Artificial Intelligence for Big Data

Number of epochs

One iteration through the entire dataset forward and backward in the deep neural network is called as an epoch. With each epoch, the network typically uses a backpropagation algorithm to adjust weights and biases. It is important to choose the right number of epochs. If the number of epochs is too high, the network will potentially overfit the data and not generalize on the new set of input.

If the number of epochs is too low, the network will underfit the data and will not perform well, even on the training data. There is no rule of thumb for selecting the number of epochs for a deep neural network. The number depends on the diversity of the dataset and the volume of the data. A recommended strategy is to start with a high ...