main.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "50295\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">╭──────────────────────────────────────────────── skimpy summary ─────────────────────────────────────────────────╮\n",
       "│ <span style=\"font-style: italic\">         Data Summary         </span> <span style=\"font-style: italic\">      Data Types       </span>                                                          │\n",
       "│ ┏━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┓ ┏━━━━━━━━━━━━━┳━━━━━━━┓                                                          │\n",
       "│ ┃<span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\"> dataframe         </span>┃<span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\"> Values </span>┃ ┃<span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\"> Column Type </span>┃<span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\"> Count </span>┃                                                          │\n",
       "│ ┡━━━━━━━━━━━━━━━━━━━╇━━━━━━━━┩ ┡━━━━━━━━━━━━━╇━━━━━━━┩                                                          │\n",
       "│ │ Number of rows    │ 58645  │ │ int32       │ 6     │                                                          │\n",
       "│ │ Number of columns │ 13     │ │ string      │ 4     │                                                          │\n",
       "│ └───────────────────┴────────┘ │ float64     │ 3     │                                                          │\n",
       "│                                └─────────────┴───────┘                                                          │\n",
       "│ <span style=\"font-style: italic\">                                                    number                                                    </span>  │\n",
       "│ ┏━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━┳━━━━━━━┳━━━━━━━━━┳━━━━━━━━━┳━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━┳━━━━━━━━┓  │\n",
       "│ ┃<span style=\"font-weight: bold\"> column_name            </span>┃<span style=\"font-weight: bold\"> NA  </span>┃<span style=\"font-weight: bold\"> NA %  </span>┃<span style=\"font-weight: bold\"> mean    </span>┃<span style=\"font-weight: bold\"> sd      </span>┃<span style=\"font-weight: bold\"> p0   </span>┃<span style=\"font-weight: bold\"> p25   </span>┃<span style=\"font-weight: bold\"> p50   </span>┃<span style=\"font-weight: bold\"> p75   </span>┃<span style=\"font-weight: bold\"> p100    </span>┃<span style=\"font-weight: bold\"> hist   </span>┃  │\n",
       "│ ┡━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━╇━━━━━━━╇━━━━━━━━━╇━━━━━━━━━╇━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━╇━━━━━━━━┩  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">id                    </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  29320</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  16930</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">14660</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">29320</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">43980</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  58640</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">▇▇▇▇▇▇</span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">person_age            </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  27.55</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  6.033</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  20</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   23</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   26</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   30</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    123</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">  ▇▁  </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">person_income         </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  64050</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  37930</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">4200</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">42000</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">58000</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">75600</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">1900000</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">  ▇   </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">person_emp_length     </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  4.701</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   3.96</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    2</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    4</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    7</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    123</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">  ▇   </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_amnt             </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   9218</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   5564</span> │ <span style=\"color: #008080; text-decoration-color: #008080\"> 500</span> │ <span style=\"color: #008080; text-decoration-color: #008080\"> 5000</span> │ <span style=\"color: #008080; text-decoration-color: #008080\"> 8000</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">12000</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  35000</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">▇▇▅▁▁ </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_int_rate         </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  10.68</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  3.035</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">5.42</span> │ <span style=\"color: #008080; text-decoration-color: #008080\"> 7.88</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">10.75</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">12.99</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  23.22</span> │ <span style=\"color: #008000; text-decoration-color: #008000\"> ▇▇▇▃ </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_percent_income   </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\"> 0.1592</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">0.09169</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\"> 0.09</span> │ <span style=\"color: #008080; text-decoration-color: #008080\"> 0.14</span> │ <span style=\"color: #008080; text-decoration-color: #008080\"> 0.21</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   0.83</span> │ <span style=\"color: #008000; text-decoration-color: #008000\"> ▇▇▂  </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">cb_person_cred_hist_le</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  5.814</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  4.029</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   2</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    3</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    4</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    8</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">     30</span> │ <span style=\"color: #008000; text-decoration-color: #008000\"> ▇▃▁  </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">ngth                  </span> │     │       │         │         │      │       │       │       │         │        │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_status           </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\"> 0.1424</span> │ <span style=\"color: #008080; text-decoration-color: #008080\"> 0.3494</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">      1</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">▇    ▁</span> │  │\n",
       "│ └────────────────────────┴─────┴───────┴─────────┴─────────┴──────┴───────┴───────┴───────┴─────────┴────────┘  │\n",
       "│ <span style=\"font-style: italic\">                                                    string                                                    </span>  │\n",
       "│ ┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━┓  │\n",
       "│ ┃<span style=\"font-weight: bold\"> column_name                           </span>┃<span style=\"font-weight: bold\"> NA    </span>┃<span style=\"font-weight: bold\"> NA %      </span>┃<span style=\"font-weight: bold\"> words per row            </span>┃<span style=\"font-weight: bold\"> total words         </span>┃  │\n",
       "│ ┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━┩  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">person_home_ownership                </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">        0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                       1</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">              58645</span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_intent                          </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">        0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                       1</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">              58645</span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_grade                           </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">        0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                       1</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">              58645</span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">cb_person_default_on_file            </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">        0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                       1</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">              58645</span> │  │\n",
       "│ └───────────────────────────────────────┴───────┴───────────┴──────────────────────────┴─────────────────────┘  │\n",
       "╰────────────────────────────────────────────────────── End ──────────────────────────────────────────────────────╯\n",
       "</pre>\n"
      ],
      "text/plain": [
       "╭──────────────────────────────────────────────── skimpy summary ─────────────────────────────────────────────────╮\n",
       "│ \u001b[3m         Data Summary         \u001b[0m \u001b[3m      Data Types       \u001b[0m                                                          │\n",
       "│ ┏━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┓ ┏━━━━━━━━━━━━━┳━━━━━━━┓                                                          │\n",
       "│ ┃\u001b[1;36m \u001b[0m\u001b[1;36mdataframe        \u001b[0m\u001b[1;36m \u001b[0m┃\u001b[1;36m \u001b[0m\u001b[1;36mValues\u001b[0m\u001b[1;36m \u001b[0m┃ ┃\u001b[1;36m \u001b[0m\u001b[1;36mColumn Type\u001b[0m\u001b[1;36m \u001b[0m┃\u001b[1;36m \u001b[0m\u001b[1;36mCount\u001b[0m\u001b[1;36m \u001b[0m┃                                                          │\n",
       "│ ┡━━━━━━━━━━━━━━━━━━━╇━━━━━━━━┩ ┡━━━━━━━━━━━━━╇━━━━━━━┩                                                          │\n",
       "│ │ Number of rows    │ 58645  │ │ int32       │ 6     │                                                          │\n",
       "│ │ Number of columns │ 13     │ │ string      │ 4     │                                                          │\n",
       "│ └───────────────────┴────────┘ │ float64     │ 3     │                                                          │\n",
       "│                                └─────────────┴───────┘                                                          │\n",
       "│ \u001b[3m                                                    number                                                    \u001b[0m  │\n",
       "│ ┏━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━┳━━━━━━━┳━━━━━━━━━┳━━━━━━━━━┳━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━┳━━━━━━━━┓  │\n",
       "│ ┃\u001b[1m \u001b[0m\u001b[1mcolumn_name           \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mNA \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mNA % \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mmean   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1msd     \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mp0  \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mp25  \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mp50  \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mp75  \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mp100   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mhist  \u001b[0m\u001b[1m \u001b[0m┃  │\n",
       "│ ┡━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━╇━━━━━━━╇━━━━━━━━━╇━━━━━━━━━╇━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━╇━━━━━━━━┩  │\n",
       "│ │ \u001b[38;5;141mid                    \u001b[0m │ \u001b[36m  0\u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m  29320\u001b[0m │ \u001b[36m  16930\u001b[0m │ \u001b[36m   0\u001b[0m │ \u001b[36m14660\u001b[0m │ \u001b[36m29320\u001b[0m │ \u001b[36m43980\u001b[0m │ \u001b[36m  58640\u001b[0m │ \u001b[32m▇▇▇▇▇▇\u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mperson_age            \u001b[0m │ \u001b[36m  0\u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m  27.55\u001b[0m │ \u001b[36m  6.033\u001b[0m │ \u001b[36m  20\u001b[0m │ \u001b[36m   23\u001b[0m │ \u001b[36m   26\u001b[0m │ \u001b[36m   30\u001b[0m │ \u001b[36m    123\u001b[0m │ \u001b[32m  ▇▁  \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mperson_income         \u001b[0m │ \u001b[36m  0\u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m  64050\u001b[0m │ \u001b[36m  37930\u001b[0m │ \u001b[36m4200\u001b[0m │ \u001b[36m42000\u001b[0m │ \u001b[36m58000\u001b[0m │ \u001b[36m75600\u001b[0m │ \u001b[36m1900000\u001b[0m │ \u001b[32m  ▇   \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mperson_emp_length     \u001b[0m │ \u001b[36m  0\u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m  4.701\u001b[0m │ \u001b[36m   3.96\u001b[0m │ \u001b[36m   0\u001b[0m │ \u001b[36m    2\u001b[0m │ \u001b[36m    4\u001b[0m │ \u001b[36m    7\u001b[0m │ \u001b[36m    123\u001b[0m │ \u001b[32m  ▇   \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_amnt             \u001b[0m │ \u001b[36m  0\u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m   9218\u001b[0m │ \u001b[36m   5564\u001b[0m │ \u001b[36m 500\u001b[0m │ \u001b[36m 5000\u001b[0m │ \u001b[36m 8000\u001b[0m │ \u001b[36m12000\u001b[0m │ \u001b[36m  35000\u001b[0m │ \u001b[32m▇▇▅▁▁ \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_int_rate         \u001b[0m │ \u001b[36m  0\u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m  10.68\u001b[0m │ \u001b[36m  3.035\u001b[0m │ \u001b[36m5.42\u001b[0m │ \u001b[36m 7.88\u001b[0m │ \u001b[36m10.75\u001b[0m │ \u001b[36m12.99\u001b[0m │ \u001b[36m  23.22\u001b[0m │ \u001b[32m ▇▇▇▃ \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_percent_income   \u001b[0m │ \u001b[36m  0\u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m 0.1592\u001b[0m │ \u001b[36m0.09169\u001b[0m │ \u001b[36m   0\u001b[0m │ \u001b[36m 0.09\u001b[0m │ \u001b[36m 0.14\u001b[0m │ \u001b[36m 0.21\u001b[0m │ \u001b[36m   0.83\u001b[0m │ \u001b[32m ▇▇▂  \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mcb_person_cred_hist_le\u001b[0m │ \u001b[36m  0\u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m  5.814\u001b[0m │ \u001b[36m  4.029\u001b[0m │ \u001b[36m   2\u001b[0m │ \u001b[36m    3\u001b[0m │ \u001b[36m    4\u001b[0m │ \u001b[36m    8\u001b[0m │ \u001b[36m     30\u001b[0m │ \u001b[32m ▇▃▁  \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mngth                  \u001b[0m │     │       │         │         │      │       │       │       │         │        │  │\n",
