Install

First clone the repository to your local machine:
git clone git@gitlab.com:qumphy/qumphy-software.git

And cd int the repository: cd qumphy-software

Then create a new pip or conda environment and activate it:
Using Conda:

conda env create -f environment.yml -n qumphy
conda activate qumphy
conda-develop .

Using pip:

python3 -m venv .venv
echo "*" > .venv/.gitignore # To exclude the pip env from git tracking
source .venv/bin/activate
pip install -r requirements.txt
pip install -e .

Install a logger of your choice (e.g. tensorboardx):

conda install tensorboardx

or

pip install tensorboardx

Now qumphy is ready to be used!

Usage

Move the dataset files that you want to use to the data directory (e.g. in a subdirectory deepbeat).
Write a config file for your model (use the templates as a basis).

Run the config file for training and testing:

python app/train.py --config app/configs/CONFIG_FILE.yml --fit --test