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_unfold_tensor_sbp_and_add_global_test #7883

Merged
merged 9 commits into from
Mar 25, 2022
Merged

fix_unfold_tensor_sbp_and_add_global_test #7883

merged 9 commits into from
Mar 25, 2022

Conversation

clackhan
Copy link
Contributor

添加unfold_tensor / unfold global 测试

@clackhan clackhan requested review from hjchen2 and wyg1997 March 23, 2022 11:31
@clackhan clackhan requested a review from oneflow-ci-bot March 24, 2022 07:23
@oneflow-ci-bot oneflow-ci-bot requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 24, 2022 15:35
@clackhan clackhan added the need-highest-priority Only add this when you really need it!!! Will block all other PRs. label Mar 25, 2022
@clackhan clackhan requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 25, 2022 01:55
@clackhan clackhan removed the need-highest-priority Only add this when you really need it!!! Will block all other PRs. label Mar 25, 2022
@clackhan clackhan requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 25, 2022 03:19
@oneflow-ci-bot oneflow-ci-bot requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 25, 2022 03:22
@github-actions
Copy link
Contributor

Speed stats: 
GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.4ms (= 12837.4ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 137.9ms (= 13789.8ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.07 (= 137.9ms / 128.4ms)

❌ OneFlow resnet50 time: 79.5ms (= 7951.0ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 84.9ms (= 8488.7ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.07 (= 84.9ms / 79.5ms)

OneFlow resnet50 time: 52.9ms (= 10572.4ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 62.4ms (= 12482.4ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.18 (= 62.4ms / 52.9ms)

OneFlow resnet50 time: 41.1ms (= 8210.5ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 50.1ms (= 10010.8ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.22 (= 50.1ms / 41.1ms)

OneFlow resnet50 time: 38.8ms (= 7760.4ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 39.7ms (= 7937.3ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.02 (= 39.7ms / 38.8ms)

md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
build_dataset >>>>>> ImageFolder
build_dataset >>>>>> ImageFolder
✔️ Relative speed: 1.07 (= 50.5s / 54.1s)

md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
build_dataset >>>>>> ImageFolder
build_dataset >>>>>> ImageFolder
✔️ Relative speed: 1.00 (= 13.3s / 13.3s)

md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
build_dataset >>>>>> ImageFolder
build_dataset >>>>>> ImageFolder
✔️ Relative speed: 1.01 (= 7.5s / 7.6s)

✔️ OneFlow resnet50 time: 135.5ms (= 13548.9ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 157.0ms (= 15700.9ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.16 (= 157.0ms / 135.5ms)

OneFlow resnet50 time: 89.2ms (= 8922.2ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 101.8ms (= 10177.2ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.14 (= 101.8ms / 89.2ms)

OneFlow resnet50 time: 62.1ms (= 12420.8ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 77.1ms (= 15423.6ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.24 (= 77.1ms / 62.1ms)

OneFlow resnet50 time: 52.0ms (= 10391.8ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 66.5ms (= 13300.7ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.28 (= 66.5ms / 52.0ms)

OneFlow resnet50 time: 50.3ms (= 10056.0ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 61.8ms (= 12365.2ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.23 (= 61.8ms / 50.3ms)

@github-actions
Copy link
Contributor

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

@clackhan clackhan requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 25, 2022 12:39
@oneflow-ci-bot oneflow-ci-bot requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 25, 2022 17:47
@github-actions
Copy link
Contributor

Speed stats:

@oneflow-ci-bot oneflow-ci-bot requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 25, 2022 22:02
@github-actions
Copy link
Contributor

Speed stats: 
GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.2ms (= 12821.5ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 140.7ms (= 14072.9ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.10 (= 140.7ms / 128.2ms)

✔️ OneFlow resnet50 time: 76.7ms (= 7666.3ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 85.3ms (= 8529.1ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.11 (= 85.3ms / 76.7ms)

OneFlow resnet50 time: 51.2ms (= 10232.0ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 61.2ms (= 12242.6ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.20 (= 61.2ms / 51.2ms)

OneFlow resnet50 time: 44.3ms (= 8861.3ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 50.7ms (= 10132.6ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.14 (= 50.7ms / 44.3ms)

OneFlow resnet50 time: 38.9ms (= 7772.3ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 39.1ms (= 7822.9ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.01 (= 39.1ms / 38.9ms)

md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
build_dataset >>>>>> ImageFolder
build_dataset >>>>>> ImageFolder
✔️ Relative speed: 1.00 (= 50.2s / 50.4s)

md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
build_dataset >>>>>> ImageFolder
build_dataset >>>>>> ImageFolder
✔️ Relative speed: 1.00 (= 13.3s / 13.2s)

md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
build_dataset >>>>>> ImageFolder
build_dataset >>>>>> ImageFolder
✔️ Relative speed: 0.97 (= 7.5s / 7.2s)

✔️ OneFlow resnet50 time: 137.1ms (= 13708.8ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 156.6ms (= 15661.1ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.14 (= 156.6ms / 137.1ms)

OneFlow resnet50 time: 86.9ms (= 8691.1ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 101.0ms (= 10104.0ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.16 (= 101.0ms / 86.9ms)

OneFlow resnet50 time: 61.4ms (= 12278.8ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 75.6ms (= 15129.1ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.23 (= 75.6ms / 61.4ms)

OneFlow resnet50 time: 49.1ms (= 9828.8ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 66.4ms (= 13275.0ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.35 (= 66.4ms / 49.1ms)

OneFlow resnet50 time: 51.3ms (= 10259.3ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 62.5ms (= 12500.0ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.22 (= 62.5ms / 51.3ms)

@mergify mergify bot merged commit af5d55e into master Mar 25, 2022
@mergify mergify bot deleted the fix_unfold_tensor_sbp_and_add_global_test branch March 25, 2022 22:39
@BBuf BBuf mentioned this pull request Apr 22, 2022
Closed
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

@wyg1997 wyg1997 Awaiting requested review from wyg1997

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

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

Successfully merging this pull request may close these issues.
3 participants
@clackhan @hjchen2 @BBuf