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

Add binary functor exception #8161

Merged
merged 39 commits into from
May 15, 2022
Merged

Add binary functor exception #8161

merged 39 commits into from
May 15, 2022

Conversation

BBuf
Copy link
Contributor

@BBuf BBuf commented May 7, 2022

No description provided.

@BBuf BBuf requested a review from oneflow-ci-bot May 9, 2022 09:24
@BBuf BBuf added the eager label May 9, 2022
@oneflow-ci-bot oneflow-ci-bot requested review from oneflow-ci-bot and removed request for oneflow-ci-bot May 9, 2022 19:47
@github-actions
Copy link
Contributor

github-actions bot commented May 9, 2022

CI failed when running job: cpu-module. PR label automerge has been removed

@github-actions github-actions bot removed the automerge label May 9, 2022
@github-actions
Copy link
Contributor

github-actions bot commented May 9, 2022

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

@BBuf BBuf added the automerge label May 10, 2022
@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 May 14, 2022 20:35
@github-actions
Copy link
Contributor

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/8161/

@github-actions
Copy link
Contributor

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

@github-actions
Copy link
Contributor

Speed stats:

@BBuf BBuf added the automerge label May 15, 2022
@github-actions
Copy link
Contributor

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/8161/

@github-actions
Copy link
Contributor

Speed stats: 
GPU Name: NVIDIA GeForce GTX 1080 

❌ OneFlow resnet50 time: 129.5ms (= 12948.0ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 143.4ms (= 14336.1ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.11 (= 143.4ms / 129.5ms)

OneFlow resnet50 time: 80.6ms (= 8063.4ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 84.0ms (= 8404.9ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.04 (= 84.0ms / 80.6ms)

OneFlow resnet50 time: 52.0ms (= 10398.1ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 63.5ms (= 12707.4ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.22 (= 63.5ms / 52.0ms)

OneFlow resnet50 time: 41.1ms (= 8228.8ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 42.6ms (= 8514.0ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.03 (= 42.6ms / 41.1ms)

OneFlow resnet50 time: 35.3ms (= 7053.9ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 38.3ms (= 7658.8ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.09 (= 38.3ms / 35.3ms)

OneFlow swin dataloader time: 0.244s (= 48.820s / 200, num_workers=1)
PyTorch swin dataloader time: 0.155s (= 31.069s / 200, num_workers=1)
Relative speed: 0.636 (= 0.155s / 0.244s)

OneFlow swin dataloader time: 0.065s (= 12.968s / 200, num_workers=4)
PyTorch swin dataloader time: 0.041s (= 8.205s / 200, num_workers=4)
Relative speed: 0.633 (= 0.041s / 0.065s)

OneFlow swin dataloader time: 0.035s (= 6.934s / 200, num_workers=8)
PyTorch swin dataloader time: 0.022s (= 4.383s / 200, num_workers=8)
Relative speed: 0.632 (= 0.022s / 0.035s)

❌ OneFlow resnet50 time: 145.5ms (= 14550.3ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 168.2ms (= 16823.9ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.16 (= 168.2ms / 145.5ms)

OneFlow resnet50 time: 99.0ms (= 9899.4ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 110.9ms (= 11094.2ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.12 (= 110.9ms / 99.0ms)

OneFlow resnet50 time: 77.2ms (= 15436.0ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 99.6ms (= 19912.4ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.29 (= 99.6ms / 77.2ms)

OneFlow resnet50 time: 65.3ms (= 13051.5ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 77.7ms (= 15544.9ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.19 (= 77.7ms / 65.3ms)

OneFlow resnet50 time: 58.5ms (= 11702.2ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 76.9ms (= 15377.5ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.31 (= 76.9ms / 58.5ms)

@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/8161/

@github-actions
Copy link
Contributor

Speed stats: 
GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.5ms (= 12853.6ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 142.2ms (= 14223.5ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.11 (= 142.2ms / 128.5ms)

OneFlow resnet50 time: 77.0ms (= 7697.5ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 86.1ms (= 8605.9ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.12 (= 86.1ms / 77.0ms)

OneFlow resnet50 time: 52.8ms (= 10561.3ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 61.1ms (= 12229.1ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.16 (= 61.1ms / 52.8ms)

OneFlow resnet50 time: 40.7ms (= 8143.3ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 46.9ms (= 9380.1ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.15 (= 46.9ms / 40.7ms)

OneFlow resnet50 time: 36.9ms (= 7381.1ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 41.9ms (= 8370.6ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.13 (= 41.9ms / 36.9ms)

