{ "cells": [ { "cell_type": "markdown", "id": "c53c9336", "metadata": {}, "source": [ "MODEL TRAINING" ] }, { "cell_type": "code", "execution_count": 1, "id": "c7a721ee", "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "import seaborn as sns\n", "%matplotlib inline\n", "import warnings\n", "warnings.filterwarnings('ignore')" ] }, { "cell_type": "code", "execution_count": null, "id": "904e70b1", "metadata": {}, "outputs": [], "source": [ "# model Packages \n", "from sklearn.model_selection import train_test_split\n", "from sklearn.linear_model import LinearRegression , Ridge, Lasso\n", "from sklearn.tree import DecisionTreeRegressor\n", "from sklearn.ensemble import RandomForestRegressor, GradientBoostingRegressor\n", "from sklearn.svm import SVR\n", "from sklearn.neighbors import KNeighborsRegressor\n", "from sklearn.preprocessing import StandardScaler, MinMaxScaler\n", "from sklearn.model_selection import GridSearchCV, RandomizedSearchCV\n", "from sklearn.metrics import mean_absolute_error, mean_squared_error, r2_score\n", "\n", "\n", "from catboost import CatBoostRegressor\n", "from xgboost import XGBRegressor" ] }, { "cell_type": "code", "execution_count": 3, "id": "b40c409b", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
genderrace/ethnicityparental level of educationlunchtest preparation coursemath scorereading scorewriting score
0femalegroup Bbachelor's degreestandardnone727274
1femalegroup Csome collegestandardcompleted699088
2femalegroup Bmaster's degreestandardnone909593
3malegroup Aassociate's degreefree/reducednone475744
4malegroup Csome collegestandardnone767875
\n", "
" ], "text/plain": [ " gender race/ethnicity parental level of education lunch \\\n", "0 female group B bachelor's degree standard \n", "1 female group C some college standard \n", "2 female group B master's degree standard \n", "3 male group A associate's degree free/reduced \n", "4 male group C some college standard \n", "\n", " test preparation course math score reading score writing score \n", "0 none 72 72 74 \n", "1 completed 69 90 88 \n", "2 none 90 95 93 \n", "3 none 47 57 44 \n", "4 none 76 78 75 " ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df = pd.read_csv(\"./data/StudentsPerformance.csv\")\n", "df.head()" ] }, { "cell_type": "code", "execution_count": 5, "id": "2007d673", "metadata": {}, "outputs": [], "source": [ "y = df[\"math score\"]\n", "X = df.drop(columns= [\"math score\"] , axis=1)\n" ] }, { "cell_type": "code", "execution_count": 6, "id": "e634f28b", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
genderrace/ethnicityparental level of educationlunchtest preparation coursereading scorewriting score
0femalegroup Bbachelor's degreestandardnone7274
1femalegroup Csome collegestandardcompleted9088
2femalegroup Bmaster's degreestandardnone9593
3malegroup Aassociate's degreefree/reducednone5744
4malegroup Csome collegestandardnone7875
\n", "
" ], "text/plain": [ " gender race/ethnicity parental level of education lunch \\\n", "0 female group B bachelor's degree standard \n", "1 female group C some college standard \n", "2 female group B master's degree standard \n", "3 male group A associate's degree free/reduced \n", "4 male group C some college standard \n", "\n", " test preparation course reading score writing score \n", "0 none 72 74 \n", "1 completed 90 88 \n", "2 none 95 93 \n", "3 none 57 44 \n", "4 none 78 75 " ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "X.head()" ] }, { "cell_type": "code", "execution_count": 7, "id": "c3776ba5", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0 72\n", "1 69\n", "2 90\n", "3 47\n", "4 76\n", " ..