Container environment using Docker for Learning IoT and AI
See Wiki
docker docker-compose
docker-compose build
Creates
iotai_infer_gpu
iotai_train_gpu
iotain_infer_gpu
: use for training models.iotain_train_gpu
builds oniotain_infer_gpu
adding development tools.iotain_infer_gpu
: use for runnign models in production.
~/.bashrc
export CURRENT_UID=$(id -u):$(id -g)
docker-compose up -d train
docker-compose up -d infer
List running containers
docker ps
Connect
docker exec -it iotai_container_env_train_1
alias iotai_infer='docker run --rm -u $(id -u):$(id -g) --name iotai_infer --runtime=nvidia -it -v "/etc/passwd:/etc/passwd:ro" -v "/etc/group:/etc/group:ro" -v "/home:/home" --memory=16g --shm-size=16g -p "780:80" -v /tmp/.X11-unix:/tmp/.X11-unix -e DISPLAY=$DISPLAY iotai_infer_gpu:latest bash'
alias iotai_train='docker run --rm -u $(id -u):$(id -g) --name iotai_train --runtime=nvidia -it -v "/etc/passwd:/etc/passwd:ro" -v "/etc/group:/etc/group:ro" -v "/home:/home" --memory=16g --shm-size=16g -p "780:80" -p "7888:8888" -v /tmp/.X11-unix:/tmp/.X11-unix -e DISPLAY=$DISPLAY iotai_train_gpu:latest bash'