Closed Mendel1 closed 7 years ago
When I'm training the retinanet,cls_loss is easy to become NaN.Does anyone know the reason?
loc_loss: 0.084 | cls_loss: nan | train_loss: nan | avg_loss: nan
See #6. I'll update the loss soon.
When I'm training the retinanet,cls_loss is easy to become NaN.Does anyone know the reason?