Youhe-Jiang/IJCAI2023-OptimalShardedDataParallel

[IJCAI2023] An automated parallel training system that combines the advantages from both data and model parallelism. If you have any interests, please visit/star/fork https://github.com/Youhe-Jiang/OptimalShardedDataParallel

34
/ 100
Emerging

This project helps machine learning engineers and researchers efficiently train very large deep learning models like GPT or OPT on multiple GPUs. It takes your existing PyTorch model and automatically optimizes how data and model components are distributed across your hardware, allowing you to train bigger models faster and with less memory overhead than standard methods. The output is a more efficient training process for large-scale AI models.

No commits in the last 6 months.

Use this if you are a machine learning engineer or researcher struggling to train large deep learning models due to memory limitations or slow training times on multi-GPU setups.

Not ideal if you are working with small models that don't require distributed training or if you are not using PyTorch.

deep-learning-training large-language-models distributed-ml gpu-optimization model-scalability
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 10 / 25

How are scores calculated?

Stars

52

Forks

5

Language

Python

License

MIT

Last pushed

May 31, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Youhe-Jiang/IJCAI2023-OptimalShardedDataParallel"

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