Where does the extra dimension come from in the target labels list?

yunjey / stargan

StarGAN - Official PyTorch Implementation (CVPR 2018)

MIT License

5.21k stars 967 forks source link

Where does the extra dimension come from in the target labels list? #139

Open aplumley opened 2 years ago

aplumley commented 2 years ago

create_labels (in the solver) is called when generating target labels for network testing.

The dimensions of the output (c_trg_list) is [c_dim x batch_size x c_dim]. Why is c_dim used for 2 dimensions, not just one? Why are the original (input) labels cloned?