https://github.com/mlysy/pfjax

Particle filtering in JAX
https://github.com/mlysy/pfjax

Last synced: 2 months ago
JSON representation

Particle filtering in JAX

• Host: GitHub
• URL: https://github.com/mlysy/pfjax
• Owner: mlysy
• Created: 2021-11-10T03:32:39.000Z (about 3 years ago)
• Default Branch: main
• Last Pushed: 2024-09-12T15:32:49.000Z (4 months ago)
• Last Synced: 2024-09-13T08:30:58.774Z (4 months ago)
• Language: Python
• Homepage: https://pfjax.readthedocs.io
• Size: 96.6 MB
• Stars: 1
• Watchers: 4
• Forks: 1
• Open Issues: 5
• Metadata Files:
  • Readme: README.md
  • Changelog: NEWS.md

Awesome Lists containing this project

README

        
# **PFJAX**: Particle Filtering in JAX

[**Quickstart**](docs/notebooks/pfjax.md)

| [**Installation**](#installation)

| [**Documentation**](#documentation)

| [**Developers**](#developers)

---

## What is **PFJAX**?

**PFJAX** is a collection of tools for estimating the parameters of state-space models using particle filtering methods, with  [**JAX**](https://jax.readthedocs.io/) as the backend for JIT-compiling models and automatic differentiation.

---

## Installation

This will clone the repo into a subfolder `pfjax`, from where you (i) issue the `git clone` command and (ii) install the package from source.

```bash

git clone https://github.com/mlysy/pfjax

cd pfjax

pip install .

``` 

## Quickstart 

A brief [introduction to **PFJAX**](docs/notebooks/pfjax.md).

## Documentation

This is a work in progress!  Current modules include:

- The [quickstart guide](docs/notebooks/pfjax.md).

- A [comparison of gradient and hessian algorithms](docs/notebooks/gradient_comparisons.md) based on particle filters, which in turn are used for conducting parameter inference.

- An example of parameter inference using [stochastic optimization](docs/notebooks/stochopt_tutorial.md).

- An example of parameter inference using [Markov chain Monte Carlo](docs/notebooks/mcmc_tutorial).

- The API [reference documentation](https://pfjax.readthedocs.io/).

## Developers

### Testing

From within `pfjax/tests`:

```bash

python3 -m unittest -v

```

Or, install [**tox**](https://tox.wiki/en/latest/index.html), then from within `pfjax` at the command line: `tox`.

### Building Documentation

From within `pfjax/docs`:

```bash

# regular build

make html

# clean build incl. repeating cached computations

make clean html

```