Implements a filtering approach with a variational update based on Wasserstein gradient flows
Create a conda environment
conda create -n NAME python=3.9
Then head to the cloned repository and execute
pip install -e .
A filtering example on a stochastic volatility model
python examples/wasserstein_filter/markov_stochastic_volatility_wf_sqrt.py