-
Notifications
You must be signed in to change notification settings - Fork 2.7k
[CPU] Weightless cache support #32143
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: master
Are you sure you want to change the base?
[CPU] Weightless cache support #32143
Conversation
b4f3134
to
12b389c
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ok for core parts
1a9a4c6
to
c118718
Compare
c118718
to
7dc1fd8
Compare
@EgorDuplensky , could you please review this PR? |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Pull Request Overview
This pull request implements weightless cache support for the CPU plugin in OpenVINO. The implementation allows for optimized cache storage when using CacheMode::OPTIMIZE_SIZE
by storing only model structure without weight data, reducing cache file sizes.
- Adds weightless serialization/deserialization capabilities to the CPU plugin
- Implements cache mode configuration and weight path handling
- Extends test coverage for the new weightless caching functionality
Reviewed Changes
Copilot reviewed 28 out of 28 changed files in this pull request and generated 5 comments.
Show a summary per file
File | Description |
---|---|
src/tests/functional/plugin/shared/src/behavior/compiled_model/model_cache.cpp | New test implementation for weightless cache accuracy validation |
src/tests/functional/plugin/shared/include/behavior/compiled_model/model_cache.hpp | Header for weightless cache test framework |
src/plugins/intel_cpu/src/utils/graph_serializer/serializer.cpp | Core serialization logic with weightless mode support |
src/plugins/intel_cpu/src/utils/graph_serializer/deserializer.cpp | Core deserialization logic with origin weights handling |
src/plugins/intel_cpu/src/plugin.cpp | Plugin integration with weightless cache mode detection and model building |
src/plugins/intel_cpu/src/config.cpp | Configuration handling for cache mode property |
src/plugins/intel_cpu/src/compiled_model.cpp | Compiled model export with weightless serialization |
src/inference/src/dev/core_impl.cpp | Core implementation updates for weightless cache support |
src/core/src/op/util/weightless_caching_attributes.cpp | Weightless cache attribute implementation |
Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.
src/plugins/intel_cpu/src/utils/graph_serializer/serializer.cpp
Outdated
Show resolved
Hide resolved
src/plugins/intel_cpu/src/utils/graph_serializer/deserializer.cpp
Outdated
Show resolved
Hide resolved
src/plugins/intel_cpu/src/utils/graph_serializer/deserializer.cpp
Outdated
Show resolved
Hide resolved
eb9f048
to
3cbf9ee
Compare
3cbf9ee
to
cd96250
Compare
Details:
Tickets: