def generate_activation()

in captum/concept/_core/tcav.py [0:0]

• 25 lines of code
• 5 McCabe index (conditional complexity)

    def generate_activation(self, layers: Union[str, List], concept: Concept) -> None:
        r"""
        Computes layer activations for the specified `concept` and
        the list of layer(s) `layers`.

        Args:
            layers (str, list[str]): A list of layer names or a layer name
                    that is used to compute layer activations for the
                    specific `concept`.
            concept (Concept): A single Concept object that provides access
                    to concept examples using a data iterator.
        """
        layers = [layers] if isinstance(layers, str) else layers
        layer_modules = [_get_module_from_name(self.model, layer) for layer in layers]

        layer_act = LayerActivation(self.model, layer_modules)
        assert concept.data_iter is not None, (
            "Data iterator for concept id:",
            "{} must be specified".format(concept.id),
        )
        for i, examples in enumerate(concept.data_iter):
            activations = layer_act.attribute.__wrapped__(  # type: ignore
                layer_act,
                examples,
                attribute_to_layer_input=self.attribute_to_layer_input,
            )
            for activation, layer_name in zip(activations, layers):
                activation = torch.reshape(activation, (activation.shape[0], -1))
                AV.save(
                    self.save_path,
                    self.model_id,
                    concept.identifier,
                    layer_name,
                    activation.detach(),
                    str(i),
                )