       "│ │ \u001b[38;5;141mloan_status           \u001b[0m │ \u001b[36m  0\u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m 0.1424\u001b[0m │ \u001b[36m 0.3494\u001b[0m │ \u001b[36m   0\u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m      1\u001b[0m │ \u001b[32m▇    ▁\u001b[0m │  │\n",
       "│ └────────────────────────┴─────┴───────┴─────────┴─────────┴──────┴───────┴───────┴───────┴─────────┴────────┘  │\n",
       "│ \u001b[3m                                                    string                                                    \u001b[0m  │\n",
       "│ ┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━┓  │\n",
       "│ ┃\u001b[1m \u001b[0m\u001b[1mcolumn_name                          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mNA   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mNA %     \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mwords per row           \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mtotal words        \u001b[0m\u001b[1m \u001b[0m┃  │\n",
       "│ ┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━┩  │\n",
       "│ │ \u001b[38;5;141mperson_home_ownership                \u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m        0\u001b[0m │ \u001b[36m                       1\u001b[0m │ \u001b[36m              58645\u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_intent                          \u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m        0\u001b[0m │ \u001b[36m                       1\u001b[0m │ \u001b[36m              58645\u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_grade                           \u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m        0\u001b[0m │ \u001b[36m                       1\u001b[0m │ \u001b[36m              58645\u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mcb_person_default_on_file            \u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m        0\u001b[0m │ \u001b[36m                       1\u001b[0m │ \u001b[36m              58645\u001b[0m │  │\n",
       "│ └───────────────────────────────────────┴───────┴───────────┴──────────────────────────┴─────────────────────┘  │\n",
       "╰────────────────────────────────────────────────────── End ──────────────────────────────────────────────────────╯\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from skimpy import skim\n",
    "\n",
    "train = pd.read_csv('train.csv')\n",
    "\n",
    "test = pd.read_csv('test.csv')\n",
    "count0 = train[train['loan_status'] == 0].shape[0]\n",
    "count1 = train[train['loan_status'] == 1].shape[0]\n",
    "\n",
    "print(count0)\n",
    "\n",
    "skim(train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = train.drop('id', axis=1)\n",
    "\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "scaler = StandardScaler()\n",
    "\n",
    "df[['person_age', 'person_income', 'person_emp_length', 'loan_amnt','loan_int_rate','loan_percent_income','cb_person_cred_hist_length']]= scaler.fit_transform(df[['person_age', 'person_income', 'person_emp_length', 'loan_amnt','loan_int_rate','loan_percent_income','cb_person_cred_hist_length']])\n",
    "\n",
    "skim(df)\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">╭──────────────────────────────────────────────── skimpy summary ─────────────────────────────────────────────────╮\n",
       "│ <span style=\"font-style: italic\">         Data Summary         </span> <span style=\"font-style: italic\">      Data Types       </span>                                                          │\n",
       "│ ┏━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┓ ┏━━━━━━━━━━━━━┳━━━━━━━┓                                                          │\n",
       "│ ┃<span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\"> dataframe         </span>┃<span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\"> Values </span>┃ ┃<span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\"> Column Type </span>┃<span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\"> Count </span>┃                                                          │\n",
       "│ ┡━━━━━━━━━━━━━━━━━━━╇━━━━━━━━┩ ┡━━━━━━━━━━━━━╇━━━━━━━┩                                                          │\n",
       "│ │ Number of rows    │ 58645  │ │ bool        │ 19    │                                                          │\n",
       "│ │ Number of columns │ 26     │ │ int32       │ 4     │                                                          │\n",
       "│ └───────────────────┴────────┘ │ float64     │ 3     │                                                          │\n",
       "│                                └─────────────┴───────┘                                                          │\n",
       "│ <span style=\"font-style: italic\">                                                    number                                                    </span>  │\n",
       "│ ┏━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━┳━━━━━━━┳━━━━━━━━━┳━━━━━━━━━┳━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━┳━━━━━━━━┓  │\n",
       "│ ┃<span style=\"font-weight: bold\"> column_name            </span>┃<span style=\"font-weight: bold\"> NA  </span>┃<span style=\"font-weight: bold\"> NA %  </span>┃<span style=\"font-weight: bold\"> mean    </span>┃<span style=\"font-weight: bold\"> sd      </span>┃<span style=\"font-weight: bold\"> p0   </span>┃<span style=\"font-weight: bold\"> p25   </span>┃<span style=\"font-weight: bold\"> p50   </span>┃<span style=\"font-weight: bold\"> p75   </span>┃<span style=\"font-weight: bold\"> p100    </span>┃<span style=\"font-weight: bold\"> hist   </span>┃  │\n",
       "│ ┡━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━╇━━━━━━━╇━━━━━━━━━╇━━━━━━━━━╇━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━╇━━━━━━━━┩  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">person_age            </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  27.55</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  6.033</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  20</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   23</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   26</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   30</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    123</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">  ▇▁  </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">person_income         </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  64050</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  37930</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">4200</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">42000</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">58000</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">75600</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">1900000</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">  ▇   </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">person_emp_length     </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  4.701</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   3.96</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    2</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    4</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    7</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    123</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">  ▇   </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_amnt             </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   9218</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   5564</span> │ <span style=\"color: #008080; text-decoration-color: #008080\"> 500</span> │ <span style=\"color: #008080; text-decoration-color: #008080\"> 5000</span> │ <span style=\"color: #008080; text-decoration-color: #008080\"> 8000</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">12000</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  35000</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">▇▇▅▁▁ </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_int_rate         </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  10.68</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  3.035</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">5.42</span> │ <span style=\"color: #008080; text-decoration-color: #008080\"> 7.88</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">10.75</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">12.99</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  23.22</span> │ <span style=\"color: #008000; text-decoration-color: #008000\"> ▇▇▇▃ </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_percent_income   </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\"> 0.1592</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">0.09169</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\"> 0.09</span> │ <span style=\"color: #008080; text-decoration-color: #008080\"> 0.14</span> │ <span style=\"color: #008080; text-decoration-color: #008080\"> 0.21</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   0.83</span> │ <span style=\"color: #008000; text-decoration-color: #008000\"> ▇▇▂  </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">cb_person_cred_hist_le</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    0</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  5.814</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">  4.029</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">   2</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    3</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    4</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">    8</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">     30</span> │ <span style=\"color: #008000; text-decoration-color: #008000\"> ▇▃▁  </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">ngth                  </span> │     │       │         │         │      │       │       │       │         │        │  │\n",
       "│ └────────────────────────┴─────┴───────┴─────────┴─────────┴──────┴───────┴───────┴───────┴─────────┴────────┘  │\n",
       "│ <span style=\"font-style: italic\">                                                     bool                                                     </span>  │\n",
       "│ ┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━┓  │\n",
       "│ ┃<span style=\"font-weight: bold\"> column_name                                     </span>┃<span style=\"font-weight: bold\"> true           </span>┃<span style=\"font-weight: bold\"> true rate              </span>┃<span style=\"font-weight: bold\"> hist           </span>┃  │\n",
       "│ ┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━┩  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">person_home_ownership_MORTGAGE                 </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">         24824</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                  0.42</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">    ▇    ▆    </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">person_home_ownership_OTHER                    </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">            89</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                0.0015</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">      ▇       </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">person_home_ownership_OWN                      </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">          3138</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                 0.054</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">      ▇       </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">person_home_ownership_RENT                     </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">         30594</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                  0.52</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">    ▇    ▇    </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_intent_DEBTCONSOLIDATION                  </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">          9133</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                  0.16</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">    ▇    ▁    </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_intent_EDUCATION                          </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">         12271</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                  0.21</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">    ▇    ▂    </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_intent_HOMEIMPROVEMENT                    </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">          6280</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                  0.11</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">    ▇    ▁    </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_intent_MEDICAL                            </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">         10934</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                  0.19</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">    ▇    ▂    </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_intent_PERSONAL                           </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">         10016</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                  