在scikit学习中使用一个类svm进行二元分类

Question

我正在研究不平衡数据集的二元分类。数据集包含 777 个少数类和 2223 个多数类。我构建了一个单类 SVM 模型，其中只有少数标记的记录。BUt 当我尝试在构建的模型上进行预测时，我得到的预测值均为 -1，因此准确性为 0.我已经缩放了我的特征。这是我的实现

ml_file_df = pd.read_csv('/data/jayashree/3000_ML_features.csv')
minority_df = ml_file_df[ml_file_df['RESULT'] == 0]
array = minority_df.values
features = array[:, 0:60630]
labels = array[:, 60630]
scaler = MinMaxScaler()
scaled_features = scaler.fit_transform(features)
features_train, features_test, labels_train, labels_test = train_test_split(
    scaled_features, labels, test_size=0.3, random_state=10)
gamma_values = [0.001, 0.005, 0.01, 0.05, 0.1, 0.5]
nu_values = [0.1, 0.3, 0.5, 0.7]
for j in nu_values:
    for i in gamma_values:
        clf = svm.OneClassSVM(nu=j, kernel='rbf', gamma=i)
        clf.fit(features_train, labels_train)
        pred = clf.predict(features_test)
        print(i, classification_report(labels_test, pred))

对于所有情况，我都会得到这样的预测

[-1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
 -1 -1 -1 -1  1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1  1 -1 -1 -1 -1 -1 -1 -1 -1 -1
 -1 -1 -1 -1 -1  1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
 -1 -1  1 -1 -1 -1 -1 -1 -1  1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1  1  1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1  1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
 -1 -1 -1  1 -1 -1 -1 -1  1]

我哪里出错了？

我已经解决了错误。一个类 svm 将预测 1 或 -1。我使用少数类为 1，将多数类用作 -1。这解决了我的问题。

Answer 1

那是因为你有0和1标签和一类SVM显示-1和1，你交换-1就足够了0。