model.tools.generator.ModelGenerator

model.tools.generator.ModelGenerator#

class ModelGenerator(cfg: List[LayerGen], in_channels: int, init_weights: bool = True)[source]#

Bases: Module

Tool for generating and processing a model from a list of layer generators

Parameters:

  • cfg (List[LayerGen]) – Description of the network configuration.

  • in_channels (int) – Number of input channels.

  • init_weights (bool, optional) – If true apply weight initialization function. Defaults to True.

Methods

forward

Define the computation performed at every call.

Attributes

training

forward(X: Tensor, state: ListState | None = None) → Tuple[Tensor, ListState][source]#

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

Contents