Update requirements.txt #7901

lixiang007666 · 2022-03-25T12:16:16Z

This PR is done:

Fix RTD [sphinx] build docstr failure.

Related:

doombeaker · 2022-03-25T14:00:39Z

可否顺便把

# above are dev dependencies
--pre
--find-links https://staging.oneflow.info/branch/master/cpu
oneflow

删除掉。原因见 https://github.com/Oneflow-Inc/OneTeam/pull/1244/files#r834922622

lixiang007666 · 2022-03-25T14:04:36Z

可否顺便把
# above are dev dependencies

--pre

--find-links https://staging.oneflow.info/branch/master/cpu

oneflow
删除掉。原因见 https://github.com/Oneflow-Inc/OneTeam/pull/1244/files#r834922622

这个如果去掉的话，大概会对readthedocs那边的更新产生影响，是依赖这个的。

github-actions · 2022-03-28T03:01:29Z

Speed stats:

GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.4ms (= 12843.5ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 140.4ms (= 14037.3ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.09 (= 140.4ms / 128.4ms)

✔️ OneFlow resnet50 time: 76.9ms (= 7692.6ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 84.8ms (= 8479.4ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.10 (= 84.8ms / 76.9ms)

OneFlow resnet50 time: 54.1ms (= 10814.9ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 57.0ms (= 11404.5ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.05 (= 57.0ms / 54.1ms)

OneFlow resnet50 time: 44.6ms (= 8914.1ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 48.9ms (= 9770.3ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.10 (= 48.9ms / 44.6ms)

OneFlow resnet50 time: 37.0ms (= 7406.7ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 43.3ms (= 8667.2ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.17 (= 43.3ms / 37.0ms)

OneFlow swin dataloader time: 0.250s (= 50.013s / 200, num_workers=1)
PyTorch swin dataloader time: 0.255s (= 50.934s / 200, num_workers=1)
✔️ Relative speed: 1.018 (= 0.255s / 0.250s)

OneFlow swin dataloader time: 0.076s (= 15.168s / 200, num_workers=4)
PyTorch swin dataloader time: 0.069s (= 13.717s / 200, num_workers=4)
❌ Relative speed: 0.904 (= 0.069s / 0.076s)

OneFlow swin dataloader time: 0.036s (= 7.277s / 200, num_workers=8)
PyTorch swin dataloader time: 0.038s (= 7.627s / 200, num_workers=8)
✔️ Relative speed: 1.048 (= 0.038s / 0.036s)

✔️ OneFlow resnet50 time: 135.4ms (= 13542.4ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 157.2ms (= 15721.5ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.16 (= 157.2ms / 135.4ms)

OneFlow resnet50 time: 89.9ms (= 8994.3ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 100.9ms (= 10090.7ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.12 (= 100.9ms / 89.9ms)

OneFlow resnet50 time: 63.1ms (= 12624.6ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 80.7ms (= 16143.2ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.28 (= 80.7ms / 63.1ms)

OneFlow resnet50 time: 51.3ms (= 10267.6ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 76.2ms (= 15246.0ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.48 (= 76.2ms / 51.3ms)

OneFlow resnet50 time: 50.7ms (= 10135.9ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 69.9ms (= 13981.1ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.38 (= 69.9ms / 50.7ms)

github-actions · 2022-03-28T03:01:30Z

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

github-actions · 2022-03-28T11:24:33Z

Speed stats:

GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.3ms (= 12833.9ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 140.9ms (= 14091.9ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.10 (= 140.9ms / 128.3ms)

✔️ OneFlow resnet50 time: 76.6ms (= 7662.4ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 85.7ms (= 8568.2ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.12 (= 85.7ms / 76.6ms)

OneFlow resnet50 time: 51.6ms (= 10327.3ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 59.8ms (= 11960.7ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.16 (= 59.8ms / 51.6ms)

OneFlow resnet50 time: 42.6ms (= 8523.1ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 47.4ms (= 9477.4ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.11 (= 47.4ms / 42.6ms)

OneFlow resnet50 time: 36.9ms (= 7381.5ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 44.0ms (= 8805.9ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.19 (= 44.0ms / 36.9ms)

OneFlow swin dataloader time: 0.253s (= 50.528s / 200, num_workers=1)
PyTorch swin dataloader time: 0.255s (= 51.008s / 200, num_workers=1)
✔️ Relative speed: 1.010 (= 0.255s / 0.253s)

OneFlow swin dataloader time: 0.069s (= 13.705s / 200, num_workers=4)
PyTorch swin dataloader time: 0.065s (= 13.039s / 200, num_workers=4)
✔️ Relative speed: 0.951 (= 0.065s / 0.069s)

