CarperAI/trlx

Add tensor padding options to improve hardware utilization

Open

#284 aberto em 6 de fev. de 2023

Ver no GitHub
 (8 comments) (2 reactions) (0 assignees)Python (450 forks)batch import
contributions welcomefeature requestgood first issue

Métricas do repositório

Stars
 (4.184 stars)
Métricas de merge de PR
 (Nenhuma PRs mesclada em 30d)

Description

🚀 The feature, motivation, and pitch

Recent discussion on Twitter highlighted the importance of tensor padding to improve hardware util. NeMo seems to already support GPU-friendly vocab size padding from discussions with @cat-state but we should also consider adding an optional pad_to_multiple_of arg to our tokenizer calls with the accelerate/transformers backend to satisfy Tensor Core requirements.

image

The implementer should also provide system plots to display any improvements/findings.

Alternatives

No response

Additional context

No response

Guia do colaborador