You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
您好,channel-wise convolution的实现方式是utils/ops.py文件里面的rev_conv2d函数么?
如果是这样,似乎实现方式有点不对。conv1d应该在[-1, num_input_channels x h x w,1]形状的数据上做卷积,而不是在[-1, num_input_channels,h x w]上做卷积。
您好,channel-wise convolution的实现方式是utils/ops.py文件里面的rev_conv2d函数么?
如果是这样,似乎实现方式有点不对。conv1d应该在[-1, num_input_channels x h x w,1]形状的数据上做卷积,而不是在[-1, num_input_channels,h x w]上做卷积。