Skip to content

[quant][graphmode] Pass debug option into insert_quant_dequant pass#39915

Closed
jerryzh168 wants to merge 7 commits intogh/jerryzh168/347/basefrom
gh/jerryzh168/347/head
Closed

[quant][graphmode] Pass debug option into insert_quant_dequant pass#39915
jerryzh168 wants to merge 7 commits intogh/jerryzh168/347/basefrom
gh/jerryzh168/347/head

Conversation

@jerryzh168
Copy link
Copy Markdown
Contributor

@jerryzh168 jerryzh168 commented Jun 12, 2020 •
edited
Loading

Stack from ghstack:

Summary:
Some of the usage, e.g. add_scalar will not be supporting the debug option,
that is, we will not have a numerically exact representation of the final quantized model
before finalize if people use add scalar.
warning will be added in a later PR.

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

Differential Revision: D22013026

@jerryzh168
[quant][graphmode] Pass debug option into insert_quant_dequant pass
11b30b1 
Summary:
Some of the usage, e.g. add_scalar will not be supporting the debug option,
that is, we will not have a numerically exact representation of the final quantized model
before finalize if people use add scalar.
warning will be added in a later PR.

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

[ghstack-poisoned]
@jerryzh168 jerryzh168 requested a review from apaszke as a code owner June 12, 2020 01:51
This was referenced Jun 12, 2020
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
@facebook-github-bot facebook-github-bot added the oncall: jit Add this issue/PR to JIT oncall triage queue label Jun 12, 2020
@jerryzh168
Update on "[quant][graphmode] Pass debug option into insert_quant_deq…
38ee56d 
…uant pass"

Summary:
Some of the usage, e.g. add_scalar will not be supporting the debug option,
that is, we will not have a numerically exact representation of the final quantized model
before finalize if people use add scalar.
warning will be added in a later PR.

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

[ghstack-poisoned]
@dr-ci
Copy link
Copy Markdown

dr-ci Bot commented Jun 12, 2020 •
edited
Loading

💊 CI failures summary and remediations

As of commit 3a547e0 (more details on the Dr. CI page):

  • 1/1 failures introduced in this PR

🕵️ 1 new failure recognized by patterns

The following CI failures do not appear to be due to upstream breakages:

See CircleCI build pytorch_linux_xenial_py3_6_gcc5_4_ge_config_simple_test (1/1)

Step: "Run tests" (full log | diagnosis details | 🔁 rerun)

Jun 15 23:57:07 [E request_callback_impl.cpp:153] Received error while processing request type 2: PickleError: ScriptModules cannot be deepcopied using copy.deepcopy or saved using torch.save. Mixed serialization of script and non-script modules is not supported. For purely script modules use my_script_module.save() instead. 
Jun 15 23:57:07   /opt/conda/lib/python3.6/site-packages/torch/distributed/rpc/internal.py(85): serialize 
Jun 15 23:57:07   /opt/conda/lib/python3.6/site-packages/torch/distributed/rpc/internal.py(137): serialize 
Jun 15 23:57:07  
Jun 15 23:57:07 [E request_callback_impl.cpp:153] Received error while processing request type 2: PickleError: ScriptModules cannot be deepcopied using copy.deepcopy or saved using torch.save. Mixed serialization of script and non-script modules is not supported. For purely script modules use my_script_module.save(<filename>) instead. 
Jun 15 23:57:07  
Jun 15 23:57:07 At: 
Jun 15 23:57:07   /opt/conda/lib/python3.6/site-packages/torch/jit/__init__.py(1896): __getstate__ 
Jun 15 23:57:07   /opt/conda/lib/python3.6/site-packages/torch/distributed/rpc/internal.py(85): serialize 
Jun 15 23:57:07   /opt/conda/lib/python3.6/site-packages/torch/distributed/rpc/internal.py(137): serialize 
Jun 15 23:57:07  
Jun 15 23:57:07 [E request_callback_impl.cpp:153] Received error while processing request type 2: PickleError: ScriptModules cannot be deepcopied using copy.deepcopy or saved using torch.save. Mixed serialization of script and non-script modules is not supported. For purely script modules use my_script_module.save(<filename>) instead. 
Jun 15 23:57:07  
Jun 15 23:57:07 At: 
Jun 15 23:57:07   /opt/conda/lib/python3.6/site-packages/torch/jit/__init__.py(1896): __getstate__ 
Jun 15 23:57:07   /opt/conda/lib/python3.6/site-packages/torch/distributed/rpc/internal.py(85): serialize 
Jun 15 23:57:07   /opt/conda/lib/python3.6/site-packages/torch/distributed/rpc/internal.py(137): serialize 
Jun 15 23:57:07  
Jun 15 23:57:07 ok (1.220s) 
Jun 15 23:57:08   test_unexepected_kwarg_is_specified (__main__.JitRpcTestWithSpawn) ... ok (1.224s) 
Jun 15 23:57:10   test_user_rrefs_confirmed (__main__.JitRpcTestWithSpawn) ... ok (1.219s) 
Jun 15 23:57:11   test_user_rrefs_confirmed_remote (__main__.JitRpcTestWithSpawn) ... ok (1.219s)
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 or post in the (internal) Dr. CI Users group.

