When training a deep learning model should the validation loss be greater than or smaller than the training loss?

While training a deep learning model I generally consider the training loss, validation loss and the accuracy as a measure to check overfitting and under fitting. And different researchers have...

23 July 2020 4,267 12 View