[quant][fx] Support override observers and fake quantize module in backend_config_dict #75135

jerryzh168 · 2022-04-01T20:48:09Z

Stack from ghstack (oldest at bottom):

Summary:
Some operators have fixed quantization parameters, this PR adds the support to override the
qconfig in the backend_config_dict

Test Plan:
python test/test_quantization.py TestQuantizeFx
python test/test_quantization.py TestQuantizeFxOps

Reviewers:

Subscribers:

Tasks:

Tags:

Differential Revision: D35334279

…ckend_config_dict Summary: Some operators have fixed quantization parameters, this PR adds the support to override the qconfig in the backend_config_dict Test Plan: python test/test_quantization.py TestQuantizeFx python test/test_quantization.py TestQuantizeFxOps Reviewers: Subscribers: Tasks: Tags: [ghstack-poisoned]

facebook-github-bot · 2022-04-01T20:48:14Z

🔗 Helpful links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/75135
📄 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

💊 CI failures summary and remediations

As of commit 7a8e312 (more details on the Dr. CI page):

💚 💚 Looks good so far! There are no failures yet. 💚 💚

This comment was automatically generated by Dr. CI (expand for details).

Please report bugs/suggestions to the (internal) Dr. CI Users group.

Click here to manually regenerate this comment.

…odule in backend_config_dict" Summary: Some operators have fixed quantization parameters, this PR adds the support to override the qconfig in the backend_config_dict Test Plan: python test/test_quantization.py TestQuantizeFx python test/test_quantization.py TestQuantizeFxOps Reviewers: Subscribers: Tasks: Tags: [ghstack-poisoned]

…ckend_config_dict Summary: Some operators have fixed quantization parameters, this PR adds the support to override the qconfig in the backend_config_dict Test Plan: python test/test_quantization.py TestQuantizeFx python test/test_quantization.py TestQuantizeFxOps Reviewers: Subscribers: Tasks: Tags: ghstack-source-id: 1dbd088 Pull Request resolved: #75135

jerryzh168 · 2022-04-02T00:53:29Z

@jerryzh168 has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

torch/ao/quantization/fx/backend_config/native.py

…odule in backend_config_dict" Summary: Some operators have fixed quantization parameters, this PR adds the support to override the qconfig in the backend_config_dict Test Plan: python test/test_quantization.py TestQuantizeFx python test/test_quantization.py TestQuantizeFxOps Reviewers: Subscribers: Tasks: Tags: Differential Revision: [D35334279](https://our.internmc.facebook.com/intern/diff/D35334279) [ghstack-poisoned]

jerryzh168 · 2022-04-04T23:31:09Z

@jerryzh168 has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

…odule in backend_config_dict" Summary: Some operators have fixed quantization parameters, this PR adds the support to override the qconfig in the backend_config_dict Test Plan: python test/test_quantization.py TestQuantizeFx python test/test_quantization.py TestQuantizeFxOps Reviewers: Subscribers: Tasks: Tags: Differential Revision: [D35334279](https://our.internmc.facebook.com/intern/diff/D35334279) [ghstack-poisoned]

vkuzo · 2022-04-05T14:20:54Z

torch/ao/quantization/fx/backend_config/native.py

+    # What we want to put here is the requirement on observers, in this case dtype,
+    # quant_min, quant_max etc., but we need to first move all configs to


quant_min, quant_max are internal variables to observers. I think restrictions in backend_config_dict should be on the public API only, such as scale and zero_point.

makes sense, yeah we can put scale/zero_point constraints here in the future

…odule in backend_config_dict" Summary: Some operators have fixed quantization parameters, this PR adds the support to override the qconfig in the backend_config_dict Test Plan: python test/test_quantization.py TestQuantizeFx python test/test_quantization.py TestQuantizeFxOps Reviewers: Subscribers: Tasks: Tags: Differential Revision: [D35334279](https://our.internmc.facebook.com/intern/diff/D35334279) [ghstack-poisoned]

jerryzh168 · 2022-04-05T17:41:11Z

@jerryzh168 has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

…ckend_config_dict (#75135) Summary: Pull Request resolved: #75135 Some operators have fixed quantization parameters, this PR adds the support to override the qconfig in the backend_config_dict Test Plan: python test/test_quantization.py TestQuantizeFx python test/test_quantization.py TestQuantizeFxOps Imported from OSS Reviewed By: vkuzo Differential Revision: D35334279 fbshipit-source-id: 390510bd8fc2d61004c36c54390989583e6519ce

github-actions · 2022-04-06T07:01:03Z

Hey @jerryzh168.
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.

