mdlearn.nn.modules.dense_net

DenseNet module.

Classes

DenseNet(*args, **kwargs)

class mdlearn.nn.modules.dense_net.DenseNet(*args: Any, **kwargs: Any)
__init__(input_dim: int, neurons: List[int] = [128], bias: bool = True, relu_slope: float = 0.0, inplace_activation: bool = False)

DenseNet module for easy feedforward network creation. Creates a neural network with Linear layers and ReLU (or LeakyReLU activation). The returned tensor from the forward function, does not pass through an activation function.

Parameters

  • input_dim (int) – Dimension of input tensor (should be flattened).

  • neurons (List[int], default=[128]) – Linear layers in_features.

  • bias (bool, default=True) – Use a bias term in the Linear layers.

  • relu_slope (float, default=0.0) – If greater than 0.0, will use LeakyReLU activiation with negative_slope set to relu_slope.

  • inplace_activation (bool, default=False) – Sets the inplace option for the activation function.

Raises

ValueError – neurons should specify atleast one layer.

forward(x: torch.Tensor) → torch.Tensor

Forward pass through dense network.

Parameters

x (torch.Tensor) – Input data.

Returns

torch.Tensor – The output of the neural network with dimension (batch size, last neuron size).