HR_data_science.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"provenance":[{"file_id":"1aX-SKVFkQDK5lPHjKJlLOtKsNZygdwCW","timestamp":1666092118390}],"collapsed_sections":["ipUkHTqivWe-","2JwwcTIHvmne","Ls-Ie4_QvoJt","vzLyL83K2tF5","LSGb3NHL3_r8","tmlnJGJM6E9j","ezWNaoNp7JZd","R07GnAb971IN","74skVGQU8U0S","gx6xykIe8tn5","oBz_sFCB8zU8","z22RYo8e9maW"]},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"}},"cells":[{"cell_type":"markdown","source":["# **0. Setting up the defaults**\n"],"metadata":{"id":"0SxGr08hd9e8"}},{"cell_type":"markdown","source":["## **Import libraries**\n","\n","First we need to import the libraries that we will be using. The used libraries and their function are listed below. \n","\n",">Pandas: used for dataset manipulation\n","\n",">Numpy: used for array manipulation\n","\n",">Matplotlib: used for visualisation\n","\n",">Seaborn: used for visualisation\n","\n",">Sklearn: used for fitting models\n","\n",">Roc_curve, Roc_auc_score: Curve of performance of a classifier based on how daring it is in calling things positive. More info in this [video](https://https://www.youtube.com/watch?v=UrqPyE4H2bI)\n","\n",">itertools: used for feature set search"],"metadata":{"id":"fGS9i3kidPqq"}},{"cell_type":"code","execution_count":1,"metadata":{"id":"F5ZrWkejc8X9","executionInfo":{"status":"ok","timestamp":1666352631955,"user_tz":-120,"elapsed":1828,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"outputs":[],"source":["import pandas as pd\n","import numpy as np\n","import matplotlib.pyplot as plt\n","import sklearn\n","import seaborn as sns\n","from sklearn.tree import DecisionTreeClassifier\n","from sklearn.ensemble import RandomForestClassifier\n","from sklearn.metrics import roc_curve, roc_auc_score, plot_confusion_matrix, confusion_matrix, accuracy_score\n","from sklearn.svm import SVC\n","from sklearn.model_selection import GridSearchCV, KFold, train_test_split\n","from itertools import combinations, chain"]},{"cell_type":"markdown","source":["# **Import dataset from drive**\n","Secondly, we're importing the dataset from drive, which can be found [here](https://drive.google.com/file/d/1C9eI8fUoEKVkbcRWCKIyK70XeI-H1XeD/view?usp=sharing)\n","\n","To do this we first need to enable access to drive. \n","Running this code will present a pop-up where you will need to grant access"],"metadata":{"id":"ZW3Sg3i3drtd"}},{"cell_type":"code","source":["from google.colab import drive\n","drive.mount('/content/drive')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"gBO9XqMAdNx1","executionInfo":{"status":"ok","timestamp":1666352651217,"user_tz":-120,"elapsed":15279,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"510dfafa-d9d9-4e2c-d3fa-22b6c35834d6"},"execution_count":2,"outputs":[{"output_type":"stream","name":"stdout","text":["Mounted at /content/drive\n"]}]},{"cell_type":"markdown","source":["Next we will grab the file from drive\n","\n","In order to be able to read it, we use the file ID from the URL with /uc?id= to get a direct download like which can be saved as a pandas dataframe."],"metadata":{"id":"aWstJj1Tex3c"}},{"cell_type":"code","source":["url='https://drive.google.com/file/d/1C9eI8fUoEKVkbcRWCKIyK70XeI-H1XeD/view?usp=sharing'\n","url='https://drive.google.com/uc?id=' + url.split('/')[-2]\n","init_df = pd.read_csv(url)"],"metadata":{"id":"FYYrsIeme50H","executionInfo":{"status":"ok","timestamp":1666352652101,"user_tz":-120,"elapsed":886,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":3,"outputs":[]},{"cell_type":"markdown","source":["In case you are running this locally, don't run the drive related blocks, but use the code block below"],"metadata":{"id":"ZEuSil5pgUUr"}},{"cell_type":"code","source":["#init_df = pd.read_csv('recruitmentdataset-2022-1.3.csv')"],"metadata":{"id":"WYV7O98igSM9","executionInfo":{"status":"ok","timestamp":1666352652104,"user_tz":-120,"elapsed":10,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":4,"outputs":[]},{"cell_type":"markdown","source":["When we've imported the dataset, let's save it under the df variable (dataframe)"],"metadata":{"id":"4jhQSTdIgdc-"}},{"cell_type":"code","source":["# df is a copy of the initial df so we don't have to redownload many times\n","df = init_df.copy()"],"metadata":{"id":"eEoBR4C2RxQ0","executionInfo":{"status":"ok","timestamp":1666352652104,"user_tz":-120,"elapsed":10,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":5,"outputs":[]},{"cell_type":"markdown","source":["Let's take an initial look at the data\n"],"metadata":{"id":"7xfK_HWhfMo3"}},{"cell_type":"code","source":["df"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":505},"id":"C8vjIRCGfbfA","executionInfo":{"status":"ok","timestamp":1666352652105,"user_tz":-120,"elapsed":10,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"cacf130f-9920-48ea-8319-4eef946b888b"},"execution_count":6,"outputs":[{"output_type":"execute_result","data":{"text/plain":["          Id  gender  age nationality     sport  ind-university_grade  \\\n","0     x8011e  female   24      German  Swimming                    70   \n","1     x6077a    male   26      German      Golf                    67   \n","2     x6006e  female   23       Dutch   Running                    67   \n","3     x2173b    male   24       Dutch   Cricket                    70   \n","4     x6241a  female   26      German      Golf                    59   \n","...      ...     ...  ...         ...       ...                   ...   \n","3995  x7640e  female   28       Dutch   Running                    63   \n","3996  x3310f  female   27       Dutch    Tennis                    62   \n","3997  x1202g    male   24     Belgian     Rugby                    60   \n","3998  x1263d  female   22       Dutch    Tennis                    66   \n","3999  x2396a    male   28       Dutch  Football                    58   \n","\n","      ind-debateclub  ind-programming_exp  ind-international_exp  \\\n","0              False                False                  False   \n","1              False                 True                  False   \n","2              False                 True                   True   \n","3              False                 True                  False   \n","4              False                False                  False   \n","...              ...                  ...                    ...   \n","3995           False                False                  False   \n","3996           False                False                  False   \n","3997            True                False                  False   \n","3998           False                 True                  False   \n","3999           False                False                  False   \n","\n","      ind-entrepeneur_exp  ind-languages  ind-exact_study ind-degree company  \\\n","0                   False              1             True        phd       A   \n","1                   False              2             True   bachelor       A   \n","2                   False              0             True     master       A   \n","3                   False              1             True     master       A   \n","4                   False              1            False     master       A   \n","...                   ...            ...              ...        ...     ...   \n","3995                False              0            False     master       D   \n","3996                 True              2             True   bachelor       D   \n","3997                 True              2            False   bachelor       D   \n","3998                False              1             True   bachelor       D   \n","3999                False              1            False   bachelor       D   \n","\n","      decision  \n","0         True  \n","1        False  \n","2        False  \n","3         True  \n","4         True  \n","...        ...  \n","3995     False  \n","3996     False  \n","3997     False  \n","3998     False  \n","3999     False  \n","\n","[4000 rows x 15 columns]"],"text/html":["\n","  <div id=\"df-22b8bc06-3d87-44ad-821a-e5c568b2985f\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>Id</th>\n","      <th>gender</th>\n","      <th>age</th>\n","      <th>nationality</th>\n","      <th>sport</th>\n","      <th>ind-university_grade</th>\n","      <th>ind-debateclub</th>\n","      <th>ind-programming_exp</th>\n","      <th>ind-international_exp</th>\n","      <th>ind-entrepeneur_exp</th>\n","      <th>ind-languages</th>\n","      <th>ind-exact_study</th>\n","      <th>ind-degree</th>\n","      <th>company</th>\n","      <th>decision</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>x8011e</td>\n","      <td>female</td>\n","      <td>24</td>\n","      <td>German</td>\n","      <td>Swimming</td>\n","      <td>70</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>True</td>\n","      <td>phd</td>\n","      <td>A</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>x6077a</td>\n","      <td>male</td>\n","      <td>26</td>\n","      <td>German</td>\n","      <td>Golf</td>\n","      <td>67</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>2</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>A</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>x6006e</td>\n","      <td>female</td>\n","      <td>23</td>\n","      <td>Dutch</td>\n","      <td>Running</td>\n","      <td>67</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>0</td>\n","      <td>True</td>\n","      <td>master</td>\n","      <td>A</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>x2173b</td>\n","      <td>male</td>\n","      <td>24</td>\n","      <td>Dutch</td>\n","      <td>Cricket</td>\n","      <td>70</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>True</td>\n","      <td>master</td>\n","      <td>A</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>x6241a</td>\n","      <td>female</td>\n","      <td>26</td>\n","      <td>German</td>\n","      <td>Golf</td>\n","      <td>59</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>False</td>\n","      <td>master</td>\n","      <td>A</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>3995</th>\n","      <td>x7640e</td>\n","      <td>female</td>\n","      <td>28</td>\n","      <td>Dutch</td>\n","      <td>Running</td>\n","      <td>63</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>0</td>\n","      <td>False</td>\n","      <td>master</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3996</th>\n","      <td>x3310f</td>\n","      <td>female</td>\n","      <td>27</td>\n","      <td>Dutch</td>\n","      <td>Tennis</td>\n","      <td>62</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>2</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3997</th>\n","      <td>x1202g</td>\n","      <td>male</td>\n","      <td>24</td>\n","      <td>Belgian</td>\n","      <td>Rugby</td>\n","      <td>60</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>2</td>\n","      <td>False</td>\n","      <td>bachelor</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3998</th>\n","      <td>x1263d</td>\n","      <td>female</td>\n","      <td>22</td>\n","      <td>Dutch</td>\n","      <td>Tennis</td>\n","      <td>66</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3999</th>\n","      <td>x2396a</td>\n","      <td>male</td>\n","      <td>28</td>\n","      <td>Dutch</td>\n","      <td>Football</td>\n","      <td>58</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>False</td>\n","      <td>bachelor</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>4000 rows × 15 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-22b8bc06-3d87-44ad-821a-e5c568b2985f')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-22b8bc06-3d87-44ad-821a-e5c568b2985f button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-22b8bc06-3d87-44ad-821a-e5c568b2985f');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":6}]},{"cell_type":"markdown","source":["Next up, it is time to get to work with this data. \n","The assignment consists out of 4 phases\n","> **Exploration:** We need to explore the dataset for information content, feature distribution and feature correlation. This means looking at the distribution of our features, learning the data type of the features and correlations\n","\n","\n","> **Data Pre-processing:** We need to make sure that we have the correct data sets that contain data of the population, our company and potentially the other companies. Then we want to make sure to create data sets with the features for each model and encode where needed. Some of this stuff can be done before the exploration, the rest will have to wait until we learn more of the data.\n","\n","\n","> **Model fitting:** We need to train 3 models consisting of\n","\n","\n","1.   Decision Trees with Random Forests to find our best indicators\n","2.   Support Vector for testing found indicators and comparison with\n","3.   Support Vector with descriptors\n","\n","\n","> **Result Analysis:** Visualize the results that we obtain from the models to interpret results and evaluate performance"],"metadata":{"id":"cDJFKVHifx_0"}},{"cell_type":"markdown","source":["# **1. Initial Data Pre-processing**"],"metadata":{"id":"VeA0i0v4hQDV"}},{"cell_type":"markdown","source":["Before we even start with data exploration, there's a few steps we can already take to tidy up out data. \n","\n","First off, we  know from *id : a unique identifier. Do not use in your algorithm* that we can drop the id feature from the dataset\n","\n","Then, we can create a couple of separate datasets for the steps that we've already seen. \n","\n","-Company dataset: Contains all the data for our company B\n","\n","-Population dataset: contains the population of A,B,C,D which we need to use for exploration\n","\n","-Indicator dataset: Contains only the indicator features and results for Model 2\n","\n","-Description dataset: Contains only the descriptor features and results for Model 3"],"metadata":{"id":"5ZL4RB4IhUja"}},{"cell_type":"markdown","source":["So let's first drop id from out dataset"],"metadata":{"id":"BSFDF_bZiPsU"}},{"cell_type":"code","source":["df = df.drop('Id', axis=1)"],"metadata":{"id":"Znin1ecvhT1K","executionInfo":{"status":"ok","timestamp":1666352652105,"user_tz":-120,"elapsed":9,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":7,"outputs":[]},{"cell_type":"markdown","source":["As you can see below here, we no longer have an Id feature"],"metadata":{"id":"Dw8Tp2tRihvj"}},{"cell_type":"code","source":["df"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":505},"id":"ox2s8NfMhBTe","executionInfo":{"status":"ok","timestamp":1666352652105,"user_tz":-120,"elapsed":9,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"36637dcc-fcf0-485c-b729-b3f62e5c4e89"},"execution_count":8,"outputs":[{"output_type":"execute_result","data":{"text/plain":["      gender  age nationality     sport  ind-university_grade  ind-debateclub  \\\n","0     female   24      German  Swimming                    70           False   \n","1       male   26      German      Golf                    67           False   \n","2     female   23       Dutch   Running                    67           False   \n","3       male   24       Dutch   Cricket                    70           False   \n","4     female   26      German      Golf                    59           False   \n","...      ...  ...         ...       ...                   ...             ...   \n","3995  female   28       Dutch   Running                    63           False   \n","3996  female   27       Dutch    Tennis                    62           False   \n","3997    male   24     Belgian     Rugby                    60            True   \n","3998  female   22       Dutch    Tennis                    66           False   \n","3999    male   28       Dutch  Football                    58           False   \n","\n","      ind-programming_exp  ind-international_exp  ind-entrepeneur_exp  \\\n","0                   False                  False                False   \n","1                    True                  False                False   \n","2                    True                   True                False   \n","3                    True                  False                False   \n","4                   False                  False                False   \n","...                   ...                    ...                  ...   \n","3995                False                  False                False   \n","3996                False                  False                 True   \n","3997                False                  False                 True   \n","3998                 True                  False                False   \n","3999                False                  False                False   \n","\n","      ind-languages  ind-exact_study ind-degree company  decision  \n","0                 1             True        phd       A      True  \n","1                 2             True   bachelor       A     False  \n","2                 0             True     master       A     False  \n","3                 1             True     master       A      True  \n","4                 1            False     master       A      True  \n","...             ...              ...        ...     ...       ...  \n","3995              0            False     master       D     False  \n","3996              2             True   bachelor       D     False  \n","3997              2            False   bachelor       D     False  \n","3998              1             True   bachelor       D     False  \n","3999              1            False   bachelor       D     False  \n","\n","[4000 rows x 14 columns]"],"text/html":["\n","  <div id=\"df-ed14b99e-a420-4e34-80b4-d38d11ded8da\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>gender</th>\n","      <th>age</th>\n","      <th>nationality</th>\n","      <th>sport</th>\n","      <th>ind-university_grade</th>\n","      <th>ind-debateclub</th>\n","      <th>ind-programming_exp</th>\n","      <th>ind-international_exp</th>\n","      <th>ind-entrepeneur_exp</th>\n","      <th>ind-languages</th>\n","      <th>ind-exact_study</th>\n","      <th>ind-degree</th>\n","      <th>company</th>\n","      <th>decision</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>female</td>\n","      <td>24</td>\n","      <td>German</td>\n","      <td>Swimming</td>\n","      <td>70</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>True</td>\n","      <td>phd</td>\n","      <td>A</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>male</td>\n","      <td>26</td>\n","      <td>German</td>\n","      <td>Golf</td>\n","      <td>67</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>2</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>A</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>female</td>\n","      <td>23</td>\n","      <td>Dutch</td>\n","      <td>Running</td>\n","      <td>67</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>0</td>\n","      <td>True</td>\n","      <td>master</td>\n","      <td>A</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>male</td>\n","      <td>24</td>\n","      <td>Dutch</td>\n","      <td>Cricket</td>\n","      <td>70</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>True</td>\n","      <td>master</td>\n","      <td>A</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>female</td>\n","      <td>26</td>\n","      <td>German</td>\n","      <td>Golf</td>\n","      <td>59</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>False</td>\n","      <td>master</td>\n","      <td>A</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>3995</th>\n","      <td>female</td>\n","      <td>28</td>\n","      <td>Dutch</td>\n","      <td>Running</td>\n","      <td>63</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>0</td>\n","      <td>False</td>\n","      <td>master</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3996</th>\n","      <td>female</td>\n","      <td>27</td>\n","      <td>Dutch</td>\n","      <td>Tennis</td>\n","      <td>62</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>2</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3997</th>\n","      <td>male</td>\n","      <td>24</td>\n","      <td>Belgian</td>\n","      <td>Rugby</td>\n","      <td>60</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>2</td>\n","      <td>False</td>\n","      <td>bachelor</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3998</th>\n","      <td>female</td>\n","      <td>22</td>\n","      <td>Dutch</td>\n","      <td>Tennis</td>\n","      <td>66</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3999</th>\n","      <td>male</td>\n","      <td>28</td>\n","      <td>Dutch</td>\n","      <td>Football</td>\n","      <td>58</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>False</td>\n","      <td>bachelor</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>4000 rows × 14 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-ed14b99e-a420-4e34-80b4-d38d11ded8da')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-ed14b99e-a420-4e34-80b4-d38d11ded8da button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-ed14b99e-a420-4e34-80b4-d38d11ded8da');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":8}]},{"cell_type":"markdown","source":["Next, let's rename some of the indicator columns such that we can use them in code later on, as the ind-name leads to Python seeing this as a df.ind function that does not exist."],"metadata":{"id":"_0mFSRAp5LsR"}},{"cell_type":"code","source":["df.rename(columns = {'ind-university_grade':'ind_university_grade', 'ind-debateclub':'ind_debateclub', 'ind-programming_exp':'ind_programming_exp', 'ind-international_exp':'ind_international_exp', 'ind-entrepeneur_exp':'ind_entrepeneur_exp', 'ind-languages':'ind_languages', 'ind-exact_study':'ind_exact_study', 'ind-degree':'ind_degree'  }, inplace = True)"],"metadata":{"id":"p6OIXzvX4nws","executionInfo":{"status":"ok","timestamp":1666352652106,"user_tz":-120,"elapsed":9,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":9,"outputs":[]},{"cell_type":"code","source":["df"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":488},"id":"_hylqRVG5J2B","executionInfo":{"status":"ok","timestamp":1666352652106,"user_tz":-120,"elapsed":8,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"ded9c68b-59b4-4726-9332-12e1c96e567e"},"execution_count":10,"outputs":[{"output_type":"execute_result","data":{"text/plain":["      gender  age nationality     sport  ind_university_grade  ind_debateclub  \\\n","0     female   24      German  Swimming                    70           False   \n","1       male   26      German      Golf                    67           False   \n","2     female   23       Dutch   Running                    67           False   \n","3       male   24       Dutch   Cricket                    70           False   \n","4     female   26      German      Golf                    59           False   \n","...      ...  ...         ...       ...                   ...             ...   \n","3995  female   28       Dutch   Running                    63           False   \n","3996  female   27       Dutch    Tennis                    62           False   \n","3997    male   24     Belgian     Rugby                    60            True   \n","3998  female   22       Dutch    Tennis                    66           False   \n","3999    male   28       Dutch  Football                    58           False   \n","\n","      ind_programming_exp  ind_international_exp  ind_entrepeneur_exp  \\\n","0                   False                  False                False   \n","1                    True                  False                False   \n","2                    True                   True                False   \n","3                    True                  False                False   \n","4                   False                  False                False   \n","...                   ...                    ...                  ...   \n","3995                False                  False                False   \n","3996                False                  False                 True   \n","3997                False                  False                 True   \n","3998                 True                  False                False   \n","3999                False                  False                False   \n","\n","      ind_languages  ind_exact_study ind_degree company  decision  \n","0                 1             True        phd       A      True  \n","1                 2             True   bachelor       A     False  \n","2                 0             True     master       A     False  \n","3                 1             True     master       A      True  \n","4                 1            False     master       A      True  \n","...             ...              ...        ...     ...       ...  \n","3995              0            False     master       D     False  \n","3996              2             True   bachelor       D     False  \n","3997              2            False   bachelor       D     False  \n","3998              1             True   bachelor       D     False  \n","3999              1            False   bachelor       D     False  \n","\n","[4000 rows x 14 columns]"],"text/html":["\n","  <div id=\"df-faa8af40-9e28-4415-8fb9-f2f0f2903d87\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>gender</th>\n","      <th>age</th>\n","      <th>nationality</th>\n","      <th>sport</th>\n","      <th>ind_university_grade</th>\n","      <th>ind_debateclub</th>\n","      <th>ind_programming_exp</th>\n","      <th>ind_international_exp</th>\n","      <th>ind_entrepeneur_exp</th>\n","      <th>ind_languages</th>\n","      <th>ind_exact_study</th>\n","      <th>ind_degree</th>\n","      <th>company</th>\n","      <th>decision</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>female</td>\n","      <td>24</td>\n","      <td>German</td>\n","      <td>Swimming</td>\n","      <td>70</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>True</td>\n","      <td>phd</td>\n","      <td>A</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>male</td>\n","      <td>26</td>\n","      <td>German</td>\n","      <td>Golf</td>\n","      <td>67</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>2</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>A</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>female</td>\n","      <td>23</td>\n","      <td>Dutch</td>\n","      <td>Running</td>\n","      <td>67</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>0</td>\n","      <td>True</td>\n","      <td>master</td>\n","      <td>A</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>male</td>\n","      <td>24</td>\n","      <td>Dutch</td>\n","      <td>Cricket</td>\n","      <td>70</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>True</td>\n","      <td>master</td>\n","      <td>A</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>female</td>\n","      <td>26</td>\n","      <td>German</td>\n","      <td>Golf</td>\n","      <td>59</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>False</td>\n","      <td>master</td>\n","      <td>A</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>3995</th>\n","      <td>female</td>\n","      <td>28</td>\n","      <td>Dutch</td>\n","      <td>Running</td>\n","      <td>63</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>0</td>\n","      <td>False</td>\n","      <td>master</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3996</th>\n","      <td>female</td>\n","      <td>27</td>\n","      <td>Dutch</td>\n","      <td>Tennis</td>\n","      <td>62</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>2</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3997</th>\n","      <td>male</td>\n","      <td>24</td>\n","      <td>Belgian</td>\n","      <td>Rugby</td>\n","      <td>60</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>2</td>\n","      <td>False</td>\n","      <td>bachelor</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3998</th>\n","      <td>female</td>\n","      <td>22</td>\n","      <td>Dutch</td>\n","      <td>Tennis</td>\n","      <td>66</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3999</th>\n","      <td>male</td>\n","      <td>28</td>\n","      <td>Dutch</td>\n","      <td>Football</td>\n","      <td>58</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>False</td>\n","      <td>bachelor</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>4000 rows × 14 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-faa8af40-9e28-4415-8fb9-f2f0f2903d87')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-faa8af40-9e28-4415-8fb9-f2f0f2903d87 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-faa8af40-9e28-4415-8fb9-f2f0f2903d87');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":10}]},{"cell_type":"markdown","source":["**Company Dataset**\n","\n","The company dataset contains all the instances where the value for the 'company' feature = B. \n","So let's select all these instances and put them in com_df"],"metadata":{"id":"F2OQJKcCiqCS"}},{"cell_type":"code","source":["com_df = df[df.company == 'B']"],"metadata":{"id":"6oROhafHjPcf","executionInfo":{"status":"ok","timestamp":1666352652106,"user_tz":-120,"elapsed":8,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":11,"outputs":[]},{"cell_type":"markdown","source":["For potential comparison stats, let's also create datasets for the other companies"],"metadata":{"id":"xrs07BVtlV_Z"}},{"cell_type":"code","source":["A_df = df[df.company == 'A']\n","C_df = df[df.company == 'C']\n","D_df = df[df.company == 'D']"],"metadata":{"id":"_dPCN8cdlUwr","executionInfo":{"status":"ok","timestamp":1666352652503,"user_tz":-120,"elapsed":11,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":12,"outputs":[]},{"cell_type":"markdown","source":["Below you can see now that com_df only contains 1000 instances instead of 4000 of the full set"],"metadata":{"id":"sOoIBgQXkK3a"}},{"cell_type":"code","source":["com_df"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":488},"id":"4OxyTXYHj_Xy","executionInfo":{"status":"ok","timestamp":1666352652503,"user_tz":-120,"elapsed":10,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"dc1b1cb3-cf02-4b67-f0ea-592204a4fd4e"},"execution_count":13,"outputs":[{"output_type":"execute_result","data":{"text/plain":["      gender  age nationality     sport  ind_university_grade  ind_debateclub  \\\n","1000  female   21       Dutch  Swimming                    58           False   \n","1001  female   27       Dutch     Chess                    67           False   \n","1002  female   25       Dutch     Rugby                    56            True   \n","1003  female   28       Dutch     Chess                    64           False   \n","1004    male   30       Dutch     Rugby                    63            True   \n","...      ...  ...         ...       ...                   ...             ...   \n","1995    male   28       Dutch    Tennis                    71           False   \n","1996  female   23     Belgian   Cricket                    55           False   \n","1997    male   26       Dutch  Football                    58            True   \n","1998  female   23      German   Running                    61           False   \n","1999  female   27       Dutch   Running                    65           False   \n","\n","      ind_programming_exp  ind_international_exp  ind_entrepeneur_exp  \\\n","1000                False                  False                False   \n","1001                False                   True                False   \n","1002                False                  False                False   \n","1003                False                   True                False   \n","1004                False                  False                 True   \n","...                   ...                    ...                  ...   \n","1995                False                  False                 True   \n","1996                False                  False                False   \n","1997                False                  False                 True   \n","1998                 True                  False                False   \n","1999                 True                  False                False   \n","\n","      ind_languages  ind_exact_study ind_degree company  decision  \n","1000              0             True   bachelor       B     False  \n","1001              0             True     master       B     False  \n","1002              2            False   bachelor       B      True  \n","1003              0             True     master       B     False  \n","1004              3            False     master       B      True  \n","...             ...              ...        ...     ...       ...  \n","1995              2             True   bachelor       B      True  \n","1996              1             True   bachelor       B     False  \n","1997              2            False   bachelor       B      True  \n","1998              1             True   bachelor       B     False  \n","1999              1             True   bachelor       B     False  \n","\n","[1000 rows x 14 columns]"],"text/html":["\n","  <div id=\"df-e84aca47-eda7-469a-a59f-0839de19fe4e\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>gender</th>\n","      <th>age</th>\n","      <th>nationality</th>\n","      <th>sport</th>\n","      <th>ind_university_grade</th>\n","      <th>ind_debateclub</th>\n","      <th>ind_programming_exp</th>\n","      <th>ind_international_exp</th>\n","      <th>ind_entrepeneur_exp</th>\n","      <th>ind_languages</th>\n","      <th>ind_exact_study</th>\n","      <th>ind_degree</th>\n","      <th>company</th>\n","      <th>decision</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>1000</th>\n","      <td>female</td>\n","      <td>21</td>\n","      <td>Dutch</td>\n","      <td>Swimming</td>\n","      <td>58</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>0</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>B</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>1001</th>\n","      <td>female</td>\n","      <td>27</td>\n","      <td>Dutch</td>\n","      <td>Chess</td>\n","      <td>67</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>0</td>\n","      <td>True</td>\n","      <td>master</td>\n","      <td>B</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>1002</th>\n","      <td>female</td>\n","      <td>25</td>\n","      <td>Dutch</td>\n","      <td>Rugby</td>\n","      <td>56</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>2</td>\n","      <td>False</td>\n","      <td>bachelor</td>\n","      <td>B</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>1003</th>\n","      <td>female</td>\n","      <td>28</td>\n","      <td>Dutch</td>\n","      <td>Chess</td>\n","      <td>64</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>0</td>\n","      <td>True</td>\n","      <td>master</td>\n","      <td>B</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>1004</th>\n","      <td>male</td>\n","      <td>30</td>\n","      <td>Dutch</td>\n","      <td>Rugby</td>\n","      <td>63</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>3</td>\n","      <td>False</td>\n","      <td>master</td>\n","      <td>B</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>1995</th>\n","      <td>male</td>\n","      <td>28</td>\n","      <td>Dutch</td>\n","      <td>Tennis</td>\n","      <td>71</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>2</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>B</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>1996</th>\n","      <td>female</td>\n","      <td>23</td>\n","      <td>Belgian</td>\n","      <td>Cricket</td>\n","      <td>55</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>B</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>1997</th>\n","      <td>male</td>\n","      <td>26</td>\n","      <td>Dutch</td>\n","      <td>Football</td>\n","      <td>58</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>2</td>\n","      <td>False</td>\n","      <td>bachelor</td>\n","      <td>B</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>1998</th>\n","      <td>female</td>\n","      <td>23</td>\n","      <td>German</td>\n","      <td>Running</td>\n","      <td>61</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>B</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>1999</th>\n","      <td>female</td>\n","      <td>27</td>\n","      <td>Dutch</td>\n","      <td>Running</td>\n","      <td>65</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>B</td>\n","      <td>False</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>1000 rows × 14 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e84aca47-eda7-469a-a59f-0839de19fe4e')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-e84aca47-eda7-469a-a59f-0839de19fe4e button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-e84aca47-eda7-469a-a59f-0839de19fe4e');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":13}]},{"cell_type":"markdown","source":["We can further tidy up this dataset by removing the company column, since we know that all our candidates are from Company B, which means that this feature will have no predictive power"],"metadata":{"id":"ExMOnLODmvrU"}},{"cell_type":"code","source":["com_df = com_df.drop('company', axis=1)"],"metadata":{"id":"eETLid7GmvOf","executionInfo":{"status":"ok","timestamp":1666352652503,"user_tz":-120,"elapsed":9,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":14,"outputs":[]},{"cell_type":"markdown","source":["**Population Dataset**\n","\n","Speaking of the full dataset, we also need to have a population dataset for our exploration comparisons. \n","\n","This pop_df contains all the instances regardless of company, and is thus just the original df"],"metadata":{"id":"Tm3njAcakSOT"}},{"cell_type":"code","source":["pop_df = df"],"metadata":{"id":"llC4XQuzkisX","executionInfo":{"status":"ok","timestamp":1666352652504,"user_tz":-120,"elapsed":10,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":15,"outputs":[]},{"cell_type":"markdown","source":["Let's check"],"metadata":{"id":"53UiQdM-km2z"}},{"cell_type":"code","source":["pop_df"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":488},"id":"ZNfy1y28kkRG","executionInfo":{"status":"ok","timestamp":1666352652504,"user_tz":-120,"elapsed":9,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"ebbd9457-90ed-4a18-9d9a-da9b60050607"},"execution_count":16,"outputs":[{"output_type":"execute_result","data":{"text/plain":["      gender  age nationality     sport  ind_university_grade  ind_debateclub  \\\n","0     female   24      German  Swimming                    70           False   \n","1       male   26      German      Golf                    67           False   \n","2     female   23       Dutch   Running                    67           False   \n","3       male   24       Dutch   Cricket                    70           False   \n","4     female   26      German      Golf                    59           False   \n","...      ...  ...         ...       ...                   ...             ...   \n","3995  female   28       Dutch   Running                    63           False   \n","3996  female   27       Dutch    Tennis                    62           False   \n","3997    male   24     Belgian     Rugby                    60            True   \n","3998  female   22       Dutch    Tennis                    66           False   \n","3999    male   28       Dutch  Football                    58           False   \n","\n","      ind_programming_exp  ind_international_exp  ind_entrepeneur_exp  \\\n","0                   False                  False                False   \n","1                    True                  False                False   \n","2                    True                   True                False   \n","3                    True                  False                False   \n","4                   False                  False                False   \n","...                   ...                    ...                  ...   \n","3995                False                  False                False   \n","3996                False                  False                 True   \n","3997                False                  False                 True   \n","3998                 True                  False                False   \n","3999                False                  False                False   \n","\n","      ind_languages  ind_exact_study ind_degree company  decision  \n","0                 1             True        phd       A      True  \n","1                 2             True   bachelor       A     False  \n","2                 0             True     master       A     False  \n","3                 1             True     master       A      True  \n","4                 1            False     master       A      True  \n","...             ...              ...        ...     ...       ...  \n","3995              0            False     master       D     False  \n","3996              2             True   bachelor       D     False  \n","3997              2            False   bachelor       D     False  \n","3998              1             True   bachelor       D     False  \n","3999              1            False   bachelor       D     False  \n","\n","[4000 rows x 14 columns]"],"text/html":["\n","  <div id=\"df-e6ffc0f0-dc9a-47cc-99af-b6dd432e0510\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>gender</th>\n","      <th>age</th>\n","      <th>nationality</th>\n","      <th>sport</th>\n","      <th>ind_university_grade</th>\n","      <th>ind_debateclub</th>\n","      <th>ind_programming_exp</th>\n","      <th>ind_international_exp</th>\n","      <th>ind_entrepeneur_exp</th>\n","      <th>ind_languages</th>\n","      <th>ind_exact_study</th>\n","      <th>ind_degree</th>\n","      <th>company</th>\n","      <th>decision</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>female</td>\n","      <td>24</td>\n","      <td>German</td>\n","      <td>Swimming</td>\n","      <td>70</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>True</td>\n","      <td>phd</td>\n","      <td>A</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>male</td>\n","      <td>26</td>\n","      <td>German</td>\n","      <td>Golf</td>\n","      <td>67</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>2</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>A</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>female</td>\n","      <td>23</td>\n","      <td>Dutch</td>\n","      <td>Running</td>\n","      <td>67</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>0</td>\n","      <td>True</td>\n","      <td>master</td>\n","      <td>A</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>male</td>\n","      <td>24</td>\n","      <td>Dutch</td>\n","      <td>Cricket</td>\n","      <td>70</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>True</td>\n","      <td>master</td>\n","      <td>A</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>female</td>\n","      <td>26</td>\n","      <td>German</td>\n","      <td>Golf</td>\n","      <td>59</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>False</td>\n","      <td>master</td>\n","      <td>A</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>3995</th>\n","      <td>female</td>\n","      <td>28</td>\n","      <td>Dutch</td>\n","      <td>Running</td>\n","      <td>63</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>0</td>\n","      <td>False</td>\n","      <td>master</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3996</th>\n","      <td>female</td>\n","      <td>27</td>\n","      <td>Dutch</td>\n","      <td>Tennis</td>\n","      <td>62</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>2</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3997</th>\n","      <td>male</td>\n","      <td>24</td>\n","      <td>Belgian</td>\n","      <td>Rugby</td>\n","      <td>60</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>2</td>\n","      <td>False</td>\n","      <td>bachelor</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3998</th>\n","      <td>female</td>\n","      <td>22</td>\n","      <td>Dutch</td>\n","      <td>Tennis</td>\n","      <td>66</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>3999</th>\n","      <td>male</td>\n","      <td>28</td>\n","      <td>Dutch</td>\n","      <td>Football</td>\n","      <td>58</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>False</td>\n","      <td>bachelor</td>\n","      <td>D</td>\n","      <td>False</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>4000 rows × 14 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e6ffc0f0-dc9a-47cc-99af-b6dd432e0510')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-e6ffc0f0-dc9a-47cc-99af-b6dd432e0510 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-e6ffc0f0-dc9a-47cc-99af-b6dd432e0510');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":16}]},{"cell_type":"markdown","source":["**Indicator Dataset**\n","\n","For Model 1 and 2 we need to create a company B dataset which only has the Indicator features, as we're not allowed to use any other features.\n","\n","So let's first list all the indicator and descriptor features mentioned in the assignment (*Notice that some of the names are not the same for the indicators as on the canvas page, as well was ind-degree being in our dataset instead of ind-univ-grade, and the existence of personality_type*)"],"metadata":{"id":"4yBWxjZgkru4"}},{"cell_type":"code","source":["indicator = [ 'ind_debateclub', 'ind_programming_exp', 'ind_international_exp', 'ind_entrepeneur_exp', 'ind_languages', 'ind_exact_study', 'ind_degree', 'ind_university_grade' ]\n","\n","descriptor = ['gender', 'age', 'nationality', 'sport']"],"metadata":{"id":"sU7MloRDlEVP","executionInfo":{"status":"ok","timestamp":1666352652504,"user_tz":-120,"elapsed":9,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":17,"outputs":[]},{"cell_type":"markdown","source":["Now that we've classified the features, we can construct ind_df by dropping all the descriptor features from our company df"],"metadata":{"id":"soHsqDYCmMbV"}},{"cell_type":"code","source":["ind_df = com_df.drop(descriptor, axis=1)"],"metadata":{"id":"tWFgUDztlETY","executionInfo":{"status":"ok","timestamp":1666352652504,"user_tz":-120,"elapsed":8,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":18,"outputs":[]},{"cell_type":"markdown","source":["Let's take a look at it"],"metadata":{"id":"prs7KGmtk5eU"}},{"cell_type":"code","source":["ind_df"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":468},"id":"GklRPwPAmeKO","executionInfo":{"status":"ok","timestamp":1666352652505,"user_tz":-120,"elapsed":9,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"2e26534f-4c95-4505-8c9a-79e68d176a36"},"execution_count":19,"outputs":[{"output_type":"execute_result","data":{"text/plain":["      ind_university_grade  ind_debateclub  ind_programming_exp  \\\n","1000                    58           False                False   \n","1001                    67           False                False   \n","1002                    56            True                False   \n","1003                    64           False                False   \n","1004                    63            True                False   \n","...                    ...             ...                  ...   \n","1995                    71           False                False   \n","1996                    55           False                False   \n","1997                    58            True                False   \n","1998                    61           False                 True   \n","1999                    65           False                 True   \n","\n","      ind_international_exp  ind_entrepeneur_exp  ind_languages  \\\n","1000                  False                False              0   \n","1001                   True                False              0   \n","1002                  False                False              2   \n","1003                   True                False              0   \n","1004                  False                 True              3   \n","...                     ...                  ...            ...   \n","1995                  False                 True              2   \n","1996                  False                False              1   \n","1997                  False                 True              2   \n","1998                  False                False              1   \n","1999                  False                False              1   \n","\n","      ind_exact_study ind_degree  decision  \n","1000             True   bachelor     False  \n","1001             True     master     False  \n","1002            False   bachelor      True  \n","1003             True     master     False  \n","1004            False     master      True  \n","...               ...        ...       ...  \n","1995             True   bachelor      True  \n","1996             True   bachelor     False  \n","1997            False   bachelor      True  \n","1998             True   bachelor     False  \n","1999             True   bachelor     False  \n","\n","[1000 rows x 9 columns]"],"text/html":["\n","  <div id=\"df-902bd9ff-8a66-4ea9-bc07-a131c1f25c84\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>ind_university_grade</th>\n","      <th>ind_debateclub</th>\n","      <th>ind_programming_exp</th>\n","      <th>ind_international_exp</th>\n","      <th>ind_entrepeneur_exp</th>\n","      <th>ind_languages</th>\n","      <th>ind_exact_study</th>\n","      <th>ind_degree</th>\n","      <th>decision</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>1000</th>\n","      <td>58</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>0</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>1001</th>\n","      <td>67</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>0</td>\n","      <td>True</td>\n","      <td>master</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>1002</th>\n","      <td>56</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>2</td>\n","      <td>False</td>\n","      <td>bachelor</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>1003</th>\n","      <td>64</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>0</td>\n","      <td>True</td>\n","      <td>master</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>1004</th>\n","      <td>63</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>3</td>\n","      <td>False</td>\n","      <td>master</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>1995</th>\n","      <td>71</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>2</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>1996</th>\n","      <td>55</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>1997</th>\n","      <td>58</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>2</td>\n","      <td>False</td>\n","      <td>bachelor</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>1998</th>\n","      <td>61</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>1999</th>\n","      <td>65</td>\n","      <td>False</td>\n","      <td>True</td>\n","      <td>False</td>\n","      <td>False</td>\n","      <td>1</td>\n","      <td>True</td>\n","      <td>bachelor</td>\n","      <td>False</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>1000 rows × 9 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-902bd9ff-8a66-4ea9-bc07-a131c1f25c84')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-902bd9ff-8a66-4ea9-bc07-a131c1f25c84 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-902bd9ff-8a66-4ea9-bc07-a131c1f25c84');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":19}]},{"cell_type":"markdown","source":["**Descriptor Dataset**\n","\n","The descriptor dataset des_df is the same process, but now we drop all the indicator features\n"],"metadata":{"id":"zlhIiLG9mWSS"}},{"cell_type":"code","source":["des_df = com_df.drop(indicator, axis=1)"],"metadata":{"id":"JRqcgyOblDb4","executionInfo":{"status":"ok","timestamp":1666352652852,"user_tz":-120,"elapsed":356,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":20,"outputs":[]},{"cell_type":"markdown","source":["Let's take a look at it"],"metadata":{"id":"6NYSLrEDnS4O"}},{"cell_type":"code","source":["des_df"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":424},"id":"bO_R31GgnU5L","executionInfo":{"status":"ok","timestamp":1666352652852,"user_tz":-120,"elapsed":4,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"f70cdc15-5c9d-40fb-8de1-4007b5aedc9f"},"execution_count":21,"outputs":[{"output_type":"execute_result","data":{"text/plain":["      gender  age nationality     sport  decision\n","1000  female   21       Dutch  Swimming     False\n","1001  female   27       Dutch     Chess     False\n","1002  female   25       Dutch     Rugby      True\n","1003  female   28       Dutch     Chess     False\n","1004    male   30       Dutch     Rugby      True\n","...      ...  ...         ...       ...       ...\n","1995    male   28       Dutch    Tennis      True\n","1996  female   23     Belgian   Cricket     False\n","1997    male   26       Dutch  Football      True\n","1998  female   23      German   Running     False\n","1999  female   27       Dutch   Running     False\n","\n","[1000 rows x 5 columns]"],"text/html":["\n","  <div id=\"df-9aa4ae9f-2c87-4449-843e-d7c3d12726d0\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>gender</th>\n","      <th>age</th>\n","      <th>nationality</th>\n","      <th>sport</th>\n","      <th>decision</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>1000</th>\n","      <td>female</td>\n","      <td>21</td>\n","      <td>Dutch</td>\n","      <td>Swimming</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>1001</th>\n","      <td>female</td>\n","      <td>27</td>\n","      <td>Dutch</td>\n","      <td>Chess</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>1002</th>\n","      <td>female</td>\n","      <td>25</td>\n","      <td>Dutch</td>\n","      <td>Rugby</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>1003</th>\n","      <td>female</td>\n","      <td>28</td>\n","      <td>Dutch</td>\n","      <td>Chess</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>1004</th>\n","      <td>male</td>\n","      <td>30</td>\n","      <td>Dutch</td>\n","      <td>Rugby</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>1995</th>\n","      <td>male</td>\n","      <td>28</td>\n","      <td>Dutch</td>\n","      <td>Tennis</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>1996</th>\n","      <td>female</td>\n","      <td>23</td>\n","      <td>Belgian</td>\n","      <td>Cricket</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>1997</th>\n","      <td>male</td>\n","      <td>26</td>\n","      <td>Dutch</td>\n","      <td>Football</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>1998</th>\n","      <td>female</td>\n","      <td>23</td>\n","      <td>German</td>\n","      <td>Running</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>1999</th>\n","      <td>female</td>\n","      <td>27</td>\n","      <td>Dutch</td>\n","      <td>Running</td>\n","      <td>False</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>1000 rows × 5 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-9aa4ae9f-2c87-4449-843e-d7c3d12726d0')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-9aa4ae9f-2c87-4449-843e-d7c3d12726d0 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-9aa4ae9f-2c87-4449-843e-d7c3d12726d0');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":21}]},{"cell_type":"markdown","source":["# **2. Data Exploration**\n","\n"],"metadata":{"id":"9eCNb5APnglW"}},{"cell_type":"markdown","source":["\n","To explore the Data, let us go and play around with a correlation matrix, missing values and feature exploration.\n","\n","First, let us take a look at the correlation matrix"],"metadata":{"id":"5y2Fm4nHqjNM"}},{"cell_type":"code","source":["# Create correlation between these columns of data\n","corr = com_df.corr()\n","\n","plt.figure(figsize=(25,20))\n","\n","# Create array of 0s in same shape as corr.shape\n","mask = np.zeros_like(corr)\n","# Take upper triangle of the shape of 0s\n","mask[np.triu_indices_from(mask)] = True\n","\n","# Change mask and correlation to remove empty rows\n","mask = mask[1:, :-1]\n","corr = corr.iloc[1:,:-1].copy()\n","\n","# Choose a diverging palette for the r values (0 should be white)\n","cmap = sns.diverging_palette(220, 20, as_cmap=True)\n","\n","# Plot the heatmap\n","sns.heatmap(corr,         # Create heatmap of the df.corr created above\n","            mask=mask,    # Use the mask to not show repeated boxes\n","            annot=True,   # Annotate the squares with the r value\n","            fmt=\".2f\",    # Use 2 decimals\n","            annot_kws={\"size\": 8}, \n","            cmap='RdYlGn',    # cmap is the diverging one created above\n","            square=True,  # Square boxes look better\n","            cbar_kws={\"shrink\": .8}, # Shrink color bar to center it\n","            vmin=-1, vmax=1) # Max and Min values for the colors\n","\n","plt.show()\n","plt.savefig('correlation_matrix.png')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000},"id":"MT-vewnWQTqq","executionInfo":{"status":"ok","timestamp":1666352653598,"user_tz":-120,"elapsed":749,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"a8190ed6-a87d-4e2e-903e-1df234741ca0"},"execution_count":22,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 1800x1440 with 2 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 0 Axes>"]},"metadata":{}}]},{"cell_type":"markdown","source":["For data exploration, it is also interesting to compare our company correlation matrix to the population dataset"],"metadata":{"id":"_GZJfe6mQTyS"}},{"cell_type":"code","source":["# Create correlation between these columns of data\n","corr = df.corr()\n","\n","plt.figure(figsize=(25,20))\n","\n","# Create array of 0s in same shape as corr.shape\n","mask = np.zeros_like(corr)\n","# Take upper triangle of the shape of 0s\n","mask[np.triu_indices_from(mask)] = True\n","\n","# Change mask and correlation to remove empty rows\n","mask = mask[1:, :-1]\n","corr = corr.iloc[1:,:-1].copy()\n","\n","# Choose a diverging palette for the r values (0 should be white)\n","cmap = sns.diverging_palette(220, 20, as_cmap=True)\n","\n","# Plot the heatmap\n","sns.heatmap(corr,         # Create heatmap of the df.corr created above\n","            mask=mask,    # Use the mask to not show repeated boxes\n","            annot=True,   # Annotate the squares with the r value\n","            fmt=\".2f\",    # Use 2 decimals\n","            annot_kws={\"size\": 8}, \n","            cmap='RdYlGn',    # cmap is the diverging one created above\n","            square=True,  # Square boxes look better\n","            cbar_kws={\"shrink\": .8}, # Shrink color bar to center it\n","            vmin=-1, vmax=1) # Max and Min values for the colors\n","\n","plt.show()\n","plt.savefig('correlation_matrix.png')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000},"id":"BvI91sh4r3Nt","executionInfo":{"status":"ok","timestamp":1666352654414,"user_tz":-120,"elapsed":817,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"60ef4e11-0533-4a6f-9d45-f2712a732283"},"execution_count":23,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 1800x1440 with 2 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 0 Axes>"]},"metadata":{}}]},{"cell_type":"markdown","source":["What we find for our company B here is that Decision is most highly correlated with ind-languages, ind-debateclub and ind-enterpeneur_exp, and the least correlated with age and ind-international_exp. This may be important to keep in mind with picking indicators for our methods.\n","\n","Next, let us check whether we have any missing data in our dataset"],"metadata":{"id":"2TSPFOultm-C"}},{"cell_type":"code","source":["missing_data = com_df.isna().sum().to_frame()\n","list_of_features = []\n","for index, row in missing_data.iterrows():\n","    if row[0] > 0:\n","        list_of_features.append((index, row[0]))\n","sorted_list = sorted(list_of_features, key=lambda index: index[1], reverse=True)\n","print(sorted_list)"],"metadata":{"id":"7t4M9UeRrwnd","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1666352654414,"user_tz":-120,"elapsed":4,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"50049d7d-7aa1-45a4-f5b0-b539137fdf06"},"execution_count":24,"outputs":[{"output_type":"stream","name":"stdout","text":["[]\n"]}]},{"cell_type":"markdown","source":["We can also check it using a heatmap of null values in our dataset. This should also show we have no missing value"],"metadata":{"id":"swM1nXs8R6LY"}},{"cell_type":"code","source":["sns.heatmap(com_df.isnull())"],"metadata":{"id":"vkP7ItOURqVt","colab":{"base_uri":"https://localhost:8080/","height":389},"executionInfo":{"status":"ok","timestamp":1666352654698,"user_tz":-120,"elapsed":287,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"f14412a4-86bc-4512-adf9-41d98d4cae71"},"execution_count":25,"outputs":[{"output_type":"execute_result","data":{"text/plain":["<matplotlib.axes._subplots.AxesSubplot at 0x7f4536f67990>"]},"metadata":{},"execution_count":25},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 2 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["Seems like there is no obvious missing data in the form of NaNs.\n","That does not mean that we don't have any missing data, we could still have 0s representing missing data (for number of languages this may very well be likely). \n","\n","So next up, let's explore all the features individually and see what data they contain"],"metadata":{"id":"DrBfx2Ovu-5V"}},{"cell_type":"markdown","source":["###gender\n","\n","The gender feature consists of nominal categoric data, so let's first see what categories are present"],"metadata":{"id":"ipUkHTqivWe-"}},{"cell_type":"code","source":["com_df.gender.unique()"],"metadata":{"id":"9-CKFMm1xWAF","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1666352654698,"user_tz":-120,"elapsed":4,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"fa4fe418-d4ee-4b2c-e0ab-c91d9c7d18a4"},"execution_count":26,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array(['female', 'male', 'other'], dtype=object)"]},"metadata":{},"execution_count":26}]},{"cell_type":"markdown","source":["Now let's see what the comparison is between them"],"metadata":{"id":"OKMEMQLdyFUs"}},{"cell_type":"code","source":["def addlabels(x,y):\n","    for i in range(len(x)):\n","        plt.text(i,y[i],y[i])\n","\n","genders = [com_df['gender'].where(com_df['gender'] == 'female').count(), com_df['gender'].where(com_df['gender'] == 'male').count(), com_df['gender'].where(com_df['gender'] == 'other').count()]\n","gender_label = ['female', 'male', 'other']\n","\n","fig = plt.figure(figsize =(5, 5))\n","plt.ylim([0,550])\n","plt.title('Gender occurrences in company B dataset')\n","plt.bar(gender_label, genders, ec='black')\n","addlabels(gender_label, genders)\n","plt.show()"],"metadata":{"id":"nkfzr9RavmX_","colab":{"base_uri":"https://localhost:8080/","height":0},"executionInfo":{"status":"ok","timestamp":1666352654989,"user_tz":-120,"elapsed":294,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"98516016-7070-48d3-81e0-424bc97800e0"},"execution_count":27,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 360x360 with 1 Axes>"],"image/png":"iVBORw0KGgoAAAANSUhEUgAAAT8AAAE/CAYAAAAwpsSrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAdFElEQVR4nO3debxVdb3/8dcbQZwQRA+KgGBBOSBh4MCt/HmdrnpJ5Kd1LRxQjGtZZmlpWb8GNad+OTzKvJQDKmWaJV5nc8jUFMFwxiBFD0cmB4bQ7Kqf+8f6Hlwcz+FM+7DP4ft+Ph77cdb6ru9a67PW3vu9hn3OPooIzMxy063aBZiZVYPDz8yy5PAzsyw5/MwsSw4/M8uSw8/MsuTwawFJEyU9WO06ciXpU5Ker3YdOZB0v6Tjq13HutBlw0/SEZIelbRK0pI0/CVJqnZtVlkR8aeI+Gi16+jsJA2RFJL+nh6LJV0qqUcHrW++pP06YtnrYj1dMvwknQJcDFwAbANsDZwAfALYsIqlfYCkDapdQ5mk7i1psy6tT0RsBuwCjAFOrHI9nVNEdKkH0BtYBRzWTL+ewI+Bl4HFwGXAxmna3sAC4BRgCbAQOLY075bAzcAKYAZwJvBgafoOwN3A68DzwGdL064Cfg7clurcr5Hatk3Lfx2YB3yhNG0D4NvA34CVwCxgUJq2c2m9i4Fvl9Z5VmkZewMLSuPzgdOAJ4G3gaFAAJPS/nkg9TsOeA54A7gTGFxaRlAcYOYCy4CfASpN/0KadyXwLPDx0rbeCCwFXgROKs2zOzAz7efFwE+aeC4b255T0/YsB34DbLSW10JTte0I3J+25xngkAbP46XA7cDfgYcoDrQXpf0zB9i1QU3fSst/A7iyviZgC+CWtA/eSMMDS/PeT/EaeyjVeBewVZp2K/CVBtvzJDC+ke0ckp6n7qW284Epa9k3+6dtWQ78FPgjcHya9mHgXuA14FVgGkWwAlwDvAe8lfbPN1P7DcCitLwHgJ1L6zo47Z+VQB1wamnaWGB2ei4eBkasbT0VyZJqhVibC4YDgXfKT3AT/S6kCJi+QC/gv4FzSm+md4AfAj3Sk/ImsEWafh1wPbApMDw9UQ+maZsCtcCxQHdg1/TC2Kn0pllOcRbajUbelOlFcSmwETAyvSn2SdO+ATwFfBQQ8DGKMO5FEdKnpPl6AXuU1tlc+M0GBgEbl94kV6ft2RgYRxHEO6bt+g7wcGkZQfGm7QNsl2o+ME37TNpHu6WahwKD0/bPAv4fxRn5h4AXgH9L8/0ZOCoNbwbs2cRz2dj2zKAI1r4UwXZCE/M2VVuPtL3fTrXtQ/Gm/Ghpn74KjEr7+16K8D6a4gB1FnBfg5qeTvu4L0WQnZWmbQkcBmySnrcbgJtK895PcbD7SHou7gfOTdM+Czxa6vsxijDasJFtrX9eu6fxbYEngOOa2DdbpW0+PO2Pr1G8L+rDbyhFOPYEaihetxc12Ob9GizzuLSNPSkOFLNL0xYCn0rDW/D+QWhXipOQPdK+PSYtu2dT66lIllQ7zFpdMBwJLGrQ9jDFEeMtYK/0Il8FfLjUZwzwYunN9BZrHiGXAHumnf8/wA6laT/i/fD7D+BPDdb/X8D3Sm+aq9dS/yDgXaBXqe0c4Ko0/DwwrpH5Pgf8pYllXkXz4Xdcabz+TfKhUtvtwKTSeDeKA8LgNB7AJ0vTrwdOT8N3Al9tpK49gJcbtH0LuDINPwD8gHSWs5Z91tj2HFkaPx+4rIl5m6rtUxRnKN1Kbb8Gvl/ap78oTfsK8FxpfBdgWYOaTiiNHwz8rYmaRgJvlMbvB75TGv8ScEca3ojibHFYGv8xcGkTy61/XpelR1C8NzZvov/RwCOlcVFcER3fRP9Dy69BmgkligNlAL3T+MvAfzash+JK6cwGbc8D/6cl62nroyve83sN2Kp8nyoi/iUi+qRp3SiOUpsAsyQtk7QMuCO1r15ORLxTGn+T4uyjhuLMp7Y07aXS8GBgj/rlpmVPoLgkqleet6FtgdcjYmWD5Q9Iw4MozgIaaqq9pRqrqdw2GLi4tE2vU7wZBpT6LCoN1++vtdU2GNi2wb76NsU9Wiguuz8CzJH0mKSxrdiepmppqKnatgVqI+K9Ulv5eYDiUrzeW42MN1xnw9fMtgCSNpH0X5JekrSCIvT7NLgf3Oj2RMQ/KC7rj5TUjeIgeE1jG1qyVXo/bEJxBnpnE/22LdccRdKsHpe0taTrJNWluq+lOFtslKQNJJ0r6W+p//z6etLPwygOCi9J+qOkMal9MHBKg9fJoFRfh+mK4fdnivtW49bS51WKF+fOEdEnPXpHcRO4OUspTv0Hldq2Kw3XAn8sLbdPRGwWEV8s9Ym1LP8VoK+kXg2WX1da/ocbma+W4rKxMasoXuj1tmmkT2M1ldtqgf9ssF0bR8TDTayzYW1N1fxig2X2ioiDASJibkR8DugHnAf8VtKmLVhfazRV2yvAoBQo9crPQ1s0fM28koZPobiNsUdEbE5xdQLFwaUlplIcYPcF3oyIP7dkpoh4i+IMdk9JjYXWwnLN6TclytvwI4rXyC6p7iMb1NzwNfV5ivflfhT35ofULzrV81hEjKN4vm+iuHqA4jk6u8HrZJOI+HUT66mILhd+EbGM4lLpUkmHS+olqZukkRT3r0hH818AF0rqByBpgKR/a8Hy3wV+B3w/HbF3orgHUe8W4COSjpLUIz12k7RjC+uvpbgUOUfSRpJGUJwBXZu6/BI4U9IwFUZI2jKtt7+kkyX1TNu9R5pnNnCwpL6StgFObkktDVwGfEvSzgCSekv6TAvn/SVwqqRRqeahkgZT3JdbKek0SRunM4PhknZL6zhSUk16vpalZb3XxDraqqnaHqU4w/pmeg73Bj5Ncb+3rU6UNFBSX+AMijM2KO6BvQUsS9O+15qFprB7D/j/NH/Wt5qknsBRFGeVrzXS5VZgZ0n/N11JncSaB85eFB8yLJc0gOJ+dNli1jwg96I4MXmN4mD8o1ItG0qaIKl3RPwPxYdc9c/1L4ATJO2RnqNNJf176QSh4XoqosuFH0BEnA98HfgmxY5ZTHHf7TSKYCENzwMeSafgf6A4+rbElykuOxZRHDmvLK17JXAAcATFkX0RxVlLz1ZswucojoqvAL+nuF/4hzTtJxRHxLsoXiCXU3xKvZLi5vOn0zrnAv+a5rmG4sb2/DRf/ZuuxSLi92k7rkv762ngoBbOewNwNvArihvoNwF904FkLMU9rhcpzsh/SXFWAMWHV89I+jvFry4dkc5WKmYttf2TYl8elOq6FDg6Iua0Y3W/otj/L1Bcap+V2i+i+CDjVeARilswrXU1xX3Ga5vrSBGyf6d4X4yh+BT7A2dPEfEqxQdC51IE1jCKy+R6PwA+TvEB3q0UJwVl5wDfSZeqp6YaX6I4e36WYlvLjgLmp9fXCRRns0TETIpP5H9KcX9zHjBxLeupCDWyT8yslSTNp/ig4A/N9W3j8o8GJkfEJzti+Tnqkmd+ZjmRtAnFJ8BTql3L+sThZ9aJpfvUSykuYX9V5XLWK77sNbMs+czPzLLk8DOzLHWKb/PYaqutYsiQIdUuw8zWM7NmzXo1Imoam9Ypwm/IkCHMnDmz2mWY2XpG0ktNTfNlr3VJQ4YMYZdddmHkyJGMHj0agO9+97uMGDGCkSNHcsABB/DKK8Vfl0UEJ510EkOHDmXEiBE8/vjj1SzdOgmHn3VZ9913H7Nnz1591fCNb3yDJ598ktmzZzN27Fh++MMfAnD77bczd+5c5s6dy5QpU/jiF7+4tsVaJhx+tt7YfPPNVw+vWrUKpf9oMH36dI4++mgkseeee7Js2TIWLlxYrTKtk3D4WZckiQMOOIBRo0YxZcr7f/hwxhlnMGjQIKZNm7b6zK+uro5Bg97/spKBAwdSV9eeL2+x9YHDz7qkBx98kMcff5zbb7+dn/3sZzzwwAMAnH322dTW1jJhwgR++tOfVrlK68wcftYlDRhQfOdov379GD9+PDNmzFhj+oQJE7jxxhtX962tff97RhcsWLB6fsuXw8+6nFWrVrFy5crVw3fddRfDhw9n7ty5q/tMnz6dHXbYAYBDDjmEq6++mojgkUceoXfv3vTv378qtVvn0Sl+z8+sNRYvXsz48eMBeOedd/j85z/PgQceyGGHHcbzzz9Pt27dGDx4MJdddhkABx98MLfddhtDhw5lk0024corr1zb4i0TneKLDUaPHh3+JefOqf/A7VhUt7Z/SWLbDBjEwgUvV7sMa4SkWRExurFpPvOztVpUV8vg026pdhmd2kvnteb/Llln4Xt+ZpYlh5+ZZcnhZ2ZZcviZWZYcfmaWJYefmWXJ4WdmWXL4mVmWHH5mlqUWhZ+k+ZKekjRb0szU1lfS3ZLmpp9bpHZJukTSPElPSvp4R26AmVlbtObM718jYmTp7+ROB+6JiGHAPWkc4CBgWHpMBn5eqWLNzCqlPZe944CpaXgqcGip/eooPAL0keTvDzKzTqWl4RfAXZJmSZqc2raOiPp/hLAI2DoNDwDKXwOyILWZmXUaLf1Wl09GRJ2kfsDdkuaUJ0ZESGrVd2OlEJ0MsN1227VmVjOzdmvRmV9E1KWfS4DfA7sDi+svZ9PPJal7HTCoNPvA1NZwmVMiYnREjK6pafQfqpuZdZhmw0/SppJ61Q8DBwBPAzcDx6RuxwDT0/DNwNHpU989geWly2Mzs06hJZe9WwO/T/8DtTvwq4i4Q9JjwPWSJgEvAZ9N/W8DDgbmAW8Cx1a8ajOzdmo2/CLiBeBjjbS/BuzbSHsAJ1akOjOzDuK/8DCzLDn8zCxLDj8zy5LDz8yy5PAzsyw5/MwsSw4/M8uSw8/MsuTwM7MsOfzMLEsOPzPLksOvwt5991123XVXxo4dC8DEiRPZfvvtGTlyJCNHjmT27NkAzJkzhzFjxtCzZ09+/OMfV7Nksyy19MtMrYUuvvhidtxxR1asWLG67YILLuDwww9fo1/fvn255JJLuOmmm9Z1iWaGz/wqasGCBdx6660cf/zxzfbt168fu+22Gz169FgHlZlZQw6/Cjr55JM5//zz6dZtzd16xhlnMGLECL72ta/x9ttvV6k6Mytz+FXILbfcQr9+/Rg1atQa7eeccw5z5szhscce4/XXX+e8886rUoVmVubwq5CHHnqIm2++mSFDhnDEEUdw7733cuSRR9K/f38k0bNnT4499lhmzJhR7VLNDIdfxZxzzjksWLCA+fPnc91117HPPvtw7bXXsnBh8e9LIoKbbrqJ4cOHV7lSMwN/2tvhJkyYwNKlS4kIRo4cyWWXXQbAokWLGD16NCtWrKBbt25cdNFFPPvss2y++eZVrtgsD102/PoP3I5FdbXNd6yi9E+fVnvmmWeYNm1ao32XL19O7969K7bubQYMYuGClyu2PLP1TZcNv0V1tQw+7ZZql9FpvXTe2GqXYNap+Z6fmWXJ4WdmWXL4mVmWHH5mliWHn5llyeFnZlly+JlZlhx+ZpYlh5+ZZcnhZ2ZZcviZWZYcfmaWJYefmWXJ4WdmWXL4mVmWHH5mliWHn5llyeFnZlly+JlZllocfpI2kPQXSbek8e0lPSppnqTfSNowtfdM4/PS9CEdU7qZWdu15szvq8BzpfHzgAsjYijwBjAptU8C3kjtF6Z+ZmadSovCT9JA4N+BX6ZxAfsAv01dpgKHpuFxaZw0fV81/B+OZmZV1tIzv4uAbwLvpfEtgWUR8U4aXwAMSMMDgFqANH156m9m1mk0G36SxgJLImJWJVcsabKkmZJmLl26tJKLNjNrVkvO/D4BHCJpPnAdxeXuxUAfSfX/9HwgUJeG64BBAGl6b+C1hguNiCkRMToiRtfU1LRrI8zMWqvZ8IuIb0XEwIgYAhwB3BsRE4D7gMNTt2OA6Wn45jROmn5vRERFqzYza6f2/J7facDXJc2juKd3eWq/HNgytX8dOL19JZqZVV735ru8LyLuB+5Pwy8AuzfS5x/AZypQm5lZh/FfeJhZlhx+ZpYlh5+ZZcnhZ2ZZcviZWZYcfmaWJYefmWXJ4WdmWXL4mVmWHH5mliWHn5llyeFnZlly+JlZlhx+ZpYlh5+ZZcnhZ2ZZcviZWZYcfmaWJYefmWXJ4WdmWXL4mVmWHH5mliWHn5llyeFnZlly+JlZlhx+ZpYlh5+ZZcnhZ2ZZcviZWZYcfmaWJYefmWXJ4WdmWXL4mVmWHH5mliWHn5llyeFnZlly+JlZlhx+ZpYlh5+ZZcnhZ2ZZcviZWZaaDT9JG0maIekJSc9I+kFq317So5LmSfqNpA1Te880Pi9NH9Kxm2Bm1notOfN7G9gnIj4GjAQOlLQncB5wYUQMBd4AJqX+k4A3UvuFqZ+ZWafSbPhF4e9ptEd6BLAP8NvUPhU4NA2PS+Ok6ftKUsUqNjOrgBbd85O0gaTZwBLgbuBvwLKIeCd1WQAMSMMDgFqANH05sGUlizYza68WhV9EvBsRI4GBwO7ADu1dsaTJkmZKmrl06dL2Ls7MrFVa9WlvRCwD7gPGAH0kdU+TBgJ1abgOGASQpvcGXmtkWVMiYnREjK6pqWlj+WZmbdOST3trJPVJwxsD+wPPUYTg4anbMcD0NHxzGidNvzciopJFm5m1V/fmu9AfmCppA4qwvD4ibpH0LHCdpLOAvwCXp/6XA9dImge8DhzRAXWbmbVLs+EXEU8CuzbS/gLF/b+G7f8APlOR6szMOoj/wsPMsuTwM7MsOfzMLEsOPzPLksPPzLLk8DOzLDn8zCxLDj8zy5LDz8yy5PAzsyw5/MwsSw4/M8uSw8/MsuTwM7MsOfzMLEsOPzPLksPPzLLk8DOzLDn8zCxLDj8zy5LDz8yy5PAzsyw5/MwsSw4/M8uSw8/MsuTwM7MsOfzMLEsOPzPLksPPzLLk8DOzLDn8zCxLDj8zy5LDz8yy5PAzsyw5/MwsSw4/M8uSw8/MsuTwM7MsOfzMLEsOPzPLUrPhJ2mQpPskPSvpGUlfTe19Jd0taW76uUVql6RLJM2T9KSkj3f0RpiZtVZLzvzeAU6JiJ2APYETJe0EnA7cExHDgHvSOMBBwLD0mAz8vOJVm5m1U7PhFxELI+LxNLwSeA4YAIwDpqZuU4FD0/A44OooPAL0kdS/4pWbmbVDq+75SRoC7Ao8CmwdEQvTpEXA1ml4AFBbmm1BajMz6zRaHH6SNgNuBE6OiBXlaRERQLRmxZImS5opaebSpUtbM6uZWbu1KPwk9aAIvmkR8bvUvLj+cjb9XJLa64BBpdkHprY1RMSUiBgdEaNramraWr+ZWZu05NNeAZcDz0XET0qTbgaOScPHANNL7UenT333BJaXLo/NzDqF7i3o8wngKOApSbNT27eBc4HrJU0CXgI+m6bdBhwMzAPeBI6taMVmZhXQbPhFxIOAmpi8byP9AzixnXWZmXUo/4WHmWXJ4WdmWXL4mVmWHH5mliWHn5llyeFnZlly+JlZlhx+ZpYlh5+ZZcnhZ2ZZcviZWZYcfmaWJYefmWXJ4WdmWXL4mVmWHH5mliWHn5llyeFnZlly+JlZlhx+ZpYlh5+ZZcnhZ2ZZcviZWZYcfmaWJYefmWXJ4WdmWXL4mVmWHH5mliWHn5llyeFnZlly+JlZlhx+ZpYlh5+ZZcnhZ2ZZcviZWZYcfmaWJYefmWXJ4WdmWXL4mVmWHH5mliWHn5llqdnwk3SFpCWSni619ZV0t6S56ecWqV2SLpE0T9KTkj7ekcWbmbVVS878rgIObNB2OnBPRAwD7knjAAcBw9JjMvDzypRpZlZZzYZfRDwAvN6geRwwNQ1PBQ4ttV8dhUeAPpL6V6pYM7NKaes9v60jYmEaXgRsnYYHALWlfgtS2wdImixppqSZS5cubWMZZmZt0+4PPCIigGjDfFMiYnREjK6pqWlvGWZmrdLW8Ftcfzmbfi5J7XXAoFK/ganNzKxTaWv43Qwck4aPAaaX2o9On/ruCSwvXR6bmXUa3ZvrIOnXwN7AVpIWAN8DzgWulzQJeAn4bOp+G3AwMA94Ezi2A2o2M2u3ZsMvIj7XxKR9G+kbwIntLcrMrKP5LzzMLEsOPzPLksPPzLLk8DOzLDn8zCxLDj8zy5LDz8yy5PAzsyw5/MwsSw4/M8uSw8/MsuTwM7MsOfzMLEsOPzPLksPPzLLk8DOzLDn8zCxLDj8zy5LDz8yy5PAzsyw5/MwsSw4/M8uSw8/MsuTwM7MsOfzMLEsOPzPLksPPzLLk8DOzJh133HH069eP4cOHr2574oknGDNmDLvssguf/vSnWbFiRRUrbDuHn5k1aeLEidxxxx1rtB1//PGce+65PPXUU4wfP54LLrigStW1j8PPzJq011570bdv3zXa/vrXv7LXXnsBsP/++3PjjTdWo7R2c/iZWavsvPPOTJ8+HYAbbriB2traKlfUNg4/M2uVK664gksvvZRRo0axcuVKNtxww2qX1Cbdq12AmXUtO+ywA3fddRdQXALfeuutVa6obXzmZ2atsmTJEgDee+89zjrrLE444YQqV9Q2PvMz6yT6D9yORXWd9/6ZpEbbr7nmGiZNmtTh699mwCAWLni5Ystz+Jl1Eovqahl82i3VLqPTeum8sRVdni97zSxLDj8zy5LDz8yy5PAzsyx1SPhJOlDS85LmSTq9I9ZhZtYeFQ8/SRsAPwMOAnYCPidpp0qvx8ysPTrizG93YF5EvBAR/wSuA8Z1wHrMzNqsI8JvAFD+Tc0Fqc3MrNNQRFR2gdLhwIERcXwaPwrYIyK+3KDfZGByGv0o8HxFC1n3tgJerXYRmfC+XjfWh/08OCJqGpvQEX/hUQcMKo0PTG1riIgpwJQOWH9VSJoZEaOrXUcOvK/XjfV9P3fEZe9jwDBJ20vaEDgCuLkD1mNm1mYVP/OLiHckfRm4E9gAuCIinqn0eszM2qNDvtggIm4DbuuIZXdi680lfBfgfb1urNf7ueIfeJiZdQX+8zYzy5LDL5F0kqTnJE3roOV/X9KpHbHsnEnaW5K/BK8NJPWR9KXSeFb70uH3vi8B+0fEhGoXYraO9KF43VeEpC715cgOP0DSZcCHgNslnSHpCkkzJP1F0rjUZ6KkmyTdLWm+pC9L+nrq84ikvqnfFyQ9JukJSTdK2qSR9X1Y0h2SZkn6k6Qd1u0Wdy6ShkiaI+kqSX+VNE3SfpIekjRX0u7p8ee0vx+W9NFGlrNpY8+dFdLr9en0OBk4F/iwpNmS6v/z+GaSfpuej2lK310vaZSkP6bX7J2S+qf2+yVdJGkm8NXqbFkbRYQfxYc+8yl+o/1HwJGprQ/wV2BTYCIwD+gF1ADLgRNSvwuBk9PwlqVlngV8JQ1/Hzg1Dd8DDEvDewD3Vnv7q7zvhwDvALtQHJBnAVcAovi78JuAzYHuqf9+wI1peG/gljTc6HNX7e3rDA9gFPBUei1vBjwD7Ao8Xeqzd3pdD0zPw5+BTwI9gIeBmtTvPyh+hQ3gfuDSam9fWx5d6jR1HTkAOKR0f24jYLs0fF9ErARWSloO/HdqfwoYkYaHSzqL4s23GcXvO64maTPgX4AbSv8QpmdHbEgX82JEPAUg6RngnogISU9RhGNvYKqkYUBQvCEbauq5e66ji+8CPgn8PiJWAUj6HfCpRvrNiIgFqc9sin2/DBgO3J1esxsAC0vz/Kbjyu44Dr8PEnBYRKzxt8aS9gDeLjW9Vxp/j/f35VXAoRHxhKSJFEfTsm7AsogYWdmyu7zm9u2ZFAef8ZKGUJxxNNToc2etUn4e3qXY9wKeiYgxTcyzqsOr6gC+5/dBdwJfKd3r2LWV8/cCFkrqAXzgw5OIWAG8KOkzafmS9LF21pyD3rz/N+ITm+jT3uduffYn4FBJm0jaFBgPPETxem3O80CNpDEAknpI2rnjSl03HH4fdCbFJdWT6fLrzFbO/13gUYoX1pwm+kwAJkl6guLei2/MN+984BxJf6HpK5b2PnfrrYh4nOKqZAbF6/OXETELeCh9AHLBWub9J3A4cF56zc6muHXTpfkvPMwsSz7zM7MsOfzMLEsOPzPLksPPzLLk8DOzLDn8zCxLDj8zy5LDz8yy9L+UeAYlG7ZNkQAAAABJRU5ErkJggg==\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["###age\n","\n","Since Age is ratio data, we can use the describe function and a histogram to take a look at the data"],"metadata":{"id":"2JwwcTIHvmne"}},{"cell_type":"code","source":["com_df.age.describe()"],"metadata":{"id":"YsnQLs8G1MZZ","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1666352654989,"user_tz":-120,"elapsed":5,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"b407d5a8-af3d-4567-9462-0678fa911944"},"execution_count":28,"outputs":[{"output_type":"execute_result","data":{"text/plain":["count    1000.00000\n","mean       26.30700\n","std         2.81081\n","min        21.00000\n","25%        24.00000\n","50%        26.00000\n","75%        28.00000\n","max        32.00000\n","Name: age, dtype: float64"]},"metadata":{},"execution_count":28}]},{"cell_type":"code","source":["com_df.age.hist(bins=25)"],"metadata":{"id":"ta157ai6vn8h","colab":{"base_uri":"https://localhost:8080/","height":0},"executionInfo":{"status":"ok","timestamp":1666352655305,"user_tz":-120,"elapsed":320,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"4264267d-fd77-48e7-f480-cc167cd75fd0"},"execution_count":29,"outputs":[{"output_type":"execute_result","data":{"text/plain":["<matplotlib.axes._subplots.AxesSubplot at 0x7f4536d6a690>"]},"metadata":{},"execution_count":29},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["What we find here is that we have ages between 21-32 with a not-entirely, but somewhat normal distribution. Furthermore, we can assume that there are no missing values indicated by a 0 or 100 in age"],"metadata":{"id":"Yz-yOpcO1civ"}},{"cell_type":"markdown","source":["###nationality\n","\n","Nationality is also nominal categorical data, so we can take a look at it again like Gender"],"metadata":{"id":"Ls-Ie4_QvoJt"}},{"cell_type":"code","source":["com_df.nationality.unique()"],"metadata":{"id":"wiJz8SYJ19bw","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1666352655305,"user_tz":-120,"elapsed":4,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"a47a3343-bc42-4752-b184-7b8be805d1e2"},"execution_count":30,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array(['Dutch', 'Belgian', 'German'], dtype=object)"]},"metadata":{},"execution_count":30}]},{"cell_type":"code","source":["nationalities = [com_df['nationality'].where(com_df['nationality'] == 'Dutch').count(), com_df['nationality'].where(com_df['nationality'] == 'Belgian').count(), com_df['nationality'].where(com_df['nationality'] == 'German').count()]\n","nationality_label = ['Dutch', 'Belgian', 'German']\n","\n","fig = plt.figure(figsize =(5, 5))\n","plt.ylim([0,850])\n","plt.title('Nationality occurences in company B dataset')\n","plt.bar(nationality_label, nationalities, ec='black')\n","addlabels(nationality_label, nationalities)\n","plt.show()"],"metadata":{"id":"MxJLqAH019Tc","colab":{"base_uri":"https://localhost:8080/","height":0},"executionInfo":{"status":"ok","timestamp":1666352655588,"user_tz":-120,"elapsed":286,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"e70aa209-0dfc-4a39-d128-6d19efc92713"},"execution_count":31,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 360x360 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["Here we see that we have an abundance of Dutch people, with similar numbers of Belgians and Germans. Again, no missing values"],"metadata":{"id":"wBXJwnau2kGa"}},{"cell_type":"markdown","source":["###sport\n","\n","Sport is nominal categoric data, so we do the same"],"metadata":{"id":"vzLyL83K2tF5"}},{"cell_type":"code","source":["com_df.sport.unique()"],"metadata":{"id":"2wyV8STd2zP1","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1666352655589,"user_tz":-120,"elapsed":7,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"c84e9b80-3b60-4403-8b52-c3d54081cdab"},"execution_count":32,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array(['Swimming', 'Chess', 'Rugby', 'Football', 'Tennis', 'Cricket',\n","       'Golf', 'Running'], dtype=object)"]},"metadata":{},"execution_count":32}]},{"cell_type":"code","source":["sports = [com_df['sport'].where(com_df['sport'] == 'Swimming').count(), com_df['sport'].where(com_df['sport'] == 'Chess').count(), com_df['sport'].where(com_df['sport'] == 'Rugby').count(), com_df['sport'].where(com_df['sport'] == 'Football').count(), com_df['sport'].where(com_df['sport'] == 'Tennis').count(), com_df['sport'].where(com_df['sport'] == 'Cricket').count(), com_df['sport'].where(com_df['sport'] == 'Golf').count(), com_df['sport'].where(com_df['sport'] == 'Running').count() ]\n","sports_label = ['Swimming', 'Chess', 'Rugby', 'Football', 'Tennis', 'Cricket','Golf', 'Running']\n","\n","fig = plt.figure(figsize =(10, 5))\n","plt.ylim([0,250])\n","plt.title('Sport occurences in company B dataset')\n","plt.bar(sports_label, sports, ec='black')\n","addlabels(sports_label, sports)\n","plt.show()"],"metadata":{"id":"pfKeBoqR23um","colab":{"base_uri":"https://localhost:8080/","height":0},"executionInfo":{"status":"ok","timestamp":1666352655813,"user_tz":-120,"elapsed":230,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"6f8b1186-b4ac-48ab-9cae-8eafdc54cfec"},"execution_count":33,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 720x360 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["What we find here is that there is a somewhat uniform distribution of sports, with Swimming, Rugby and Football standing out. No missing data again."],"metadata":{"id":"UzFbQLjL3yA_"}},{"cell_type":"markdown","source":["###ind_university_grade\n","\n","ind_university_grade is ratio data, so we do the same as age\n"],"metadata":{"id":"LSGb3NHL3_r8"}},{"cell_type":"code","source":["com_df.ind_university_grade.describe()"],"metadata":{"id":"mLwJTwIs4LoF","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1666352655814,"user_tz":-120,"elapsed":4,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"254f5807-32dd-4c75-8d29-251f9043ef6d"},"execution_count":34,"outputs":[{"output_type":"execute_result","data":{"text/plain":["count    