First of all, thank you very much. This project has helped me a lot.
I have a few questions and would like to receive your reply:
for the dictionary size of different types of tokens, your code is not consistent with the paper and the open-source code of the author. For example, in your code, the dictionary size for chords is 18 (+2), whereas in the original paper, the dictionary size for chords is 133 (+2).
can you tell me how you convert the open source dataset of the original author into your dataset?
First of all, thank you very much. This project has helped me a lot.
I have a few questions and would like to receive your reply:
Hope to get your reply