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 circular import issue #8268

Merged
merged 3 commits into from
May 20, 2022
Merged

fix circular import issue #8268

merged 3 commits into from
May 20, 2022

Conversation

xiacijie
Copy link
Contributor

Bug report: #8265

@xiacijie xiacijie requested review from strint and marigoold May 20, 2022 04:04
@xiacijie xiacijie requested a review from oneflow-ci-bot May 20, 2022 04:15
@xiacijie xiacijie enabled auto-merge (squash) May 20, 2022 04:15
@xiacijie xiacijie added the graph graph mode label May 20, 2022
@github-actions
Copy link
Contributor

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/8268/

@github-actions
Copy link
Contributor

CI failed when running job: cuda-module. 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/8268/

@github-actions
Copy link
Contributor

Speed stats: 
GPU Name: NVIDIA GeForce GTX 1080 

❌ OneFlow resnet50 time: 130.0ms (= 12999.8ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 145.5ms (= 14549.9ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.12 (= 145.5ms / 130.0ms)

OneFlow resnet50 time: 77.5ms (= 7748.2ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 84.5ms (= 8454.4ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.09 (= 84.5ms / 77.5ms)

OneFlow resnet50 time: 51.7ms (= 10336.6ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 60.3ms (= 12060.7ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.17 (= 60.3ms / 51.7ms)

OneFlow resnet50 time: 42.4ms (= 8472.0ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 44.8ms (= 8961.9ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.06 (= 44.8ms / 42.4ms)

OneFlow resnet50 time: 37.4ms (= 7471.0ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 36.6ms (= 7314.6ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 0.98 (= 36.6ms / 37.4ms)

OneFlow swin dataloader time: 0.247s (= 49.484s / 200, num_workers=1)
PyTorch swin dataloader time: 0.150s (= 30.096s / 200, num_workers=1)
Relative speed: 0.608 (= 0.150s / 0.247s)

OneFlow swin dataloader time: 0.068s (= 13.617s / 200, num_workers=4)
PyTorch swin dataloader time: 0.043s (= 8.505s / 200, num_workers=4)
Relative speed: 0.625 (= 0.043s / 0.068s)

OneFlow swin dataloader time: 0.054s (= 10.882s / 200, num_workers=8)
PyTorch swin dataloader time: 0.023s (= 4.521s / 200, num_workers=8)
Relative speed: 0.415 (= 0.023s / 0.054s)

❌ OneFlow resnet50 time: 146.0ms (= 14599.8ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 170.8ms (= 17077.1ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.17 (= 170.8ms / 146.0ms)

OneFlow resnet50 time: 97.6ms (= 9758.1ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 112.7ms (= 11274.7ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.16 (= 112.7ms / 97.6ms)

OneFlow resnet50 time: 75.0ms (= 15006.1ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 95.4ms (= 19077.8ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.27 (= 95.4ms / 75.0ms)

OneFlow resnet50 time: 60.6ms (= 12118.3ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 73.9ms (= 14772.7ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.22 (= 73.9ms / 60.6ms)

OneFlow resnet50 time: 54.4ms (= 10890.0ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 69.2ms (= 13834.1ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.27 (= 69.2ms / 54.4ms)

@github-actions
Copy link
Contributor

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/8268/

@github-actions
Copy link
Contributor

Speed stats: 
GPU Name: NVIDIA GeForce GTX 1080 

❌ OneFlow resnet50 time: 130.1ms (= 13008.2ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 143.1ms (= 14305.9ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.10 (= 143.1ms / 130.1ms)

OneFlow resnet50 time: 77.5ms (= 7754.6ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 84.5ms (= 8445.0ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.09 (= 84.5ms / 77.5ms)

OneFlow resnet50 time: 53.3ms (= 10656.2ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 59.6ms (= 11917.0ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.12 (= 59.6ms / 53.3ms)

OneFlow resnet50 time: 43.6ms (= 8715.2ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 45.8ms (= 9152.0ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.05 (= 45.8ms / 43.6ms)

OneFlow resnet50 time: 39.5ms (= 7908.3ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 35.7ms (= 7141.0ms / 200, input_shape=[1, 3, 224, 224])
❌ Relative speed: 0.90 (= 35.7ms / 39.5ms)

OneFlow swin dataloader time: 0.248s (= 49.572s / 200, num_workers=1)
PyTorch swin dataloader time: 0.152s (= 30.428s / 200, num_workers=1)
Relative speed: 0.614 (= 0.152s / 0.248s)

OneFlow swin dataloader time: 0.101s (= 20.206s / 200, num_workers=4)
PyTorch swin dataloader time: 0.043s (= 8.628s / 200, num_workers=4)
Relative speed: 0.427 (= 0.043s / 0.101s)

OneFlow swin dataloader time: 0.036s (= 7.221s / 200, num_workers=8)
PyTorch swin dataloader time: 0.022s (= 4.350s / 200, num_workers=8)
Relative speed: 0.602 (= 0.022s / 0.036s)

❌ OneFlow resnet50 time: 147.5ms (= 14746.5ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 167.6ms (= 16758.8ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.14 (= 167.6ms / 147.5ms)

OneFlow resnet50 time: 97.0ms (= 9698.1ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 112.2ms (= 11216.8ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.16 (= 112.2ms / 97.0ms)

OneFlow resnet50 time: 73.8ms (= 14761.0ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 88.1ms (= 17626.7ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.19 (= 88.1ms / 73.8ms)

OneFlow resnet50 time: 61.6ms (= 12315.9ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 74.7ms (= 14933.5ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.21 (= 74.7ms / 61.6ms)

OneFlow resnet50 time: 53.7ms (= 10733.2ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 72.7ms (= 14539.5ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.35 (= 72.7ms / 53.7ms)

@xiacijie xiacijie merged commit f8cf5cc into master May 20, 2022
@xiacijie xiacijie deleted the fix_python3_6_circular_import_issue_8265 branch May 20, 2022 13:58
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@strint strint strint approved these changes

@marigoold marigoold marigoold approved these changes

@BBuf BBuf Awaiting requested review from BBuf

@daquexian daquexian Awaiting requested review from daquexian

@jackalcooper jackalcooper Awaiting requested review from jackalcooper

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

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

Successfully merging this pull request may close these issues.
3 participants
@xiacijie @strint @marigoold