search

taohan10200 / IIM

PyTorch implementations of the paper: "Learning Independent Instance Maps for Crowd Localization"
MIT License
163 stars 39 forks source link

the weights of Threshold Encoder be NaN when training #6

Closed s1702319 closed 3 years ago

s1702319 commented 3 years ago

Hello, your work is interesting and inspired. I'm trying to re-implement your paper using keras, but the weights of threshold encoder would be nan after some epochs. Have this phenomenon happened to you, or could you give some suggestions?

thank you~

taohan10200 commented 3 years ago

This problem occurs in some datasets where you can lower the learning rate of the threshold encoder, such as 1e-7.

Thanks for your attention!

s1702319 commented 3 years ago

This problem occurs in some datasets where you can lower the learning rate of the threshold encoder, such as 1e-7.

Thanks for your attention!

Thanks for your reply, I will try it.