What's a "good" value for the loss function?

[jasondoule]

I collected ~1500 labelled data and train with yolo, got a training loss of ~10, validation loss ~ 16. Obviously the fact is to use real test data to evaluate the performance, but is there a way to tell if this training loss = 10 is a "good" one? Or does it indicate I need to use more training data to see if I can push it down to 5 or even less?

Ultimately my question is, for a well-known model with a pre-defined loss function, is there a "good" standard value for the training loss?

thanks.

[kilanny]

I am training 25,000 images on 36 classes, getting 100+ loss after 10 Epochs !! Maybe it is also not good sign for lower loss due to overfitting