Create a quantized version ReLU function for CUDA #85502

fufeisi · 2022-09-22T20:39:49Z

Summary:
this is to allow the relu function to run on a quantized tensor on cuda. That is torch.relu(qa) for a quantized tensor qa on cuda.
Test Plan:
python test/test_quantization.py

pytorch-bot · 2022-09-22T20:39:51Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/85502

📄 Preview Python docs built from this PR
📄 Preview C++ docs built from this PR
❓ Need help or want to give feedback on the CI? Visit our office hours

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit ed6bd36:
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

dzdang · 2022-09-22T21:15:13Z

is this PR ready for review?

fufeisi · 2022-09-22T21:20:54Z

is this PR ready for review?

Yes.

test/quantization/core/test_quantized_op.py

jerryzh168 · 2022-09-23T00:58:23Z

We extend the relu function for quantized tensor on cuda. Now we can call torch.relu(qa) for a quantized tensor qa on cuda.

Can you use standard format for summary, e.g. #84984? i.e. include a Summary and Test Plan at least

fufeisi · 2022-09-23T01:04:09Z

We extend the relu function for quantized tensor on cuda. Now we can call torch.relu(qa) for a quantized tensor qa on cuda.

Can you use standard format for summary, e.g. #84984? i.e. include a Summary and Test Plan at least

Sure, I changed and will use a standard format for future PRs.

test/quantization/core/test_quantized_op.py

relu_quantized_cuda.

…d_cuda_.

fufeisi · 2022-09-24T05:57:33Z

@pytorchbot merge

pytorchmergebot · 2022-09-24T05:59:10Z

@pytorchbot successfully started a merge job. Check the current status here.
The merge job was triggered without a flag. This means that your change will be merged once all checks on your PR have passed (ETA: 0-4 Hours). If this is not the intended behavior, feel free to use some of the other merge options in the wiki.
Please reach out to the PyTorch DevX Team with feedback or questions!

github-actions · 2022-09-24T06:00:04Z

Hey @fufeisi.
You've committed this PR, but it does not have both a 'release notes: ...' and 'topics: ...' label. Please add one of each to the PR. The 'release notes: ...' label should represent the part of PyTorch that this PR changes (fx, autograd, distributed, etc) and the 'topics: ...' label should represent the kind of PR it is (not user facing, new feature, bug fix, perf improvement, etc). The list of valid labels can be found here for the 'release notes: ...' and here for the 'topics: ...'.
For changes that are 'topic: not user facing' there is no need for a release notes label.

janeyx99 · 2022-09-24T12:33:57Z

@pytorchbot revert -m "Sorry, reverting as 10.2 builds on trunk broke due to this change, see https://hud.pytorch.org/pytorch/pytorch/commit/93a53ff4d92c883d87cc7aee35af719039b481a8" -c nosignal

Running with land checks would prevent a revert like this in the future, for more details see https://github.com/pytorch/pytorch/wiki/Bot-commands.

pytorchmergebot · 2022-09-24T12:35:18Z

@pytorchbot successfully started a revert job. Check the current status here.
Please reach out to the PyTorch DevX Team with feedback or questions!

pytorchmergebot · 2022-09-24T12:35:24Z

@fufeisi your PR has been successfully reverted.

This reverts commit 93a53ff. Reverted #85502 on behalf of https://github.com/janeyx99 due to Sorry, reverting as 10.2 builds on trunk broke due to this change, see https://hud.pytorch.org/pytorch/pytorch/commit/93a53ff4d92c883d87cc7aee35af719039b481a8

…eLU function, relu_quantized_cuda_." Summary: this and #85669 are to allow the relu function to run on a quantized tensor on cuda. That is torch.relu(qa) for a quantized tensor qa on cuda. Test Plan: python test/test_quantization.py Previous PR that has been reverted: #85502. [ghstack-poisoned]

…DA ReLU function," Summary: this and #85670 are to allow the relu function to run on a quantized tensor on cuda. That is torch.relu(qa) for a quantized tensor qa on cuda. Test Plan: python test/test_quantization.py Previous PR that has been reverted: #85502. [ghstack-poisoned]

