Add support for (de/re)quantization with shift#27481
Merged
luotao1 merged 1 commit intoPaddlePaddle:developfrom Sep 29, 2020
Merged
Add support for (de/re)quantization with shift#27481luotao1 merged 1 commit intoPaddlePaddle:developfrom
luotao1 merged 1 commit intoPaddlePaddle:developfrom
Conversation
Author
|
@luotao1 , |
Contributor
#27502 is merged, you can restart it again. |
Author
|
@luotao1 , |
Contributor
|
I see all required CI passed |
wojtuss
force-pushed
the
wojtuss/quantization-with-shift
branch
from
6296583 to
842c9ee
Compare
luotao1
approved these changes
Sep 29, 2020
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
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
PR types
New features
PR changes
OPs
Describe
This patch adds support for quantization with shift (aka asymmetric quantization). The support is required by the
fusion_gruop quantization (#27330).Quantization with shift is performed according to the formula:
When shift is nonzero, the output of quantization is always of unsigned int8 data type.
Dequantization formula is:
Dequantization with shift expects unsigned int8 input.
Support in quantization passes will come in a separate PR.