Reference: Http://scikit-learn.org/stable/modules/model_evaluation.html#scoring-parameter
Three methods to evaluate the predictive quality of the model:
- Estimator Score Method: estimators have score method as the default evaluation criteria, not part of this section, specific reference to different estimators documents.
- scoring parameter : model-evaluation tools using Cross-validation (Such as cross_validation.cross_val_score and grid_search. GRIDSEARCHCV ) rely on a internal scoring strategy. This section discusses the scoring Parameter:defining model evaluation rules . (Refer to the first section)
- Metric functions: The Metrics module can evaluate the predictive quality more comprehensively, this section discusses classification metrics, Multilabel Ranking metrics, Regression metrics and Clustering metrics. (Refer to section two or three, four or five)
Finally, the Dummy estimators is introduced, and the strategy of stochastic guessing can be used as the baseline of predictive quality evaluation. (refer to section Sixth)
See Also
For "pairwise" metrics, between samples and no estimators or predictions, see the pairwise metrics, Affiniti Es and Kernels section.
The specific content has time to write again ...
1.
ThescoringParameter:defining Model Evaluation Rules
2.
Classification metrics
3.
Multilabel ranking Metrics
4.
Regression Metrics
5.
Clustering metrics
6.
Dummy estimators
Copyright NOTICE: This article for Bo Master original article, without Bo Master permission not reproduced.
scikit-learn:3.3. Model evaluation:quantifying the quality of predictions