sklearn

[yufang67]

TL;DR

Classification, Regression, Clustering, Dimensionality reduction, Model selection, Preprocessing

Load and read data

API

• estimator

• model.fit()

• model.coef_ , model.labels_ , model.intercept_ , model.cluster_centers_ , model.inertia_

• input X should be two dimension: X = [:,np.newaxis]

• predictor

• model.predict()

• model.predict_proba()

• model.score(trainX, testX)

• model.decision_function(testX)

• transformer (.fit() .transform())

• LabelEncoder: 1d, LE.classes_

• OrdinalEncoder: nd, OE.categories_ , 有顺序

• OneHotEncoder: 2d

• MinMaxScaler: 2d, X.reshape(-1,1)

• StandardScaler: 2d

• meta estimator

• ensemble.BaggingClassifier

• ensemble.VotingClassifier

• multiclass.OneVsOneClassifier

• multiclass.OneVsRestClassifier

• multioutput.MultiOutputClassifier

• impute.SimpleImputer

• model_selection.GridSearchCV

• model_selection.RandomizedSearchCV

• pipeline.Pipeline
  

Ref

非常详细的sklearn介绍