Jfdong/quantize conv integer bias#1827
Merged
askhade merged 4 commits intomicrosoft:askhade/quantization_and_caliberationfrom Oct 21, 2019
Merged
Jfdong/quantize conv integer bias#1827askhade merged 4 commits intomicrosoft:askhade/quantization_and_caliberationfrom
askhade merged 4 commits intomicrosoft:askhade/quantization_and_caliberationfrom
Conversation
Author
|
Added a _convert_clip_node function to change clip node according to Opset 11 requirement. The convert pull min and max values from Clip operators attributes and put them into inputs field. After the convert, Clip node should be able to run correctly in Opset 11 runtime. |
…r quantized model
jfdong
force-pushed
the
jfdong/quantize_convInteger_bias
branch
from
fdc9233 to
e3b1111
Compare
askhade
approved these changes
Oct 21, 2019
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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description: The changes fix quantized model "opset" to 11 to support DynamicQuantizeLinear operator and add bias in convInteger node to make resnet50 be quantized correctly.
Motivation and Context
After this commit, we got 77% accuracy on quantized resnet50 model in MLPerf with 5000 images.