140 const Tensor& input, uint64_t seed);
146 uint64_t seed,
const std::vector<std::string>& layers);
First-class explanation result type – values, method, and metadata together.
Base class for every layer type (LinearModule, Conv2DModule, activations, ...).
Definition module.hpp:78
N-dimensional tensor. Owns its data buffer exclusively; a DeviceBackend* is injected (not owned) – th...
Definition tensor.hpp:29
Abstract base every layer subclasses – NVI forward(), pure-virtual LRP contract.
Definition acquisition_functions.hpp:16
float Sparseness(const Attribution &attribution)
Gini index of the attribution magnitudes (Chalasani et al. 2020; Quantus Sparseness): 0 when relevanc...
PerturbationCurve DeletionCurve(const PredictFn &predict, const Tensor &input, const Attribution &attribution, const PerturbationOptions &options={})
Deletion (MoRF): removes features from most to least relevant, imputing them, and records the target ...
std::function< Tensor(const Tensor &input)> PredictFn
The model as a function: input (with a leading batch dimension of 1) to outputs.
Definition explanation_metrics.hpp:24
PerturbationCurve InsertionCurve(const PredictFn &predict, const Tensor &input, const Attribution &attribution, const PerturbationOptions &options={})
Insertion: starts from the fully imputed input and restores features from most to least relevant....
RandomizationResult ModelParameterRandomizationTest(Module &model, const ExplainFn &explain, const Tensor &input, uint64_t seed)
The model-parameter randomization test (Adebayo et al., "Sanity Checks for Saliency Maps",...
float Complexity(const Attribution &attribution)
Entropy of the fractional contributions |a_i| / sum |a| (Bhatt et al. 2020; Quantus Complexity),...
float SpearmanRankCorrelation(const std::vector< float > &x, const std::vector< float > &y)
Spearman rank correlation, with tied values given their average rank.
Tensor Impute(const Tensor &input, const std::vector< bool > &removed, const Imputation &imputation)
Returns input with every element whose removed flag is set filled by imputation.
std::function< Attribution(const Tensor &input)> ExplainFn
Explains input using the model in its current state.
Definition explanation_metrics.hpp:27
An explanation result: the raw attribution values, the method that produced them, and any relevant me...
Definition attribution.hpp:23
How a removed feature is filled in when perturbing an input.
Definition explanation_metrics.hpp:55
Kind
Definition explanation_metrics.hpp:56
static Imputation Constant(float value=0.0f)
Every removed element becomes value.
Definition explanation_metrics.hpp:63
float noise_std
NoisyLinear.
Definition explanation_metrics.hpp:59
Kind kind
Definition explanation_metrics.hpp:57
static Imputation NoisyLinear(float noise_std=0.01f, uint64_t seed=0)
ROAD (Rong et al., "A Consistent and Efficient Evaluation Strategy for Attribution Methods",...
Definition explanation_metrics.hpp:73
float value
Constant.
Definition explanation_metrics.hpp:58
uint64_t seed
NoisyLinear.
Definition explanation_metrics.hpp:60
A perturbation curve: the target score after each fraction of features was changed.
Definition explanation_metrics.hpp:93
std::vector< float > scores
Definition explanation_metrics.hpp:95
float auc
trapezoid-rule area under scores over fractions
Definition explanation_metrics.hpp:96
std::vector< float > fractions
0, 1/steps, ..., 1
Definition explanation_metrics.hpp:94
Options for DeletionCurve() and InsertionCurve().
Definition explanation_metrics.hpp:86
Imputation imputation
Definition explanation_metrics.hpp:89
int64_t target
which output (flat index into the prediction) is scored
Definition explanation_metrics.hpp:88
int64_t steps
points on the curve after the first
Definition explanation_metrics.hpp:87
The model-parameter randomization test's result.
Definition explanation_metrics.hpp:121
std::vector< float > similarity
After randomizing layers[0..i] (cascading), the Spearman rank correlation of the new attribution magn...
Definition explanation_metrics.hpp:126
std::vector< std::string > layers
Layers randomized, in order: top-level name prefixes from the output layer down.
Definition explanation_metrics.hpp:123
N-dimensional tensor – owns a buffer via DeviceBackend*, RAII (Rule of Five).