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

Skip Conv cases failed #10383

Merged
merged 10 commits into from
Dec 21, 2023
Merged

Skip Conv cases failed #10383

merged 10 commits into from
Dec 21, 2023

Conversation

jackalcooper
Copy link
Collaborator

No description provided.

@github-actions GitHub Actions
Copy link
Contributor

View latest API docs preview at: https://oneflow-staging.oss-cn-beijing.aliyuncs.com/docs/Oneflow-Inc/oneflow/pr/10383/

@github-actions GitHub Actions
Copy link
Contributor

View latest API docs preview at: https://oneflow-staging.oss-cn-beijing.aliyuncs.com/docs/Oneflow-Inc/oneflow/pr/10383/

@github-actions GitHub Actions
Copy link
Contributor

Speed stats: 
GPU Name: NVIDIA GeForce RTX 3080 Ti 

❌ OneFlow resnet50 time: 43.6ms (= 4357.5ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 57.3ms (= 5725.4ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.31 (= 57.3ms / 43.6ms)

OneFlow resnet50 time: 26.1ms (= 2605.3ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 37.6ms (= 3761.6ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.44 (= 37.6ms / 26.1ms)

OneFlow resnet50 time: 18.5ms (= 3697.4ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 36.7ms (= 7345.6ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.99 (= 36.7ms / 18.5ms)

OneFlow resnet50 time: 17.1ms (= 3419.7ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 31.9ms (= 6385.3ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.87 (= 31.9ms / 17.1ms)

OneFlow resnet50 time: 16.4ms (= 3283.9ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 31.3ms (= 6269.2ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.91 (= 31.3ms / 16.4ms)

OneFlow swin dataloader time: 0.199s (= 39.750s / 200, num_workers=1)
PyTorch swin dataloader time: 0.128s (= 25.527s / 200, num_workers=1)
Relative speed: 0.642 (= 0.128s / 0.199s)

OneFlow swin dataloader time: 0.053s (= 10.682s / 200, num_workers=4)
PyTorch swin dataloader time: 0.033s (= 6.569s / 200, num_workers=4)
Relative speed: 0.615 (= 0.033s / 0.053s)

OneFlow swin dataloader time: 0.030s (= 5.932s / 200, num_workers=8)
PyTorch swin dataloader time: 0.017s (= 3.341s / 200, num_workers=8)
Relative speed: 0.563 (= 0.017s / 0.030s)

❌ OneFlow resnet50 time: 49.3ms (= 4925.0ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 66.1ms (= 6609.2ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.34 (= 66.1ms / 49.3ms)

OneFlow resnet50 time: 37.1ms (= 3705.9ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 46.7ms (= 4667.0ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.26 (= 46.7ms / 37.1ms)

OneFlow resnet50 time: 27.9ms (= 5582.3ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 42.3ms (= 8460.3ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.52 (= 42.3ms / 27.9ms)

OneFlow resnet50 time: 25.4ms (= 5073.8ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 38.5ms (= 7696.6ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.52 (= 38.5ms / 25.4ms)

OneFlow resnet50 time: 24.0ms (= 4808.1ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 36.7ms (= 7334.9ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.53 (= 36.7ms / 24.0ms)

@github-actions 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 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 GitHub Actions
Copy link
Contributor

View latest API docs preview at: https://oneflow-staging.oss-cn-beijing.aliyuncs.com/docs/Oneflow-Inc/oneflow/pr/10383/

@github-actions GitHub Actions
Copy link
Contributor

Speed stats:

@jackalcooper jackalcooper requested review from oneflow-ci-bot and removed request for oneflow-ci-bot December 21, 2023 03:48
@github-actions GitHub Actions
Copy link
Contributor

View latest API docs preview at: https://oneflow-staging.oss-cn-beijing.aliyuncs.com/docs/Oneflow-Inc/oneflow/pr/10383/

@github-actions GitHub Actions
Copy link
Contributor

View latest API docs preview at: https://oneflow-staging.oss-cn-beijing.aliyuncs.com/docs/Oneflow-Inc/oneflow/pr/10383/

@github-actions GitHub Actions
Copy link
Contributor

Speed stats: 
GPU Name: NVIDIA GeForce RTX 3080 Ti 

❌ OneFlow resnet50 time: 43.9ms (= 4387.1ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 57.8ms (= 5783.6ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.32 (= 57.8ms / 43.9ms)

OneFlow resnet50 time: 26.2ms (= 2621.7ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 37.6ms (= 3760.4ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.43 (= 37.6ms / 26.2ms)

OneFlow resnet50 time: 19.4ms (= 3870.2ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 35.0ms (= 6997.3ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.81 (= 35.0ms / 19.4ms)

OneFlow resnet50 time: 17.9ms (= 3575.5ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 34.6ms (= 6919.7ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.94 (= 34.6ms / 17.9ms)

OneFlow resnet50 time: 18.3ms (= 3652.6ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 29.3ms (= 5855.8ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.60 (= 29.3ms / 18.3ms)

OneFlow swin dataloader time: 0.201s (= 40.184s / 200, num_workers=1)
PyTorch swin dataloader time: 0.129s (= 25.785s / 200, num_workers=1)
Relative speed: 0.642 (= 0.129s / 0.201s)

OneFlow swin dataloader time: 0.053s (= 10.651s / 200, num_workers=4)
PyTorch swin dataloader time: 0.033s (= 6.528s / 200, num_workers=4)
Relative speed: 0.613 (= 0.033s / 0.053s)

OneFlow swin dataloader time: 0.030s (= 5.977s / 200, num_workers=8)
PyTorch swin dataloader time: 0.017s (= 3.313s / 200, num_workers=8)
Relative speed: 0.554 (= 0.017s / 0.030s)

❌ OneFlow resnet50 time: 49.2ms (= 4918.2ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 64.9ms (= 6493.5ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.32 (= 64.9ms / 49.2ms)

OneFlow resnet50 time: 36.4ms (= 3643.2ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 44.6ms (= 4461.9ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.22 (= 44.6ms / 36.4ms)

OneFlow resnet50 time: 28.1ms (= 5619.8ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 40.4ms (= 8087.3ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.44 (= 40.4ms / 28.1ms)

OneFlow resnet50 time: 25.1ms (= 5013.0ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 39.4ms (= 7878.6ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.57 (= 39.4ms / 25.1ms)

OneFlow resnet50 time: 24.0ms (= 4796.2ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 38.0ms (= 7592.3ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.58 (= 38.0ms / 24.0ms)

@jackalcooper jackalcooper enabled auto-merge (squash) December 21, 2023 05:42
levi131
levi131 approved these changes Dec 21, 2023
View reviewed changes
Copy link
Contributor

@levi131 levi131 left a comment

Choose a reason for hiding this comment

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

LGTM，跳过case后续再研究怎么修复

@jackalcooper jackalcooper merged commit 6f53149 into master Dec 21, 2023
21 checks passed
@jackalcooper jackalcooper deleted the skip-test_convmixer_compatibility branch December 21, 2023 05:49
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@marigoold marigoold marigoold approved these changes

@levi131 levi131 levi131 approved these changes

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

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

Successfully merging this pull request may close these issues.
4 participants
@jackalcooper @marigoold @levi131 @oneflow-ci-bot