遷移學習調用預訓練模型出現Given input size: (512x3x3). Calculated output size: (512x-3x-3)錯誤

調用預訓練模型 ResNet18 時報錯：RuntimeError: Given input size: (512x3x3). Calculated output size: (512x-3x-3).
Output size is too small at /opt/conda/conda-bld/pytorch_1535491974311/work/aten/src/THCUNN/generic/SpatialAveragePooling.cu:63

其解決方法是將倒數第二層的 self.avgpool = nn.AdaptiveAvgPool2d((1, 1)) 替換為 self.avgpool = nn.AdaptiveAvgPool2d(1) 即可。具體原因可能是一個bug.

參考：https://github.com/pytorch/vision/issues/696