seleccion_modelo.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Partición estratificada"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Una técnica para manejar el desbalance de clases en problemas de clasificación supervisada, consiste en garantizar que todos los subconjuntos y grupos de datos tengan la misma proporción de desbalance de clase. De esta forma, no es necesario agregar ni eliminar ninguna columna. \n",
    "\n",
    "Para construir el modelo predictivo y reportar su desempeño usaremos particiones de la forma **Train - Validation - Test**. En donde los conjuntos tendrán los siguientes tamaños:\n",
    "\n",
    "<img src=\"imgs/train-val-test.png\" alt=\"Drawing\" style=\"width: 600px;\"/>"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Debido al desbalance de clases, tanto el conjunto de test como el de train deben tener la misma proporción de desbalance de clase, que en nuestro caso es 0.21.\n",
    "\n",
    "Adicionalmente, para evitar el sobreajuste y tener buena capacidad de generalización, en la calibración de hiperparámetros se implementa la técnica Stratified Kfold Cross Validation, que consiste en la partición de los conjuntos train y validation en grupos o folds que preservan la proporción de desbalance de clases."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"imgs/skf.png\" alt=\"Drawing\" style=\"width: 600px;\"/>"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- EL XGBoost es un algoritmo de ensamble basado en árboles de decisión, es un algoritmo del estado del arte que han demostrado excelente desempeño en tareas de clasificación, además, dada su naturaleza de árobl de decisión, se trata de un modelo interpretable. Probaremos las diferentes ténicas para controlar el desbalance de clases mediante un XGboost cuyos hiperparámetros son ajustados usando GridSearch.\n",
    "\n",
    "- La métrica principal de desempeño es el **f1-score**, ya que esta es el promedio armónico entre recall y precision, lo que significa que se brinda igual impoprtancia a la correcta clasificación de pagos e impagos. Esta es la métrica más adecuada en un contexto de desbalance de clase. "
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Librerías e importación de datos preprocesados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "\n",
    "from sklearn.model_selection import train_test_split, StratifiedKFold, GridSearchCV\n",
    "from xgboost import XGBClassifier\n",
    "\n",
    "from utils import dar_estilo_visual, reporte_desempeno\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "dar_estilo_visual()\n",
    "# Cargamos los datos preparados para la selección del modelo\n",
    "df_modelo = pd.read_csv(\"cleaned_data/data_preparada.csv\")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Estructuración train - test estratificada"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "target = \"default payment next month\"\n",
    "X = df_modelo.drop(columns=target)\n",
    "y = df_modelo[target]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "seed = 42\n",
    "# Partición train-test preservando la proporción de impagos en cada parte\n",
    "X_train, X_test, y_train, y_test= train_test_split(X, y, \n",
    "                                                    stratify=y, # Con este parámetro se garantiza la divisón estratificada\n",
    "                                                    test_size=0.15, # 15% de registros destinados para test\n",
    "                                                    shuffle=True,\n",
    "                                                    random_state= seed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Proporcion de impagos en el conjunto de entrenamiento: 0.21\n",
      "Proporción de impagos en el conjunto de prueba (test): 0.21\n"
     ]
    }
   ],
   "source": [
    "# Validación de que se preserva la proporción de impagos\n",
    "prop_train = len(y_train[y_train==1]) / len(y_train)\n",
    "prop_test = len(y_test[y_test==1]) / len(y_test) \n",
    "print(\"Proporcion de impagos en el conjunto de entrenamiento: {:.2f}\".format(prop_train))\n",
    "print(\"Proporción de impagos en el conjunto de prueba (test): {:.2f}\".format(prop_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "skf = StratifiedKFold(n_splits=5, \n",
    "                      random_state=seed, \n",
    "                      shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;background-color: white;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GridSearchCV(cv=StratifiedKFold(n_splits=5, random_state=42, shuffle=True),\n",
       "             estimator=XGBClassifier(base_score=None, booster=None,\n",
       "                                     callbacks=None, colsample_bylevel=None,\n",
       "                                     colsample_bynode=None,\n",
       "                                     colsample_bytree=None,\n",
       "                                     early_stopping_rounds=None,\n",
       "                                     enable_categorical=False, eval_metric=None,\n",
       "                                     feature_types=None, gamma=None,\n",
       "                                     gpu_id=None, grow_policy=None,\n",
       "                                     importance_typ...\n",
       "                                     learning_rate=None, max_bin=None,\n",
       "                                     max_cat_threshold=None,\n",
       "                                     max_cat_to_onehot=None,\n",
       "                                     max_delta_step=None, max_depth=None,\n",
