Closed yangzhao1230 closed 1 year ago
In your paper, you mentioned: However, my train set and valid set are respectively: which are far less than mentioned in paper. All the code about data preprocessing and spliting is based on the default setting. I wonder what causes this problem?
Yes, you are right see #10 , I plan to change this on the arXiv version.
In your paper, you mentioned: However, my train set and valid set are respectively: which are far less than mentioned in paper. All the code about data preprocessing and spliting is based on the default setting. I wonder what causes this problem?