WeightNet icon indicating copy to clipboard operation
WeightNet copied to clipboard

Published 20 hours ago • verified publisher icon megvii-model
→

Metadata

Results 2 WeightNet issues
Sort by recently updated
recently updated
newest added

in the WeightNet_DW，is that "x_w = x_w.reshape(-1, 1, 1, self.ksize, self.ksize)"wrong？why the input and output channel is 1?

class WeightNet_DW(M.Module): r""" Here we show a grouping manner when we apply WeightNet to a depthwise convolution. The grouped fc layer directly generates the convolutional kernel, has fewer parameters while...

kairenchen123 avatar kairenchen123

Is this actually attention of different channels for different kernels?

Hi, thanks for the great work on unifying SENet and CondConv. I just want to confirm if my understanding is correct: In SENet, attention is computed across the C input...

Haus226 avatar Haus226

← Metadata

169
Stars
24
Forks
Watchers

Owner

verified publisher icon megvii-model

Metadata

Back