[quant][graphmode] Fold quantized prepacking ops by jerryzh168 · Pull Request #35077 · pytorch/pytorch

jerryzh168 · 2020-03-20T00:21:05Z

Stack from ghstack:

[quant][graphmode] Quantization support for at::sort #35571 [quant][graphmode] Quantization support for at::sort
[quant][graphmode][refactor] getGeneralTensorInputs(Node*) -> getPassThroughInputs(Value*) #35558 [quant][graphmode][refactor] getGeneralTensorInputs(Node*) -> getPassThroughInputs(Value*)
[quant][graphmode][fix] docs for InsertObservers #35557 [quant][graphmode][fix] docs for InsertObservers
[quant][graphmode] Quantization support for quantized::add_scalar_relu and quantized::add_scalar_relu_out #35509 [quant][graphmode] Quantization support for quantized::add_scalar_relu and quantized::add_scalar_relu_out
[quant][graphmode] Remove unused patterns #35385 [quant][graphmode] Remove unused patterns
[quant][graphmode] Quantization support for aten::conv3d #35347 [quant][graphmode] Quantization support for aten::conv3d
[quant][graphmode][fix] Add filter for quantized::add #35345 [quant][graphmode][fix] Add filter for quantized::add
[quant][graphmode] qconfig_dict support None #35336 [quant][graphmode] qconfig_dict support None
[quant][graphmode] Make aten::relu a general op #35420 [quant][graphmode] Make aten::relu a general op
[quant][graphmode] prepare_script takes original qconfig_dict #35335 [quant][graphmode] prepare_script takes original qconfig_dict
[quant][graphmode] Quantization support for quantized::add_scalar #35334 [quant][graphmode] Quantization support for quantized::add_scalar
[quant][graphmode][refactor] Support filter function in quant fusion patterns #35333 [quant][graphmode][refactor] Support filter function in quant fusion patterns
[quant][graphmode] Quantization support for permute and repeat_interleave #35332 [quant][graphmode] Quantization support for permute and repeat_interleave
[quant][graphmode] Make interpolate/upsample work again #35130 [quant][graphmode] Make interpolate/upsample work again
[quant][graphmode] SwapDeQuant support prim::If #35142 [quant][graphmode] SwapDeQuant support prim::If
[quant][graphmode][refactor] getGeneralOpTensorInputIndexes -> getGeneralOpTensorInputs #35141 [quant][graphmode][refactor] getGeneralOpTensorInputIndexes -> getGeneralOpTensorInputs
[quant][graphmode][refactor] swapDeQuant takes block as arugment #35135 [quant][graphmode][refactor] swapDeQuant takes block as arugment
[quant][graphmode] Fold quantized prepacking ops #35077 [quant][graphmode] Fold quantized prepacking ops

Summary:
Fold the prepack ops: quantized::linear_prepack and quantized::conv2d_prepack after
freeze

Test Plan:
python test/test_jit.py

Reviewers:
.

Subscribers:

Tasks:

Tags:

Differential Revision: D20655301

Summary: Fold the prepack ops: `quantized::linear_prepack` and `quantized::conv2d_prepack` after `freeze` Test Plan: python test/test_jit.py Reviewers: . Subscribers: Tasks: Tags: [ghstack-poisoned]

…per_channel pure" Summary: We want to do constant propagation through these ops Test Plan: see next PR: #35077 Reviewers: . Subscribers: Tasks: Tags: [ghstack-poisoned]

Summary: We want to do constant propagation through these ops Test Plan: see next PR: #35077 Reviewers: . Subscribers: Tasks: Tags: [ghstack-poisoned]

Summary: Fold the prepack ops: `quantized::linear_prepack` and `quantized::conv2d_prepack` after `freeze` Test Plan: python test/test_jit.py Reviewers: . Subscribers: Tasks: Tags: ghstack-source-id: bcfc205 Pull Request resolved: #35077

Summary: Fold the prepack ops: `quantized::linear_prepack` and `quantized::conv2d_prepack` after `freeze` Test Plan: python test/test_jit.py Reviewers: . Subscribers: Tasks: Tags: [ghstack-poisoned]

