About Experiment Setting

[ZhijieXiong]

Hello, thank you very much for your open source. I would like to ask how you preprocess the data set, such as what is the fixed length of the sequence (the paper only says that the shortest sequence length is 3)? Is some data being discarded? How to deal with sequences that are too long (For example, cutting an overly long sequence into multiple short sequences of the same length) ? When dividing the dataset, should we divide it only once or divide it by 5 fold and finally average the results?