       "                                     max_leaves=None, min_child_weight=None,\n",
       "                                     missing=nan, monotone_constraints=None,\n",
       "                                     n_estimators=100, n_jobs=None,\n",
       "                                     num_parallel_tree=None, predictor=None,\n",
       "                                     random_state=None, ...),\n",
       "             param_grid={&#x27;learning_rate&#x27;: [0.1, 0.01], &#x27;max_depth&#x27;: [3, 5],\n",
       "                         &#x27;n_estimators&#x27;: [100, 500]},\n",
       "             scoring=&#x27;f1&#x27;)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" ><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GridSearchCV</label><div class=\"sk-toggleable__content\"><pre>GridSearchCV(cv=StratifiedKFold(n_splits=5, random_state=42, shuffle=True),\n",
       "             estimator=XGBClassifier(base_score=None, booster=None,\n",
       "                                     callbacks=None, colsample_bylevel=None,\n",
       "                                     colsample_bynode=None,\n",
       "                                     colsample_bytree=None,\n",
       "                                     early_stopping_rounds=None,\n",
       "                                     enable_categorical=False, eval_metric=None,\n",
       "                                     feature_types=None, gamma=None,\n",
       "                                     gpu_id=None, grow_policy=None,\n",
       "                                     importance_typ...\n",
       "                                     learning_rate=None, max_bin=None,\n",
       "                                     max_cat_threshold=None,\n",
       "                                     max_cat_to_onehot=None,\n",
       "                                     max_delta_step=None, max_depth=None,\n",
       "                                     max_leaves=None, min_child_weight=None,\n",
       "                                     missing=nan, monotone_constraints=None,\n",
       "                                     n_estimators=100, n_jobs=None,\n",
       "                                     num_parallel_tree=None, predictor=None,\n",
       "                                     random_state=None, ...),\n",
       "             param_grid={&#x27;learning_rate&#x27;: [0.1, 0.01], &#x27;max_depth&#x27;: [3, 5],\n",
       "                         &#x27;n_estimators&#x27;: [100, 500]},\n",
       "             scoring=&#x27;f1&#x27;)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" ><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: XGBClassifier</label><div class=\"sk-toggleable__content\"><pre>XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
       "              colsample_bylevel=None, colsample_bynode=None,\n",
       "              colsample_bytree=None, early_stopping_rounds=None,\n",
       "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
       "              gamma=None, gpu_id=None, grow_policy=None, importance_type=None,\n",
       "              interaction_constraints=None, learning_rate=None, max_bin=None,\n",
       "              max_cat_threshold=None, max_cat_to_onehot=None,\n",
       "              max_delta_step=None, max_depth=None, max_leaves=None,\n",
       "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
       "              n_estimators=100, n_jobs=None, num_parallel_tree=None,\n",
       "              predictor=None, random_state=None, ...)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-3\" type=\"checkbox\" ><label for=\"sk-estimator-id-3\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">XGBClassifier</label><div class=\"sk-toggleable__content\"><pre>XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
       "              colsample_bylevel=None, colsample_bynode=None,\n",
       "              colsample_bytree=None, early_stopping_rounds=None,\n",
       "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
       "              gamma=None, gpu_id=None, grow_policy=None, importance_type=None,\n",
       "              interaction_constraints=None, learning_rate=None, max_bin=None,\n",
       "              max_cat_threshold=None, max_cat_to_onehot=None,\n",
       "              max_delta_step=None, max_depth=None, max_leaves=None,\n",
       "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
       "              n_estimators=100, n_jobs=None, num_parallel_tree=None,\n",
       "              predictor=None, random_state=None, ...)</pre></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "GridSearchCV(cv=StratifiedKFold(n_splits=5, random_state=42, shuffle=True),\n",
       "             estimator=XGBClassifier(base_score=None, booster=None,\n",
       "                                     callbacks=None, colsample_bylevel=None,\n",
       "                                     colsample_bynode=None,\n",
       "                                     colsample_bytree=None,\n",
       "                                     early_stopping_rounds=None,\n",
       "                                     enable_categorical=False, eval_metric=None,\n",
       "                                     feature_types=None, gamma=None,\n",
       "                                     gpu_id=None, grow_policy=None,\n",
       "                                     importance_typ...\n",
       "                                     learning_rate=None, max_bin=None,\n",
       "                                     max_cat_threshold=None,\n",
       "                                     max_cat_to_onehot=None,\n",
