splineLearning

This repository contains PyTorch implementation of the submission: Bayesian Spline Learning for Equation Discovery of Nonlinear Dynamics with Quantified Uncertainty.

0. Environment Setup

enviroment setup: "conda create --name python36 --file requirements.txt" conda activate python36

Cases

It contains two demos cases: Demo/ODE.py and Demo/PDE.py, which coorespond to Lotka Volterra and Burgers'

1. Training

To train the ODE demo, run the following command:

cd/Demo
python ODE.py

To train the PDE demo model on Lung dataset with GraphRNN/DAGG, run the following:

cd/Demo
python PDE.py