Fix conv act int8 scale #38331
Merged
Fix conv act int8 scale #38331
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
|
Thanks for your contribution! |
sfraczek
force-pushed
the
fix-conv-act-int8-scale
branch
from
b729e9f
to
a3abf4a
Compare
|
@Aganlengzi Hi, could you please merge this PR? It fix conv int8 +activation accuracy bug issue. |
|
Sorry to inform you that 7c1441d's CIs have passed for more than 7 days. To prevent PR conflicts, you need to re-run all CIs manually. |
|
Related PR #38023 |
|
@baoachun Please review this PR |
|
@baoachun hi could you please help merge this PR? |
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Outdated suggestions cannot be applied.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.
PR types
Bug fixes
PR changes
OPs
Describe
Make
Scale_outscale apply to output after activation if fused with activation. When no activation, pass the scale as usual to output_shift_scale.For conv we had:
act(Scale_out/scale_in_eltwiseresidual + Scale_out/(input_scaleweights_scale) * (a*w+b))
So we had to move Scale_out out of the act.
So we want to have:
Scale_outact(1/scale_in_eltwiseresidual + 1/(data_scaleweights_scale)(a*w+b))