       "                                     max_delta_step=None, max_depth=None,\n",
       "                                     max_leaves=None, min_child_weight=None,\n",
       "                                     missing=nan, monotone_constraints=None,\n",
       "                                     n_estimators=100, n_jobs=None,\n",
       "                                     num_parallel_tree=None, predictor=None,\n",
       "                                     random_state=None, ...),\n",
       "             param_grid={'learning_rate': [0.1, 0.01], 'max_depth': [3, 5],\n",
       "                         'n_estimators': [100, 500]},\n",
       "             scoring='f1')"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xg_params = {\n",
    "    'max_depth': [3, 5],\n",
    "    'learning_rate': [0.1, 0.01],\n",
    "    'n_estimators': [100,500],\n",
    "}\n",
    "\n",
    "tuned_xgboost = GridSearchCV(XGBClassifier(), \n",
    "                             param_grid=xg_params, \n",
    "                             cv=skf, \n",
    "                             scoring=\"f1\")\n",
    "tuned_xgboost.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "reporte_desempeno(y_test, tuned_xgboost.predict(X_test))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Podemos observar que el modelo no tiene un buen desempeño, pues las métricas f1 y recall se encuentras por debajo de 0.5, lo anterior se debe en gran parte al desbalance de clase, por lo tanto usaremos otras técnicas."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Ovarsampling"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "El oversampling consiste en agregar filas con muestras de la clase menos favorecida al conjunto de entrenamiento, de tal forma que la proporción de clases negativas y positivas sea la misma.\n",
    "\n",
    "En nuestro caso, usaremos una ténicnica llamada SMOTE  (Synthetic Minority Oversampling Technique) en la que se realiza la imputación de datos agregando varianza usando el algortimo KNN para crear la nueva data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "from imblearn.over_sampling import SMOTE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>Cupo_TC</th>\n",
       "      <th>SEX_1</th>\n",
       "      <th>SEX_2</th>\n",
       "      <th>EDUCATION_1</th>\n",
       "      <th>EDUCATION_2</th>\n",
       "      <th>EDUCATION_3</th>\n",
       "      <th>EDUCATION_4</th>\n",
       "      <th>MARRIAGE_1</th>\n",
       "      <th>MARRIAGE_2</th>\n",
       "      <th>MARRIAGE_3</th>\n",
       "      <th>...</th>\n",
       "      <th>PAGO_MES_t-1</th>\n",
       "      <th>PAGO_MES_t-2</th>\n",
       "      <th>PAGO_MES_t-3</th>\n",
       "      <th>PAGO_MES_t-4</th>\n",
       "      <th>PAGO_MES_t-5</th>\n",
       "      <th>default payment next month</th>\n",
       "      <th>AGE_DECILES</th>\n",
       "      <th>PROMEDIO_DEUDA</th>\n",
       "      <th>CUPO_DECILES</th>\n",
       "      <th>DEUDA/CUPO</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>80000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1740</td>\n",
       "      <td>1200</td>\n",
       "      <td>366</td>\n",
       "      <td>7432</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>29068.500000</td>\n",
       "      <td>3</td>\n",
       "      <td>0.363356</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>210000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>5082</td>\n",
       "      <td>6000</td>\n",
       "      <td>5900</td>\n",
       "      <td>4544</td>\n",
       "      <td>4700</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>143074.666667</td>\n",
       "      <td>7</td>\n",
       "      <td>0.681308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>230000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>4987</td>\n",
       "      <td>6222</td>\n",
       "      <td>15121</td>\n",
       "      <td>17425</td>\n",
       "      <td>17007</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>18364.166667</td>\n",
       "      <td>7</td>\n",
       "      <td>0.079844</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>130000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>4400</td>\n",
       "      <td>5000</td>\n",
       "      <td>4000</td>\n",
       "      <td>3500</td>\n",
       "      <td>4000</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>103045.666667</td>\n",
       "      <td>5</td>\n",
       "      <td>0.792659</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>470000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>3000</td>\n",
       "      <td>3016</td>\n",
       "      <td>2200</td>\n",
       "      <td>3000</td>\n",
       "      <td>1600</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>60088.833333</td>\n",
       "      <td>10</td>\n",
       "      <td>0.127849</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19090</th>\n",
       "      <td>280000</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>208141</td>\n",
       "      <td>5003</td>\n",
       "      <td>5002</td>\n",
       "      <td>5000</td>\n",
       "      <td>3000</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>122055.000000</td>\n",
       "      <td>8</td>\n",
       "      <td>0.435911</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19091</th>\n",
       "      <td>180000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1300</td>\n",
       "      <td>7504</td>\n",
       "      <td>1123</td>\n",
       "      <td>13000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>29093.166667</td>\n",
       "      <td>6</td>\n",
       "      <td>0.161629</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19092</th>\n",
       "      <td>180000</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1400</td>\n",
       "      <td>2200</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>26947.500000</td>\n",
       "      <td>6</td>\n",
       "      <td>0.149708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19093</th>\n",
       "      <td>320000</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1224</td>\n",
       "      <td>739</td>\n",
       "      <td>742</td>\n",
       "      <td>751</td>\n",
       "      <td>703</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>28293.333333</td>\n",
       "      <td>9</td>\n",
       "      <td>0.088417</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19094</th>\n",
       "      <td>260000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>300</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>351.166667</td>\n",
       "      <td>8</td>\n",
       "      <td>0.001351</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>19095 rows × 34 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Cupo_TC  SEX_1  SEX_2  EDUCATION_1  EDUCATION_2  EDUCATION_3   \n",
       "0        80000      1      0            1            0            0  \\\n",
       "1       210000      1      0            0            1            0   \n",
       "2       230000      1      0            1            0            0   \n",
       "3       130000      1      0            0            0            1   \n",
       "4       470000      1      0            0            1            0   \n",
       "...        ...    ...    ...          ...          ...          ...   \n",
       "19090   280000      0      1            0            1            0   \n",
       "19091   180000      1      0            0            1            0   \n",
       "19092   180000      0      1            0            1            0   \n",
       "19093   320000      0      1            0            1            0   \n",
       "19094   260000      1      0            0            1            0   \n",
       "\n",
       "       EDUCATION_4  MARRIAGE_1  MARRIAGE_2  MARRIAGE_3  ...  PAGO_MES_t-1   \n",
       "0                0           1           0           0  ...          1740  \\\n",
       "1                0           0           1           0  ...          5082   \n",
       "2                0           0           1           0  ...          4987   \n",
       "3                0           0           1           0  ...          4400   \n",
       "4                0           0           1           0  ...          3000   \n",
       "...            ...         ...         ...         ...  ...           ...   \n",
       "19090            0           1           0           0  ...        208141   \n",
       "19091            0           0           1           0  ...          1300   \n",
       "19092            0           0           1           0  ...          1400   \n",
       "19093            0           1           0           0  ...          1224   \n",
       "19094            0           1           0           0  ...             0   \n",
       "\n",
       "       PAGO_MES_t-2  PAGO_MES_t-3  PAGO_MES_t-4  PAGO_MES_t-5   \n",
       "0              1200           366          7432             0  \\\n",
       "1              6000          5900          4544          4700   \n",
       "2              6222         15121         17425         17007   \n",
       "3              5000          4000          3500          4000   \n",
       "4              3016          2200          3000          1600   \n",
       "...             ...           ...           ...           ...   \n",
       "19090          5003          5002          5000          3000   \n",
       "19091          7504          1123         13000             0   \n",
       "19092          2200             0             0             0   \n",
       "19093           739           742           751           703   \n",
       "19094             0             0             0           300   \n",
       "\n",
       "       default payment next month  AGE_DECILES  PROMEDIO_DEUDA  CUPO_DECILES   \n",
       "0                               0           10    29068.500000             3  \\\n",
       "1                               0            6   143074.666667             7   \n",
       "2                               0            9    18364.166667             7   \n",
       "3                               0            6   103045.666667             5   \n",
       "4                               0            6    60088.833333            10   \n",
       "...                           ...          ...             ...           ...   \n",
       "19090                           0            9   122055.000000             8   \n",
       "19091                           0            9    29093.166667             6   \n",
       "19092                           0            6    26947.500000             6   \n",
       "19093                           0            4    28293.333333             9   \n",
       "19094                           0            7      351.166667             8   \n",
       "\n",
       "       DEUDA/CUPO  \n",
       "0        0.363356  \n",
       "1        0.681308  \n",
       "2        0.079844  \n",
       "3        0.792659  \n",
       "4        0.127849  \n",
       "...           ...  \n",
       "19090    0.435911  \n",
       "19091    0.161629  \n",
       "19092    0.149708  \n",
       "19093    0.088417  \n",
       "19094    0.001351  \n",
