第十五周作业 sklearn

最新推荐文章于 2018-06-20 22:57:03 发布

cjf16337023

最新推荐文章于 2018-06-20 22:57:03 发布

阅读量202

点赞数

分类专栏：作业

本文链接：https://blog.csdn.net/cjf16337023/article/details/80731876

版权

作业专栏收录该内容

16 篇文章 0 订阅

订阅专栏

题目：

import numpy as np  
from sklearn import datasets, metrics
from sklearn.naive_bayes import GaussianNB  
from sklearn.svm import SVC  
from sklearn.ensemble import RandomForestClassifier  
from sklearn.model_selection import KFold
  
def evaluate(algorithm_name, test_y, predict_x):
    print(algorithm_name)
    print("Accuracy:    "+str(metrics.accuracy_score(test_y, predict_x)))  
    print("F1-score:    "+str(metrics.f1_score(test_y, predict_x)))
    print("AUC ROC:     " +str(metrics.roc_auc_score(test_y, predict_x)))
    print("——————————————————————————————————————")
  
#生成创建分类数据集
dataset = datasets.make_classification(n_samples=1000, n_features=10)  
  
#使用10重交叉验证对数据集进行拆分
fold = KFold(n_splits=10)  
  
#Train the algorithms  
time = 1  
for train, test in fold.split(dataset[0]): 
    print("\n\n【Test "+str(time) + "】\n")  
    time += 1  
    train_x, test_x = dataset[0][train], dataset[0][test]  
    train_y, test_y = dataset[1][train], dataset[1][test]  
  
    # GaussianNB  
    algorithm = GaussianNB()  
    algorithm.fit(train_x, train_y)  
    predict_x = algorithm.predict(test_x)  
    evaluate("[GaussianNB]:", test_y, predict_x)  
  
    # SVC (possible C values [1e-02, 1e-01, 1e00, 1e01, 1e02], RBF kernel)  
    algorithm = SVC(C=1e-2, gamma=0.1, kernel='rbf')  
    algorithm.fit(train_x, train_y)  
    predict_x = algorithm.predict(test_x)  
    evaluate("[SVC]:", test_y, predict_x)  
  
    # RandomForestClassifier (possible n estimators values [10, 100, 1000])  
    algorithm = RandomForestClassifier(n_estimators=100)  
    algorithm.fit(train_x, train_y)  
    predict_x = algorithm.predict(test_x)  
    evaluate("[RandomForestClassifier]:", test_y, predict_x)

测试几次的结果如下图