Docker-compose project with Jupyterlab as a service
Check jupyterlab_gpu/Dockerfile to see all libraries installed.
Mos important ones:
- scikit-learn
- fastgpu
- nbdev
- pandas
- transformers
- tensorflow-addons
- tensorflow
- pymongo
- emoji
- python-dotenv
- plotly
- torch
- torchvision
- torchaudio
To keep same user and permissions between docker and host, you have to run:
echo -e "\nUID=$(id -u)\nGID=$(id -g)\nUSER=$(whoami)" >> .env