Deep Learning in Pytorch#

Instructor#

  • Instructor: Carl Ehrett

  • Office: 2105 Barre Hall, Clemson University

  • Email: cehrett@clemson.edu

Workshop Description#

This workshop series 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 sessions, 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, create or navigate to an empty folder. Then run the following command:

wget https://raw.githubusercontent.com/clemsonciti/rcde_workshops/master/pytorch/download.sh

This will download a script named download.sh. Run it with:

bash download.sh

This will create a folder named pytorch 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. You can create this environment by submitting a job to the cluster.

From the terminal, after downloading the workshop files:

cd pytorch
sbatch create_env_pip.sh

This will submit a job that creates a Conda environment named PytorchWorkshop 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”, which you can select in JupyterLab to run the notebooks.

Note: Notebook 8 contains some code that requires you to have a free Weights & Biases (WandB) account.