su

Results 7 comments of su

I am interested in this issue. I am working a Model-Quant project based on Pytorch and tensorflow. the techniques include CLE, BC and QAT.

@quic-akhobare have you met this problem? Any suggestion? Thanks.

@quic-akhobare we checked the details and found that it seems to be a mismatch between 'libpymo.QUANTIZATION_TF' and 'DlQuantization::QuantizationMode'. But we do not know how to solve it.

> _No description provided._ hi, did you compare the performance using kmeans and using EM for initializing the model? the corresponding NDB values based on the two methods. I test...