Neural Network – Loss

Let’s look at loss function with Keras in detail and implement it in our neural network

In our previous posts we have seen MSE (means square gradient) as a loss function, but in practice there are many more efficient loss functions available to use https://keras.io/losses/

In our code we use a loss function as this

model.compile(Adam(lr=.0001), loss="sparse_categorical_crossentropy" , .....   )

There are many different types of loss function to be used in different cases for optimal output.

Its again important to understand them and these are excellent blogs to follow for the same

Loss and Loss Functions for Training Deep Learning Neural Networks

How to Choose Loss Functions When Training Deep Learning Neural Networks