CommonObjectDetectionMetrics

class ignite.metrics.CommonObjectDetectionMetrics(num_classes=80, output_transform=<function CommonObjectDetectionMetrics.<lambda>>, device=device(type='cpu'), skip_unrolling=True)

Common Object Detection metrics. Included metrics are as follows:

Metric name	Description
AP@50..95	Average precision averaged over .50 to.95 IOU thresholds
AR-100	Average recall with maximum 100 detections
AP@50	Average precision with IOU threshold=.50
AP@75	Average precision with IOU threshold=.75
AP-S	Average precision over small objects (< 32px * 32px)
AR-S	Average recall over small objects
AP-M	Average precision over medium objects (S < . < 96px * 96px)
AR-M	Average recall over medium objects
AP-L	Average precision over large objects (M < . < 1e5px * 1e5px)
AR-L	Average recall over large objects greater than zero)
AR-1	Average recall with maximum 1 detection
AR-10	Average recall with maximum 10 detections

New in version 0.5.2.

Methods

compute	Computes the metric based on its accumulated state.
reset	Resets the metric to its initial state.
update	Updates the metric's state using the passed batch output.

Parameters

• num_classes (int) –

• output_transform (Callable) –

• device (Union[str, device]) –

• skip_unrolling (bool) –

compute()

Computes the metric based on its accumulated state.

By default, this is called at the end of each epoch.

Returns

the actual quantity of interest. However, if a Mapping is returned, it will be (shallow) flattened into engine.state.metrics when completed() is called.

Return type

Any

Raises

NotComputableError – raised when the metric cannot be computed.

reset()

Resets the metric to its initial state.

By default, this is called at the start of each epoch.

Return type

None

update(output)

Updates the metric’s state using the passed batch output.

By default, this is called once for each batch.

Parameters

output (Sequence[Tensor]) – the is the output from the engine’s process function.

Return type

None