OneFlow swin dataloader time: 0.390s (= 77.936s / 200, num_workers=1)
PyTorch swin dataloader time: 0.153s (= 30.683s / 200, num_workers=1)
Relative speed: 0.394 (= 0.153s / 0.390s)

OneFlow swin dataloader time: 0.100s (= 20.089s / 200, num_workers=4)
PyTorch swin dataloader time: 0.043s (= 8.575s / 200, num_workers=4)
Relative speed: 0.427 (= 0.043s / 0.100s)

OneFlow swin dataloader time: 0.037s (= 7.362s / 200, num_workers=8)
PyTorch swin dataloader time: 0.022s (= 4.380s / 200, num_workers=8)
Relative speed: 0.595 (= 0.022s / 0.037s)

❌ OneFlow resnet50 time: 137.4ms (= 13738.4ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 163.1ms (= 16312.1ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.19 (= 163.1ms / 137.4ms)

OneFlow resnet50 time: 86.2ms (= 8615.4ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 101.3ms (= 10134.8ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.18 (= 101.3ms / 86.2ms)

OneFlow resnet50 time: 60.7ms (= 12144.3ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 87.8ms (= 17554.7ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.45 (= 87.8ms / 60.7ms)

OneFlow resnet50 time: 50.3ms (= 10060.7ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 69.5ms (= 13891.2ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.38 (= 69.5ms / 50.3ms)

OneFlow resnet50 time: 48.1ms (= 9629.5ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 69.6ms (= 13913.4ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.44 (= 69.6ms / 48.1ms)

@github-actions
Copy link
Contributor

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

@BBuf BBuf added the automerge label May 15, 2022
@BBuf BBuf requested review from oneflow-ci-bot and removed request for oneflow-ci-bot May 15, 2022 14:22
@github-actions
Copy link
Contributor

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/8161/

@github-actions
Copy link
Contributor

Speed stats: 
GPU Name: NVIDIA GeForce GTX 1080 

❌ OneFlow resnet50 time: 129.7ms (= 12968.0ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 142.4ms (= 14237.6ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.10 (= 142.4ms / 129.7ms)

OneFlow resnet50 time: 76.3ms (= 7628.7ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 86.6ms (= 8656.2ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.13 (= 86.6ms / 76.3ms)

OneFlow resnet50 time: 50.9ms (= 10173.9ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 56.5ms (= 11294.6ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.11 (= 56.5ms / 50.9ms)

OneFlow resnet50 time: 42.2ms (= 8431.2ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 46.7ms (= 9332.5ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.11 (= 46.7ms / 42.2ms)

OneFlow resnet50 time: 35.9ms (= 7172.5ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 37.4ms (= 7486.1ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.04 (= 37.4ms / 35.9ms)

OneFlow swin dataloader time: 0.246s (= 49.107s / 200, num_workers=1)
PyTorch swin dataloader time: 0.149s (= 29.783s / 200, num_workers=1)
Relative speed: 0.606 (= 0.149s / 0.246s)

OneFlow swin dataloader time: 0.063s (= 12.516s / 200, num_workers=4)
PyTorch swin dataloader time: 0.041s (= 8.235s / 200, num_workers=4)
Relative speed: 0.658 (= 0.041s / 0.063s)

OneFlow swin dataloader time: 0.057s (= 11.302s / 200, num_workers=8)
PyTorch swin dataloader time: 0.022s (= 4.356s / 200, num_workers=8)
Relative speed: 0.385 (= 0.022s / 0.057s)

❌ OneFlow resnet50 time: 145.5ms (= 14553.2ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 173.2ms (= 17324.0ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.19 (= 173.2ms / 145.5ms)

OneFlow resnet50 time: 95.7ms (= 9567.3ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 111.5ms (= 11154.6ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.17 (= 111.5ms / 95.7ms)

OneFlow resnet50 time: 75.2ms (= 15033.2ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 98.3ms (= 19658.4ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.31 (= 98.3ms / 75.2ms)

OneFlow resnet50 time: 63.3ms (= 12659.5ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 75.2ms (= 15049.7ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.19 (= 75.2ms / 63.3ms)

OneFlow resnet50 time: 55.5ms (= 11096.1ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 69.0ms (= 13806.9ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.24 (= 69.0ms / 55.5ms)

@mergify mergify bot merged commit 5e6451c into master May 15, 2022
@mergify mergify bot deleted the add_binary_functor_exception branch May 15, 2022 16:10
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@MARD1NO MARD1NO MARD1NO approved these changes

@liufengwei0103 liufengwei0103 liufengwei0103 approved these changes

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

@daquexian daquexian Awaiting requested review from daquexian

@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
api automerge bug eager enhancement
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@BBuf @liufengwei0103 @MARD1NO