[TorchFX] Experimental quantization using torch.ao quantizer #32
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
daniil-lyakhov
force-pushed
the
dl/fx/experimental_quantization
branch
from
3b1865c to
e1c1fba
Compare
daniil-lyakhov
changed the title
daniil-lyakhov
force-pushed
the
dl/fx/experimental_quantization
branch
2 times, most recently
from
bb8f475 to
e6bf1d5
Compare
### Changes Redundant code is removed from `tests/cross_fw/test_templates/test_quantizer_config.py` ### Reason for changes To decrease amount of code in tests
daniil-lyakhov
force-pushed
the
dl/fx/experimental_quantization
branch
3 times, most recently
from
efd3367 to
d1941f3
Compare
### Changes - Added BF16/FP16 models to PTQ tests scope. ### Reason for changes - Extend e2e with different precisions. ### Related tickets - 147481 ### Tests - manual post_training_quantization/562/ - passed
### Changes Removed TensorBackendType. ### Reason for changes Got TensorBackendType and TensorBackend with the same definition
Workaround for issue huggingface/optimum-intel#993
daniil-lyakhov
force-pushed
the
dl/fx/experimental_quantization
branch
from
7701b88 to
bea7751
Compare
Reverts openvinotoolkit#2922 ### Related tickets 158536
### Changes Removed requirements.txt ### Reason for changes - requirements.txt that install package is not expected as default behavior to install module in editable mode. - editable mode is deprecated - requirements.txt not actual after pyproject.toml
### Changes Update readme.md and installation.md
### Changes - Add pattern based on file extension to gitignore for examples - Use pathlib style instead os.path - Fix issue with creating ../nncf/~/.cache/datasets directory - Replace tqdm to nncf.common.logging.track_progress - Remove unused function arguments and if conditions - Clean logs, disable progress bar, remove log of validataion - Pass list of command arguments to subprocess instead of string - Remove nncf dependencies from tests/cross_fw/examples - Add `--reuse-venv` to use venv from example directory - Add progress bar to collect_values_for_each_item_using_prepared_model ### Tests https://github.com/openvinotoolkit/nncf/actions/runs/11900403097 nightly/job/test_examples/618 nightly/job/windows/job/test-examples/278/
…notoolkit#3027) ### Changes * Bias fusing is removed from default transformations * `constant_folding` is updated to remove inplace operations without users * `extract_model` is updated to support original model output as a subgraph output ### Reason for changes To make it possible to apply quantization the same way it done by X86Quantizer ### Related tickets openvinotoolkit#2766 110985 ### Tests * All int8 references are updated and checked manually * `test_constant_folding` and `test_constant_folding_with_constraints` are updated with a constant subgraph which contains an inplace op (`relu_`) * `test_model_extraction_with_original_output` is introduced * conformance test post_training_quantization/557 have finished successfully
daniil-lyakhov
force-pushed
the
dl/fx/experimental_quantization
branch
from
aea0bdf to
52e80c8
Compare
github-actions
bot
added
documentation
daniil-lyakhov
force-pushed
the
dl/fx/experimental_quantization
branch
2 times, most recently
from
9178921 to
43bc251
Compare
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.
Experimental quantization:
Changes
Reason for changes
Related tickets
Tests