WSciML SIAM Workshop 2026

Welcome to the repository accompanying the 2026 SIAM workshop on Weak-form Scientific Machine Learning (WSciML)!

Initialization

Start by running the following to allow git to properly initialize submodules:

git clone --recurse-submodules https://github.com/MathBioCU/WSciML-SIAMWorkshop2026.git

Running example scripts

1. Scripts to run WSINDy and WENDy algorithms on test sets are found in the python_scripts and matlab_scripts folders.
2. Corresponding source files are found in the libs folder.
3. To run the python scripts, it is recommended to create a conda env using conda env create -f environment.yml
4. If you would prefer to use pip, starting from a python>=3.14 environment run pip install -r requirements.txt

Additional data

1. Several PDE datasets are located in libs/wsindy_obj_data/pde_data
2. A collection of other PDE datasets is found at https://zenodo.org/records/20787783. This can be obtained from terminal in python by

pip install zenodo-get

zenodo_get 20787783

3. It is recommended that workshop attendees download the data from Zenodo, can change script paths accordingly
4. This is necessary to utilize larger datasets, such as the Navier-Stokes flow passed a cylinder data (Nav_Stokes.mat)

Discord channel

https://discord.gg/VxYtxwx8Q

Further communication

David Bortz: [email protected]
Dan Messenger: [email protected]