\n", "995 88\n", "996 62\n", "997 59\n", "998 68\n", "999 77\n", "Name: math score, Length: 1000, dtype: int64" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "y " ] }, { "cell_type": "code", "execution_count": 8, "id": "0f3f604e", "metadata": {}, "outputs": [], "source": [ "# create column transformer for with 3 type of transformers \n", "from sklearn.compose import ColumnTransformer \n", "from sklearn.preprocessing import OneHotEncoder , StandardScaler\n", "\n", "numerical_features = X.select_dtypes(exclude=[\"object\"]).columns.tolist()\n", "categorical_features = X.select_dtypes(include=['object']).columns.tolist()\n", "\n", "oh_transformer = OneHotEncoder()\n", "numeric_transformer = StandardScaler()\n", "\n", "\n", "preprocessor = ColumnTransformer(\n", " transformers=[\n", " ('num', numeric_transformer, numerical_features),\n", " ('cat', oh_transformer, categorical_features)\n", " ])\n" ] }, { "cell_type": "code", "execution_count": 9, "id": "6d1274f9", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(1000, 19)" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "X = preprocessor.fit_transform(X)\n", "X.shape" ] }, { "cell_type": "code", "execution_count": 10, "id": "f30338ba", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([[ 0.19399858, 0.39149181, 1. , ..., 1. ,\n", " 0. , 1. ],\n", " [ 1.42747598, 1.31326868, 1. , ..., 1. ,\n", " 1. , 0. ],\n", " [ 1.77010859, 1.64247471, 1. , ..., 1. ,\n", " 0. , 1. ],\n", " ...,\n", " [ 0.12547206, -0.20107904, 1. , ..., 0. ,\n", " 1. , 0. ],\n", " [ 0.60515772, 0.58901542, 1. , ..., 1. ,\n", " 1. , 0. ],\n", " [ 1.15336989, 1.18158627, 1. , ..., 0. ,\n", " 0. , 1. ]], shape=(1000, 19))" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "X" ] }, { "cell_type": "code", "execution_count": 11, "id": "8365c79c", "metadata": {}, "outputs": [], "source": [ "# train test split \n", "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)" ] }, { "cell_type": "code", "execution_count": 12, "id": "d8d7a165", "metadata": {}, "outputs": [], "source": [ "def evaluate_model(true , predicted):\n", " mae = mean_absolute_error(true , predicted)\n", " mse = mean_squared_error(true , predicted)\n", " rmse = np.sqrt(mse)\n", " r2_square = r2_score(true , predicted)\n", " return mae , mse , rmse , r2_square" ] }, { "cell_type": "code", "execution_count": 14, "id": "476ee5fd", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "-----------------------------------------\n", "--------------------- Linear Regression ---------------------\n", "---------- TRAINING METRICS ----------\n", "Train MAE: 4.266711846071956\n", "Train MSE: 28.33487038064859\n", "Train RMSE: 5.323050852720514\n", "Train R2: 0.8743172040139593\n", "---------- TESTING METRICS ----------\n", "Test MAE: 4.214763142474852\n", "Test MSE: 29.095169866715487\n", "Test RMSE: 5.393993869732843\n", "Test R2: 0.8804332983749565\n", "-----------------------------------------\n", "\n", "\n", "\n", "-----------------------------------------\n", "--------------------- Ridge Regression ---------------------\n", "---------- TRAINING METRICS ----------\n", "Train MAE: 4.264987823725977\n", "Train MSE: 28.337788233082456\n", "Train RMSE: 5.323324922741656\n", "Train R2: 0.8743042615212908\n", "---------- TESTING METRICS ----------\n", "Test MAE: 4.211100688014261\n", "Test MSE: 29.05627219234826\n", "Test RMSE: 5.390387016935636\n", "Test R2: 0.880593148502874\n", "-----------------------------------------\n", "\n", "\n", "\n", "-----------------------------------------\n", "--------------------- Lasso Regression ---------------------\n", "---------- TRAINING METRICS ----------\n", "Train MAE: 5.206296077972952\n", "Train MSE: 43.47829788272618\n", "Train RMSE: 6.593807540619166\n", "Train R2: 0.8071466723085148\n", "---------- TESTING METRICS ----------\n", "Test MAE: 5.157879138921815\n", "Test MSE: 42.50633235127343\n", "Test RMSE: 6.5196880562856245\n", "Test R2: 0.825320079562973\n", "-----------------------------------------\n", "\n", "\n", "\n", "-----------------------------------------\n", "--------------------- Decision Tree ---------------------\n", "---------- TRAINING METRICS ----------\n", "Train MAE: 0.01875\n", "Train MSE: 0.078125\n", "Train RMSE: 0.2795084971874737\n", "Train R2: 0.9996534669718089\n", "---------- TESTING METRICS ----------\n", "Test MAE: 6.165\n", "Test MSE: 59.735\n", "Test RMSE: 7.728842086625913\n", "Test R2: 0.7545188100192982\n", "-----------------------------------------\n", "\n", "\n", "\n", "-----------------------------------------\n", "--------------------- Random Forest ---------------------\n", "---------- TRAINING METRICS ----------\n", "Train MAE: 1.8297778273809524\n", "Train MSE: 5.342042640217546\n", "Train RMSE: 2.311285927837044\n", "Train R2: 0.9763047140756385\n", "---------- TESTING METRICS ----------\n", "Test MAE: 4.61957619047619\n", "Test MSE: 35.32140070861678\n", "Test RMSE: 5.943181026068176\n", "Test R2: 0.8548465811042697\n", "-----------------------------------------\n", "\n", "\n", "\n", "-----------------------------------------\n", "--------------------- Gradient Boosting ---------------------\n", "---------- TRAINING METRICS ----------\n", "Train MAE: 3.722632404265115\n", "Train MSE: 21.408568924292386\n", "Train RMSE: 4.626939477050936\n", "Train R2: 0.9050396644022572\n", "---------- TESTING METRICS ----------\n", "Test MAE: 4.315982000012644\n", "Test MSE: 31.16465127053362\n", "Test RMSE: 5.582530901887926\n", "Test R2: 0.8719287573579277\n", "-----------------------------------------\n", "\n", "\n", "\n", "-----------------------------------------\n", "--------------------- Support Vector Regressor ---------------------\n", "---------- TRAINING METRICS ----------\n", "Train MAE: 4.869189452384868\n", "Train MSE: 43.257024268031365\n", "Train RMSE: 6.57700724251018\n", "Train R2: 0.8081281585902299\n", "---------- TESTING METRICS ----------\n", "Test MAE: 5.4015392444969965\n", "Test MSE: 66.04200493745648\n", "Test RMSE: 8.126623218622633\n", "Test R2: 0.7286001513223705\n", "-----------------------------------------\n", "\n", "\n", "\n", "-----------------------------------------\n", "--------------------- K-Neighbors Regressor ---------------------\n", "---------- TRAINING METRICS ----------\n", "Train MAE: 4.5165\n", "Train MSE: 32.6339\n", "Train RMSE: 5.712608861107156\n", "Train R2: 0.8552483303848109\n", "---------- TESTING METRICS ----------\n", "Test MAE: 5.619\n", "Test MSE: 52.617\n", "Test RMSE: 7.253757646902741\n", "Test R2: 0.7837702557426202\n", "-----------------------------------------\n", "\n", "\n", "\n", "-----------------------------------------\n", "--------------------- XGBoost Regressor ---------------------\n", "---------- TRAINING METRICS ----------\n", "Train MAE: 0.687466561794281\n", "Train MSE: 1.0146163702011108\n", "Train RMSE: 1.0072816737145132\n", "Train R2: 0.9954995512962341\n", "---------- TESTING METRICS ----------\n", "Test MAE: 5.1036295890808105\n", "Test MSE: 43.50392150878906\n", "Test RMSE: 6.595750261250729\n", "Test R2: 0.8212205171585083\n", "-----------------------------------------\n", "\n", "\n", "\n", "-----------------------------------------\n", "--------------------- CatBoost Regressor ---------------------\n", "---------- TRAINING METRICS ----------\n", "Train MAE: 2.405393926779502\n", "Train MSE: 9.257805405523678\n", "Train RMSE: 3.042664195326799\n", "Train R2: 