OneFlow swin dataloader time: 0.036s (= 7.299s / 200, num_workers=8)
PyTorch swin dataloader time: 0.038s (= 7.504s / 200, num_workers=8)
✔️ Relative speed: 1.028 (= 0.038s / 0.036s)

✔️ OneFlow resnet50 time: 136.0ms (= 13605.0ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 158.2ms (= 15824.4ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.16 (= 158.2ms / 136.0ms)

OneFlow resnet50 time: 87.2ms (= 8720.2ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 100.0ms (= 9998.2ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.15 (= 100.0ms / 87.2ms)

OneFlow resnet50 time: 62.5ms (= 12507.4ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 82.3ms (= 16465.5ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.32 (= 82.3ms / 62.5ms)

OneFlow resnet50 time: 54.0ms (= 10802.9ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 66.9ms (= 13377.7ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.24 (= 66.9ms / 54.0ms)

OneFlow resnet50 time: 50.7ms (= 10131.4ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 72.7ms (= 14533.3ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.43 (= 72.7ms / 50.7ms)

github-actions · 2022-03-28T11:25:56Z

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

github-actions · 2022-03-28T20:35:19Z

Speed stats:

GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.3ms (= 12825.6ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 140.0ms (= 14004.9ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.09 (= 140.0ms / 128.3ms)

✔️ OneFlow resnet50 time: 77.3ms (= 7730.0ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 82.7ms (= 8265.6ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.07 (= 82.7ms / 77.3ms)

OneFlow resnet50 time: 51.5ms (= 10306.1ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 57.0ms (= 11401.2ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.11 (= 57.0ms / 51.5ms)

OneFlow resnet50 time: 44.6ms (= 8918.7ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 48.2ms (= 9645.5ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.08 (= 48.2ms / 44.6ms)

OneFlow resnet50 time: 38.9ms (= 7780.0ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 44.3ms (= 8858.0ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.14 (= 44.3ms / 38.9ms)

OneFlow swin dataloader time: 0.252s (= 50.382s / 200, num_workers=1)
PyTorch swin dataloader time: 0.250s (= 49.985s / 200, num_workers=1)
✔️ Relative speed: 0.992 (= 0.250s / 0.252s)

OneFlow swin dataloader time: 0.066s (= 13.208s / 200, num_workers=4)
PyTorch swin dataloader time: 0.069s (= 13.896s / 200, num_workers=4)
✔️ Relative speed: 1.052 (= 0.069s / 0.066s)

OneFlow swin dataloader time: 0.038s (= 7.554s / 200, num_workers=8)
PyTorch swin dataloader time: 0.038s (= 7.543s / 200, num_workers=8)
✔️ Relative speed: 0.999 (= 0.038s / 0.038s)

✔️ OneFlow resnet50 time: 135.2ms (= 13522.4ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 152.6ms (= 15263.0ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
❌ Relative speed: 1.13 (= 152.6ms / 135.2ms)

OneFlow resnet50 time: 87.7ms (= 8769.4ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 101.1ms (= 10107.5ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.15 (= 101.1ms / 87.7ms)

OneFlow resnet50 time: 60.6ms (= 12115.3ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 76.7ms (= 15349.2ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.27 (= 76.7ms / 60.6ms)

OneFlow resnet50 time: 50.3ms (= 10059.6ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 66.5ms (= 13309.6ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.32 (= 66.5ms / 50.3ms)

OneFlow resnet50 time: 49.7ms (= 9946.2ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 60.9ms (= 12186.8ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.23 (= 60.9ms / 49.7ms)

github-actions · 2022-03-28T20:35:20Z

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

github-actions · 2022-03-29T04:26:32Z

Speed stats:

GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.4ms (= 12838.0ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 139.8ms (= 13976.1ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.09 (= 139.8ms / 128.4ms)

✔️ OneFlow resnet50 time: 78.1ms (= 7813.1ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 84.8ms (= 8483.5ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.09 (= 84.8ms / 78.1ms)

OneFlow resnet50 time: 52.3ms (= 10466.6ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 58.8ms (= 11767.3ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.12 (= 58.8ms / 52.3ms)

OneFlow resnet50 time: 43.6ms (= 8719.4ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 46.3ms (= 9257.4ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.06 (= 46.3ms / 43.6ms)

OneFlow resnet50 time: 37.7ms (= 7530.7ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 44.9ms (= 8974.3ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.19 (= 44.9ms / 37.7ms)

OneFlow swin dataloader time: 0.256s (= 51.162s / 200, num_workers=1)
PyTorch swin dataloader time: 0.254s (= 50.752s / 200, num_workers=1)
✔️ Relative speed: 0.992 (= 0.254s / 0.256s)

