https://img.shields.io/pypi/v/bayes_window.svg https://github.com/mmyros/bayes-window/actions/workflows/pytest.yaml/badge.svg Documentation Status https://codecov.io/gh/mmyros/bayes-window/branch/master/graph/badge.svg?token=CQMHJRNC9I Language grade: Python

Bayes Window

bw_logo_drawing.png

Pretty and easy hierarchical Bayesian or linear mixed effects estimation with data overlay

TODO

  • Replace dummified self.posterior axes with original without filling in data

  • Fit checks for LME under plot_diagnostics()

  • Posterior predictive to mimic data plot

  • Shade slope HDI

  • Less haphazard testing

  • Shrinkage layer: Should be as easy as alt.layer with unpooled model

    • Make sure + works well with facet

  • Prior predictive and posterior predictive plots

  • wait for resolution to https://github.com/vega/vega-lite/issues/4373#issuecomment-447726094

  • delete facet attribute of chart if independent_axes=True - how? Ask in altair github

  • Random-effect plots (eg intercepts): bw.plot_extras?

    • see lfp notebook

  • Decide on Vega theme

INSTALLATION

For gpu: `pip install --upgrade "jax[cuda102]" -f https://storage.googleapis.com/jax-releases/jax_releases.html` or cudaXXX from nvcc –version

Features

See extensive examples https://bayes-window.readthedocs.io/en/latest/index.html