MCPcopy Create free account
hub / github.com/NVIDIA/semantic-segmentation / forward

Method forward

network/utils.py:207–218  ·  view source on GitHub ↗
(self, x)

Source from the content-addressed store, hash-verified

205 Norm2d(reduction_dim), nn.ReLU(inplace=True))
206
207 def forward(self, x):
208 x_size = x.size()
209
210 img_features = self.img_pooling(x)
211 img_features = self.img_conv(img_features)
212 img_features = Upsample(img_features, x_size[2:])
213 out = img_features
214
215 for f in self.features:
216 y = f(x)
217 out = torch.cat((out, y), 1)
218 return out
219
220
221class ASPP_edge(AtrousSpatialPyramidPoolingModule):

Callers

nothing calls this directly

Calls 1

UpsampleFunction · 0.90

Tested by

no test coverage detected