{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ " " ] }, { "cell_type": "markdown", "metadata": { "id": "C1LAAic_C1Lt" }, "source": [ "# Clasificación\n", "\n", "## Práctica" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "zI4khfezCNJZ" }, "outputs": [], "source": [ "import pandas as pd\n", "import numpy as np\n", "from sklearn import preprocessing\n", "import matplotlib.pyplot as plt \n", "plt.rc(\"font\", size=14)\n", "from sklearn.linear_model import LogisticRegression\n", "from sklearn.model_selection import train_test_split\n", "import seaborn as sns\n", "sns.set(style=\"white\")\n", "sns.set(style=\"whitegrid\", color_codes=True)" ] }, { "cell_type": "markdown", "metadata": { "id": "KC8azJ77DEcE" }, "source": [ "El dataset contiene datos relacionados con las campañas de marketing dirigidas (llamadas telefónicas) de una entidad bancaria portuguesa. El objetivo de la clasificación es predecir si el cliente suscribirá (1/0) un depósito a plazo (variable y)." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "uKg2Cc3oCzkq" }, "outputs": [], "source": [ "data = pd.read_csv('https://raw.githubusercontent.com/madmashup/targeted-marketing-predictive-engine/master/banking.csv')" ] }, { "cell_type": "markdown", "metadata": { "id": "apZho91aDatY" }, "source": [ "El conjunto de datos proporciona la información de los clientes del banco. Incluye 41.188 registros y 21 campos." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "SodO6-7uC2hA", "outputId": "59f808d0-1783-451c-c0ba-d92ddac992d1" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(41188, 21)\n", "Index(['age', 'job', 'marital', 'education', 'default', 'housing', 'loan',\n", " 'contact', 'month', 'day_of_week', 'duration', 'campaign', 'pdays',\n", " 'previous', 'poutcome', 'emp_var_rate', 'cons_price_idx',\n", " 'cons_conf_idx', 'euribor3m', 'nr_employed', 'y'],\n", " dtype='object')\n" ] } ], "source": [ "data = data.dropna()\n", "print(data.shape)\n", "print(data.columns)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 235 }, "id": "Nb4sPjRTDS6d", "outputId": "2472fb8c-31df-489f-d304-98bb3507827a" }, "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", "
agejobmaritaleducationdefaulthousingloancontactmonthday_of_week...campaignpdayspreviouspoutcomeemp_var_ratecons_price_idxcons_conf_idxeuribor3mnr_employedy
044blue-collarmarriedbasic.4yunknownyesnocellularaugthu...19990nonexistent1.493.444-36.14.9635228.10
153technicianmarriedunknownnononocellularnovfri...19990nonexistent-0.193.200-42.04.0215195.80
228managementsingleuniversity.degreenoyesnocellularjunthu...362success-1.794.055-39.80.7294991.61
\n", "

3 rows × 21 columns

\n", "
\n", " \n", " \n", " \n", "\n", " \n", "
\n", "
\n", " " ], "text/plain": [ " age job marital education default housing loan \\\n", "0 44 blue-collar married basic.4y unknown yes no \n", "1 53 technician married unknown no no no \n", "2 28 management single university.degree no yes no \n", "\n", " contact month day_of_week ... campaign pdays previous poutcome \\\n", "0 cellular aug thu ... 1 999 0 nonexistent \n", "1 cellular nov fri ... 1 999 0 nonexistent \n", "2 cellular jun thu ... 3 6 2 success \n", "\n", " emp_var_rate cons_price_idx cons_conf_idx euribor3m nr_employed y \n", "0 1.4 93.444 -36.1 4.963 5228.1 0 \n", "1 -0.1 93.200 -42.0 4.021 5195.8 0 \n", "2 -1.7 94.055 -39.8 0.729 4991.6 1 \n", "\n", "[3 rows x 21 columns]" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data.head(3)" ] }, { "cell_type": "markdown", "metadata": { "id": "O2L0WCNyD3aL" }, "source": [ "**Variables de entrada**\n", "\n", "* edad (numérica)\n", "* trabajo : tipo de trabajo (categórico: \"administrativo\", \"obrero\", \"empresario\", \"empleado doméstico\", \"directivo\", \"jubilado\", \"autónomo\", \"servicios\", \"estudiante\", \"técnico\", \"desempleado\", \"desconocido\")\n", "* marital : estado civil (categórico: \"divorciado\", \"casado\", \"soltero\", \"desconocido\")\n", "* educación (categórica: \"básico.4y\", \"básico.6y\", \"básico.9y\", \"bachillerato\", \"analfabeto\", \"curso.profesional\", \"título.universitario\", \"desconocido\")\n", "* morosidad: ¿tiene crédito en mora? (categórico: \"no\", \"sí\", \"desconocido\")\n", "* vivienda: ¿tiene préstamo para la vivienda? (categórico: \"no\", \"sí\", \"desconocido\")\n", "* préstamo: ¿tiene un préstamo personal? (categórico: \"no\", \"sí\", \"desconocido\")\n", "* contacto: tipo de comunicación del contacto (categórico: \"celular\", \"teléfono\")\n", "* mes: último mes del año del contacto (categórico: \"ene\", \"feb\", \"mar\", ..., \"nov\", \"dec\")\n", "* día_de_la_semana: día de la semana del último contacto (categórico: \"mon\", \"tue\", \"wed\", \"thu\", \"fri\")\n", "* duración: duración del último contacto, en segundos (numérico). Nota importante: este atributo afecta en gran medida al objetivo de salida (por ejemplo, si duration=0 entonces y='no'). La duración no se conoce antes de que se realice una llamada, además, tras el final de la llamada, y obviamente se conoce. Por lo tanto, esta entrada sólo debería incluirse a efectos de referencia y debería descartarse si se pretende tener un modelo predictivo realista\n", "* campaña: número de contactos realizados durante esta campaña y para este cliente (numérico, incluye el último contacto)\n", "* pdays: número de días transcurridos desde que se contactó por última vez con el cliente en una campaña anterior (numérico; 999 significa que no se contactó con el cliente previamente)\n", "* anterior: número de contactos