Choosing Layers

Overview


Choosing the number and type of layer in a neural network model can be one of the trickier parts of model design. If only 1 layer is chosen, the model will likely be equivalent to other well known models, such as regression.

These models are well understood. However, these models are also limited in the complexity of the patterns that they can recognize. Adding layers to a model will increase the model space and the resulting complexity of the model.

Topics


Contents