training error

[deokisys]

I train ./experiments/scripts/faster_rcnn_end2end.sh 0 VGG16 pascal_voc.
if just run, error message is

so, I fixed faster_rcnn_end2end.sh ITERS=2000000 ->500 then it worked.
big ITERS occur error
small ITERS not occur error, but it is not good learning.
error message is

Floating point exception

1 ITERS is almost spend 3s. is it good? or this error is common error?

[lsj910128]

Hi, you can decrease the base_lr in the solver.prototxt. Change the base_lr: 0.001 to 0.0001, repeating this until you find a base_lr value that work.

And the time of 1 iters is ok. Mine is 2.45/iter.