Clone the repo and install virtualenv
git clone <REMOTE_GIT>
sudo apt install python3-virtualenv
Create the virtual environment with python3
virtualenv --python=/usr/bin/python3 venv
Activate the virtual environment
source venv/bin/activate
Install the required libraries listed in requirements.txt
pip install -r requirements.txt
python main.py -m BASIC -d DATASETS/dataset_test/ -i 50x1 -e 3
Check the input parameters with
python main.py --help
The performances results are printed on file in the logs folder,
Moreover, tensorboard data is stored in the folder tensorboard and can be used to print graphs on the training
loss/acc trend by running
tensorboard --logdir tensorboard/
and then visualize the graphs at http://localhost:6006/.