rasbt/pytorch-memory-optim

This code repository contains the code used for my "Optimizing Memory Usage for Training LLMs and Vision Transformers in PyTorch" blog post.

38
/ 100
Emerging

This project offers practical code examples and scripts for PyTorch developers who are training large language models (LLMs) and vision transformers. It demonstrates techniques to reduce the memory footprint during model training, helping you work with larger models or limited GPU resources. You provide existing PyTorch training code and get back insights and modified code that uses less GPU memory.

No commits in the last 6 months.

Use this if you are a PyTorch developer encountering 'out of memory' errors or want to optimize GPU memory usage when training large AI models.

Not ideal if you are not a PyTorch developer or are looking for a fully automated, black-box memory optimization solution without diving into code.

PyTorch-development GPU-memory-optimization large-model-training machine-learning-engineering deep-learning-optimization
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 9 / 25
Maturity 16 / 25
Community 13 / 25

How are scores calculated?

Stars

92

Forks

11

Language

Python

License

Apache-2.0

Last pushed

Jul 14, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/rasbt/pytorch-memory-optim"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.