Layer

A set of neurons in a neural network. Three common types of layers are as follows:

The input layer, which provides values for all the features.

One or more hidden layers, which find nonlinear relationships between the features and the label.

The output layer, which provides the prediction.

For example, the following illustration shows a neural network with one input layer, two hidden layers, and one output layer:

In TensorFlow, layers are also Python functions that take Tensors and configuration options as input and produce other tensors as output.

1.
For example, the following illustration shows a neural network with one input layer, two hidden layers, and one output layer:
2.
In TensorFlow, layers are also Python functions that take Tensors and configuration options as input and produce other tensors as output.

Created for this library

1.
A research team measures the contribution of each layer in its transformer by probing layer outputs on a downstream task.
2.
An ML platform team logs activation statistics per layer to spot vanishing or exploding signals early during training.
3.
A computer vision team experiments with layer count from 18 to 152 to find the smallest model that meets its accuracy targets.