Income_Pred.ipynb

{"cells":[{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"# Data Analysis\nimport pandas as pd\nimport numpy as np\n\n# Visualization\nfrom matplotlib import pyplot as plt\nimport seaborn as sns\n\n# Data Prediction\nfrom sklearn.preprocessing import LabelEncoder\nimport xgboost\nfrom sklearn.metrics import mean_squared_error\n\n# Ignore warnings\nimport warnings\nwarnings.filterwarnings(\"ignore\")\n\nimport os\nprint(os.listdir(\"../input/inc-wlabels\"))","execution_count":199,"outputs":[{"output_type":"stream","text":"['inc_wlabels.csv', 'inc_wo_labels.csv']\n","name":"stdout"}]},{"metadata":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","trusted":true},"cell_type":"code","source":"# Importing the dataset\ndataset = pd.read_csv('../input/inc-wlabels/inc_wlabels.csv') # 111993\ndt = pd.read_csv('../input/inc-wlabels/inc_wo_labels.csv') #73230\n\n# Remove negative incomes\n# dataset.drop(dataset.loc[dataset['Income in EUR']<0].index, inplace=True)\n\norg_dt = dataset # 111827\n# print(dataset.shape, dt.shape)\ndesc_X = dataset.describe()\n\ny = org_dt.iloc[:,11].values.reshape(-1,1) # 111827\n# y = np.log(y)\n\n# Merge both labelled and unlabelles so they may be processed together\ndataset = pd.concat([dataset,dt],axis='rows')\nprint(dataset.shape)\nprint(org_dt.shape)\nprint(y.shape)\ndesc_X\ny","execution_count":200,"outputs":[{"output_type":"stream","text":"(185223, 12)\n(111993, 12)\n(111993, 1)\n","name":"stdout"},{"output_type":"execute_result","execution_count":200,"data":{"text/plain":"array([[ 61031.94416],\n       [ 91001.32764],\n       [157982.1767 ],\n       ...,\n       [289951.3294 ],\n       [100046.5278 ],\n       [145886.2885 ]])"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"dataset.nunique()","execution_count":201,"outputs":[{"output_type":"execute_result","execution_count":201,"data":{"text/plain":"Instance             185223\nYear of Record           40\nGender                    5\nAge                     102\nCountry                 166\nSize of City         165260\nProfession             1351\nUniversity Degree         5\nWears Glasses             2\nHair Color                6\nBody Height [cm]        156\nIncome in EUR        111991\ndtype: int64"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"## Null Counting Fucntion\ndef null_values(df):\n    \n    sum_null = df.isnull().sum()\n    total = df.isnull().count()\n    percent_nullvalues = 100* sum_null / total \n    df_null = pd.DataFrame()\n    df_null['Total'] = total\n    df_null['Null_Count'] = sum_null\n    df_null['Percent'] = round(percent_nullvalues,2)\n    df_null = df_null.sort_values(by='Null_Count',ascending = False)\n    df_null = df_null[df_null.Null_Count > 0]\n    \n    return(df_null)\n\nprint(null_values(org_dt))\nnull_values(dataset)","execution_count":202,"outputs":[{"output_type":"stream","text":"                    Total  Null_Count  Percent\nGender             111993        7432     6.64\nUniversity Degree  111993        7370     6.58\nHair Color         111993        7242     6.47\nAge                111993         494     0.44\nYear of Record     111993         441     0.39\nProfession         111993         322     0.29\n","name":"stdout"},{"output_type":"execute_result","execution_count":202,"data":{"text/plain":"                    Total  Null_Count  Percent\nIncome in EUR      185223       73230    39.54\nGender             185223       12294     6.64\nUniversity Degree  185223       12227     6.60\nHair Color         185223       12158     6.56\nAge                185223         773     0.42\nYear of Record     185223         736     0.40\nProfession         185223         517     0.28","text/html":"<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>Total</th>\n      <th>Null_Count</th>\n      <th>Percent</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <td>Income in EUR</td>\n      <td>185223</td>\n      <td>73230</td>\n      <td>39.54</td>\n    </tr>\n    <tr>\n      <td>Gender</td>\n      <td>185223</td>\n      <td>12294</td>\n      <td>6.64</td>\n    </tr>\n    <tr>\n      <td>University Degree</td>\n      <td>185223</td>\n      <td>12227</td>\n      <td>6.60</td>\n    </tr>\n    <tr>\n      <td>Hair Color</td>\n      <td>185223</td>\n      <td>12158</td>\n      <td>6.56</td>\n    </tr>\n    <tr>\n      <td>Age</td>\n      <td>185223</td>\n      <td>773</td>\n      <td>0.42</td>\n    </tr>\n    <tr>\n      <td>Year of Record</td>\n      <td>185223</td>\n      <td>736</td>\n      <td>0.40</td>\n    </tr>\n    <tr>\n      <td>Profession</td>\n      <td>185223</td>\n      <td>517</td>\n      <td>0.28</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"null_values(dataset)\n\n# Fill missing values\ndataset = dataset.fillna({\n    'University Degree':'No',\n    'Gender':'unknown',\n    'Hair Color':'None',\n    'Profession':'None',\n    'Age': org_dt['Age'].mean(),\n    'Year of Record': org_dt['Year of Record'].mean()\n})\n\ndataset['Gender'] = dataset['Gender'].replace({'0': 'unknown'})\ndataset['Hair Color'] = dataset['Hair Color'].replace({'0': 'Unknown'})\ndataset['University Degree'] = dataset['University Degree'].replace({'0': 'No'})\n\ndataset.columns = ['Instance', 'Year of Record', 'Gender','Age','Country','Size of City','Profession','University Degree','Wears Glasses','Hair Color','Body Height cm','Income in EUR']\ny = pd.DataFrame.from_records(y)\n\nnull_values(dataset)","execution_count":203,"outputs":[{"output_type":"execute_result","execution_count":203,"data":{"text/plain":"                Total  Null_Count  Percent\nIncome in EUR  185223       73230    39.54","text/html":"<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>Total</th>\n      <th>Null_Count</th>\n      <th>Percent</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <td>Income in EUR</td>\n      <td>185223</td>\n      <td>73230</td>\n      <td>39.54</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"comm_profess = pd.Series()\ncomm_profess['comm_profess'] = pd.Series(' '.join(dataset['Profession']).lower().split()).value_counts()[:500].index.tolist()\ncomm_profess['comm_profess'].remove('and')\ncomm_profess['comm_profess'].remove('&')\ncomm_profess['comm_profess'].remove('-')\n\ndef replaceProf(x):\n    arr = x.split() \n    for p in range(0,len(comm_profess['comm_profess'])) :\n        for a in range(0,len(arr)):\n            if (comm_profess['comm_profess'][p]==arr[a]):\n                # print(x,comm_profess['comm_profess'][p])\n                x = comm_profess['comm_profess'][p]\n                return comm_profess['comm_profess'][p]\n            else:\n                x = 'worker'\n\n# Extracts the most common words from the Profession\n\n\"\"\"\ndataset['Profession'] = dataset['Profession'].apply(lambda x: replaceProf(x))\n# org_dt['Profession'].apply(lambda x: replaceProf(x))\ndataset['Profession'].to_csv('datprof.csv')\n\"\"\"\n","execution_count":204,"outputs":[{"output_type":"execute_result","execution_count":204,"data":{"text/plain":"\"\\ndataset['Profession'] = dataset['Profession'].apply(lambda x: replaceProf(x))\\n# org_dt['Profession'].apply(lambda x: replaceProf(x))\\ndataset['Profession'].to_csv('datprof.csv')\\n\""},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"dataset['Uni Degree'] = pd.Categorical(dataset['University Degree']).rename_categories({'No': 0, 'Bachelor': 1,'Master':2,'PhD':3})\ndataset['Uni Degree'].value_counts()\ndataset['Uni Degree'] = dataset['Uni Degree'].astype('int64')\ndataset['Uni Degree']\n","execution_count":205,"outputs":[{"output_type":"execute_result","execution_count":205,"data":{"text/plain":"0        1\n1        2\n2        3\n3        0\n4        0\n        ..