1000.000000\n","mean       62.348000\n","std         6.015237\n","min        47.000000\n","25%        58.000000\n","50%        62.000000\n","75%        67.000000\n","max        78.000000\n","Name: ind_university_grade, dtype: float64"]},"metadata":{},"execution_count":34}]},{"cell_type":"code","source":["com_df.ind_university_grade.hist(bins=31)"],"metadata":{"id":"OIIL0-_P5xGd","colab":{"base_uri":"https://localhost:8080/","height":0},"executionInfo":{"status":"ok","timestamp":1666352656063,"user_tz":-120,"elapsed":252,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"96c55479-59cb-4419-aba6-6b5e3b18413a"},"execution_count":35,"outputs":[{"output_type":"execute_result","data":{"text/plain":["<matplotlib.axes._subplots.AxesSubplot at 0x7f4537046750>"]},"metadata":{},"execution_count":35},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["What we see here is that university grades are distributed normally with no missing vlaues or extreme outliers"],"metadata":{"id":"4iMdaWOj55O_"}},{"cell_type":"markdown","source":["###ind_debateclub\n","\n","With ind_debateblub being a boolean category, we expect to only see True and False, let's see whether this is actually the case"],"metadata":{"id":"tmlnJGJM6E9j"}},{"cell_type":"code","source":["com_df.ind_debateclub.unique()"],"metadata":{"id":"_hyGL4CT6SeH","executionInfo":{"status":"ok","timestamp":1666352656063,"user_tz":-120,"elapsed":4,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"colab":{"base_uri":"https://localhost:8080/"},"outputId":"48df9693-0ef1-43ec-ad63-c58830796356"},"execution_count":36,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([False,  True])"]},"metadata":{},"execution_count":36}]},{"cell_type":"code","source":["debate_club = [com_df['ind_debateclub'].where(com_df['ind_debateclub'] == False).count(), com_df['ind_debateclub'].where(com_df['ind_debateclub'] == True).count()]\n","debate_labels = [\"False\", \"True\"]\n","plt.ylim([0,800])\n","plt.title('Debate Club Experience in company B dataset')\n","plt.bar(debate_labels, debate_club, ec='black')\n","addlabels(debate_labels, debate_club)\n","plt.show()"],"metadata":{"id":"yg2aKuZE6ZUz","executionInfo":{"status":"ok","timestamp":1666352656427,"user_tz":-120,"elapsed":367,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"colab":{"base_uri":"https://localhost:8080/","height":0},"outputId":"e82b0e5f-46b1-44a7-fb81-29382d7a8435"},"execution_count":37,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["Here we see that most candidates are not in a debateclub and there are no missing values"],"metadata":{"id":"hB9lFF9Z7Chr"}},{"cell_type":"markdown","source":["###ind_programming_exp\n","\n","ind_programming_exp is another boolean category, so let's check for missing values and distribution"],"metadata":{"id":"ezWNaoNp7JZd"}},{"cell_type":"code","source":["com_df.ind_programming_exp.unique()"],"metadata":{"id":"eLdRXfYA7TS1","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1666352656427,"user_tz":-120,"elapsed":6,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"8ce11b70-2e6a-439d-8cc7-c2f464a5a19d"},"execution_count":38,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([False,  True])"]},"metadata":{},"execution_count":38}]},{"cell_type":"code","source":["programming = [com_df['ind_programming_exp'].where(com_df['ind_programming_exp'] == False).count(), com_df['ind_programming_exp'].where(com_df['ind_programming_exp'] == True).count()]\n","programming_labels = [\"False\", \"True\"]\n","plt.ylim([0,720])\n","plt.title('Programming Experience in company B dataset')\n","plt.bar(programming_labels, programming, ec='black')\n","addlabels(programming_labels, programming)\n","plt.show()"],"metadata":{"id":"VreJ4TW27XrX","colab":{"base_uri":"https://localhost:8080/","height":0},"executionInfo":{"status":"ok","timestamp":1666352656428,"user_tz":-120,"elapsed":6,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"31314037-19aa-4d29-cd3d-3c48975640ef"},"execution_count":39,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["We see here again no missing values and more no experience than programming exprience"],"metadata":{"id":"4llR5dVL7uaY"}},{"cell_type":"markdown","source":["###ind_international_exp\n","\n","ind_international_exp is another boolean category, so let's check for missing values and distribution\n","\n","\n"],"metadata":{"id":"R07GnAb971IN"}},{"cell_type":"code","source":["com_df.ind_international_exp.unique()"],"metadata":{"id":"1J7T092e78ok","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1666352656428,"user_tz":-120,"elapsed":6,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"e4267a7a-20de-4adf-aa7a-921a3044deaf"},"execution_count":40,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([False,  True])"]},"metadata":{},"execution_count":40}]},{"cell_type":"code","source":["international = [com_df['ind_international_exp'].where(com_df['ind_international_exp'] == False).count(), com_df['ind_international_exp'].where(com_df['ind_international_exp'] == True).count()]\n","international_labels = [\"False\", \"True\"]\n","plt.ylim([0,800])\n","plt.title('International Experience in company B dataset')\n","plt.bar(international_labels, international, ec='black')\n","addlabels(international_labels, international)\n","plt.show()"],"metadata":{"id":"axNz7_mm7_Mk","colab":{"base_uri":"https://localhost:8080/","height":0},"executionInfo":{"status":"ok","timestamp":1666352656798,"user_tz":-120,"elapsed":375,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"90827c9b-6725-4077-8a1e-3c6838e31414"},"execution_count":41,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["We see here again no missing values and more no experience than international experience"],"metadata":{"id":"mCsXxa5Q8QKR"}},{"cell_type":"markdown","source":["###ind_enterpreneur_exp\n","\n","ind_entrepeneur_exp is another boolean category, so let's check for missing values and distribution"],"metadata":{"id":"74skVGQU8U0S"}},{"cell_type":"code","source":["com_df.ind_entrepeneur_exp.unique()"],"metadata":{"id":"0j-XFBKu8axT","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1666352656799,"user_tz":-120,"elapsed":7,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"1d498f5d-e159-4bdb-f31a-5e3e4704b5b1"},"execution_count":42,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([False,  True])"]},"metadata":{},"execution_count":42}]},{"cell_type":"code","source":["entrepeneur = [com_df['ind_entrepeneur_exp'].where(com_df['ind_entrepeneur_exp'] == False).count(), com_df['ind_entrepeneur_exp'].where(com_df['ind_entrepeneur_exp'] == True).count()]\n","entrepeneur_labels = [\"False\", \"True\"]\n","plt.ylim([0,800])\n","plt.title('Entrepeneur Experience in company B dataset')\n","plt.bar(entrepeneur_labels, entrepeneur, ec='black')\n","addlabels(entrepeneur_labels, entrepeneur)\n","plt.show()"],"metadata":{"id":"A6lvQk6q8enf","colab":{"base_uri":"https://localhost:8080/","height":0},"executionInfo":{"status":"ok","timestamp":1666352656799,"user_tz":-120,"elapsed":6,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"9b1b2b20-32d8-4e8d-ebb6-f2222573af39"},"execution_count":43,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["We see here again no missing values and more no experience than entrepeneur experience"],"metadata":{"id":"syC0GEJY8rU4"}},{"cell_type":"markdown","source":["###ind_languages\n","\n","This should be the number of languages spoken fluently by the candidate. Assuming that everyone speaks at least 1 language in order to be able to communicate, we can assume that a 0 here consitutes missing data. "],"metadata":{"id":"gx6xykIe8tn5"}},{"cell_type":"code","source":["com_df.ind_languages.describe()"],"metadata":{"id":"Oz7gT0Dn-tZv","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1666352656800,"user_tz":-120,"elapsed":7,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"b50b723b-773b-4467-c0ca-ff5aa5da83d1"},"execution_count":44,"outputs":[{"output_type":"execute_result","data":{"text/plain":["count    1000.000000\n","mean        1.375000\n","std         0.856217\n","min         0.000000\n","25%         1.000000\n","50%         1.000000\n","75%         2.000000\n","max         3.000000\n","Name: ind_languages, dtype: float64"]},"metadata":{},"execution_count":44}]},{"cell_type":"code","source":["com_df.ind_languages.hist()"],"metadata":{"id":"Q8zUjOd--2CG","colab":{"base_uri":"https://localhost:8080/","height":0},"executionInfo":{"status":"ok","timestamp":1666352657109,"user_tz":-120,"elapsed":315,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"a06237bc-a496-4d14-dab2-292a3266402b"},"execution_count":45,"outputs":[{"output_type":"execute_result","data":{"text/plain":["<matplotlib.axes._subplots.AxesSubplot at 0x7f4536b4bad0>"]},"metadata":{},"execution_count":45},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["What we find here is that there are about 150 instances of candidates speaking no language. This could be a case of missing data, but without access to the domain experts, we cannot be sure. "],"metadata":{"id":"kezDd_bZ_E9N"}},{"cell_type":"markdown","source":["###ind_exact_study\n","\n","ind_exact_study is another boolean category, so let's check for missing values and distribution"],"metadata":{"id":"oBz_sFCB8zU8"}},{"cell_type":"code","source":["com_df.ind_exact_study.unique()"],"metadata":{"id":"qg6rXFW889ZR","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1666352657110,"user_tz":-120,"elapsed":10,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"b81b6209-8dba-420f-d3b1-45b2d3205732"},"execution_count":46,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([ True, False])"]},"metadata":{},"execution_count":46}]},{"cell_type":"code","source":["exact_study = [com_df['ind_exact_study'].where(com_df['ind_exact_study'] == False).count(), com_df['ind_exact_study'].where(com_df['ind_exact_study'] == True).count()]\n","exact_study_labels = [\"False\", \"True\"]\n","fig = plt.figure(figsize =(5, 5))\n","plt.ylim([0,550])\n","plt.title('Exact Study Experience in company B dataset')\n","plt.bar(exact_study_labels, exact_study, ec='black')\n","addlabels(exact_study_labels, exact_study)\n","plt.show()"],"metadata":{"id":"_Hav5u8I9Fu5","colab":{"base_uri":"https://localhost:8080/","height":0},"executionInfo":{"status":"ok","timestamp":1666352657110,"user_tz":-120,"elapsed":9,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"7cd3dacf-5b72-43eb-c034-2ac6fa7b126c"},"execution_count":47,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 360x360 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["We see here again no missing values and almost a half split of exact study participation"],"metadata":{"id":"z4YqSDIm9cmv"}},{"cell_type":"markdown","source":["###ind_degree\n","\n","ind_degree is an ordinal feature that should only contain bachelor, masters and phd, let's see if that's the case and what the distribution is\n"],"metadata":{"id":"z22RYo8e9maW"}},{"cell_type":"code","source":["com_df.ind_degree.unique()"],"metadata":{"id":"GVpQ54339wJH","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1666352657111,"user_tz":-120,"elapsed":9,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"13d9147e-b80f-4aa3-abff-ba585103a174"},"execution_count":48,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array(['bachelor', 'master', 'phd'], dtype=object)"]},"metadata":{},"execution_count":48}]},{"cell_type":"code","source":["degrees = [com_df['ind_degree'].where(com_df['ind_degree'] == 'bachelor').count(), com_df['ind_degree'].where(com_df['ind_degree'] == 'master').count(), com_df['ind_degree'].where(com_df['ind_degree'] == 'phd').count()]\n","degree_label = ['bachelor', 'master', 'phd']\n","\n","fig = plt.figure(figsize =(5, 5))\n","plt.ylim([0,500])\n","plt.title('Degree occurences in company B dataset')\n","plt.bar(degree_label, degrees, ec='black')\n","addlabels(degree_label, degrees)\n","plt.show()"],"metadata":{"id":"JLBB6OA894ls","colab":{"base_uri":"https://localhost:8080/","height":0},"executionInfo":{"status":"ok","timestamp":1666352657372,"user_tz":-120,"elapsed":269,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"2e705ef7-f872-4a13-94a3-75939457f6fc"},"execution_count":49,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 360x360 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":[">"],"metadata":{"id":"v2O3TZ3ayvXH"}},{"cell_type":"markdown","source":["#**2.5 Data Preprocessing 2**\n","\n","Now that we have insight on the data, let us create a basic preprocessing function that 0/1 encodes the Boolean data, One-hot encodes the nominal data and rank encodes the ordinal data\n","\n"],"metadata":{"id":"xDg9DQfYBqw0"}},{"cell_type":"markdown","source":["First define 4 categories on which we can seperate the features in our dataframe. This is useful for preprocessing the data and might be useful later on as well.\n","\n","> Boolean data: Data classified with True/False.\n","\n","> Nominal data: Data classified into different categories without ranking.\n","\n","> Ordinal data: Data classified into different categories with ranking.\n","\n","> Numeric data: Numeric data based on normally valued real numbers.\n"],"metadata":{"id":"oOjSsTQd0oCp"}},{"cell_type":"code","source":["boo_col = ['ind_debateclub', 'ind_programming_exp', 'ind_international_exp', 'ind_entrepeneur_exp', 'ind_exact_study', 'decision']\n","\n","nom_col = ['gender', 'nationality', 'sport']\n","\n","ord_col = ['ind_degree']\n","\n","num_col= ['ind_university_grade', 'age', 'ind_languages']"],"metadata":{"id":"Tbr8rpT4inRN","executionInfo":{"status":"ok","timestamp":1666352800298,"user_tz":-120,"elapsed":404,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":64,"outputs":[]},{"cell_type":"markdown","source":["Second, we create a function that will automatically preprocess any data set that we feed into it. We will use one-hot, 0/1 and label encoding here"],"metadata":{"id":"LZW_TFj1FIdE"}},{"cell_type":"code","source":["def preprocess(df):\n","  #Check which data is in the dataframe\n","    features = list(df.columns[:])\n","\n","\n","    #Check which features are in the supplied dataframe\n","    df_boo_col = list(boo_col)\n","    to_boo_remove = []\n","    for i in df_boo_col:\n","      if i not in features:\n","        to_boo_remove.append(i)\n","    for i in to_boo_remove:\n","      df_boo_col.remove(i)\n","\n","    df_nom_col = list(nom_col)\n","    to_nom_remove = []\n","    for i in df_nom_col:\n","      if i not in features:\n","        to_nom_remove.append(i)\n","    for i in to_nom_remove:\n","      df_nom_col.remove(i)\n","\n","\n","    df_ord_col = list(ord_col)\n","    to_ord_remove = []\n","    for i in df_ord_col:\n","      if i not in features:\n","        to_ord_remove.append(i)\n","    for i in to_ord_remove:\n","      df_ord_col.remove(i)\n","\n","    df_num_col = list(num_col)\n","    to_num_remove = []\n","    for i in df_num_col:\n","      if i not in features:\n","        to_num_remove.append(i)\n","    for i in to_num_remove:\n","      df_num_col.remove(i)\n","\n","  # Split boolean, nominal, ordinal, and numerical columns\n","    df_boo = df[df_boo_col]\n","    df_nom = df[df_nom_col]\n","    df_ord = df[df_ord_col]\n","    df_num = df[df_num_col]\n","\n","    #0/1 encode boolean columns\n","    if len(df_boo_col) > 0:\n","      df_boo = df_boo.astype(int)\n","\n","    # One-hot encode nominal columns\n","    if len(df_nom_col) > 0:\n","      df_nom = pd.get_dummies(data=df_nom)\n","\n","    # Label encode ordinal columns\n","    if len(df_ord_col) > 0:\n","      df_ord = df_ord.replace({'bachelor' : 0, 'master' : 1, 'phd' : 2})\n","\n","    # Merge nominal, ordinal, and numerical back together\n","    df = pd.concat([df_boo, df_num, df_nom, df_ord], axis=1)\n","\n","    return df\n"],"metadata":{"id":"sGK5EonWCkrM","executionInfo":{"status":"ok","timestamp":1666352800496,"user_tz":-120,"elapsed":2,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":65,"outputs":[]},{"cell_type":"markdown","source":["##Full Correlation Map\n","\n","This may seem like a weird tagent, but now that we have our preprocess function, we can preprocess our company dataset to create a correlation heatmap with the categorical features (which were excluded from our initial heatmap)"],"metadata":{"id":"5v-uNwOxRKjp"}},{"cell_type":"code","source":["# Create correlation between these columns of data\n","corr = preprocess(com_df).corr()\n","\n","\n","plt.figure(figsize=(25,15))\n","\n","# Create array of 0s in same shape as corr.shape\n","mask = np.zeros_like(corr)\n","# Take upper triangle of the shape of 0s\n","mask[np.triu_indices_from(mask)] = True\n","\n","# Change mask and correlation to remove empty rows\n","mask = mask[1:, :-1]\n","corr = corr.iloc[1:,:-1].copy()\n","\n","# Choose a diverging palette for the r values (0 should be white)\n","cmap = sns.diverging_palette(220, 20, as_cmap=True)\n","\n","# Plot the heatmap\n","sns.heatmap(corr,         # Create heatmap of the df.corr created above\n","            mask=mask,    # Use the mask to not show repeated boxes\n","            annot=True,   # Annotate the squares with the r value\n","            fmt=\".2f\",    # Use 2 decimals\n","            annot_kws={\"size\": 8}, \n","            cmap='RdYlGn',    # cmap is the diverging one created above\n","            square=True,  # Square boxes look better\n","            cbar_kws={\"shrink\": .8}, # Shrink color bar to center it\n","            vmin=-1, vmax=1) # Max and Min values for the colors\n","\n","plt.show()\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":965},"id":"B3N97U0oRHme","executionInfo":{"status":"ok","timestamp":1666352802668,"user_tz":-120,"elapsed":2173,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"1089a4f5-f6da-4382-9dca-e786c1632416"},"execution_count":66,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 1800x1080 with 2 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["# **3. Model Fitting**\n","Our Research question is as follows:\n","*Which set of four features best determines an applicant’s hiring for company B?*\n","\n","In order to answer this we will try to find the best 4 indicator features and the best 4 desciptor features and compare their performance. \n","\n","1. Find the best indicator features with DT with RF\n","> Model type: Decision trees with Random Forests (i.e. sklearn.tree.DecisionTreeClassifier)\n","\n","  > Features: All combinations of 4 inds\n","\n","  > Steps:\n","  > *   Use the ind_df\n","  > *   Use grid search to prepare all possible set of 4 indicator combinations \n","  > *   Fit and evaluate model on each set and pick best accuracy\n","\n","\n","2. Use SVM to model the found best 4 inds\n","> Model type: SVM\n","\n","  > Features: Best 4 inds from M1\n","\n","  > Steps:\n","  > *   Use the ind_df\n","  > *   Fit model with best 4 features from M1\n","  > *   In case of large disreptancy, can always fit model on second-best M1 set to provide context\n","\n","3. Use the same SVM model as M2 with descriptor features to get proper comparison of feature effects\n","> Model type: SVM\n","\n","  > Features: all descriptors\n","\n","  > Steps:\n","  > *   Use the des_df\n","  > *   Fit model with best all descriptors\n","  > *   Compare accuracy of M2 and M3\n"],"metadata":{"id":"XgR60AutnkAp"}},{"cell_type":"markdown","source":["For each model we will undergo a few steps\n","1. df creation: create a new df that we can mess around with\n","2. feature selection: drop features that we don't want to use\n","3. data preprocessing: deal with missing values, use techniques to adjust data\n","4. Hyperparameter tuning: tune the hyperparameters of the model\n","4. Train/Test split the data, Train the Model and Validate it"],"metadata":{"id":"O3_PEHjczlz0"}},{"cell_type":"markdown","source":["To run these models for step 4, let's create a classifier function that does the train-test split of the data for us and give us a ROC graph as a curve of performance."],"metadata":{"id":"z2dEp6My2nqP"}},{"cell_type":"code","source":["def classifier(model, df, drop=['decision'], target=['decision'], frac=0.3):\n","    # Training set is a sample of 30%. Seed is used to get consistent results while testing\n","\n","\n","    val = df.sample(frac=frac,random_state=200)\n","    # training is set acquired by dropping the validation set\n","    train = df.drop(val.index)\n","\n","    x = train.drop(drop, axis=1)\n","    y = np.ravel(train[target].values) # transforming the array to a 1D array\n","    \n","    clas = model.fit(x, y)\n","\n","    valx = val.drop(drop, axis=1)\n","    valy = val[target].values\n","    print('-----------------------')\n","    print(\"Training results\")\n","    print(\"Accuracy:\")\n","    print(clas.score(x, y))    \n","\n","    print(\"Validation results\")\n","    print(\"Accuracy:\")\n","    print(clas.score(valx, valy))\n","\n","\n","    predictions = clas.predict(val.drop(['decision'], axis=1))\n","\n","    result = pd.DataFrame(columns=['Prediction', 'Actual'])\n","\n","    result = result.assign(Actual=val['decision'])  \n","    result = result.assign(Prediction=predictions)\n","\n","    \n","    print(\"Confusion matrix:\")\n","    val_matrix = plot_confusion_matrix(model, valx,valy, cmap=plt.cm.Reds)\n","    val_matrix.ax_.set_title('Confusion Matrix Validation Set')\n","    plt.xlabel('Predicted label')\n","    plt.ylabel('True label')\n","    plt.gcf().axes[0].tick_params(colors='black')\n","    plt.gcf().axes[1].tick_params(colors='black')\n","    plt.gcf().set_size_inches(10,6)\n","    plt.show()\n","    \n","    try:   \n","      #Prediction probabilties\n","      base_probs = [0 for _ in range(len(valy))]\n","      model_probs = clas.predict_proba(valx)\n","      #Keeping probabilities for positive outcomes\n","      model_probs = model_probs[:,1]\n","      #Compute AUROC values\n","      base_auc = roc_auc_score(valy, base_probs)\n","      model_auc = roc_auc_score(valy, model_probs)\n","      #Display AUROC scores\n","      print(\"Random prediction: AUROC = %.3f\" %(base_auc))\n","      print(\"Model prediction: AUROC = %.3f\" %(model_auc))\n","      #Calculate ROC curve\n","      base_fpr, base_tpr, _ = roc_curve(valy, base_probs)\n","      model_fpr, model_tpr, _ = roc_curve(valy, model_probs)\n","      #Plot the ROC curve\n","      plt.figure(figsize=(15,10))\n","      plt.plot(base_fpr, base_tpr, linestyle='--', label='Random prediction (AUROC = %.3f)' %base_auc)\n","      plt.plot(model_fpr, model_tpr, linestyle='--', label='Random prediction (AUROC = %.3f)' %model_auc)\n","      plt.title('ROC Plot')\n","      plt.xlabel('False Positive Rate')\n","      plt.ylabel('True Positive Rate')\n","      plt.legend()\n","      plt.show()\n","    except:\n","      pass\n","\n","    return confusion_matrix(valy, clas.predict(valx)), result\n","    \n","    "],"metadata":{"id":"ed-gOPhG25ib","executionInfo":{"status":"ok","timestamp":1666355414995,"user_tz":-120,"elapsed":405,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":167,"outputs":[]},{"cell_type":"markdown","source":["We'll also create an evaluation function which calculates all the statistical concepts we can derive from the confusion matrix. This is something that we can use to evaluate the general performance of the model or compare the performance of the model on sub-groups"],"metadata":{"id":"iVxd1nsKg-Wt"}},{"cell_type":"code","source":["def evaluation(results):\n","  true_positive = results[1,1]\n","  false_positive = results[0,1]\n","  false_negative = results[1,0]\n","  true_negative = results[0,0]\n","  actual_positive = true_positive + false_negative\n","  actual_negative = false_positive + true_negative\n","  precision = true_positive / (true_positive + false_positive)\n","  false_discovery_rate = false_positive / (true_positive + false_positive)\n","  false_omission_rate = false_negative / (true_negative + false_negative)\n","  negative_predictive_value = true_negative / (true_negative + false_negative)\n","  recall = true_positive / actual_positive\n","  false_positive_rate = false_positive / actual_negative\n","  false_negative_rate = false_negative / actual_positive\n","  true_negative_rate = true_negative / actual_negative\n","\n","  return [true_positive, false_positive, false_negative, true_negative, actual_positive, actual_negative, precision, false_discovery_rate, false_omission_rate, negative_predictive_value, recall, false_positive_rate, false_negative_rate, true_negative_rate]"],"metadata":{"id":"oXjdALNXeKg0","executionInfo":{"status":"ok","timestamp":1666355415361,"user_tz":-120,"elapsed":20,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":168,"outputs":[]},{"cell_type":"markdown","source":["And a function that prints the evaluation"],"metadata":{"id":"5veGQXFSghal"}},{"cell_type":"code","source":["def print_eval(model_result):\n","  print('---------------------------')\n","  print('True Positive =  ', model_result[0])\n","  print('False Positive = ', model_result[1])\n","  print('False Negative = ', model_result[2])\n","  print('True Negative = ', model_result[3])\n","  print('Actual Positive = ', model_result[4])\n","  print('Actual Negative = ', model_result[5])\n","  print('Precision =', model_result[6])\n","  print('False Discovery Rate = ', model_result[7])\n","  print('False Omission Rate = ', model_result[8])\n","  print('Negative Predictive Value = ', model_result[10])\n","  print('Recall = ', model_result[11])\n","  print('False Positive Rate = ', model_result[12])\n","  print('True Negative Rate = ', model_result[13])"],"metadata":{"id":"ABDjrORSARFQ","executionInfo":{"status":"ok","timestamp":1666355415361,"user_tz":-120,"elapsed":19,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":169,"outputs":[]},{"cell_type":"markdown","source":["Also, we need to make sure that we find the right hyperparameters for our data. This will be done with the following function using gridsearch"],"metadata":{"id":"jAQoclXrChSp"}},{"cell_type":"code","source":["def classifierRF(model, df, kfold=5, ratio=0.3, state=200):\n","    # Separate the test set for final model evaluation\n","    #test = df.sample(frac=frac,random_state=200)\n","    # training is set acquired by dropping the validation set\n","\n","    # splitting data\n","    x,y = df.drop('decision', axis='columns'), df['decision']\n","    