Yufeng Li
61ba5b501a
Fix bug in the back to back quantization of matmul and conv ( #5264 )
...
* fix bug in the back to back quantization of matmul and conv
* fix bug in back to back gather
2020-09-23 08:47:20 -07:00
Yufeng Li
3068a835f1
Fix quantization of 1-D conv with bias ( #5157 )
2020-09-14 18:07:14 -07:00
Yufeng Li
20b2f45b24
Support per-channel quantization of weight tensor ( #5057 )
...
* Support per-channel quantization of weight tensor
* rename util functions
* fix bugs in calibrate
* add support of reduce_range
* refine opset check
2020-09-14 11:53:50 -07:00
Yufeng Li
ffc2b25a3a
Quantization tool improvement ( #4933 )
...
Improve quantization tools:
1. Support QAT
2. Make quantization tool to register Operators.
3. Make the API clear to use
Co-authored-by: t-yguo <t-yguo@microsoft.com>
2020-09-01 09:07:46 -07:00
