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.

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

Feat/mut resource eagerly #7267

Merged
merged 65 commits into from
Jan 26, 2022
Merged

Feat/mut resource eagerly #7267

merged 65 commits into from
Jan 26, 2022

Conversation

strint
Copy link
Contributor

@strint strint commented Jan 15, 2022

Set resource eagerly after session init.

@oneflow-ci-bot oneflow-ci-bot requested review from oneflow-ci-bot and removed request for oneflow-ci-bot January 25, 2022 14:55
@oneflow-ci-bot oneflow-ci-bot requested review from oneflow-ci-bot and removed request for oneflow-ci-bot January 25, 2022 17:27
@oneflow-ci-bot oneflow-ci-bot requested review from oneflow-ci-bot and removed request for oneflow-ci-bot January 25, 2022 18:16
@github-actions
Copy link
Contributor

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

@oneflow-ci-bot oneflow-ci-bot removed their request for review January 25, 2022 19:46
@strint strint requested a review from oneflow-ci-bot January 26, 2022 06:46
@github-actions
Copy link
Contributor

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.

@strint strint requested review from oneflow-ci-bot and removed request for oneflow-ci-bot January 26, 2022 06:48
@oneflow-ci-bot oneflow-ci-bot requested review from oneflow-ci-bot and removed request for oneflow-ci-bot January 26, 2022 07:53
@oneflow-ci-bot oneflow-ci-bot requested review from oneflow-ci-bot and removed request for oneflow-ci-bot January 26, 2022 08:53
@github-actions
Copy link
Contributor

Speed stats:
GPU Name: GeForce GTX 1080 

OneFlow resnet50 time: 136.7ms (= 13668.1ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 139.9ms (= 13990.6ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.02 (= 139.9ms / 136.7ms)

OneFlow resnet50 time: 78.5ms (= 7852.9ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 81.4ms (= 8142.1ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.04 (= 81.4ms / 78.5ms)

OneFlow resnet50 time: 54.1ms (= 10810.4ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 61.6ms (= 12315.9ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.14 (= 61.6ms / 54.1ms)

OneFlow resnet50 time: 41.2ms (= 8238.6ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 46.7ms (= 9342.2ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.13 (= 46.7ms / 41.2ms)

OneFlow resnet50 time: 34.6ms (= 6919.3ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 42.2ms (= 8435.0ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.22 (= 42.2ms / 34.6ms)

OneFlow resnet50 time: 148.3ms (= 14825.8ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 158.6ms (= 15860.6ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.07 (= 158.6ms / 148.3ms)

OneFlow resnet50 time: 90.9ms (= 9088.4ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 100.5ms (= 10053.9ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.11 (= 100.5ms / 90.9ms)

OneFlow resnet50 time: 66.3ms (= 13254.8ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 78.5ms (= 15709.1ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.19 (= 78.5ms / 66.3ms)

OneFlow resnet50 time: 58.9ms (= 11785.7ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 70.4ms (= 14085.7ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.20 (= 70.4ms / 58.9ms)

OneFlow resnet50 time: 61.5ms (= 12291.1ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 57.2ms (= 11435.0ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 0.93 (= 57.2ms / 61.5ms)

@oneflow-ci-bot oneflow-ci-bot removed their request for review January 26, 2022 11:42
@oneflow-ci-bot oneflow-ci-bot self-requested a review January 26, 2022 11:42
@oneflow-ci-bot oneflow-ci-bot merged commit bdaac0f into master Jan 26, 2022
@oneflow-ci-bot oneflow-ci-bot deleted the feat/mut_resource_eagerly branch January 26, 2022 13:47
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

6 participants