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.

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

Cherry-pick IR changes #9430

Merged
merged 20 commits into from
Nov 18, 2022
Merged

Cherry-pick IR changes #9430

merged 20 commits into from
Nov 18, 2022

Conversation

jackalcooper
Copy link
Collaborator

No description provided.

@jackalcooper jackalcooper marked this pull request as ready for review November 17, 2022 03:51
@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.

@jackalcooper jackalcooper requested review from oneflow-ci-bot and removed request for oneflow-ci-bot November 17, 2022 03:52
@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.

@jackalcooper jackalcooper changed the title Cherry pick IR changes Cherry-pick IR changes Nov 17, 2022
@github-actions
Copy link
Contributor

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

@github-actions
Copy link
Contributor

Speed stats:

@github-actions
Copy link
Contributor

Speed stats:
GPU Name: GeForce GTX 1080 









❌ OneFlow resnet50 time: 141.8ms (= 14175.4ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 164.6ms (= 16458.2ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.16 (= 164.6ms / 141.8ms)

OneFlow resnet50 time: 86.6ms (= 8657.8ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 112.0ms (= 11204.8ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.29 (= 112.0ms / 86.6ms)

OneFlow resnet50 time: 58.3ms (= 11655.6ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 87.8ms (= 17569.8ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.51 (= 87.8ms / 58.3ms)

OneFlow resnet50 time: 45.0ms (= 8993.2ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 71.6ms (= 14325.6ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.59 (= 71.6ms / 45.0ms)

OneFlow resnet50 time: 41.6ms (= 8319.6ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 66.6ms (= 13312.1ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.60 (= 66.6ms / 41.6ms)

@github-actions
Copy link
Contributor

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/9430/

@github-actions
Copy link
Contributor

Speed stats:
GPU Name: GeForce GTX 1080 









❌ OneFlow resnet50 time: 140.4ms (= 14044.3ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 161.8ms (= 16178.8ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.15 (= 161.8ms / 140.4ms)

OneFlow resnet50 time: 85.2ms (= 8515.8ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 111.4ms (= 11135.5ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.31 (= 111.4ms / 85.2ms)

OneFlow resnet50 time: 57.8ms (= 11553.6ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 87.3ms (= 17468.1ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.51 (= 87.3ms / 57.8ms)

OneFlow resnet50 time: 44.2ms (= 8833.2ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 71.0ms (= 14198.4ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.61 (= 71.0ms / 44.2ms)

OneFlow resnet50 time: 41.4ms (= 8289.5ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 69.4ms (= 13882.9ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.67 (= 69.4ms / 41.4ms)

@github-actions
Copy link
Contributor

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

@github-actions
Copy link
Contributor

Speed stats:
GPU Name: GeForce GTX 1080 









❌ OneFlow resnet50 time: 139.8ms (= 13980.9ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 161.0ms (= 16100.3ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.15 (= 161.0ms / 139.8ms)

OneFlow resnet50 time: 85.7ms (= 8574.4ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 105.9ms (= 10593.6ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.24 (= 105.9ms / 85.7ms)

OneFlow resnet50 time: 57.6ms (= 11526.0ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 77.3ms (= 15468.4ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.34 (= 77.3ms / 57.6ms)

OneFlow resnet50 time: 45.8ms (= 9152.3ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 68.5ms (= 13692.6ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.50 (= 68.5ms / 45.8ms)

OneFlow resnet50 time: 40.6ms (= 8114.7ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 73.4ms (= 14687.1ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.81 (= 73.4ms / 40.6ms)

@github-actions
Copy link
Contributor

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/9430/

@mergify mergify bot merged commit 0185f14 into master Nov 18, 2022
@mergify mergify bot deleted the cherry-pick-ir-changes branch November 18, 2022 00:52
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants