mmpretrain.evaluation

This package includes metrics and evaluators for classification tasks.

Single Label Metric

Accuracy	Accuracy evaluation metric.
SingleLabelMetric	A collection of precision, recall, f1-score and support for single-label tasks.
ConfusionMatrix	A metric to calculate confusion matrix for single-label tasks.

Multi Label Metric

AveragePrecision	Calculate the average precision with respect of classes.
MultiLabelMetric	A collection of precision, recall, f1-score and support for multi-label tasks.
VOCAveragePrecision	Calculate the average precision with respect of classes for VOC dataset.
VOCMultiLabelMetric	A collection of metrics for multi-label multi-class classification task based on confusion matrix for VOC dataset.

Retrieval Metric

RetrievalRecall	Recall evaluation metric for image retrieval.
RetrievalAveragePrecision	Calculate the average precision for image retrieval.