See how this bot performed.

This comment has been revised 24 times.

@jerryzh168
Update on "[quant][graphmode] Pass debug option into insert_quant_deq…
09d7da3 
…uant pass"

Summary:
Some of the usage, e.g. add_scalar will not be supporting the debug option,
that is, we will not have a numerically exact representation of the final quantized model
before finalize if people use add scalar.
warning will be added in a later PR.

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

Differential Revision: [D22013026](https://our.internmc.facebook.com/intern/diff/D22013026)

[ghstack-poisoned]
@jerryzh168 jerryzh168 mentioned this pull request Jun 12, 2020
Closed
@jerryzh168
Update on "[quant][graphmode] Pass debug option into insert_quant_deq…
c2f950e 
…uant pass"

Summary:
Some of the usage, e.g. add_scalar will not be supporting the debug option,
that is, we will not have a numerically exact representation of the final quantized model
before finalize if people use add scalar.
warning will be added in a later PR.

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

Differential Revision: [D22013026](https://our.internmc.facebook.com/intern/diff/D22013026)

[ghstack-poisoned]
@jerryzh168
Update on "[quant][graphmode] Pass debug option into insert_quant_deq…
7782577 
…uant pass"

Summary:
Some of the usage, e.g. add_scalar will not be supporting the debug option,
that is, we will not have a numerically exact representation of the final quantized model
before finalize if people use add scalar.
warning will be added in a later PR.

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

Differential Revision: [D22013026](https://our.internmc.facebook.com/intern/diff/D22013026)

[ghstack-poisoned]
This was referenced Jun 12, 2020
Closed
Closed
Closed
@jerryzh168
Update on "[quant][graphmode] Pass debug option into insert_quant_deq…
4bfacf5 
…uant pass"

Summary:
Some of the usage, e.g. add_scalar will not be supporting the debug option,
that is, we will not have a numerically exact representation of the final quantized model
before finalize if people use add scalar.
warning will be added in a later PR.

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

Differential Revision: [D22013026](https://our.internmc.facebook.com/intern/diff/D22013026)

[ghstack-poisoned]
@jerryzh168 jerryzh168 mentioned this pull request Jun 15, 2020
Closed
@jerryzh168
Update on "[quant][graphmode] Pass debug option into insert_quant_deq…
3a547e0 
…uant pass"

Summary:
Some of the usage, e.g. add_scalar will not be supporting the debug option,
that is, we will not have a numerically exact representation of the final quantized model
before finalize if people use add scalar.
warning will be added in a later PR.

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

Differential Revision: [D22013026](https://our.internmc.facebook.com/intern/diff/D22013026)

[ghstack-poisoned]
@facebook-github-bot
Copy link
Copy Markdown
Contributor

facebook-github-bot commented Jun 16, 2020

This pull request has been merged in ee5ad6c.

@cpuhrsch
Copy link
Copy Markdown
Contributor

cpuhrsch commented Jun 16, 2020

seems to have broken test_lstm_api - quantization.test_quantized_module.TestDynamicQuantizedModule

@cpuhrsch
Copy link
Copy Markdown
Contributor

cpuhrsch commented Jun 16, 2020

please attempt to merge again

xwang233 pushed a commit to xwang233/pytorch that referenced this pull request Jun 20, 2020
@jerryzh168 @xwang233
[quant][graphmode] Pass debug option into insert_quant_dequant pass (p…
dc3d4e8 
…ytorch#39915)

Summary:
Pull Request resolved: pytorch#39915

Some of the usage, e.g. add_scalar will not be supporting the debug option,
that is, we will not have a numerically exact representation of the final quantized model
before finalize if people use add scalar.
warning will be added in a later PR.

Test Plan: Imported from OSS

Differential Revision: D22013026

fbshipit-source-id: 714b938f25c10fad3dfc79f095356b9803ef4b47
@facebook-github-bot facebook-github-bot deleted the gh/jerryzh168/347/head branch June 20, 2020 14:16
laurentdupin pushed a commit to laurentdupin/pytorch that referenced this pull request Apr 24, 2026
@jerryzh168 @facebook-github-bot
[quant][graphmode] Pass debug option into insert_quant_dequant pass (p…
8964e10 
…ytorch#39915)

Summary:
Pull Request resolved: pytorch#39915

Some of the usage, e.g. add_scalar will not be supporting the debug option,
that is, we will not have a numerically exact representation of the final quantized model
before finalize if people use add scalar.
warning will be added in a later PR.

Test Plan: Imported from OSS

Differential Revision: D22013026

fbshipit-source-id: 714b938f25c10fad3dfc79f095356b9803ef4b47
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@vkuzo vkuzo vkuzo approved these changes

@apaszke apaszke Awaiting requested review from apaszke

Assignees

No one assigned

Labels

Merged oncall: jit Add this issue/PR to JIT oncall triage queue

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

5 participants

@jerryzh168 @facebook-github-bot @cpuhrsch @vkuzo @mruberry