Struct ConfusionMatrix 

pub struct ConfusionMatrix {
    pub num_classes: usize,
    pub counts: Vec<u64>,
}

Confusion matrix for multi-class classification.

Fields

num_classes: usize

Number of classes.

counts: Vec<u64>

Matrix data: counts\[true_label * num_classes + predicted_label\].

Implementations

impl ConfusionMatrix

pub fn new(num_classes: usize) -> Self

Create a zero-filled confusion matrix for num_classes classes.

pub fn record(&mut self, true_label: usize, predicted: usize)

Record a single prediction.

pub fn accuracy(&self) -> f64

Overall accuracy: fraction of correct predictions.

pub fn precision(&self, class: usize) -> f64

Per-class precision: TP / (TP + FP).

pub fn recall(&self, class: usize) -> f64

Per-class recall: TP / (TP + FN).

pub fn f1(&self, class: usize) -> f64

Per-class F1 score.

pub fn to_csv(&self) -> String

Export as a CSV string.

Trait Implementations

impl Clone for ConfusionMatrix

fn clone(&self) -> ConfusionMatrix

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for ConfusionMatrix

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.