fiftyone.utils.torch.OutputProcessor

class documentation

class OutputProcessor(object): (source)

Known subclasses: fiftyone.utils.torch.ClassifierOutputProcessor, fiftyone.utils.torch.DetectorOutputProcessor, fiftyone.utils.torch.InstanceSegmenterOutputProcessor, fiftyone.utils.torch.KeypointDetectorOutputProcessor, fiftyone.utils.torch.SemanticSegmenterOutputProcessor, fiftyone.utils.ultralytics.UltralyticsOutputProcessor

Constructor: OutputProcessor(classes, **kwargs)

View In Hierarchy

Interface for processing the outputs of Torch models.

Parameters
classes	the list of class labels for the model. This may not be required or used by some models

Method	`__call__`	Parses the model output.
Method	`__init__`	Undocumented

def __call__(self, output, frame_size, confidence_thresh=None): (source) ¶

overridden in fiftyone.utils.torch.ClassifierOutputProcessor, fiftyone.utils.torch.DetectorOutputProcessor, fiftyone.utils.torch.InstanceSegmenterOutputProcessor, fiftyone.utils.torch.KeypointDetectorOutputProcessor, fiftyone.utils.torch.SemanticSegmenterOutputProcessor, fiftyone.utils.ultralytics.UltralyticsOutputProcessor

Parses the model output.

Parameters
output	the model output for the batch of predictions
frame_size	the `(width, height)` of the frames in the batch
confidence_thresh:`None`	an optional confidence threshold to use to filter any applicable predictions
Returns
a list of `fiftyone.core.labels.Label` instances

def __init__(self, classes=None, **kwargs): (source) ¶

Undocumented