       "\n",
       "[19095 rows x 34 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_modelo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "smote = SMOTE(random_state=seed)\n",
    "X, y = smote.fit_resample(df_modelo.drop(columns=target), df_modelo[target])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test= train_test_split(X, y, \n",
    "                                                    test_size=0.15, # 15% de registros destinados para test\n",
    "                                                    shuffle=True,\n",
    "                                                    random_state= seed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Proporcion de impagos en el conjunto de entrenamiento: 0.50\n",
      "Proporción de impagos en el conjunto de prueba (test): 0.50\n"
     ]
    }
   ],
   "source": [
    "# Nueva proporción\n",
    "prop_train = len(y_train[y_train==1]) / len(y_train)\n",
    "prop_test = len(y_test[y_test==1]) / len(y_test) \n",
    "print(\"Proporcion de impagos en el conjunto de entrenamiento: {:.2f}\".format(prop_train))\n",
    "print(\"Proporción de impagos en el conjunto de prueba (test): {:.2f}\".format(prop_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {color: black;background-color: white;}#sk-container-id-2 pre{padding: 0;}#sk-container-id-2 div.sk-toggleable {background-color: white;}#sk-container-id-2 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-2 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-2 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-2 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-2 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-2 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-2 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-2 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-2 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-2 div.sk-item {position: relative;z-index: 1;}#sk-container-id-2 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-2 div.sk-item::before, #sk-container-id-2 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-2 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-2 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-2 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-2 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-2 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-2 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-2 div.sk-label-container {text-align: center;}#sk-container-id-2 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-2 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GridSearchCV(cv=3,\n",
       "             estimator=XGBClassifier(base_score=None, booster=None,\n",
       "                                     callbacks=None, colsample_bylevel=None,\n",
       "                                     colsample_bynode=None,\n",
       "                                     colsample_bytree=None,\n",
       "                                     early_stopping_rounds=None,\n",
       "                                     enable_categorical=False, eval_metric=None,\n",
       "                                     feature_types=None, gamma=None,\n",
       "                                     gpu_id=None, grow_policy=None,\n",
       "                                     importance_type=None,\n",
       "                                     interaction_constraints=None,\n",
       "                                     learning_rate=None,...n=None,\n",
       "                                     max_cat_threshold=None,\n",
       "                                     max_cat_to_onehot=None,\n",
       "                                     max_delta_step=None, max_depth=None,\n",
       "                                     max_leaves=None, min_child_weight=None,\n",
       "                                     missing=nan, monotone_constraints=None,\n",
       "                                     n_estimators=100, n_jobs=None,\n",
       "                                     num_parallel_tree=None, predictor=None,\n",
       "                                     random_state=None, ...),\n",
       "             param_grid={&#x27;learning_rate&#x27;: [0.1, 0.01], &#x27;max_depth&#x27;: [3, 5],\n",
       "                         &#x27;n_estimators&#x27;: [100, 500]},\n",
       "             scoring=&#x27;f1&#x27;)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-4\" type=\"checkbox\" ><label for=\"sk-estimator-id-4\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GridSearchCV</label><div class=\"sk-toggleable__content\"><pre>GridSearchCV(cv=3,\n",
       "             estimator=XGBClassifier(base_score=None, booster=None,\n",
       "                                     callbacks=None, colsample_bylevel=None,\n",
       "                                     colsample_bynode=None,\n",
       "                                     colsample_bytree=None,\n",
       "                                     early_stopping_rounds=None,\n",
       "                                     enable_categorical=False, eval_metric=None,\n",
       "                                     feature_types=None, gamma=None,\n",
       "                                     gpu_id=None, grow_policy=None,\n",
       "                                     importance_type=None,\n",
       "                                     interaction_constraints=None,\n",
       "                                     learning_rate=None,...n=None,\n",
       "                                     max_cat_threshold=None,\n",
       "                                     max_cat_to_onehot=None,\n",
