2.10.0_cuda11.8

PaddleSeg

Uses PaddleSeg (documentation).

Uses PyTorch 2.5.1, CUDA 11.8 and PaddleSeg 2.10.0.

Quick start

Inhouse registry

• Log into registry using public credentials:

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

• Pull and run image (adjust volume mappings -v):

  docker run --gpus=all --shm-size 8G --net=host \
    -v /local/dir:/container/dir \
    -it public.aml-repo.cms.waikato.ac.nz:443/paddle/paddleseg:2.10.0_cuda11.8

Docker hub

• Pull and run image (adjust volume mappings -v):

  docker run --gpus=all --shm-size 8G --net=host \
    -v /local/dir:/container/dir \
    -it waikatodatamining/paddleseg:2.10.0_cuda11.8

Build local image

• Build the image from Docker file (from within /path_to/paddleseg/2.10.0_cuda11.8)

  docker build -t paddleseg .

• Run the container

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

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

Publish images

Build

docker build -t paddleseg:2.10.0_cuda11.8 .

Inhouse registry

• Tag

  docker tag \
    paddleseg:2.10.0_cuda11.8 \
    public-push.aml-repo.cms.waikato.ac.nz:443/paddle/paddleseg:2.10.0_cuda11.8

• Push

  docker push public-push.aml-repo.cms.waikato.ac.nz:443/paddle/paddleseg:2.10.0_cuda11.8

  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

Docker hub

• Tag

  docker tag \
    paddleseg:2.10.0_cuda11.8 \
    waikatodatamining/paddleseg:2.10.0_cuda11.8

• Push

  docker push waikatodatamining/paddleseg:2.10.0_cuda11.8

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

  docker login

Requirements

docker run --rm \
  -it public.aml-repo.cms.waikato.ac.nz:443/paddle/paddleseg:2.10.0_cuda11.8 \
  pip freeze > requirements.txt

Permissions

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 ...

Caching

PaddleSeg will download pretrained models and cache them locally. To avoid having to download them constantly, you can the cache directory to the host machine:

• when running the container as current user

  -v /some/where/cache:/.paddleseg \

Scripts

The following additional scripts are available:

• paddleseg_export_config - for exporting template config files and setting parameters (calls the /opt/PaddleSeg/tools/export_config.py script)
• paddleseg_train - for training models (calls the /opt/PaddleSeg/tools/train.py script)
• paddleseg_export - for exporting models (calls the /opt/PaddleSeg/tools/export.py script)
• paddleseg_val - for evaluating models (calls the /opt/PaddleSeg/tools/val.py script)
• paddleseg_predict - for generating predictions of supplied files (calls the /opt/PaddleSeg/tools/predict.py script)
• paddleseg_predict_poll - for generating predictions of supplied files in batch/poll mode (calls the /opt/PaddleSeg/tools/predict_poll.py script)
• paddleseg_predict_redis - for generating predictions via Redis (calls the /opt/PaddleSeg/tools/predict_redis.py script)