;vrH;oYo(cMnKd~zX^AfMbB1Oo%P;PWCp=?aStfRpva@9qFT6epW zc&|)mAn(tcVNfPz*YPTA;lioBDoD3T)>vf%zB+c@3P(EJ%c$=W@2WkF*aMt+z~5l_ zvWSW=KCS{*aN1&LWnc=@jfLpee-%Z-E38}Ff(Z##k9T`oqsQcoBc*EDYUR0r2BLmq zyPoBX%L|zgLWmeIvL1ENlO0d?VRgDBwVEH5Upc}=Q5+?SRN%!JFCOS^`tY?V-b3~e zmC3S)$iX>Xg7ISG!qo*J4U6GNyZE?+V*VIOJicY+d++9@+lkb}Zz(c<;-eG$ajX*h z`Bd~~gXMlxC(Np&mm;j|V7e*zLHSH}<>VTx#N&+#CE!69`UA@yWauol=MJ%Jzqb?b zz?V>P=%t4=ja6ZGI0+v#*qi@P|M)X!|1tjd3cJu4Ft?gxwlR9{BOoJcj1vTf#tX4I z%ZavoA`1UCu!;mxdhD#q6BzYkb7MCoMX32Wpnm}ACA>j>cMHO}vscOg@v~y&dYT=w-CXfs=Fjxy&ME%9lkBF_L$#f?g*hf zO4+;mH5n1LHznTI7-_V+WjxrQ_}N|d2NQy0o{%;DnAj@CF1!tt3?bC@&DD}!|F#r# zLfzL8DnCW+<0NI1k8Kal8%PkG;;I|2_xGr7W{21FQP24#ka4SqlIf)5Gi_lYAvIcl zM^OLnBBoHB1ovCN?G_JfL=p+pMB&5oi+%M|tf5)PG5oLbim@V~!3 zSe5whKY-WSm!Bc;z-aGi(o@%FFS~=d$%U`K)#gIq3m?@9?tunXZ*PCK6N$6tyIdlq zK^bF6F>`Ok>^3`{_0pbJk*sAUz1gpghYJ1#P`N~~3}+_E$gI%M!={?>t#qr}_1Ixo(QFwxIhOiu927msO<-WpAu`1h;%-(xsm3 zomqe>zc($f36X;5SaJysfQQE=6RKKK6UI~yesV~O$0@3}BNzAg&d^6!Y1(%v0QBG*)jhcrhVV4|Oq3~EO=g+E7cC%P)Ur++(BWhA*ikycG9H9PL~ z%~qHkI!7!ER({C}*#2jo5{+_5l_n>*S+QW&KHx!d$Z~MJsPDex=J7GxD^}zk4oGWH zje5tJaB4$21sK`3Ktqm?FxC%KbpqRdu(060eHiG?Uj{!DFnJBT20s&6^nUog8eSc_ z-&bZ`6l9+^u4>Ao04q3QHeyZ|%Ie!#lucgyaPyO>*QoG`GDn_mlwx~xm#HVKp|sXR z-cD%)?GT^vQGwf%88m&sifD@~qBcKb`v2s_YTHzaY=-ZuWZ+x(UMu4-Lpj=285@xf ze%0d&X7#hGRoYa&v|#R2d;BD9W*!arPc>Hgi%9$7hBBWBIQ;~32`6)h!m1vo^zL1h zT@UB#Ekv5ZMP&CRxe4QnJhV1WkmS3_OEu5fZD;oX;Fo_CtA7k>zv!vxfDxcqvjGXr zF#VnW_LkG^7qLyj`3M#hdjY0Iqm|3c)!#WB36rVr$5%c%C>yN3HK1(vxj7bw*PfBi zhjRF>ELFk8UsZF0WoU~4J6ZhEjLxTD0u3GwmS}JC&o3<4&m}@zh5DWTww~7&gh*`&~y z?FXFuQ4Oi%ZCx?a>*!99V~HC-ieyi*yQ2ONbV}^Qyy(d& zlDB18x6YO%vZyY<3%~mgf9#<#zWSN4<#lig`WBz-kLg+6U?z>8#6Qv7|B-NvkG*+x zB?swut1<4RTpmb*PlZ$$f4qMKIG9+FH-d!xA;&M0IgCOjN4jRB8+AYU)XPhk%r6A! z15T^)7YO&3VqbY1yp*$M1h5IYjiYy8BD%^#1Q!cN^#{K zPkdT6q4qb_i)^Ao*3KhfZrcUi=KwC0k!{TVAO>V zNcIL%XR8##V`K#~2>og%`nXB&ba|M8&f4EXxd`a=7t8rK=YBy@ufFT;0&_?fkW{61 zYRd-Zb#LYPPuy~PN8(7a#^zr`xxikskn03V=(ELmmvJw%p7kZ&F~(5_J>NzNvrh@S zL5TOem3>#fZ(Jh;1_;Wl9#0mH7{WU;y_v`pMVV#($=d}owu^8l5tMzl`3PO(4JQ6# z$HeV-oY2doMfS+h|KVL#$y-Ws+3k+;j#h>*>}lN~6Q*41ny=KYo|GP9=(BKKc9d;i zt7Zs*uBfMWLGLFRG)8*W(5ze_wPTZFDi73OXCI6-uGTKBE+6~&6YQl#=KinkWLwuP zWrFPyqF&Ww?Esm;!*cBqbz4r~9;pw6cM z+o~?>rG4r2e1GedmH$?7Z{s+Nq{741ta7p@k9LnSqs%tmoGT+$eXqkBvQ?5qokuYy zss9V`cL#9Z<;wStEqv9i&E1|SdAU>5bas4usSH|PyDR#GiRDfH8AnVrBy(;}kEUlQ z(j^)e=F-Oh3CUi(_y;5_o?%?}s$;7n#h58lVIc~bX#<_#h{B+P=K+QcVm?~~Ri}ph zvI@#39Bmj)q%Q`=K7XKSFj4-IJv6ATWW)CFQWbMZW>=+Jx06}xoe-R_p$=Vo8`s?- z7@*NH|6R#!5w^g(?|xrolCS^mFAhE&?W6}4S;081OxFh0SSsl6J{`7Okvjxk;1dTO)@07&Clkm z`)~+}*yLQ>UOu$qy&W=S8M3R0 z{n^|8Gynb39JNb1>b2@%_W52yRa|SdkMeTC_vhK!BCH;RwHr14zwzG`5?LR4P<5hk zW8mlowb;V~(1+*A(El_#_VkUPkh{WQepCpQ3Z?Z!h=Rh~Guuh%WWxC^O(Y4VbUTZ< zJIq(YZO*P-Ka~XoNa8Xu6ZvBl6Z>LMoU-#@(i;ew`DRA1CgQeeppX5U=6ytxlm+K} z$5M4_q;-I73y)^deX^~EGktx<{^G-hheMw~_5AGKvrYxWLb4XGjO_S;l-3OA8w||_zRLtFL=GKbTE+>W<3;xpUF-f z{-$vBDIWyPQi3;@GahNWpW3(|D)e8NV)MM}lIDjv5*YQ{b}$EAo22c8aL>lxWJcK{ z?*^k2nf@sEAqo9Ux&NQ#N~)u&y>{ZzeOVT9oKaaHhOdx#PA_24DM*9%r77%;wMj*QjF zNu*Z5UEC>zi{vN2i@C^u`Pl6-2{WVA+teM-FP06#=-^@F4fv*qeLBDbSdStHS`meb0QG< zbDntHtJyjP^%Zk}_0AEjNt50P2yVAk!`_Tr(dZq<`1R&EFm|qM0;sNIWHa+0v z!L83%c@&B$Y+Y^+XiOk_X;?TF-sU4$viX6`gu`X2{=t$|12dY~cfTq2dlEiU15v-y z<79-kPFQA1O42p29FxZL(1sPk#z0%Gbk#l006WNae?^hbxE?e$e z{N&>p(KtU%*Jwv2zc}w%S{sOQp;ES@mJ%?(!fz@ z5WmLM0m?s`@cNYkcCcDStcMHt@Xy|jdjDaHL9NZ!=#i1K3RtL*sDeuC&x#%(cDmef|--?-hBER zR)Qg4)oU1!?UlM^i|pZ;(Bu5ycg$a|@0a_B$oMM+{-4Ez%Yr!y8`IqrVU3E+Nn44s zRCa|9z0BG2^Ypm3L_ci_eAF88TW+o7qNr5uDCvo9Qip*B%xGVTghNTI6{N}#9_z%>gi7#H@ETdxIsGrJSO+C}QwJIRu>sLM% z_`AK+${iBe-q>%ZXd-2)XP*|8;!NEl&Rip^joq7c*tW6kSsWhL4}A&xmR+??2Mf?g z2HzMYX-a@!(-WX}fyTQ%$KoralnG-TS8e_5JESb=ICH|beQL3j%##>Qjhx;%&mm;i z;0z&>#R=P22|pkc_b@A&;X>83(6uUZ@>ImkJR{sIcCXv{3W~XxNQ|+&j7ojNB)w7g z1gAMxVqncQBto`l8*)sd3Pv`*8#_8G}#UyeFG*EI#Hf`DfVQ{*tE-fAl%U+(9$6mf2o zM?;}letB4Xpdd66wc3&An0^3b%CWuM(v0hAA#Au;PAlJe3J9gGq;FZda2Gp6~6H7}?8B|aq7QViSVZ^#PcGLV^|2!$s)cW6wXFs|44{Sj@j;ZuRkTlH+c zM5&7+Ekg-hrdWxcu*8Wo+78tS5Xpr<_mzu55zwLv|LrKdBxd9WFLB;qZ}%aLvyzXT zy#O>4oW6D#o=rXQb?s}j%;X%clnF#dUzT!c6$61>)M3zEzp>nCPAf{qt}W@uh(aIQ zK61vFgy${1xRN;C)?&gq{oLRN22=VgAzH+OjMIKOukf?@G*4Ebo@^O*u!F9%flQ?0RJp8eRD^qL zTQrhFD%8-v99>F{y|hxou-3q6ss_9x2wx4h&Hlk<8MjPee`a|nu-|`&{XQ@b`-A;2 z9IwrMU)+=p=vCNk?CiAS*z#Fh&!Z}w+WN}ZIl<9gxUtR@<)~P?rEB8%+tkFY*Lr~U zWaTh5$_l->p&;erI!e|{^sIzZ@f>VUXI~Pv{fz;-wA*>YX}g{-CCI9tocjQ;J!fub zRWF5DgJGPq`ZE!BP-|2A!4!k3FDT|jQozRC;!MDnn?v8y=J`EFB+R>PX&JLebp%I$1y%IOpoei%&Ay?8Dh1Q%k5-}RSNzchOldKKsw}d zW!t;1mcC9qjvZ7!?&2?kdpJ-s|C>B2e5rS(b1|xX-E3|4|*WQ0eHMwmI!?+*{iVCQJZiHaPf`9@dAP{8>DosVC7aJf5s39ON0THnQx>2e$ zaf{NGUV{~t4hEz~MOrAKg%Sw)=6Z6@8TY>Tc*Z;KZ;apg#`yf}c-(P4YxcQjo#|qD z2QKO^ai0&ZVDe4PXcl(WsLbu6@vBTswD7H3W@aaOa!ir?;nNsr=E&Jd2K#%B?q~HI zLpQA{(suTpQr7u<19~s~eU&y%hCO#2j@6#)djDK*2dnKK`FyKWA4|8(y%UXekSR zeCAt*)L(sRFL+xtKu;aX%uFTJsZMLMXB}y&#l*Uzx~yV)QSYt7|wUn^DW(6M<}1{Elt!r{5Vd7>a>jFtKX;6$g{~#3Wa4w zes^6w(xs*(80q=NO>#BF!Os1|KRso;daf&K=5>{{B`gXM<+JzGH2a?2kTvBi9QSpR z#@eqg9CcQpu2Bna@_zLWtTs<*iQ!zMB*)tO8AopT%&a>3GwXG25yD%wz6T|Ty(DN?37G)enom6zjK(d zgHRp+6)VX+8n0M3%bM%}9X&j};md9W9l2{7?$w$m%fq%aXYRG+<@>xw!0!e=r?hQ% z!b?1%=txWzldty^=bHKam~&)!SZ ztWtqMqfWf;I4XG`e}She5TOAsv0GRSq7=@St;q)f5njrQ=Kv9_F+`*nhs0*!36GD> zwZK3d6*s?k=Dhdh%NYYygBI)Z{e3GX_N9h;jis3SRrbK&oc0^@t8=U!GgKcrA?Y>G zA5%5hDr}do3nN9gt%>THwRDEGU1wp)Vb$SAD>27sSKNqSE-t<&kyVzCw{vU10IgfvQu7gBANR5&a>Lknd-7Z0oajTL4~e{2d+70{paRR~UUP`l9=7(LSw zU17!}Xm-xrid)lplkd%;r+C7x_d%fo{ru`DkE9&*$LI4Vg+Xz*<&)bo+v*=j?WDsT zhUcYT8+_Xqbw43+UZ^pjwy4OQAb+{UevsqLe(c?dJ1rhJ$Fwg9)*9@jPL*_|6df1< zemuEjP9)08uB|WdlkTmYl)k|InW1fN@O^eBB?nSoivm7zzrR=B=h4o!Im(G2B|xpihcfDkl$8o3mK)hJZY+%9+FML}8*e+zuHg7q+7sTs3Q^ExoymeK2toa&g1ss<`VJXQ!w+biYss~rpph*xHmmI=D zF3vL8>z-WwmJzdwd`F#|Ad-$4k{9}2&jj5c~kQl zsboQ=zNGzgd-!5lW?@lj;|DJqN1wWmuiEt;dp|d&C4koef)at3fDKqLK)na(^0E#9;n! zjMo<^Wzm&*+7)`W^S z-X=|4y2sCbV#OM`^wT^AB2k9YA9$Npv}W(nUkR4C==YZkk= zvPGQdhoituGhYi#)ENsEh?)8lm_c+}`DA`?@7#t3b!wTAM&|!1a#ZLt%zW0p4@L`{ zReUbkXOC5?%?f}42(Z^w-V;6xN~e#T$6O~P%!0Se{nXKT#4>e;9Cs!=t30`A)*rF77bQgrv5cePFIh&`&Y-W?2E}aZmAhL32JA^jeUVwTcFkPxaW14 zA2F}pFVq!O-<6RUGN-g=KFr0$3X!TArX|DE{k`K)C3x>0h2*W>)qsl$Lfdc~YvbdE zeiW)XQI^QsF-sO|Q6miKW{h7yl@6Ii_>;2Jy`y*Y@b)abp?N6L+-5kU&MdLWYD*0o zWHfzRH*R0WOzS{`#kG_kPD?dYQ^?DZ$j~OG0V;IZBh@Tx5bgUi1c!W7SpFtgPBs&KAnbB+ce9 zlv<>q!0MDs$=i{do}*v~!=cZ$;wR!*ea%CUHNYBQA$}nVf;IMJ$7vu#sJ{&}g#VWw zfs6l@i4ap0+nKxPVvD*iR%)be4kMEZUh^LkS@Q{6f3v_u%663w^jex$md-Q@wo80q zlz-0Sy=QW{<4?O|h#t>2xVm~y_$sIzL@{Spc@f_Jq9vqgtDsjQU+TTpoI znV~;k1hWNVCo)ZBcGO!vO39O05|eQvMZz&+QFj$iEfehC-I1vIoAj6RGwn!$TaKO2 zET8P4 zJRgb+hxF|>Azy9VAJUD6@j{zkt6j=MLQ*^rqbHD%aFzP9*F&ctf{;AF8ZFDiyJR^g zBz9vF2Ih%HE4DNo4zS-evoAPKxx#EAvnpg?+f%jC2l81Soqf-p!VL=2U5f_V=%tE= zsie^m)*Q4rR;SmwWq1-YFGyFI>3tmLYnE3SoOsRBeO#w;09Reva6&*D0n9m-! z;)U^~iRXbeyBWCz;#;b?3ghz%^`XlJ?9{gjQ!>;Kv6rMWEv6ABK6M0C#PH^TW>>{O zl{|p^^E&aX4U!Y$Y#eHL6VTs|-C@o3){O1oFrMhy3v*eQuf-h3iXXdUU*QSsk36{x zgIGshR;h?a6ryxn}Vb|J=BB)O1_VM^+Zcx3U$MBI>$B$`e)i{U@oEo56k4Qd&ctRdN=otQP}V6G+BNFOBA4RXeblbtdHcK;c4!8j(e{|B z>%S*0ac<2z(WsV**ZmK=ZMtWrVRY`aG3Y0)K{P@NIwkyg&uoM^m!^tOBz#;p+0g!| zKfZ|ToLRnAAOzUoKQd>T7~M>H)Nqb(or;amMIu$6e9D^4JdEBer(AN_8mX+f=p5g7 zLswVue7GtSxvW}FxO>)a$~?o551uI!t$`KZz;GF+)eD$q(B)c&bIcNQE$j72%h&Z_ z8BSxCA(t3c39b8H+iOWH0(ysTw&D)@Kk&P)e;@D68!u#$yiYh_2#&g7yJ-zJnQH}R zZxsX2M!o4+G8ssaA>x6KD=-bO`tIb&mQ7s-owkTX>L!o+NLL+ z>c1rYdFd)><%V%mA{Y@u*M7OU0u*)dek&Bu=)fGC$LcPF{A(Ta4K+O=fMFKlYQ3YP z|FlNj$^dgdVC5%$<$T3k?Re5GneO`#fWln2t$Rkz$pYBO0?i?-_rrX)s;oS`V%Q(c zl_s8Re=!$OcrvcGKi~=Ta=Ow@ctUA!kVEtu!AMgZ zm(N_hT**=fr=Ex>$m$hjuC;Sdi?6*|Aq!}-Zy?z6cCOC$r)GGoA@@W0)@F>culPzT zXZVv8GAIo&%CDU3BDGk6ceHW|xJeu7vV7}Kcu79SpMX>DAX{~7ryb9(omt$DG+hkY zIh&Ud|Ip@Tz23M=9Dr5u!~p9vY-jO++71tsJf5f;@wcVva)n|OBL{6eF`hd9oc<#D z0M}(8Y6xwJ>)&zly}vv(<}_X_su|6%b+^TMVBC3keRc9p7}`b`CW`UiLnKPv4yPP` zRCRvYpP)2EAGKiZZHb#P?rnW=T`goI5Br;NFdBXlRe}LVNO57qbh748m8+luzN#@} zm$v^2y1Dh`4=KmuioMo+G?YGoI{*=c8dujegZ|ub+RJM>eSy@69XcQtV%g1o7u*o; z`cv>`5C)veS+Le}X4|#w8k&a*F}sm7Wluj=tyzvIc)a!6a4=C-=zVpV2=Be>l%gM# zeCu{K76w!lFL^T&fG2dikKmz1O)Oi@^rQ@Qdhzgza2c03Q+u0uz0fxK0FcJ|D_kCt zhvrN!j|7?n%@akTn|T+^4LyDX7S+! z);>9O6K>GB3x|zD^tUznZCZc(8NZ#b-yZ01cl`I;kl zZowV6dF$H6FqzXc1MbjLIfSxQrhrH*Rmpq!29yAiOZbG-Suos#XxI7(h|K4`3&UoapMX2- z9WyXaE-QR9UV1T>Pg30eVY#ueedeO@0V;2ezTi0vxcV(v6Lt^nxcIy_+uy{#vpNPH z(pnebf!^E=b4oN`ukOO7C4rw{3zm)BvBi6W$gVtDGT zQA#;cg^SeKzozMbG6TzY7KFo2=Kw4*wQRIy-N~q;rN^JC2r4P=zIPEuElIh=iRoyfqBCF$w!UHMt#+b{g*xOKJsyV~^2d|8b!LNmV3Aas)L?W7o`9-7fSj`bCXlNQ#Z5)6lHFFb_Z5}7PJRe z)WJ*#*aLQ)sLy)=%Bgl@6nmqg77=7`y9KSoduV ziXP;va`*h-U*#fh@tf-0g$bAut!iXY3Ckq$y?~Tz!SmjH7C>X_ru-y~W6dy*mCe7g zt^AHZ>9Xhs6(EA$DsEnxhwvG&l5(SJ()EAODoqG07dy^{!hWL{gD~e^_#<#iwE02( zVEiFq{AFV+F{HfbX1MokRcs&v3S>Rg3q zz+Bj}%XO)tM-%k{q6^J6SUE+aHCG+9bUF^hG3{6O!2%67R{WO}R|zA%CHo%2A`fG_ zL@Y4YZn%o&Qk4zR-U1~JRvV15DQ=gznl<*XsE->^Ny9sL*cd2qPXtw=5LVPa&RT6> zyI!g3EiBpiUkcE@xpl(!ADA@*P-*cv^^6=+2eP0JRt$~uEeEWa9Dk{gX_dcFC3%4T ztJUlFKY_A=O_|!Df$R=qS^8^2{AcVBv~LKLt*?H$vlopoJp%l8Bzu)!A|6n(M6Bi^ zrVfZ)l*D!U?*^1u;`i~vKvOZS9-M92&+LVQ>YXQ&3MZ3wsyI1#lv zV2ss5G-Ei)QNFr3Z+jx{l1Te;7|ebWyT@##ZImjxX?5uKdfB97y+`04s<7{?k8ePM zPfZ#O&f>$c#vC|bm|H2DY%C!i0Ko7RZ-g3%N0=B->@HCtl*ezCghgT}^QsZ-im(XI zfzREOR}$4xB%M_g1tJLF;ug|1+>+d%1ZQk4Pl7XoxM%DtP#};Zwn`mMG(StNj)pUg z;f$0U1QlLMB!91;sBJ}OKzRbHC#a2U=1~-`7qUOjXADZSi7S68)o-e2_ec1yg;DBM zMNjyBI;F21vZ|pgX;pgK3(99F7Cm>OTM=%6z6 zMg$h)J%@FEP{?uv(7G&XNGFjW%%Au1^C^HQ08$Xqx3~J?s{06>w@;#0!y#CKJR$!T z95ON3SO#NxQlAl^?6~3&I7HdMLqAa$9dhbY3N#4;l-xtU+!%>^35Rt5i24H#5yse+ zb#=L7O2Oa{4LBq{m4G(fE{8)z?W_zEgV7;o?kSi%0n1WQVKK3(NSg=LJ@1|KcR-b5 zMHap87uMPv_BHqTEK@46vhG8}f-k@!7o_D@_rC<3NQu*40AKo7yu%ES>Yivx!PWIb zVy9${x8f6zVAiX}R`X0RtA=%1D8L?t|K#;M&dioy~m@F zj+2>S06W^ERiQfyvlvC?-pdqQ`lu;@5x>-Y0-QY!*XYvXTO{lV^<2tp_aXi2UvUS! zM$x7R63fNl8W;~J3Sifek&jd$Tusv)*-<}qjvA*3k^q~Rrt&8ZA6MK0P()Unq9Yas zh+smqA?eCteo$$W+n)fBz#L>$_ARq3T!FARMU%0$eu%(~Q-?FqRnFFv2k~K0AEL1{-*{{G zO}GXkG9oRX7RT1g?;TXH8nI4WRDh{*c0DI0Rh~)$5^urZwyvEqWDV zi4IXkH#tVXA)vn-ZsMfoiMh>ETy++9rz_c*Dm3-{tKp+X0b4M^3mq3!)HSfm$%i@j zE}MOj8>0xst$o;dTR6%bE>eAC70?BIJ7J7brPh`Nr##0E!CXI>Vwl0bNVQz`hN;UJ zuL3YXPfx)o@ndiX<|9OJ-~lthM3011utDskTN&!4vfAKT6z$`x7xW;*niw_@YtBTNth^or#f#^If+3 z-rTjpw++Ck&tDPXA{CObJwOO1;$8>pN`F^@Q=yyG9CNWu_g&R10`gvUcIf33c!&|k z-ej3#)wSwgO+DBiBf1x==mn}zfbXiSAH3;!A)JBvS_sp3xVHC=%l!y2R=WcwV6$Tv z_^wYF0nZe2JcZbfGO(diXRrc3_|XWw09%9VTlk#-$cvvLSV#U=E>(`2D7GAhBTzLn zAgKwRQ3q!biYr5XLSO||sOJ?Z-sV$-Rog@7&EwVvya1syc;9OXCrj5WVgfT{{@v$l zG3#~w+O~^OQ>YIX!f&NgQ1mA$?>I@b#MahEHEe$|cbnmG(hpT8XAT8W~muXY_p@u;aTU1{TwORv)XS zwOEa@M?}TSrc}@$Ca!@nfnFQrc5-+9TKlQd*IkV}1eGkYJ-&Qb!QJdVzJwrsAGz}X zsF9MsUx1;s*IFhg))23?KQ7h>8OD?7kR5P{A*(-|rUdM9@krc zato0wk}>!*0Lt=RQ3Dt)_7=k5<2u-F8^|HXO2LXem5V76rWFk|g_0^vIuZjoVQ}$Z2|k?vV%gP-SYrk#txK zT}A@^Bku+^28 zVSvg1-NWd~E`-z|{*aBQLF;6G?$!ZTaMy!(9#%@4gWh>q-wh~8d5pTPk$0vC4=@AS zsRcP;5>c6}`$j-XK4)gnkLPfaFYi$6dVf0XkIDI)c$TjjspQV{BV2XLtv#0#qwuGs z>a)e4#=OM02sw!-M>`PQk(mu$miBTJBkxO8a@D*NGyBXjaL$g*(WV9 zRp`T3=pXtt$ggf`A*17u-v*qzT^V4pAII+v(E%hJ|B8&|G%#6b7A!(mcu{dcJlnN= z;SN2rFYctZ=lfY|o~h08PLDwYZgF&3XC^gsTJ!AK^wWt~wwk>37^B_Quyw>8yK{8r zQk_s4g*uW#m#if67<<5wJb4GqU_R2FS*0lM=uuQ$i$<1KfqkyRTWzU77vLpp*GU@8 zO2v_<*HDX6eAA1W5fY|fI=rU&P53qFqSVu##e;X|3OC|-h&3r<_V8}@=JLO-pkfrI z;sERQY~5!OKwZ5g5bL3s<8sB4iFHG5(IC#JPxPVvtR{TD>mBoFYvY;>NtHjan%UE; z>hMs^mI^Qz&!*Q8CK&m8K_PfuwBuKUeeJ1gL!US)#Z{U7hPjluMuF$fe|RQ?0?y)! zjpb&=SR1=$(q-0z&TxQ+jfis;aT+eH=49Q;M02N}YD67lf$Gim`eU_{0ljPAdx-No z3se{TGdtx=i;BMm%Ls+M8n*Vx@+hr52n`OdCf0=6x>P-J&6^$y>y+j$<@dy;Z6*zB z0Om(JtQ=H_S*jk{ORm(^A106A%2PgdwFab*oRK{+jV1=xp8MHqdQD4ZEkv{cHq|Ry1FQMA-Yh|=$pGoe8@eT_IiF60r@y=_* zf7cL8aHE4%wiNSo@2e1BzAwrN7E#NfrvJb8WmPu{)w+p=BIzALRKI~#QD^0u$Xflx z*1w(9#DOi-H!3(MM$_Ze!S7x^XphkI%Afc-#n#e#KHNu{9z7$Hm2^eX6R)~#^nUv#G*q7tqts!23!PwXH!>F`)GtqCX{m}3$3Ul z<(y5S@Nw7T_jht;S(P@0qt>xnPOe(xGt;J?ZCC0536sG2nFfHFBpKRFM%m6aZ`U9x^&a91YFvN+9f2kt=9s zYJJs^`1xp-v!;$mUnkBCb_idY zyB5u|=rL6!3+T3Jb}nxp-$kCaO&!cEO^mJ)sVplfo-mwtl578{i?^P55flw}BbX}y zCxmjMvMPY{d>y(S5$B-};HxVT=lPT}zu)8gFOJ$jzdI?sO@SPcFP1G=^38_|orj-; z)~u6_|C8O28|L$!Hn*x~HZqsGIn_B0t;ms1!oIXk_IgtCvRtZ=jyexNzyUF!q_B_3 z_po26*i$A7nhVvXgQ#_`iP{HE*cg+q5FL%JnCELVt4P((V)(L@P1tgb(VRYZf`#Pe zcY1FQQv{(WmG`dwX9&-5o2rf16rcOR>oOT=bN@%{qD2~MXt~|`gmUycl`clmEeGft zs$BhVfkEYl!C=3M{YoVnjRY?WaFptcQq;1G`e56t>e4bPtRHk%np~=!+Vmjms8Ll) zQfK6vGViBQ{Jx)aDp+6K6;!(9i!y^11${GDxK%%yImsy^(DwOEbq(knytEPGU)KHAli zmjxwzZ9#ir4V8FbOlt1Ax!jmqau(uC2|-ACf^GvK1pvrdO~dL-|IO*&15447@{nCf z>f_tH_SQ>(=HJt79bz|;qd9e{p#1{PsV*?yYB-;kT-i19*%zRqwZXnQ-Z3sz<#{4) z;s8}XSxg5#Bg%n@M?C#hr(PO#x8LjNy|V#6T$viR24b9CC)}r3 z$H~ZUfoxtU4W_#k2=WJH}f`sn`%!r)|_ag|CK;W6KTD4*Cj1Efp91f z=0GU7v~JgfR&T8DUZD(=5kt5B_}S$kBfY!q>_HfB?Gk3)T8mr-R~Jc*5VUNmwAh~T>v;=_o2yNEA@s#zUIuxCn!3u7E)QiAGkd7j z_m&BpQLry%(rPuv#T|2S)wXX}z=snWo= z3+cH~V-LxIR=zfy3f71??WFZ+puBQ;dEIx4apzopcg%n75C{1|@8a;-)|?Fi`E#$Y zp07r2AQSMI{4*9nB539j9v%Rx|5q*shM45U&mTaV?cVt1%k}Ys)gTElI3sgk+qE30 zw>MWhJH`9Zmr;ev$I9&~ozq&+kX-zCwPu64OHD0rR`kj%gpn>UuE)&CfU6#fue=9{&G4%6o!_lo`Q64;l*KzSyano!g(-vF5u^(Di z*S=*@^4pkQy6Il~wB~#HQ*m+Q4fYfb2Kf@PLWGl1bZK*jfAKS>3akTx4mlK@ELYx} zGqB{@Fn!nMq`6(&lJo@M zX}2^*%EE}f2Xor4nHF5sTvxzZm`<~e_dJ%Ym#&wA&%y})plCnuQ_MN5B{AIj2Z;{U+on-qq5oAV+$ zr7U_dxg@$9Pmw-Of&H{S6;UwCN`8*y9{}B&V*JS3m&2Y zwk&U)8?E+rK5yS7sNXJRsP)7LVc8b+#I5xh=uz}J1p0!&W$3w^JZ9I1S7ah-E%T02 zYmMNuH?M#x6Qtzgv0}2DeQ}cp?zrT_mxA3^Qf#@JShqr_J=LWnpKX=q+zKYTsK~3T zxgg;Mew`FiL6DbkocqizrJvn5ArkdB?s46U4JeM@331&&(~&1LTp5-l0s0~|rVwQz z<3z#Q=(CZca=`s%D4X=(94f~tm4LPxac0$*?Ei}?x2KBwxf)4S6Db2jox7(zq z9QicFcZZ^iYF15O%FKBX3~tQ{{v1&H+)nfAG&y@k9Ol+kS3rtJy50qm1Zn6_($M}A z6pMi{5y}?HEP1w4$@w1URCtVk?|-^l5d2Ky2d(ln)&Paw`O*B+%bHN?vsgD{Ycf?n zwcV|R_;YGbgZ_v*!avXY$jtnx4JIDzNV*@!HSxx46r&T6iJ!OsaRiz8XJ8+Ck*t@@ zUb}MI8Djk{`(B}Vj@o7gnBhN;yt>*0d~ru~cyr|JsAqcC+xj>#Cls~RCp)8d#r^2D zbg>2NOHP9p+mDYtuJ|nlAL^ipK13|jv6rM`+N-GQ$9(p~{C!F*gV9_F2k>V0)FJe@ z2H#y1i-(8}&tKLE72v+`*0f1e5ery*@u`J9=IhP#C&H*1`K7T;#uY6Ix8lRZFh87m z=kytQGvB`6LO-j>1VFLr8hD3GhhRWH8&h$z3h-hEUU3*e`A=-1%LFp?J(^^-1eUynh za2vA|cCa61L*cOJh=5XHDUGjJ240$^2;^e}YjR~~q`9!p0dj%d$f*lNcF#5=gz4nn zJ8>P(y0J4F&EV|aqL77G)KAI4&MN%BJ}Uw{>+gFf?!Z|Ok44L%v!oRmc)}{<7??R# zgI@i1)+o;M{sK1hss9Cr+NnvN`mv^FA&=4u!>~mGKS7c~V%FMklTTW99#V?c>C(^8 z%Pb0T990}@kSGg5feX~COQ3uH8+jmdWo4@iSrcHtXXtGsrNyTM1|WG`CYMEETj=hL zHJh_t;ef$y1b)!l!m!j1>v7DU(b#{1Z6|bRXzeoLBRjz^GM9oje|kaj+6VaXy}Rci zb}JlJ00z11NvLijicIu0*H)uTBx+!O(xtkNBg4=^+n%!zJP0;qB;I+Wkdqc}SFQJ6 zun6Z3r;sI=0WSv2fGdc@A%rvT!q`O$v5R<=)@x{JujBbqI-Rx{@P0< z#H;3N$jNyH!9{!2P+T7a(VFJxD=~gqg2kQeUmsRtib7cOA6_#74C^}!WLV`oRy8h$ zumTrX(!v1F!8h=dk+t($0oVXiHUTJ)0F+*V z=v@No@Gq?%RfZrkMgZ4gUC8z#5Foi5KL=L9A8*SIJjshPTD<7|b0%&MFo0(cF0xJ> z)Khy_ut*i4Ai`GL0bTnx;CGF2>jQirWZ;P9?+ovwoAgR!js~H^{nijv6Iv7+?-yMR)wooA%rFQxD2V8rdU+DeFprr-S; zkLOTib1$fASrnxWZ*9F>u^(7uzi12mhs0!{QK(q+MUkjO5ItY(ID*ZpE91`e=z3Hx zE(0N>!_jfjMY;@>f(Rn+o3rbR*TBCseAsXl$YA$Rj;rSf4*5qkd{KZomo%r3PS>4) z0-AFs)wdS~$f1xkvf$RK>4c+O0fMb=R=|{7AY`-+ud*dTAEjxzndtvR2Sb|ny8Al# zZ8~I(Z|yE9(gR;V5{7#H^rycQT4-4AQvyRB^MFU3mldHno+QLvJ)TDy236bxU0{zA zjfj3hTrqsoQKo*BZO1q^as#;E9JrsE{iZr%Y-P();5eBo`cuROHqv}C-EK9UHo@AN zQw!lr+PpymH&`@v?WoGoqDLt$%kPfLb1)=1!