PyTorch Model Porting

This guide provides the steps required to set up the GPU Migration toolkit in your model. The GPU Migration toolkit helps to automate the process of migrating GPU-based models to Intel® Gaudi® AI accelerator, requiring minimal modifications to your model.

This guide also provides the initial steps and modifications needed to manually migrate an existing PyTorch model to Gaudi. The modifications include setting up the Intel Gaudi Torch library, enabling mixed precision and initializing HCCL for distributed training.

Using the initial steps detailed in the below sections ensures that your migrated model is functional. You can then proceed to the documents listed here.

• GPU Migration Toolkit
• Importing PyTorch Models Manually

Note

This guide details the steps required to prepare PyTorch models to run on Gaudi. To create and execute a simple PyTorch model with Gaudi, see Getting Started with Training on Intel Gaudi.