Base Classes¶
Attribution¶
 class captum.attr.Attribution(forward_func)[source]¶
All attribution algorithms extend this class. It enforces its child classes to extend and override core attribute method.
 Parameters:
forward_func (Callable or torch.nn.Module) – This can either be an instance of pytorch model or any modification of model’s forward function.

attribute:
Callable
¶ This method computes and returns the attribution values for each input tensor. Deriving classes are responsible for implementing its logic accordingly.
Specific attribution algorithms that extend this class take relevant arguments.
 Parameters:
inputs (Tensor or tuple[Tensor, ...]) – Input for which attribution is computed. It can be provided as a single tensor or a tuple of multiple tensors. If multiple input tensors are provided, the batch sizes must be aligned across all tensors.
 Returns:
 attributions (Tensor or tuple[Tensor, …]):
Attribution values for each input tensor. The attributions have the same shape and dimensionality as the inputs. If a single tensor is provided as inputs, a single tensor is returned. If a tuple is provided for inputs, a tuple of corresponding sized tensors is returned.
 Return type:
Tensor or tuple[Tensor, …] of attributions

compute_convergence_delta:
Callable
¶ The attribution algorithms which derive Attribution class and provide convergence delta (aka approximation error) should implement this method. Convergence delta can be computed based on certain properties of the attribution alogrithms.
 Parameters:
attributions (Tensor or tuple[Tensor, ...]) – Attribution scores that are precomputed by an attribution algorithm. Attributions can be provided in form of a single tensor or a tuple of those. It is assumed that attribution tensor’s dimension 0 corresponds to the number of examples, and if multiple input tensors are provided, the examples must be aligned appropriately.
*args (Any, optional) – Additonal arguments that are used by the subclasses depending on the specific implementation of compute_convergence_delta.
 Returns:
 deltas (Tensor):
Depending on specific implementaion of subclasses, convergence delta can be returned per sample in form of a tensor or it can be aggregated across multuple samples and returned in form of a single floating point tensor.
 Return type:
Tensor of deltas
 classmethod get_name()[source]¶
Create readable class name by inserting a space before any capital characters besides the very first.
 Returns:
a readable class name
 Return type:
Example
for a class called IntegratedGradients, we return the string ‘Integrated Gradients’
 has_convergence_delta()[source]¶
This method informs the user whether the attribution algorithm provides a convergence delta (aka an approximation error) or not. Convergence delta may serve as a proxy of correctness of attribution algorithm’s approximation. If deriving attribution class provides a compute_convergence_delta method, it should override both compute_convergence_delta and has_convergence_delta methods.
 Returns:
Returns whether the attribution algorithm provides a convergence delta (aka approximation error) or not.
 Return type:
Layer Attribution¶
 class captum.attr.LayerAttribution(forward_func, layer, device_ids=None)[source]¶
Layer attribution provides attribution values for the given layer, quantifying the importance of each neuron within the given layer’s output. The output attribution of calling attribute on a LayerAttribution object always matches the size of the layer output.
 Parameters:
forward_func (Callable or torch.nn.Module) – This can either be an instance of pytorch model or any modification of model’s forward function.
layer (torch.nn.Module) – Layer for which output attributions are computed. Output size of attribute matches that of layer output.
device_ids (list[int]) – Device ID list, necessary only if forward_func applies a DataParallel model, which allows reconstruction of intermediate outputs from batched results across devices. If forward_func is given as the DataParallel model itself, then it is not necessary to provide this argument.
 static interpolate(layer_attribution, interpolate_dims, interpolate_mode='nearest')[source]¶
Interpolates given 3D, 4D or 5D layer attribution to given dimensions. This is often utilized to upsample the attribution of a convolutional layer to the size of an input, which allows visualizing in the input space.
 Parameters:
layer_attribution (Tensor) – Tensor of given layer attributions.
interpolate_dims (int or tuple) – Upsampled dimensions. The number of elements must be the number of dimensions of layer_attribution  2, since the first dimension corresponds to number of examples and the second is assumed to correspond to the number of channels.
interpolate_mode (str) – Method for interpolation, which must be a valid input interpolation mode for torch.nn.functional. These methods are “nearest”, “area”, “linear” (3Donly), “bilinear” (4Donly), “bicubic” (4Donly), “trilinear” (5Donly) based on the number of dimensions of the given layer attribution.
 Returns:
 attributions (Tensor):
Upsampled layer attributions with first 2 dimensions matching slayer_attribution and remaining dimensions given by interpolate_dims.
 Return type:
Tensor of upsampled attributions
Neuron Attribution¶
 class captum.attr.NeuronAttribution(forward_func, layer, device_ids=None)[source]¶
Neuron attribution provides input attribution for a given neuron, quantifying the importance of each input feature in the activation of a particular neuron. Calling attribute on a NeuronAttribution object requires also providing the index of the neuron in the output of the given layer for which attributions are required. The output attribution of calling attribute on a NeuronAttribution object always matches the size of the input.
 Parameters:
forward_func (Callable or torch.nn.Module) – This can either be an instance of pytorch model or any modification of model’s forward function.
layer (torch.nn.Module) – Layer for which output attributions are computed. Output size of attribute matches that of layer output.
device_ids (list[int]) – Device ID list, necessary only if forward_func applies a DataParallel model, which allows reconstruction of intermediate outputs from batched results across devices. If forward_func is given as the DataParallel model itself, then it is not necessary to provide this argument.

