ML之xgboost：利用xgboost算法(sklearn+7CrVa)训练mushroom蘑菇数据集(22+1,6513+1611)来预测蘑菇是否毒性(二分类预测)

输出结果

设计思路

核心代码

kfold = StratifiedKFold(n_splits=10, random_state=7)                                

#fit_params = {'eval_metric':"logloss"}

#results = cross_val_score(bst, X_train, y_train, cv=kfold, fit_params)

results = cross_val_score(bst, X_train, y_train, cv=kfold)                          

print(results)

print("7-CrVa Accuracy Mean(STD): %.2f%% (%.2f%%)" % (results.mean()*100, results.std()*100))  #输出

x = range(0,len(results))

y1 = results

y2 = [results.mean()]*10  

Xlabel = 'n_splits'

Ylabel = 'Accuracy'

title = 'mushroom datase: xgboost(sklearn+7CrVa) model'

plt.plot(x,y1,'g')                      #绘制曲线

plt.plot(x,y2,'r--')                    #平均值曲线

plt.xlabel(Xlabel)

plt.ylabel(Ylabel)

plt.title(title)

plt.show()