{lwa}} z5IK0#zsQ|{*V`|Dc03W_uXqv-38Y`0vnZe()<@5syx7Mb<@Jke;{pDF!-^MveGU(= zAB3l;{%)*A@wZ=n4j%yUbD#OY#w2SKBw+OZjzgyH-`@wAiWivbpA`VV!oY7v`prnc z8R`F@8Y#$>RXS0dW-}4fl0GQAuhYHBsbn}m&ExBtV5hm^f~-|){fYE?ZJ~^P?U!CP5`Uw2!w2|+RHEF>bLuaTrS4vpKg}-iGK4U4#to!apbU$rEe6vhOR@3cELD^JZ z7EGa}Kfh8^gtj8d^)x3lkm;MtChni-`@aXg#^2r>yxtE!#N1saC3L-(%y>%WqdwGfPG+ zhRdeAsEYAp_vkj$#V!5stih><@C&NMl1X6r)8>0s(tIf3OtQR9n4Jug=li5pnls<% zgoL?{A7|_Cs23#}^Ddm$t%_y=>^dgIhQs0}(`5|5w6x4(r?FlkzkY2hnVrQUc4TH^ zC14RtN+UwAr}*wJo9&HF^DLi;qsDekKBLpqvZUfVBwH?(&-PL|^9h_BZ)TcpCrfLu z)s>U3T-$e>Z&ERNbjMtDtZVTLR+ARHiC*JIA`m-?Eg^;@q0UkhQI)a_Z~HnuX@2kh zIxzb*2HJWoBso+eOJ}?O_(37><4k52xvc;F&4(e!*_qWcTC<<1E(M)pM@;9fIfI$< zFV95 z=S_}R$=mw&U0{b+A0{sf9a@*`7tGWzf}jvqc=` zsM%0C6p)KxWE+=7Xbv1TO4Vc!7zr6T7Ib=8u23B;pAx6c9A?UM#^p`eFTM51tJH_G z-**)(xQS_D7Bn%8mlfTRnnHJoXAe1;;$}bU-rSZ8F_7Jv!_+}9hj=f0W{|vZQZaj~ zJ=Tw&6`o45NnJR$1N+za@pc=(jT>iI99p-}?Ali)P6>z2&}wLvz2Ovyt=;z$ zz2GUIPkJ_cbHfBX*NzsbS+txYr&T$;8Z{hv{ia0c=7QW?!uLaoMZTVq$nTPV;NQ}C zVrFgD-{dI=^1jKKqEFjmS0GihLaeJtLFsi+*i*^U?R@LW`#p=dd0#_80Xq$kib<2qn0#Qv)#th4QKluE2lQK&uk@sxAj{orb)j#RGv8$>QA9ISUj7tv{t(t#Q3z~z;sca%Ys;>T>ecL81lTQ z9q4pq*S9?PFIw-N2i};TT2bb)@ESM&y8m5-)kJYinX4m*g%g>D7dB`UKhgTSd#5!I zYvqg*PiV70>DjUx8*X=Asd-~cSeX%v*fMZ1i$Z$O?&OLepYPA1H>t!=T2)%9XTM9v z+{HN;cM-f_3)&nn14ZF7;_-1~@F_vYwo7u6xetCM?~I?Vb~ufwEAfmnbC>emsp%1^ zYvzN>=`2u~v9UDvh^MI_Bf>?l9Ref(S`2`;GUvJ0HpXFj?-NJ?1*Lshn0k=I9TE#g zP{6*e?WEmhAdP9m1+?gkT;<97J9(?)Twd--U6^Wbl>H_nGOH*Ddw`%T&x5m{Da!$+i<4U!z7b)=;8p9Wj(rUFh^Y1?$ zU^Rqgme0JG4>i2>N!iHjvvDeUCYbD)`NHQ(B`7M0+hU6KFMO#AS?#I>@5IdWJKphv z4B-u|=RqZ%!>+pwImLsS5?^q;^Mt>y>~#3PQT?hHhdqsxEnVaE{+!0hl`!WK0_b%i zI%CSW%ChSf{Uwf79~|nd($mPZjs5%%E7?~3kEizY-V4m0pquss-7W6OT(_c+q_hebNR|>b3r-L{p7h!bF)ziZ@-MN zZdCsRUI0Ak6)l+gbI}TcEN4f0)_sd>fd>nn-qXNws5>q{;l`PA+ch${6e*uuGT+pd zx?A(F{`68lB*?hh(jC0@z)h3iPkmP%Z4r1iY6D6A61IjE0Z4*dJBgdpSIAF^&| zKS-vPkH~ozReejdWz)Z&lhe|Xw2|eg4yNZU6~xK!ahEC?dR3KJPAbz%n0Y62GgNV4 zI!tcfIXh|2Z?2tY!>FHz$VY zjg!0O1<0nut8wLmj;mF6;9p&0bs-fwuU-DAtOE;)bc%t(X;zsgJ+mRF&qZeI6pc3I zmTMDo?_n{&UGS*VH2t3MoXey|&0bb7SA{Iq`feJUs{Pg2#! z@4ZDxy%2fk?Lm!%%a2*pZz?Hp+&<5tvF74&NyfFx`j^|-k@UQ=MpfULw>TT;?FHhp zhy|{3S%9&ZTuX>ooave_7teGe2`Wh}M=bl%5E1|A$+C#h+Vefm5#-g?<#Qt?@%B*FR|w+)*T=mr-VBV%rK`+%f}ztCuG00*nBEDf>}8rr%^jn!qA-ZeLL37Vh4i= zYFEyEt<=nKwia{aglo-)C(W@boN3Os;)J-*U{7Nf+=zeahT;jUx6W~I-O{+%MlVuR zjZPvBlqe2L&A*o?jWt@Mt{?C)Y?k#qjE^|bnU$=xoG$Rlg`~#s2d~Izr}|YNS=`{fm+3lKG5h9Lu<$ z4SDm7Mo&k>C-%td=WhO%mwT_Jan$j-&-7a%E#K!xnVj!G0p7dWcOmu<2ubt=r>Jx; z8adwL6_Rz zJU72VOKnDBI_FWNg%;S~$eS_#4w@6un?22?9HrpJ6WH!X@jFks{60mK`6B6@X3q|z zWh%-PpgL3Ca4_tVwbMy*(sYox&DRdMwvax}=_2h)$>~h$mCA)G%6t`2lfhwADu;WE zLj`6m4LAjApR|uAs^0#~4{m>7r^5N-`mDd1$;k^{)sNvMEvxIV%RRg&?@}-mtF*om|DeDH;Rn!L&&)uI?+OAw`fn#fgxqz9&t$7 z`%n5kfI^r*KcQmCBmbZsi?Tj`VKPOCSwzh8ogb@A@%pw7>lzU_za=r@Q$fNDn%&Eq zeeEkou11!~l5#JxcU;qJnS0~GzIX+B#yVj;jZsE|456~17H=y6)W~3l2F|g<*Yn{9 zAIcZC$iK@Xvj_RL=347|pIl}$TZGEN9a^agCT~?dXTDDeS^kwF>LZ;Gs}|QF?n$fd zfky^Tt#>nc(GS>>{4w2s+|Q|1bN)$cvN@69?C3 zPk9V}^xx!h-c~QG@W-9kZ>6TPlcWkpJO??eW~f=DN4){(P9ggteIJ@HwlPS(@P=`b z+D)vOjbr#aT;|Nt$WIHrCSqwFLk@w+EVV(2e!hd`g{M96pxUH9(icgY?@1)R%-}p~ zXcW#^h}3QJwR$t#m!S2WW-$9hoO12*&j-4zK+zT?g*~-a)A<^|rTzd|uu20;QRXUx zw_|i#IP?Tg3$+{hv^B+?t1TY8V_JIF(U4jB!*G1U0i1B{X`eoIn(C*{A>-%Z1m?u{G)_)T2^ryfsH34xx;a68L@-q}>^u2R}i-s(+z1XAQimRwB#mU{$LJmP{&o zO)GXwDKm8kJ@cto?yjNW@D*0V`lPO%BS+Q<8ce4 z!7{~ek_z(7=xt{4j?(HEHiEe*<~T<8(db)*_yT&j9GgO{^9T4;v*i)--t-rNC0)!+ z?J^hkA2#H_K7kI1>E-z+ji9$>8;xY27hLsT-1fj%(%f*ccN}<{XYxu4v3>BHAjrBo zG2iPWbKfCm2mLr<_Uvn+$j{! z93`QjD3T9sQ?6dt}ltPXJk=gAONpF$Nz)uhZakxx1F%8iO(UUK8A3Jv; zl0V94=w;C8{XeiaLq2=)ruP zyp+Ud$5wHAW?oiGNj%%jWT5AUcAq&%Z}KQ96r9?mIcmVmgb|4JSkGx$oaJN~j`yzy zrGFkN{ihZ4`hHVlDX59| zKZR};l=ke_)~F-IPdo;lB!dJmDgiQoq==)TN-mpE8Rf6q6e*|qr5$TDb3MEy1Y^$_9TeYb$j*w+8529>f}8hf&$iCR|V<@#*eBFz5) zks1TjHOa2l)f|8D`g<@=^DBh1#9Aga7skTEF+2;r2MLGd{-B%xnPy#+*lZS ze%yMg3ttTJTj8X`tZrTN>mOSqtzNlD65SYiEjBepu3xjy)P%hPE~K;%*~a4hCL0E& znj}B3=E9UaR#4oWY(M*B2$|WukslrixsFG{dTD0;jbr!g@)PSyCUHz^LdR8Lqk>6C zg%z0bf(05{;pGWl;LK`%dHcmSR{2w1H(WYtu7HHfJzub zOq(R$?S10O!m4VdLq48G&x|$(ij>t4SIK{&85Ak~ObeTfsytvn>w1*8hGlKXR0rZ# z+?(6q@!(VD)%VEXPU@y|oSgF#y2G?cjvUS$nZs}@Dalw2&#@K`ZucJeAMfqUlv-9_ zh3+W6#%0%b*L3IQF2vl($ZvCc9toj0hI(QDle>)?`|ok{LCZzz>z*f*no#Cb0Iq`p%ts}e>(OprKv#)UG$6ejt{R5)# zVCkqIqchg~(dN2T3c+I(7dpXT=0=)`)zyr5h6(Y8YUS*4F^DjAf{=mc7=Ey+-pZiT z1^F5fb5UYD575}1w3W;fVgk9Kq|;>#cchtf$@3%QVb7ks`G6@rPzbuMStCOIp$ylo zz1XF}pirq&8oV`B_YRbkMd8FIN`xGvGjh5t+-54y_#!9b9yX|afD7euc4XPmh0V^_ zi1Nl{QB(F67E+~O+Xs#Q8plG~7s0K%*wMF-4p9~FL%Nd0FCPV&gAnw#&t&SpYG!mi zKA-etVSbjvN~8R70OI-*Z5(i#ju1+{&&@vh_{h{o)qQSUu*H_l2J(?L^5*-N8C`o* zuNi6MJuq$^Q|0E<-hOi`qt0b^Lu{PP*XI=V{?WHWIMv?$Ieycfo|juML>gQ^RpNfI zg63pHPTQsN3~G-*6{0q&0f=uSwiTcGz|0#ggZ7;T#-|m?h{5BHw9QXz%}!g&?e5t+ z^1@dY>3f4N^srVA@)9n4qObr=6M)P%59>hifu4uPH^oF(X-w9?5Yig4KF*}*Sg_ck z%t~!4El71FsI1%-jEF=jdgy+^owMZlj)x~eBypxY3>Ra|4*hsJJw&DJ+T|MMmKB}l z^Kan)>_^^b%-k2+FmO6jz62k3=0{etjVY^$*=)%#bgg1>412KX%gWCX63=h6ZZ0i% zy=Aj79vSAljxkON_>DZJ403cvG5YbbwkIYWZLhoMQ%$u^AVL6eMaB}@;fwiQ-h=N|E~PWK z-Mdohfo%||Hnr3CD-4pksKwX3_44Z_;-?^9ZzRSk5MF*7(mJ+zqmt3oLuLh;-R^#U z6Lne?wc*^R)C6$esc*zXO$BW?*$n5@X;R{ze6)=yKL#kC#=^(q~p7Nwe0gob%N<&VSMs93v;I6S?K1C-X=QZDzK^# zVHn=A9u}z~cNvq@fArFD*hg>X%17Lq*EF`bhujHv)e` zy^D^_J}hAL%cGj0h9Tt+p-zwlD-QvN?HLYdfy`3z%bV znCQ6UOZaMbIzV^huTQVA#@CmwQU7;m|LdP6E*FR~sl~E+zamsXiGQM(jK_Zj7B4ES zl=hgrBjZ=!^W-@=VL0@!FaNtg1niV{gyyAR;|qWt;l17)_m#V?`&XYb!Plo>_4p;$ z5z0CMf9fw$tK~15iZ9ha(YyzZr1>@Mg@OVxo@e?0(cU&-;DD= z)b4Ln_1jea4{P(=$N3MA&~G2-fAGt{{QKXG^Zz}@S-0z?H&x@Q^-_2SE;@SH@XyRc HcGv$O=^!?a literal 71931 zcmeEu1ydE^7VV+CyHiBEOS(}h=`IQB?(XhVQb6gJ?naRk5RvZg?)RO0=g#{NZ=Qn# z;)yT!UVE*zPn3$190n>WDg*+-P>`2?4}rk>Kp-$J$cW%K{LZXy;2Vmgyp9V5f`$kE z3j_I_O$>ohLlmSX)IGBgay)$1HSV6pJdSccD=5nc@!(1#qr5MIuMMOo{BYQuPSA$j zhFl?rK#+-N!Ud+8T_3T(mAmd{f4 zY&){rn8@{lTJE;+CP%Oyb-oHCK5Cfc7x3*(mgKpdR^Wer{_n45Da#I2|L^~kg);g4 z@4={HzY)Lur2OwQp^yG9QBo%deVPA0)eigrKlMM0@&EcV-4`;8f4s+*?B@7I5FmTI z+`WEUGMDo%bzuHyw zxncb|A^V;&wGmb0im?mrC%n?jv&IP7Le;$eXEzz;)qs|+`#&No9exBpKWEnG8|_+H z$(H7ye{1JIJn>{3Sp|Gi@fT5_D}1o$`3`%73WMJF{bNuu>v}o9%3C{;(aYH?rB2`l6y6 z&g7H2j1X?p?L?mcYZV?_<`k?&**MXCqp-ETU%Txi)pGvTCVEW-#Nn#<+=uMXSv4%>SIXgb903Mq zk2xm9UWz6<;&71<4svH#%Y;mBCjre3?iA_X|dARzeQ+kd2KG zuJQ+vUVe6c@UP@#Ty+hN9$P-T7cUTWE3Z7tT73Q57URu^;wA89K3G`LatYzu7n6Ax zQu5C5W{L%fuC1*(&Kh{o2=EXmPw=IsF5@aBCuU{EG}z3p{n2xfS?=^(J@w3=+NX}P zD;KfjBRe}khk+Qhx=N_1Oi=05y|Gypz>E!tAr73WDPgJZkEeL;c(fomJu^e|`ZZb8 zf-_ohNJx(}i@5(IVwz@k*rN4N6yD2~Q%`I%vY4V&dih}Zbn_YB=h^&WB>N^Z2?>dS z;cuj2XFif?B0n9%>$1whw$z}1Q_6{U+yB$Ep2Xp}Lj5a;pQyXK{^ZpxBwv?&<%?$D zataBzh4Iqtc{{3~AcIo&jhxmlR@l10^q)Vbf41vu{TGGgDY1^4*D-%R`8KwmPA*vV zSLoE&xX!&_5Ll7@mBoMTQPJ{FRTXD2mISVRCVF?O*z886?+gFl>i5^2?g!cx25p48 zIfQ&!1mnWz2gin8&u-tnS;q+T25wH*o$6NTJMWKP`#xUI?ku+PM;0HVWWQ9-?N3n_ zMS(!q+52+u5@|N_x309Rx(L7Kqa}gpy)6SODk^m(?E)K7K|w*S)4HrxgKn2%snNPO z+ZpXxu1HMR(_wY9z8{#|FY)(pl6CWJ(%|0}PEs=pAU6t;xe8Il@fUypXm$MU!2(-S z{krJ<>$cOuU(Om9w7BqrFH#YXtt%J2sz0Sw`_DEESFlhn z%*Q*i>xwOd!`C)?d`Fpp6ebg9xStYPP_QP;EbqdZ)z#F{@zv7(>-zbdfTs%i-JpcDZW#3YTWlHRn80~JXqs4uT6s@L2YfS5T68wNy(UUB>JzC{tTRE7xlPH|kLw@Xci)eiL8hO9w*B#aHM)#69)ZC@s> zj4`=pnNe|rdW^eW#ziCCCl4ozm5qciCZrty<@i5JM>T!43vehMtK3D>x_==1)7G%u z@t)Lv0XaBuEa&xl4-%EMr?$XR^CRZP;XDy7yI~rdWnXKl-`jO|V*VxtXV$bf@i*(o zudwf#nlQF+nJcBF24o@OQ-$qf`9v)=3Lp1)r9*)kB`8qZ7E8<_#K z?1?4eK3Pu?TPuEk2xaejz(ov!0bkx7ltMX2WqAz_darYoGDAiofy2Y{EtJC5Rg0`% zayL~UJFg2KG;!&F<-8_R7QLk>WYG^i-yXkjaprp`^zDJn>d4lf^_$9*Uj9$YP5y|P zy1(B);?b-2=Szf~^y4Xur_w)MOeuF0R7m}l9{DI(!@{gQ%mNqEL;W&-ML=*c(2m85 zuVDw_{h%+=k8l1%y8_LN_6_I*BICQRWoW8>6mFZF%-zgwy>H&fhsWqp!^CZSaoGBa z_@!yT_$E<{ji4JSK=JZfwl8qo3sUEnx-QUQ4^qwK@TnfF)9j*$$y}YecB3uHQR{t` zY#wvErb2S}d3zZ;Z$+2nTUb;YX+?V+`5yZ?=O4QQ^2yWkb7yt|%Z*ky2cKKR9YlCy z5eeV2wz}o3O0STf1h!5D2U zNJy7Q8ihf|XFORTGyL;saF6jxy}JN;ysVqC4k7u_Es8<5(veMeI$ZJQw9PBCXmg!r z{VL6dw=Q<23=$s$xe!0sA24ci477qViOFd89J=3Mnk|dIQb^7_t8O3XKxa_SjRjGB z6>O)hv^qR8jiXQw%yv$=OYqb#{P;Nxt6R|BN2!@a zKRq=A%1iUiKF{Ny`E|06zU&;YZupscT)*K(36f1PHP2px;^Ww=qj>pj!I){!(PEo~ zFeN%E53&bl8(uQK#0o;u*RM2;jL0Bn6)={$SgWnSk-&!Ux>V9BPi+93_Sf#oCgFEy z=iyI|SfN4Kf#dl@cMeinzdw+CkU2n>1A)?|hNOVHz{Xu`2jh$(YtjC?4Qbit)}Frn z*d|jJb$N#AAO24VggAh7whrd&u_CajWK~qK5i!XlF)4&r<{i3ZoSbkP9hT#t9`Bo4 zStOA?`LS{gi~iq$lo=}|C?S!|hLgXk6f+X6fBct>h4=Wt;J%?ct)df(ZNkdHqj(Cy~MTktM_$X<4>I`){xzxdP1sgA`;$hItdo3({Vg>1T8k|5hlO{+ktpkTxFP{lE*TE+u zEHu-Vv%1s`-R)Hu6kohB+i?DJ8gIk?bD^gOT{}E%YS)%WPKEIqW5Q6OW$T75GZY(& zv6AU&n3zcY?k*%eKM2&-*VlSqT32Y-N@`q@wkq<3)p?=?L$XE@Bl=bi)I~T*%r|9D z&-t@P_cr#mst`*`t9ruFh-y5J4YeyZ>0i8nf>0C`6i7a5Z8YTTLc)x-M!n}NSg9X# zxa4B~2o%L34Cy)T0K?=H})CZhQD|-n_}5+D%MP4{vT3 z03*c7jz7nXOcUL5Zc)+Gi>ayM(CPjInaL`d=4ROovD^sp^v zHV{qlx7G}4$^YICL_Yb)gC^%^)qkqRDFh2rVe~t!I z6zhq?R_>TlN+zJl-BwjIfK$N7Atl_h)Nm8Z!A5}Tp`j_`Fm$SgML<9h@P81Hl$1 zFafVKdPym%q}0?~8Z^*VLrt=O-d!)3=)LN)r^XHO7`yQB$LHpxtrhK+5W= z%`PjtTZMsw?m9FgM`3m>mJwPS8W^YA+MVQ2yR+2)Av{WJjlOho;-p!3? zs9bnC30_t`xxp)GS^<|W^q)U}CT3(HNh0Io;o*R~sLG&?ua1=2g?0R#k)1u_hn2MY zJ06HmkC68YBNH>T2@mOHxfVMyI--M>1gmGR+=mbJAP&+?yuVMKv6hjQMfASd3*s=y z6wYl&jhWi*FI6j6QCAN=J9F+I8Y+~Ez})#;OGjHsr?Dxcdcn#|_}9*zI2MsuQ8rR? z#A%-|3j?G(*}~K4Itn~m=K8OeZ=JS&k!);jfqU(`{|BR#LJcPm8;c__-0A^eR3RCx zBV30eR{0E5f|ElVsyz%ZFYjLXKpF%8B#CQ{-XS9!Q}0|dEBY%z)BWM?$wlIa(mwkeE>5$i>>!oGj! zX@7pYzq#&u#(0~9AUs~a34#d{-b*Fvp?Hd!yOE=Bb>bbMY}0P{MZnSX4ja97>9@`0*TTSf+cd1 zNP3>X2J`0aTh6l|N!SRgieI0Xe2z^H|1~%Y95}Y>6Q52XzsnQt7}z#&JfqL0x`c(S z#LVT<)!S%wjVEkU4p`P(m|Pz%u6R*m7E2M9(b7U+T#m{QKj3aE51Wz%v*!g5vjF2` z@RZZvCQGWV(X0p&pR1e2E(!Lb#08(@4$rMmme^EO$6Dc<)dL4u7SQ74<~;8?HsEPL zPrJ)qMD*##EL->!*5tl@*|bA{WW>qE)Hu(-Fb4$`%}y;nJv7Y`(}$Z=Miv(7o>Exf zvmY;sU%MX=9R4<9jc%hNHXgd7$Q1G6J3Bj5(J(+z#m2_Pjh>iLS@1m{^*kLQdfYo| z>j6AZjY%KkyIf%s6BCmgrvcPZIqATexCNqSp40B{7bopxl#=;Vd-*fgoW|XdSc=!0kE|+rj{*V$#L#0-$ZiW+HZs1iGy!-EdeXvolsabuXm)daY|sS9`;>+uT{BV`5fa zg;rc!x;fAbXbUI4z9ot!7sQovph8VFD&FArcvi|#)A;PBOM!?^9C*4uNCS?W*noQZ ztm)`y-b|j4xS$BaPleNyw9oC66# zqIimq^A!#qC27zQRY3dG5)V3b)>xu95D;r_Zq^efEA^Tz=Al)Y=lLi<4lZdh3Du~0 zo~YkveL6VK-5CY#dJ7C|Yn$noz;yxXP`Q=(*GJtulZ8o%sSrZv`;V8p>KgjJW(gwU zt*t`H!4L^-Y;yA0uWvKfmYJV1f{0ce1x^-R#!W7ECbcenuaA~ubm}cvZ)(>|%FCtf z>{u@lMNIz<7_W2(WeT}j$C#_@S9EXZ1;odZ@vkl1x|EeSXqL;=2oj6YL)u#1}Zk1rA7X_j-Snul63x z-$py;u{^PaPu!NYuU?fJMwtZaa9f%H!mp(M4-F4JK3F@+cWuO}>FH}AylQt?@LnCx zM?hDuFB}UlDblC{df$!rra#tbvv44M0--EHDMzSB*HQ)$hi~4zJz1K*uJJmz;B{+Q z`;8d)5DVBNeKsPHvQMjLpTkD8_~^K}a4lRov^AFy>@buv&@#%|ea1ihzG>;*+}>Ot z0OtjyLO^5yp@$k!xV#O#fa)a%q-r$(GK(a57Y4=8@PKI3h}e0$=}Yhge4{FebcH>U z-=}#%t=eN+n_p9dJu{lH&uGRI2w5u!6netX)T>-4u%d?Y7IqvQ{@?76L1%bn}wSX6BEx1I2E8k!|SAvi@ zERVJgA8bOL8T?MW@x!MD3yln8u1^!3U+I4T{td7}2p}KK zRtv%eye}9H{ckrdMqJt!Yji6e1#%*%tv+PDt0c@|H%ey>;N{i+ZiU0}Y{rG|4Ods+ zCQ&_0CniP(DA#57CrZG<-2g_0)91?8_i12z!ddr&r8%9DkZ!{oxsWe1dz=r7OJ{aS zY%F@0kfRXhP(1m-4GNy4@UcMfXpj7xHN)cTp4vajg8ichXAtp31vK>$<5 zuy-O$lz4Qpva=7JkBcRQAYpae@)erBkt_Q4t#{XT4bZ~FVxJP4mc7V)t`2$t$rlaK z_~rE*a*dn@%+k`*vd!s3cRE14LxrGtlZZ61q6?;EGP%rA{pU{6q*bNccHDPn*^vNrvK-6qnX9*~b(vKn7xg7x zA4(W<85i@Asr`$gs;}jOwQYK~@k79Og9D6BU!NTD115x~?rU|?1`D6UEZXyDDm!0jw8QxDIZ8+=Kj522Ds>z8_V!K^UOji0X_T+H3gF=3g#e0?j)@5s>^~zL+u)aw zh(vHlqN7pYWpYSVYPo9?0)@e8>wu&}w*gh8`nXir)o-Y)fW!Cs@w%j}H8fYmhxYaB zwgHzd{ES3F+pbwg{qV^JoG4Lbn-4&QwJZ%eGUcH<(k^}p?Ma>Nm zECSJljcd~f6LiG7A+hdTx;9yvw z+nDr)V+@NWB>;hB4Is0$fRd!~(1tlX>xa5Ud$_*gs?WNWgdilGg$K#n2LMVLY@6A? z`>7S@R%}Yj_=5^tcqF7gi~27W73hFr_dKuUueCg;eCOxDWNKzMa{^PTG|x1`8g`~9 z1sH1HAt36pg#$n<2MG4Ni=&WVzuw6Qhl2d{;e&;7cOcBocz}N|AQ4Z7=@P^(6V*ET zJtDuinhipfMK1B}yB^&^jotouHQ%i6hSGOT86P1R3t>kFG(Qjy%2t4M!g)O+c(YgD z;@Gm-{M>ne`ipyXp$dcujY7K7=CEw);VPF=+fmG&TCx`Uw2xPf;_`B;WNPXy{X8 zhzXpJA-mRXE<%`=G2^BOZ`oJ-ql8`eQscWG4;_d-M_41Hq_gM*1C(Sk zo+|>4Y!;)RHI|?KVuyULj(Px4o&5Ti;oG}U!C;eiT%iZ+aK33BN_zK9Ctm70)fwQ zyXN)gnTWWxH4~(p92YtqGk_;|ri!r)N$+vJuMSY+*NyAvYTOTXeNTFjN-8UREc5)r z{U7gWSb!E40;8)LbGbij2Es&SR8$XGGgeC+46F>Z)&={6xq77I&TF=Ka=~E!Wp6wk zgSLWa41ON#?evqyUn%e?dX2U{t8@`K1SDY~6_GYAx|4zQ-2G8-TE7aqTBmcCd8>bG z3q1qw?xD1kh2}RSaw_UJ+W-I{YUU3;uRd8i-5)mf8FS>9m7&TW|Gl8_0e84>b0lN( zMIoAO0%hF6fh`cBK_x#^4<%d5H7kRa$_Gwx`c4n($7wk^aaLDX0}=mrd5DQVoJf%i zI6^@5lkoKwemW!4Avq4{!yxARfDBb??B*Lyf)zF4)97_TReZEI-{c@CAprw;q|*s0 ztey{iyscf~J8|9{+nZ(!H>UyttHSTq)J|u+o(SL&(Z%1t$Ay=0?|T3hPA#~|_yk1I zGg+7A_Dqf)MGY1cTFbm+zyrnbW=ml!>h-dR7^Uxt1n;D#uC8tjQVw(UQY-KN$sn0S z_fZX0Oiy7@l6-!8oI%fh@!|!v#)MWvO-tTrkZ#12GiasgbaP_TKR9UG8;U|K>MQWe zLNI*Bq8^GwGw)~g$pw9hrT~Ukz9(keRP%(?oWGq`S z^KAT;b(V@u!kJOX3L&!Jx7+tUat z1>GT@-_;Ic&NnM2lE^fgF%5HdGKT8U8jgbcMGkqL^xq87y|ybokZD3ar4#Nx&F?aV zVmn_rvh6NazE;>t6d!CCkbT5VLZ+c?5O{B>sT95_7~4PCpX?L9EVXC7k9@p_LKBLg z%S5m9Nr%4Dj8sA44Z~EiqSij`S;0dSeTJa-wAJq~en`$zV;DtireP1FG|(XyN!?>^ zv>xma*DjdtIUqFU|0z6Cr@0*}3P_Yf!o<cKdt7EWM>Z0S4hT9k>6Iz`2p}L-rA|x6E2M!gECDmXYGcv$8!|mnJ(+g zy}(FK%j=U&JTU1F*XL8R{RKC6R`na$fN8)S~ zVj=|*6%S4?po7uR@S}Jos8(==MT^5_vH*`uhWA$J`>{A>`!h_7KKsvQ&6qUkG?G&@ zrA-Uwla@NXQ-U90TiU$ti{&n`lZbx24e*JAPWIw(Dnp;Cy23#`DiM$&bgDS2MV8mO zdkbGKw33KsI61CITAi6xr;=t z=jC$fNxmIf4^Yj6IbMsJF?~@e7stRkc_@EvzxZP>PwQTUeff+%)sVVr!4(a9vJo$+ z#F^MMr>|Y25;I=5zr)4DRnWI3Cvo1#ey^ePBH*aBnwci_R~~;J^oT&mkoWYMa@PW8 z-!!Gz`T3kc=XU*#+mqw#MgJ~#K{ZLlMXpF`IoTAj2r2X9um89oFDPR+sHiP&FIq^QZi zdCvviE0za$lSipbQ6uv)@Ma(SL+GYE#SX?ytn+LH*#ChDfK0Q@ewXXZd5_lRUph8} zOVf~H^P?l6H3NF&z)dZTcloN=YVJk{PSJ z-mQI^XN3V8E4thybc*4d#4#xZq^pj-L_QJ`D1xj1j-r??;;mg_bPNIX_R53x`gFRA z{u}#i=jiCHMw-S({FkN_a4((DhlDKqRurYS`lYl9xQ&esYAz<+uV26ZZTI4qHni0^X~Feb zulezbzY?frNJXNTZKj`qRAa0oc-mVwQ?9C}7m~VxA4|&n#+{em+1o-Ddxl~vvzhqe z&>div<miNeOy&Dyj88?qD#Xl$T! zl(j7+(@R!Q0X==BmMW2Ztd@nj&Dbk&`n!7j?N-*#zXs;u;1G3>Pf#HBYmOTbg+t}l zozd>KhtowkN)c1+tk-Z@B7v5=|H2NmEDto*78- z@~yLoriE7xT{O%~HkunCT!0mb?<-PBIXUUWk}96!-6t#>T<;uAXij#&w%-##=RzutD0!P48kO>L?HidxF!i0Kmn}}7y*bxs z4*v)EIrI5@mpECtrYQ`UNRkfk%V2s1WK7JS;Uv1G+$7A1(J!J@;c482m?T{A42394 z5qKG8W$pYg+g{o)))4@u*K_Af!~{szCv24MdPi*ll5lDyh+GjPE^%Q8Gny{8x?P?p zUwUocGqb@HXv zrH(xQ`Q9^@{I)2Yq=sN)a}$agCU!49_ezVQU|k%Se#U^XKdUH;2*@r8P(BK9azKRY z#}a$Uw^&inMJb=8rWJ;T)|faX6ez$jVUwewWV*IyQeJNYcrYkr?qf?J5T^u7V{~3x zoZEXMhM~1s?}nJOOi~)95M_w=78_c!12JY3OZVzb?ewr7h(Jk`NyNO~_;;84NF0|x zWMdf$b=}AJeM|ujIZ?xciiQUFa{ZK&AV{INLfdYsP@^rt+mq=I;{J)3hTi96>2Zc-@|^ZNph4h<&*i%&BQWcGS1)L8cH88 zbxbxn7y&OF5L~pcLX9_X-+wB6`{vEb$dwWH2h3hT-(!}lxwX|m#VD@ zU;>k)%ysUfMgnp1;;W&*-s@9P8SItXH)>&F*xTEe8|#UX%N;mc9NsbIs4Rqy*v$5r zh~b#Zvp0X6o#@`Y9q-w&o1a!H#a;;htnmrE;A$T{H{auUsb*S-Iy!Es{KtphOinYX zoYVDGGBsjTx8pFHHKoAaYz28_UOdV$It%4MBNt$_6!{lJe^tD@^`V4K`B9X6U5M;CYZM^T+t&x)+;OEF zrpqM@AHnWqr3V=kBb^ATdTlGqU6_VOBHqj_^X`Tp_F~UXgjg0XYcV|F!Hf$~8EhG& zKR)Q&{ES9gv@8-r&+V7zVamow8AaBg+AF z0LPXsu0Itzc#i$s7vmBU{g~EkaGTmlYwMbu<>KE%oi}9U<&lB<4}maiR&;;XumKXT z;6S^oQC{QXHcQ=B)#`PN2j`}NFjPAA5g>Zqc-+0T=#vC~B#c2=O!Vg&79|BsNdu!( zqN1O0)OOryVMHevAG&CdRJ@+gy;Ea0ggHH`yNwL>h{nIcfnF8mW_o5u>E1pO7G@r>AATUvih_0s(SgeYicJ zux=pcG!4y$UD}RS7(h^qA{!eUL!^OkX=xGgIMOe?3k?k|nNxkC5Ug3QabGkXV;!!w zIrnNvq<>&&6_hCO$6K}jkKVr?oiPEkXJu7#z;+;V2ExXzzC^s8#-Co0+IrqwN|;L0 z9(5i2p9ohEx^SYyDWp#OmoZ|3;1x3P9$HwApvG#=(S?ei{l}cU)Zx=-;!$7cdyWB`2he^>hw^k2<1196(_3tZ-zJB`a z8e1Pv34V@Ra9fwzsHms}LfN+}iq}qi8?V$2TH(b-UG?y#|JpVNW~mi0t^p==l=fUmf)Q8Pz)-X*6DnP;ko3hi+1P#Rb*|}}exPb*)D8+0N;gH}&AyN5a2Oi)qy+2C; zF6Um;QkjG!5snYT*j?