scoreLog_S23_V_1_0.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#!/usr/bin/env python\n",
    "# -*- coding: utf-8 -*-\n",
    "\n",
    "import os\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# 切换绝对路径.\n",
    "%pwd\n",
    "os.chdir(r\"D:/ai/project/multiappl/src/验证代码\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# 读取各客群的训练测试集与验证集\n",
    "Train_Test_Cash_Off = pd.read_csv(\"Train_Test_Cash_Off.csv\")\n",
    "Train_Test_Cons_On = pd.read_csv(\"Train_Test_Cons_On.csv\")\n",
    "Train_Test_Revoloan = pd.read_csv(\"Train_Test_Revoloan.csv\")\n",
    "Validation_Cash_Off = pd.read_csv(\"Validation_Cash_Off.csv\")\n",
    "Validation_Cons_On = pd.read_csv(\"Validation_Cons_On.csv\")\n",
    "Validation_Revoloan = pd.read_csv(\"Validation_Revoloan.csv\")\n",
    "\n",
    "Train_Test_Cash_Off.columns = [\"numb\", \"probability\", \"y\", \"cus_num\"]\n",
    "Train_Test_Cons_On.columns = [\"numb\", \"probability\", \"y\", \"cus_num\"]\n",
    "Train_Test_Revoloan.columns = [\"numb\", \"probability\", \"y\", \"cus_num\"]\n",
    "Validation_Cash_Off.columns = [\"numb\", \"probability\", \"y\", \"cus_num\"]\n",
    "Validation_Cons_On.columns = [\"numb\", \"probability\", \"y\", \"cus_num\"]\n",
    "Validation_Revoloan.columns = [\"numb\", \"probability\", \"y\", \"cus_num\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# 去除输出为-1的样本\n",
    "Train_Test_Cash_Off = Train_Test_Cash_Off[Train_Test_Cash_Off[\"probability\"] != -1]\n",
    "Train_Test_Cons_On = Train_Test_Cons_On[Train_Test_Cons_On[\"probability\"] != -1]\n",
    "Train_Test_Revoloan = Train_Test_Revoloan[Train_Test_Revoloan[\"probability\"] != -1]\n",
    "Validation_Cash_Off = Validation_Cash_Off[Validation_Cash_Off[\"probability\"] != -1]\n",
    "Validation_Cons_On = Validation_Cons_On[Validation_Cons_On[\"probability\"] != -1]\n",
    "Validation_Revoloan = Validation_Revoloan[Validation_Revoloan[\"probability\"] != -1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# 对概率值进行max-min归一化.\n",
    "max_prob = Train_Test_Cash_Off[\"probability\"].max()\n",
    "min_prob = Train_Test_Cash_Off[\"probability\"].min()\n",
    "Train_Test_Cash_Off[\"probability\"] = Train_Test_Cash_Off[\"probability\"].map(lambda x: (x - min_prob)/(max_prob - min_prob))\n",
    "\n",
    "max_prob = Train_Test_Cons_On[\"probability\"].max()\n",
    "min_prob = Train_Test_Cons_On[\"probability\"].min()\n",
    "Train_Test_Cons_On[\"probability\"] = Train_Test_Cons_On[\"probability\"].map(lambda x: (x - min_prob)/(max_prob - min_prob))\n",
    "\n",
    "max_prob = Train_Test_Revoloan[\"probability\"].max()\n",
    "min_prob = Train_Test_Revoloan[\"probability\"].min()\n",
    "Train_Test_Revoloan[\"probability\"] = Train_Test_Revoloan[\"probability\"].map(lambda x: (x - min_prob)/(max_prob - min_prob))\n",
    "\n",
    "max_prob = Validation_Cash_Off[\"probability\"].max()\n",
    "min_prob = Validation_Cash_Off[\"probability\"].min()\n",
    "Validation_Cash_Off[\"probability\"] = Validation_Cash_Off[\"probability\"].map(lambda x: (x - min_prob)/(max_prob - min_prob))\n",
    "\n",
    "max_prob = Validation_Cons_On[\"probability\"].max()\n",
    "min_prob = Validation_Cons_On[\"probability\"].min()\n",
    "Validation_Cons_On[\"probability\"] = Validation_Cons_On[\"probability\"].map(lambda x: (x - min_prob)/(max_prob - min_prob))\n",
    "\n",
    "max_prob = Validation_Revoloan[\"probability\"].max()\n",
    "min_prob = Validation_Revoloan[\"probability\"].min()\n",
    "Validation_Revoloan[\"probability\"] = Validation_Revoloan[\"probability\"].map(lambda x: (x - min_prob)/(max_prob - min_prob))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEKCAYAAAD9xUlFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x99e0a90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0.21139125887949395"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Cash Off验证集的KS表现(注：KS函数与PSI函数均在代码块末尾).\n",
    "ks_stats(Validation_Cash_Off)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEKCAYAAAD9xUlFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3Xd4FFUXwOHfTSGdBEIIHULvNTRpoQp8igIiVYpKL9IR\nUDrSkSpFKRYUGyAooihEQIr03nsNIRBCervfHxNCaMkSskk2Oe/z8OzM7OzuGQg5O7ecq7TWCCGE\nEABWaR2AEEKI9EOSghBCiHiSFIQQQsSTpCCEECKeJAUhhBDxJCkIIYSIJ0lBCCFEPEkKQggh4klS\nEEIIEc8mrQN4UTly5NCFChVK1mtDQkJwcnJK2YDSObnmzEGuOXN4mWvev3//Ha21R1LnWVxSKFSo\nEPv27UvWa319ffHx8UnZgNI5uebMQa45c3iZa1ZKXTblPGk+EkIIEU+SghBCiHiSFIQQQsSzuD6F\nZ4mKiuLatWuEh4cnep6rqysnT55MpajSB0u8Znt7e/Lly4etrW1ahyJEppMhksK1a9dwcXGhUKFC\nKKWee96DBw9wcXFJxcjSnqVds9aagIAArl27hpeXV1qHI0SmY7bmI6XUcqXUbaXUsec8r5RS85RS\n55RSR5RSlZP7WeHh4bi7uyeaEIRlUErh7u6e5F2fEMI8zNmnsBJomsjzzYBicX96AIte5sMkIWQc\n8m8pRNoxKSkopRyUUiVe5I211tuAu4mc8gbwlTbsBtyUUrlf5DOEECIzCAgKpcH4sfy864bZPyvJ\nPgWl1OvATCAL4KWUqghM0Fq3eMnPzgtcTbB/Le7YzWfE0APjbgJPT098fX0fe97V1ZUHDx4k+YEx\nMTEmnZccfn5+jBgxggMHDuDm5oaHhwdTp06lWLFiKfo5bm5ulClTBq011tbWzJw5k+rVqz/3/Cev\n+ZNPPsHZ2ZkBAwY8dt4nn3zCl19+SY4cOYiMjGT48OG0adMm0Vh+/fVXihYtSsmSJQGYNGkStWrV\non79+i9xhYbw8PCn/p1NFRwcnOzXWiq55ozr+P3jDD80kVD8OHW2Cy235MLKyoyNPFrrRP8A+wFX\n4GCCY0eTel3ceYWAY8957legdoL9vwHvpN6zSpUq+kknTpx46tizBAUFmXTei4qNjdU1atTQixYt\nij926NAhvW3bthT/LCcnp/jtTZs26bp16yZ6/pPXPHbsWD1jxoynzkt4/MyZM9rFxUVHRkYm+t5d\nunTRP/74o6mhvxBT/02fZevWrSkXiIWQa854rgRe0Q3mddSMRTPEXVNmsV6/5vdkvx+wT5vwe9uU\ndBOltb7/ZC5JgXx0HcifYD9f3DGLs3XrVmxtbenVq1f8sQoVKlCnTh18fX157bXX4o/369ePlStX\nAkbJjpEjR1KxYkW8vb05cOAAr776KkWKFGHx4sVJfm5QUBDZsmUDjG9NDRs2pHLlypQrV45ffvkl\n/rzJkydTvHhxateuzenTp5N832LFiuHo6Mi9e/cA+Pzzz6latSoVKlSgdevWhIaGsnPnTtavX8+w\nYcOoWLEi58+fp2vXrvz0008A/P3331SqVIly5crx7rvvEhERkfRfpBCCwPBApu2YRoE5Bdjivwr2\ntaPolo10Ld4MFzc7s3++KUNSjyulOgDWSqliwABgZwp89nqgn1JqNVAduK+1fqrp6EUNHAiHDj37\nuZgYB6ytX/w9K1aEOXOe//yxY8eoUqXKi78xUKBAAQ4dOsSgQYPo2rUr//77L+Hh4ZQtW/axJPNQ\nWFgYFStWJDw8nJs3b7JlyxbAGNu/du1asmbNyp07d6hRowYtWrTg4MGDrF69mkOHDhEdHU3lypWT\njPXAgQMUK1aMnDlzAtCqVSu6d+8OwEcffcSyZcvo378/LVq04LXXXuOtt9567PXh4eF07dqVv//+\nm+LFi9O5c2cWLVrEwIEDk/V3JERmEBMbw6J9ixj15ygexDzA/pY34b9OYnCzosxYVxArGxt8fS+Y\nPQ5TkkJ/YDQQAXwH/AFMTOpFSqnvAB8gh1LqGjAWsAXQWi8GNgLNgXNAKNDtxcN/ObGxYM6mOVO0\naGF0zZQrV47g4GBcXFxwcXHBzs6OwMBA3NzcHjvfwcGBQ3FZb9euXXTu3Jljx46htWbUqFFs27YN\nKysrrl+/jp+fHzt37qRly5Y4Ojo+9nnP8umnn7JixQrOnDnDhg0b4o8fO3aMjz76iMDAQIKDg3n1\n1VcTvabTp0/j5eVF8eLFAejSpQsLFy6UpCDEc1wOvEynNZ3YcXUHZbJ4UT9mCCt/7kceKztmzbcF\nm9SbUpbkJ2mtQzGSwugXeWOtdfskntdA3xd5T1Mk9o3+wYMws0zkKlOmTHyzyZNsbGyIjY2N339y\n/L2dnXE7aGVlFb/9cD86OjrRz61ZsyZ37tzB39+fjRs34u/vz/79+7G1taVQoUIvPNZ/0KBBDB06\nlPXr1/Pee+9x/vx57O3t6dq1K+vWraNChQqsXLkyU3TuCZEatNasOrqK99a+RySRjM7fmRxXhzJo\nXAkgC8P73AK7XKka03O/JyulNiil1j/vT2oGmd41aNCAiIgIli5dGn/syJEjbN++nYIFC3LixAki\nIiIIDAzk77//TrHPPXXqFDExMbi7u3P//n1y5syJra0tW7du5fJlo0purVq1WLduHWFhYTx48OCx\nO4DnadGiBd7e3nz55ZeAMSs6d+7cREVFsWrVqvjzXFxcnjmaq0SJEly6dIlz584B8PXXX1OvXr2U\nuGQhMoxLgZdosLIB76x9h6L2efij3GxGlO3PtIWlcLS35tBmf0ZOyZrqcSV2pzAz1aKwcEop1q5d\ny8CBA5k2bRr29vYUKlSIOXPmkD9/ft5++23Kli2Ll5cXlSpVeqnPetinAMa3jC+//BJra2s6duzI\n66+/Trly5fD29o4fJlqxYkXatm1LhQoVyJkzJ1WrVjXpc8aMGUOHDh3o3r07EydOpHr16nh4eFC9\nevX4RNCuXTu6d+/OvHnzHrtTsre3Z8WKFbRp04bo6GiqVq36zP4RITKjWB3LykMr6bG+BzHEMCDP\nW4ws8h7unvmZvDgXt/xtqFg0mAo+2VK12eghZbTiJHGSUlmAkhijjk5rrSPNHdjzeHt76ycX2Tl5\n8iSlSpVK8rWWVgcoJVjqNZv6b/ossvhK5mCJ1/z72d8Z8NsAzt0/R3nnYswvPICIgIYMnuFFTLTi\n5AWjCfn2cX88Sj+9SNpLLrKzX2vtndR5pkxe+x+wGDgPKIwJbD211r8nKzIhhMhkQqNCGfrHUBbt\nX4SnysHw3J3p4vY+i9aVYsFXOQBoUPUeuXOE0qt9CB6l86VZrKbcm8wC6mutzwEopYoAvwGSFIQQ\nIgkHbh6gzeo2XAi6QBvPBozLNZgyLRsyHfv4c2aPvMGgCTnjmouypV2wmJYUHjxMCHEuAOapFSGE\nEBnEtaBrjNsyjmWHl+GGC4uLDKVj0dZM/6YoYI+PdxD1XonCp04MPm/lSetw45mSFPYppTYCP2D0\nKbQB9iqlWgFordeYMT4hhLAoWuv4juRoomnhXovJXr0olr88ysqe2SuyY6Vi2fgrOHi6p3W4TzEl\nKdgDfsDDMYX+gAPwOkaSkKQghBAYTUVDNg3B94ovFRyLMrVwb+oUqMP9cDf6TcqJ738OhIRa8eor\n93DI6Zb0G6YBUyavpfpMYyGEsCTBkcEM2TSEpQeXYos1/fO0ZmKJ/py8mY/mfXOzba9RUcDJLooy\nXg/4+dsYSKfrhiRZ5EEp5aWUmq2UWiOT157v1q1btGvXjiJFilClShWaN2/OmTNnUvxz/Pz86NCh\nA4ULF6ZKlSrUrFmTtWvXpsh7+/j48ORw34fHS5QoQYUKFahatWp8mY3EzJkzh9DQ0Pj95s2bExgY\nmCJxCpGebDi9geKfFmfpwaW0y9mY3VWXMbXKKFxz5aZl/4Js2+tItZL3ebuhPw9uh3PsjANOBXOk\nddjPZUrz0TpgGbABiE3i3ExJa03Lli3p0qULq1evBuDw4cP4+fnF1/9Jqc9588036dKlC99++y0A\nly9fZv168+foVatW4e3tzYoVKxg2bBibN29O9Pw5c+bQqVOn+JpLGzduNHuMQqSmC/cuMGTTENad\nWUdO5c4XhYfTrkRr7F1c0Nqa4dNzcMvfhmqlHrDniBPYuKZ1yCYxpRxcuNZ6ntZ6q9b6n4d/zB6Z\nBUmt0tlbtmwhS5Ysj31OwYIF6d+/P2DUVerWrRvlypWjUqVKbN26NdHjYWFhtGvXjlKlStGyZUvC\nwsKSvNaaNWty/fqjCue9e/fG29ubMmXKMHbsWADmzZvHjRs3qF+/fvyCO4UKFeLOnTsAzJ49m7Jl\ny1K2bFnmJFasSoh0KDImkrm751JkXhHWnVlH7zwtOVXnezpX7oyTmxuHTzlSv3M+ZizLDsA3y8PT\nZGZycpkS6Vyl1FjgT4xKqQBorQ+YLaqXMHDTQA7denbzRkxMDNYJamdrrU1aD7hirorMafr8X16p\nVTr7+PHjVK5c+bnvtXDhQpRSHD16lFOnTtGkSRP279/P0qVLnzp+5swZFi1ahKOjIydPnuTIkSOJ\nvvdDmzZt4s0334zfnzx5MtmzZycmJoaGDRty5MgRBgwYwOzZs9m6dSs5cjx+m7x//35WrFjBnj17\n0FpTvXp16tWr99LlP4RIDUf9jtL2h7acvHsSb+fSzC7cF++CNXBwdgbghp81VVoVjDs7lt+/9qdY\n9ZxpF3AymJIUygHvAA141Hyk4/YtltYajQadtgvFv2jp7IT69u3Ljh07yJIlC3v37mXHjh3xdw0l\nS5akYMGCnDt37pnHz5w5w7Zt2+KX5Sxfvjzly5d/7md17NiRyMhIgoODH+tT+OGHH1i6dCnR0dHc\nvHmTEydOJPo+O3bsoGXLljg5OQHGWg3bt2+XpCDStVgdy9zdcxn852AcsGNawd70KtoBJ3f3+C+a\nWsMbffICMGHAHfoNtiVbQc+0DDtZTEkKbYDCaVnv6EUk9o3+yTpApt4pJCW1SmeXKVOGn3/+OX5/\n4cKF3LlzB2/vJMuZvLRVq1ZRpUoVhg0bRv/+/VmzZg0XL15k5syZ7N27l2zZstG1a9cXLtctRHp3\nNuAsPX7pge9VX2q4lGdu0X6UK1AVh7j+sofaDcrNvmP2WKlYRo/PgpVb6lc4TQmm9CkcA9LngNqX\nlFJ3CKlVOrtBgwaEh4ezaNGi+GMJR/jUqVMnvrT1mTNnuHLlCsWKFXvm8RIlSlC3bt34Dutjx45x\n5MiRRD9fKcXEiRPZvXs3p06dIigoCCcnJ1xdXfHz8+P33x9VPnleWe06deqwbt06QkNDCQkJYe3a\ntdSpUyfZfydCmEusjmXennkUX1Ac36u+jMjbkT9rLKJKsdqPJYQvfsxKofpe/PC78YXz3C5/i00I\nYNqdghtwSim1l8f7FJ6/hFcmk1qls5VSrFu3jkGDBjF9+nQ8PDxwcnJi2rRpAPTp04fevXtTrlw5\nbGxsWLlyJXZ2ds893rt3b7p160apUqUoVaqUSf0iDg4ODBkyhBkzZrBs2TIqVapEyZIlyZ8/P7Vq\n1Yo/r0ePHjRt2pQ8efLEd2wDVK5cma5du1KtWjUA3n//fWk6EunOjQc36LymM39f+psaWcszu3Af\nKhaohkNcs+dD/netWfaTKyEhmiavBLBgZixe1S2vySihJEtnK6WeuTpKWo1AktLZL8ZSr1lKZ78Y\nueaUEatj+fLQl/Rd35cwwhiStx1jivfGycPjsUEqsbEw6BMP5n1tFK/r0fI2S9aYv0M5XZTOluGn\nQojM4HrQdTqv6cyWy1soYV+AWUX64+NVH6cnvlSFhineHeXJ9xuNJqKxfW/T+m3rZ72lRTJlPYUa\nwHygFJAFsAZCtNaW22gmhBBxIqIjmLNrDmO3jCWCCIbka8+Yoj1x9PDA5on5BSt+zsqUpdk5eykL\nAH+uukXj9p7ptmRFcpjSp7AAaAf8CHgDnYGUm6abQlJqJJFIe6asBihESjgTcIZW37bi+N3jlLPz\nYlrRvtQtVA+nrI++8+7YZ8/nP7pyO8CGTduNPoWcbhGc/i8ItyI5M1RCANOSAlrrc0opa611DLBC\nKXUQGGne0Exnb29PQEAA7u7ukhgsnNaagIAA7O3tkz5ZiGTSWvPV4a/o9ks3smDDnEIDeL94O+xc\nXePvDjZtc+STJdnZvs8YaZQ7eziF8oTyyxf3KFcvO8rx6eUyMwJTkkJo3BrNh5RS04GbmDaUNdXk\ny5ePa9eu4e/vn+h54eHhme6XjSVes729Pfnypd1yhCJjC4oIoteGXnx3/DsqOhVjYbFBVC70CvYO\nDkRFQc8xOdmwxYmb/rYAVCt5nzcbBTFylicoa7DJk+HuDhIyJSm8g5EE+gGDgPxAa3MG9aJsbW3x\n8vJK8jxfX99MN/wxM16zEM+z9/peWn/XmqshV3kvz+vMLD4Il5y5iImx5su1Wen6Ya74c9/wCeBV\nn3B6j8kDyjKK2aUEU0YfXQZQSsUA64HrWuvb5g5MCCFSSqyOZfau2QzbPAx3XFlRbCRvFXkD52zG\nkNLXe+Xl921Gf0GXFncYOkhT1idjNg8l5blJQSm1GJivtT6ulHIFdgExQHal1FCt9XepFaQQIvO5\ncQPWrDHmBDzL2bN5SWISPgBBsX5886Azp2P/pEB0VdqETsT/eCmWnzCaVaNjiEsImq/nBdCpt5tF\nVTVNaYldeR2t9cMynd2AM1rrN5VSuYDfAUkKQlioJUvg5Mm0jiJxCxZATExiZxRL+k0Kb4ZWb4Nd\nIOwaypUtHzELF57VLfrltFt06p87ueFmGIklhYQF8BpjDElFa31LRvgIYbn27YOHVdntsxi/dZUy\nqnyae/tFzrWxgkZVA/l20hWwfnpy2I7bt6md89mziKNio5h8aQHzr64kNx7MLD2fOu1b4+RiDYQ+\ndb6NDWTN4Z7o31tmkVhSCFRKvQZcB2oB7wEopWwAh1SITQiRwsaNg/Hjje3/lhyk6iu2aRqPSbI4\nPfNw1ogYsud5emTdheCrtNk2gAP3T9AiRy1m1J9NoaIVyZIli7kjzRASSwo9gXlALmCg1vpW3PGG\nwG/mDkwIkbICAx8lhCFtL1K1VhawtYCk8AJWX/6Nd3YORRPLlJIf0LfBaJzc3bGySlej6NO15yYF\nrfUZoOkzjv8B/GHKmyulmgJzMUpjfKG1nvrE867AN0CBuFhmaq1XmBy9ECJJV68af+JWRmV6nwsM\n+yAayDgJISQ6lH57J7Dy0lpKZ8nPojqzqFqp+VNVTUXSzNbFrpSyBhZi9EdcA/YqpdZrrU8kOK0v\ncEJr/bpSygM4rZRaZSkL+giR3mkNVapo/P2NfsBSXiH06fiAjNQCfOjeSVpt7cvFiOt0zf0aU5rM\nIke+wk/VLRKmMeffWjXgnNb6AoBSajXwBpAwKWjARRk9187AXSD6yTcSQiTPlSvg76/o3PgaNSuG\n82bDEJxyZIyEoLVm/pmvGbB/Ei7Ys6TSeDrW6Y+jm5uUu3kJ5kwKeYGrCfavAdWfOGcBxoS4G4AL\n0FZr/ZxRyUKIF3XsmPHY9Y0g6jfOOGNEAiLuMerUWHbf/48azmWZ7zOdsqV9sHfIGNeXlhKbvDY4\nsRdqrWenwOe/ChwCGgBFgM1Kqe1a66AnYukB9ADw9PTE19c3WR8WHByc7NdaKrnmzOF517x+fX6g\nCCEe1/C9mDGqzx66f5ixJycTRBAtPV6na553uR+Yhd179qR1aGaXGj/bid0pPFxZogRQFeMbPcDr\nwH8mvPd1jDpJD+WLO5ZQN2CqNmoln1NKXQRKPvn+WuulwFIwVl5L7spDsjpV5iDX/Mi770Jezyhe\nK5Ub7OxSP7AUFB0bzYRjC5l48jNykZ3+BUbwYcsPcXbLkEvIP1Nq/GwnNvpoPIBSahtQWWv9IG5/\nHKYNSd0LFFNKeWEkg3ZAhyfOuYIxxHW7UsoTIwFdeMFrEEI8QWvo1w8uXoR6FUMtfujplZAbtP93\nMDsDDtI0WzXmNJzDpWtBmSohpBZT+hQ8eXx2c2TcsURpraOVUv0whq9aA8vj6ij1int+MTARWKmU\nOgooYITW+s4LXoMQIoE9e+Drr+Gzz4z9hSMugZXltrWvvbqZTjuGEko4Y4r2ZEij8Th7eHDTf1ta\nh5YhmZIUvgL+U0qtjdt/E/jSlDfXWm8ENj5xbHGC7RtAE9NCFUIk5dw5qFHj0f5fsw5TprJlJoSw\n6HAGH5jC4vOrKWqdm4W1p1Gr8huPrYomUp4ppbMnK6U2AbXjDnXTWh80b1hCiOT46Sfj8duPTlCr\nWhQFilpmaYfj98/Sams/zoRdoq1nIz7xmUa+wmWlVEUqMHU5zv1KqauAPYBSqoDW+opZIxNCvLCV\nK43HVv+LwC675c3m1Vrz+fkf6Ll3DI7YsrDiGDrXHoBjtmxSqiKVJJkUlFItgFlAHuA2RkmKU0AZ\n84YmhHgRY8fC6dNQu/x97Nwsr8koMDKId3eNZO2Nv6jiWJwljRZQtnRd7Cx81JSlMeVOYSJQA/hL\na11JKVUf6GTesIQQpvr1V2jXrjYhIcb+9H5Xwcqymll2+h+g1Za++MXepV/+tkz63xxcPHLK3UEa\nMOVvPEprHQBYKaWstNZbAW8zxyWEMNGqVRAVBd2aXmP74uPUrGc5CSEmNoZJxz6j1l/t0bExfFdj\nLtPafI6rZy5JCGnElDuFQKWUM7ANWKWUug2EmDcsIYQptIbVq6GIVwjLZwWBjeXMR7gR6ke77YPY\nfnc/jdyqMrfxPIoWrSydyWnMlKTwBhAGDAI6Aq7ABHMGJYRI2urV8F3corhFvO5a1LrCv17fSrtt\nAwkhnI+K9WR4k0k4Zc8udwfpgClDUh/eFcQqpX4DAuLKUggh0kB4uDFbedmyh0di6dntHFAwDaMy\nTURMJMMPTmfe2a8pZJ2L7+tMpX6V1jg6O6d1aCJOYgXxagBTMcpZTwS+BnJg9C101lpvSp0QhRAJ\nNWsGRk00zb+fHaFGDdjml+gK9+nCmaCLtP6nP8eCz9LK04e5r36GZ76i2Fp4CY6MJrE7hQXAKIzm\noi1AM631bqVUSeA7QJKCEGYWGAh162ru3jUWtI+KAj8/ha11DIFbDuLo6WI8kY7XD9Ba89XFdXTd\n8yF2WDOz3HB6+IzAOVs2WfcgHUosKdhorf8EUEpN0FrvBtBan5J/SCFSx/LlcPSoonD+EMoXDiUm\nFuzLxDK2ux+OudJ/uYegqGB67v6Y1dc2UsHeiy8aLqRsGVn3ID1LLCkkXOwm7InnpE9BiFTwww/G\n45mfTmHt5pLgGfs0iedF7A04QsstfbkefZue+VoztdlcsubKLZ3J6VxiSaGCUioIo3qpQ9w2cfvp\n/ydSCAu2axd88YVR8bRs4WCsXRzTOiSTxepYZp1czvDDM8iOM19Xm0mrWt1xlEJ2FiGx9RSsUzMQ\nIYTh8mV45ZVH+z+MOwXWlvEL1S/sDh13DOHvO7upnbUSixrPpXjx6jL3wIJYzsBmITKB4GAoVcrY\nnt7zHH3eCcbJQgrbbb75L218B3CfYIZ7deOjZtNwcneX5iILI0lBiHTknXcgLAzqlLvHsD6hYJ/+\nm42iYqMYffhTZpxaRl6Vky9rf0rjau1k7oGFkqQgRDrx4AGsWweKWDYvvQT26f8O4ULwVdpsG8CB\n+yd43aMWC5osJnfBEjL3wIKZfF+nlHJJsF3UPOEIkXktWmQ8zux30SLWQlh9+TdKbGjC4fsn+KT0\nIFa1/YX8RcpIQrBwL3KnsEMpdRH4FpgCFDFPSEJkPlFRMGaMsd2/YyDgkuj5aSkkOpS+/43ny8vr\nKJUlH4t95lK1QlMcHNN/U5dIWmJlLhyBSK11NIDWuoJSqjfGbOZ2qRSfEBna+fNGYbvffoOICKhZ\nOhBbt/R7l3Do3klabunDpcgbvJPnf8xp9hmuufNibS2DFTOKxO4UtgBvArcAlFItgd7AqxgVU380\ne3RCZCBaw7hxcPXqo2MrVjzatrWOYfOCM2DlluqxJUVrzYKz3zBg/ySy4sBy7ym8XasXTm7pL1bx\nchJLCg5a64cJoQfQHWiotfZXSk1NleiEyAAuXYJeveDiRThzxjiWM1skSoGHG7xe8w5Lxt3CytYa\nK0fXNI31WQIi7tFl5wh+u/UP1Z3LsLjBHEqXrkMWWSYzQ0osKQQopcYC+YGWQFGt9T2lVG5AZqII\nYaING+CPP6B84SBql4vlq7Hn8SqZoCiAUmCTPodv+vrt4a2t/QnQ9xlcoBNjm83AOacsk5mRJZYU\n2mA0F50BegB/KqWOAvWB0akQmxAW79QpGDAAnB1jOPTzRZSjA+m5E/mh6Nhoxh9dwKQTi8hFdn6s\nuYDmNTvj6JL+YxcvJ7EyFwHApIf7SqldQC1gmtb6dCrEJoRFi4iAevWM7UFvXUTZW0Zzy5WQG7y9\n7QP2BB6hefYazG44B6+ilaRURSZh8pBUrfUNpHNZCJNoDS1awO3b4JU3lAnDwsEq/f9SXXP1Tzru\nGEI4kYwv3pfBTcbjmC2bNBdlIjKjWYgUEBEBU6cas5IBDh+Gv/6K2/7yGGRJ36N0wqLDGXxgCovP\nr6aYTR7m15lGncpvSqmKTEiSghAv6eRJaN7cGGUEYGNtLEXi7BDL8a8O4JIv/Y0oSuj4/bO09u3P\n6dCLtMvVhFlN5uKRr4jMTM6kJCkI8ZKGDjUSgnfxIP5ZcgpH9wSrimVxTbdLZWqt+fz8D/TcOwZH\nbJlTfiTv1h2Cc/bsskxmJpbYjOYHJLLCmtbaMgq8C5GCYmPh7t1H+6+9ZiyEky9nBHt/ugx26buZ\n6KF7kfd5f/do1lzfTBXH4ixtOJ/SperIMpki0dFHLgBKqYnATeBrjFXXOgK5UyU6IdKRmBho1Ah8\nfR8/ns05iuWjz4MFTObSWrPq0np67R5DCOH0zvcWnzSbI8tkinimNB+10FpXSLC/SCl1GBiT1AuV\nUk2BuYA18IXW+qmZ0EopH2AOYAvc0VrXMyVwIVJbz55GQshiE8PAdkatCvssmvdaBVOgWPpPCLvv\nHKLfnglj9UZoAAAgAElEQVTsDzpOSdu8TKgyguY1OuPkmr77PETqMiUphCilOgKrMZqT2gMhSb1I\nKWUNLAQaA9eAvUqp9VrrEwnOcQM+A5pqra8opXIm4xqEMKuffjKaiJYtM/YPfnWY0lUSjspJ3wkh\nMiaS8ccW8MmJJbjiwID87RnXdAYuOT2xsZFuRfE4U34iOmB8258bt78j7lhSqgHntNYXAJRSq4E3\ngBMJzukArNFaXwHQWt82MW4hUsWNG9CmzaP99VOPUbpy+q1imtCZoIuMOjibX29sJYIomrnXYEGj\nheQpWEr6DsRzKa2f25f8cm+s1FsYdwDvx+2/A1TXWvdLcM7DZqMyGHP/52qtv3rGe/XAKLWBp6dn\nldWrVycrpuDgYJwz2bhrueYXt2FDbrZuNW5ajx1zJSrKirHD91Gnzh2sbUiXo4mCIyJwjuvTiNEx\n/HhzDUuuGLc2VWwrUjWHN81zvY6zi0uGGVkkP9svpn79+vu11t5JnZfknYJSKh8wH6PEBcB24AOt\n9bVkRfb051cBGgIOwC6l1G6t9ZmEJ2mtlwJLAby9vbWPj0+yPszX15fkvtZSZfZr1tr48yxnzkDL\nlpqwsMePX75s/NKsVDSI0gWCye0ZycdtrLHOWtiMUb8c34sXqV0wPysurGHBia85EnKGV1zKM6v2\nZCqUqY+1rW2GK1OR2X+2zcWU5qMVGKutPbyJ7hR3rHESr7uOUWH1oXxxxxK6BgRorUMw+i62ARUw\nivAJ8VK0hgoV4OjRxM5S1C5zj6zZogGwtoIy+aBnizu0eDPhwjHpu8noSthVqm8azIH7J/AgG2MK\n92TIqxNwcneXBXDECzElKXhorRMsBcJKpdRAE163FyimlPLCSAbteLov4hdggVLKBqMcd3XgUxPe\nW4gk3bplJIQGle9SrEAo1tbGsFIgfjt/zghG9w+Dp75FW8Yv0hP3zzH+8Hx+uL4JWxRji/dhsM9H\n2GfPLusdiGQxJSkEKKU6YSzDCcboo4CkXqS1jlZK9QP+wPgftlxrfVwp1Svu+cVa65NKqU3AESAW\nY9jqseRciBBPeriq2fDOt3m1WWJj8C2vWSVWxzLn1JcMOTQVBXjbV2JJoymUKlVH1koWL8WUpPAu\nRp/Cw2/w/wLdTHlzrfVGYOMTxxY/sT8DmGHK+wnxIubO1YCiWrkw0nvzz4s4cu8UPXZ/zJ7AI9Rw\nKsuculMIuGNFxUqNZQKaeGlJJgWt9WWgRSrEIkSK+fhjuH1bMbrzFbLlyRjDLxPeHdhjw6jC3fmw\nyUQcc+Rg+/btkhBEikjr0UdCpLj7922ZFLc8VOuG98HK8tvWb4T60XHHUHwD/qOmcznmNpxB2WK1\ncXDKOHdAIn0w5avFCmA9kCfuz4a4Y0KkS3v2ZAfgy1GnqVTd8voLEoqJjeGHKxsp+ksjfAP+48Mi\n7/FHly1ULtdIEoIwC3OOPhIi1e3dC1OmlAKgVbMwUJbb6frz1T8Y+N9krkX6UcQmD4vqzKB2lTck\nGQizMtvoIyHSwgcfGI+T3z+Hc3bLvEu4EnKDofun8eP1TRQkJx8V602vmgPwzF9UahUJs3vR0Uca\n2ImJo4+ESC0xMTBrFuzaBTY20YwaEAY2ltWXcDH4Kp+f/YGpp5aigXdzv86kRtPJlqcQdnZ2GaY8\nhUjfZPSRyBBmzYIRI4ztcSMOWMxiN2D0G8w6tZwRh2cC8IpjecZXH0HNCv/DMWtWSQYiVSW28loZ\noIjWen3c/qfAw8LrC7TWB1IhPiGea/t2mDDBWA1tyxbj2J1fd3PUOhCwjKRwNPA03Xd+xJ77R6ju\nVJbxNUZRp0xTsri4SFORSBOJ/dRNBaYk2H8V+BhwxFhg500zxiVEkpo1g5AQKF84iPKFoVszP9yL\nZoNL99M6tCTF6lhmnlzGiMMzsQbGF+vDoCbjcXBzk2Qg0lRiP325tdY7E+wHaa1/BlBK9TRvWEIk\n7vZtIyE08r7L5q/8jGJGgLFibPp2PdSPjv8O4Z87e6mVtQKf+cyiRMlXsJM1DkQ6kFhScEm4o7Wu\nkWBXVkgTaeb8eRg0yNge1P52goSQ/q29uplOO4YQSgSjvd5neJOJOOfMKbORRbqRWFK4oZSqrrXe\nk/CgUqoGcMO8YQkBly/Dpk2PH9Maevd+tP9q7RCe+P6SLoVEh/LBvsksu/gTRaxzM7/mJOpVa4Oj\nS/qPXWQuiSWFEcD3SqmVwMNO5SpAF6CtmeMSghYt4MiRZz/3XrMbjO19G+us6XsiV0xsDMsu/MT4\ng/O5Ee1Ph1zNmFT/E/IWKp3hFr0RGcNzk4LW+j+lVHWgH9A17vBxoIbW2i8VYhOZWHQ0HD+uKZo/\nnG2fnyFhX4GNDXh4WkGW9D1b+Z/b//H+ztGcC7tCIeXJ4gpj6eQzEIesWaW5SKRbiQ5z0Frfxhhp\nJESqat4cYmIUH3a6Tm4vy+qAvR/5gL77xrPq8gay48TkEgPoXmsgrp555e5ApHsy9k08JjDQGPef\n2kaNgu++0ygF4eEQEaGws42hySshGMt3W4Yd/vt4+5+B3Izyp3Ou/zGqzscU9CqPnb29TEITFkGS\ngog3dqwxGSztKDo1uUVMLNhaxTKx/x3yF7eMhBAVG8X4owuYfGIxHmTlS+9ptHqlO46urtJUJCxK\nYjOav9Zav6OU+kBrPTc1gxKpLyTkUUIY1uECKGtsrCE6bk3jh9vPOpZS220aB1O1pm2CqOzNcq0p\nbc+dw/Tc9TGHg0/TNFs1ZvvMonBxb+zsLSN+IRJK7E6hilIqD/CuUuornpgVpLW+a9bIRKr6/HPj\ncXDbS0wfEwUqOg2isE36lHQkIOIeww/OYPnFn3HElmklB9Gn/oc45sghdwfCYiWWFBYDfwOFgf08\nnhR03HGRAezaZUwGs7LSzBwaAErGziflr1s7affPQAJi79Pasz7TfWaSz6sMWewsqzKrEE9KbEjq\nPGCeUmqR1rr3884TlmvSJPD1hb//Nva7vHoT5ZS+h3mmtfCYCEYdns2np1eSmxz8VHM+Tat1wtHV\nVTqSRYZgSuns3kqpCkCduEPbtNbPmVIkLMXmzcbi9gAVijzgLR8/PhoUAdaW1YSTmnz99tB791hO\nhV6kpUcdptabQaFilWSYqchQkkwKSqkBQA9gTdyhVUqppVrr+WaNTJhFYCD07w/ffGPs71pyhBr1\n7EApLK1NP7WERocx4tBMFpz9Bjcc+bTscN6rOxQnd3fpOxAZjilDUt8HqmutQwCUUtOAXRirsQkL\ncvkyFCr0aH/JsHPUqGMblxDEs2y88Q9dt4/AP/YerXL6MKX+DAoULIO9VDQVGZQpSUEBMQn2Y7CE\n+sTiKVWqGI8NKvszfcAtqtS0AyynwmhqCokOZciBqSw5/z158eDzKpNoV6uPzDsQGZ4pSWEFsEcp\ntTZu/01gmflCEilt7VpYtw4CAqBs4WA2L7+FlYOMknkeX789dN0xgsuRN2mbqwkzmszBM28R6TsQ\nmYIpHc2zlVK+QO24Q9201gfNGpVIMQEBWXjrLWPb1jqGLz86j5U0fTxTeEwEww9OZ/7Zb8iGE0sq\nT6RDnX44ycgikYmYVOYibj1mWZPZgly7Bh06wPbtrwCwYsQpunaMAlu5Q3iWk/fP89Y//TkRcp63\nPOszu8FcPAsWl3kHItOR2kcZSGQk3L0Ldetqzp41vtnmz3ufOiVD6NIuAuzkDuFJWms+P/8DPfeO\nwQFbFpQfRZd6Q3DKlk3uDkSmJEkhg4iIADc3TXi4AhReeUJ5o44/LTqfoH7RIlhSpdHUcjcikHd3\njeKXm39TxaEY8+vNoFL5JjKySGRqJiUFpZQnUDVu97+4dRZMeV1TYC7GEJcvtNZTn3NeVYxhru20\n1j+Z8t7iccuWQXi4olzhIFrVv8OQbsG4eNjje1FGyjzLttt7ab2lH3d0IP3zteGjJlPJnrsANjby\nPUlkbqZMXnsbmAH4YgxFna+UGpbUL2+llDWwEGgMXAP2KqXWa61PPOO8acCfyboCQXQ0jBhhbG9d\ndBb3Qi5YSoXR1BYdG834owuYdGIRucjODzXn878anXFwcZHmIiEw7U5hNFD14d2BUsoD+AtI6ht9\nNeCc1vpC3OtWA28AJ544rz/wM4/uRMQL2rIFgoOhWskg3PNJ08fzXAq+RtvtA/kv8Cj/y16TWY3m\n4FWkogw1FSIBU5KC1RPNRQGAKW0SeYGrCfavAdUTnqCUygu0BOojSSFZJkyA1auN7d9mnwYb17QN\nKJ36/vJGuuwcTiRRTCo5gA8ajcExWzaZiCbEE0xJCpuUUn8A38XttwU2ptDnzwFGaK1jE7t1V0r1\nwKi/hKenJ76+vsn6sODg4GS/Nr2JjFQMG1aBI0fcAPCpc4Vj3IGLjy9zERwRge/Fi2kRYppJeM1h\nMWHMubCQPwP+Iq/KzQCvfpR1qci+o0fTOMqUlZF+tk0l12wepkxeG6aUag3Uiju0VGu9NrHXxLkO\n5E+wny/uWELewOq4hJADaK6UitZar3sihqXAUgBvb2/t4+Njwsc/zdfXl+S+Nr355hs4cgRyZY/g\nlxlnqVbLFlSRp87zvXgRHy+vNIgw7fhevEjFvO5MPLKQBWe/IZJoOuVuzvQmn5Ijrxe2thmv8F9G\n+tk2lVyzeZg6ee1njHb/F7EXKKaU8sJIBu2ADk+8b/xvK6XUSuDXJxOCeFpkJAwcaGwfWXUMj8Ky\nKM5DN0L9mH9+EWt2/wLAK47l6V22My1rvidrHghhgsTWaN6hta6tlHqAsdJa/FOA1lpnTeyNtdbR\nSql+wB8YQ1KXa62PK6V6xT2/+OXDz5y6dDHqGJUrHIxHAelYBqNExdTjSxl/fAEA9Z0r06d8d5pV\nbYeds7MMNRXCRImtvFY77jHZX0O11ht5ov/heclAa901uZ+TWUREwEcfPepY3r7kpHQsA//c/o/2\n2wZzM8ofn6yVae7+Fv3e6IOtk5MkAyFeUJJDL5RSX5tyTJjX6dNQpgzMnGns/z3nKK75E71Zy/Ai\nYiIZfnA6Pn+/Q0xUFIsrjWNDp01UzlsNB1dXSQhCJIMp/2vKJNxRStkAVcwTjniW2FioVg2CgqBi\nkQf8+8VJHD1dMu3iOLE6lrmnv2LqkSXcjrlLixy1mFZvOl7FqmBnZ4e1tawRIURyJdanMBIYBTgo\npYIeHgYiiRsJJFJH27ZGQqhWMog9P1wGO7e0DinN3Aj1452dw9niv5tS1nkZUXYE3esMxilHDplz\nIEQKSKxPYQowRSk1RWs9MhVjEgkcOQI//QSg+WXOxUxd6XT9tb9pv30QoUQwrFAXRjeZgoO7u8xI\nFiIFmTJPYaRSKhtQjAQFdbTW28wZmDA8HHr6y5QT5PLKnAkhNDqMwfunsOTC93hZ52axzyxqV2qB\no5NTWocmRIZjSkG894EPMCafHQJqYFQ0bWDe0MT167B1K0AsLZrHABlv0lVSDt87RautfbkQcY23\nPRsxq8k8chUoJp3IQpiJKY2wH2DUJbqsta4PVAICzRqVwN8fypUztr8dcwbsM1fV05tht+n333gq\nbnqDOxEBfF55Esvb/Uher5KSEIQwI1P+d4VrrcOVUiil7LTWp5RSJcweWSZ26hRUrgxhYeBdIoj2\nLcMBx7QOK1VorVl07lv67psAgI9LJT6tP53SpevI0phCpAJTksI1pZQbsA7YrJS6B1w2b1iZ208/\nGQmhcfU7/LHwCjg6p3VIqeJuRCDddo9k/Y0teDuWYGrN8dQs3wx7Z2cZWSREKjGlo7ll3OY4pdRW\nwBX43axRZXLbtkGhvJH8ufQG2GeOhPDYSmj52zKh6SxcPHPJnAMhUtkLff3SWv8DbAOGmyeczC08\nHKZMgc2boXi+UMgEzSXRsdGMPTKPen93wlorfqwxjymtl+CWJ68kBCHSQGKT1/IDHwN5MJqOvgMm\nAO/waG0FkYLefBP++MPYHtH2MqiMPQT1SsgN2m4byO7AwzR1r8HCJovI71UmQ5a2FsJSJNZ89BXw\nD0bJ7KbAPowhqeW11rdSIbZM4/hxmDbtUUI49e0BSlTI2KON1l7dTPsdA4kgmsmlBjKg0cc4Zcsm\npa2FSGOJJYXsWutxcdt/KKXaAB211rHmDytzmTTJqHxqpWLZvfgwJapk3H6EsOhwBu3/hCUXvqeo\nTR6WNJjLKxX/h71Dxr4rEsJSJNrRHDeT+eFXtwDAVcV9ldNa333uC0WSLl+Gzp0hLExz8iQ0rxXI\nbwuvZOgyFscCz9Bya1/OhV+hY+7mTG86h5x5vGTegRDpSGL/G12B/TxKCgAH4h41UNhcQWUG3bsb\no4yKFwyhfMEo3mnkBxn027LWmsVnv6PP/vE4YsvCyuPoUncgjlmzSnOREOlMYgXxCqViHJmK1vDv\nvxp312hOrb+IsrfjBQeCWYy7EYF02zmS9be2UNWpFEuaLKB0iVrYZYKRVUJYIrlvT2Xh4dC/P4SG\nKoa+eysuIWRMO/z30WpLP/xj79E/fzsmvTYH5xweMhFNiHRMkkIqGzAAvvjC2H7v9dtAslc7Tbdi\nYmOYdPwzxh1bgCfZ+L7mfF6r2QVHl4x3rUJkNJIUUsmOHfDtt/D558b+zV/+I1fxjLec5rXQW7Td\nNpCd9w7SKFtVljRdKnMPhLAgJiUFpVRtoJjWeoVSygNw1lpfNG9oGcO4cbB796M5CK5OUczpf4Fc\nJVwz3HKav1z7i/bbBxFGJOOK92Fw4/E4u7tLZ7IQFsSU9RTGAt5ACWAFRlH/b4Ba5g3N8kVGwuTJ\nmhyuUZQpHEmnhrf4cEA4WFuDyjglHMKiwxl6cBqfnfuWwta5WVr/U16p9DoOjpmjsqsQGYkpdwot\nMdZQOACgtb6hlJLGYRNs2gTR0YrJva7y7jvRxkGVsZaOPHH/HK19+3Eq9CJv52rErMayCI4QlsyU\n/7mRWmutlNIASilZAzEJsbGwdy9s2GDsN652H1TGmqWsteaL8z/SY+/HOGDLgkof06XOIJzc3KS5\nSAgLZkpS+EEptQRwU0p1B94FPjdvWJZt5Up47z1ju3D+MPIXzlh3B4GRQby3axRrbmymkkNRFjaY\nQ6UyDaRUhRAZgCnrKcxUSjUGgjD6FcZorTebPTILtHu3cXfwySfG/m9Tj1KhTDTYZpy29Z3+B2i1\npS9+sXfpl/9tJjWfg7NHTilzLUQGYUpH82Dge0kEidMaGjeG4GAATZsGt2n+ujXYZIzJaTGxMUw5\nsYSPj84lJ658V30OLWq9K3MPhMhgTGk+cgH+VErdBb4HftRa+5k3LMtz+7aREDo08mPVLD+wsgKb\njNFsdD3Uj3bbB7Lj7gEauVVlbqO5FC1WhSxZMsb1CSEeSbLegNZ6vNa6DNAXyA38o5T6y+yRWYCo\nKBgyBHLk0JQsqQHo2Owu2NtDBvmFueH6For/0pgddw/wcdGerOnwGyVLVZeEIEQG9SLjBm8DtzBK\naOc0TziWIyAAcuR4uKdo19APhyzRNKodAVh+H0J4TATDDk5nwdlvKGSdi6V1Z1K7yps4OMngMyEy\nMlP6FPoAbwMewI9Ad631CXMHlt517248vlLqHrOG3KBGnYdlHCw/IZwKOk9r3/6cCDnPW54N+LTx\nfHIVLC5zD4TIBEz5X54fGKi1PvSib66UagrMBayBL7TWU594viMwAmPNhgdAb6314Rf9nNR27x6s\nXQt5ckSweflVHLNnjKUztdYsv/AT7//3EfbYMK/8KLrVGyLLZAqRiTw3KSilsmqtg4AZcfvZEz6f\n1MprSilrYCHQGLgG7FVKrX/iLuMiUE9rfU8p1QxYClRP1pWkAq1h6lSYMcPY79PqZoZICFprVl/+\njSlHF3M0+CyV7Iuw0GcWlco3kbkHQmQyid0pfAu8hrH6mubxFdhMWXmtGnBOa30BQCm1GngDiE8K\nWuudCc7fDeQzOfI0sHAhjBplbFcvdZ/RPe8Clj1T+U7EXbrv/ph1N/4iJ9n4sGA3BjYYjXuegtJc\nJEQmpLTW5nljpd4Cmmqt34/bfweorrXu95zzhwIlH57/xHM9gB4Anp6eVVavXp2smIKDg3F2Tt4v\n8Vu37GjfviYA3yzdQt78kRZR5TQ4IgLnZ6xyprXG9+52pp+dRTgRvJa9Ke/k7YKrUzayZMli0c1F\nL/PvbKnkmjOHl7nm+vXr79daeyd1nikdzX9rrRsmdexlKKXqA+8BtZ/1vNZ6KUbTEt7e3trHxydZ\nn+Pr60tyXhsVBW+/bWxP63GOjvXS9Q3NY3wvXsTHy+uxY2eCLtL3vwn85b+LfFaefFZzIg2rt8PB\n2dmik8FDyf13tmRyzZlDalxzYn0K9hhDaXIopbLxqPkoK5DXhPe+jtFJ/VC+uGNPfk554AugmdY6\nwMS4U1WrVvDrr2BrE8uw3iGAZbazR8dGM/3EF4w++ikAH+Rvz9hGU8iaN5+UqRBCAInfKfQEBgJ5\nMPoVHiaFIGCBCe+9FyimlPLCSAbtgA4JT1BKFQDWAO9orc+8WOip48gRIyHULnefsd1voBwtMyHs\n8N/HO9uHcSniBnWzVuLTulMoVbK2zDsQQjzmuUlBaz0XmKuU6q+1nv+ib6y1jlZK9QP+wBiSulxr\nfVwp1Svu+cXAGMAd+Cyu2SLalDav1HL6NFSoYGy/3cifRq9a3rfpoOgHvLdrFMsv/UwOsjK9zFB6\n1R2GU44cWFklOaFdCJHJmFIldb5SqixQGrBPcPwrE167Edj4xLHFCbbfB57qWE4PvvsOOsTd14zs\neIl+3cIByyntoLXm1xtb6bBvCCGE0sqjHrMazSV3wZLYPaPjWQghwPTlOH0wksJGoBmwA0gyKViq\nlSuhWzdju0/Lq3zyYbBF1TK6HR7Au7tH8tvNf/AkJ1/WmkXTqu1xzJo1Q3QkCyHMx5SB6G8BFYCD\nWutuSilPjDWaMyStHyWE78ef5O02gLXlJIRfr2+l07ah3CeYPvnaUCfbm7So1wZbW9ukXyyEyPRM\naVQO01rHAtFKqawYhfHyJ/Eai3XjhvH4/v9u8HY7a7CQUTlHA0/TftsgXt/WCzcrJ36sOZ9prZfi\nmS23JAQhhMlMuVPYp5Ryw1iCcz8QDOwya1RpaM4c4/HtV4MwLWemrVth/ry3exQbb20DoE3Ohnz6\n6nw88hWx+EloQojUZ0pHc5+4zcVKqU1AVq31EfOGlXZWrjQeG70SSnouYREVG8WPVzbx/q5RhBFJ\n9zxv0r/GQIoWqox9BpmEJoRIfYlNXquc2HNa6wPmCSntnD8Pd+5A/UqBKOf0O35/zdU/GfzfJ1yO\nvImXdS7mVJ9Ao+ptsXd2lmGmQoiXktidwqxEntNAgxSOJU0FBkL1uPqsIzpdB5X+2uHPPbjM4P1T\n2HBzK7lwZ1LJD+hdazBZc+WR4nVCiBSR2OS1+qkZSFrbutVYTa1UoWAa14sE0k9SCIsO59NTK+PL\nU/TI14opjWfhnCuPLIsphEhRpsxT6Pys46ZMXrMUsbFGfSOAvV+fxcolfTQdhUSHsvjsamYdW8bN\n6DtUdyzD7NoTqVi2EY4uLmkdnhAiAzKlzaFqgm17oCFwgAw0eW3kSOOxQeW7OGVL+2/eWmv+vLWD\nbv+O5GaUP8WtczOq/Gi61P4Ax+zZpXidEMJsTBl91D/hftzw1OQtaJBO7dsHoPlp5hWwTds1lnf4\n76PLjhFcCL9GTtxYVGkcnWr2IYurqzQVCSHMLjm9kyGAV5JnWZBjxzSdXvUnW960Swh3Iu7Sb+9E\nvr+6EXecGVawKz1e6U++AqWxt7f8JT+FEJbBlD6FDRijjcCYzVUa+MGcQaWmjz+G27cVFYqFpcnn\nR8VGMfvkCiYfWcQDQung2ZSP646jUOEKZMmSRYaYCiFSlSl3CjMTbEcDl7XW18wUT6qKiIBJk4zt\nRtWCgNStHnri/jnabx/MkQenKWObn4nVZtOkWnscZL6BECKNmNKn8A9AXN0jm7jt7Frru2aOzew2\nxhX1/urjs1Sslnrt9ddD/Zh9Yjmzz67EHhumlRpI73of4uDuLvMNhBBpypTmox7ABCAciMVYgU0D\nhc0bmnlpDZ3jBtv6eD8AZf6SFlGxUcZ8g8OziSaWWk7lmVlrAhXKN5IV0IQQ6YIpX0uHAWW11nfM\nHUxq2rEDgoOhUrEH5C9q/o7cM0EXabttIIcenKKqQ0kmVhtF7QqvYZ81qwwxFUKkG6YkhfNAqLkD\nSW3ff288rvzoHNiY71u61prPz/9Az71jsAE+KTGA/g0/wj5bNmkqEkKkO6b8VhoJ7FRK7QEiHh7U\nWg8wW1SpYO9eUEpTvqL5vqWvu/YXEw8t5MCDE1RxLM68ejOoVKahNBUJIdItU5LCEmALcBSjT8Hi\naQ0nTmjef80PzDAH4HLIdQbvm8KaG5txx4V++doy6X+f4uyRU5qKhBDpmilJwVZrPdjskaSiH3+E\n4GBF+eIpOzdBa83yCz/R87+PiUHzbp43mNFsHk4entjZpe5wVyGESA5TksLvcSOQNvB485HFDkl9\nODfhtdqBQMo05fzrv59eu8ZyLOQsFR2KMKfudKqWaSyF64QQFsWUpNA+7nFkgmMWOyQ1JgaOHoVq\npR5QqMTLNx2Fx0Qw5vBcZpxeRjYcGVygE2OaTZemIiGERTJl8lqGqnM0erTx+HajAHjJX9qbbmyj\ny44R3I65yxs5avNpkwXkzl9CahUJISxWpltPYdUqDSg6Nw8Akte0c/L+ebrvGs2/9w6SmxwsqTyB\njrX74ejmJmsjCyEsWqZaTyEszJpr1xSD217Dw+vFZzCHRocx6+RyxhybRxas6Jm7JcPqjiafV1np\nSBZCZAiZaj2FS5eM0tg1yofBC3yj11qz6tJ6Bv83Bf/Ye7ziUp75DWZRukQtstjZSfE6IUSGkanW\nU7hwwRhpVKlEKOBg0mu23NpF7z3jOBN6icJWuZhceSLta/bCMXt2SQZCiAwnU62ncOmSE/Z2sRT2\n0v3kOJwAAAzMSURBVEme6xd2hz57x7Hm+mY8cGV4oa582Ggizjk9sbW1TYVohRAi9WWq9RQuXnSi\ntFc4VnbP/6V+MfgqX1/4henHPyeEcLrmfo0J9T/BI38xGVUkhMjwnpsUlFJFAc+H6ykkOF5LKWWn\ntT6f1Jur/7d37jFWVVcc/n4yYEEQUBQRH4DgA634ANEJ2sEn0LTUlFYrqdGKBl+IRos1ajXViDGp\n1lpFao0lUZGq+EQpiYyOCgoiOAxYi1QBH/gCddSqU1b/2Huuh/HOzBm4D+7c9SU355y999l7rXNm\n9tqvs7Y0CvgT0AG4y8ymNolXjB9DcLp3hpktabsa6Vi9uiujh30KWVr6yze+wQ3L7+S+tU8AsN92\nu3P9UVdy0tBfsUP37r6qyHGcsqClnsItbP7BWiOfxbiftJSxpA7AX4ATgHXAIkmPmdmKRLLRwKD4\nGw7cEY8558MPYcOGTgwZ+J3D14ZNDTz7wSKuefVWnt8YbNHJvX7ExEPPZcT+J9CxWzcfKnIcp6xo\nySj0NrPapoFmViupX4q8jwBWmdlqAEkzgbFA0iiMBWaYmQELJfWQ1MfM3kurQFrq6sLxgAFfseCj\nZcxaPYfZa+fx9jfv0YWO/KLXsUw56jIOGFhJpy5d3K214zhlSUs1X48W4tIs3ekLrE1cr+P7vYBs\nafoCOTcKz7//FB0mT+LU+nV8Oe+/AOzNLlw18FwmHnUhO/cdQEVFhbumcBynrGnJKCyWdLaZ/TUZ\nKGkC8Ep+xdqc6JDvHIDevXtTXV3d5jx6dl7NkN12RLY//Xfox5jeP6ZPt75s16EDb6xZD2vW51jq\nbYP6+votel6ljOtcHrjO+aElozAZmC1pPN8ZgaFAJ+DkFHm/A+yZuN4jhrU1DWY2HZgOMHToUKuq\nqkpR/OZUUcXgHQdTWVmJJDp27FgWk8fV1dVsyfMqZVzn8sB1zg/NGgUzWw9UShoJHBSDnzSzZ1Lm\nvQgYJKk/oaI/FTitSZrHgAvifMNw4NN8zCc0IsndUTiO47RAGjcX84H5bc3YzBokXQDMJSxJvdvM\n6iRNjPHTgDmE5airCEtSz2xrOY7jOE7uyOsSGzObQ6j4k2HTEucGnJ9PGRzHcZz0uPMex3EcJ4Mb\nBcdxHCeDGwXHcRwngxsFx3EcJ4MbBcdxHCeDwgKg0kHSh8DbW3h7L+CjHIpTCrjO5YHrXB5sjc57\nm9kurSUqOaOwNUhabGZDiy1HIXGdywPXuTwohM4+fOQ4juNkcKPgOI7jZCg3ozC92AIUAde5PHCd\ny4O861xWcwqO4zhOy5RbT8FxHMdpgXZpFCSNkvQvSaskXZ4lXpJujfGvSTqsGHLmkhQ6j4+61kp6\nUdKQYsiZS1rTOZFumKQGSeMKKV8+SKOzpCpJSyXVSXq20DLmmhR/290lPS5pWdS5pL0tS7pb0geS\nljcTn9/6y8za1Y/gpvtNYABhQ6BlwOAmacYATwECjgReKrbcBdC5EugZz0eXg86JdM8QvPWOK7bc\nBXjPPQj7oO8Vr3ctttwF0PkK4MZ4vgvwCdCp2LJvhc7HAIcBy5uJz2v91R57CkcAq8xstZl9A8wE\nxjZJMxaYYYGFQA9JfQotaA5pVWcze9HMNsTLhYRd7kqZNO8Z4ELgIeCDQgqXJ9LofBrwsJmtATCz\nUtc7jc4GdFPYSrErwSg0FFbM3GFmzxF0aI681l/t0Sj0BdYmrtfFsLamKSXaqs9ZhJZGKdOqzpL6\nEraOvaOAcuWTNO95X6CnpGpJr0g6vWDS5Yc0Ot8GHAC8C9QCF5nZpsKIVxTyWn/ldZMdZ9sjbq96\nFjCi2LIUgFuAKWa2qRz2445UAIcDxwGdgQWSFprZG8UVK6+cBCwFjgX2AeZJqjGzz4orVmnSHo3C\nO8Ceies9Ylhb05QSqfSRdDBwFzDazD4ukGz5Io3OQ4GZ0SD0AsZIajCzRwojYs5Jo/M64GMz+wL4\nQtJzwBCgVI1CGp3PBKZaGHBfJek/wP7Ay4URseDktf5qj8NHi4BBkvpL6gScCjzWJM1jwOlxFv9I\n4FMze6/QguaQVnWWtBfwMPDrdtJqbFVnM+tvZv3MrB/wIHBeCRsESPe3/SgwQlKFpC7AcGBlgeXM\nJWl0XkPoGSGpN7AfsLqgUhaWvNZf7a6nYGYNki4A5hJWLtxtZnWSJsb4aYSVKGOAVcCXhJZGyZJS\n56uBnYHbY8u5wUrYmVhKndsVaXQ2s5WSngZeAzYBd5lZ1qWNpUDK9/wH4B5JtYQVOVPMrGS9p0q6\nH6gCeklaB/we6AiFqb/8i2bHcRwnQ3scPnIcx3G2EDcKjuM4TgY3Co7jOE4GNwqO4zhOBjcKjuM4\nTgY3Cg6SdpM0U9Kb0TXCHEn75rnMakktLomVNDmutW+8niOpRw7Kfit6i31N0rOS9k5xzxVNrl/c\nWjliPpMkrZR0byLspOjldKmk+ughdKmkGS3kc5ikUSnKO17S977VkNQ1/g3USlouqSb57HNN/I5i\nY77yd7YcNwplTnQiNhuoNrN9zOxw4HdA7+JKBsBkIFMxmdkYM8tVRTLSzA4GqoErU6TfzCiYWWWO\n5DgPOMHMxifynmtmh5jZIcBiYHy8bsmP0WFAq0ahBS4G1pjZD83sIOBs4NutyM8pUdwoOCOBb5Mf\ne5nZMjOrUfDL/0RjuKTbJJ0Rz9+SdENswS6OLdW5sbcxMaZp9v4kku6IedRJujaGTQJ2B+ZLmp8o\ns5ekqZLOT9x/jaRL4/llkhbFXsC1KfRfQMKZmKRHYm+pTtI5MWwq0Dnqem8Mq49HSboptq5rJZ2S\nrRBJl8Q0yyVNjmHTCC6hn5J0cQpZkdRZ0t9jWUskHSOpM+HjxPFRxnGSjpS0QNKrkl6QNKiVrPuQ\ncJVgZq+b2bexzMcTz2RCDKuQtFHSH2P4XEnDY89rtaQxMd0ESbNj+L8lZTXAki6X9HJ8b1eneRZO\nnsinX3D/bfs/YBJwczNxVcATievbgDPi+VvAufH8ZsIXtN0I/uzXp7i/Ghgaz3eKxw4x/OBEGb0S\n979F8GF0KPBsInwFwRfMiYQ9bEVo8DwBHJNFr0y+BKd55yTiGmXpDCwHdo7X9U3yqI/HnwPzouy9\nCS4X+jRJezjBe+cOBNfOdcCh2XTMImvmOcXrKcD0eH4g8DZhn4EJwC2JdN2Bing+Cnggnh8PPJKl\nnMOBD4EXCV8ID8zyTLrEZ92T4A3BCL0cgMcJnncbHfItjuETCMamZ9R/BXBITLcxphkD3J54b08D\nlcX+3yjXX7tzc+EUlEYfNLVAVzP7HPhc0tdtHPv/ZWyVVxBarIMJRiYrZvaqpF0l7U4wQhvMbK2k\niwiG4dWYtCswCHguSzbzJe0E1ANXJcInSTo5nu8Z72/JeeAI4H4z+x+wXmGns2Fs7p9nBDDbgpM6\nJD0MHJ2Qsy2MAG4CsODu4V1gYJZ0PYAZkvZJk6mZvSJpAOH5HQ8slnSEBT9ZF0v6aUy6B8ET6VLg\nKzObF8NrCT54GhTcTfRLZD/X4l4ecT5jBMHgNnIiYeOn5Hvbl2CgnALjRsGpA5rbprKBzYcYf9Ak\n/ut43JQ4b7yuSHE/kvoDlwLDzGyDpHuypcvCP6LcuwEPNGYH3GBmd6a4fySwEbgXuBa4RFIVoUI8\nysy+lFSdUpZtkesJlfHtkgYSWt8tEo36Q8BDca5ptIIjxWOAI83sK0nP890z+SZxe/JvoPH9Z7Ju\nWlSTawHXmdnfUujl5BmfU3CeAbZvHD+H4GJb0tGEoYnBkraPLf/j2ph3mvt3BL4APlXwcDk6Efc5\nYUgqGw8QPGaOIxgICE7TfiOpa9Sjr6RdmxPOzBoIk9mnx15Dd0Kv40tJ+xO2OmzkW0kds2RTA5wi\nqYOkXQgVaFOXzTXAzyR1kbQDYeOfmubkaoUaYHzU7wBCz2oV339W3flujuCM1jKVNKKxdydpe8Km\nNW/HfD6JBuFAQi+orZwoqYfCaqaxwAtN4ucCZ8Vng6Q9JPXagnKcHOBGocyxMKh7MnC8wiRxHXAD\n8L6ZrQVmEbr6s2jjcEea+81sWQx/HbiPzSuM6cDTjRPNTe6rI1SC71h0G2xm/4x5LIhDGA/SvFFp\nzOc94H7gfEJrukLSSmAqYdvSpCyvKbF0NDKbMNS1jGBgf2tm7zcpYwlwD8FYvETwXLolQ0cAfyZM\netcSejmnW9im8hlgSJxYHgfcCNwkaQmhJd4ag4CamO8SwgT8o8CTQBdJK4DrovxtZVHMaxlhqG1p\nMtLM5hDe1cJY/izCEJJTBNxLquM4eSOuVjrIzCYXWxYnHd5TcBzHcTJ4T8FxHMfJ4D0Fx3EcJ4Mb\nBcdxHCeDGwXHcRwngxsFx3EcJ4MbBcdxHCeDGwXHcRwnw/8BYCTEiyLNWtQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa301d68>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0.27850877192982459"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Cons On验证集的KS表现.\n",
    "ks_stats(Validation_Cons_On)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEKCAYAAAD9xUlFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nGwXHcRwngxsFx3EcJ8P/Ay5Bl3DZHYmqAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa8abd68>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0.46953550118772142"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Revoloan验证集的KS表现.\n",
    "ks_stats(Validation_Revoloan)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "d:\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:64: RuntimeWarning: divide by zero encountered in log\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAEYCAYAAAAJeGK1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa6024a8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average PSI:0.070026\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>y</th>\n",
       "      <th>0_left</th>\n",
       "      <th>1_left</th>\n",
       "      <th>bad_ratio_left</th>\n",
       "      <th>good_ratio_left</th>\n",
       "      <th>0_right</th>\n",
       "      <th>1_right</th>\n",
       "      <th>bad_ratio_right</th>\n",
       "      <th>good_ratio_right</th>\n",
       "      <th>psi_bad</th>\n",
       "      <th>psi_good</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>quantile</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10%</th>\n",
       "      <td>236.0</td>\n",
       "      <td>59.0</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.088789</td>\n",
       "      <td>556</td>\n",
       "      <td>133</td>\n",
       "      <td>0.191092</td>\n",
       "      <td>0.089764</td>\n",
       "      <td>0.000406</td>\n",
       "      <td>0.000011</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20%</th>\n",
       "      <td>259.0</td>\n",
       "      <td>36.0</td>\n",
       "      <td>0.122034</td>\n",
       "      <td>0.097442</td>\n",
       "      <td>587</td>\n",
       "      <td>102</td>\n",
       "      <td>0.146552</td>\n",
       "      <td>0.094769</td>\n",
       "      <td>0.004489</td>\n",
       "      <td>0.000074</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30%</th>\n",
       "      <td>271.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>0.081356</td>\n",
       "      <td>0.101956</td>\n",
       "      <td>599</td>\n",
       "      <td>90</td>\n",
       "      <td>0.129310</td>\n",
       "      <td>0.096706</td>\n",
       "      <td>0.022221</td>\n",
       "      <td>0.000278</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40%</th>\n",
       "      <td>255.0</td>\n",
       "      <td>41.0</td>\n",
       "      <td>0.138983</td>\n",
       "      <td>0.095937</td>\n",
       "      <td>617</td>\n",
       "      <td>72</td>\n",
       "      <td>0.103448</td>\n",
       "      <td>0.099613</td>\n",
       "      <td>0.010493</td>\n",
       "      <td>0.000138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>263.0</td>\n",
       "      <td>32.0</td>\n",
       "      <td>0.108475</td>\n",
       "      <td>0.098947</td>\n",
       "      <td>626</td>\n",
       "      <td>63</td>\n",
       "      <td>0.090517</td>\n",
       "      <td>0.101066</td>\n",
       "      <td>0.003250</td>\n",
       "      <td>0.000045</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60%</th>\n",
       "      <td>266.0</td>\n",
       "      <td>29.0</td>\n",
       "      <td>0.098305</td>\n",
       "      <td>0.100075</td>\n",
       "      <td>643</td>\n",
       "      <td>46</td>\n",
       "      <td>0.066092</td>\n",
       "      <td>0.103810</td>\n",
       "      <td>0.012790</td>\n",
       "      <td>0.000137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>70%</th>\n",
       "      <td>279.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>0.057627</td>\n",
       "      <td>0.104966</td>\n",
       "      <td>637</td>\n",
       "      <td>52</td>\n",
       "      <td>0.074713</td>\n",
       "      <td>0.102841</td>\n",
       "      <td>0.004436</td>\n",
       "      <td>0.000043</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80%</th>\n",
       "      <td>250.0</td>\n",
       "      <td>45.0</td>\n",
       "      <td>0.152542</td>\n",
       "      <td>0.094056</td>\n",
       "      <td>639</td>\n",
       "      <td>50</td>\n",
       "      <td>0.071839</td>\n",
       "      <td>0.103164</td>\n",
       "      <td>0.060771</td>\n",
       "      <td>0.000842</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90%</th>\n",
       "      <td>295.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.110986</td>\n",
       "      <td>652</td>\n",
       "      <td>37</td>\n",
       "      <td>0.053161</td>\n",
       "      <td>0.105263</td>\n",
       "      <td>inf</td>\n",
       "      <td>0.000303</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100%</th>\n",
       "      <td>284.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.040678</td>\n",
       "      <td>0.106847</td>\n",
       "      <td>638</td>\n",
       "      <td>51</td>\n",
       "      <td>0.073276</td>\n",
       "      <td>0.103003</td>\n",
       "      <td>0.019185</td>\n",
       "      <td>0.000141</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "y         0_left  1_left  bad_ratio_left  good_ratio_left  0_right  1_right  \\\n",
       "quantile                                                                      \n",
       "10%        236.0    59.0        0.200000         0.088789      556      133   \n",
       "20%        259.0    36.0        0.122034         0.097442      587      102   \n",
       "30%        271.0    24.0        0.081356         0.101956      599       90   \n",
       "40%        255.0    41.0        0.138983         0.095937      617       72   \n",
       "50%        263.0    32.0        0.108475         0.098947      626       63   \n",
       "60%        266.0    29.0        0.098305         0.100075      643       46   \n",
       "70%        279.0    17.0        0.057627         0.104966      637       52   \n",
       "80%        250.0    45.0        0.152542         0.094056      639       50   \n",
       "90%        295.0     0.0        0.000000         0.110986      652       37   \n",
       "100%       284.0    12.0        0.040678         0.106847      638       51   \n",
       "\n",
       "y         bad_ratio_right  good_ratio_right   psi_bad  psi_good  \n",
       "quantile                                                         \n",
       "10%              0.191092          0.089764  0.000406  0.000011  \n",
       "20%              0.146552          0.094769  0.004489  0.000074  \n",
       "30%              0.129310          0.096706  0.022221  0.000278  \n",
       "40%              0.103448          0.099613  0.010493  0.000138  \n",
       "50%              0.090517          0.101066  0.003250  0.000045  \n",
       "60%              0.066092          0.103810  0.012790  0.000137  \n",
