Robusta-HMF

jax implementation of robust heteroskedastic matrix factorisation. Robusta like the coffee bean, get it?

Installation

Easiest is from PyPI either with pip

pip install robusta-hmf

or uv (recommended)

uv add robusta-hmf

Or, you can clone and build from source

git clone git@github.com:TomHilder/robusta-hmf.git
cd robusta-hmf
pip install -e .

TODO

TODO

TODO

More interpretable loss, maybe normalised in some sensible way (maybe proper NLL)
Type checking with mypy