[quant][fx] Relax the constraint for input of custom module nodes #74510
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Summary: Previously we require the dequantize before custom module to have one user, this is because we are removing the dequantize node before custom module while we transform an observed custom module to a quantized custom module, but actually we don't need to remove it, we can just change the input of custom module with quantize node instead. If the dequantize node only has one user, it will be removed by the dead code elimination pass that was added recently. Test Plan: python test/test_quantization.py TestQuantizeFx.test_custom_module_class_input_has_multiple_users Reviewers: Subscribers: Tasks: Tags: [ghstack-poisoned]
jerryzh168
added a commit
that referenced
this pull request
Mar 22, 2022
Summary: Previously we require the dequantize before custom module to have one user, this is because we are removing the dequantize node before custom module while we transform an observed custom module to a quantized custom module, but actually we don't need to remove it, we can just change the input of custom module with quantize node instead. If the dequantize node only has one user, it will be removed by the dead code elimination pass that was added recently. Test Plan: python test/test_quantization.py TestQuantizeFx.test_custom_module_class_input_has_multiple_users Reviewers: Subscribers: Tasks: Tags: ghstack-source-id: fef8bb6 Pull Request resolved: #74510
Contributor
🔗 Helpful links
💊 CI failures summary and remediationsAs of commit 8597311 (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. |
Contributor
Author
|
@jerryzh168 has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
vkuzo
reviewed
Mar 22, 2022
vkuzo
reviewed
Mar 22, 2022
…e nodes" Summary: Previously we require the dequantize before custom module to have one user, this is because we are removing the dequantize node before custom module while we transform an observed custom module to a quantized custom module, but actually we don't need to remove it, we can just change the input of custom module with quantize node instead. If the dequantize node only has one user, it will be removed by the dead code elimination pass that was added recently. Test Plan: python test/test_quantization.py TestQuantizeFx.test_custom_module_class_input_has_multiple_users Reviewers: Subscribers: Tasks: Tags: Differential Revision: [D35034626](https://our.internmc.facebook.com/intern/diff/D35034626) [ghstack-poisoned]
jerryzh168
added a commit
that referenced
this pull request
Mar 23, 2022
Summary: Previously we require the dequantize before custom module to have one user, this is because we are removing the dequantize node before custom module while we transform an observed custom module to a quantized custom module, but actually we don't need to remove it, we can just change the input of custom module with quantize node instead. If the dequantize node only has one user, it will be removed by the dead code elimination pass that was added recently. Test Plan: python test/test_quantization.py TestQuantizeFx.test_custom_module_class_input_has_multiple_users Reviewers: Subscribers: Tasks: Tags: ghstack-source-id: 33ff7f1 Pull Request resolved: #74510
Contributor
Author
|
@jerryzh168 has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
Contributor
Author
|
thanks for the suggestion @vkuzo, I think it makes sense, also added an item in our BE task list to remove DCE in our code base |
andrewor14
approved these changes
Mar 23, 2022
Contributor
andrewor14
left a comment
andrewor14
left a comment
There was a problem hiding this comment.
Does the latest code not rely on DCE? If so I think the PR description is out of date now.
Contributor
Author
yeah, we remove the dequantize op directly after that now, will update the PR description |
facebook-github-bot
pushed a commit
that referenced
this pull request
Mar 23, 2022
…4510) Summary: Pull Request resolved: #74510 Previously we require the dequantize before custom module to have one user, this is because we are removing the dequantize node before custom module while we transform an observed custom module to a quantized custom module, but actually we don't need to remove it, we can just change the input of custom module with quantize node instead. If the dequantize node only has one user, it will be removed by the dead code elimination pass that was added recently. Test Plan: python test/test_quantization.py TestQuantizeFx.test_custom_module_class_input_has_multiple_users Imported from OSS Reviewed By: dzdang Differential Revision: D35034626 fbshipit-source-id: eea9fbf9fb34c61f114c6431377be347632ce36d
Contributor
|
Hey @jerryzh168. |
jerryzh168
added
release notes: quantization
shahofblah
pushed a commit
that referenced
this pull request
Mar 25, 2022
…4510) Summary: Pull Request resolved: #74510 Previously we require the dequantize before custom module to have one user, this is because we are removing the dequantize node before custom module while we transform an observed custom module to a quantized custom module, but actually we don't need to remove it, we can just change the input of custom module with quantize node instead. If the dequantize node only has one user, it will be removed by the dead code elimination pass that was added recently. Test Plan: python test/test_quantization.py TestQuantizeFx.test_custom_module_class_input_has_multiple_users Imported from OSS Reviewed By: dzdang Differential Revision: D35034626 fbshipit-source-id: eea9fbf9fb34c61f114c6431377be347632ce36d (cherry picked from commit 2878085)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Stack from ghstack (oldest at bottom):
Summary:
Previously we require the dequantize before custom module to have one user, this is because we are removing the dequantize node
before custom module while we transform an observed custom module to a quantized custom module, but actually we don't need to remove it,
we can just change the input of custom module with quantize node instead. If the dequantize node only has one user, it will be removed. We are removing the node directly instead of relying on dead code elimination to improve the debugability of the system, since people will see the final graph after the conversion, instead of introducing the dce as a step that changes the graph. We will remove the dependence on DCE in the future as well.
Test Plan:
python test/test_quantization.py TestQuantizeFx.test_custom_module_class_input_has_multiple_users
Reviewers:
Subscribers:
Tasks:
Tags:
Differential Revision: D35034626