Jintai Chen

Why use nn.Conv in the ResNet?

2

Hello! Why do you use nn.Conv in the ResNet (self.conv1 and self.fc)? Shouldn't they be implemented with adder?

why not implement the same padding in ConvTranspose2d?

1

Good Job! Why not implement the same padding in ConvTranspose2d?

Is the code for "Introducing Routing Uncertainty in Capsule Networks" available?

Why use tanh? The value of a pixel must be 0-1.

Why are the inputs of HierVAE.decoder different from that of other models?

2

The decoder in HierVAE : self.decoder((root_vecs, root_vecs, root_vecs), graphs, tensors, orders), with only the vectors of roots are fed. But the inputs of the decoder in other models are something...

What are the meanings of the metrics implemented in the /generation/decoder.py?

1

What are the meanings of the metrics (cls_acc, icls_acc, topo_acc, assm_acc) implemented in the /generation/decoder.py?

What is the technical differences between the models in ./generation/poly_hgraph and /hgraph?

1

What is the technical differences between the models in ./generation/poly_hgraph and /hgraph?

Can GradCAM be used in Transformer?

2

GradCAM is initially devised for CNNs, but can GradCAM be available for Transformer or some other architectures with self-attention?