amiralansary/rl-medical

Improve accuracy

Opened this issue · 0 comments

Hi,

First of all, thank you so much for your great and useful code. I'm trying to apply this code to my dataset, 37 pancreas ducts (small size organ) and I have 2 landmarks. training error is around 0.5, but test error is around 16-20. I tried with the default parameters and sum change on, step_per_episode, batch size, target update freq, and learning rate. Do you have any suggestions for decreasing the gap between these two errors?

Thanks in advance