api current_device #7856

liufengwei0103 · 2022-03-22T04:47:22Z

python/oneflow/cuda/__init__.py

python/oneflow/test/misc/test_env_cuda.py

Co-authored-by: daquexian <daquexian566@gmail.com>

github-actions · 2022-03-22T13:03:17Z

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.

github-actions · 2022-03-23T00:25:04Z

Speed stats:

GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.2ms (= 12820.1ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 140.1ms (= 14014.9ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.09 (= 140.1ms / 128.2ms)

✔️ OneFlow resnet50 time: 77.7ms (= 7769.9ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 82.8ms (= 8277.9ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.07 (= 82.8ms / 77.7ms)

OneFlow resnet50 time: 52.0ms (= 10403.2ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 60.1ms (= 12023.7ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.16 (= 60.1ms / 52.0ms)

OneFlow resnet50 time: 41.1ms (= 8214.4ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 46.4ms (= 9271.6ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.13 (= 46.4ms / 41.1ms)

OneFlow resnet50 time: 38.4ms (= 7676.5ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 41.1ms (= 8211.3ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.07 (= 41.1ms / 38.4ms)

md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
build_dataset >>>>>> ImageFolder
build_dataset >>>>>> ImageFolder
✔️ Relative speed: 1.02 (= 49.4s / 50.3s)

md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
build_dataset >>>>>> ImageFolder
build_dataset >>>>>> ImageFolder
✔️ Relative speed: 1.06 (= 13.1s / 13.9s)

md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
build_dataset >>>>>> ImageFolder
build_dataset >>>>>> ImageFolder
✔️ Relative speed: 0.98 (= 7.4s / 7.2s)

✔️ OneFlow resnet50 time: 135.7ms (= 13571.7ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 158.6ms (= 15859.2ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.17 (= 158.6ms / 135.7ms)

OneFlow resnet50 time: 87.6ms (= 8756.5ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 100.6ms (= 10057.0ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.15 (= 100.6ms / 87.6ms)

OneFlow resnet50 time: 58.8ms (= 11764.3ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 76.7ms (= 15335.8ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.30 (= 76.7ms / 58.8ms)

OneFlow resnet50 time: 52.8ms (= 10555.9ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 66.7ms (= 13334.9ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.26 (= 66.7ms / 52.8ms)

OneFlow resnet50 time: 47.9ms (= 9579.0ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 62.8ms (= 12565.2ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.31 (= 62.8ms / 47.9ms)

github-actions · 2022-03-23T00:32:16Z

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

github-actions · 2022-03-23T16:04:30Z

Speed stats:

GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.3ms (= 12834.2ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 139.9ms (= 13985.7ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.09 (= 139.9ms / 128.3ms)

✔️ OneFlow resnet50 time: 77.5ms (= 7753.7ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 80.8ms (= 8080.8ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.04 (= 80.8ms / 77.5ms)

OneFlow resnet50 time: 48.8ms (= 9761.5ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 58.1ms (= 11627.0ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.19 (= 58.1ms / 48.8ms)

OneFlow resnet50 time: 42.5ms (= 8495.4ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 43.5ms (= 8695.0ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.02 (= 43.5ms / 42.5ms)

OneFlow resnet50 time: 36.0ms (= 7190.1ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 39.5ms (= 7897.9ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.10 (= 39.5ms / 36.0ms)

md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
build_dataset >>>>>> ImageFolder
build_dataset >>>>>> ImageFolder
✔️ Relative speed: 0.99 (= 50.9s / 50.4s)

md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
build_dataset >>>>>> ImageFolder
build_dataset >>>>>> ImageFolder
✔️ Relative speed: 1.04 (= 13.3s / 13.9s)

md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
md5 /home/ci-user/ci-cache/test_cache/onerec_test/nanodataset.zip 7f5cde8b5a6c411107517ac9b00f29db
build_dataset >>>>>> ImageFolder
build_dataset >>>>>> ImageFolder
✔️ Relative speed: 0.98 (= 7.4s / 7.2s)

