Deep Learning in Pytorch#

Instructor#

  • Instructor: Benjamin Formby

  • Office: 2002 Barre Hall, Clemson University

  • Email: bformby@clemson.edu

Workshop Description#

This workshop introduces the essential concepts in deep learning and walks through the common steps in a deep learning workflow from data loading and preprocessing to training and model evaluation. Throughout the workshop, students participate in writing and executing simple deep learning programs using Pytorch – a popular Python library for developing, training, and deploying deep learning models.

Prerequisites#

All students should have a Palmetto Cluster account. If you do not already have an account, you can visit our getting started page. Students should be familiar with the Python programming language. This requirement could be fulfilled by personal projects, coursework, or completion of the Introduction to Python Programming workshop series. Experience in numeric computing with the numpy library is helpful but not required.

Accessing Workshop Files#

You can download the notebooks and their contents as follows. If you are unfamiliar with using the terminal on Palmetto, please take our self-guided onboarding training.

In the terminal on the cluster, run the following command:

cp -R /project/rcde/workshops/pytorch_intro/workshop/ ~/

This will create a folder named workshop in your home directory containing the notebooks and workshop materials.

Environment#

To run the code in this workshop, you’ll need a Python environment with the appropriate libraries installed. For the workshop, I have already created an environment that you can use. It is located at /project/rcde/workshops/pytorch-intro/env. You can use it in JupyterLab via Open OnDemand by supplying the “Virtual environment activation command”:

source activate /project/rcde/workshops/pytorch_intro/env

Long-term, if you want to be using Pytorch, you should have your own environment that you use. You can create this environment by submitting a job to the cluster. (You do not need to do this for the workshop, since you can activate the environment as described above.)

From the terminal, after copying the workshop files:

cd workshop
sbatch create_env.sh

This will submit a job that creates a Conda environment named PytorchWorkshopFA26 and installs all required libraries. The job may take up to 30 minutes to complete. You can monitor its progress using:

squeue -u $USER

Once the job finishes, the environment will be available as a Jupyter kernel named “Pytorch Workshop Fall 2026”, which you can select in JupyterLab to run the notebooks.

JupyterLab Open OnDemand settings for the workshop#

To run the notebooks during the workshop, you should start a JupyterLab session in Open OnDemand at ondemand.rcd.clemson.edu. Once there, use the menu at the top to go to “Interactive Apps” -> “Jupyter Notebook”, and use the below settings.

  • Partition: workshop

  • Account: rcd_workshop

  • Slurm Reservation: pytorch

  • CPU cores: 8

  • Memory (GB): 32

  • GPUs: 1

  • GPU Model: V100

  • Number of hours: 2

  • Environment: Standard Jupyter Notebook

  • Use a custom environment not listed below?: Yes

  • Virtual environment activation command: source activate /project/rcde/workshops/pytorch_intro/env