Request for Detailed Data Preprocessing Code and Training Pipeline

[aidarikako]

Hello, your work is highly valuable. I am currently trying to reproduce your results, but I need clarification on the data pipeline.

Could you please share the preprocessing code used to convert the raw BEAT data into the model's input format? Additionally, a brief, step-by-step executable flow of the full training procedure would be immensely helpful for successful replication.

Thank you very much!