Skip to main content
Run your code in a docker container. W&B docker lets you run your code in a docker image ensuring wandb is configured. It adds the WANDB_DOCKER and WANDB_API_KEY environment variables to your container and mounts the current directory in /app by default. You can pass additional args which will be added to docker run before the image name is declared, we’ll choose a default image for you if one isn’t passed: 
wandb docker -v /mnt/dataset:/app/data
wandb docker gcr.io/kubeflow-images-public/tensorflow-1.12.0-notebook-cpu:v0.4.0 --jupyter
wandb docker wandb/deepo:keras-gpu --no-tty --cmd "python train.py --epochs=5"
By default, we override the entrypoint to check for the existence of wandb and install it if not present. If you pass the —jupyter flag we will ensure jupyter is installed and start jupyter lab on port 8888. If we detect nvidia-docker on your system we will use the nvidia runtime. If you just want wandb to set environment variable to an existing docker run command, see the wandb docker-run command. Run your code in a docker container. W&B docker lets you run your code in a docker image ensuring wandb is configured. It adds the WANDB_DOCKER and WANDB_API_KEY environment variables to your container and mounts the current directory in /app by default. You can pass additional args which will be added to docker run before the image name is declared, we’ll choose a default image for you if one isn’t passed: sh wandb docker -v /mnt/dataset:/app/data wandb docker gcr.io/kubeflow-images-public/tensorflow-1.12.0-notebook-cpu:v0.4.0 --jupyter wandb docker wandb/deepo:keras-gpu --no-tty --cmd "python train.py --epochs=5" By default, we override the entrypoint to check for the existence of wandb and install it if not present. If you pass the —jupyter flag we will ensure jupyter is installed and start jupyter lab on port 8888. If we detect nvidia-docker on your system we will use the nvidia runtime. If you just want wandb to set environment variable to an existing docker run command, see the wandb docker-run command.

​
Usage

wandb docker [DOCKER_RUN_ARGS] [DOCKER_IMAGE] [OPTIONS]

​
Arguments

ArgumentDescriptionRequired
DOCKER_RUN_ARGSNo description availableNo
DOCKER_IMAGENo description availableNo

​
Options

OptionDescription
--nvidiaUse the nvidia runtime, defaults to nvidia if nvidia-docker is present (default: False)
--digestOutput the image digest and exit (default: False)
--jupyterRun jupyter lab in the container (default: False)
--dirWhich directory to mount the code in the container (default: /app)
--no-dirDon’t mount the current directory (default: False)
--shellThe shell to start the container with (default: /bin/bash)
--portThe host port to bind jupyter on (default: 8888)
--cmdThe command to run in the container
--no-ttyRun the command without a tty (default: False)