Cross-modal Evaluation code

[HsiaoYingLu]

Hi,

I found that you have provided the second part of the cross-modal evaluation in another github page.

I wonder if it is possible for you to further provide the followings:

1. the first part of the cross-modal evaluation. (The part that create a dictionary mapping music embedding to dance sequence)
2. For now, there is only training code on github for the second part of the cross-modal evaluation. How can I use this trained model to get the score reported in the paper? Also, can you provide the pretrain model? (the one you used for generating the result in paper)

Thanks!