Optimum documentation

Single-HPU Training

Optimum

Overview

🤗 Optimum Installation Quick tour Notebooks

Conceptual guides

Nvidia

AMD

Intel

AWS Trainium/Inferentia

Google TPUs

for Intel Gaudi

🤗 Optimum for Intel Gaudi Installation Quickstart

Tutorials

Overview Single-HPU Training Distributed Training Run Inference Stable Diffusion TGI on Gaudi

How-To Guides

Reference

ExecuTorch

ONNX

Furiosa

Exporters

Torch FX

LLM quantization

Utilities

You are viewing main version, which requires installation from source. If you'd like regular pip install, checkout the latest stable version (v1.27.0).

Join the Hugging Face community

and get access to the augmented documentation experience

Collaborate on models, datasets and Spaces

Faster examples with accelerated inference

Switch between documentation themes

to get started

Single-HPU Training

Training on a single device is as simple as in Transformers:

You need to replace the Transformers’ Trainer class with the GaudiTrainer class,
You need to replace the Transformers’ TrainingArguments class with the GaudiTrainingArguments class and add the following arguments:
- use_habana to execute your script on an HPU,
- use_lazy_mode to use lazy mode (recommended) or not (i.e. eager mode),
- gaudi_config_name to give the name of (Hub) or the path to (local) your Gaudi configuration file.

To go further, we invite you to read our guides about accelerating training and pretraining.

Update on GitHub

←Overview Distributed Training→