- Make equation discovery more transparent and illustrative
- Combine power of pytorch, numerical methods and math overall to conquer and solve ALL {O,P}DEs. There are some examples to provide a little insight to an operator form
- Core features
- Installation
- Examples
- Project Structure
- Documentation
- Getting started
- License
- Contacts
- Citation
- Solve ODE initial or boundary-value problems
- Solve PDE initial-boundary value problems
- Use variable models and their differentiation methods
- Faster solution using cache
TEDEouS can be installed with pip
:
$ git clone https://github.com/ITMO-NSS-team/torch_DE_solver.git $ cd torch_DE_solver $ pip install -r requirements.txt
After the TEDEouS is installed the user may refer to various examples that are in examples forlder.
$ cd examples
Every example is designed such that the boxplots of the launches are commented and the preliminary results are not shown, but stored in separate folders.
- Legendre polynomial equation
$ python example_ODE_Legendre.py
or
$ python example_ODE_Legendre_autograd.py
- Panleve transcendents (others are places in 'examples\to_renew' folder due to the architecture change)
$ python example_Painleve_I.py
- Wave equation (non-physical conditions for equation discovery problem)
$ python example_wave_paper_autograd.py
- Wave equation (initial-boundary value problem)
$ python example_wave_physics.py
- Heat equation
$ python example_heat.py
- KdV equation (non-physical conditions for equation discovery problem)
$ python example_KdV.py
- KdV equation (solitary solution with periodic boundary conditions)
$ python example_KdV_periodic.py
- Burgers equation and DeepXDE comparison
$ python WIP
Stable version is located in the master branch.
https://torch-de-solver.readthedocs.io/en/docs/index.html
Schroedinger equation example step-by-step https://torch-de-solver.readthedocs.io/en/docs/tedeous/examples/schrodinger.html
TEDEouS is distributed under BSD-3 licence found in LICENCE file
- Feel free to make issues or contact @SuperSashka directly
- @article{"name",
- title = {}, author = {}, journal = {}, year = {}, issn = {}, doi = {}}