Method forward

script/feature/model.py:321–367

(self, x, upsampleH, upsampleW)

        )

    def forward(self, x, upsampleH, upsampleW): #
        feat = []
        feat_out = [] # we only use high level features
        for i in range(len(self.encoder)):
            # print("layer {} encoder layer: {}".format(i, self.encoder[i]))
            x = self.encoder[i](x)
            if i == 3: # ReLU-4
                feat.append(x)
            elif i == 8: # ReLU-9
                feat.append(x)
            elif i == 17: # ReLU-18
                feat.append(x)
            elif i == 26: # ReLU-27
                feat.append(x)
            elif i == 35: # ReLU-36
                feat.append(x)

        for i in range(len(self.decoder)):
            # print("layer {} decoder layer: {}".format(i, self.decoder[i]))
            x = self.decoder[i](x)
            if i == 1:
                _, _, h, w = feat[4].shape
                x = nn.UpsamplingBilinear2d(size=(h,w))(x)
                x = x + feat[4]
            elif i == 3:
                _, _, h, w = feat[3].shape
                x = nn.UpsamplingBilinear2d(size=(h,w))(x)
                x = x + feat[3]
            elif i == 5:
                _, _, h, w = feat[2].shape
                x = nn.UpsamplingBilinear2d(size=(h,w))(x)
                x = x + feat[2]
                feature = torch.mean(torch.nn.UpsamplingBilinear2d(size=(upsampleH, upsampleW))(x), dim=1)
                feat_out.append(feature)
            elif i == 7:
                _, _, h, w = feat[1].shape
                x = nn.UpsamplingBilinear2d(size=(h,w))(x)
                x = x + feat[1]
                feature = torch.mean(torch.nn.UpsamplingBilinear2d(size=(upsampleH, upsampleW))(x), dim=1)
                feat_out.append(feature)
            elif i == 9:
                _, _, h, w = feat[0].shape
                x = nn.UpsamplingBilinear2d(size=(h,w))(x)
                x = x + feat[0]
                feature = torch.mean(torch.nn.UpsamplingBilinear2d(size=(upsampleH, upsampleW))(x), dim=1)
                feat_out.append(feature)
        return feat_out, x

class autoencoder_vgg7(nn.Module): # no decoder
    ''' vgg encoder with bilinear upsampling'''