\n73225    1\n73226    0\n73227    2\n73228    2\n73229    2\nName: Uni Degree, Length: 185223, dtype: int64"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Creating Dummy Variables\ndummy_gender = (pd.get_dummies(dataset['Gender'])).iloc[:, :-1] # 4 gender\ndummy_country = pd.get_dummies(dataset['Country']).iloc[:, :-1] #159 countries\ndummy_profession = pd.get_dummies(dataset['Profession']).iloc[:, :-1] # 1339 professions\n# dummy_degree = pd.get_dummies(dataset['University Degree']).iloc[:, :-1] # 4 degrees \ndummy_haircolor = pd.get_dummies(dataset['Hair Color']).iloc[:, :-1] # 5 hair\n\n\ndataset['CitySizeLog'] = np.log(dataset['Size of City'])\ndataset['IncomeLog'] = np.log(dataset['Income in EUR'])\n# dataset['Year of Record'] = dataset['Year of Record'].apply(lambda x: x - 1980)\ndataset['Year of Record']\n\n\"\"\"dummy_gender = (pd.get_dummies(dataset['Gender'])) # 4 gender\ndummy_country = pd.get_dummies(dataset['Country']) #159 countries\ndummy_profession = pd.get_dummies(dataset['Profession']) # 1339 professions\n# dummy_degree = pd.get_dummies(dataset['University Degree']).iloc[:, :-1] # 4 degrees \ndummy_haircolor = pd.get_dummies(dataset['Hair Color']) # 5 hair\"\"\"","execution_count":206,"outputs":[{"output_type":"execute_result","execution_count":206,"data":{"text/plain":"\"dummy_gender = (pd.get_dummies(dataset['Gender'])) # 4 gender\\ndummy_country = pd.get_dummies(dataset['Country']) #159 countries\\ndummy_profession = pd.get_dummies(dataset['Profession']) # 1339 professions\\n# dummy_degree = pd.get_dummies(dataset['University Degree']).iloc[:, :-1] # 4 degrees \\ndummy_haircolor = pd.get_dummies(dataset['Hair Color']) # 5 hair\""},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"org_dt['Country']","execution_count":207,"outputs":[{"output_type":"execute_result","execution_count":207,"data":{"text/plain":"0                      Belarus\n1                    Singapore\n2                       Norway\n3                         Cuba\n4         United Arab Emirates\n                  ...         \n111988                Honduras\n111989                 Tunisia\n111990               Sri Lanka\n111991                 Denmark\n111992      State of Palestine\nName: Country, Length: 111993, dtype: object"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"y.columns = ['Income']\ncountryAves = pd.concat([org_dt['Country'],\n                        org_dt['Size of City'],\n                        org_dt['Age'],\n                        y['Income']], axis=1)\n\nb = countryAves.groupby('Country').mean()\nb = b.rename({ 'Size of City':'CountryAveCitySize',\n               'Income':'CountryAveEarning',\n               'Age':'CountryAveAge'}, axis=1)\n\nb","execution_count":208,"outputs":[{"output_type":"execute_result","execution_count":208,"data":{"text/plain":"             CountryAveCitySize  CountryAveAge  CountryAveEarning\nCountry                                                          \nAfghanistan        6.373037e+05      39.285714       8.879485e+05\nAlbania            8.724576e+05      37.535176       3.093651e+05\nAlgeria            9.825891e+05      26.285714       5.836315e+05\nAngola             1.190360e+06      35.792453       6.436873e+05\nArgentina          1.019613e+06      30.833333       1.104589e+06\n...                         ...            ...                ...\nVenezuela          8.810847e+05      36.173469       5.409539e+05\nVietnam            9.138400e+04      31.000000       3.213682e+06\nYemen              1.002108e+06      36.551724       5.469064e+05\nZambia             8.646738e+05      37.023355       3.035599e+05\nZimbabwe           8.230789e+05      38.113953       1.815273e+05\n\n[160 rows x 3 columns]","text/html":"<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>CountryAveCitySize</th>\n      <th>CountryAveAge</th>\n      <th>CountryAveEarning</th>\n    </tr>\n    <tr>\n      <th>Country</th>\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <td>Afghanistan</td>\n      <td>6.373037e+05</td>\n      <td>39.285714</td>\n      <td>8.879485e+05</td>\n    </tr>\n    <tr>\n      <td>Albania</td>\n      <td>8.724576e+05</td>\n      <td>37.535176</td>\n      <td>3.093651e+05</td>\n    </tr>\n    <tr>\n      <td>Algeria</td>\n      <td>9.825891e+05</td>\n      <td>26.285714</td>\n      <td>5.836315e+05</td>\n    </tr>\n    <tr>\n      <td>Angola</td>\n      <td>1.190360e+06</td>\n      <td>35.792453</td>\n      <td>6.436873e+05</td>\n    </tr>\n    <tr>\n      <td>Argentina</td>\n      <td>1.019613e+06</td>\n      <td>30.833333</td>\n      <td>1.104589e+06</td>\n    </tr>\n    <tr>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <td>Venezuela</td>\n      <td>8.810847e+05</td>\n      <td>36.173469</td>\n      <td>5.409539e+05</td>\n    </tr>\n    <tr>\n      <td>Vietnam</td>\n      <td>9.138400e+04</td>\n      <td>31.000000</td>\n      <td>3.213682e+06</td>\n    </tr>\n    <tr>\n      <td>Yemen</td>\n      <td>1.002108e+06</td>\n      <td>36.551724</td>\n      <td>5.469064e+05</td>\n    </tr>\n    <tr>\n      <td>Zambia</td>\n      <td>8.646738e+05</td>\n      <td>37.023355</td>\n      <td>3.035599e+05</td>\n    </tr>\n    <tr>\n      <td>Zimbabwe</td>\n      <td>8.230789e+05</td>\n      <td>38.113953</td>\n      <td>1.815273e+05</td>\n    </tr>\n  </tbody>\n</table>\n<p>160 rows × 3 columns</p>\n</div>"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Empty spaces can be added in later by autoclean...\nX = org_dt['Country']\n#X = pd.merge(X, b, on='Profession', how='left', sort=False)\nX = X.reset_index().merge(b, on='Country', how=\"left\").set_index('index')\nX = X.drop('Country', axis=1)\nX","execution_count":209,"outputs":[{"output_type":"execute_result","execution_count":209,"data":{"text/plain":"        CountryAveCitySize  CountryAveAge  CountryAveEarning\nindex                                                       \n0            804670.998106      37.633159       33116.813053\n1            952551.623932      38.180011       56285.026965\n2            869125.012500      38.064686       87116.830844\n3            789873.203729      37.777547      100607.155049\n4            836528.687774      37.061125       42773.138452\n...                    ...            ...                ...\n111988       840913.833568      37.213712       40478.734970\n111989       841826.691016      37.655434      135724.673788\n111990       994180.887805      38.240196      429979.701959\n111991       867784.190106      37.725293       61135.778415\n111992       914492.440765      36.680678      100229.495759\n\n[111993 rows x 3 columns]","text/html":"<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>CountryAveCitySize</th>\n      <th>CountryAveAge</th>\n      <th>CountryAveEarning</th>\n    </tr>\n    <tr>\n      <th>index</th>\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <td>0</td>\n      <td>804670.998106</td>\n      <td>37.633159</td>\n      <td>33116.813053</td>\n    </tr>\n    <tr>\n      <td>1</td>\n      <td>952551.623932</td>\n      <td>38.180011</td>\n      <td>56285.026965</td>\n    </tr>\n    <tr>\n      <td>2</td>\n      <td>869125.012500</td>\n      <td>38.064686</td>\n      <td>87116.830844</td>\n    </tr>\n    <tr>\n      <td>3</td>\n      <td>789873.203729</td>\n      <td>37.777547</td>\n      <td>100607.155049</td>\n    </tr>\n    <tr>\n      <td>4</td>\n      <td>836528.687774</td>\n      <td>37.061125</td>\n      <td>42773.138452</td>\n    </tr>\n    <tr>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <td>111988</td>\n      <td>840913.833568</td>\n      <td>37.213712</td>\n      <td>40478.734970</td>\n    </tr>\n    <tr>\n      <td>111989</td>\n      <td>841826.691016</td>\n      <td>37.655434</td>\n      <td>135724.673788</td>\n    </tr>\n    <tr>\n      <td>111990</td>\n      <td>994180.887805</td>\n      <td>38.240196</td>\n      <td>429979.701959</td>\n    </tr>\n    <tr>\n      <td>111991</td>\n      <td>867784.190106</td>\n      <td>37.725293</td>\n      <td>61135.778415</td>\n    </tr>\n    <tr>\n      <td>111992</td>\n      <td>914492.440765</td>\n      <td>36.680678</td>\n      <td>100229.495759</td>\n    </tr>\n  </tbody>\n</table>\n<p>111993 rows × 3 columns</p>\n</div>"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"prof = org_dt['Profession'].str.lower()\nX = pd.concat([X, prof], axis=1)\nprofessionAves = pd.concat([org_dt['Profession'].str.lower(),\n                            org_dt['Size of City'],\n                            org_dt['Age'],\n                            X['CountryAveCitySize'],\n                            X['CountryAveEarning'],\n                            X['CountryAveAge'],\n                            y['Income']], axis=1)\n\nc = professionAves.groupby('Profession').mean().reset_index()\nc = c.rename({ 'Size of City':'ProfessionAveCitySize',\n               'Income':'ProfessionAveEarning',\n               'Age':'ProfessionAveAge',\n             'CountryAveAge':'CountryProfAveAge',\n             'CountryAveEarning':'CountryProfAveEarning',\n             'CountryAveAge':'CountryProfAveAge'}, axis=1)\n\nX = X.reset_index().merge(c, on='Profession', how=\"left\").set_index('index')\nX = X.drop('Profession', axis=1)\nX","execution_count":210,"outputs":[{"output_type":"execute_result","execution_count":210,"data":{"text/plain":"        CountryAveCitySize_x  CountryAveAge  CountryAveEarning  \\\nindex                                                            \n0              804670.998106      37.633159       33116.813053   \n1              952551.623932      38.180011       56285.026965   \n2              869125.012500      38.064686       87116.830844   \n3              789873.203729      37.777547      100607.155049   \n4              836528.687774      37.061125       42773.138452   \n...                      ...            ...                ...   \n111988         840913.833568      37.213712       40478.734970   \n111989         841826.691016      37.655434      135724.673788   \n111990         994180.887805      38.240196      429979.701959   \n111991         867784.190106      37.725293       61135.778415   \n111992         914492.440765      36.680678      100229.495759   \n\n        ProfessionAveCitySize  ProfessionAveAge  CountryAveCitySize_y  \\\nindex                                                                   \n0                5.690987e+05         39.706667         830823.825962   \n1                1.319762e+06         37.440594         831883.473583   \n2                7.289293e+05         36.838542         835229.543686   \n3                1.457333e+06         34.608696         842784.843305   \n4                1.007441e+06         37.871166         837108.021998   \n...                       ...               ...                   ...   \n111988           9.117592e+05         35.419753         834623.977935   \n111989           1.050179e+06         37.561983         832605.251928   \n111990           1.101049e+06         36.674286         832521.726789   \n111991           1.173424e+06         37.238889         835985.511602   \n111992           6.615599e+05         37.628959         841142.617690   \n\n        CountryProfAveEarning  CountryProfAveAge  ProfessionAveEarning  \nindex                                                                   \n0               108403.292627          37.247088         151466.154967  \n1               114493.990230          37.291437         133995.974029  \n2               106792.495712          37.318456         117286.626556  \n3                95755.675523          37.389970          56144.047562  \n4                93881.829786          37.368673          88209.535957  \n...                       ...                ...                   ...  \n111988           99084.553195          37.297375          59594.668358  \n111989          111781.435088          37.442488         135245.179782  \n111990          114335.205695          37.319183          98273.245191  \n111991          101666.245926          37.279530         114634.753398  \n111992          114481.966930          37.311485         119996.468881  \n\n[111993 rows x 9 columns]","text/html":"<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>CountryAveCitySize_x</th>\n      <th>CountryAveAge</th>\n      <th>CountryAveEarning</th>\n      <th>ProfessionAveCitySize</th>\n      <th>ProfessionAveAge</th>\n      <th>CountryAveCitySize_y</th>\n      <th>CountryProfAveEarning</th>\n      <th>CountryProfAveAge</th>\n      <th>ProfessionAveEarning</th>\n    </tr>\n    <tr>\n      <th>index</th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <td>0</td>\n      <td>804670.998106</td>\n      <td>37.633159</td>\n      <td>33116.813053</td>\n      <td>5.690987e+05</td>\n      <td>39.706667</td>\n      <td>830823.825962</td>\n      <td>108403.292627</td>\n      <td>37.247088</td>\n      <td>151466.154967</td>\n    </tr>\n    <tr>\n      <td>1</td>\n      <td>952551.623932</td>\n      <td>38.180011</td>\n      <td>56285.026965</td>\n      <td>1.319762e+06</td>\n      <td>37.440594</td>\n      <td>831883.473583</td>\n      <td>114493.990230</td>\n      <td>37.291437</td>\n      <td>133995.974029</td>\n    </tr>\n    <tr>\n      <td>2</td>\n      <td>869125.012500</td>\n      <td>38.064686</td>\n      <td>87116.830844</td>\n      <td>7.289293e+05</td>\n      <td>36.838542</td>\n      <td>835229.543686</td>\n      <td>106792.495712</td>\n      <td>37.318456</td>\n      <td>117286.626556</td>\n    </tr>\n    <tr>\n      <td>3</td>\n      <td>789873.203729</td>\n      <td>37.777547</td>\n      <td>100607.155049</td>\n      <td>1.457333e+06</td>\n      <td>34.608696</td>\n      <td>842784.843305</td>\n      <td>95755.675523</td>\n      <td>37.389970</td>\n      <td>56144.047562</td>\n    </tr>\n    <tr>\n      <td>4</td>\n      <td>836528.687774</td>\n      <td>37.061125</td>\n      <td>42773.138452</td>\n      <td>1.007441e+06</td>\n      <td>37.871166</td>\n      <td>837108.021998</td>\n      <td>93881.829786</td>\n      <td>37.368673</td>\n      <td>88209.535957</td>\n    </tr>\n    <tr>\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      <td>111988</td>\n      <td>840913.833568</td>\n      <td>37.213712</td>\n      <td>40478.734970</td>\n      <td>9.117592e+05</td>\n      <td>35.419753</td>\n      <td>834623.977935</td>\n      <td>99084.553195</td>\n      <td>37.297375</td>\n      <td>59594.668358</td>\n    </tr>\n    <tr>\n      <td>111989</td>\n      <td>841826.691016</td>\n      <td>37.655434</td>\n      <td>135724.673788</td>\n      <td>1.050179e+06</td>\n      <td>37.561983</td>\n      <td>832605.251928</td>\n      <td>111781.435088</td>\n      <td>37.442488</td>\n      <td>135245.179782</td>\n    </tr>\n    <tr>\n      <td>111990</td>\n      <td>994180.887805</td>\n      <td>38.240196</td>\n      <td>429979.701959</td>\n      <td>1.101049e+06</td>\n      <td>36.674286</td>\n      <td>832521.726789</td>\n      <td>114335.205695</td>\n      <td>37.319183</td>\n      <td>98273.245191</td>\n    </tr>\n    <tr>\n      <td>111991</td>\n      <td>867784.190106</td>\n      <td>37.725293</td>\n      <td>61135.778415</td>\n      <td>1.173424e+06</td>\n      <td>37.238889</td>\n      <td>835985.511602</td>\n      <td>101666.245926</td>\n      <td>37.279530</td>\n      <td>114634.753398</td>\n    </tr>\n    <tr>\n      <td>111992</td>\n      <td>914492.440765</td>\n      <td>36.680678</td>\n      <td>100229.495759</td>\n      <td>6.615599e+05</td>\n      <td>37.628959</td>\n      <td>841142.617690</td>\n      <td>114481.966930</td>\n      <td>37.311485</td>\n      <td>119996.468881</td>\n    </tr>\n  </tbody>\n</table>\n<p>111993 rows × 9 columns</p>\n</div>"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"professionAves\nc","execution_count":211,"outputs":[{"output_type":"execute_result","execution_count":211,"data":{"text/plain":"                              Profession  ProfessionAveCitySize  \\\n0                         .net developer           1.098149e+06   \n1                .net software developer           1.012420e+06   \n2     accessibility outreach coordinator           8.526100e+04   \n3          accessibility program manager           1.960634e+06   \n4                     account executive            4.966050e+05   \n...                                  ...                    ...   \n1323           workforce planning intern           8.511467e+05   \n1324                              writer           7.228182e+05   \n1325                    x-ray technician           8.561219e+05   \n1326                          yardmaster           2.085673e+06   \n1327      youth initiatives lead advisor           5.418901e+05   \n\n      ProfessionAveAge  CountryAveCitySize  CountryProfAveEarning  \\\n0            33.200000       826409.729624          122657.951457   \n1            31.000000       808918.400061           88529.179164   \n2            33.000000       800567.723309           64396.592903   \n3            33.500000       779880.022599           31732.578983   \n4            20.000000       881084.653061          540953.871906   \n...                ...                 ...                    ...   \n1323         39.875000       838281.283661          100872.505565   \n1324         34.781250       846329.868988          119371.558010   \n1325         36.862069       831321.006622          125210.072567   \n1326         39.428571       856728.115393           72191.563054   \n1327         43.062500       832088.259729          104960.509795   \n\n      CountryProfAveAge  ProfessionAveEarning  \n0             36.748935         194211.087120  \n1             37.385940          63816.304430  \n2             36.960718          17204.790070  \n3             37.378408          37212.198420  \n4             36.173469         644524.183000  \n...                 ...                   ...  \n1323          37.345718         164529.109083  \n1324          37.372702         170922.079052  \n1325          37.307195         219642.858725  \n1326          37.166622          96012.661454  \n1327          37.353625         182604.361828  \n\n[1328 rows x 7 columns]","text/html":"<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>Profession</th>\n      <th>ProfessionAveCitySize</th>\n      <th>ProfessionAveAge</th>\n      <th>CountryAveCitySize</th>\n      <th>CountryProfAveEarning</th>\n      <th>CountryProfAveAge</th>\n      <th>ProfessionAveEarning</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <td>0</td>\n      <td>.net developer</td>\n      <td>1.098149e+06</td>\n      <td>33.200000</td>\n      <td>826409.729624</td>\n      <td>122657.951457</td>\n      <td>36.748935</td>\n      <td>194211.087120</td>\n    </tr>\n    <tr>\n      <td>1</td>\n      <td>.net software developer</td>\n      <td>1.012420e+06</td>\n      <td>31.000000</td>\n      <td>808918.400061</td>\n      <td>88529.179164</td>\n      <td>37.385940</td>\n      <td>63816.304430</td>\n    </tr>\n    <tr>\n      <td>2</td>\n      <td>accessibility outreach coordinator</td>\n      <td>8.526100e+04</td>\n      <td>33.000000</td>\n      <td>800567.723309</td>\n      <td>64396.592903</td>\n      <td>36.960718</td>\n      <td>17204.790070</td>\n    </tr>\n    <tr>\n      <td>3</td>\n      <td>accessibility program manager</td>\n      <td>1.960634e+06</td>\n      <td>33.500000</td>\n      <td>779880.022599</td>\n      <td>31732.578983</td>\n      <td>37.378408</td>\n      <td>37212.198420</td>\n    </tr>\n    <tr>\n      <td>4</td>\n      <td>account executive</td>\n      <td>4.966050e+05</td>\n      <td>20.000000</td>\n      <td>881084.653061</td>\n      <td>540953.871906</td>\n      <td>36.173469</td>\n      <td>644524.183000</td>\n    </tr>\n    <tr>\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      <td>1323</td>\n      <td>workforce planning intern</td>\n      <td>8.511467e+05</td>\n      <td>39.875000</td>\n      <td>838281.283661</td>\n      <td>100872.505565</td>\n      <td>37.345718</td>\n      <td>164529.109083</td>\n    </tr>\n    <tr>\n      <td>1324</td>\n      <td>writer</td>\n      <td>7.228182e+05</td>\n      <td>34.781250</td>\n      <td>846329.868988</td>\n      <td>119371.558010</td>\n      <td>37.372702</td>\n      <td>170922.079052</td>\n    </tr>\n    <tr>\n      <td>1325</td>\n      <td>x-ray technician</td>\n      <td>8.561219e+05</td>\n      <td>36.862069</td>\n      <td>831321.006622</td>\n      <td>125210.072567</td>\n      <td>37.307195</td>\n      <td>219642.858725</td>\n    </tr>\n    <tr>\n      <td>1326</td>\n      <td>yardmaster</td>\n      <td>2.085673e+06</td>\n      <td>39.428571</td>\n      <td>856728.115393</td>\n      <td>72191.563054</td>\n      <td>37.166622</td>\n      <td>96012.661454</td>\n    </tr>\n    <tr>\n      <td>1327</td>\n      <td>youth initiatives lead advisor</td>\n      <td>5.418901e+05</td>\n      <td>43.062500</td>\n      <td>832088.259729</td>\n      <td>104960.509795</td>\n      <td>37.353625</td>\n      <td>182604.361828</td>\n    </tr>\n  </tbody>\n</table>\n<p>1328 rows × 7 columns</p>\n</div>"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"import seaborn as sns\n# Identify Numeric features\nnumeric_features = ['Year of Record','Age','Size of City','Uni Degree','Wears Glasses','Body Height cm','Income in EUR','IncomeLog']\n\n# Identify Categorical features\ncat_features = ['Gender','Country','Profession','Hair Color']\n\n# Correlation matrix between numerical values\ng = sns.heatmap(dataset[numeric_features].corr(),annot=True, fmt = \".2f\", cmap = \"coolwarm\")","execution_count":212,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 432x288 with 2 Axes>","image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"X","execution_count":213,"outputs":[{"output_type":"execute_result","execution_count":213,"data":{"text/plain":"        CountryAveCitySize_x  CountryAveAge  CountryAveEarning  \\\nindex                                                            \n0              804670.998106      37.633159       33116.813053   \n1              952551.623932      38.180011       56285.026965   \n2              869125.012500      38.064686       87116.830844   \n3              789873.203729      37.777547      100607.155049   \n4              836528.687774      37.061125       42773.138452   \n...                      ...            ...                ...   \n111988         840913.833568      37.213712       40478.734970   \n111989         841826.691016      37.655434      135724.673788   \n111990         994180.887805      38.240196      429979.701959   \n111991         867784.190106      37.725293       61135.778415   \n111992         914492.440765      36.680678      100229.495759   \n\n        ProfessionAveCitySize  ProfessionAveAge  CountryAveCitySize_y  \\\nindex                                                                   \n0                5.690987e+05         39.706667         830823.825962   \n1                1.319762e+06         37.440594         831883.473583   \n2                7.289293e+05         36.838542         835229.543686   \n3                1.457333e+06         34.608696         842784.843305   \n4                1.007441e+06         37.871166         837108.021998   \n...                       ...               ...                   ...   \n111988           9.117592e+05         35.419753         834623.977935   \n111989           1.050179e+06         37.561983         832605.251928   \n111990           1.101049e+06         36.674286         832521.726789   \n111991           1.173424e+06         37.238889         835985.511602   \n111992           6.615599e+05         37.628959         841142.617690   \n\n        CountryProfAveEarning  CountryProfAveAge  ProfessionAveEarning  \nindex                                                                   \n0               108403.292627          37.247088         151466.154967  \n1               114493.990230          37.291437         133995.974029  \n2               106792.495712          37.318456         117286.626556  \n3                95755.675523          37.389970          56144.047562  \n4                93881.829786          37.368673          88209.535957  \n...                       ...                ...                   ...  \n111988           99084.553195          37.297375          59594.668358  \n111989          111781.435088          37.442488         135245.179782  \n111990          114335.205695          37.319183          98273.245191  \n111991          101666.245926          37.279530         114634.753398  \n111992          114481.966930          37.311485         119996.468881  \n\n[111993 rows x 9 columns]","text/html":"<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>CountryAveCitySize_x</th>\n      <th>CountryAveAge</th>\n      <th>CountryAveEarning</th>\n      <th>ProfessionAveCitySize</th>\n      <th>ProfessionAveAge</th>\n      <th>CountryAveCitySize_y</th>\n      <th>CountryProfAveEarning</th>\n      <th>CountryProfAveAge</th>\n      <th>ProfessionAveEarning</th>\n    </tr>\n    <tr>\n      <th>index</th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <td>0</td>\n      <td>804670.998106</td>\n      <td>37.633159</td>\n      <td>33116.813053</td>\n      <td>5.690987e+05</td>\n      <td>39.706667</td>\n      <td>830823.825962</td>\n      <td>108403.292627</td>\n      <td>37.247088</td>\n      <td>151466.154967</td>\n    </tr>\n    <tr>\n      <td>1</td>\n      <td>952551.623932</td>\n      <td>38.180011</td>\n      <td>56285.026965</td>\n      <td>1.319762e+06</td>\n      <td>37.440594</td>\n      <td>831883.473583</td>\n      <td>114493.990230</td>\n      <td>37.291437</td>\n      <td>133995.974029</td>\n    </tr>\n    <tr>\n      <td>2</td>\n      <td>869125.012500</td>\n      <td>38.064686</td>\n      <td>87116.830844</td>\n      <td>7.289293e+05</td>\n      <td>36.838542</td>\n      <td>835229.543686</td>\n      <td>106792.495712</td>\n      <td>37.318456</td>\n      <td>117286.626556</td>\n    </tr>\n    <tr>\n      <td>3</td>\n      <td>789873.203729</td>\n      <td>37.777547</td>\n      <td>100607.155049</td>\n      <td>1.457333e+06</td>\n      <td>34.608696</td>\n      <td>842784.843305</td>\n      <td>95755.675523</td>\n      <td>37.389970</td>\n      <td>56144.047562</td>\n    </tr>\n    <tr>\n      <td>4</td>\n      <td>836528.687774</td>\n      <td>37.061125</td>\n      <td>42773.138452</td>\n      <td>1.007441e+06</td>\n      <td>37.871166</td>\n      <td>837108.021998</td>\n      <td>93881.829786</td>\n      <td>37.368673</td>\n      <td>88209.535957</td>\n    </tr>\n    <tr>\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      <td>111988</td>\n      <td>840913.833568</td>\n      <td>37.213712</td>\n      <td>40478.734970</td>\n      <td>9.117592e+05</td>\n      <td>35.419753</td>\n      <td>834623.977935</td>\n      <td>99084.553195</td>\n      <td>37.297375</td>\n      <td>59594.668358</td>\n    </tr>\n    <tr>\n      <td>111989</td>\n      <td>841826.691016</td>\n      <td>37.655434</td>\n      <td>135724.673788</td>\n      <td>1.050179e+06</td>\n      <td>37.561983</td>\n      <td>832605.251928</td>\n      <td>111781.435088</td>\n      <td>37.442488</td>\n      <td>135245.179782</td>\n    </tr>\n    <tr>\n      <td>111990</td>\n      <td>994180.887805</td>\n      <td>38.240196</td>\n      <td>429979.701959</td>\n      <td>1.101049e+06</td>\n      <td>36.674286</td>\n      <td>832521.726789</td>\n      <td>114335.205695</td>\n      <td>37.319183</td>\n      <td>98273.245191</td>\n    </tr>\n    <tr>\n      <td>111991</td>\n      <td>867784.190106</td>\n      <td>37.725293</td>\n      <td>61135.778415</td>\n      <td>1.173424e+06</td>\n      <td>37.238889</td>\n      <td>835985.511602</td>\n      <td>101666.245926</td>\n      <td>37.279530</td>\n      <td>114634.753398</td>\n    </tr>\n    <tr>\n      <td>111992</td>\n      <td>914492.440765</td>\n      <td>36.680678</td>\n      <td>100229.495759</td>\n      <td>6.615599e+05</td>\n      <td>37.628959</td>\n      <td>841142.617690</td>\n      <td>114481.966930</td>\n      <td>37.311485</td>\n      <td>119996.468881</td>\n    </tr>\n  </tbody>\n</table>\n<p>111993 rows × 9 columns</p>\n</div>"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Dataset copy to ocmpare \ndataset2 = dataset\n\n#Drop categorical columns\ndataset2 = dataset2.drop(['Gender','University Degree','IncomeLog','Country','Profession','Size of City','Hair Color','Instance','Wears Glasses','Income in EUR'],axis='columns')\n# dataset2 = dataset2.drop(['Gender','Country','Profession','University Degree','Hair Color','Instance'],axis='columns')\n\n# Merge dummy variables\nmerged  = pd.concat([dataset2,dummy_gender],axis=\"columns\")\n# merged  = pd.concat([dataset2,dummy_gender,dummy_degree,dummy_haircolor],axis=\"columns\")\n\nmerged = pd.concat([merged[0:111993],X],axis=1)\nmerged","execution_count":214,"outputs":[{"output_type":"execute_result","execution_count":214,"data":{"text/plain":"        Year of Record   Age  Body Height cm  Uni Degree  CitySizeLog  female  \\\n0               1997.0  41.0             193           1    14.030565       0   \n1               1996.0  41.0             186           2    14.287702       0   \n2               2018.0  28.0             170           3    14.076348       0   \n3               2006.0  33.0             171           0    13.530363       0   \n4               2010.0  46.0             188           0    11.465719       1   \n...                ...   ...             ...         ...          ...     ...   \n111988          1980.0  29.0             135           0    12.232101       1   \n111989          2006.0  36.0             147           2    13.828255       1   \n111990          1993.0  36.0             153           2    11.016134       1   \n111991          2019.0  54.0             190           0    13.651390       0   \n111992          2017.0  27.0             174           0    12.693394       1   \n\n        male  other  CountryAveCitySize_x  CountryAveAge  CountryAveEarning  \\\n0          0      0         804670.998106      37.633159       33116.813053   \n1          0      1         952551.623932      38.180011       56285.026965   \n2          0      1         869125.012500      38.064686       87116.830844   \n3          0      1         789873.203729      37.777547      100607.155049   \n4          0      0         836528.687774      37.061125       42773.138452   \n...      ...    ...                   ...            ...                ...   \n111988     0      0         840913.833568      37.213712       40478.734970   \n111989     0      0         841826.691016      37.655434      135724.673788   \n111990     0      0         994180.887805      38.240196      429979.701959   \n111991     0      1         867784.190106      37.725293       61135.778415   \n111992     0      0         914492.440765      36.680678      100229.495759   \n\n        ProfessionAveCitySize  ProfessionAveAge  CountryAveCitySize_y  \\\n0                5.690987e+05         39.706667         830823.825962   \n1                1.319762e+06         37.440594         831883.473583   \n2                7.289293e+05         36.838542         835229.543686   \n3                1.457333e+06         34.608696         842784.843305   \n4                1.007441e+06         37.871166         837108.021998   \n...                       ...               ...                   ...   \n111988           9.117592e+05         35.419753         834623.977935   \n111989           1.050179e+06         37.561983         832605.251928   \n111990           1.101049e+06         36.674286         832521.726789   \n111991           1.173424e+06         37.238889         835985.511602   \n111992           6.615599e+05         37.628959         841142.617690   \n\n        CountryProfAveEarning  CountryProfAveAge  ProfessionAveEarning  \n0               108403.292627          37.247088         151466.154967  \n1               114493.990230          37.291437         133995.974029  \n2               106792.495712          37.318456         117286.626556  \n3                95755.675523          37.389970          56144.047562  \n4                93881.829786          37.368673          88209.535957  \n...                       ...                ...                   ...  \n111988           99084.553195          37.297375          59594.668358  \n111989          111781.435088          37.442488         135245.179782  \n111990          114335.205695          37.319183          98273.245191  \n111991          101666.245926          37.279530         114634.753398  \n111992          114481.966930          37.311485         119996.468881  \n\n[111993 rows x 17 columns]","text/html":"<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>Year of Record</th>\n      <th>Age</th>\n      <th>Body Height cm</th>\n      <th>Uni Degree</th>\n      <th>CitySizeLog</th>\n      <th>female</th>\n      <th>male</th>\n      <th>other</th>\n      <th>CountryAveCitySize_x</th>\n      <th>CountryAveAge</th>\n      <th>CountryAveEarning</th>\n      <th>ProfessionAveCitySize</th>\n      <th>ProfessionAveAge</th>\n      <th>CountryAveCitySize_y</th>\n      <th>CountryProfAveEarning</th>\n      <th>CountryProfAveAge</th>\n      <th>ProfessionAveEarning</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <td>0</td>\n      <td>1997.0</td>\n      <td>41.0</td>\n      <td>193</td>\n      <td>1</td>\n      <td>14.030565</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>804670.998106</td>\n      <td>37.633159</td>\n      <td>33116.813053</td>\n      <td>5.690987e+05</td>\n      <td>39.706667</td>\n      <td>830823.825962</td>\n      <td>108403.292627</td>\n      <td>37.247088</td>\n      <td>151466.154967</td>\n    </tr>\n    <tr>\n      <td>1</td>\n      <td>1996.0</td>\n      <td>41.0</td>\n      <td>186</td>\n      <td>2</td>\n      <td>14.287702</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>952551.623932</td>\n      <td>38.180011</td>\n      <td>56285.026965</td>\n      <td>1.319762e+06</td>\n      <td>37.440594</td>\n      <td>831883.473583</td>\n      <td>114493.990230</td>\n      <td>37.291437</td>\n      <td>133995.974029</td>\n    </tr>\n    <tr>\n      <td>2</td>\n      <td>2018.0</td>\n      <td>28.0</td>\n      <td>170</td>\n      <td>3</td>\n      <td>14.076348</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>869125.012500</td>\n      <td>38.064686</td>\n      <td>87116.830844</td>\n      <td>7.289293e+05</td>\n      <td>36.838542</td>\n      <td>835229.543686</td>\n      <td>106792.495712</td>\n      <td>37.318456</td>\n      <td>117286.626556</td>\n    </tr>\n    <tr>\n      <td>3</td>\n      <td>2006.0</td>\n      <td>33.0</td>\n      <td>171</td>\n      <td>0</td>\n      <td>13.530363</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>789873.203729</td>\n      <td>37.777547</td>\n      <td>100607.155049</td>\n      <td>1.457333e+06</td>\n      <td>34.608696</td>\n      <td>842784.843305</td>\n      <td>95755.675523</td>\n      <td>37.389970</td>\n      <td>56144.047562</td>\n    </tr>\n    <tr>\n      <td>4</td>\n      <td>2010.0</td>\n      <td>46.0</td>\n      <td>188</td>\n      <td>0</td>\n      <td>11.465719</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>836528.687774</td>\n      <td>37.061125</td>\n      <td>42773.138452</td>\n      <td>1.007441e+06</td>\n      <td>37.871166</td>\n      <td>837108.021998</td>\n      <td>93881.829786</td>\n      <td>37.368673</td>\n      <td>88209.535957</td>\n    </tr>\n    <tr>\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      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <td>111988</td>\n      <td>1980.0</td>\n      <td>29.0</td>\n      <td>135</td>\n      <td>0</td>\n      <td>12.232101</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>840913.833568</td>\n      <td>37.213712</td>\n      <td>40478.734970</td>\n      <td>9.117592e+05</td>\n      <td>35.419753</td>\n      <td>834623.977935</td>\n      <td>99084.553195</td>\n      <td>37.297375</td>\n      <td>59594.668358</td>\n    </tr>\n    <tr>\n      <td>111989</td>\n      <td>2006.0</td>\n      <td>36.0</td>\n      <td>147</td>\n      <td>2</td>\n      <td>13.828255</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>841826.691016</td>\n      <td>37.655434</td>\n      <td>135724.673788</td>\n      <td>1.050179e+06</td>\n      <td>37.561983</td>\n      <td>832605.251928</td>\n      <td>111781.435088</td>\n      <td>37.442488</td>\n      <td>135245.179782</td>\n    </tr>\n    <tr>\n      <td>111990</td>\n      <td>1993.0</td>\n      <td>36.0</td>\n      <td>153</td>\n      <td>2</td>\n      <td>11.016134</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>994180.887805</td>\n      <td>38.240196</td>\n      <td>429979.701959</td>\n      <td>1.101049e+06</td>\n      <td>36.674286</td>\n      <td>832521.726789</td>\n      <td>114335.205695</td>\n      <td>37.319183</td>\n      <td>98273.245191</td>\n    </tr>\n    <tr>\n      <td>111991</td>\n      <td>2019.0</td>\n      <td>54.0</td>\n      <td>190</td>\n      <td>0</td>\n      <td>13.651390</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>867784.190106</td>\n      <td>37.725293</td>\n      <td>61135.778415</td>\n      <td>1.173424e+06</td>\n      <td>37.238889</td>\n      <td>835985.511602</td>\n      <td>101666.245926</td>\n      <td>37.279530</td>\n      <td>114634.753398</td>\n    </tr>\n    <tr>\n      <td>111992</td>\n      <td>2017.0</td>\n      <td>27.0</td>\n      <td>174</td>\n      <td>0</td>\n      <td>12.693394</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>914492.440765</td>\n      <td>36.680678</td>\n      <td>100229.495759</td>\n      <td>6.615599e+05</td>\n      <td>37.628959</td>\n      <td>841142.617690</td>\n      <td>114481.966930</td>\n      <td>37.311485</td>\n      <td>119996.468881</td>\n    </tr>\n  </tbody>\n</table>\n<p>111993 rows × 17 columns</p>\n</div>"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"dataset['Profession'] ","execution_count":215,"outputs":[{"output_type":"execute_result","execution_count":215,"data":{"text/plain":"0                            steel workers\n1                   safe event coordinator\n2             receivables/payables analyst\n3                          fleet assistant\n4                             lead trainer\n                       ...                \n73225                   payroll supervisor\n73226             purchasing agent - buyer\n73227    neigborhood resiliency specialist\n73228    it infrastructure project manager\n73229                   materials engineer\nName: Profession, Length: 185223, dtype: object"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Finding Outliers, +0.7 or higher: Very strong positive relationship\n\n\"\"\"\norg_dt = merged.iloc[0:111827,:]\norg_dt = pd.DataFrame.from_records(org_dt)\norg_dt['IncomeLog'] = np.log(org_dt['Income in EUR'])\n\ncorr = org_dt.corr().abs()\ncorr.IncomeLog[corr.IncomeLog >= 0.005].sort_values(ascending=False)\n\"\"\"","execution_count":216,"outputs":[{"output_type":"execute_result","execution_count":216,"data":{"text/plain":"\"\\norg_dt = merged.iloc[0:111827,:]\\norg_dt = pd.DataFrame.from_records(org_dt)\\norg_dt['IncomeLog'] = np.log(org_dt['Income in EUR'])\\n\\ncorr = org_dt.corr().abs()\\ncorr.IncomeLog[corr.IncomeLog >= 0.005].sort_values(ascending=False)\\n\""},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"\"\"\"org_dt.plot.scatter(x='Age', y='IncomeLog', color = 'green')\nplt.show()\"\"\"\n","execution_count":217,"outputs":[{"output_type":"execute_result","execution_count":217,"data":{"text/plain":"\"org_dt.plot.scatter(x='Age', y='IncomeLog', color = 'green')\\nplt.show()\""},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Splitting the dataset into the Training set and Test set manually\n\n\"\"\"X_train = merged.iloc[0:78000,:] \nX_test = merged.iloc[78000:111827,:]\ny_train = y[0:78000]\ny_test = y[78000:111827]\"\"\"\n\n\nX_train = merged.iloc[0:111993,:] \nX_test = merged.iloc[100000:111993,:]\ny_train = y[0:111993]\ny_test = y[100000:111993]\n","execution_count":218,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"\"\"\"\nxgb = xgboost.XGBRegressor(colsample_bytree=0.8, subsample=0.5,\n                             learning_rate=0.05, max_depth=8, \n                             min_child_weight=2.8, n_estimators=1000,\n                             reg_alpha=0.3, reg_lambda=0.75, gamma=0.01, \n                             silent=1, random_state =7, nthread = -1)\n\"\"\"\n\nxgb = xgboost.XGBRegressor(colsample_bytree=0.4,\n                   eval_metric='rmse',\n                 gamma=0.01,\n                   nthread=-1,\n                 learning_rate=0.05,\n                 max_depth=4, \n                 min_child_weight=1.8,\n                 n_estimators=3000,                                                     \n                 reg_alpha=0.3,\n                 reg_lambda=0.75,\n                 subsample=0.5,\n                 seed=42)\n\n\"\"\"# Fitting Simple Linear Regression to the Training set\nfrom sklearn.linear_model import LinearRegression\nxgb = LinearRegression()\"\"\"\n\nxgb.fit(X_train, y_train)\nxgb_pred = xgb.predict(X_test)\ny_t2 = xgb.predict(X_train)","execution_count":219,"outputs":[{"output_type":"stream","text":"[23:21:10] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n","name":"stdout"},{"output_type":"error","ename":"KeyboardInterrupt","evalue":"","traceback":["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m","\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)","\u001b[0;32m<ipython-input-219-f4e18a0f6f02>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     24\u001b[0m xgb = LinearRegression()\"\"\"\n\u001b[1;32m     25\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 26\u001b[0;31m \u001b[0mxgb\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_train\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     27\u001b[0m \u001b[0mxgb_pred\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mxgb\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_test\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     28\u001b[0m \u001b[0my_t2\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mxgb\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/opt/conda/lib/python3.6/site-packages/xgboost/sklearn.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, X, y, sample_weight, eval_set, eval_metric, early_stopping_rounds, verbose, xgb_model, sample_weight_eval_set, callbacks)\u001b[0m\n\u001b[1;32m    394\u001b[0m                               \u001b[0mevals_result\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mevals_result\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mobj\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeval\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mfeval\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    395\u001b[0m                               \u001b[0mverbose_eval\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mverbose\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mxgb_model\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mxgb_model\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 396\u001b[0;31m                               callbacks=callbacks)\n\u001b[0m\u001b[1;32m    397\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    398\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mevals_result\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/opt/conda/lib/python3.6/site-packages/xgboost/training.py\u001b[0m in \u001b[0;36mtrain\u001b[0;34m(params, dtrain, num_boost_round, evals, obj, feval, maximize, early_stopping_rounds, evals_result, verbose_eval, xgb_model, callbacks, learning_rates)\u001b[0m\n\u001b[1;32m    214\u001b[0m                            \u001b[0mevals\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mevals\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    215\u001b[0m                            \u001b[0mobj\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeval\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mfeval\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 216\u001b[0;31m                            xgb_model=xgb_model, callbacks=callbacks)\n\u001b[0m\u001b[1;32m    217\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    218\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/opt/conda/lib/python3.6/site-packages/xgboost/training.py\u001b[0m in \u001b[0;36m_train_internal\u001b[0;34m(params, dtrain, num_boost_round, evals, obj, feval, xgb_model, callbacks)\u001b[0m\n\u001b[1;32m     72\u001b[0m         \u001b[0;31m# Skip the first update if it is a recovery step.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     73\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mversion\u001b[0m \u001b[0;34m%\u001b[0m \u001b[0;36m2\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 74\u001b[0;31m             \u001b[0mbst\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mupdate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdtrain\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mobj\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     75\u001b[0m             \u001b[0mbst\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msave_rabit_checkpoint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     76\u001b[0m             \u001b[0mversion\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/opt/conda/lib/python3.6/site-packages/xgboost/core.py\u001b[0m in \u001b[0;36mupdate\u001b[0;34m(self, dtrain, iteration, fobj)\u001b[0m\n\u001b[1;32m   1107\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mfobj\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1108\u001b[0m             _check_call(_LIB.XGBoosterUpdateOneIter(self.handle, ctypes.c_int(iteration),\n\u001b[0;32m-> 1109\u001b[0;31m                                                     dtrain.handle))\n\u001b[0m\u001b[1;32m   1110\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1111\u001b[0m             \u001b[0mpred\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdtrain\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;31mKeyboardInterrupt\u001b[0m: "]}]},{"metadata":{"trusted":true},"cell_type":"code","source":"RMSE = np.sqrt(mean_squared_error(y_train, y_t2))\nprint(np.sqrt(mean_squared_error(xgb_pred, y_test)))\nprint(RMSE.round(4))\nscore = xgb.score(X_test,y_test)\nprint(score)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"y_t2","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"y_dt_pred = xgb.predict(merged.iloc[111827:,:])\ncorrected_pred = y_dt_pred\nfor i in range(0,len(corrected_pred)):\n    if corrected_pred[i] < 0:\n        corrected_pred[i] = corrected_pred[i]*(-1)\n        \ncorrected_pred.flatten()        \ny_dt_pred.flatten()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"submission = pd.DataFrame({\n        \"Instance\": dt['Instance'],\n        \"Income\": y_dt_pred.flatten()\n    })\nsubmission = pd.DataFrame.from_records(submission)\nsubmission.to_csv('submission.csv')\n\nsubmission_corr = pd.DataFrame({\n        \"Instance\": dt['Instance'],\n        \"Income\": corrected_pred.flatten()\n    })\nsubmission_corr = pd.DataFrame.from_records(submission)\nsubmission.to_csv('submission_corr.csv')\nprint(submission_corr)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"merged","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"dt['Instance'].values\nsubmission","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat":4,"nbformat_minor":1}