Skip to content

Latest commit

 

History

History
246 lines (166 loc) · 5.93 KB

README.md

File metadata and controls

246 lines (166 loc) · 5.93 KB

MMDetection

Allows processing of images with MMSegmentation.

Uses PyTorch 1.9.0 and CPU support.

Version

MMDetection github repo tag/hash:

v0.25.0
46326f63ce411c794d237e986dd3924590d0e75e

and timestamp:

June 3rd, 2022

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 --shm-size 8G \
  -v /local/dir:/container/dir \
  -it public.aml-repo.cms.waikato.ac.nz:443/open-mmlab/mmsegmentation:0.25.0_cpu

Docker hub

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

    docker run --shm-size 8G \
  -v /local/dir:/container/dir \
  -it waikatodatamining/mmsegmentation:0.25.0_cpu

Build local image

  • Build the image from Docker file (from within /path_to/mmsegmentation/0.25.0_cpu)

    docker build -t mmseg .

  • Run the container

    docker run --shm-size 8G -v /local/dir:/container/dir -it mmseg

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

Publish images

Build

docker build -t mmsegmentation:0.25.0_cpu .

Inhouse registry

  • Tag

    docker tag \
  mmsegmentation:0.25.0_cpu \
  public-push.aml-repo.cms.waikato.ac.nz:443/open-mmlab/mmsegmentation:0.25.0_cpu

  • Push

    docker push public-push.aml-repo.cms.waikato.ac.nz:443/open-mmlab/mmsegmentation:0.25.0_cpu

    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 \
  mmsegmentation:0.25.0_cpu \
  waikatodatamining/mmsegmentation:0.25.0_cpu

  • Push

    docker push waikatodatamining/mmsegmentation:0.25.0_cpu

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

    docker login

Scripts

The following scripts are available:

  • mmseg_config - for expanding/exporting default configurations (calls /mmsegmentation/tools/misc/print_config.py)
  • mmseg_predict_poll - for applying a model to images (uses file-polling, calls /mmsegmentation/tools/predict_poll.py)
  • mmseg_predict_redis - for applying a model to images (via Redis backend), add --net=host to the Docker options (calls /mmsegmentation/tools/predict_redis.py)

Usage

  • Predict and produce PNG files

    mmseg_predict_poll \
    --model /path_to/epoch_n.pth \
    --config /path_to/your_data_config.py \
    --prediction_in /path_to/test_imgs \
    --prediction_out /path_to/test_results

    Run with -h for all available options.

  • Predict via Redis backend

    You need to start the docker container with the --net=host option if you are using the host's Redis server.

    The following command listens for images coming through on channel images and broadcasts predicted images on channel predictions:

    mmseg_predict_redis \
    --model /path_to/epoch_n.pth \
    --config /path_to/your_data_config.py \
    --redis_in images \
    --redis_out predictions

    Run with -h for all available options.

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 models

PyTorch downloads base models, if necessary. However, by using Docker, this means that models will get downloaded with each Docker image, using up unnecessary bandwidth and slowing down the startup. To avoid this, you can map a directory on the host machine to cache the base models for all processes (usually, there would be only one concurrent model being trained):

-v /somewhere/local/cache:/.cache

Or specifically for PyTorch:

-v /somewhere/local/cache/torch:/.cache/torch

NB: When running the container as root rather than a specific user, the internal directory will have to be prefixed with /root.

Testing Redis

You can use simple-redis-helper to broadcast images and listen for image segmentation results when testing.

Testing inference

You can test the inference of your container with the image_demo2.py script as follows:

  • create a test directory and change into it

    mkdir test_inference
cd test_inference

  • create cache directory

    mkdir -p cache/torch

  • start the container in interactive mode

    docker run --shm-size 8G -u $(id -u):$(id -g) -e USER=$USER \
  -v `pwd`:/workspace \
  -v `pwd`/cache:/.cache \
  -v `pwd`/cache/torch:/.cache/torch \
  -it public.aml-repo.cms.waikato.ac.nz:443/open-mmlab/mmsegmentation:0.25.0_cuda11.1

  • download a pretrained model

    cd /workspace
mim download mmsegmentation --config pspnet_r50-d8_512x1024_40k_cityscapes --dest .

  • perform inference

    python /mmsegmentation/demo/image_demo2.py \
  --img /mmsegmentation/demo/demo.png \
  --config /mmsegmentation/configs/pspnet/pspnet_r50-d8_512x1024_40k_cityscapes.py \
  --checkpoint pspnet_r50-d8_512x1024_40k_cityscapes_20200605_003338-2966598c.pth \
  --output_file /workspace/demo_out.png

  • the model saved the result of the segmentation in test_inference/demo_out.png (in grayscale)

Troubleshooting

  • ValueError: SyncBatchNorm expected input tensor to be on GPU

    Replace SyncBN with BN in the config file (see here).