Skip to content

v0.0.1: Training on AWS Trainium

Choose a tag to compare

View all tags
@michaelbenayoun michaelbenayoun released this 13 Mar 14:06
· 1750 commits to main since this release
v0.0.1
13d792c

The following architectures can be trained on AWS Trainium instances (trn1.2xlarge and trn1.32xlarge) :

  • ALBERT
  • BERT
  • DistilBERT
  • RoBERTa
  • XLM-RoBERTa
  • CamemBERT
  • Electra
  • GPT-2
  • GPT-Neo
  • MarianMT
  • T5
  • BART
  • ViT

Training examples for many tasks are provided here.

Assets 2
Loading