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 isnan isinf #7943

Merged
merged 23 commits into from
Apr 5, 2022
Merged

add isnan isinf #7943

merged 23 commits into from
Apr 5, 2022

Conversation

liufengwei0103
Copy link
Contributor

No description provided.

@liufengwei0103 liufengwei0103 marked this pull request as ready for review April 1, 2022 10:26
@liufengwei0103 liufengwei0103 marked this pull request as draft April 1, 2022 10:42
@liufengwei0103 liufengwei0103 marked this pull request as ready for review April 2, 2022 03:25
@github-actions github-actions bot removed the automerge label Apr 5, 2022
@liufengwei0103 liufengwei0103 requested review from oneflow-ci-bot and removed request for oneflow-ci-bot April 5, 2022 00:35
@github-actions
Copy link
Contributor

github-actions bot commented Apr 5, 2022

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

@github-actions github-actions bot removed the automerge label Apr 5, 2022
@github-actions
Copy link
Contributor

github-actions bot commented Apr 5, 2022

Speed stats: 
GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.3ms (= 12825.0ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 140.5ms (= 14052.3ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.10 (= 140.5ms / 128.3ms)

✔️ OneFlow resnet50 time: 78.7ms (= 7867.2ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 84.6ms (= 8462.6ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.08 (= 84.6ms / 78.7ms)

OneFlow resnet50 time: 51.4ms (= 10280.1ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 54.5ms (= 10890.5ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.06 (= 54.5ms / 51.4ms)

OneFlow resnet50 time: 42.1ms (= 8417.3ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 47.6ms (= 9521.5ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.13 (= 47.6ms / 42.1ms)

OneFlow resnet50 time: 37.6ms (= 7520.1ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 41.0ms (= 8200.1ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.09 (= 41.0ms / 37.6ms)

OneFlow swin dataloader time: 0.250s (= 49.963s / 200, num_workers=1)
PyTorch swin dataloader time: 0.251s (= 50.275s / 200, num_workers=1)
✔️ Relative speed: 1.006 (= 0.251s / 0.250s)

OneFlow swin dataloader time: 0.067s (= 13.333s / 200, num_workers=4)
PyTorch swin dataloader time: 0.067s (= 13.491s / 200, num_workers=4)
✔️ Relative speed: 1.012 (= 0.067s / 0.067s)

OneFlow swin dataloader time: 0.037s (= 7.410s / 200, num_workers=8)
PyTorch swin dataloader time: 0.038s (= 7.602s / 200, num_workers=8)
✔️ Relative speed: 1.026 (= 0.038s / 0.037s)

✔️ OneFlow resnet50 time: 135.7ms (= 13572.5ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 155.7ms (= 15566.1ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.15 (= 155.7ms / 135.7ms)

OneFlow resnet50 time: 86.3ms (= 8629.4ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 99.9ms (= 9992.6ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.16 (= 99.9ms / 86.3ms)

OneFlow resnet50 time: 59.5ms (= 11897.2ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 85.5ms (= 17094.3ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.44 (= 85.5ms / 59.5ms)

OneFlow resnet50 time: 52.1ms (= 10423.5ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 66.3ms (= 13261.5ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.27 (= 66.3ms / 52.1ms)

OneFlow resnet50 time: 50.4ms (= 10076.3ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 66.4ms (= 13282.2ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.32 (= 66.4ms / 50.4ms)

@liufengwei0103 liufengwei0103 requested review from oneflow-ci-bot and removed request for oneflow-ci-bot April 5, 2022 01:35
@github-actions
Copy link
Contributor

github-actions bot commented Apr 5, 2022

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/7943/

@oneflow-ci-bot oneflow-ci-bot requested review from oneflow-ci-bot and removed request for oneflow-ci-bot April 5, 2022 02:47
@github-actions
Copy link
Contributor

github-actions bot commented Apr 5, 2022

Speed stats:

@github-actions
Copy link
Contributor

github-actions bot commented Apr 5, 2022

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/7943/

@github-actions
Copy link
Contributor

github-actions bot commented Apr 5, 2022

Speed stats: 
GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.1ms (= 12812.3ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 140.7ms (= 14069.4ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.10 (= 140.7ms / 128.1ms)

✔️ OneFlow resnet50 time: 77.9ms (= 7786.8ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 83.3ms (= 8332.4ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.07 (= 83.3ms / 77.9ms)

OneFlow resnet50 time: 52.0ms (= 10402.0ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 53.7ms (= 10741.9ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.03 (= 53.7ms / 52.0ms)

OneFlow resnet50 time: 43.8ms (= 8768.3ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 51.0ms (= 10200.4ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.16 (= 51.0ms / 43.8ms)

OneFlow resnet50 time: 38.7ms (= 7740.9ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 42.1ms (= 8428.3ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.09 (= 42.1ms / 38.7ms)

