ThetaGPU (Argonne LCF)

ThetaGPU is an extension of Theta and is comprised of 24 NVIDIA DGX A100 nodes at Argonne Leadership Computing Facility (ALCF). See the documentation of ThetaGPU from the Datascience group at Argonne National Laboratory for more information. The system documentation from the ALCF can be accessed here.

Already installed module

This installation procedure shows you how to access the installed DeepHyper module on ThetaGPU. After logging in Theta, connect to a ThetaGPU service node:

$ ssh thetagpusn1

Then, to access Deephyper run the following commands:

$ module load conda/2021-09-22
$ conda activate base

Finally, to verify the installation do:

$ python
>>> import deephyper
>>> deephyper.__version__
'0.3.0'

Conda environment

This installation procedure shows you how to create your own Conda virtual environment and install DeepHyper in it.

After logging in Theta, locate yourself on one of the ThetaGPU service node (thetagpusnX) and move to your project folder (replace PROJECTNAME by your own project name):

$ ssh thetagpusn1
$ cd /lus/theta-fs0/projects/PROJECTNAME

Then create the dhgpu environment:

$ module load conda/2021-09-22
$ conda create -p dhgpu --clone base
$ conda activate dhgpu/

Finally install DeepHyper in the previously created dhgpu environment:

$ pip install pip --upgrade
$ # DeepHyper + Analytics Tools (Parsing logs, Plots, Notebooks)
$ pip install deephyper["analytics"]

Developer installation

Follow the Conda environment installation and replace pip install deephyper[analytics] by:

$ git clone https://github.com/deephyper/deephyper.git
$ cd deephyper/ && git checkout develop
$ pip install -e ".[dev,analytics]"

Internet Access

If the node you are on does not have outbound network connectivity, set the following to access the proxy host:

$ export http_proxy=http://proxy.tmi.alcf.anl.gov:3128
$ export https_proxy=http://proxy.tmi.alcf.anl.gov:3128