foxai.explainer package

Subpackages

• foxai.explainer.computer_vision package
  • Subpackages
    • foxai.explainer.computer_vision.algorithm package
      • Submodules
      • foxai.explainer.computer_vision.algorithm.conductance module
      • foxai.explainer.computer_vision.algorithm.deconv module
      • foxai.explainer.computer_vision.algorithm.deeplift module
      • foxai.explainer.computer_vision.algorithm.deeplift_shap module
      • foxai.explainer.computer_vision.algorithm.gradcam module
      • foxai.explainer.computer_vision.algorithm.gradient_shap module
      • foxai.explainer.computer_vision.algorithm.guided_backprop module
      • foxai.explainer.computer_vision.algorithm.input_x_gradient module
      • foxai.explainer.computer_vision.algorithm.integrated_gradients module
      • foxai.explainer.computer_vision.algorithm.lrp module
      • foxai.explainer.computer_vision.algorithm.noise_tunnel module
      • foxai.explainer.computer_vision.algorithm.occlusion module
      • foxai.explainer.computer_vision.algorithm.saliency module
      • Module contents
    • foxai.explainer.computer_vision.object_detection package
      • Submodules
      • foxai.explainer.computer_vision.object_detection.base_object_detector module
      • foxai.explainer.computer_vision.object_detection.types module
      • foxai.explainer.computer_vision.object_detection.utils module
      • Module contents
  • Submodules
  • foxai.explainer.computer_vision.model_utils module
    • get_last_conv_model_layer()
    • modify_modules()
  • Module contents

Submodules

foxai.explainer.base_explainer module

Abstract Explainer class.

foxai.explainer.base_explainer.CVExplainerT

CVExplainer subclass type.

alias of TypeVar(‘CVExplainerT’, bound=Explainer)

class foxai.explainer.base_explainer.Explainer

Bases: ABC

Abstract explainer class.

property algorithm_name: str

Get algorithm name.

Returns:

Name of algorithm.

Return type:

str

abstract calculate_features(model: Module, input_data: Tensor, pred_label_idx: Union[None, int, Tuple[int, ...], Tensor, List[Tuple[int, ...]], List[int]] = None, **kwargs) → Tensor

Calculate features of given explainer.

Parameters:

• model – Neural network model You want to explain.

• input_data – Input image.

• pred_label_idx – Predicted label.

Returns:

Tensor of attributes.

Module contents