…lu_quantized_cuda_." Summary: this and #85669 are to allow the relu function to run on a quantized tensor on cuda. That is torch.relu(qa) for a quantized tensor qa on cuda. Test Plan: python test/test_quantization.py Previous PR that has been reverted: #85502. [ghstack-poisoned]

Summary: this and #85670 are to allow the relu function to run on a quantized tensor on cuda. That is torch.relu(qa) for a quantized tensor qa on cuda. Test Plan: python test/test_quantization.py Previous PR that has been reverted: #85502. [ghstack-poisoned]

…DA ReLU function," Summary: this and #85670 are to allow the relu function to run on a quantized tensor on cuda. That is torch.relu(qa) for a quantized tensor qa on cuda. Test Plan: python test/test_quantization.py Previous PR that has been reverted: #85502. [ghstack-poisoned]

…eLU function, relu_quantized_cuda_." Summary: this and #85669 are to allow the relu function to run on a quantized tensor on cuda. That is torch.relu(qa) for a quantized tensor qa on cuda. Test Plan: python test/test_quantization.py Previous PR that has been reverted: #85502. [ghstack-poisoned]

…lu_quantized_cuda_." Summary: this and #85669 are to allow the relu function to run on a quantized tensor on cuda. That is torch.relu(qa) for a quantized tensor qa on cuda. Test Plan: python test/test_quantization.py Previous PR that has been reverted: #85502. [ghstack-poisoned]

Summary: this and #85670 are to allow the relu function to run on a quantized tensor on cuda. That is torch.relu(qa) for a quantized tensor qa on cuda. Test Plan: python test/test_quantization.py Previous PR that has been reverted: #85502. [ghstack-poisoned]

…DA ReLU function," Summary: this and #85670 are to allow the relu function to run on a quantized tensor on cuda. That is torch.relu(qa) for a quantized tensor qa on cuda. Test Plan: python test/test_quantization.py Previous PR that has been reverted: #85502. [ghstack-poisoned]

Summary: this and #85670 are to allow the relu function to run on a quantized tensor on cuda. That is torch.relu(qa) for a quantized tensor qa on cuda. Test Plan: python test/test_quantization.py Previous PR that has been reverted: #85502. [ghstack-poisoned]

Summary: this and #85670 are to allow the relu function to run on a quantized tensor on cuda. That is torch.relu(qa) for a quantized tensor qa on cuda. Test Plan: python test/test_quantization.py Previous PR that has been reverted: #85502. Pull Request resolved: #85669 Approved by: https://github.com/dzdang

…eLU function, relu_quantized_cuda_." Summary: this and #85669 are to allow the relu function to run on a quantized tensor on cuda. That is torch.relu(qa) for a quantized tensor qa on cuda. Test Plan: python test/test_quantization.py Previous PR that has been reverted: #85502. [ghstack-poisoned]

…lu_quantized_cuda_." Summary: this and #85669 are to allow the relu function to run on a quantized tensor on cuda. That is torch.relu(qa) for a quantized tensor qa on cuda. Test Plan: python test/test_quantization.py Previous PR that has been reverted: #85502. [ghstack-poisoned]

…d_cuda_. (#85670) Summary: this and #85669 are to allow the relu function to run on a quantized tensor on cuda. That is torch.relu(qa) for a quantized tensor qa on cuda. Test Plan: python test/test_quantization.py Previous PR that has been reverted: #85502. Pull Request resolved: #85670 Approved by: https://github.com/dzdang, https://github.com/z-a-f

Summary: this is to allow the relu function to run on a quantized tensor on cuda. That is torch.relu(qa) for a quantized tensor qa on cuda. Test Plan: python test/test_quantization.py Pull Request resolved: #85502 Approved by: https://github.com/dzdang

This reverts commit 93a53ff. Reverted #85502 on behalf of https://github.com/janeyx99 due to Sorry, reverting as 10.2 builds on trunk broke due to this change, see https://hud.pytorch.org/pytorch/pytorch/commit/93a53ff4d92c883d87cc7aee35af719039b481a8