~OLOEm4&8Sf;Q0p^rpBj4zK)ob^)65H<)kb|lmxuBpV8!}Y zdO~LI7_TLelM?t(Dbdb$=o0==;r)iBSBlwx`PPbvbWPR26_-`(Hlj%`=n9{~wCSJp z=~HQLbqQlrQ0fQPm)BlP8UYqXP+hTDjo-OJmlbf((Mm)n%S(QL+dp!x7YX}w&_mFM~(m@18zFtl>!A;ka(0NfmHRJ3FT|7Y&2Z9rjYGzz|Qe$ID<7DITRMyb_xQV$G6$Igdu7w1p=7814^tjW{^PqNU zqTW)usJOT=;Q2{jNlDh!nf9Fp-hxu~Wda0{`?4}J2(gH8Rc%K?f^nVHcZ*GR`HUa` zn5~%t-%TXg7ga+;%9=O8Sn%jiC2!oC{Fn7BCLjxlMZD3KvWepWxo5J_>;!ZSxdcj) zi9$In*Ta7V;0{laT?VO*@du4;26GNaQp|f814ac1J!J3_fg_utj0!8-gr%h)n*mO8 zYT2q&9>_<$|hHoIc2w!ffLAAbGAP+BiZRZg4c&c~feF9=Ik5Crd z#I>R<9*}1C@Yw?)mk7E|c;j^yOgmza7gGYx8^~wp%Ete`U)Q?sa)C=$hAO45p89XA zbLcBiM6ztBTCoyH^Y51(GHbQHYIO<0ogV?qhL35wjd!bB_Jxqea*_ zq9vgr@bv)lcL)&eD!RI`O|x*_R*7j}B|;EsK-^+$dTTdg1UKXqh$vv-NE$=P+6!nz z=&Knf5*VrCCuL{Hg0d0_DzYg!?rkFZTok0Z(TR!JmjYqo=kNO5kufmR#-}f)9i(9{ z50Al$5#k`W&BQ}7D&U`>QWm^~3SI(``YI}_hKF~4mxv1G__ECdS`#X&L5&7bwhj0j zk_fm$FoX5FKl9pvM9K*+s#i#nL~_+PMrLLyd3i>VCUD3}4~h?7-8@m?lZRehxRQ%} zjscu{WUy_A`ncoe$KrBTZD~Vka0)LF;ka5qln@4VUDoyZ&NtX-WrqA(@$y)QRy4kmH|a*k$9$&S^=#R%}&shV;59nX8<;X&Z5FI6ljO7}rLODs!`DTrj% zYcvLh!|?cc#K3#BrM6Q`uFr{|K1JCtw)U8K+&DD9-;6&qf(K17v*>QLGm!m1m3^C^ zMF5YV8+=+=?k6*4Ea5!^ez}8q56HG*|1kuZd0}^nGTwHTim&h>h&L>~3J$lXL4%xH ztY$qV`~X*SaG}Koc))_)_Ge8lKhHpwwqU7XPRAAThuc*#x^lhrJ3Y`T0@#0IblLv? z{t_h4`}E51LA1ZPDG{ThqW(7c5V3iA^pX`U$f+HAeT_+XZp@)Z-dXMv+xqn>1c<6@ z-9c~yo;%c6hs#JnoJd^3hPuH4&BY#$P6FGZx?J`y8!@COtH^?>j9rTk09K0EKXtH4 z!om9XtoB8m?*3AKBs4MtQA{{F^yI|xM0k0l>ewO3RKfo!Y`ey8p3io^6Jx5lD*?#7 zJVIouLD+Ne?}5&_4!BFZMqf!}U=*8?>Wvx4@3Z168_K=$AHOydC$yg*8uAIVoY!bj z08T}L_z-Gj7%XDVgxR?=eG(9xHnA7!e_BKY-1n5x;xHyD_gf*fx zG&GFDexwFm+;H*5vYwYHO9O>AvsE{1G3+N#_s8oQM*clf4|oFZ55(!uHRRxBK%g1| z4S95J=e4#m(UjB<(#;0;{R0D~!Cl9HH7mPCfmAC|Xwq}u-uej`KWEZ}XlW&@S`)aY z|2g@A_L*4HOIX0>!-M^_mb-m7=zjPQC{KizZ^ujzTN}y<4z$)A-W>hygiB#il-zB3 z4**V)#~V4IlK-;c298DGx~R3i<=XkKJL5KM-N5IZp@GZK+uzi6t81igt^tRg#B*m4 z)&knT0#Za2FgJAXI@17Z#rt9h;3i~+WXH1&g@c6#G{C%SO`VoEZL_`vy+eHwSaL^` z`nHR$1h-zdJJY3j5aaW$QK(4|Fx(ub{m7T?nqT8b<5B=~4pJ)yG4ASq{O&vvQ13yK z0!2&c$fhLp>Nwsv#yRADJX}!G&2qYO?5|j!5S^FR+Lyyg@=a~Y&l~9#oBBnowcOA{$|lkEnTuh2@BM6-q(<`&sF5F zxPkECd%frZ;5#-hc`(pXXc-t}Mcp?ayr4za_83KEy#01l60}_-01tGU>Erpn^an92 zFW`v;R9H3tDxe+!c*8*8(mz%D*1rx@oQ5^ehm-}5ML4+jKxtQT`_xMMz@VKcMsj?s zEqpxqx0c;xIFY6VP~-wM`>sO3_0`<*ajvhgZzjig0JeuCwl$JpUqDaNVH$8%o)@)p zoEMGrju(G950*N}fJd+wsDB{LhF5pqp_2V^=pJG3ibaweW)lV-58%tbNw4u;ZR}^J zHk!_=4FA`kmhMjmfIQY-p>NP#MetL}dkjMmYNrGN7Mz+eU^xYoUH#CbP_bM?Dfaw8 zC49Yb?zx$!Wz0jWr%!(k91cy({$y@{gg~V0+4zxSPQ%fU-opx7{_?+lLj=hOkC072 zaw@EtEDiK=2~)zx3^f?SCUmy>7ny)eYcU?Ak!iW!j zj~_LP6+gU=-cO!14*lVyS>({t37ZZ(QhO#CvUJl_u6mWNRJ?~suygGXTNvJAO2nC> zpaipAZG;JI4%4RYZ}Xo>fD;0~!a>rsVEI~4azkdOV7b;z?$>8tG#MG0zjfwl0R1Fo zWJLV_{hnOF6B2<*)eG{($!>98fq`us;kqaRk$k$jM$<|9-V0<{PYKO}Mf+-A>DpXR zGt@<;Yl2a=H~MQ})=u>NM4X(p0C4c+daj|&ENYa`^qSVL0Ez%qU^MPW3v+vCS-~D7 zf7*b*-=jDtRr|*s99BIWS1QFP+VxV&PdV=N3NfMngF@rIKWkdB_=XXuj zF9Z_TiY}A~;=U&lp^O*S@RuTJ;b z3kjgBFNU=kjShVM`KCg@1sBw~2*vAAg-E;74eG|2F_Y6R7p@^k%Q0NuxgXW}y_?>TMF%&aoxOBop7TmTCpt@V~ zYlaZ{OQ%+VmwS_6yydVQBR_!+P@IpKY|#Xk%d>Kj=@$hzUCA}84O#W@&}D_McFfA! z%+x^CVlimN-z3=7?(!Gm;^KZ7qp(((P$disZ;S8zm{3?y&|Sv@>=?jL(CTV;U|6^o z)b4Y9UQIo0pB==4t<@?})=LQo9}#Liu_6J`8a8}WAZUTCFzTX2BVv^{mqA$aT*sh) z`LfVsc?lU6^-LFWk@3~5q+=cH_Nl%7pS!L?(B>GhD1hUEF|)ex7AYt!)cDf&C2J(D zpnnrZ)RJNE(%bcTl48-v9SNds%>wNbg5C$e`|H&#hb{?eY4|}!;@JW>2Vhq#X_yAi zA{3}XoBTd`di=Y$U?VM`DB*a=2{{ygtIVxyd+nJCHr8@@#IgEDZ_W|{t_Fi=XQ|^g zrT>iya9>g^2Rw^}V~~Y|{HLREuIig6`iFF+m>nqpKvfseam3x++$?KQo~xmj7M=Qa z=sNQH7TvM54mX%O(1b^82K4Fl7<2HH_TR?um)DXrrxQi=1p#57*Iu)H_9qI@=cs!P z_#2xbXg9~%L*KcukjGJM>@W0gx;&WApzrgz^Bxnb>c!{01`;Hd@!q|fUi&ft$2%C_ z8+pHw3X|r$R(%T@-J%;Bi;Ccbsl}^j3k_U(w3U?9`<-nIy=uSimk-(3oH@ZEVUh>_ zumZhQ!a>`Ho>i*937`cD(o!tI7$6vl8j~-iK}%|BfV zV*a$O{6wXlF7svxKU+ux(k2z3vt`h3bbp`zm462mG*yjX&pIcq`hkN&MO`}-;Gx*Z zdF_SA;(SVard*Z6K+C1$%R~z@39WF_9q=he{(FwS(;0Y-q(U24|;_a z17Xc7eUxyM2~cd@Y@{dyht;&8c9Iu3XD0xMtbwvu&S24@%S--?J_fXF+L}~%iqByQ zT>?9C=@kXkQtW7T&Pfgo=b&oJNSQG$vX2{;g_9hv=Ln2{%TD@IhO9q%1 z;RSTi21J zaDY=~gApqjIy3AGe zZJ8521n_Ctl8z@6642HFK0eL3%d=(?YcciEhr_N1*oZ#NxUDYk6#dyM0~}J)a9~`g z0bVq}yQ?IAk9Nt-V}_9GOK}BcSJo8)rO|j0Ni<+|qRvq2;r;DwUPqr7NcoC4%)cc z6%cUg_kwWFB!ktJWvGlZ9`f1Mpgnzhdb)envHJs$3E=mf33VjOJgQ%)m@zatEK}B? zFB7it# zrM0gO8a+U_3@}JbX=tL{;%XaVR z`S3w<6q!Rir}1Kymu--iih$P|J+$i`nCO80spkV%X_Rg1#R;G`p{<#t7Es^C%s`hQ z0Q|yBwlP5!av+TXM>=Tqb~-iHc|?lsnmH>=ZPR=x`sz$J*exwB19ePe<4^{TX7QVV zGJ*r?Q&#L{%X_GkD>Nc(9WarZ{0@|0ni|(eUonNpLuTq2Mf9S9eT6caKPcMao^SK$ z4}O`c#6!S*@{9H_VJ)jZaQ4j-qaTDEmV^ZkFdcO?6esn`1+04)JuxM6BLc2v>gs-T< z)bf?ln#ko9uri6EquBD2R|*zwvo2$o_vPY zO4aw{W!^g%7jCdOn|jN4z;Ia^{RhTJwKvw5k4&+t@(v`7sQs$8c)%P6rl(sd37r=4 zz{obdr{F`G`Qu=nd96j4F^cia$}wV22{FB^IlJwsHO5ylUqCbGrGo6#)6&y-bV)N8 zI{U%RqzI~2(C+c*{`Tr%SdeXDuytJb(>n$cKBlyv`8YIY?{r0rZOn+oZht5v#!+Hg z0u$qQ^Ll(?Q*j2UfYQnAo59CY2%P`o-r3n9zP!ENDNOiQ*$}s#8xUV!-GWHWNCID^ zrkJ8bGpp}{5up3@lEFa(ekTorUDWJLNHDfp3m|)A>{5 z6`x}``v6RcoNd#c-oSS2ewEqVYr?Y6gRk&XgGEU26xJV`x*l0OIz+xVHl7}NborF0 zUiJA&sE=3!ze?&)b70_tU5!^xu(fe8ySvzh2wzUC!M!>g6%Q$VU*>qXTg>gSwq z=zBY2wAJZz6eBwq?px_?PC)hCKXkjv9`J*|>vq&iL`qR7yx)q=w?SeM&`0CS0e5Ou zop+>1E&GcHEVu7x1c$Dp0B`Q) zhb?~2Dh0Alg%f>NEuB`^SDjQG9!Cpy3^58kO+4$(Y@m@t0QlKgH1Efo!}FFd)T%nJ z{^1xc=Tphy-5lE1SjyTzXpU$o30~qWi9NaoHzR&t7XTl*Q<+SA5E~}PvH>=E@Y=8Q z`OyYYVhaDPuz~~i+2r{0%r{4vZ>Nhz2bh_SzR)ZuDw&1cXp@`I`xRZcaZ89F@?{F! zp#lpTU;H}sLq44T4|{L;=tAx%(CoTz#beW{HSrK$5O8B*W0{^Ri(R9>FaoRt94_Mg`=QO$hn zd&&_{7PJ;+Y;tJ=WMug(;dU`{(*JOOf4wsGfodQ{oq3ndXW>R;dL#M7nG7KRtS>(t zZ*9eD>MuWEI4mqk2xDXgZO5@WudOBh*^a-RqGVZiAYQqfwjXC&?vDE|>)G-^xU%KT z^GH;3Uj55;i6@a`JUm) z&r~7Zj4`mpoSyJf4xgqwz$j1VEALeeaYh27#urqQ$J1}zudDYX4$<1`k6x_Xl;Cz8 z`(a~acRcGo-1;33ztiPrrAJos$sIttr^29dhLx1ipmN`84Ri09JO+$pQAE~0o)6UR zW@4T7_bb+(wj;4^MdgKa{W;eS==u`Mc@F3`@yOIXv$!~zFTE9j^7f*zEVkBP z9mMgx*fu_tE^`?=&H^H9`+=q2iA3An+J&V#C`^{z2}GK`fXfA@SFHl(eRGHBh0X18 zV?vh^d-)>e#$woML|BMZodt*KiZhfhNK_% zmY@9l6G>#R03^|)4JBu-`ap>1UivjR1R7XB^3?(=bpL7|p6^fMFOly|9@$DDg59%9t zxK4y;BuGvu5{Jge7u$ibCZ5O572K~ZUHz+>v^MGO8})Y zK0n_Ng#C%lEx+-I<& z*FU?ywwGP4CL~|2vN_My(JM$si9kLx%@l=EUA?*9)z$ljyolu@`c}4Xupx5NE!?7Yv6PqbVX$fFk zUP1TL`EsnAtNaeP;l2|a$S~S;bo^Rp>s^nB)v4uXPwd=}YiQQv!enC;3mpe&4qQ7) z)~M4e!)!Zvt~aeGO7)l6+?O?k>rXpPsXqcSfys8Sbi~@N7}eUHlm@ozSslQ1XR8LI zemjly=W9zjWY9rol|E67c-r^EkVvazQgQkeQF9IDz)hP z6noKQkg^7xTOuCdluig<#T--LHjg|%9F45JJgr>JoID-a)Z@~T<8CV;HE+O*qrWv8 zm7$4U&p08P&C|ZRXK7A3aXE2vUdKpXy~9e`InaE;0pRkc>|A;P23@ZI6DsE^Z}Vj~ zExq-$SU}e=FxQwZ+wG|59pHYt^kUV!J<+&&gIShaUmizBjDG*lQKPBU8_);JJ4PlZ z>*mkf=C_+sQy;)_R<%yD&wI*jh}BBQ0?)nW$7+`sM`34CqZ4 zsq<^DgiIxidMsPjiRCM0$mC~b4iPiQPz}a0dA|CDP?txYGx8tw;w5F@0OynBvTZRX zueo8J`hbVV^@#q8Y_EB_w*vy-_8S*wThk06KE4>{3I!anCx9qo9THSsI28{VyScv3 zA;zixRi2(*Qxhwa4a`XN``y1~b$N3&F*_T_-2U|2X7Tlpm06y{YsnlW(uFsl{XE07 z(rRO`Zh_GWl0eSz%7P2!zK0LxIb$^~Hr1_EW;UDunXQ&ySs9rb%AN47v@s$?7~gZ6 z#@60Gx2raiSGzXtD?D`n7K8Wmua7_Vn#k!zySK(gX}drew!ar4UH}1=b4hb)Np+X@ zUAp8ek{1G5 zmqZ3N{wm`7hK4%pzT`z4YdOV|@^X3_yA?^A*qw?2n%5Bh5zqwE-K)$C-bWkZr@GCD zqkM%zz)GBY(>lAq24iF|r(b7YZXM0)wlL4{IGax<7G^{Zgveg*OJDlOvqkUDf_e0x z4>V$FDBkQI9Aws%#`-w(%fWu#KoOab{>k3#=;So9QyTTnlO4+EzE_C*wVYpjM#f?V zG#6goT1202hGXj1hpFGuZnd|Btan+FwLsA_!wTmNIXoRdP1xTvQi9!o<>}Z}B|(+- zpr~>@zcv#EB$EAsx|mwcc1r&7q2+}GfbM^*S$Y0;tY$@)jGru5C=cWrACcYxBZ}IA znOZ$SI3eKV#No|HBwi-Vfuo!+M}GN9kgEa?X2t^uhGJjitH%wVXPv0@YnGPn6sO~y z?CfYjW9^d`Hy_bbM^;>S&Hvo4`R*N@gkGyp!{FMEssZFjHLE@n&g)wpHJ72OS6Gm= zfpt1>x{_A^CWZ6E!pgRN%LD$+iW^V1jQ_9lpMf(T5N2v>O8^OIwOqGjIU&xo23Q1x zxypfEm6(B)-jYY=;{44} zJ;<6$bqd_6xiTemw)=9F(TL`b6N$4xQo*`<1*bLZ$_M3{UI&Sn#Y*j;q>wc-HPv}d zS<^gG*C%Bq^;Vv)#Wj6Vs7selAXMC2aygQTsGw>S|D3;DR%{v<2{WfD_AS1mv`tZ& zp^Sarsh?Xj$5!uI3$q-ob-Zfhmy{$KM8 z3+2-l0O9s?zlr^5;&P|v^OolMo}^m>KPH|BiYYsE_$BX15?AC#C`SrgVuA z<8)Y8ZQ+nb3c0MW#a^8#VDVc`xizz8w4WRA@T>>v2N&Fy?P`QTK1RuP!^091oM)X? zNk5Nlmc8gnkEBz963fug5J1xn1u#107HjRv65c7`$sl^xZteam3My_r`iUHgt~he- z5J=2+j6N#cEjmr(DIZPksYg}pW0)t}8UTdJt;}R}M4m}k$seDb?Mz&G)CcY(V0DKF z8(E^60JMHyZ57b|6@>}Oa+R+z>$AjDQB{5Qxyk^4PT`<9;;N<3u@Z>dyn_ zYuR;Pdb&ICR8~NLVPbk(FkemtcTzT2Y^!!7bfaU(EiTmMPzQ(y)eN*$ephUp0&y~) z%cUyY?W*S)r+H%Lob~qbE33Tmtd&KECOVU;ts4m5B>aRfDopm4`O2j_ZPoP$=!Lxb6Yj%vo5g|_IRsuPQytvV6~f8Z0sByud8HR{R<f~~y$fc+-0Z+W16)Ca30mOv4sg~>|8|xt7 z4m8c)HD7!QYisL|i}Sl@XsQ-UHUuWdxOTWR!(PDi3?^^|Vo0)vhuk)q&sBJfk@5p( zi7eHbLGylH4-atpqYD!QgV>C0=e@)eU*$?Ra)^6;@iDo~MXd+QX1>rKXm=Pc?E~Ns zw&&Z8Ir|m&4`lqB#^q93>G|$DA>2MQ&D)UMX32D^pF-ZDjswza_uisxJ54(RZJPO= z01Z>Q%5ba*3~ovQeKpAKB0fa1-)8OHxQuM-6Y1*pJO$#r~(Yl;t~Y4nqH*_XAX z9{CHi;_w*=T=8Uc#h22n1A86lr>x{t1?Ncm5Wq%7iXzcq;*HU-Xv7E|bm|UhV{$29 z%log#hsZ)xhknR1^>;B${k~ib9tt+979F`5kD3rBt@vHnzduzt0XqvYLNmt#YA^XK za`J(&hhn;N&V;>mrS#S_r)iQj;^%ZbNsTB@J%r1xWJiYTBQWMdTqVqYJWF_QGWP3e zVUwZ5oV(cHzJdjDT;^=93sX~5X>Dz|GU;tdI@b2Q?T>63ir>f$$E}V%L=X0=5n>s_ zu&u3VFte~s%+Dv7ousEmhEI=7!t%U&@r6Xvv3Owg1bar3 z?w;87cPW%{+s}}_JE41Lk`cuQH2becQW5lY*Z@%C^CzUP@5=hdskuPmCirRoEy%Jj zR2|{^cPjA^_yFhL>0$5SK=kGYs3Bztnn#$@yY~NbyG&rL6H8FxR%TS7_|fB)XqkBv zcevxAO2W0hNrP$aD1^!#@S2GMhn*XY#L4@E^qoTw^DjXU;j z&>w8tWSThVk_<9;J~x{|xxy)%ghG0ls}9EoCuLxY3Hqn&~#j*y47YJd;JpwWr#7BO;0@yfD1ta&i2NcnwRn=7j@dQL%J7f{oEgJ&(;(OJo za578g#x8ishn3yig0I`Kzn@WB8VUdkvbTgIpbsGr7Nd8zUe*JnH^&c5BpmCB3Y5`p z2aOR628-YYi;7}^y#a)4gn((6u9o-x-V>pDR(Qv@pMr-kPQK`!;$PvmY!C`h>oTt;hnSV!3Oo2wN{Cp%*bP+uPj&8`M^MYea@{;{RgXTGb{Wb_^kGqfjV01*!N^G*Z zdMsr|@PhsKi^c;^cZWr4;aZMDdu#8s#Y`4H{UW5a+KaFKg0mSYMV>zqI;ZlZu|>Uh z24Z4j)G03*7hRWQ)|MAWhx_JdaHrN5%a5d;29Nh$m!mhDukUJZ=8DvFSrZPrs#9cj zCC3xE0Dk6 zqr-!_;aZ(kWU+HM_Nyg8|3DmJyB+xOy$lZOUy5h~o5~$`UbHb)}(TBw~C2d)Ha&Rko zFrZ@R-5+vjy4n2W!@U9|k}qW@a|{|U_bjLXbIr9B9ibE12PTEkCKlJ~?tP|{;1l)z zNWp8}8f??hU4cX4<*DVRbUS^KNhL{dyn`OxK7r*_r+qfkb>_L0GuNSphr8<>7rGc? zrw!A_VaA4mi%_ z6Rk@d_BB`dwOa^`2lN0YxmW^u|8bil>6a@p1Qp#HTych=wdT}bS)%SQ;cWtp3J59! zy6bsdA-)_rzQeyjG!I8g^B4#!cHK(B$CDBpViB&d7g}D{NJvPCW1*!y$T$rC!Z7{u zu+Qvf2lFskk0IJ$>x;SD`#JhdF0{~{kJbLpA=n2;Qi(=Pkpof`@@34FOp}v~V3Rfb z_u=Z&g#7QnVDN%`Y_WFz{d#8ybQvyhu62TlX2OU8{>9aY9|GMGg4Y6RpqJSH#R7DP z?eF{f^;+1tpoepPCwBVs@5lCMTS_lTrnK144_Rmi?)@Cb$3uU=*PltEcmJQy{QIY$ z_`sR^_ca;c|NAjuOaAYM{J&cC|62{|U6`QdeLd8i@mx>P`%z0>4~axh-QA&v95Q|O-{p%U zUJH=EffVR>D*k_bB=+eCT?!gYL-mqJZEKx@iYE&-ydodClen z{+z8{C+^-N%X^Ah3~U1?`lU3@`mf(T-x>7!gZGY#xQNJF3fN?QwOK8&3L}rrw5j64 zM&Gh_Ye55GW%|7F$x%O9>+jKy+P;tGBdPD z`6|r1wX%(b^dws|gx$!Q?kg z+GUS>yG=TOUxMp~IyuLgP7dri8fgN9G{-xV`12jFRZwIIY;CvR>_8m;^SV(muDZrX z!&_##n{Bg*OAU_c)wgHd2%l%-))YgRx<&|}QM{)S^-3#zHS>d*a9MORjp}_yL5rhW zi1ORpIj3kRjeqBYC(c~n-D3oopM9*Jmte$5wg}eAG!K^fdzf&0ubc5Al(Nmz-V~{& zTln3E!Q^NSE!;(9p}~$zw%~clKvgF_*pN z%s8Lr(pc7jRm|x`5RC2yDQ5>Dnltc}Yr(q62c$`D^ODNsf{W@=0o@qLLFfl0eK*74=Y)mg?WCf#t`Qfp# z)a+*;(3VnRE`olWV?e~)3%V>b$i}%7K!8sP@^!#YfoO??ZbI0``o_zZXji=PNA@^=otm1q3T<@G4_{GbxSUwGCw8p4; z3*k_P(H_mkk~Ii(dJssGIZfSI1kzBaV2~Ih-~~bUsu<%z8Bf}$;=HSTT_}4wolHD; zw9|H_aC{Z%UFA$2qVv8SRPoyek+h6-(BOGE5g~sMbh0vCK2Ka$Dac?ly{~=BbFA22 zk_Sp8jC3^_%JDkk_*-*cxALFHsz{hhvWvZv&o|7#6?1D!6MeclrwRz+0!WVxnUSlw ztfsXJFgH{g(g!rDT;Y71&%$aa{yeS}B1-}u*`d61qSb^mmn-EXNb=0taPLIa{?cs)!YUEMWJj&2@nAX~Qxldr4&O~i z=;h$)!8Hmx3Mq%UFQ;jl8xVZ^#TSUhyKH|}5R)DH60YicVJ*RP5oBYU7E&u+kwf~; z0IW*1GRvEKH7B#T!(-kQ5=&`JzU7kCw_5we8-?5R3!Z*V7ZWLjNtcfwgJx>^iv-Cg z3(Hf$@9g$=7MiqQPrG+P!a_Hb1asD}dXT9bOb~O~X++Jci;%#1I7m7@bgCSKYt>T~ zNTuAAq8l10h;fRqq39ix!ugalsEZVm*(a<+Td(8|f}UwZXbXhXR)=O_4$83n^sPXL zDL&dSM*+_AvwJ$}obMq;x8T8VBIR%R$viG(x+H~n?QUv)-HjuM#Mha$s-!jrv#p`H z*@&lh!>hcscQ`1A6S9|$rhRY+gvC7^rOy-tllqeZ?7kY~YnNL_68NSRjZjn)fuoZ7=BW(DD%#FA~~lR}b-5J}rv-%P%*fr~dp zEQS&Tm8RQlXe}b`<+f+P-09tuvWd;xW8yYKyO?>8%I%ST&b-)?!0v4~f)&5RO?E(O zpm)HmNKfAf5jfnV6@9a{+-+lh5)qCdRZ$#?Ye6W%-na-An1+NL{`b@*L>!eSC1#Q#|c|xi%qB@kp z<;B4{<|=_$35HS3(hYQc3oWQ9bNQByJDo<6u!0YhwR5smX9vaaOJ?=rVTXQKp zcm1I)!Y{~hIV@7W(PgN*9-l*I@qOYsX*20Y@0ji}(sYnZm~{F1yK8Er|MonWMcFnz z*+>}W8oCrnfYfn{=Z3#Z;Cu2b8)s&jS~9C_{2g+1-l;ZLQc)A`k)D{1JLtYZjx@@v zBZheZ(@4g*r9;8>pw(S1!)Xl(&Kh8PKU0#@OvZ}M6(x65IE8p{gpQ?=$!%rgnE;&zn!VD$FwkkrK1$$QW?8B9=irA1V9_S$aUar!-7(eNQBFZ4w zu#!bPxurmTEqUykKa=E{IZp#lr6i+RwVWzmYbXq|sLZsWlf_}sJ{2fcXbHVSUR1T( zz32HvSVN@}N(Fb%N~;0^p@v0Bg~LKO#9y$1D8XU9-EHxf>-c3N(*~~`ZQx$feJJc( z?cT@TbYy9=fDJe=jTm6c9U+D3VkcB-YENYomkC8IJwIakx6#CZ7Q80?>YUFrl9VCF zpN8VPu<2@thCQSsSuORe&Ikl{qhXAX3t5{~Yc6Kzt|nDoTsuL?OpusV7OMCrL8>S? zBIdov89#x`{>$w~!4w??!-JAlx;>_mssFSVBcd^Tx^qehPpQcag{Se%k$GsqY9yJ~ zL)GsC;#=auAEiUC`wnH=j5fB6=4Y0T^Goki_+#GuSj30~cABd@i`kf`l^;w)E4BTwBU(4Yb5pdp z3JQe%L@NKs^HTo>o4KS){03*F)f)hP~g_AJs$+K*h0mlvTn~ zW`Bez=6j*O^FsH>oQ@d~jYq&WZ5 zJBD-;@N7#vQlx>pij-EHuA=JLZ~z z*6V4qt-K#J98eWX-D9fT>@6}r>tvSKhK#*kRM#z2!uU{2L_(WIrbC3$&6M8$USFG( zNe^$F&&BL=WBx0Bbh#WH?XOq_fwmOCUL}zj-TeeI22xmYYZm>^za#cA)3>;<&(L+B z!!WTk@~d(snBGT!rK?=((j;n%UarHY5m`B#+pv%$q{D@l%2(c4OU6kpw`$vmwEtMi zA|u~}QoBCzuyu=qPAX{T6yXbsvRaueEa)s6^Vg}pdC&~S3(J3cu^G@zsY-EqZ=Y@r`nSzbH0Tm~_rC%nPRC`z2x zLUu^$_%;PmbG%aAP2L+XGe;^6Q1?=XBQs0GT+Ex~lgEIvuQV0XQD5R3G&Iem@Y;!O zenVU8S&iANo6ZE&+`{jx08+sd3SZe3z$YOM$wI*#nRnNsIYPzOZxF<)VMS-rb;-N+ z(Kp@;fN5*LZF-MQV(`H7{(bm7$UHUy^gH!%t9tA+P_dQjU?|qZM%x47Mk`wIW6WSf z!mJ(Y9w{{+m!qR{9KnY(eWbZTsf=FTIpu$*yrK}`KV_ZR>F+9XYHg~-&CLtac*SZc zwGpPM4sXWjpK_8brtKT=n$=W#UI$}@o}*mmv39tjAYxG^pShk#zUSFUyIAl1C)pt6 z--54-rH>3tW?@lCa7&<779SSd54d_JO>Ob!)b1y;33Y0e`cxEt679K2KHQj2n*t;UiI0ca_EHo$pX5?m z!NIZux`5B;vEhfGZ90-R9;=Z#5L<>300#!G=^ zzJ>h3r;jcMyJg*P)b!K>dD~eIVWdB5B%2pwZi|=_5)zqGHz#m+a-)(b(9{AXz z{BRaN5-2u7ZI!%Jx!L*dpWIa17)PG{L-HfxaVub~5vk9(W&g5p4>G&XyEmW~q-z@v zf$1~X!{!bd-KIIYZ&^-!(E;2CIM^Cm`d^g^BLG7PLtFf#v@O4aoWl83Vl)5BM}D(Z zg?sV$NL9Bq-F$sW>+8c64W06|uw+F6xh1Nw{nc!|vivZ#oHt+4xUa!*hb*xM{z4}a zbJf~b2AOj;4XPCz76_in$Ym2=b<&t4E=g25g|Rb^k#~b|v{_s$UJwQM<+nik3K(b7 zzYYO$OrQgS7mk=#l>B-k7m0XU*Ox*$1O46+qs;mJ_fZ+h$=%%tGT@(8s*@p`%CK-w zkZdD79Sdkt(xB;G>X!*C0+)58?coFoMJewf3^+$St@<1Y;$^oxy>L38XRxQq3~9V6 z{^z2;0yd8@!xnE|tzO&cxa5P)mxtCKR)_Ngx@i*R_RG7Em*U7k*#RhGgjRO{N=Ao+ z`3^>uro84a>?!DXq-oj~p)4K35HtTgcsEx68TW$l4xio^kTt*1`EPI7Aj^!G zjrA*KWR3j({aD3|$8UwblJTA2*x(kkBvc!>zkPBu2iMm$fOImB%raa?HL1oIC&emb zbBBzxDCj=I&#pV1xq>Q)i-LdO`(Qe=wi>T0CvJ(GR^GL)+gsmm^23w!?@Hy%_3APp zUXr5!^DPxZQ#s!D<2WNkW`R8c?Km_KrGXVWFV;~IzH6QcyYd}1Dy{eP<*}yEFaTKd$p0U#@SpDX)ig^p6D`Nn1{rR zXInqKZDVZ_#rFSK^8xZ>yKG|=($mjxsJ!C;Mo(R-U27_qH|KX2%?zave1vZgqr+#C zvZx{R>eIqooz1IPp#g*8%8sMuIZrnUwN=Qnok29cl@&4ePdpSP}9<1@S-~0 zWuV#FMfjWr-Gm?rI7mOg!#^<&3Cg^2!3Vo87a}5z41=6nsENrRvF&SS<#_!6=k^L~Mb~d^%z4$nYHTo{KM# z9}FMW6=tcX(A*dSDJ@FRgv1ZkA)|YUf+?f)mjv9G3D8MKf&{gU2Hpcba!$e(l-sQd zd+USDn~8HD>>}nSxjX^;0HKj$|K<@T{v4cluAh3^=&>oAJ{>!qvlWyUZCo+tgc%wB zOqF$C?o9u8c)L4$BPlamW^1#}eI8BR5uucO*7RQFxo?)fkxN*YGvZP(ae(-eSRVp@ zd7oh^Dtt!8vJkHl965twNa^WURJiEOlx(ds+N_LkW15T5r7Ey&=>aNhN#w$!N_jF& zP;52f3#reaW`GIYm7y*ei3}t`8yj8e?$}WRKvo!|w%ge-Epz1(L#2wyrK)wTd#@0S zfRCSy5&oaj=|MnJxD0OSNpPL+37bCJX{#XegD*rg-h6mRnw6DnSr) zBz?r9OVpxUr3H9iD~^CaSC_%vMd7G7KuL$@a6&cU(3 z+bJr$ilrdZ6h7qq21yl~Y|3^RmGVkVHnJSN)2zB4VQTP+GiICqLeFUW;=Vp=&V?4M zExI;_s&_Vr@Emga4?R+xGz+WHWjO|1nh(o0w;P9V2viaN5F)-yb2pu)&6i6rfA1Jr zW=0$t${iH^{SywuB?rmJI7u)8TPa%h{<#w01z- z{Z#hatiV7Qe^<4IICKY_P}JcgZ>)6RZe?iXsYc-85Uf#*ucBrI+fVOhmsXaFeK(Ie zWibIU?