Metadata-Version: 2.1
Name: fastvpinns
Version: 1.0.2
Summary: A fast tensor-driven variational physics-informed neural network library for solving PDEs.
Author-email: Thivin Anandh <thivinanandh@gmail.com>, Divij Ghose <divijghose@gmail.com>, Sashikumaar Ganesan <sashi@iisc.ac.in>
Project-URL: Homepage, https://github.com/cmgcds/fastvpinns
Project-URL: Issues, https://github.com/cmgcds/fastvpinns/issues
Project-URL: Documentation, https://cmgcds.github.io/fastvpinns/
Project-URL: logo, https://raw.githubusercontent.com/cmgcds/fastvpinns/main/Fastvpinns_logo.png
Keywords: PDE,Neural Networks,Physics,AI,Machine Learning,variational PINNs,VPINNs,PINNs,Physics-informed neural networks
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Education
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3 :: Only
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Scientific/Engineering
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Topic :: Scientific/Engineering :: Mathematics
Classifier: Topic :: Software Development
Classifier: Topic :: Software Development :: Libraries
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Requires-Python: >=3.8
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: numpy<=1.23.5,>=1.21
Requires-Dist: pandas<1.4.4
Requires-Dist: matplotlib
Requires-Dist: tensorflow<=2.13.0,>=2.9.1
Requires-Dist: pytest==8.0.2
Requires-Dist: meshio==5.3.4
Requires-Dist: pyDOE==0.3.8
Requires-Dist: tqdm==4.66.2
Requires-Dist: scipy
Requires-Dist: gmsh==4.11.0
Requires-Dist: pyyaml
Requires-Dist: rich==12.5.1
Requires-Dist: black
Requires-Dist: pre-commit


[![Unit tests](https://github.com/cmgcds/fastvpinns/actions/workflows/unit-tests.yml/badge.svg)](https://github.com/cmgcds/fastvpinns/actions/workflows/unit-tests.yml)
[![Integration tests](https://github.com/cmgcds/fastvpinns/actions/workflows/integration-tests.yml/badge.svg)](https://github.com/cmgcds/fastvpinns/actions/workflows/integration-tests.yml)
[![Compatability check](https://github.com/cmgcds/fastvpinns/actions/workflows/compatibility-tests.yml/badge.svg)](https://github.com/cmgcds/fastvpinns/actions/workflows/compatibility-tests.yml)
[![codecov](https://codecov.io/gh/cmgcds/fastvpinns/graph/badge.svg?token=NI9G37R2Q7)](https://codecov.io/gh/cmgcds/fastvpinns)
[![PyPI](https://badge.fury.io/py/fastvpinns.svg)](https://badge.fury.io/py/fastvpinns)

[![MIT License](https://img.shields.io/badge/License-MIT-blue.svg)](https://opensource.org/licenses/MIT)
[![Code style: black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/psf/black)
![Python Versions](https://img.shields.io/badge/python-3.8%20|%203.9%20|%203.10%20|%203.11-blue)
[![status](https://joss.theoj.org/papers/173453d08813f66a718f583502351747/status.svg)](https://joss.theoj.org/papers/173453d08813f66a718f583502351747)


<br />
<div align="center">
  <a href="https://github.com/cmgcds/fastvpinns">
    <img alt="FastVPINNs logo" src="https://raw.githubusercontent.com/cmgcds/fastvpinns/main/Fastvpinns_logo.png" width="500">
  </a>

<h3 align="center">Tensor-driven accelerated framework for hp-variational pinns</h3>

  <p align="center">
    <br />
    <a href="https://cmgcds.github.io/fastvpinns"><strong>Link to Documentation 📚</strong></a>
    <br />

  </p>
</div>

A robust tensor-based deep learning framework for solving partial differential equations using hp-Variational Physics-Informed Neural Networks (hp-VPINNs). The framework is based on the methodology presented in the [FastVPINNs Paper](https://arxiv.org/abs/2404.12063).


