Scientific images

This page is mostly related to our Official JupyterHub, but all images can be also used in other pods deployed directly on the cluster.

There are two main projects providing the stack of images:

Docker Stack

B-Data

Both projects now support CUDA and ARM. Docker stack only has CUDA in TensorFlow and PyTorch variants.

The list of NRP-provided images and registry links are available in our GitLab registry:

The NRP image with additional libraries (gitlab-registry.nrp-nautilus.io/nrp/scientific-images/python) is based on Docker Stack TensorFlow with the addition of:

    torch
    torchvision
    torchaudio
    'jax[cuda12]'
    flax
    keras
    keras-tuner
    'tensorflow-probability[tf]'
    fastai
    'jupyter-ai[all]'
    astropy
    bowtie
    imutils
    nbgitpuller
    opencv-python
    psycopg2-binary
    pydot
    requests
    visualdl
    git+https://github.com/veeresht/CommPy.git
    bash_kernel
    'code-server==4.10.0'
    'jupyter-vscode-proxy'
    'ghostscript'
    'graphviz'
    'rasterio'
    'gdal'
    'libgdal-arrow-parquet'

We can add more libraries to this image by requesting in Matrix.

The Desktop image has X Window system installed and you can launch the GUI interface in Jupyter with this image. It's based on Minimal stack.