JupyterLab
To support machine learning code development, our users can deploy one or more private JupyterLab applications.
To encourage fair sharing these applications are time limited. We also ask users to request only the resources that they need.
Selecting GPU memory and instances
The AF cluster has four NVIDIA A100 GPUs. Each GPU can be partitioned into seven GPU instances. This means the AF cluster can have up to 28 GPU instances running in parallel.
A user can request 0 to 7 GPU instances as a resource for the notebook. A user can request 40,836 MB of memory for an entire A100 GPU, or 4864 MB of memory for a MIG instance.
Selecting a Docker image
Users can choose from five images:
ml_platform:latest
- has most of the ML packages (Tensorflow, Keras, ScikitLearn,...) preinstalled, and a small tutorial with example codes in /ML_platform_tests/tutorial, it supports NVidia GPUs and has ROOT preinstalled.ml_platform:conda
- comes with full anaconda.ml_platform:julia
- with Julia programming langugeml_platform:lava
- with Intel Lava neuromorphic computing frameworkml_platform:centos
AB-stable
- based on AnalysisBaseAB-dev
- based on AnalysisBase but with cutting edge uproot, dask, awkward arrays, etc.
Users can choose between two images: One with full anaconda (ivukotic/ml_platform:conda) and one with NVidia GPU and ROOT support (ivukotic/ml_platform:latest). The later has most of the ML packages (Tensorflow, Keras, ScikitLearn,...) preinstalled, and a small tutorial with example codes in /ML_platform_tests/tutorial. For software additions and upgrades please contact ivukotic@uchicago.edu.
Tutorials
Basic usage of the platform can be exerienced by running the set of tutorials that come preinstalled with both latest and conda image.
Running in conda
To run tutorial in conda environment, one first has to initialize conda. Simply open a jupyter lab terminal, and execute: conda init. Close that terminal and open a new one. This will drop you in (base) conda environment. You may now switch to a HEP relevant environment by executing: conda activate codas-hep.