dlpy.model.Model.evaluate_object_detection

Model.evaluate_object_detection(ground_truth, coord_type, detection_data=None, classes=None, iou_thresholds=array([ 0.5, 0.55, 0.6, 0.65, 0.7, 0.75, 0.8, 0.85, 0.9, 0.95]))

Evaluate the deep learning model on a specified validation data set.

Parameters:
ground_truth : string or CASTable, optional

Specifies a ground truth table to evaluate its corresponding prediction results

coord_type : string, optional

Specifies the format of how ground_truth table to represent bounding boxes.
Valid Values: ‘yolo’, ‘coco’

detection_data : string or CASTable, optional

Perform evaluation on the table. If the parameter is not specified, the function evaluates the last prediction performed by the model.

classes : string or list-of-strings, optional

The classes are selected to be evaluated. If you never set it, then it will perform on all of classes in ground truth table and detection_data table.

iou_thresholds : float or list-of-floats, optional

Specifying an iou threshold or a list of iou thresholds that determines what is counted as a model predicted positive detection of the classes defined by classes parameter.

Returns:
list containing calculated results.