[CPU] Disable ConvertGatherToGatherCompressed optimization for quantized models #25478
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
Reasons: 1: Small size weights have little impact on compile_model performance; 2: There is a high probability that constfolding will not be performed when compile_model; Signed-off-by: xipingya <[email protected]>
github-actions
bot
added
category: CPU
… model versa. disable it when useLpt == false. Signed-off-by: xipingya <[email protected]>
xipingyan
force-pushed
the
xp/fix_versa_acc_degradation
branch
from
c39a9bd to
1420794
Compare
github-actions
bot
removed
the
category: transformations
dmitry-gorokhov
changed the title
src/plugins/intel_cpu/src/transformations/transformation_pipeline.cpp
Outdated
Show resolved
Hide resolved
1: Add test 2: When u8/i8 + useLpt, disable "ConvertGatherToGatherCompressed". Signed-off-by: xipingya <[email protected]>
github-actions
bot
added
the
category: transformations
Signed-off-by: xipingya <[email protected]>
xipingyan
commented
Jul 12, 2024
..._cpu/tests/functional/custom/subgraph_tests/src/disable_gathercompressed_quantized_model.cpp
Outdated
Show resolved
Hide resolved
|
@v-Golubev could you please also take a look? |
..._cpu/tests/functional/custom/subgraph_tests/src/disable_gathercompressed_quantized_model.cpp
Outdated
Show resolved
Hide resolved
Signed-off-by: xipingya <[email protected]>
v-Golubev
reviewed
Jul 18, 2024
..._cpu/tests/functional/custom/subgraph_tests/src/disable_gathercompressed_quantized_model.cpp
Outdated
Show resolved
Hide resolved
..._cpu/tests/functional/custom/subgraph_tests/src/disable_gathercompressed_quantized_model.cpp
Outdated
Show resolved
Hide resolved
..._cpu/tests/functional/custom/subgraph_tests/src/disable_gathercompressed_quantized_model.cpp
Outdated
Show resolved
Hide resolved
Signed-off-by: xipingya <[email protected]>
…edding precision is I8) Signed-off-by: xipingya <[email protected]>
2: u8/i8: check if exist Gather 3: i8: check MatMul's runtime precision is u8; Signed-off-by: xipingya <[email protected]>
xipingyan
force-pushed
the
xp/fix_versa_acc_degradation
branch
from
e73b9e1 to
6b30374
Compare
dmitry-gorokhov
approved these changes
Jul 22, 2024
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.
Details:
useLPTis falseTickets: