Merge branch 'development'

.gitignore

 # Created by .ignore support plugin (hsz.mobi)
 .idea
+.idea/workspace.xml
 **/.ipynb_checkpoints
 .ipynb_checkpoints
 startup.py

J_feature/Untitled2.ipynb

+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.6.0"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

J_feature/V7模型分-豪举.ipynb

@@ -732,13 +732,13 @@
     {
      "ename": "NameError",
      "evalue": "name 'params' is not defined",
-     "output_type": "error",
      "traceback": [
       "---------------------------------------------------------------------------",
       "NameError                                 Traceback (most recent call last)",
       "<ipython-input-9-d29c09fe160d> in <module>()\n      6 test_result = []\n      7 \n----> 8 pbar = tqdm(range(len(params)),'测试接口调用')\n      9 for i in pbar:\n     10     semlock.acquire()\n",
       "NameError: name 'params' is not defined"
-     ]
+     ],
+     "output_type": "error"
     }
    ],
    "source": [

J_feature/紧急联系人运营商特征_test.ipynb

@@ -1470,7 +1470,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.0"
+   "version": "3.6.6"
   }
  },
  "nbformat": 4,

config/concat_sql.py

@@ -75,4 +75,8 @@ con_vccthree_vcc_talos = create_engine('mysql+pymysql://qa:qatest@172.17.5.8:311
 
 con_app_fraud = create_engine('mysql+pymysql://peng.xiong:RMxyUTYgm73TkkQb@fengkong-tidb.quantgroups.com:4010/rc_user_behavior?charset=utf8',echo=False)
 
-con_app_applied_type = create_engine('mysql+pymysql://b_events_w:q3ReZoL9x5asCxcx@10.17.115.6:4010/dp_analysis?charset=utf8',echo=False)
\ No newline at end of file
+con_app_applied_type = create_engine('mysql+pymysql://b_events_w:q3ReZoL9x5asCxcx@10.17.115.6:4010/dp_analysis?charset=utf8',echo=False)
+
+"""催收二期"""
+con_collection = create_engine('mysql+pymysql://rc-meta-user-pre:YqnqVfIwZFMlkGya@10.17.115.6:4010/rc_meta_data_pre?charset=utf8',echo=False)
+

