相关性算法--皮尔逊算法实现(python,java)

python

1.数据处理工具

#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @File  : CasicMathBase.py
# @Author: lh
# @Date  : 2018/6/14
# @Desc  : 算法数据预处理工具
import numpy as tz_num

class Tz_math_base(object):
    # 定义数据预处理方法
    # 数列len
    # 不符合规则的数据
    # target 要处理的数列
    # 要处理的一个或多个数列(self param is touple)
    def tz_trans(self, standard, param):
        if len(standard) == 0:
            # 如果目标数列为空，任何计算都是没有意义的
            pass
        else:
            # 目标数列长度
            tar_length = len(standard)
            for list_i in param:
                if len(list_i) < tar_length and len(list_i) != 0:
                    len_i = tar_length - len(list_i)
                    while len_i > 0:
                        list_i.append(tz_num.average(list_i))
                        len_i -= 1

                elif len(list_i) > tar_length:
                    len_i = len(list_i) - tar_length
                    while len_i > 0:
                        list_i.pop()
                        len_i -= 1
                else:
                    pass

    # Z-score标准化方法
    def tz_score(self, list):
        if len(list) == 0:
            pass
        else:
            # 均值
            mean = tz_num.average(list)
            # 标准差
            std = tz_num.std(list)
            res = []
            for i in list:
                res = (i - mean) / std
                res.append(res)
            return res

    # 归1标准化
    def tz_ineins(self, list):
        if len(list) == 0:
            pass
        else:
            res = []
            tz_max = tz_num.max(list)
            tz_min = tz_num.min(list)
            tz_range = tz_max - tz_min
            if tz_range == 0:
                for i in list:
                    res.append(i -