Training Compiler

Train deep learning (DL) models faster on scalable GPU instances managed by SageMaker.

Get started with SageMaker Training Compiler

PyTorch with Hugging Face Transformers

For single-node single-GPU training:

• Compile and Train a Hugging Face Transformers Trainer Model for Question and Answering with the SQuAD dataset
• Compile and Train a Hugging Face Transformer BERT Model with the SST Dataset using SageMaker Training Compiler
• Compile and Train a Binary Classification Trainer Model with the SST2 Dataset for Single-Node Single-GPU Training
• Compile and Train a Vision Transformer Model on the Caltech-256 Dataset using a Single Node

For single-node multi-GPU training:

• Compile and Train a Vision Transformer Model on the Caltech-256 Dataset using a Single Node with Multiple GPUs

For multi-node multi-GPU training:

• Compile and Train the GPT2 Model using the Transformers Trainer API with the wikitext Dataset for Multi-Node Multi-GPU Training

For more information, see Amazon SageMaker Training Compiler in the Amazon SageMaker Developer Guide.