{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.14","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":84896,"databundleVersionId":10305135,"sourceType":"competition"},{"sourceId":9178166,"sourceType":"datasetVersion","datasetId":5547076}],"dockerImageVersionId":30804,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"![Insurance1.jpg](attachment:38c41682-521b-4069-af29-bcc50d1a5390.jpg)\n\n# <p style=\"background-color:#E34234; color:white; padding:10px 20px; width:95%; border-radius:15px; text-align:center;margin: 0 auto; box-shadow:0 4px 8px rgba(0, 0, 0, 0.1);font-weight: bold;\">Introduction</p>\n\nThis notebook will begin with a comprehensive <strong>Exploratory Data Analysis (EDA)</strong> on the original dataset. Based on the findings, both the synthetic (competition) data and the original data will be processed. Finally, a predictive model will be built and trained using the processed datasets.","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19"},"attachments":{"38c41682-521b-4069-af29-bcc50d1a5390.jpg":{"image/jpeg":"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"}}},{"cell_type":"code","source":"import numpy as np\nimport pandas as pd\nfrom scipy.stats import kruskal\n\nimport matplotlib \nimport matplotlib.pyplot as plt\nimport seaborn as sns\n\nfrom lightgbm import LGBMRegressor\nfrom sklearn.model_selection import cross_validate\nfrom sklearn.metrics import mean_squared_log_error\n\nimport warnings\nwarnings.filterwarnings('ignore')\n\ncmap = matplotlib.colormaps.get_cmap('rocket_r')","metadata":{"trusted":true,"_kg_hide-input":true,"execution":{"iopub.status.busy":"2024-12-17T16:00:47.064128Z","iopub.execute_input":"2024-12-17T16:00:47.06482Z","iopub.status.idle":"2024-12-17T16:00:51.33914Z","shell.execute_reply.started":"2024-12-17T16:00:47.064781Z","shell.execute_reply":"2024-12-17T16:00:51.33813Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# <p style=\"background-color:#E34234; color:white; padding:10px 20px; width:95%; border-radius:15px; text-align:center; margin: 0 auto; box-shadow:0 4px 8px rgba(0, 0, 0, 0.1);font-weight: bold;\">Original Data Overview</p>","metadata":{}},{"cell_type":"code","source":"original  = pd.read_csv('/kaggle/input/insurance-premium-prediction/Insurance Premium Prediction Dataset.csv')\noriginal.info()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T16:00:57.795774Z","iopub.execute_input":"2024-12-17T16:00:57.796292Z","iopub.status.idle":"2024-12-17T16:00:59.440964Z","shell.execute_reply.started":"2024-12-17T16:00:57.796237Z","shell.execute_reply":"2024-12-17T16:00:59.439935Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"original['Policy Start Date'] = pd.DatetimeIndex(original['Policy Start Date'])\noriginal['year'] = original['Policy Start Date'].dt.year\noriginal['month'] = original['Policy Start Date'].dt.month\noriginal['date'] = original['Policy Start Date'].dt.day\noriginal['day'] = original['Policy Start Date'].dt.weekday","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T16:00:59.442445Z","iopub.execute_input":"2024-12-17T16:00:59.442756Z","iopub.status.idle":"2024-12-17T16:00:59.635789Z","shell.execute_reply.started":"2024-12-17T16:00:59.442724Z","shell.execute_reply":"2024-12-17T16:00:59.634772Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"original.head()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T16:01:01.558201Z","iopub.execute_input":"2024-12-17T16:01:01.558591Z","iopub.status.idle":"2024-12-17T16:01:01.591694Z","shell.execute_reply.started":"2024-12-17T16:01:01.558554Z","shell.execute_reply":"2024-12-17T16:01:01.590677Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"original.describe().style.background_gradient()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T16:01:02.853004Z","iopub.execute_input":"2024-12-17T16:01:02.853781Z","iopub.status.idle":"2024-12-17T16:01:03.111149Z","shell.execute_reply.started":"2024-12-17T16:01:02.853741Z","shell.execute_reply":"2024-12-17T16:01:03.11014Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"original.describe(include='O')","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T16:01:03.372156Z","iopub.execute_input":"2024-12-17T16:01:03.372875Z","iopub.status.idle":"2024-12-17T16:01:03.76077Z","shell.execute_reply.started":"2024-12-17T16:01:03.372818Z","shell.execute_reply":"2024-12-17T16:01:03.759813Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"## <p style=\"background-color:#E34234; color:white; padding:10px 20px; width:95%; border-radius:15px; text-align:center; margin: 0 auto; box-shadow:0 4px 8px rgba(0, 0, 0, 0.1);font-weight: bold;\">Missing Values</p>","metadata":{}},{"cell_type":"code","source":"missing = original.isna().sum().reset_index()\nmissing.columns = ['features','missing_count']\nmissing['percentage'] = missing['missing_count']/original.shape[0]*100\n(missing[missing['missing_count']>0]\n .sort_values(by='missing_count',ascending=False)\n .reset_index(drop=True)\n .style.background_gradient())","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T16:01:04.331939Z","iopub.execute_input":"2024-12-17T16:01:04.332607Z","iopub.status.idle":"2024-12-17T16:01:04.48167Z","shell.execute_reply.started":"2024-12-17T16:01:04.332568Z","shell.execute_reply":"2024-12-17T16:01:04.480725Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"#### **Observation:** \n- **Total 9 features(except Premium Amount)** have NA values.\n- Need to investigate whether there is any relationship between the missing values in these features and the `Premium Amount`.\n- **Premium Amount** itself has 1,841 missing values, which will be dropped from the analysis as it is the target variable.","metadata":{}},{"cell_type":"code","source":"original.dropna(subset='Premium Amount',inplace=True)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T16:01:05.607178Z","iopub.execute_input":"2024-12-17T16:01:05.607976Z","iopub.status.idle":"2024-12-17T16:01:05.6439Z","shell.execute_reply.started":"2024-12-17T16:01:05.607933Z","shell.execute_reply":"2024-12-17T16:01:05.642875Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# <p style=\"background-color:#E34234; color:white; padding:10px 20px; width:95%; border-radius:15px; text-align:center; margin: 0 auto; box-shadow:0 4px 8px rgba(0, 0, 0, 0.1);font-weight: bold;\">Exploratory Data Analysis</p>","metadata":{}},{"cell_type":"code","source":"discrete_feat = [feat for feat in original.select_dtypes(exclude='O') if original[feat].nunique()<20]\ncontinuous_feat = [feat for feat in original.select_dtypes(include='number') if feat not in discrete_feat]\ncategorical_feat = [feat for feat in original.select_dtypes(include='O')]\ncontinuous_feat.remove('Premium Amount')","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T16:01:06.681165Z","iopub.execute_input":"2024-12-17T16:01:06.682057Z","iopub.status.idle":"2024-12-17T16:01:06.861347Z","shell.execute_reply.started":"2024-12-17T16:01:06.682016Z","shell.execute_reply":"2024-12-17T16:01:06.860282Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"## <p style=\"background-color:#E34234; color:white; padding:10px 20px; width:95%; border-radius:15px; text-align:center; margin: 0 auto; box-shadow:0 4px 8px rgba(0, 0, 0, 0.1);font-weight: bold;\">Discrete Feature Analysis</p>","metadata":{}},{"cell_type":"markdown","source":"Lets check the data distribution for the discrete features","metadata":{}},{"cell_type":"code","source":"for feat in discrete_feat:\n    agg_data = original.groupby(feat).size()\n    norm = plt.Normalize(agg_data.min(), agg_data.max())\n    \n    plt.figure(figsize=(8,3))\n    \n    plt.subplot(1,2,1)\n    plt.pie(agg_data,\n            autopct='%.2f',\n            labels=agg_data.index,\n            colors=sns.color_palette(\"rocket_r\", len(agg_data)+2),\n            radius=1.25)\n    plt.pie([1],radius=0.65,colors=['white'])\n    \n    plt.subplot(1,2,2)\n    bars = sns.barplot(x=agg_data.index,y=agg_data)\n    bars.bar_label(bars.containers[0], \n                   rotation=90,fontsize=10,\n                   label_type='center',\n                   labels=[f'{x/1e3:.1f}K' for x in agg_data])\n    \n    for bar in bars.patches:\n        height = bar.get_height()\n        bar.set_facecolor(cmap(norm(height)*0.75))\n    \n    plt.legend([])\n    plt.yticks([])\n    plt.ylabel('')\n    plt.xlabel('')\n    plt.box(False)\n    \n    plt.suptitle(feat+' Distribution')\n    plt.tight_layout()\n    plt.show()","metadata":{"trusted":true,"_kg_hide-input":true,"execution":{"iopub.status.busy":"2024-12-17T06:25:33.042318Z","iopub.execute_input":"2024-12-17T06:25:33.043592Z","iopub.status.idle":"2024-12-17T06:25:35.860671Z","shell.execute_reply.started":"2024-12-17T06:25:33.043544Z","shell.execute_reply":"2024-12-17T06:25:35.859636Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"Lets check discrete features relation with the Premium Amount","metadata":{}},{"cell_type":"code","source":"for feat in discrete_feat:\n    agg_data = original.groupby(feat)['Premium Amount']\n    groups = [group.values for _, group in agg_data]\n    _, p_value = kruskal(*groups)\n    \n    plt.figure(figsize=(8,3))\n    sns.boxplot(data=original, x=feat, y='Premium Amount',palette='rocket_r')\n    plt.xlabel(f'{feat} P Value :{p_value:0.2f}')\n    plt.box(False)\n    \n    plt.title('Premium Amount by '+feat)\n    plt.tight_layout()\n    plt.show()","metadata":{"trusted":true,"_kg_hide-input":true,"execution":{"iopub.status.busy":"2024-12-17T06:25:35.862615Z","iopub.execute_input":"2024-12-17T06:25:35.864354Z","iopub.status.idle":"2024-12-17T06:25:38.32982Z","shell.execute_reply.started":"2024-12-17T06:25:35.864277Z","shell.execute_reply":"2024-12-17T06:25:38.328665Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"#### **Observation:**\n- Since there’s no significant difference in the `Premium Amount` across the different categories of **Number of Dependents**, and the categories are reasonably well distributed, it suggests that the number of dependents may not be a strong factor influencing the Premium Amount\n- Despite the varying number of claims, the absence of a significant difference in `Premium Amount` suggests that **Previous Claims** does not have a clear effect on the premium, even though some categories have small sample sizes.\n- The distribution of **Insurance Duration** is evenly spread, but there is no significant relationship with the `Premium Amount`.\n- **year**,**month** and **day** are also well distributed but no significant association found.","metadata":{}},{"cell_type":"markdown","source":"Let's examine the impact of missing values in discrete features on the Premium Amount","metadata":{}},{"cell_type":"code","source":"for feat in discrete_feat: \n    if original[feat].isna().sum()>0:\n        data = original.copy()\n        data['is_na'] = original[feat].isna().astype(int)\n        agg_data = data.groupby('is_na')['Premium Amount']\n        groups = [group.values for _, group in agg_data]\n        _, p_value = kruskal(*groups)\n        \n        plt.figure(figsize=(8,3))\n        sns.boxplot(data=data, x='is_na', y='Premium Amount',palette='rocket_r')\n        plt.xlabel(f'{feat} NA P Value :{p_value:0.2f}')\n        plt.box(False)\n        \n        plt.title('Premium Amount by '+feat+' NA Values')\n        plt.tight_layout()\n        plt.show()","metadata":{"trusted":true,"_kg_hide-input":true,"execution":{"iopub.status.busy":"2024-12-17T06:25:38.331101Z","iopub.execute_input":"2024-12-17T06:25:38.331466Z","iopub.status.idle":"2024-12-17T06:25:39.250134Z","shell.execute_reply.started":"2024-12-17T06:25:38.331431Z","shell.execute_reply":"2024-12-17T06:25:39.248973Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"#### **Observation:**\n- **`NA`** values does not have any effect in `Premium Amount` for **Number of Dependents**, **Previous Claims**","metadata":{}},{"cell_type":"markdown","source":"## <p style=\"background-color:#E34234; color:white; padding:10px 20px; width:95%; border-radius:15px; text-align:center; margin: 0 auto; box-shadow:0 4px 8px rgba(0, 0, 0, 0.1);font-weight: bold;\">Continuous Feature Analysis</p>","metadata":{}},{"cell_type":"markdown","source":"Let see data distribution of continuous features","metadata":{}},{"cell_type":"code","source":"for feat in continuous_feat:\n    counts, bins = np.histogram(original[feat].dropna(), bins=50)\n    norm = plt.Normalize(counts.mean()-counts.std(), counts.max())\n    \n    plt.figure(figsize=(8,3))\n    \n    plt.subplot(1,2,1)\n    hist = sns.histplot(x=original[feat].dropna(),bins=50)\n    for patch in hist.patches:\n        height = patch.get_height()  \n        patch.set_facecolor(cmap(norm(height)*0.75))\n    plt.box(False)\n    plt.ylabel('')\n    \n    plt.subplot(1,2,2)\n    sns.boxplot(x=original[feat].dropna(),palette='rocket_r')\n    plt.box(False)\n    plt.ylabel('')\n    \n    plt.suptitle(feat+' Distribution')\n    plt.tight_layout()\n    plt.show()","metadata":{"trusted":true,"_kg_hide-input":true,"execution":{"iopub.status.busy":"2024-12-17T06:25:39.253124Z","iopub.execute_input":"2024-12-17T06:25:39.253651Z","iopub.status.idle":"2024-12-17T06:25:42.040701Z","shell.execute_reply.started":"2024-12-17T06:25:39.253597Z","shell.execute_reply":"2024-12-17T06:25:42.039379Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"Let's check continuous features correlation with Premium Amount","metadata":{}},{"cell_type":"code","source":"for feat in continuous_feat:\n    corr = original[['Premium Amount']].corrwith(original[feat])\n    plt.figure(figsize=(4,4))\n    # plt.subplot(121)\n    g = sns.JointGrid(data=original.sample(frac=0.1), x='Premium Amount', y=feat, space=0)\n    g.plot_joint(sns.kdeplot,fill=True,levels=20, cmap=\"rocket_r\")\n    