@#iEdO5g?xuVLR?d5#KEt7T18#`DoAqi~jR3S3L zf@yrn+vlA{mwL#MSdu2@8iGvibgaxTN-Bv&u!PHOvC|Y9mOB1jR3}BFgQn`DrkrrY zR=*-pOb&_OKh=e0pLg$O9oj$S9Y(}7m%H}0flZjA<(}<2^pWtHz`>;MNx1+2*d6Y% zu=g5q@;o zL_^b=PD`+^cLyBW`(H&&gQM&yMn#b^eo^+~VqlI-eDF04Hre%h8$zNSfJdk@12x53T{vBcJV>Z^gjHg8W&N>4i5n=2zA)3}KY+Vx)Wd;GW?T$~PLQ{vr7 zkFJDM$;Mow+Or{=>M3kGx{S;WQp(IFvH5~A%`9{L+W82A!bPu_wm6*xvf6pcm&pA? z&gUw;{wVeCK1v%{8iQC5LjBT7_pQgV?a)xq2+TvHq}gMsh{NAZkD?yku!3_ zpI6^-=NLn(BA@AlB_s$1bBM|waq2l9wx=pno zKYbF{{j5|Yys7vgfAN{u{=;q;@u?D0U;X@M4!>&*m(T?34 z<1MAQFlTvzvr|Y}yrhyaNLgMUav)+&_biwA$nPcX5%YaR;4IHmNZkJ>#ZjANN*R zP@Wd7Ta>OwbvB^62!&5=MPVC&JCN7*YtT8poMrFUUai#o^o2uo~2#jF0JbH4Q%+{9=^0 zz=BSvH)@6JG^Q~fLgPR7^C#}S9Z<4>J}rcq z&36jsUNF}>zlCrQmj0$#(I-^r3O;BU?FKD;XR{sGp$t0ML8;5Ng}rX@m#s>05Gbwl z?&kBRXil{22Apn{LeAlz|F*=WOMcq(bLTr|^W*hi9n{=fHw?Fa)VEH3Q5ERqsQ1(z zZ=E`xWV%HPZ4FWJa~x0{NCr0Sx!#KZVb)iNryR>0QgHu!NL>W+V*c7mR5d1NM5ZpA zINA79;Q~dwJz*i9e6|u(EM?Bcr@EO6@usSehqxHm!)7V$kbpFW{h)p@p2{R} zTJ_n34*slypc{1m@X)tqwQ)FitAar+Cy`#%8U9W4L>| z5jgl@ORvXW=Wwxg%9Nma-i&Cuhz2I=*y$jl!KsfbhE&wN#fD#8i#!4$v5E~bjp4Oh ztq`k%)o#;~ZVOqcVOuRTdB-gA7~%(6oB8K3|0~Wg7v&4zOGWB-ndyLgrI3Wj;h++w zNOJ`Zm2dRd|6&1X?aRyS0)1aj5Ko2(ZeE#pO z`q=9WLfiCYPaI3??+-pM@T*u+U(S7)8#S|S83lRNcK`bM6I*nu^z`=(4nflck)e6U zChOZX$3;3kef5$<`u|m=ETdJ#7=Jh`A z1HF^t*hU-%3CzD0xaaPYu5l~HjyP^+-x3wxgz)GhUQK@|oT3D8rb~H=Cu*;cM`*3!BehX6{%Ch|Nfq`t4#^)m>hy3!voFc+#fpSavt4+5z>tFUk z@TPJZ+M+x~>Y~QR7`P^HQU>&++mg!uol^imLa&1&!P%_Az}+-QxjczS%c9cjAYF#_ zR$9)mqqc+qMo+egA1pKp#q+mFDAeDWJt&MH?4vinA9ak)MLIni%)LOM;O9DB<_hp+*g(@52-0IEBa)*{DWlGrq+@KW|wNFfp&15l_lr!E^^XeGZ;LOPuH~Ie; z|M>k}s?^F%qY79VY(3G2+t!{&-9UM^9R>|5WO^{X2U@xRL?dlD|Tr7BdP4~a6>O{U1alU7|e+^3tqkTnKpsNgV{oSau21KR$BW*5G|rt zv>b5w<-hM#K*01uM8Ga-lSe|myPvmfP4~~*}nhJ zY9okJJWkMfz`A0K!-S-(jQXQ_p+P*A!9(WsZ%OY1JetCS&f3mu>PgnZ29x^yG<7g; zQ9nMks^lnG*B%5uw#36-)Lqc(2h6ymP(_P)1CUYlw}`i-Z~j~Ohti#jOYiUZg?KUN zX#sOMa|^i2pTX$fmasGI;CD2)1hZdjk&Hc@kc7vh>9Ds@;$nt2^b(ZR?JP8@Ejp86 z?iL?-`fcn#plrgM@RkteQnXj8Bn4jywI>^Y5T z3egERAI3Teem%b-gnpCHw|i7ug+$UVy>(y1R~8e0c!&Gch}}pewZl>k0CF1rHXF}d zew(gk4$aWL`{wm{o?)Syxy^2B$~NeWUQ@zSF{{`b5E(Oq<#z1~ zY-FKu1t{?F;es%@-k>UVSU0-|xKl5gP}*2;-86^naL83hIqc-SwPo;Qh7B0$ntzcl zt4UNcw!|~Y^&*kf2fxE8p13IDFI?y}efdM27@%%~>NL0h>Fyk| z@-=SaCuGi_!La-8o*F%Cf@fnzxtG>XN!5lT>Vdp;Zm>i2xUH>XH|*VBT9?Inx47SCkB~DnRA(6?fV(+j~ zhdu1HqYr;%V=qCbmN^UXUU@N67+%s?>?ar{!$C0WB$xK*wPI-Q*V8rkUbrMW3&jW} zj9yNRmt{%q;Q-g6@$^F_YLXBel{Cp`c>bgyQLR?X?kN>%c>L7ZX3ndm&-X9S!b6HUbk-~BY1I9?<+aiNdT$J~>DWJ8h-4M%rKPS% z*KlXd-6DfjMhxCY;bpl>@{qh2FqevnxK>V}IF~5i%9MMf!5iq0mU@d`k$l0B>HUp| zn$nKSqu3grE~143xGOwSbD2ZdFpHkmodPD4Yx~j7QWx&Q!|8OcmP%R<4kGHT;k~rp zsyUW9ZSrIHVElVg_IFNy6|~ei z(Ku&JRaa#Gt<8^?m*Ug#)C9rRTwCManUL-bZf_fg-L`)Y+GzaW1hE8?Kc|+4d*{OK z?t$<7byhi3V+o(Iil?w!|DN#Ai;uJZc=^(%p-KM0i(af=`HR>oifM}>YQQzMS^jnF z;LG);RWhqGJ_V{qr*Bd&tJjD9sD6_VMiP6^Nj7|UxQ;;e_T6@E0^3kVE03w@!0Xg9==Hl|Yq|@w zh2_Hrq>0Tw2Z`D%1Taph!9l|q)lidzN!6UytymwZ5w0CQQ9?ib*ntLKBma7d%TGGU z7P z9&b48I8+uif7l>p05K%*Xx=idUGSySOEzL1TbQ)|jc0K*REVpH)mtb33PimQjVOz( z(v7W)K6>JRf_3yVO_a~dI1n)ABomj_;m};4nk_DXf22B4SWKZ$)8*QA6&Ad9j0|D&D1o>8BR^SSZ*<^~$7XImYl(^6;Dkb_|G4Luf0yG9B`Jaek;B{EPIP}-CHk$4 zQ;!E10MlOWhYsJX!e3kt{uW4Pl=Kwy)~QM0edwVrJs!bQ=DAQ}b!L}cqa0^6p47h$ z`QX5ILIg+x zqFir8U@7n40kRY}&sHx$w)1~rM`29_#Xwxj8fJXfK^O3g-Q)8Sn_#33? zl+mz^<}JZ3MUoy=BHE&#Y!80dv~A-1Thb)xyehEN!+YQ%+}qaQCd~f|Kv7`|Ig8tY z?c}0VQKrHq#4Y`S;N=?*6{TPzb8blfR#%jOh8_9TWZ`KO^x@yVQhQWz^>hSv{wRba zf?-5xfVn1lyCMxAlJQj&5Tjr>#?fo37*KK1u!qm!2!=vnYM@%B4jXtxis)X~K5vh` z%g&7i5Dhs4ND~pDJ)5Z#vSF5y>oW?;tW(Ys4I=-*VRt3Pj9cdWh=k6ul`f z;F0*J`9+xPpV5jfY!j+Pq!VU>Zd{2~_wkTS-EB;LB$r*0Oy1PtArruV_l*+MX(h*` z9&BsGBXj}#HXJwj!kq)JMZ3Oz)_5piidk6wgX(RCFmj>LmHLf6Osf=*BD?+bAP%e^ z?CHgZn}-Ns4nU^#pmzEux+U57u&!g7($bc4aNf-5AD#CrZFhNPvFB~;m;?R^5&g)^ z`l9`w5)n_yytO_Fd}g?;bPh#a;d6uK(vKne9IJl)K&hb)%giF?iBjE%2Kc$>TXO=*a<;BKX z{GQ~nI<_enE~z{yZ~v51?*h9Xl$WWL3B1|eSbVnRT@VvFRZ8L61xXc}d{$rBRwNtJ z?LtP)oZ+D0+dYfXcG#G-jl47^J;1mbAzIbp?TH}@v8@Rx>3s9qpe!WFmk1>hxta{& zZadahBoLwPj_PJ67=S;b3?1s6|d#d zN&gAFW5k81IL}v98NZzwih|_H zgYWN0G5_(C0c=UQ3jYFF0@HNtG)zxxiqe{^Lg##U|Rt6W-kGt!W}V_t7>pO-RBT7aaKU&V?0qx!u`Sc6a`I)%41Ukj`1BeHjB`)=J^y z@NDQOo@g)We;gz5@CW0sKMYi6Vkh(MS5DQ}_*1T-omm8LzWJCD^c#Pa+Wzp-T96rL zrLCPpg*|Pa-#9VlwF;>neoW}Wg~;pui;oCKg3Q0egCn`4UeJ~F&N&2JN^8-S{9uX6 zNHV)i`{$Mf+<3(q(>lZzF1|y+xCV$Qb{yj6Hq)_#N2ZCL!!(11?Hoqggdi(0IgS+QHJM$(^5O!*g7ExG@ zE32wI!jkz1?DukPGu(HsCn-cI;pl`p;UQfJwD7Q0h58>e(xS9V+3YGS9k~)`x`-8f zH2+zF7bp6iOhGWu!)ml;|7_GaY5iVMkig6->9P%$Q7Qam(k7!%Q#cgzn3bgZDdfW2MVY0kA?4Q1nGG5fk#C55<}%-Xw>Jr$u|{FKu}H+ke7HOo58Fq<~2&ze1i z{!30OzEnFOjSLqoWU3Gx28ITUUZ58n2tpX&TlJ1u>a=@6INVkQPsDliC6HS;7>mAbA=hCu+LjO|E_l^7>;< zWJYFtbBl>T_pcM^NaoPKfMqRSEWEi|S890-D^KQm6e`GrQx1R&80+vQYqL&nxq)=rh?e~uif(xn zZ{!gpT403L8M2cH7&AO56*;;BDm4wzhiH%cSL6QB!FtSOx~lp5;0}D&=*Y-!V!47V zx$d}7sP-~pB&vI2uUU5a79;n1=i=S#e@!D$vGhAaFTI5{vOvKm9UT3>7W`;Z=D;up2*ZY%oXR2bKE z+`MNPr^p`77d2QR!s;_!{hV$OKL1+|9c+1XzN$B>s|&A-UmD7)l(DL!r9X5Oqh8Vl z_l_3^^WtLPQBaq;sZ+uLppj}koXfWQ$#LVDN4{1lBwp+}2Rt1T-+x`E zhgd}V1-IX~qDL7Se=zxr6vvJc#l8njK3QU8E>_BbwOG~|F=2KsHG&d8PBr~zH&;%% znUKHX(*^u59K2uNI=@52lCj^F-MrST_({{gu8+Fx{sAWd+m=$b@}Bm4dH>h{*}D?P z6pj{Zp1*L|tE{;k6+aKY9?|DSBV(?TZ45Y8v6u+HQE%(k+izTu;rg}g+A{9tOujIG z)G3v9lTloCA9>#MoG`aXNp4JC9F5&9i#v|aL}%`acVL|i2DJ_yUWrwlkNi*SzY&vX zEqHmLUJy%$)wx-McK5LIsUl zysR$tOf)k3C9ij8D-NPdxKBnkJ6=BRwZ3pdsSz(o& zo^>>X5keN!gx#+Zi3&%J%+zUw%&RP{?7O392{+}sk zTy@#D+QE41ysIU!cE9oETP70s5{M=?TosE#vE6_!|%vkp-gx}iBkke;Ye98weVk~ARG%MY3cCFhW*nh zo^anW@vE%kCFRyxdi&>>WFPsl*zQ*0yu!hZ-{v43#HNqX>PDGb9$| zRmD|`xM=a+7FE|2=p&fa9_<$p~;4SZI;Q!_GCZl=4*hpiDJ zeYljJYYV;57w+htkm@J#icXH{%|LCAs&kZ%ywPn3!WiP@Ugn1h><_pZb z^_kzeBqOfqR~m@#K4jnT=S{G?)FS0oz{1tc&MO&Lu9L6#Cz;Q1HKnTXQ>z{dX!9FC zV5J4EwH?#k9?gynK4{4o8NTG;CP$Pm;FjZ06-XaSr!;&^1{QQbQ@$ zi@~M)gZmU3ZxE=N<9_n9M(xe(9d~OSGkh8NaKY8AF;-%aEh&zsrO^vR(@c(dZ=@?Y z29&I?wOO<5JU=GTQ#+M*W#O-})PfZh75lgcJcRxy(;}_bX{g246cRuFDpdL0*f>=~ zGc#stMjnmtRoj`%?TOEa@5wFVX8H^-@1)|oB|>)8Ln#s-iH9yUO_6!_@9GVSJ%1dm zA2VHvK~QFU7_Xtm3~B4Qx#{j&Tg9`D{un|Ox3lN2<-=`%nrc_SPZtskn48nF{xqNc zZ$aJ!q7<^v2FJ9Fh_7VP*~=rW<)tFGEAUy8p`pN?FFwna;Pxgx#6Wa0>aE?d{;nKc zdX9u0HfWHZl}|IX>ap1Gf}*|rEv;-go!N5e4(s&Bf}~`@$AbZ3-gm8|Z`*e9Y39y( z=djV~L6gCDTGr6&_9S;~a#G0N-hSmVDC66I`s9ZXHf^_D-xvPWu>Ol+B{?!vIX~3ESu9@v;6OWNvMUJ z5XbgS5=cVnCU}a2S}W#T8>A9CSI=H@-LjbMNb@brWi295&gsl57h`_A{I0bKlJCF}4q@Fsbp+;5=TX$9xec zzbAjO^X{aomlmFmoJg*T{Yk&1c}tCT&KV0Op6+*j_vpSKe_hEey_k0R~>Kaz?xN0IoptdbwU z?nwx_$%OI!G?Dl>wd&zJ(txdcV!SB-Rg;o+3N8k+s(U5%GXZ;#{c7J2pnwC&21bDj&h?48eL7S zqZ;qKhbc*|SF0F`3I`?}Xzk|2@N3eV{PJkD*QkzP;XGD-WwZJb$qB19|8HTDQ{>n) z{f5oYQLw}A;bBN<2Pe)Q-r*8kAqE_=kcc6(DPre1Ag*S7WGx!b`^hqPr_uD#rbQ{2 z({)Dmve0+lhB=L)Wc{_}xkc)`Z9yup29tXy;xAZJCu$y`0{mZoaSJzMmKGMLg%3uo zO&s)jduXUn&wj{Fap(jSgO!EFFOJc3);9mXS^AtO#Jo1tzd8=aw-i;e%m1G8zh?oA zKM*-c`nV#fO;D{e$-c_NQpjzi5?HOZe-`K=`MCulH=B2DhQ~SK+s^a=(#Y$C(f$w(yrTD8QULwLdJ^Z@REBkI2T?b9B zK2l$|l=3D^Nq5>uNn7g`d#F)G%!+2W>T}|eHCIvzMb%879bSGsJj&vEx;XlUR?k2t z2JBS|+r#vQby^s)+V7>Pr|-Qq>HhhEul98%)D)8;iaARZKjVf7q%v+dLr|QBVsT^L z)Cv&_GyYN)6w6xDd~rkldDhoF)2YwLnL>KtA%nN$tR;ydj9e6cSdw*W;Q5*~>IZCE z&3AL$>SPvd2qEVpC9G!a2M`1bC4>cE_$pdcciD7ix>(KB`2FV z3YQv1e_>U48aJ+_GlZ))eKEHoK3TAXS3=~h8KOXfjvV3}=JVs4$T0Ixfr#rnuPj5W z>==)@oGRlSI%fkOO3MJ^P~5nyc>QtjL_*bqkOIeNd|?@*F6t1$?aW-=bcR@VQNiDzGx{~^gV=hiVZHE0j^_^ahkvN(we8rlUj)A^B%~mq7`0hEyY_!~^YE@{2LVcbE z5cfuB0)76Zfc3}d3tw}efQy2Ver9ElGo3-TEJ9s-R&By z^(*Oh5m#cLQ`iq-N6*&|E$q9dEn3~V$oT^mKM<756SpSM4=wqka|tZ)RhtmvOk%6j zx9D`tyY&0XF+9Pp4}dZ_kaz6yeL<0M?H>PxR?2b6lu1Uc2~8*Lmt;H3jyuXJ)&(`e zJ|sP!ObrY+?aYwzH43Ix;iR#dxTqR!X~U+6thEN0K>3*lamIaie0h-+x@qlvhZk0( z?3Cwf2{l-r+~Kf2Us6=2UtivXr{AitsX2M8T9XYi3u7?D&SQ`YEw+gNo)I{DFd0g6 zr=B#13f`m>QC7u`s@3itk2dV3TO4pv-=D8@KY=v_dX|(cb51_mL&hmtWPIv< z{SX%oM)di*gR>*y2;%{du~*7*@1kb&#!x9d$tH122C* z>iqA~`HL6fRcFc!HK$^=9~h?Aq@;ey>Jh&fH_IR{;G?s>zIwUMrTIWnzeNPN6K;A< z`~E7~8=*8pX@SyFIQlH+nYLcjMs3FBsKdD@r@{7|==;JSzG?Dnc+2`%v8bf0aw$Bg z-;0q)eIKGrg!)CQjE55ad>yel>@AO97pF}7bJFo{Uz^csacU*ZSpM5nmzqpK#)I?d zM;DX`u2N*jW2PY3w7OlFyt2fWca4F=CnfMMFTF;2IV7_<<6tZLs+B=~&7_RGK3l>JI^Qp|)$WM+*PbLf6!e&#|a zp|7S8G~f?Psp@z;B91H{IOV0-`D9zP>S9|3))v7>D>$uJAC=rX9s03wo|m z9=aU-yiS!_>y(_ZW?W5hjCZ2fr9N}?U?zHmeM*gwRRRI!O$2f zH`rD7)^CjE^d(bDx@E)9^Dfez>izVBQ1xYHRh>;)=!3t1G4r#a$}nw5lvT#w5qip6 zPhPZ`u*zJ2i~a6&{>N_8ykjq_W8O=|^#ZK!5_jRKs3rv9%(K*fx(e6UvHkZObGg4$qhd&2eSy#06*Hp3_pu*RdM8toJ6*GpN&Z;2_V zuPQ?AQ{JObTK6b7s`12;ePkvaC@j%YDH%lEWFFZ7Z%rMeh2IWqD=bLFG&esJ{~E-b z+uv1ThW+MNf}KLo4sSHSjvq_qneJC-5)(gpn2v74riO3ogNqFKL?RgQDK`9F{c<>& z65sN12#3FomiWr*6Or#4nBUi+`&7}r#urn7o1*CXDrdSfDm~&b&t`K5S5WB zc2}>e_K~JLuAlj!)wDZx z$(B($Cu(<$VqUQ2$j88)+cBj`g#`uKqk^FEoS3Pxv#%v?rGZ}@%$U6y@X>;2P(v(~ zO&7|B2{P{v;>`Zmcw*%01!VVRHi&}RFAtIN$(e0qB;J~c44;ktK3OnZs-ClbQ4K}E z($ZgV{;mz@i%uKv@M8Z|kOS%RmP$X<*bzAsRX_Ku_|MF_5y%3JsoAFj+?;Ah0|k|I z#oxUu0ih>(Cnln6Yr;W3Uux;zk-uww2mp z(N#6`@ydns0NBwRW;7bm)wk zb5$VuSQ#s4mw)kTo|5Yd>eFGA@{-e@LnCb*#R^dJ;a1>^mGdOIMcjX9SWo@J`o z>kWg&qf-6ZXl|FjL?w$s*GG1nlJ`TE+lAoawT$_~L{#Yci*luqIIr@)*Gy_(nw90n zx)v&mUqz@8`>iM=k5b$(SKvIu)GFdv^*i(1p6F+62Ird6fh{>vj}yqJ2lZ-Y1tGta z`nA44<9YL%RSYRK0!4IBjgCCGiqfyNu^2D6F65E6Un0M#E*}dy>`{O`E0ZWIJY~Wr z^h*5SDl!91gnR7z6|U$n+?t^fZ|+{4$r1PcLO!@!t;gB9!u4GGf7RxT@|vHKm({`l zYlyAeQe`(3SLBtAskvo5MN>UD#e`-eQwI+6pv{fO#2t^jxkF3r!u-R(!MzSRrg1y_ z7xgQGr^3}PYLECdyd&;&!R@iJ|2(k)T_yDMILA-1M`MrX94mU(=^r#5OjD%>XnU<2 z1e9toc$FQ@%VHl~m3Qa$aLNYKl#O>?G&Ju%U_1;XyHP3Ptl!8n$ew(vz&&RGmpeAI zL0k7FdWhzSiftdhh2&&OrVTlGw|o2bL&*OwU!bVVx&v*ySD3Z6I9Hm{3{6V3}f z;Ql?Lc=5(1^}pYKQI7aurSjs<^ymMUEdp_2E znBmmP)Z7L`ATYJ2$yQuC*)c|r=Tcpu^KK$|wBagILM)%*T%#qg|Hy;GLCEvNozX~? z))C6i5N7i$fk5nuq%fUwhcUI%FqOTn75b{iYb)LpFcOnyD9#nd%!Om;Q$GIXwR;6<0=~d*LY$zAiYE z?g<$-X64Ss*Nr+$VMgxk4{V)@m8;7y6UB=0wIVHH&s}KqQS->Pyyq@k?}^!Kh#AmV zpUGCCjT+ARBe43MJr%R%&_;uXBp=1cOMx1J)2;NxQ9uU$*u8b>u7n$X^Smk(rs^5-j4#xEt6i_=yjQ`SjfZr^OyuN_QFb@p0wrZ4VaUvzy1bu)?}|3$p_n%2!Oe_o$ILLx4I2@Z+G6Qsi~jNVM$gMDDhkV>H=xl&R%-Vq%a2czBW}(oq+#(m?I=4$P+6>b}>upHf)i?N2>X6g}%+RV7~4RB*jxNI*jc%V1t=< zjFsIT-eFGUN^duv4MSFaUm{9Iu!gStzDtn=!{vKZ9GCk5MZMLk4EM^-^ZGGB-DNi$ zjFf)@w<|*(i^uKs5g?0r?+$BAnSZ+coF-tVRjUf*TfBVkU&QqTK#oXP7%<`eMYg&7 z0gr^^unxk0P&E-{zBJ)Y^0{7NUHa-VNUpr|8VSO8CEg4X<%t!=h=GHSMTR0apva?N zjJi?Xox!Z5!cSVEfBvlH=cejFxfQJsmsu0L8w2Ow2`{YxJ(meETv{GdEHlUYzRvXr zZxeTI_W{t%FV~~`faaERu1sD;5m-j&zxZH#V8EPnf%3NV-P&?6MiUAsV$|#X8S4^OLNL*H zgbk)zV{+dgYjWJ}SLPQKh#gig`d1-fF|)Bk;l4SNNz;d0kO;oii4ksMwoS3NT0|K! zM%PQTM~=HRv+60Om|$|E9&G?@gYq&4fnBXUUe=oNnw)7EpHUGXc0hL*80aPdDjEPT zo>4JZHMSBnjD)^i)77vmbj~&aH)uYd;v%9EOw!*86(v`5&V~>oO{@JyoV4z^n5o>{ z8N)(G0PzV5{e+?^aMi6~lguiR7LMw8QnxRzcT|#XFkb31T(x|6(`HZYXtZn{0OX6F z^{x;ONGQ`RVAMcKDu98)m7C%~fJ@ut^)8$H%9O=N6Vj}BFjM+AiU4^AhYo3=g_N{iAj$`+J>W@h)RwK<@7~n7w_q z-Iu`F*SRQ$m8uSN`nWj~2?Crc6fd5Ex!}$lks@`gK7HzTC&KFXn~Hjsb0l=k3VmnX z6>|krS0Ps?^dlz(Af7=fOEcd+i35miUOXpudIiJu#2sCjtB%&TYnV#E@?`oV8{74; z`~%yP9p#gp4s5RP zF|1iJZ)3&EZ}Nt2mX16Ej_#mnt2Ioc<8f`sym)|j8iZYx#GXEWZ&I9OWy{5ay&>8} zJTe5Xrys6Mvc9e)DNCR$p6o>fk?OfPFCcl)k$T=UvaQ%*en@tDrt3&D;MjWg9gGSe zK5RT1hH_E@jyz}Sds-VqRA3Ry-Y;-2dw~0#vBw9S7XRG^f5EI74GIa^)Zv($)|l@Q z4=_eCSN)c7Wo9h0F#O9B3b!cPEp#yGzWNgh6N=Ql`E%v@+1{Q-XDDjOAiuQqL#&i6TUq!)%}Tg9Lw$_%+4cz-%8Fp; zL|?!?!yX>JuL%HfI_0w#ErChF!}_TU4m>Wj1v;9coG6?)QF(4DVxk}MQ~(u$1MCGV z^p*i-h7EeF_=YKeMuwm8LDWSZ0Ykw_-FdP zx~DV13Gn;cwAX+{6Egrro~mieAZ%zv*!4XLn;CJ>Zx!109Q9Df7_sb)J2GC4bt!(+ z4riZmT!SOK8KVgK6$+_ey^6}3U4UBCf2g^VvZDS-csb`vYSDb5wB^aF`Q^{dE46C# z9tBQ62nR?FmrItKQ2tM<^CiB9`-NgUc$@5wH+=7hJ#ofEpLI&s1G4${Zr!Dz8(%R? zadEM^Z7$9GOv^;6gHR_8ZIN)udwjcWfuOs?#t1-3h&!lRC>gjDU$EO0t>sB^WhpQ% zK)UzRF^!CTCsJ{3}wBY{{*PiR&KD?)6KMQG8a9i?=# z{^S}727X(g%Zi+gC3D=LM0o0fDfp8!GnF7Ze6m=qEBP=&06amju(FD|`WR5j04ea7 zRWCduI>Btd_j&hT=|5Pg{qmlQg_#zy{8y~!)zbi20%gvjbMAa5aoKqz1-M-#PxnHu zC4{UVg(d~F7L5mhqvP7UY16mZx1HvmiqB+FW;FH|hILj8?(%8*3a@cOsxRKd!a6+3 zq9F*-+Ftiq|6(;Ku-zK`qtd~4bwlx-*zVQT)D*ZVI}bQkR8EL9B?=C$R$B9w!DQUG z8gYoMSY)5xtVv-pbR0Z%_&NZD9Y&9QE#l>k zu!+ZP^y-^=H8!L>EVHHt$CNI7`#7TK@ejaBbQ*)X^L(UF9{{^34NA34GOxjh;H%d)rOAo$mBSZi78dKXA-1!An$q_P9GZ9YGqWH@ zHoCYt2yV}LueeqPUjo3L(oD~SUDT!8VJmHk2F z1|lH&A-eMB;mDKp^3`AT*m7oI>_80I?qG?YFrV&PVL}Pn z&?UZ6?Xo}UHXhXI-Vr;j0x;ZH7wz{iZwJILV|PT9jjOFcKLXS>_yocq#{8RJyAzg> z-ggljP!Qm3nKpRcJ#V&tXPNT04nKJn!hM*h(G-3x`|p|Pn;ySuyR zF8GqAmzPH-!m{)c<&(Xj#0j2ltI+!MtiNT%AjzEK?5A+V?jaXrtmw(&*@!$fc79mw2H>W>%LnD~wiq$(bVIlSPtAJ< zLh~v{L4VZ9m~9p>AaREa0S^MaAqFy4zbp{eT!5XkM+ zefZ0-(cna-tZt0}k^+(4DnEOwUaXNsLHLUo(#5wUPq@w&qZdjP(x7qHauF4Y}n5pMu}3DR=+*jYWJJ^(a* zM=j2*l;9!Qe__+5={u64gcJ8PvKL_Sbt)y+h0GNw*DV$ZJ9A zD-PG%Fj!~80F4B8^Q9FCO`i+`7(tjMT@u%qTXcE|yFXQK-8d#C83t9X70aw|!9q9uw zTsA;ZQ+|)u4+xc-kGHrM4Z9RI_hRhZM@zLUn*g7(4v?>Cz?TUx&@R3oN<-Iw2NUiZ6M$Ff`&k8zd)hiURE5LTu?#n;u;)9_^hK4!oM7)|}+Q<*IKoHoaipAuB@M1u!4Km<-Y1)xHps<7ViBf;mZ6W_{C% z023#_q<9O!dsYF?aSI@V!24SLEH3nqU0wGFmYxtGIkACmbiC$Vva_RAulpIsHuuZl z-{0Y^i740n2gn@M$yTW&Oje`m8@wkg49%1@h5&_jzrFh;LyMJPT)YV?H zGO;raR3Dy?XKA5x+7b)*^4Y*g0UGEzG~Rtkv-`xCbkFp0txwT(DL=;X6?`A>Tjvwg zoDlkkKwyMcXr*?ExoI8*UQlvsam_sS9BE12DY91<6IrY%qr-MN$WTzO^iN?FTt++J z-=c=&ks?HWahgc-zQjG+l|yX0pxtvDe@F7)vjCepNfzr9(tMIW&2GHXoGCd9l-9s` zPyZH{+Cb3QYyhwI{dQIV=hDpS>FKghzx_48U(%#HZyD3qtqS;Cd{hv#4Pl^r9G+4O zIE$-V2U%`wkZB~VK0CpTLL98gG^7%b$AuYJq6LO+Gs<+h`~iKT@&R4DgM+?F5o~XM zZdyOIi2y+w>|4qv*UAmL=gojd6UbD3uH$MGlZ!&Y z3`yP3F`%5FJv7w_nvyynG}c+N!7k+aC$!kJE?=8AD5VHwJ81K!uX>9eu=aS`0};6;5k9z|Xh= zyG}k=N`E3U>6v-Ymu$IwA@lEarOl@V6FHKQWZPlU`GK&ou!BpaKj0F<=+tixc9sO}uY4S4#R#O;_r6_n2adhK%*Kj(2r41@Pk>+@;y zNsv?J5UaQcE?ZI6^nnYuFPRqT^UJPeN#gz+rni;JCUr1wnU!snv^R_3Idrv>pxNMVouI!KV(srNI{ zZl@<7V#LT<7p(+Ym+a&}B}g=yCrcSZw|uI4s_$*%QqXqj{Y9zgrVk2(R;lHF7g%NS zdKIbZq98^3Zj>>P`Gb$ODoZSEY?;+tez+Q|z!ukzJ3KqHM5FPVr;JYPxnJdk;hqT} zw%l1jIqzFTV63edC+mtEAa|vRgG2r98pYWnQiYWRz1wgTQ&T{t)B}0uex&GWZ9Rm3 ztIhhk+#qgrr2l3RZf(n217Y2x{~pKYhCI8;xa=(1{^+SY)dz-_x;ondaaZ$veTzE2 zSaE%g`=b!hilJ{>eRjEUSY(Z4c)RWFPx_z^Zei6@*59Fh-Q$G!zSQ#}Bzqw4 zAABF+(|)pA{R|*b`F6peJ(tX3>jb{%1PNm^GB(x*9jH)Zt~l>lRSj88tsY_rJy5lX zZozt_LffH*r7ekoS{MSQ4oiwy^xuzceW1FHC4MbNWnyCTq8}C@OF6g>>)rwSmjRu2 zM_ai9wJcXU>pd3@^_IBEd4Hrx4gC!Wc zzd6INzxkT#b}AZvy%Sg_5^w z9s5G)=iOr5D#>M`n~eqFo0L#ubdp&U5<~0#r_!cVWL*cmXlB}VOoN{4QR=7g@R5b+ zJ+a^x!{W`qg(U_%h=R_b#DWF{5N$)3O_*8B`fjv9(@cVM3vb!;$^CNK)zV?v*wAfd z8`^URMsB1muyExb+$HdfmBj54*NQ$~w?Bpo&_sV>5Qb_^1H*+CyG!=U&@TBizM`&=7t{GLi#!ku`i7miIIWI0QL|xyNyumnf1bTVVc{>H#nqLG`Evs^kyIHP0W->vgObZrN02twr zuLT}2f&JD*Tl|)R(Yg*?OLjf1C$imVROi}Mlj4F@Wr?lpdU}TTJU-&hNYqm#h~PHv z`djIk_Hf8hyg+HBEVi<0v(s>^Q=kR~R05DAP;UV%2LX|*cS^owtAM8TS%YSb{r+tE zqDRj`6KKtV%FnK$xxPFB3j1c%?Un%$`b<}zLA+>qh|bCcM2Qi2gH`f6~~1kL*Aq0GbcbCjb6uUIf683{93PkhLZiwRJB46DPG+M36xH{ z;XMA|Eb9FLS&2iH)W$iaVZ{wH1l70fz#HnX*F&PmFT)|dZ^v}9DYK8ckDe+Wm)B$` zcj2BS7NX2JMPGQodhv#gO!CF67m;s++t+^E1fq?tP&t%)A#`yyQs!-^AYil8Q0wqx zl*_^+@MryaP3DgRJI*%8%hk$9!9;~943J7`+b4|g8<{G(j;^>YJ%<>-Iy8PS{ZO%R z1A4gq^VOyafL=+=X-CG(n>y>nqw@(g^nL!20q3qzbwyh(zOVrR@ZIl5dtZKbc6Yy) z6Ld0dHbAzUFmx8wWxzrnqsT^sg0Mk#wE6h_(6IOC8@q#fY|+T@eJu>TcVtB9nX}j? z1*Wgyy^$~}j*mNDo`B}WuT4sv9!C3GzE{Uds+yoQ+St3$JOe+Rs5F$mz1$0-3>Nw# zeA-QDxq9NH0#?*~Vm>i|ipS~Kf<jXH2;x_lWZY9TKV`rWmeetro}uw zcpVy-u*8IA6LDqAd39c6X9pCfNPVcFWdiWlZqWFJ^+mKAVwpJ|_k0=v&4nSZsYUK4 z?GAr{qbQ!MP*OUdQ}O-`Ge=Qrw=3sxINz5in<13Snn#PtT&|#(?