x_train, x_test, y_train, y_test = train_test_split(x,\n","                                                          y,\n","                                                          random_state=state,\n","                                                          test_size=ratio)\n","    \n","    # hyperparameters\n","    params = {'n_estimators':[50,60,70,80,90,100,110,120,130,140,150,160,170,180,190,200],\n","              'criterion':['gini', 'entropy','log_loss'],\n","              'max_features':['log2', 'sqrt'],\n","              'max_samples':[10,20,30,40,50,60,70,80,90,100]}\n","\n","\n","\n","          \n","    # gridsearch with kfold\n","    grid = GridSearchCV(estimator=model,\n","                        param_grid=params\n","                        ,cv=kfold)\n","    \n","    # fitting data\n","    grid.fit(x_train, y_train)\n","    grid.score(x_test, y_test)\n","\n","    # presenting result\n","    print('------------------------------------------------------')\n","    print(\"Hyperparameter optimization for RandomForest: \")\n","    print('Grid Search Result')\n","    print(f'Best estimator: {grid.best_estimator_},\\nBest score: {grid.best_score_},\\nBest params: {grid.best_params_}')\n","    #print(pd.DataFrame(grid.cv_results_))\n"],"metadata":{"id":"3MitK75fBPbW","executionInfo":{"status":"ok","timestamp":1666355415362,"user_tz":-120,"elapsed":20,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":170,"outputs":[]},{"cell_type":"code","source":["def classifierSVM(model, df, kfold=5, ratio=0.3, state=200):\n","\n","    # Separate the test set for final model evaluation\n","    #test = df.sample(frac=frac,random_state=200)\n","    # training is set acquired by dropping the validation set\n","    #train = df.drop(test.index)\n","\n","    x,y = df.drop('decision', axis='columns'), df['decision']\n","    x_train, x_test, y_train, y_test = train_test_split(x,\n","                                                          y,\n","                                                          random_state=state,\n","                                                          test_size=ratio)\n","    \n","    # hyperparameters\n","    params = {'C':[1.0, 2.0, 3.0, 4.0, 5.0],\n","          'kernel':['poly','rbf'],\n","          'gamma':['scale', 'auto']}\n","    # gridsearch with kfold\n","    grid = GridSearchCV(estimator=model,\n","                        param_grid=params\n","                        ,cv=KFold(n_splits=kfold))\n","    \n","    # fitting data\n","    grid.fit(x_train, y_train)\n","    grid.score(x_test, y_test)\n","\n","    # presenting result\n","    print(\"Hyperparameter optimization for SVM: \")\n","    print('Grid Search Result')\n","    print(f'Best estimator: {grid.best_estimator_},\\nBest score: {grid.best_score_},\\nBest params: {grid.best_params_}')\n","    #print(pd.DataFrame(grid.cv_results_))"],"metadata":{"id":"xEo7QfRHlSyl","executionInfo":{"status":"ok","timestamp":1666355415363,"user_tz":-120,"elapsed":20,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":171,"outputs":[]},{"cell_type":"markdown","source":["## **Model 1: Decision Trees with Random Forests - 4 Indicators**"],"metadata":{"id":"vak6c46M0L2I"}},{"cell_type":"markdown","source":["### 1. Create Dataset\n","Since Model 1 is only allowed to use indicators, m1_df will be made up from the indicator dataframe"],"metadata":{"id":"UVFbKB_20gI2"}},{"cell_type":"code","source":["m1_df = ind_df"],"metadata":{"id":"HTMxR5cp0kIl","executionInfo":{"status":"ok","timestamp":1666355415363,"user_tz":-120,"elapsed":19,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":172,"outputs":[]},{"cell_type":"markdown","source":["2. Feature selection\n","\n","Here we want to find what set of indicators gives us the best prediction accuracy. To do this, we take the list of all indicators and make all possible unique sets of 4 from them. \n","\n","\n","We do this by using the combination function from itertools and thus create a set list of all possible unique combinations of sets, where order of items is not accounted for in uniqueness (so [A,B,C] and [A,C,B] are see as the same list)."],"metadata":{"id":"N4rDPON_0l-n"}},{"cell_type":"code","source":["sets = combinations(indicator,4)\n","set_list = []\n","for i in sets:\n","  set_list.append(i)"],"metadata":{"id":"b7Xjsp4E1qxA","executionInfo":{"status":"ok","timestamp":1666355415364,"user_tz":-120,"elapsed":20,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":173,"outputs":[]},{"cell_type":"markdown","source":["Once this has been created, let's take a look at how many sets of indicators we have"],"metadata":{"id":"cJTHJZJO5ASZ"}},{"cell_type":"code","source":["len(set_list)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"wbYIoBrV10GK","executionInfo":{"status":"ok","timestamp":1666355415364,"user_tz":-120,"elapsed":19,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"9830c8fb-d0b0-4c36-8e95-5161817636a8"},"execution_count":174,"outputs":[{"output_type":"execute_result","data":{"text/plain":["70"]},"metadata":{},"execution_count":174}]},{"cell_type":"markdown","source":["In order to test these 70 sets, we need to create a few functions which do the following.\n","\n","\n","*  Takes all the sets and rank orders them based on performance\n","\n","> 1. Drop the all the indicators from the dataset apart from the ones in the set\n","> 2. Preprocess the dataset \n","> 3. Split Training and Test set as per the Classifier function and split the training set into training and validation, as we're not testing the model, just using validation set to find the best indicator set as hyperparameter\n","\n","\n","\n"],"metadata":{"id":"VWXSRV5t53NG"}},{"cell_type":"code","source":["def evaluate_inds(setlist):\n","  results= []\n","  for i in setlist: \n","    i_df = drop_features(i)\n","    i_df = preprocess(i_df)\n","    accuracy = eval_classifier(RandomForestEntropy(), i_df)\n","    results.append((accuracy, i))\n","  results.sort(reverse=True)\n","  return results"],"metadata":{"id":"fPQa9ycK6y8I","executionInfo":{"status":"ok","timestamp":1666355415365,"user_tz":-120,"elapsed":14,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":175,"outputs":[]},{"cell_type":"code","source":["def drop_features(setlist):\n","  to_drop_inds = list(indicator)\n","  for i in setlist:\n","    to_drop_inds.remove(i)\n","  i_df = ind_df.drop(to_drop_inds, axis=1)\n","  return i_df\n"],"metadata":{"id":"B0V36CmR7XFS","executionInfo":{"status":"ok","timestamp":1666355415365,"user_tz":-120,"elapsed":13,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":176,"outputs":[]},{"cell_type":"code","source":["def eval_classifier(model, df, drop=['decision'], target=['decision'], frac=0.3):\n","    # Test set is a sample of 30%. Seed is used to get consistent results while testing\n","    # Separate the test set for final model evaluation\n","    test = df.sample(frac=frac,random_state=200)\n","    # training is set acquired by dropping the validation set\n","    train = df.drop(test.index)\n","    #Sample 20% of training data for validation set for evaluation\n","    val = df.sample(frac=0.2,random_state=200)\n","    train = df.drop(val.index)\n","\n","    x = train.drop(drop, axis=1)\n","    y = np.ravel(train[target].values) # transforming the array to a 1D array\n","    \n","    clas = model.fit(x, y)\n","\n","    valx = val.drop(drop, axis=1)\n","    valy = val[target].values\n","    return clas.score(valx, valy)"],"metadata":{"id":"-4iSKtfA8ukY","executionInfo":{"status":"ok","timestamp":1666355415366,"user_tz":-120,"elapsed":14,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":177,"outputs":[]},{"cell_type":"markdown","source":["We now also introduce the Decision Tree model with Random forests using Entropy with default hyperparameters. We will refine these hyperparameters with grid search once we have our chosen set of features"],"metadata":{"id":"gwG6BSEN-8WJ"}},{"cell_type":"code","source":["\n","def RandomForestEntropy():\n","  \"\"\"\n","  A Random forest classifier using entropy as quality of a split\n","  \"\"\"\n","  return RandomForestClassifier(\n","      n_estimators=200,\n","      bootstrap=True,\n","      criterion='entropy',\n","      max_features='sqrt',\n","      max_samples=50\n","  )\n"],"metadata":{"id":"N6CYNkFi9r0K","executionInfo":{"status":"ok","timestamp":1666355415366,"user_tz":-120,"elapsed":13,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":178,"outputs":[]},{"cell_type":"markdown","source":["Then we run the evaluate_inds function"],"metadata":{"id":"OfuSAZhn_Jda"}},{"cell_type":"code","source":["set_results_RF = evaluate_inds(set_list)\n","print(set_results_RF)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"JItpT0CW9iML","executionInfo":{"status":"ok","timestamp":1666355439077,"user_tz":-120,"elapsed":23723,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"fe225099-a89a-4f0c-c56c-2e39d009a01f"},"execution_count":179,"outputs":[{"output_type":"stream","name":"stdout","text":["[(0.91, ('ind_programming_exp', 'ind_international_exp', 'ind_languages', 'ind_university_grade')), (0.91, ('ind_programming_exp', 'ind_international_exp', 'ind_languages', 'ind_degree')), (0.91, ('ind_programming_exp', 'ind_entrepeneur_exp', 'ind_languages', 'ind_degree')), (0.905, ('ind_languages', 'ind_exact_study', 'ind_degree', 'ind_university_grade')), (0.905, ('ind_international_exp', 'ind_entrepeneur_exp', 'ind_languages', 'ind_university_grade')), (0.905, ('ind_debateclub', 'ind_programming_exp', 'ind_international_exp', 'ind_languages')), (0.9, ('ind_programming_exp', 'ind_languages', 'ind_exact_study', 'ind_degree')), (0.9, ('ind_programming_exp', 'ind_international_exp', 'ind_entrepeneur_exp', 'ind_languages')), (0.9, ('ind_programming_exp', 'ind_entrepeneur_exp', 'ind_languages', 'ind_university_grade')), (0.9, ('ind_entrepeneur_exp', 'ind_languages', 'ind_degree', 'ind_university_grade')), (0.895, ('ind_programming_exp', 'ind_languages', 'ind_degree', 'ind_university_grade')), (0.895, ('ind_entrepeneur_exp', 'ind_languages', 'ind_exact_study', 'ind_degree')), (0.89, ('ind_programming_exp', 'ind_entrepeneur_exp', 'ind_languages', 'ind_exact_study')), (0.89, ('ind_international_exp', 'ind_languages', 'ind_exact_study', 'ind_degree')), (0.89, ('ind_debateclub', 'ind_languages', 'ind_degree', 'ind_university_grade')), (0.885, ('ind_international_exp', 'ind_languages', 'ind_degree', 'ind_university_grade')), (0.885, ('ind_international_exp', 'ind_entrepeneur_exp', 'ind_languages', 'ind_exact_study')), (0.885, ('ind_debateclub', 'ind_programming_exp', 'ind_languages', 'ind_exact_study')), (0.885, ('ind_debateclub', 'ind_international_exp', 'ind_languages', 'ind_exact_study')), (0.885, ('ind_debateclub', 'ind_entrepeneur_exp', 'ind_languages', 'ind_exact_study')), (0.88, ('ind_debateclub', 'ind_programming_exp', 'ind_languages', 'ind_university_grade')), (0.88, ('ind_debateclub', 'ind_languages', 'ind_exact_study', 'ind_university_grade')), (0.875, ('ind_programming_exp', 'ind_international_exp', 'ind_languages', 'ind_exact_study')), (0.875, ('ind_debateclub', 'ind_international_exp', 'ind_languages', 'ind_university_grade')), (0.87, ('ind_programming_exp', 'ind_languages', 'ind_exact_study', 'ind_university_grade')), (0.87, ('ind_international_exp', 'ind_languages', 'ind_exact_study', 'ind_university_grade')), (0.865, ('ind_entrepeneur_exp', 'ind_languages', 'ind_exact_study', 'ind_university_grade')), (0.86, ('ind_debateclub', 'ind_programming_exp', 'ind_languages', 'ind_degree')), (0.86, ('ind_debateclub', 'ind_entrepeneur_exp', 'ind_languages', 'ind_university_grade')), (0.855, ('ind_debateclub', 'ind_programming_exp', 'ind_entrepeneur_exp', 'ind_languages')), (0.85, ('ind_debateclub', 'ind_languages', 'ind_exact_study', 'ind_degree')), (0.85, ('ind_debateclub', 'ind_international_exp', 'ind_languages', 'ind_degree')), (0.845, ('ind_international_exp', 'ind_entrepeneur_exp', 'ind_languages', 'ind_degree')), (0.845, ('ind_debateclub', 'ind_international_exp', 'ind_entrepeneur_exp', 'ind_languages')), (0.845, ('ind_debateclub', 'ind_entrepeneur_exp', 'ind_languages', 'ind_degree')), (0.84, ('ind_debateclub', 'ind_programming_exp', 'ind_international_exp', 'ind_entrepeneur_exp')), (0.835, ('ind_debateclub', 'ind_entrepeneur_exp', 'ind_exact_study', 'ind_university_grade')), (0.83, ('ind_debateclub', 'ind_programming_exp', 'ind_entrepeneur_exp', 'ind_university_grade')), (0.83, ('ind_debateclub', 'ind_programming_exp', 'ind_entrepeneur_exp', 'ind_exact_study')), (0.83, ('ind_debateclub', 'ind_entrepeneur_exp', 'ind_degree', 'ind_university_grade')), (0.825, ('ind_debateclub', 'ind_programming_exp', 'ind_entrepeneur_exp', 'ind_degree')), (0.825, ('ind_debateclub', 'ind_international_exp', 'ind_entrepeneur_exp', 'ind_university_grade')), (0.825, ('ind_debateclub', 'ind_international_exp', 'ind_entrepeneur_exp', 'ind_exact_study')), (0.825, ('ind_debateclub', 'ind_entrepeneur_exp', 'ind_exact_study', 'ind_degree')), (0.82, ('ind_debateclub', 'ind_programming_exp', 'ind_international_exp', 'ind_university_grade')), (0.82, ('ind_debateclub', 'ind_programming_exp', 'ind_international_exp', 'ind_degree')), (0.82, ('ind_debateclub', 'ind_programming_exp', 'ind_exact_study', 'ind_university_grade')), (0.82, ('ind_debateclub', 'ind_programming_exp', 'ind_exact_study', 'ind_degree')), (0.82, ('ind_debateclub', 'ind_programming_exp', 'ind_degree', 'ind_university_grade')), (0.82, ('ind_debateclub', 'ind_international_exp', 'ind_degree', 'ind_university_grade')), (0.82, ('ind_debateclub', 'ind_exact_study', 'ind_degree', 'ind_university_grade')), (0.815, ('ind_debateclub', 'ind_programming_exp', 'ind_international_exp', 'ind_exact_study')), (0.81, ('ind_programming_exp', 'ind_international_exp', 'ind_entrepeneur_exp', 'ind_university_grade')), (0.81, ('ind_debateclub', 'ind_international_exp', 'ind_exact_study', 'ind_university_grade')), (0.81, ('ind_debateclub', 'ind_international_exp', 'ind_exact_study', 'ind_degree')), (0.81, ('ind_debateclub', 'ind_international_exp', 'ind_entrepeneur_exp', 'ind_degree')), (0.8, ('ind_programming_exp', 'ind_international_exp', 'ind_entrepeneur_exp', 'ind_degree')), (0.8, ('ind_programming_exp', 'ind_entrepeneur_exp', 'ind_exact_study', 'ind_university_grade')), (0.8, ('ind_international_exp', 'ind_entrepeneur_exp', 'ind_degree', 'ind_university_grade')), (0.795, ('ind_programming_exp', 'ind_entrepeneur_exp', 'ind_degree', 'ind_university_grade')), (0.79, ('ind_programming_exp', 'ind_entrepeneur_exp', 'ind_exact_study', 'ind_degree')), (0.79, ('ind_international_exp', 'ind_entrepeneur_exp', 'ind_exact_study', 'ind_degree')), (0.79, ('ind_entrepeneur_exp', 'ind_exact_study', 'ind_degree', 'ind_university_grade')), (0.785, ('ind_programming_exp', 'ind_international_exp', 'ind_entrepeneur_exp', 'ind_exact_study')), (0.78, ('ind_international_exp', 'ind_entrepeneur_exp', 'ind_exact_study', 'ind_university_grade')), (0.725, ('ind_international_exp', 'ind_exact_study', 'ind_degree', 'ind_university_grade')), (0.72, ('ind_programming_exp', 'ind_international_exp', 'ind_exact_study', 'ind_university_grade')), (0.72, ('ind_programming_exp', 'ind_international_exp', 'ind_degree', 'ind_university_grade')), (0.705, ('ind_programming_exp', 'ind_international_exp', 'ind_exact_study', 'ind_degree')), (0.675, ('ind_programming_exp', 'ind_exact_study', 'ind_degree', 'ind_university_grade'))]\n"]}]},{"cell_type":"markdown","source":["What we find is that programming experience, international experience, languages and university grade return the highest validation accuracy. \n","\n","So we will use this set of features to test our model with"],"metadata":{"id":"BKJbz7CU_Pxb"}},{"cell_type":"code","source":["m1_df = drop_features(set_results_RF[0][1])"],"metadata":{"id":"MA7o_s6i_kMH","executionInfo":{"status":"ok","timestamp":1666355439078,"user_tz":-120,"elapsed":16,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":180,"outputs":[]},{"cell_type":"markdown","source":["3. Data Preprocessing\n","\n","Now we preprocess the data from the chosen dataset"],"metadata":{"id":"8Ks4H8kQPFhJ"}},{"cell_type":"code","source":["m1_df = preprocess(m1_df)\n"],"metadata":{"id":"KEC2Y8ep0qaT","executionInfo":{"status":"ok","timestamp":1666355439078,"user_tz":-120,"elapsed":16,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":181,"outputs":[]},{"cell_type":"markdown","source":["**4**. Hyperparameter Tuning\n","\n","Before we run our RandomForestClassifier, we need to find the best hyperparameters\n","\n","**Now that this has been run, don't run it again (takes 20 mins)**"],"metadata":{"id":"OAqqNi7e0q4B"}},{"cell_type":"code","source":["# commented out so it won't retrain\n","#classifierRF(RandomForestClassifier(),m1_df)"],"metadata":{"id":"ZHpR1TbYBTJ-","executionInfo":{"status":"ok","timestamp":1666355741872,"user_tz":-120,"elapsed":393,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":219,"outputs":[]},{"cell_type":"markdown","source":["This gives us the following best hyperparameters\n","\n","Best score: 0.8757142857142857,\n","Best params: {'criterion': 'gini', 'max_features': 'sqrt', 'max_samples': 50, 'n_estimators': 200}\n","\n","Given these hyperparameters, we can now create a RandomForestClassifier with these params to fit our model\n"],"metadata":{"id":"jzvLgbH3CyJo"}},{"cell_type":"code","source":["def RandomForestBest():\n","  return RandomForestClassifier(\n","      n_estimators=200,\n","      bootstrap=True,\n","      criterion='gini',\n","      max_features='sqrt',\n","      max_samples=50\n","  )"],"metadata":{"id":"b7g8VqjBCxzO","executionInfo":{"status":"ok","timestamp":1666355439080,"user_tz":-120,"elapsed":17,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":183,"outputs":[]},{"cell_type":"markdown","source":["**5**. Classifier\n","Then we run our classifier to get our accuracy, confusion matrix and ROC curve"],"metadata":{"id":"vS8JMLd8FHxJ"}},{"cell_type":"code","source":["m1_results = classifier(RandomForestBest(), m1_df)\n","m1_2nd_results = classifier(RandomForestBest(), m1_df)\n","m1_csv = pd.merge(m1_results[1], init_df['Id'], left_index=True, right_index=True).to_csv('m1.csv')\n"],"metadata":{"id":"KCwSGeEg0s2R","colab":{"base_uri":"https://localhost:8080/","height":1000},"executionInfo":{"status":"ok","timestamp":1666355441352,"user_tz":-120,"elapsed":2289,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"974de342-f03d-4d31-f64c-b7231a412333"},"execution_count":184,"outputs":[{"output_type":"stream","name":"stdout","text":["-----------------------\n","Training results\n","Accuracy:\n","0.8757142857142857\n","Validation results\n","Accuracy:\n","0.8566666666666667\n","Confusion matrix:\n"]},{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.7/dist-packages/sklearn/utils/deprecation.py:87: FutureWarning: Function plot_confusion_matrix is deprecated; Function `plot_confusion_matrix` is deprecated in 1.0 and will be removed in 1.2. Use one of the class methods: ConfusionMatrixDisplay.from_predictions or ConfusionMatrixDisplay.from_estimator.\n","  warnings.warn(msg, category=FutureWarning)\n"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 720x432 with 2 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"stream","name":"stdout","text":["Random prediction: AUROC = 0.500\n","Model prediction: AUROC = 0.940\n"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 1080x720 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"stream","name":"stdout","text":["-----------------------\n","Training results\n","Accuracy:\n","0.8771428571428571\n","Validation results\n","Accuracy:\n","0.8866666666666667\n","Confusion matrix:\n"]},{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.7/dist-packages/sklearn/utils/deprecation.py:87: FutureWarning: Function plot_confusion_matrix is deprecated; Function `plot_confusion_matrix` is deprecated in 1.0 and will be removed in 1.2. Use one of the class methods: ConfusionMatrixDisplay.from_predictions or ConfusionMatrixDisplay.from_estimator.\n","  warnings.warn(msg, category=FutureWarning)\n"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 720x432 with 2 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"stream","name":"stdout","text":["Random prediction: AUROC = 0.500\n","Model prediction: AUROC = 0.943\n"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 1080x720 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["We can also acquire the full statistical results under m1_evalaution"],"metadata":{"id":"N-JvhK2edjwj"}},{"cell_type":"code","source":["m1_evaluation = evaluation(m1_results[0])"],"metadata":{"id":"3yg7IGCQdjWm","executionInfo":{"status":"ok","timestamp":1666355470085,"user_tz":-120,"elapsed":791,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":186,"outputs":[]},{"cell_type":"code","source":["print_eval(m1_evaluation)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"pHK2708xAOua","executionInfo":{"status":"ok","timestamp":1666355475921,"user_tz":-120,"elapsed":720,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"d9143308-c208-415c-e0cc-100c87177f36"},"execution_count":187,"outputs":[{"output_type":"stream","name":"stdout","text":["---------------------------\n","True Positive =   84\n","False Positive =  25\n","False Negative =  18\n","True Negative =  173\n","Actual Positive =  102\n","Actual Negative =  198\n","Precision = 0.7706422018348624\n","False Discovery Rate =  0.22935779816513763\n","False Omission Rate =  0.09424083769633508\n","Negative Predictive Value =  0.8235294117647058\n","Recall =  0.12626262626262627\n","False Positive Rate =  0.17647058823529413\n","True Negative Rate =  0.8737373737373737\n"]}]},{"cell_type":"markdown","source":["Okay, so we have a test accuracy of 0.8766666666666667. Now we move on to Model 2 to see what the differences are for this feature set on a different model. \n","\n","If we get wildly different results, then that may indicate that there is bias from the model to favour these features for prediction and that they may actually not be the best predictors."],"metadata":{"id":"k8bWoc2laIif"}},{"cell_type":"markdown","source":["## **Model 2: Choice Model - Choice Indicators**"],"metadata":{"id":"K1pShurh0V_e"}},{"cell_type":"markdown","source":["### 1. Create Dataset + Feature Selection\n","The second model will be the actual experimental model that will be compared to the other experimental model of the descriptors. This will allow us to properly test with only the features as variables. \n","\n","So first, let's get the features that we found from M1"],"metadata":{"id":"ZTefz9B60zol"}},{"cell_type":"code","source":["m2_df = drop_features(set_results_RF[0][1])"],"metadata":{"id":"dtMkOeck0zol","executionInfo":{"status":"ok","timestamp":1666355481139,"user_tz":-120,"elapsed":709,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":188,"outputs":[]},{"cell_type":"markdown","source":["2. Data Preprocessing\n","\n","We preprocess the data just like M1"],"metadata":{"id":"8sAJG-7R0zon"}},{"cell_type":"code","source":["m2_df = preprocess(m2_df)"],"metadata":{"id":"qrsxn77v0zon","executionInfo":{"status":"ok","timestamp":1666355481451,"user_tz":-120,"elapsed":2,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":189,"outputs":[]},{"cell_type":"markdown","source":["3. Hyperparameter tuning\n","\n","\n","Before we run the classifier, we need to tune the SVM hyperparamters by using grid search to find the best\n","\n","**Now that this has been run, don't run it again (takes 20 mins)**"],"metadata":{"id":"TvgwG7zG0zoo"}},{"cell_type":"code","source":["# commented out so it won't retrain\n","#classifierSVM(SVC(),m2_df)"],"metadata":{"id":"ee3Bt7xPQU-G","executionInfo":{"status":"ok","timestamp":1666355723711,"user_tz":-120,"elapsed":331,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":218,"outputs":[]},{"cell_type":"markdown","source":["This gives us the following best hyperparameters\n","Best params: {'C': 2.0, 'gamma': 'auto', 'kernel': 'rbf'}\n"],"metadata":{"id":"XkUcGpJWRd8_"}},{"cell_type":"markdown","source":["4. Classifier\n","\n","Given these hyperparameters, we can now create a SVM with these params to fit our model"],"metadata":{"id":"lyqXB1HHhyab"}},{"cell_type":"code","source":["def SVBest():\n","\n","  return SVC(\n","      C= 2.0,\n","      kernel='rbf',\n","      gamma='auto',\n","      probability=True\n","\n","  )"],"metadata":{"id":"fKcwoYvZRakG","executionInfo":{"status":"ok","timestamp":1666355502439,"user_tz":-120,"elapsed":275,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":191,"outputs":[]},{"cell_type":"code","source":["m2_results = classifier(SVBest(), m2_df)\n","m2_csv = pd.merge(m2_results[1], init_df['Id'], left_index=True, right_index=True).to_csv('m2.csv')\n"],"metadata":{"id":"D5SMDi7H0zoo","colab":{"base_uri":"https://localhost:8080/","height":1000},"executionInfo":{"status":"ok","timestamp":1666355505374,"user_tz":-120,"elapsed":898,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"0e8cac20-532d-4426-efda-2856eb7e3a0c"},"execution_count":192,"outputs":[{"output_type":"stream","name":"stdout","text":["-----------------------\n","Training results\n","Accuracy:\n","0.8871428571428571\n","Validation results\n","Accuracy:\n","0.8766666666666667\n","Confusion matrix:\n"]},{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.7/dist-packages/sklearn/utils/deprecation.py:87: FutureWarning: Function plot_confusion_matrix is deprecated; Function `plot_confusion_matrix` is deprecated in 1.0 and will be removed in 1.2. Use one of the class methods: ConfusionMatrixDisplay.from_predictions or ConfusionMatrixDisplay.from_estimator.