       "70%              0.074713          0.102841  0.004436  0.000043  \n",
       "80%              0.071839          0.103164  0.060771  0.000842  \n",
       "90%              0.053161          0.105263       inf  0.000303  \n",
       "100%             0.073276          0.103003  0.019185  0.000141  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 验证Cash Off验证集与训练测试集的PSI表现，其中左参数为验证集，右参数为训练测试集.\n",
    "psi_stats_prob(Validation_Cash_Off, Train_Test_Cash_Off)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAEYCAYAAAAJeGK1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\njIEyGAwGgy0xBspgMBgMtuT/Abn037Cy9EkdAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa5ec860>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average PSI:0.180986\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>y</th>\n",
       "      <th>0_left</th>\n",
       "      <th>1_left</th>\n",
       "      <th>bad_ratio_left</th>\n",
       "      <th>good_ratio_left</th>\n",
       "      <th>0_right</th>\n",
       "      <th>1_right</th>\n",
       "      <th>bad_ratio_right</th>\n",
       "      <th>good_ratio_right</th>\n",
       "      <th>psi_bad</th>\n",
       "      <th>psi_good</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>quantile</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10%</th>\n",
       "      <td>141</td>\n",
       "      <td>61</td>\n",
       "      <td>0.267544</td>\n",
       "      <td>0.078333</td>\n",
       "      <td>272</td>\n",
       "      <td>195</td>\n",
       "      <td>0.386139</td>\n",
       "      <td>0.065243</td>\n",
       "      <td>0.043514</td>\n",
       "      <td>0.002393</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20%</th>\n",
       "      <td>166</td>\n",
       "      <td>37</td>\n",
       "      <td>0.162281</td>\n",
       "      <td>0.092222</td>\n",
       "      <td>402</td>\n",
       "      <td>65</td>\n",
       "      <td>0.128713</td>\n",
       "      <td>0.096426</td>\n",
       "      <td>0.007779</td>\n",
       "      <td>0.000187</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30%</th>\n",
       "      <td>190</td>\n",
       "      <td>13</td>\n",
       "      <td>0.057018</td>\n",
       "      <td>0.105556</td>\n",
       "      <td>435</td>\n",
       "      <td>33</td>\n",
       "      <td>0.065347</td>\n",
       "      <td>0.104342</td>\n",
       "      <td>0.001136</td>\n",
       "      <td>0.000014</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40%</th>\n",
       "      <td>190</td>\n",
       "      <td>13</td>\n",
       "      <td>0.057018</td>\n",
       "      <td>0.105556</td>\n",
       "      <td>418</td>\n",
       "      <td>49</td>\n",
       "      <td>0.097030</td>\n",
       "      <td>0.100264</td>\n",
       "      <td>0.021273</td>\n",
       "      <td>0.000272</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>182</td>\n",
       "      <td>21</td>\n",
       "      <td>0.092105</td>\n",
       "      <td>0.101111</td>\n",
       "      <td>427</td>\n",
       "      <td>41</td>\n",
       "      <td>0.081188</td>\n",
       "      <td>0.102423</td>\n",
       "      <td>0.001377</td>\n",
       "      <td>0.000017</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60%</th>\n",
       "      <td>177</td>\n",
       "      <td>25</td>\n",
       "      <td>0.109649</td>\n",
       "      <td>0.098333</td>\n",
       "      <td>402</td>\n",
       "      <td>65</td>\n",
       "      <td>0.128713</td>\n",
       "      <td>0.096426</td>\n",
       "      <td>0.003056</td>\n",
       "      <td>0.000037</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>70%</th>\n",
       "      <td>183</td>\n",
       "      <td>20</td>\n",
       "      <td>0.087719</td>\n",
       "      <td>0.101667</td>\n",
       "      <td>426</td>\n",
       "      <td>41</td>\n",
       "      <td>0.081188</td>\n",
       "      <td>0.102183</td>\n",
       "      <td>0.000505</td>\n",
       "      <td>0.000003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80%</th>\n",
       "      <td>202</td>\n",
       "      <td>1</td>\n",
       "      <td>0.004386</td>\n",
       "      <td>0.112222</td>\n",
       "      <td>466</td>\n",
       "      <td>2</td>\n",
       "      <td>0.003960</td>\n",
       "      <td>0.111777</td>\n",
       "      <td>0.000043</td>\n",
       "      <td>0.000002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90%</th>\n",
       "      <td>196</td>\n",
       "      <td>7</td>\n",
       "      <td>0.030702</td>\n",
       "      <td>0.108889</td>\n",
       "      <td>460</td>\n",
       "      <td>7</td>\n",
       "      <td>0.013861</td>\n",
       "      <td>0.110338</td>\n",
       "      <td>0.013392</td>\n",
       "      <td>0.000019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100%</th>\n",
       "      <td>173</td>\n",
       "      <td>30</td>\n",
       "      <td>0.131579</td>\n",
       "      <td>0.096111</td>\n",
       "      <td>461</td>\n",
       "      <td>7</td>\n",
       "      <td>0.013861</td>\n",
       "      <td>0.110578</td>\n",
       "      <td>0.264923</td>\n",
       "      <td>0.002029</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "y         0_left  1_left  bad_ratio_left  good_ratio_left  0_right  1_right  \\\n",
       "quantile                                                                      \n",
       "10%          141      61        0.267544         0.078333      272      195   \n",
       "20%          166      37        0.162281         0.092222      402       65   \n",
       "30%          190      13        0.057018         0.105556      435       33   \n",
       "40%          190      13        0.057018         0.105556      418       49   \n",
       "50%          182      21        0.092105         0.101111      427       41   \n",
       "60%          177      25        0.109649         0.098333      402       65   \n",
       "70%          183      20        0.087719         0.101667      426       41   \n",
       "80%          202       1        0.004386         0.112222      466        2   \n",
       "90%          196       7        0.030702         0.108889      460        7   \n",
       "100%         173      30        0.131579         0.096111      461        7   \n",
       "\n",
       "y         bad_ratio_right  good_ratio_right   psi_bad  psi_good  \n",
       "quantile                                                         \n",
       "10%              0.386139          0.065243  0.043514  0.002393  \n",
       "20%              0.128713          0.096426  0.007779  0.000187  \n",
       "30%              0.065347          0.104342  0.001136  0.000014  \n",
       "40%              0.097030          0.100264  0.021273  0.000272  \n",
       "50%              0.081188          0.102423  0.001377  0.000017  \n",
       "60%              0.128713          0.096426  0.003056  0.000037  \n",
       "70%              0.081188          0.102183  0.000505  0.000003  \n",
       "80%              0.003960          0.111777  0.000043  0.000002  \n",
       "90%              0.013861          0.110338  0.013392  0.000019  \n",
       "100%             0.013861          0.110578  0.264923  0.002029  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 验证Cons On验证集与训练测试集的PSI表现，其中左参数为验证集，右参数为训练测试集.\n",
    "psi_stats_prob(Validation_Cons_On, Train_Test_Cons_On)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAEYCAYAAAAJeGK1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nsVhcyf8D0Eo7vCv7G6oAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa39e668>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average PSI:0.011768\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>y</th>\n",
