Merge pull request #36 from TomF98/main

Fix small Bug and add literature information

.gitignore

@@ -16,7 +16,9 @@ __pycache__/*
 
 # Log folders
 **/lightning_logs/**
-
+**/fluid_logs/**
+**/Multiscale/**
+**/bosch/**
 # Project files
 .ropeproject
 .project

README.rst

@@ -71,9 +71,9 @@ to have a look at the following sections:
 - `Examples: Different applications with detailed explanations`_
 - Documentation_
 
-.. _`Tutorial: Understanding the structure of TorchPhysics`: https://torchphysics.readthedocs.io/en/latest/tutorial/tutorial_start.html
+.. _`Tutorial: Understanding the structure of TorchPhysics`: https://boschresearch.github.io/torchphysics/tutorial/tutorial_start.html
 .. _`Examples: Different applications with detailed explanations`: https://github.com/boschresearch/torchphysics/tree/main/examples
-.. _Documentation: https://torchphysics.readthedocs.io/en/latest/api/modules.html
+.. _Documentation: https://boschresearch.github.io/torchphysics/index.html
 
 
 Installation
@@ -133,6 +133,23 @@ since we aim to develop a reliable code basis and extend the library to include
 
 .. _GitHub: https://github.com/boschresearch/torchphysics
 
+Cite TorchPhysics
+=================
+If TorchPhysics has been helpful for your research, please cite:
+
+.. code-block:: latex
+
+  @article{TorchPhysics,
+      author = {Derick Nganyu Tanyu and Jianfeng Ning and Tom Freudenberg and Nick Heilenkötter and Andreas Rademacher and Uwe Iben and Peter Maass},
+      title = {Deep learning methods for partial differential equations and related parameter identification problems},
+      journal = {Inverse Problems},
+      doi = {10.1088/1361-6420/ace9d4},
+      year = {2023},
+      publisher = {IOP Publishing},
+      volume = {39},
+      number = {10},
+      pages = {103001}}
+
 License
 =======
 TorchPhysics uses an Apache License, see the LICENSE_ file.

docs/tutorial/applied_tutorial_start.rst

@@ -42,4 +42,4 @@ More applications can be found on the `example page`_.
 
 .. _`PINNs for Poisson`: https://github.com/boschresearch/torchphysics/blob/main/examples/tutorial/solve_pde.ipynb
 .. _`DRM for Poisson`: https://github.com/boschresearch/torchphysics/blob/main/examples/tutorial/solve_pde_drm.ipynb
-.. _`example page`: https://torchphysics.readthedocs.io/en/latest/examples.html
+.. _`example page`: https://boschresearch.github.io/torchphysics/examples.html

docs/tutorial/condition_tutorial.rst

@@ -8,7 +8,7 @@ Many kinds of conditions are pre implemented, they can be found in
 the docs_. Depending on the used type, the condition get different inputs. The five arguments 
 that almost all conditions need, are:
 
-.. _docs: https://torchphysics.readthedocs.io/en/latest/api/torchphysics.problem.conditions.html
+.. _docs: https://boschresearch.github.io/torchphysics/api/torchphysics.problem.conditions.html
 
 module
     Your model that should be trained, with the condition you are creating. The definition

docs/tutorial/differentialoperators.rst

@@ -27,7 +27,7 @@ entry is specified. E.g. :math:`u: \mathbb{R}^3 \to \mathbb{R}^3` then do
 ``laplacian`` (:math:`u[:, 0], x`), to get the laplacian of the first entry.
 The newest version of all implemented operators can be found under the docs_.
 
-.. _docs: https://torchphysics.readthedocs.io/en/latest/api/torchphysics.utils.html
+.. _docs: https://boschresearch.github.io/torchphysics/api/torchphysics.utils.html
 
 Since ``autograd`` is used, the differential operators can work with neural networks and functions
 that use PyTorch-Tensors. It follow a short example of the usage:

docs/tutorial/model_creation.rst

@@ -6,7 +6,7 @@ What basics network structures are implemented, can be found under the `model-do
 All networks can found under ``torchphysics.models``. For example, a simple fully 
 connected neural network can be created with:
 
-.. _`model-docs`: https://torchphysics.readthedocs.io/en/latest/api/torchphysics.models.html
+.. _`model-docs`: https://boschresearch.github.io/torchphysics/api/torchphysics.models.html
 
 .. code-block:: python
 

docs/tutorial/sampler_tutorial.rst

@@ -13,7 +13,7 @@ are implemented, for example:
 
 And many more, see `sampler-docs`_. The samplers lay all under ``torchphysics.samplers``.
 
-.. _`sampler-docs`: https://torchphysics.readthedocs.io/en/latest/api/torchphysics.problem.samplers.html
+.. _`sampler-docs`: https://boschresearch.github.io/torchphysics/api/torchphysics.problem.samplers.html
 
 In general, a sampler gets the following arguments:
 

docs/tutorial/solve_pde.rst

@@ -209,4 +209,4 @@ be found under the `example-folder`_.
 And more in-depth information can be found on the `tutorial page`_.
 