✔️ OneFlow resnet50 time: 135.3ms (= 13532.6ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 158.3ms (= 15826.8ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.17 (= 158.3ms / 135.3ms)

OneFlow resnet50 time: 88.0ms (= 8796.9ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 105.5ms (= 10545.3ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.20 (= 105.5ms / 88.0ms)

OneFlow resnet50 time: 58.0ms (= 11604.3ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 85.8ms (= 17167.8ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.48 (= 85.8ms / 58.0ms)

OneFlow resnet50 time: 53.5ms (= 10694.7ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 76.0ms (= 15196.0ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.42 (= 76.0ms / 53.5ms)

OneFlow resnet50 time: 50.2ms (= 10049.5ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 63.4ms (= 12682.4ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.26 (= 63.4ms / 50.2ms)

api current_device

41460d0

liufengwei0103 requested review from BBuf, daquexian and jackalcooper as code owners March 22, 2022 04:47

add rst

42f7fbd

liufengwei0103 requested a review from doombeaker as a code owner March 22, 2022 05:14

BBuf reviewed Mar 22, 2022

View reviewed changes

python/oneflow/cuda/__init__.py Show resolved Hide resolved

liufengwei0103 added 4 commits March 22, 2022 18:39

add test case about env cuda

72e1d8f

add cpu only

cfd6739

add more test

c9cb669

refine

8d38f03

BBuf approved these changes Mar 22, 2022

View reviewed changes

daquexian reviewed Mar 22, 2022

View reviewed changes

python/oneflow/test/misc/test_env_cuda.py Outdated Show resolved Hide resolved

Update python/oneflow/test/misc/test_env_cuda.py

0ce2842

Co-authored-by: daquexian <daquexian566@gmail.com>

daquexian approved these changes Mar 22, 2022

View reviewed changes

liufengwei0103 added automerge api python labels Mar 22, 2022

Merge branch 'master' into api_current_device

90b4f2a

liufengwei0103 requested a review from oneflow-ci-bot March 22, 2022 11:12

oneflow-ci-bot requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 22, 2022 11:22

oneflow-ci-bot requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 22, 2022 12:42

auto format by CI

80a862f

mergify bot added 2 commits March 22, 2022 12:26

Merge branch 'master' into api_current_device

6899fed

Merge branch 'master' into api_current_device

9543e3c

clackhan added system feature labels Mar 23, 2022

github-actions bot removed the automerge label Mar 23, 2022

liufengwei0103 requested review from oneflow-ci-bot and removed request for jackalcooper, doombeaker and oneflow-ci-bot March 23, 2022 01:47

liufengwei0103 enabled auto-merge (squash) March 23, 2022 02:01

Merge branch 'master' into api_current_device

a4c23b7

liufengwei0103 requested review from oneflow-ci-bot and removed request for oneflow-ci-bot March 23, 2022 14:17

liufengwei0103 merged commit 08b1773 into master Mar 23, 2022

liufengwei0103 deleted the api_current_device branch March 23, 2022 16:26

BBuf mentioned this pull request Apr 22, 2022

OneFlow 算子对齐 PyTorch 完备计划推进表 #4936

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

api current_device #7856

api current_device #7856

liufengwei0103 commented Mar 22, 2022 •

edited

Loading

github-actions bot commented Mar 22, 2022

github-actions bot commented Mar 23, 2022

github-actions bot commented Mar 23, 2022

github-actions bot commented Mar 23, 2022

api current_device #7856

api current_device #7856

Conversation

liufengwei0103 commented Mar 22, 2022 • edited Loading

github-actions bot commented Mar 22, 2022

github-actions bot commented Mar 23, 2022

github-actions bot commented Mar 23, 2022

github-actions bot commented Mar 23, 2022

liufengwei0103 commented Mar 22, 2022 •

edited

Loading