hyunsu-yang / PyTorch

Deep Learning Zero to All - Pytorch
0 stars 0 forks source link

pytorch study 9주차 #11

Open hyunsu-yang opened 3 years ago

hyunsu-yang commented 3 years ago

강의

목차

hyunsu-yang commented 3 years ago

Lab-10-6-1 Advanced CNN(RESNET-1)

핵심

Bottleneck

  if self.downsample is not None:
            identity = self.downsample(x)
  out += identity

ResNet

downsample = nn.Sequential(
                conv1x1(self.inplanes, planes * block.expansion, stride), #conv1x1(256, 512, 2)
                nn.BatchNorm2d(planes * block.expansion), #batchnrom2d(512)
            )
hyunsu-yang commented 3 years ago

Lab-10-6-2 Advanced CNN(RESNET-2)

핵심

hyunsu-yang commented 3 years ago

Lab-10-7 Next step of CNN

핵심