Summary: Fold the prepack ops: `quantized::linear_prepack` and `quantized::conv2d_prepack` after `freeze` Test Plan: python test/test_jit.py Reviewers: . Subscribers: Tasks: Tags: ghstack-source-id: bcfc205 Pull Request resolved: #35077

dr-ci · 2020-03-20T09:56:25Z

💊 CircleCI build failures summary and remediations

As of commit a65ba73 (more details on the Dr. CI page):

✅ None of the build failures appear to be your fault 💚

1/1 broken upstream at merge base 04a3345 from Mar 27 until Mar 28 (23 commits; 0c16ced - a9b540d)
Please rebase on the viable/strict branch (expand for instructions)

If your commit is newer than viable/strict, you can try basing on an older, stable commit:
```
git fetch https://github.com/pytorch/pytorch viable/strict
git rebase --onto FETCH_HEAD $(git merge-base origin/master HEAD)
```
If your commit is older than viable/strict:
```
git fetch https://github.com/pytorch/pytorch viable/strict
git rebase FETCH_HEAD
```
Check out the recency history of this "viable master" tracking branch.

🚧 1 upstream failure:

These were probably caused by upstream breakages:

pytorch_macos_10_13_py3_test from Mar 27 until Mar 28 (23 commits; 0c16ced - a9b540d)

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

Follow this link to opt-out of these comments for your Pull Requests.

Please report bugs/suggestions on the GitHub issue tracker.

See how this bot performed.

This comment has been revised 110 times.

kimishpatel · 2020-03-20T15:05:28Z

  d.dedup();
 }

+void FoldQuantizedPrepackingOps(Module& module) {


I thought coding convention was camelcase for function names.

I guess I also have mixed that up on some of my passes.

yeah we are following camelcase for methods and helper functions I think. For jit passes I saw we are using Pascal case, e.g.: https://bddppq.github.io/codebrowser/pytorch/pytorch/torch/csrc/jit/passes/constant_propagation.h.html
I'm not sure why though

Summary: Fold the prepack ops: `quantized::linear_prepack` and `quantized::conv2d_prepack` after `freeze` Test Plan: python test/test_jit.py Reviewers: . Subscribers: Tasks: Tags: [ghstack-poisoned]

Summary: Fold the prepack ops: `quantized::linear_prepack` and `quantized::conv2d_prepack` after `freeze` Test Plan: python test/test_jit.py Reviewers: . Subscribers: Tasks: Tags: Differential Revision: [D20655301](https://our.internmc.facebook.com/intern/diff/D20655301) [ghstack-poisoned]

facebook-github-bot · 2020-03-28T04:02:47Z

This pull request has been merged in 76a8d30.

Summary: Pull Request resolved: pytorch#35077 Fold the prepack ops: `quantized::linear_prepack` and `quantized::conv2d_prepack` after `freeze` Test Plan: python test/test_jit.py Imported from OSS Differential Revision: D20655301 fbshipit-source-id: fbb4223323f788c88db7b55cfafda46fad106d49

[quant][graphmode] Fold quantized prepacking ops

f5026ba

Summary: Fold the prepack ops: `quantized::linear_prepack` and `quantized::conv2d_prepack` after `freeze` Test Plan: python test/test_jit.py Reviewers: . Subscribers: Tasks: Tags: [ghstack-poisoned]

jerryzh168 requested a review from apaszke as a code owner March 20, 2020 00:21

facebook-github-bot added the oncall: jit Add this issue/PR to JIT oncall triage queue label Mar 20, 2020

jerryzh168 requested a review from kimishpatel March 20, 2020 00:25

Update on "[quant][graphmode] Fold quantized prepacking ops"

454a2da

Summary: Fold the prepack ops: `quantized::linear_prepack` and `quantized::conv2d_prepack` after `freeze` Test Plan: python test/test_jit.py Reviewers: . Subscribers: Tasks: Tags: [ghstack-poisoned]

Update on "[quant][graphmode] Fold quantized prepacking ops"

ea91b7d

Summary: Fold the prepack ops: `quantized::linear_prepack` and `quantized::conv2d_prepack` after `freeze` Test Plan: python test/test_jit.py Reviewers: . Subscribers: Tasks: Tags: [ghstack-poisoned]

kimishpatel reviewed Mar 20, 2020

View reviewed changes

Update on "[quant][graphmode] Fold quantized prepacking ops"

5828d76

Summary: Fold the prepack ops: `quantized::linear_prepack` and `quantized::conv2d_prepack` after `freeze` Test Plan: python test/test_jit.py Reviewers: . Subscribers: Tasks: Tags: [ghstack-poisoned]

This was referenced Mar 24, 2020

[reland][quant][graphmode] Add quantization support for aten::cat (#34346) #35337

Closed

[quant][graphmode][fix] Add filter for quantized::add #35345

Closed

[quant][graphmode] Quantization support for aten::conv3d #35347

Closed

raghuramank100 approved these changes Mar 25, 2020

View reviewed changes

Update on "[quant][graphmode] Fold quantized prepacking ops"

60e8ec1

Summary: Fold the prepack ops: `quantized::linear_prepack` and `quantized::conv2d_prepack` after `freeze` Test Plan: python test/test_jit.py Reviewers: . Subscribers: Tasks: Tags: [ghstack-poisoned]

jerryzh168 mentioned this pull request Mar 25, 2020

[quant][graphmode] Remove unused patterns #35385

Closed

Update on "[quant][graphmode] Fold quantized prepacking ops"

72531fb

Summary: Fold the prepack ops: `quantized::linear_prepack` and `quantized::conv2d_prepack` after `freeze` Test Plan: python test/test_jit.py Reviewers: . Subscribers: Tasks: Tags: [ghstack-poisoned]

jerryzh168 mentioned this pull request Mar 25, 2020

[quant][graphmode] Make aten::relu a general op #35420

Closed

jerryzh168 added 3 commits March 25, 2020 14:12

Update on "[quant][graphmode] Fold quantized prepacking ops"

6638c8f

Summary: Fold the prepack ops: `quantized::linear_prepack` and `quantized::conv2d_prepack` after `freeze` Test Plan: python test/test_jit.py Reviewers: . Subscribers: Tasks: Tags: [ghstack-poisoned]

jerryzh168 mentioned this pull request Mar 26, 2020

[quant][graphmode] Quantization support for quantized::add_scalar_relu and quantized::add_scalar_relu_out #35509

Closed

jerryzh168 added 2 commits March 26, 2020 20:05

jerryzh168 mentioned this pull request Mar 27, 2020

[fix] at::print for quantized Tensor #35545

Closed

This was referenced Mar 27, 2020

[quant][graphmode][fix] docs for InsertObservers #35557

Closed

[quant][graphmode][refactor] getGeneralTensorInputs(Node*) -> getPassThroughInputs(Value*) #35558

Closed

jerryzh168 mentioned this pull request Mar 27, 2020

[quant][graphmode] Quantization support for at::sort #35571

Closed

jerryzh168 added 2 commits March 27, 2020 14:16

facebook-github-bot closed this in 76a8d30 Mar 28, 2020

facebook-github-bot added the merged label Mar 28, 2020

facebook-github-bot deleted the gh/jerryzh168/258/head branch March 31, 2020 14:18

jerryzh168 mentioned this pull request Apr 13, 2020

[jit] constant prop works for ops that output Objects #33910

Closed

mruberry added the Merged label Oct 28, 2020

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[quant][graphmode] Fold quantized prepacking ops#35077

[quant][graphmode] Fold quantized prepacking ops#35077
jerryzh168 wants to merge 26 commits intogh/jerryzh168/258/basefrom
gh/jerryzh168/258/head

jerryzh168 commented Mar 20, 2020 •

edited

Loading

Uh oh!

dr-ci Bot commented Mar 20, 2020 •

edited

Loading

Uh oh!

kimishpatel Mar 20, 2020

Uh oh!

kimishpatel Mar 20, 2020

Uh oh!

jerryzh168 Mar 20, 2020 •

edited

Loading

Uh oh!

facebook-github-bot commented Mar 28, 2020

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

Conversation

jerryzh168 commented Mar 20, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

dr-ci Bot commented Mar 20, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

💊 CircleCI build failures summary and remediations

🚧 1 upstream failure:

Uh oh!

kimishpatel Mar 20, 2020

Choose a reason for hiding this comment

Uh oh!

kimishpatel Mar 20, 2020

Choose a reason for hiding this comment

Uh oh!

jerryzh168 Mar 20, 2020 • 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!

facebook-github-bot commented Mar 28, 2020

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

jerryzh168 commented Mar 20, 2020 •

edited

Loading

dr-ci Bot commented Mar 20, 2020 •

edited

Loading

jerryzh168 Mar 20, 2020 •

edited

Loading