fx quant: do not insert observers at quantized inputs #49239
Closed
Conversation
This file contains 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: Context: the existing implementation of `quantized_input_idxs` is convert-only. Therefore, observers are inserted between the input and the first quantized node. This is a problem during QAT, because the initial input is a fake_quant, and it starts with scale=1 and zp=0. This does not match the quantization parameters of the graph input, which can lead to incorrect numerics. Fix: do not insert observer for a quantized input. Test Plan: ``` python test/test_quantization.py TestQuantizeFx ``` Reviewers: Subscribers: Tasks: Tags: [ghstack-poisoned]
This was referenced Dec 11, 2020
💊 CI failures summary and remediationsAs of commit bd25bc0 (more details on the Dr. CI page):
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 to the (internal) Dr. CI Users group. This comment has been revised 21 times. |
vkuzo
added a commit
that referenced
this pull request
Dec 11, 2020
Summary: Context: the existing implementation of `quantized_input_idxs` is convert-only. Therefore, observers are inserted between the input and the first quantized node. This is a problem during QAT, because the initial input is a fake_quant, and it starts with scale=1 and zp=0. This does not match the quantization parameters of the graph input, which can lead to incorrect numerics. Fix: do not insert observer for a quantized input. Test Plan: ``` python test/test_quantization.py TestQuantizeFx ``` Reviewers: Subscribers: Tasks: Tags: ghstack-source-id: c9f76756f5c34e1f65289984421e4a9df4fc2744 Pull Request resolved: #49239
Summary: Context: the existing implementation of `quantized_input_idxs` is convert-only. Therefore, observers are inserted between the input and the first quantized node. This is a problem during QAT, because the initial input is a fake_quant, and it starts with scale=1 and zp=0. This does not match the quantization parameters of the graph input, which can lead to incorrect numerics. Fix: do not insert observer for a quantized input. Test Plan: ``` python test/test_quantization.py TestQuantizeFx ``` Reviewers: Subscribers: Tasks: Tags: Differential Revision: [D25499486](https://our.internmc.facebook.com/intern/diff/D25499486) [ghstack-poisoned]
vkuzo
added a commit
that referenced
this pull request
Dec 15, 2020
Summary: Context: the existing implementation of `quantized_input_idxs` is convert-only. Therefore, observers are inserted between the input and the first quantized node. This is a problem during QAT, because the initial input is a fake_quant, and it starts with scale=1 and zp=0. This does not match the quantization parameters of the graph input, which can lead to incorrect numerics. Fix: do not insert observer for a quantized input. Test Plan: ``` python test/test_quantization.py TestQuantizeFx ``` Reviewers: Subscribers: Tasks: Tags: ghstack-source-id: 936518ba30591e75900c36168183f68f7e97e5ab Pull Request resolved: #49239
This was referenced Dec 15, 2020
Summary: Context: the existing implementation of `quantized_input_idxs` is convert-only. Therefore, observers are inserted between the input and the first quantized node. This is a problem during QAT, because the initial input is a fake_quant, and it starts with scale=1 and zp=0. This does not match the quantization parameters of the graph input, which can lead to incorrect numerics. Fix: do not insert observer for a quantized input. Test Plan: ``` python test/test_quantization.py TestQuantizeFx ``` Reviewers: Subscribers: Tasks: Tags: Differential Revision: [D25499486](https://our.internmc.facebook.com/intern/diff/D25499486) [ghstack-poisoned]
Summary: Context: the existing implementation of `quantized_input_idxs` is convert-only. Therefore, observers are inserted between the input and the first quantized node. This is a problem during QAT, because the initial input is a fake_quant, and it starts with scale=1 and zp=0. This does not match the quantization parameters of the graph input, which can lead to incorrect numerics. Fix: do not insert observer for a quantized input. Test Plan: ``` python test/test_quantization.py TestQuantizeFx ``` Reviewers: Subscribers: Tasks: Tags: Differential Revision: [D25499486](https://our.internmc.facebook.com/intern/diff/D25499486) [ghstack-poisoned]
Codecov Report
@@ Coverage Diff @@
## gh/vkuzo/187/base #49239 +/- ##
==================================================
Coverage 80.62% 80.62%
==================================================
Files 1875 1875
Lines 202790 202797 +7
==================================================
+ Hits 163501 163510 +9
+ Misses 39289 39287 -2 |
|
This pull request has been merged in 7542076. |
hwangdeyu
pushed a commit
to hwangdeyu/pytorch
that referenced
this pull request
Jan 6, 2021
Summary: Pull Request resolved: pytorch#49239 Context: the existing implementation of `quantized_input_idxs` is convert-only. Therefore, observers are inserted between the input and the first quantized node. This is a problem during QAT, because the initial input is a fake_quant, and it starts with scale=1 and zp=0. This does not match the quantization parameters of the graph input, which can lead to incorrect numerics. Fix: do not insert observer for a quantized input. Test Plan: ``` python test/test_quantization.py TestQuantizeFx ``` Imported from OSS Reviewed By: jerryzh168 Differential Revision: D25499486 fbshipit-source-id: 303b49cc9d95a9fd06fef3b0859c08be34e19d8a
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:
Summary:
Context: the existing implementation of
quantized_input_idxsis convert-only.Therefore, observers are inserted between the input and the first
quantized node. This is a problem during QAT, because the initial
input is a fake_quant, and it starts with scale=1 and zp=0. This does
not match the quantization parameters of the graph input, which can
lead to incorrect numerics.
Fix: do not insert observer for a quantized input.
Test Plan:
Reviewers:
Subscribers:
Tasks:
Tags:
Differential Revision: D25499486