fiftyone.utils.eval.openimages.OpenImagesEvaluationConfig

class documentation

class OpenImagesEvaluationConfig(DetectionEvaluationConfig): (source)

Constructor: OpenImagesEvaluationConfig(pred_field, gt_field, iou, classwise, ...)

View In Hierarchy

Open Images-style evaluation config.

Parameters
pred_fieldthe name of the field containing the predicted fiftyone.core.labels.Detections, fiftyone.core.labels.Polylines, or fiftyone.core.labels.Keypoints
gt_fieldthe name of the field containing the ground truth fiftyone.core.labels.Detections, fiftyone.core.labels.Polylines, or fiftyone.core.labels.Keypoints
iouthe IoU threshold to use to determine matches
classwisewhether to only match objects with the same class label (True) or allow matches between classes (False)
iscrowdthe name of the crowd attribute
use_maskswhether to compute IoUs using the instances masks in the mask attribute of the provided objects, which must be fiftyone.core.labels.Detection instances
use_boxeswhether to compute IoUs using the bounding boxes of the provided fiftyone.core.labels.Polyline instances rather than using their actual geometries
tolerancea tolerance, in pixels, when generating approximate polylines for instance masks. Typical values are 1-3 pixels. By default, IoUs are computed directly on the dense pixel masks
max_predsthe maximum number of predicted objects to evaluate when computing mAP and PR curves
error_level

the error level to use when manipulating instance masks or polylines. Valid values are:

  • 0: raise geometric errors that are encountered
  • 1: log warnings if geometric errors are encountered
  • 2: ignore geometric errors

If error_level > 0, any calculation that raises a geometric error will default to an IoU of 0

hierarchyan optional dict containing a hierarchy of classes for evaluation following the structure {"LabelName": label, "Subcategory": [{...}, ...]}
pos_label_fieldthe name of a field containing image-level fiftyone.core.labels.Classifications that specify which classes should be evaluated in the image
neg_label_fieldthe name of a field containing image-level fiftyone.core.labels.Classifications that specify which classes should not be evaluated in the image
expand_gt_hierarchywhether to expand ground truth objects and labels according to the provided hierarchy
expand_pred_hierarchywhether to expand predicted objects and labels according to the provided hierarchy
custom_metricsan optional list of custom metrics to compute or dict mapping metric names to kwargs dicts
Method __init__ Undocumented
Instance Variable error_level Undocumented
Instance Variable expand_gt_hierarchy Undocumented
Instance Variable expand_pred_hierarchy Undocumented
Instance Variable hierarchy Undocumented
Instance Variable iscrowd Undocumented
Instance Variable max_preds Undocumented
Instance Variable neg_label_field Undocumented
Instance Variable pos_label_field Undocumented
Instance Variable tolerance Undocumented
Instance Variable use_boxes Undocumented
Instance Variable use_masks Undocumented
Property method The name of the method.
Property requires_additional_fields Whether fields besides pred_field and gt_field are required in order to perform evaluation.
Instance Variable _hierarchy_keyed_child Undocumented
Instance Variable _hierarchy_keyed_parent Undocumented

Inherited from DetectionEvaluationConfig:

Instance Variable gt_field Undocumented
Instance Variable pred_field Undocumented
Property type The type of run.

Inherited from BaseRunConfig (via DetectionEvaluationConfig, BaseEvaluationMethodConfig, EvaluationMethodConfig):

Class Method from_dict Constructs a BaseRunConfig from a serialized JSON dict representation of it.
Static Method base_config_cls Returns the config class for the given run type.
Method attributes Returns the list of class attributes that will be serialized by serialize.
Method build Builds the BaseRun instance associated with this config.
Method load_credentials Loads any necessary credentials from the given keyword arguments or the relevant FiftyOne config.
Property cls The fully-qualified name of this BaseRunConfig class.
Property run_cls The BaseRun class associated with this config.
Class Method _virtual_attributes A list of attributes that are serialized but should not be treated as parameters when loading the config class from the database.

Inherited from Config (via DetectionEvaluationConfig, BaseEvaluationMethodConfig, EvaluationMethodConfig, BaseRunConfig):

Method __repr__ Undocumented
def __init__(self, pred_field, gt_field, iou=None, classwise=None, iscrowd='IsGroupOf', use_masks=False, use_boxes=False, tolerance=None, max_preds=None, error_level=1, hierarchy=None, pos_label_field=None, neg_label_field=None, expand_gt_hierarchy=True, expand_pred_hierarchy=False, custom_metrics=None, **kwargs): (source)
overrides fiftyone.utils.eval.detection.DetectionEvaluationConfig.__init__

Undocumented

error_level: 1 = (source)

Undocumented

expand_gt_hierarchy: True = (source)

Undocumented

expand_pred_hierarchy: False = (source)

Undocumented

hierarchy: None = (source)

Undocumented

iscrowd: "IsGroupOf" = (source)

Undocumented

max_preds: None = (source)

Undocumented

neg_label_field: None = (source)

Undocumented

pos_label_field: None = (source)

Undocumented

tolerance: None = (source)

Undocumented

use_boxes: False = (source)

Undocumented

use_masks: False = (source)

Undocumented

@property
method = (source)
overrides fiftyone.core.evaluation.EvaluationMethodConfig.method

The name of the method.

@property
requires_additional_fields = (source)
overrides fiftyone.utils.eval.detection.DetectionEvaluationConfig.requires_additional_fields

Whether fields besides pred_field and gt_field are required in order to perform evaluation.

If True then the entire samples will be loaded rather than using select_fields() to optimize.

_hierarchy_keyed_child = (source)

Undocumented

_hierarchy_keyed_parent = (source)

Undocumented