g.plot_marginals(sns.histplot, color=\"tomato\", alpha=1, bins=25)\n    plt.box(False)\n    \n    plt.suptitle(f'Premium Amount X {feat} [Correlation:{corr.values[0]:0.2f}]')\n    plt.tight_layout()\n    plt.show()\n    ","metadata":{"trusted":true,"_kg_hide-input":true,"execution":{"iopub.status.busy":"2024-12-17T06:25:42.043411Z","iopub.execute_input":"2024-12-17T06:25:42.044148Z","iopub.status.idle":"2024-12-17T06:27:22.92798Z","shell.execute_reply.started":"2024-12-17T06:25:42.044088Z","shell.execute_reply":"2024-12-17T06:27:22.926687Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"#### **Observation:** \n- It's seems not even one continuous values have any correlation with **Premium Amount**.\n- **Annual Income** and **Credit Score** are having a skewed distribution.","metadata":{}},{"cell_type":"markdown","source":"Let's examine the impact of missing values in continuous features on the Premium Amount","metadata":{}},{"cell_type":"code","source":"for feat in continuous_feat: \n    if original[feat].isna().sum()>0:\n        data = original.copy()\n        data['is_na'] = original[feat].isna().astype(int)\n        agg_data = data.groupby('is_na')['Premium Amount']\n        groups = [group.values for _, group in agg_data]\n        _, p_value = kruskal(*groups)\n        \n        plt.figure(figsize=(8,3))\n        sns.boxplot(data=data, x='is_na', y='Premium Amount',palette='rocket_r')\n        plt.xlabel(f'{feat} NA P Value :{p_value:0.2f}')\n        plt.box(False)\n        \n        plt.title('Premium Amount by '+feat+' NA Values')\n        plt.tight_layout()\n        plt.show()","metadata":{"trusted":true,"_kg_hide-input":true,"execution":{"iopub.status.busy":"2024-12-17T06:27:22.929437Z","iopub.execute_input":"2024-12-17T06:27:22.929775Z","iopub.status.idle":"2024-12-17T06:27:24.488572Z","shell.execute_reply.started":"2024-12-17T06:27:22.929741Z","shell.execute_reply":"2024-12-17T06:27:24.487398Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"#### **Observation:**\n- **Health Score** `NA` values have a significant effect on Premium Amount.","metadata":{}},{"cell_type":"markdown","source":"## <p style=\"background-color:#E34234; color:white; padding:10px 20px; width:95%; border-radius:15px; text-align:center; margin: 0 auto; box-shadow:0 4px 8px rgba(0, 0, 0, 0.1);font-weight: bold;\">Categorical Feature Analysis</p>","metadata":{}},{"cell_type":"code","source":"data = original.copy()\ndata = data[categorical_feat].fillna('NA')\n\nfor feat in categorical_feat:\n    agg_data = data.groupby(feat).size()\n    norm = plt.Normalize(agg_data.min(), agg_data.max())\n    \n    plt.figure(figsize=(8,3))\n    \n    plt.subplot(1,2,1)\n    plt.pie(agg_data,\n            autopct='%.2f',\n            labels=agg_data.index,\n            colors=sns.color_palette(\"rocket_r\", len(agg_data)+2),\n            radius=1.25)\n    plt.pie([1],radius=0.65,colors=['white'])\n    \n    plt.subplot(1,2,2)\n    bars = sns.barplot(x=agg_data.index,y=agg_data)\n    bars.bar_label(bars.containers[0], \n                   rotation=90,fontsize=10,\n                   label_type='center',\n                   labels=[f'{x/1e3:.1f}K' for x in agg_data])\n    \n    for bar in bars.patches:\n        height = bar.get_height()\n        bar.set_facecolor(cmap(norm(height)*0.75))\n    \n    plt.legend([])\n    plt.yticks([])\n    plt.ylabel('')\n    plt.xlabel('')\n    plt.box(False)\n    \n    plt.suptitle(feat+' Distribution')\n    plt.tight_layout()\n    plt.show()\n","metadata":{"trusted":true,"_kg_hide-input":true,"execution":{"iopub.status.busy":"2024-12-17T06:27:24.489977Z","iopub.execute_input":"2024-12-17T06:27:24.490283Z","iopub.status.idle":"2024-12-17T06:27:27.703895Z","shell.execute_reply.started":"2024-12-17T06:27:24.490253Z","shell.execute_reply":"2024-12-17T06:27:27.702873Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"data = original.copy()\ndata[categorical_feat] =  data[categorical_feat].fillna('NA')\n\nfor feat in categorical_feat:\n    agg_data = data.groupby(feat)['Premium Amount']\n    groups = [group.values for _, group in agg_data]\n    _, p_value = kruskal(*groups)\n    \n    plt.figure(figsize=(8,3))\n    plt.subplot(121)\n    sns.kdeplot(data=data, x='Premium Amount',hue=feat,palette='rocket_r')\n    plt.box(False)\n    \n    plt.subplot(122)\n    sns.boxplot(data=data, x='Premium Amount',y=feat,palette='rocket_r')\n    plt.box(False)\n    \n    plt.suptitle(f'Premium Amount by {feat} [P Value:{p_value:0.2f}]')\n    plt.tight_layout()\n    