Loading…
Glossary term
A group of units in a neural network that process data at the same stage.
A layer is a group of units that process data at the same stage. A typical network has an input layer, several hidden layers and an output layer.
Layers come in types: fully connected, convolutional, recurrent and attention layers, each suited to different data.
One fully connected layer
import torch.nn as nn
layer = nn.Linear(in_features=10, out_features=4)Depth, the number of layers, is what gives deep learning its name and much of its power.