-
Notifications
You must be signed in to change notification settings - Fork 66
/
Copy pathsk-交叉验证.py
109 lines (88 loc) · 5.58 KB
/
sk-交叉验证.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
from sklearn.model_selection import train_test_split,cross_val_score,cross_validate # 交叉验证所需的函数
from sklearn.model_selection import KFold,LeaveOneOut,LeavePOut,ShuffleSplit # 交叉验证所需的子集划分方法
from sklearn.model_selection import StratifiedKFold,StratifiedShuffleSplit # 分层分割
from sklearn.model_selection import GroupKFold,LeaveOneGroupOut,LeavePGroupsOut,GroupShuffleSplit # 分组分割
from sklearn.model_selection import TimeSeriesSplit # 时间序列分割
from sklearn import datasets # 自带数据集
from sklearn import svm # SVM算法
from sklearn import preprocessing # 预处理模块
from sklearn.metrics import recall_score # 模型度量
iris = datasets.load_iris() # 加载数据集
print('样本集大小：',iris.data.shape,iris.target.shape)
# ===================================数据集划分,训练模型==========================
X_train, X_test, y_train, y_test = train_test_split(iris.data, iris.target, test_size=0.4, random_state=0) # 交叉验证划分训练集和测试集.test_size为测试集所占的比例
print('训练集大小：',X_train.shape,y_train.shape) # 训练集样本大小
print('测试集大小：',X_test.shape,y_test.shape) # 测试集样本大小
clf = svm.SVC(kernel='linear', C=1).fit(X_train, y_train) # 使用训练集训练模型
print('准确率：',clf.score(X_test, y_test)) # 计算测试集的度量值（准确率）
# 如果涉及到归一化，则在测试集上也要使用训练集模型提取的归一化函数。
scaler = preprocessing.StandardScaler().fit(X_train) # 通过训练集获得归一化函数模型。（也就是先减几，再除以几的函数）。在训练集和测试集上都使用这个归一化函数
X_train_transformed = scaler.transform(X_train)
clf = svm.SVC(kernel='linear', C=1).fit(X_train_transformed, y_train) # 使用训练集训练模型
X_test_transformed = scaler.transform(X_test)
print(clf.score(X_test_transformed, y_test)) # 计算测试集的度量值（准确度）
# ===================================直接调用交叉验证评估模型==========================
clf = svm.SVC(kernel='linear', C=1)
scores = cross_val_score(clf, iris.data, iris.target, cv=5) #cv为迭代次数。
print(scores) # 打印输出每次迭代的度量值（准确度）
print("Accuracy: %0.2f (+/- %0.2f)" % (scores.mean(), scores.std() * 2)) # 获取置信区间。（也就是均值和方差）
# ===================================多种度量结果======================================
scoring = ['precision_macro', 'recall_macro'] # precision_macro为精度，recall_macro为召回率
scores = cross_validate(clf, iris.data, iris.target, scoring=scoring,cv=5, return_train_score=True)
sorted(scores.keys())
print('测试结果：',scores) # scores类型为字典。包含训练得分，拟合次数， score-times （得分次数）
# ==================================K折交叉验证、留一交叉验证、留p交叉验证、随机排列交叉验证==========================================
# k折划分子集
kf = KFold(n_splits=2)
for train, test in kf.split(iris.data):
print("k折划分：%s %s" % (train.shape, test.shape))
break
# 留一划分子集
loo = LeaveOneOut()
for train, test in loo.split(iris.data):
print("留一划分：%s %s" % (train.shape, test.shape))
break
# 留p划分子集
lpo = LeavePOut(p=2)
for train, test in loo.split(iris.data):
print("留p划分：%s %s" % (train.shape, test.shape))
break
# 随机排列划分子集
ss = ShuffleSplit(n_splits=3, test_size=0.25,random_state=0)
for train_index, test_index in ss.split(iris.data):
print("随机排列划分：%s %s" % (train.shape, test.shape))
break
# ==================================分层K折交叉验证、分层随机交叉验证==========================================
skf = StratifiedKFold(n_splits=3) #各个类别的比例大致和完整数据集中相同
for train, test in skf.split(iris.data, iris.target):
print("分层K折划分：%s %s" % (train.shape, test.shape))
break
skf = StratifiedShuffleSplit(n_splits=3) # 划分中每个类的比例和完整数据集中的相同
for train, test in skf.split(iris.data, iris.target):
print("分层随机划分：%s %s" % (train.shape, test.shape))
break
# ==================================组 k-fold交叉验证、留一组交叉验证、留 P 组交叉验证、Group Shuffle Split==========================================
X = [0.1, 0.2, 2.2, 2.4, 2.3, 4.55, 5.8, 8.8, 9, 10]
y = ["a", "b", "b", "b", "c", "c", "c", "d", "d", "d"]
groups = [1, 1, 1, 2, 2, 2, 3, 3, 3, 3]
# k折分组
gkf = GroupKFold(n_splits=3) # 训练集和测试集属于不同的组
for train, test in gkf.split(X, y, groups=groups):
print("组 k-fold分割：%s %s" % (train, test))
# 留一分组
logo = LeaveOneGroupOut()
for train, test in logo.split(X, y, groups=groups):
print("留一组分割：%s %s" % (train, test))
# 留p分组
lpgo = LeavePGroupsOut(n_groups=2)
for train, test in lpgo.split(X, y, groups=groups):
print("留 P 组分割：%s %s" % (train, test))
# 随机分组
gss = GroupShuffleSplit(n_splits=4, test_size=0.5, random_state=0)
for train, test in gss.split(X, y, groups=groups):
print("随机分割：%s %s" % (train, test))
# ==================================时间序列分割==========================================
tscv = TimeSeriesSplit(n_splits=3)
TimeSeriesSplit(max_train_size=None, n_splits=3)
for train, test in tscv.split(iris.data):
print("时间序列分割：%s %s" % (train, test))