about the loss about dasiamrpnwithofflinetraining HOT 9 OPEN

@hjjlovecyy how to slove the problem of nan, i met this problem too.

from dasiamrpnwithofflinetraining.

And I find that if I change total_loss.backward() in train.py to another loss, such as total_reg_loss.backward()
"an illegal memory access was encountered" error no longer happens.
And the result no longer appear nan like my question.

from dasiamrpnwithofflinetraining.

nan still appear, and the model did not converge.
this is also related to the batch_size

from dasiamrpnwithofflinetraining.

Maybe you can check your pretrained model.

from dasiamrpnwithofflinetraining.

@hjjlovecyy
hello, I also met this problem (nan appears in loss). Have you solved this problem? thanks a lot!!

from dasiamrpnwithofflinetraining.

@liushhAlex I follow the repos of songdejia, foolwood and this. Finally I write a demo which can train successfully. And I implement the local-to-global search strategy.

from dasiamrpnwithofflinetraining.

@liushhAlex I follow the repos of songdejia, foolwood and this. Finally I write a demo which can train successfully. And I implement the local-to-global search strategy.

what's your auc on otb2015?

from dasiamrpnwithofflinetraining.

@hjjlovecyy How to solve the problem of nan, I met this problem too. Thanks a lot!

from dasiamrpnwithofflinetraining.

I met the same proboelm how to solve that?

from dasiamrpnwithofflinetraining.