*This library is a highly optimised version of the the initial implementation of hp-VPINNs by [Kharazmi et al.](https://github.com/ehsankharazmi/hp-VPINNs). Refer the [hp-VPINNs Paper](https://arxiv.org/abs/2003.05385).*

## Authors 👨‍💻
---

[Thivin Anandh](https://github.com/thivinanandh), [Divij Ghose](https://divijghose.github.io/), [Sashikumaar Ganesan](https://cds.iisc.ac.in/faculty/sashi)

STARS Lab, Department of Computational and Data Sciences, Indian Institute of Science, Bangalore, India

## Installation 🛠️
---

The build of the code is currently tested on Python versions (3.8, 3.9, 3.10, 3.11), on OS Ubuntu 20.04 and Ubuntu 22.04, MacOS-latest and Windows-latest (refer compatibility build [Compatability check](https://github.com/cmgcds/fastvpinns/actions/workflows/compatibility-tests.yml)).

You can install the package using pip as follows:

```bash
pip install fastvpinns
```

 On ubuntu systems with libGL issues caused due to matplotlib or gmsh, please run the following command to install the required dependencies:
```bash
sudo apt-get install -y libglu1-mesa 
```

For more information on the installation process, please refer to our documentation [here](https://cmgcds.github.io/fastvpinns/).

## Citing 📜
---

If you use this code in your research, please consider citing the following paper:

```bibtex
@misc{anandh2024fastvpinns,
      title={FastVPINNs: Tensor-Driven Acceleration
             of VPINNs for Complex Geometries}, 
      author={Thivin Anandh, Divij Ghose, Himanshu Jain
               and Sashikumaar Ganesan},
      year={2024},
      eprint={2404.12063},
      archivePrefix={arXiv},
      primaryClass={cs.LG}
}
```

## Usage 🚀
---

For detailed usage, please refer to our documentation [here](https://cmgcds.github.io/fastvpinns/).

The package provides a simple API to train and solve PDE using VPINNs. The following code snippet demonstrates how to train a hp-VPINN model for the 2D Poisson equation for a structured grid. We could observe that we can solve a PDE using fastvpinns using 15 lines of code.

```python
#load the geometry 
domain = Geometry_2D("quadrilateral", "internal", 100, 100, "./")
cells, boundary_points = domain.generate_quad_mesh_internal(x_limits=[0, 1],y_limits=[0, 1],n_cells_x=4, n_cells_y=4, num_boundary_points=400)

# load the FEspace
fespace = Fespace2D(domain.mesh,cells,boundary_points,domain.mesh_type,fe_order=5,fe_type="jacobi",quad_order=5,quad_type="legendre", fe_transformation_type="bilinear",bound_function_dict=bound_function_dict,bound_condition_dict=bound_condition_dict,
forcing_function=rhs,output_path=i_output_path,generate_mesh_plot=True)

# Instantiate Data handler 
datahandler = DataHandler2D(fespace, domain, dtype=tf.float32)

# Instantiate the model with the loss function for the model 
model = DenseModel(layer_dims=[2, 30, 30, 30, 1],learning_rate_dict=0.01,params_dict=params_dict,
        loss_function=pde_loss_poisson,  ## Loss function of poisson2D
        input_tensors_list=[in_tensor, dir_in, dir_out],
        orig_factor_matrices=[datahandler.shape_val_mat_list,datahandler.grad_x_mat_list, datahandler.grad_y_mat_list],
        force_function_list=datahandler.forcing_function_list, tensor_dtype=tf.float32,
        use_attention=i_use_attention, ## Archived (not in use)
        activation=i_activation,
        hessian=False)

# Train the model
for epoch in range(1000):
    model.train_step()
```

Note : Supporting functions which define the actual solution and boundary conditions have to be passed to the main code.

## Contributing 🤝
---
This code is currently maintained by the authors as mentioned in the section above. We welcome contributions from the community. Please refer to the [documentation](https://cmgcds.github.io/fastvpinns/) for guidelines on contributing to the project.

## License 📑
---

This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details. 