plt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T06:27:27.706947Z","iopub.execute_input":"2024-12-17T06:27:27.707388Z","iopub.status.idle":"2024-12-17T06:27:51.076022Z","shell.execute_reply.started":"2024-12-17T06:27:27.707319Z","shell.execute_reply":"2024-12-17T06:27:51.074684Z"},"_kg_hide-input":true},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"#### **Observation:**\n- All categorical features are well distributed\n- **Exercise Frequency**, **Education Level**, and **Marital Status** have sigthly effect on `Premium Amount`","metadata":{}},{"cell_type":"markdown","source":"## <p style=\"background-color:#E34234; color:white; padding:10px 20px; width:95%; border-radius:15px; text-align:center; margin: 0 auto; box-shadow:0 4px 8px rgba(0, 0, 0, 0.1);font-weight: bold;\">Target Feature Analysis</p>","metadata":{}},{"cell_type":"code","source":"plt.figure(figsize=(8,6))\n\nplt.subplot(221)\ncounts, bins = np.histogram(original['Premium Amount'].dropna(), bins=100)\nnorm = plt.Normalize(counts.min(), counts.max())\nhist = sns.histplot(data=original, x='Premium Amount',bins=100)\nfor patch in hist.patches:\n    height = patch.get_height()  \n    patch.set_facecolor(cmap(norm(height)*0.75))\nplt.box(False)\nplt.xticks([])\nplt.yticks([])\nplt.xlabel('')\nplt.ylabel('')\nplt.title(f\"Original Distribution [Skew:{original['Premium Amount'].skew():0.3f}]\")\n\nplt.subplot(222)\namount = np.log1p(original['Premium Amount'].dropna())\ncounts, bins = np.histogram(amount, bins=100)\nnorm = plt.Normalize(counts.min(), counts.max())\nhist = sns.histplot( x=amount,bins=100)\nfor patch in hist.patches:\n    height = patch.get_height()  \n    patch.set_facecolor(cmap(norm(height)*0.75))\nplt.box(False)\nplt.xticks([])\nplt.yticks([])\nplt.xlabel('')\nplt.ylabel('')\nplt.title(f\"Log1p Distribution [Skew:{amount.skew():0.3f}]\")\n\nplt.subplot(223)\namount = np.sqrt(original['Premium Amount'].dropna())\ncounts, bins = np.histogram(amount, bins=100)\nnorm = plt.Normalize(counts.min(), counts.max())\nhist = sns.histplot( x=amount,bins=100)\nfor patch in hist.patches:\n    height = patch.get_height()  \n    patch.set_facecolor(cmap(norm(height)*0.75))\nplt.box(False)\nplt.xticks([])\nplt.yticks([])\nplt.xlabel('')\nplt.ylabel('')\nplt.title(f\"SQRT Distribution [Skew:{amount.skew():0.3f}]\")\n\nplt.subplot(224)\namount = np.cbrt(original['Premium Amount'].dropna())\ncounts, bins = np.histogram(amount, bins=100)\nnorm = plt.Normalize(counts.min(), counts.max())\nhist = sns.histplot( x=amount,bins=100)\nfor patch in hist.patches:\n    height = patch.get_height()  \n    patch.set_facecolor(cmap(norm(height)*0.75))\nplt.box(False)\nplt.xticks([])\nplt.yticks([])\nplt.xlabel('')\nplt.ylabel('')\nplt.title(f\"CBRT Distribution [Skew:{amount.skew():0.3f}]\")\n\nplt.tight_layout()\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T06:27:51.077919Z","iopub.execute_input":"2024-12-17T06:27:51.078432Z","iopub.status.idle":"2024-12-17T06:27:52.521074Z","shell.execute_reply.started":"2024-12-17T06:27:51.07838Z","shell.execute_reply":"2024-12-17T06:27:52.519911Z"},"_kg_hide-input":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"data = original.copy()\nfor feat in (discrete_feat+continuous_feat):\n    data[feat] = data[feat].fillna(data[feat].median())\ndata[categorical_feat] = data[categorical_feat].fillna('NA')\ndata = pd.get_dummies(data,columns=categorical_feat,drop_first=True,dtype=int)\n\nplt.figure(figsize=(10,8))\nsns.heatmap(data.corr(),cmap='rocket_r',cbar=False)\nplt.title('Heatmap')\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T06:27:52.522785Z","iopub.execute_input":"2024-12-17T06:27:52.523241Z","iopub.status.idle":"2024-12-17T06:27:55.476774Z","shell.execute_reply.started":"2024-12-17T06:27:52.523191Z","shell.execute_reply":"2024-12-17T06:27:55.475196Z"},"_kg_hide-input":true},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# <p style=\"background-color:#E34234; color:white; padding:10px 20px; width:95%; border-radius:15px; text-align:center; margin: 0 auto; box-shadow:0 4px 8px rgba(0, 0, 0, 0.1);font-weight: bold;\">Data Processing</p>","metadata":{}},{"cell_type":"code","source":"train_data = pd.read_csv('/kaggle/input/playground-series-s4e12/train.csv',index_col='id')\ntest_data = pd.read_csv('/kaggle/input/playground-series-s4e12/test.csv',index_col='id')\noriginal  = pd.read_csv('/kaggle/input/insurance-premium-prediction/Insurance Premium Prediction Dataset.csv')\noriginal.dropna(subset='Premium Amount',inplace=True)\ncombined_data  = pd.concat([original,train_data,test_data])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T16:06:16.877455Z","iopub.execute_input":"2024-12-17T16:06:16.87824Z","iopub.status.idle":"2024-12-17T16:06:27.723905Z","shell.execute_reply.started":"2024-12-17T16:06:16.8782Z","shell.execute_reply":"2024-12-17T16:06:27.723055Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"combined_data['Policy Start Date'] = pd.DatetimeIndex(combined_data['Policy Start Date'])\ncombined_data['year'] = combined_data['Policy Start Date'].dt.year\ncombined_data['month'] = combined_data['Policy Start Date'].dt.month\ncombined_data['date'] = combined_data['Policy Start Date'].dt.day\ncombined_data['day'] = combined_data['Policy Start Date'].dt.weekday\ncombined_data.drop('Policy Start Date',axis=1,inplace=True)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T16:06:27.725759Z","iopub.execute_input":"2024-12-17T16:06:27.726451Z","iopub.status.idle":"2024-12-17T16:06:29.718587Z","shell.execute_reply.started":"2024-12-17T16:06:27.726402Z","shell.execute_reply":"2024-12-17T16:06:29.717745Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"for feat in discrete_feat: \n    if combined_data[feat].isna().sum()>0:\n        combined_data[feat+'_is_na'] = combined_data[feat].isna().astype(int)\n        impute = np.exp( np.mean( np.log1p(combined_data[feat]) ) )-1\n        combined_data[feat] = combined_data[feat].fillna(impute)\n\nfor feat in continuous_feat: \n    if combined_data[feat].isna().sum()>0:\n        combined_data[feat+'_is_na'] = combined_data[feat].isna().astype(int)\n        impute = np.exp( np.mean( np.log1p(combined_data[feat]) ) )-1\n        combined_data[feat] = combined_data[feat].fillna(impute)\n\ncombined_data[categorical_feat] = combined_data[categorical_feat].fillna('NA')\ncombined_data = pd.get_dummies(combined_data,columns=categorical_feat,dtype=int) ","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T16:06:29.719697Z","iopub.execute_input":"2024-12-17T16:06:29.720003Z","iopub.status.idle":"2024-12-17T16:06:35.922722Z","shell.execute_reply.started":"2024-12-17T16:06:29.719973Z","shell.execute_reply":"2024-12-17T16:06:35.921898Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# <p style=\"background-color:#E34234; color:white; padding:10px 20px; width:95%; border-radius:15px; text-align:center; margin: 0 auto; box-shadow:0 4px 8px rgba(0, 0, 0, 0.1);font-weight: bold;\">Feature Selection & Model Evaluation</p>","metadata":{}},{"cell_type":"code","source":"X = combined_data[:len(test_data)].drop('Premium Amount',axis=1) \ny = combined_data[:len(test_data)]['Premium Amount']\n\nX_train_only = combined_data[len(original):len(test_data)].drop('Premium Amount',axis=1)\ny_train_only = combined_data[len(original):len(test_data)]['Premium Amount']\n\nmodel = LGBMRegressor(n_estimators=3000,learning_rate=0.01,lambda_l1=0.5,lambda_l2=0.9,verbose=-1,n_jobs=4)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T16:06:35.924421Z","iopub.execute_input":"2024-12-17T16:06:35.924724Z","iopub.status.idle":"2024-12-17T16:06:36.031414Z","shell.execute_reply.started":"2024-12-17T16:06:35.924695Z","shell.execute_reply":"2024-12-17T16:06:36.030369Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"## <p style=\"background-color:#E34234; color:white; padding:10px 20px; width:95%; border-radius:15px; text-align:center; margin: 0 auto; box-shadow:0 4px 8px rgba(0, 0, 0, 0.1);font-weight: bold;\">Base Model</p>","metadata":{}},{"cell_type":"code","source":"def validation_scores(score):\n    mean_scores = {metric: np.mean(values) for metric, values in score.items()}\n    return pd.DataFrame(mean_scores, index=['Score'])\n    \nscore = cross_validate(model,X,y,cv=10,\n                       return_train_score=True,\n                       return_estimator=True,\n                       scoring=['r2','neg_mean_squared_log_error'])\n\nestimators = score.pop('estimator')\nprint('RMSLE:',np.mean([np.sqrt(mean_squared_log_error(y,i.predict(X))) for i in estimators]))\nprint('Train Only RMSLE:',np.mean([np.sqrt(mean_squared_log_error(y_train_only,i.predict(X_train_only))) for i in estimators]))\n\nvalidation_scores(score)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T16:06:42.288961Z","iopub.execute_input":"2024-12-17T16:06:42.289863Z","iopub.status.idle":"2024-12-17T16:07:29.755258Z","shell.execute_reply.started":"2024-12-17T16:06:42.289805Z","shell.execute_reply":"2024-12-17T16:07:29.754296Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"## <p style=\"background-color:#E34234; color:white; padding:10px 20px; width:95%; border-radius:15px; text-align:center; margin: 0 auto; box-shadow:0 4px 8px rgba(0, 0, 0, 0.1);font-weight: bold;\">Experiment 01</p>","metadata":{"execution":{"iopub.status.busy":"2024-12-12T06:51:25.77357Z","iopub.execute_input":"2024-12-12T06:51:25.774183Z","iopub.status.idle":"2024-12-12T06:51:25.82526Z","shell.execute_reply.started":"2024-12-12T06:51:25.774133Z","shell.execute_reply":"2024-12-12T06:51:25.823948Z"}}},{"cell_type":"markdown","source":"Let's evaluate whether applying a transformation to the **Premium Amount** improves performance.","metadata":{}},{"cell_type":"code","source":"# using log1p because its beter performing than sqrt and cbrt\nscore = cross_validate(model,X,np.log1p(y),cv=10,\n                       return_train_score=True,\n                       return_estimator=True,\n                       scoring=['r2','neg_mean_squared_log_error'])\nestimators_1 = score.pop('estimator')\nprint('RMSLE:',np.mean([np.sqrt(mean_squared_log_error(y,np.exp(i.predict(X))-1)) for i in estimators_1]))\nprint('Train Only RMSLE:',np.mean([np.sqrt(mean_squared_log_error(y_train_only,np.exp(i.predict(X_train_only))-1)) for i in estimators_1]))\n\nvalidation_scores(score)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T16:07:29.756899Z","iopub.execute_input":"2024-12-17T16:07:29.757204Z","iopub.status.idle":"2024-12-17T16:08:14.710821Z","shell.execute_reply.started":"2024-12-17T16:07:29.757175Z","shell.execute_reply":"2024-12-17T16:08:14.709902Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"#### **Observation:**\nAs observed, there is approximately a **7.9%** reduction in train only RMSLE, going from **1.14** to **1.05**.","metadata":{}},{"cell_type":"markdown","source":"## <p style=\"background-color:#E34234; color:white; padding:10px 20px; width:95%; border-radius:15px; text-align:center; margin: 0 auto; box-shadow:0 4px 8px rgba(0, 0, 0, 0.1);font-weight: bold;\">Experiment 02</p>","metadata":{}},{"cell_type":"markdown","source":"Let's evaluate whether applying transformations to the **Skewed Continous Features** improves performance or not.","metadata":{}},{"cell_type":"code","source":"X['Annual Income'] = np.cbrt(X['Annual Income']) # Skewness 1.34 to 0.007\nX['Health Score'] = np.sqrt(X['Health Score']+1) # Skewness 0.57 to -0.056\n\nscore = cross_validate(model,X,np.log1p(y),cv=10,\n                       return_train_score=True,\n                       return_estimator=True,\n                       scoring=['r2','neg_mean_squared_log_error'])\nestimators_2 = score.pop('estimator')\nprint('RMSLE:',np.mean([np.sqrt(mean_squared_log_error(y,np.exp(i.predict(X))-1)) for i in estimators_2]))\nprint('Train Only RMSLE:',np.mean([np.sqrt(mean_squared_log_error(y_train_only,np.exp(i.predict(X_train_only)))) for i in estimators_2]))\n\nvalidation_scores(score)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T10:50:57.939405Z","iopub.execute_input":"2024-12-17T10:50:57.939919Z","iopub.status.idle":"2024-12-17T10:51:14.978909Z","shell.execute_reply.started":"2024-12-17T10:50:57.939879Z","shell.execute_reply":"2024-12-17T10:51:14.977867Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"#### **Observation:**\nTrain only RMSLE increased in this case","metadata":{}},{"cell_type":"markdown","source":"## <p style=\"background-color:#E34234; color:white; padding:10px 20px; width:95%; border-radius:15px; text-align:center; margin: 0 auto; box-shadow:0 4px 8px rgba(0, 0, 0, 0.1);font-weight: bold;\">Experiment 03</p>","metadata":{}},{"cell_type":"markdown","source":"Let's evaluate whether **OneHot Encoding** on **Discrete Features** improves performance or not.","metadata":{}},{"cell_type":"code","source":"X_discrete_enc = pd.get_dummies(X,columns=discrete_feat,dtype=int) \nX_train_dscrt_enc = pd.get_dummies(X_train_only,columns=discrete_feat,dtype=int) \n\nscore = cross_validate(model,X_discrete_enc,np.log1p(y),cv=10,\n                       return_train_score=True,\n                       return_estimator=True,\n                       scoring=['r2','neg_mean_squared_log_error'])\nestimators_3 = score.pop('estimator')\nprint('RMSLE:',np.mean([np.sqrt(mean_squared_log_error(y,np.exp(i.predict(X_discrete_enc))-1)) for i in estimators_3]))\nprint('Train Only RMSLE:',np.mean([np.sqrt(mean_squared_log_error(y_train_only,np.exp(i.predict(X_train_dscrt_enc)))) for i in estimators_3]))\n\nvalidation_scores(score)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T10:53:25.094876Z","iopub.execute_input":"2024-12-17T10:53:25.096205Z","iopub.status.idle":"2024-12-17T10:53:50.113781Z","shell.execute_reply.started":"2024-12-17T10:53:25.096153Z","shell.execute_reply":"2024-12-17T10:53:50.112654Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"#### **Observation:**\nNo better result than **Experiment 01**","metadata":{}},{"cell_type":"markdown","source":"## <p style=\"background-color:#E34234; color:white; padding:10px 20px; width:95%; border-radius:15px; text-align:center; margin: 0 auto; box-shadow:0 4px 8px rgba(0, 0, 0, 0.1);font-weight: bold;\">Experiment 04</p>","metadata":{}},{"cell_type":"markdown","source":"Let's evaluate with only statistically significant features","metadata":{}},{"cell_type":"code","source":"data = combined_data[:len(test_data)]\nsignificant_feat = []\nfor feat in X.columns:\n    n = X[feat].nunique()\n    if n>1 and n<20:\n        groups = data.groupby(feat)['Premium Amount'].apply(list)\n        _, p_value = kruskal(*groups)\n        print(f'[{p_value<0.05}] {feat}:{p_value:0.2f}(P Value)')\n        if(p_value<=0.05):\n            significant_feat.append(feat)\n    else:\n        significant_feat.append(feat)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T10:54:08.89416Z","iopub.execute_input":"2024-12-17T10:54:08.894678Z","iopub.status.idle":"2024-12-17T10:54:26.062279Z","shell.execute_reply.started":"2024-12-17T10:54:08.894633Z","shell.execute_reply":"2024-12-17T10:54:26.060889Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"score = cross_validate(model,X[significant_feat],np.log1p(y),cv=10,\n                       return_train_score=True,\n                       return_estimator=True,\n                       scoring=['r2','neg_mean_squared_log_error'])\nestimators_4 = score.pop('estimator')\nprint('RMSLE:',np.mean([np.sqrt(mean_squared_log_error(y,np.exp(i.predict(X[significant_feat])))) for i in estimators_4]))\nprint('Train Only RMSLE:',np.mean([np.sqrt(mean_squared_log_error(y_train_only,np.exp(i.predict(X_train_only[significant_feat])))) for i in estimators_4]))\n\nvalidation_scores(score)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T10:54:29.734937Z","iopub.execute_input":"2024-12-17T10:54:29.735431Z","iopub.status.idle":"2024-12-17T10:54:39.281247Z","shell.execute_reply.started":"2024-12-17T10:54:29.73539Z","shell.execute_reply":"2024-12-17T10:54:39.280153Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"#### **Observation:**\nThe RMSLE value increased instead of decreasing during training","metadata":{}},{"cell_type":"markdown","source":"## <p style=\"background-color:#E34234; color:white; padding:10px 20px; width:95%; border-radius:15px; text-align:center; margin: 0 auto; box-shadow:0 4px 8px rgba(0, 0, 0, 0.1);font-weight: bold;\">Experiment 05</p>","metadata":{}},{"cell_type":"markdown","source":"Lets check with only competition data without including the original dataset","metadata":{}},{"cell_type":"code","source":"score = cross_validate(model,X_train_only,np.log1p(y_train_only),cv=10,\n                       return_train_score=True,\n                       return_estimator=True,\n                       scoring=['r2','neg_mean_squared_log_error'])\nestimators_5 = score.pop('estimator')\nprint('RMSLE:',np.mean([np.sqrt(mean_squared_log_error(y,np.exp(i.predict(X)))) for i in estimators_5]))\nprint('Train Only RMSLE:',np.mean([np.sqrt(mean_squared_log_error(y_train_only,np.exp(i.predict(X_train_only)))) for i in estimators_5]))\nvalidation_scores(score)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T16:12:54.563407Z","iopub.execute_input":"2024-12-17T16:12:54.56418Z","iopub.status.idle":"2024-12-17T16:13:30.293006Z","shell.execute_reply.started":"2024-12-17T16:12:54.56414Z","shell.execute_reply":"2024-12-17T16:13:30.292061Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"#### **Observation:**\nTill now this is the best **RMSLE** value.","metadata":{}},{"cell_type":"markdown","source":"# <p style=\"background-color:#E34234; color:white; padding:10px 20px; width:95%; border-radius:15px; text-align:center; margin: 0 auto; box-shadow:0 4px 8px rgba(0, 0, 0, 0.1);font-weight: bold;\">Submission</p>","metadata":{}},{"cell_type":"code","source":"for k, estimator in enumerate([estimators_1,estimators_5]):\n    pred = [np.exp(i.predict(combined_data[-len(test_data):].drop('Premium Amount',axis=1)))-1 for i in estimator]\n    pred = np.mean(pred,axis=0)\n    test_data['Premium Amount'] = pred\n    test_data['Premium Amount'].to_csv(f'submission_{k}.csv')","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-17T16:14:23.864782Z","iopub.execute_input":"2024-12-17T16:14:23.865661Z","iopub.status.idle":"2024-12-17T16:14:42.134744Z","shell.execute_reply.started":"2024-12-17T16:14:23.865624Z","shell.execute_reply":"2024-12-17T16:14:42.133856Z"}},"outputs":[],"execution_count":null}]}