code-of-learn-deep-learning-with-pytorch 关于cat vs dog中的维度问题

关于cat vs dog中的维度问题

Open zhangjinchu opened this issue 3 years ago • 0 comments

net.py文件中的这段： vgg = models.vgg19(pretrained=True) self.feature = nn.Sequential(*list(vgg.children())[:-1]) self.feature.add_module('global average', nn.AvgPool2d(9))

前两行代码的输出尺寸是7*7，而第三行平均池化的卷积核是9，运行后会出现这样的错误：Given input size: (512x7x7). Calculated output size: (512x0x0). Output size is too small

我的pytorch版本是1.6，是不是版本的问题？但按理说，Pytorch版本的迭代不会影响VGG这种经典网络的参数啊。这到底是怎么回事呢？

May 13 '22 11:05 zhangjinchu

code-of-learn-deep-learning-with-pytorch code-of-learn-deep-learning-with-pytorch copied to clipboard

关于cat vs dog中的维度问题

code-of-learn-deep-learning-with-pytorch
code-of-learn-deep-learning-with-pytorch copied to clipboard