cnn.py里conv2的输出是不是算错了？

DSE-MSU / DeepRobust

A pytorch adversarial library for attack and defense methods on images and graphs

MIT License

995 stars 192 forks source link

Closed ColdAsYou165 closed 3 years ago

ColdAsYou165 commented 3 years ago

self.fc1 = nn.Linear(8 * 8 * out_channel2, 1024) conv2的输出应为7*7*out_channel2，所以应将上述代码的8改为7 （菜鸡一枚，班门弄斧了）

YaxinLi0-0 commented 3 years ago

Thank you for opened this issue. I've updated this file.