Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/leiterrl/porch
A PyTorch library for Physics-Informed Neural Networks (PINNs)
https://github.com/leiterrl/porch
gpu machine-learning neural-networks physics-informed-neural-networks python pytorch
Last synced: 3 months ago
JSON representation
A PyTorch library for Physics-Informed Neural Networks (PINNs)
- Host: GitHub
- URL: https://github.com/leiterrl/porch
- Owner: leiterrl
- License: mpl-2.0
- Created: 2021-06-07T09:16:17.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2024-04-26T10:43:36.000Z (9 months ago)
- Last Synced: 2024-10-04T21:55:09.327Z (4 months ago)
- Topics: gpu, machine-learning, neural-networks, physics-informed-neural-networks, python, pytorch
- Language: Python
- Homepage:
- Size: 295 KB
- Stars: 9
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Porch
Porch is a Python library that enables users to train Physics-Informed Neural Networks (PINNs) using the PyTorch framework. PINNs combine the power of neural networks with the governing laws of physics, allowing for the solution of complex physical problems with limited or noisy data.
## Features
- Flexible and intuitive API for defining and training Physics-Informed Neural Networks.
- Verbose definition of a PINN [BaseModel](porch/model.py). No details are hidden from the user.
- Built-in utilities for data preprocessing, visualization, and evaluation of trained models.
- Utility functionality for generating geometry and dealing with boundary and initial conditions.
## Installation
Porch requires PyTorch to be installed as well. Please refer to the official [PyTorch website](https://pytorch.org/get-started/locally/) for installation instructions specific to your platform. This avoids pip installing the CPU-only version of PyTorch on a system with a GPU.
To install Porch, you can use pip locally:
```shell
pip install .
```
## Usage
For a detailed example of how to use Porch, please refer to the [experiments](experiments) directory.
A good place to start is the [Burgers' equation](experiments/burgers.py) example, which demonstrates how to use Porch to solve a simple non-linear PDE.
## License
Porch is released under the Mozilla Public License 2.0 (MPL 2.0). See the [LICENSE](LICENSE) file for more details.
## Contact
If you have any questions, suggestions, or comments, please feel free open an issue here on GitHub.