Highlights
- Arctic Code Vault Contributor
Pinned
196 contributions in the last year
Less
More
Activity overview
Contributed to
microsoft/onnxruntime,
microsoft/onnxruntime-openenclave,
onnx/onnx
and 5 other
repositories
Contribution activity
February 2021
Created 2 commits in 1 repository
Created a pull request in microsoft/onnxruntime that received 3 comments
Plug and Allocate using external CUDA allocator.
The motivation for this change is to share allocator with external entities like PyTorch to prevent having multiple allocators hiding free spaces f…
+890
−530
•
3
comments
Opened 12 other pull requests in 1 repository
microsoft/onnxruntime
5
merged
1
open
6
closed
- Sync from master
- Do not create compute stream when external CUDA allocator is used.
- [DO NOT REVIEW] VERY VERY CRUDE PROOF OF CONCEPT FOR PAUSE n Resume aka Partial Graph Run.
- Enable external CUDA allocator in ORTModule.
- Sync from master
- Thiagofc/merge from master
- DO NOT REVIEW.....
- DO NOT REVIEW
- Rebase ortmodule-api branch on master
- Plug n Allocate with external CUDA allocator via PyBind.
- DO NOT REVIEW .....
- [DO NO REVIEW] Mzs/external cuda allocator 2