+w z7*50WfW^Hg>TS8}l|}p7N6>|XlP%3Wn{s z+_VC{sPS@n5vZ(Tb*uE|u-)?NeJQt;C|{J+AE&*enz{->(G6ph@MiqW@j)uNSJg1J z3!L5bwD8%%>lqsE$!;_x%C%eVjJ|>APY0)U=@zWNQco?iya`UTXZWZ1 z7STAG*Mn*Bm<&+giG=C3J7fN0azx*2eVn&@&m7a}&-b(!2NZj|M9kav^ z$&pg;Rd4xSD*U=eeruZ~6coTegJmNAgf}^{Lh8ED*gr5Zl!w&j4flMu$`}*c>Ol%Z z0x|X%qJpt8mFwj`qoyel?q9jZ$~!HXk1&Y;lH4RGY!vm0upprd1bpW#>CaN|_bq8#G&qITJa z#`?q>gh^t9cYoTO{fabOQx_K(zxZnuH{x^|WTvNo<}>i+nK8Pi5WU^!6=aALdU(<( zZP_OL6AbE)#Lz)ha9sAq#f1|jOPPnf3WK*k^zk83p7#E}nOe1pu$UNfety25!=Z$n z&1>E?+<*>I9i6u%5!XwFoCeyO$hB30*CU*R}%$oxAebzB zwP`%Bdb7psO;X761~`+mU}h`Lms@_7Ntp3_^6tSy;P@FD;5%HnOcnc3bv+r|bm4iXgEGU;xK!Z07G`PbRHH+SUq8<;0A)!*7*gnHRMZEZG zx5T2dR2I&M3OD$xdiMNr3A8JPgoStgF6}c$IqO-TqhnP9nQ3@cLC-xb zMmU%0SwM$3(|Ayhv4XAd9PK|re5Ca{ z836?c1}vt8jm;<184pol3(8d}a}=pQ!utBz9_j&TjZ5K_fd*7x^_{cx z)m@u4u=R@=s7WGvm@h+T@J0|sqki$ab3Waz zc#wcn1(=JP3e&r}?n+8>azD_e`FM2WJm{`NZ%dNa(fatnaM*mLfrv)*qhn3)aIs#Z zK`Uzc*Fc6adwE)neF}5O+Q&aKG7$wvE{A1c(G+dog1QYV*#Jiw4$9rW-(>JyJE&dR ziu-1vy>S}(o}9eg{m!A~>Cx5yCFyW5HqGF3f7*eI<+_H4sjqy&p`pxZP5xN8xO;qo z4RW3!jM@A_z_kL$*k@@^_qlV4JS=%OW5T%Qw%bw$e4I&&h02k!vBLVov{V{n2Q86V%XGiG?&=KJzz zTAK9UbTNnB^((^Gxn2-?oWZUR-qE~xR$a{!mi!!1F*f!`hqi4cE<9}MOi1puuEV~9 zV(f0Ml;Ok7UIK?LMQO{eNhQ+|fllSz3qOG$d%WL-gnqCDUa@!P>7L!Vg>6w2=o#qS zY{n|v+!5EFbz(|e0qOeDdR>V3>E5;k{54n(DIpk3372UG4PWEo^|{&!=%UZ-@l_Jl1v%JCifCMF8&e5)$HVZypbkxTJ`=?-HB21 z0%Tq%pmy@|ipco*(m*VLNgI*Y(|f;W^((j^`H^+aQr-2Tx^=+OZkuCDW@O< z(4l~--2qOQ+S9$^hK3hjav#HhBn_Swcr!UUc?_Bu z!a5Kg{!x)bFs5oUVtB0WJ0z#3W?G@`L4(gQ*%@)$&4J#s`2r_XB!X3b1C?2^1N&gT zB*D=doIltCjRAHdAcGP-=i>ix7XqG_KzyjU`;DkL;Ov+@A>+y9fsPg4Sm=E$P24(e8Y5O0!@z_az;kRjSW=RP3Z|{P_I!k(Bw80 z1S|CgDJUvn=!T4;)ikug0(VU33mn<*u&7N!Ovo`tC1U@XFJHQzpZh3WYq3-ZG8zGo zYiOj_m-9vWZ{md;Lhgg%lrkC`QqIEl&b*oM-Rm>&hMH`*mU?dAdRU^qtiTWkU9xtI zN+zdcEgfB*NIj&|o{$VmyZ#*^J3D63DnCD5O3IoRJwF1kqpgep^)+{XKruc({tFn- z!`)TS_E}_fw4j6pszzzN3}q*X&t~Y)S&@}!sHtTXW5F{3k-uUBX~h&2m7(Hh9kPy8 z*zhSWHC4*Qgu1%A+U{_1t|-a)S~V21wYfR=rWpMRsaxYS2`HDgIX5ec%8o$=6=>@^-QIgfnyCij9p0!eN4{ z5>uN+0vsUB&CTgeqe;h`e~K*mI-grT!_`%7EC}3O|!oZ^x#n@;Bo9o@vE zza}R=AMjqh4SGE_<$`kM@oIMY3{o0)mjn=pB9z^KRNg@LB@+c8e z@x0!P#l*))CnkznSd?(Q7KjnLz1$6%*~@L+g%{?PR!TKYaKUQj51APQZZbx+M$f;ygXxhKh8R?1a0qNX(Rc#}JO3 zpI39a*cXnEZWwJ;wiFW+yS@(@p`xXI&&C!{XxW6J8V611TwH|wuV!04`I=oW-nzEI zz{3L#2Qp@bBt?-w^cNTyUV;u{8!6e_w?07UynXv_uqbr7*%jij2N}54zj{|RJyOoB z@owOgLoPAM+Wc65O(n4r;s=Dm4Nggv`C<4xt|Admu%WiRqpGIk@GRN&zZH*IkTbTI zH(2 zx^sb)fOLs;zhmy_zK`Sm_WT3S`mlBHO{_KNbzNhO^Zbo5X;b4UPuq8k5DO&DerEROa@)pQE}%#J^+Mq4_WtFH5@`d^IxYt(u1q_e-w+AG05!j@9n z?l2+tk3wg2)L}LHSx0!wmSXd4pZtyi$@Xni2Ujw9wQ;t26H1OQRTUq8 z)hj(|YvS2w|Np(tU+T3c!wUM}S4#>}*H~HcVMMVl&HeLsKD1Q(4|vT*$~1HQYTP{| zBa1CLxK>Qc7@jek_3PRbJu0QQi`QcZtV~Qyw%fkUDqF*6?Zg%IPP4IMQ_Rt9Zw`Dd z@rj43@WSto?PLBKE3ZeY%5PP4->=rh=Mso~;U5bhNi`4*)P2s!+4^IyN={pQUDfBT zF~l~bo_(1xee7nx6hTf;uUu)d5HZ)mj!05!^-@F7%v#%Kx!~Ysc6NV4%3i~VSbV9y zf>}8UIgy{NQ(x`rN3@bBGPt=^2ssJOzO4_vnuuOo0QwVqm+oCDyPyGiZtmQ{2?YQE z0pd$V!!>qq-l7;u4=pb1+<*5JzwP@!i!LM+cp;L66?`e(Q2llKYxLR`spz{8rT?Ud zbI_N$FsWwU(5@iG3ZIFEEt$>?3V0YuH_1|5fhr~O)Fx)1${@_Fxbgs7dAzq4MvI#{ z1eft)(^%hAzawq%`EN5S{#e^1p`nS)%{}y$2u`C-d-nR9)Q^;uoZzF#NdMc}dGqESow#_?W9FALTUl@fzQm+kcNa~Q zTGl0$yZCxK7=KsNy*`%s-_AG`nWgj@?ReOP)PJNpE{RyVUzLirAeEhkv>%I>^w_CQ z%H?1&Yg1RL)7XXgR^)C?3h(-_xdZw4SYp(Ka6F?AgyO|DoZYQQ_8Kv;#Wksk@<5H3Dnzn2nlE^M)2@!54y5$m!v z9FpnGFouuRTwhLmBtyhfscT~(U&icRGI2qb#>F|+;Tn}I{|LYRbY1-Esqd-3WJyWM z`b1Tvtq1=-w+$3y$MERrT(Kz5l?M;yPf(qiS34dV-s$ zi|;J%XQnU_;Sdp1G}Ug^b(Px;c_R8YxD;?N#_R>4gfC^k_;Y5f8tcC*!>!4~Zn8#v#prGy_AZ`iMmiRfG%1CmH0QBBDj$O2pwwzyhUE(Z0dr{*+Wgzm zr7BMod2ZleU1Q*ObG9D6KH3pZi@WIeJiDEIC9{;>tZcj+J1lCD6z}EkjzIQP ze)F-^>IzdQduwmbx7g<9=3o4uV&Pl~dHH}&^E!MG4Ms+NOqH352TjK((~ogUv4W&% z18H#%&;02#=!05$^aHqS*u&FbGb#;_4nwWb?U!YegiCqo=a1o%!t-ZpOx{BOxt5H> zW}drjoSO7S-}{}%%;qduWRlmK$7vO#^=q@3@zNgSB=&?t)k~=b7Yzfl?oX}6=^Q?g zd4W)uaFV5dJ>AA2<9Ixf^Vuq;WrQ~QP|)a$sVgDUTG}l)H!yd1$Zf_#oow6d+UB{@ zUn#Y2v-AfZDG%@!sEeno`;7IERUDd;;IH;Ql_q*2M1_&5sTR%XE5QjvVHHqN2y)#a zC{>Zu+;jelql=YH8M!i9A-ZAdK=W)g&)cqEh#|>3H*>2eN_9sBKWXGWg?=F~4ds|c zC z&Z4+vWXN4?b?q1qqZ%%gA6OzDoA#rI2ayyReXnhbMXrGw5Cef|BX0Wbd^%;6C#JFyZ-c*cMw&+b?z`?@I z&PBnk`10swFr9J0nVZDft|s{9PSxq;fa5?YXjrKz1Hoah^~uNBSm1`@{+`!v?EoP# zzwN1@sAzI}xQ*XU@tPU6xYjsP9xv-y*R(lRON#Hp(Hwqd_{lGFiAD;(#O;9d zgfJ9%*RHMfr(H%+9$*s&2_6OnD(EtvgzxYCvMWN0dUy{C3JT8#kCu3?72bIjsB^1l zOUJ~d6qy4V4kbMUieB7j{|_(-nVL(gHS~{{0zVzBi~Xu}=z32l;|VuMuhCx|`I6Av z&uNKp^!P`}?!RM3WrnCgDB-6j<1lOwCAj_9yB_W^l1?9P&)*h~1mE81p}IL#g6#UG zFFmHcuwkM|clRG~A~T*uKKB%(Hk~uY82e_z;8=Hb^-AqEBH1Y9+MXXYLWXTS>L8q? z`hf1ttJHhX_R;lNq6L&Gc)y=>sVA@eFAETGe)P?d!)1n(g=Mw^ZEUu?JfIN$v(9Tr z=*4dgv$l5@li!?b=KD=#2&j#0b6ZLltPebzxry+Wz3og)=ny++WYgByn8rb4;&0Us zQ-21!1I(rjW<20EyHg}nG}XjodXq$Vz>wbovBb1$Vr3u$`@=moCJv5XSHA4rPJox% z!I88OS%>C|=Ha|7%+xD*d9AQ2!ur2!OE&{1)WekK}4k zf}Y0ywDR*^rC1^yEQHH~iW9rIG~EBnV7I8%v|m2wTx4+Hk{N)m=9ZQ`|H`9MN3tYf zW#_m5$FI1JXhRfyA0PjwYheyd4HMWk@P*OwF>h$`)>+1yL4%NnIv2cvwDi*GeVdK( zdLKc~#(zAJG?@QskFrkKlt!2^&;4;Lg-4Il!SkmQdJ*W=*OyUT{Qk4F@M7~~Goi#E zSV!+Z-D8 z`6E7GN~sA&fMi5&F0;U^BvJQL-y`SNNE)?AiKIVl{k$U>A8G06IB!&-dEn%L!rK~M zcn9k8y5m(g`>C23z=T%^r1a+`FU|?KIQ>#e2YJZY^`ziXaIZ}QWA-ijX+Tn;QO#hr zk`<%AX*2rvw|}`g>ysrPKB$V|%Bt(tTk>DK#m>mUAR~DA!~0j;?vpHDpM8e*(g!E| z8@N|Pki`kgUSOe7ZS?1lZ<&$7H`@nn=kL(-3kVzntPx*EV0Cr%*oHgirKv85j&_>J zP<3yHQr%rY;;n`g4)}k&Le*jV?T#K}xL0?(1AQ*MI1i>hI8pgTNKL2#=Rx{x)DQvW=7K|1^%k8%D7BvF?Qn4`I-XVp%Sr5~%}k207}XYiNpvO;K#8n>SytBp6hOe?agL za%V6UGT#cHG3Xxrs&@s3^Z+x;)G|TuJT?E@$`0`Af+OB6H0|HKeT6^MlnjS!uh7I zk=xB$k1eJO2Ug^=0$t2*c=(k5Gcla`IFuh_Vz~EmbEqU-KY_pgy8dK;u))s<6&|h+ z9PyJ%dlNWUEX}tNY7@j)2=39KL3?=#Blnq2!ni4WNNyJ8K3gn$c$`;M?b`WueX{z) z*eAWY`T35%z9mrUtw8GdgWoxDnfm+f@af-Rv8Se|2Uoi+&Fvh)L5hDzWZ3@QFxTy# zxh1C`vitrl?paCq6RWuFM%j<|w%VEr#g)OG2xfe&VfX_qZVv#DrQ6=!SsBY=K&69y zfXMMkA4!~Y0~VHex47T2>olClpd!jl7mGEVAKS%p_@w}{u7B`$_!T_o&2k6A0e_wI62Gohq1s81Lbn+D0AbGYX_y8+finFNTR8q7>YOdNz>vRLSh zLGW-(mH4Pwi$88p%EdxUfuK_>JDpz;JNo(IEd_6@avoB!eFG{MAwE7qD-FiapFjVe z9n6$HpAta=bhM^x$Z4%g;!z{pc`qEipg?e>Gb2PU2D0JN&y6lIv$D?($kTw@@+csi zjRb+ni|UT?E>7Dd(5f#5ms>hn-OXZ&SC1v z>7UIg%nRZ7jJW(?(wXr%>vu^w>kcsmRchbZ71p-y{wp%BPXK?18!k7(EHTQe;kwt_ z>D7Dl11UxBcd~co88nL-`fj%%VZxu9<#=ZbxHB2?`|iI3u)iIz<{>&YE-tP)VR!3_E6MHKU7$j)po?gb z66<xL#G4J|k{WO072>%1fp4I;P_*)85IkiZ#*)_MPE+8|y^ziE!(A@@AHG z_x&7M5a1c2^v}CLoxlV|Fj;$41lXzlwT_Pcsdf3EfiU1*OV(#(WJH_=T7??AC)!3W zU@-Op6#8sc1jD!py)I$9-BsLBozesZkCPCU*v>9*jk94~$YhO*9hf#Mpp$Na1C+`s zl+vfv=zsPJmRZ`fc0;jS#-4QkjgfHpV1oMx=%kc;Y}6i=PkiyT5G5NRB$VCVq;Mz)g**$ z-pX$NVvn5q+Yi525Ln+U-FwZxeP42 zxq`h101>^Px9*=EH? zIl)!J=U~HPsW%A&s(yMD{Q$Q&%7B@Oo`dckvSI*pB-7T;$tj!#wcUKVUknbnAVx)S znt$hPzmHD%(RjYci6#9EvZ6q06+$dbX4DVygGm{~2UJk`@}pvv4a)h_nG^W^5(S={ zBHfRk^8|122v+(pQoLBPh;!cmR^|nfFeg7BD>XCosyZp~E0hohYHsL`kn_0WFv!0{ zEZkPjHEDHH%uDsPrjYbtpp4QWZ57)a*8B{i8{?1TOfL`w)j;h~?8I>g7K&o*c5Sx7 z8)>fGn9wg@cH40TQ|IkFib?;84G?>nQG?d?ZZO9!_cZ`~3lIvlh zjOQ0!9(lH_pWy`ztn$ucoVShg@lovj=Er($ncD`JXp%9w!TI`~vwRia5&L!h#M(584vsXN>+r z%O1no>b`%QG?;*%fOjJUjs%a08xMm=sfs1!+&IDCpI#j65=l#OO@xIfC_o%u885pv z<9|{x9ivVH2_d*7H*V8}XWDxEJu|{{Y9<&(B~vvjlsitd-D~kT^-MiVg3TiFugYnO}UN>+422LN-)r%{%va2+h9xbVg1RuzSs#kOK5YW`i z%j?V6uTjwn!Hmmp`L_k1!!6^r>?ZH*?0g*=S>(tOvia~f^g7z13tEg8T4!IR=F-)8 zY^iE!Xk=vS6D(}&MYNV2?wFWcV<3W>6xa7q$A+~1l4{5S6W@I}JoMb(m}qPG;fLh% zu%s9j)BR3d;U_$|bmbGZlqV57S#NX(gy6*{FkGO30xGYZrJB?o`W%C%NH3_iwFkJD z?nMrQHGLOnXKQpNthnXtw=x=vjB4GFUFc4mdy^wz`alY%8KLB3XN)3r7c`x{99nru zgjJ>=H$eI|H8n7hGjL)V>9&nLHQ8dy<({`^^%Gd~Qg$*HKiaAr4) zWT-k2ml+5CcxcD;@QZa!YR^ zHqNM>CB_;1zo{Z?Y$>(q0u2yh4stZJ2FpCu>Wn$DA6+N9bqjRi*Cxb@QQg;-=;>iS zx0fnIwt7}=PEjJw*ZTM^`{kkq>s))vBrQT33HhU~P>vL#DouO*y1iY2SG*`q zxBvAVoZ~&+JAyY~*Vx%VXRfOlKQG}n(1lqHkI^edbo`r?h65bF5;W0_ng#})AY=pd z?@%rXMKSzcP1(q`H~ZiXYA`VujM_2^)X>V$+3c-J0WK?AD4p1mFiaH`?oAFq*p5#Usk3?oyGn?T}$7y&96Es7qt*ZuA7?av7wx#Rr`L|9Pb?41o2 ztytWcBWznI3Vri|3G%S;Tbzr~9`SoNj3(3XWFLV80not!2Mcm*rxGWS`DeD!OXr~v zAo9fTV2HXdBbR(YtH*V)+R$59EklN3H8Rq^Sf~L9ZSjIaA^PRaT2^T~^2Xgae!W7k zjhv{q^OUBeQ{7hvU%^UAx1j`c?}d=PTx?>qCkN;kNEaxQ@{!e-y#A^bhlz|vKn_c1 z7ng0&?59S`9VU9&m3V>)dll}bNxg>6hIvU6;HLped85ByW+%~^HM#$1aTdL!xj}C> z?<%)!guf)geao@nmaIDPPas0iqPSvcswN^lu`+UPHa_I$mfN616CLa0kw|$N`+6~G z9>)Gh796$fJV+C6J$aoPDqZqW9~pXvJ)MmxVsBDUNRpwH`3PpMrUZsd;fFMZ z<&EhO?{fVpoZ~go3>MTN^9a-LYyF(HG)Pcc+lz+Da*mw(KOm*OT5d@z+IoMh% z)c1%F&7j3O7D^GJXJAkXn}pQ{m3UcouuKP4@eHYP}%5VZY;$w$9)VVG54vbg=^x2UXSZ`8!| z+K4u4zmL)(E}-~%)*EX^a-9Q82o%`woue1k>D?Nnpe?hHWW6%*@#!)hSs#I*2g2{- z-$&M5*bhnJyZ#XLm_aG_@eW5bMNVI+vk+A~3Y;&}b=Lk{vUM3@X-Lq^g9BG~b zB@%>D$Gd%@ZMY_OAJiL)YpcVs0afFjaK47gNaU7Z+4?ENQK zhY26tlqJR*t9KEkDLVWF@)iXa4%&}w-cT;WxLHIzT#|!e$UY*hQ9aSF{S$ZdHQE9{ zpXlglz4L8n$gCb~R}2)pR86Sp>&I84$+vb+&a7?c6L9rodxBs}P?!=62Wu;2F%opi z(lXAT%TZz9_J6G%?oX)L|0;je-K`9smSe*a({{mJPxbp&C+Gx5@bl%-f<$2#{MDrl zfscKeZuiN^cc-+J%J$IB<|q-EngIt8LHCY=?>-lqmXCD~hPNcoh z&H9GDc3`C+t7wz-1ByzP<8vR<+F<}OL9=OLLus9Ii;W4e2?#xkhyn7EZ-m9?^ zGwftaLa12t^VykH6WZB|>VClP0BLH1RtjBsZySVXbv?aoD^pJ~oM5SkUpCDR%dLrn zq`p(-rCMbyQ?Rt2N`}Wa^Xz9wu6osbm-jw*6ywsnW=H!2x|g6y`EHqI&kv)8t6bN# zE&v}xvXFmM9y3AN*9tQf6>4UWDCHiEYr1yLW6#bI*5Q~~Dw2gcT8x^bYPjC=lhewzJ-5!{H-Sp4iZM50fL#lvJJUJs- zb`%H03Uu;AL$yc;-5hk_#b1b-30Osu;sT%rh+!wl)!|ukkz#Akd(8@T%l~YIN&rK= z_;x@8EFZ}$pKEKHv)yPGX4C^HYrw9MP?|SC(AxmAGAy6`6 zHy~rnjO1u9O&2VDf&R8P`g{0xP;PGeE51@}OBb|}$4+FxX$WYRJl6ey$Iy?JT3ln( zE%;N)5;suq;~}MMTPH#DysEaN!?nn#qV|`e6m;}GMS7JGG+hr=)Qw*i88tNt3te$T zS&mOtY1~9?A6xu=-!$pemISJ-BuIS-p_ACt3Yi9?LoX%yP(y?Ct9a^cLY0HWr3!0d ze1_?z9rd-G5kVo9gJe)F%f5L+Gy+!!(OCfd#R0YnoH7Jf8e12z?5*p{7tP%pCc)!( ze2A%4n+FygxP6Eo2jc1-s|ew6;%0e>eSHx1kgOX1YrSx#2?!?;*&Qm`7A0TUV<6kbbv(S^#U_=}}gm3jb zm}x^QdnnCtUp+uXUzxD$+b}JFNIh=68?6`;?8qyiCH|Hlu%Nq!fm%UALH zH4{Rlfu-70c!hJxF;&0iJ8DxO9&eZ!gOoc)`QQnj8aKuh6N81~cEHY2>x-vr>~$A9 z?o%Oq27ja&90yQd_e!o|U@q3!4D%QdUFhiRzhP&C^76q0%>R7+8UNb|(GJ0wwdvV= z_(lZlhv^>zyfx>@LV^0`BhqG^CKEQus_vM*z5Oq8+f$5PulW1qi8wW3P~Z&TUkE$h zJU7DUZ<)0w_gFGe>43ifv!#Xj@5*%(wSr`(_^9|MIJw|5PPH8^rSt*O3e@`-D3;VD zPd@UhN`?%niW2l6jOMPg@)|YJr9FEOF&aEYIFGYFJL5s&IO%pU<-0$=1XfqbzkkjR zdPi5{;Q<1m*(c;i#U{5x4{0^@?!d%^YwaB!O|vf^8ya?pbxAPR1_*!$mz8YHzWVOR zOC4Ox@*0~WjPR#n{8~vh1W$SEyd`SfHy#xw*ArPp6;l>E?oo~k0gAmcl6MRCvo8oH zq@<+pKYeO1zn#SmQ?crmtc8{JSys6zabYwnib(!cXwDLTudVaUsjYn|$1q0Yn zBWty*_z{4?!JR`|NxS!t zGIF73M35I89j_H3nV`e3@>ysqUWDHKwxzwaSNu?Fo%8H-{%mP2@z^pREqv0I77|AO zk@t1&RzqTt@MYYq!Ko|P@fsylyDW5KSvUzKtG3HPE6V_Ay$R-vJp#+?q2h2@e1?45 zo$J_${Gq6*2p4tPX4S$1CY8X`^=iWFOuUA_*%gmF9M(xd(z!BG#p&ea6v%O{m0MQ* zi&}~VBh%gJUji{;&Vh#$os!ZABCf2VAr&GHAqyY$M!0MM4S|fs2q@9UL{%9G;NV^e z!|n9?+k{QwH)>s8Y?+zwB}UD|%e&&PAxJ7~Q%T(!trngVLW^U0(3p@wnV$ZiRjxju z{&K`SE0p1fw3ad{gZ$@E+U34gm>h(`}uZ^aho|+ z8^ZM{wq0Na90#fqMCHye)W`C@H2fYl^&AR>sKi92uy=R5NqZMXf(@u7$yJ86w7=H; zQn9pT@;lzmvHNz{q-t&4*uN-+THO0Re6^DkH(1e0;|lGcO6-Jku1J4h+a4@awrgeQN^)%6gvzCLY%4NxEuACZ^~{h7|rh^ka4PCh)(%*)1BGyYXPvxY8`S0_c3CfLzoY`z4xu!?2MXcc$bQR zK{(cziiXcgmY=9XT3vC5#^X`Mb$8tQs0 zeyX5P(wOyNVppscCnpnaw%wLl%@ISer==ftJLco*nA7jDnz^XMdh?O@N=jr;uCqn# z+_SbZDYH#I7dwr;ijM5)*ebbXAft5Q%eOwZ@3Q5fsed7E(fo_|zQ2-xH|{Tz&Vmu4 z_Vlkdy(az3GddwR!kP41r?YWFi!}JDOa9CayRA|o zvhrc-vHt!t^YOW=pYJO&Vx}Z>#SD(+yUZwjsEmH8DR~~StKu8{Gk)T1HZWNJQTc4T zbbRtCe^HvO@#g)QQp=I{C#Tb=IzDqHV><VsXEBmH?Y)5D3gIEi5{mOJ9vWR{@0(DZ*8y`ND z3&gN}T=D1U+$U3VTfn0_Nc>YSvahP!{a5ya+fEPa3^3Pcqj9#tahjMo{dgw3pOAJp zW-&uphl?CkH`p#mM@OV&WOFZsipT9GkNlOIF6MV!GH%Mds1e*fKKZdVUTJZgIug5| zp}BFlPCngaxCK>rT5a8NVuPPH4Tm%k7~RX$YS0w}As7U|VmP(QYN@aNpI6tIXeN25 z3;eQ@6}(8|j%z?ThDb}%aM>_xkOi>&7nDwN)%fs*)<+dm;De<83NSyNLPJGnQq1qq zt1fcaN`y)dD|{ZgIM=$Z=`@^diZUvS)a=ncfB+nVO~`cb-V$~|B^5pyStEhYapS9& zH$gG2`tr?}auYJsfHH+m3m=V$@yyZov8=W>Z>i~*Vd=BhoUxk^f*xngD@DJg#Rcga zqrEggZE%+Y$z|^`5A%?0k8&=cMv=qwJ6M|eZdap3aa(=D%8l7gBGdP zVHQGEt&7EJISF2lSkf58m<@p}b^>7pjp!MBzs&r0TB&{-MpC@l7ed`9EM286k=Mg6 zs|-Qe-ClZ5-KZv+$&A8D$H0WM3^ND=m7Y3ATK=Mb*4j|58IA3G2>>s63})z;mhD&S zyjIf_A{4_~1kxnk;*TmjhzdQJ$b0W5PKBc9Wz|Xj9Eq<42hk>ELJL;OxP3(LJvJVL zZ~in62|sMs)rQAIZ52i=B=au{JZ7ekVPXNTo08HHpO*GnQ;C+gY-I=5SySxAKX&%> zmwayBdNO(U6HGgrQ*LyP6{#)@h-`2}A-gfTd}PzFU@=R84>f z21pNt9ngkAiv*m4`Tl(q0gEG(=ib_f+LG-K8a9ZnrA5yZfW?AouMgyuWy2v1!4e${tMq5u* z-iIG=Q+a#sJpBwRcihIOf|GM|a|&u|S)d`(e&0;09S1@lkP?7AM+=RVXo+aq*usa0 zwT9m3F&9}gra&K{YLdtRIQ$RQhY230LOKSS6}aEhAjg6)`Y@WYw5&`~LBZP9VYK=8 z;rIFNzW`JqNB{)racEtBC<2%>K(B;%k)LlT0_G60n29ei$8@jb0{({yNVK~j5g zYG;9mQ&YPKs2(CrI?hPsva+rCbh3b?3~oE36M>!twGab*1rTuXhU7u4gY=uLei_8o z&(a1Kf!5w{n3CE%bD#rP&VngsAbY4?ls&y!{MP)3o*4j_gB2?OFgOy*=p z1rM!=jXvE-xt=xDku2`R%*r|kCBnJ%OW>&PAR3J-MG^)`A4nAfOB$4)2O1hAaGw!T z?Z=M?q4?CwhQ4Bti&7oOORpsM$U(|6z!w2JW0@UV^(sNc+;s@Qz`lK`Th* z^C?trF65L$0tNR$+l99XN0?1#65JKIX~-`E8jRO<`LR@FGypIq4if@EmjDn#2TK{w zmcAuB(!EL{3Fxhc6bxC#a3!FekfN0WpBu@dw_9!jbXwNRAuJpZo&;#PeUK|qfLjJv zP=-_%E=dz4hQRcG$;<>0m6?;%YGiffBb6A^{sxrz`c!QkeA@I}R&8x<==}l{jD>{- ziVSr_LuK&FpmPd7yS+eSd?7X7E5wRxZ*Qk%;JpkNKwqB%glZGzVRBPbQw(Od`({V~ z77)D>uy#hpQjGz?_32{gdwfHVmXbv>_|fGykzDHQSDH{dYCjM<0Ik9ZPtL4!h+HoZoZ%Wo7F&E>2TR{X$X zE4M?xgFZn7I+@szXq{P4A*^ji>U}WK{(v$Ga$@Dcj0dbKSRXbq+46UgLKCiTM^DdJ z-7RCR%ii@CIezRA+aL(gmHaaSxD=kEvamiW!XT9nLIKg6wVImz@9`w%_mQ~00Mzv z2n97W`H$ySAL8QVjg6_nQwP%YqX5fWQ6DNuDTbfk8q}03!^WrO;CKgK?hfVKW768@ zcVJ^ci)dL}ii(Ctw0`4PW;{pIsobvWaS<_!FonWguW%SVfHaj{mLB@{fe|8_G<1ueCvzcj@fp`cAd*U) zoN6Ya@6lcWLc&F3y^jNG&R{~we^0qD$mGSgH%^Fp{VXBx;xFY1!GW6$X`C=(EOks+y5E z_n)rUJ)f%i3Lnj&0PzKL>g$av9daHsXY1s6RlI^kL36G5_jEne)s?&=FP6zFl6ZK4Y%6*qe>t3n32+HHEiu)f^^ILw$T=6+!L8!W4^XFPKqBs zd@#`0S7#)NvPoE7T|JgmVI;uE_eA%PS$8f7&}&TP+^3?Z9tGj)$DZ;Dj-u)3} z+}5)6#63OTD?qA5e-2Xh@u}@xD1AQDP`~E+O8y9r47Y{On{KYI%rTET_*52lyr#Z6 z$y6l0d!K33S-FvP$)`-v)7h#kr`o%6as=JfB(0*NV&mkbE)y17{%;3mX=&+d?y0x! z=hP8fSy^eGkSQQ2SYNzJo%p-Ze|LAcz0%=kxU6=bHh-HbUV=2VIuuEVI!qX|`5eT1 zO+G_cIygLko(uO=ns-?|h44Js<)KW)s_XJc#yp)57UOtWmvgo4#5g6u_+)JMGPWH_ zIG}p+d;iK?8FpMKJPu(nhYy^P zm{>0<`23~-zD$($!`!YlYMEqsoAs)5+Rd1j(Yj zp`j4k%>4klt?}GWNASr#!u`DJYWWy_JQqT~zbKyO}lu5N5 z*GD=dxHc0L6URXGQhSNR?`+vfM{1rsLNBDiRLV4%FwkbvC4Cp_7T~2#bGnn#i@x4| z8-u^nU}oYjNuM12OlB$#fy5EmHONj&laZN@?Tx#B=ij@=7rItz$zm%G^KMmb)dz5e z`S}HHyllpEZcho*8admF(%c-Sx!F2-v-yQkKFp+zd2~1H)tT`nD^ms{98Ut@w@FJ` zh8$bRlP?=?^-a@~;$OAUE-<1YTvw)qA<|*ze)CTEV=Zi)oi$)EkvDw$7<(lVd3vlb z6(jsfk^$#YPs_Glgab?t+F)3M{K!ikwqA2NW>SqDt--xD^v&=VEoM?aZtgT#z-Jx) zL>L%>3F3;iI^S3KYWx!}6pwt3VY$HLEb4iVQPUR_Gv*VOi;Ih2YFu#suCM1;Ih&D( zqFU0GhqOu_55s>D?1usFu%TgF=YC|J4urOZqQyd3^6$0F%8QFrBO(Z_YM0M-a?YRq zEib8F>G`Mx4;$$Ayg0{ihA$z-d*yog*QBSf&+fUU>dwv%{3w{YrdhH2Sw2>Q(Of^) zCFOGO8Mn&NM5}&|R^25JJG(sHX8lPH?~0_^7h|G}W)KQ>HSwx8JEpv?CMxMwh~3=W zY@Ric!7R$0ly2~~{0^<_S(y{8zJ*ch}U_!<*OwQrw<;{4^`12ta28P;iPMZDwLet55AKhCVhW#M; zX|o7h%?Lk1eZHSuVW7hE1yXs~-bM|+`gV49c%i|zisjh`1*_n&AA|Gjw$yt^HbcE- z-l^!4ybz^MzHla7Rik=u?ZM3;z4=t}m`5y83PP7pok#38ODbzJIU7$kVlBK>#D2^c zyAb_LHpBRP@)v8Ez+;F78Q~8vXdT z78bH*W@a_c3#grZ*=9;O?tqlc&(kgPzHCFGnhhhR`gQ^S^ZamS^K5!@Qj0ESrlQ`# z`h@}yu&uwd_rfe2s!{EFf z#0mmhj0(oD*woG6qJo*Y7Z?~v~`|bzD zz5RXL-JjVtyi+1Bi*KP}>;RsNFf=!p?p*y@PF`L+Z`eFk+3^*Z_0@oR z(bk<_zqpY#HwQ;YpR`;d$4A_@F9f!}2 z_Uhk=UZJL{gM$e26IYP}ur63WbO<$d7UpGS$gH^jy7nwDFApV`A>zH84Kl+5Kbht>7qP0u@lim_UW-Y{f#usGq6b=#U2hs-8P$mzj~s{y>@ zQ2e{mxGf2$#+TFm-uRUkOqs-3kzxd5zqPG?6oqzKwh^(eiLlroZ4t9Mo0cg!qF_I3 ztg~)2Ivw5H^rJTZvPJN&)u-;j|AM&BZhjNDtF*3pG^J(l?