sklearn学习笔记

sklearn学习笔记模型验证方法1.学习率曲线（learn_curve）2.交叉验证得分（cross_val_score）3.验证曲线（validation_curve）一.学习率曲线计算指定的学习器模型在不同大小的训练集上经过交叉验证的训练得分和测试得分首先，用一个交叉验证生成器划分整体数据集K次，每一次划分都有一个训练集和测试集。然后从每次划分的训练集中拿出若干个数量不断增加的子集，在这些训练集上训练模型。然后再计算模型在对应的子训练集和测试集上的得分，最后，对于在每种子训练集大小下，将K次训练集得分和测试集得分分别进行平均。[python]viewplaincopyprint?#importnumpyasnp#fromsklearn.model_selectionimportlearning_curve,ShuffleSplit#fromsklearn.datasetsimportload_digits#fromsklearn.naive_bayesimportGaussianNB#fromsklearnimportsvm#importmatplotlib.pyplotasplt#defplot_learning_curve(estimator,title,X,y,ylim=None,cv=None,n_jobs=1,train_size=np.linspace(.1,1.0,5)):#if__name__=='__main__':#plt.figure()#plt.title(title)#ifylimisnotNone:#plt.ylim(*ylim)#plt.xlabel('Trainingexample')#plt.ylabel('score')#train_sizes,train_scores,test_scores=learning_curve(estimator,X,y,cv=cv,n_jobs=n_jobs,train_sizes=train_size)#train_scores_mean=np.mean(train_scores,axis=1)#train_scores_std=np.std(train_scores,axis=1)#test_scores_mean=np.mean(test_scores,axis=1)#test_scores_std=np.std(test_scores,axis=1)#plt.grid()#区域#plt.fill_between(train_sizes,train_scores_mean-train_scores_std,#train_scores_mean+train_scores_std,alpha=0.1,#color=r)#plt.fill_between(train_sizes,test_scores_mean-test_scores_std,#test_scores_mean+test_scores_std,alpha=0.1,#color=g)#plt.plot(train_sizes,train_scores_mean,'o-',color='r',#label=Trainingscore)#plt.plot(train_sizes,test_scores_mean,'o-',color=g,#label=Cross-validationscore)#plt.legend(loc=best)#returnplt#digits=load_digits()#X=digits.data#y=digits.target#cv=ShuffleSplit(n_splits=100,test_size=0.2,random_state=0)#切割100ci#estimator=GaussianNB()#title=LearningCurves(naive_bayes)#plot_learning_curve(estimator,title,X,y,ylim=(0.7,1.01),cv=cv,n_jobs=4)#title=LearningCurves(SVM,RBFkernel,$\gamma=0.001$)#cv=ShuffleSplit(n_splits=10,test_size=0.2,random_state=0)#交叉验证传入别的方法，而不是默认的k折交叉验证#estimator=svm.SVC(gamma=0.001)#plot_learning_curve(estimator,title,X,y,(0.7,1.01),cv=cv,n_jobs=4)#plt.show()二.交叉验证得分[python]viewplaincopyprint?#importmatplotlib.pyplotasplt#fromsklearn.model_selectionimportcross_val_score#importnumpyasnp#fromsklearnimportdatasets,svm#digits=datasets.load_digits()#x=digits.data#y=digits.target#vsc=svm.SVC(kernel='linear')#if__name__=='__main__':#c_S=np.logspace(-10,0,10)#在范围内取是个对数##print(length,len(c_S))#scores=list()#scores_std=list()#forcinc_S:#vsc.C=c#this_scores=cross_val_score(vsc,x,y,n_jobs=4)#多线程n_jobs，默认三次交叉验证#scores.append(np.mean(this_scores))#scores_std.append(np.std(this_scores))#plt.figure(1,figsize=(4,3))#绘图#plt.clf()#plt.semilogx(c_S,scores)#划线#plt.semilogx(c_S,np.array(scores)+np.array(scores_std),'b--')#plt.semilogx(c_S,np.array(scores)-np.array(scores_std),'b--')#locs,labels=plt.yticks()#plt.yticks(locs,list(map(lambdaX:%g%X,locs)))#阶段点#plt.ylabel('CVscore')#plt.xlabel('parameterC')#plt.ylim(0,1.1)#范围#plt.show()三.验证曲线当某个参数不断变化是，在每一个取值上计算出的模型在训练集和测试集上的得分在一个不断变化的参数上计算学习器的得分，类似于只有一个参数的网格搜索，但是这个函数也会计算训练集上的得分[python]viewplaincopyprint?#fromsklearnimportsvm#fromsklearn.model_selectionimportvalidation_curve#fromsklearn.datasetsimportload_digits#importnumpyasnp#importmatplotlib.pyplotasplt#digits=load_digits()#X=digits.data#y=digits.target#param_range=np.logspace(-6,-1,5)#vsc=svm.SVC()#train_score,test_score=validation_curve(vsc,X,y,param_name='gamma',param_range=param_range,cv=10,scoring=accuracy,n_jobs=1)#train_score_mean=np.mean(train_score,axis=1)#train_score_std=np.std(train_score,axis=1)#test_score_mean=np.mean(test_score,axis=1)#test_score_std=np.std(test_score,axis=1)#plt.title(validationcurvewithSVM)#plt.xlabel($\gamma%)#plt.ylabel(Score)#plt.ylim()#lw=2#plt.semilogx(param_range,train_score_mean,label=trainingscore,color=darkorange,lw=lw)#plt.fill_between(param_range,train_score_mean-train_score_std,train_score_mean+train_score_std,alpha=0.2,color=navy,lw=lw)#plt.legend(loc=best)#plt.show()