归一化处理的三种类型

#以下定义的三个函数，分别用于不同类型的NumPy数组归一化处理
    def maxminnorm(self, array):
        """
        Numpy数组的归一化处理，把数据调到[0,1]
        :param array: 需要归一化的数组
        :return: 归一化后的数组
        """
        maxcols = array.max(axis=0)# 计算每列最大值
        mincols = array.min(axis=0)# 计算每列最小值
        data_rows = array.shape[0]# 获取数据集的行数
        data_cols = array.shape[1]# 获取数据集的列数
        t = np.empty((data_rows, data_cols))# 创建一个空的NumPy数组用于存储归一化后的数据
        for i in range(data_cols):# 遍历每一列
            t[:, i] = (array[:, i] - mincols[i]) / (maxcols[i] - mincols[i])
        return t

    def meannorm(self, array):
        """
        Numpy数组的归一化处理--Mean类型。把数据调到[-1,1]，平均值为0
        :param array: 需要归一化的数组
        :return: 归一化后的数组
        """
        maxcols = array.max(axis=0)
        mincols = array.min(axis=0)
        meancols = array.mean(axis=0)# 计算每列的平均数
        data_rows = array.shape[0]
        data_cols = array.shape[1]
        t = np.empty((data_rows, data_cols))
        for i in range(data_cols):
            t[:, i] = (array[:, i] - meancols[i]) / (maxcols[i] - mincols[i])
        return t

    def zeronorm(self, array):
        """
        Numpy数组的归一化处理--zero类型。把数据调到[-1,1]，平均值为0
        :param array: 需要归一化的数组
        :return: 归一化后的数组
        """
        std = array.std(axis=0)# 计算每列的平均差
        mean = array.mean(axis=0)
        data_rows = array.shape[0]
        data_cols = array.shape[1]
        t = np.empty((data_rows, data_cols))
        for i in range(data_cols):
            t[:, i] = (array[:, i] - mean[i]) / std[i]
        return t