0.9589358676277713\n", "---------- TESTING METRICS ----------\n", "Test MAE: 4.612531714976557\n", "Test MSE: 36.10365799356841\n", "Test RMSE: 6.008631956907363\n", "Test R2: 0.8516318920747058\n", "-----------------------------------------\n", "\n", "\n", "\n" ] } ], "source": [ "models = {\n", " \"Linear Regression\": LinearRegression(),\n", " \"Ridge Regression\": Ridge(),\n", " \"Lasso Regression\": Lasso(),\n", " \"Decision Tree\": DecisionTreeRegressor(),\n", " \"Random Forest\": RandomForestRegressor(),\n", " \"Gradient Boosting\": GradientBoostingRegressor(),\n", " \"Support Vector Regressor\": SVR(),\n", " \"K-Neighbors Regressor\": KNeighborsRegressor(),\n", " \"XGBoost Regressor\": XGBRegressor(),\n", " \"CatBoost Regressor\": CatBoostRegressor(verbose=False)\n", "}\n", "\n", "model_report = {}\n", "\n", "for model_name, model in models.items():\n", " # training the model\n", " model.fit(X_train, y_train)\n", " \n", " # predicting the model\n", " y_train_pred = model.predict(X_train)\n", " y_test_pred = model.predict(X_test)\n", " \n", " # evaluating the model\n", " train_mae, train_mse, train_rmse, train_r2 = evaluate_model(y_train, y_train_pred)\n", " test_mae, test_mse, test_rmse, test_r2 = evaluate_model(y_test, y_test_pred)\n", "\n", " print(\"-----------------------------------------\")\n", " print(\"--------------------- {0} ---------------------\".format(model_name))\n", " print(\"---------- TRAINING METRICS ----------\")\n", " print(\"Train MAE: \", train_mae)\n", " print(\"Train MSE: \", train_mse)\n", " print(\"Train RMSE: \", train_rmse)\n", " print(\"Train R2: \", train_r2)\n", " print(\"---------- TESTING METRICS ----------\")\n", " print(\"Test MAE: \", test_mae)\n", " print(\"Test MSE: \", test_mse)\n", " print(\"Test RMSE: \", test_rmse)\n", " print(\"Test R2: \", test_r2)\n", " print(\"-----------------------------------------\")\n", " print(\"\\n\\n\")\n", " \n", " model_report[model_name] = {\n", " \"Train MAE\": train_mae,\n", " \"Train MSE\": train_mse,\n", " \"Train RMSE\": train_rmse,\n", " \"Train R2\": train_r2,\n", " \"Test MAE\": test_mae,\n", " \"Test MSE\": test_mse,\n", " \"Test RMSE\": test_rmse,\n", " \"Test R2\": test_r2\n", " }\n" ] }, { "cell_type": "code", "execution_count": 15, "id": "39e7570f", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Train MAETrain MSETrain RMSETrain R2Test MAETest MSETest RMSETest R2
Linear Regression4.26671228.3348705.3230510.8743174.21476329.0951705.3939940.880433
Ridge Regression4.26498828.3377885.3233250.8743044.21110129.0562725.3903870.880593
Lasso Regression5.20629643.4782986.5938080.8071475.15787942.5063326.5196880.825320
Decision Tree0.0187500.0781250.2795080.9996536.16500059.7350007.7288420.754519
Random Forest1.8297785.3420432.3112860.9763054.61957635.3214015.9431810.854847
Gradient Boosting3.72263221.4085694.6269390.9050404.31598231.1646515.5825310.871929
Support Vector Regressor4.86918943.2570246.5770070.8081285.40153966.0420058.1266230.728600
K-Neighbors Regressor4.51650032.6339005.7126090.8552485.61900052.6170007.2537580.783770
XGBoost Regressor0.6874671.0146161.0072820.9955005.10363043.5039226.5957500.821221
CatBoost Regressor2.4053949.2578053.0426640.9589364.61253236.1036586.0086320.851632
\n", "
