What are some common issues you might run into while training a neural network with Keras?
Some common issues you might run into while training a neural network with Keras include overfitting, underfitting, and vanishing gradients. Overfitting occurs when your model has too much capacity and begins to learn the noise in your training data instead of the signal. Underfitting occurs when your model doesn’t have enough capacity to learn the signal in your training data. Vanishing gradients occur when the gradients of the weights in your neural network become very small, making it difficult for the network to learn.
