Partition score¶
Quality measures usable with partition assessment |
|
|
Overloaded function. |
-
class
PartitionMeasure
¶ Quality measures usable with partition assessment
Members:
BCE
DHamming
DCovering
-
BCE
= <PartitionMeasure.BCE: 0>¶
-
DCovering
= <PartitionMeasure.DCovering: 2>¶
-
DHamming
= <PartitionMeasure.DHamming: 1>¶
-
property
name
¶
-
property
value
¶
-
-
assess_partition
(*args, **kwargs)¶ Overloaded function.
assess_partition(candidate: numpy.ndarray[numpy.int8], ground_truth: numpy.ndarray[numpy.int8], partition_measure: higra.higram.PartitionMeasure) -> float
Assess a given candidate partition with a ground-truth segmentation and an evaluation measure (see enumeration PartitionMeasure). The candidate and ground-truth partitions must be given as labelisations with integers values between 0 (included) and the number of regions in the partition (excluded).
assess_partition(candidate: numpy.ndarray[numpy.uint8], ground_truth: numpy.ndarray[numpy.uint8], partition_measure: higra.higram.PartitionMeasure) -> float
assess_partition(candidate: numpy.ndarray[numpy.int16], ground_truth: numpy.ndarray[numpy.int16], partition_measure: higra.higram.PartitionMeasure) -> float
assess_partition(candidate: numpy.ndarray[numpy.uint16], ground_truth: numpy.ndarray[numpy.uint16], partition_measure: higra.higram.PartitionMeasure) -> float
assess_partition(candidate: numpy.ndarray[numpy.int32], ground_truth: numpy.ndarray[numpy.int32], partition_measure: higra.higram.PartitionMeasure) -> float
assess_partition(candidate: numpy.ndarray[numpy.uint32], ground_truth: numpy.ndarray[numpy.uint32], partition_measure: higra.higram.PartitionMeasure) -> float
assess_partition(candidate: numpy.ndarray[numpy.int64], ground_truth: numpy.ndarray[numpy.int64], partition_measure: higra.higram.PartitionMeasure) -> float
assess_partition(candidate: numpy.ndarray[numpy.uint64], ground_truth: numpy.ndarray[numpy.uint64], partition_measure: higra.higram.PartitionMeasure) -> float