"""
This is the module main file
"""
from __future__ import print_function
import numpy as np


def mean_std(p_array):
    """
    result = (p_array - np.mean(p_array)) / np.std(p_array) \n
    :rtype: numpy.array:
    :param numpy.array p_array: the array needed to be normalized
    """
    check_params(p_array)
    mean = np.mean(p_array)
    std = np.std(p_array)
    result = (p_array - mean) / std
    return result


def mean_std_back(p_array, y_predict):
    """
    数组归一化的反向求解\n
    y_predict = (p_array - np.mean(p_array)) / np.std(p_array) \n
    p_array = (y_predict * np.std(p_array)) + np.mean(p_array)
    :rtype: numpy.array:
    :param numpy.array|float y_predict: one of the data that has been normalized
    :param numpy.array p_array: original entire array
    """
    check_params(p_array)
    mean = np.mean(p_array)
    std = np.std(p_array)
    result = (y_predict * std) + mean
    return result


def max_interval_divided(p_array):
    """
    把数组中元素按照数组的最大间隔范围映射为 0 - 1 区间的 float \n
    :rtype: numpy.array
    :param numpy.array p_array: the array needed to be normalized
    公式：\n
    result = (p_array - arr_min) / (arr_max - arr_min)
    """
    check_params(p_array)
    arr_max = np.max(p_array)
    arr_min = np.min(p_array)
    result = (p_array - arr_min) / (arr_max - arr_min)
    return result


def max_interval_divided_back(p_array, y_predict):
    """
    数组归一化的反向求解\n
    数组间隔范围映射为 0 - 1，求给定的值的真实值 \n
    :rtype: float
    :param numpy.array|float y_predict: one of the data that has been normalized
    :param numpy.array p_array: original entire array
    公式如下：   \n
    y_predict = (result - arr_min) / (arr_max - arr_min)   \n
    result = (y_predict * (arr_max - arr_min)) + arr_min   \n
    """
    check_params(p_array)
    arr_max = np.max(p_array)
    arr_min = np.min(p_array)
    result = y_predict * (arr_max - arr_min) + arr_min
    return result


def check_params(params):
    """
    Check the type and content of the parameters \n
    :param numpy.array params: the given parameter
    """
    # check type
    if not isinstance(params, np.ndarray):
        print("Wrong parameter type, Needed Numpy array")
        exit()

    # check data type
    p_type = params.dtype
    isdigit = p_type in ['int32', 'int64', 'float32', 'float64']
    if not isdigit:
        print("Wrong Type of data，Needed int32,int64,float32,float64")
        exit()

    # Check whether the Nan is included
    if np.isnan(params).sum() > 0:
        print("Parameters include NAN, Please Check your data")
        exit()


if __name__ == '__main__':
    a = [[x, x + 1, x + 2, x + 3, x + 4] for x in range(10)]
    a = np.array(a)
    print(a)
    normalized = mean_std(a)
    latest_item = normalized[-1,-1]
    print(normalized)
    print(latest_item)
    restore = mean_std_back(a, latest_item)
    print("restore: ", restore)
    print("original: ", a[-1, -1])
    # normalized = max_interval_divided(p_array)
    # test = normalized[:3]
    # back = max_interval_divided_back(p_array, test)
    # back = max_interval_divided_back(p_array, 0.125)
    # print(p_array, normalized, test, back)