【Deep-ML系列】Feature Scaling Implementation（特征缩放：standardization & min-max）

最新推荐文章于 2024-09-14 19:18:15 发布

py明天会更好

最新推荐文章于 2024-09-14 19:18:15 发布

阅读量222

点赞数 3

分类专栏：深度学习/机器学习文章标签：特征缩放机器学习 MIn-Max 标准化

本文链接：https://blog.csdn.net/weixin_56029873/article/details/141054315

版权

深度学习/机器学习专栏收录该内容

14 篇文章 0 订阅

订阅专栏

题目链接：Deep-ML

import numpy as np
def feature_scaling(data: np.ndarray) -> (np.ndarray, np.ndarray):

    mean = np.mean(data, axis=0)
    std = np.std(data, axis=0)
    standardized_data = (data - mean) / std

    maxnum = np.max(data, axis=0)
    minnum = np.min(data, axis=0)
    normalized_data = (data - minnum) / (maxnum - minnum) * (maxnum - minnum) + minnum


    return np.round(standardized_data, 4).tolist(), np.round(normalized_data, 4).tolist()

if __name__ == '__main__':
    data = np.array([[1, 2], [3, 4], [5, 6]])
    print(feature_scaling(data))