search

Zhen-Dong / HAWQ

Quantization library for PyTorch. Support low-precision and mixed-precision quantization, with hardware implementation through TVM.
MIT License
410 stars 83 forks source link

Update QuantConv2d & QuantBnConv2d export in QONNX #41

Closed jicampos closed 1 year ago

jicampos commented 1 year ago

We are adding bias in QuantConv2d and QuantBnConv2d when exporting to the QONNX format. Batchnorm. folding with QuantBnConv2d is now included as well. If flag _fixBN is not set, exporting will include the Batchnorm layer.