model compression based on pytorch (1、quantization: 8/4/2bits(dorefa)、ternary/binary value(twn/bnn/xnor-net);2、 pruning: normal、regular and group convolutional pruning;3、 group convolution structure;4、BN fusion for binary value of feature(A))
Geek Repo:Geek Repo
Github PK Tool:Github PK Tool