CatFood

@goldservice2017 I am confused about the 'weight_u' and 'weight_v' in conv. layer. Could you please tell me what do '_u' and '_v' means? thank you!

just refer to the that link, and you will find self._dict = yaml.safe_load(self._yaml) fixes this.

Also, I found another issue, in `Attention.py `, transposed conv. is used to reconstruct the feature map, as shown in this line: ` feature_map = F.conv_transpose2d(attention_scores, conv_kernels, stride = 1,...