文章作者：Tyan
部落格：noahsnail.com  |  CSDN  |  簡書

本文主要介紹scikit-learn中的交叉驗證。

• Demo 1

import numpy as np
from sklearn import datasets
from sklearn.cross_validation import train_test_split
from sklearn.neighbors import KNeighborsClassifier
from sklearn.cross_validation import cross_val_score 

# 載入iris資料集
 

iris = datasets.load_iris()
# 讀取特徵
X = iris.data
# 讀取分類標籤
y = iris.target
# 定義分類器
knn = KNeighborsClassifier(n_neighbors = 5)
# 進行交叉驗證資料評估, 資料分為5部分, 每次用一部分作為測試集
scores = cross_val_score(knn, X, y, cv = 5, scoring = 'accuracy')
# 輸出5次交叉驗證的準確率
print scores

• 結果

[ 0.96666667  1.          0.93333333  0.96666667
 
  1.        ]

• Demo 2

import numpy as np
import matplotlib.pyplot as plt
from sklearn import datasets
from sklearn.cross_validation import train_test_split
from sklearn.neighbors import KNeighborsClassifier
from sklearn.cross_validation import cross_val_score 

# 確定knn中k的取值

# 載入iris資料集
iris = datasets.load_iris()
# 讀取特徵
 

X = iris.data
# 讀取分類標籤
y = iris.target
# 定義knn中k的取值, 0-10
k_range = range(1, 30)
# 儲存k對應的準確率
k_scores = []
# 計算每個k取值對應的準確率
for k in k_range:
    # 獲得knn分類器
    knn = KNeighborsClassifier(n_neighbors = k)
    # 對資料進行交叉驗證求準確率
    scores = cross_val_score(knn, X, y, cv = 10, scoring = 'accuracy')
    # 儲存交叉驗證結果的準確率均值
    k_scores.append(scores.mean())

# 繪製k取不同值時的準確率變化影象
plt.plot(k_range, k_scores)
plt.xlabel('K Value in KNN')
plt.ylabel('Cross-Validation Mean Accuracy')
plt.show()

• 結果

參考資料