Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

fix kMaxInputCount check in GroupedMatmulFunctor #10322

Merged
merged 3 commits into from
Aug 26, 2023
Merged

fix kMaxInputCount check in GroupedMatmulFunctor #10322

merged 3 commits into from
Aug 26, 2023

Conversation

jackalcooper
Copy link
Collaborator

No description provided.

@github-actions
Copy link
Contributor

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/10322/

@github-actions
Copy link
Contributor

Speed stats:

@github-actions
Copy link
Contributor

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/10322/

@github-actions
Copy link
Contributor

Speed stats: 
GPU Name: NVIDIA GeForce RTX 3080 Ti 

❌ OneFlow resnet50 time: 43.7ms (= 4369.5ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 61.8ms (= 6181.2ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.41 (= 61.8ms / 43.7ms)

OneFlow resnet50 time: 26.2ms (= 2615.8ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 37.8ms (= 3783.2ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.45 (= 37.8ms / 26.2ms)

OneFlow resnet50 time: 19.5ms (= 3895.1ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 34.7ms (= 6934.5ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.78 (= 34.7ms / 19.5ms)

OneFlow resnet50 time: 18.9ms (= 3774.2ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 31.9ms (= 6385.9ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.69 (= 31.9ms / 18.9ms)

OneFlow resnet50 time: 18.6ms (= 3723.5ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 28.2ms (= 5641.8ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.52 (= 28.2ms / 18.6ms)

OneFlow swin dataloader time: 0.201s (= 40.114s / 200, num_workers=1)
PyTorch swin dataloader time: 0.129s (= 25.730s / 200, num_workers=1)
Relative speed: 0.641 (= 0.129s / 0.201s)

OneFlow swin dataloader time: 0.054s (= 10.719s / 200, num_workers=4)
PyTorch swin dataloader time: 0.032s (= 6.460s / 200, num_workers=4)
Relative speed: 0.603 (= 0.032s / 0.054s)

OneFlow swin dataloader time: 0.030s (= 6.073s / 200, num_workers=8)
PyTorch swin dataloader time: 0.016s (= 3.291s / 200, num_workers=8)
Relative speed: 0.542 (= 0.016s / 0.030s)

❌ OneFlow resnet50 time: 47.7ms (= 4768.2ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 65.3ms (= 6525.7ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.37 (= 65.3ms / 47.7ms)

OneFlow resnet50 time: 31.9ms (= 3185.9ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 45.0ms (= 4497.5ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.41 (= 45.0ms / 31.9ms)

OneFlow resnet50 time: 24.1ms (= 4816.0ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 40.7ms (= 8140.4ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.69 (= 40.7ms / 24.1ms)

OneFlow resnet50 time: 20.7ms (= 4138.1ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 36.3ms (= 7259.8ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.75 (= 36.3ms / 20.7ms)

OneFlow resnet50 time: 21.6ms (= 4320.8ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 34.2ms (= 6836.6ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.58 (= 34.2ms / 21.6ms)

@github-actions
Copy link
Contributor

Static analysis with clang failed. PR label automerge has been removed

This was referenced Aug 25, 2023
Merged
Closed
@jackalcooper jackalcooper enabled auto-merge (squash) August 25, 2023 12:03
@github-actions
Copy link
Contributor

Speed stats: 
GPU Name: NVIDIA GeForce RTX 3080 Ti 

❌ OneFlow resnet50 time: 43.1ms (= 4308.3ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 57.1ms (= 5707.2ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.32 (= 57.1ms / 43.1ms)

OneFlow resnet50 time: 25.9ms (= 2585.2ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 37.8ms (= 3779.5ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.46 (= 37.8ms / 25.9ms)

OneFlow resnet50 time: 19.7ms (= 3949.9ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 35.3ms (= 7066.4ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.79 (= 35.3ms / 19.7ms)

OneFlow resnet50 time: 19.5ms (= 3898.3ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 30.8ms (= 6151.5ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.58 (= 30.8ms / 19.5ms)

OneFlow resnet50 time: 17.5ms (= 3494.8ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 29.4ms (= 5876.3ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.68 (= 29.4ms / 17.5ms)

OneFlow swin dataloader time: 0.202s (= 40.371s / 200, num_workers=1)
PyTorch swin dataloader time: 0.128s (= 25.660s / 200, num_workers=1)
Relative speed: 0.636 (= 0.128s / 0.202s)

OneFlow swin dataloader time: 0.053s (= 10.700s / 200, num_workers=4)
PyTorch swin dataloader time: 0.033s (= 6.550s / 200, num_workers=4)
Relative speed: 0.612 (= 0.033s / 0.053s)

OneFlow swin dataloader time: 0.031s (= 6.209s / 200, num_workers=8)
PyTorch swin dataloader time: 0.016s (= 3.296s / 200, num_workers=8)
Relative speed: 0.531 (= 0.016s / 0.031s)

❌ OneFlow resnet50 time: 47.6ms (= 4764.3ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 65.4ms (= 6544.6ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.37 (= 65.4ms / 47.6ms)

OneFlow resnet50 time: 30.9ms (= 3090.3ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 45.0ms (= 4496.2ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.45 (= 45.0ms / 30.9ms)

OneFlow resnet50 time: 24.4ms (= 4889.2ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 41.2ms (= 8236.8ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.68 (= 41.2ms / 24.4ms)

OneFlow resnet50 time: 22.8ms (= 4569.3ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 38.1ms (= 7629.5ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.67 (= 38.1ms / 22.8ms)

OneFlow resnet50 time: 18.8ms (= 3755.9ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 33.8ms (= 6763.9ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.80 (= 33.8ms / 18.8ms)

@github-actions
Copy link
Contributor

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/10322/

@jackalcooper jackalcooper requested review from oneflow-ci-bot and removed request for oneflow-ci-bot August 26, 2023 03:19
@jackalcooper jackalcooper merged commit 57c4a49 into master Aug 26, 2023
38 of 39 checks passed
@jackalcooper jackalcooper deleted the add-kMaxInputCount-in-GroupedMatmulFunctor branch August 26, 2023 03:56
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@liujuncheng liujuncheng liujuncheng approved these changes

@strint strint strint approved these changes

@hjchen2 hjchen2 hjchen2 approved these changes

@oneflow-ci-bot oneflow-ci-bot Awaiting requested review from oneflow-ci-bot

Assignees
No one assigned
Labels
automerge bug system
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
4 participants
@jackalcooper @liujuncheng @strint @hjchen2