假设我有一个具有 2 个独立特征和 6 个从属特征的数据集,如下所示,其中AA and BB
独立和A,B,C,D,E,F
因变量。
AA BB A B C D E F
teacher No 1 0 0 1 0 0
student Yes 1 1 0 1 0 0
officer Yes 0 0 1 0 0 0
engineer Maybe 1 1 1 0 0 0
student Sure 0 1 1 0 0 1
lawyer Maybe 1 1 0 0 1 0
student Yes 1 0 0 1 1 0
teacher No 1 1 1 0 0 0
doctor No 0 1 1 0 0 0
engineer Maybe 1 0 0 1 1 0
如果假设我想通过上采样或下采样或过采样来平衡我的数据,我该怎么做?普通的 SMOTE 和 sklearn 技术不会采用多个目标并引发大量错误。有人可以帮我解决这个问题。