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

eager global moving average max min observer #7726

Merged
merged 6 commits into from
Mar 29, 2022
Merged

eager global moving average max min observer #7726

merged 6 commits into from
Mar 29, 2022

Conversation

liufengwei0103
Copy link
Contributor

No description provided.

@liufengwei0103 liufengwei0103 requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 28, 2022 03:23
@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.

@liufengwei0103 liufengwei0103 requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 28, 2022 05:56
@github-actions
Copy link
Contributor

Speed stats: 
GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.3ms (= 12833.1ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 140.7ms (= 14069.8ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.10 (= 140.7ms / 128.3ms)

✔️ OneFlow resnet50 time: 77.2ms (= 7720.7ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 86.2ms (= 8616.9ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.12 (= 86.2ms / 77.2ms)

OneFlow resnet50 time: 51.1ms (= 10226.3ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 59.0ms (= 11807.2ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.15 (= 59.0ms / 51.1ms)

OneFlow resnet50 time: 41.6ms (= 8327.9ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 48.0ms (= 9592.0ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.15 (= 48.0ms / 41.6ms)

OneFlow resnet50 time: 39.4ms (= 7888.2ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 37.8ms (= 7559.0ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 0.96 (= 37.8ms / 39.4ms)

OneFlow swin dataloader time: 0.258s (= 51.500s / 200, num_workers=1)
PyTorch swin dataloader time: 0.253s (= 50.557s / 200, num_workers=1)
✔️ Relative speed: 0.982 (= 0.253s / 0.258s)

OneFlow swin dataloader time: 0.069s (= 13.741s / 200, num_workers=4)
PyTorch swin dataloader time: 0.069s (= 13.865s / 200, num_workers=4)
✔️ Relative speed: 1.009 (= 0.069s / 0.069s)

OneFlow swin dataloader time: 0.036s (= 7.216s / 200, num_workers=8)
PyTorch swin dataloader time: 0.037s (= 7.383s / 200, num_workers=8)
✔️ Relative speed: 1.023 (= 0.037s / 0.036s)

✔️ OneFlow resnet50 time: 135.6ms (= 13560.2ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 163.2ms (= 16319.1ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.20 (= 163.2ms / 135.6ms)

OneFlow resnet50 time: 87.9ms (= 8787.5ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 100.8ms (= 10079.0ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.15 (= 100.8ms / 87.9ms)

OneFlow resnet50 time: 61.4ms (= 12284.0ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 75.4ms (= 15071.5ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.23 (= 75.4ms / 61.4ms)

OneFlow resnet50 time: 50.7ms (= 10139.5ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 77.9ms (= 15570.9ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.54 (= 77.9ms / 50.7ms)

OneFlow resnet50 time: 47.9ms (= 9583.7ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 62.0ms (= 12407.0ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.29 (= 62.0ms / 47.9ms)

@github-actions
Copy link
Contributor

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

@liufengwei0103
Copy link
Contributor Author

action time-out

@liufengwei0103 liufengwei0103 removed the request for review from oneflow-ci-bot March 29, 2022 03:39
@github-actions
Copy link
Contributor

Speed stats: 
GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.4ms (= 12835.5ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 139.4ms (= 13943.5ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.09 (= 139.4ms / 128.4ms)

✔️ OneFlow resnet50 time: 78.5ms (= 7854.7ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 85.3ms (= 8531.4ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.09 (= 85.3ms / 78.5ms)

OneFlow resnet50 time: 51.5ms (= 10304.8ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 61.6ms (= 12311.6ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.19 (= 61.6ms / 51.5ms)

OneFlow resnet50 time: 42.5ms (= 8500.0ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 45.3ms (= 9064.7ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.07 (= 45.3ms / 42.5ms)

OneFlow resnet50 time: 39.5ms (= 7908.0ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 38.3ms (= 7664.1ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 0.97 (= 38.3ms / 39.5ms)

OneFlow swin dataloader time: 0.252s (= 50.404s / 200, num_workers=1)
PyTorch swin dataloader time: 0.255s (= 50.944s / 200, num_workers=1)
✔️ Relative speed: 1.011 (= 0.255s / 0.252s)

OneFlow swin dataloader time: 0.066s (= 13.107s / 200, num_workers=4)
PyTorch swin dataloader time: 0.066s (= 13.260s / 200, num_workers=4)
✔️ Relative speed: 1.012 (= 0.066s / 0.066s)

OneFlow swin dataloader time: 0.037s (= 7.305s / 200, num_workers=8)
PyTorch swin dataloader time: 0.038s (= 7.549s / 200, num_workers=8)
✔️ Relative speed: 1.033 (= 0.038s / 0.037s)

✔️ OneFlow resnet50 time: 135.8ms (= 13581.9ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 155.6ms (= 15562.0ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.15 (= 155.6ms / 135.8ms)

OneFlow resnet50 time: 87.3ms (= 8731.7ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 102.2ms (= 10219.6ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.17 (= 102.2ms / 87.3ms)

OneFlow resnet50 time: 61.2ms (= 12246.2ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 76.6ms (= 15320.0ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.25 (= 76.6ms / 61.2ms)

OneFlow resnet50 time: 53.0ms (= 10596.4ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 76.0ms (= 15201.3ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.43 (= 76.0ms / 53.0ms)

OneFlow resnet50 time: 51.3ms (= 10251.3ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 62.5ms (= 12492.7ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.22 (= 62.5ms / 51.3ms)

@mergify mergify bot merged commit 0f45edd into master Mar 29, 2022
@mergify mergify bot deleted the eager_global_moving_average_min_max_observer branch March 29, 2022 06:10
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@hjchen2 hjchen2 hjchen2 approved these changes

@BBuf BBuf BBuf approved these changes

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

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

Successfully merging this pull request may close these issues.
4 participants
@liufengwei0103 @hjchen2 @BBuf @oneflow-ci-bot