Convolutional neural network model compression mainly resorts to parameter quantization [13], pruning [14-16], low-rank decomposition [17] and ...
確定！ 回上一頁
