4.5 Deep Learning Transfer Learning Hands On

Source repo: ciml-summer-institute-2021 | Branch: main | Last synced: 2026-04-24 10:27:17.425 UTC

4.5_deep_learning_transfer_learning_hands_on/

Running on Expanse Portal

1. Log in to the Expanse Portal

Use your XSEDE username (trainXX) and password to login to the portal.

https://portal.expanse.sdsc.edu

2. Select Jupyter from the Interactive Apps menu on the Expanse Portal.

3. Specify your Job Parameters

Account: `account`

Partition: gpu-shared

Time limit (min): 180

Number of cores: 10

Memory required per node: 93

GPUs: 1

Singularity Image File: /cm/shared/apps/containers/singularity/tensorflow/tensorflow-latest.sif

Environment modules to be loaded: singularitypro

Reservation: `reservation`

Working directory: home

Type: JupyterLab

4. Click on URL to start Jupyter

Running on Terminal

1. Log in to Expanse in terminal window

ssh `username`@login.expanse.sdsc.edu

2. Launch Jupyter server with GPU

start_gpu

This is an alias for

export PATH="/cm/shared/apps/sdsc/galyleo:${PATH}"

galyleo.sh launch --account \`account\` --reservation \`reservation\` --partition 'gpu-shared' --cpus-per-task 10 --memory-per-node 93 --gpus 1 --time-limit 03:00:00 --jupyter 'lab' --notebook-dir "/home/${USER}" --env-modules 'singularitypro' --sif '/cm/shared/apps/containers/singularity/tensorflow/tensorflow-latest.sif' --bind '/expanse,/scratch,/cvmfs' --nv --quiet

3. Copy URL in a web browser