交叉表PD.crosstab(index=y_test，columns=y_，rownames=['True']，colnames=['Predict']，margins )=true
froms klearn.metricsimportconfusion _ matrix confusion _ matrix (y_ test，y _ ) #真值y_test.value_counts )预测
#准确率、召回率、f1-score调和平均值froms klearn.metricsimportclassification _ report print (class ification _ report (y _ test，y，yy (m ) ) )提高精度、精度、召回率的方法(归一化操作x_norm1=(x- X.min ) )/)/(X.max )-x.min ) ) X_train，x _ tesess y_ ) ) 65 test_size=0.2 ) KNN=kneighborsclassifier (params={ ' n _ neighbors ' : [ iforinrange (1，30 ) ] ) scoring='accuracy '，cv=6) gcv.fit(x_train，y _ train (y _=GCV.) ) ) ) 65y _ # 0.98245 pring
# Z-Score归一化、标准化x_norm2=(x-x.mean ()/X.std ) ) X_train、X_test、y_train、 y_test=train_test test_size=0.2 ) KNN=kneighborsclassifier (params={ ' n _ neighbors ' 3360 [ iforinrange ] scoring='accuracy '，cv=6) gcv.fit(x_train，y_train ) y_=gcv.) ) ) ) 65
规范化和标准化在“prepocessing”数据预处理模块中进行
sklearn.prepocessing.StandardScaler
sklearn.prepocessing.MinMaxScaler
froms klearn.preprocessingimportminmaxscaler， 标准scaler # minmax scaler为， 与最大值、最小值标准化效果同样，与MMS=minmaxscaler(MMS.fit ) x ) X2=mms.transform(X ) x )为) X - X.min )相当的ss=standardscaler (xx nd=X.get_values ) ) nd-nd.mean ) axis=0) )/将与nd .相当的salary演习——字符串转换为数字importnumpyasnpimportpandasaspdfromsklearn.neighborsimportkneighborsclassifierfromsklearn.moded plit# cv int 6数据来自froms klearn.model _ selectionimportcross _ val _ score，gridsearchcvdata=PD.read_score，gridide
data.columns
data.drop (labels=[ ' final _ weight '，' education '，' capital_gain '，' capital_loss']，axis=1， inplacccs 0:-1 ) y=data [ ' salary ' ] KNN=kneighborsclassifier (KNN.fit (x，y ) #方法将数据中的str转换为int，float
NP.argwhere(u=='local-gov ' ) 0，0 ) #4defconvert ) x ) :returnNP.argwhere(u==x ) ) 0， 0 ) x(workclass )、x ) workclass )、map ) convert ) cols )、relationshatus )、occupation )、relations hard ' native _ native . unique(defconvert ) x ) :returnNP.argwhere ) u==x )0 0 ) x[col]=x[col] . map(convert ) knn=KNeighborsClassifier ) (KFold=KFold(10 ) knn=KNeighborsClassifier ) ) ) kfold ) knn y )