Giters

YyzHarry / imbalanced-regression

[ICML 2021, Long Talk] Delving into Deep Imbalanced Regression

Home Page:http://dir.csail.mit.edu

Repository from Github https://github.comYyzHarry/imbalanced-regressionRepository from Github https://github.comYyzHarry/imbalanced-regression

Question about validation

ChengkaiYang opened this issue · comments

commented

Dear Author:
I hope this issue finds you well! First of all, your work is constructive.
I want to use FDS in my project but I found so many bins are empty near bins which minority samples located. Could the method be used in such case?
Also, i find a possible more correct sequence for running the code. Why FDS.update_running_stats is after FDS.update_last_epoch_stats? if FDS.update_running_stats is after FDS.update_last_epoch_stats, it leads model uses smoothed running statistics the epoch before last epoch but not the running statistics updated last epoch. So if it is correct to exchange the sequence of FDS.update_running_stats and FDS.update_last_epoch_stats?
Looking forward for your reply.Thanks!