ComputePerInstanceStatistics¶
-
class
ComputePerInstanceStatistics.
ComputePerInstanceStatistics
(evaluationMetric='all', labelCol=None, scoredLabelsCol=None, scoredProbabilitiesCol=None, scoresCol=None)[source]¶ Bases:
mmlspark.Utils.ComplexParamsMixin
,pyspark.ml.util.JavaMLReadable
,pyspark.ml.util.JavaMLWritable
,pyspark.ml.wrapper.JavaTransformer
Evaluates the given scored dataset with per instance metrics.
The Regression metrics are:
- “L1_loss”
- “L2_loss”
The Classification metrics are:
- “log_loss”
Parameters: - evaluationMetric (str) – Metric to evaluate models with (default: all)
- labelCol (str) – The name of the label column
- scoredLabelsCol (str) – Scored labels column name, only required if using SparkML estimators
- scoredProbabilitiesCol (str) – Scored probabilities, usually calibrated from raw scores, only required if using SparkML estimators
- scoresCol (str) – Scores or raw prediction column name, only required if using SparkML estimators
-
getEvaluationMetric
()[source]¶ Returns: Metric to evaluate models with (default: all) Return type: str
-
getScoredLabelsCol
()[source]¶ Returns: Scored labels column name, only required if using SparkML estimators Return type: str
-
getScoredProbabilitiesCol
()[source]¶ Returns: Scored probabilities, usually calibrated from raw scores, only required if using SparkML estimators Return type: str
-
getScoresCol
()[source]¶ Returns: Scores or raw prediction column name, only required if using SparkML estimators Return type: str
-
setEvaluationMetric
(value)[source]¶ Parameters: evaluationMetric (str) – Metric to evaluate models with (default: all)
-
setParams
(evaluationMetric='all', labelCol=None, scoredLabelsCol=None, scoredProbabilitiesCol=None, scoresCol=None)[source]¶ Set the (keyword only) parameters
Parameters: - evaluationMetric (str) – Metric to evaluate models with (default: all)
- labelCol (str) – The name of the label column
- scoredLabelsCol (str) – Scored labels column name, only required if using SparkML estimators
- scoredProbabilitiesCol (str) – Scored probabilities, usually calibrated from raw scores, only required if using SparkML estimators
- scoresCol (str) – Scores or raw prediction column name, only required if using SparkML estimators
-
setScoredLabelsCol
(value)[source]¶ Parameters: scoredLabelsCol (str) – Scored labels column name, only required if using SparkML estimators