PyTorch is a powerful machine learning library for Python. It comes in CPU and GPU enabled versions. Currently, the RCC only supports the CPU version, using the anaconda3.7.3 module. Users are welcome to install the GPU version from within their own home directories using a conda environment or a virtualenv environment. For more information about creating these environments, see the official venv and conda documentation pages.

Using PyTorch on RCC Resources

Before you can use PyTorch on the HPC, you need to load the appropriate anaconda module. From the command line, simply type:

$ module load anaconda/3.7.3

Then you can load PyTorch from within your Python code. Include the following line near the top of your code:

$ import torch

Then you can run your Python program. A wealth of official tutorials are available to help you get started with using PyTorch. For more information about PyTorch, please refer to the official documentation.