features/calc_loanPostFeatures/primary_v2.py

+# -*- coding:utf-8 -*-
+from features.Base_Features import BaseFeatures as BF
+from service.Sql_server import *
+import datetime
+from service.Sql_collection import *
+from dateutil.relativedelta import relativedelta
+from service.Sql_complexOrder import *
+
+class __INIT__(BF):
+    def __init__(self):
+
+        time_now = datetime.datetime.strptime(datetime.datetime.strftime(datetime.datetime.now(),'%Y-%m-%d %H:%M:%S'),'%Y-%m-%d %H:%M:%S')
+        BF._apply_at = time_now
+        BF._apply_df = pd.DataFrame()
+        df_get_order = get_order_id(self.loanId)
+        BF.order_id = df_get_order['order_no'].values[0]
+        BF._df= self.get_init(BF.user_id,BF.loanId,BF.termNo)
+
+
+    def get_init(self,user_id,loanId = '',termNo = ''):
+        df_empty = pd.DataFrame()
+        if loanId == '' or termNo == '':
+            return None
+        df = getSql_loanAll(str(tuple([self.user_id])).replace(',)', ')'))
+        if df.empty == False:
+            deadline_current = df.loc[(df['loan_id'] == self.loanId) & (df['term_no'] == self.termNo)]['deadline']
+            if len(deadline_current) != 0:
+                deadline_current_time = deadline_current.values[0]
+                df = get_Loan_rollBack(df,deadline_current_time,passdueDay_Negative=True)
+                return df
+            else:
+                return df_empty
+        else:
+            return df_empty
+
+
+class 用户申请时改订单的申请类型状态值():
+    """申请类型： 用户申请时改订单的申请类型状态值   current_loan_post_user_loan_type"""
+    def feature(self):
+        _用户申请时改订单的申请类型状态值 = -9999999
+        df_temp_v4 = get_Application_type(self.order_id,'user_loan_type_v4')
+        df_temp_v5 = get_Application_type(self.order_id, 'user_loan_type_v5')
+        if df_temp_v4.empty ==False:
+            df_temp_v4 = df_temp_v4.sort_values(['value'])
+            applied_from  = df_temp_v4['value'].values[0]
+            _用户申请时改订单的申请类型状态值 = applied_from
+        if df_temp_v5.empty ==False:
+            df_temp_v5 = df_temp_v5.sort_values(['value'])
+            applied_from  = df_temp_v5['value'].values[0]
+            _用户申请时改订单的申请类型状态值 = applied_from
+        self._result["用户申请时改订单的申请类型状态值"]["value"] = _用户申请时改订单的申请类型状态值
+        return self._result
+
+
+class 历史最大逾期天数v2():
+    """历史最大逾期天数  current_loan_post_term_delq_days"""
+    def feature(self):
+        _历史最大逾期天数v2 = -9999999
+        if self._df.empty == False:
+            self._df['deadline'] = self._df['deadline'].apply(lambda x: str(x))
+            df_histroy = self._df.loc[(self._df['loan_id'] < self.loanId)&(self._df['progress']== 16)]
+            df_current = self._df.loc[(self._df['loan_id'] == self.loanId) & (self._df['term_no']< self.termNo)]
+            df_concat = pd.concat([df_histroy,df_current],axis=0)
+            deadline_current = self._df.loc[(self._df['loan_id'] == self.loanId) & (self._df['term_no'] == self.termNo)]['deadline'].values[0]
+            df_concat = df_concat.loc[(df_concat['deadline'] < deadline_current)]
+            if df_concat.empty ==False:
+                max_delq_days = df_concat['passdue_day'].max()
+                _历史最大逾期天数v2 = max_delq_days
+        self._result["历史最大逾期天数v2"]["value"] = _历史最大逾期天数v2
+        return self._result
+
+class 当前订单月还款额():
+    """当前订单月还款额   current_loan_post_loan_info#monthlyRepayment"""
+    def feature(self):
+        _当前订单月还款额 = -9999999
+        df_temp = get_order_monthly_payment(self.loanId)
+        if df_temp.empty ==False:
+            monthly_payment = df_temp['monthly_repayment'].sum()
+            _当前订单月还款额 = monthly_payment
+        self._result["当前订单月还款额"]["value"] = _当前订单月还款额
+        return self._result
+
+
+class 到期期数非本期中提前还款的占比():
+    """到期期数非本期中提前还款的占比  current_loan_post_advance_repay_rate"""
+    def feature(self):
+        _到期期数非本期中提前还款的占比 = -9999999
+        if self._df.empty == False:
+            self._df['deadline'] = self._df['deadline'].apply(lambda x: str(x))
+            df_histroy = self._df.loc[(self._df['loan_id'] < self.loanId)&(self._df['progress']== 16)]
+            df_current = self._df.loc[(self._df['loan_id'] == self.loanId) & (self._df['term_no']< self.termNo)]
+            df_concat = pd.concat([df_histroy,df_current],axis=0)
+            deadline_current = self._df.loc[(self._df['loan_id'] == self.loanId) & (self._df['term_no'] == self.termNo)]['deadline'].values[0]
+            df_concat = df_concat.loc[(df_concat['deadline'] < deadline_current) ]
+            if df_concat.empty ==False:
+                df_temp_first = df_concat[df_concat['passdue_day'] < 0]
+                monthly_payment = len(df_temp_first) / len(df_concat)
+                _到期期数非本期中提前还款的占比 = round(monthly_payment,6)
+        self._result["到期期数非本期中提前还款的占比"]["value"] = _到期期数非本期中提前还款的占比
+        return self._result
+
+
+
+
+

requirements.txt

-requests=2.18.4
-tornado=5.0.1
+requests==2.18.4
+tornado==5.0.1
 logging
 redis
-PyMysql=0.8.0
-pandas=0.22.0
+PyMysql==0.8.0
+pandas==0.22.0
 numpy=1.14.2
-tornado=5.0.1
+tornado==5.0.1
 xlrd=1.1.0
-scrapy=1.6.0
-SQLAlchemy=1.2.5
+scrapy==1.6.0
+SQLAlchemy==1.2.5
 jupyter
 paramiko
 cpca
-lxml=3.8.0
\ No newline at end of file
+lxml==3.8.0
\ No newline at end of file