attribute:
Callable
¶ This method computes and returns the neuron attribution values for each input tensor. Deriving classes are responsible for implementing its logic accordingly.
Specific attribution algorithms that extend this class take relevant arguments.
 Parameters:
inputs – A single high dimensional input tensor or a tuple of them.
neuron_selector (int or tuple) – Tuple providing index of neuron in output of given layer for which attribution is desired. Length of this tuple must be one less than the number of dimensions in the output of the given layer (since dimension 0 corresponds to number of examples).
 Returns:
 attributions (Tensor or tuple[Tensor, …]):
Attribution values for each input vector. The attributions have the dimensionality of inputs.
 Return type:
Tensor or tuple[Tensor, …] of attributions
Gradient Attribution¶
 class captum.attr.GradientAttribution(forward_func)[source]¶
All gradient based attribution algorithms extend this class. It requires a forward function, which most commonly is the forward function of the model that we want to interpret or the model itself.
 Parameters:
forward_func (Callable or torch.nn.Module) – This can either be an instance of pytorch model or any modification of model’s forward function.
 compute_convergence_delta(attributions, start_point, end_point, target=None, additional_forward_args=None)[source]¶
Here we provide a specific implementation for compute_convergence_delta which is based on a common property among gradientbased attribution algorithms. In the literature sometimes it is also called completeness axiom. Completeness axiom states that the sum of the attribution must be equal to the differences of NN Models’s function at its end and start points. In other words: sum(attributions)  (F(end_point)  F(start_point)) is close to zero. Returned delta of this method is defined as above stated difference.
This implementation assumes that both the start_point and end_point have the same shape and dimensionality. It also assumes that the target must have the same number of examples as the start_point and the end_point in case it is provided in form of a list or a nonsingleton tensor.
 Parameters:
attributions (Tensor or tuple[Tensor, ...]) – Precomputed attribution scores. The user can compute those using any attribution algorithm. It is assumed the shape and the dimensionality of attributions must match the shape and the dimensionality of start_point and end_point. It also assumes that the attribution tensor’s dimension 0 corresponds to the number of examples, and if multiple input tensors are provided, the examples must be aligned appropriately.
start_point (Tensor or tuple[Tensor, ...], optional) – start_point is passed as an input to model’s forward function. It is the starting point of attributions’ approximation. It is assumed that both start_point and end_point have the same shape and dimensionality.
end_point (Tensor or tuple[Tensor, ...]) – end_point is passed as an input to model’s forward function. It is the end point of attributions’ approximation. It is assumed that both start_point and end_point have the same shape and dimensionality.
target (int, tuple, Tensor, or list, optional) –
Output indices for which gradients are computed (for classification cases, this is usually the target class). If the network returns a scalar value per example, no target index is necessary. For general 2D outputs, targets can be either:
a single integer or a tensor containing a single integer, which is applied to all input examples
a list of integers or a 1D tensor, with length matching the number of examples in inputs (dim 0). Each integer is applied as the target for the corresponding example.
For outputs with > 2 dimensions, targets can be either:
A single tuple, which contains #output_dims  1 elements. This target index is applied to all examples.
A list of tuples with length equal to the number of examples in inputs (dim 0), and each tuple containing #output_dims  1 elements. Each tuple is applied as the target for the corresponding example.
Default: None
additional_forward_args (Any, optional) – If the forward function requires additional arguments other than the inputs for which attributions should not be computed, this argument can be provided. It must be either a single additional argument of a Tensor or arbitrary (nontuple) type or a tuple containing multiple additional arguments including tensors or any arbitrary python types. These arguments are provided to forward_func in order following the arguments in inputs. For a tensor, the first dimension of the tensor must correspond to the number of examples. additional_forward_args is used both for start_point and end_point when computing the forward pass. Default: None
 Returns:
 deltas (Tensor):
This implementation returns convergence delta per sample. Deriving subclasses may do any type of aggregation of those values, if necessary.
 Return type:
Tensor of deltas
Perturbation Attribution¶
 class captum.attr.PerturbationAttribution(forward_func)[source]¶
All perturbation based attribution algorithms extend this class. It requires a forward function, which most commonly is the forward function of the model that we want to interpret or the model itself.
 Parameters:
forward_func (Callable or torch.nn.Module) – This can either be an instance of pytorch model or any modification of model’s forward function.