From reading the sklearn docs, I believe the RobustScaler is more immune to outliers that the StandardScaler. In that case, why not just use ...
確定! 回上一頁