realizados antes de esta campaña y para este cliente (numérico)\n", "* poutcome: resultado de la campaña de marketing anterior (categórico: \"fracaso\", \"inexistente\", \"éxito\")\n", "* emp.var.rate: tasa de variación del empleo - (numérico)\n", "* cons.price.idx: índice de precios al consumo - (numérico)\n", "* cons.conf.idx: índice de confianza del consumidor - (numérico)\n", "* euribor3m: índice euribor a 3 meses - (numérico)\n", "* nr.employed: número de empleados - (numérico)" ] }, { "cell_type": "markdown", "metadata": { "id": "0m4EvSGWNnc9" }, "source": [ "**Variable de predicción (objetivo deseado):**\n", "y - ¿ha suscrito el cliente un depósito a plazo? (binario: \"1\", significa \"Sí\", \"0\" significa \"No\")" ] }, { "cell_type": "markdown", "metadata": { "id": "Zj8Jy0hjN1OD" }, "source": [ "### Pre-processing + EDA" ] }, { "cell_type": "markdown", "metadata": { "id": "-RmzOPbAN23m" }, "source": [ "La columna de educación del conjunto de datos tiene muchas categorías y necesitamos reducir las categorías para una mejor modelización. La columna de educación tiene las siguientes categorías:" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "FVoxteXTNrP2", "outputId": "fcc04323-bffe-4be0-93af-f77d35920136" }, "outputs": [ { "data": { "text/plain": [ "array(['basic.4y', 'unknown', 'university.degree', 'high.school',\n", " 'basic.9y', 'professional.course', 'basic.6y', 'illiterate'],\n", " dtype=object)" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data.education.unique()" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "l6wm91HQN7Qj" }, "outputs": [], "source": [ "data['education']=np.where(data['education'] =='basic.9y', 'Basic', data['education'])\n", "data['education']=np.where(data['education'] =='basic.6y', 'Basic', data['education'])\n", "data['education']=np.where(data['education'] =='basic.4y', 'Basic', data['education'])" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "frUwZ7TJN8MC", "outputId": "d999c6b9-ec6f-4910-e906-1d116099fef8" }, "outputs": [ { "data": { "text/plain": [ "array(['Basic', 'unknown', 'university.degree', 'high.school',\n", " 'professional.course', 'illiterate'], dtype=object)" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data.education.unique()" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "e1Sffy6dN9Lk", "outputId": "4cc2eb46-f70c-4248-e254-48d583a148e8" }, "outputs": [ { "data": { "text/plain": [ "0 36548\n", "1 4640\n", "Name: y, dtype: int64" ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data.y.value_counts()" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "p4eZWeBuOTLq", "outputId": "805e6a63-c3b2-498d-f301-38c986892bad" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "porcentaje de no subscripcion: 88.73458288821988\n", "porcentaje de subscripcion 11.265417111780131\n" ] } ], "source": [ "count_no_sub = len(data[data['y']==0])\n", "count_sub = len(data[data['y']==1])\n", "pct_of_no_sub = count_no_sub/(count_no_sub+count_sub)\n", "print(\"porcentaje de no subscripcion: \", pct_of_no_sub*100)\n", "pct_of_sub = count_sub/(count_no_sub+count_sub)\n", "print(\"porcentaje de subscripcion\", pct_of_sub*100)" ] }, { "cell_type": "markdown", "metadata": { "id": "EWsu0uKzPZmp" }, "source": [ "Nuestras clases están desequilibradas, y la proporción de instancias sin suscripción y con suscripción es de 89:11. Antes de seguir adelante con el equilibrio de las clases, vamos a hacer un poco más de exploración." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 186 }, "id": "ZxWrGQ88Pcbo", "outputId": "7e396cac-940c-4b39-d418-bd913a992726" }, "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", "
agedurationcampaignpdayspreviousemp_var_ratecons_price_idxcons_conf_idxeuribor3mnr_employed
y
039.911185220.8448072.633085984.1138780.1323740.24887593.603757-40.5930973.8114915176.166600
140.913147553.1911642.051724792.0355600.492672-1.23344893.354386-39.7897842.1231355095.115991
\n", "
\n", " \n", " \n", " \n", "\n", " \n", "
\n", "
\n", " " ], "text/plain": [ " age duration campaign pdays previous emp_var_rate \\\n", "y \n", "0 39.911185 220.844807 2.633085 984.113878 0.132374 0.248875 \n", "1 40.913147 553.191164 2.051724 792.035560 0.492672 -1.233448 \n", "\n", " cons_price_idx cons_conf_idx euribor3m nr_employed \n", "y \n", "0 93.603757 -40.593097 3.811491 5176.166600 \n", "1 93.354386 -39.789784 2.123135 5095.115991 " ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data.groupby(\"y\").mean()" ] }, { "cell_type": "markdown", "metadata": { "id": "DsvM_zpQPl3C" }, "source": [ "**Observaciones:**\n", "\n", "* La edad media de los clientes que compraron el depósito a \n", "plazo es mayor que la de los clientes que no lo hicieron.\n", "* El pdays (días transcurridos desde que se contactó con el cliente por última vez) es comprensiblemente más bajo para los clientes que lo compraron. Cuanto más bajo sea el pdays, mejor será el recuerdo de la última llamada y, por tanto, mayores serán las posibilidades de venta.\n", "* Sorprendentemente, las campañas (número de contactos o llamadas realizadas durante la campaña actual) son menores para los clientes que compraron el depósito a plazo." ] }, { "cell_type": "markdown", "metadata": { "id": "y3GeX4SDP6it" }, "source": [ "### Visualizaciones" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 369 }, "id": "KyqOHRcxPvMJ", "outputId": "e8601d73-2288-428a-fd14-d56a6ad087f6" }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "%matplotlib inline\n", "pd.crosstab(data.job,data.y).plot(kind='bar')\n", "plt.title('Purchase Frequency for Job Title')\n", "plt.xlabel('Job')\n", "plt.ylabel('Frequency of Purchase')\n", "plt.savefig('purchase_fre_job')" ] }, { "cell_type": "markdown", "metadata": { "id": "lTWqu3wBP-YF" }, "source": [ "La frecuencia de compra del depósito depende en gran medida del cargo. Así pues, el título de trabajo puede ser un buen predictor de la variable de resultado." