We use metrics in keras for the following reasons:
- They measure the performance of your network using non-differentiable functions.
- Help to obtain values of different loss functions when your final loss is a combination of a few of them
- Track a measure with respect to which you don't want to directly optimize your model
Hope this answer helps you! Going through the Tensorflow Tutorial will also affect the way this problem could be solved.
Watch this video to know more about Keras: