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 docs ZZK Part #7848

Merged
merged 58 commits into from
Apr 11, 2022
Merged

Add docs ZZK Part #7848

merged 58 commits into from
Apr 11, 2022

Conversation

MARD1NO
Copy link
Contributor

@MARD1NO MARD1NO commented Mar 21, 2022 •
edited
Loading

Adamw support_sparse

image

Optimizer

step

image

support sparse

image

SGD

step
support sparse

image

LambdaLR

step

image

ReduceLROnPlateau

in_cooldown
is_better

image

fused mlp

image

TODO：FusedMLP 其实很多多余的方法被导出，用户无需感知，暂未解决屏蔽文档部分API的方法

@MARD1NO MARD1NO requested a review from oneflow-ci-bot March 21, 2022 08:13
@oneflow-ci-bot oneflow-ci-bot requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 21, 2022 10:27
@github-actions
Copy link
Contributor

CI failed when running job: cuda-module-distributed-rank-0. PR label automerge has been removed

@MARD1NO MARD1NO requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 23, 2022 01:00
@github-actions
Copy link
Contributor

github-actions bot commented Apr 2, 2022

Speed stats: 
GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.5ms (= 12845.3ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 142.2ms (= 14221.9ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.11 (= 142.2ms / 128.5ms)

✔️ OneFlow resnet50 time: 77.2ms (= 7721.8ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 86.4ms (= 8641.8ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.12 (= 86.4ms / 77.2ms)

OneFlow resnet50 time: 53.5ms (= 10700.8ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 59.0ms (= 11796.3ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.10 (= 59.0ms / 53.5ms)

OneFlow resnet50 time: 44.3ms (= 8854.1ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 49.4ms (= 9883.7ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.12 (= 49.4ms / 44.3ms)

OneFlow resnet50 time: 39.9ms (= 7980.0ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 40.0ms (= 8007.2ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.00 (= 40.0ms / 39.9ms)

OneFlow swin dataloader time: 0.259s (= 51.835s / 200, num_workers=1)
PyTorch swin dataloader time: 0.255s (= 50.983s / 200, num_workers=1)
✔️ Relative speed: 0.984 (= 0.255s / 0.259s)

OneFlow swin dataloader time: 0.066s (= 13.296s / 200, num_workers=4)
PyTorch swin dataloader time: 0.066s (= 13.250s / 200, num_workers=4)
✔️ Relative speed: 0.997 (= 0.066s / 0.066s)

OneFlow swin dataloader time: 0.037s (= 7.491s / 200, num_workers=8)
PyTorch swin dataloader time: 0.039s (= 7.714s / 200, num_workers=8)
✔️ Relative speed: 1.030 (= 0.039s / 0.037s)

❌ OneFlow resnet50 time: 138.9ms (= 13894.4ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 158.0ms (= 15800.6ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.14 (= 158.0ms / 138.9ms)

OneFlow resnet50 time: 88.6ms (= 8861.7ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 102.7ms (= 10267.8ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.16 (= 102.7ms / 88.6ms)

OneFlow resnet50 time: 63.1ms (= 12614.3ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 76.2ms (= 15247.2ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.21 (= 76.2ms / 63.1ms)

OneFlow resnet50 time: 52.6ms (= 10513.3ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 66.5ms (= 13292.7ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.26 (= 66.5ms / 52.6ms)

OneFlow resnet50 time: 52.5ms (= 10509.1ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 66.8ms (= 13368.2ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.27 (= 66.8ms / 52.5ms)

@github-actions
Copy link
Contributor

github-actions bot commented Apr 2, 2022

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

@github-actions github-actions bot removed the automerge label Apr 2, 2022
@MARD1NO MARD1NO removed the request for review from oneflow-ci-bot April 6, 2022 03:37
@MARD1NO MARD1NO requested a review from oneflow-ci-bot April 6, 2022 03:37
@github-actions
Copy link
Contributor

github-actions bot commented Apr 6, 2022

CI failed when running job: cuda-module-distributed-rank-0. PR label automerge has been removed

@github-actions github-actions bot removed the automerge label Apr 6, 2022
@github-actions
Copy link
Contributor

github-actions bot commented Apr 6, 2022

CI failed when running job: cuda-module-distributed-rank-1. PR label automerge has been removed

@MARD1NO MARD1NO removed the request for review from oneflow-ci-bot April 10, 2022 02:30
@MARD1NO MARD1NO requested a review from oneflow-ci-bot April 10, 2022 02:30
@github-actions
Copy link
Contributor

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/7848/

@github-actions
Copy link
Contributor

Speed stats: 
GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.6ms (= 12864.9ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 139.2ms (= 13915.0ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.08 (= 139.2ms / 128.6ms)

OneFlow resnet50 time: 76.2ms (= 7624.7ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 84.6ms (= 8456.2ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.11 (= 84.6ms / 76.2ms)

OneFlow resnet50 time: 54.9ms (= 10985.9ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 55.5ms (= 11090.4ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.01 (= 55.5ms / 54.9ms)

OneFlow resnet50 time: 41.6ms (= 8327.1ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 45.2ms (= 9038.2ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.09 (= 45.2ms / 41.6ms)

OneFlow resnet50 time: 37.9ms (= 7586.5ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 42.0ms (= 8394.7ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.11 (= 42.0ms / 37.9ms)

OneFlow swin dataloader time: 0.251s (= 50.273s / 200, num_workers=1)
PyTorch swin dataloader time: 0.253s (= 50.604s / 200, num_workers=1)
✔️ Relative speed: 1.007 (= 0.253s / 0.251s)

OneFlow swin dataloader time: 0.070s (= 14.076s / 200, num_workers=4)
PyTorch swin dataloader time: 0.067s (= 13.378s / 200, num_workers=4)
✔️ Relative speed: 0.950 (= 0.067s / 0.070s)

OneFlow swin dataloader time: 0.035s (= 7.080s / 200, num_workers=8)
PyTorch swin dataloader time: 0.037s (= 7.417s / 200, num_workers=8)
✔️ Relative speed: 1.048 (= 0.037s / 0.035s)

✔️ OneFlow resnet50 time: 135.4ms (= 13540.9ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 156.0ms (= 15597.2ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.15 (= 156.0ms / 135.4ms)

OneFlow resnet50 time: 85.5ms (= 8553.6ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 102.4ms (= 10242.3ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.20 (= 102.4ms / 85.5ms)

OneFlow resnet50 time: 61.7ms (= 12331.0ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 85.9ms (= 17187.7ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.39 (= 85.9ms / 61.7ms)

OneFlow resnet50 time: 49.2ms (= 9846.3ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 66.2ms (= 13234.8ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.34 (= 66.2ms / 49.2ms)

OneFlow resnet50 time: 46.2ms (= 9240.0ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 61.2ms (= 12238.7ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.32 (= 61.2ms / 46.2ms)

@mergify mergify bot merged commit 397600b into master Apr 11, 2022
@mergify mergify bot deleted the add_docs_zzk_part branch April 11, 2022 13:06
Flowingsun007 added a commit that referenced this pull request Apr 12, 2022
@Flowingsun007
Revert "add docs (#7848)"
bccba42 
This reverts commit 397600b.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@simonJJJ simonJJJ simonJJJ approved these changes

@BBuf BBuf BBuf approved these changes

@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 documentation op
Projects
None yet
Milestone
No milestone
Development

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