OneFlow swin dataloader time: 0.249s (= 49.705s / 200, num_workers=1)
PyTorch swin dataloader time: 0.252s (= 50.303s / 200, num_workers=1)
✔️ Relative speed: 1.012 (= 0.252s / 0.249s)

OneFlow swin dataloader time: 0.068s (= 13.568s / 200, num_workers=4)
PyTorch swin dataloader time: 0.066s (= 13.106s / 200, num_workers=4)
✔️ Relative speed: 0.966 (= 0.066s / 0.068s)

OneFlow swin dataloader time: 0.037s (= 7.315s / 200, num_workers=8)
PyTorch swin dataloader time: 0.037s (= 7.340s / 200, num_workers=8)
✔️ Relative speed: 1.003 (= 0.037s / 0.037s)

✔️ OneFlow resnet50 time: 135.3ms (= 13529.2ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 156.1ms (= 15612.8ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.15 (= 156.1ms / 135.3ms)

OneFlow resnet50 time: 87.6ms (= 8759.0ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 98.8ms (= 9878.1ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.13 (= 98.8ms / 87.6ms)

OneFlow resnet50 time: 61.9ms (= 12386.6ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 75.0ms (= 15007.5ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.21 (= 75.0ms / 61.9ms)

OneFlow resnet50 time: 53.3ms (= 10664.2ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 65.2ms (= 13049.4ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.22 (= 65.2ms / 53.3ms)

OneFlow resnet50 time: 51.9ms (= 10371.5ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 61.9ms (= 12376.7ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.19 (= 61.9ms / 51.9ms)

@github-actions
Copy link
Contributor

github-actions bot commented Apr 5, 2022

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

@github-actions github-actions bot removed the automerge label Apr 5, 2022
@liufengwei0103 liufengwei0103 requested review from oneflow-ci-bot and removed request for oneflow-ci-bot April 5, 2022 11:45
@oneflow-ci-bot oneflow-ci-bot requested review from oneflow-ci-bot and removed request for oneflow-ci-bot April 5, 2022 18:50
@github-actions
Copy link
Contributor

github-actions bot commented Apr 5, 2022

Speed stats: 
GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.1ms (= 12805.4ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 136.4ms (= 13639.1ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.07 (= 136.4ms / 128.1ms)

OneFlow resnet50 time: 76.7ms (= 7669.9ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 85.6ms (= 8556.4ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.12 (= 85.6ms / 76.7ms)

OneFlow resnet50 time: 52.1ms (= 10429.4ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 52.9ms (= 10576.3ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.01 (= 52.9ms / 52.1ms)

OneFlow resnet50 time: 42.1ms (= 8423.7ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 50.8ms (= 10152.1ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.21 (= 50.8ms / 42.1ms)

OneFlow resnet50 time: 35.9ms (= 7188.2ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 38.2ms (= 7635.3ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.06 (= 38.2ms / 35.9ms)

OneFlow swin dataloader time: 0.258s (= 51.554s / 200, num_workers=1)
PyTorch swin dataloader time: 0.257s (= 51.382s / 200, num_workers=1)
✔️ Relative speed: 0.997 (= 0.257s / 0.258s)

OneFlow swin dataloader time: 0.067s (= 13.375s / 200, num_workers=4)
PyTorch swin dataloader time: 0.065s (= 12.970s / 200, num_workers=4)
✔️ Relative speed: 0.970 (= 0.065s / 0.067s)

OneFlow swin dataloader time: 0.036s (= 7.289s / 200, num_workers=8)
PyTorch swin dataloader time: 0.037s (= 7.403s / 200, num_workers=8)
✔️ Relative speed: 1.016 (= 0.037s / 0.036s)

✔️ OneFlow resnet50 time: 135.4ms (= 13543.6ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 155.7ms (= 15573.7ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.15 (= 155.7ms / 135.4ms)

OneFlow resnet50 time: 86.6ms (= 8663.6ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 99.0ms (= 9899.5ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.14 (= 99.0ms / 86.6ms)

OneFlow resnet50 time: 60.7ms (= 12139.4ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 75.5ms (= 15107.3ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.24 (= 75.5ms / 60.7ms)

OneFlow resnet50 time: 52.4ms (= 10470.6ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 76.4ms (= 15281.4ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.46 (= 76.4ms / 52.4ms)

OneFlow resnet50 time: 48.8ms (= 9766.4ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 61.4ms (= 12277.4ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.26 (= 61.4ms / 48.8ms)

@github-actions
Copy link
Contributor

github-actions bot commented Apr 5, 2022

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/7943/

@mergify mergify bot merged commit 7fe29cb into master Apr 5, 2022
@mergify mergify bot deleted the support_isnan_isinf branch April 5, 2022 20:37
@zhongshsh zhongshsh 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

@Flowingsun007 Flowingsun007 Flowingsun007 approved these changes

@BBuf BBuf BBuf approved these changes

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

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

Assignees
No one assigned
Labels
api automerge feature op
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@liufengwei0103 @Flowingsun007 @BBuf