       "                                     max_delta_step=None, max_depth=None,\n",
       "                                     max_leaves=None, min_child_weight=None,\n",
       "                                     missing=nan, monotone_constraints=None,\n",
       "                                     n_estimators=100, n_jobs=None,\n",
       "                                     num_parallel_tree=None, predictor=None,\n",
       "                                     random_state=None, ...),\n",
       "             param_grid={&#x27;learning_rate&#x27;: [0.1, 0.01], &#x27;max_depth&#x27;: [3, 5],\n",
       "                         &#x27;n_estimators&#x27;: [100, 500]},\n",
       "             scoring=&#x27;f1&#x27;)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-5\" type=\"checkbox\" ><label for=\"sk-estimator-id-5\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: XGBClassifier</label><div class=\"sk-toggleable__content\"><pre>XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
       "              colsample_bylevel=None, colsample_bynode=None,\n",
       "              colsample_bytree=None, early_stopping_rounds=None,\n",
       "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
       "              gamma=None, gpu_id=None, grow_policy=None, importance_type=None,\n",
       "              interaction_constraints=None, learning_rate=None, max_bin=None,\n",
       "              max_cat_threshold=None, max_cat_to_onehot=None,\n",
       "              max_delta_step=None, max_depth=None, max_leaves=None,\n",
       "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
       "              n_estimators=100, n_jobs=None, num_parallel_tree=None,\n",
       "              predictor=None, random_state=None, ...)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-6\" type=\"checkbox\" ><label for=\"sk-estimator-id-6\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">XGBClassifier</label><div class=\"sk-toggleable__content\"><pre>XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
       "              colsample_bylevel=None, colsample_bynode=None,\n",
       "              colsample_bytree=None, early_stopping_rounds=None,\n",
       "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
       "              gamma=None, gpu_id=None, grow_policy=None, importance_type=None,\n",
       "              interaction_constraints=None, learning_rate=None, max_bin=None,\n",
       "              max_cat_threshold=None, max_cat_to_onehot=None,\n",
       "              max_delta_step=None, max_depth=None, max_leaves=None,\n",
       "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
       "              n_estimators=100, n_jobs=None, num_parallel_tree=None,\n",
       "              predictor=None, random_state=None, ...)</pre></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "GridSearchCV(cv=3,\n",
       "             estimator=XGBClassifier(base_score=None, booster=None,\n",
       "                                     callbacks=None, colsample_bylevel=None,\n",
       "                                     colsample_bynode=None,\n",
       "                                     colsample_bytree=None,\n",
       "                                     early_stopping_rounds=None,\n",
       "                                     enable_categorical=False, eval_metric=None,\n",
       "                                     feature_types=None, gamma=None,\n",
       "                                     gpu_id=None, grow_policy=None,\n",
       "                                     importance_type=None,\n",
       "                                     interaction_constraints=None,\n",
       "                                     learning_rate=None,...n=None,\n",
       "                                     max_cat_threshold=None,\n",
       "                                     max_cat_to_onehot=None,\n",
       "                                     max_delta_step=None, max_depth=None,\n",
       "                                     max_leaves=None, min_child_weight=None,\n",
       "                                     missing=nan, monotone_constraints=None,\n",
       "                                     n_estimators=100, n_jobs=None,\n",
       "                                     num_parallel_tree=None, predictor=None,\n",
       "                                     random_state=None, ...),\n",
       "             param_grid={'learning_rate': [0.1, 0.01], 'max_depth': [3, 5],\n",
       "                         'n_estimators': [100, 500]},\n",
       "             scoring='f1')"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xg_params = {\n",
    "    'max_depth': [3, 5],\n",
    "    'learning_rate': [0.1, 0.01],\n",
    "    'n_estimators': [100,500],\n",
    "}\n",
    "\n",
    "tuned_xgboost = GridSearchCV(XGBClassifier(), \n",
    "                             param_grid=xg_params,  \n",
    "                             cv=3,\n",
    "                             scoring=\"f1\")\n",
    "tuned_xgboost.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "reporte_desempeno(y_test, tuned_xgboost.predict(X_test))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Observamos que al aplicar la técnica SMOTE de oversampling obtenemos resultados satisfactorios, con un Accuracy cercano al 90%, recall en 83% y F1-Score en 89%. Por lo tanto, es posible explicar los impagos del próximo mes mediante este modelo."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "venv",
   "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": "3.11.3"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}