Python基于sklearn库的分类算法简单应用示例
本文实例讲述了Python基于sklearn库的分类算法简单应用。分享给大家供大家参考，具体如下：
scikit-learn已经包含在Anaconda中。也可以在官方下载源码包进行安装。本文代码里封装了如下机器学习算法，我们修改数据加载函数，即可一键测试：
#coding=gbk
'''
Createdon2016年6月4日
@author:bryan
'''
importtime
fromsklearnimportmetrics
importpickleaspickle
importpandasaspd
#MultinomialNaiveBayesClassifier
defnaive_bayes_classifier(train_x,train_y):
fromsklearn.naive_bayesimportMultinomialNB
model=MultinomialNB(alpha=0.01)
model.fit(train_x,train_y)
returnmodel
#KNNClassifier
defknn_classifier(train_x,train_y):
fromsklearn.neighborsimportKNeighborsClassifier
model=KNeighborsClassifier()
model.fit(train_x,train_y)
returnmodel
#LogisticRegressionClassifier
deflogistic_regression_classifier(train_x,train_y):
fromsklearn.linear_modelimportLogisticRegression
model=LogisticRegression(penalty='l2')
model.fit(train_x,train_y)
returnmodel
#RandomForestClassifier
defrandom_forest_classifier(train_x,train_y):
fromsklearn.ensembleimportRandomForestClassifier
model=RandomForestClassifier(n_estimators=8)
model.fit(train_x,train_y)
returnmodel
#DecisionTreeClassifier
defdecision_tree_classifier(train_x,train_y):
fromsklearnimporttree
model=tree.DecisionTreeClassifier()
model.fit(train_x,train_y)
returnmodel
#GBDT(GradientBoostingDecisionTree)Classifier
defgradient_boosting_classifier(train_x,train_y):
fromsklearn.ensembleimportGradientBoostingClassifier
model=GradientBoostingClassifier(n_estimators=200)
model.fit(train_x,train_y)
returnmodel
#SVMClassifier
defsvm_classifier(train_x,train_y):
fromsklearn.svmimportSVC
model=SVC(kernel='rbf',probability=True)
model.fit(train_x,train_y)
returnmodel
#SVMClassifierusingcrossvalidation
defsvm_cross_validation(train_x,train_y):
fromsklearn.grid_searchimportGridSearchCV
fromsklearn.svmimportSVC
model=SVC(kernel='rbf',probability=True)
param_grid={'C':[1e-3,1e-2,1e-1,1,10,100,1000],'gamma':[0.001,0.0001]}
grid_search=GridSearchCV(model,param_grid,n_jobs=1,verbose=1)
grid_search.fit(train_x,train_y)
best_parameters=grid_search.best_estimator_.get_params()
forpara,valinlist(best_parameters.items()):
print(para,val)
model=SVC(kernel='rbf',C=best_parameters['C'],gamma=best_parameters['gamma'],probability=True)
model.fit(train_x,train_y)
returnmodel
defread_data(data_file):
data=pd.read_csv(data_file)
train=data[:int(len(data)*0.9)]
test=data[int(len(data)*0.9):]
train_y=train.label
train_x=train.drop('label',axis=1)
test_y=test.label
test_x=test.drop('label',axis=1)
returntrain_x,train_y,test_x,test_y
if__name__=='__main__':
data_file="H:\\Research\\data\\trainCG.csv"
thresh=0.5
model_save_file=None
model_save={}
test_classifiers=['NB','KNN','LR','RF','DT','SVM','SVMCV','GBDT']
classifiers={'NB':naive_bayes_classifier,
'KNN':knn_classifier,
'LR':logistic_regression_classifier,
'RF':random_forest_classifier,
'DT':decision_tree_classifier,
'SVM':svm_classifier,
'SVMCV':svm_cross_validation,
'GBDT':gradient_boosting_classifier
}
print('readingtrainingandtestingdata...')
train_x,train_y,test_x,test_y=read_data(data_file)
forclassifierintest_classifiers:
print('*******************%s********************'%classifier)
start_time=time.time()
model=classifiers[classifier](train_x,train_y)
print('trainingtook%fs!'%(time.time()-start_time))
predict=model.predict(test_x)
ifmodel_save_file!=None:
model_save[classifier]=model
precision=metrics.precision_score(test_y,predict)
recall=metrics.recall_score(test_y,predict)
print('precision:%.2f%%,recall:%.2f%%'%(100*precision,100*recall))
accuracy=metrics.accuracy_score(test_y,predict)
print('accuracy:%.2f%%'%(100*accuracy))
ifmodel_save_file!=None:
pickle.dump(model_save,open(model_save_file,'wb'))
测试结果如下：
readingtrainingandtestingdata...
*******************NB********************
trainingtook0.004986s!
precision:78.08%,recall:71.25%
accuracy:74.17%
*******************KNN********************
trainingtook0.017545s!
precision:97.56%,recall:100.00%
accuracy:98.68%
*******************LR********************
trainingtook0.061161s!
precision:89.16%,recall:92.50%
accuracy:90.07%
*******************RF********************
trainingtook0.040111s!
precision:96.39%,recall:100.00%
accuracy:98.01%
*******************DT********************
trainingtook0.004513s!
precision:96.20%,recall:95.00%
accuracy:95.36%
*******************SVM********************
trainingtook0.242145s!
precision:97.53%,recall:98.75%
accuracy:98.01%
*******************SVMCV********************
Fitting3foldsforeachof14candidates,totalling42fits
[Parallel(n_jobs=1)]:Done 42outof 42|elapsed: 6.8sfinished
probabilityTrue
verboseFalse
coef00.0
degree3
tol0.001
shrinkingTrue
cache_size200
gamma0.001
max_iter-1
C1000
decision_function_shapeNone
random_stateNone
class_weightNone
kernelrbf
trainingtook7.434668s!
precision:98.75%,recall:98.75%
accuracy:98.68%
*******************GBDT********************
trainingtook0.521916s!
precision:97.56%,recall:100.00%
accuracy:98.68%
更多关于Python相关内容感兴趣的读者可查看本站专题：《Python数学运算技巧总结》、《Python数据结构与算法教程》、《Python函数使用技巧总结》、《Python字符串操作技巧汇总》、《Python入门与进阶经典教程》及《Python文件与目录操作技巧汇总》
希望本文所述对大家Python程序设计有所帮助。