gnn_tracking.models.mlp#

Module Contents#

Classes#

MLP

Multi Layer Perceptron, using ReLu as activation function.
class gnn_tracking.models.mlp.MLP(input_size: int, output_size: int, hidden_dim: int | None, L=3, *, bias=True, include_last_activation=False)#

Bases: torch.nn.Module, pytorch_lightning.core.mixins.HyperparametersMixin

Multi Layer Perceptron, using ReLu as activation function.

Parameters:

  • input_size – Input feature dimension

  • output_size – Output feature dimension

  • hidden_dim – Feature dimension of the hidden layers. If None: Choose maximum of input/output size

  • L – Total number of layers (1 initial layer, L-2 hidden layers, 1 output layer)

  • bias – Include bias in linear layer?

  • include_last_activation – Include activation function for the last layer?

reset_parameters()#
forward(x)#