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 341 }, "id": "gXpthcIXP3A1", "outputId": "fab70700-6af1-49a7-e804-dd330911e085" }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "table=pd.crosstab(data.marital,data.y)\n", "table.div(table.sum(1).astype(float), axis=0).plot(kind='bar', stacked=True)\n", "plt.title('Stacked Bar Chart of Marital Status vs Purchase')\n", "plt.xlabel('Marital Status')\n", "plt.ylabel('Proportion of Customers')\n", "plt.savefig('mariral_vs_pur_stack')" ] }, { "cell_type": "markdown", "metadata": { "id": "VNGfYl_0QGhT" }, "source": [ "El estado civil no parece ser un fuerte predictor de la variable de resultado." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 398 }, "id": "sTk4NRh7QCOD", "outputId": "07ff71d4-36b0-4923-d8cc-f05269cdf951" }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYkAAAF9CAYAAADx3pSyAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3dd5hcddnG8W8SUuiRQAKEXnILBFCCIkhvUsVXRaRIUVEsFAEBETH40lGUDgLSiVRBkCYqRYpCFDACD7zUJCQQQk2AhJT3j98ZmEx2ds8mM3t2zt6f69prd86cmfOc2Zl5zq/3mj17NmZmZm3pXXQAZmbWfTlJmJlZXU4SZmZWl5OEmZnV5SRhZmZ1OUmYmVldThIlIOlSScc34Hn2lfT3RsTUDJJWkjRb0gJFx1Ih6X8kjZU0RdKn5/O5Npc0rlGxdeK4R0u6qKuP250U9dq3gm7zYSsjSRsDpwJrATOBp4BDIuIRSfsC346IjQsMMTdJKwEvAFOzTVOBG4CDI+LDBh5nGHACsAXQF3gJuBQ4o1HHqDnebGD1iPi/eXyKXwI/jIib23n+94DqAUm/iIhT5/F480XS5sCVEbFcZVtEnFhELHnVvIZvA9cAP46ImYUG1kO4JNEkkhYDbgXOApYAhgLHAdOKjKsBBkbEIsDawIbAD+blSdoqDUhaFfgHMBZYOyIWB3YF1gcWneeIcx5/Hq0I/LeDfdaNiEWqfgpJEC1u3ex9txWwB7B/Z5+gO5VAW4lftOYZBhARo7Lb7wN3AUhaAzgf6CtpCjAjIgZK2hE4HliVdMV0cUSMrDxhVclkTeBd4GcRcWn1QSUtCvwR+A9wMCBSohoBTMoec2227yDgEmBz4GngzrwnFxGvSfpzFkvl2EeRPryDSV/0P42IP2T37Zvd909gb+A84Jiapz0OeDAiDq06TpC+FJA0MNu8p6T/BRYCfh0RJ2T3f5ZU4liD9HrfABwaEdOz+2cDPwQOARaQNDZ7vsez+74VEddUBySpN3B0FvuCwB3AgcAHwGSgT/b4iRGxat7XL3vuBbPXYRdgAul/UX3/HKUcSZcC4yLimOz2Ltlrtgrpf/uDiLhD0n7AEcBy2fZTIuICSQsDtwP9s/cdpPfpd4DVImKv7Hm/CJxEurB5DPheRDyV3fcicDbpf7hi9nrsExEf1MTeH3gV2DgixmTblgJezh43i1RC3Dj7+7/AZhExq73XLCKelnQ/MLytUlEW37cj4m5JI4HhpP/VF4FDJd0I/Ar4Aun/eW9EfKnq8YcBR5JK/kdHxCXZ9rqfTUkDgIuA7Unvh2eBnSLiVUmLA6cDO2TneQnw81YqBbkk0TzPADMlXSZpe0mfqNyRfeAOAB7KriwrX35TSR++gcCOwPckfQlA0oqkD/hZwFLAp0gf4I9kX/p/AR6IiINIX6J/Bq4mfXF/HThXUuWL/RzSB2gZ4JvZTy6SliV90B6u2vwcsAmwOOnL60pJy1TdvwHwPDCEVKVUa2vg+hyH35iU/LYCjs2SLqQP9o+AJUmlnK2A79c89ktZHGtGxKbZtsqV/jXMbd/sZwvSl/EiwNkRMS27sq08vlMJIvNz0pfOqqTXcp+8D8wS4uXAj0nvl02BF7O7XwN2AhYD9gN+LWm9iJhK+iJ7papU80rN8w4DRpES6VLAbcAtkvpV7fY1YDtgZWAd0uszh4iYBtwI7F7zuHsj4jXgMGBcdowhpETc4RxB2Xt3E+DfHe2b2YX0nhoIXAVcQfpcrEX6TPy6at+lSe/docC3gHOqPrd1P5uk/9viwPLAINJn+/3svkuBGcBqwKeBbYFv54y9W3BJokki4p3syv9I4EJgaUm3AftHxKt1HnNP1c0nJI0CNgNuIl1N311VMpmc/VQsC9wLXBYRp2XbdgJerFwNAf+WdAOwa9bQ/RVStc5UYIyky0hfNu15XRKkD8VDVH2pR8R1VftdI+knwGeBSn39KxFxVvb3jDaeexDpirojx0XE+6Qr+MeBdYGnImJ01T4vSrqA9Pr9pmr7SRHxRo5jVOwJnB4RzwNk5zRG0n4R0dY5tOVfkqqvkHeLiDtJX5rfz+J5Q9KZwLE5n/NbwO8i4s/Z7fGVOyLiT1X73SvpLtIX679yPO9uwJ8qzyvpl6QS6UbAPdk+Z1aSi6RbSBcsbbkauAD4aXZ7j+w2wIeki5MVs5LS/R3E9S9JM4E3SFftl9DxexXShdhNWawDSUlyUES8md1/b9W+H5Lai2YAt2WlLQEPd/DZ/JD03l0tIp4ARmfHG0IqQQzM3q9TJf2aVHK7gBbhJNFEWYlhXwBJnwSuJH1h7d7W/pI2AE4mFZH7Af2Byhfv8qQr9Xp2BKaQqrEqVgQ2kPRW1bYFSFdTS2V/j62676Ucp7VkRMzIqkp+Qaqi2jCLf2/gUGClbN9FSFf1FdXHastk0hdHRyZW/f1edpzKVfDppDaMhUjnN7rmsR3FUGtZ5nxdXsqedwhVX8wdWK9Ow/iydP71r1iedJU/F0nbk0opw0i1BQuRqh/zmON8I2JWVi03tGqf2td/2TrP9Tdgoex9/Sopmfwhu+80YCRwV3bR8duIOLmduOZ6DbPHdaT69V0eeKMqQdSaXJP4q99b7X02r8ie+/dZIrqSlBhXJHW+mFAVa286/x4slKubukhEPE0qeg7PNrVVtL6a1J6wfNZoez7QK7tvLKlaop4LSfXDt2V1z5XH3BsRA6t+FomI75HqqmeQ3twVK3TifN7PzudzkpbMqsMuJNX5D8qq0MZUxV/vnKvdTSrdzKvzSG0rq0fEYqQqjF41+3R22uNXSB/2ihVIr1ubpcFOmkD7r/97pC/4iqWr/m7z/ZC1BdxA6nU1JPs/3MbHr0NH5z/H+UrqlcWYNyF+JKt3v5Z0UbQ7cGtEvJvd925EHBYRq/Bxe8FWnTzEVKpeH0l9SBc/1arPdyywRFXbVmfU/WxGxIcRcVxErEkqce1EqpoaS+qosmTV52+xiFhrHo5fGJckmiQrOewIXBMR4yQtT/qgVOrwXwWWk9Sv0rBK6sHzRkR8kNU570HW2E2qTz1a0tdIdb2Lk96w1e0SPyR9Ud+SNbTdCpws6RvA77N9PgVMiYinska8kZK+Sbr634eP67U7Or/+wDdIV5WTSfW7s0nJh6zxdHjdJ2jbz4FHJJ0G/CoiJkpajXTF+cMcj18UeAeYkr3+lWTYnldJbQ31usCOAo6UdHv2XCeS/qd5q5racy3wE0n/ABYmNYhXewzYQ9J/gW1I1RuPZvddTLoKv5V0xb4M6fzHk65yJwEzslLFtqSEDel8B0laPCLerhPTUdkX9n2kqqZpwIPzeI5Xk6pkJvNxtROSdiIl9OdIDcEzSQ27nfEMMCB7r99FuijoX2/niJiQ/R/PlfQDUsl7w4i4L8ex6n42JW0BvA48SXr/fQjMyo53F/ArST/LjrcysFxE3NvWQbojlySa511SA+k/JE0lJYcxpAY7gL+SenRMlPR6tu37wC8kvUuqm7628mQR8TKpfvMwUr3sY6S6eKr2mU2q7xxHagf4kPQF8XXSFeJE4BQ+/iD9kFScnkgqFczRu6aOt7K62ldJ1UxfjIjZEfEkqdfIQ9l9awMP5Hi+6vify55zJeC/kt4mXRU/Sno9O3I46cP7LilZttUQXWskcJmkt7IEXOt3pOqE+0jjRD5g7i/zjjyuNNiu8lNpIzmOVLXzAukL54qaxx0M7Ay8RWobualyR0T8k6xRmvQley+pfv9d4CDSe+dN0uvxx6rHPU1KfM9n5zxHVVHWm2wvUgeJ17Pj71x1IdMpEfEP0hX/sqSOFxWrk0qOU0jvmXMj4m+dfO63SZ+Zi0jJcSrpvd+eb5A+F0+TGvgPyXm4up9NUgnvelKCeIr0v6j8L/cmVU89Sfp/XE++KtVuo5cXHTIzs3pckjAzs7qcJMzMrC4nCTMzq6s0vZtGjx7dH/gMqVthywx5NzMrWB9SY/ojI0aMmGtuudIkCVKC6GjUppmZtW0TYK6lAsqUJCYADBs2jH79+nW0b0OMGTOG4cM7OxSgdfj8WleZzw18fo00ffp0nnnmGagzJU6ZksRMgH79+tG/f93xNA3Xlccqgs+vdZX53MDn1wRtVtO74drMzOpykjAzs7qcJMzMrC4nCTMzq8tJwszM6uqS3k3Z6lZfIc3uuXZlzduaffoAZ5KWRZwNnBwRF3VFfGZm1rauKkncRFpqsL2Vt/YkrQO7Omm66JGSVmp+aGZmVk+XJImI+HtEdLRk327AhRExKyImkRLLrs2PzszM6ulOg+lWYM6SxsvMubRjLmPGzFWT1aG11/wk/RZcuOMda4wYMaLTjwGY/v5U/vPk0/P02Hnh82tbK5xfmc8NfH71dKfz605JoiGGDx8+TyMVnz9hfpZW7pxVfnrDPL8J5pXPr3G6+vzKfG7g82ukeTm/adOmtXtx3Z16N73M3AvOd1RFZWZmTdSdShLXAftLuhEYBHyJNCuhmZkVpEtKEpLOlDQOWA64W9J/s+23SVo/2+0K4HngWeBh4BcR8UJXxGdmZm3rkpJERBwEHNTG9h2q/p4JfK8r4jEzs3y6U5uEmZl1M04SZmZWl5OEmZnV5SRhZmZ1OUmYmVldThJmZlaXk4SZmdXlJGFmZnU5SZiZWV1OEmZmVpeThJmZ1eUkYWZmdXWnqcILM+vD6azy0xu69Hi9+/brsuOZmc0rJwmgd99+7HzYzV12vFt+tUuXHcvMbH64usnMzOpykjAzs7pc3dQDuM3FzOaVk0QP4DaX1uUEb0VzkjDrxpzgrWhukzAzs7qcJMzMrK55ShKSVpG0UoNjMTOzbiZXm4SkUcBZEfGgpP2Ac4FZkg6KiIubGqGZlZYb5ru/vA3XWwH7ZH8fCmwNvAXcBDhJmNk8ccN895c3SfSLiOmShgJLRMQDAJKGNC80M7PWVoaSUt4k8ZiknwArAn8CyBLGOw2NxsysRMpQUsrbcP0tYG1gQeCYbNuGwFUNj8jMzLqNDksSkvqQ2iO+GREfVLZHxPXA9U2MzczMCtZhSSIiZgLfB6Y3PxwzM+tO8lY3XQ4c0MxAzMys+8nbcP1Z4EBJRwBjgdmVOyJi02YEZmZmxcubJC7MfszMrAfJlSQi4rL5PZCkYcBlwCBgMrB3RDxbs89g4BJgeaAv8DfgoIiYMb/HNzOzzsvVJiGpl6T9Jf1V0hPZtk0lfa0TxzofOCcihgHnABe0sc/RwFMRsQ6wDjAC+HInjmFmZg2Ut+H6F6SxEr8FVsi2jQOOzPPgrISwHjAq2zQKWE/SUjW7zgYWldQb6A/0A8bnjNHMzBosb5vEvsCnI+J1Sedl214AVsn5+OWB8Vl3WiJipqRXsu2Tqvb7X+AGYAKwMHB2ZQqQvMaMGdOZ3QEYMWJEpx8zv0aPHt1lx/L5NV5XnV+Zzw18fs3Q6PPLmyT6AFOyvys9mxap2tYouwJPkCYUXBS4XdJXs4F7uQwfPpz+/fs3OKzGK+LN05V8fq2rzOcGPr9a06ZNa/fiOm91023A6ZL6Q2qjIF3135Lz8WOBodno7coo7mWz7dUOBK6KiFkR8TZwM7BFzmOYmVmD5U0ShwLLAG8Di5NKECuSs00iIl4DHgN2zzbtDvw7IibV7PoCsB2ApH6kKck7X39kZmYNkbcL7DvA/2RTg68AjI2IiZ081gHAZZKOBd4E9gaQdBtwbEQ8ChwCnC/pP6Qqrr/h8RlmZoXJ2yZR8T6pt1FvScsCRMQreR4YEU8DG7SxfYeqv58DtulkTNbDlWHOfrPuKu/ypVuTur+uCPSqums26YrfrDBlmLPfrLvK2yZxMXAiqT2ib9WPL6fMzEosb3XTAOCSyjgHMzPrGfKWJH4NHJF1fTUzsx4ib0niBuBO4CeSXq++IyLyjro2M7MWkzdJXA/cD1xH6uFkZmY9QN4ksTJp7qZZzQzGzMy6l7xtEjcDWzYzEDMz637yliT6A3+UdD/wavUdEbF3w6MyM7NuIW+S+G/2Y2ZmPUjeuZuOa3YgZmbW/eSeu0nS5qRJ+YaS5m+6IiL+1qS4zMysG8i7xvW3gWuBicCNpJXjRknav4mxmZlZwfKWJI4AtomIxysbJF1DGmTnqbzNzEoqbxfYQcCTNdsCWKKx4ZiZWXeSN0n8nbR86UIAkhYGTgMebFZgZmZWvLxJ4gBgXeBtSa8Cb2W3D2hWYGZmVry8XWAnAJtKWg5YFnglIsY1NTIzMytc3t5N/waIiHER8c9KgpD0aDODMzOzYuWtblqtdkO2toSnCTczK7F2q5skXZ792a/q74qV8FQdZmal1lGbxHN1/p4NPEBaX8LMzEqq3SRRmbNJ0sMRcWfXhGRmZt1F3hHX0yWtHBEvSFoaOAWYBfwkIiY2LzwzMytS3obrc4GZ2d+nA31JSeK3zQjKzMy6h7wliaER8bKkBYAvACsC04FXmhaZmZkVLm9J4h1JQ4DNgCcjYkq2vW9zwjIzs+4gb0niLOARoB9wSLbt88DTzQjKzMy6h1wliYg4Bdga+HxE/D7bPB74drMCMzOz4uVemS4inmnvtpmZlU+uJCFpLGkA3VwiYoWGRmRmZt1G3pLEXjW3lwEOBn7fxr5mZlYSeacKv7d2m6R7gDuAM/I8h6RhwGWkVe4mA3tHxLNt7Pc14GdAL1LpZeuIeDXPMczMrLHydoFtyzRg5U7sfz5wTkQMA84BLqjdQdL6wEjSetrDgY2Bt+cjRjMzmw952yR+UbNpIWAH4Pacjx8MrAdsk20aBZwtaamImFS164+AX1am+ogIJwgzswLlbZNYvub2VNL0HFd04vHjI2ImQETMlPRKtr06SawJvCDpPmAR4EbghIhos9G8LWPGjMm760dGjBjR6cfMr9GjR3fZsXx+jddV51fmcwOfXzM0+vzytkns19Cj1tcHWIdU4uhHavN4Gahdy6Ku4cOH079//+ZE10BFvHm6ks+vdZX53MDnV2vatGntXly32yYh6fOSTqlz38mSPpczjrHAUEl9ssf2Ia2VPbZmv5eB6yNiWkS8C9wMfDbnMczMrME6arg+Grivzn33AD/Nc5CIeA14DNg927Q78O+a9giAq4FtJfWS1BfYCng8zzHMzKzxOkoSnyJV+bTlbqAz5ZoDgAMlPQMcmN1G0m1ZryZI4y5eA54kJZX/Ahd34hhmZtZAHbVJLEZqG3i/jfv6AovmPVBEPA1s0Mb2Har+ngUcmv2YmVnBOipJPA1sW+e+bfEssGZmpdZRSeLXwAVZQ/NNETFLUm/gS6QBcb7iNzMrsXaTRERcna1pfRnQX9LrwJKk0dY/j4hRXRCjmZkVpMNxEhFxuqSLgA35eN6lhyLinWYHZ2Zmxco7mO4d4M4mx2JmZt3M/EzwZ2ZmJeckYWZmddVNEpK+WPV3364Jx8zMupP2ShJXVv09udmBmJlZ99New/VEST8kTZGxgKQtSKvFzSEi/tqs4MzMrFjtJYl9gV+Q1rLuB/yujX1mA6s0PiwzM+sO6iaJiHgQ2BpA0v9FxGpdFpWZmXULecdJrAYgaQVgKDAuImrXgjAzs5LJu8b10sA1pFHXk4FBkh4Gvh4RrzQxPjMzK1DecRLnkxb/+URELAN8Avh3tt3MzEoqV0kC2BhYJiI+BIiIqZKOAMY3LTIzMytc3pLEm8CaNdsEvNXYcMzMrDvJW5I4Fbhb0sXAS8CKwH7Az5oVmJmZFS9XSSIiLgR2I60lsXP2e4+I+G0TYzMzs4LlLUlURlZ7dLWZWQ/iWWDNzKwuJwkzM6vLScLMzOpykjAzs7ryTsuxBHA48Clgker7ImLTJsRlZmbdQN7eTVcD/YFrgfeaF46ZmXUneZPERsBSETGtmcGYmVn3krdN4glguWYGYmZm3U/eksRfgTskXQJMrL4jItpasc7MzEogb5LYBBgHbFOzfTZtL2tqZmYlkHdlui2aHYiZmXU/uedukvQJ0uR+Q0nrSNwSEW82KzAzMyteroZrSRsCzwEHAOsA3wWey7abmVlJ5S1J/Ab4fkT8vrJB0m7AmcBn8jyBpGHAZcAg0jrZe0fEs3X2FWl51HMj4vCcMZqZWYPl7QI7jDSQrtr1wGqdONb5wDkRMQw4B7igrZ0k9cnuu6kTz21mZk2QN0k8C3y9ZtuupCqoDkkaDKwHjMo2jQLWk7RUG7sfBdwKPJMzNjMza5K81U2HALdKOoi0fOlKwOrATjkfvzwwPiJmAkTETEmvZNsnVXaStC7wBWAL5nFp1DFjxnT6MSNGjJiXQ82X0aNHd9mxfH6N11XnV+ZzA59fMzT6/PJ2gX1Q0qrAjsCywC3AbRHxRqMCkdQX+C2wX5ZE5ul5hg8fTv/+/RsVVtMU8ebpSj6/1lXmcwOfX61p06a1e3HdmeVL3wSu7NTRPzYWGCqpT5YA+pCSzdiqfZYBVgVuyxLEQKCXpMUi4jvzeFwzM5sPdZOEpDsiYrvs7/tJo6vnkmeq8Ih4TdJjwO6kRLM78O+ImFS1z8vAklXHHwks4t5NZmbFaa8kcXnV3xc14FgHAJdJOhZ4E9gbQNJtwLER8WgDjmFmZg1UN0lExNVVN5+OiH/U7iPps3kPFBFPAxu0sX2HOvuPzPvcZmbWHHm7wP65zvY7GhWImZl1P+02XEvqDfQiNSD3yv6uWBWY0cTYzMysYB31bppBarDuxdwJYRZwQjOCMjOz7qGjJLEyKUHcC1T3YpoNTIqI95sVmJmZFa/dJBERL2VjGl4AJnqNazOznqXDhutsKo2V8+xrZmblknfE9XHAeZJ+TlrG9KOBdRExqxmBmZlZ8fImicpgum9UbetFShZ9GhqRmZl1G3mTxMpNjcLMzLqlvLPAvgQfjZsYArzqaiYzs/LLu8b1YpIuBz4AxgPvS7pM0uJNjc7MzAqVt8fSmcDCwHBgQWBtYKFsu5mZlVTeNontgFUi4r3s9jOS9iPn8qVmZtaa8pYkPgBq16NeEvDgOjOzEutMF9g/SzqdtMb1isCPSMuNmplZSeVNEicArwB7kJYdfQU4Ffhdk+IyM7NuIG8X2NmkhOCkYGbWg+QtSSDpm6S1qSslid8Dv8sSiJmZlVCuJCHpVGAX4Dd83CZxOCDgiKZF1yCzZs1i3LhxTJ06te4+h39luS6L56mnnupwn4UXXpjllluO3r09r6KZFSdvSWJfYL2IGFfZIOlW4F+0QJJ4/fXX6dWrF5LqfukuMPatLotn9eUHtnv/rFmzGD9+PK+//jqDBw/uoqjMzOaW9zL13eyndts7jQ2nOd566y2GDBnSMlflvXv3ZsiQIbz99ttFh2JmPVzeksRvgBslnUyaKnx54MfAryWtUtkpIp5vfIjzb+bMmfTt27foMDqlb9++zJjhJcTNrFh5k8QZ2e8tarZvxcdTc3TracN79epVdAid0mrxmlk55e0C2xr1NGZm1lC5u8ACSFoBGAqMi4ixzQnJzMy6i7xdYJchjYvYEJgMDJL0MPD1iHilifGZmVmB8lYjnQc8DnwiIpYBPgH8Gzi/WYGZmVnx8iaJjYHDImIqQPb7CGCjZgXWnd1wzRWcOPLIObZdcPavuOCcXxUUkZlZc+RNEm8Ca9ZsE9B1I9C6kS223p7Rjz7MlClp6MjMmTO4754/s9U2OxQcmZlZY+VtuD4VuFvSxXw8Lcd+wM+aFVh3tsSgJRm+9qf5+71/Ybsdv8ToRx5mscUWZ7VhaxQdmplZQ+UqSUTEhcBupIWGds5+7xERPXY9iS233YF7/nIHAH+7+3a22Hr7giMyM2u8DksSkvoAzwBrRsRfmx9Sa9jw85tx7hmn8uILz/HIww+w3/4HFh2SmVnDdZgkImKmpJnAAOZjuVJJw4DLgEGkbrR7R8SzNfv8DPg6MBP4EDg6Iu6c12M2U79+/fn8JlvyyxOPZdgn