declair 0.1.2

Package for declarative hyperparameter search experiments.

Declair :cake:

Declair is a framework for declaratively defining hyperparameter optimization experiments. It uses Sacred for storing experiment results and supports hyperopt for optimization.

It came about from attempts to recreate DeepSolaris results in PyTorch instead of Keras. However, it grew to be a more extensive and general framework than originally planned.

Usage

For detailed instructions on how to use Declair, see the documentation.

Installation

Install required Python packages in your favourite virtual environment

pip install -r requirements.txt

Running the tests

Go into the root of the repository (i.e. where this README.md is) and run

python -m pytest

Credits

This project has been heavily inspired by cbds_common.