 .. _`example-folder`: https://github.com/boschresearch/torchphysics/tree/main/examples
-.. _`tutorial page`: https://torchphysics.readthedocs.io/en/latest/tutorial/tutorial_start.html
+.. _`tutorial page`: https://boschresearch.github.io/torchphysics/tutorial/tutorial_start.html

docs/tutorial/tutorial_domain_basics.rst

@@ -16,7 +16,7 @@ For explicit documentation of each method and all implemented domains, see the d
 All domains can be found under ``torchphysics.domains``.
 Some pre implemented domains are: ``Point, Interval, Parallelogram, Circle...``
 
-.. _docs: https://torchphysics.readthedocs.io/en/latest/api/torchphysics.problem.domains.html
+.. _docs: https://boschresearch.github.io/torchphysics/api/torchphysics.problem.domains.html
 
 Here we focus now on the creation of different domains. To create a domain,
 we need a space to which the domain belongs. 

docs/tutorial/tutorial_start.rst

@@ -62,7 +62,7 @@ Utils
 .. _`PointSampler tutorial`: sampler_tutorial.html
 .. _`model creation tutorial`: model_creation.html
 .. _`condition tutorial`: condition_tutorial.html
-.. _docs: https://torchphysics.readthedocs.io/en/latest/api/torchphysics.utils.html
+.. _docs: https://boschresearch.github.io/torchphysics/api/torchphysics.utils.html
 .. _`differential operators`: differentialoperators.html
 .. _`Creating plots`: plotting.html
 .. _`solving a simple PDE`: solve_pde.html

examples/tutorial/solve_pde_drm.ipynb

@@ -6,7 +6,7 @@
    "source": [
     "Solving a simple PDE with the DeepRitz-approach\n",
     "===========================================\n",
-    "Here we give a beginner-friendly introduction to TorchPhysics, going over all the basic concepts and steps. For a more in-depth explanation, we recommend the [tutorial page](https://torchphysics.readthedocs.io/en/latest/tutorial/tutorial_start.html).\n",
+    "Here we give a beginner-friendly introduction to TorchPhysics, going over all the basic concepts and steps. For a more in-depth explanation, we recommend the [tutorial page](https://boschresearch.github.io/torchphysics/tutorial/tutorial_start.html).\n",
     "We introduce the library with the aim to solve the following PDE:\n",
     "\n",
     "\\begin{align*}\n",

examples/workshop/Exercise3_2.ipynb

@@ -209,7 +209,7 @@
    "metadata": {},
    "source": [
     "##### c) Construct the DeepONet\n",
-    "Build a DeepONet consisting of a fully connected TrunkNet and BranchNet. Check the `TorchPhysics` [documentation](https://torchphysics.readthedocs.io/en/latest/api/torchphysics.models.deeponet.html), to see how this has to implemented. \n",
+    "Build a DeepONet consisting of a fully connected TrunkNet and BranchNet. Check the `TorchPhysics` [documentation](https://boschresearch.github.io/torchphysics/tutorial/tutorial_start.html), to see how this has to implemented. \n",
     "\n",
     "Both (TrunkNet and BranchNet) should have 3 hidden layers with 25 neurons each. The output of both networks should be 30 neurons."
    ]

examples/workshop/Sol3_2.ipynb

@@ -217,7 +217,7 @@
    "metadata": {},
    "source": [
     "##### c) Construct the DeepONet\n",
-    "Build a DeepONet consisting of a fully connected TrunkNet and BranchNet. Check the `TorchPhysics` [documentation](https://torchphysics.readthedocs.io/en/latest/api/torchphysics.models.deeponet.html), to see how this has to implemented. \n",
+    "Build a DeepONet consisting of a fully connected TrunkNet and BranchNet. Check the `TorchPhysics` [documentation](https://boschresearch.github.io/torchphysics/tutorial/tutorial_start.html), to see how this has to implemented. \n",
     "\n",
     "Both (TrunkNet and BranchNet) should have 3 hidden layers with 25 neurons each. The output of both networks should be 30 neurons."
    ]

setup.cfg

@@ -13,7 +13,7 @@ long_description_content_type = text/x-rst; charset=UTF-8
 url = https://github.com/boschresearch/torchphysics
 # Add here related links, for example:
 project_urls =
-    Documentation = https://torchphysics.readthedocs.io/en/latest/
+    Documentation = https://boschresearch.github.io/torchphysics/index.html
     Source = https://github.com/boschresearch/torchphysics
     Changelog = https://github.com/boschresearch/torchphysics/blob/main/CHANGELOG.rst
 #    Tracker = https://github.com/pyscaffold/pyscaffold/issues

src/torchphysics/models/deeponet/branchnets.py

@@ -27,7 +27,7 @@ def __init__(self, function_space, discretization_sampler):
         super().__init__(function_space, output_space=None)
         self.output_neurons = 0
         self.discretization_sampler = discretization_sampler
-        self.input_dim = len(self.discretization_sampler)
+        self.input_dim = len(self.discretization_sampler) * function_space.output_space.dim
         self.current_out = torch.empty(0)
 
     def finalize(self, output_space, output_neurons):

src/torchphysics/problem/conditions/__init__.py

@@ -4,7 +4,7 @@
 
 A tutorial on the usage of Conditions can be found here_.
 
-.. _here: https://torchphysics.readthedocs.io/en/latest/tutorial/condition_tutorial.html
+.. _here: https://boschresearch.github.io/torchphysics/tutorial/tutorial_start.html
 """
 
 from .condition import (Condition,