service/Sql_collection.py

+import cpca
+import pandas as pd
+from config.concat_sql import *
+import re
+import requests
+
+
+# 用户申请表
+def get_sql_apply_quota_record(user_id,apply_time):
+    """ 用户申请放款历史 """
+    sql_online_apply_quota_record = """
+    select * from apply_quota_record where user_id = {0} and apply_time > '{1}';
+    """
+    _df = pd.read_sql(sql_online_apply_quota_record.format(user_id,apply_time),con=con_tuomin_xyqb)
+    return _df
+"""通过loan_id 获取 order_id"""
+
+def get_order_id(loan_id):
+    sql_orderid = """select e.order_no,apply_time from apply_quota_record a JOIN loan_account_ext e on
+     a.order_no = e.order_no where e.loan_id ={0};""".format(loan_id)
+    _df = pd.read_sql(sql=sql_orderid,con=con_tuomin_xyqb)
+    return _df
+
+def get_Application_type(order_id,code):
+    Sql = """select code,name,value from calc_batch_info b JOIN calc_unit_record r on b.batch_uuid = r.batch_uuid
+     where order_id = '{0}' and code ='{1}' order by r.id desc """.format(order_id,code)
+    _df = pd.read_sql(sql=Sql,con=con_collection)
+    return _df
+
+
+def get_order_monthly_payment(loan_id):
+    Sql = """SELECT * FROM loan_application_manifest_history WHERE loan_application_history_id ={0};""".format(loan_id)
+    _df = pd.read_sql(sql=Sql,con=con_tuomin_xyqb)
+    return _df
\ No newline at end of file

service/Sql_contactsOperator.py

@@ -802,7 +802,7 @@ def get_Loan_rollBack(df_loan,apply_time):
     df_loan['observationTime'] = apply_time
     df_loan = df_loan.loc[(df_loan['apply_time'] < df_loan['observationTime']) & (df_loan['loan_paid_at'] <= df_loan['observationTime'] )]
     if df_loan.empty == False:
-        df_loan['repaid_at'] = df_loan['repaid_at'].apply(lambda x : datetime.datetime.strptime(datetime.datetime.strftime(x,'%Y-%m-%d %H:%M:%S'),'%Y-%m-%d %H:%M:%S'))
+        df_loan['repaid_at'] = df_loan['repaid_at'].apply(lambda x : datetime.datetime.strptime(datetime.datetime.strftime(x,'%Y-%m-%d 00:00:00'),'%Y-%m-%d 00:00:00'))
         df_loan['deadline'] = df_loan['deadline'].apply(lambda x : datetime.datetime.strptime(datetime.datetime.strftime(x,'%Y-%m-%d 00:00:00'),'%Y-%m-%d 00:00:00'))
         df_loan['deadline'] = df_loan['deadline'].apply(lambda x : datetime.datetime.strptime(datetime.datetime.strftime(x,'%Y-%m-%d 00:00:00'),'%Y-%m-%d 00:00:00'))
         df_loan['apply_time'] = df_loan['apply_time'].apply(lambda x : datetime.datetime.strptime(datetime.datetime.strftime(x,'%Y-%m-%d 00:00:00'),'%Y-%m-%d 00:00:00'))
@@ -955,3 +955,48 @@ def get_billDay_list(ApplyTime, IntervalDay):
                 month_list.append(str(start_.year) + str(start_.month))
             start_ = start_ + relativedelta(months=1)
     return month_list
+
+def get_phone_black(phone,first_time = False,end_time = False,type_tuple = False,else_type =False):
+    """获取手机号黑名单"""
+
+    if first_time:
+        _sql = (""" select id,major_type majorType, type ,created_at createdAt from black_list_new where phone_no in ('{0}') and status = 1 and created_at >= '{1}' and  created_at < '{2}' and type in """ + type_tuple).format(phone,first_time,end_time)
+    if else_type:
+        _sql = (""" select id,major_type majorType, type ,created_at createdAt from black_list_new where phone_no  in ('{0}') and status = 1 and created_at >= '{1}' and  created_at < '{2}' and type not in """ + type_tuple).format(phone,first_time,end_time)
+    if first_time == False:
+        _sql = (""" select id,major_type majorType, type ,created_at createdAt from black_list_new where phone_no in ('{0}') and created_at < '{1}'  and status = 1  and type in """ + type_tuple).format(phone,end_time)
+    if (else_type ==True) and (first_time == False):
+        _sql = (""" select id,major_type majorType, type ,created_at createdAt from black_list_new where phone_no in ('{0}') and created_at < '{1}'  and status = 1  and type not in """ + type_tuple).format(phone,end_time)
+    _df = pd.read_sql(_sql,con_blacklist3_read)
+    return _df
+
+def get_emergency_contacts(phone,first_time = False,end_time = False,type_tuple = False,else_type =False):
+    """获取紧急联系人"""
+    if first_time:
+        _sql = (""" select id,major_type majorType, type ,created_at createdAt from black_list_new where phone_no in {0} and status = 1 and created_at >= '{1}' and  created_at < '{2}' and type in """ + type_tuple).format(phone, first_time, end_time)
+    if else_type:
+        _sql = (""" select id,major_type majorType, type ,created_at createdAt from black_list_new where phone_no in {0} and status = 1 and created_at >= '{1}' and  created_at < '{2}' and type not in """ + type_tuple).format(phone,first_time,end_time)
+    if first_time == False:
+        _sql =( """select id,major_type majorType, type ,created_at createdAt from black_list_new where phone_no in {0}  and created_at < '{1}' and status = 1 and type in """ + type_tuple).format(phone,end_time)
+    if (else_type ==True) and (first_time == False):
+        _sql = (""" select id,major_type majorType, type ,created_at createdAt from black_list_new where phone_no in {0} and created_at < '{1}'  and status = 1  and type not in """ + type_tuple).format(phone,end_time)
+
+    _df = pd.read_sql(_sql,con_blacklist3_read)
+    return _df
+
+
+
+def get_id_no_black(phone,first_time = False,end_time = False,type_tuple = False,else_type =False):
+    """获取身份证黑名单"""
+
+    if first_time:
+        _sql = (""" select id,major_type majorType, type ,created_at createdAt from black_list_new where id_no in ('{0}') and status = 1 and created_at >= '{1}' and  created_at < '{2}' and type in """ + type_tuple).format(phone,first_time,end_time)
+    if else_type:
+        _sql = (""" select id,major_type majorType, type ,created_at createdAt from black_list_new where id_no in ('{0}') and status = 1 and created_at >= '{1}' and  created_at < '{2}' and type not in """ + type_tuple).format(phone,first_time,end_time)
+    if first_time == False:
+        _sql = (""" select id,major_type majorType, type ,created_at createdAt from black_list_new where id_no in ('{0}')  and created_at < '{1}' and status = 1  and type in """ + type_tuple).format(phone,end_time)
+    if (else_type == True) and (first_time == False):
+        _sql = (""" select id,major_type majorType, type ,created_at createdAt from black_list_new where id_no in ('{0}') and created_at < '{1}'  and status = 1  and type not in """ + type_tuple).format(phone,end_time)
+
+    _df = pd.read_sql(_sql,con_blacklist3_read)
+    return _df
\ No newline at end of file

