Pulbished to here
pull:
docker pull ebrown/nlp-gpu-jupyter:latest
Replace Local_Path_to_Notebooks with the path on your system. This will be mounted to /tf/notebooks in the container. This is the jupyter root directory. run in Windows:
# in powershell, into Jupyter:
wt -p "PowerShell" docker run -it --rm --gpus all -p 8888:8888 -p 6006:6006 -v "Local_Path_to_Notebooks:/tf/notebooks" ebrown/nlp-gpu-jupyter:latest
# in powershell, into bash:
wt -p "PowerShell" docker run --entrypoint /bin/bash -it --rm --gpus all -p 8888:8888 -p 6006:6006 -v "Local_Path_to_Notebooks:/tf/notebooks" ebrown/nlp-gpu-jupyter:latest
If you're using docker compose you'll need to adjust the volume mount to reflect the location of jupyter notebooks on your system.
run in Linux:
docker-compose up -d
docker-compose down