I observed the same behavior described in #10 . The second-stage model outputs almost identical values for different samples if the predicted ages by the first-stage model fall into the same discretization interval. Do you have an explanation of the possible causes of this behavior? @JianCheng @liuziyang1106
Thank you for your interest in our project. You can contact me via email (liuziyang1106@buaa.edu.cn)to provide more details, and I will try to help you solve this problem.
Dear authors,
I observed the same behavior described in #10 . The second-stage model outputs almost identical values for different samples if the predicted ages by the first-stage model fall into the same discretization interval. Do you have an explanation of the possible causes of this behavior? @JianCheng @liuziyang1106
Thank you!