Struct IoU 

pub struct IoU {
    pub threshold: f64,
    pub epsilon: f64,
}

Intersection over Union (IoU) metric for segmentation tasks.

IoU = (Intersection) / (Union) = TP / (TP + FP + FN)

Also known as Jaccard Index, this is a key metric for:

• Semantic segmentation
• Instance segmentation
• Object detection (bounding box overlap)

Fields

threshold: f64

Threshold for converting predictions to binary

epsilon: f64

Small epsilon to avoid division by zero

Implementations

impl IoU

pub fn new(threshold: f64) -> Self

Create a new IoU metric with custom threshold.

Trait Implementations

impl Clone for IoU

fn clone(&self) -> IoU

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for IoU

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

Formats the value using the given formatter.

impl Default for IoU

fn default() -> Self

Returns the “default value” for a type.

impl Metric for IoU

fn compute( &self, predictions: &ArrayView<'_, f64, Ix2>, targets: &ArrayView<'_, f64, Ix2>, ) -> TrainResult<f64>

Compute metric value.

fn name(&self) -> &str

Get metric name.

fn reset(&mut self)

Reset metric state (for stateful metrics).