{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "          member_id   funded_amnt          term      int_rate    emp_length  \\\n",
      "count  42535.000000  42535.000000  42535.000000  42535.000000  42535.000000   \n",
      "mean   21268.000000  10821.585753     42.207218      0.121650      2.612789   \n",
      "std    12278.941187   7146.914675     10.509342      0.037079     15.223510   \n",
      "min        1.000000    500.000000     36.000000      0.054200    -88.000000   \n",
      "25%    10634.500000   5000.000000     36.000000      0.096300      2.000000   \n",
      "50%    21268.000000   9600.000000     36.000000      0.119900      4.000000   \n",
      "75%    31901.500000  15000.000000     60.000000      0.147200      9.000000   \n",
      "max    42535.000000  35000.000000     60.000000      0.245900     10.000000   \n",
      "\n",
      "             income   loan_status     grade_num  home_ownership_num  \n",
      "count  4.253500e+04  42535.000000  42535.000000        42535.000000  \n",
      "mean   6.913005e+04      0.848807      5.329141            1.600623  \n",
      "std    6.409685e+04      0.358241      1.438428            0.626047  \n",
      "min   -8.800000e+01      0.000000      1.000000            1.000000  \n",
      "25%    4.000000e+04      1.000000      4.000000            1.000000  \n",
      "50%    5.900000e+04      1.000000      6.000000            2.000000  \n",
      "75%    8.250000e+04      1.000000      6.000000            2.000000  \n",
      "max    6.000000e+06      1.000000      7.000000            3.000000  \n",
      "level_1         25%      50%      75%    count      max          mean    min  \\\n",
      "loan_status                                                                    \n",
      "0            5200.0  10000.0  15650.0   6431.0  35000.0  11492.190173  500.0   \n",
      "1            5000.0   9350.0  15000.0  36104.0  35000.0  10702.134805  500.0   \n",
      "\n",
      "level_1              std  \n",
      "loan_status               \n",
      "0            7655.264857  \n",
      "1            7045.939943  \n",
      "level_1           25%      50%      75%    count        max          mean  \\\n",
      "loan_status                                                                 \n",
      "0            37000.00  53000.0  75000.0   6431.0  1250000.0  63366.870316   \n",
      "1            41155.75  60000.0  84000.0  36104.0  6000000.0  70156.608354   \n",
      "\n",
      "level_1         min           std  \n",
      "loan_status                        \n",
      "0            2000.0  49684.103617  \n",
      "1             -88.0  66284.371098  \n",
      "level_1      25%  50%  75%    count  max      mean  min       std\n",
      "loan_status                                                      \n",
      "0            4.0  5.0  6.0   6431.0  7.0  4.643135  1.0  1.502201\n",
      "1            5.0  6.0  7.0  36104.0  7.0  5.451335  1.0  1.391751\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0xa875d50>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np \n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.model_selection import train_test_split\n",
    "import seaborn as sns\n",
    "\n",
    "\n",
    "#Read CSV file \t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\n",
    "df = pd.read_csv(\"C:/Users/ERice.ERP/Desktop/Building Knowledge/Python_Projects/Loan.csv\")\n",
    "                 \n",
    "\n",
    "#Look for missing data\n",
    "sum(df.isnull().values.ravel())\n",
    "\n",
    "\n",
    "#Creating numeric grade column\n",
    "\n",
    "grade_num = []\n",
    "\n",
    "for row in df['grade']:\n",
    "    if row == 'A':\n",
    "        grade_num.append(7)\n",
    "    elif row == 'B':\n",
    "        grade_num.append(6)\n",
    "    elif row == 'C':\n",
    "        grade_num.append(5)\n",
    "    elif row == 'D':\n",
    "        grade_num.append(4)\n",
    "    elif row == 'E':\n",
    "        grade_num.append(3)\n",
    "    elif row == 'F':\n",
    "        grade_num.append(2)\n",
    "    else:\n",
    "        grade_num.append(1)\n",
    "    \n",
    "df['grade_num'] = grade_num\n",
    "\n",
    "\n",
    "#Creating numeric home_ownership column\n",
    "\n",
    "home_ownership_num = []\n",
    "\n",
    "for row in df['home_ownership']:\n",
    "    if row == 'OWN':\n",
    "        home_ownership_num.append(3)\n",
    "    elif row == 'MORTGAGE':\n",
    "        home_ownership_num.append(2)\n",
    "    elif row == 'RENT':\n",
    "        home_ownership_num.append(1)\n",
    "    elif row == 'NONE':\n",
    "        home_ownership_num.append(-88)\n",
    "    else:\n",
    "        home_ownership_num.append(-88)\n",
    "    \n",
    "df['home_ownership_num'] = home_ownership_num\n",
    "\n",
    "  \n",
    "# Mean imputation for missing home ownership \n",
    "from sklearn.preprocessing import Imputer  # imputer library #class\n",
    "imputer = Imputer(missing_values = -88, strategy = 'mean', axis = 0) # receongnize what a missing value is , taking the mean of the column\n",
    "imputer = imputer.fit(df.iloc[:,10:12]) #Upper ban is not included, so we get all rows,and last to columns\n",
    "df.iloc[:,10:12] = imputer.transform(df.iloc[:,10:12])  #method\n",
    "\n",
    "       \n",
    "#Making loan status numeric\n",
    "\n",
    "\n",
    "def func(row):\n",
    "    if row['loan_status'] == 'Fully Paid':\n",
    "        return 1\n",
    "    else:\n",
    "        return 0\n",
    "\n",
    "df['loan_status'] = df.apply(func, axis=1)\n",
    "\n",
    "   \n",
    "# Descriptive Stats \n",
    "\n",
    "summary_stats = df.describe()\n",
    "print(summary_stats)\n",
    "\n",
    "summary_funded_amnt_by_status = df.groupby('loan_status').describe().reset_index().pivot(index='loan_status', values='funded_amnt', columns='level_1')\n",
    "print(summary_funded_amnt_by_status)\n",
    "\n",
    "status_by_income = df.groupby('loan_status').describe().reset_index().pivot(index='loan_status', values='income', columns='level_1')\n",
    "print(status_by_income)\n",
    "\n",
    "status_by_grade = df.groupby('loan_status').describe().reset_index().pivot(index='loan_status', values='grade_num', columns='level_1')\n",
    "print(status_by_grade)\n",
    "\n",
    "#Graphic Descriptives\n",
    "\n",
    "\n",
    "#Bar Chart for Interest Rate by Credit Grade\n",
    "means = df['int_rate'].groupby(df['grade']).mean()\n",
    "my_plot = means.plot(kind='bar',legend=None,title=\"Interest Rate by Credit Grade\")\n",
    "my_plot.set_xlabel(\"Credit Grade\")\n",
    "my_plot.set_ylabel(\"Mean Interest Rate\")\n",
    "\n",
    "#Bar Chart for Income by home ownership status\n",
    "income_median_by_home = df['income'].groupby(df['home_ownership']).median()\n",
    "my_plot = income_median_by_home.plot(kind='bar',legend=None,title=\"Median Income by Home Ownership Status\")\n",
    "my_plot.set_xlabel(\"Status\")\n",
    "my_plot.set_ylabel(\"Median Income\")\n"
   ]
  }
 ],
 "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
}