\n","  warnings.warn(msg, category=FutureWarning)\n"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 720x432 with 2 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"stream","name":"stdout","text":["Random prediction: AUROC = 0.500\n","Model prediction: AUROC = 0.941\n"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 1080x720 with 1 Axes>"],"image/png":"iVBORw0KGgoAAAANSUhEUgAAA3gAAAJcCAYAAACrJAbaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd3wUdf7H8dekB1IgIUAoIYHQ0mihBESaFBVRUIqFokgRT9A7D9QT+52o90MQ8QRUkCIgIqgcoiJwAiKQYKiCgCAloSUBEkJCyvz+mLAkbOiETXk/H4997Ox8ZmY/u0lgP/tthmmaiIiIiIiISMnn5OgERERERERE5OZQgSciIiIiIlJKqMATEREREREpJVTgiYiIiIiIlBIq8EREREREREoJFXgiIiIiIiKlhAo8ERGRImQYxn7DMO5wdB4iIlI2qMATEZFSI6+YOmsYRpphGEcMw5hhGIbXRce0NgxjhWEYqYZhnDIM4xvDMMIuOsbHMIwJhmEcyLvW3rzHlS7xvKZhGGfyjj1sGMZ4wzCcrzH39oZhHLr2Vy0iInKBCjwRESlt7jFN0wtoDDQBnj8fMAwjBvge+AqoBoQAm4G1hmHUzjvGDfgRCAe6AT5ADJAEtLjM8zbKe95OwEPAkJv7skRERK5MBZ6IiJRKpmkeAb7DKvTOexuYaZrmRNM0U03TTDZN80XgF+CVvGMGAEFAT9M0d5immWua5jHTNF83TXPpVTzvTmA1EHFxzDAM97yWwIS824S8feWBb4Fqea2AaYZhVLuR1y8iImWTCjwRESmVDMOoAdwJ7Ml7XA5oDSwo5PDPgc5523cAy0zTTLvO5w0D2gK/FhL+B9AKq+hshNUi+KJpmmfyck0wTdMr75ZwPc8vIiJlmwo8EREpbRYbhpEKHASOAS/n7ffD+n8vsZBzEoHz4+v8L3HMlWwyDCMF+Ab4CJheyDEPA6/ltQgeB14F+l/Hc4mIiBRKBZ6IiJQ295mm6Q20BxpwoXBLAXKBwELOCQRO5G0nXeKYK2lqmmZF0zTrmKb5ommauYUcUw34M9/jP/P2iYiI3BQq8EREpFQyTfN/wAzg33mPzwDrgN6FHN4Ha2IVgOVA17xxcTdbAlAr3+OgvH0AZhE8n4iIlDEq8EREpDSbAHQ2DKNR3uPngIGGYYw0DMPbMIyKhmG8gTVL5qt5x8zC6t650DCMBoZhOBmG4W8YxguGYdx1g/nMBV40DCMgb8mFl4DZebGjgL9hGL43+BwiIlKGqcATEZFSK2+c20ysQgrTNNcAXYFeWOPs/sRaSuE20zR35x2TiTXRyk7gB+A0sAGrq+f6G0zpDSAW2AJsBTbl7Ts/++Zc4A/DME5qFk0REbkehmmqR4iIiIiIiEhpoBY8ERERERGRUkIFnoiIiIiISCmhAk9ERERERKSUUIEnIiIiIiJSSrg4OoFrValSJTM4ONjRaYiIiIiIiDhEXFzcCdM0AwqLlbgCLzg4mNjYWEenISIiIiIi4hCGYfx5qZi6aIqIiIiIiJQSKvBERERERERKCRV4IiIiIiIipUSJG4NXmKysLA4dOkRGRoajUxEp1Tw8PKhRowaurq6OTkVEREREClEqCrxDhw7h7e1NcHAwhmE4Oh2RUsk0TZKSkjh06BAhISGOTkdEREREClEqumhmZGTg7++v4k6kCBmGgb+/v1rKRURERIqxUlHgASruRG4B/Z2JiIiIFG+lpsATEREREREp61Tg3STOzs40btyYiIgI7rnnHk6ePHlTrjtjxgz+8pe/3JRrFSUvLy8AEhISeOCBBy577IQJE0hPT7c9vuuuu27a+zVhwgRmzpxpe5ydnU1AQADPPfdcgeOCg4M5ceKE7fGqVavo3r07YL3nAQEBNG7cmAYNGvDuu+8WOHfq1Kk0aNCABg0a0KJFC9asWWOLZWVl8dxzz1G3bl2aNm1KTEwM33777Q2/rjfffJPQ0FDq16/Pd999V+gxgwYNIiQkhMaNG9O4cWPi4+MBa+zcyJEjCQ0NJSoqik2bNtnO+fTTT6lbty5169bl008/te2/4447SElJueG8RUREROTWUoF3k3h6ehIfH8+2bdvw8/Nj8uTJjk7phmVnZ1/zOdWqVeOLL7647DEXF3hLly6lQoUK1/xcF8vOzuaTTz7hoYcesu374YcfqFevHgsWLMA0zau+Vt++fYmPj2ft2rX885//5ODBgwAsWbKEKVOmsGbNGnbu3MmHH37IQw89xJEjRwAYO3YsiYmJbNu2jU2bNrF48WJSU1Nv6HXt2LGDefPmsX37dpYtW8aIESPIyckp9Nh33nmH+Ph44uPjady4MQDffvstu3fvZvfu3UydOpUnnngCgOTkZF599VXWr1/Phg0bePXVV21FXf/+/fnggw9uKG8RERERufVU4BWBmJgYDh8+DMCGDRuIiYmhSZMmtG7dml27dgFWK1GvXr3o1q0bdevWZfTo0bbzp0+fTr169WjRogVr16617d+/fz8dO3YkKiqKTp06ceDAAcBquXniiSdo1aoVtWvXZtWqVTz22GM0bNiQQYMGFZpjcHAwo0ePJjIykhYtWrBnzx7btYYPH07Lli0ZPXo0e/fupVu3bjRr1oy2bduyc+dOAPbt20dMTAyRkZG8+OKLBXKMiIgAICcnh2effZaIiAiioqKYNGkS7733HgkJCXTo0IEOHTrYcjnfmjZ+/HgiIiKIiIhgwoQJtms2bNiQIUOGEB4eTpcuXTh79qzda1qxYgVNmzbFxeXC5LBz585l1KhRBAUFsW7duqv9Edr4+/sTGhpKYmIiAG+99RbvvPMOlSpVAqBp06YMHDiQyZMnk56ezrRp05g0aRLu7u4AVKlShT59+lzz8+b31Vdf0a9fP9zd3QkJCSE0NJQNGzZc0/kDBgzAMAxatWrFyZMnSUxM5LvvvqNz5874+flRsWJFOnfuzLJlywDo0aMHc+fOvaG8RUREROTWKxXLJFys7xT7D/LdowLpHxPM2XM5DJpu/+H4gWY16B1dk+Qz53hidlyB2PxhMVf93Dk5Ofz4448MHjwYgAYNGrB69WpcXFxYvnw5L7zwAgsXLgQgPj6eX3/9FXd3d+rXr89TTz2Fi4sLL7/8MnFxcfj6+tKhQweaNGkCwFNPPcXAgQMZOHAgn3zyCSNHjmTx4sUApKSksG7dOr7++mt69OjB2rVr+eijj2jevHmB1pz8fH192bp1KzNnzuTpp59myZIlgLXsxM8//4yzszOdOnXiww8/pG7duqxfv54RI0awYsUKRo0axRNPPMGAAQMu2Vo5depU9u/fT3x8PC4uLiQnJ+Pn58f48eNZuXKlrUg6Ly4ujunTp7N+/XpM06Rly5a0a9eOihUrsnv3bubOncu0adPo06cPCxcu5JFHHilw/tq1a2nWrJntcUZGBsuXL2fKlCmcPHmSuXPn0rp166v+WQIcOHCAjIwMoqKiANi+fXuB5wCIjo7m008/Zc+ePQQFBeHj43PF6z7zzDOsXLnSbn+/fv3supMePnyYVq1a2R7XqFHD9gXCxf7xj3/w2muv0alTJ8aNG4e7uzuHDx+mZs2adudfaj9AxYoVyczMJCkpCX9//yu+HhEREREpHtSCd5OcPXuWxo0bU7VqVY4ePUrnzp0BOHXqFL179yYiIoJnnnmG7du3287p1KkTvr6+eHh4EBYWxp9//sn69etp3749AQEBuLm50bdvX9vx69ats3U/7N+/f4GxX/fccw+GYRAZGUmVKlWIjIzEycmJ8PBw9u/fX2jODz74oO0+f+tW7969cXZ2Ji0tjZ9//pnevXvTuHFjhg0bZmvJWrt2re38/v37F3r95cuXM2zYMFuLmp+f32XfwzVr1tCzZ0/Kly+Pl5cXvXr1YvXq1QC2sWUAzZo1K/Q1JSYmEhAQYHu8ZMkSOnTogKenJ/fffz+LFy+2dW0sbDbI/Pvmz59PVFQUoaGhjBgxAg8Pj8vmfq3effddW1fK/LeLi7tr8eabb7Jz5042btxIcnIyb7311g3lWLlyZRISEm7oGiIiIiJya5XKFrzLtbh5ujlfNu5X3u2aWuxs180bg5eenk7Xrl2ZPHkyI0eOZOzYsXTo0IFFixaxf/9+2rdvbzvnfDc+sCZpuZ4xbxdfy8nJqcB1nZycLnnd/AVN/u3y5csDkJubS4UKFWyTdVzu/KJ28XtVWBdNT0/PAmu0zZ07lzVr1hAcHAxAUlISK1asoHPnzvj7+5OSkmJrRUxOTi7Qoti3b1/ef/99YmNj6dKlCz169KBq1aqEhYURFxdHx44dbcfGxcURHh5OaGgoBw4c4PTp01dsxbuWFrzq1avbxgCC1cJavXp1u3MDAwNt79Wjjz7Kv//978ueX716dVatWlVgf/7fz4yMDDw9PS/7OkRERESkeFEL3k1Wrlw53nvvPf7v//6P7OxsTp06ZfswPmPGjCue37JlS/73v/+RlJREVlYWCxYssMVat27NvHnzAJgzZw5t27a9oVznz59vu4+JsS9qfXx8CAkJseVgmiabN28GoE2bNgVyKUznzp2ZMmWKrcBMTk4GwNvbu9CJR9q2bcvixYtJT0/nzJkzLFq06JpeY8OGDW1jCU+fPs3q1as5cOAA+/fvZ//+/UyePNk2rqx9+/bMmjULsLrVzp492zYmML/o6Gj69+/PxIkTARg9ejRjxowhKSkJsLrZzpgxgxEjRlCuXDkGDx7MqFGjOHfuHADHjx8v8DM871pa8Hr06MG8efPIzMxk37597N69mxYtWtgdd7511TRNFi9ebBsL2aNHD2bOnIlpmvzyyy/4+voSGBhI165d+f7770lJSSElJYXvv/+erl272q5x5MgRW3EsIiIiIiWDCrwi0KRJE6Kiopg7dy6jR4/m+eefp0mTJlfVQhcYGMgrr7xCTEwMbdq0oWHDhrbYpEmTmD59OlFRUcyaNctWdFyvlJQUoqKimDhxot1SAOfNmTOHjz/+mEaNGhEeHs5XX30FwMSJE5k8eTKRkZGXHA/2+OOPExQURFRUFI0aNeKzzz4DYOjQoXTr1s2uoGratCmDBg2iRYsWtGzZkscff9w2/vBq3Hnnnfz0008ALFq0iI4dOxZo+bv33nv55ptvyMzMZOzYsezZs4dGjRrRpEkTQkND7cb0nTdmzBimT59OamoqPXr04LHHHqN169Y0aNCAIUOGMHv2bFvr2RtvvEFAQABhYWFERETQvXv3qxqTdznh4eH06dOHsLAwunXrxuTJk3F2dgasJSbOd6N8+OGHiYyMJDIykhMnTtgmv7nrrruoXbs2oaGhDBkyxDY7pp+fH2PHjqV58+Y0b96cl156ydaNNi4ujlatWhWYsEZEREREij/jWqaOLw6io6PN2NjYAvt+++23AoWQXFlwcDCxsbF2E52UdD179uTtt9+mbt26jk6lRBs1ahQ9evSgU6dOdjH9vYmIiIg4lmEYcaZpRhcWUwuelCrjxo2zdVWU6xcREVFocSciIiIixZv6X5VRl5pZs6SrX78+9evXd3QaJd6QIUMcnYKIiIiIXAe14ImIiIiIiJQSRVbgGYbxiWEYxwzD2HaJuGEYxnuGYewxDGOLYRhNiyoXERERERGRsqAoW/BmAN0uE78TqJt3Gwr8pwhzERERERERKfWKbAyeaZo/GYYRfJlD7gVmmtY0nr8YhlHBMIxA0zQ1Q4aIiIhIUUj4Fb76C1SJgF5TrH2zesGpgwWPC74NuuctofRJN0hPKhiv2wW6/tPannI7ZJ0tGA/vCR1esLbfb26fR+OH4ban4dwZmNrePt58CLQcCmnHYcZd9vE2o6DJI5CyH+b0to+3fx4iesHRHbBgoH28yxtQryscioPFw+3jd4+HkLaw7yf479/s4/d9CDWawa5l8MNY+3ifmVC5IWxbCKvG2ccf/gIq1oJNs+Dn9+zjj34L5SvB+qmwcZp9fOgqcCsPa96F+M/s43/ZaN2v+CfsWFww5uoJw6xlpfjuH7D7+4LxcpXgsW+t7W+ehj/XFoxXCIJHFlrbXw6DhE0F4wENoK+11jDzH4HjuwrGqzUtEb97Z5ITyZzWjQrPrMfJzcP+uGLMkZOsVAfy/0QP5e2zK/AMwxiK1cpHUFDQLUnuWjk7OxMZGUl2djYhISHMmjWLChUq3PB1Z8yYQWxsLO+///5NyLLoeHl5kZaWRkJCAiNHjuSLL7645LETJkxg6NChlCtXDrDWafvss89uyvs1YcIE/Pz8GDBgAADZ2dkEBgYyePBgxo278A/sxctErFq1in//+98sWbKEGTNm8Pe//53q1auTkZHBsGHDeOaZZ2znTp06lfHjxwPWYvDjx4/ntttuAyArK4uxY8eycOFCvL29cXd356WXXuLOO++8odf15ptv8vHHH+Ps7Mx7771nW5A8vxUrVvDss89y7tw5mjVrxscff1xgHbuNGzcSExPDvHnzeOCBBwDo1q0bv/zyC7fddhtLliyxHduvXz9ef/11LTchIlLaJMTD0W0QeseFfZXqgcdFa7ZWyPd5K6A+ZJwqGPetmS/eEHIyC8a9Ay9sVwm3z8O7qnVvOBUe9wqw7p2cC4+Xy1vmydm98LhnReve1aPwuIevde9WrvC4u1fevXfhcbdyF65TWNwlryDw9Cs87uxm3ZevVHjcyVrvFq+AwuPG+XjVwuPn+VSzj7vkK1Z8a9jHPfJ9HqtYCzJOFox7Vb2w7Rdi/7OvGJwvXgecLio3/EIubBfj3z0Pd3c2m0F4J54mslbJKvAwTbPIbkAwsO0SsSXAbfke/whEX+mazZo1My+2Y8cOu323Wvny5W3bAwYMMN94442bct3p06ebTz755E251rXKysq66mPzv/4rqVWrlnn8+PHrSemysrKyzMjIyAJ5L1261GzdurVZu3ZtMzc395I5rFy50rz77rtN0yz4np84ccL09/c3Dxw4YJqmaX7zzTdm06ZNbefGxcWZNWvWNBMTE03TNM0xY8aYAwYMMDMyMkzTNM0jR46Y8+fPv6HXtX37djMqKsrMyMgw//jjD7N27dpmdnZ2gWNycnLMGjVqmLt27TJN0zTHjh1rfvTRR7Z4dna22aFDB/POO+80FyxYYNu/fPly8+uvv7a99vNWrVplPv7444XmUxz+3kRE5Dpt/MQ0X/YxzVMJjs5EpNhJOJluPj3vV/N4qvU5Ljsn9wpnOA4Qa16iXnLkLJqHgXwlODXy9pV4MTExHD5svZQNGzYQExNDkyZNaN26Nbt2Wc3UM2bMoFevXnTr1o26desyevRo2/nTp0+nXr16tGjRgrVrLzSL79+/n44dOxIVFUWnTp04cOAAAIMGDeKJJ56gVatW1K5dm1WrVvHYY4/RsGFDBg0aVGiOwcHBjB49msjISFq0aMGePXts1xo+fDgtW7Zk9OjR7N27l27dutGsWTPatm3Lzp07Adi3bx8xMTFERkby4osvFsgxIiICgJycHJ599lkiIiKIiopi0qRJvPfeeyQkJNChQwc6dOhgy+XEiRMAjB8/noiICCIiIpgwYYLtmg0bNmTIkCGEh4fTpUsXzp69qDkeqwWradOmBVqt5s6dy6hRowgKCmLdunVX+yO08ff3JzQ01La23ltvvcU777xja/lr2rQpAwcOZPLkyaSnpzNt2jQmTZqEu7s7AFWqVKFPnz7X/Lz5ffXVV/Tr1w93d3dCQkIIDQ1lw4YNBY5JSkrCzc2NevXqAdC5c2cWLlxoi0+aNIn777+fypUrFzivU6dOeHt72z1n27ZtWb58OdnZ2TeUu4iIiEhxl5tr8tn6A3QZ/xPfbktkyyGr1dLZyXBwZtfHkV00vwb+YhjGPKAlcMq8WePvpt9tvy/8PmgxBM6lF95Xu/FD0ORhOJMEnw8oGHv0v1f91Dk5Ofz4448MHjwYgAYNGrB69WpcXFxYvnw5L7zwgu2Dd3x8PL/++ivu7u7Ur1+fp556ChcXF15++WXi4uLw9fWlQ4cONGnSBICnnnqKgQMHMnDgQD755BNGjhzJ4sVWv+qUlBTWrVvH119/TY8ePVi7di0fffQRzZs3Jz4+nsaNG9vl6uvry9atW5k5cyZPP/20rYveoUOH+Pnnn3F2dqZTp058+OGH1K1bl/Xr1zNixAhWrFjBqFGjeOKJJxgwYACTJ08u9L2YOnUq+/fvJz4+HhcXF5KTk/Hz82P8+PGsXLnSViSdFxcXx/Tp01m/fj2madKyZUvatWtHxYoV2b17N3PnzmXatGn06dOHhQsX8sgjjxQ4f+3atTRr1sz2OCMjg+XLlzNlyhROnjzJ3Llzad269VX/LAEOHDhARkYGUVFRAGzfvr3AcwBER0fz6aefsmfPHoKCgvDx8SnsUgU888wzrFy50m5/v379eO655wrsO3z4MK1atbI9rlGjhu0LhPMqVapEdnY2sbGxREdH88UXX3Dw4EHb+YsWLWLlypVs3Ljxql63k5MToaGhbN682e71ioiIiJQW+0+c4bkvt/DLH8nE1PZn3P2R1PIv7+i0bkiRFXiGYcwF2gOVDMM4BLwMuAKYpvkhsBS4C9gDpAOPFlUut8LZs2dp3Lgxhw8fpmHDhnTu3BmAU6dOMXDgQHbv3o1hGGRlZdnO6dSpE76+Vh/wsLAw/vzzT06cOEH79u0JCLD6//bt25fff/8dgHXr1vHll18C0L9//wKtfvfccw+GYRAZGUmVKlWIjIwEIDw8nP379xda4D344IO2+/xjzHr37o2zszNpaWn8/PPP9O59oSDOzLT6Oa9du9ZWqPbv358xY8bYXX/58uUMHz7c1qLm5+d32fdwzZo19OzZk/LlrT+qXr16sXr1anr06EFISIjtNTRr1qzQhdoTExNp2LCh7fGSJUvo0KEDnp6e3H///bz++utMmDABZ2dnDMP+G5n8++bPn89PP/3Ezp07ef/99/HwuLl9r999992bej3DMJg3bx7PPPMMmZmZdOnSBWdnq3/+008/zVtvvYWT07U12FeuXJmEhAQVeCJy86Qdt8Y2uXrCls/hyyH2xwz9H1RrDBs/hv/+1T7+1CbwrwNrJ8IPL9nH//Y7eFexJpf46W37+AsJ1uQUy16AXy7+gtKAV/LGG339FGyaWTDs5g0vHLK2FzwK278sGPcOhL9ZPV2Y09t+8gr/uvBUrLX9yZ1w4OeC8WpNrMkzAD5sC0e2FIwHt4VBeeOl32sCyX8UjNe7Ex6aZ22/UxfOHCsYj+wN9062JqFwv/KXkSJlwYTlv7P98GnG9Yqkb/OahX5GLGmKchbNB68QN4Eni+TJL9fi5lbu8vHy/tfUYneep6cn8fHxpKen07VrVyZPnszIkSMZO3YsHTp0YNGiRezfv5/27dvbzjnfjQ+sSVpupDvc+Ws5OTkVuK6Tk9Mlr5v/Fzj/9vkCKzc3lwoVKhAfH3/F84vaxe9VYV00PT09ycjIsD2eO3cua9asITg4GLC6Ma5YsYLOnTvj7+9PSkqKrRUxOTm5QIti3759ef/994mNjaVLly706NGDqlWrEhYWRlxcHB07drQdGxcXR3h4OKGhoRw4cIDTp09fsRXvWlrwqlevbmuNA6uFtXr16nbnxsTEsHr1agC+//572xcDsbGx9OvXD4ATJ06wdOlSXFxcuO+++y6bY0ZGBp6enpc9RkTkqpgm/DoLvh8LzQdDp5egchi0s/9yEK8q1n21xoXHz0+eUbNl4fHzk18E3waF/T/l5Grd1+l4YSINm3zH1+tWcLIGuDAxBkBYD6h00URUbvmuF/GAVbDlV87/wnbjh6xZGvPzzjd5RbNBkHa0YLxCrQvbLYbC2ZSCcf98+bT+izVLYH6Vw8DF3bqJlGG7jqTi5uJESKXyvNg9jOfubEhV3xI2kcplOLKLZqlUrlw53nvvPe677z5GjBjBqVOnbB/GZ8yYccXzW7ZsyahRo0hKSsLHx4cFCxbQqFEjAFq3bs28efPo378/c+bMoW3btle42uXNnz+f5557jvnz5xMTE2MX9/HxISQkhAULFtC7d29M02TLli00atSINm3aMG/ePB555BHmzJlT6PU7d+7MlClT6NChQ4Eumt7e3qSmptp10Wzbti2DBg3iueeewzRNFi1axKxZs6769TRs2NA2lvD06dOsXr2agwcP2orD6dOnM3fuXDp37kz79u2ZNWsWr732Gjk5OcyePbvQgic6Opr+/fszceJE3nzzTUaPHs2YMWNYtmwZ/v7+xMfHM2PGDNavX0+5cuUYPHgwo0aNYsqUKbi5uXH8+HFWrVpVoBUUrq0Fr0ePHjz00EP89a9/JSEhgd27d9OiRQu7444dO0blypXJzMzkrbfe4h//+AdgjZc8b9CgQXTv3v2KxR3A77//bhtPKSJy3U7szptqfQ0EtYaovtb+qhHW7VKqN7NulxLUyrpdSu121u1S6t5h3S6lwd3W7VLCe1q3S2nU99IxgKb9Lx9vPvjy8VZPXD7eZtTl4yJl0LnsXCav3MMHq/bQrl5lPhoYTSWv0veFhyMnWSm1mjRpQlRUFHPnzmX06NE8//zzNGnS5Kpa6AIDA3nllVeIiYmhTZs2BbocTpo0ienTpxMVFcWsWbOYOHHiDeWZkpJCVFQUEydOvGTBMWfOHD7++GMaNWpEeHg4X331FQATJ05k8uTJREZG2o0HO+/xxx8nKCiIqKgoGjVqxGefWeu0DB06lG7dutkmWTmvadOmDBo0iBYtWtCyZUsef/xx2/jDq3HnnXfy00/Wui6LFi2iY8eOBVr+7r33Xr755hsyMzMZO3Yse/bsoVGjRjRp0oTQ0FC7MX3njRkzhunTp5OamkqPHj147LHHaN26NQ0aNGDIkCHMnj2bwEDrW9433niDgIAAwsLCiIiIoHv37lc1Ju9ywsPD6dOnD2FhYXTr1o3Jkyfbul/eddddJCQkAPDOO+/QsGFDoqKiuOeeewq0Ml5K27Zt6d27Nz/++CM1atTgu+++A+Do0aN4enpStWrVK1xBROQy4ufCf1rD0a1wz3sw6L/WtOciIrdY/MGTdJ+0mok/7ubuyEDefiDK0SkVGcPqKVlyREdHm7GxsQX2/fbbbwUKIWFwhsAAACAASURBVLmyi9eBKy169uzJ22+/rfXbbtC7776Lj4+PbbKg/PT3JrdEbg6cS7PWmMo6C6cO2R/jVcVaP+ncGTidYB/3DrS64GWmQuoR+7hPdas7X8Zp+65wYK215OoBZ0/CmeP28Qq1wMUN0pPtF+IFqBgCzi7W5F1nk+3jfnXAyckal3bxOlNwoftf6lHIPF0wZjhZ49AATida71V+Ts7gV9vaPnUYstILxp1dL6xVdfIgZGcUjLu4X1iLKuVPyDlXMO7qaa2fBZC8D3Iv+gKzfAB4VoBjv1kLMXd5A7wKzuIrInKrLN9xlKGzYqni48E/e0bQsUEVR6d0wwzDiDNNM7qwmLpoSqkybtw4EhMTVeDdoAoVKtC//xW6D4kUpRWvW+OZbn8Wju2AaYW0SN//MUQ+AIdiYWYP+/hDn0O9rrDvJ5j3kH180FIIbgO/Lyt8so9hP0FgI9i28PKTffw66/KTfaz/8PKTfawZD798cFEw32QfK9+wn+zD3Qeezxub+90Ll5/sY8nTl5/s48uhl5/sY/7DcGRrwXj+yT5m97Kf7CPiAXjgY6jcEHpNtX/tIiK3QFpmNl7uLrQJrcSI9qEMa1cbbw9XR6dV5FTglVGFzUJZGtSvX5/69dX950Y9+miJntRWSoP0JIj71CrwKoZYxdzFauaNRa3csPB4VWs2YQIbFx6vVC/vOi0Lj/vmLdVau33h8fMtUvW6Wa2BF/PI654ddm/h3RKd87qQR/W9/FizpoMg5KKxZE75/vtuOcx+rJhLvskC2oy6MO7tvPyTgbQbbd8C6VnhwnanV+xbGMvn6/3R9V/2k3k4u0JurtVCKSJyi506m8WbS39j/b5klo5si6ebM892LTufD0tNF80GDRqUimlNRYoz0zTZuXOnumhK0fv6Kdj9w4VWKBERkavww46jvLh4K8dTMxlye22euaMeHq7Ojk7rpiv1XTQ9PDxISkrC399fRZ5IETFNk6SkpJu+JqCIiIjIjTqTmc2YhVtYsiWRBlW9mTYgmqgaFa58YilUKgq8GjVqcOjQIY4fL2QQvIhcmmlakyPk5gD5WvOd3axJGnJzCkyu4GFkUcMlGdJ9oZyfNfFF4hb76wa3sSbHSPkTjm63j9duZ409StoLx3fZx+t0tCa3OP47JO2xj9ftYk1ecXQHpOy3jze4y7pP3GI/OYeTC9TrYm0f3mQ/+YaLO4R2srYPbrSfXMOt/IWp1/9cZ78OlYev9foB9q22JvjIr5w/BLW0tveutCYQyc+rCtTI6663ZzlkXzS5hU81a30wgN+/y/vZ5VOhptU10TRh17fY8QuxujTmZFktZBerVNe6ZZ218rtY5QbW5B2Zqdbru1jVCGtyjrMp1vtzsWqNrddw5gQc3GAfrxFtdX08c8L+tYmIiFyCh6szJ9Iy+VvnegxvXwdX57LbRbxUFHiurq6EhIQ4Og2R4ic3B04dhJMHrGLr5AHr1mwQ1IqBPT9aEyRc7KEFVhH02xJY8LB9/NFvoVZr2L8GFg2zjw9bDYFRsOcH+O/f7OMjf7WKhN++geUv28ef3W0VeFs/h5/esY+/kGgVeJtmwvr/FIwZTvByXtG1YQr8Ortg3N0Xnj9gbf/8HmxfVDDuUx3+usPa/t9b1mvIr1I9+MtGa/vHV+HARUVMtaYwNK8wWva8NT18fiG3w8BvrO0lT9sXqPXvhgetJUVYNNy+wIzqe2HSis8HQvZFBWL0Y9D9XTBzYd6D2Gk9Erq8bo2ZKize4R8XxmQVFu/6JsSMsGatLCzeYxI0HQBJfxQef+ATiLgfjm4rPH7+d69OR/uJO0RERPJJOHmWt5btZGz3MCp5ufPZ461wclJvvlIxBk+kzMrNhdOH7Au4BndDw+5W69jkfIuiG05WAdPldWuB3jMnrFagCjWtWfnO8wuxWqLOniy8hcw/1Jp+Pj3Zer6LVapnTT9/5kTh09tXbmi1lKUehdRE+3iVcGuShtOJhU9fXzXKmrzh1CHrOS52voXr5AErx/ycnC9MvpG8DzJOFYw7u1rPD1YL48UtcC4eVisWWAs4Xzy5hGs5CMibvOP4LvsWOnfvC9PbH91hP/28h8+F6e2PbLOfft6zIlSsZW0nbrZa6vIr52/9PE3Til/Mq7LVgpaTbRVZF/Ouat2yz1mzV17Mpzp4BVivq7DWV9+aUN7fel9O7LaPV6xlvYaM04UXcPl/98xcq6VYREQkn9xck882HGDctzvJyTX54JGmdKhftpZiudwYPBV4IsXdqUIKuJrNrVa4zFR4s0a+gw3rw3ubp6HlUMjKgK0LrC5zFYKsdaucS//0wCIiIlI67TtxhjELt7BhXzK3hVbizV6R1PQr5+i0brlSP8mKSImWesRqJTt5AE7+aRVyfiHQNq9r439aF2xl8g4En0Br290b7vuPta9CkNV64uJ24VhXD2iq9exERESkdJi4/Hd+SzzN2/dH0Tu6hiZYLIRa8ESK2vbFF7o55mRZY+LcvKDbv6x9H8QU7ApXvjLU72aNZQJrjJi7D1SoZbXAuWoWSxERESk7diScxsPVidoBXiSlZZKda1LFp2x/HlILnogjxc+B3d9feFw+AKrn+3u841UwjAsFnNtF3QzCe96aPEVERESKkczsHN5fsYf/rNpLhwaVmTYgGn8vd0enVeypwBMpan1mWZNFgDXBh8tF/zCdn7JfRERERACI+zOFMQu3sOdYGr2aVmfs3WGOTqnEKLsLRIgUtZQ/4TV/2P6l1SrnVs6+uBMRERGRAn7YcZQHPvyZs+dymPFoc8b3aUzF8m5XPlEAteCJFK3cbPtp7EVERETETlpmNl7uLtwWWomnOoQytF0dvNxVrlwrteCJiIiIiIjDnErPYvQXm+n+3mrOnsvB082Zv3apr+LuOuldExERERERh1i27Qhjv9pG8plzDL29Nlr14MapwBMRERERkVsqLTObMV9s4b9bEwkL9GH6oOZEVPd1dFqlggo8kaLi7g0thkFAfUdnIiIiIlKseLo6k3zmHH/vWp+ht9fG1Vkjx24WLXQuIiIiIiJF7lBKOuO+3cnL94QT4O1Obq6Jk5P6ZF4PLXQucrUSN8MPL0HWWetxtaZw5zhr+/MBkHqk4PG1WsMdr1jbc3pDxql8QQO6vA7Vm1nr34mIiIiUQbm5JrPX/8lb3+7EBB5oVoP29SuruCsiKvBEzks7DnMfgpxMqBJu7XPJt+aKiwe4ehY8x9n9ovi5gvGf/g3d3gT/OkWTs4iIiEgxtvd4Gs8t3MLG/SncXi+Af/WMoEbFco5Oq1RTgSdy3oldVnH38BdQrbF9vNfUy5/fd1bR5CUiIiJSQr2/Yg+/H03j370bcX/T6hiaJrPIaQyeSH7n0sFN3yqJiIiIXK/tCafwcHWmToAXSWmZ5Jgmlb09HJ1WqXK5MXiarkZk6xewfgqYpoo7ERERkeuUkZXDO9/tpMf7axn37U4A/L3cVdzdYuqiKWXbmRPw9VMQ2AiiB4Oz/iRERERErlXs/mRGL9zCH8fP0LtZDV68O8zRKZVZ+jQrZVtqImSlQ6sRKu5ERERErsP3248wbHYc1Xw9mflYC26vF+DolMo0faIVAdCAXxEREZFrcjojCx8PV9rWDWBUp7oMaVub8u4qLxxNY/BEREREROSqnUw/x18/j+eeSWs4ey4HTzdnnr6jnoq7YkI/BSnbXDygSgS4ezs6ExEREZFib+nWRF76ahsn07MY3q4OTmouKnZU4EnZVqkuPLHW0VmIiIiIFGtpmdk8+/lmlm0/QkR1Hz59rAXh1XwdnZYUQgWeiIiIiIhcVjlXZ9Iys3nuzgY8flsILs5quiuu9JORsi1pL0zrBPtWOzoTERERkWLlYHI6T87ZxPHUTJycDGYNbsHwdnVU3BVzasGTsi0rHQ7HQsZJR2ciIiIiUizk5JrMXLefd77bhQHsSKxJO+8ADM06XiKowJPizzThyFY4l2Y9DoqxljVI2gtpRwseazhBUCtr+8RuOHO8YNzJFWo2t7aP7YTEzUWbu4iIiEgJsudYKmMWbiXuzxTa1QvgX70iqV7B09FpyTVQgSfF35GtMKXthccvJYPhDOsmQ+zHBY918YQXj1jbP70DW+YXjJerBKP3Wts/vga7/mttu5UvmtxFRERESpDJK/ey93ga4/s0omeT6mq1K4EM0zQdncM1iY6ONmNjYx2dhtxK+9fCjLug8+sQGAUh7awWvOO/Q2pCwWMNJwi53do+9pt9C5+TKwS3sbaPbIX0JKsorNEczfMrIiIiZdHWQ6fwdHMmtLIXyWfOkZNrEuDt7ui05DIMw4gzTTO6sJha8KTkCGwEtdtdeBxQz7pdSuWG1u1SqkbevNxERERESpiMrBzeXf470376gzsaVmHqgGj8yrs5Oi25QSrwpPjzCYRWI8C3hqMzERERESkV1v+RxHNfbmXfiTP0a16T5++6zJfiUqKowJPiz682dHvT0VmIiIiIlArfbT/CsFlx1PTzZM7jLWkTWsnRKclNpAJPir+zJ8HJBdy9HJ2JiIiISIl16mwWvp6utKsXwN8612Nw2xDKuakcKG00q4QUfxumwbggOHfG0ZmIiIiIlDjJZ87xzPx47pm0hvRz2Xi4OvNUp7oq7kop/VSl+EuMB78QLWUgIiIicg1M0+S/WxN5+avtnDqbxYgOoTg7admD0k4FnhR/CfEQ1NLRWYiIiIiUGGmZ2TwzP54fdhwlqoYvsx9vScNAH0enJbeACjwp3s6cgNOHIHCYozMRERERKTHKuTqTkZXDC3c14LE2Ibg4a2RWWaGftBRvifHWfbXGjs1DREREpJg7kJTO8FlxHEvNwMnJYOZjLRh6ex0Vd2WMWvCkeKtUH7q9BYEq8EREREQKk5NrMn3tPv79/S5cnJzYmZhKZW8PDEPj7coiFXhSvFWoCa2GOzoLERERkWLp96OpjP5iC/EHT9KxQWX+2TOCQF9PR6clDqQCT4q3Xcus7pneVR2diYiIiEix8+GqvRxITmdiv8b0aFRNrXaiMXhSjKUnw9y+sHmeozMRERERKTY2HzzJnmOpAIztHsYPz9zOvY2rq7gTQAWeFGeaYEVERETE5uy5HP619Dd6frCWt5ftAqBieTf8vdwdnJkUJ+qiKcVXQl6BF9jIsXmIiIiIONi6vUk8/+UW9iel82CLIJ6/q4GjU5JiSgWeFF+J8VAxGDwrOjoTEREREYdZtu0Iw2fHUcu/HJ8NaUnrOpUcnZIUYyrwpPhK3KzlEURERKTMOpWehW85V9rXD+DvXevzWJsQPN2cHZ2WFHMq8KT46r8IcrIdnYWIiIjILZWUlsmr3+zg14MpfPf07ZRzc+HJDqGOTktKCBV4Unz51XZ0BiIiIiK3jGmafL05gVe+3k5aZjZ/6VAXFyfNiSjXRgWeFE+7l0PKPmj+OGjKXxERESnlUjOyeHpePD/uPEbjmhV4+4Eo6lXxdnRaUgKpwJPiKX42HIqDFkMcnYmIiIhIkSvv5kJ2rsmLdzfk0TYhODvpC265PmrzleIpcTNU0/IIIiIiUnrtP3GGITNjOZaagZOTwYxHm/N429oq7uSGqAVPip/MNEj+Axo/5OhMRERERG66nFyTj9f8wf99/ztuzk78fiSNyt4eGBqWIjeBCjwpfrLOWvceFRybh4iIiMhNtutIKqO/2MzmQ6e4o2EV3rgvgqq+Ho5OS0oRFXgiIiIiIrfIlP/t5VDKWSY92ITuUYFqtZObzjBN09E5XJPo6GgzNjbW0WlIUTJNyM4EJxdw1ncQIiIiUrJtOpCCt7sLdat4k3LmHCbgV97N0WlJCWYYRpxpmtGFxTTJihQ/hgGuHiruREREpERLP5fN60t2cP9/fubf3+8CoGJ5NxV3UqT0CVqKn8xUWPEGhN0LtVo7OhsRERGRa7Z2zwme+3ILB5PP0r9VLUZ3q+/olKSMUIEnxU9WBqz/EPxDVeCJiIhIibNsWyLDZ28ipFJ55g9tRcva/o5OScoQFXgiIiIiIjdByplzVCzvRvv6lRnTrQGPtgnGw9XZ0WlJGaMxeCIiIiIiN+BEWiZPfraJe95fw5nMbDxcnXmifR0Vd+IQasGT4uHoDsg8DUGtHJ2JiIiIyFUxTZPF8Yd59ZsdpGfm8FTHUFyd1X4ijqUCT4qH2fdbyyI8sxXcvcDNCzwrOjorERERkUKlZmQxcu6vrNx1nCZBFXj7/ijqVvF2dFoiKvCkmDiXBkEx1rarJwxfDRVDHJuTiIiIyCWUd7M+Rr98TxgDYoJxdtKC5VI8qA1ZigcnZ6hY68Jjv9rWengiIiIixcQfx9MYPGMjR09n4ORk8Mmg5jzaJkTFnRQrasETEREREbmM7JxcPlqzj3d/+B13Fyf2HEujio8Hhr6MlmJIBZ7cGif2wI7F9vurRkG9Lrc+HxEREZGrsCPhNKMXbmbb4dN0Da/C6/dGUNnHw9FpiVySCjy5NTJPw+E42LW04P6mA6wC78mN4OLumNxERERELuGjNX9w5FQm/3m4KXdGBjo6HZErMkzTdHQO1yQ6OtqMjY11dBpyPbLP2e8znMBZ3zOIiIhI8RH3ZzLeHq7Uq+LNyXTr80uFcm4OzkrkAsMw4kzTjC4spklW5NY4sRsSN4OLW8GbijsREREpJs5kZvPK19t54MN1jP/+d8Aq7FTcSUmiT9dya6ydAHtXwl93ODoTERERETurdx/n+S+3cijlLANjavH3bg0cnZLIdVGBJyIiIiJl2rdbE3liziZqB5RnwfAYmgf7OTolkeumAk9EREREyqTkM+fwK+9GhwaVef7OBgxsHYyHq7Oj0xK5IRqDJyIiIiJlyrHUDJ6YHcc9k9ZwJjMbD1dnhrWro+JOSgW14ImIiIhImWCaJgs3Heb1JTs4m5XDqE51cXNRe4eULirw5NZoMQzCezk6CxERESmjTmdk8eScTazefYLoWhUZd38UoZW9HJ2WyE2nAk9ujcAoR2cgIiIiZZiXmwtuzk682iOc/q1q4eRkODolkSKhNmm5NY5stZZJEBEREblF9h5P47EZGzl6OgMnJ4OPBkYzsHWwijsp1dSCJ7fG+g+1Dp6IiIjcElk5uUz96Q8m/rgbT1dn9h5Lo4qPB4ahwk5KPxV4IiIiIlJqbDt8ijELt7A94TR3RVbl1R4RBHi7OzotkVtGBZ6IiIiIlBrT1+7nWGomHz7SlG4RgY5OR+SWK9ICzzCMbsBEwBn4yDTNcRfFg4BPgQp5xzxnmubSosxJisD+NTD3Qfv9vadD6B3w+3eweR546x9ZERERufk27k/G19OVelW8eal7GAC+5VwdnJWIYxRZgWcYhjMwGegMHAI2GobxtWma+QdhvQh8bprmfwzDCAOWAsFFlZPcRMn7YFoHuHs8BDaCJo/YH+NTw7r3rQEthkKN6Fubo4iIiJRqaZnZvL1sJzPX/Um38Kp82L+ZCjsp84qyBa8FsMc0zT8ADMOYB9wL5C/wTMAnb9sXSCjCfORmMnPhbArkZIF/Hej25qWPrRJ++biIiIjINVq16xj/WLSNhFNnGdQ6mL93re/olESKhaIs8KoDB/M9PgS0vOiYV4DvDcN4CigP3FHYhQzDGAoMBQgKCrrpiYqIiIhIybF0ayIj5myiTkB5vhgeQ7Nafo5OSaTYcPQ6eA8CM0zTrAHcBcwyDMMuJ9M0p5qmGW2aZnRAQMAtT1JEREREHC8pLROAjg0q8+LdDVk6qq2KO5GLFGWBdxiome9xjbx9+Q0GPgcwTXMd4AFUKsKcRERERKSEOXY6g2GzYunx/lrSMrPxcHXm8ba1cXdxdnRqIsVOURZ4G4G6hmGEGIbhBvQDvr7omANAJwDDMBpiFXjHizAnuVncvaHRQ1Ax2NGZiIiISCllmiafxx7kjvH/Y9Wu4/SPqYWHi6M7oIkUb0U2Bs80zWzDMP4CfIe1BMInpmluNwzjNSDWNM2vgb8B0wzDeAZrwpVBpmmaRZWT3ERelaHnfxydhYiIiJRSpzOyGDF7E2v2nKBFsB/j7o+kdoCXo9MSKfaKdB28vDXtll6076V82zuANkWZg4iIiIiUPF5uLni4OvH6fRE83CIIJyfD0SmJlAhq45brk7QXXqsEWxY4OhMREREpJfYcS2XQ9A0cOZWBk5PBtAHR9G9VS8WdyDUo0hY8KeVys6z18ERERERuQFZOLh+u2sukFXso5+7MHyfSqOrrgWGosBO5VirwRERERMRhthw6yegvtrDzSCrdowJ5pUc4lbzcHZ2WSImlAk9EREREHObTn/8kJf0cU/s3o0t4VUenI1LiqcATERERkVtq/R9JVCjnRv2q3rzUPQwM8PV0dXRaIqWCJlmR6+PhCy2GQqVQR2ciIiIiJURqRhYvLt5K36m/MGH57wD4lnNVcSdyE6kFT65P+Upw1zuOzkJERERKiJU7j/HCoq0cOZ3B4NtC+FuXeo5OSaRUUoEn18c0IeccOLmAk7OjsxEREZFi7L9bEnnys03UrezFwida0zSooqNTEim11EVTrk/yH/BGZdj6haMzERERkWLINE2Op2YCcEdYZcZ2D2PJyNtU3IkUMRV4IiIiInJTHT2dwdBZcdw3eS1pmdm4uzgz+LYQ3F3U60ekqKmLpoiIiIjcFKZpMn/jQf659DfOZefybJf6eLioPUHkVlKBJyIiIiI37HRGFsNnxfHz3iRa1fZjXK8ogiuVd3RaImWOCjwRERERuWHe7i54e7jwr56R9GteEycnw9EpiZRJajOX6+NZEW77K1Ru4OhMRERExEF2HUml/8frOXIqA8MwmNI/modaBqm4E3EgteDJ9SnnB3e87OgsRERExAHOZefywao9TF65B28PV/adOENVXw9HpyUiqMCT65WbAxmnwK08uLg7OhsRERG5RTYfPMnoL7aw62gq9zauxkvdw/D30mcBkeJCXTTl+uxZDm+HwIFfHJ2JiIiI3EKzfvmTU2ez+GhANBP7NVFxJ1LMqAVPrs8fq8DFA2q2dHQmIiIiUsTW7U2iYnlXGlT1YWz3MAwDfDxcHZ2WiBRCLXhyff74HwS1Alf1txcRESmtTmdk8fyXW3lw2i+89+NuAHw9XVXciRRjasGTa5d6FI5th6hXHJ2JiIiIFJHlO47yj8VbOZ6aydDba/PMHfUcnZKIXAUVeHLt9v1k3ddu78gsREREpIgs2ZLAXz77lfpVvJnSP5rGNSs4OiURuUoq8OTa1W4PPadA1ShHZyIiIiI3iWmaHE/LpLK3B53DqvDyPWE83LIWbi4a0SNSkugvVq6dVwA06gdOzo7ORERERG6CxFNnefzTWO57fy1pmdm4uzjzaJsQFXciJZBa8OTanDoEu76FiPutxc5FRESkxMrNNZm78QBvLt1JTq7Js13r4+mqL3BFSjIVeHJtfl8GS5+FOh1V4ImIiJRgp85mMXRmLOv3JdMm1J83e0YR5F/O0WmJyA1SgSfX5o//gW9N8Kvt6ExERETkBvh4uFCxnBtv3R9Jn+iaGIbh6JRE5CZQx2q5erk51gyatduB/hMQEREpcXYeOc3DH/1C4qmzGIbBh/2b0bd5kIo7kVJELXhy9Y5sgYyTENLe0ZmIiIjINcjMzmHyij18sGovvp6uHEhKJ9DX09FpiUgRUIEnV+/INjCcrBY8ERERKRE2HUhhzBdb2H0sjZ5NqvNS9zAqlndzdFoiUkRU4Im904mQfsJ+f5NHoOE94KnFTkVEREqKz9YfIC0zm+mDmtOhQWVHpyMiRUwFntj7ZTL8PMl+/0spKu5ERERKgLV7TuBX3o2GgT6M7R6GkwHeHq6OTktEbgEVeHLBnuWQcQoaPQQ1W9rHNQBbRESkWDt1Not//fc35sce5O7IQCY/3BRfTxV2ImWJCjy5IO5TOLEbnvwFqoQ5OhsRERG5Bt9vP8KLi7eRdOYcw9vV4ek76jo6JRFxABV4IiIiIiXcN5sTeGrurzQM9OHjgc2JrOHr6JRExEFU4ImIiIiUQKZpciw1kyo+HnQJr8Jr94bzYIsgXJ21zLFIWaZ/AURERERKmMMnz/LojI30nLyWtMxs3F2cGRATrOJORNSCJyIiIlJS5OaazNlwgHFLfyPXhNHd6uPp6uzotESkGFGBJxfc9Q7kZDk6CxERESnEqbNZDJkZy4Z9ydwWWok3e0VS06+co9MSkWJGBZ5c4F3V0RmIiIjIJfh4uFDJy423H4iid7MaGFq+SEQKoY7acsGOr2HTLEdnISIiInm2J5yi39R1JJw8i2EYfPBwM/pE11RxJyKXpAJPLEl7YfETsH2RozMREREp8zKycnjnu530eH8te46lcTA53dEpiUgJoS6aAplpMP8RcHaF7u86OhsREZEyLe7PZEZ/sYW9x89wf9MajO3ekArl3BydloiUECrwyjrThK//Asd3wiMLoWItR2ckIiJSps3feJCMrFw+fawF7eoFODodESlhVOCVdX+stLpl3vEK1Ono6GxERETKpNW7j+Nf3p2waj682D0MJ8PAy10f00Tk2ulfjrKuTkfovxhqt3d0JiIiImXOqfQs3vjvDhbEHeLuyEAmP9wUHw9XR6clIiWYCjyBOh0cnYGIiEiZs2xbImO/2k7ymXOMaF+HkZ3qOjolESkFNItmWbZuMrzf3NFZiIiIlDlfb05g+OxNBHi589WTbRjdrQEers6OTktESgG14JVl6cnW8ggiIiJS5EzT5MjpDAJ9PekWXpXX74ugX/OauDrr+3YRuXn0L4qIiIhIETuYnM6ATzbQ64OfSc3Iws3Fif6taqm4E5GbTi14IiIiIkUkN9dk5rr9vP3dLgzg/9m77yi5y4L9/+97Zkuy2fTee+8h9F4SSkINRYpd8bE86pciIIgIqBSlihQVFcuDSu+gtJAAgQAhPSGFkEp62832+/fHRn9RSbLAzn52oCaaJwAAIABJREFUd96vc+bsTslyHU4y+7nmbpccP4hmeV5+Scoc32EkSZIyYMuOcr78uzeZvmwThw9oz49PHUa31gVJx5LUyFnwslm7/jDg2KRTSJLUKLVokkOnlk34+RkjOW1MV0IISUeSlAVCjDHpDB/L2LFj4/Tp05OOIUmS9F9mr9zC1U/M5ZazRtGlVdOk40hqpEIIb8UYx37Uc47gSZIkfUol5ZXc+vx73DN5CW2a5bFy8w4LnqREWPCy2VMXw4Kn4f/NTjqJJEkN1pvvb+SSB2ayZH0RZ+zTjSsmDKFlQW7SsSRlKQtetlrwDLxxD4w6L+kkkiQ1aH+bvpyyyir+8OX9OLR/+6TjSMpyFrxstGExPHQ+dBoBE36WdBpJkhqclxaspX3zfIZ2ackPJg4hFQLN8r2skpQ8T9fMNqXb4f5zIJWGs/4Iua4PkCSppjYVlXHBX2fwhd++yZ0vLQageZNcy52kesN3o2xTth3yCuH066B1z6TTSJLUIMQYeXr2Gq58dDabi8v59lH9+OZR/ZKOJUn/xYKXbZp3gi//HVIO3kqSVFOPz1zNt//vHYZ3bcl9X9qfIV1aJB1Jkj6SV/nZYvGLcP+5ULLFcidJUg3EGFm1eQcAxw3txE9OHc7D3zjIciepXvNKPxtsWgYPfBE2LoGQTjqNJEn13vKNxXz2N29w2i9fZVtJOXk5Kc7Zvwc5aS+dJNVvTtFs7Mp3wF/Og6qq6k1V8guTTiRJUr1VWRX5/avvc+OzC0inApceP4hmeV4uSWo4fMdqzGKEx78Da2bBOX+Btn2TTiRJUr21pbicL/7uDd7+YDNHDGzPT04dTpdW7jYtqWGx4DVm29ZUr7074jIYcGzSaSRJqtdaNM2hW+sCPntgT04Z1ZUQQtKRJOljs+A1Zi06w9dfhYK2SSeRJKlemrliM1c/Ppdbzx5N11ZNue3s0UlHkqRPxZXCjdHWVTD5RqiqhML27popSdJ/KCmv5KdPzeOUO6ayfFMxq3fulilJDZ0jeI1NRSn89XOwdh4MOx3a9E46kSRJ9crrSzZw6YMzeX9DMWfv151Ljx9My6a5SceSpFphwWtsnv4erHgTzvyD5U6SpI/w8NsrqYrw56/sz0H92iUdR5JqlQWvMXnr9/DW7+CQC2DISUmnkSSp3nhx/lraN89nWNeWXDFxMOlUoMDjDyQ1Qi7Oaix2bIJnL4e+R8NRVySdRpKkemFjURnfvf8dvvi7N7nr5cUANG+Sa7mT1Gj57tZYNG0Nn3u0elpmKp10GkmSEhVj5ImZq7nqsTls2VHOd47uzzeO9DxYSY2fBa+hqyyHpZOh39HQbZ+k00iSVC889u4qvnP/DEZ2a8mfvro/gzq1SDqSJNUJC15D99wPYNqdcP7L0GVU0mkkSUpMjJGVm3fQrXUBxw/rzI7TKjljbHfSKQ8sl5Q9XIPXkL37l+pyt//XLXeSpKy2bEMR5/xqGpPufJVtJeXk5aT4zH49LHeSso4jeA3V6nfh8W9Dz0Ng/DVJp5EkKRGVVZHfTl3Kz55bQG4qxfcnDKYw38sbSdnLd8CGqKwY/nIeNG0DZ/wW0h7OKknKPluKy/ncb9/g3eWbOWZwB649ZTidWjZJOpYkJcqC1xDlFcAR34d2A6CwQ9JpJElKRIumOfRuW8CXD+nNiSM6E4LTMSXJNXgNzbYPq7+OOttdMyVJWWfG8s2c9suprNhUTAiBWz4zmpNGdrHcSdJOFryGZM4jcOtI+GBa0kkkSapTO8oq+fGTczntl1NZtbmED7eWJB1Jkuolp2g2FGvnwSPfgE7D3DFTkpRVXl28nksfnMUHG4s5Z/8eXHr8IFo0cf25JH0UC15DsGMz3H8u5DWDM/8AOflJJ5Ikqc48NmMVIcD/ffUADuzbNuk4klSvWfDqu6oqePhrsHkZfP4JaNE56USSJGXcP+Z+SKeWTRjWtSVXTBxCOgSa5qWTjiVJ9Z4Fr76LVdC2H/Q9GnoemHQaSZIyasP2Uq56fC6Pv7uKk0Z24bazR3uunSR9DL5j1nfpHDj2x0mnkCQpo2KMPPbuKq56bA5FpZVcOG4AXzu8b9KxJKnBseDVd1tXVa/B6zgk6SSSJGXMY++u4jv3z2BU91bcePoI+ndsnnQkSWqQLHj13ZRbYOZf4NJlSSeRJKlWVVVFVm7eQfc2BZwwvDOlFVVMGtONdMoz7STpk/IcPEmSVOeWri/i7F+9zul3vcq2knJy0ynOHNvdcidJn5IjeJIkqc5UVFZx79Sl/Py5heTlpLhiwmA3UZGkWuQ7qiRJqhNbisv57L3TmLliC+OGdOTaU4bRsUWTpGNJUqOS0SmaIYTjQggLQgiLQgiX7uY1Z4YQ5oYQ5oQQ/pzJPJIkqe7FGAFo0TSHfu0L+cU5o7nns/tY7iQpAzI2ghdCSAN3AOOAFcCbIYTHYoxzd3lNf+Ay4OAY46YQQodM5WmwRpwF3fdLOoUkSZ/I2x9s4kePz+UXZ4+me5sCbjprVNKRJKlRy+QUzf2ARTHGJQAhhPuBk4G5u7zmq8AdMcZNADHGtRnM0zB126f6JklSA1JcVsHPnl3Ib19dSucWTVi3vZTubQqSjiVJjV4mC15XYPku91cA+//HawYAhBCmAmngqhjjM//5g0II5wPnA/To0SMjYeutjUugaAN03zfpJJIk1ciU99Zz6UMzWbFpB587sCffO26QG6lIUh1J+t02B+gPHAF0AyaHEIbHGDfv+qIY4z3APQBjx46NdR0yUa/f5Tl4kqQG5clZq8lNp/jr1w5kv95tko4jSVklkwVvJdB9l/vddj62qxXAtBhjObA0hLCQ6sL3ZgZzSZKkWvbcnDV0btmU4d1acsWEwaRTgSa56aRjSVLWyeQumm8C/UMIvUMIecBngMf+4zWPUD16RwihHdVTNpdkMJMkSapF67aV8s0/v835f3iLX0+p/hXeLD/HcidJCcnYCF6MsSKE8C3gWarX190bY5wTQrgamB5jfGznc+NDCHOBSuDiGOOGTGWSJEm1I8bIw++s5Oon5lJcWsnFxw7k/MP6JB1LkrJeRtfgxRifAp76j8eu3OX7CFyw8yZJkhqIR2es4oK/vsuYHq244fQR9OvQPOlIkiQ+RsELIRTEGIszGUYfYcxnoe9RSaeQJImqqsiKTTvo0baACSM6U1EVOXV0V9KpkHQ0SdJOe12DF0I4aOcUyvk7748MIfwy48lUrdNwGHhc0ikkSVlu8brtnHXPa5x+16tsKyknN53i9H26We4kqZ6pySYrNwPHAhsAYozvAodlMpR2sXY+LH4x6RSSpCxVXlnFL19axPG3vsKCNdu4+NiBnmknSfVYjd6hY4zLQ/i3T+gqMxNH/2X6vZ6DJ0lKxObiMs799TTmrNrKcUM7cfUpQ+nQvEnSsSRJe1CTgrc8hHAQEEMIucB3gHmZjaV/2b4GmrRIOoUkKYvEGAkh0LJpLoM7t+BbR/bj+OGdk44lSaqBmkzR/B/gm0BXqg8qHwV8I5OhtFNVFSydDL0OTTqJJClLvLVsIyffMZXlG4sJIfCzM0Za7iSpAanJCN7AGOO5uz4QQjgYmJqZSPqXNTNhxyboc0TSSSRJjVxRaQU3PruA37/2Pl1aNmX99lK6tylIOpYk6WOqScG7HRhTg8dU25ZOrv7a+/Bkc0iSGrXJC9dx2UOzWLVlB58/sBcXHzuQZm6kIkkN0m7fvUMIBwIHAe1DCLseRN4CSGc6mIADvg69D4PmHZNOIklqxJ6Zs4b83BR//dqB7NurTdJxJEmfwp4+nssDCne+pvkuj28FTs9kKO2UzoUuo5JOIUlqhJ6etZqurZsyolsrLj9hMOlUoEmun99KUkO324IXY3wZeDmE8LsYo3v017UVb8HsB+CQC6CwfdJpJEmNxNqtJVz56ByembOGU0Z14ZbPjHY6piQ1IjV5Ry8OIdwIDAX+dfhNjPGojKUSLHgSpt0NR1yWdBJJUiMQY+SBt1ZwzRNzKamo4pLjBvHVQ3snHUuSVMtqUvD+BPwFmEj1kQmfB9ZlMpSAJS9Dt7GegSdJqhWPzljFxQ/MZN9erblu0gj6ti9MOpIkKQNqUvDaxhh/E0L4zi7TNt/MdLCstmMzrHobDrs46SSSpAasqiqyfFMxPds2Y8KIzkQiJ4/sSioVko4mScqQmhx0Xr7z6+oQwoQQwmjALbYy6f0pEKs8HkGS9IktWruNM+5+jdPveo1tJeXkplOcOrqb5U6SGrmajOBdG0JoCVxI9fl3LYDvZjRVtivZDK16Qrd9k04iSWpgyiuruGfyEm79x3sU5Ke5cuIQCt1ERZKyxl7f8WOMT+z8dgtwJEAI4eBMhsp6o8+DUedC8FNWSVLNbS4u45xfTWPu6q1MGN6Zq04aSvvm+UnHkiTVoT0ddJ4GzgS6As/EGGeHECYC3weaAqPrJmKWqaqCVMpyJ0mqsRgjIQRaNs1laJcWfPvo/hw3rFPSsSRJCdjTGrzfAF8B2gK3hRD+CPwMuCHGaLnLlNkPwC0jYOuqpJNIkhqAN5ZuZOLtU1i+sZgQAjeeMdJyJ0lZbE9TNMcCI2KMVSGEJsAaoG+McUPdRMtSJVtg8zJI5SadRJJUj20vreD6p+fzh9eX0a11UzYUldG9TUHSsSRJCdtTwSuLMVYBxBhLQghLLHeSJCXvxQVrufyhWazeWsKXDu7NRccOoCDPjVQkSXsueINCCDN3fh+AvjvvByDGGEdkPJ0kSfov/5j7Ic3yc3jw6wcxpkfrpONIkuqRPRW8wXWWQpIk7VaMkadmraFb66aM7N6KyycMJp0K5Oekk44mSapndlvwYozL6jKIdmrTG4acAjl5SSeRJNUDa7eWcMUjs3lu7oecOrorN581yumYkqTd8jdEfdPvmOqbJCmrxRj52/QVXPPkXMoqqrjs+EF8+ZDeSceSJNVzFjxJkuqhR2as5HsPzmS/3m24ftIIerdrlnQkSVIDUKOCF0JoCvSIMS7IcB69fR88dwX879vQrF3SaSRJdaiyKvLBxmJ6t2vGxBFdSIXAiSO6kEqFpKNJkhqIPR10DkAI4URgBvDMzvujQgiPZTpY1qoorT4LL8akk0iS6tDCD7cx6c5XOfPu19hWUk5uOsXJo7pa7iRJH0tNRvCuAvYDXgKIMc4IIbgIQJKkWlBWUcVdLy/m9hfeozA/hx+eOJTCfFdQSJI+mZr8BimPMW4J4d8+QXR4SZKkT2lzcRmfued15q/Zxokju/DDE4fQrjA/6ViSpAasJgVvTgjhHCAdQugPfBt4NbOxJElqvGKMhBBo2TSXUd1bceH4gYwb0jHpWJKkRmCva/CA/wWGAqXAn4EtwHczGSqrtRsAo86DHD/BlaTG6PUlGzjhtiks31hMCIHrJo2w3EmSak1NRvAGxRgvBy7PdBgBfQ6vvkmSGpVtJeVc9/R8/jTtA3q0KWBTcRnd2xQkHUuS1MjUpOD9PITQCXgA+EuMcXaGM0mS1Ki8MP9DLn94Nh9uLeErh/TmwvEDaZqXTjqWJKkR2usUzRjjkcCRwDrg7hDCrBDCFRlPlq3e/DX8qA1sX5d0EklSLXlh/lqaN8nhwa8fxBUTh1juJEkZU6N9mGOMa4DbQggvAt8DrgSuzWSwrBUjxMqkU0iSPoUYI0/MXE33NgWM6t6K758wmHQqkJ9jsZMkZVZNDjofHEK4KoQwC7id6h00u2U8mSRJDdCaLSV89b63+N//e4f7Xn0fgIK8HMudJKlO1GQE717gL8CxMcZVGc4jSVKDFGPk/jeX85Mn51FeVcUVEwbzxYN7Jx1LkpRl9lrwYowH1kUQSZIaskdmrOSyh2ZxYJ+2XDdpOD3bNks6kiQpC+224IUQ/hpjPHPn1My461NAjDGOyHi6bNRxKOz/P5DbJOkkkqS9qKyKLNtQRJ/2hZw4ogs5qRQTR3QmhJB0NElSltrTCN53dn6dWBdBtFPPg6pvkqR6bcGabXzvwZms3LSDFy86nOZNcjlxZJekY0mSstxuN1mJMa7e+e03YozLdr0B36ibeFmosgLKd1TvpilJqnfKKqq45R8LmXj7KyzfWMyVJw6hML9Gm1JLkpRxe91FExj3EY8dX9tBtNNbv4Ufd4Ki9UknkST9h01FZZx4+xRu+cd7TBjemX9ccDgnjezilExJUr2xpzV4X6d6pK5PCGHmLk81B6ZmOpgkSfVFjJEQAq0KchnTszWXHD+QowZ1TDqWJEn/ZU9zSv4MPA38FLh0l8e3xRg3ZjSVJEn1xKuL13P143O557Nj6dG2gJ+eNjzpSJIk7daeCl6MMb4fQvjmfz4RQmhjyZMkNWZbdpRz3dPz+L83ltOrbQGbd5TRg4KkY0mStEd7G8GbCLxF9TEJuy4wiECfDOaSJCkxf5/7IVc8Mot120r52mF9+O4xA2ial046liRJe7XbghdjnLjza++6iyO6jIZDL4LcpkknkaSsNXnhOloX5PGrz41lRLdWSceRJKnG9rqvcwjhYGBGjLEohHAeMAa4Jcb4QcbTZaNuY6tvkqQ6E2Pk0Rmr6Nm2gNE9WvP9EwaTTgXycmqy2bQkSfVHTX5z3QkUhxBGAhcCi4E/ZDRVNisrhu3roKoq6SSSlBVWbd7Bl38/ne/+ZQZ/fL36s8umeWnLnSSpQarJb6+KGGMETgZ+EWO8g+qjEpQJM/4EP+sHxRuSTiJJjVpVVeSPry9j/M2TeW3xBq6cOIQbTh+RdCxJkj6VvU7RBLaFEC4DPgscGkJIAbmZjSVJUmY9MmMlVzwym0P6teOnpw2next3yJQkNXw1KXhnAecAX4oxrgkh9ABuzGwsSZJqX0VlFcs2FtO3fSEnjexCk9w0xw/rRAhh739YkqQGYK9TNGOMa4A/AS1DCBOBkhjjfRlPJklSLZq3eiun3fkqZ939OttLK8hJpzhheGfLnSSpUdlrwQshnAm8AZwBnAlMCyGcnulgkiTVhtKKSm56bgEn3j6FVZt38KOThtLMM+0kSY1UTaZoXg7sG2NcCxBCaA/8A3ggk8GyVrd94ZirIM+1IJL0aW0qKuOMu19j0drtnDamKz+YMITWzfKSjiVJUsbUpOCl/lnudtpAzXbf1CfRZVT1TZL0icUYCSHQqiCXA/q04fIJgzlyYIekY0mSlHE1KWrPhBCeDSF8IYTwBeBJ4KnMxspiJVtg41Koqkw6iSQ1SK+8t45jb5nMBxuKCSFw7SnDLXeSpKxRk01WLgbuBkbsvN0TY7wk08Gy1sy/wm2joHhj0kkkqUHZsqOc7z3wLp/9zRtUVEa2lpQnHUmSpDq32ymaIYT+wM+AvsAs4KIY48q6CiZJUk09O2cNP3hkNhuKyvj6EX35ztH9aZLrRiqSpOyzpzV49wL3AZOBE4HbgdPqIpQkSR/H1EXraVeYz71f2JdhXVsmHUeSpMTsqeA1jzH+auf3C0IIb9dFIEmS9ibGyMPvrKRXu2aM6dGay44fTE46kJt2DzBJUnbbU8FrEkIYDfzzBNimu96PMVr4JEl1bsWmYi5/eDYvL1zH6ft0Y0yP1jT1XDtJkoA9F7zVwE273F+zy/0IHJWpUFmt58Fwws8gr1nSSSSpXqmqivxx2jKuf3o+EbjqxCF89sBeSceSJKle2W3BizEeWZdBtFPHIdU3SdK/eWTGSq58dA6H9m/HT04dTvc2BUlHkiSp3qnJQeeqS0UbYNsqaD8I0rlJp5GkRJVXVrFsQxH9OjTnpJFdKMhLc+zQToQQ9v6HJUnKQq5Gr2/mPAR3HQI7NiedRJISNXvlFk65Yyqfued1tpdWkJNOcdywzpY7SZL2wBE8SVK9UlJeye0vvMddLy+hdUEe15w8jMJ8f11JklQTe/2NGao/Kj0X6BNjvDqE0APoFGN8I+PpJElZZVNRGaff9SqL1xVx+j7duGLCYFoV5CUdS5KkBqMmH4n+EqiietfMq4FtwIPAvhnMJUnKIlVVkVQq0Kogl4P6tuOHJw7lsAHtk44lSVKDU5M1ePvHGL8JlADEGDcBfpwqSaoVkxeuY/wtk3l/fREhBK45ZZjlTpKkT6gmI3jlIYQ01WffEUJoT/WInjKhzxFwyl2QX5h0EknKqM3FZVz75DweeGsFfdo3Y3tpRdKRJElq8GpS8G4DHgY6hBB+DJwOXJHRVNmsXf/qmyQ1Yk/PWs0PHp3DpuIyvnVkP751VD+a5KaTjiVJUoO314IXY/xTCOEt4GggAKfEGOdlPFm22rYGNi6FrvtAjjNhJTVOry/ZQMcW+fz+S/sytEvLpONIktRo1GQXzR5AMfD4ro/FGD/IZLCsNe9xeOoiuGgRFLoGRVLjEGPcORWzkH16tuayEwaTkwrkpD2OVZKk2lSTKZpPUr3+LgBNgN7AAmBoBnNJkhqJ5RuL+f7Ds3jlvfWcObYb+/Rs7XRMSZIypCZTNIfvej+EMAb4RsYSSZIahcqqyH2vvc+Nzy4gANecPJRz9++ZdCxJkhq1mozg/ZsY49shhP0zEUaS1Hg88s5KfvT4XA4f0J6fnDacrq2aJh1JkqRGryZr8C7Y5W4KGAOsylgiSVKDVV5Zxfvri+jfsTknj+pC8yY5jBvSkRBC0tEkScoKNVnd3nyXWz7Va/JOzmSorNbvGDjrj9CkRdJJJOljmb1yCyf9Yipn/+p1tpdWkJNOMX5oJ8udJEl1aI8jeDsPOG8eY7yojvKoTe/qmyQ1ECXlldz6/HvcM3kJbZrl8eNTh1OY/7FXAEiSpFqw29/AIYScGGNFCOHgugyU9da/B5vehz5HQDo34TCStGebisqYdOerLFlfxFlju/P9EwbTssD3LkmSkrKnj1jfoHq93YwQwmPA34Cifz4ZY3wow9my06PfhOIN8K3pSSeRpN2qqoqkUoFWBbkc2r8dV588jEP6t0s6liRJWa8ma/CaABuAo4CJwIk7v6q2rZ0Hy6fBPl8A16xIqqdeXLCWY25+mffXFxFC4EeWO0mS6o09jeB12LmD5mz+/4PO/ylmNFW2euv3kMqFkWcnnUSS/sumojKueWIuD72zkv4dCikuq0w6kiRJ+g97KnhpoJB/L3b/ZMGrbeUlMPN+GDwRmvlJuKT65cmZq7ny0dls2VHOt4/uzzeP7Et+TjrpWJIk6T/sqeCtjjFeXWdJst26+dVf9/lCojEk6aO8+f5GurRqyh+/sj+DO3uMiyRJ9dWeCp6LwOpSl1FwwXxI5yWdRJKIMfK36Svo26EZ+/Rsw6XHDyInFchJ12TptiRJSsqeCt7RdZYi25UVQ04TyG2SdBJJ4oMNxVz28EymLtrAWWO7s0/PNjTJdTqmJEkNwW4/io0xbvy0PzyEcFwIYUEIYVEI4dI9vG5SCCGGEMZ+2v9mg/Tij+GXB0BFWdJJJGWxyqrIb6Ys5dhbJvPu8i1ce8owfnra8KRjSZKkj2FPI3ifSgghDdwBjANWAG+GEB6LMc79j9c1B74DTMtUlnqtohTe/T/oeRDkOD1TUnIeeWcl1zwxl6MGdeDaU4bRpVXTpCNJkqSPKWMFD9gPWBRjXAIQQrgfOBmY+x+vuwa4Hrg4g1nqr5VvVR9sPuIzSSeRlIXKKqpYur6IgZ2ac/KoLrRsmsvRgzsQPItTkqQGKZOr5bsCy3e5v2LnY/8SQhgDdI8xPrmnHxRCOD+EMD2EMH3dunW1nzRJleXVX5u2TjaHpKzz7vLNnPSLKZzzq9fZXlpBTjrFMUM6Wu4kSWrAEtsOLYSQAm4CLtzba2OM98QYx8YYx7Zv3z7z4SSpEdtRVslPnprHqb+cyubicq6fNILC/ExO6JAkSXUlk7/RVwLdd7nfbedj/9QcGAa8tPPT4k7AYyGEk2KM0zOYq35p3QuOugJa9Ug6iaQssLGojNN+OZX3NxRz9n49uOyEQbRokpt0LEmSVEsyWfDeBPqHEHpTXew+A5zzzydjjFuAdv+8H0J4Cbgoq8odQOuecFh2Lj+UVHcqqyLpVKB1QS5HDurAuCEdOahvu73/QUmS1KBkbIpmjLEC+BbwLDAP+GuMcU4I4eoQwkmZ+u82OOUlsHl59VdJyoAX5n/IMTe9zNL1RYQQ+OGJQy13kiQ1UhlddBFjfAp46j8eu3I3rz0ik1nqreXT4L6T4AtPQa+Dk04jqRHZsL2Uq5+Yy6MzVjGgYyEl5ZVJR5IkSRnmqnpJaoQef3cVP3xsDttKyvnuMf35xhH9yMtJbF8tSZJURyx4ktQIvbVsE93bFHDDpBEM7NQ86TiSJKmOWPASF5MOIKkRqKqK3P/mcgZ0LGRsrzZcevwgctMp0inPtJMkKZtY8JK26B8QUtCy695fK0kf4f31RVz60ExeX7KRs/frzthebWiSm046liRJSoAFL2n7/w+0H1x9Hp4kfQyVVZF7pyzl539fQG4qxXWnDeesfbvv/Q9KkqRGy4KXtJbdYPS5SaeQ1AA98s5KfvzUPI4Z3IFrTxlOp5ZNko4kSZISZsFLyvr34KmLYcLPoW3fpNNIaiDKKqpYsn47gzq14JTRXWnTLI8jBrYnBNfaSZKkDB50rr144VpY/gbkt0g6iaQGYsbyzUy8/RXO/dU0tpdWkE4FjhzUwXInSZL+xRG8JKx6B+Y+Aod9DwrbJ51GUj23o6ySnz+3gHunLqVjiybceMYICvN9+5YkSf/NK4QkPH81NG0DB30r6SSS6rmNRWWccsdUPthYzHkH9OCS4wbRvElu0rEkSVI9ZcGra0tfgcUvwPhroUnLpNNIqqcqqyLpVKB1QS7jhnRk3JCOHNCnbdKxJElSPecavLrWZTSMuwb2/UrSSSTVU8/NWcORP3uJJeu2E0LgBxOHWO4kSVKNOIJX1/IL4eBvJ51CUj20fnspVz02hydmrmZQp+aUVVYlHUmSJDUwFry6UlUJf/0c7PNF6H9M0mkk1TOPzlgEBtocAAAgAElEQVTJVY/Noai0kgvHDeB/juhLbtpJFpIk6eOx4NWVmX+F+U/A8NOTTiKpHnp3+RZ6tWvGDZNG0L9j86TjSJKkBsqCVxcqyuCln0DnUTD45KTTSKoHqqoif37jAwZ2as6+vdrwveMGkptOkU55pp0kSfrkLHh14a3fweYPYOItkHLKlZTtlq4v4pIHZ/LG0o2cvV8P9u3Vhia56aRjSZKkRsCCVxdevR16HAR9j0o6iaQEVVRW8espS7n57wvJy0lxw6QRnDG2W9KxJElSI2LBqwuHfBc6DIbg1Cspmz0yYxXXPT2f8UM6cs0pw+jYoknSkSRJUiNjwasL+3456QSSElJaUcnitUUM6dKCU0d3pV1hHocPaE/wAx9JkpQBLgirC2vnwbY1SaeQVMfeWraJCbdN4bzfTGN7aQXpVOCIgR0sd5IkKWMseHXh1+Ng6m1Jp5BUR4pKK/jR43M4/a5X2VFWyU1njqQw3wkTkiQp87zikKRatGF7KSffMZUVm3bw+QN7cvFxgyx3kiSpznjVIUm1oKKyipx0iraF+Rw/rBPjh3Zi315tko4lSZKyjFM0JelTemb2Go78+UssWbcdgMsnDLHcSZKkRDiCJ0mf0LptpVz12ByenLWaIZ1bUF4Zk44kSZKynAWvLky8Cdr0TTqFpFr08DsruOqxuewoq+TiYwdy/mF9yE07KUKSJCXLglcXRpyZdAJJtWzWiq3061DI9ZNG0K9DYdJxJEmSAAte3VgxHZq1g9a9kk4i6ROqqor8cdoyBnVqwX6923DJ8QPJSaVIpzzTTpIk1R/OJ6oL950C0+5JOoWkT2jxuu2cefdrXPnoHB6dsRKA/Jy05U6SJNU7juBJ0m6UV1Zxz+Ql3Pr8ezTNTfOzM0YyaUzXpGNJkiTtlgVPknbj0RmruPHZBZwwvBNXnTSUDs2bJB1JkiRpjyx4krSLkvJKFq/bztAuLTl1dFc6tsjn0P7tk44lSZJUI67Bk6Sdpr+/kRNue4XP/uYNikorSKeC5U6SJDUojuDVhdPuhlY9kk4haTe2l1Zw4zPzue/1ZXRp2ZRbzhpFs3zfHiVJUsPjFUxdGDQh6QSSdmNjURkn3j6FVVt28PkDe3HxsQMtd5IkqcHyKqYuLH4RWnSF9gOSTiJpp4rKKnLSKdo0y2PiiM6MG9KRsb3aJB1LkiTpU3ENXl34y2fhrd8lnULSTk/NWs3hN77EorXbAbjshMGWO0mS1Cg4gicpa6zdWsKVj87hmTlrGNa1BTHGpCNJkiTVKguepKzwwFsruPrxOZRWVHHp8YP4yiG9yUk7iUGSJDUuFjxJWWH+6q0M6tSC6yYNp0/7wqTjSJIkZYQFT1KjVFkV+cNr7zO4cwv279OWi48bSG4qRSoVko4mSZKUMRa8uvCZP1bvoimpTixau41LHpzFW8s2cd4BPdi/T1vyc9JJx5IkSco4C15d6HNE0gmkrFBeWcXdLy/mtucXUZCf5uazRnLKKD9ckSRJ2cOCVxfmPwmtekCn4UknkRq1R2es4mfPLWTCiM786KShtCvMTzqSJElSnbLg1YWHvgZjPgfHWfCk2lZSXsl7H25neLeWnDq6K51bNuHgfu2SjiVJkpQI9wiX1GBNW7KB4299hc/dO42i0grSqWC5kyRJWc0RPEkNzraScq5/Zj5/fP0Durdpyi/OGUOzfN/OJEmSvCKS1KBs2F7KibdPYc3WEr5ySG8uGD+AgjzfyiRJksCCJ6mBKK+sIjedom1hPieP7sr4IR0Z3aN10rEkSZLqFdfg1YXPPQr7fy3pFFKDFGPkiZmrOPyGF1m0djsAlxw3yHInSZL0ERzBqwvd9kk6gdQgfbi1hCsemc3f537IiG4tgZh0JEmSpHrNgpdpC5+FTctg7BchnZt0GqnB+Ov05VzzxFzKKqr4/gmD+NLBvclJO+lAkiRpTyx4mTblZti+Fvb7atJJpAZl4ZptDOncgusnjaBXu2ZJx5EkSWoQLHiZtHY+fPAajLsaQkg6jVSvVVZFfjt1KcO6tuSAPm353nGDyEkFUin/7UiSJNWUBS+T3r4PUrkw8pykk0j12sIPt/G9B2YyY/lmPn9gTw7o05a8HKdjSpIkfVwWvEwpL4F3/wyDJkBh+6TTSPVSWUUVd760mF+8+B7Nm+Ry62dGcdLILknHkiRJarAseJmy/UNoPxj2+ULSSaR667F3V3HzPxZy8qguXDlxCG0L85OOJEmS1KBZ8DKldU/40tNJp5DqnR1llby3dhsjurXitNFd6dqqKQf2bZt0LEmSpEbBRS6ZsH1d9U3Sv3lt8QaOu3Uyn7/3DYpKK0ilguVOkiSpFlnwMuHV2+DWEVCyNekkUr2wtaScyx6axdm/eh2AO84dQ7N8JxBIkiTVNq+wMmHZq9BlDDRpkXQSKXEbtpcy4bYprN1WwlcP7c0F4wbSNC+ddCxJkqRGyYKXERFymyQdQkpUWUUVeTkp2hbmM2mfrowb0olR3VslHUuSJKlRc4qmpFoVY+TRGSs57IYXWbR2GwAXHzvIcidJklQHHMGTVGtWb9nBFQ/P5vn5axnVvRUhhKQjSZIkZRULXiYceiGkPc9L2eX+Nz7g2ifnUVFVxRUTBvPFg3uTTlnwJEmS6pIFLxMGTUg6gVTnlqwvYkS3llx32gh6tC1IOo4kSVJWsuBlwtp5kMqBdv2TTiJlTEVlFb+d+j7DurbkwL5tuWj8QHLTwWmZkiRJCXKTlUx49JvwzKVJp5AyZv6arUy681V+/NQ8np2zBoC8nJTlTpIkKWGO4EmqsdKKSu54cTG/fHERLZvmcvvZo5k4onPSsSRJkrSTBU9SjT02YxW3Pf8ep47uyg8mDqFNs7ykI0mSJGkXFjxJe1RcVsHCD7czqnsrJo3pRo82Bezfp23SsSRJkvQRXIMnabemLlrPsbdM5ou/fYOi0gpSqWC5kyRJqsccwcuEo66AtFPX1HBt2VHOT56cx1+mL6d3u2bced4+NMv37UKSJKm+84otE/oelXQC6RNbv72UE259hfXbS/na4X34f8cMoEluOulYkiRJqgELXiasfLv6HLzOI5JOItVYWUUVeTkp2hXmc9a+3Rk3pCMjurVKOpYkSZI+BtfgZcJTF8HzP0o6hVQjMUYefmcFh97wAu99uA2AC8cPtNxJkiQ1QI7gSVls1eYdXP7wLF5csI4xPVqRTnlQuSRJUkNmwZOy1J+nfcBPnppHZVXkhycO4XMH9rLgSZIkNXAWPClLLdtYxKjurfjpacPp3qYg6TiSJEmqBRY8KUtUVFbx6ylLGdGtJQf1bcdF4weSkwqE4KidJElSY2HBy4RjfwLp3KRTSP8yd9VWvvfgu8xeuZUvHtyLg/q2IzftHkuSJEmNjQUvE3ockHQCCYDSikp+8cIi7nxpMa0KcvnluWM4flinpGNJkiQpQyx4mfD+FEjnQff9kk6iLPf4u6u5/YVFTBrTjR9MHEyrgrykI0mSJCmDLHiZ8PcroWlrOO/BpJMoCxWVVrDww22M7tGa00Z3pVfbAsb2apN0LEmSJNUBF+FIjcgr763j2Fsm86XfvUlxWQWpVLDcSZIkZRFH8KRGYEtxOdc+OZe/vbWCPu2bcfNZYynI85+3JElStvEKUGrg1m8v5fhbX2FjURnfOKIv3z66P01y00nHkiRJUgIseFIDVVpRSX5OmnaF+Zy9Xw/GD+nIsK4tk44lSZKkBFnwMmHizZDyf60yI8bIQ2+v5Lpn5vPnr+xP/47NuWDcgKRjSZIkqR7I6CYrIYTjQggLQgiLQgiXfsTzF4QQ5oYQZoYQng8h9MxknjrTeSR0HJp0CjVCKzYV8/nfvsmFf3uXnm0KyPGwckmSJO0iY8NMIYQ0cAcwDlgBvBlCeCzGOHeXl70DjI0xFocQvg7cAJyVqUx15r2/V5+D1+fwpJOoEfnD68v46VPzAPjRSUP57AE9SaVCwqkkSZJUn2RyHuF+wKIY4xKAEML9wMnAvwpejPHFXV7/OnBeBvPUnZd+Wn0OngVPtWjlph2M7dWGn5w6jG6tC5KOI0mSpHookwWvK7B8l/srgP338PovA09/1BMhhPOB8wF69OhRW/mkeq28sop7Ji9hdPdWHNSvHReNH0A6FQjBUTtJkiR9tHqxgCeEcB4wFrjxo56PMd4TYxwbYxzbvn37ug0nJWD2yi2c/Iup3PjsAp6fvxaAnHTKcidJkqQ9yuQI3kqg+y73u+187N+EEI4BLgcOjzGWZjCPVO+VlFdy6/Pvcc/kJbRplsdd543huGGdk44lSZKkBiKTBe9NoH8IoTfVxe4zwDm7viCEMBq4Gzguxrg2g1mkBuGJmau586XFnLFPN66YMISWBblJR5IkSVIDkrGCF2OsCCF8C3gWSAP3xhjnhBCuBqbHGB+jekpmIfC3nVPPPogxnpSpTHXm1Lsh1IvZr2oAtpdWsPDDbYzp0ZrTRneld7tm7NOzddKxJEmS1ACFGGPSGT6WsWPHxunTpycdQ6oVLy1Yy+UPz6a4rIKplx5FQV4mB9UlSZLUGIQQ3ooxjv2o57yazIS5j0I6HwYel3QS1VObisq45sm5PPT2Svq2b8ZtZ4+13EmSJOlT84oyE6beWn0OngVPH2H99lKOu2Uym4vL+d+j+vHNI/vRJDeddCxJkiQ1AhY8qY6UlFfSJDdNu8J8zjugJ+OHdGJIlxZJx5IkSVIj4k4gUobFGPnr9OUccv0LLPxwGwDfPWaA5U6SJEm1zhE8KYOWbyzmsodmMWXRevbr1Yb8HD9TkSRJUuZY8KQMue+19/npU/NJBbjmlGGcu18PUqmQdCxJkiQ1Yha8TDjzPghumpHtVm8pYf8+bfjxqcPp2qpp0nEkSZKUBSx4mdCyW9IJlIDyyiruemkxY3q25uB+7bhw3ADSqUAIjtpJkiSpbrggKBPevR/mPJx0CtWhWSu2cOLtU/j53xfy4vy1AOSkU5Y7SZIk1SlH8DLhjXuqz8EbemrSSZRhJeWV3PyPhfxq8hLaFeZz92f34dihnZKOJUmSpCxlwZM+hSdmrubul5dw1tjufH/CYFo2zU06kiRJkrKYBU/6mLaVlLNgzTbG9mrDaaO70qd9M8b0aJ10LEmSJMk1eNLH8eL8tRx782S+et90issqSKWC5U6SJEn1hiN4Ug1sLCrjmifm8vA7K+nfoZA7zh1DQZ7/fCRJklS/eIWaCec+AO6e2Gis317KsTdPZsuOcr5zdH++cWRf8nM851CSJEn1jwUvEwraJJ1AtaCkvJImuWnaFebz+YN6MX5oRwZ1apF0LEmSJGm3XIOXCdPvrT4LTw1SjJH73/iAg697gQVrtgHw7aP7W+4kSZJU7zmClwnv/LH6HLyRn0k6iT6mZRuKuOyhWby6eAP7925D01ynYkqSJKnhsOBJO/126lKuf2Y+OakUPzl1OJ/ZtzuplGspJUmS1HBY8KSd1m4r5aC+7fjxqcPo3LJp0nEkSZKkj82Cp6xVVlHFnS8tZkzPVhzavz0XjhtAOhUI7oAqSZKkBspNVpSV3l2+mRNvn8LN/1jIlPfWA5CTTlnuJEmS1KA5gpcJX3gq6QTajR1lldz09wX8ZspSOjRvwq8/N5ZjhnRMOpYkSZJUKyx4mZDbJOkE2o0nZ63mV68s5Zz9e3Dp8YNo0SQ36UiSJElSrbHgZcJrd0BuUxj7paSTCNhaUs6CNdvYt1cbThvdlf4dChnZvVXSsSRJkqRa5xq8TJj9IMx/MukUAp6f9yHjb5rM+fdNp7isglQqWO4kSZLUaDmCp0Zpw/ZSfvT4XB57dxUDOzbnhtP3oSDPv+6SJElq3LziVaOzfnsp42+ezLaScv7fMQP4+hF9yctxsFqSJEmNnwVPjcaOskqa5qVpV5jPlw7uxbghnRjYqXnSsSRJkqQ647BGRoSdN9WFqqrIn6Yt46Drnmf+mq0AfOuo/pY7SZIkZR1H8DLhq88nnSBrLF1fxKUPzmTa0o0c1LctzVxnJ0mSpCzm1bAarN9MWcoNz8wnLyfF9ZOGc+bY7oTgyKkkSZKylwWvtr31e1jyEpx6N+TkJZ2mUdtYVMphA9pz7SnD6NjCw+UlSZIkC15tm/c4bHrfcpcBpRWV3PHiYvbt1ZpD+7fngnEDSQUctZMkSZJ2cpOV2lRRBsumQp8jkk7S6Lz9wSYm3jaF255/jymL1gOQTgXLnSRJkrQLR/Bq04o3obzYgleLissq+PlzC7l36lI6tWjCb7+wL0cO6pB0LEmSJKlesuDVpiUvQUhBr0OSTtJoPDVrDb+ZspTzDujBJccNonmT3KQjSZIkSfWWBa82NWsPw8+Apq2STtKgbdlRzvzVW9m/T1tOG92VgR2bM7xby6RjSZIkSfWeBa827X8+cH7SKRq05+as4YpHZlNeWcXUS4+iIC/HcidJkiTVkAWvtuzYDHnNIO0Uwk9i3bZSrnp8Dk/OXM3gzi24YdIICjy0XJIkSfpYvIKuLS9dB7MfgAsXQCqddJoGZd22Usbf/DJFpZVcNH4AXzu8L7lpN3iVJEmSPi4LXm1Z8hJ0HGa5+xiKyyooyMuhffN8vnpYH8YP6Ui/Ds2TjiVJkiQ1WA6T1IZta2DdPI9HqKGqqsgfXl/GQde9wLzVWwH4xhH9LHeSJEnSp+QIXm1YOrn6a5/Dk83RACxZt51LH5zFG+9v5ND+7SjM96+gJEmSVFu8uq4NH7wG+S2h04ikk9Rrv35lCTc+u4D8nBQ3nj6C0/fpRggh6ViSJElSo2HBqw3DJkHPg11/txebiss4YmB7rjl5GB1aNEk6jiRJktToWPBqQ69Dkk5QL5WUV/KLFxaxb+82HD6gPReMG0g65YidJEmSlCluslIbNi6F1TOTTlGvvLVsIxNue4VfvLiI15dsALDcSZIkSRnmCF5tePl6WDYVvjsr6SSJKyqt4MZnF/D7196nS8um/P5L+3H4gPZJx5IkSZKyggVPterp2Wv4/Wvv87kDenLxcYPcJVOSJEmqQ15961PbUlzOvDVbOaBPWyaN6crgzs0Z2qVl0rEkSZKkrOMaPH0qz8xezTE3v8zX//gWO8oqCSFY7iRJkqSEOIKnT2TtthJ++Ogcnp69hiGdW3DD6SNomucxEZIkSVKSLHi1Yd+vwJBTkk5RZ9ZtK2XcTZPZUV7JxccO5PzD+pCbdjBYkiRJSpoFrzZ0G5t0gjpRVFpBs/wc2jfP5+tH9OWYwR3p16Ew6ViSJEmSdnLYpTasnQ8fTEs6RcZUVUV+N3UpB/70eeau2grA/xze13InSZIk1TOO4NWGqbc02nPwFq3dzqUPzmT6sk0cNqA9LQtyk44kSZIkaTcseNqtu19ezM//vpCmuWl+fsZIThvTlRBC0rEkSZIk7YYFT7u1vbSCcYM7ctVJQ2nfPD/pOJIkSZL2woKnfykpr+S2599jv95tOGJgB/7fMQNIpRyxkyRJkhoKC54AePP9jVzywEyWrC8iBDhiYAfLnSRJktTAWPBqw4HfhFHnJJ3iE9leWsENz8znvteW0a11U/7w5f04tH/7pGNJkiRJ+gQseLWh0/CkE3xiz8xewx9eX8YXD+7FReMH0izfvxKSJElSQ+XVfG1YNQN2bIS+RyWdpEY2FZUxb81WDurbjkljujKkcwuGdGmRdCxJkiRJn5IHndeGaXfB499JOsVexRh5cuZqxt38Mt/809vsKKskhGC5kyRJkhoJR/CyxNqtJVzxyGyem/shw7u25PpJI2ial046liRJkqRaZMHLAuu2lXLMTS9TWlHFZccP4suH9CYn7eCtJEmS1NhY8Bqx7aUVFObn0L55Pt86qh/HDO5In/aFSceSJEmSlCEO4zRClVWRe6cs5cCfPs+cVVsAOP+wvpY7SZIkqZFzBK82HHoh7PuVpFMA8N6H27jkwZm8/cFmjhjYntYFeUlHkiRJklRHLHi1oV3/pBMAcOdLi7n57wtplp/mlrNGcfKoLoQQko4lSZIkqY5Y8GrDB9OgaC0MPjHRGDvKKhg/tCNXnTSUdoX5iWaRJEmSVPcseLXhrd/Csql1XvBKyiu5+e8LOaBvW44c2IHvHjOAVMoRO0mSJClbWfAaqNeXbODSB2fy/oZi8nNSHDmwg+VOkiRJynIWvAZmW0k51z09nz9N+4AebQr481f256B+7ZKOJUmSJKkesOA1MM/N+ZD/e+MDvnJIby4cP5CmeemkI0mSJEmqJyx4DcDGojLmr97KQf3acdqYrgzr2pKBnZonHUuSJElSPWPBqw1Hfh/Kimr9x8YYeWLmaq56bA4RmHrJUTTNS1vuJEmSJH0kC15taNWj1n/kh1tLuPzh2fxj3oeM7NaS608f4XRMSZIkSXtkwasNS16CbR/CyLNq5cet21bKMTe9THllFZefMJgvHdKbtDtkSpIkSdoLC15tePf+6nPwPmXB21ZSTvMmubRvns93ju7PMYM70qtds1oKKUmSJKmxSyUdQFBZFfn1K0s46KcvMHvlFgC+cmgfy50kSZKkj8URvIQtWLON7z04k3eXb+boQR1oV5ifdCRJkiRJDZQFL0F3vLiIW/6xkOZNcrnt7NGcOKIzIbjWTpIkSdInY8FLUFlFFScM78wPTxxKm2Z5SceRJEmS1MBZ8GrDuKuhfMdeX7ajrJKb/r6Ag/q248hBHfjuMf0dsZMkSZJUayx4taGww15f8uri9Vz6/7V398Fy1fUdx98fE8Jz40ySOhYI0ZJUU6E8pAh1EG0ylIeWlIIC1RE0Ix2nUBHrlLYMOqhYRHF0pK0ITKi1JEorkxZsRJ7CKI9DYiBUmAxYBK0g0FAE5OnbP/aks73c5O6F7G725P2ayWT3nN8553t3vrP3fvb8ztl/vosHH3+aHadN5Z1v+lXDnSRJkqQtyoC3Jdy3Ep58GBZ84GWrnnz2eT5z9Q+5/LYH2XPGTlz+wYM4+NdnDKFISZIkSW1nwNsS1n2r8z144wS8a9b9jOW3P8gpb38jH1k0jx2nTRlCgZIkSZK2BQa8PnjsqV9yz0+f5JC5s/ij/Xdj792nM+91uw67LEmSJEkt19cvOk9yeJJ7k6xPcuY467dPsrxZf2uSOf2sp9+qiitXP8yiC27kw8vW8MxzL5LEcCdJkiRpIPoW8JJMAS4EjgDmAycmmT9m2BLgiaraC/gCcF6/6um3F14qllx2B6cvX8OcmTuz/JSDnI4pSZIkaaD6OUXzQGB9Vd0PkGQZsBi4p2vMYuATzeMrgC8nSVVVH+va4p55/kUe2/AsNz/xGGf//nxO+p05THmNd8iUJEmSNFj9DHi7AT/uev4Q8NZNjamqF5JsAGYAP+8elOQU4BSA2bNn96veV2zHoz/HjTPXs/K33szsGTsNuxxJkiRJ26i+XoO3pVTVRVW1oKoWzJo1a9jlvNwO03nP7x5guJMkSZI0VP0MeA8De3Q9371ZNu6YJFOB6cBjfaxJkiRJklqrnwHvdmBukjckmQacAKwYM2YFcFLz+DjgulG7/k6SJEmSthZ9uwavuabuVGAlMAW4tKrWJTkHuKOqVgCXAF9Lsh54nE4IlCRJkiS9An39ovOquhq4esyys7sePwu8q581SJIkSdK2YiRusiJJkiRJmpgBT5IkSZJawoAnSZIkSS1hwJMkSZKkljDgSZIkSVJLGPAkSZIkqSUMeJIkSZLUEgY8SZIkSWoJA54kSZIktYQBT5IkSZJawoAnSZIkSS1hwJMkSZKkljDgSZIkSVJLGPAkSZIkqSUMeJIkSZLUEgY8SZIkSWoJA54kSZIktYQBT5IkSZJawoAnSZIkSS2Rqhp2DZOS5FHgP4ddxzhmAj8fdhFqLftL/WR/qd/sMfWT/aV+2lr7a8+qmjXeipELeFurJHdU1YJh16F2sr/UT/aX+s0eUz/ZX+qnUewvp2hKkiRJUksY8CRJkiSpJQx4W85Fwy5ArWZ/qZ/sL/WbPaZ+sr/UTyPXX16DJ0mSJEkt4Rk8SZIkSWoJA54kSZIktYQBb5KSHJ7k3iTrk5w5zvrtkyxv1t+aZM7gq9So6qG/zkhyT5K1Sa5Nsucw6tRomqi/usYdm6SSjNRtoTVcvfRXknc372HrkvzToGvUaOvhd+TsJNcnWd38njxyGHVq9CS5NMkjSe7exPok+VLTe2uT7D/oGifDgDcJSaYAFwJHAPOBE5PMHzNsCfBEVe0FfAE4b7BValT12F+rgQVVtQ9wBfDZwVapUdVjf5FkV+DDwK2DrVCjrJf+SjIX+EvgbVX1m8DpAy9UI6vH97CzgG9U1X7ACcDfDrZKjbClwOGbWX8EMLf5dwrwdwOo6RUz4E3OgcD6qrq/qp4DlgGLx4xZDFzWPL4CWJgkA6xRo2vC/qqq66vq6ebpLcDuA65Ro6uX9y+AT9L5YOrZQRankddLf30QuLCqngCoqkcGXKNGWy89VsCvNI+nAz8ZYH0aYVW1Cnh8M0MWA/9QHbcAr03y+sFUN3kGvMnZDfhx1/OHmmXjjqmqF4ANwIyBVKdR10t/dVsCfLuvFalNJuyvZsrJHlV11SALUyv08v41D5iX5HtJbkmyuU/LpbF66bFPAO9N8hBwNXDaYErTNmCyf6MN1dRhFyBp8pK8F1gAHDrsWtQOSV4DXACcPORS1F5T6Uxveged2QerkuxdVf891KrUJicCS6vq80kOBr6W5C1V9dKwC5MGyTN4k/MwsEfX892bZeOOSTKVzhSBxwZSnUZdL/1FkkXAXwNHV9UvB1SbRt9E/bUr8BbghiQ/Ag4CVnijFfWol/evh4AVVfV8VT0A3Ecn8Em96KXHlgDfAKiqm4EdgJkDqU5t19PfaFsLA97k3A7MTfKGJNPoXMC7YsyYFcBJzePjgOvKb5NXbybsryT7AV+hE+68fkWTsdn+qqoNVTWzquZU1Rw613geXeeNygUAAAUcSURBVFV3DKdcjZhefj9eSefsHUlm0pmyef8gi9RI66XHHgQWAiR5M52A9+hAq1RbrQDe19xN8yBgQ1X9dNhFbYpTNCehql5IciqwEpgCXFpV65KcA9xRVSuAS+hMCVhP52LNE4ZXsUZJj/11PrAL8M3m3j0PVtXRQytaI6PH/pJekR77ayVwWJJ7gBeBj1WVM1zUkx577KPAV5N8hM4NV072Q3b1IsnldD6Amtlcw/lxYDuAqvp7Otd0HgmsB54G3j+cSnsT+16SJEmS2sEpmpIkSZLUEgY8SZIkSWoJA54kSZIktYQBT5IkSZJawoAnSZIkSS1hwJMkDU2SF5Os6fo3ZzNjn9oCx1ua5IHmWHcmOfgV7OPiJPObx381Zt33X22NzX42vi53J/nXJK+dYPy+SY7cEseWJI02vyZBkjQ0SZ6qql229NjN7GMp8G9VdUWSw4DPVdU+r2J/r7qmifab5DLgvqr69GbGnwwsqKpTt3QtkqTR4hk8SdJWI8kuSa5tzq7dlWTxOGNen2RV1xmuQ5rlhyW5udn2m0kmCl6rgL2abc9o9nV3ktObZTsnuSrJD5rlxzfLb0iyIMnfADs2dXy9WfdU8/+yJEd11bw0yXFJpiQ5P8ntSdYm+ZMeXpabgd2a/RzY/Iyrk3w/yW8kmQacAxzf1HJ8U/ulSW5rxr7sdZQktdPUYRcgSdqm7ZhkTfP4AeBdwDFV9WSSmcAtSVbU/59u8sfAyqr6dJIpwE7N2LOARVX1iyR/AZxBJ/hsyh8AdyU5AHg/8FYgwK1JbgTeCPykqo4CSDK9e+OqOjPJqVW17zj7Xg68G7iqCWALgQ8BS4ANVfXbSbYHvpfkO1X1wHgFNj/fQuCSZtEPgUOq6oUki4Bzq+rYJGfTdQYvybnAdVX1gWZ6521JvltVv9jM6yFJagEDniRpmJ7pDkhJtgPOTfJ24CU6Z65eB/xX1za3A5c2Y6+sqjVJDgXm0wlMANPonPkaz/lJzgIepRO4FgLf2hh+kvwLcAjw78Dnk5xHZ1rnTZP4ub4NfLEJcYcDq6rqmWZa6D5JjmvGTQfm0gm33TYG392A/wCu6Rp/WZK5QAHbbeL4hwFHJ/nz5vkOwOxmX5KkFjPgSZK2Ju8BZgEHVNXzSX5EJ5z8n6pa1QTAo4ClSS4AngCuqaoTezjGx6rqio1Pkiwcb1BV3Zdkf+BI4FNJrq2qzZ0R7N722SQ3AL8HHA8s23g44LSqWjnBLp6pqn2T7ASsBP4U+BLwSeD6qjqmuSHNDZvYPsCxVXVvL/VKktrDa/AkSVuT6cAjTbh7J7Dn2AFJ9gR+VlVfBS4G9gduAd6WZOM1dTsnmdfjMW8C/jDJTkl2Bo4Bbkrya8DTVfWPwPnNccZ6vjmTOJ7ldKZ+bjwbCJ2w9qGN2ySZ1xxzXFX1NPBnwEeTTKXz+jzcrD65a+j/ALt2PV8JnJbmdGaS/TZ1DElSuxjwJElbk68DC5LcBbyPzjVnY70D+EGS1XTOjn2xqh6lE3guT7KWzvTMN/VywKq6E1gK3AbcClxcVauBvelcu7YG+DjwqXE2vwhYu/EmK2N8BzgU+G5VPdcsuxi4B7gzyd3AV5hgNk1Ty1rgROCzwGean717u+uB+RtvskLnTN92TW3rmueSpG2AX5MgSZIkSS3hGTxJkiRJagkDniRJkiS1hAFPkiRJklrCgCdJkiRJLWHAkyRJkqSWMOBJkiRJUksY8CRJkiSpJf4Xbaov6UCbhkQAAAAASUVORK5CYII=\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["Statistical Evaluation"],"metadata":{"id":"RBufAMa4jf5z"}},{"cell_type":"code","source":["m2_evaluation = evaluation(m2_results[0])"],"metadata":{"id":"eTEwbX4Wjf50","executionInfo":{"status":"ok","timestamp":1666355520147,"user_tz":-120,"elapsed":301,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":194,"outputs":[]},{"cell_type":"code","source":["print_eval(m2_evaluation)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"gDeiUP-KUvwZ","executionInfo":{"status":"ok","timestamp":1666355521188,"user_tz":-120,"elapsed":5,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"bd7204e2-d91a-47b8-d638-3224c2ac1524"},"execution_count":195,"outputs":[{"output_type":"stream","name":"stdout","text":["---------------------------\n","True Positive =   85\n","False Positive =  20\n","False Negative =  17\n","True Negative =  178\n","Actual Positive =  102\n","Actual Negative =  198\n","Precision = 0.8095238095238095\n","False Discovery Rate =  0.19047619047619047\n","False Omission Rate =  0.08717948717948718\n","Negative Predictive Value =  0.8333333333333334\n","Recall =  0.10101010101010101\n","False Positive Rate =  0.16666666666666666\n","True Negative Rate =  0.898989898989899\n"]}]},{"cell_type":"markdown","source":["Here we also see a accuracy of 0.8766666666666667, which means that regardless of the model, this feature set is a good predictor. So now let's compare this feature set to the descriptor feature set"],"metadata":{"id":"C1XuCbKgagjq"}},{"cell_type":"markdown","source":["## **Model 3: Choice Model - Choice Descriptors**"],"metadata":{"id":"SYGKR6Gu0aU3"}},{"cell_type":"markdown","source":["### 1. Create Dataset\n","Now we will use the same model as M2, but then use 4 descriptors to compare the difference in results of the best indicators with descriptors\n","\n","First let's create the dataset our of the descriptor dataset"],"metadata":{"id":"SLFSQyCe00FV"}},{"cell_type":"code","source":["m3_df = des_df\n"],"metadata":{"id":"JRETqaa100FV","executionInfo":{"status":"ok","timestamp":1666355523658,"user_tz":-120,"elapsed":229,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":196,"outputs":[]},{"cell_type":"markdown","source":["2. Feature selection\n","\n","As we used 4 indicators and there are only 4 descriptors, we do not drop any descriptors"],"metadata":{"id":"3sv9200-00FV"}},{"cell_type":"code","source":["to_drop_des = [] #keep all descriptors\n","m3_df = m3_df.drop(to_drop_des, axis=1)"],"metadata":{"id":"1DvJ62LogscP","executionInfo":{"status":"ok","timestamp":1666355542584,"user_tz":-120,"elapsed":414,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":198,"outputs":[]},{"cell_type":"markdown","source":["3. Data Preprocessing\n","\n","Now we preprocess our data"],"metadata":{"id":"Qs_KEYt100FW"}},{"cell_type":"code","source":["m3_df = preprocess(m3_df)\n","m3_df"],"metadata":{"id":"QOMjvGU500FW","executionInfo":{"status":"ok","timestamp":1666355544840,"user_tz":-120,"elapsed":301,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"colab":{"base_uri":"https://localhost:8080/","height":488},"outputId":"66f50eb8-afb0-4ade-ddd7-6257dceb4bd9"},"execution_count":199,"outputs":[{"output_type":"execute_result","data":{"text/plain":["      decision  age  gender_female  gender_male  gender_other  \\\n","1000         0   21              1            0             0   \n","1001         0   27              1            0             0   \n","1002         1   25              1            0             0   \n","1003         0   28              1            0             0   \n","1004         1   30              0            1             0   \n","...        ...  ...            ...          ...           ...   \n","1995         1   28              0            1             0   \n","1996         0   23              1            0             0   \n","1997         1   26              0            1             0   \n","1998         0   23              1            0             0   \n","1999         0   27              1            0             0   \n","\n","      nationality_Belgian  nationality_Dutch  nationality_German  sport_Chess  \\\n","1000                    0                  1                   0            0   \n","1001                    0                  1                   0            1   \n","1002                    0                  1                   0            0   \n","1003                    0                  1                   0            1   \n","1004                    0                  1                   0            0   \n","...                   ...                ...                 ...          ...   \n","1995                    0                  1                   0            0   \n","1996                    1                  0                   0            0   \n","1997                    0                  1                   0            0   \n","1998                    0                  0                   1            0   \n","1999                    0                  1                   0            0   \n","\n","      sport_Cricket  sport_Football  sport_Golf  sport_Rugby  sport_Running  \\\n","1000              0               0           0            0              0   \n","1001              0               0           0            0              0   \n","1002              0               0           0            1              0   \n","1003              0               0           0            0              0   \n","1004              0               0           0            1              0   \n","...             ...             ...         ...          ...            ...   \n","1995              0               0           0            0              0   \n","1996              1               0           0            0              0   \n","1997              0               1           0            0              0   \n","1998              0               0           0            0              1   \n","1999              0               0           0            0              1   \n","\n","      sport_Swimming  sport_Tennis  \n","1000               1             0  \n","1001               0             0  \n","1002               0             0  \n","1003               0             0  \n","1004               0             0  \n","...              ...           ...  \n","1995               0             1  \n","1996               0             0  \n","1997               0             0  \n","1998               0             0  \n","1999               0             0  \n","\n","[1000 rows x 16 columns]"],"text/html":["\n","  <div id=\"df-3b2ee0bd-333a-4dcb-90c8-ae60e8610531\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>decision</th>\n","      <th>age</th>\n","      <th>gender_female</th>\n","      <th>gender_male</th>\n","      <th>gender_other</th>\n","      <th>nationality_Belgian</th>\n","      <th>nationality_Dutch</th>\n","      <th>nationality_German</th>\n","      <th>sport_Chess</th>\n","      <th>sport_Cricket</th>\n","      <th>sport_Football</th>\n","      <th>sport_Golf</th>\n","      <th>sport_Rugby</th>\n","      <th>sport_Running</th>\n","      <th>sport_Swimming</th>\n","      <th>sport_Tennis</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>1000</th>\n","      <td>0</td>\n","      <td>21</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","    </tr>\n","    <tr>\n","      <th>1001</th>\n","      <td>0</td>\n","      <td>27</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","    </tr>\n","    <tr>\n","      <th>1002</th>\n","      <td>1</td>\n","      <td>25</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","    </tr>\n","    <tr>\n","      <th>1003</th>\n","      <td>0</td>\n","      <td>28</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","    </tr>\n","    <tr>\n","      <th>1004</th>\n","      <td>1</td>\n","      <td>30</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>1995</th>\n","      <td>1</td>\n","      <td>28</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","    </tr>\n","    <tr>\n","      <th>1996</th>\n","      <td>0</td>\n","      <td>23</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","    </tr>\n","    <tr>\n","      <th>1997</th>\n","      <td>1</td>\n","      <td>26</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","    </tr>\n","    <tr>\n","      <th>1998</th>\n","      <td>0</td>\n","      <td>23</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","    </tr>\n","    <tr>\n","      <th>1999</th>\n","      <td>0</td>\n","      <td>27</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>1000 rows × 16 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-3b2ee0bd-333a-4dcb-90c8-ae60e8610531')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-3b2ee0bd-333a-4dcb-90c8-ae60e8610531 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-3b2ee0bd-333a-4dcb-90c8-ae60e8610531');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":199}]},{"cell_type":"markdown","source":["4. Classifier\n","\n","\n","Then we run the classifier according to the hyperparams from M2 to ensure we have the same model"],"metadata":{"id":"olcaG4ND00FW"}},{"cell_type":"code","source":["m3_results = classifier(SVBest(), m3_df)\n","m3_csv = pd.merge(m3_results[1], init_df['Id'], left_index=True, right_index=True).to_csv('m3.csv')\n"],"metadata":{"id":"9eTdkFkH00FW","colab":{"base_uri":"https://localhost:8080/","height":1000},"executionInfo":{"status":"ok","timestamp":1666355548233,"user_tz":-120,"elapsed":841,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"c26a8c94-5af8-4c77-8907-a6b24c7ce2f3"},"execution_count":200,"outputs":[{"output_type":"stream","name":"stdout","text":["-----------------------\n","Training results\n","Accuracy:\n","0.7828571428571428\n","Validation results\n","Accuracy:\n","0.7466666666666667\n","Confusion matrix:\n"]},{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.7/dist-packages/sklearn/utils/deprecation.py:87: FutureWarning: Function plot_confusion_matrix is deprecated; Function `plot_confusion_matrix` is deprecated in 1.0 and will be removed in 1.2. Use one of the class methods: ConfusionMatrixDisplay.from_predictions or ConfusionMatrixDisplay.from_estimator.\n","  warnings.warn(msg, category=FutureWarning)\n"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 720x432 with 2 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"stream","name":"stdout","text":["Random prediction: AUROC = 0.500\n","Model prediction: AUROC = 0.779\n"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 1080x720 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["Statistical Evaluation"],"metadata":{"id":"O3-ShegljzhD"}},{"cell_type":"code","source":["m3_evaluation = evaluation(m3_results[0])"],"metadata":{"id":"hNlr4AoZjzhE","executionInfo":{"status":"ok","timestamp":1666355559889,"user_tz":-120,"elapsed":241,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":209,"outputs":[]},{"cell_type":"code","source":["print_eval(m3_evaluation)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"bvdjc7rmGayp","executionInfo":{"status":"ok","timestamp":1666355561205,"user_tz":-120,"elapsed":198,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"e34137c8-8243-4827-a44d-3fde8ea294f0"},"execution_count":210,"outputs":[{"output_type":"stream","name":"stdout","text":["---------------------------\n","True Positive =   67\n","False Positive =  41\n","False Negative =  35\n","True Negative =  157\n","Actual Positive =  102\n","Actual Negative =  198\n","Precision = 0.6203703703703703\n","False Discovery Rate =  0.37962962962962965\n","False Omission Rate =  0.18229166666666666\n","Negative Predictive Value =  0.6568627450980392\n","Recall =  0.20707070707070707\n","False Positive Rate =  0.3431372549019608\n","True Negative Rate =  0.7929292929292929\n"]}]},{"cell_type":"markdown","source":["There we see that the accuracy of descriptors is lower, as was expected. "],"metadata":{"id":"qmYbnDjdax4e"}},{"cell_type":"markdown","source":["# **4. Result analysis**\n","\n","Now that we have our results and evaluations for each of the models, we can use this to create data visualizations/comparisons for our fairness analysis and reporting.\n"],"metadata":{"id":"Sh_tHpy6a7dW"}},{"cell_type":"code","source":["bias_df = ind_df\n","bias_df = preprocess(drop_features(set_results_RF[0][1]))\n","bias_df = pd.concat([bias_df, des_df.drop(['decision'], axis=1)], axis=1)\n","bias_df"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":424},"id":"nWguLi1T3-yN","executionInfo":{"status":"ok","timestamp":1666355563663,"user_tz":-120,"elapsed":311,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"bee2dfd1-40cb-4a1c-c2f6-46dd61efe7b0"},"execution_count":211,"outputs":[{"output_type":"execute_result","data":{"text/plain":["      ind_programming_exp  ind_international_exp  decision  \\\n","1000                    0                      0         0   \n","1001                    0                      1         0   \n","1002                    0                      0         1   \n","1003                    0                      1         0   \n","1004                    0                      0         1   \n","...                   ...                    ...       ...   \n","1995                    0                      0         1   \n","1996                    0                      0         0   \n","1997                    0                      0         1   \n","1998                    1                      0         0   \n","1999                    1                      0         0   \n","\n","      ind_university_grade  ind_languages  gender  age nationality     sport  \n","1000                    58              0  female   21       Dutch  Swimming  \n","1001                    67              0  female   27       Dutch     Chess  \n","1002                    56              2  female   25       Dutch     Rugby  \n","1003                    64              0  female   28       Dutch     Chess  \n","1004                    63              3    male   30       Dutch     Rugby  \n","...                    ...            ...     ...  ...         ...       ...  \n","1995                    71              2    male   28       Dutch    Tennis  \n","1996                    55              1  female   23     Belgian   Cricket  \n","1997                    58              2    male   26       Dutch  Football  \n","1998                    61              1  female   23      German   Running  \n","1999                    65              1  female   27       Dutch   Running  \n","\n","[1000 rows x 9 columns]"],"text/html":["\n","  <div id=\"df-0a116014-4ffb-44ec-bf6c-a2cecc57f7f6\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>ind_programming_exp</th>\n","      <th>ind_international_exp</th>\n","      <th>decision</th>\n","      <th>ind_university_grade</th>\n","      <th>ind_languages</th>\n","      <th>gender</th>\n","      <th>age</th>\n","      <th>nationality</th>\n","      <th>sport</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>1000</th>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>58</td>\n","      <td>0</td>\n","      <td>female</td>\n","      <td>21</td>\n","      <td>Dutch</td>\n","      <td>Swimming</td>\n","    </tr>\n","    <tr>\n","      <th>1001</th>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>67</td>\n","      <td>0</td>\n","      <td>female</td>\n","      <td>27</td>\n","      <td>Dutch</td>\n","      <td>Chess</td>\n","    </tr>\n","    <tr>\n","      <th>1002</th>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>56</td>\n","      <td>2</td>\n","      <td>female</td>\n","      <td>25</td>\n","      <td>Dutch</td>\n","      <td>Rugby</td>\n","    </tr>\n","    <tr>\n","      <th>1003</th>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>64</td>\n","      <td>0</td>\n","      <td>female</td>\n","      <td>28</td>\n","      <td>Dutch</td>\n","      <td>Chess</td>\n","    </tr>\n","    <tr>\n","      <th>1004</th>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>63</td>\n","      <td>3</td>\n","      <td>male</td>\n","      <td>30</td>\n","      <td>Dutch</td>\n","      <td>Rugby</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>1995</th>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>71</td>\n","      <td>2</td>\n","      <td>male</td>\n","      <td>28</td>\n","      <td>Dutch</td>\n","      <td>Tennis</td>\n","    </tr>\n","    <tr>\n","      <th>1996</th>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>55</td>\n","      <td>1</td>\n","      <td>female</td>\n","      <td>23</td>\n","      <td>Belgian</td>\n","      <td>Cricket</td>\n","    </tr>\n","    <tr>\n","      <th>1997</th>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>1</td>\n","      <td>58</td>\n","      <td>2</td>\n","      <td>male</td>\n","      <td>26</td>\n","      <td>Dutch</td>\n","      <td>Football</td>\n","    </tr>\n","    <tr>\n","      <th>1998</th>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>61</td>\n","      <td>1</td>\n","      <td>female</td>\n","      <td>23</td>\n","      <td>German</td>\n","      <td>Running</td>\n","    </tr>\n","    <tr>\n","      <th>1999</th>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>0</td>\n","      <td>65</td>\n","      <td>1</td>\n","      <td>female</td>\n","      <td>27</td>\n","      <td>Dutch</td>\n","      <td>Running</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>1000 rows × 9 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-0a116014-4ffb-44ec-bf6c-a2cecc57f7f6')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-0a116014-4ffb-44ec-bf6c-a2cecc57f7f6 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-0a116014-4ffb-44ec-bf6c-a2cecc57f7f6');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":211}]},{"cell_type":"code","source":["def bias_classifier(model, df, drop=['decision'], target=['decision'], frac=0.3):\n","    # Training set is a sample of 30%. Seed is used to get consistent results while testing\n","\n","\n","    val = df.sample(frac=frac,random_state=200)\n","    # training is set acquired by dropping the validation set\n","    train = df.drop(val.index)\n","\n","    x = train.drop(drop, axis=1)\n","    y = np.ravel(train[target].values) # transforming the array to a 1D array\n","    \n","    clas = model.fit(x, y)\n","\n","    valx = val.drop(drop, axis=1)\n","    valy = val[target].values\n"," \n","    return clas.score(valx, valy), confusion_matrix(valy, clas.predict(valx))\n"],"metadata":{"id":"frZoB5c_MabI","executionInfo":{"status":"ok","timestamp":1666355563890,"user_tz":-120,"elapsed":4,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":212,"outputs":[]},{"cell_type":"code","source":["# Model 1 and 2\n","# genders\n","score_male = bias_df[bias_df['gender'] == 'male'].drop(descriptor, axis=1)\n","score_female = bias_df[bias_df['gender'] == 'female'].drop(descriptor, axis=1)\n","score_other = bias_df[bias_df['gender'] == 'other'].drop(descriptor, axis=1)\n","\n","# nationalities\n","score_dutch =  bias_df[bias_df['nationality'] == 'dutch'].drop(descriptor, axis=1)\n","score_german = bias_df[bias_df['nationality'] == 'german'].drop(descriptor, axis=1)\n","score_belgian = bias_df[bias_df['nationality'] == 'belgian'].drop(descriptor, axis=1)\n","\n","# sports\n","score_chess = bias_df[bias_df['sport'] == 'chess'].drop(descriptor, axis=1)\n","score_cricket = bias_df[bias_df['sport'] == 'cricket'].drop(descriptor, axis=1)\n","score_football =bias_df[bias_df['sport'] == 'football'].drop(descriptor, axis=1)\n","score_golf = bias_df[bias_df['sport'] == 'golf'].drop(descriptor, axis=1)\n","score_rugby = bias_df[bias_df['sport'] == 'rugby'].drop(descriptor, axis=1)\n","score_running = bias_df[bias_df['sport'] == 'running'].drop(descriptor, axis=1)\n","score_swimming = bias_df[bias_df['sport'] == 'swimming'].drop(descriptor, axis=1)\n","score_tennis = bias_df[bias_df['sport'] == 'tennis'].drop(descriptor, axis=1)\n","\n","# age groups\n","score_age_young = bias_df[bias_df['age'] < 26].drop(descriptor, axis=1)\n","score_age_old = bias_df[bias_df['age'] >= 26].drop(descriptor, axis=1)"],"metadata":{"id":"2QQgyleuOD5L","executionInfo":{"status":"ok","timestamp":1666355563891,"user_tz":-120,"elapsed":4,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":213,"outputs":[]},{"cell_type":"code","source":["# Model 3\n","\n","# gender\n","score_male = bias_classifier(SVBest(),m3_df[(m3_df['gender_male'] == 1 )])\n","score_female = bias_classifier(SVBest(),m3_df[(m3_df['gender_female'] == 1 )])\n","score_other = bias_classifier(SVBest(),m3_df[(m3_df['gender_other'] == 1 )])\n","\n","#nationality\n","score_dutch = bias_classifier(SVBest(),m3_df[(m3_df['nationality_Dutch'] == 1)])\n","score_belgian = bias_classifier(SVBest(),m3_df[(m3_df['nationality_Belgian'] == 1)])\n","score_german = bias_classifier(SVBest(),m3_df[(m3_df['nationality_German'] == 1)])\n","\n","\n","# gender-age\n","score_male_old = bias_classifier(SVBest(),m3_df[(m3_df['gender_male'] == 1 ) & (m3_df['age'] >= 26)])\n","score_male_young = bias_classifier(SVBest(),m3_df[(m3_df['gender_male'] == 1 ) & (m3_df['age'] < 26)])\n","\n","score_female_old = bias_classifier(SVBest(),m3_df[(m3_df['gender_female'] == 1 ) & (m3_df['age'] >= 26)])\n","score_female_young = bias_classifier(SVBest(),m3_df[(m3_df['gender_female'] == 1 ) & (m3_df['age'] < 26)])\n","\n","# score_other_old = bias_classifier(SVBest(),m3_df[(m3_df['gender_other'] == 1 ) & (m3_df['age'] >= 26)]) Not applicable\n","score_other_young = bias_classifier(SVBest(),m3_df[(m3_df['gender_other'] == 1 ) & (m3_df['age'] < 26)])\n","\n","# gender-nationality\n","score_male_dutch = bias_classifier(SVBest(),m3_df[(m3_df['gender_male'] == 1 ) & (m3_df['nationality_Dutch'] == 1)])\n","score_male_belgian = bias_classifier(SVBest(),m3_df[(m3_df['gender_male'] == 1 ) & (m3_df['nationality_Belgian'] == 1)])\n","score_male_german = bias_classifier(SVBest(),m3_df[(m3_df['gender_male'] == 1 ) & (m3_df['nationality_German'] == 1)])\n","\n","score_female_dutch = bias_classifier(SVBest(),m3_df[(m3_df['gender_female'] == 1 ) & (m3_df['nationality_Dutch'] == 1)])\n","score_female_belgian = bias_classifier(SVBest(),m3_df[(m3_df['gender_female'] == 1 ) & (m3_df['nationality_Belgian'] == 1)])\n","score_female_german = bias_classifier(SVBest(),m3_df[(m3_df['gender_female'] == 1 ) & (m3_df['nationality_German'] == 1)])\n","\n","score_other_dutch = bias_classifier(SVBest(),m3_df[(m3_df['gender_other'] == 1 ) & (m3_df['nationality_Dutch'] == 1)])\n","# score_other_belgian = bias_classifier(SVBest(),m3_df[(m3_df['gender_other'] == 1 ) & (m3_df['nationality_Belgian'] == 1)]) NON EXISTENT\n","# score_other_german = bias_classifier(SVBest(),m3_df[(m3_df['gender_other'] == 1 ) & (m3_df['nationality_German'] == 1)]) NON EXISTENT\n","\n","# age-nationality\n","score_young_dutch = bias_classifier(SVBest(),m3_df[(m3_df['age'] < 26) & (m3_df['nationality_Dutch'] == 1)])\n","score_young_belgian = bias_classifier(SVBest(),m3_df[(m3_df['age'] < 26) & (m3_df['nationality_Belgian'] == 1)])\n","score_young_german = bias_classifier(SVBest(),m3_df[(m3_df['age'] < 26) & (m3_df['nationality_German'] == 1)])\n","\n","score_old_dutch = bias_classifier(SVBest(),m3_df[(m3_df['age'] >= 26) & (m3_df['nationality_Dutch'] == 1)])\n","score_old_belgian = bias_classifier(SVBest(),m3_df[(m3_df['age'] >= 26)& (m3_df['nationality_Belgian'] == 1)])\n","score_old_german = bias_classifier(SVBest(),m3_df[(m3_df['age'] >= 26) & (m3_df['nationality_German'] == 1)])"],"metadata":{"id":"RqfnfpfeE-br","executionInfo":{"status":"ok","timestamp":1666355564776,"user_tz":-120,"elapsed":888,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":214,"outputs":[]},{"cell_type":"code","source":["def bias_eval(results):\n","  true_positive = results[1,1]\n","  false_positive = results[0,1]\n","  false_negative = results[1,0]\n","  true_negative = results[0,0]\n","  actual_positive = true_positive + false_negative\n","  precision = true_positive / (true_positive + false_positive)\n","  recall = true_positive / actual_positive\n","  return precision,recall\n","\n"],"metadata":{"id":"TxR7OqxPJVLT","executionInfo":{"status":"ok","timestamp":1666355564779,"user_tz":-120,"elapsed":10,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}}},"execution_count":215,"outputs":[]},{"cell_type":"code","source":["bias = pd.DataFrame(\n","    {\n","    'male':[score_male[0], bias_eval(score_male[1])[0], bias_eval(score_male[1])[1]],\n","    'other':[score_other[0], bias_eval(score_other[1])[0], bias_eval(score_other[1])[1]],\n","    'female':[score_female[0], bias_eval(score_female[1])[0], bias_eval(score_female[1])[1]],\n","    'dutch':[score_dutch[0], bias_eval(score_dutch[1])[0], bias_eval(score_dutch[1])[1]],\n","    'belgian':[score_belgian[0], bias_eval(score_belgian[1])[0], bias_eval(score_belgian[1])[1]],\n","    'german':[score_german[0], bias_eval(score_german[1])[0], bias_eval(score_german[1])[1]],\n","    'male_old':[score_male_old[0], bias_eval(score_male_old[1])[0], bias_eval(score_male_old[1])[1]],\n","    'male_young':[score_male_young[0], bias_eval(score_male_young[1])[0], bias_eval(score_male_young[1])[1]],\n","    'female_old':[score_female_old[0], bias_eval(score_female_old[1])[0], bias_eval(score_female_old[1])[1]],\n","    'female_young':[score_female_young[0], bias_eval(score_female_young[1])[0], bias_eval(score_female_young[1])[1]],\n","    'other_young':[score_other_young[0], bias_eval(score_other_young[1])[0], bias_eval(score_other_young[1])[1]],\n","    'male_dutch':[score_male_dutch[0], bias_eval(score_male_dutch[1])[0], bias_eval(score_male_dutch[1])[1]],\n","    'male_belgian':[score_male_belgian[0], bias_eval(score_male_belgian[1])[0], bias_eval(score_male_belgian[1])[1]],\n","    'male_german':[score_male_german[0], bias_eval(score_male_german[1])[0], bias_eval(score_male_german[1])[1]],\n","    'female_dutch':[score_female_dutch[0], bias_eval(score_female_dutch[1])[0], bias_eval(score_female_dutch[1])[1]],\n","    'female_belgian':[score_female_belgian[0], bias_eval(score_female_belgian[1])[0], bias_eval(score_female_belgian[1])[1]],\n","    'female_german':[score_female_german[0], bias_eval(score_female_german[1])[0], bias_eval(score_female_german[1])[1]],\n","     'other_dutch':[score_other_dutch[0], bias_eval(score_other_dutch[1])[0], bias_eval(score_other_dutch[1])[1]],\n","     'young_dutch':[score_young_dutch[0], bias_eval(score_young_dutch[1])[0], bias_eval(score_young_dutch[1])[1]],\n","     'young_belgian':[score_young_belgian[0], bias_eval(score_young_belgian[1])[0], bias_eval(score_young_belgian[1])[1]],\n","     'young_german':[score_young_german[0], bias_eval(score_young_german[1])[0], bias_eval(score_young_german[1])[1]],\n","     'old_dutch':[score_old_dutch[0], bias_eval(score_old_dutch[1])[0], bias_eval(score_old_dutch[1])[1]],\n","     'old_belgian':[score_old_belgian[0], bias_eval(score_old_belgian[1])[0], bias_eval(score_old_belgian[1])[1]],\n","     'old_german':[score_old_german[0], bias_eval(score_old_german[1])[0], bias_eval(score_old_german[1])[1]]\n","     }\n","     )\n","\n","\n","df_bias = pd.DataFrame({\n","    'subgroups': bias.columns,\n","    'accuracy':bias.iloc[0].values,\n","    'precision':bias.iloc[1].values,\n","    'recall':bias.iloc[2].values}).fillna(0)\n","\n","df_bias"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":837},"id":"KV87loy9HgfC","executionInfo":{"status":"ok","timestamp":1666355564779,"user_tz":-120,"elapsed":9,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"955b9300-6b6d-4ccc-86f1-32f05f8c67a4"},"execution_count":216,"outputs":[{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:7: RuntimeWarning: invalid value encountered in long_scalars\n","  import sys\n"]},{"output_type":"execute_result","data":{"text/plain":["         subgroups  accuracy  precision    recall\n","0             male  0.704403   0.607143  0.957746\n","1            other  0.666667   0.000000  0.000000\n","2           female  0.844444   0.000000  0.000000\n","3            dutch  0.771784   0.716667  0.530864\n","4          belgian  0.727273   1.000000  0.181818\n","5           german  0.851852   0.000000  0.000000\n","6         male_old  0.648649   0.538462  0.795455\n","7       male_young  0.604167   0.558824  0.826087\n","8       female_old  0.811594   0.000000  0.000000\n","9     female_young  0.820896   0.000000  0.000000\n","10     other_young  0.000000   0.000000  0.000000\n","11      male_dutch  0.684615   0.602564  0.824561\n","12    male_belgian  0.466667   0.500000  0.125000\n","13     male_german  0.428571   0.200000  0.200000\n","14    female_dutch  0.801887   0.000000  0.000000\n","15  female_belgian  0.941176   0.000000  0.000000\n","16   female_german  0.750000   0.000000  0.000000\n","17     other_dutch  0.250000   0.000000  0.000000\n","18     young_dutch  0.842105   0.555556  0.588235\n","19   young_belgian  0.833333   0.000000  0.000000\n","20    young_german  0.500000   0.200000  1.000000\n","21       old_dutch  0.737931   0.657143  0.469388\n","22     old_belgian  0.600000   0.500000  0.125000\n","23      old_german  0.736842   0.000000  0.000000"],"text/html":["\n","  <div id=\"df-511f2ef4-9127-4741-807e-78516361f372\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>subgroups</th>\n","      <th>accuracy</th>\n","      <th>precision</th>\n","      <th>recall</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>male</td>\n","      <td>0.704403</td>\n","      <td>0.607143</td>\n","      <td>0.957746</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>other</td>\n","      <td>0.666667</td>\n","      <td>0.000000</td>\n","      <td>0.000000</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>female</td>\n","      <td>0.844444</td>\n","      <td>0.000000</td>\n","      <td>0.000000</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>dutch</td>\n","      <td>0.771784</td>\n","      <td>0.716667</td>\n","      <td>0.530864</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>belgian</td>\n","      <td>0.727273</td>\n","      <td>1.000000</td>\n","      <td>0.181818</td>\n","    </tr>\n","    <tr>\n","      <th>5</th>\n","      <td>german</td>\n","      <td>0.851852</td>\n","      <td>0.000000</td>\n","      <td>0.000000</td>\n","    </tr>\n","    <tr>\n","      <th>6</th>\n","      <td>male_old</td>\n","      <td>0.648649</td>\n","      <td>0.538462</td>\n","      <td>0.795455</td>\n","    </tr>\n","    <tr>\n","      <th>7</th>\n","      <td>male_young</td>\n","      <td>0.604167</td>\n","      <td>0.558824</td>\n","      <td>0.826087</td>\n","    </tr>\n","    <tr>\n","      <th>8</th>\n","      <td>female_old</td>\n","      <td>0.811594</td>\n","      <td>0.000000</td>\n","      <td>0.000000</td>\n","    </tr>\n","    <tr>\n","      <th>9</th>\n","      <td>female_young</td>\n","      <td>0.820896</td>\n","      <td>0.000000</td>\n","      <td>0.000000</td>\n","    </tr>\n","    <tr>\n","      <th>10</th>\n","      <td>other_young</td>\n","      <td>0.000000</td>\n","      <td>0.000000</td>\n","      <td>0.000000</td>\n","    </tr>\n","    <tr>\n","      <th>11</th>\n","      <td>male_dutch</td>\n","      <td>0.684615</td>\n","      <td>0.602564</td>\n","      <td>0.824561</td>\n","    </tr>\n","    <tr>\n","      <th>12</th>\n","      <td>male_belgian</td>\n","      <td>0.466667</td>\n","      <td>0.500000</td>\n","      <td>0.125000</td>\n","    </tr>\n","    <tr>\n","      <th>13</th>\n","      <td>male_german</td>\n","      <td>0.428571</td>\n","      <td>0.200000</td>\n","      <td>0.200000</td>\n","    </tr>\n","    <tr>\n","      <th>14</th>\n","      <td>female_dutch</td>\n","      <td>0.801887</td>\n","      <td>0.000000</td>\n","      <td>0.000000</td>\n","    </tr>\n","    <tr>\n","      <th>15</th>\n","      <td>female_belgian</td>\n","      <td>0.941176</td>\n","      <td>0.000000</td>\n","      <td>0.000000</td>\n","    </tr>\n","    <tr>\n","      <th>16</th>\n","      <td>female_german</td>\n","      <td>0.750000</td>\n","      <td>0.000000</td>\n","      <td>0.000000</td>\n","    </tr>\n","    <tr>\n","      <th>17</th>\n","      <td>other_dutch</td>\n","      <td>0.250000</td>\n","      <td>0.000000</td>\n","      <td>0.000000</td>\n","    </tr>\n","    <tr>\n","      <th>18</th>\n","      <td>young_dutch</td>\n","      <td>0.842105</td>\n","      <td>0.555556</td>\n","      <td>0.588235</td>\n","    </tr>\n","    <tr>\n","      <th>19</th>\n","      <td>young_belgian</td>\n","      <td>0.833333</td>\n","      <td>0.000000</td>\n","      <td>0.000000</td>\n","    </tr>\n","    <tr>\n","      <th>20</th>\n","      <td>young_german</td>\n","      <td>0.500000</td>\n","      <td>0.200000</td>\n","      <td>1.000000</td>\n","    </tr>\n","    <tr>\n","      <th>21</th>\n","      <td>old_dutch</td>\n","      <td>0.737931</td>\n","      <td>0.657143</td>\n","      <td>0.469388</td>\n","    </tr>\n","    <tr>\n","      <th>22</th>\n","      <td>old_belgian</td>\n","      <td>0.600000</td>\n","      <td>0.500000</td>\n","      <td>0.125000</td>\n","    </tr>\n","    <tr>\n","      <th>23</th>\n","      <td>old_german</td>\n","      <td>0.736842</td>\n","      <td>0.000000</td>\n","      <td>0.000000</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-511f2ef4-9127-4741-807e-78516361f372')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-511f2ef4-9127-4741-807e-78516361f372 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-511f2ef4-9127-4741-807e-78516361f372');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":216}]},{"cell_type":"code","source":["df_bias[['subgroups', 'accuracy', 'precision', 'recall']].plot(x='subgroups', kind='bar', figsize=(30,7), color=['green','red','blue'],width=0.5)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":405},"id":"lHlHtPSOTxyG","executionInfo":{"status":"ok","timestamp":1666355565485,"user_tz":-120,"elapsed":713,"user":{"displayName":"Varun Fuego","userId":"17764027894283918060"}},"outputId":"6171f5ea-5bf5-42a8-c010-193c133b67ac"},"execution_count":217,"outputs":[{"output_type":"execute_result","data":{"text/plain":["<matplotlib.axes._subplots.AxesSubplot at 0x7f453386e1d0>"]},"metadata":{},"execution_count":217},{"output_type":"display_data","data":{"text/plain":["<Figure size 2160x504 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]}]}