Method __call__

monai/visualize/class_activation_maps.py:302–316

Compute the activation map with upsampling and postprocessing. Args: x: input tensor, shape must be compatible with `nn_module`. class_idx: index of the class to be visualized. Default to argmax(logits) layer_idx: index of the target layer if the

(self, x, class_idx=None, layer_idx=-1, **kwargs)

        return output.reshape(b, 1, *spatial)  # resume the spatial dims on the selected class

    def __call__(self, x, class_idx=None, layer_idx=-1, **kwargs):  # type: ignore[override]
        """
        Compute the activation map with upsampling and postprocessing.

        Args:
            x: input tensor, shape must be compatible with `nn_module`.
            class_idx: index of the class to be visualized. Default to argmax(logits)
            layer_idx: index of the target layer if there are multiple target layers. Defaults to -1.
            kwargs: any extra arguments to be passed on to the module as part of its `__call__`.

        Returns:
            activation maps
        """
        acti_map = self.compute_map(x, class_idx, layer_idx, **kwargs)
        return self._upsample_and_post_process(acti_map, x)


class GradCAM(CAMBase):