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 fill test #8118

Merged
merged 42 commits into from
May 11, 2022
Merged

Add fill test #8118

merged 42 commits into from
May 11, 2022

Conversation

ShawnXuan
Copy link
Collaborator

No description provided.

liujuncheng and others added 22 commits April 26, 2022 18:49
@liujuncheng
[Primitive] Unit test
1f2cd7d
@liujuncheng
Merge branch 'master' into dev_primitive_unit_test
a5fd294
@liujuncheng
Merge branch 'master' into dev_primitive_unit_test
9a88d41
@mergify
Merge branch 'master' into dev_primitive_unit_test
b0fd5a8
@mergify
Merge branch 'master' into dev_primitive_unit_test
2165165
@mergify
Merge branch 'master' into dev_primitive_unit_test
cc3a6c3
@mergify
Merge branch 'master' into dev_primitive_unit_test
cd7f5ed
@guo-ran
Merge branch 'master' into dev_primitive_unit_test
75b2fbe
@mergify
Merge branch 'master' into dev_primitive_unit_test
21c9470
@mergify
Merge branch 'master' into dev_primitive_unit_test
d7e5f26
@mergify
Merge branch 'master' into dev_primitive_unit_test
c0fe5fa
@mergify
Merge branch 'master' into dev_primitive_unit_test
2ad3fdf
@mergify
Merge branch 'master' into dev_primitive_unit_test
21462a5
@mergify
Merge branch 'master' into dev_primitive_unit_test
e156cb1
@mergify
Merge branch 'master' into dev_primitive_unit_test
7902483
@mergify
Merge branch 'master' into dev_primitive_unit_test
f2bdf3d
@ShawnXuan
add fill test
228d523
@ShawnXuan
update
faee075
@ShawnXuan
Merge branch 'master' of github.com:Oneflow-Inc/oneflow into dev_fill…
6546553 
…_1hot_test
@ShawnXuan
update
daf0c4b
@ShawnXuan
Merge branch 'master' of github.com:Oneflow-Inc/oneflow into dev_fill…
8dfe9ea 
…_1hot_test
@ShawnXuan
update
0b7941e
@ShawnXuan ShawnXuan marked this pull request as ready for review May 6, 2022 08:45
@ShawnXuan ShawnXuan requested a review from liujuncheng as a code owner May 6, 2022 08:45
@ShawnXuan ShawnXuan requested a review from guo-ran May 6, 2022 10:48
@ShawnXuan ShawnXuan changed the title Dev fill 1hot test Add fill test May 6, 2022
@github-actions
Copy link
Contributor

github-actions bot commented May 9, 2022

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/8118/

@github-actions
Copy link
Contributor

github-actions bot commented May 9, 2022

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

@github-actions github-actions bot removed the automerge label May 9, 2022
@oneflow-ci-bot oneflow-ci-bot requested review from oneflow-ci-bot and removed request for oneflow-ci-bot May 9, 2022 19:47
@github-actions
Copy link
Contributor

github-actions bot commented May 9, 2022

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/8118/

@github-actions
Copy link
Contributor

github-actions bot commented May 9, 2022

Speed stats: 
GPU Name: NVIDIA GeForce GTX 1080 

❌ OneFlow resnet50 time: 129.4ms (= 12938.3ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 143.3ms (= 14334.9ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.11 (= 143.3ms / 129.4ms)

OneFlow resnet50 time: 77.5ms (= 7750.1ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 85.7ms (= 8570.8ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.11 (= 85.7ms / 77.5ms)

OneFlow resnet50 time: 54.1ms (= 10826.7ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 57.9ms (= 11585.3ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.07 (= 57.9ms / 54.1ms)

OneFlow resnet50 time: 42.7ms (= 8537.6ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 44.3ms (= 8852.3ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.04 (= 44.3ms / 42.7ms)

OneFlow resnet50 time: 37.9ms (= 7586.8ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 38.8ms (= 7766.5ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.02 (= 38.8ms / 37.9ms)

OneFlow swin dataloader time: 0.258s (= 51.679s / 200, num_workers=1)
PyTorch swin dataloader time: 0.151s (= 30.226s / 200, num_workers=1)
Relative speed: 0.585 (= 0.151s / 0.258s)

OneFlow swin dataloader time: 0.069s (= 13.712s / 200, num_workers=4)
PyTorch swin dataloader time: 0.044s (= 8.776s / 200, num_workers=4)
Relative speed: 0.640 (= 0.044s / 0.069s)

OneFlow swin dataloader time: 0.036s (= 7.209s / 200, num_workers=8)
PyTorch swin dataloader time: 0.022s (= 4.423s / 200, num_workers=8)
Relative speed: 0.614 (= 0.022s / 0.036s)

