search

JaveyWang / Pyramid-Attention-Networks-pytorch

Implementation of Pyramid Attention Networks for Semantic Segmentation.
GNU General Public License v3.0
235 stars 55 forks source link

FPA block #3

Closed summerrr closed 5 years ago

summerrr commented 5 years ago

In network.py x_gpb = nn.AvgPool2d(x.shape[2:])(x).view(x.shape[0], self.channels_cond, 1, 1)， Assume the shape of x is Nx2048xHxW, so after this operation，whether the shape of x is Nx2048x1x1？

JaveyWang commented 5 years ago

I am sorry to hear you so late. Yes, it is a global average pooling operation.