[quant][observer] Add l2 norm minimization#24022
Closed
hx89 wants to merge 28 commits intogh/hx89/3/basefrom
Closed
[quant][observer] Add l2 norm minimization#24022hx89 wants to merge 28 commits intogh/hx89/3/basefrom
hx89 wants to merge 28 commits intogh/hx89/3/basefrom
Conversation
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
This was referenced Aug 8, 2019
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
hx89
pushed a commit
that referenced
this pull request
Aug 8, 2019
Pull Request resolved: #24022 In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc ghstack-source-id: 87960774 Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
hx89
pushed a commit
that referenced
this pull request
Aug 16, 2019
Pull Request resolved: #24022 In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc ghstack-source-id: 88490681 Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
hx89
pushed a commit
that referenced
this pull request
Aug 19, 2019
Pull Request resolved: #24022 In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc ghstack-source-id: 88591184 Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
hx89
pushed a commit
that referenced
this pull request
Aug 20, 2019
Pull Request resolved: #24022 In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc ghstack-source-id: 88653245 Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
hx89
pushed a commit
that referenced
this pull request
Aug 22, 2019
Pull Request resolved: #24022 In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc ghstack-source-id: 88762358 Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
hx89
pushed a commit
that referenced
this pull request
Aug 23, 2019
Pull Request resolved: #24022 In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc ghstack-source-id: 88912056 Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
hx89
pushed a commit
that referenced
this pull request
Aug 24, 2019
Pull Request resolved: #24022 In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc ghstack-source-id: 88916776 Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
hx89
pushed a commit
that referenced
this pull request
Aug 27, 2019
Pull Request resolved: #24022 In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc ghstack-source-id: 89092376 Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
hx89
pushed a commit
that referenced
this pull request
Aug 28, 2019
Pull Request resolved: #24022 In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc ghstack-source-id: 89127537 Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
hx89
pushed a commit
that referenced
this pull request
Sep 14, 2019
Pull Request resolved: #24022 In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc ghstack-source-id: 90128157 Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
Author
|
Addressed comments. |
dzhulgakov
requested changes
Sep 15, 2019
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
hx89
pushed a commit
that referenced
this pull request
Sep 16, 2019
Pull Request resolved: #24022 In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc ghstack-source-id: 90145338 Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
Author
|
Addressed comments. |
dzhulgakov
approved these changes
Sep 16, 2019
Collaborator
dzhulgakov
left a comment
There was a problem hiding this comment.
Looks good in general, though I didn't look through each bit of the histogram merging
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/) [ghstack-poisoned]
hx89
pushed a commit
that referenced
this pull request
Sep 16, 2019
Pull Request resolved: #24022 In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc ghstack-source-id: 90182493 Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/) [ghstack-poisoned]
hx89
pushed a commit
that referenced
this pull request
Sep 17, 2019
Pull Request resolved: #24022 In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc ghstack-source-id: 90215720 Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/) [ghstack-poisoned]
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/) [ghstack-poisoned]
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/) [ghstack-poisoned]
In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/) [ghstack-poisoned]
hx89
pushed a commit
that referenced
this pull request
Sep 18, 2019
Pull Request resolved: #24022 In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc ghstack-source-id: 90298789 Differential Revision: [D16713239](https://our.internmc.facebook.com/intern/diff/D16713239/)
Contributor
|
This pull request has been merged in f2e9622. |
laurentdupin
pushed a commit
to laurentdupin/pytorch
that referenced
this pull request
Apr 24, 2026
Summary: Pull Request resolved: pytorch#24022 In histogram observer add an approximation for L2 error minimization for selecting min/max. By selecting new min/max, we filter out outliers in input distribution. This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc ghstack-source-id: 90298789 Test Plan: buck test mode/dev caffe2/test:quantization -- 'test_histogram_observer' Differential Revision: D16713239 fbshipit-source-id: 82631ba47974e25689c9c66bc3088117090e26d4
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
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:
In histogram observer add an approximation for L2 error minimization for selecting min/max.
By selecting new min/max, we filter out outliers in input distribution.
This follows the implementation of NormMinimization::NonlinearQuantizationParamsSearch in caffe2/quantization/server/norm_minimization.cc
Differential Revision: D16713239