Neural network caffe CNN：跨频道共享
我有一个nxmx16x1 conv层，我想跨通道进行池处理，因此结果的维数为nxmx1x1Neural network caffe CNN：跨频道共享,neural-network,deep-learning,caffe,conv-neural-network,pycaffe,Neural Network,Deep Learning,Caffe,Conv Neural Network,Pycaffe,我有一个nxmx16x1 conv层，我想跨通道进行池处理，因此结果的维数为nxmx1x1 有什么建议吗？据我所知，池没有axis参数，对吗？您不需要axisparam。只需使用非均匀内核/步幅： layer { name: "pool16x1" type: "Pool" bottom: "input" top: "output" pooling_param { kernel_size: 16 kernel_size: 1 stride: 16
有什么建议吗？据我所知，池没有axis参数，对吗？您不需要
axislayer {
name: "pool16x1"
type: "Pool"
bottom: "input"
top: "output"
pooling_param {
kernel_size: 16
kernel_size: 1
stride: 16
stride: 1
# ...
}
}