12TwkKWLDsnMrOHyNlz/BrhW0maSVpW0SuWnE8c6HzgnIoYB5wAXtLHPP4HPRMQ6wDeBayQt2IljdKmttt2BF1/4P1c1mVlp5W24Pjv7vU3N9lzzNUkaDKxX9fhRwNmSloqISZX9akoNTwC9SCWPcXRDSw1emv79+/P5TWqntDIzK4eumrtpeWB8RMzMnm+mpFey7ZPqPGZv4LnqNSzyGDNmzFzbFlhggXYXHFp44YU7cwggrflw0w1Xs8nm27DQwot0+vHtxVMxffp0Ro8e3ennrjVixIj5fo7OakTceZX5/Mp8buDza4ZGn1+7SULSQsAxwHDSAkMnRcQ8t0vkJWkz4H+Zu+TSoeHDh9O/f/85tj311FPzlAjq+eD999lr1+0ZPGRpjjvpjI4f0IY88fTr14911113np6/aEV8OLpSmc+vzOcGPr9a06ZNa/PiuqKjksQ5wPrA7cBXSVU/89KNZywwVFKfrBTRh7RW9lyTBEraELgS2CUiYh6O1XQDFlyQ62+9p+gwzMyarqNqpO2AbSPiCGB7YKd5OUhEvAY8Buyebdod+Hd1ewSApM8A1wBfjYh/zcuxzMyscTpKEgtHxASAbGrwxefjWAcAB0p6hlQaOQBA0m2S1s/2ORdYELhA0mPZz9rzcUwzM5sPHVU3LSBpC1Ivo7Zuk3eAXUQ8DWzQxvYdqv7+TJ7nMjOzrtFRkngN+F3V7ck1t2cDnRkrYWZmLaTdJBERK3VRHF1q+ocz6dd3zuEdqy8/cL6f970PZjB+0pRc+77wwgscddRRvPXWWwwcOJBTTjmFlVZaab5jMDNrpE4tX1oW/fr2YefDbm74897yq11y7/vzn/+cPfbYg1122YWbb76ZY489lssvv7zhMZmZzY/5HSRn82Dy5Mk8+eST7LRT6iy200478eSTT/LGG28UHJmZ2ZycJAowYcIEhgwZQp8+qcqrT58+DB48mAkTJhQcmZnZnJwkzMysLieJAiyzzDK8+uqrzJw5E4CZM2fy2muvscwyyxQcmZnZnJwkCjBo0CDWWGMNbr31VgBuvfVW1lhjDZZYYomCIzMzm1OP7N00/cOZneqJlNd7H8zIve/IkSM56qijOPfcc1lsscU45ZRTGh6Pmdn86pFJonaMBMCzY9/q0hhWXXVVrrvuui49pplZZ7m6yQXkCssAABqaSURBVMzM6nKSMDOzupwkzMysLicJMzOry0nCzMzqcpIwM7O6emQX2FkzptN7gX5zbGvEVOEfTpvGi6+9P9/PY2bWXfTIJNF7gX48f8JXGv68q/z0BqDjJHHKKadw5513Mn78eG655RaGDRvW8FjMzBrB1U0F2GqrrbjqqqsYOnRo0aGYmbWrR5Ykirb++usXHYKZWS4uSZiZWV1OEmZmVpeThJmZ1eUkYWZmdfXIhutZM6Zn3VUb68Np03Ltd/zxx3PXXXfx+uuvs99++zFw4ED+9Kc/NTweM7P51SOTRO1AOuja9SSOOeYYjjnmmC47npnZvHJ1k5mZ1eUkYWZmdfWYJDF79uyiQ+iUVovXzMqpRySJAQMGMHny5Jb54p09ezaTJ09mwIABRYdiZj1cj2i4Xm655Rg3bhyTJk2qu89rb77XZfHMmLJQh/sMGDCA5ZZbrguiMTOrr0ckib59+7Lyyiu3u88Rh93cRdHALb/apcuOZWY2P3pEdZOZmc2bLitJSBoGXAYMAiYDe0fEszX79AHOBLYDZgMnR8RFXRWjmZnNqStLEucD50TEMOAc4II29tkTWA1YHdgQGClppS6L0MzM5tAlJQlJg4H1gG2yTaOAsyUtFRHVrcm7ARdGxCxgkqSbgF2B03Icpg/A9OnT5ynGgQv3mafHzYtpOafvaCSfX+N09fmV+dzA59dI83J+Vd+ZbQbaqyu6hUoaAVweEWtVbXsS2Csi/lW17T/ANyPikez2EcByEXFQR8cYPXr0xsD9DQ/ezKxn2GTEiBF/r91Ypt5NjwCbABOAmQXHYmbWKvoAy5C+Q+fSVUliLDBUUp+ImJk1UC+bba/2MrAiHwe7AvBSngOMGDFiGjBXFjQzsw49V++OLmm4jojXgMeA3bNNuwP/rmmPALgO2F9Sb0lLAV8Cru+KGM3MbG5d2bvpAOBASc8AB2a3kXSbpPWzfa4AngeeBR4GfhERL3RhjGZmVqVLGq7NzKw1ecS1mZnV5SRhZmZ1OUmYmVldThJmZlaXk4SZmdXlJGFmZnWVaVoOM7OWJ6kfVd/NEdF1y2a2wUmikyR9A7g1It7Mbi8BbB8RVxUb2fyRNIm0hketXsDsiBjcxSE1haSFgKOBVSJiD0mfBD4ZETcVHNp8k7Q6cAkwNCJWlrQe8MWIGFlsZI2RzSZ9OrBCRGwqaR1go4g4v+DQGkLS/wBnkeZRguyzR53ZWbuKq5s67/BKggCIiDeAwwuMp1HWBz7Txk9le1mcR7o4Wje7PQ74eXHhNNR5wPHA29ntx0hT7ZfFhaT52QZmt58Gvl9cOA13GvA1oG9E9ImI3hFRaIIAJ4lGKfwfOb8i4qXKDzAeWCT7GZ9tK4t1IuIoYDpAREyhPJ+DxSPiDrISYbYuy7wtsNI9Dc1KDTMBImI6MKvYkBrqjYh4MPu/dRuubuq8iZK+HBE3Akj6CvBawTE1TDaP1g3ANFJxdwFJX6le96PFzbEqi6QBlCdJzJTUlyxJSBpKub5EZ1TfkDSQ9B4tiz9I+h5wDfBBZaPbJFrPwcDNkk7Nbs8AdikwnkY7g7Tw018AJG1Jqif9fKFRNc59ko4G+kvaHDgUuLnYkBrmXOAPwJKSRgJ7Az8tNKLGulHSBcCikvYlVTX9rtiQGuqE7Pc5pETfLdokPMHfPMjWw1B2MyKiNIscSXosIj7V0bZWlV1pHwF8kfQh/CNwUln+h5I2BnYmndstEVGq1Rol7UnV/y4iriw4pNJzkshJUv+ImJb1jplL0UXCRpH0IHB0RNyT3d6M9CW6UaGBNYikT0bE0x1ta3VZN8olImJi0bFYPpK+DdwdES8WHUs1Vzfl9xCwHjCFObuKdosiYQMdDFwvqdIm0Rf4SrEhNdTVpP9jR9tajqTfA98lNVY/Tqp2OjEifllsZI0haRipeqmUXXyBTwOHZwn+r8BfgL9GxKtFBuUkkVNErJf9LksjZ5si4hFJqzFnddqHRcbUCJKWBAYDAyStwccNnosDCxcWWGMpIt6W9FXSl8yhpMW7SpEkSG0uxwMnZ7cfIy1UNrKogBopIn4AIGk5YEfgRNISzoVegDpJdJKkxYApETFL0nBgOHBj1h2vLFYHNsv+nkHqj97q9gQOIa2tflvV9reBU9t8ROvpm/3eDLgtIt6TVKbeTYtHxB2SToLUxVdSaT53kkYAWwFbkwbU3UkqTRTKSaLz/gZsKmlR0j9xDLAdsG+RQTVKNqL8ZD7+Iv2JpCNbfUR5RJwBnCHp6Ig4seh4muRJSbcDawBHSVqw6IAarOxdfB8BKm2C9xUdTIWTROf1ioipknYHLoyIkZL+U3RQDXQ4MKLS4ClpaVIybOkkUVFJENkUDwOqtr9cWFCNsw/wBeDx7D06FDiq4JgaqexdfD8HbAn8LHt/PkBqyL6xyKCcJDpvgKT+wDbA2dm2UnSfrKjuERMREyW1t3tLkbQFcDkwhPR/6wdMJrVXtKysW/YDlbYzgIgYTxo9XwoRcbmk50ldfBcC9ilTF9+I+CfwT0lXADuREvx3cZtEy7kGmAg8CzyQXWl/0P5DWspzko4DLshu7w88X2A8jfZLUr3vNaQeTd8CVioyoEaIiJmSpkgaEBFlej8CHyXBR7Ik+Pei42kGSWeTShILkjoeHJP9LpSTRCdFxHGSzgTezhrOplCuLqIHAGcCT2S3/0y6mimNiHhGUt+ImA1cJOlR0gey1QVpRPn1pK7aaWPEucWF1BhlT4KZJ4BfR8RzRQdSzUli3swC1s/m/akoRbE+Il4Dvl50HE1U6c47XtLOwIvAEsWF01ALAP8lNVxXlGm0bGmTIEBE/FbSIEk7ZpsejojJhQaFk0SnSdqNVGXxCVJiWI00cKnlB2NVSNoKWJU5Fz4pxQeR1MPpE6SSwyjSOIlDig2pMSJiv6JjaLJSJ0FJXwCuJI3/AFhH0l4R8ecCw3KSmAdHAyOAOyPi05K2Ab5acEwNI+lS0hoS/+LjBvlSfBAl9SZVE75J6m64WsEhNZSkNtdWKEuC7wFJ8ARg04h4CiAb9HkFqcq3ME4SnTcjIl6TtABARPxZ0ilFB9VAGwFrlWGUda2sDel45hxMVybVi0MNALYgjbguRZKokwTfBv4ZEc92dTxN0LeSIAAi4qlsXEihnCQ6b5qkXsCzkg4k1WkvUmxIDTW26ACa7DFJn826G5ZK7ZW2pGVI006XxXak0eSVUchbAv8ATpI0MiJafdrwSZL2jYhLASTtA0wqNiQniXlxDLAYcCRpucjFKcESilVXac8Af5F0E3MufFKKq1FSVeEDkp5lzsbPzxYXUnNExIRsUryymA2sXRn4KGl5UhL8LHA3rb+2xHeBqySdTzrXx4C9ig3JSSK3rJ/2l4E3s0nU9iF9iT5Oqt9uddVVFc8Ba1fdLkWbROagogNolprqmN6k/2lpVk0EVqkeGR8RYyWtnA34nNHeA1tB1vX1c5IWyW5P6eAhXcLrSeSUrYg1nFTX+zxpwMutpOJvr4goc7fR0pG0VEQUXpRvJEmXVN2cAfwfaeqYNwoKqaEk3U0aXFY5z31Jk+FtCzwaEZ8uKLSGkbQqc/csLLQNzSWJ/DYB1iJNB/AKsFRETJf0Wz4eeNbyJB1J1ReLpEGk5UxPKzayxpC0AXAt6Up7eaU1vb8TEd8pNrL51wN6/+xNGuh5GKl0ew9pvqq+2X0tLZvd9tvAU8zZs9BJokVMy0boTpX0fGVq8LJNVwzsHhEf9daKiMmS9gBKkSSA04HtySYsjIhHJV1WbEiNka2a+DPS1TXAXcAJZVk1MSJeoX538zJMsrkrsGpEvFN0INWcJPLrX7VYTfXfUDWbaAn0amNbmd4n/SLiyZpJC8uS5M8i/a8qgwO/TZqE8puFRdRAWRL8CaltYk9JnwQ+GRE3FRxao0zobgkCyvXhb7aFmLPYV/13mRp2npV0KPBrUsL4EaluuyymZQ2DlTUJ1qQ8EzR+JiLWqdzI1it/vMB4Gu08YALwqez2ONKo+bIkiYckjQKuY86eha5uagURsVLRMXSRg0hTA5xI+iJ9EPhGoRE11gmkaphls9Hl29ENuhk2SC9JC0fE1Oz2QrRdMmxV60TEPtn0FUTElGwUfVlUehgeWLXNbRLWvWT1vltKWji7PbWDh7SUiLhdUpAW5+kFHB8RZSkpXUm6Gv19dns30toZZTGt+kY2wWZpkkREbFG7LVu7plCleYGtMSRtKmmRLDl8XdL5klYuOq5GkXRERDwfEedFxLkR8X+Sjig6rkbIOhwcQZrVdgngyLL0SsvcJ+loUpvg5qReajcXG1LjSDqt5vYCwPUFhfMRJwmrdTapB9dapK6GLwMXFxtSQ7U1nqU0Y1wi4o6IODz7uaPoeBrsp6TS37vAqcA/gZFFBtRgy0v6AXw0GeUo4OliQ3J1k81tRkTMlrQ9cF5EnCVp16KDml/ZbL3bktoiTq26a3FKUm8v6e/Aztkst0haArgpIjYtNrLGyCadPCH7KaN9gDskTQS+CLwaET8uOCYnCZvLAtmAsy+Tli6FgtfYbZDppLmaZgPV7SwTgJMKiajxFqkkCICIeEPSokUG1Ag1SX0uEVGW6sJpkr4M3E9acOiHRccEThI2t5+R1rf+S0T8N5sgruUbdiPiXuBeSTdExJii42mS3pIWqgyey7r6Fj7VdANUkvqqpGlwbsxu/w9wbyERNZCkSczZjX5BYBlJXwSIiMGFBJbx3E3WI0jaNSKuK/PCPJJ+AuxEGk8A8D3glog4ubioGkfSX4FdK0t6ZlPGXBcRWxYb2fyRtGJ790fES10VS1tckrAOSdopIm4tOo75NJw0SOkzbdxXiiuliDhJ0iuk+myACyKiTF1gl65e8zmbMmbpIgNqhKKTQEdckrAOSbowIvbveE+z5pF0HWklukpvu/2AJSKiFMsHS9qI1GtrFdIFfC9gdtHVTS5JWIfKkCAk7dDe/UVPfdAskr4TEb8tOo4G+RZwLKmbNqRpww8vLpyGuxj4X9KSszM72LfLOEnYXLKJ1JZjzjntnywuooZoryth4VMfNNHQogNolGzyuzIlhVrvR8TVRQdRy9VNNodsMM/JwBvArGzz7IhYpbiozNpWkvYyACQdDzwQEbcXHUs1lySs1mHA8O7emNZZ2WyvdbVySanM51ZN0uci4uGazbuQVogsg+8CR0t6lzRPldskrFuaWLYEkflTO/fNJjUWtqoyn1u18yXNBM4Bro6ID8rQXlZl/aIDaIurmwyY42p0N9Jgnt8z55z2pbgatdYmaRPg+8DmpFlvz4uI5wsNqkHaGFQHFD+YziUJq6i9Gq2er6nlr0Yl9c+mPViorfvLssQngKTBVK2WGBEvFxhOQ0XE/cD9kj4F/BH4kaTbgR9HROGT4c2n6pLEAGBP4MOCYvmIk4QBEBGlmQ68joeA9fh4/qbqSf1mU4L5qSRtCVwGDCF1oewHTAYKvRJtpKwk8QNgA+Ci7GdLUsIYVmBo862Nat5jJT1M6hZbGCcJm0OdRtC3I2J8lwfTQBGxXva7zNPjnwZsBVxDSojfAlYqMqBGkvQf0jThZwF7RcSM7K4rJZVldcGPSFqFbpDgnSSs1m3A8qSRrZCm0n5N0gfA7m30LrFuJCKekdQ3ImYDF0l6FDim6LgaZL+IeLStOyJiu64OptFq2iR6kyZnPLi4iBInCat1E3BPRNwEIGkXUnH+D8AZpGK+dU+V+uvxknYGXiStUFcWvwE2rt4g6e8RsXGd/VtNdZvEDFJPw8JHXjtJWK3NI+KQyo2IuFnScRFxsKQFiwzMOnSGpE+QSg6jSKXAHxUbUkPN0ekgW72tNEmwu3Y9d5KwWr0lbRQRDwJI2pCPl7mdVf9hVrSIGJX9+QiwWpGxNJKkH5PW7l5c0mtVdy0EXFVMVD2Hk4TV+gFwjaRKl9CFgD2yBWx+XVxYloekrUiL81TPu9Xqa2X8ljTN+9mk92fFO9Ur8VlzeDCdzUVSP0DZzYiI6UXGY/lIugwYAfyLj2cRnR0R3ywuKmt1ThIG9KzBZmUl6RlgrYgofABWI0m6IiK+IekR2h6R/NkCwuoxXN1kFW0NNqv+3fKDzXqAsUUH0CS/yX6XeZrwbsslCbOSkHQesBapG3P1vFut3iYxl6xKdImImFh0LGXnkoS1KfsQVjd+urqp+xsAPAesXbWtNFeBkn5Pmk57OvA4sKSkEyPil8VGVm5OEjYHSV8GzgSWxdVNLSUi9is6hiZTRLwt6aukpUsPJS316STRRE4SVutU4GvAwxHhcREtRFIv4DvA1tmmu4CLsik6yqBv9nsz4LaIeE+S36NN5iRhtd6oDKSzlnMq8Gngkuz2PsDqpIFoZfBkNi34GsBRngGgazhJGABVXV//IOl7pJlEqxs/3SbR/X0BWK8yO6qka4HRlCdJ7EM6x8cjYqqkocBRBcdUek4SVlG7zsI5uE2i1VT+VxW162a0LEl9gAcqU74DZNPXt/QU9q3AXWDNSkLSL4F1gEuzTfsAT0TEjwsLqoEk3QdsGxEfdLizNYxLEmblcQSpi+iXs9t/IM17VBYB3CfpelLJN20s4TiQ7sRJwqwkst5o52U/ZbQA8F9Sw3WFq0KazNVNZi1O0sERcYak02h7bqOyNFxbAVySMGt9lTr6Ke3u1eKyHng/AVaJiD0lfRL4ZGUVRWsOlyTMrCVkU6FPAHaOiLWyNU7uj4hPFxxaqfXueBczawWSDpW0ePb3FZKelrRt0XE10DoRcRRp7iYiYgr+Dms6v8Bm5bFvNrfRFsBg4JvAiQXH1EjTqm9IGoC/w5rOL7BZeVRWo9sCuCqbXqVMn/H7JB0N9Je0OXAtcHOxIZVfmd5AZj3d+5KOBHYH7som/OtXcEyN9FPSCPJ3SfNU/RMYWWRAPYF7N5mVx77A94EjI2KipFWBq4oNqaEWiIgTgBOKDqQnce8msxIq48ptkiaRqpcuiYgHio6np3CSMCuJtlZuA0qzcpukJYA9SCWmxUhzVF0eEeMKDKv03CZhVh6KiLeBHUkrty0H7F1sSI0TEW9ExNkRsT5pfqrVgRcKDqv03CZhVh6lX7lNUm9gB1JpYlM+nvHWmsTVTWYlkS0ytChpAry1gFnAQxHxqUIDaxBJpwO7kSb5uwy4MSLeLzaq8nNJwqw8yr5y22TgcxExtuhAehKXJMxKRNIwYI2IuFnSokDfiHij6LisdbkkYVYSkvYhzZLaj9RVdFnSMrRbFxnX/JL0l4jYKusCW31V2wuYHRGDCwqtR3CSMCuPQ4D1gfsBIiIkLV1sSA2xV/Z7/UKj6KGcJMzKY3pETJFUvW1GUcE0SkRMyH6/VHQsPZGThFl5TM7aJGYDSNoLKM1AM0kbkeZsWoX03eXqpi7gJGFWHocAVwOS9CLwHrBzkQE12MXA/wIP8/GMt9ZkThJmLU7SZhFxL/ASsAEwjHSVHRFRpi/T9yPi6qKD6GmcJMxa3+nACNLAufWApwqOp1luk7R9RNxedCA9icdJmLU4Sf8hTU9xCHBS7f0RcW5Xx9QMWRfYQaT1JKbhNoku4ZKEWev7DvANYCHgMzX3lekq0F1gC+AkYdbiIuIh4CFJz5dlWvA6HqXtpOeSRBM5SZiVRET8UtKapDWuAf4SEU8XGVODVZckBgB7Ah8WFEuP4TYJs5KQ9A3gZOC2bNP2pKVMy7SE6RwkPRwRnys6jjJzScKsPA4HRlSWLM2m5LiTcq1z/RFJq+CqpqZzkjArkeo1rSNiYs0UHS2tZoK/3qRFlg4uLqKewUnCrDyek3QccEF2e3/g+QLjabTqNokZwMSSDRbslpwkzMrjAOBM4AnSFffdwHcLjaiBPMFfMZwkzEpAUh/gBxHx9aJjsXLpXXQAZjb/smqX7YuOw8rHXWDNSkLSz4GpwOXAlMr2iHivsKCs5TlJmJWEpFltbJ4dEX26PBgrDScJMzOryw3XZiUiaUmgMgL5oYiYXGQ81vrccG1WEpK+DDwNHJT9PCXpS8VGZa3OJQmz8jgB2CgingGQtDrwR+CmQqOyluaShFl5fFBJEAAR8SzwfoHxWAm44dqsJCSNJE2dfTFp1bb9SLUFpwG93BXW5oWThFlJ1OkCW+GusDZPnCTMzKwut0mYmVldThJmZlaXk4RZOyStJGm2pC7tLi5pT0l3deUxzdricRLWI0l6ERgCVC9ac2lE/LCAWFYCXgD6RsQMgGxd6lIuO2qtxUnCerKdI+LuooMw686cJMyqZIv3nALsC7wD/Krm/heBb1eSSzY2YbWI2Cu7vTFwKrAm8C7ws4i4VNKOwPHAqsDbwMURMTJ72vuy329la1JvAyg7zsbZ824EnAEMA54BDo6IB7P77gHuB7YE1gEeAvaIiNcb8qJYj+Y2CbM57Q/sBHyatKbyV/M+UNKKwO3AWcBSwKeAx7K7pwJ7AwOBHYHvVc2rtGn2e2BELBIRD9U87xLAn0hLkw4CTgf+JGlQ1W57kAbPDQb6AYfnjdusPS5JWE92k6QZVbd/DHwN+E1EjAWQdBKwec7n2wO4OyJGZbcnZz9ExD1V+z0haRSwGfnmVdoReDYirshuj5J0ELAzcGm27ZKqOZuuBb6YM2azdjlJWE/2pdo2CUmHAWOrNr3UiedbHniurTskbQCcDAwnXen3B67L+bzLthHHS8DQqtsTq/5+D1gk53ObtcvVTWZzmkD6sq9Yoeb+qcBCVbeXrvp7LKnNoS1Xk2ZkXT4iFgfOJ82vBNDRtAevACvWbFsBGN/B48zmm0sSZnO6FjhI0q2khHBUzf2PAV+XdDuwLqnN4o7svquAoyV9DbgRWJyUFB4DFgXeiIgPJH2WVDVVGQcxCZgFrEJqlK51G3CWpD2y+L5Cahi/tQHna9YulySsJ7tF0pSqnz8AFwJ3Ao8D/yJ92Vf7Gam08CZwHKmEAEBEvAzsABwGvEFKKOtmd38f+IWkd4FjSV/2lce9R1oL4gFJb0mqrCxXuX8yqTH9MFIbxxHATu69ZF3BE/yZmVldLkmYmVldThJmZlaXk4SZmdXlJGFmZnU5SZiZWV1OEmZmVpeThJmZ1eUkYWZmdf0/KzmgoqyISpcAAAAASUVORK5CYII=", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "table=pd.crosstab(data.education,data.y)\n", "table.div(table.sum(1).astype(float), axis=0).plot(kind='bar', stacked=True)\n", "plt.title('Stacked Bar Chart of Education vs Purchase')\n", "plt.xlabel('Education')\n", "plt.ylabel('Proportion of Customers')\n", "plt.savefig('edu_vs_pur_stack')" ] }, { "cell_type": "markdown", "metadata": { "id": "QGKuaESvQJFh" }, "source": [ "La educación parece ser un buen predictor de la variable de resultado." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 315 }, "id": "5vuTl9clQIWO", "outputId": "345fbed4-28f7-4913-d9d8-3221e0478f08" }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "pd.crosstab(data.day_of_week,data.y).plot(kind='bar')\n", "plt.title('Purchase Frequency for Day of Week')\n", "plt.xlabel('Day of Week')\n", "plt.ylabel('Frequency of Purchase')\n", "plt.savefig('pur_dayofweek_bar')" ] }, { "cell_type": "markdown", "metadata": { "id": "_DhYmQxsQUD9" }, "source": [ "El día de la semana puede no ser un buen indicador del resultado." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 315 }, "id": "4n_OHUrOQRTM", "outputId": "620128ac-368b-45d0-da53-fc3a4f96ac5b" }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZoAAAEqCAYAAAAszJYWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deXxdVbn/8U/SNmkpSi2lhbYMMvShl9RyCQpeUVBEFKllULxlqMAFFZmuogiKDN6CVJlkEgTBMsgVRAaLID+vMgkohuFlBZ+ChYJA6WQLLTZpk/z+WOu0u6FJdk7O3mf39Pt+vfJKzl777P2ck+Q8ew17rbrOzk5ERESyUl/tAEREpLYp0YiISKaUaEREJFNKNCIikiklGhERyZQSjYiIZGpgtQMQScvMtgFeBAa5+6oqh7NBM7MhwK3AR4D73f1zVQ6pLGZ2JHCMu+9R7VhqmRKNVISZvQSMAtqB5cC9wAnuvqyKYVVc/GD6CfCvxOafuvsJ1Ymoaj5L+H1vWomkb2Z7Ab8H7nT3AxPbJwJPAw+6+179PMc26EKlKtR0JpU0yd03BnYBdgXO6OsBzGx9uPh5zN03Tny9I8msJ6+jP7YGZpfzgd3De7MA+KCZbZrY9gVgdhnxSYHU+j+DVIG7v2pm9wJN67qKNLMHgJvc/dpYQzgW+BMwFfiRmZ0LTCNcNQ8D/gLskzjFYWb2P8BGwMXufm487geAHwLjCTWO24GvuXubmdUBFwGHAYOBucAUd59lZo3AucAhQCNwB/BVd0/WWnpkZmcDTcAK4DPA18zstnjO/YAO4HrgLHdvN7MBwHTgSOBN4ELg8tL7FGuIx7j7bxPH397dD4+Pd4/H/rf4Wk529wcS7+/DwMeA9wGPAYe6+8JYvgfw/fjct4DvAH8FZgKj3b097ndQjHdil9d6DnA6UGdmBwAnx9f2LcLvcghwH3Ciuy9N/A0cA5wFvERocuuqLcbwn8AV8T36PHB1fC2l8/8H4fc8jpCETnb3R1O89ofiIZaYGST+pszsAuC/gCXAV9z93nXEJ2VSjUYqzsy2JHy4PpXyKbsBcwhNMecCFwDNwH8Aw4FTCR/UJXsABuwNnGlm4+P2duCrwAjgg7H8K7HsE4QPt3HAJoSksiiWnR+37wxsD4wBzkz7ehMmA78gJMebgZ8Cq+Ix/z3GcEzc91hg/7h9V0JSTcXMxgD3EJLxcODrwO1mtllit0OBo4CRQEPcBzPbmtCseRmwGeE1P+3uTxDej08kjnEEcEPX87v7WcB5wM9jje4nhIR5JPBRYFtgY0LiTNqTcBGwbw8v7wbCBQdxv1nAa4nXPjy+9kuBTQnJ9p4utaB1vnbWJLdhMe7H4uPdACf83Xwf+Em8MJEKUY1GKulOM1sFLCV8GJxHSB69ec3dLwMwsw7gaGB3d381lpeuVkv7nxNrG8+Y2TPAROA5d29JHPMlM7ua8OF2CbASeBewI/And38uHrMO+CLwPndfHLedB/yMcNW+Lrub2ZLE40/G74+5+53xGO8mJNthMdblZnZxPNfVhER3ibu/Evf/HrBXivcK4HDg1+7+6/j4/5nZn+P5ZsRt17v77HjsWwm1LAgfwr9191vi40WsSbgz4rHvjR/o+7ImUffmMOAid58Tz3k6MMvMjkrsc7a7L+/pIO7+qJkNt/DLnkpIPEMSu3waeN7db4yPbzGzk4BJhMTe02vvzlx3vybuPwO4kvB3O6+X50lKSjRSSQeUmnpKEsmhJ68kfh5BaNr6ew/7Jz8A3iZcPWNm4whXuLsSmtUGAi0A7v47M7scuALY2sx+SbjSHRz3bUnEWgcM6OH8j3cdpWRmn+zyOrYGBgGvJ45bn9hndJf95/Zwvq62Bj5nZpMS2wYROtNL1vkeAVvS/Xt7E/CcmQ0lJMKH3f31lDGNZu3XMJfw/icvNF4hnRuBEwi1o6MJybG785TONSbxuLvX3p3V+7v72/H31dtzpA+UaCRrpSvYjQh9EQCbd9knOYX4QkI/x3bAM308148IzXVT3P0tM/tvEk1S7n4pcKmZjSQMzf0Goc/gX8BOiRpUuZKv4xWgFRjRTYf564QP/ZKtupQvJ7xnJcn37BXgRnc/towYXwE+sK6C2Lf2GHAQodnsR3047muEBFiyFaHZ8A1gbNyWdqr4G4EXgBsSH/zdnad0rvtSHFdT1VeJEo1kyt0XmNmrwOGxKesLhCTS3f4dZnYdcJGZHUH4oPoA8GSK072LkMyWmdmOwHGEkUyY2fsJNYonCR/iK4COeL5rgIvN7AR3nx/7QJrc/Tdlvmzc/XUzux+40My+AywD3guMdfcHCYnuJDObGeM5rcshngb+Mw6qmEhImKUP05uAJ8xsX+C3hNrM7sAL7v6PXkK7GfiWmR0C/JLQX7Wluz8dy2+IsWwdy9O6BfhmjHcBa/pwVqWs1a7m7i+a2Z6Efruufg1cZmaHEt7DgwmDGmamOPQCQl/ftmgkW640GEDycCyh9rAI2InY59KDrxNGmj0BLCaMzkrzt/p1QjPLW8A1wM8TZe+O2/5JaGpZBPwgln2TcAX9uJm9Sfjw7tun47pNJXRGPxvP+wtgi1h2DfAbQq3tSd75of4dQkL+J3AOoc8IgNivM5kwymsBoZbyDVK8R+7+MqEv5xTCe/s0IZGV3EFIMne4+9upXylcR6iJPEQYYbYCOLEPz+8a5yPu/to6ti8iDKI4hfA7PBXYvzSirpdjvk0YbPIHM1sSR+5JDuq08JlI9RXpZkIz+zvwpa79bSLlUo1GRFYzs4MJfRm/q3YsUjvURyMiwOqbHf8NOMLdO3rZXSQ1NZ2JiEim1HQmIiKZUtNZFy0tLY3A+wn3ObRXORwRkfXFAMKoyieam5tbkwVKNO/0fsKkfCIi0ncfBh5JblCieafXAcaNG0dDQ0NZB5g1axZNTU0VDWp9jaMIMRQljiLEUJQ4ihBDUeIoQgyViKOtrY3Zs2dD/AxNUqJ5p3aAhoYGGhsbyz5If55bSUWIowgxQDHiKEIMUIw4ihADFCOOIsQAFYvjHV0OGgwgIiKZUqIREZFMqelMRKRK3nzzTQYOHMhzzz1X7VBSxzF06FDGjh1LfX36eooSjYhIFbz55pu88cYbjB07luHDh1NXV91FPZcvX87QoUN73Kejo4NXX32VhQsXMnLkyNTHVtOZiEgVzJ8/nzFjxjB48OCqJ5m06uvrGTVqFEuXLu3b8zKKR0REerBy5UqGDBnS+44FM2jQIFat6tsE40o0IiJVsr7UZJLKiVmJRkQqrm1lz7M37Th+p5wikSLQYAARqbiGQQOYdMpd3Zb/6sLJOUYj1aYajYiIZEqJRkSkBlx77bWceOKJa22bNm0a06ZNq1JEayjRiIjUgM985jM8/PDDvPnmmwCsWrWKe+65hwMOOKDKkSnRiIjUhJEjR7Lrrrty3333AfDwww/znve8pxAzQyvRiIjUiAMPPJC7774bgLvvvpvJk4sx6EKJRkSkRnz84x/H3Zk9ezYPPPAAkyZNqnZIgBKNiEjNaGxsZN999+WUU05hwoQJjB49utohAUo0IiI15YADDmD27NmFaTYDJRoRkZoyevRoBg8ezL777lvtUFbLbWYAM7sAOBjYBpjg7rPMbFPgRmA7oA14HviSuy+Iz9kduBoYArwEHO7u8/tTJiJSqzo6Orj++uvZb7/92Hjjjasdzmp51mjuBD4CzE1s6wS+7+7m7hOAvwPnA5hZPXATcLy7jwMe6m+ZiEitevvtt2lububRRx/lpJNOqnY4a8mtRuPujwCYWXLbYuCBxG6PA8fFn5uBFaXnAVcRaidH96NMRKQmbbTRRjz11FPVDmOdCjOpZqyJHAfcHTdtRaL24+4LzazezIaXWxYTWyqzZs3q1+tpaWnp1/MrpQhxFCEGKEYcRYgBso+jubm56jGkVa04Bg4cyPLlywFWf6+2tHG0tbX16X0rTKIBLgOWAZdXOxCApqYmGhsby3puS0tLqn+0rBUhjiLEUJQ4ihBDkeIoQgzVfC+ee+45hg4dmmoJ5Tz0JY6GhgYmTpy41rbW1tZuL9ALMeosDhTYAfi8u3fEzS8DWyf2GQF0xFpJuWUiIpKzqicaMzuP0K9ygLu3JopagCFmtkd8/GXgtn6WiYhIzvIc3nwpcBCwOfBbM1sEHAKcDswGHo0DBV509wPdvcPMjgCuNrPBxGHKAOWWiYhI/vIcdXYSsK4xd90uQO3ujwITKlkmIlJUbSvbaRg0oGrHffHFFznttNNYsmQJw4YNY/r06WyzzTb9Pn+RBgOIiGzQelsCu1xpl84+66yzOPTQQ5k8eTJ33XUXZ555JjfccEO/z1/1PhoREam+xYsX8+yzz7L//vsDsP/++/Pss8+yeHH/x1Ep0YiICPPmzWPUqFEMGBCa2AYMGMDIkSN5/fXX+31sJRoREcmUEo2IiLD55pvzxhtv0N7eDkB7ezvz589niy226PexlWhERIThw4czfvx4Zs6cCcDMmTMZP348w4cP7/exNepMRKQg2la2px4h1tfjphnefPbZZ3Paaadx5ZVX8u53v5vp06dX5PxKNCIiBZHFPTR9Oe52223HbbdVfiIVNZ2JiEimlGhERCRTSjQiIpIpJRoREclUnxJNXKmy/4OqRURkg5Eq0ZjZMDP7GbACeCFu+4yZTcsyOBERWf+lrdFcBSwlrFzZFrc9Bnw+i6BERDZEHavaet8pw+NOnz6dj33sY5gZs2fPrtj5095Hszcw2t1XmlkngLsvMLORFYtERGQDVz+wgTnnHlzx42777dtT7bf33nszdepUDjvssIqeP22iWQqMAFZP42lmWyUfi4jI+m3XXXfN5Lhpm86uBW43s48C9Wb2QWAGoUlNRESkW2lrNNOBfwFXAIOA64CrgR9mFJeIiNSIVInG3TsJSUWJRURE+iRVoolNZi+5+4tmtjmhhtMBnO7u87IMUERE1m9p+2iuBNrjzxcRms86gB9nEZSIiORv2rRpfOQjH2HevHkcddRRfPrTn67IcdP20Yxx95fNbCCwL2vup3mtIlGIiAgdq9pSD0Xu63HrBzb0ut8ZZ5zBGWecUfHzp000b5rZKKAJeNbdl5lZA6Fm0yszuwA4GNgGmODus