>Q#M@9pz{{67B0g3F5d z?U%&jCFhI=zG=H|Nw3xAVPG7d5$z{ys_}~UJX<;Xmp&DFYr}hcmvU=6#i=pHa^c+) zF-D+OSTQyRyd-wxwpwv*4gb!R3;L(d zhhvwKf2W~ezXX2-5)$tt|N1=)2UKzljNdXZ|KGR&-!9!TVTfhKqsK4B1d&UoD69Ig J@WHd-{|~Zw7pMRL diff --git a/references.bib b/references.bib index 07868fed..919d0e36 100644 --- a/references.bib +++ b/references.bib @@ -220,3 +220,13 @@ @misc{Team_2023 @misc{VinBrain, url={https://vinbrain.net/aiscaler}, journal={VinBrain}} @article{Ardila_Branson_Davis_Henretty_Kohler_Meyer_Morais_Saunders_Tyers_Weber_2020, title={ Common Voice: A Massively-Multilingual Speech Corpus }, journal={Proceedings of the 12th Conference on Language Resources and Evaluation}, author={Ardila, Rosana and Branson, Megan and Davis, Kelly and Henretty, Michael and Kohler, Michael and Meyer, Josh and Morais, Reuben and Saunders, Lindsay and Tyers, Francis M. and Weber, Gregor}, year={2020}, month={May}, pages={4218–4222}} + +@ARTICLE{6296535, + author={Deng, Li}, + journal={IEEE Signal Processing Magazine}, + title={The MNIST Database of Handwritten Digit Images for Machine Learning Research [Best of the Web]}, + year={2012}, + volume={29}, + number={6}, + pages={141-142}, + doi={10.1109/MSP.2012.2211477}} From 3bb9b37fe35f90f0ffca58e81cb3802920224412 Mon Sep 17 00:00:00 2001 From: Henry Bae <69275685+BaeHenryS@users.noreply.github.com> Date: Mon, 23 Oct 2023 02:22:27 -0400 Subject: [PATCH 09/22] Updated References --- references.bib | 78 ++++++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 69 insertions(+), 9 deletions(-) diff --git a/references.bib b/references.bib index 919d0e36..59f53bc3 100644 --- a/references.bib +++ b/references.bib @@ -221,12 +221,72 @@ @misc{VinBrain @article{Ardila_Branson_Davis_Henretty_Kohler_Meyer_Morais_Saunders_Tyers_Weber_2020, title={ Common Voice: A Massively-Multilingual Speech Corpus }, journal={Proceedings of the 12th Conference on Language Resources and Evaluation}, author={Ardila, Rosana and Branson, Megan and Davis, Kelly and Henretty, Michael and Kohler, Michael and Meyer, Josh and Morais, Reuben and Saunders, Lindsay and Tyers, Francis M. and Weber, Gregor}, year={2020}, month={May}, pages={4218–4222}} -@ARTICLE{6296535, - author={Deng, Li}, - journal={IEEE Signal Processing Magazine}, - title={The MNIST Database of Handwritten Digit Images for Machine Learning Research [Best of the Web]}, - year={2012}, - volume={29}, - number={6}, - pages={141-142}, - doi={10.1109/MSP.2012.2211477}} +@article{vinuesa2020role, + title={The role of artificial intelligence in achieving the Sustainable Development Goals}, + author={Vinuesa, Ricardo and Azizpour, Hossein and Leite, Iolanda and Balaam, Madeline and Dignum, Virginia and Domisch, Sami and Fell{\"a}nder, Anna and Langhans, Simone Daniela and Tegmark, Max and Fuso Nerini, Francesco}, + journal={Nature communications}, + volume={11}, + number={1}, + pages={1--10}, + year={2020}, + publisher={Nature Publishing Group} +} + +@inproceedings{altayeb2022classifying, + title={Classifying mosquito wingbeat sound using TinyML}, + author={Altayeb, Moez and Zennaro, Marco and Rovai, Marcelo}, + booktitle={Proceedings of the 2022 ACM Conference on Information Technology for Social Good}, + pages={132--137}, + year={2022} +} + +@article{yamashita2023coffee, + title={Coffee disease classification at the edge using deep learning}, + author={Yamashita, Jo{\~a}o Vitor Yukio Bordin and Leite, Jo{\~a}o Paulo RR}, + journal={Smart Agricultural Technology}, + volume={4}, + pages={100183}, + year={2023}, + publisher={Elsevier} +} + +@inproceedings{bamoumen2022tinyml, + title={How TinyML Can be Leveraged to Solve Environmental Problems: A Survey}, + author={Bamoumen, Hatim and Temouden, Anas and Benamar, Nabil and Chtouki, Yousra}, + booktitle={2022 International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies (3ICT)}, + pages={338--343}, + year={2022}, + organization={IEEE} +} + +@inproceedings{ooko2021tinyml, + title={TinyML in Africa: Opportunities and challenges}, + author={Ooko, Samson Otieno and Ogore, Marvin Muyonga and Nsenga, Jimmy and Zennaro, Marco}, + booktitle={2021 IEEE Globecom Workshops (GC Wkshps)}, + pages={1--6}, + year={2021}, + organization={IEEE} +} + + +@inproceedings{zennaro2022tinyml, + title={TinyML: applied AI for development}, + author={Zennaro, Marco and Plancher, Brian and Reddi, V Janapa}, + booktitle={The UN 7th Multi-stakeholder Forum on Science, Technology and Innovation for the Sustainable Development Goals}, + pages={2022--05}, + year={2022} +} + +@article{zennarobridging, + title={Bridging the Digital Divide: the Promising Impact of TinyML for Developing Countries}, + author={Zennaro, Marco and Plancher, Brian and Reddi, Vijay Janapa} +} + + +@misc{Sheth_2022, title={Eletect - TinyML and IOT based Smart Wildlife Tracker}, url={https://www.hackster.io/dhruvsheth_/eletect-tinyml-and-iot-based-smart-wildlife-tracker-c03e5a}, journal={Hackster.io}, author={Sheth, Dhruv}, year={2022}, month={Mar}} + +@misc{Verma_2022, title={Elephant AI}, url={https://www.hackster.io/dual_boot/elephant-ai-ba71e9}, journal={Hackster.io}, author={Verma, Team Dual_Boot: Swapnil}, year={2022}, month={Mar}} + +@misc{Rao_2021, url={https://www.wevolver.com/article/tinyml-unlocks-new-possibilities-for-sustainable-development-technologies}, journal={www.wevolver.com}, author={Rao, Ravi}, year={2021}, month={Dec}} + + From 90b4b0d28d20eb4137d7f4d3beea9d9f367a3350 Mon Sep 17 00:00:00 2001 From: sophiacho1 <67521139+sophiacho1@users.noreply.github.com> Date: Mon, 23 Oct 2023 04:40:27 -0400 Subject: [PATCH 10/22] added links for frameworks when first mentioned --- frameworks.qmd | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) diff --git a/frameworks.qmd b/frameworks.qmd index 10274413..e022bf33 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -23,7 +23,8 @@ Learning Objectives Machine learning frameworks provide the tools and infrastructure to efficiently build, train, and deploy machine learning models. In this chapter, we will explore the evolution and key capabilities of major -frameworks like TensorFlow, PyTorch, and specialized frameworks for +frameworks like [[TensorFlow]{.underline}](https://www.tensorflow.org/) +, [[PyTorch]{.underline}](https://pytorch.org/), and specialized frameworks for embedded devices. We will dive into the components like computational graphs, optimization algorithms, hardware acceleration, and more that enable developers to quickly construct performant models. Understanding @@ -46,13 +47,13 @@ specialized hardware like NVIDIA GPUs to further accelerate training via optimizations like parallelization and efficient matrix operations. In addition, frameworks simplify deploying finished models into -production through tools like TensorFlow Serving for scalable model -serving and TensorFlow Lite for optimization on mobile and edge devices. +production through tools like [[TensorFlow Serving]{.underline}](https://www.tensorflow.org/tfx/guide/serving) for scalable model +serving and [[TensorFlow Lite]{.underline}](https://www.tensorflow.org/lite) for optimization on mobile and edge devices. Other valuable capabilities include visualization, model optimization techniques like quantization and pruning, and monitoring metrics during training. -Leading open source frameworks like TensorFlow, PyTorch, and MXNet power +Leading open source frameworks like TensorFlow, PyTorch, and [[MXNet]{.underline}](https://mxnet.apache.org/versions/1.9.1/) power much of AI research and development today. Commercial offerings like Amazon SageMaker and Microsoft Azure Machine Learning integrate these open source frameworks with proprietary capabilities and enterprise @@ -85,14 +86,14 @@ But the release of large datasets like ImageNet in 2009 and advancements in parallel GPU computing unlocked the potential for far deeper neural networks. -The first ML frameworks, Theano (2007) and Caffe (2014), were developed +The first ML frameworks, [[Theano]{.underline}](https://pypi.org/project/Theano/#:~:text=Theano%20is%20a%20Python%20library,a%20similar%20interface%20to%20NumPy's.) (2007) and [[Caffe]{.underline}](https://caffe.berkeleyvision.org/) (2014), were developed by academic institutions (Montreal Institute for Learning Algorithms, Berkeley Vision and Learning Center). Amid a growing interest in deep learning due to state-of-the-art performance of AlexNet (2012) on the ImageNet dataset, private companies and individuals began developing ML -frameworks, resulting in frameworks such as Keras by Google researcher -François Chollet (2015), Chainer by Preferred Networks (2015), -TensorFlow by Google (2015), CNTK by Microsoft (2016), and PyTorch by +frameworks, resulting in frameworks such as [[Keras]{.underline}](https://keras.io/) by Google researcher +François Chollet (2015), [[Chainer]{.underline}](https://chainer.org/) by Preferred Networks (2015), +TensorFlow by Google (2015), [[CNTK]{.underline}](https://learn.microsoft.com/en-us/cognitive-toolkit/) by Microsoft (2016), and PyTorch by Facebook (2016). Many of these ML frameworks can be divided into categories, namely @@ -1799,4 +1800,4 @@ embedded ML application. [^6]: [[https://www.geeksforgeeks.org/batch-processing-operating-system/#]{.underline}](https://www.geeksforgeeks.org/batch-processing-operating-system/#) -[^7]: [[https://www.tensorflow.org/tensorboard/scalars_and_keras]{.underline}](https://www.tensorflow.org/tensorboard/scalars_and_keras) \ No newline at end of file +[^7]: [[https://www.tensorflow.org/tensorboard/scalars_and_keras]{.underline}](https://www.tensorflow.org/tensorboard/scalars_and_keras) From d522a9053aaa52a0a988ee33e241e962ee5e50de Mon Sep 17 00:00:00 2001 From: Henry Bae <69275685+BaeHenryS@users.noreply.github.com> Date: Mon, 23 Oct 2023 19:20:35 -0400 Subject: [PATCH 11/22] Further refinement in Basic and Advanced Framework Components, Image/Section Fix --- frameworks.qmd | 74 ++++++++++++++++++++++---------------------------- 1 file changed, 32 insertions(+), 42 deletions(-) diff --git a/frameworks.qmd b/frameworks.qmd index e022bf33..3ff57772 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -448,8 +448,7 @@ other in another direction. The figure below demonstrates this step. Therefore, vectors and matrices can be considered special cases of tensors, with 1D and 2D dimensions respectively. -![](images_ml_frameworks/image2.png){width="3.9791666666666665in" -height="1.9672287839020122in"} +![Visualization of Tensor Data Structure](images_ml_frameworks/image2.png){width="3.9791666666666665in" height="1.9672287839020122in" caption="Visualization of Tensor Data Structure" align="center"} Defining formally, in machine learning, tensors are a multi-dimensional array of numbers. The number of dimensions defines the rank of the @@ -494,8 +493,7 @@ taking two input matrices (or tensors) and producing an output matrix directed acyclic graph above computes $z = x \times y$, where each of the variables are just numbers. -![](images_ml_frameworks/image1.png){width="1.9947922134733158in" -height="1.6791174540682414in"} +![Basic Example of Computational Graph](images_ml_frameworks/image1.png){width="50%" height="auto" align="center" caption="Basic Example of Computational Graph"} Underneath the hood, the computational graphs represent abstractions for common layers like convolutional, pooling, recurrent, and dense layers, @@ -846,57 +844,44 @@ After training deep learning models, frameworks provide utilities to evaluate performance and gain insights into the models\' workings. These tools enable disciplined experimentation and debugging. -##### Evaluation Metrics +#### Evaluation Metrics Frameworks include implementations of common evaluation metrics for validation: -- Accuracy - Fraction of correct predictions overall. Widely used for - > classification. +- Accuracy - Fraction of correct predictions overall. Widely used for classification. -- Precision - Of positive predictions, how many were actually - > positive. Useful for imbalanced datasets. +- Precision - Of positive predictions, how many were actually positive. Useful for imbalanced datasets. -- Recall - Of actual positives, how many did we predict correctly. - > Measures completeness. +- Recall - Of actual positives, how many did we predict correctly. Measures completeness. -- F1-score - Harmonic mean of precision and recall. Combines both - > metrics. +- F1-score - Harmonic mean of precision and recall. Combines both metrics. -- AUC-ROC - Area under ROC curve. Used for classification threshold - > analysis. +- AUC-ROC - Area under ROC curve. Used for classification threshold analysis. -- MAP - Mean Average Precision. Evaluates ranked predictions in - > retrieval/detection. +- MAP - Mean Average Precision. Evaluates ranked predictions in retrieval/detection. -- Confusion Matrix - Matrix that shows the true positives, true - > negatives, false positives, and false negatives. Provides a more - > detailed view of classification performance. +- Confusion Matrix - Matrix that shows the true positives, true negatives, false positives, and false negatives. Provides a more detailed view of classification performance. These metrics quantify model performance on validation data for comparison. -##### Visualization +#### Visualization Visualization tools provide insight into models: -- Loss curves - Plot training and validation loss over time to spot - > overfitting. +- Loss curves - Plot training and validation loss over time to spot overfitting. -![](images_ml_frameworks/image7.png){width="3.5052088801399823in" -height="2.3043503937007874in"} -Example of a loss curve from TensorBoard [^7] - -- Activation grids - Illustrate features learned by convolutional - > filters. +- Activation grids - Illustrate features learned by convolutional filters. - Projection - Reduce dimensionality for intuitive visualization. - Precision-recall curves - Assess classification tradeoffs. -Tools like TensorBoard (TensorFlow) and TensorWatch (PyTorch) enable +Tools like [[TensorBoard]{.underline}](https://www.tensorflow.org/tensorboard/scalars_and_keras) +for TensorFlow and [[TensorWatch]{.underline}](https://github.com/microsoft/tensorwatch)for PyTorch enable real-time metrics and visualization during training. ### Differentiable programming @@ -1591,7 +1576,9 @@ goal is to balance model complexity, hardware limitations, and software integration to design a tailored ML pipeline for embedded and edge devices. -### ![](images_ml_frameworks/image4.png){width="6.5in" height="3.1666666666666665in"}Model +![TensorFlow Framework Comparison - General](images_ml_frameworks/image4.png){width="100%" height="auto" align="center" caption="TensorFlow Framework Comparison - General"} + +### Model TensorFlow supports significantly more ops than TensorFlow Lite and TensorFlow Lite Micro as it is typically used for research or cloud @@ -1604,10 +1591,10 @@ native quantization tooling and support, where quantization refers to the process of transforming an ML program into an approximated representation with available lower