python作业 sklearn习题

Posted on | In | | Visitors:

学习sklearn库，进行数据的生成、分组、回归及算法性能分析。

步骤

  • 数据生成

    1
    dataset = datasets.make_classification(n_samples=1000, n_features=10,n_informative=2, n_redundant=2, n_repeated=0, n_classes=2)

  • 数据分组

    1
    2
    3
    4
    5
    kf = cross_validation.KFold(len(data), n_folds=10, shuffle=True)

    for train_index, test_index in kf:
    X_train, y_train = data[train_index], target[train_index]
    X_test, y_test = data[test_index], target[test_index]

  • 对于每组数据，进行回归：

    • Gaussian Navie Bayes：

      1
      2
      3
      clf = GaussianNB()
      clf.fit(X_train, y_train)
      pred = clf.predict(X_test)

    • SVC:

      1
      2
      3
      clf = SVC(C=1, kernel='rbf')
      clf.fit(X_train, y_train)
      pred = clf.predict(X_test)

    • Random Forest:

      1
      2
      3
      clf = RandomForestClassifier(n_estimators=100)
      clf.fit(X_train, y_train)
      pred = clf.predict(X_test)

  • 数据分析：

    • accuracy：

      1
      metrics.accuracy_score(y_test, pred)

    • F1-score：

      1
      metrics.f1_score(y_test, pred)

    • AUC ROC：

      1
      metrics.roc_auc_score(y_test, pred)

总代码

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
from sklearn import datasets, cross_validation
from sklearn.naive_bayes import GaussianNB
from sklearn.svm import SVC
from sklearn.ensemble import RandomForestClassifier
from sklearn import metrics
# create a classification dataset
dataset = datasets.make_classification(n_samples=1000, n_features=10,n_informative=2, n_redundant=2, n_repeated=0, n_classes=2)
data = dataset[0]
target = dataset[1]

# Split the dataset using 10-fold cross validation
kf = cross_validation.KFold(len(data), n_folds=10, shuffle=True)

acc = {"GaussianNB":[], "SVC":[], "RandomForest":[]}
f1 = {"GaussianNB":[], "SVC":[], "RandomForest":[]}
auc = {"GaussianNB":[], "SVC":[], "RandomForest":[]}

for train_index, test_index in kf:
    X_train, y_train = data[train_index], target[train_index]
    X_test, y_test = data[test_index], target[test_index]

    # GaussianNB
    clf = GaussianNB()
    clf.fit(X_train, y_train)
    pred = clf.predict(X_test)
    acc["GaussianNB"].append( metrics.accuracy_score(y_test, pred) )
    f1["GaussianNB"].append( metrics.f1_score(y_test, pred) )
    auc["GaussianNB"].append( metrics.roc_auc_score(y_test, pred) )
    
    # SVC (possible C values [1e-02, 1e-01, 1e00, 1e01, 1e02], RBF kernel)
    clf = SVC(C=1, kernel='rbf')
    clf.fit(X_train, y_train)
    pred = clf.predict(X_test)
    acc["SVC"].append( metrics.accuracy_score(y_test, pred) )
    f1["SVC"].append( metrics.f1_score(y_test, pred) )
    auc["SVC"].append( metrics.roc_auc_score(y_test, pred) )

    # RandomForestClassifier (possible n estimators values [10, 100, 1000])
    clf = RandomForestClassifier(n_estimators=100)
    clf.fit(X_train, y_train)
    pred = clf.predict(X_test)
    acc["RandomForest"].append( metrics.accuracy_score(y_test, pred) )
    f1["RandomForest"].append( metrics.f1_score(y_test, pred) )
    auc["RandomForest"].append( metrics.roc_auc_score(y_test, pred) )

print("acc: ")
for item in acc.items():
    print("--- "+item[0] +": " + str(sum(item[1])/len(item[1])) )

print("f1: ")
for item in f1.items():
    print("--- "+item[0] +": " + str(sum(item[1])/len(item[1])) )

print("auc: ")
for item in auc.items():
    print("--- "+item[0] +": " + str(sum(item[1])/len(item[1])) )

结果

1
2
3
4
5
6
7
8
9
10
11
12
acc:
--- GaussianNB: 0.885
--- SVC: 0.9109999999999999
--- RandomForest: 0.9199999999999999
f1:
--- GaussianNB: 0.8814437089165666
--- SVC: 0.907042930534323
--- RandomForest: 0.9176129601622535
auc:
--- GaussianNB: 0.8832256906743698
--- SVC: 0.9101616490472964
--- RandomForest: 0.9190646809857602