Obtains the most frequent class (0 or 1) for types.BinaryAnnotation datasets
Obtains the most frequent class (0 or 1) for types.BinaryAnnotation datasets
The annotations dataset (spark Dataset of type types.BinaryAnnotation) to be aggregated
Obtain the most frequent class for each example of the a types.MulticlassAnnotation dataset.
Obtain the most frequent class for each example of the a types.MulticlassAnnotation dataset.
The annotations dataset (spark Dataset of type types.MulticlassAnnotation) to be aggregated
Obtain the mean of the annotations for each example from a types.RealAnnotation.
Obtain the mean of the annotations for each example from a types.RealAnnotation.
The annotations dataset (spark Dataset of type types.RealAnnotation) to be aggregated
Obtains probability of the class being positive for types.BinaryAnnotation datasets
Obtains probability of the class being positive for types.BinaryAnnotation datasets
The annotations dataset (spark Dataset of type types.BinaryAnnotation) to be aggregated
Obtain a list of datasets resulting of applying transformSoftBinary to each class against the others (One vs All) on a types.MulticlassAnnotation dataset.
Obtain a list of datasets resulting of applying transformSoftBinary to each class against the others (One vs All) on a types.MulticlassAnnotation dataset.
It supposes classes go from 0 to nClasses. For example, for a three class problem, there should be classes {0,1,2}.
The annotations dataset (spark Dataset of type types.MulticlassAnnotation) to be aggregated
Provides functions for transforming an annotation dataset into a standard label dataset using the majority voting approach
This object provides several functions for using majority voting style algorithms over annotations datasets (spark datasets with types types.BinaryAnnotation, types.MulticlassAnnotation, or types.RealAnnotation). For discrete types (types.BinaryAnnotation, types.MulticlassAnnotation) the method uses the most frequent class. For continuous types, the mean is used.
The object also provides methods for estimating the probability of a class for the discrete type, computing, for the binary case, the proportion of the positive class and, for the multiclass case, the proportion of each of the classes.
The next example can be found in the examples folder of the project.
0.1.3