       "      <th>0_left</th>\n",
       "      <th>1_left</th>\n",
       "      <th>bad_ratio_left</th>\n",
       "      <th>good_ratio_left</th>\n",
       "      <th>0_right</th>\n",
       "      <th>1_right</th>\n",
       "      <th>bad_ratio_right</th>\n",
       "      <th>good_ratio_right</th>\n",
       "      <th>psi_bad</th>\n",
       "      <th>psi_good</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>quantile</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10%</th>\n",
       "      <td>375</td>\n",
       "      <td>195</td>\n",
       "      <td>0.341506</td>\n",
       "      <td>0.073028</td>\n",
       "      <td>818</td>\n",
       "      <td>513</td>\n",
       "      <td>0.385135</td>\n",
       "      <td>0.068263</td>\n",
       "      <td>0.005245</td>\n",
       "      <td>3.214989e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20%</th>\n",
       "      <td>439</td>\n",
       "      <td>132</td>\n",
       "      <td>0.231173</td>\n",
       "      <td>0.085492</td>\n",
       "      <td>1057</td>\n",
       "      <td>275</td>\n",
       "      <td>0.206456</td>\n",
       "      <td>0.088208</td>\n",
       "      <td>0.002795</td>\n",
       "      <td>8.497827e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30%</th>\n",
       "      <td>488</td>\n",
       "      <td>82</td>\n",
       "      <td>0.143608</td>\n",
       "      <td>0.095034</td>\n",
       "      <td>1160</td>\n",
       "      <td>171</td>\n",
       "      <td>0.128378</td>\n",
       "      <td>0.096804</td>\n",
       "      <td>0.001707</td>\n",
       "      <td>3.265275e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40%</th>\n",
       "      <td>521</td>\n",
       "      <td>50</td>\n",
       "      <td>0.087566</td>\n",
       "      <td>0.101461</td>\n",
       "      <td>1212</td>\n",
       "      <td>120</td>\n",
       "      <td>0.090090</td>\n",
       "      <td>0.101143</td>\n",
       "      <td>0.000072</td>\n",
       "      <td>9.937194e-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>538</td>\n",
       "      <td>33</td>\n",
       "      <td>0.057793</td>\n",
       "      <td>0.104771</td>\n",
       "      <td>1252</td>\n",
       "      <td>79</td>\n",
       "      <td>0.059309</td>\n",
       "      <td>0.104481</td>\n",
       "      <td>0.000039</td>\n",
       "      <td>8.028697e-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60%</th>\n",
       "      <td>548</td>\n",
       "      <td>22</td>\n",
       "      <td>0.038529</td>\n",
       "      <td>0.106719</td>\n",
       "      <td>1277</td>\n",
       "      <td>55</td>\n",
       "      <td>0.041291</td>\n",
       "      <td>0.106568</td>\n",
       "      <td>0.000191</td>\n",
       "      <td>2.136944e-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>70%</th>\n",
       "      <td>559</td>\n",
       "      <td>12</td>\n",
       "      <td>0.021016</td>\n",
       "      <td>0.108861</td>\n",
       "      <td>1291</td>\n",
       "      <td>40</td>\n",
       "      <td>0.030030</td>\n",
       "      <td>0.107736</td>\n",
       "      <td>0.003217</td>\n",
       "      <td>1.168242e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80%</th>\n",
       "      <td>557</td>\n",
       "      <td>13</td>\n",
       "      <td>0.022767</td>\n",
       "      <td>0.108471</td>\n",
       "      <td>1304</td>\n",
       "      <td>28</td>\n",
       "      <td>0.021021</td>\n",
       "      <td>0.108821</td>\n",
       "      <td>0.000139</td>\n",
       "      <td>1.124643e-06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90%</th>\n",
       "      <td>558</td>\n",
       "      <td>13</td>\n",
       "      <td>0.022767</td>\n",
       "      <td>0.108666</td>\n",
       "      <td>1304</td>\n",
       "      <td>27</td>\n",
       "      <td>0.020270</td>\n",
       "      <td>0.108821</td>\n",
       "      <td>0.000290</td>\n",
       "      <td>2.203973e-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100%</th>\n",
       "      <td>552</td>\n",
       "      <td>19</td>\n",
       "      <td>0.033275</td>\n",
       "      <td>0.107498</td>\n",
       "      <td>1308</td>\n",
       "      <td>24</td>\n",
       "      <td>0.018018</td>\n",
       "      <td>0.109155</td>\n",
       "      <td>0.009359</td>\n",
       "      <td>2.534878e-05</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "y         0_left  1_left  bad_ratio_left  good_ratio_left  0_right  1_right  \\\n",
       "quantile                                                                      \n",
       "10%          375     195        0.341506         0.073028      818      513   \n",
       "20%          439     132        0.231173         0.085492     1057      275   \n",
       "30%          488      82        0.143608         0.095034     1160      171   \n",
       "40%          521      50        0.087566         0.101461     1212      120   \n",
       "50%          538      33        0.057793         0.104771     1252       79   \n",
       "60%          548      22        0.038529         0.106719     1277       55   \n",
       "70%          559      12        0.021016         0.108861     1291       40   \n",
       "80%          557      13        0.022767         0.108471     1304       28   \n",
       "90%          558      13        0.022767         0.108666     1304       27   \n",
       "100%         552      19        0.033275         0.107498     1308       24   \n",
       "\n",
       "y         bad_ratio_right  good_ratio_right   psi_bad      psi_good  \n",
       "quantile                                                             \n",
       "10%              0.385135          0.068263  0.005245  3.214989e-04  \n",
       "20%              0.206456          0.088208  0.002795  8.497827e-05  \n",
       "30%              0.128378          0.096804  0.001707  3.265275e-05  \n",
       "40%              0.090090          0.101143  0.000072  9.937194e-07  \n",
       "50%              0.059309          0.104481  0.000039  8.028697e-07  \n",
       "60%              0.041291          0.106568  0.000191  2.136944e-07  \n",
       "70%              0.030030          0.107736  0.003217  1.168242e-05  \n",
       "80%              0.021021          0.108821  0.000139  1.124643e-06  \n",
       "90%              0.020270          0.108821  0.000290  2.203973e-07  \n",
       "100%             0.018018          0.109155  0.009359  2.534878e-05  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 验证Revoloan验证集与训练测试集的PSI表现，其中左参数为验证集，右参数为训练测试集.\n",
    "psi_stats_prob(Validation_Revoloan, Train_Test_Revoloan)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def ks_stats(data, non_computed=None, plot_image=True):\n",
    "    \"\"\"Calculate Kolmogorov-Smirnov value.\n",
    "    \n",
    "    Parameters\n",
    "    ----------\n",
    "    data : pandas.DataFrame\n",
    "        The necessary columns ‘probability’, ‘y’ must be in data.\n",
    "        \n",
    "    non_computed : str or None\n",
    "        The column name of non-computed probability indicators.\n",
    "        \n",
    "    plot_image : bool (default True)\n",
    "        Plot image.\n",
    "        \n",
    "    Returns\n",
    "    -------\n",
    "    ks_max : float\n",
    "        The max of Kolmogorov-Smirnov value.\n",
    "    \"\"\"\n",
    "    \n",
    "    \"\"\"Check columns of data.\"\"\"\n",
    "    check_cols = ['probability', 'y']\n",
