Python Homework（第十五周，周三）

题目

代码

import sklearn
from sklearn import metrics
#load a default dataset
from sklearn import datasets
iris = datasets.load_iris()
dataset = datasets.make_classification(n_samples=1000,n_features=10,n_informative=2,n_redundant=2,n_repeated=0,n_classes=2)
#dataset description
print(iris.DESCR)
#data examples (features)
print(iris.data)
#data target labels (classes)
print(iris.target)

from sklearn import cross_validation
kf = cross_validation.KFold(len(iris.data),n_folds=10,shuffle=True)
for train_index,test_index in kf:
	X_train,X_test = dataset[0][train_index],dataset[0][test_index]
	y_train,y_test = dataset[1][train_index],dataset[1][test_index]
print(X_train)
print(y_train)
print(X_test)
print(y_test)

from sklearn.naive_bayes import GaussianNB
clf = GaussianNB()
clf.fit(X_train,y_train)
pred = clf.predict(X_test)
print(pred)
print(y_test)
acc = metrics.accuracy_score(y_test,pred)
print("Gaussian acc: "+str(acc))
f1 = metrics.f1_score(y_test,pred)
print("Gaussian f1: "+str(f1))
auc = metrics.roc_auc_score(y_test,pred)
print("Gaussian auc: "+str(auc))

from sklearn.svm import SVC
clf = SVC(C=1e-00,kernel='rbf',gamma=0.1)
clf.fit(X_train,y_train)
pred = clf.predict(X_test)
print(pred)
print(y_test)
acc = metrics.accuracy_score(y_test,pred)
print("SVC acc: "+str(acc))
f1 = metrics.f1_score(y_test,pred)
print("SVC f1: "+str(f1))
auc = metrics.roc_auc_score(y_test,pred)
print("SVC auc: "+str(auc))

from sklearn.ensemble import RandomForestClassifier
clf = RandomForestClassifier(n_estimators=10)
clf.fit(X_train,y_train)
pred = clf.predict(X_test)
print(pred)
print(y_test)
acc = metrics.accuracy_score(y_test,pred)
print("RandomForestClassifier acc: "+str(acc))
f1 = metrics.f1_score(y_test,pred)
print("RandomForestClassifier f1: "+str(f1))
auc = metrics.roc_auc_score(y_test,pred)
print("RandomForestClassifier auc: "+str(auc))

训练结果和评估结果

总结分析

由评估结果可知，三种算法中Random Forest的性能最好，SVM次之，naïve bayes稍微差一点