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

How well a classifier does on a labeled dataset. More...

#include <performance.hpp>

Public Attributes

int64_t n = 0
 
int64_t classes = 0
 
double accuracy = 0.0
 The fraction of inputs whose most probable class is the label.
 
double log_loss = 0.0
 The mean negative log probability of the label (scikit-learn's log_loss). A probability of 0 counts as 1e-15.
 
double brier = 0.0
 Brier's multi-class score: the mean over inputs of sum_k (p_k - y_k)^2, from 0 to 2 (scikit-learn's brier_score_loss with scale_by_half=False).
 
double ece = 0.0
 The expected calibration error: sum over bins of count/n |accuracy - confidence|.
 
double mce = 0.0
 The maximum calibration error: the largest |accuracy - confidence| of a non-empty bin.
 
std::vector< ReliabilityBin > bins
 The reliability diagram: one entry per bin, from confidence 0 to 1.
 
std::vector< std::vector< int64_t > > confusion
 confusion[true class][predicted class]: how many inputs.
 
std::vector< double > precision
 Per class: the fraction of its predictions that are right, of its inputs found, and their harmonic mean; 0 where there is nothing to divide by, as scikit-learn's zero_division=0.
 
std::vector< double > recall
 
std::vector< double > f1
 
std::vector< int64_t > support
 Per class: how many inputs have it as their label.
 
double macro_f1 = 0.0
 The unweighted mean of f1 over the classes.
 

Detailed Description

How well a classifier does on a labeled dataset.

Member Data Documentation

◆ accuracy

double pulsatrix::ClassificationReport::accuracy = 0.0

The fraction of inputs whose most probable class is the label.

◆ bins

std::vector<ReliabilityBin> pulsatrix::ClassificationReport::bins

The reliability diagram: one entry per bin, from confidence 0 to 1.

◆ brier

double pulsatrix::ClassificationReport::brier = 0.0

Brier's multi-class score: the mean over inputs of sum_k (p_k - y_k)^2, from 0 to 2 (scikit-learn's brier_score_loss with scale_by_half=False).

◆ classes

int64_t pulsatrix::ClassificationReport::classes = 0

◆ confusion

std::vector<std::vector<int64_t> > pulsatrix::ClassificationReport::confusion

confusion[true class][predicted class]: how many inputs.

◆ ece

double pulsatrix::ClassificationReport::ece = 0.0

The expected calibration error: sum over bins of count/n |accuracy - confidence|.

◆ f1

std::vector<double> pulsatrix::ClassificationReport::f1

◆ log_loss

double pulsatrix::ClassificationReport::log_loss = 0.0

The mean negative log probability of the label (scikit-learn's log_loss). A probability of 0 counts as 1e-15.

◆ macro_f1

double pulsatrix::ClassificationReport::macro_f1 = 0.0

The unweighted mean of f1 over the classes.

◆ mce

double pulsatrix::ClassificationReport::mce = 0.0

The maximum calibration error: the largest |accuracy - confidence| of a non-empty bin.

◆ n

int64_t pulsatrix::ClassificationReport::n = 0

◆ precision

std::vector<double> pulsatrix::ClassificationReport::precision

Per class: the fraction of its predictions that are right, of its inputs found, and their harmonic mean; 0 where there is nothing to divide by, as scikit-learn's zero_division=0.

◆ recall

std::vector<double> pulsatrix::ClassificationReport::recall

◆ support

std::vector<int64_t> pulsatrix::ClassificationReport::support

Per class: how many inputs have it as their label.


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