transformers
HFQuantizer implementation for compressed-tensors library
#31704
Merged
Merged
HFQuantizer implementation for compressed-tensors library #31704
SunMarc
merged
42 commits into
huggingface:main
from
neuralmagic:compressed-tensors-quantizer
Add compressed-tensors HFQuantizer implementation
d695ec3e
flag serializable as False
f4689647
run
41224d3d
revive lines deleted by ruff
b61bfb96
fixes to load+save from sparseml, edit config to quantization_config,…
ff8f1c5a
address satrat comment
c1cb55de
compressed_tensors to compressed-tensors and revert back is_serializable
ef9d3f17
rename quant_method from sparseml to compressed-tensors
117d0504
tests
1901c3e5
edit tests
3ca270df
clean up tests
9a14b092
make style
ec59052d
cleanup
520ded87
cleanup
7dec8fc8
Merge branch 'main' into compressed-tensors-quantizer
afb550da
add test skip for when compressed tensors is not installed
d9b36601
remove pydantic import + style
e51ac594
delay torch import in test
ccb54423
initial docs
bfd9220b
update main init for compressed tensors config
71a80f92
make fix-copies
547f9cce
docstring
8acbc090
remove fill_docstring
eaa5f20b
bfineran
commented
on 2024-08-06
Apply suggestions from code review
4ba75fbc
review comments
94ea0d3c
review comments
c48840d0
Merge branch 'main' into compressed-tensors-quantizer
ab74d26c
comments - suppress warnings on state dict load, tests, fixes
2ecf7110
bug-fix - remove unnecessary call to apply quant lifecycle
e1ae5049
run_compressed compatability
ea9e927c
SunMarc
commented
on 2024-09-02
SunMarc
commented
on 2024-09-02
SunMarc
commented
on 2024-09-02
revert changes not needed for compression
1c3ad5c9
no longer need unexpected keys fn
aa1a4f97
unexpected keys not needed either
81a13dd7
Apply suggestions from code review
f53d7b99
add to_diff_dict
d8f7073c
update docs and expand testing
c4fbf70f
Merge remote-tracking branch 'upstream/main' into compressed-tensors-…
1992a887
Update _toctree.yml with compressed-tensors
298a6387
Update src/transformers/utils/quantization_config.py
3cb44153
Merge branch 'main' into compressed-tensors-quantizer
a9431576
update doc
64f475ad
add note about saving a loaded model
fabe8a31
SunMarc
approved these changes
on 2024-09-25
Reviewers
Assignees
No one assigned
Labels
None yet
Milestone
No milestone
Login to write a write a comment.
Login via GitHub