precision operations. -![](images_ml_frameworks/image5.png){width="6.5in" -height="2.0833333333333335in"} - ### Software +![TensorFlow Framework Comparison - Software](images_ml_frameworks/image5.png){width="100%" height="auto" align="center" caption="TensorFlow Framework Comparison - Model"} + + TensorFlow Lite Micro does not have OS support, while TensorFlow and TensorFlow Lite do, in order to reduce memory overhead, make startup @@ -1621,7 +1608,9 @@ schedule code to different accelerators, whereas TensorFlow Lite Micro does not, as embedded systems tend not to have a rich array of specialized accelerators. -### Hardware![](images_ml_frameworks/image3.png){width="6.5in" height="2.2083333333333335in"} +### Hardware + +![TensorFlow Framework Comparison - Hardware](images_ml_frameworks/image3.png){width="100%" height="auto" align="center" caption="TensorFlow Framework Comparison - Hardware"} TensorFlow Lite and TensorFlow Lite Micro have significantly smaller base binary sizes and base memory footprints compared to TensorFlow. For @@ -1702,9 +1691,7 @@ tutorials and example projects. Community support provides some assurance that the framework will continue to be supported for future updates. There are only a handful of frameworks that cater to TinyML needs. Of that, TensorFlow Lite Micro is the most popular and has the -most community support. - -## +most community support. ## Future Trends in ML Frameworks @@ -1713,7 +1700,9 @@ most community support. Currently, the ML system stack consists of four abstractions, namely (1) computational graphs, (2) tensor programs, (3) libraries and runtimes, and (4) hardware -primitives.![](images_ml_frameworks/image8.png){width="2.557292213473316in" +primitives. + +![](images_ml_frameworks/image8.png){width="2.557292213473316in" height="2.9092125984251966in"} This has led to vertical (i.e. between abstraction levels) and @@ -1796,8 +1785,9 @@ embedded ML application. [^4]: [[https://proceedings.neurips.cc/paper_files/paper/2014/file/1ff1de774005f8da13f42943881c655f-Paper.pdf ]{.underline}](https://proceedings.neurips.cc/paper_files/paper/2014/file/1ff1de774005f8da13f42943881c655f-Paper.pdf ) -[^5]: [[https://www.tensorflow.org/datasets/catalog/mnist]{.underline}](https://www.tensorflow.org/datasets/catalog/mnist) +[^5]: [[TensorFlow: Large-scale machine learning on heterogeneous systems, +2015.]{.underline}](https://www.tensorflow.org/datasets/catalog/mnist) + +[^6]: [[Patrick McClanahan, Introduction to Operating Systems, 2023]{.underline}](https://eng.libretexts.org/Courses/Delta_College/Introduction_to_Operating_Systems/03%3A_The_Operating_System/3.06%3A_Types_of_Operating_Systems) -[^6]: [[https://www.geeksforgeeks.org/batch-processing-operating-system/#]{.underline}](https://www.geeksforgeeks.org/batch-processing-operating-system/#) -[^7]: [[https://www.tensorflow.org/tensorboard/scalars_and_keras]{.underline}](https://www.tensorflow.org/tensorboard/scalars_and_keras) From 0dd41d4a1cca0ce476c719f259c2bc0cb1a2a6e8 Mon Sep 17 00:00:00 2001 From: sophiacho1 <67521139+sophiacho1@users.noreply.github.com> Date: Mon, 23 Oct 2023 20:07:58 -0400 Subject: [PATCH 12/22] 7.1 done --- frameworks.qmd | 17 ++++++----------- 1 file changed, 6 insertions(+), 11 deletions(-) diff --git a/frameworks.qmd b/frameworks.qmd index 3ff57772..145a99d2 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -2,29 +2,24 @@ Learning Objectives -- The evolution, core components, and advanced features of ML - > frameworks +- The evolution, core components, and advanced features of ML frameworks - How frameworks specialize for cloud, edge, and tinyML environments - Challenges of embedded ML and how frameworks optimize models -- Criteria for selecting the right framework based on models, - > hardware, software factors +- Criteria for selecting the right framework based on models, hardware, software factors -- How to match framework capabilities to the constraints and - > requirements of a project +- How to match framework capabilities to the constraints and requirements of a project -- Ongoing innovations in frameworks for next-generation machine - > learning +- Ongoing innovations in frameworks for next-generation machine learning ## Introduction Machine learning frameworks provide the tools and infrastructure to efficiently build, train, and deploy machine learning models. In this chapter, we will explore the evolution and key capabilities of major -frameworks like [[TensorFlow]{.underline}](https://www.tensorflow.org/) -, [[PyTorch]{.underline}](https://pytorch.org/), and specialized frameworks for +frameworks like [[TensorFlow]{.underline}](https://www.tensorflow.org/), [[PyTorch]{.underline}](https://pytorch.org/), and specialized frameworks for embedded devices. We will dive into the components like computational graphs, optimization algorithms, hardware acceleration, and more that enable developers to quickly construct performant models. Understanding @@ -55,7 +50,7 @@ training. Leading open source frameworks like TensorFlow, PyTorch, and [[MXNet]{.underline}](https://mxnet.apache.org/versions/1.9.1/) power much of AI research and development today. Commercial offerings like -Amazon SageMaker and Microsoft Azure Machine Learning integrate these +[[Amazon SageMaker]{.underline}](https://aws.amazon.com/pm/sagemaker/?trk=b6c2fafb-22b1-4a97-a2f7-7e4ab2c7aa28&sc_channel=ps&ef_id=CjwKCAjws9ipBhB1EiwAccEi1JpbBz6j4t7sRUoAiKFDc0mi59faZYge5MuFecAU6zGDQYTFz9NnaBoCV-wQAvD_BwE:G:s&s_kwcid=AL!4422!3!651751060692!e!!g!!amazon%20sagemaker!19852662230!145019225977) and [[Microsoft Azure Machine Learning]{.underline}](https://azure.microsoft.com/en-us/free/machine-learning/search/?ef_id=_k_CjwKCAjws9ipBhB1EiwAccEi1JVOThls797Sj3Li96_GYjoJQDx_EWaXNsDaEWeFbIaRkESUCkq64xoCSmwQAvD_BwE_k_&OCID=AIDcmm5edswduu_SEM__k_CjwKCAjws9ipBhB1EiwAccEi1JVOThls797Sj3Li96_GYjoJQDx_EWaXNsDaEWeFbIaRkESUCkq64xoCSmwQAvD_BwE_k_&gad=1&gclid=CjwKCAjws9ipBhB1EiwAccEi1JVOThls797Sj3Li96_GYjoJQDx_EWaXNsDaEWeFbIaRkESUCkq64xoCSmwQAvD_BwE) integrate these open source frameworks with proprietary capabilities and enterprise tools. From 7e6182480241bb3118e78c87a6792f3b8f18889a Mon Sep 17 00:00:00 2001 From: sophiacho1 <67521139+sophiacho1@users.noreply.github.com> Date: Mon, 23 Oct 2023 20:26:29 -0400 Subject: [PATCH 13/22] 7.2 done --- frameworks.qmd | 31 ++++++++++++------------------- 1 file changed, 12 insertions(+), 19 deletions(-) diff --git a/frameworks.qmd b/frameworks.qmd index 145a99d2..91e8ed3b 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -77,7 +77,7 @@ meet the expanding needs of practitioners and rapid advances in deep learning techniques. Early neural network research was constrained by insufficient data and compute power. Building and training machine learning models required extensive low-level coding and infrastructure. -But the release of large datasets like ImageNet in 2009 and advancements +But the release of large datasets like [[ImageNet]{.underline}](https://www.image-net.org/) in 2009 and advancements in parallel GPU computing unlocked the potential for far deeper neural networks. @@ -113,7 +113,7 @@ construct on-the-fly for more iterative development. But around 2016, frameworks began adopting dynamic graphs like PyTorch and TensorFlow 2.0 which can construct graphs on-the-fly. This provides greater flexibility for model development. We will discuss these concepts and details later -on in AI Training Section. +on in the AI Training section. The development of these frameworks facilitated an explosion in model size and complexity over time---from early multilayer perceptrons and @@ -127,20 +127,15 @@ TPUs. Each generation of frameworks unlocked new capabilities that powered advancement: -- Theano and TensorFlow (2015) introduced computational graphs and - > automatic differentiation to simplify model building. +- Theano and TensorFlow (2015) introduced computational graphs and automatic differentiation to simplify model building. -- CNTK (2016) pioneered efficient distributed training by combining - > model and data parallelism. +- CNTK (2016) pioneered efficient distributed training by combining model and data parallelism. -- PyTorch (2016) provided imperative programming and dynamic graphs - > for flexible experimentation. +- PyTorch (2016) provided imperative programming and dynamic graphs for flexible experimentation. -- TensorFlow 2.0 (2019) made eager execution default for intuitiveness - > and debugging. +- TensorFlow 2.0 (2019) made eager execution default for intuitiveness and debugging. -- TensorFlow Graphics (2020) added 3D data structures to handle point - > clouds and meshes. +- TensorFlow Graphics (2020) added 3D data structures to handle point clouds and meshes. In recent years, there has been a convergence on the frameworks. TensorFlow and PyTorch have become the overwhelmingly dominant ML @@ -150,12 +145,10 @@ Networks transitioned Chainer to PyTorch in 2019; and Microsoft stopped actively developing CNTK in 2022 in favor of supporting PyTorch on Windows. -![](images_ml_frameworks/image6.png){width="3.821385608048994in" +![Popularity of ML frameworks in the United States as measured by Google +web searches](images_ml_frameworks/image6.png){width="3.821385608048994in" height="2.5558081802274715in"} -Popularity of ML frameworks in the United States as measured by Google -web searches - However, a one-size-fits-all approach does not work well across the spectrum from cloud to tiny edge devices. Different frameworks represent various philosophies around graph execution, declarative versus @@ -767,11 +760,11 @@ the value of the loss function. Knowing this gradient, the model moves in the direction that reduces the gradient. There are many challenges associated with this, however, primarily stemming from the fact that the optimization problem is not convex, making it very easy to solve, and -more details about this will come in the AI training sections. Modern +more details about this will come in the AI Training section. Modern frameworks come equipped with efficient implementations of several optimization algorithms, many of which are variants of gradient descent algorithms with stochastic methods and adaptive learning rates. More -information with clear examples can be found in the AI training section. +information with clear examples can be found in the AI Training section. Last but not least, overly complex models tend to overfit, meaning they perform well on the training data but fail to generalize to new, unseen @@ -780,7 +773,7 @@ employed to penalize model complexity and encourage it to learn simpler patterns. Dropout for instance randomly sets a fraction of input units to 0 at each update during training, which helps prevent overfitting. -However, there are cases where the problem is more complex than what the model can represent, and this may result in underfitting. Therefore, choosing the right model architecture is also a critical step in the training process. Further heuristics and techniques are discussed in the AI training section. +However, there are cases where the problem is more complex than what the model can represent, and this may result in underfitting. Therefore, choosing the right model architecture is also a critical step in the training process. Further heuristics and techniques are discussed in the AI Training section. Frameworks also provide efficient implementations of gradient descent, Adagrad, Adadelta, and Adam. Adding regularization like dropout and From 3f57539629c758e89a23e8abee3dad2ac793ee23 Mon Sep 17 00:00:00 2001 From: sophiacho1 <67521139+sophiacho1@users.noreply.github.com> Date: Mon, 23 Oct 2023 20:41:37 -0400 Subject: [PATCH 14/22] Center image in 7.2 --- frameworks.qmd | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/frameworks.qmd b/frameworks.qmd index 91e8ed3b..a5758925 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -146,8 +146,7 @@ actively developing CNTK in 2022 in favor of supporting PyTorch on Windows. ![Popularity of ML frameworks in the United States as measured by Google -web searches](images_ml_frameworks/image6.png){width="3.821385608048994in" -height="2.5558081802274715in"} +web searches](images_ml_frameworks/image6.png) However, a one-size-fits-all approach does not work well across the spectrum from cloud to tiny edge devices. Different frameworks represent From 55c85cb834020382515cdd91fdf1f86037a37b41 Mon Sep 17 00:00:00 2001 From: sophiacho1 <67521139+sophiacho1@users.noreply.github.com> Date: Mon, 23 Oct 2023 20:46:29 -0400 Subject: [PATCH 15/22] Resizing image in 7.2 --- frameworks.qmd | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/frameworks.qmd b/frameworks.qmd index a5758925..91e8ed3b 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -146,7 +146,8 @@ actively developing CNTK in 2022 in favor of supporting PyTorch on Windows. ![Popularity of ML frameworks in the United States as measured by Google -web searches](images_ml_frameworks/image6.png) +web searches](images_ml_frameworks/image6.png){width="3.821385608048994in" +height="2.5558081802274715in"} However, a one-size-fits-all approach does not work well across the spectrum from cloud to tiny edge devices. Different frameworks represent From 20277b53ac0cbb1260dff33fb1d5d6050f18e775 Mon Sep 17 00:00:00 2001 From: sophiacho1 <67521139+sophiacho1@users.noreply.github.com> Date: Mon, 23 Oct 2023 20:52:14 -0400 Subject: [PATCH 16/22] Gave Summary a heading --- frameworks.qmd | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/frameworks.qmd b/frameworks.qmd index 91e8ed3b..4dc5cbe2 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -1362,7 +1362,7 @@ improved portability, performance profiling, and benchmarking support. But ongoing innovation is still needed to enable seamless, cost-effective deployment of AI to edge devices. -Summary +### Summary The absence of standardized frameworks, benchmarks, and infrastructure for embedded ML has traditionally hampered adoption. However, recent From e10686655c46e4014ad7be3c2428beb8ebe84313 Mon Sep 17 00:00:00 2001 From: sophiacho1 <67521139+sophiacho1@users.noreply.github.com> Date: Mon, 23 Oct 2023 21:06:00 -0400 Subject: [PATCH 17/22] Got rid of > --- frameworks.qmd | 81 +++++++++++++------------------------------------- 1 file changed, 20 insertions(+), 61 deletions(-) diff --git a/frameworks.qmd b/frameworks.qmd index 4dc5cbe2..5760dd4d 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -19,7 +19,7 @@ Learning Objectives Machine learning frameworks provide the tools and infrastructure to efficiently build, train, and deploy machine learning models. In this chapter, we will explore the evolution and key capabilities of major -frameworks like [[TensorFlow]{.underline}](https://www.tensorflow.org/), [[PyTorch]{.underline}](https://pytorch.org/), and