pytorch-deep-learning

This repository provides a comprehensive guide and exercises on the practical aspects of deep learning. We will start with learning PyTorch, basic tensor operations, backpropagation, and then developing various neural network models. It is designed for beginners to understand the basic concepts clearly.

1. 00_tensor_fundamentals.ipynb - This repository provides a comprehensive guide and exercises on basic tensor operations in PyTorch. It is designed for beginners to understand and practice various tensor manipulations essential for deep learning and machine learning tasks.
2. 00_tensor_exercises.ipynb - After completing 00_tensor_fundamentals.ipynb, you can try out the simple exercises provided in this notebook to test your understanding.
3. 01_simple_neural_network_iris.ipynb - This notebook introduces your first neural network model. Using the Iris dataset, it demonstrates how to build a neural network to classify different flower species.
4. 01_overfitting_underfitting.ipynb - This notebook explains the concepts of overfitting and underfitting, common issues encountered when training machine learning and deep learning models.
5. 02_fully_connected_mnist.ipynb - In this notebook, we will walk through the process of building, training, and evaluating a fully connected neural network for classifying handwritten digits from the MNIST dataset using PyTorch. We'll cover data loading and preprocessing, model definition, training, evaluation, and visualization of the results.
6. 03_optimize_nn_exercises.ipynb - This notebook provides a series of incremental exercises designed to deepen your understanding of various aspects of neural network training and optimization. Each exercise introduces a small modification to the existing model or training process. These modifications will help you explore the effects of different model architectures, activation functions, optimizers, learning rates, regularization techniques, data augmentation, loss functions, and evaluation metrics. By completing these exercises, you will gain hands-on experience in tuning neural networks and improving model performance.
7. 04_cnn_concepts.ipynb - This notebook will cover the fundamental concepts that underpin CNNs, including convolution operations, padding, stride, feature maps, and filters. By understanding these basics, students will be well-equipped to delve into more complex topics and applications of CNNs.
8. 04_cnn_mnist.ipynb - In this notebook, we will walk through the process of building, training, and evaluating a Convolutional Neural Network (CNN) for classifying handwritten digits from the MNIST dataset using PyTorch. We'll cover data loading and preprocessing, model definition, training, evaluation, and visualization of the results.
9. 04_cnn_cifar_10.ipynb - This notebook demonstrates how to build a Convolutional Neural Network (CNN) using PyTorch to classify images from the CIFAR-10 dataset.