Skip to main content

Hidden Markov Models in PyMC3

Project description

Build Status Binder

PyMC3 HMM

Hidden Markov models in PyMC3.

Features

  • Fully implemented PyMC3 Distribution classes for HMM state sequences (DiscreteMarkovChain) and mixtures that are driven by them (SwitchingProcess)
  • A forward-filtering backward-sampling (FFBS) implementation (FFBSStep) that works with NUTS—or any other PyMC3 sampler
  • A conjugate Dirichlet transition matrix sampler (TransMatConjugateStep)
  • Support for time-varying transition matrices in the FFBS sampler and all the relevant Distribution classes

To use these distributions and step methods in your PyMC3 models, simply import them from the pymc3_hmm package.

See the examples directory for demonstrations of the aforementioned features. You can also use Binder to run the examples yourself.

Installation

Currently, the package can be installed via pip directly from GitHub

$ pip install git+https://github.com/AmpersandTV/pymc3-hmm

Development

First, pull in the source from GitHub:

$ git clone git@github.com:AmpersandTV/pymc3-hmm.git

Next, you can run make conda or make venv to set up a virtual environment.

Once your virtual environment is set up, install the project, its dependencies, and the pre-commit hooks:

$ pip install -r requirements.txt
$ pre-commit install --install-hooks

After making changes, be sure to run make black in order to automatically format the code and then make check to run the linters and tests.

License

Apache License, Version 2.0

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

pymc3-hmm-0.1.4.tar.gz (32.8 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page