OneFlow swin dataloader time: 0.071s (= 14.235s / 200, num_workers=4)
PyTorch swin dataloader time: 0.070s (= 13.933s / 200, num_workers=4)
✔️ Relative speed: 0.979 (= 0.070s / 0.071s)

OneFlow swin dataloader time: 0.038s (= 7.548s / 200, num_workers=8)
PyTorch swin dataloader time: 0.036s (= 7.296s / 200, num_workers=8)
✔️ Relative speed: 0.967 (= 0.036s / 0.038s)

✔️ OneFlow resnet50 time: 135.8ms (= 13577.9ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 157.5ms (= 15754.7ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.16 (= 157.5ms / 135.8ms)

OneFlow resnet50 time: 87.4ms (= 8738.2ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 101.1ms (= 10111.1ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.16 (= 101.1ms / 87.4ms)

OneFlow resnet50 time: 62.0ms (= 12395.8ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 78.1ms (= 15628.5ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.26 (= 78.1ms / 62.0ms)

OneFlow resnet50 time: 51.8ms (= 10354.5ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 66.1ms (= 13223.1ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.28 (= 66.1ms / 51.8ms)

OneFlow resnet50 time: 51.6ms (= 10311.1ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 72.6ms (= 14529.4ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.41 (= 72.6ms / 51.6ms)

Update requirements.txt

f379857

lixiang007666 requested review from doombeaker and jackalcooper as code owners March 25, 2022 12:16

lixiang007666 added automerge bug documentation labels Mar 25, 2022

doombeaker approved these changes Mar 25, 2022

View reviewed changes

Update requirements.txt

c322d78

Update requirements.txt

a137037

lixiang007666 added the ci label Mar 26, 2022

Merge branch 'master' into lixiang007666-patch-1

32d892d

MARD1NO approved these changes Mar 28, 2022

View reviewed changes

Merge branch 'master' into lixiang007666-patch-1

d54ce0d

lixiang007666 requested a review from oneflow-ci-bot March 28, 2022 01:02

github-actions bot removed the automerge label Mar 28, 2022

lixiang007666 added the automerge label Mar 28, 2022

lixiang007666 requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 28, 2022 03:08

oneflow-ci-bot requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 28, 2022 04:04

lixiang007666 requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 28, 2022 04:47

lixiang007666 requested a review from oneflow-ci-bot March 28, 2022 10:10

github-actions bot removed the automerge label Mar 28, 2022

Merge branch 'master' into lixiang007666-patch-1

f3dd65a

lixiang007666 added the automerge label Mar 28, 2022

lixiang007666 requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 28, 2022 13:04

mergify bot added 2 commits March 28, 2022 16:52

Merge branch 'master' into lixiang007666-patch-1

7f3e788

Merge branch 'master' into lixiang007666-patch-1

a65f91a

github-actions bot removed the automerge label Mar 28, 2022

lixiang007666 requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 28, 2022 23:09

lixiang007666 added the automerge label Mar 28, 2022

mergify bot and others added 2 commits March 29, 2022 00:27

Merge branch 'master' into lixiang007666-patch-1

2dfd896

Merge branch 'master' into lixiang007666-patch-1

25a1f39

lixiang007666 requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 29, 2022 01:49

Merge branch 'master' into lixiang007666-patch-1

d4a9d30

mergify bot merged commit a910d9f into master Mar 29, 2022

mergify bot deleted the lixiang007666-patch-1 branch March 29, 2022 04:50

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Update requirements.txt #7901

Update requirements.txt #7901

lixiang007666 commented Mar 25, 2022 •

edited

Loading

doombeaker commented Mar 25, 2022

lixiang007666 commented Mar 25, 2022

github-actions bot commented Mar 28, 2022

github-actions bot commented Mar 28, 2022

github-actions bot commented Mar 28, 2022

github-actions bot commented Mar 28, 2022

github-actions bot commented Mar 28, 2022

github-actions bot commented Mar 28, 2022

github-actions bot commented Mar 29, 2022

Update requirements.txt #7901

Update requirements.txt #7901

Conversation

lixiang007666 commented Mar 25, 2022 • edited Loading

doombeaker commented Mar 25, 2022

lixiang007666 commented Mar 25, 2022

github-actions bot commented Mar 28, 2022

github-actions bot commented Mar 28, 2022

github-actions bot commented Mar 28, 2022

github-actions bot commented Mar 28, 2022

github-actions bot commented Mar 28, 2022

github-actions bot commented Mar 28, 2022

github-actions bot commented Mar 29, 2022

lixiang007666 commented Mar 25, 2022 •

edited

Loading