pulsatrix 1.1.0
Loading...
Searching...
No Matches
pulsatrix::Explanation Struct Reference

An explanation and what it does and doesn't support (XAI-3). More...

#include <explanation_report.hpp>

Collaboration diagram for pulsatrix::Explanation:

Public Member Functions

bool supports_reason () const
 Whether nothing blocks using it for its reason.
 
const CheckResult * check (std::string_view name) const
 The result of the check named name, or nullptr.
 
std::optional< std::string > setting (std::string_view key) const
 The value of setting key, or std::nullopt.
 

Public Attributes

XaiQuestion question = XaiQuestion::Why
 
XaiReason reason = XaiReason::Justify
 
ModelTraits traits
 
std::string method
 The catalog name of the explainer: "IntegratedGradients".
 
int plan_rank = 0
 Its place in the plan for this question, reason and model, from 1; 0 if the planner excluded it, and then excluded_reason says why.
 
std::string excluded_reason
 
std::optional< Attribution > attribution
 What the explainer returned; empty for explainers that don't return an attribution.
 
std::vector< std::pair< std::string, std::string > > settings
 The explainer's metadata, sorted by key, then the caller's settings, then the checks' own settings under "check.": everything needed to reproduce it.
 
std::vector< std::pair< std::string, double > > metrics
 Properties of the attribution itself: "sparseness" and "complexity".
 
std::vector< CheckResult > checks
 One result per check the plan asks for, required first, plus any extra checks the caller ran.
 
std::optional< Attribution > aggregate
 The mean absolute attribution over the dataset, when the aggregate check ran.
 
std::vector< Caveat > caveats
 DeriveCaveats(*this): blocking first, then warnings, then notes.
 

Detailed Description

An explanation and what it does and doesn't support (XAI-3).

Member Function Documentation

◆ check()

const CheckResult * pulsatrix::Explanation::check ( std::string_view  name) const

The result of the check named name, or nullptr.

◆ setting()

std::optional< std::string > pulsatrix::Explanation::setting ( std::string_view  key) const

The value of setting key, or std::nullopt.

◆ supports_reason()

bool pulsatrix::Explanation::supports_reason ( ) const

Whether nothing blocks using it for its reason.

Member Data Documentation

◆ aggregate

std::optional<Attribution> pulsatrix::Explanation::aggregate

The mean absolute attribution over the dataset, when the aggregate check ran.

◆ attribution

std::optional<Attribution> pulsatrix::Explanation::attribution

What the explainer returned; empty for explainers that don't return an attribution.

◆ caveats

std::vector<Caveat> pulsatrix::Explanation::caveats

DeriveCaveats(*this): blocking first, then warnings, then notes.

◆ checks

std::vector<CheckResult> pulsatrix::Explanation::checks

One result per check the plan asks for, required first, plus any extra checks the caller ran.

◆ excluded_reason

std::string pulsatrix::Explanation::excluded_reason

◆ method

std::string pulsatrix::Explanation::method

The catalog name of the explainer: "IntegratedGradients".

◆ metrics

std::vector<std::pair<std::string, double> > pulsatrix::Explanation::metrics

Properties of the attribution itself: "sparseness" and "complexity".

◆ plan_rank

int pulsatrix::Explanation::plan_rank = 0

Its place in the plan for this question, reason and model, from 1; 0 if the planner excluded it, and then excluded_reason says why.

◆ question

XaiQuestion pulsatrix::Explanation::question = XaiQuestion::Why

◆ reason

XaiReason pulsatrix::Explanation::reason = XaiReason::Justify

◆ settings

std::vector<std::pair<std::string, std::string> > pulsatrix::Explanation::settings

The explainer's metadata, sorted by key, then the caller's settings, then the checks' own settings under "check.": everything needed to reproduce it.

◆ traits

ModelTraits pulsatrix::Explanation::traits

The documentation for this struct was generated from the following file: