openvino/model-optimizer/mo
Edward Shogulin f57f1fafc7
[LPT] Q/DQ support integration (#3575)
* [ MO ] New INT weights form

	Ticket #34096

	FQ = Quantize -> Dequantize =
	= FQ` -> Convert(int8) -> Convert(src_dtype) -> FQ =
	= Constant(int8) -> Convert(src_dtype) -> FQ =
	= Constant(int8) -> Convert(src_dtype) -> Subtract(zero_point) -> Multiply(scale)

* [LPT] Q/DQ representation support

Co-authored-by: Evgenya Stepyreva <evgenya.stepyreva@intel.com>
2021-02-03 17:05:34 +03:00
..
back soft_get fix (#3662) 2020-12-21 14:05:41 +03:00
front Re-implement caffe old-style extractors with extractor extensions (#3675) 2021-02-01 13:17:17 +03:00
graph [MO] Implement TensorFlow 2 While and Keras RNN support in MO (#3573) 2021-01-21 17:39:57 +03:00
middle [LPT] Q/DQ support integration (#3575) 2021-02-03 17:05:34 +03:00
ops Update operation attributes (#3814) 2021-01-29 10:08:06 +03:00
pipeline [MO] Fix many spelling and grammar errors (#2543) 2020-10-09 12:16:12 +03:00
utils [LPT] Q/DQ support integration (#3575) 2021-02-03 17:05:34 +03:00
__init__.py Publishing R3 2018-10-16 13:45:03 +03:00
bom_test.py Publishing 2020.2 content 2020-04-13 21:17:23 +03:00
main.py [ MO ] KSO=ON for Kaldi (#2028) 2020-10-23 13:14:00 +03:00
main_test.py Cleanup IR v7 from the MO (#1008) 2020-06-22 11:52:00 +03:00