Jiarui Fang(方佳瑞)

Results 6 repositories owned by Jiarui Fang(方佳瑞)

Distributed-ResNet-Tensorflow

20
Stars
6
Forks
Watchers

A Distributed ResNet on multi-machines each with one GPU card.

SWCaffe

39
Stars
20
Forks
Watchers

A Deep Learning Framework customized for Sunway TaihuLight

long-context-attention

317
Stars
20
Forks
Watchers

USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference

LLMRoofline

62
Stars
3
Forks
Watchers

Compare different hardware platforms via the Roofline Model for LLM inference tasks.

LLMSpeculativeSampling

431
Stars
47
Forks
Watchers

Fast inference from large lauguage models via speculative decoding

Odysseus-Transformer

47
Stars
1
Forks
Watchers

Odysseus: Playground of LLM Sequence Parallelism