" ], "text/plain": [ " Train MAE Train MSE Train RMSE Train R2 \\\n", "Linear Regression 4.266712 28.334870 5.323051 0.874317 \n", "Ridge Regression 4.264988 28.337788 5.323325 0.874304 \n", "Lasso Regression 5.206296 43.478298 6.593808 0.807147 \n", "Decision Tree 0.018750 0.078125 0.279508 0.999653 \n", "Random Forest 1.829778 5.342043 2.311286 0.976305 \n", "Gradient Boosting 3.722632 21.408569 4.626939 0.905040 \n", "Support Vector Regressor 4.869189 43.257024 6.577007 0.808128 \n", "K-Neighbors Regressor 4.516500 32.633900 5.712609 0.855248 \n", "XGBoost Regressor 0.687467 1.014616 1.007282 0.995500 \n", "CatBoost Regressor 2.405394 9.257805 3.042664 0.958936 \n", "\n", " Test MAE Test MSE Test RMSE Test R2 \n", "Linear Regression 4.214763 29.095170 5.393994 0.880433 \n", "Ridge Regression 4.211101 29.056272 5.390387 0.880593 \n", "Lasso Regression 5.157879 42.506332 6.519688 0.825320 \n", "Decision Tree 6.165000 59.735000 7.728842 0.754519 \n", "Random Forest 4.619576 35.321401 5.943181 0.854847 \n", "Gradient Boosting 4.315982 31.164651 5.582531 0.871929 \n", "Support Vector Regressor 5.401539 66.042005 8.126623 0.728600 \n", "K-Neighbors Regressor 5.619000 52.617000 7.253758 0.783770 \n", "XGBoost Regressor 5.103630 43.503922 6.595750 0.821221 \n", "CatBoost Regressor 4.612532 36.103658 6.008632 0.851632 " ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model_report_df = pd.DataFrame(model_report).T\n", "model_report_df.sort_values(by=\"Test R2\", ascending=False)\n", "model_report_df" ] }, { "cell_type": "code", "execution_count": 16, "id": "00751ba7", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(4.214763142474852,\n", " 29.095169866715487,\n", " np.float64(5.393993869732843),\n", " 0.8804332983749565)" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# lets use Linear Regressior \n", "final_model = LinearRegression()\n", "final_model.fit(X_train, y_train)\n", "y_pred = final_model.predict(X_test)\n", "\n", "evaluate_model(y_test, y_pred)" ] }, { "cell_type": "code", "execution_count": 19, "id": "03cd811e", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# plot graph for data \n", "plt.figure(figsize=(8,6))\n", "sns.scatterplot(x=y_test, y=y_pred)\n", "plt.xlabel(\"Actual Math Scores\")\n", "plt.ylabel(\"Predicted Math Scores\")\n", "plt.title(\"Actual vs Predicted Math Scores\")\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": null, "id": "9f3c355a", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# lets show best fit line \n", "plt.figure(figsize=(8,6))\n", "sns.regplot(x=y_test, y=y_pred, line_kws={\"color\":\"red\"})\n", "plt.xlabel(\"Actual Math Scores\")\n", "plt.ylabel(\"Predicted Math Scores\")\n", "plt.title(\"Actual vs Predicted Math Scores with Best Fit Line\")\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 28, "id": "4d932ec8", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DifferenceActual ValuePredicted Value
014.6120309176.387970
1-5.8859705358.885970
23.0097358076.990265
3-2.8518047476.851804
4-3.6273788487.627378
\n", "
" ], "text/plain": [ " Difference Actual Value Predicted Value\n", "0 14.612030 91 76.387970\n", "1 -5.885970 53 58.885970\n", "2 3.009735 80 76.990265\n", "3 -2.851804 74 76.851804\n", "4 -3.627378 84 87.627378" ] }, "execution_count": 28, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Evaluation difference DF\n", "\n", "# // actual value predict value and difference \n", "\n", "# Convert both to Series with identical positional indices\n", "y_test_series = pd.Series(y_test).reset_index(drop=True)\n", "y_pred_series = pd.Series(y_pred).reset_index(drop=True)\n", "\n", "difference = y_test_series - y_pred_series\n", "\n", "evaluated_difference = pd.DataFrame({\n", " \"Difference\": difference,\n", " \"Actual Value\": y_test_series,\n", " \"Predicted Value\": y_pred_series\n", "})\n", "\n", "evaluated_difference.head()" ] }, { "cell_type": "code", "execution_count": null, "id": "95f8218e", "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "myenv (3.13.7)", "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": "3.13.7" } }, "nbformat": 4, "nbformat_minor": 5 }