0.17</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">    ▇    ▂    </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_intent_VENTURE                            </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">         10011</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                  0.17</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">    ▇    ▂    </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_grade_A                                   </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">         20984</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                  0.36</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">    ▇    ▅    </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_grade_B                                   </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">         20400</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                  0.35</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">    ▇    ▅    </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_grade_C                                   </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">         11036</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                  0.19</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">    ▇    ▂    </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_grade_D                                   </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">          5034</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                 0.086</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">    ▇    ▁    </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_grade_E                                   </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">          1009</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                 0.017</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">      ▇       </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_grade_F                                   </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">           149</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                0.0025</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">      ▇       </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">loan_grade_G                                   </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">            33</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">               0.00056</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">      ▇       </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">cb_person_default_on_file_N                    </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">         49943</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                  0.85</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">    ▁    ▇    </span> │  │\n",
       "│ │ <span style=\"color: #af87ff; text-decoration-color: #af87ff\">cb_person_default_on_file_Y                    </span> │ <span style=\"color: #008080; text-decoration-color: #008080\">          8702</span> │ <span style=\"color: #008080; text-decoration-color: #008080\">                  0.15</span> │ <span style=\"color: #008000; text-decoration-color: #008000\">    ▇    ▁    </span> │  │\n",
       "│ └─────────────────────────────────────────────────┴────────────────┴────────────────────────┴────────────────┘  │\n",
       "╰────────────────────────────────────────────────────── End ──────────────────────────────────────────────────────╯\n",
       "</pre>\n"
      ],
      "text/plain": [
       "╭──────────────────────────────────────────────── skimpy summary ─────────────────────────────────────────────────╮\n",
       "│ \u001b[3m         Data Summary         \u001b[0m \u001b[3m      Data Types       \u001b[0m                                                          │\n",
       "│ ┏━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┓ ┏━━━━━━━━━━━━━┳━━━━━━━┓                                                          │\n",
       "│ ┃\u001b[1;36m \u001b[0m\u001b[1;36mdataframe        \u001b[0m\u001b[1;36m \u001b[0m┃\u001b[1;36m \u001b[0m\u001b[1;36mValues\u001b[0m\u001b[1;36m \u001b[0m┃ ┃\u001b[1;36m \u001b[0m\u001b[1;36mColumn Type\u001b[0m\u001b[1;36m \u001b[0m┃\u001b[1;36m \u001b[0m\u001b[1;36mCount\u001b[0m\u001b[1;36m \u001b[0m┃                                                          │\n",
       "│ ┡━━━━━━━━━━━━━━━━━━━╇━━━━━━━━┩ ┡━━━━━━━━━━━━━╇━━━━━━━┩                                                          │\n",
       "│ │ Number of rows    │ 58645  │ │ bool        │ 19    │                                                          │\n",
       "│ │ Number of columns │ 26     │ │ int32       │ 4     │                                                          │\n",
       "│ └───────────────────┴────────┘ │ float64     │ 3     │                                                          │\n",
       "│                                └─────────────┴───────┘                                                          │\n",
       "│ \u001b[3m                                                    number                                                    \u001b[0m  │\n",
       "│ ┏━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━┳━━━━━━━┳━━━━━━━━━┳━━━━━━━━━┳━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━┳━━━━━━━━┓  │\n",
       "│ ┃\u001b[1m \u001b[0m\u001b[1mcolumn_name           \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mNA \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mNA % \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mmean   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1msd     \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mp0  \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mp25  \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mp50  \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mp75  \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mp100   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mhist  \u001b[0m\u001b[1m \u001b[0m┃  │\n",
       "│ ┡━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━╇━━━━━━━╇━━━━━━━━━╇━━━━━━━━━╇━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━╇━━━━━━━━┩  │\n",
       "│ │ \u001b[38;5;141mperson_age            \u001b[0m │ \u001b[36m  0\u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m  27.55\u001b[0m │ \u001b[36m  6.033\u001b[0m │ \u001b[36m  20\u001b[0m │ \u001b[36m   23\u001b[0m │ \u001b[36m   26\u001b[0m │ \u001b[36m   30\u001b[0m │ \u001b[36m    123\u001b[0m │ \u001b[32m  ▇▁  \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mperson_income         \u001b[0m │ \u001b[36m  0\u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m  64050\u001b[0m │ \u001b[36m  37930\u001b[0m │ \u001b[36m4200\u001b[0m │ \u001b[36m42000\u001b[0m │ \u001b[36m58000\u001b[0m │ \u001b[36m75600\u001b[0m │ \u001b[36m1900000\u001b[0m │ \u001b[32m  ▇   \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mperson_emp_length     \u001b[0m │ \u001b[36m  0\u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m  4.701\u001b[0m │ \u001b[36m   3.96\u001b[0m │ \u001b[36m   0\u001b[0m │ \u001b[36m    2\u001b[0m │ \u001b[36m    4\u001b[0m │ \u001b[36m    7\u001b[0m │ \u001b[36m    123\u001b[0m │ \u001b[32m  ▇   \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_amnt             \u001b[0m │ \u001b[36m  0\u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m   9218\u001b[0m │ \u001b[36m   5564\u001b[0m │ \u001b[36m 500\u001b[0m │ \u001b[36m 5000\u001b[0m │ \u001b[36m 8000\u001b[0m │ \u001b[36m12000\u001b[0m │ \u001b[36m  35000\u001b[0m │ \u001b[32m▇▇▅▁▁ \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_int_rate         \u001b[0m │ \u001b[36m  0\u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m  10.68\u001b[0m │ \u001b[36m  3.035\u001b[0m │ \u001b[36m5.42\u001b[0m │ \u001b[36m 7.88\u001b[0m │ \u001b[36m10.75\u001b[0m │ \u001b[36m12.99\u001b[0m │ \u001b[36m  23.22\u001b[0m │ \u001b[32m ▇▇▇▃ \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_percent_income   \u001b[0m │ \u001b[36m  0\u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m 0.1592\u001b[0m │ \u001b[36m0.09169\u001b[0m │ \u001b[36m   0\u001b[0m │ \u001b[36m 0.09\u001b[0m │ \u001b[36m 0.14\u001b[0m │ \u001b[36m 0.21\u001b[0m │ \u001b[36m   0.83\u001b[0m │ \u001b[32m ▇▇▂  \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mcb_person_cred_hist_le\u001b[0m │ \u001b[36m  0\u001b[0m │ \u001b[36m    0\u001b[0m │ \u001b[36m  5.814\u001b[0m │ \u001b[36m  4.029\u001b[0m │ \u001b[36m   2\u001b[0m │ \u001b[36m    3\u001b[0m │ \u001b[36m    4\u001b[0m │ \u001b[36m    8\u001b[0m │ \u001b[36m     30\u001b[0m │ \u001b[32m ▇▃▁  \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mngth                  \u001b[0m │     │       │         │         │      │       │       │       │         │        │  │\n",
       "│ └────────────────────────┴─────┴───────┴─────────┴─────────┴──────┴───────┴───────┴───────┴─────────┴────────┘  │\n",
       "│ \u001b[3m                                                     bool                                                     \u001b[0m  │\n",
       "│ ┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━┓  │\n",
       "│ ┃\u001b[1m \u001b[0m\u001b[1mcolumn_name                                    \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mtrue          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mtrue rate             \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mhist          \u001b[0m\u001b[1m \u001b[0m┃  │\n",
       "│ ┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━┩  │\n",
       "│ │ \u001b[38;5;141mperson_home_ownership_MORTGAGE                 \u001b[0m │ \u001b[36m         24824\u001b[0m │ \u001b[36m                  0.42\u001b[0m │ \u001b[32m    ▇    ▆    \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mperson_home_ownership_OTHER                    \u001b[0m │ \u001b[36m            89\u001b[0m │ \u001b[36m                0.0015\u001b[0m │ \u001b[32m      ▇       \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mperson_home_ownership_OWN                      \u001b[0m │ \u001b[36m          3138\u001b[0m │ \u001b[36m                 0.054\u001b[0m │ \u001b[32m      ▇       \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mperson_home_ownership_RENT                     \u001b[0m │ \u001b[36m         30594\u001b[0m │ \u001b[36m                  0.52\u001b[0m │ \u001b[32m    ▇    ▇    \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_intent_DEBTCONSOLIDATION                  \u001b[0m │ \u001b[36m          9133\u001b[0m │ \u001b[36m                  0.16\u001b[0m │ \u001b[32m    ▇    ▁    \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_intent_EDUCATION                          \u001b[0m │ \u001b[36m         12271\u001b[0m │ \u001b[36m                  0.21\u001b[0m │ \u001b[32m    ▇    ▂    \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_intent_HOMEIMPROVEMENT                    \u001b[0m │ \u001b[36m          6280\u001b[0m │ \u001b[36m                  0.11\u001b[0m │ \u001b[32m    ▇    ▁    \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_intent_MEDICAL                            \u001b[0m │ \u001b[36m         10934\u001b[0m │ \u001b[36m                  0.19\u001b[0m │ \u001b[32m    ▇    ▂    \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_intent_PERSONAL                           \u001b[0m │ \u001b[36m         10016\u001b[0m │ \u001b[36m                  0.17\u001b[0m │ \u001b[32m    ▇    ▂    \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_intent_VENTURE                            \u001b[0m │ \u001b[36m         10011\u001b[0m │ \u001b[36m                  0.17\u001b[0m │ \u001b[32m    ▇    ▂    \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_grade_A                                   \u001b[0m │ \u001b[36m         20984\u001b[0m │ \u001b[36m                  0.36\u001b[0m │ \u001b[32m    ▇    ▅    \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_grade_B                                   \u001b[0m │ \u001b[36m         20400\u001b[0m │ \u001b[36m                  0.35\u001b[0m │ \u001b[32m    ▇    ▅    \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_grade_C                                   \u001b[0m │ \u001b[36m         11036\u001b[0m │ \u001b[36m                  0.19\u001b[0m │ \u001b[32m    ▇    ▂    \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_grade_D                                   \u001b[0m │ \u001b[36m          5034\u001b[0m │ \u001b[36m                 0.086\u001b[0m │ \u001b[32m    ▇    ▁    \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_grade_E                                   \u001b[0m │ \u001b[36m          1009\u001b[0m │ \u001b[36m                 0.017\u001b[0m │ \u001b[32m      ▇       \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_grade_F                                   \u001b[0m │ \u001b[36m           149\u001b[0m │ \u001b[36m                0.0025\u001b[0m │ \u001b[32m      ▇       \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mloan_grade_G                                   \u001b[0m │ \u001b[36m            33\u001b[0m │ \u001b[36m               0.00056\u001b[0m │ \u001b[32m      ▇       \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mcb_person_default_on_file_N                    \u001b[0m │ \u001b[36m         49943\u001b[0m │ \u001b[36m                  0.85\u001b[0m │ \u001b[32m    ▁    ▇    \u001b[0m │  │\n",
       "│ │ \u001b[38;5;141mcb_person_default_on_file_Y                    \u001b[0m │ \u001b[36m          8702\u001b[0m │ \u001b[36m                  0.15\u001b[0m │ \u001b[32m    ▇    ▁    \u001b[0m │  │\n",
       "│ └─────────────────────────────────────────────────┴────────────────┴────────────────────────┴────────────────┘  │\n",
       "╰────────────────────────────────────────────────────── End ──────────────────────────────────────────────────────╯\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.952510870491943\n",
      "Classification Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.99      0.97     10087\n",
      "           1       0.92      0.73      0.81      1642\n",
      "\n",
      "    accuracy                           0.95     11729\n",
      "   macro avg       0.94      0.86      0.89     11729\n",
      "weighted avg       0.95      0.95      0.95     11729\n",
      "\n",
      "Confusion Matrix:\n",
      "[[9980  107]\n",
      " [ 450 1192]]\n",
      "Feature Importance:\n",
      "                           Feature  Importance\n",
      "5              loan_percent_income    0.244473\n",
      "4                    loan_int_rate    0.122395\n",
      "1                    person_income    0.103419\n",
      "20                    loan_grade_D    0.082313\n",
      "3                        loan_amnt    0.066954\n",
      "2                person_emp_length    0.061223\n",
      "10      person_home_ownership_RENT    0.057547\n",
      "0                       person_age    0.045721\n",
      "6       cb_person_cred_hist_length    0.036149\n",
      "7   person_home_ownership_MORTGAGE    0.025637\n",
      "11   loan_intent_DEBTCONSOLIDATION    0.021950\n",
      "19                    loan_grade_C    0.020946\n",
      "14             loan_intent_MEDICAL    0.018722\n",
      "9        person_home_ownership_OWN    0.014685\n",
      "13     loan_intent_HOMEIMPROVEMENT    0.012956\n",
      "21                    loan_grade_E    0.012725\n",
      "12           loan_intent_EDUCATION    0.008536\n",
      "15            loan_intent_PERSONAL    0.008333\n",
      "17                    loan_grade_A    0.007530\n",
      "16             loan_intent_VENTURE    0.007141\n",
      "24     cb_person_default_on_file_N    0.006908\n",
      "25     cb_person_default_on_file_Y    0.006454\n",
      "18                    loan_grade_B    0.005245\n",
      "22                    loan_grade_F    0.001129\n",
      "8      person_home_ownership_OTHER    0.000647\n",
      "23                    loan_grade_G    0.000262\n"
     ]
    }
   ],
   "source": [
    "### Random forest method without feature scaling\n",
    "\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import accuracy_score, classification_report, confusion_matrix\n",
    "from sklearn.preprocessing import LabelEncoder, OneHotEncoder\n",
    "label_encoder = LabelEncoder()\n",
    "\n",
    "\n",
    "# Assuming your train data is in a Pandas DataFrame called 'df_train'\n",
    "# and the target variable is 'loan_status'\n",
    "\n",
    "# Split the data into features (X) and target (y)\n",
    "X = train.drop('loan_status', axis=1).drop('id',axis = 1)\n",
    "y = train['loan_status']\n",
    "\n",
    "categorical_columns = ['person_home_ownership', 'loan_intent', 'loan_grade', 'cb_person_default_on_file']\n",
    "X = pd.get_dummies(X, columns=categorical_columns)\n",
    "\n",
    "\n",
    "skim(X)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# Split the data into training and validation sets\n",
    "X_train, X_val, y_train, y_val = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "\n",
    "# Initialize the Random Forest Classifier\n",
    "rf_clf = RandomForestClassifier(n_estimators=100, random_state=42)\n",
    "\n",
    "# Train the model on the training data\n",
    "rf_clf.fit(X_train, y_train)\n",
    "\n",
    "# Make predictions on the validation data\n",
    "y_pred = rf_clf.predict(X_val)\n",
    "\n",
    "# Evaluate the model\n",
    "accuracy = accuracy_score(y_val, y_pred)\n",
    "print(\"Accuracy:\", accuracy)\n",
    "print(\"Classification Report:\")\n",
    "print(classification_report(y_val, y_pred))\n",
    "print(\"Confusion Matrix:\")\n",
    "print(confusion_matrix(y_val, y_pred))\n",
    "\n",
    "\n",
    "feature_importance = rf_clf.feature_importances_\n",
    "feature_importance_df = pd.DataFrame({'Feature': X.columns, 'Importance': feature_importance})\n",
    "feature_importance_df = feature_importance_df.sort_values(by='Importance', ascending=False)\n",
    "\n",
    "print(\"Feature Importance:\")\n",
    "print(feature_importance_df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "# Define the hyperparameter grid\n",
    "param_grid = {\n",
    "    'n_estimators': [100, 200, 300],\n",
    "    'max_depth': [None, 5, 10],\n",
    "    'min_samples_split': [2, 5, 10],\n",
    "    'min_samples_leaf': [1, 5, 10],\n",
    "}\n",
    "\n",
    "# Initialize the Random Forest model\n",
    "rf = RandomForestClassifier(random_state=42)\n",
    "\n",
    "# Perform grid search with cross-validation\n",
    "grid_search = GridSearchCV(estimator=rf, param_grid=param_grid, cv=5, scoring='accuracy')\n",
    "grid_search.fit(X_train, y_train)\n",
    "\n",
    "# Get the best-performing model\n",
    "best_model = grid_search.best_estimator_\n",
    "\n",
    "# Evaluate the best model on the test set\n",
    "y_pred = best_model.predict(X_val)\n",
    "accuracy = accuracy_score(y_val, y_pred)\n",
    "print(\"Accuracy:\", accuracy)\n",
    "\n",
    "print(\"Classification Report:\")\n",
    "print(classification_report(y_val, y_pred))\n",
    "print(\"Confusion Matrix:\")\n",
    "print(confusion_matrix(y_val, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load the test data\n",
    "\n",
    "# Apply the same preprocessing steps as before\n",
    "categorical_columns = ['person_home_ownership', 'loan_intent', 'loan_grade', 'cb_person_default_on_file']\n",
    "test_data_encoded = pd.get_dummies(test, columns=categorical_columns)\n",
    "test_data_encoded = test_data_encoded.drop('id', axis= 1)\n",
    "print(test.columns)\n",
    "# Make sure the test data has the same columns as the training data\n",
    "\n",
    "print(X.columns)\n",
    "\n",
    "skim(test_data_encoded)\n",
    "# test_data_encoded = test_data_encoded[X.columns]\n",
    "\n",
    "# Make predictions on the test data\n",
    "predictions = rf_clf.predict(test_data_encoded)\n",
    "\n",
    "# Save the predictions to a submission file\n",
    "submission_df = pd.DataFrame({'loan_id': test['id'], 'loan_status': predictions})\n",
    "submission_df.to_csv('submission.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1 0 1 ... 0 0 1]\n"
     ]
    }
   ],
   "source": [
    "print(predictions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "### Random forest with defualt settings  gave 0.85 score on test , Now we choose best model from grid search\n",
    "\n",
    "\n",
    "# Apply the same preprocessing steps as before\n",
    "categorical_columns = ['person_home_ownership', 'loan_intent', 'loan_grade', 'cb_person_default_on_file']\n",
    "test_data_encoded = pd.get_dummies(test, columns=categorical_columns)\n",
    "test_data_encoded = test_data_encoded.drop('id', axis= 1)\n",
    "print(test.columns)\n",
    "# Make sure the test data has the same columns as the training data\n",
    "\n",
    "print(X.columns)\n",
    "\n",
    "skim(test_data_encoded)\n",
    "# test_data_encoded = test_data_encoded[X.columns]\n",
    "\n",
    "# Make predictions on the test data\n",
    "predictions = best_model.predict(test_data_encoded)\n",
    "\n",
    "# Save the predictions to a submission file\n",
    "submission_df = pd.DataFrame({'loan_id': test['id'], 'loan_status': predictions})\n",
    "submission_df.to_csv('submission_best.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\rushi\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\sklearn\\ensemble\\_weight_boosting.py:527: FutureWarning: The SAMME.R algorithm (the default) is deprecated and will be removed in 1.6. Use the SAMME algorithm to circumvent this warning.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "AdaBoost Accuracy: 0.9298320402421348\n",
      "[0]\tvalidation_0-error:0.14238\tvalidation_0-logloss:0.37095\tvalidation_1-error:0.14238\tvalidation_1-logloss:0.37138\n",
      "[1]\tvalidation_0-error:0.14238\tvalidation_0-logloss:0.33857\tvalidation_1-error:0.14238\tvalidation_1-logloss:0.33923\n",
      "[2]\tvalidation_0-error:0.14238\tvalidation_0-logloss:0.31380\tvalidation_1-error:0.14238\tvalidation_1-logloss:0.31469\n",
      "[3]\tvalidation_0-error:0.14238\tvalidation_0-logloss:0.29390\tvalidation_1-error:0.14238\tvalidation_1-logloss:0.29484\n",
      "[4]\tvalidation_0-error:0.06676\tvalidation_0-logloss:0.27701\tvalidation_1-error:0.06548\tvalidation_1-logloss:0.27832\n",
      "[5]\tvalidation_0-error:0.05915\tvalidation_0-logloss:0.26294\tvalidation_1-error:0.05943\tvalidation_1-logloss:0.26454\n",
      "[6]\tvalidation_0-error:0.05427\tvalidation_0-logloss:0.25098\tvalidation_1-error:0.05593\tvalidation_1-logloss:0.25282\n",
      "[7]\tvalidation_0-error:0.05267\tvalidation_0-logloss:0.24056\tvalidation_1-error:0.05474\tvalidation_1-logloss:0.24257\n",
      "[8]\tvalidation_0-error:0.05197\tvalidation_0-logloss:0.23163\tvalidation_1-error:0.05329\tvalidation_1-logloss:0.23385\n",
      "[9]\tvalidation_0-error:0.05098\tvalidation_0-logloss:0.22344\tvalidation_1-error:0.05184\tvalidation_1-logloss:0.22581\n",
      "[10]\tvalidation_0-error:0.05092\tvalidation_0-logloss:0.21636\tvalidation_1-error:0.05167\tvalidation_1-logloss:0.21881\n",
      "[11]\tvalidation_0-error:0.05092\tvalidation_0-logloss:0.21022\tvalidation_1-error:0.05150\tvalidation_1-logloss:0.21273\n",
      "[12]\tvalidation_0-error:0.05073\tvalidation_0-logloss:0.20469\tvalidation_1-error:0.05141\tvalidation_1-logloss:0.20744\n",
      "[13]\tvalidation_0-error:0.05075\tvalidation_0-logloss:0.19984\tvalidation_1-error:0.05141\tvalidation_1-logloss:0.20286\n",
      "[14]\tvalidation_0-error:0.05064\tvalidation_0-logloss:0.19553\tvalidation_1-error:0.05107\tvalidation_1-logloss:0.19876\n",
      "[15]\tvalidation_0-error:0.05060\tvalidation_0-logloss:0.19165\tvalidation_1-error:0.05098\tvalidation_1-logloss:0.19520\n",
      "[16]\tvalidation_0-error:0.05043\tvalidation_0-logloss:0.18834\tvalidation_1-error:0.05107\tvalidation_1-logloss:0.19212\n",
      "[17]\tvalidation_0-error:0.05035\tvalidation_0-logloss:0.18537\tvalidation_1-error:0.05081\tvalidation_1-logloss:0.18936\n",
      "[18]\tvalidation_0-error:0.05028\tvalidation_0-logloss:0.18236\tvalidation_1-error:0.05064\tvalidation_1-logloss:0.18668\n",
      "[19]\tvalidation_0-error:0.05022\tvalidation_0-logloss:0.17997\tvalidation_1-error:0.05064\tvalidation_1-logloss:0.18450\n",
      "[20]\tvalidation_0-error:0.05015\tvalidation_0-logloss:0.17771\tvalidation_1-error:0.05064\tvalidation_1-logloss:0.18257\n",
      "[21]\tvalidation_0-error:0.05011\tvalidation_0-logloss:0.17551\tvalidation_1-error:0.05056\tvalidation_1-logloss:0.18069\n",
      "[22]\tvalidation_0-error:0.05000\tvalidation_0-logloss:0.17368\tvalidation_1-error:0.05056\tvalidation_1-logloss:0.17900\n",
      "[23]\tvalidation_0-error:0.04998\tvalidation_0-logloss:0.17197\tvalidation_1-error:0.05056\tvalidation_1-logloss:0.17741\n",
      "[24]\tvalidation_0-error:0.04990\tvalidation_0-logloss:0.17044\tvalidation_1-error:0.05047\tvalidation_1-logloss:0.17599\n",
      "[25]\tvalidation_0-error:0.04960\tvalidation_0-logloss:0.16910\tvalidation_1-error:0.05047\tvalidation_1-logloss:0.17482\n",
      "[26]\tvalidation_0-error:0.04958\tvalidation_0-logloss:0.16784\tvalidation_1-error:0.05039\tvalidation_1-logloss:0.17378\n",
      "[27]\tvalidation_0-error:0.04934\tvalidation_0-logloss:0.16642\tvalidation_1-error:0.05030\tvalidation_1-logloss:0.17253\n",
      "[28]\tvalidation_0-error:0.04934\tvalidation_0-logloss:0.16498\tvalidation_1-error:0.05030\tvalidation_1-logloss:0.17139\n",
      "[29]\tvalidation_0-error:0.04930\tvalidation_0-logloss:0.16385\tvalidation_1-error:0.05030\tvalidation_1-logloss:0.17050\n",
      "[30]\tvalidation_0-error:0.04892\tvalidation_0-logloss:0.16234\tvalidation_1-error:0.05013\tvalidation_1-logloss:0.16913\n",
      "[31]\tvalidation_0-error:0.04881\tvalidation_0-logloss:0.16155\tvalidation_1-error:0.04988\tvalidation_1-logloss:0.16848\n",
      "[32]\tvalidation_0-error:0.04862\tvalidation_0-logloss:0.16045\tvalidation_1-error:0.04988\tvalidation_1-logloss:0.16763\n",
      "[33]\tvalidation_0-error:0.04853\tvalidation_0-logloss:0.15876\tvalidation_1-error:0.04962\tvalidation_1-logloss:0.16600\n",
      "[34]\tvalidation_0-error:0.04828\tvalidation_0-logloss:0.15790\tvalidation_1-error:0.04971\tvalidation_1-logloss:0.16532\n",
      "[35]\tvalidation_0-error:0.04804\tvalidation_0-logloss:0.15712\tvalidation_1-error:0.04954\tvalidation_1-logloss:0.16480\n",
      "[36]\tvalidation_0-error:0.04785\tvalidation_0-logloss:0.15628\tvalidation_1-error:0.04911\tvalidation_1-logloss:0.16417\n",
      "[37]\tvalidation_0-error:0.04770\tvalidation_0-logloss:0.15504\tvalidation_1-error:0.04868\tvalidation_1-logloss:0.16303\n",
      "[38]\tvalidation_0-error:0.04760\tvalidation_0-logloss:0.15427\tvalidation_1-error:0.04860\tvalidation_1-logloss:0.16254\n",
      "[39]\tvalidation_0-error:0.04751\tvalidation_0-logloss:0.15359\tvalidation_1-error:0.04834\tvalidation_1-logloss:0.16203\n",
      "[40]\tvalidation_0-error:0.04734\tvalidation_0-logloss:0.15229\tvalidation_1-error:0.04834\tvalidation_1-logloss:0.16083\n",
      "[41]\tvalidation_0-error:0.04689\tvalidation_0-logloss:0.15153\tvalidation_1-error:0.04843\tvalidation_1-logloss:0.16036\n",
      "[42]\tvalidation_0-error:0.04679\tvalidation_0-logloss:0.15102\tvalidation_1-error:0.04860\tvalidation_1-logloss:0.16007\n",
      "[43]\tvalidation_0-error:0.04672\tvalidation_0-logloss:0.15043\tvalidation_1-error:0.04843\tvalidation_1-logloss:0.15971\n",
      "[44]\tvalidation_0-error:0.04664\tvalidation_0-logloss:0.14986\tvalidation_1-error:0.04851\tvalidation_1-logloss:0.15949\n",
      "[45]\tvalidation_0-error:0.04657\tvalidation_0-logloss:0.14934\tvalidation_1-error:0.04868\tvalidation_1-logloss:0.15924\n",
      "[46]\tvalidation_0-error:0.04653\tvalidation_0-logloss:0.14875\tvalidation_1-error:0.04868\tvalidation_1-logloss:0.15899\n",
      "[47]\tvalidation_0-error:0.04644\tvalidation_0-logloss:0.14817\tvalidation_1-error:0.04877\tvalidation_1-logloss:0.15875\n",
      "[48]\tvalidation_0-error:0.04632\tvalidation_0-logloss:0.14752\tvalidation_1-error:0.04860\tvalidation_1-logloss:0.15838\n",
      "[49]\tvalidation_0-error:0.04623\tvalidation_0-logloss:0.14708\tvalidation_1-error:0.04843\tvalidation_1-logloss:0.15829\n",
      "[50]\tvalidation_0-error:0.04602\tvalidation_0-logloss:0.14614\tvalidation_1-error:0.04834\tvalidation_1-logloss:0.15749\n",
      "[51]\tvalidation_0-error:0.04600\tvalidation_0-logloss:0.14579\tvalidation_1-error:0.04843\tvalidation_1-logloss:0.15735\n",
      "[52]\tvalidation_0-error:0.04589\tvalidation_0-logloss:0.14532\tvalidation_1-error:0.04843\tvalidation_1-logloss:0.15719\n",
      "[53]\tvalidation_0-error:0.04581\tvalidation_0-logloss:0.14495\tvalidation_1-error:0.04843\tvalidation_1-logloss:0.15706\n",
      "[54]\tvalidation_0-error:0.04568\tvalidation_0-logloss:0.14456\tvalidation_1-error:0.04843\tvalidation_1-logloss:0.15700\n",
      "[55]\tvalidation_0-error:0.04549\tvalidation_0-logloss:0.14366\tvalidation_1-error:0.04843\tvalidation_1-logloss:0.15623\n",
      "[56]\tvalidation_0-error:0.04546\tvalidation_0-logloss:0.14306\tvalidation_1-error:0.04834\tvalidation_1-logloss:0.15590\n",
      "[57]\tvalidation_0-error:0.04542\tvalidation_0-logloss:0.14281\tvalidation_1-error:0.04834\tvalidation_1-logloss:0.15582\n",
      "[58]\tvalidation_0-error:0.04536\tvalidation_0-logloss:0.14240\tvalidation_1-error:0.04834\tvalidation_1-logloss:0.15573\n",
      "[59]\tvalidation_0-error:0.04523\tvalidation_0-logloss:0.14146\tvalidation_1-error:0.04860\tvalidation_1-logloss:0.15501\n",
      "[60]\tvalidation_0-error:0.04514\tvalidation_0-logloss:0.14114\tvalidation_1-error:0.04851\tvalidation_1-logloss:0.15476\n",
      "[61]\tvalidation_0-error:0.04508\tvalidation_0-logloss:0.14079\tvalidation_1-error:0.04851\tvalidation_1-logloss:0.15461\n",
      "[62]\tvalidation_0-error:0.04506\tvalidation_0-logloss:0.14051\tvalidation_1-error:0.04851\tvalidation_1-logloss:0.15459\n",
      "[63]\tvalidation_0-error:0.04500\tvalidation_0-logloss:0.14022\tvalidation_1-error:0.04851\tvalidation_1-logloss:0.15445\n",
      "[64]\tvalidation_0-error:0.04489\tvalidation_0-logloss:0.13943\tvalidation_1-error:0.04817\tvalidation_1-logloss:0.15393\n",
      "[65]\tvalidation_0-error:0.04480\tvalidation_0-logloss:0.13910\tvalidation_1-error:0.04817\tvalidation_1-logloss:0.15376\n",
      "[66]\tvalidation_0-error:0.04470\tvalidation_0-logloss:0.13879\tvalidation_1-error:0.04817\tvalidation_1-logloss:0.15359\n",
      "[67]\tvalidation_0-error:0.04476\tvalidation_0-logloss:0.13861\tvalidation_1-error:0.04817\tvalidation_1-logloss:0.15356\n",
      "[68]\tvalidation_0-error:0.04472\tvalidation_0-logloss:0.13823\tvalidation_1-error:0.04809\tvalidation_1-logloss:0.15352\n",
      "[69]\tvalidation_0-error:0.04474\tvalidation_0-logloss:0.13785\tvalidation_1-error:0.04809\tvalidation_1-logloss:0.15340\n",
      "[70]\tvalidation_0-error:0.04463\tvalidation_0-logloss:0.13756\tvalidation_1-error:0.04817\tvalidation_1-logloss:0.15339\n",
      "[71]\tvalidation_0-error:0.04459\tvalidation_0-logloss:0.13746\tvalidation_1-error:0.04817\tvalidation_1-logloss:0.15339\n",
      "[72]\tvalidation_0-error:0.04446\tvalidation_0-logloss:0.13726\tvalidation_1-error:0.04800\tvalidation_1-logloss:0.15331\n",
      "[73]\tvalidation_0-error:0.04451\tvalidation_0-logloss:0.13699\tvalidation_1-error:0.04800\tvalidation_1-logloss:0.15319\n",
      "[74]\tvalidation_0-error:0.04442\tvalidation_0-logloss:0.13657\tvalidation_1-error:0.04783\tvalidation_1-logloss:0.15305\n",
      "[75]\tvalidation_0-error:0.04436\tvalidation_0-logloss:0.13620\tvalidation_1-error:0.04809\tvalidation_1-logloss:0.15292\n",
      "[76]\tvalidation_0-error:0.04414\tvalidation_0-logloss:0.13596\tvalidation_1-error:0.04792\tvalidation_1-logloss:0.15286\n",
      "[77]\tvalidation_0-error:0.04419\tvalidation_0-logloss:0.13575\tvalidation_1-error:0.04792\tvalidation_1-logloss:0.15285\n",
      "[78]\tvalidation_0-error:0.04406\tvalidation_0-logloss:0.13530\tvalidation_1-error:0.04792\tvalidation_1-logloss:0.15268\n",
      "[79]\tvalidation_0-error:0.04391\tvalidation_0-logloss:0.13455\tvalidation_1-error:0.04783\tvalidation_1-logloss:0.15221\n",
      "[80]\tvalidation_0-error:0.04387\tvalidation_0-logloss:0.13423\tvalidation_1-error:0.04783\tvalidation_1-logloss:0.15212\n",
      "[81]\tvalidation_0-error:0.04391\tvalidation_0-logloss:0.13406\tvalidation_1-error:0.04783\tvalidation_1-logloss:0.15201\n",
      "[82]\tvalidation_0-error:0.04391\tvalidation_0-logloss:0.13386\tvalidation_1-error:0.04783\tvalidation_1-logloss:0.15194\n",
      "[83]\tvalidation_0-error:0.04387\tvalidation_0-logloss:0.13363\tvalidation_1-error:0.04792\tvalidation_1-logloss:0.15191\n",
      "[84]\tvalidation_0-error:0.04378\tvalidation_0-logloss:0.13317\tvalidation_1-error:0.04783\tvalidation_1-logloss:0.15165\n",
      "[85]\tvalidation_0-error:0.04372\tvalidation_0-logloss:0.13294\tvalidation_1-error:0.04783\tvalidation_1-logloss:0.15164\n",
      "[86]\tvalidation_0-error:0.04365\tvalidation_0-logloss:0.13271\tvalidation_1-error:0.04809\tvalidation_1-logloss:0.15164\n",
      "[87]\tvalidation_0-error:0.04361\tvalidation_0-logloss:0.13245\tvalidation_1-error:0.04809\tvalidation_1-logloss:0.15150\n",
      "[88]\tvalidation_0-error:0.04365\tvalidation_0-logloss:0.13230\tvalidation_1-error:0.04809\tvalidation_1-logloss:0.15144\n",
      "[89]\tvalidation_0-error:0.04342\tvalidation_0-logloss:0.13200\tvalidation_1-error:0.04800\tvalidation_1-logloss:0.15132\n",
      "[90]\tvalidation_0-error:0.04340\tvalidation_0-logloss:0.13168\tvalidation_1-error:0.04800\tvalidation_1-logloss:0.15106\n",
      "[91]\tvalidation_0-error:0.04346\tvalidation_0-logloss:0.13146\tvalidation_1-error:0.04800\tvalidation_1-logloss:0.15107\n",
      "[92]\tvalidation_0-error:0.04342\tvalidation_0-logloss:0.13123\tvalidation_1-error:0.04800\tvalidation_1-logloss:0.15104\n",
      "[93]\tvalidation_0-error:0.04338\tvalidation_0-logloss:0.13109\tvalidation_1-error:0.04800\tvalidation_1-logloss:0.15100\n",
      "[94]\tvalidation_0-error:0.04333\tvalidation_0-logloss:0.13086\tvalidation_1-error:0.04800\tvalidation_1-logloss:0.15098\n",
      "[95]\tvalidation_0-error:0.04323\tvalidation_0-logloss:0.13061\tvalidation_1-error:0.04800\tvalidation_1-logloss:0.15090\n",
      "[96]\tvalidation_0-error:0.04320\tvalidation_0-logloss:0.13015\tvalidation_1-error:0.04800\tvalidation_1-logloss:0.15059\n",
      "[97]\tvalidation_0-error:0.04310\tvalidation_0-logloss:0.12995\tvalidation_1-error:0.04783\tvalidation_1-logloss:0.15062\n",
      "[98]\tvalidation_0-error:0.04308\tvalidation_0-logloss:0.12979\tvalidation_1-error:0.04792\tvalidation_1-logloss:0.15062\n",
      "[99]\tvalidation_0-error:0.04303\tvalidation_0-logloss:0.12945\tvalidation_1-error:0.04800\tvalidation_1-logloss:0.15054\n",
      "XGBoost Accuracy: 0.9519993179299173\n",
      "Best Threshold: 0.393939393939394\n",
      "Best F-score: 0.8139303482587065\n"
     ]
    }
   ],
   "source": [
    "#### Trying boosting algorithms\n",
    "\n",
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "from sklearn.model_selection import train_test_split\n",
    "import xgboost as xgb\n",
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "# Split the data into training and testing sets\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y,stratify=y ,test_size=0.2, random_state=42)\n",
    "eval_set = [(X_train, y_train), (X_test, y_test)]\n",
    "\n",
    "# AdaBoost\n",
    "ada_boost = AdaBoostClassifier(n_estimators=100, random_state=42)\n",
    "ada_boost.fit(X_train, y_train)\n",
    "y_pred_ada = ada_boost.predict(X_test)\n",
    "accuracy_ada = accuracy_score(y_test, y_pred_ada)\n",
    "print(\"AdaBoost Accuracy:\", accuracy_ada)\n",
    "\n",
    "# Gradient Boosting\n",
    "# grad_boost = xgb.XGBClassifier(n_estimators=100, random_state=42)\n",
    "# grad_boost.fit(X_train, y_train)\n",
    "# y_pred_grad = grad_boost.predict(X_test)\n",
    "# accuracy_grad = accuracy_score(y_test, y_pred_grad)\n",
    "# print(\"Gradient Boosting Accuracy:\", accuracy_grad)\n",
    "# Define the lambda function as a keyword argument\n",
    "params = {\n",
    "    'n_estimators':100,\n",
    "    'eval_metric': ['error','logloss'],\n",
    "    # 'max_depth': 4,\n",
    "    # 'lambda': 1,  # Regularization parameter\n",
    "    # 'subsample': 0.3,\n",
    "    # 'colsample_bytree': 0.5,\n",
    "    'eta': 0.1,\n",
    "    'seed': 42\n",
    "}\n",
    "# XGBoost\n",
    "# xgb_boost = xgb.XGBClassifier(n_estimators=400, random_state=42, device = 'cuda', eval_metric=[\"error\", \"logloss\"],max_depth = 5)\n",
    "xgb_boost = xgb.XGBClassifier(**params)\n",
    "\n",
    "xgb_boost.fit(X_train, y_train, eval_set=eval_set, verbose=True)\n",
    "y_pred_xgb = xgb_boost.predict(X_test)\n",
    "accuracy_xgb = accuracy_score(y_test, y_pred_xgb)\n",
    "print(\"XGBoost Accuracy:\", accuracy_xgb)\n",
    "\n",
    "from sklearn.metrics import f1_score\n",
    "\n",
    "\n",
    "# get the predicted probabilities\n",
    "y_pred_proba = xgb_boost.predict_proba(X_test)[:, 1]\n",
    "\n",
    "# define a range of thresholds\n",
    "thresholds = np.linspace(0, 1, 100)\n",
    "\n",
    "# calculate the F-score for each threshold\n",
    "f_scores = []\n",
    "for threshold in thresholds:\n",
    "    y_pred = (y_pred_proba >= threshold).astype(int)\n",
    "    f_score = f1_score(y_test, y_pred)\n",
    "    f_scores.append(f_score)\n",
    "\n",
    "# find the index of the maximum F-score value\n",
    "max_f_score_idx = np.argmax(f_scores)\n",
    "\n",
    "# get the corresponding threshold value\n",
    "best_threshold = thresholds[max_f_score_idx]\n",
    "\n",
    "print(\"Best Threshold:\", best_threshold)\n",
    "print(\"Best F-score:\", f_scores[max_f_score_idx])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'validation_0': OrderedDict([('error', [0.14238212976383324, 0.14238212976383324, 0.14238212976383324, 0.14103930428851563, 0.13620086963935543, 0.13140506437036406, 0.12682240600221673, 0.12025748145621963, 0.10256628868616251, 0.09316651035893939, 0.07720180748571916, 0.07342910734077926, 0.06822832296018416, 0.06839884048085941, 0.06603290988149033, 0.06364566459203683, 0.0645621962656663, 0.06370960866229006, 0.06430641998465342, 0.06170602779435587, 0.05916957967431154, 0.05708074004603973, 0.05673970500468923, 0.05563134112030011, 0.05505584448802114, 0.05437377440532015, 0.05422457157472931, 0.05399010998380083, 0.05347855742177509, 0.05328672521101543, 0.05307357831017137, 0.05279648733907409, 0.05243413760763919, 0.05241282291755478, 0.05234887884730156, 0.05228493477704834, 0.05215704663654191, 0.05217836132662631, 0.05209310256628868, 0.05200784380595106, 0.05181601159519141, 0.051794696905107, 0.05162417938443175, 0.05160286469434734, 0.05158155000426294, 0.05153892062409413, 0.05141103248358769, 0.05104868275215278, 0.0507715917810555, 0.05068633302071788, 0.05062238895046466, 0.05062238895046466, 0.05051581550004263, 0.05030266859919857, 0.05023872452894535, 0.0500468923181857, 0.0500682070082701, 0.0500468923181857, 0.05002557762810129, 0.04994031886776366, 0.04996163355784807, 0.0497271719669196, 0.0497271719669196, 0.0493648222354847, 0.04934350754540029, 0.04913036064455623, 0.04923693409497826, 0.04910904595447182, 0.04893852843379657, 0.04900247250404979, 0.04876801091312132, 0.04855486401227726, 0.04853354932219286, 0.04849091994202404, 0.04844829056185523, 0.04861880808253048, 0.04859749339244607, 0.04855486401227726, 0.04851223463210845, 0.04842697587177083, 0.04808594083042032, 0.04806462614033592, 0.04808594083042032, 0.04808594083042032, 0.04806462614033592, 0.04785147923949186, 0.04787279392957627, 0.04783016454940745, 0.04783016454940745, 0.04780884985932304, 0.04776622047915424, 0.04772359109898542, 0.04778753516923864, 0.04789410861966067, 0.04780884985932304, 0.0476596470287322, 0.0476596470287322, 0.0476383323386478, 0.04748912950805696, 0.04719072384687527, 0.04725466791712848, 0.04740387074771933, 0.04733992667746611, 0.0473186119873817, 0.04727598260721289, 0.04727598260721289, 0.04727598260721289, 0.04712677977662205, 0.04716940915679086, 0.04710546508653764, 0.04708415039645324, 0.04710546508653764, 0.04710546508653764, 0.04704152101628442, 0.04699889163611561, 0.0469562622559468, 0.04682837411544036, 0.04682837411544036, 0.04676443004518714, 0.04670048597493393, 0.04678574473527155, 0.04680705942535596, 0.04672180066501833, 0.04672180066501833, 0.04672180066501833, 0.04670048597493393, 0.04684968880552477, 0.04682837411544036, 0.04680705942535596, 0.04672180066501833, 0.04680705942535596, 0.04682837411544036, 0.04678574473527155, 0.04674311535510273, 0.04674311535510273, 0.04678574473527155, 0.04670048597493393, 0.04674311535510273, 0.04672180066501833, 0.04674311535510273, 0.04667917128484952, 0.04676443004518714, 0.04665785659476511, 0.04663654190468071, 0.04670048597493393, 0.04646602438400545, 0.04655128314434308, 0.04648733907408986, 0.04638076562366783, 0.04640208031375224, 0.04640208031375224, 0.04631682155341461, 0.04629550686333021, 0.04633813624349902, 0.04621024810299258, 0.04614630403273937, 0.04618893341290817, 0.04618893341290817, 0.04616761872282377, 0.0462528774831614, 0.04616761872282377, 0.04601841589223293, 0.04597578651206412, 0.04591184244181089, 0.04597578651206412, 0.04589052775172649, 0.04591184244181089, 0.0459331571318953, 0.0459331571318953, 0.04599710120214852, 0.04599710120214852, 0.0459331571318953, 0.04595447182197971, 0.04586921306164209, 0.04586921306164209, 0.04591184244181089, 0.0459331571318953, 0.04591184244181089, 0.04589052775172649, 0.04584789837155768, 0.04582658368147327, 0.04582658368147327, 0.04582658368147327, 0.04582658368147327, 0.04580526899138886, 0.04574132492113565, 0.04569869554096683, 0.04569869554096683, 0.04565606616079802, 0.04561343678062921, 0.04565606616079802, 0.04567738085088243, 0.04572001023105124, 0.04569869554096683, 0.04574132492113565, 0.04578395430130446, 0.04580526899138886, 0.04582658368147327, 0.04589052775172649, 0.04580526899138886, 0.04578395430130446, 0.04572001023105124, 0.04565606616079802, 0.04569869554096683, 0.04563475147071362, 0.04569869554096683, 0.04567738085088243, 0.04567738085088243, 0.0455708074004604, 0.04559212209054481, 0.04550686333020718, 0.04544291925995396, 0.04537897518970074, 0.04533634580953193, 0.04531503111944752, 0.04529371642936312, 0.04529371642936312, 0.04529371642936312, 0.04518714297894109, 0.04533634580953193, 0.04531503111944752, 0.04527240173927871, 0.04529371642936312, 0.04529371642936312, 0.04516582828885668, 0.04516582828885668, 0.0452297723591099, 0.04514451359877227, 0.0452510870491943, 0.0452297723591099, 0.0452084576690255, 0.04531503111944752, 0.0452297723591099, 0.0452510870491943, 0.04516582828885668, 0.04516582828885668, 0.0452510870491943, 0.0452297723591099, 0.0452084576690255, 0.04518714297894109, 0.04518714297894109, 0.0452084576690255, 0.04512319890868786, 0.04497399607809702, 0.04508056952851906, 0.04505925483843465, 0.04510188421860346, 0.04505925483843465, 0.04508056952851906, 0.04486742262767499, 0.04484610793759058, 0.04491005200784381, 0.04491005200784381, 0.04486742262767499, 0.04493136669792822, 0.04486742262767499, 0.04482479324750618, 0.04480347855742178, 0.04482479324750618, 0.04480347855742178, 0.04476084917725296, 0.04486742262767499, 0.04480347855742178, 0.0448887373177594, 0.04497399607809702, 0.04480347855742178, 0.04482479324750618, 0.04482479324750618, 0.04482479324750618, 0.04491005200784381, 0.04469690510699974, 0.04491005200784381, 0.04486742262767499, 0.04478216386733737, 0.04480347855742178, 0.04482479324750618, 0.04480347855742178, 0.04478216386733737, 0.04469690510699974, 0.04471821979708415, 0.04473953448716856, 0.04478216386733737, 0.04476084917725296, 0.04471821979708415, 0.04478216386733737, 0.04469690510699974, 0.04467559041691534, 0.04467559041691534, 0.04469690510699974, 0.04463296103674653, 0.04465427572683093, 0.04463296103674653, 0.04463296103674653, 0.04456901696649331, 0.0445263875863245, 0.04459033165657771, 0.04463296103674653, 0.04461164634666212, 0.04444112882598687, 0.04444112882598687]), ('logloss', [0.38808610241535235, 0.36670979639815904, 0.35053663104372373, 0.3324065787003228, 0.31335758723897267, 0.3013234309110813, 0.2922534957054552, 0.2840333938147939, 0.2723046130736483, 0.2639778412324913, 0.25368382920017096, 0.2480758105590717, 0.2431332275094505, 0.23869101435323384, 0.2327301370150536, 0.2277266068404061, 0.22404698389817768, 0.22085967160240788, 0.21812480016993369, 0.21328235018213243, 0.20953158123661822, 0.2064367658615472, 0.20534859083021823, 0.20245352883683704, 0.20074564185342403, 0.1984013691664035, 0.1966055613994533, 0.19495428378770543, 0.19306265216474636, 0.1920567120070708, 0.19075680755389102, 0.18967747793626047, 0.18839981003986572, 0.18700770593763794, 0.18615325972395697, 0.1856230903063011, 0.1846383159475728, 0.18354954625876022, 0.18254283617770872, 0.1819424455776318, 0.1808692616036484, 0.17996347516693187, 0.1790384394051986, 0.17840022855298424, 0.17787506915656304, 0.17750327546069772, 0.17717001973362717, 0.17560916223878445, 0.17461400129458138, 0.17394250373557174, 0.1731886363937291, 0.17266598072465206, 0.17237845050555609, 0.17202820747865477, 0.17162093651461996, 0.1710488286148721, 0.17038236612868385, 0.16991737260833692, 0.16970076069288167, 0.16932564395113828, 0.16908341876309008, 0.16871644480665315, 0.1684772112348905, 0.1679068385223381, 0.16772943402187665, 0.1674435431351048, 0.16730921647154465, 0.16677798880756167, 0.16639570272295678, 0.16619487888707277, 0.16587244099322798, 0.16469673785316286, 0.1644658760572824, 0.16404862378709567, 0.16385460719461148, 0.16364157747368202, 0.1634539763504248, 0.16315411067007496, 0.16295895522225368, 0.16280310967478479, 0.16238693558726985, 0.16224506820226706, 0.1617020265970147, 0.16148188605876132, 0.16137724458067865, 0.16075783200260943, 0.1605841754901464, 0.16032949286447504, 0.1602098316518186, 0.15998773169731126, 0.15989925789513992, 0.15952640585958894, 0.15931209213312797, 0.15924162002423747, 0.15912687706762194, 0.15873428846067741, 0.15866149277791633, 0.15860933808984667, 0.15807593111762736, 0.15754511240628802, 0.15747481083589002, 0.15722750485489903, 0.15703550258598165, 0.15691376456338404, 0.1568223704774429, 0.1567622980871462, 0.1564508834810146, 0.15610283613608425, 0.15604990302461672, 0.15597999831177792, 0.15554120371826036, 0.15544303640887663, 0.1551625562809921, 0.15511866029014926, 0.15505433896661527, 0.15493389836783558, 0.1547723053085397, 0.15465892450729246, 0.1545368470122666, 0.1544653129021313, 0.15433415372655235, 0.15417883956163375, 0.15406737206537668, 0.1540112888227041, 0.1538990040552394, 0.15373996505883758, 0.15360338927292447, 0.1535295235141639, 0.15338261649399607, 0.1532363439339898, 0.15317945144599518, 0.15297127065236812, 0.15286683995495626, 0.1527530528352995, 0.1526102791942416, 0.1521798416932796, 0.15215251303741462, 0.15209447331913, 0.15202585940799576, 0.15194026799512347, 0.15175035325656766, 0.15166708995686093, 0.15152143187546122, 0.15142840753330328, 0.1513455628388088, 0.15125552439795487, 0.15108303516218274, 0.15097170582357122, 0.1508805914161295, 0.15059162512827132, 0.15042534489543646, 0.1499738531294612, 0.14984180638060415, 0.1496690661218595, 0.1494369851194251, 0.14934288797502301, 0.14925238949168035, 0.14897282478822377, 0.1489159666907168, 0.14881824396789073, 0.14875465959678935, 0.14861092151569874, 0.14842256597092227, 0.1483528450645474, 0.1482010838516858, 0.14812339834493662, 0.1480956993075057, 0.14806030318273736, 0.14781080077934794, 0.14758125744898326, 0.14752191150159052, 0.14748313876182062, 0.147438031720629, 0.1473200685274444, 0.14726874404017679, 0.14710586441192697, 0.14687699656055556, 0.14682916907935797, 0.1466039185764514, 0.14652020060758192, 0.14641066991768537, 0.14633989879446757, 0.14625607067096702, 0.14605915840654443, 0.1459738323342578, 0.14592332248932655, 0.14584697849589265, 0.14573588852839284, 0.14560760883860943, 0.1455723892948182, 0.14553493982741872, 0.1453565526104237, 0.14528653521280332, 0.14508155488079177, 0.1450258761642722, 0.14498509009901833, 0.14490421944696685, 0.144879861801556, 0.14473154587927625, 0.14468291178310122, 0.14438462075559885, 0.14409917209636275, 0.14402949410484642, 0.14396469778899315, 0.14387279029176747, 0.14370354943186242, 0.1435358288565038, 0.14350252736713814, 0.14328440478540908, 0.14309580050114237, 0.14304488506399538, 0.14282834475197753, 0.14270311681023284, 0.14264799607271864, 0.14239863786677331, 0.1422885400794954, 0.14225230903356453, 0.14219616033824042, 0.14210737223514747, 0.14205118889846652, 0.14181818498354454, 0.1417772454851818, 0.14165615802697934, 0.14150157439775352, 0.14147079684495986, 0.14141037963065145, 0.1411961602250413, 0.14110850709704031, 0.14104914628609597, 0.14099597237104888, 0.1409299274799397, 0.1407361221688079, 0.14058361289230417, 0.14055113806371333, 0.14047537417498215, 0.14043976247893655, 0.14035075567944705, 0.14026992723005796, 0.14024970144806537, 0.14008254857341906, 0.14004498690900505, 0.13996018431733406, 0.13978760166577753, 0.13972862869287986, 0.13963866676503406, 0.1395523416059417, 0.13952462063795787, 0.1394608310082523, 0.13944376877148637, 0.1393163236349585, 0.1392807066583158, 0.13923614916354063, 0.1391473863566564, 0.13903861229192718, 0.13898728409782904, 0.13891729954995968, 0.13877031555627592, 0.13875683346935055, 0.13870894969767028, 0.1386850251936126, 0.13858918950243362, 0.13854454493042886, 0.138497474202944, 0.1384451309339511, 0.1384168742251357, 0.13835015283059315, 0.1383336000465045, 0.13827543044671897, 0.13822992962470632, 0.13819046621046702, 0.13815233682026903, 0.1380814224978587, 0.13800962955028434, 0.13795364827728585, 0.13792305211451594, 0.1378122921065203, 0.1377098010917076, 0.1376823294068333, 0.1376007798943262, 0.13755046660173367, 0.1374747100671523, 0.13741970016415278, 0.13737480214738615, 0.1373315613779297, 0.1372717795993347, 0.1371749050124367, 0.137040782710355, 0.13697593911805897, 0.13689011455167016, 0.13685316534326591, 0.13682341615981358, 0.1367849396133701, 0.13672563553854702, 0.13667825459740093, 0.1366074433425198, 0.13656835120193875, 0.13650429286310245, 0.13645606750306866, 0.13635820713003072, 0.13630389596548526])]), 'validation_1': OrderedDict([('error', [0.14238212976383324, 0.14238212976383324, 0.14238212976383324, 0.14127376587944412, 0.13726660414357575, 0.13308892488703214, 0.1278028817460994, 0.12123795720010232, 0.10384517009122687, 0.09310256628868616, 0.0752834853781226, 0.07332253389035723, 0.0680364907494245, 0.06820700827009975, 0.06607553926165914, 0.06292096512916702, 0.06437036405490665, 0.06292096512916702, 0.06360303521186803, 0.06087475488106403, 0.05823173331059767, 0.05618552306249467, 0.05652655810384517, 0.05533293545911842, 0.05482138289709267, 0.05430983033506693, 0.05371301901270355, 0.05362776025236593, 0.05303094893000256, 0.05294569016966493, 0.05320146645067781, 0.05277517264898968, 0.05192258504561344, 0.05183732628527581, 0.05175206752493819, 0.05158155000426294, 0.05149629124392531, 0.05124051496291244, 0.05115525620257481, 0.05141103248358769, 0.05132577372325006, 0.05115525620257481, 0.05115525620257481, 0.05106999744223719, 0.05098473868189957, 0.05098473868189957, 0.05072896240088669, 0.05081422116122432, 0.05038792735953619, 0.05013215107852332, 0.05038792735953619, 0.05013215107852332, 0.05013215107852332, 0.04979111603717282, 0.04987637479751045, 0.04979111603717282, 0.04979111603717282, 0.04962059851649757, 0.04962059851649757, 0.04962059851649757, 0.04970585727683519, 0.04919430471480945, 0.04910904595447182, 0.04953533975615994, 0.04945008099582232, 0.0493648222354847, 0.0493648222354847, 0.04927956347514707, 0.0493648222354847, 0.0493648222354847, 0.04962059851649757, 0.04962059851649757, 0.04945008099582232, 0.0493648222354847, 0.04945008099582232, 0.04953533975615994, 0.04945008099582232, 0.04919430471480945, 0.04910904595447182, 0.04919430471480945, 0.04945008099582232, 0.04962059851649757, 0.04953533975615994, 0.04945008099582232, 0.04927956347514707, 0.0493648222354847, 0.0493648222354847, 0.0493648222354847, 0.04902378719413419, 0.04910904595447182, 0.04910904595447182, 0.04902378719413419, 0.04893852843379657, 0.04902378719413419, 0.04893852843379657, 0.04893852843379657, 0.04902378719413419, 0.04885326967345895, 0.04885326967345895, 0.04885326967345895, 0.04885326967345895, 0.04885326967345895, 0.04876801091312132, 0.04859749339244607, 0.0486827521527837, 0.04876801091312132, 0.0486827521527837, 0.0486827521527837, 0.04859749339244607, 0.0486827521527837, 0.04851223463210845, 0.04885326967345895, 0.04842697587177083, 0.04885326967345895, 0.0486827521527837, 0.04859749339244607, 0.0486827521527837, 0.0486827521527837, 0.04876801091312132, 0.0486827521527837, 0.04851223463210845, 0.04851223463210845, 0.04876801091312132, 0.04876801091312132, 0.04876801091312132, 0.04876801091312132, 0.04859749339244607, 0.04902378719413419, 0.04876801091312132, 0.04910904595447182, 0.04893852843379657, 0.04885326967345895, 0.04902378719413419, 0.04902378719413419, 0.04893852843379657, 0.04902378719413419, 0.04902378719413419, 0.04910904595447182, 0.04893852843379657, 0.04885326967345895, 0.04902378719413419, 0.04919430471480945, 0.04902378719413419, 0.04902378719413419, 0.04902378719413419, 0.04910904595447182, 0.04910904595447182, 0.04876801091312132, 0.04859749339244607, 0.0486827521527837, 0.04876801091312132, 0.0486827521527837, 0.04825645835109558, 0.0483417171114332, 0.04851223463210845, 0.04851223463210845, 0.04851223463210845, 0.0483417171114332, 0.0483417171114332, 0.04851223463210845, 0.04825645835109558, 0.04859749339244607, 0.04851223463210845, 0.04842697587177083, 0.04842697587177083, 0.04851223463210845, 0.0483417171114332, 0.04825645835109558, 0.0483417171114332, 0.0480006820700827, 0.0483417171114332, 0.04851223463210845, 0.04851223463210845, 0.04842697587177083, 0.04842697587177083, 0.0483417171114332, 0.04851223463210845, 0.04842697587177083, 0.0483417171114332, 0.04825645835109558, 0.04817119959075795, 0.04825645835109558, 0.04825645835109558, 0.04825645835109558, 0.04842697587177083, 0.04851223463210845, 0.04842697587177083, 0.04825645835109558, 0.0483417171114332, 0.04817119959075795, 0.04808594083042032, 0.04808594083042032, 0.04825645835109558, 0.04808594083042032, 0.0480006820700827, 0.04825645835109558, 0.0483417171114332, 0.04825645835109558, 0.04808594083042032, 0.04808594083042032, 0.04825645835109558, 0.04808594083042032, 0.0480006820700827, 0.04808594083042032, 0.04817119959075795, 0.04817119959075795, 0.0483417171114332, 0.0483417171114332, 0.04842697587177083, 0.04859749339244607, 0.04859749339244607, 0.04851223463210845, 0.04851223463210845, 0.04859749339244607, 0.04859749339244607, 0.04842697587177083, 0.04842697587177083, 0.04859749339244607, 0.04876801091312132, 0.0486827521527837, 0.0486827521527837, 0.04876801091312132, 0.04876801091312132, 0.0486827521527837, 0.04876801091312132, 0.04893852843379657, 0.04876801091312132, 0.0486827521527837, 0.0486827521527837, 0.04885326967345895, 0.04876801091312132, 0.04893852843379657, 0.04876801091312132, 0.04876801091312132, 0.0486827521527837, 0.04876801091312132, 0.04876801091312132, 0.0483417171114332, 0.04842697587177083, 0.04851223463210845, 0.04851223463210845, 0.04842697587177083, 0.0483417171114332, 0.04842697587177083, 0.04859749339244607, 0.04851223463210845, 0.04851223463210845, 0.04859749339244607, 0.04859749339244607, 0.04851223463210845, 0.0483417171114332, 0.04851223463210845, 0.04851223463210845, 0.0483417171114332, 0.04825645835109558, 0.04808594083042032, 0.04825645835109558, 0.04825645835109558, 0.04825645835109558, 0.04825645835109558, 0.0483417171114332, 0.04825645835109558, 0.04842697587177083, 0.0483417171114332, 0.04851223463210845, 0.04851223463210845, 0.04859749339244607, 0.04859749339244607, 0.04859749339244607, 0.04859749339244607, 0.04859749339244607, 0.04851223463210845, 0.04859749339244607, 0.0486827521527837, 0.04859749339244607, 0.04876801091312132, 0.04876801091312132, 0.04885326967345895, 0.04893852843379657, 0.04910904595447182, 0.04919430471480945, 0.04910904595447182, 0.04902378719413419, 0.04902378719413419, 0.04893852843379657, 0.04910904595447182, 0.04893852843379657, 0.04876801091312132, 0.04902378719413419, 0.04885326967345895, 0.04893852843379657, 0.04902378719413419, 0.04885326967345895, 0.04893852843379657, 0.04893852843379657, 0.04885326967345895, 0.04893852843379657, 0.04885326967345895, 0.04885326967345895, 0.04893852843379657]), ('logloss', [0.388483770589957, 0.3672322924420783, 0.3509481034424006, 0.3328954863781026, 0.31397002650700684, 0.3017171401953248, 0.29269187761949794, 0.2843878452922339, 0.2726277510975163, 0.2642516620620426, 0.25401286421168523, 0.24846882326595923, 0.24370915405925495, 0.23927156243945583, 0.23311282553859242, 0.22820091215638424, 0.22453300719236674, 0.22139909744843292, 0.21883949441508938, 0.2140493114632214, 0.21020677100852106, 0.20715682205997146, 0.20611100074995437, 0.20318786906784672, 0.20150022142102944, 0.199184194443496, 0.1973782825221851, 0.19584597088809447, 0.19401850466945103, 0.19293296070157548, 0.19169368127341724, 0.19074396434505966, 0.18945016395178346, 0.18801309481045278, 0.18701345542585052, 0.18645705440192742, 0.18544214922186472, 0.18444349033333554, 0.18350897456288798, 0.182807421486106, 0.1816704914641432, 0.1809728645566101, 0.18011074770726018, 0.1794112512462034, 0.17883208246447602, 0.17848490789225052, 0.17813217113152946, 0.1764840585984698, 0.1756170655793915, 0.17500328764929873, 0.1743385794639072, 0.1738978281212763, 0.17363765608402276, 0.17335269085456106, 0.17298857928556635, 0.1725433020530446, 0.17198581812270122, 0.17149512826571708, 0.1711917686426131, 0.1708629754032809, 0.17062886534386562, 0.1702213247395537, 0.16997610824059553, 0.16961354321370758, 0.16946830772257618, 0.16919098421609682, 0.16906157110554348, 0.1686848195665371, 0.16847793032773797, 0.16826335867957076, 0.16790993669866244, 0.16657327582955808, 0.16637444632027912, 0.16597426716214916, 0.16586389834956855, 0.16562825110732624, 0.16554026277661985, 0.16521229103608537, 0.16506872797574512, 0.16498524387110913, 0.16469326860002256, 0.16456535286931315, 0.16402863005867552, 0.16376439146976624, 0.16370772574169096, 0.16321552137530462, 0.16316920044325617, 0.1630567840836533, 0.16295637512918296, 0.1627195742949116, 0.16262332008978994, 0.16229514581107157, 0.1620314648843971, 0.16201927263627908, 0.16195735269080944, 0.16164320786010938, 0.16159503375300877, 0.1615316158397527, 0.16091686151417736, 0.16025884648666405, 0.16022956217090073, 0.16002897926619755, 0.15975328733584795, 0.1597460564164821, 0.15971910153397406, 0.15968962433871697, 0.15941665275407493, 0.159193762634392, 0.1590785446597677, 0.15912354886008562, 0.15874135781493362, 0.15866450517758368, 0.15848732672015411, 0.15846792028043244, 0.15840943402871102, 0.15828992747173276, 0.15822683596638426, 0.1582027412470205, 0.15817118546822678, 0.15819178761426975, 0.1581407203677547, 0.15811672448785272, 0.15810049999349274, 0.15807559904962398, 0.15799939386546097, 0.15798957500447391, 0.1579075584415386, 0.15793430854633927, 0.15781940966861333, 0.15771239081191443, 0.15777981318681011, 0.15774458161492919, 0.15768915731449634, 0.1576276487715285, 0.157543385476985, 0.15710220002407554, 0.15708943003207973, 0.1570102059688481, 0.15700307324131002, 0.1568971565930421, 0.1567177186326014, 0.1566056935488604, 0.1564888805018928, 0.15659478431941856, 0.1566308931437015, 0.15658640055671846, 0.15653615975171933, 0.15648609031616817, 0.15643623556250005, 0.1562900652725724, 0.15627384523607357, 0.15600057525608854, 0.1558601894466217, 0.15578276712668604, 0.15557177937945116, 0.15554196188780905, 0.15553651300765686, 0.15538582953876573, 0.15537471375681883, 0.1553723185995264, 0.15526259463057196, 0.15511303681185892, 0.15501167516883324, 0.1549551505443686, 0.15480452153903193, 0.1547447397813045, 0.1547386545940898, 0.15471755052182104, 0.15454882676900958, 0.15436559000365221, 0.15448649655569546, 0.1544973644162991, 0.1544304875738991, 0.1542979794152403, 0.1542285009143151, 0.15402552063781302, 0.1538320056266386, 0.1537915228995483, 0.15363657192378902, 0.15361379476574527, 0.15350724525036155, 0.15349174364179846, 0.15351456714231634, 0.15351182734715088, 0.15361350832706425, 0.1535568424231427, 0.15356992146712853, 0.1535253503636064, 0.15347710242251966, 0.15345042207272971, 0.15342651794620532, 0.15340504481123024, 0.15342003042498334, 0.15328360849059533, 0.1532626803580883, 0.15318038772907563, 0.15321523683025276, 0.15321072027762705, 0.15320927267796086, 0.15322396599236168, 0.15291409808643153, 0.1528656266385322, 0.15288198502293396, 0.15289653367099248, 0.15270006301492653, 0.1525542241085771, 0.15242238318726087, 0.15240539233346256, 0.15229855829445915, 0.1521850553625196, 0.15210304212969128, 0.1519650348614877, 0.15195743224783276, 0.15200175552431558, 0.15186338674949604, 0.15192959016550195, 0.15196995786592188, 0.15196764494978146, 0.15195685486647617, 0.15194685720653228, 0.1517452000686878, 0.15185888219909996, 0.1518387680440963, 0.15167600536407186, 0.15165690901157952, 0.15167458410947884, 0.15155917495802299, 0.1515974395909621, 0.15157187037429562, 0.15149469591837147, 0.15148039977588793, 0.15153013876658036, 0.15146600503191956, 0.15151169900303404, 0.1514547176844802, 0.15145295790626265, 0.15140508593209048, 0.15139391443699782, 0.15147160923256983, 0.15146019409080397, 0.15143723310455773, 0.15147055867012543, 0.15120447483126684, 0.1512182215599565, 0.1511528188525564, 0.1511205933102553, 0.15115706369902931, 0.15112191340394102, 0.15110519586586216, 0.15113997537585766, 0.15124165463175804, 0.15121630598326216, 0.15126677117442566, 0.15112712467970052, 0.15119497214889693, 0.15127634777779791, 0.1512149172719339, 0.15124286908428977, 0.15119423829774867, 0.15120068728520136, 0.15126646973029825, 0.15122512741728145, 0.15123697036590986, 0.15119100439374575, 0.15118523909650194, 0.15120279591166635, 0.1511952345418277, 0.15117611490494098, 0.15116225566459085, 0.1511883188937859, 0.15113209712619532, 0.15104238810796627, 0.15105403676746837, 0.15096302579954374, 0.15098277394200388, 0.15086912702286065, 0.15088243157498973, 0.15092061405893198, 0.150837806257148, 0.15087666094991992, 0.1508306189941432, 0.15087989960414983, 0.15090983547616446, 0.15084740735778926, 0.15078695854719323, 0.150657136117557, 0.15072925642182594, 0.15070947017597291, 0.1506600285577897, 0.1506480989647739, 0.15070675918807236, 0.1506665993936531, 0.15063945683137603, 0.15069972535075096, 0.15075718875229757, 0.15071610398868054, 0.15072710760018823, 0.15072456395924086, 0.15057603999958816, 0.15052250353640176])])}\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                           Feature  Importance\n",
      "20                    loan_grade_D    0.176359\n",
      "10      person_home_ownership_RENT    0.092828\n",
      "5              loan_percent_income    0.076535\n",
      "9        person_home_ownership_OWN    0.060660\n",
      "21                    loan_grade_E    0.056858\n",
      "7   person_home_ownership_MORTGAGE    0.051386\n",
      "11   loan_intent_DEBTCONSOLIDATION    0.042608\n",
      "17                    loan_grade_A    0.042399\n",
      "19                    loan_grade_C    0.040823\n",
      "4                    loan_int_rate    0.040051\n",
      "24     cb_person_default_on_file_N    0.035946\n",
      "25     cb_person_default_on_file_Y    0.035711\n",
      "14             loan_intent_MEDICAL    0.034473\n",
      "13     loan_intent_HOMEIMPROVEMENT    0.032933\n",
      "1                    person_income    0.032689\n",
      "16             loan_intent_VENTURE    0.020622\n",
      "3                        loan_amnt    0.019059\n",
      "2                person_emp_length    0.017998\n",
      "12           loan_intent_EDUCATION    0.015370\n",
      "18                    loan_grade_B    0.012589\n",
      "22                    loan_grade_F    0.012216\n",
      "15            loan_intent_PERSONAL    0.011899\n",
      "0                       person_age    0.011793\n",
      "6       cb_person_cred_hist_length    0.009636\n",
      "8      person_home_ownership_OTHER    0.009502\n",
      "23                    loan_grade_G    0.007055\n"
     ]
    }
   ],
   "source": [
    "xgb_boost.save_model(\"categorical-model.json\")\n",
    "results = xgb_boost.evals_result()\n",
    "print(results)\n",
    "epochs = len(results['validation_0']['error'])\n",
    "x_axis = range(0, epochs)\n",
    "# plot log loss\n",
    "import matplotlib.pyplot as pyplot\n",
    "fig, ax = pyplot.subplots()\n",
    "ax.plot(x_axis, results['validation_0']['logloss'], label='Train')\n",
    "ax.plot(x_axis, results['validation_1']['logloss'], label='Test')\n",
    "ax.legend()\n",
    "pyplot.ylabel('Log Loss')\n",
    "pyplot.title('XGBoost Log Loss')\n",
    "pyplot.show()\n",
    "# plot classification error\n",
    "fig, ax = pyplot.subplots()\n",
    "ax.plot(x_axis, results['validation_0']['error'], label='Train')\n",
    "ax.plot(x_axis, results['validation_1']['error'], label='Test')\n",
    "ax.legend()\n",
    "pyplot.ylabel('Classification Error')\n",
    "pyplot.title('XGBoost Classification Error')\n",
    "pyplot.show()\n",
    "\n",
    "feature_importances = xgb_boost.feature_importances_\n",
    "\n",
    "# Create a DataFrame with feature names and importance values\n",
    "feature_importance_df = pd.DataFrame({\n",
    "    'Feature': X_train.columns,\n",
    "    'Importance': feature_importances\n",
    "})\n",
    "\n",
    "# Sort the DataFrame by importance in descending order\n",
    "feature_importance_df = feature_importance_df.sort_values(by='Importance', ascending=False)\n",
    "\n",
    "# Print the feature importance\n",
    "print(feature_importance_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_curve, auc\n",
    "\n",
    "\n",
    "# Get the predicted probabilities on the test set\n",
    "y_pred_prob = xgb_boost.predict_proba(X_test)[:, 1]\n",
    "\n",
    "# Calculate the false positive rates and true positive rates\n",
    "fpr, tpr, thresholds = roc_curve(y_test, y_pred_prob)\n",
    "\n",
    "# Calculate the AUC\n",
    "roc_auc = auc(fpr, tpr)\n",
    "\n",
    "# Plot the ROC curve\n",
    "import matplotlib.pyplot as plt\n",
    "plt.figure(figsize=(8, 8))\n",
    "plt.plot(fpr, tpr, label='ROC Curve (AUC = %0.2f)' % roc_auc, color='darkorange', lw=2)\n",
    "plt.plot([0, 1], [0, 1], color='navy', lw=2, linestyle='--')\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('ROC Curve')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "### Random forest with defualt settings  gave 0.85 score on test , Now we choose best model from grid search\n",
    "\n",
    "\n",
    "# Apply the same preprocessing steps as before\n",
    "categorical_columns = ['person_home_ownership', 'loan_intent', 'loan_grade', 'cb_person_default_on_file']\n",
    "test_data_encoded = pd.get_dummies(test, columns=categorical_columns)\n",
    "test_data_encoded = test_data_encoded.drop('id', axis= 1)\n",
    "print(test.columns)\n",
    "# Make sure the test data has the same columns as the training data\n",
    "\n",
    "print(X.columns)\n",
    "\n",
    "skim(test_data_encoded)\n",
    "# test_data_encoded = test_data_encoded[X.columns]\n",
    "\n",
    "# Make predictions on the test data\n",
    "predictions = xgb_boost.predict(test_data_encoded)\n",
    "\n",
    "# prediction39 = xgb_boost.predict_proba(test_data_encoded)[:, 1]\n",
    "\n",
    "# use the threshold value to convert probabilities to class labels\n",
    "predictions = (predictions >= best_threshold).astype(int)\n",
    "\n",
    "print(best_threshold)\n",
    "\n",
    "\n",
    "# Save the predictions to a submission file\n",
    "submission_df = pd.DataFrame({'id': test['id'], 'loan_status': predictions})\n",
    "submission_df.to_csv('submission_xgb_predproba.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "14337    0\n",
      "17117    0\n",
      "32590    0\n",
      "55811    0\n",
      "40120    0\n",
      "        ..\n",
      "54343    0\n",
      "38158    0\n",
      "860      0\n",
      "15795    0\n",
      "56422    0\n",
      "Name: loan_status, Length: 46916, dtype: int64\n",
      "Epoch 1/100\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\rushi\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\keras\\src\\layers\\core\\dense.py:87: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
      "  super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1ms/step - accuracy: 0.8077 - loss: 147.9066 - val_accuracy: 0.8850 - val_loss: 2.6936\n",
      "Epoch 2/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 910us/step - accuracy: 0.8023 - loss: 5.4733 - val_accuracy: 0.8601 - val_loss: 10.1877\n",
      "Epoch 3/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 792us/step - accuracy: 0.8082 - loss: 5.6678 - val_accuracy: 0.8655 - val_loss: 5.5272\n",
      "Epoch 4/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 830us/step - accuracy: 0.8181 - loss: 4.8334 - val_accuracy: 0.8865 - val_loss: 2.1477\n",
      "Epoch 5/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 753us/step - accuracy: 0.8134 - loss: 4.5831 - val_accuracy: 0.8858 - val_loss: 3.0675\n",
      "Epoch 6/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 766us/step - accuracy: 0.8182 - loss: 4.5640 - val_accuracy: 0.8858 - val_loss: 4.9781\n",
      "Epoch 7/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 761us/step - accuracy: 0.8185 - loss: 4.2189 - val_accuracy: 0.1925 - val_loss: 6.5612\n",
      "Epoch 8/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 760us/step - accuracy: 0.8178 - loss: 5.3128 - val_accuracy: 0.8924 - val_loss: 4.9853\n",
      "Epoch 9/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 807us/step - accuracy: 0.8351 - loss: 4.4008 - val_accuracy: 0.9045 - val_loss: 1.3827\n",
      "Epoch 10/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 756us/step - accuracy: 0.8344 - loss: 3.7756 - val_accuracy: 0.8647 - val_loss: 0.7381\n",
      "Epoch 11/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 760us/step - accuracy: 0.8358 - loss: 3.4302 - val_accuracy: 0.8600 - val_loss: 13.3471\n",
      "Epoch 12/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 787us/step - accuracy: 0.8404 - loss: 3.7836 - val_accuracy: 0.8895 - val_loss: 1.6092\n",
      "Epoch 13/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 768us/step - accuracy: 0.8389 - loss: 3.3729 - val_accuracy: 0.8982 - val_loss: 1.2125\n",
      "Epoch 14/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 921us/step - accuracy: 0.8461 - loss: 2.5625 - val_accuracy: 0.8959 - val_loss: 1.5111\n",
      "Epoch 15/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 767us/step - accuracy: 0.8515 - loss: 2.9568 - val_accuracy: 0.8277 - val_loss: 2.2340\n",
      "Epoch 16/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 812us/step - accuracy: 0.8450 - loss: 2.4048 - val_accuracy: 0.8662 - val_loss: 1.8900\n",
      "Epoch 17/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 790us/step - accuracy: 0.8490 - loss: 1.9319 - val_accuracy: 0.1442 - val_loss: 10.3055\n",
      "Epoch 18/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 796us/step - accuracy: 0.8433 - loss: 1.8898 - val_accuracy: 0.9065 - val_loss: 0.6973\n",
      "Epoch 19/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 774us/step - accuracy: 0.8519 - loss: 1.4615 - val_accuracy: 0.8743 - val_loss: 0.4557\n",
      "Epoch 20/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 810us/step - accuracy: 0.8725 - loss: 0.4249 - val_accuracy: 0.8733 - val_loss: 0.3745\n",
      "Epoch 21/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 843us/step - accuracy: 0.8743 - loss: 0.3698 - val_accuracy: 0.8735 - val_loss: 0.3684\n",
      "Epoch 22/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 766us/step - accuracy: 0.8743 - loss: 0.3653 - val_accuracy: 0.8644 - val_loss: 0.3669\n",
      "Epoch 23/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 773us/step - accuracy: 0.8701 - loss: 0.3720 - val_accuracy: 0.8750 - val_loss: 0.3597\n",
      "Epoch 24/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 767us/step - accuracy: 0.8723 - loss: 0.3615 - val_accuracy: 0.8634 - val_loss: 0.3638\n",
      "Epoch 25/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 808us/step - accuracy: 0.8700 - loss: 0.3637 - val_accuracy: 0.8782 - val_loss: 0.3646\n",
      "Epoch 26/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 764us/step - accuracy: 0.8722 - loss: 0.3595 - val_accuracy: 0.8633 - val_loss: 0.3667\n",
      "Epoch 27/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 762us/step - accuracy: 0.8705 - loss: 0.3624 - val_accuracy: 0.8781 - val_loss: 0.3602\n",
      "Epoch 28/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 782us/step - accuracy: 0.8718 - loss: 0.3587 - val_accuracy: 0.8711 - val_loss: 0.3516\n",
      "Epoch 29/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 756us/step - accuracy: 0.8696 - loss: 0.3622 - val_accuracy: 0.8785 - val_loss: 0.3625\n",
      "Epoch 30/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 787us/step - accuracy: 0.8732 - loss: 0.3581 - val_accuracy: 0.8695 - val_loss: 0.3503\n",
      "Epoch 31/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 765us/step - accuracy: 0.8722 - loss: 0.3559 - val_accuracy: 0.8778 - val_loss: 0.3482\n",
      "Epoch 32/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 782us/step - accuracy: 0.8705 - loss: 0.3618 - val_accuracy: 0.8605 - val_loss: 0.3935\n",
      "Epoch 33/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 812us/step - accuracy: 0.8702 - loss: 0.3576 - val_accuracy: 0.8725 - val_loss: 0.3447\n",
      "Epoch 34/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 766us/step - accuracy: 0.8698 - loss: 0.3584 - val_accuracy: 0.8721 - val_loss: 0.3470\n",
      "Epoch 35/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 784us/step - accuracy: 0.8754 - loss: 0.3480 - val_accuracy: 0.8764 - val_loss: 0.3355\n",
      "Epoch 36/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 762us/step - accuracy: 0.8730 - loss: 0.3595 - val_accuracy: 0.8842 - val_loss: 0.3286\n",
      "Epoch 37/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 774us/step - accuracy: 0.8748 - loss: 0.3455 - val_accuracy: 0.8601 - val_loss: 0.4011\n",
      "Epoch 38/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 807us/step - accuracy: 0.8779 - loss: 0.3368 - val_accuracy: 0.8654 - val_loss: 0.3696\n",
      "Epoch 39/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 813us/step - accuracy: 0.8732 - loss: 0.3580 - val_accuracy: 0.8600 - val_loss: 0.4176\n",
      "Epoch 40/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 844us/step - accuracy: 0.8567 - loss: 0.4160 - val_accuracy: 0.8601 - val_loss: 0.4047\n",
      "Epoch 41/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 766us/step - accuracy: 0.8732 - loss: 0.3566 - val_accuracy: 0.8680 - val_loss: 0.3661\n",
      "Epoch 42/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 771us/step - accuracy: 0.8806 - loss: 0.3319 - val_accuracy: 0.8864 - val_loss: 0.3080\n",
      "Epoch 43/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 825us/step - accuracy: 0.8836 - loss: 0.3185 - val_accuracy: 0.8940 - val_loss: 0.3100\n",
      "Epoch 44/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 807us/step - accuracy: 0.8833 - loss: 0.3212 - val_accuracy: 0.8880 - val_loss: 0.3043\n",
      "Epoch 45/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 839us/step - accuracy: 0.8781 - loss: 0.3396 - val_accuracy: 0.8975 - val_loss: 0.2846\n",
      "Epoch 46/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 744us/step - accuracy: 0.8864 - loss: 0.3104 - val_accuracy: 0.8985 - val_loss: 0.2805\n",
      "Epoch 47/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 823us/step - accuracy: 0.8868 - loss: 0.3014 - val_accuracy: 0.8887 - val_loss: 0.2825\n",
      "Epoch 48/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 778us/step - accuracy: 0.8875 - loss: 0.3042 - val_accuracy: 0.9047 - val_loss: 0.2751\n",
      "Epoch 49/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 805us/step - accuracy: 0.8910 - loss: 0.2913 - val_accuracy: 0.8727 - val_loss: 0.3418\n",
      "Epoch 50/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 802us/step - accuracy: 0.8854 - loss: 0.3059 - val_accuracy: 0.9018 - val_loss: 0.2711\n",
      "Epoch 51/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 802us/step - accuracy: 0.8934 - loss: 0.2840 - val_accuracy: 0.9050 - val_loss: 0.2695\n",
      "Epoch 52/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 797us/step - accuracy: 0.8945 - loss: 0.2848 - val_accuracy: 0.9003 - val_loss: 0.2806\n",
      "Epoch 53/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 798us/step - accuracy: 0.8952 - loss: 0.2829 - val_accuracy: 0.9047 - val_loss: 0.2676\n",
      "Epoch 54/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 781us/step - accuracy: 0.8960 - loss: 0.2802 - val_accuracy: 0.8995 - val_loss: 0.2726\n",
      "Epoch 55/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 799us/step - accuracy: 0.8975 - loss: 0.2804 - val_accuracy: 0.8950 - val_loss: 0.2590\n",
      "Epoch 56/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 767us/step - accuracy: 0.9020 - loss: 0.2671 - val_accuracy: 0.9066 - val_loss: 0.2553\n",
      "Epoch 57/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 791us/step - accuracy: 0.8978 - loss: 0.2761 - val_accuracy: 0.9084 - val_loss: 0.2524\n",
      "Epoch 58/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 813us/step - accuracy: 0.8984 - loss: 0.2797 - val_accuracy: 0.8947 - val_loss: 0.2619\n",
      "Epoch 59/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 762us/step - accuracy: 0.8969 - loss: 0.2758 - val_accuracy: 0.9106 - val_loss: 0.2476\n",
      "Epoch 60/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 757us/step - accuracy: 0.8995 - loss: 0.2736 - val_accuracy: 0.8933 - val_loss: 0.2646\n",
      "Epoch 61/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 722us/step - accuracy: 0.9014 - loss: 0.2689 - val_accuracy: 0.8777 - val_loss: 0.3453\n",
      "Epoch 62/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 848us/step - accuracy: 0.8954 - loss: 0.2782 - val_accuracy: 0.9019 - val_loss: 0.2738\n",
      "Epoch 63/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 742us/step - accuracy: 0.8993 - loss: 0.2719 - val_accuracy: 0.9113 - val_loss: 0.2487\n",
      "Epoch 64/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 745us/step - accuracy: 0.9044 - loss: 0.2601 - val_accuracy: 0.9076 - val_loss: 0.2444\n",
      "Epoch 65/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 727us/step - accuracy: 0.8993 - loss: 0.2682 - val_accuracy: 0.9037 - val_loss: 0.2464\n",
      "Epoch 66/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 723us/step - accuracy: 0.9016 - loss: 0.2687 - val_accuracy: 0.8369 - val_loss: 0.4114\n",
      "Epoch 67/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 745us/step - accuracy: 0.9049 - loss: 0.2608 - val_accuracy: 0.8945 - val_loss: 0.2893\n",
      "Epoch 68/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 713us/step - accuracy: 0.9007 - loss: 0.2749 - val_accuracy: 0.9086 - val_loss: 0.2552\n",
      "Epoch 69/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 748us/step - accuracy: 0.9028 - loss: 0.2665 - val_accuracy: 0.9105 - val_loss: 0.2652\n",
      "Epoch 70/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 735us/step - accuracy: 0.9031 - loss: 0.2628 - val_accuracy: 0.9078 - val_loss: 0.2422\n",
      "Epoch 71/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 741us/step - accuracy: 0.9058 - loss: 0.2626 - val_accuracy: 0.9191 - val_loss: 0.2399\n",
      "Epoch 72/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 753us/step - accuracy: 0.9065 - loss: 0.2624 - val_accuracy: 0.8998 - val_loss: 0.2521\n",
      "Epoch 73/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 726us/step - accuracy: 0.9088 - loss: 0.2513 - val_accuracy: 0.9076 - val_loss: 0.2423\n",
      "Epoch 74/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 760us/step - accuracy: 0.9050 - loss: 0.2690 - val_accuracy: 0.8912 - val_loss: 0.2954\n",
      "Epoch 75/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 714us/step - accuracy: 0.9059 - loss: 0.2620 - val_accuracy: 0.9135 - val_loss: 0.2508\n",
      "Epoch 76/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 776us/step - accuracy: 0.9059 - loss: 0.2621 - val_accuracy: 0.8941 - val_loss: 0.2885\n",
      "Epoch 77/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 753us/step - accuracy: 0.9075 - loss: 0.2564 - val_accuracy: 0.9142 - val_loss: 0.2401\n",
      "Epoch 78/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 756us/step - accuracy: 0.9082 - loss: 0.2592 - val_accuracy: 0.9129 - val_loss: 0.2352\n",
      "Epoch 79/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 729us/step - accuracy: 0.9048 - loss: 0.2643 - val_accuracy: 0.9139 - val_loss: 0.2501\n",
      "Epoch 80/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 713us/step - accuracy: 0.9094 - loss: 0.2530 - val_accuracy: 0.9191 - val_loss: 0.2595\n",
      "Epoch 81/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 735us/step - accuracy: 0.9084 - loss: 0.2569 - val_accuracy: 0.9187 - val_loss: 0.2386\n",
      "Epoch 82/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 722us/step - accuracy: 0.9088 - loss: 0.2541 - val_accuracy: 0.9187 - val_loss: 0.2482\n",
      "Epoch 83/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 739us/step - accuracy: 0.9077 - loss: 0.2575 - val_accuracy: 0.9144 - val_loss: 0.2395\n",
      "Epoch 84/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 727us/step - accuracy: 0.9080 - loss: 0.2596 - val_accuracy: 0.9172 - val_loss: 0.2330\n",
      "Epoch 85/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 753us/step - accuracy: 0.9087 - loss: 0.2537 - val_accuracy: 0.9194 - val_loss: 0.2360\n",
      "Epoch 86/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 720us/step - accuracy: 0.9087 - loss: 0.2525 - val_accuracy: 0.9163 - val_loss: 0.2365\n",
      "Epoch 87/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 746us/step - accuracy: 0.9063 - loss: 0.2561 - val_accuracy: 0.9037 - val_loss: 0.2634\n",
      "Epoch 88/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 722us/step - accuracy: 0.9077 - loss: 0.2581 - val_accuracy: 0.9179 - val_loss: 0.2381\n",
      "Epoch 89/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 738us/step - accuracy: 0.9104 - loss: 0.2543 - val_accuracy: 0.9167 - val_loss: 0.2376\n",
      "Epoch 90/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 719us/step - accuracy: 0.9098 - loss: 0.2539 - val_accuracy: 0.9068 - val_loss: 0.2494\n",
      "Epoch 91/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 744us/step - accuracy: 0.9057 - loss: 0.2630 - val_accuracy: 0.9209 - val_loss: 0.2311\n",
      "Epoch 92/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 724us/step - accuracy: 0.9144 - loss: 0.2456 - val_accuracy: 0.9132 - val_loss: 0.2485\n",
      "Epoch 93/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 730us/step - accuracy: 0.9131 - loss: 0.2492 - val_accuracy: 0.9148 - val_loss: 0.2444\n",
      "Epoch 94/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 709us/step - accuracy: 0.9106 - loss: 0.2565 - val_accuracy: 0.8961 - val_loss: 0.3121\n",
      "Epoch 95/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 746us/step - accuracy: 0.9130 - loss: 0.2479 - val_accuracy: 0.9012 - val_loss: 0.2701\n",
      "Epoch 96/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 755us/step - accuracy: 0.9086 - loss: 0.2523 - val_accuracy: 0.9175 - val_loss: 0.2368\n",
      "Epoch 97/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 730us/step - accuracy: 0.9127 - loss: 0.2524 - val_accuracy: 0.9042 - val_loss: 0.2725\n",
      "Epoch 98/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 757us/step - accuracy: 0.9126 - loss: 0.2483 - val_accuracy: 0.9089 - val_loss: 0.2641\n",
      "Epoch 99/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 713us/step - accuracy: 0.9133 - loss: 0.2466 - val_accuracy: 0.9229 - val_loss: 0.2261\n",
      "Epoch 100/100\n",
      "\u001b[1m1467/1467\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 771us/step - accuracy: 0.9131 - loss: 0.2457 - val_accuracy: 0.8947 - val_loss: 0.3251\n",
      "[[0]\n",
      " [0]\n",
      " [1]\n",
      " ...\n",
      " [0]\n",
      " [0]\n",
      " [0]]\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "from keras.callbacks import EarlyStopping\n",
    "\n",
    "# assume you have your training and test data\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "# skim(X)\n",
    "print(y_train)\n",
    "# define the neural network model\n",
    "model = Sequential()\n",
    "model.add(Dense(16, activation='relu', input_shape=(X_train.shape[1],)))\n",
    "model.add(Dense(8, activation='relu'))\n",
    "model.add(Dense(1, activation='sigmoid'))\n",
    "\n",
    "# compile the model\n",
    "model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])\n",
    "\n",
    "# set up early stopping to stop training when the validation loss stops improving\n",
    "# early_stopping = EarlyStopping(monitor='val_loss', patience=10, verbose=1)\n",
    "\n",
    "# train the model\n",
    "history = model.fit(X_train, y_train, epochs=100, batch_size=32, validation_data=(X_test, y_test))\n",
    "\n",
    "# make predictions on the test data\n",
    "y_pred = model.predict(X_test, verbose=0)\n",
    "y_pred = (y_pred >= 0.5).astype(int)\n",
    "\n",
    "print(y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.8947054309830335\n",
      "Classification Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.92      0.94     10087\n",
      "           1       0.60      0.77      0.67      1642\n",
      "\n",
      "    accuracy                           0.89     11729\n",
      "   macro avg       0.78      0.84      0.80     11729\n",
      "weighted avg       0.91      0.89      0.90     11729\n",
      "\n",
      "Confusion Matrix:\n",
      "[[9231  856]\n",
      " [ 379 1263]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "accuracy = accuracy_score(y_test, y_pred)\n",
    "print(\"Accuracy:\", accuracy)\n",
    "\n",
    "print(\"Classification Report:\")\n",
    "print(classification_report(y_test, y_pred))\n",
    "\n",
    "print(\"Confusion Matrix:\")\n",
    "print(confusion_matrix(y_test, y_pred))\n",
    "\n",
    "# plot the training and validation loss\n",
    "import matplotlib.pyplot as plt\n",
    "plt.plot(history.history['loss'], label='Training Loss')\n",
    "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "# plot the training and validation accuracy\n",
    "plt.plot(history.history['accuracy'], label='Training Accuracy')\n",
    "plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1 0 1 ... 0 1 1]\n"
     ]
    }
   ],
   "source": [
    "# make predictions on the test data\n",
    "y_pred = model.predict(test_data_encoded, verbose=0)\n",
    "y_pred = (y_pred >= 0.5).astype(int)\n",
    "y_pred = y_pred.flatten()\n",
    "print(y_pred)\n",
    "\n",
    "# create a DataFrame with the test data and predictions\n",
    "\n",
    "\n",
    "# write the DataFrame to a CSV file\n",
    "\n",
    "# Save the predictions to a submission file\n",
    "submission_df = pd.DataFrame({'id': test['id'], 'loan_status': y_pred})\n",
    "submission_df.to_csv('neural_network_small.csv', index=False)"
   ]
  }
 ],
 "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": "3.11.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}