A Jupyter Server container image with Langchain and Deno pre-installed.
This project can be used to run a custom image on AWS SageMaker Studio. See this guide for more information.
export WORKDIR=$(pwd) # <-- set this to whatever
export OPENAI_API_KEY # <-- set this to the OpenAI API Key
docker run \
--rm \
--interactive \
--tty \
--publish 8888:8888 \
--user root \
--env OPENAI_API_KEY \
--env NB_UID=$(id -u) \
--env NB_GID=$(id -g) \
--volume "${WORKDIR}:/home/jovyan/work" \
andreswebs/jupyter-langchain
Edit the jupyter_server_config.py
file before
building to customize the server. See this
guide for an example of how to generate
that file.
Andre Silva andreswebs
https://docs.conda.io/projects/conda/en/latest/commands/index.html
https://jupyter-docker-stacks.readthedocs.io
https://github.com/jupyter/docker-stacks/blob/main/images/docker-stacks-foundation/Dockerfile
https://jupyter-server.readthedocs.io/en/latest/other/full-config.html
https://jupyter-docker-stacks.readthedocs.io/en/latest/using/common.html
https://github.com/aws-samples/sagemaker-studio-custom-image-samples
https://github.com/aws-samples/sagemaker-studio-custom-image-samples/blob/main/DEVELOPMENT.md
The Dockerfile is based on this recipe: https://jupyter-docker-stacks.readthedocs.io/en/latest/using/recipes.html
This project is licensed under the Unlicense.