#!/usr/bin/env python
# -*- coding:utf-8 -*- 
# @Time    : 2018/4/5 8:09
# @Author  : liujiantao
# @Site    : 特征集成
# @File    : feature_integrate.py
# @Software: PyCharm
import time

from data_helper import DataHelper
from ljt_feature_etl02 import LjtFeatureEtl02
from threading_util import ThreadingUtil
import pandas as pd
import numpy as np
from config import *

class FeatureIntegrate02(object):
    """
    特征汇总
    """
    def __init__(self):
        self.d_h = DataHelper()

    def get_all_features(self,path=path_train01):
        start = time.time()
        data = self.d_h.get_data(path)
        ft_Liyang = LjtFeatureEtl02(self.d_h, data)
        user_Y_list = self.d_h.get_user_Y_list(data)
        min_Y = data[data['Y'] != 0]['Y'].min()
        max_Y = data[data['Y'] != 0]['Y'].max()
        all_features_list = []
        pre_features_list = []
        act_class_Y = []
        act_Y = []
        all_Y = []
        pre_Y = []
        mt = ThreadingUtil()
        g_func_list = []
        # 用户行程占比
        g_func_list.append({"func": ft_Liyang.get_features01, "args": (data,)})
        g_func_list.append({"func": ft_Liyang.get_features02, "args": (data,)})
        g_func_list.append({"func": ft_Liyang.get_features03, "args": (data,)})
        g_func_list.append({"func": ft_Liyang.get_features04, "args": (data,)})
        g_func_list.append({"func": ft_Liyang.get_features05, "args": (data,)})
        g_func_list.append({"func": ft_Liyang.get_features06, "args": (data,)})

        mt.set_thread_func_list(g_func_list)
        mt.start()
        m = 0
        n = 0
        # for j in range(mt.data_list.__len__()):
        #     all_features_list.extend(mt.data_list[j])
        # res = np.array(mt.data_list)
        all_features_list = [[row[col] for row in mt.data_list] for col in range(len(mt.data_list[0]))]
        # pre_features_list = [all_features_list[i] for i in range(len(user_Y_list)) if user_Y_list[i]>0]
        # pre_Y = [user_Y_list[i] for i in range(len(user_Y_list)) if user_Y_list[i]>0]
        # act_class_Y = list(map(lambda x :1 if x>0 else 0,user_Y_list))
        # df_features = pd.DataFrame(mt.data_list,columns=predictors1)
        # df_features = df_features[predictors]
        self.d_h.print_str += " get_train_features cost time: " + str(time.time() - start)+" "
        return all_features_list, user_Y_list, pre_features_list, pre_Y, act_class_Y, min_Y, max_Y

    def get_test_features02(self, path=path_test01):
        """

        :param path:
        :return:
        """
        start = time.time()
        data = self.d_h.get_test_data(path)
        userid_list = self.d_h.get_userlist(data)

        ft_Liyang = LjtFeatureEtl02(self.d_h, data)
        mt = ThreadingUtil()
        g_func_list = []
        g_func_list.append({"func": ft_Liyang.get_features01, "args": (data,)})
        g_func_list.append({"func": ft_Liyang.get_features02, "args": (data,)})
        g_func_list.append({"func": ft_Liyang.get_features03, "args": (data,)})
        g_func_list.append({"func": ft_Liyang.get_features04, "args": (data,)})
        g_func_list.append({"func": ft_Liyang.get_features05, "args": (data,)})
        g_func_list.append({"func": ft_Liyang.get_features06, "args": (data,)})
        mt.set_thread_func_list(g_func_list)
        mt.start()

        test_features = [[row[col] for row in mt.data_list] for col in range(len(mt.data_list[0]))]
        self.d_h.print_str += " get_test_features cost time: " + str(time.time() - start)+" "
        return userid_list, test_features