Summary: this and #85670 are to allow the relu function to run on a quantized tensor on cuda. That is torch.relu(qa) for a quantized tensor qa on cuda. Test Plan: python test/test_quantization.py Previous PR that has been reverted: #85502. Pull Request resolved: #85669 Approved by: https://github.com/dzdang

…d_cuda_. (#85670) Summary: this and #85669 are to allow the relu function to run on a quantized tensor on cuda. That is torch.relu(qa) for a quantized tensor qa on cuda. Test Plan: python test/test_quantization.py Previous PR that has been reverted: #85502. Pull Request resolved: #85670 Approved by: https://github.com/dzdang, https://github.com/z-a-f

fufeisi requested review from dzdang, jerryzh168, vkuzo and z-a-f as code owners September 22, 2022 20:39

pytorch-bot bot added the release notes: quantization release notes category label Sep 22, 2022

facebook-github-bot added the cla signed label Sep 22, 2022

fufeisi force-pushed the cuda_qrelu branch from c940327 to c2bc5d9 Compare September 22, 2022 20:42

jerryzh168 reviewed Sep 22, 2022

View reviewed changes

test/quantization/core/test_quantized_op.py Outdated Show resolved Hide resolved

fufeisi force-pushed the cuda_qrelu branch from c2bc5d9 to 5fb5b24 Compare September 23, 2022 00:30

dzdang approved these changes Sep 23, 2022

View reviewed changes

dzdang reviewed Sep 23, 2022

View reviewed changes

test/quantization/core/test_quantized_op.py Outdated Show resolved Hide resolved

fufeisi force-pushed the cuda_qrelu branch from 5fb5b24 to 5af8d82 Compare September 23, 2022 03:18

jerryzh168 reviewed Sep 23, 2022

View reviewed changes

test/quantization/core/test_quantized_op.py Outdated Show resolved Hide resolved

fufeisi force-pushed the cuda_qrelu branch from 5af8d82 to bf07512 Compare September 23, 2022 17:09

fufeisi added 2 commits September 23, 2022 17:12

Create a quantized non-in-palce version CUDA ReLU function,

fa41b7c

relu_quantized_cuda.

Create a quantized in-palce version CUDA ReLU function, relu_quantize…

ed6bd36

…d_cuda_.

fufeisi force-pushed the cuda_qrelu branch from bf07512 to ed6bd36 Compare September 23, 2022 17:16

fufeisi requested a review from jerryzh168 September 23, 2022 20:50

pytorchmergebot added the Merged label Sep 24, 2022

pytorchmergebot closed this in 93a53ff Sep 24, 2022

pytorchmergebot added the Reverted label Sep 24, 2022

This was referenced Sep 28, 2022

Create a quantized non-in-palce version CUDA ReLU function, #85669

Closed

Create a quantized in-palce version CUDA ReLU function, relu_quantized_cuda_. #85670

Closed

Create a quantized version ReLU function for CUDA #85502

Create a quantized version ReLU function for CUDA #85502

Uh oh!

Conversation

fufeisi commented Sep 22, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pytorch-bot bot commented Sep 22, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/85502

✅ No Failures

Uh oh!

dzdang commented Sep 22, 2022

Uh oh!

fufeisi commented Sep 22, 2022

Uh oh!

Uh oh!

Uh oh!

jerryzh168 commented Sep 23, 2022

Uh oh!

fufeisi commented Sep 23, 2022

Uh oh!

Uh oh!

fufeisi commented Sep 24, 2022

Uh oh!

pytorchmergebot commented Sep 24, 2022

Uh oh!

github-actions bot commented Sep 24, 2022

Uh oh!

janeyx99 commented Sep 24, 2022

Uh oh!

pytorchmergebot commented Sep 24, 2022

Uh oh!

pytorchmergebot commented Sep 24, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

fufeisi commented Sep 22, 2022 •

edited

Loading

pytorch-bot bot commented Sep 22, 2022 •

edited

Loading