Val loss increases after 13 epochs

[harshays572]

val loss increases after 13 epochs

[pythonlessons]

Why is this an issue? Maybe the model is overfitting with your data?

[harshays572]

even i though the same so i tried with 3 classes but then i got Train and Val loss as NaN.

[harshays572]

i used init and end learning rate 1e-4, 1e-6 respectively.

[harshays572]

The problem is in Calculating Location Loss Because when i print them while model Training, Location Loss gives NaN.