/libre

Primary LanguagePythonMIT LicenseMIT

LIBRE -- Learning Interpretable Biological REpresentations

Codebase for the Koo Lab at Cold Spring Harbor Laboratory.

This is a work in progress.

Results of continuous integration Codecov

Install

One can install the master branch (the bleeding edge) with pip. One must also install TensorFlow.

pip install https://github.com/p-koo/libre/tarball/master tensorflow

Developers

Clone and install the package as below. This also creates a virtual environment, so the dependencies of this package are contained within this directory.

Note regarding TensorFlow: the installation of TensorFlow depends on the environment. For example, if there is no GPU, one can install tensorflow-cpu. If there is a GPU, then one must take care to install the proper TensorFlow version for the given CUDA and cuDNN versions.

git clone https://github.com/p-koo/libre.git
cd libre
python -m venv venv
source ./venv/bin/activate
python -m pip install --no-cache-dir --upgrade pip setuptools wheel
python -m pip install --no-cache-dir --editable .[dev] tensorflow

It is also useful to use pre-commit to run the styler (black) and lint the code on every commit.

pre-commit

Running tests

Run tests with pytest:

pytest libre