The shapes in netscope are wrong?

shicai / DenseNet-Caffe

DenseNet Caffe Models, converted from https://github.com/liuzhuang13/DenseNet

357 stars 273 forks source link

The shapes in netscope are wrong? #20

Closed bluemandora closed 7 years ago

bluemandora commented 7 years ago

In pool1, the input spatial shape is 112 and ceil_mode = false. So the output shape will be floor((112-3+2)/2+1) = 56 Does this equal to "padding = 'SAME'" in tensorflow? Which output_spatial_shape[i] = ceil(input_spatial_shape[i] / strides[i])

shicai commented 7 years ago

The output shape of this pooling layer is the same as 3x3 convolution with stride=2 in 'same' mode. So you may be right.