创建均匀分布的RandomOverSampler条件

Question

我目前在一个基于ML的项目中工作，我的数据中有轻微的不平衡，需要over_sampling技术。特征(X_train)维度为(90664,190)，目标(Y_binary_train_trans)为(90664，)。但是，代码运行后仍然输出相同的、不均匀的目标分布。这是用于RandomOverSampler的代码，它也曾尝试使用smote；

counter= Counter(Y_binary_train_trans)
ros= RandomOverSampler(random_state=42)
X_train, Y_binary_train_trans = ros.fit_resample(X_train,Y_binary_train_trans)
counter = Counter(Y_binary_test_trans)

Answer 1

counter= Counter(Y_binary_train_trans)
ros= RandomOverSampler(random_state=42)
X_train, Y_binary_train_trans = ros.fit_resample(X_train,Y_binary_train_trans)
counter = Counter(Y_binary_test_trans)

对于这段代码，您的第二个计数器计算的是测试样本，而不是您实际更改的训练样本！

相反，它应该是：

counter= Counter(Y_binary_train_trans)
ros= RandomOverSampler(random_state=42)
X_train, Y_binary_train_trans = ros.fit_resample(X_train,Y_binary_train_trans)
counter = Counter(Y_binary_train_trans)