vamagoul

Hello, In process_skeleton.py the split in training and test samples is determined by 'S' letter in skeleton name, Instead, it should change to 'P' for cross_subject and 'C' for cross_view...