What to explain, why, and what the caller has.
More...
#include <explanation_report.hpp>
|
| XaiQuestion | question = XaiQuestion::Why |
| |
| XaiReason | reason = XaiReason::Justify |
| |
| ModelTraits | traits |
| |
| std::string | method |
| | The catalog name of the explainer the ExplainFn runs.
|
| |
| int64_t | target = 0 |
| | The output (flat index into the prediction) the explanation is for.
|
| |
| std::optional< Tensor > | baseline |
| | The baseline a complete explainer measured from (Integrated Gradients' path start, KernelSHAP's absent features): its attributions should add up to f(input) - f(baseline). LRP reports its own totals and needs none.
|
| |
| std::vector< Tensor > | dataset |
| | Inputs to aggregate explanations over, each shaped like the one explained.
|
| |
| std::vector< int64_t > | labels |
| | For the per-slice error check: the true class of each dataset input, and the slice it belongs to.
|
| |
| std::vector< std::string > | slices |
| |
| std::vector< std::pair< std::string, std::string > > | settings |
| | Anything about how the explanation was made that the explainer's metadata leaves out.
|
| |
What to explain, why, and what the caller has.
◆ baseline
| std::optional<Tensor> pulsatrix::ExplanationRequest::baseline |
The baseline a complete explainer measured from (Integrated Gradients' path start, KernelSHAP's absent features): its attributions should add up to f(input) - f(baseline). LRP reports its own totals and needs none.
◆ dataset
| std::vector<Tensor> pulsatrix::ExplanationRequest::dataset |
Inputs to aggregate explanations over, each shaped like the one explained.
◆ labels
| std::vector<int64_t> pulsatrix::ExplanationRequest::labels |
For the per-slice error check: the true class of each dataset input, and the slice it belongs to.
◆ method
| std::string pulsatrix::ExplanationRequest::method |
The catalog name of the explainer the ExplainFn runs.
◆ question
◆ reason
◆ settings
| std::vector<std::pair<std::string, std::string> > pulsatrix::ExplanationRequest::settings |
Anything about how the explanation was made that the explainer's metadata leaves out.
◆ slices
| std::vector<std::string> pulsatrix::ExplanationRequest::slices |
◆ target
| int64_t pulsatrix::ExplanationRequest::target = 0 |
The output (flat index into the prediction) the explanation is for.
◆ traits
The documentation for this struct was generated from the following file: