{"cells":[{"cell_type":"code","execution_count":1,"id":"0f51d59d-06e7-410c-b351-230c1b358909","metadata":{},"outputs":[],"source":"import pandas as pd\nimport numpy as np\nfrom sklearn.model_selection import StratifiedKFold\nfrom sklearn.metrics import mean_squared_error , accuracy_score\nfrom sklearn.preprocessing import StandardScaler \nimport lightgbm as lgb\nimport matplotlib.pyplot as plt\nimport seaborn as sns\nimport warnings\nwarnings.filterwarnings('ignore')"},{"cell_type":"code","execution_count":31,"id":"837f2bd9-8ed7-48c6-b109-701206c2d6d6","metadata":{},"outputs":[{"data":{"text/plain":["((720000, 20), (480000, 19))"]},"execution_count":31,"metadata":{},"output_type":"execute_result"}],"source":"df_train = pd.read_csv('train.csv')\ndf_test = pd.read_csv('test.csv')\ndf_train.shape , df_test.shape"},{"cell_type":"code","execution_count":32,"id":"c616ee46-51c5-4a44-bcf7-4aa6a7bedcda","metadata":{},"outputs":[{"data":{"text/plain":["Index(['id', 'Age', 'Annual Income', 'Marital Status', 'Number of Children',\n","       'Education Level', 'Occupation', 'Health Score', 'Location',\n","       'Policy Type', 'Previous Claims', 'Vehicle Age', 'Credit Score',\n","       'Insurance Duration', 'Policy Start Date', 'Customer Feedback',\n","       'Smoking Status', 'Exercise Frequency', 'Property Type',\n","       'Premium Amount'],\n","      dtype='object')"]},"execution_count":32,"metadata":{},"output_type":"execute_result"}],"source":"df_train.columns"},{"cell_type":"code","execution_count":33,"id":"691763ab-634e-42d7-8633-1e153ab65cd9","metadata":{},"outputs":[{"data":{"text/plain":["id                      int64\n","Age                   float64\n","Annual Income         float64\n","Marital Status         object\n","Number of Children    float64\n","Education Level        object\n","Occupation             object\n","Health Score          float64\n","Location               object\n","Policy Type            object\n","Previous Claims       float64\n","Vehicle Age           float64\n","Credit Score          float64\n","Insurance Duration    float64\n","Policy Start Date      object\n","Customer Feedback      object\n","Smoking Status         object\n","Exercise Frequency     object\n","Property Type          object\n","Premium Amount        float64\n","dtype: object"]},"execution_count":33,"metadata":{},"output_type":"execute_result"}],"source":"df_train.dtypes"},{"cell_type":"code","execution_count":34,"id":"4d69285b-695f-4985-b735-520b01120900","metadata":{},"outputs":[],"source":"categorical_features = ['Marital Status','Education Level','Occupation','Location','Policy Type','Customer Feedback','Smoking Status','Exercise Frequency','Property Type']\nnumerical_features = ['Age','Annual Income','Number of Children','Health Score','Previous Claims','Vehicle Age','Credit Score','Insurance Duration']\ndate_feature = 'Policy Start Date'"},{"cell_type":"code","execution_count":35,"id":"6aeb2cc1-4ad1-4ce9-844a-054459c65435","metadata":{},"outputs":[],"source":"from sklearn.impute import SimpleImputer\ndef missing_values(df) :\n    df = df.copy()\n    imputer1 = SimpleImputer(strategy = 'median')\n    df[numerical_features] = imputer1.fit_transform(df[numerical_features])\n\n    imputer2 = SimpleImputer(strategy = 'most_frequent')\n    df[categorical_features] = imputer2.fit_transform(df[categorical_features])\n    return df"},{"cell_type":"code","execution_count":36,"id":"68f55204-831d-49d3-bd81-d89d34e90cf5","metadata":{},"outputs":[],"source":"df_train = missing_values(df_train)\ndf_test = missing_values(df_test)"},{"cell_type":"code","execution_count":37,"id":"245f0819-5be7-448e-aea3-c8e6b7d3c449","metadata":{},"outputs":[{"data":{"text/plain":["id                    0\n","Age                   0\n","Annual Income         0\n","Marital Status        0\n","Number of Children    0\n","Education Level       0\n","Occupation            0\n","Health Score          0\n","Location              0\n","Policy Type           0\n","Previous Claims       0\n","Vehicle Age           0\n","Credit Score          0\n","Insurance Duration    0\n","Policy Start Date     0\n","Customer Feedback     0\n","Smoking Status        0\n","Exercise Frequency    0\n","Property Type         0\n","Premium Amount        0\n","dtype: int64"]},"execution_count":37,"metadata":{},"output_type":"execute_result"}],"source":"df_train.isna().sum()"},{"cell_type":"code","execution_count":38,"id":"445ced59-39ae-4f2d-a334-e0fa539d2f07","metadata":{},"outputs":[{"data":{"image/png":"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","text/plain":["<Figure size 1200x900 with 2 Axes>"]},"metadata":{},"output_type":"display_data"}],"source":"numeric_df = df_train.select_dtypes(include=['int64', 'float64'])\nnumeric_df = numeric_df.drop('id' , axis = 1)\n# Compute correlation matrix\ncorr = numeric_df.corr()\n\n# Mask upper triangle to make it cleaner\nmask = np.triu(np.ones_like(corr, dtype=bool))\n\nplt.figure(figsize=(12, 9))\nsns.heatmap(\n    corr.round(2),\n    annot=True,\n    fmt=\".2f\",\n    mask=mask,\n    cmap=\"RdBu_r\",\n    center=0,\n    linewidths=0.5,\n    cbar_kws={'shrink': 0.8}\n)\nplt.title(\"Correlation Matrix (Numeric Features)\", fontsize=14, weight='bold')\nplt.xticks(rotation=45, ha='right')\nplt.yticks(rotation=0)\nplt.tight_layout()\nplt.show()"},{"cell_type":"code","execution_count":39,"id":"e41ca189-a365-4fd4-8b8e-e60a40c9ea40","metadata":{},"outputs":[],"source":"def extract_date_features(df, date_col='Policy Start Date'):\n    df = df.copy()\n    if date_col in df.columns:\n        df[date_col] = pd.to_datetime(df[date_col])\n        df['Policy_Year'] = df[date_col].dt.year\n        df['Policy_Month'] = df[date_col].dt.month\n        df['Policy_Quarter'] = df[date_col].dt.quarter\n        df['Policy_DayOfYear'] = df[date_col].dt.dayofyear\n        df['Policy_DayOfWeek'] = df[date_col].dt.dayofweek\n        reference_date = pd.Timestamp('2025-10-05')\n        df['Policy_Age_Days'] = (reference_date - df[date_col]).dt.days\n        df.drop(columns=[date_col], inplace=True)\n    return df\n"},{"cell_type":"code","execution_count":40,"id":"8515ab87-a866-41f0-a443-6332891e321e","metadata":{},"outputs":[],"source":"df_train = extract_date_features(df_train)\ndf_test = extract_date_features(df_test)"},{"cell_type":"code","execution_count":41,"id":"a245b67e-c9bf-4af8-8dec-3d8c56c7c9fd","metadata":{},"outputs":[{"data":{"image/png":"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","text/plain":["<Figure size 640x480 with 1 Axes>"]},"metadata":{},"output_type":"display_data"}],"source":"plt.scatter(df_train['Age'], df_train['Smoking Status'].map({'Yes': 1, 'No': 0}))\nplt.xlabel('Age')\nplt.ylabel('Smoking Status')\nplt.show()\n"},{"cell_type":"code","execution_count":42,"id":"5e96c9d4-6333-42d3-a900-727d345cf57e","metadata":{},"outputs":[{"name":"stdout","output_type":"stream","text":["-7.120933824769506e-05\n"]}],"source":"df_train['Smoking Status Encoded'] = df_train['Smoking Status'].map({'Yes': 1, 'No': 0})\ncorr = df_train['Age'].corr(df_train['Smoking Status Encoded'])\nprint(corr)\n"},{"cell_type":"code","execution_count":43,"id":"a11f6cd8-05b9-48ec-a1be-545d42c6cbd4","metadata":{},"outputs":[{"data":{"text/plain":["<Axes: xlabel='Smoking Status', ylabel='Age'>"]},"execution_count":43,"metadata":{},"output_type":"execute_result"},{"data":{"image/png":"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","text/plain":["<Figure size 640x480 with 1 Axes>"]},"metadata":{},"output_type":"display_data"}],"source":"import seaborn as sns\nsns.boxplot(x='Smoking Status', y='Age', data=df_train)\n"},{"cell_type":"code","execution_count":44,"id":"a3796315-7608-4ff5-80e6-d9b01c664e54","metadata":{},"outputs":[{"name":"stdout","output_type":"stream","text":["id vs Married: 0.00231600278111599\n","Age vs Married: 0.0013061894778608535\n","Annual Income vs Married: -0.0010629019087048403\n","Number of Children vs Married: 0.0021782868809723742\n","Health Score vs Married: -0.0008994167704971719\n","Previous Claims vs Married: -0.0029531831539511702\n","Vehicle Age vs Married: 0.0016717734019098878\n","Credit Score vs Married: -0.0020833121387160924\n","Insurance Duration vs Married: -0.0011755067005152092\n","Premium Amount vs Married: -0.00041526401625489466\n","Policy_Age_Days vs Married: -0.001717337637129298\n","Smoking Status Encoded vs Married: -0.0019091387830951973\n","id vs Single: -0.0025161991032488476\n","Age vs Single: -0.0002585153125829145\n","Annual Income vs Single: 0.0018189267557813394\n","Number of Children vs Single: -0.001605313316958315\n","Health Score vs Single: 0.0020537161637704457\n","Previous Claims vs Single: 0.0032539618351505616\n","Vehicle Age vs Single: -0.0005779142433717687\n","Credit Score vs Single: -0.0007778194161859399\n","Insurance Duration vs Single: -0.001638222807760255\n","Premium Amount vs Single: 0.002588706084028865\n","Policy_Age_Days vs Single: -1.5484913426239326e-05\n","Smoking Status Encoded vs Single: 0.0003498039995840718\n","id vs High School: 0.0009639624406202897\n","Age vs High School: 0.0005381161522286112\n","Annual Income vs High School: -0.0005414614794195927\n","Number of Children vs High School: -0.001992748489069502\n","Health Score vs High School: -0.0021536459079059423\n","Previous Claims vs High School: -0.0006323804606523979\n","Vehicle Age vs High School: 0.0005751962464374948\n","Credit Score vs High School: -0.0006734161290377884\n","Insurance Duration vs High School: 5.5443134916302e-05\n","Premium Amount vs High School: -8.287263424929489e-05\n","Policy_Age_Days vs High School: 0.0006866804101423502\n","Smoking Status Encoded vs High School: -0.0009621208117924983\n","id vs Master's: 4.462547139612964e-05\n","Age vs Master's: 0.00029256760079805804\n","Annual Income vs Master's: 0.0017612770080325413\n","Number of Children vs Master's: -0.0009220584908373609\n","Health Score vs Master's: -0.0010471261421286292\n","Previous Claims vs Master's: -0.00031000719419155384\n","Vehicle Age vs Master's: -0.0031806206048055896\n","Credit Score vs Master's: -0.0022052462508693966\n","Insurance Duration vs Master's: 0.0013067555924708833\n","Premium Amount vs Master's: 0.0004822290664744612\n","Policy_Age_Days vs Master's: -0.0008000402976091781\n","Smoking Status Encoded vs Master's: -0.0017764717593150418\n","id vs PhD: -0.0003086273883386312\n","Age vs PhD: 0.0001389984469905451\n","Annual Income vs PhD: -0.0005625449371024754\n","Number of Children vs PhD: 0.0016768236040665544\n","Health Score vs PhD: 0.0015376485480454074\n","Previous Claims vs PhD: 0.0017505588459048509\n","Vehicle Age vs PhD: 0.002703155194697701\n","Credit Score vs PhD: 0.00031212929492129834\n","Insurance Duration vs PhD: -0.0008141720172550013\n","Premium Amount vs PhD: -9.894618567914392e-05\n","Policy_Age_Days vs PhD: -0.000610134516123337\n","Smoking Status Encoded vs PhD: 0.0019711134295046276\n","id vs Self-Employed: 0.0008044251399739186\n","Age vs Self-Employed: 0.0012651996281340834\n","Annual Income vs Self-Employed: 0.0025468262052005994\n","Number of Children vs Self-Employed: 0.001005839968273568\n","Health Score vs Self-Employed: -0.0014223261906312747\n","Previous Claims vs Self-Employed: -0.0036976293681588008\n","Vehicle Age vs Self-Employed: -0.0004461104130165693\n","Credit Score vs Self-Employed: 0.002496520687633577\n","Insurance Duration vs Self-Employed: -0.0008751317935437611\n","Premium Amount vs Self-Employed: -0.001870818801973453\n","Policy_Age_Days vs Self-Employed: 0.0032108825871518605\n","Smoking Status Encoded vs Self-Employed: -0.0009031248425085621\n","id vs Unemployed: 0.00041715513912509735\n","Age vs Unemployed: -0.000624133305508612\n","Annual Income vs Unemployed: -0.0012013429837329657\n","Number of Children vs Unemployed: -0.0017815279402330565\n","Health Score vs Unemployed: 0.0027016336754560975\n","Previous Claims vs Unemployed: 0.0025984254032035826\n","Vehicle Age vs Unemployed: 0.0005943372718729155\n","Credit Score vs Unemployed: -0.0023206291134483617\n","Insurance Duration vs Unemployed: 0.0017684853390423877\n","Premium Amount vs Unemployed: -0.000419302318894769\n","Policy_Age_Days vs Unemployed: -0.0030500813343141446\n","Smoking Status Encoded vs Unemployed: -0.0006868602635611968\n","id vs Suburban: -0.0024248616407398778\n","Age vs Suburban: -0.001124329802006672\n","Annual Income vs Suburban: -3.90596313558616e-06\n","Number of Children vs Suburban: 0.0005644452426428164\n","Health Score vs Suburban: 0.0025027991967386362\n","Previous Claims vs Suburban: 0.0012265851984811854\n","Vehicle Age vs Suburban: -0.0009753482010276799\n","Credit Score vs Suburban: -0.000326001872363154\n","Insurance Duration vs Suburban: -0.0021550387182678785\n","Premium Amount vs Suburban: -0.0012458094476737363\n","Policy_Age_Days vs Suburban: -0.0006141162582290455\n","Smoking Status Encoded vs Suburban: 0.0005834619751561271\n","id vs Urban: 0.0011630678036913327\n","Age vs Urban: 0.0018615414395956045\n","Annual Income vs Urban: 0.00035958982244662334\n","Number of Children vs Urban: -0.00011214645432198825\n","Health Score vs Urban: -0.0038950438100415637\n","Previous Claims vs Urban: -0.001431964604716356\n","Vehicle Age vs Urban: 0.0013126741743393162\n","Credit Score vs Urban: -0.0019813387828195635\n","Insurance Duration vs Urban: 0.0035163085045266003\n","Premium Amount vs Urban: 0.001529031735758216\n","Policy_Age_Days vs Urban: 0.00040874628718526985\n","Smoking Status Encoded vs Urban: 0.0011043259706428816\n","id vs Comprehensive: 0.0001355309191765383\n","Age vs Comprehensive: 0.0010513294259612816\n","Annual Income vs Comprehensive: -0.0019532455534755316\n","Number of Children vs Comprehensive: -0.0011626429154546142\n","Health Score vs Comprehensive: -0.0007458476422183294\n","Previous Claims vs Comprehensive: 0.00168775447625721\n","Vehicle Age vs Comprehensive: -0.0013199509955600696\n","Credit Score vs Comprehensive: -0.0002633639620910799\n","Insurance Duration vs Comprehensive: -0.0011722650322107306\n","Premium Amount vs Comprehensive: 0.002136249290610793\n","Policy_Age_Days vs Comprehensive: 0.001415736530765529\n","Smoking Status Encoded vs Comprehensive: -0.0011384944726733925\n","id vs Premium: -0.00022100856096183525\n","Age vs Premium: -0.0010605272339345257\n","Annual Income vs Premium: 0.0009004523432829517\n","Number of Children vs Premium: -0.00021603424650061822\n","Health Score vs Premium: -0.002570246939762108\n","Previous Claims vs Premium: -0.0026053880878229474\n","Vehicle Age vs Premium: 0.0020744631852916817\n","Credit Score vs Premium: -0.0016530873385217153\n","Insurance Duration vs Premium: -0.0001719484181062405\n","Premium Amount vs Premium: -0.0013751707145711403\n","Policy_Age_Days vs Premium: -0.0021636007252291253\n","Smoking Status Encoded vs Premium: -6.004164043614749e-05\n","id vs Good: 0.0017968168671440988\n","Age vs Good: 0.00038364911973022873\n","Annual Income vs Good: -0.001395138126313431\n","Number of Children vs Good: 0.0003096344283959102\n","Health Score vs Good: -0.00034611999978263665\n","Previous Claims vs Good: 0.0008502663952515067\n","Vehicle Age vs Good: -0.001036697275939039\n","Credit Score vs Good: -0.004456504244017792\n","Insurance Duration vs Good: 0.0003387616980212284\n","Premium Amount vs Good: 0.0002642449180091094\n","Policy_Age_Days vs Good: -0.0009773481130173095\n","Smoking Status Encoded vs Good: -0.001283027742888445\n","id vs Poor: -0.0007470047111536027\n","Age vs Poor: -0.0004597106917409184\n","Annual Income vs Poor: -0.003994823867108519\n","Number of Children vs Poor: -0.0017628994377212651\n","Health Score vs Poor: -0.0015548933167507718\n","Previous Claims vs Poor: -0.005462085476542101\n","Vehicle Age vs Poor: 0.0025697282092431613\n","Credit Score vs Poor: -0.00042294294369541255\n","Insurance Duration vs Poor: 0.001027484751332248\n","Premium Amount vs Poor: -0.0033642586115780024\n","Policy_Age_Days vs Poor: 0.0031736612385186254\n","Smoking Status Encoded vs Poor: 6.654296823168175e-05\n","id vs Smoking Status: 0.0005450857511852445\n","Age vs Smoking Status: 7.120933824767816e-05\n","Annual Income vs Smoking Status: 0.0018366630882670966\n","Number of Children vs Smoking Status: 0.00019549452495091802\n","Health Score vs Smoking Status: 0.0016003535915102615\n","Previous Claims vs Smoking Status: 0.000981394962599775\n","Vehicle Age vs Smoking Status: 0.0014486698551966584\n","Credit Score vs Smoking Status: -0.0007917115849195108\n","Insurance Duration vs Smoking Status: 0.00041944261045833136\n","Premium Amount vs Smoking Status: -0.00012878395474512003\n","Policy_Age_Days vs Smoking Status: 0.001929665978956282\n","Smoking Status Encoded vs Smoking Status: -1.0\n","id vs Monthly: 0.0014130677979773867\n","Age vs Monthly: -0.0006347420105771654\n","Annual Income vs Monthly: -0.0010982620513047274\n","Number of Children vs Monthly: -0.0005327658094634311\n","Health Score vs Monthly: -0.0024238794892247222\n","Previous Claims vs Monthly: 0.0006482440399124182\n","Vehicle Age vs Monthly: 0.002457077308900813\n","Credit Score vs Monthly: -0.00013888021355323236\n","Insurance Duration vs Monthly: -0.00020507314893258055\n","Premium Amount vs Monthly: 0.0007707653766071442\n","Policy_Age_Days vs Monthly: 0.0019829555011912953\n","Smoking Status Encoded vs Monthly: -4.899819728929184e-05\n","id vs Rarely: -0.0022613410974818113\n","Age vs Rarely: -0.0003982452270772881\n","Annual Income vs Rarely: 0.0034471409103096983\n","Number of Children vs Rarely: 0.0036094261731393702\n","Health Score vs Rarely: 0.0009311027932500568\n","Previous Claims vs Rarely: -0.0010438153392441172\n","Vehicle Age vs Rarely: -0.002502567210839779\n","Credit Score vs Rarely: 3.374745607768912e-05\n","Insurance Duration vs Rarely: -0.0019319982601308681\n","Premium Amount vs Rarely: -0.002326574317834105\n","Policy_Age_Days vs Rarely: 0.0009643997036574496\n","Smoking Status Encoded vs Rarely: 0.0009397483309216962\n","id vs Weekly: -5.363806655501239e-06\n","Age vs Weekly: 0.0004559429456286251\n","Annual Income vs Weekly: 0.0007979484765377231\n","Number of Children vs Weekly: -0.002055136369550404\n","Health Score vs Weekly: 4.824549289969084e-05\n","Previous Claims vs Weekly: 0.0006964404821003901\n","Vehicle Age vs Weekly: 0.000858064488996469\n","Credit Score vs Weekly: -0.0002917996542374275\n","Insurance Duration vs Weekly: 0.0003963087009045379\n","Premium Amount vs Weekly: 0.0006785807912258907\n","Policy_Age_Days vs Weekly: -0.0022071691127550776\n","Smoking Status Encoded vs Weekly: 0.00041075392165820573\n","id vs Condo: 0.0011395177094425694\n","Age vs Condo: -0.000356044011531248\n","Annual Income vs Condo: -0.0002677942513713827\n","Number of Children vs Condo: -0.0014324468836083941\n","Health Score vs Condo: 0.0014373021372072253\n","Previous Claims vs Condo: -0.0012381640838111253\n","Vehicle Age vs Condo: 0.0014247157951723715\n","Credit Score vs Condo: 0.0008893730434192784\n","Insurance Duration vs Condo: -0.0006548892561250075\n","Premium Amount vs Condo: 0.00032872935603571734\n","Policy_Age_Days vs Condo: -0.0007550609448688852\n","Smoking Status Encoded vs Condo: 0.0002672977755328698\n","id vs House: 0.00020385104106605467\n","Age vs House: 0.0009394554066038673\n","Annual Income vs House: 0.0006533558384984353\n","Number of Children vs House: 0.0013079504063203302\n","Health Score vs House: -0.0013974544567535634\n","Previous Claims vs House: 0.0009924723380400004\n","Vehicle Age vs House: -0.0038087003920513705\n","Credit Score vs House: 0.0007383446437143838\n","Insurance Duration vs House: 0.00021511255060980433\n","Premium Amount vs House: 0.000494986549803765\n","Policy_Age_Days vs House: 0.0006882942433389272\n","Smoking Status Encoded vs House: 0.0001005169891516072\n"]}],"source":"from scipy.stats import pointbiserialr\n\ndef correlate_categorical_numeric(df):\n    correlations = {}\n\n    # Identify numeric and categorical columns\n    numeric_cols = df.select_dtypes(include=['int64', 'float64']).columns\n    categorical_cols = df.select_dtypes(include=['object', 'category']).columns\n\n    for cat_col in categorical_cols:\n        # Encode categorical column to numeric\n        if df[cat_col].nunique() == 2:\n            # Binary categorical → map to 0/1\n            mapping = {val: idx for idx, val in enumerate(df[cat_col].unique())}\n            encoded = df[cat_col].map(mapping)\n            \n            for num_col in numeric_cols:\n                corr, p = pointbiserialr(encoded, df[num_col])\n                correlations[f\"{num_col} vs {cat_col}\"] = corr\n        else:\n            # Multi-class → one-hot encode temporarily\n            encoded_df = pd.get_dummies(df[cat_col], drop_first=True)\n            for dummy_col in encoded_df.columns:\n                for num_col in numeric_cols:\n                    corr = df[num_col].corr(encoded_df[dummy_col])\n                    correlations[f\"{num_col} vs {dummy_col}\"] = corr\n\n    return correlations\n\n# Example usage\ncorrelations = correlate_categorical_numeric(df_train)\nfor k, v in correlations.items():\n    print(f\"{k}: {v}\")"},{"cell_type":"code","execution_count":45,"id":"72db01ce-6b84-4300-96d1-4541616cef3f","metadata":{},"outputs":[{"data":{"image/png":"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","text/plain":["<Figure size 640x480 with 1 Axes>"]},"metadata":{},"output_type":"display_data"}],"source":"sns.histplot(df_train['Vehicle Age'])\nplt.show()"},{"cell_type":"code","execution_count":47,"id":"e5424aea-fb50-411f-b0c9-617b57b58d3d","metadata":{},"outputs":[{"data":{"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>id</th>\n","      <th>Age</th>\n","      <th>Annual Income</th>\n","      <th>Marital Status</th>\n","      <th>Number of Children</th>\n","      <th>Education Level</th>\n","      <th>Occupation</th>\n","      <th>Health Score</th>\n","      <th>Location</th>\n","      <th>Policy Type</th>\n","      <th>...</th>\n","      <th>Exercise Frequency</th>\n","      <th>Property Type</th>\n","      <th>Premium Amount</th>\n","      <th>Policy_Year</th>\n","      <th>Policy_Month</th>\n","      <th>Policy_Quarter</th>\n","      <th>Policy_DayOfYear</th>\n","      <th>Policy_DayOfWeek</th>\n","      <th>Policy_Age_Days</th>\n","      <th>Smoking Status Encoded</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>476061</td>\n","      <td>53.0</td>\n","      <td>6.091910</td>\n","      <td>Single</td>\n","      <td>0.0</td>\n","      <td>PhD</td>\n","      <td>Employed</td>\n","      <td>34.727739</td>\n","      <td>Rural</td>\n","      <td>Premium</td>\n","      <td>...</td>\n","      <td>Rarely</td>\n","      <td>Condo</td>\n","      <td>2.685987</td>\n","      <td>2023</td>\n","      <td>7</td>\n","      <td>3</td>\n","      <td>207</td>\n","      <td>2</td>\n","      <td>801</td>\n","      <td>1</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>186666</td>\n","      <td>29.0</td>\n","      <td>6.265551</td>\n","      <td>Married</td>\n","      <td>2.0</td>\n","      <td>Bachelor's</td>\n","      <td>Self-Employed</td>\n","      <td>29.684751</td>\n","      <td>Suburban</td>\n","      <td>Premium</td>\n","      <td>...</td>\n","      <td>Weekly</td>\n","      <td>Condo</td>\n","      <td>2.227834</td>\n","      <td>2024</td>\n","      <td>2</td>\n","      <td>1</td>\n","      <td>43</td>\n","      <td>0</td>\n","      <td>600</td>\n","      <td>0</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>742870</td>\n","      <td>25.0</td>\n","      <td>5.136855</td>\n","      <td>Single</td>\n","      <td>2.0</td>\n","      <td>Master's</td>\n","      <td>Self-Employed</td>\n","      <td>12.953414</td>\n","      <td>Urban</td>\n","      <td>Basic</td>\n","      <td>...</td>\n","      <td>Monthly</td>\n","      <td>Condo</td>\n","      <td>0.352669</td>\n","      <td>2024</td>\n","      <td>4</td>\n","      <td>2</td>\n","      <td>108</td>\n","      <td>2</td>\n","      <td>535</td>\n","      <td>0</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>173677</td>\n","      <td>56.0</td>\n","      <td>6.589426</td>\n","      <td>Single</td>\n","      <td>1.0</td>\n","      <td>Master's</td>\n","      <td>Self-Employed</td>\n","      <td>47.739073</td>\n","      <td>Rural</td>\n","      <td>Basic</td>\n","      <td>...</td>\n","      <td>Weekly</td>\n","      <td>House</td>\n","      <td>2.548228</td>\n","      <td>2023</td>\n","      <td>6</td>\n","      <td>2</td>\n","      <td>165</td>\n","      <td>2</td>\n","      <td>843</td>\n","      <td>0</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>215979</td>\n","      <td>26.0</td>\n","      <td>4.467350</td>\n","      <td>Married</td>\n","      <td>2.0</td>\n","      <td>PhD</td>\n","      <td>Self-Employed</td>\n","      <td>22.087884</td>\n","      <td>Rural</td>\n","      <td>Comprehensive</td>\n","      <td>...</td>\n","      <td>Monthly</td>\n","      <td>House</td>\n","      <td>2.885021</td>\n","      <td>2023</td>\n","      <td>4</td>\n","      <td>2</td>\n","      <td>96</td>\n","      <td>3</td>\n","      <td>912</td>\n","      <td>1</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>5 rows × 26 columns</p>\n","</div>"],"text/plain":["       id   Age  Annual Income Marital Status  Number of Children  \\\n","0  476061  53.0       6.091910         Single                 0.0   \n","1  186666  29.0       6.265551        Married                 2.0   \n","2  742870  25.0       5.136855         Single                 2.0   \n","3  173677  56.0       6.589426         Single                 1.0   \n","4  215979  26.0       4.467350        Married                 2.0   \n","\n","  Education Level     Occupation  Health Score  Location    Policy Type  ...  \\\n","0             PhD       Employed     34.727739     Rural        Premium  ...   \n","1      Bachelor's  Self-Employed     29.684751  Suburban        Premium  ...   \n","2        Master's  Self-Employed     12.953414     Urban          Basic  ...   \n","3        Master's  Self-Employed     47.739073     Rural          Basic  ...   \n","4             PhD  Self-Employed     22.087884     Rural  Comprehensive  ...   \n","\n","   Exercise Frequency  Property Type  Premium Amount  Policy_Year  \\\n","0              Rarely          Condo        2.685987         2023   \n","1              Weekly          Condo        2.227834         2024   \n","2             Monthly          Condo        0.352669         2024   \n","3              Weekly          House        2.548228         2023   \n","4             Monthly          House        2.885021         2023   \n","\n","  Policy_Month Policy_Quarter Policy_DayOfYear Policy_DayOfWeek  \\\n","0            7              3              207                2   \n","1            2              1               43                0   \n","2            4              2              108                2   \n","3            6              2              165                2   \n","4            4              2               96                3   \n","\n","   Policy_Age_Days  Smoking Status Encoded  \n","0              801                       1  \n","1              600                       0  \n","2              535                       0  \n","3              843                       0  \n","4              912                       1  \n","\n","[5 rows x 26 columns]"]},"execution_count":47,"metadata":{},"output_type":"execute_result"}],"source":"df_train.head()"},{"cell_type":"code","execution_count":49,"id":"83f5caa3-a5ed-4f36-9ad3-eab179e1d306","metadata":{},"outputs":[],"source":"def featured_engineered(df):\n    df = df.copy()\n    if 'Annual Income' in df.columns and 'Number of Children' in df.columns :\n        df['Income_per_Child'] = df['Annual Income'] / (df['Number of Children'] + 1)\n\n    if 'Premium Amount' in df.columns and 'Annual Income' in df.columns :\n        df['Income_Premium_Ratio'] = df['Premium Amount'] / df['Annual Income']\n\n    if 'Previous Claims' in df.columns and 'Insurance Duration' in df.columns:\n        df['Claim_Frequency'] = df['Previous Claims'] / (df['Insurance Duration'] + 1e-6)\n\n    if 'Age' in df.columns and 'Vehicle Age' in df.columns :\n        df['Age_Vehicle_Age_Ratio'] = df['Age'] / (df['Vehicle Age'] + 1e-6)\n\n    if 'Age' in df.columns and 'Smoking Status' in df.columns:\n        df['Age_x_Smoking'] = df['Age'] * (df['Smoking Status'] == 'Yes').astype(int)\n    if 'Health Score' in df.columns and 'Exercise Frequency' in df.columns:\n        df['Health_x_Exercise'] = df['Health Score'] * df['Exercise Frequency'].map({'Daily':4,'Weekly':3,'Monthly':2,'Rarely':1,'Never':0}).fillna(0)\n    if 'Age' in df.columns:\n        df['Age_Group'] = pd.cut(df['Age'], bins=[0,25,35,45,55,65,100], labels=[1,2,3,4,5,6], right=False)\n        df['Age_Group'] = df['Age_Group'].astype(float).fillna(0).astype(int)\n        \n    return df\n    "},{"cell_type":"code","execution_count":50,"id":"c86de15f-b134-45cb-b315-3fb6cb7cb61c","metadata":{},"outputs":[],"source":"df_train = featured_engineered(df_train)\ndf_test = featured_engineered(df_test)"},{"cell_type":"code","execution_count":51,"id":"47d3480a-0ea2-4a18-b228-14aeab63e5ce","metadata":{},"outputs":[],"source":"# One Hot Encoding\ndf_train = pd.get_dummies(df_train, columns=categorical_features, drop_first=True)\ndf_test = pd.get_dummies(df_test, columns=categorical_features, drop_first=True)\n"},{"cell_type":"code","execution_count":52,"id":"d0ba3c4b-e47a-4353-97b7-d5a73c42730a","metadata":{},"outputs":[],"source":"X = df_train.drop(columns=['id','Premium Amount'])\ny = df_train['Premium Amount']\nX_test = df_test.drop(columns=['id'])\ntest_ids = df_test['id'].values"},{"cell_type":"code","execution_count":54,"id":"b01ab1d8-1840-469a-8117-df6c1dc1d7d5","metadata":{},"outputs":[{"data":{"image/png":"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","text/plain":["<Figure size 640x480 with 1 Axes>"]},"metadata":{},"output_type":"display_data"}],"source":"sns.histplot(y,kde=True)\nplt.plot(1)\nplt.title('Premium Amount Distribution')\nplt.show()"},{"cell_type":"code","execution_count":56,"id":"62eaf65b-8382-4604-a5e1-322e7a717e03","metadata":{},"outputs":[{"data":{"text/plain":["(720000, 41)"]},"execution_count":56,"metadata":{},"output_type":"execute_result"}],"source":"X.shape"},{"cell_type":"code","execution_count":61,"id":"2af447ca-6af2-4ad9-82f4-80b5e24fcfc2","metadata":{},"outputs":[],"source":"# --- 1. Define X, y, X_test (Assuming OHE has been applied) ---\n# NOTE: Ensure df_train and df_test are aligned on columns before running this.\n# You MUST ensure df_train and df_test have the same columns, \n# especially after OHE, before defining X and X_test.\n\n# Align columns (Crucial Step after OHE)\ntrain_cols = set(df_train.columns)\ntest_cols = set(df_test.columns)\n\ncommon_cols = list(train_cols.intersection(test_cols))\nX = df_train[common_cols].drop(columns=['Premium Amount', 'id'], errors='ignore')\ny = df_train['Premium Amount']\nX_test = df_test[common_cols].drop(columns=['id', 'Premium Amount'], errors='ignore')\ntest_ids = df_test['id'].values \n\n# After OHE, LightGBM treats OHE columns as numerical, \n# so we pass an empty list to avoid errors.\ncurrent_cat_features = []"},{"cell_type":"code","execution_count":62,"id":"bb7d10a2-13d3-4088-b1f4-4f233127cfa8","metadata":{},"outputs":[{"name":"stdout","output_type":"stream","text":["Starting Gating Network Training with OHE features...\n","Gating Network Accuracy: 0.7000\n","--------------------------------------------------\n"]}],"source":"# --- 2. Gating Network (Classifier) Setup ---\n\n# Define segmentation target and Stratified K-Fold setup\nTHRESHOLD = 1.0\ny_class = (y > THRESHOLD).astype(int)\n\n# Create 10 target bins for Stratified K-Fold stability\ny_bins = pd.qcut(y.rank(method='first'), q=10, labels=False, duplicates='drop')\ny_bins = y_bins.fillna(0).astype(int)\nskf = StratifiedKFold(n_splits=5, shuffle=True, random_state=42)\n\nclassifier_params = {\n    'objective':'binary', 'metric':'binary_logloss', 'boosting_type':'gbdt',\n    'num_leaves':63, 'max_depth':8, 'learning_rate':0.05,\n    'min_data_in_leaf':100, 'feature_fraction':0.8, 'bagging_fraction':0.8,\n    'bagging_freq':3, 'lambda_l1':0.5, 'lambda_l2':2.0, 'verbose':-1, \n    'random_state':42, 'force_col_wise':True\n}\n\noof_class_probs = np.zeros(len(X))\ntest_class_probs = np.zeros(len(X_test))\n\nprint(\"Starting Gating Network Training with OHE features...\")\n\nfor fold, (train_idx, val_idx) in enumerate(skf.split(X, y_bins), 1):\n    X_train_fold, y_class_fold = X.iloc[train_idx], y_class.iloc[train_idx]\n    X_val_fold, y_class_val = X.iloc[val_idx], y_class.iloc[val_idx]\n    \n    # NOTE: current_cat_features is [] here\n    train_data = lgb.Dataset(X_train_fold, label=y_class_fold, categorical_feature=current_cat_features)\n    val_data = lgb.Dataset(X_val_fold, label=y_class_val, categorical_feature=current_cat_features, reference=train_data)\n    \n    clf = lgb.train(\n        classifier_params, train_data, num_boost_round=1000, valid_sets=[val_data],\n        callbacks=[lgb.early_stopping(50, verbose=False)]\n    )\n    \n    # Store probability (P_High) for Soft MoE\n    oof_class_probs[val_idx] = clf.predict(X_val_fold, num_iteration=clf.best_iteration)\n    test_class_probs += clf.predict(X_test, num_iteration=clf.best_iteration) / skf.n_splits\n\nprint(f\"Gating Network Accuracy: {accuracy_score(y_class, (oof_class_probs>0.5).astype(int)):.4f}\")\nprint(\"-\" * 50)"},{"cell_type":"code","execution_count":67,"id":"a0b1a4f3-6846-4a9a-9299-3385939a4fde","metadata":{},"outputs":[{"name":"stdout","output_type":"stream","text":["Starting Expert Network Training...\n","Expert Networks Training Complete.\n","Combined Soft MoE OOF RMSE: 0.7123\n","Final Soft MoE (OHE) Predictions saved to 'submission_soft_moe_ohe.csv'.\n"]}],"source":"# --- NOTE: This code assumes the necessary variables (X, y, X_test, oof_class_probs, test_class_probs, etc.)\n#           are available from the Gating Network stage and that OHE was used (current_cat_features = []).\n\n# --- 1. Data Splits and Initialization (Stage 2 Setup) ---\nTHRESHOLD = 1.0\nlow_premium_mask = y <= THRESHOLD\nhigh_premium_mask = y > THRESHOLD\n\nX_low = X[low_premium_mask]\ny_low = y[low_premium_mask]\nX_high = X[high_premium_mask]\ny_high = y[high_premium_mask]\n\nskf_low = KFold(n_splits=5, shuffle=True, random_state=42)\nskf_high = KFold(n_splits=5, shuffle=True, random_state=42)\n\nparams_expert = {\n    'objective':'regression', 'metric':'rmse', 'boosting_type':'gbdt',\n    'num_leaves':100, 'max_depth':10, 'learning_rate':0.03,\n    'min_data_in_leaf':50, 'feature_fraction':0.9, 'bagging_fraction':0.85,\n    'bagging_freq':5, 'lambda_l1':0.3, 'lambda_l2':1.5, 'min_gain_to_split':0.01,\n    'verbose':-1, 'force_col_wise':True\n}\n\noof_low_pred = np.zeros(len(y_low))\ntest_low_pred = np.zeros(len(X_test))\noof_high_pred = np.zeros(len(y_high))\ntest_high_pred = np.zeros(len(X_test))\ncurrent_cat_features = [] \n\nprint(\"Starting Expert Network Training...\")\n\n# --- Expert 1: Low Premium Model Training ---\nfor train_idx, val_idx in skf_low.split(X_low):\n    X_train_fold, y_train_fold = X_low.iloc[train_idx], y_low.iloc[train_idx]\n    X_val_fold, y_val_fold = X_low.iloc[val_idx], y_low.iloc[val_idx]\n    train_data = lgb.Dataset(X_train_fold, label=y_train_fold, categorical_feature=current_cat_features)\n    val_data = lgb.Dataset(X_val_fold, label=y_val_fold, categorical_feature=current_cat_features, reference=train_data)\n    model_low = lgb.train(params_expert.copy(), train_data, num_boost_round=3000, valid_sets=[val_data], callbacks=[lgb.early_stopping(100, verbose=False)])\n    oof_low_pred[val_idx] = np.nan_to_num(model_low.predict(X_val_fold, num_iteration=model_low.best_iteration), nan=y_low.median())\n    test_low_pred += np.nan_to_num(model_low.predict(X_test, num_iteration=model_low.best_iteration), nan=y_low.median()) / skf_low.n_splits\n\n# --- Expert 2: High Premium Model Training ---\nparams_high = params_expert.copy()\nparams_high['random_state'] = 123 \n\nfor train_idx, val_idx in skf_high.split(X_high):\n    X_train_fold, y_train_fold = X_high.iloc[train_idx], y_high.iloc[train_idx]\n    X_val_fold, y_val_fold = X_high.iloc[val_idx], y_high.iloc[val_idx]\n    train_data = lgb.Dataset(X_train_fold, label=y_train_fold, categorical_feature=current_cat_features)\n    val_data = lgb.Dataset(X_val_fold, label=y_val_fold, categorical_feature=current_cat_features, reference=train_data)\n    model_high = lgb.train(params_high, train_data, num_boost_round=3000, valid_sets=[val_data], callbacks=[lgb.early_stopping(100, verbose=False)])\n    oof_high_pred[val_idx] = np.nan_to_num(model_high.predict(X_val_fold, num_iteration=model_high.best_iteration), nan=y_high.median())\n    test_high_pred += np.nan_to_num(model_high.predict(X_test, num_iteration=model_high.best_iteration), nan=y_high.median()) / skf_high.n_splits\n\nprint(\"Expert Networks Training Complete.\")\n\n# --- 3. Soft MoE Combination and RMSE Calculation (Corrected Logic) ---\n\n# Gating Network Weights\nP_high = test_class_probs \nP_low = 1.0 - P_high\n\n# Final Test Predictions (Weighted Average)\ntest_pred_final_moe = (P_low * test_low_pred) + (P_high * test_high_pred)\ntest_pred_final_moe = np.clip(test_pred_final_moe, 0, None)\n\n# --- Corrected OOF Combination for RMSE ---\n\n# 1. Align predictions and probabilities to the full index using Series\noof_predictions_low = pd.Series(oof_low_pred, index=X_low.index)\noof_predictions_high = pd.Series(oof_high_pred, index=X_high.index)\noof_probs = pd.Series(oof_class_probs, index=X.index)\ny_series = pd.Series(y.values, index=X.index)\n\n# 2. Reindex OOF predictions to the full size (filling NaNs with 0)\nOOF_low_aligned = oof_predictions_low.reindex(X.index).fillna(0)\nOOF_high_aligned = oof_predictions_high.reindex(X.index).fillna(0)\n\n# 3. Get OOF Weights\nP_high_oof_aligned = oof_probs.loc[X.index]\nP_low_oof_aligned = 1.0 - P_high_oof_aligned\n\n# 4. Final Soft MoE OOF Prediction\noof_final = (P_low_oof_aligned * OOF_low_aligned) + (P_high_oof_aligned * OOF_high_aligned)\noof_final = np.clip(oof_final, 0, None) # Ensure non-negative\n\n# 5. Calculate RMSE (as requested)\ncombined_rmse_moe = np.sqrt(mean_squared_error(y_series, oof_final.values))\nprint(f\"Combined Soft MoE OOF RMSE: {combined_rmse_moe:.4f}\")\n\n# --- Create Final Submission File ---\nsubmission_moe = pd.DataFrame({'id': test_ids, 'Premium Amount': test_pred_final_moe})\nsubmission_moe.to_csv('submission_soft_moe_ohe.csv', index=False)\nprint(\"Final Soft MoE (OHE) Predictions saved to 'submission_soft_moe_ohe.csv'.\")"},{"cell_type":"code","execution_count":null,"id":"bad26c33-34bc-483f-aa23-966b4cbfc730","metadata":{},"outputs":[],"source":""}],"metadata":{"kernelspec":{"display_name":"Python 3 (ipykernel)","language":"python","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.12.7"}},"nbformat":4,"nbformat_minor":5}