If you run build and create a our Docker image using docker compose, this could happen. The solution is exit the container, stop the container with docker compose down and then start the container with docker compose up -d again.
docker-compose.yml contains scripts for running with GPU(s). If you do not have GPU(s), please command out code from line 12 them re-run docker compose up -d command. And pip install only the cpu-only-requiremnts.txt.
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: all
capabilities: [gpu]
PySZZ is surpose to create its own out folder. If it does not create the folder, then you have to create for it.
mkdir <path/to/PySZZ>/out
Some models still require you to download its hyperparameters, pretrain is also downloaded with it. The pretrain will not be load into the model.