Link Search Menu Expand Document

Average Precision

It represents the weighted mean of precision achieved at each threshold, with the increase in recall from the previous threshold used as weight. It is defined as:

ap = sum[(recall(n) - recall(n-1))*precision(n)]

Example

from odin.classes import Metrics

my_metric = Metrics.AVERAGE_PRECISION_SCORE
# use my_metric as 'metric' parameter in the analyses

Tasks supported

Binary Classification Single-label Classification Multi-label Classification Object Detection Instance Segmentation
yes yes yes yes yes