❌ OneFlow resnet50 time: 145.2ms (= 14518.3ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 167.8ms (= 16782.6ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.16 (= 167.8ms / 145.2ms)

OneFlow resnet50 time: 100.9ms (= 10090.2ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 113.1ms (= 11314.3ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.12 (= 113.1ms / 100.9ms)

OneFlow resnet50 time: 74.5ms (= 14895.0ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 87.3ms (= 17454.7ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
❌ Relative speed: 1.17 (= 87.3ms / 74.5ms)

OneFlow resnet50 time: 62.0ms (= 12403.2ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 74.6ms (= 14922.8ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.20 (= 74.6ms / 62.0ms)

OneFlow resnet50 time: 56.7ms (= 11345.1ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 78.5ms (= 15703.2ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.38 (= 78.5ms / 56.7ms)

@github-actions
Copy link
Contributor

github-actions bot commented May 9, 2022

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

@github-actions github-actions bot removed the automerge label May 9, 2022
@github-actions
Copy link
Contributor

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/8118/

@github-actions
Copy link
Contributor

Speed stats: 
GPU Name: NVIDIA GeForce GTX 1080 

❌ OneFlow resnet50 time: 129.3ms (= 12933.3ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 142.6ms (= 14257.7ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.10 (= 142.6ms / 129.3ms)

OneFlow resnet50 time: 84.1ms (= 8412.1ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 86.5ms (= 8645.8ms / 100, input_shape=[8, 3, 224, 224])
❌ Relative speed: 1.03 (= 86.5ms / 84.1ms)

OneFlow resnet50 time: 54.0ms (= 10797.1ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 62.3ms (= 12466.2ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.15 (= 62.3ms / 54.0ms)

OneFlow resnet50 time: 44.7ms (= 8946.8ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 42.6ms (= 8516.4ms / 200, input_shape=[2, 3, 224, 224])
❌ Relative speed: 0.95 (= 42.6ms / 44.7ms)

OneFlow resnet50 time: 40.6ms (= 8126.1ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 39.3ms (= 7860.3ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 0.97 (= 39.3ms / 40.6ms)

OneFlow swin dataloader time: 0.259s (= 51.867s / 200, num_workers=1)
PyTorch swin dataloader time: 0.151s (= 30.211s / 200, num_workers=1)
Relative speed: 0.582 (= 0.151s / 0.259s)

OneFlow swin dataloader time: 0.069s (= 13.756s / 200, num_workers=4)
PyTorch swin dataloader time: 0.042s (= 8.455s / 200, num_workers=4)
Relative speed: 0.615 (= 0.042s / 0.069s)

OneFlow swin dataloader time: 0.060s (= 11.910s / 200, num_workers=8)
PyTorch swin dataloader time: 0.024s (= 4.769s / 200, num_workers=8)
Relative speed: 0.400 (= 0.024s / 0.060s)

❌ OneFlow resnet50 time: 145.4ms (= 14542.1ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 169.1ms (= 16907.9ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.16 (= 169.1ms / 145.4ms)

OneFlow resnet50 time: 99.4ms (= 9942.1ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 112.5ms (= 11245.3ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.13 (= 112.5ms / 99.4ms)

OneFlow resnet50 time: 75.1ms (= 15011.6ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 88.6ms (= 17719.8ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.18 (= 88.6ms / 75.1ms)

OneFlow resnet50 time: 63.5ms (= 12695.5ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 72.9ms (= 14570.2ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
❌ Relative speed: 1.15 (= 72.9ms / 63.5ms)

OneFlow resnet50 time: 54.6ms (= 10924.0ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 78.8ms (= 15751.0ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.44 (= 78.8ms / 54.6ms)

@github-actions
Copy link
Contributor

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/8118/

@github-actions
Copy link
Contributor

Speed stats: 
GPU Name: NVIDIA GeForce GTX 1080 

❌ OneFlow resnet50 time: 129.2ms (= 12919.2ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 142.7ms (= 14268.9ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.10 (= 142.7ms / 129.2ms)

OneFlow resnet50 time: 84.1ms (= 8411.8ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 84.4ms (= 8436.3ms / 100, input_shape=[8, 3, 224, 224])
❌ Relative speed: 1.00 (= 84.4ms / 84.1ms)

OneFlow resnet50 time: 54.4ms (= 10872.2ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 58.9ms (= 11781.1ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.08 (= 58.9ms / 54.4ms)

OneFlow resnet50 time: 41.8ms (= 8351.2ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 39.8ms (= 7950.6ms / 200, input_shape=[2, 3, 224, 224])
❌ Relative speed: 0.95 (= 39.8ms / 41.8ms)

OneFlow resnet50 time: 39.5ms (= 7901.6ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 34.6ms (= 6920.4ms / 200, input_shape=[1, 3, 224, 224])
❌ Relative speed: 0.88 (= 34.6ms / 39.5ms)

