mednet.engine.detect.evaluator

Defines functionality for the evaluation of object detection predictions.

Functions

make_plots(results[, iou_threshold])	Create plots for all curves and score distributions in results.
make_table(data, fmt)	Tabulate summaries from multiple splits.
run(predictions, binning[, iou_threshold])	Run inference and calculates measures for multilabel object detection.

mednet.engine.detect.evaluator.run(predictions, binning, iou_threshold=None)

Run inference and calculates measures for multilabel object detection.

Parameters:

• predictions (Sequence[tuple[str, Sequence[tuple[Sequence[int], int]], Sequence[tuple[Sequence[int | float], int, float]]]]) – A list of predictions to consider for measurement.

• binning (str | int) – The binning algorithm to use for computing the bin widths and distribution for histograms. Choose from algorithms supported by numpy.histogram().

• iou_threshold (float | None) – IOU threshold by which we consider successful object detection. If set to None, then apply no thresholding.

Return type:

dict[str, Any]

Returns:

A dictionary containing the performance summary on the specified threshold, general performance curves (under the key curves), and score histograms (under the key score-histograms).

mednet.engine.detect.evaluator.make_table(data, fmt)

Tabulate summaries from multiple splits.

This function can properly tabulate the various summaries produced for all the splits in a prediction database.

Parameters:

• data (Mapping[str, Mapping[str, Any]]) – An iterable over all summary data collected.

• fmt (str) – One of the formats supported by python-tabulate.

Returns:

A string containing the tabulated information.

Return type:

str

mednet.engine.detect.evaluator.make_plots(results, iou_threshold=None)

Create plots for all curves and score distributions in results.

Parameters:

• results (dict[str, dict[str, Any]]) – Evaluation data as returned by run().

• iou_threshold (float | None) – IOU threshold by which we consider successful object detection. If set, it is shown on plots.

Return type:

list

Returns:

A list of figures to record to file