This was referenced Apr 1, 2022

[quant][fx] Remove is_output_quantized from QuantizeHandler #74843

Closed

[quant][refactor] Refactor find_matches for easier future extension #74878

Closed

[quant][fx] Add support for BinarOpQuantizeHandler in backend_config_dict #74882

Closed

facebook-github-bot added the cla signed label Apr 1, 2022

facebook-github-bot added the module: fx label Apr 1, 2022

jerryzh168 requested review from andrewor14 and vkuzo April 1, 2022 20:56

jerryzh168 mentioned this pull request Apr 2, 2022

fx quant: add quantized Softmax workflow integration #75106

Closed

vkuzo reviewed Apr 4, 2022

View reviewed changes

torch/ao/quantization/fx/backend_config/native.py Outdated Show resolved Hide resolved

jerryzh168 mentioned this pull request Apr 4, 2022

[quant][fx] Remove Standalone and CustomModule QuantizeHandler type checks in prepare #75202

Closed

jerryzh168 added 4 commits April 4, 2022 14:17

jerryzh168 requested a review from vkuzo April 4, 2022 23:30

jerryzh168 added 2 commits April 4, 2022 16:32

jerryzh168 mentioned this pull request Apr 5, 2022

[quant][fx] Move all binary op configs to backend_config_dict #75241

Closed

vkuzo reviewed Apr 5, 2022

View reviewed changes

vkuzo approved these changes Apr 5, 2022

View reviewed changes

jerryzh168 mentioned this pull request Apr 5, 2022

[quant][fx] Remove the remaining registrations in BinaryOpQuantizeHandler #75258

Closed

This was referenced Apr 5, 2022

[quant][fx] Add cat to backend_config_dict #75259

Closed

[quant][fx] Add BatchNorm ops to backend_config_dict #75260

Closed

jerryzh168 added 2 commits April 5, 2022 09:52

pytorchmergebot closed this in a90bcd2 Apr 6, 2022

facebook-github-bot deleted the gh/jerryzh168/765/head branch April 9, 2022 14:16

WBobby mentioned this pull request Aug 17, 2022

Add ROCm5.2.3/AMDGPU support for PyTorch WBobby/pytorch#2

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[quant][fx] Support override observers and fake quantize module in backend_config_dict #75135

[quant][fx] Support override observers and fake quantize module in backend_config_dict #75135

Uh oh!

jerryzh168 commented Apr 1, 2022 •

edited

Loading

Uh oh!

facebook-github-bot commented Apr 1, 2022 •

edited

Loading

Uh oh!

jerryzh168 commented Apr 2, 2022

Uh oh!

Uh oh!

jerryzh168 commented Apr 4, 2022

Uh oh!

vkuzo Apr 5, 2022

Uh oh!

jerryzh168 Apr 5, 2022 •

edited

Loading

Uh oh!

jerryzh168 commented Apr 5, 2022

Uh oh!

github-actions bot commented Apr 6, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

		# What we want to put here is the requirement on observers, in this case dtype,
		# quant_min, quant_max etc., but we need to first move all configs to

[quant][fx] Support override observers and fake quantize module in backend_config_dict #75135

[quant][fx] Support override observers and fake quantize module in backend_config_dict #75135

Uh oh!

Conversation

jerryzh168 commented Apr 1, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

facebook-github-bot commented Apr 1, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful links

💊 CI failures summary and remediations

Uh oh!

jerryzh168 commented Apr 2, 2022

Uh oh!

Uh oh!

jerryzh168 commented Apr 4, 2022

Uh oh!

vkuzo Apr 5, 2022

Choose a reason for hiding this comment

Uh oh!

jerryzh168 Apr 5, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

jerryzh168 commented Apr 5, 2022

Uh oh!

github-actions bot commented Apr 6, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

jerryzh168 commented Apr 1, 2022 •

edited

Loading

facebook-github-bot commented Apr 1, 2022 •

edited

Loading

jerryzh168 Apr 5, 2022 •

edited

Loading