    "    for col in check_cols:\n",
    "        if col not in data.columns:\n",
    "            raise ValueError('There is no column %s of data.' % col)\n",
    "            \n",
    "    \"\"\"Drop NaN by column 'probability'.\"\"\"\n",
    "    data = data.loc[data['probability'].notnull(), check_cols]\n",
    "    \n",
    "    \"\"\"Define function for swapping label 'y'.\"\"\"\n",
    "    def swap_label(data):\n",
    "        bad_index = (data == 1)\n",
    "        data[bad_index], data[~bad_index] = 0, 1\n",
    "    \n",
    "    \"\"\"Compute cum bad ratio, cum good ratio and max KS.\"\"\"\n",
    "    data_sorted = data.sort_values(by=['probability'], ascending=False)['y']\n",
    "    total_bad, total_good = data_sorted.sum(), data_sorted.shape[0] - data_sorted.sum()\n",
    "    cum_bad_amount = data_sorted.cumsum()\n",
    "    swap_label(data_sorted)\n",
    "    cum_good_amount = data_sorted.cumsum()\n",
    "    cum_bad_ratio, cum_good_ratio = cum_bad_amount / float(total_bad), cum_good_amount / float(total_good)\n",
    "    ks_max = np.fabs(cum_good_ratio - cum_bad_ratio).max()\n",
    "    \n",
    "    \"\"\"Plot image.\"\"\"\n",
    "    if plot_image == True:\n",
    "        plt.figure()\n",
    "        rate_range = np.arange(0.0, 1.0, 1.0/data_sorted.shape[0])\n",
    "        plt.plot(rate_range, cum_bad_ratio, color='blue', marker='o', \n",
    "                 markersize=0.01, label='Cum Bad Ratio')\n",
    "        plt.plot(rate_range, cum_good_ratio, color='green', marker='s', \n",
    "                 markersize=0.01, label='Cum Good Ratio') \n",
    "        plt.fill_between(rate_range, cum_good_ratio, cum_bad_ratio, \n",
    "                         alpha=0.15, color='red')\n",
    "        plt.grid()\n",
    "        plt.legend(loc='upper left')\n",
    "        plt.xlabel('Cumulative Ratio of Total Sample')\n",
    "        plt.ylabel('Cumulative Ratio of Good & Bad Sample')\n",
    "        plt.ylim([-0.05, 1.05])\n",
    "        plt.show()\n",
    "    return ks_max"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def psi_stats_prob(data_left, data_right,non_computed=None, plot_image=True):\n",
    "    \"\"\"Calculate Average PSI value.\n",
    "    \n",
    "    Parameters\n",
    "    ----------\n",
    "    data_left: pandas.DataFrame\n",
    "        The necessary columns ‘probability’, ‘y’, ‘cus_num’ must be in data.\n",
    "        \n",
    "    data_right: pandas.DataFrame\n",
    "        The necessary columns ‘probability’, ‘y’, ‘cus_num’ must be in data.\n",
    "        \n",
    "    plot_image : bool (default True)\n",
    "        Plot image.\n",
    "        \n",
    "    Returns\n",
    "    -------\n",
    "    psi_table : pandas.DataFrame\n",
    "        The PSI value of probability interval.\n",
    "    \"\"\"\n",
    "    \n",
    "    \"\"\"Check columns of data.\"\"\"\n",
    "    check_cols = ['probability', 'y', 'cus_num']\n",
    "    if non_computed != None and type(non_computed) == str:\n",
    "        check_cols += [non_computed]\n",
    "        data_left = data_left[~data_left[non_computed] == True].copy()\n",
    "        data_right = data_right[~data_right[non_computed] == True].copy()\n",
    "    elif non_computed == None:\n",
    "        data_left = data_left.copy()\n",
    "        data_right = data_right.copy()\n",
    "    else:\n",
    "        raise ValueError('non_computed must be a str.')\n",
    "    for col in check_cols:\n",
    "        if col not in data_left.columns or col not in data_right.columns:\n",
    "            raise ValueError('There is no column %s of data' % col) \n",
    "            \n",
    "    \"\"\"Drop NaN and sort values by column 'probability'.\"\"\"\n",
    "    data_left = data_left.loc[data_left['probability'].notnull(), check_cols]\n",
    "    data_right = data_right.loc[data_right['probability'].notnull(), check_cols]\n",
    "    data_left.sort_values(by=['probability'], inplace=True, ascending=False)\n",
    "    data_right.sort_values(by=['probability'], inplace=True, ascending=False)\n",
    "    data_left.reset_index(drop=True, inplace=True) \n",
    "    data_right.reset_index(drop=True, inplace=True) \n",
    "    \n",
    "    \"\"\"Discrete probability value.\"\"\"\n",
    "    break_points_left = [int(data_left.shape[0]*i) for i in np.arange(0.0, 1.1, 0.1)]\n",
    "    break_points_right = [int(data_right.shape[0]*i) for i in np.arange(0.0, 1.1, 0.1)]\n",
    "    quantile_str = [str(i)+'%' for i in np.arange(10, 110, 10)]\n",
    "    for i in range(len(break_points_left)):\n",
    "        if i == len(break_points_left) - 1:\n",
    "            break\n",
    "        else:\n",
    "            data_left.loc[break_points_left[i]:break_points_left[i+1], 'quantile'] = quantile_str[i]  \n",
    "            data_right.loc[break_points_right[i]:break_points_right[i+1], 'quantile'] = quantile_str[i]       \n",
    "    \n",
    "    \"\"\"Count psi of bad & good sample.\"\"\"\n",
    "    count_left = data_left.groupby(['quantile', 'y']).count()['cus_num'].unstack().fillna(value=0.0)\n",
    "    count_right = data_right.groupby(['quantile', 'y']).count()['cus_num'].unstack().fillna(value=0.0)\n",
    "    count_left['bad_ratio'] = count_left[1]/count_left[1].sum()\n",
    "    count_right['bad_ratio'] = count_right[1]/count_right[1].sum()\n",
    "    count_left['good_ratio'] = count_left[0]/count_left[0].sum()\n",
    "    count_right['good_ratio'] = count_right[0]/count_right[0].sum()\n",
    "    count_final = pd.merge(count_left, count_right, left_index=True, \n",
    "                           right_index=True, suffixes=['_left', '_right'])\n",
    "    count_final['psi_bad'] = (count_left['bad_ratio'] - count_right['bad_ratio']) *\\\n",
    "                             np.log(count_left['bad_ratio'] / count_right['bad_ratio'])\n",
    "    count_final['psi_good'] = (count_left['good_ratio'] - count_right['good_ratio']) * \\\n",
    "                             np.log(count_left['good_ratio'] / count_right['good_ratio'])\n",
    "    count_final = count_final.reindex(quantile_str)\n",
    "    average_psi = (count_final['psi_bad'].replace([np.inf, np.nan], 0.0).sum() + \\\n",
    "                   count_final['psi_good'].replace([np.inf, np.nan], 0.0).sum())/2\n",
    "    \n",
    "    \"\"\"Plot image\"\"\"   \n",
    "    if plot_image == True:\n",
    "        plot_range = ['bad_ratio_left', 'good_ratio_left', \n",
    "                      'bad_ratio_right', 'good_ratio_right']\n",
    "        plot_label = ['Bad Ratio of Out Sample', 'Good Ratio of Out Sample', \n",
    "                      'Bad Ratio of Train Set', 'Good Ratio of Train Set']\n",
    "        color = ['blue', 'red', 'green', 'cyan']\n",
    "        marker = ['s', 'x', 'o', 'v']\n",
    "        plt.figure()\n",
    "        for p, l, c, m in zip(plot_range, plot_label, color, marker):\n",
    "            value = count_final[p].values\n",
    "            prob_range = range(len(count_final.index))\n",
    "            quantile_label = quantile_str\n",
    "            plt.plot(prob_range, value, color=c, marker=m, \n",
    "                     markersize=1, label=l)\n",
    "        plt.grid()\n",
    "        plt.legend(loc='best')\n",
    "        plt.xticks(prob_range, quantile_label, rotation=45)\n",
    "        plt.title('PSI of Score Card')\n",
    "        plt.ylabel('Ratio')\n",
    "        plt.tight_layout()\n",
    "        plt.show()     \n",
    "    print ('Average PSI:%f' % average_psi )\n",
    "    return count_final"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}