specialized frameworks for +frameworks like [[TensorFlow (TF)]{.underline}](https://www.tensorflow.org/), [[PyTorch]{.underline}](https://pytorch.org/), and specialized frameworks for embedded devices. We will dive into the components like computational graphs, optimization algorithms, hardware acceleration, and more that enable developers to quickly construct performant models. Understanding @@ -184,64 +184,23 @@ package. ### TF Ecosystem -1. [[TensorFlow - > Core]{.underline}](https://www.tensorflow.org/tutorials): primary - > package that most developers engage with. It provides a - > comprehensive, flexible platform for defining, training, and - > deploying machine learning models. It includes tf.keras as its - > high-level API. - -2. [[TensorFlow Lite]{.underline}](https://www.tensorflow.org/lite): - > designed for deploying lightweight models on mobile, embedded, and - > edge devices. It offers tools to convert TensorFlow models to a - > more compact format suitable for limited-resource devices and - > provides optimized pre-trained models for mobile. - -3. [[TensorFlow.js]{.underline}](https://www.tensorflow.org/js): - > JavaScript library that allows training and deployment of machine - > learning models directly in the browser or on Node.js. It also - > provides tools for porting pre-trained TensorFlow models to the - > browser-friendly format. - -4. [[TensorFlow on Edge Devices - > (Coral)]{.underline}](https://developers.googleblog.com/2019/03/introducing-coral-our-platform-for.html): - > platform of hardware components and software tools from Google - > that allows the execution of TensorFlow models on edge devices, - > leveraging Edge TPUs for acceleration. - -5. [[TensorFlow Federated - > (TFF)]{.underline}](https://www.tensorflow.org/federated): - > framework for machine learning and other computations on - > decentralized data. TFF facilitates federated learning, allowing - > model training across many devices without centralizing the data. - -6. [[TensorFlow - > Graphics]{.underline}](https://www.tensorflow.org/graphics): - > library for using TensorFlow to carry out graphics-related tasks, - > including 3D shapes and point clouds processing, using deep - > learning. - -7. [[TensorFlow Hub]{.underline}](https://www.tensorflow.org/hub): - > repository of reusable machine learning model components to allow - > developers to reuse pre-trained model components, facilitating - > transfer learning and model composition - -8. [[TensorFlow - > Serving]{.underline}](https://www.tensorflow.org/tfx/guide/serving): - > framework designed for serving and deploying machine learning - > models for inference in production environments. It provides tools - > for versioning and dynamically updating deployed models without - > service interruption. - -9. [[TensorFlow Extended - > (TFX)]{.underline}](https://www.tensorflow.org/tfx): end-to-end - > platform designed to deploy and manage machine learning pipelines - > in production settings. TFX encompasses components for data - > validation, preprocessing, model training, validation, and - > serving. - -Source: -[[https://www.tensorflow.org/learn]{.underline}](https://www.tensorflow.org/learn) +1. [[TensorFlow Core]{.underline}](https://www.tensorflow.org/tutorials): primary package that most developers engage with. It provides a comprehensive, flexible platform for defining, training, and deploying machine learning models. It includes tf.keras as its high-level API. + +2. [[TensorFlow Lite]{.underline}](https://www.tensorflow.org/lite): designed for deploying lightweight models on mobile, embedded, and edge devices. It offers tools to convert TensorFlow models to a more compact format suitable for limited-resource devices and provides optimized pre-trained models for mobile. + +3. [[TensorFlow.js]{.underline}](https://www.tensorflow.org/js): JavaScript library that allows training and deployment of machine learning models directly in the browser or on Node.js. It also provides tools for porting pre-trained TensorFlow models to the browser-friendly format. + +4. [[TensorFlow on Edge Devices (Coral)]{.underline}](https://developers.googleblog.com/2019/03/introducing-coral-our-platform-for.html): platform of hardware components and software tools from Google that allows the execution of TensorFlow models on edge devices, leveraging Edge TPUs for acceleration. + +5. [[TensorFlow Federated (TFF)]{.underline}](https://www.tensorflow.org/federated): framework for machine learning and other computations on decentralized data. TFF facilitates federated learning, allowing model training across many devices without centralizing the data. + +6. [[TensorFlow Graphics]{.underline}](https://www.tensorflow.org/graphics): library for using TensorFlow to carry out graphics-related tasks, including 3D shapes and point clouds processing, using deep learning. + +7. [[TensorFlow Hub]{.underline}](https://www.tensorflow.org/hub): repository of reusable machine learning model components to allow developers to reuse pre-trained model components, facilitating transfer learning and model composition + +8. [[TensorFlow Serving]{.underline}](https://www.tensorflow.org/tfx/guide/serving): framework designed for serving and deploying machine learning models for inference in production environments. It provides tools for versioning and dynamically updating deployed models without service interruption. + +9. [[TensorFlow Extended (TFX)]{.underline}](https://www.tensorflow.org/tfx): end-to-end platform designed to deploy and manage machine learning pipelines in production settings. TFX encompasses components for data validation, preprocessing, model training, validation, and serving. TensorFlow was developed to address the limitations of DistBelief[^2]---the framework in use at Google from 2011 to 2015---by providing flexibility @@ -249,7 +208,7 @@ along three axes: 1) defining new layers, 2) refining training algorithms, and 3) defining new training algorithms. To understand what limitations in DistBelief led to the development of TensorFlow, we will first give a brief overview of the Parameter Server Architecture that -DistBelief employed[^3]. +DistBelief employed.[^3] The Parameter Server (PS) architecture is a popular design for distributing the training of machine learning models, especially deep @@ -266,7 +225,7 @@ maintain and manage this state across the training process. **Computation**: The worker processes, which could be run in parallel, were stateless and purely computational, processing data and computing -gradients without maintaining any state or long-term memory[^4]. +gradients without maintaining any state or long-term memory.[^4] DistBelief and its architecture defined above were crucial in enabling distributed deep learning at Google but also introduced limitations that From a7f5a186d421dd1166f88893fea72c2eb24eb1a7 Mon Sep 17 00:00:00 2001 From: sophiacho1 <67521139+sophiacho1@users.noreply.github.com> Date: Mon, 23 Oct 2023 21:10:08 -0400 Subject: [PATCH 18/22] Citation 1 --- frameworks.qmd | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/frameworks.qmd b/frameworks.qmd index 5760dd4d..f6595430 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -1727,7 +1727,7 @@ will guide developers towards picking the ideal framework for their embedded ML application. [^1]: [[https://tvm.apache.org/2021/12/15/tvm-unity]{.underline}](https://tvm.apache.org/2021/12/15/tvm-unity) -[^2]: [[https://arxiv.org/pdf/1603.04467.pdf]{.underline}](https://arxiv.org/pdf/1603.04467.pdf) +[^2]: Abadi et al. [[https://arxiv.org/pdf/1603.04467.pdf]{.underline}](https://arxiv.org/pdf/1603.04467.pdf) [^3]: [[https://storage.googleapis.com/pub-tools-public-publication-data/pdf/40565.pdf]{.underline}](https://storage.googleapis.com/pub-tools-public-publication-data/pdf/40565.pdf) [^4]: [[https://proceedings.neurips.cc/paper_files/paper/2014/file/1ff1de774005f8da13f42943881c655f-Paper.pdf ]{.underline}](https://proceedings.neurips.cc/paper_files/paper/2014/file/1ff1de774005f8da13f42943881c655f-Paper.pdf From f1b7c57dc81f3c446f0a6c68d5c093c2da415184 Mon Sep 17 00:00:00 2001 From: sophiacho1 <67521139+sophiacho1@users.noreply.github.com> Date: Mon, 23 Oct 2023 21:16:07 -0400 Subject: [PATCH 19/22] Citation 2 done --- frameworks.qmd | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/frameworks.qmd b/frameworks.qmd index f6595430..f66855de 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -1727,7 +1727,7 @@ will guide developers towards picking the ideal framework for their embedded ML application. [^1]: [[https://tvm.apache.org/2021/12/15/tvm-unity]{.underline}](https://tvm.apache.org/2021/12/15/tvm-unity) -[^2]: Abadi et al. [[https://arxiv.org/pdf/1603.04467.pdf]{.underline}](https://arxiv.org/pdf/1603.04467.pdf) +[^2]: Abadi et al. 2015. "TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems." [[https://arxiv.org/pdf/1603.04467.pdf]{.underline}](https://arxiv.org/pdf/1603.04467.pdf). [^3]: [[https://storage.googleapis.com/pub-tools-public-publication-data/pdf/40565.pdf]{.underline}](https://storage.googleapis.com/pub-tools-public-publication-data/pdf/40565.pdf) [^4]: [[https://proceedings.neurips.cc/paper_files/paper/2014/file/1ff1de774005f8da13f42943881c655f-Paper.pdf ]{.underline}](https://proceedings.neurips.cc/paper_files/paper/2014/file/1ff1de774005f8da13f42943881c655f-Paper.pdf From f589cf32e7d6faf0acf745bc93a1dfcdc465ecb9 Mon Sep 17 00:00:00 2001 From: sophiacho1 <67521139+sophiacho1@users.noreply.github.com> Date: Mon, 23 Oct 2023 21:21:03 -0400 Subject: [PATCH 20/22] Citation 3 done --- frameworks.qmd | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/frameworks.qmd b/frameworks.qmd index f66855de..70836979 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -1728,7 +1728,7 @@ embedded ML application. [^1]: [[https://tvm.apache.org/2021/12/15/tvm-unity]{.underline}](https://tvm.apache.org/2021/12/15/tvm-unity) [^2]: Abadi et al. 2015. "TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems." [[https://arxiv.org/pdf/1603.04467.pdf]{.underline}](https://arxiv.org/pdf/1603.04467.pdf). -[^3]: [[https://storage.googleapis.com/pub-tools-public-publication-data/pdf/40565.pdf]{.underline}](https://storage.googleapis.com/pub-tools-public-publication-data/pdf/40565.pdf) +[^3]: Dean et al. 2012. "Large Scale Distributed Deep Networks." *Proceedings of the 25th International Conference on Neural Information Processing Systems* 1: 1223–1231. [[https://storage.googleapis.com/pub-tools-public-publication-data/pdf/40565.pdf]{.underline}](https://storage.googleapis.com/pub-tools-public-publication-data/pdf/40565.pdf). [^4]: [[https://proceedings.neurips.cc/paper_files/paper/2014/file/1ff1de774005f8da13f42943881c655f-Paper.pdf ]{.underline}](https://proceedings.neurips.cc/paper_files/paper/2014/file/1ff1de774005f8da13f42943881c655f-Paper.pdf ) From 549f8190638022fecd306fde230bde0ee6eb6429 Mon Sep 17 00:00:00 2001 From: sophiacho1 <67521139+sophiacho1@users.noreply.github.com> Date: Mon, 23 Oct 2023 21:26:22 -0400 Subject: [PATCH 21/22] Citation 4 done --- frameworks.qmd | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/frameworks.qmd b/frameworks.qmd index 70836979..fe7ff9ef 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -1729,9 +1729,9 @@ embedded ML application. [^1]: [[https://tvm.apache.org/2021/12/15/tvm-unity]{.underline}](https://tvm.apache.org/2021/12/15/tvm-unity) [^2]: Abadi et al. 2015. "TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems." [[https://arxiv.org/pdf/1603.04467.pdf]{.underline}](https://arxiv.org/pdf/1603.04467.pdf). [^3]: Dean et al. 2012. "Large Scale Distributed Deep Networks." *Proceedings of the 25th International Conference on Neural Information Processing Systems* 1: 1223–1231. [[https://storage.googleapis.com/pub-tools-public-publication-data/pdf/40565.pdf]{.underline}](https://storage.googleapis.com/pub-tools-public-publication-data/pdf/40565.pdf). -[^4]: [[https://proceedings.neurips.cc/paper_files/paper/2014/file/1ff1de774005f8da13f42943881c655f-Paper.pdf +[^4]: Li et al. 2014. "Communication Efficient Distributed Machine Learning with the Parameter Server." *Proceedings of the 27th International Conference on Neural Information Processing Systems* 1: 19–27. [[https://proceedings.neurips.cc/paper_files/paper/2014/file/1ff1de774005f8da13f42943881c655f-Paper.pdf ]{.underline}](https://proceedings.neurips.cc/paper_files/paper/2014/file/1ff1de774005f8da13f42943881c655f-Paper.pdf -) +). [^5]: [[TensorFlow: Large-scale machine learning on heterogeneous systems, 2015.]{.underline}](https://www.tensorflow.org/datasets/catalog/mnist) From f4ce7e49ed0569d3d7c7c577c440d04dcbfa0bd8 Mon Sep 17 00:00:00 2001 From: sophiacho1 <67521139+sophiacho1@users.noreply.github.com> Date: Mon, 23 Oct 2023 21:33:44 -0400 Subject: [PATCH 22/22] Citation 1 done --- frameworks.qmd | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/frameworks.qmd b/frameworks.qmd index fe7ff9ef..1c64f52b 100644 --- a/frameworks.qmd +++ b/frameworks.qmd @@ -1726,7 +1726,7 @@ intended models, use cases, and evaluating options against key metrics will guide developers towards picking the ideal framework for their embedded ML application. -[^1]: [[https://tvm.apache.org/2021/12/15/tvm-unity]{.underline}](https://tvm.apache.org/2021/12/15/tvm-unity) +[^1]: Sampson et al. 2021. "Apache TVM Unity: a vision for the ML software & hardware ecosystem in 2022." [[https://tvm.apache.org/2021/12/15/tvm-unity]{.underline}](https://tvm.apache.org/2021/12/15/tvm-unity). [^2]: Abadi et al. 2015. "TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems." [[https://arxiv.org/pdf/1603.04467.pdf]{.underline}](https://arxiv.org/pdf/1603.04467.pdf). [^3]: Dean et al. 2012. "Large Scale Distributed Deep Networks." *Proceedings of the 25th International Conference on Neural Information Processing Systems* 1: 1223–1231. [[https://storage.googleapis.com/pub-tools-public-publication-data/pdf/40565.pdf]{.underline}](https://storage.googleapis.com/pub-tools-public-publication-data/pdf/40565.pdf). [^4]: Li et al. 2014. "Communication Efficient Distributed Machine Learning with the Parameter Server." *Proceedings of the 27th International Conference on Neural Information Processing Systems* 1: 19–27. [[https://proceedings.neurips.cc/paper_files/paper/2014/file/1ff1de774005f8da13f42943881c655f-Paper.pdf @@ -1736,5 +1736,3 @@ embedded ML application. 2015.]{.underline}](https://www.tensorflow.org/datasets/catalog/mnist) [^6]: [[Patrick McClanahan, Introduction to Operating Systems, 2023]{.underline}](https://eng.libretexts.org/Courses/Delta_College/Introduction_to_Operating_Systems/03%3A_The_Operating_System/3.06%3A_Types_of_Operating_Systems) - -