Seeing maskrcnn_resnet50_fpn + FP32 performance drop by 15% after #5443
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 17.9k
- Forks
- 7.3k
- Avg merge
- 1d 15h
- Merged PRs (30d)
- 13
Description
🐛 Describe the bug
Seeing maskrcnn_resnet50_fpn + FP32 performance drop by 15% after #5443
We're running benchmarks for torchvision maskrcnn_resnet50_fpn + FP32 model with pretrained = False, pretrained_backbone = False. After the PR, we saw training throughput drops by ~15% on various GPU models (A100, V100, 3090).
Versions
commit: after #5443
GPU: Nvidia A100, V100, 3090
Library: cuda 11.6 + cudnn 8.3.2
cc @ptrblck @ngimel
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Begin by reviewing PR #5443 and the maskrcnn_resnet50_fpn training benchmark using pretrained = False and pretrained_backbone = False. Reproduce the reported throughput on the listed A100, V100, or 3090 systems with CUDA 11.6 and cuDNN 8.3.2, then identify and document the change responsible for the regression.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- computer-vision, machine-learning, performance
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100