service/Sql_vcc.py

@@ -591,23 +591,16 @@ def id_user_credited(validity_date,user_id):
 
 def day_incoming_parts(Same_day_first,Same_day_next):
     """当日客户数"""
-    print(Same_day_first,Same_day_next)
     _Sql = """select count(distinct user_id) count, apply_status applyStatus from risk_record
      where created_at >= "{0}" and created_at < "{1}" group by apply_status;""".format(Same_day_first,Same_day_next)
-    print('----2----')
-    print(_Sql)
     _df = pd.read_sql(_Sql,con_vccthree_vcc_talos)
-    print('----3----')
     return _df
 
 def day_amount_credit(Same_day_first,Same_day_next):
     """当日授信总额度"""
-    print(Same_day_first,Same_day_next)
     _Sql = """select sum(limit_amount) totalCount from acs_limit_record where limit_record_type = 'New'
      and limit_amount > 0 and create_time >= "{0}"and create_time < "{1}";""".format(Same_day_first,Same_day_next)
-    print('----2----')
     _df = pd.read_sql(_Sql,con_vccthree_acsdb)
-    print('----3----')
     return _df
 
 def sale_phone_white(phone_no):

startup.py

@@ -37,7 +37,7 @@ def apps():
         (URL.url_new_audit, NewAudit_Handler.New_Audit),
         (URL.url_calc_features, Calc_Features_Handler.CalcFeatures),
         (URL.url_complex_order, ComplexOrder_Handler.ComplexOrder),
-        (URL.url_loanPostFeatures, LoanPostFeatures_Handler.loanPostFeatures),
+        (URL.url_loanPostFeatures, LoanPostFeatures_Handler.loanPostFeatures)
     ])
 
 

utils/tools.py

@@ -3,7 +3,7 @@ import json
 import pandas as pd
 import sys,datetime,time
 import threading
-import paramiko
+# import paramiko
 import inspect
 import requests