mIoU¶

ignite.metrics.mIoU(cm, ignore_index=None)[source]¶

Calculates mean Intersection over Union using ConfusionMatrix metric.

Parameters

cm (ignite.metrics.confusion_matrix.ConfusionMatrix) – instance of confusion matrix metric
ignore_index (Optional[int]) – index to ignore, e.g. background index

Returns

MetricsLambda

Return type

ignite.metrics.metrics_lambda.MetricsLambda

Examples:

train_evaluator = ...

cm = ConfusionMatrix(num_classes=num_classes)
mIoU(cm, ignore_index=0).attach(train_evaluator, 'mean IoU')

state = train_evaluator.run(train_dataset)
# state.metrics['mean IoU'] -> scalar