Evaluation¶
This module provides evaluation methods for classification, regression and clustering. Available metrics include:
AUC: Compute AUC for binary classification.
KS: Compute Kolmogorov-Smirnov for binary classification.
LIFT: Compute lift of binary classification.
PRECISION: Compute the precision for binary and multi-classification
RECALL: Compute the recall for binary and multi-classification
ACCURACY: Compute the accuracy for binary and multi-classification
EXPLAINED_VARIANCE: Compute explain variance for regression tasks
MEAN_ABSOLUTE_ERROR: Compute mean absolute error for regression tasks
MEAN_SQUARED_ERROR: Compute mean square error for regression tasks
MEAN_SQUARED_LOG_ERROR: Compute mean squared logarithmic error for regression tasks
MEDIAN_ABSOLUTE_ERROR: Compute median absolute error for regression tasks
R2_SCORE: Compute R^2 (coefficient of determination) score for regression tasks
ROOT_MEAN_SQUARED_ERROR: Compute the root of mean square error for regression tasks
JACCARD_SIMILARITY_SCORE:Compute Jaccard similarity score for clustering tasks (labels are needed)
ADJUSTED_RAND_SCORE:Compute adjusted rand score for clustering tasks (labels are needed)
FOWLKES_MALLOWS_SCORE:Compute Fowlkes Mallows score for clustering tasks (labels are needed)
DAVIES_BOULDIN_INDEX:Compute Davies Bouldin index for clustering tasks
DISTANCE_MEASURE:Compute cluster information in clustering algorithms
CONTINGENCY_MATRIX:Compute contingency matrix for clustering tasks (labels are needed)
Param¶
-
class
EvaluateParam
(eval_type='binary', pos_label=1, need_run=True, metrics=None, run_clustering_arbiter_metric=False, unfold_multi_result=False)¶ Define the evaluation method of binary/multiple classification and regression
- Parameters
eval_type (string, support 'binary' for HomoLR, HeteroLR and Secureboosting. support 'regression' for Secureboosting. 'multi' is not support these version) –
unfold_multi_result (bool, unfold multi result and get several one-vs-rest binary classification results) –
pos_label (specify positive label type, can be int, float and str, this depend on the data's label, this parameter effective only for 'binary') –
need_run (bool, default True) – Indicate if this module needed to be run