nn.build_mlp

nn.build_mlp(n_in: int, n_out: int, n_hidden: int | Sequence[int] | None = None, n_layers: int = 2, activation: Callable = torch.nn.functional.silu, last_bias: bool = True, last_zero_init: bool = False) → torch.nn.Module

Build multiple layer fully connected perceptron neural network.

Parameters:

• n_in – number of input nodes.

• n_out – number of output nodes.

• n_hidden – number hidden layer nodes. If an integer, same number of node is used for all hidden layers resulting in a rectangular network. If None, the number of neurons is divided by two after each layer starting n_in resulting in a pyramidal network.

• n_layers – number of layers.

• activation – activation function. All hidden layers would the same activation function except the output layer that does not apply any activation function.