OneFlow swin dataloader time: 0.259s (= 51.764s / 200, num_workers=1)
PyTorch swin dataloader time: 0.151s (= 30.123s / 200, num_workers=1)
Relative speed: 0.582 (= 0.151s / 0.259s)

OneFlow swin dataloader time: 0.069s (= 13.791s / 200, num_workers=4)
PyTorch swin dataloader time: 0.042s (= 8.481s / 200, num_workers=4)
Relative speed: 0.615 (= 0.042s / 0.069s)

OneFlow swin dataloader time: 0.059s (= 11.765s / 200, num_workers=8)
PyTorch swin dataloader time: 0.023s (= 4.538s / 200, num_workers=8)
Relative speed: 0.386 (= 0.023s / 0.059s)

❌ OneFlow resnet50 time: 144.8ms (= 14479.4ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 168.2ms (= 16815.8ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.16 (= 168.2ms / 144.8ms)

OneFlow resnet50 time: 96.3ms (= 9628.3ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 111.2ms (= 11118.1ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.15 (= 111.2ms / 96.3ms)

OneFlow resnet50 time: 78.3ms (= 15659.3ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 88.4ms (= 17675.0ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
❌ Relative speed: 1.13 (= 88.4ms / 78.3ms)

OneFlow resnet50 time: 64.7ms (= 12939.0ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 76.7ms (= 15332.6ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.18 (= 76.7ms / 64.7ms)

OneFlow resnet50 time: 56.1ms (= 11223.1ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 69.2ms (= 13830.3ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.23 (= 69.2ms / 56.1ms)

@github-actions
Copy link
Contributor

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/8118/

@github-actions
Copy link
Contributor

Speed stats: 
GPU Name: NVIDIA GeForce GTX 1080 

❌ OneFlow resnet50 time: 129.1ms (= 12912.0ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 142.7ms (= 14274.6ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.11 (= 142.7ms / 129.1ms)

OneFlow resnet50 time: 81.6ms (= 8155.1ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 83.4ms (= 8342.2ms / 100, input_shape=[8, 3, 224, 224])
❌ Relative speed: 1.02 (= 83.4ms / 81.6ms)

OneFlow resnet50 time: 54.5ms (= 10903.2ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 56.9ms (= 11373.8ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.04 (= 56.9ms / 54.5ms)

OneFlow resnet50 time: 41.5ms (= 8306.0ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 40.9ms (= 8173.5ms / 200, input_shape=[2, 3, 224, 224])
❌ Relative speed: 0.98 (= 40.9ms / 41.5ms)

OneFlow resnet50 time: 36.6ms (= 7310.8ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 42.1ms (= 8427.6ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.15 (= 42.1ms / 36.6ms)

OneFlow swin dataloader time: 0.258s (= 51.516s / 200, num_workers=1)
PyTorch swin dataloader time: 0.150s (= 30.092s / 200, num_workers=1)
Relative speed: 0.584 (= 0.150s / 0.258s)

OneFlow swin dataloader time: 0.114s (= 22.805s / 200, num_workers=4)
PyTorch swin dataloader time: 0.042s (= 8.425s / 200, num_workers=4)
Relative speed: 0.369 (= 0.042s / 0.114s)

OneFlow swin dataloader time: 0.036s (= 7.263s / 200, num_workers=8)
PyTorch swin dataloader time: 0.022s (= 4.409s / 200, num_workers=8)
Relative speed: 0.607 (= 0.022s / 0.036s)

❌ OneFlow resnet50 time: 145.6ms (= 14558.4ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 169.0ms (= 16903.4ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.16 (= 169.0ms / 145.6ms)

OneFlow resnet50 time: 98.9ms (= 9892.1ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 119.3ms (= 11933.5ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.21 (= 119.3ms / 98.9ms)

OneFlow resnet50 time: 76.9ms (= 15374.8ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 88.4ms (= 17671.4ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
❌ Relative speed: 1.15 (= 88.4ms / 76.9ms)

OneFlow resnet50 time: 62.7ms (= 12531.4ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 74.9ms (= 14985.9ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.20 (= 74.9ms / 62.7ms)

OneFlow resnet50 time: 55.8ms (= 11150.4ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 68.9ms (= 13789.4ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.24 (= 68.9ms / 55.8ms)

@ShawnXuan ShawnXuan merged commit fa1d9b8 into master May 11, 2022
@ShawnXuan ShawnXuan deleted the dev_fill_1hot_test branch May 11, 2022 08:51
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@guo-ran guo-ran guo-ran approved these changes

@MARD1NO MARD1NO MARD1NO 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
enhancement op
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
4 participants
@ShawnXuan @guo-ran @MARD1NO @liujuncheng