如何知道0级有多少，1级有多少？

Question

我有一个代码可以提供 SVM 的准确性，但我想知道 0 类和 1 类有多少。

这里是代码

from sklearn.svm import SVC
from sklearn.metrics import accuracy_score

clf = SVC(C=10000.0, kernel='rbf')
t0 = time()
clf.fit(features_train, labels_train)
print "training_time:", round(time()-t0, 3), "s"
t0 = time()
pred = clf.predict(features_test)
print "prediction time:", round(time()-t0, 3), "s"
acc = accuracy_score(pred, labels_test)
print acc

我试过下面这段代码，但没有成功...

from sklearn.svm import SVC
from sklearn.metrics import accuracy_score

clf = SVC(C=10000.0, kernel='rbf', probability=True)
t0 = time()
clf.fit(features_train, labels_train)
print "training_time:", round(time()-t0, 3), "s"
t0 = time()
pred = clf.predict(features_test)
class = clf.predict_proba(features_test)
print sum(class)
print "prediction time:", round(time()-t0, 3), "s"
acc = accuracy_score(pred, labels_test)
print acc

我错过了什么？太棒了！

Answer 1

您可以创建混淆矩阵来理解您的预测

from sklearn.metrics import confusion_matrix
confusion_matrix(labels_test, pred)