Docker image for faster-whisper 1.0.2.

Uses PyTorch 2.3.0, CUDA 12.1.

• Log into registry using public credentials:

  docker login -u public -p public public.aml-repo.cms.waikato.ac.nz:443 

• Create the following directory:

  mkdir cache

• Launch docker container

  docker run \
    -u $(id -u):$(id -g) -e USER=$USER \
    --gpus=all \
    --shm-size 8G \
    -v `pwd`:/workspace \
    -v `pwd`/cache:/.cache \
    -it public.aml-repo.cms.waikato.ac.nz:443/pytorch/python-faster-whisper:1.0.2_cuda12.1

• Create the following directory:

  mkdir cache

• Launch docker container

  docker run \
    -u $(id -u):$(id -g) -e USER=$USER \
    --gpus=all \
    --shm-size 8G \
    -v `pwd`:/workspace \
    -v `pwd`/cache:/.cache \
    -it waikatodatamining/python-faster-whisper:1.0.2_cuda12.1

• Build the image from Docker file (from within /path_to/faster-whisper-1.0.2_cuda12.1)

  docker build -t hf .

• Run the container

  docker run --gpus=all --shm-size 8G -v /local/dir:/container/dir -it hf

  /local/dir:/container/dir maps a local disk directory into a directory inside the container

docker build -t python-faster-whisper:1.0.2_cuda12.1 .

• Tag

  docker tag \
    python-faster-whisper:1.0.2_cuda12.1 \
    public-push.aml-repo.cms.waikato.ac.nz:443/pytorch/python-faster-whisper:1.0.2_cuda12.1

• Push

  docker push public-push.aml-repo.cms.waikato.ac.nz:443/pytorch/python-faster-whisper:1.0.2_cuda12.1

  If error "no basic auth credentials" occurs, then run (enter username/password when prompted):

  docker login public-push.aml-repo.cms.waikato.ac.nz:443

• Tag

  docker tag \
    python-faster-whisper:1.0.2_cuda12.1 \
    waikatodatamining/python-faster-whisper:1.0.2_cuda12.1

• Push

  docker push waikatodatamining/python-faster-whisper:1.0.2_cuda12.1

  If error "no basic auth credentials" occurs, then run (enter username/password when prompted):

  docker login

docker run --rm --pull=always \
  -it public.aml-repo.cms.waikato.ac.nz:443/pytorch/python-faster-whisper:1.0.2_cuda12.1 \
  pip freeze > requirements.txt

The following scripts are available:

• fw_predict_poll - for applying a model to audio files (uses file-polling, calls /opt/faster-whisper/predict_poll.py)
• fw_predict_redis - for applying a model to images (via Redis backend), add --net=host to the Docker options (calls /opt/faster-whisper/predict_redis.py)

When running the docker container as regular use, you will want to set the correct user and group on the files generated by the container (aka the user:group launching the container):

docker run -u $(id -u):$(id -g) -e USER=$USER ...