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

Dev optimize pooling grad #7980

Merged
merged 54 commits into from
Apr 24, 2022
Merged

Dev optimize pooling grad #7980

merged 54 commits into from
Apr 24, 2022

Conversation

Flowingsun007
Copy link
Contributor

@Flowingsun007 Flowingsun007 commented Apr 8, 2022 •
edited
Loading

@Flowingsun007 Flowingsun007 marked this pull request as ready for review April 8, 2022 02:30
@Flowingsun007 Flowingsun007 requested a review from hjchen2 as a code owner April 8, 2022 02:30
@Flowingsun007 Flowingsun007 enabled auto-merge (squash) April 8, 2022 02:30
MARD1NO
MARD1NO approved these changes Apr 11, 2022
View reviewed changes
oneflow/core/autograd/gradient_funcs/pooling.cpp
@@ -29,7 +29,6 @@ namespace {
struct PoolingCaptureState : public AutoGradCaptureState {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

把相关名字改成MaxPooling吧，包括文件名，类名这些

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

建议单独开一个 PR 来弄， review 会方便一些

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

好的

@github-actions
Copy link
Contributor

Code got formatted by CI. Please request CI again if you still want to have this PR merged. If the PR is from a forked repo, please download the patch files from the GitHub Actions web page and apply them locally.

@Flowingsun007 Flowingsun007 requested review from oneflow-ci-bot and removed request for oneflow-ci-bot April 11, 2022 06:22
@github-actions
Copy link
Contributor

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/7980/

@github-actions
Copy link
Contributor

Speed stats:

@github-actions
Copy link
Contributor

CI failed when running job: cuda-speed-test. PR label automerge has been removed

@Flowingsun007 Flowingsun007 requested review from oneflow-ci-bot and removed request for oneflow-ci-bot April 22, 2022 07:08
@Flowingsun007 Flowingsun007 requested review from oneflow-ci-bot and removed request for oneflow-ci-bot April 22, 2022 10:13
@github-actions
Copy link
Contributor

Speed stats:

@github-actions
Copy link
Contributor

CI failed when running job: cuda-speed-test. PR label automerge has been removed

@github-actions
Copy link
Contributor

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/7980/

@github-actions
Copy link
Contributor

Code got formatted by CI. Please request CI again if you still want to have this PR merged. If the PR is from a forked repo, please download the patch files from the GitHub Actions web page and apply them locally.

@github-actions
Copy link
Contributor

Speed stats: 
GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.6ms (= 12860.0ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 142.2ms (= 14222.8ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.11 (= 142.2ms / 128.6ms)

OneFlow resnet50 time: 76.4ms (= 7644.9ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 83.6ms (= 8363.3ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.09 (= 83.6ms / 76.4ms)

OneFlow resnet50 time: 54.0ms (= 10803.4ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 57.6ms (= 11512.2ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.07 (= 57.6ms / 54.0ms)

OneFlow resnet50 time: 41.9ms (= 8388.2ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 44.7ms (= 8944.1ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.07 (= 44.7ms / 41.9ms)

OneFlow resnet50 time: 36.7ms (= 7336.9ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 34.5ms (= 6899.8ms / 200, input_shape=[1, 3, 224, 224])
❌ Relative speed: 0.94 (= 34.5ms / 36.7ms)

OneFlow swin dataloader time: 0.259s (= 51.700s / 200, num_workers=1)
PyTorch swin dataloader time: 0.150s (= 29.932s / 200, num_workers=1)
Relative speed: 0.579 (= 0.150s / 0.259s)

OneFlow swin dataloader time: 0.068s (= 13.505s / 200, num_workers=4)
PyTorch swin dataloader time: 0.041s (= 8.121s / 200, num_workers=4)
Relative speed: 0.601 (= 0.041s / 0.068s)

OneFlow swin dataloader time: 0.037s (= 7.406s / 200, num_workers=8)
PyTorch swin dataloader time: 0.022s (= 4.330s / 200, num_workers=8)
Relative speed: 0.585 (= 0.022s / 0.037s)

✔️ OneFlow resnet50 time: 136.1ms (= 13610.6ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 168.2ms (= 16821.2ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.24 (= 168.2ms / 136.1ms)

OneFlow resnet50 time: 87.6ms (= 8755.1ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 101.1ms (= 10114.9ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.16 (= 101.1ms / 87.6ms)

OneFlow resnet50 time: 64.4ms (= 12885.9ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 79.1ms (= 15828.9ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.23 (= 79.1ms / 64.4ms)

OneFlow resnet50 time: 51.2ms (= 10235.6ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 69.3ms (= 13854.3ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.35 (= 69.3ms / 51.2ms)

OneFlow resnet50 time: 46.1ms (= 9227.1ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 72.3ms (= 14453.7ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.57 (= 72.3ms / 46.1ms)

@github-actions
Copy link
Contributor

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/7980/

@Flowingsun007 Flowingsun007 merged commit 28dcad1 into master Apr 24, 2022
@Flowingsun007 Flowingsun007 deleted the dev_optimize_pooling_grad branch April 24, 2022 17:56
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@wyg1997 wyg1997 wyg1997 approved these changes

@BBuf BBuf BBuf approved these changes

@MARD1NO MARD1NO MARD1NO approved these changes

@hjchen2 hjchen2 Awaiting requested review from hjchen2 hjchen2 is a code owner

@jackalcooper jackalcooper Awaiting requested review from jackalcooper jackalcooper is a code owner

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

Assignees
No one assigned
Labels
automerge enhancement op refactor
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
5 participants
@Flowingsun007 @BBuf @wyg1997 @MARD1NO @oneflow-ci-bot