onnxruntime
Longformer Attention CUDA kernel memory Improvements
#6646

Merged

Longformer Attention CUDA kernel memory Improvements #6646

tianleiwu merged 6 commits into master from tlwu/longformer_attention_v2

Integrate memory improvements from NVidia

9a8b1e81

tianleiwu added model:Longformer

tianleiwu requested a review 5 years ago

tianleiwu marked this pull request as draft 5 years ago

add comments

440f59e0

Merge remote-tracking branch 'origin/master' into tlwu/longformer_att…

ccd76de4

Compute max_global_num

67e37208

update conversion script to support transformers 4.0

0f358b03

Use a wrapper of cuda event to avoid memory leak

8a2e2da9

tianleiwu requested a review from

gh-yewang 5 years ago

tianleiwu marked this pull request as ready for review 5 years ago

tianleiwu changed the title ~~[WIP] Longformer Attention CUDA kernel Improvements~~ Longformer Attention CUDA kernel memory Improvements 5 years ago

gh-yewang approved these changes on 2021-02-16

tianleiwu added release:1.7

tianleiwu merged 9b446d5f into master 5 years ago

tianleiwu deleted the tlwu/longformer_attention_v2 branch 5 years ago

oliviajain removed release:1.7

Reviewers

gh-yewang

Assignees

No one assigned

Labels

None yet

Milestone

No milestone

onnxruntime Longformer Attention CUDA kernel memory Improvements #6646 Merged

Longformer Attention CUDA kernel memory Improvements #6646

onnxruntime
Longformer Attention CUDA kernel memory Improvements
#6646

Merged