Skip to content

Latest commit

 

History

History
20 lines (10 loc) · 466 Bytes

README.md

File metadata and controls

20 lines (10 loc) · 466 Bytes

Variational Wasserstein-Gradient-Flow Filter

Implements a filtering approach with a variational update based on Wasserstein gradient flows

Installation

Create a conda environment

conda create -n NAME python=3.9

Then head to the cloned repository and execute

pip install -e .

Examples

A filtering example on a stochastic volatility model

python examples/wasserstein_filter/markov_stochastic_volatility_wf_sqrt.py