FeaturesType
- Type of input features. E.g., Vector
M
- Concrete Model typepublic abstract class ProbabilisticClassificationModel<FeaturesType,M extends ProbabilisticClassificationModel<FeaturesType,M>> extends ClassificationModel<FeaturesType,M> implements ProbabilisticClassifierParams
Model produced by a ProbabilisticClassifier
.
Classes are indexed {0, 1, ..., numClasses - 1}.
Constructor and Description |
---|
ProbabilisticClassificationModel() |
Modifier and Type | Method and Description |
---|---|
static void |
normalizeToProbabilitiesInPlace(DenseVector v)
Normalize a vector of raw predictions to be a multinomial probability vector, in place.
|
M |
setProbabilityCol(String value) |
M |
setThresholds(double[] value) |
Dataset<Row> |
transform(Dataset<?> dataset)
Transforms dataset by reading from
featuresCol , and appending new columns as specified by
parameters:
- predicted labels as predictionCol of type Double
- raw predictions (confidences) as rawPredictionCol of type Vector
- probability of each class as probabilityCol of type Vector . |
numClasses, predict, setRawPredictionCol
numFeatures, setFeaturesCol, setPredictionCol, transformSchema
transform, transform, transform
equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
validateAndTransformSchema
getLabelCol, labelCol
featuresCol, getFeaturesCol
getPredictionCol, predictionCol
clear, copy, copyValues, defaultCopy, defaultParamMap, explainParam, explainParams, extractParamMap, extractParamMap, get, getDefault, getOrDefault, getParam, hasDefault, hasParam, isDefined, isSet, paramMap, params, set, set, set, setDefault, setDefault, shouldOwn
toString, uid
getRawPredictionCol, rawPredictionCol
getProbabilityCol, probabilityCol
getThresholds, thresholds
initializeLogging, initializeLogIfNecessary, initializeLogIfNecessary, isTraceEnabled, log_, log, logDebug, logDebug, logError, logError, logInfo, logInfo, logName, logTrace, logTrace, logWarning, logWarning
public static void normalizeToProbabilitiesInPlace(DenseVector v)
The input raw predictions should be nonnegative. The output vector sums to 1.
NOTE: This is NOT applicable to all models, only ones which effectively use class instance counts for raw predictions.
v
- (undocumented)IllegalArgumentException
- if the input vector is all-0 or including negative valuespublic M setProbabilityCol(String value)
public M setThresholds(double[] value)
public Dataset<Row> transform(Dataset<?> dataset)
featuresCol
, and appending new columns as specified by
parameters:
- predicted labels as predictionCol
of type Double
- raw predictions (confidences) as rawPredictionCol
of type Vector
- probability of each class as probabilityCol
of type Vector
.
transform
in class ClassificationModel<FeaturesType,M extends ProbabilisticClassificationModel<FeaturesType,M>>
dataset
- input dataset