+L2cYTRa5sCi4Cp7v58VmUiIkWWJhkU6bipz59yv8uAJ4CbCSPPAD4E/C3l8+8EPgLM7bL9KuAKdx8Xj3t1xmUiIpKztKPOppvZHUC7u/89bn4VOCbl8x8BMLPV2+KsArsA+8RNtwCXm9lmQF2ly9x9QZpYRUSkslLP3uzusxNJpvT4L/0495bAq+7eHo/XTujz2TKjMhGRQuno6Kh2CH3W2dnZ5+ekHd78buBsYE/CVDR1pTJ336rPZ10PzJo1q1/Pb2lpqVAk/VOEOIoQAxQjjiLEANnH0dzcXPUY0qpWHPX19cyZM4cRI0awbNky6urqen9SxpYvX95jeWdnJ0uXLmXFihV9et/SDga4EhgLfBe4CTgc+AbQn+ERrwBjzGyAu7eb2QBgdNxel0FZnzQ1NdHY2FjWC2tpaUn1j5a1IsRRhBiKEkcRYihSHEWIoZrvRUdHBwsXLmTu3Lk0NFS3sx6gra0tVRyDBw+mqamJQYPWHgvW2tra7QV62kTzCWC8uy8ys3Z3v8vM/gz8Crg45THW4u7zzexpYAoheU0Bnir1pWRRJiJSFPX19YwcOZJXXnmFiRMnVjscWlpaMosjbaKpJ8zgDLDMzDYhzNy8fZonm9mlwEHA5sBvzWyRu+8EfBmYYWZnAv8EpiaelkWZiIjkLG2ieYbQP/N/wMOEprRlQKqVcdz9JOCkdWz/G7BbN8+peJmIiOQv7aizY4GX4s8nE2ZyHoZqCyIi0ou099HMSfw8n5T3z4iIiKRtOsPMPgHsDGyc3O7uZ1Y6KBERqR1p76O5HDgE+D3wdqYRiYhITUlbozkUmOjufb4fRURENmxpBwMsBJZkGYiIiNSmbms0ZrZt4uGFwM1m9j3gjeR+yYECIiIiXfXUdPYC0EliXjNg/y77dAIDKh2UiIjUjm4TjbunntlZRESkO6mSiZmNMbP3dNn2HjMbnU1YIiJSK9LWWu4kzN6cNBa4o7LhiIhIrUmbaMZ1XeQsPt6x8iGJiEgtSZtoFpjZWjM1x8eLKh+SiIjUkrQ3bF4H3G5m3wbmANsB/wNcm1VgIiJSG9ImmvOBlcAFwJbAy8BPgIsyiktERGpEr4kmLod8HfBFd/9B9iGJiEgt6bWPxt3bCUs5d2QfjoiI1Jq0gwEuBs4xs4YsgxERkdqTto/mRGBz4GtmtoAw9QwA7r5VFoGJiEhtSJtoDs80ChERqVlpl3J+MOtARESkNqVdYfO73ZVpKWcREelJ2qazLbs83hzYE811JiIivUjbdHZU121m9klgSiWCMLP9CTMN1MWvc9z9l2Y2DpgBbEqY7maquz8fn1NWmYiI5Ks/a87cDxzQ3wDMrA64ETjC3XcGjgBmmFk9cBVwhbuPA64Ark48tdwyERHJUdo+mm27bNoIOBR4pUJxdACbxJ+HAa8DI4BdgH3i9luAy81sM0Ktp89l7r6gQvGKiEhKaWs0L3T5ehz4MPCF/gbg7p3AIcBdZjaXsPbNVEK/0KtxZoLSDAWvxe3llomISM7S9tFktqyzmQ0ETgcmu/sfzOxDwK2EJrSqmTVrVr+e39LSUqFI+qcIcRQhBihGHEWIAbKPo7m5ueoxpFWEOIoQA2QXR4+JxsxGEaafaQKeBE5x90qvQbMzMNrd/wAQk81yYAUwxswGuHt7nNxzNKG5rq7MstSamppobGws6wW1tLSk+kfLWhHiKEIMRYmjCDEUKY4ixFCE96IIMVQijtbW1m4v0HurqVwJjCJ0rm8JXFJ2FN37BzDWzAzAzMbHcz4PPM2akW1TgKfcfYG7zy+nLIPYRUSkF701nX2YsIzzEjO7FXiq0gG4+zwzOw74hZmVZog+2t0Xm9mXCSPQzgT+Sei7KSm3TEREctRbohns7ksA3H2hmQ3NIgh3vxm4eR3b/wbs1s1zyioTEZF89ZZoBpnZUYR+D4BGMzs6uYO7X5dJZCIiUhN6SzR/ZO1mpz+x9miwTsLqmyIiIuvUY6Jx971yikNERGpUZvfHiIiIgBKNiIhkTIlGREQy1W2iMbMTEj9vn084IiJSa3qq0Zyb+PnJrAMREZHa1NOoszlmdiHwV8L9NEevayfdRyMiIj3pKdF8HjiVMFfYINY9m7LuoxERkR51m2jcfTZwDICZ/Z+7751bVCIiUjPSrkezd1w35j+AMYQZlx9z91VZBiciIuu/tEs5GzATGEJY12VLYIWZTXL35zKMT0RE1nNp76P5EfBjYEt3/6C7jyWsUXNlZpGJiEhNSJtodgYucvfOxLZL4nYREZFupU00rwF7dtn24bhdRESkW6n6aIBvAXeb2UxgLrA18Gng8KwCExGR2pCqRuPudwO7ALOAd8Xvze5+V4axiYhIDUhboyndVzMtw1hEyta2sp2GQQO6Ld9x/E45RiMiSakTjUiRNQwawKRTuq9g/+rCyTlGIyJJWiZAREQypUQjIiKZSpVozGxynIJGRESkT9Imj+8C15rZz4Eb3f2PlQzCzAYDFwMfB1YQ5lH7opmNA2YAmwKLgKnu/nx8TlllIiKSr7TDmycSksC/gNvNzM3sDDPbpkJxfJ+QYMa5+wTgO3H7VcAV7j4OuAK4OvGccstERCRHfRne/AzwjJmdCuwNXAicY2Z/IHyQ3+LuHX0NwMw2BqYCY0tT3Lj7G2Y2knDvzj5x11uAy81sM6CunDJ3X9DX+EREpH/6NBjAzLYDziRMsjk4/nwNcALwizJj2I7QvHWWmf3ZzB4wsz0IM0S/6u7tAPH7a3F7uWUiIpKztMsEHE9YYXMH4OfAEe7+eKL8dmB+mTEMALYFnnL3b5jZbsCvgM+VebyKmDVrVr+e39LSUqFI+qcIceQRQ3NzcyHiWB9igOzjWF9+H1CMOIoQA2QXR9qms08RmsrudvfWroXu/raZHVRmDC8DqwhNXLj7H81sIaE/aIyZDXD3djMbAIwmrIdTV2ZZak1NTTQ2Npb1glpaWlL9o2WtCHEUIYaSasdRlPeiKHEUIYYivBdFiKEScbS2tnZ7gZ626eyzwJ3JJGNmg8xs9Sexu99fTnDuvhD4PbFPJY4YGwnMBp4GpsRdpxBqPQvcfX45ZeXEJyIi/ZM20dwPdE11zcBvKhTHl4FvmdlfgP8lNM0tidtPNLPZwInxcfI55ZSJiEiO0jadvQ/oeu/Mn4CJlQjC3ecAe61j+9+A3bp5TlllIiKSr7Q1miXAqC7bRgHLKxuOiIjUmrQ1mtuBn5nZScAcwpDki4BbswpMRERqQ9oazbeB5wjNZW8BjwNOWHlTRESkW6lqNO6+AjjezE4ARgALS3fxS/VosS8RWR+knoLGzDYBDNg4PgbA3X+XSWTSKy32JSLrg7QzAxxJmJxyGfB2oqiTcFe/iIjIOqWt0ZwLfNbd780yGBERqT1pBwMMJNy0KSIi0idpE8104Awz09LPIiLSJ2mbzr4KbA6camaLkgXuvlXFoxIRkZqRNtEcnmkUIiJSs9LeR/Ng1oGIiEhtSju8uZGwmuYUYFN338TMPgGMc/fLswxQRETWb2k79y8GmoDDCPfOAPwVOC6LoEREpHakTTQHAoe6+2NAB4C7vwqMySowERGpDWkTTRtdmtnMbDNg0bp3F9nwtK1s77Fcc8/JhirtqLPbgBlm9lUAM9sCuISwGqaIoLnnRLqTtkbzLeBF4C/AMOB54DXgnIziEhGRGpF2eHMb4abNr8YmMy0TICIiqaQd3tx1huZ3JZYJmFPpoEREpHak7aN5gTCsuS6xrVSj6X7lLRER2eClbTpbqy/HzDYHzgIeziIoERGpHWXNxuzu84D/Br5X2XBERKTWpF7KeR0M2KhSgQCY2VnA2cAEd59lZrsDVwNDgJeAw919fty3rDIREclXqhqNmT1sZg8lvv4M/BG4qFKBmNkuwO7A3Pi4HrgJON7dxwEPAef3p0xERPKXtkZzbZfHy4Fn3P35SgQRJ+28gjBp5wNxczOwwt0fiY+vItROju5HmYiI5CztYIAZGcfxXeAmd3+pNGwa2IpYu4kxLDSzejMbXm6Zuy/O+HWIiEgXae+j+W6a/dz9zL4GYGYfBHYFTuvrc7M0a9asfj2/paWlQpF0r7m5uRBxFCGGIrwXRYghLb0XaxQhjiLEANnFkbbpbAfgYOAJQm1hK+ADwO3AirhPuTMF7AmMB16MtZmxwG+AS4GtSzuZ2Qigw90Xm9nL5ZT1JaimpiYaGxvLekEtLS2p/tHyUO049F4UL4ai/E6KEEMR3osixFCJOFpbW7u9QE+baOqAKe5+e2mDmR0EfM7djyo7MsDdzyfRWW9mLwH7A88CXzSzPWJ/y5cJk3sCtABDyigTEZGcpb2P5lPAnV223Q3sV9lw1nD3DuAI4Edm9jyh5nNaf8pERCR/fZmC5nhCc1bJccDfKx2Qu2+T+PlRYEI3+5VVJiIi+UqbaI4B7jCzU4HSypqrgIOyCkxEpD/aVrbTMKjnqRi1GF0+0g5vfsrMdiDcUDkaeB14zN1XZhmciEi5eluIDrQYXV7KnevsIaDBzIZWOB4REakxaaegmQDMBq4BfhI37wlcl1FcIiJSI9LWaH4EnOnuOwKl5rIHgT0yiUpERGpG2kSzE2GiSog3Zrr7csLsyCIiIt1Km2heIkxWuZqZfYAw7FlERKRbaYc3fwe4x8yuIgwCOJ1wx/2xmUUmIiI1IVWNxt1nAp8ENiP0zWwNHOTu92cYm4iI1IBeazRmNoAw4uzf3P0r2YckIiK1pNcajbu3A+3A4OzDERGRWpO2j+YS4FYzOw/4B4klAdx9ThaBiYhIbUibaC6P3/fpsr0T6HkyIRER2aD1mGjMbHN3n+fuZU1VIyIi0lsCmZ18YGa/zDAWERGpQb0lmrouj/fKKA4REalRvSWazl7KRUREetTbYICBZvZR1tRsuj7G3X+XVXAiIrL+6y3RzGftpQAWdXncCWxb6aBERKR29Jho3H2bnOIQEZEapWHLIiKSKSUaERHJlBKNiIhkKu0UNJkxs02BG4HtgDbgeeBL7r7AzHYHrias5PkScLi7z4/PK6tMRETyVYQaTSfwfXc3d58A/B0438zqCctHH+/u44CHgPMByi0TEZH8VT3RuPtid38gselxwsJqzcAKd38kbr8KOCT+XG6ZiIjkrOqJJinWRo4D7ga2AuaWytx9IVBvZsP7UVYRbSvbeyzfcfxOlTqViMh6r+p9NF1cBiwjLEtwYDUDmTVrVrdlzc3NTDrlrm7Lf3XhZFpaWrII6x1x9CaPOIoQQxHeiyLEkNaG8F6kiSGPONaXGCC7OAqTaMzsAmAHYJK7d5jZy4QmtFL5CKDD3ReXW9aXeJqammhsbCz79aT9I89ateNoaWmpegwlRYijCDEU5XdShBig+nEU5ffR3zhaW1u7vUAvRNNZXLmzGTjA3Vvj5hZgiJntER9/Gbitn2UiIpKzqtdozGwn4HTC2jePmhnAi+5+oJkdAVxtZoOJw5QBYo2nz2UiIpK/qicad/8r71z3plT2KDChkmUiIpKvQjSdiYhI7VKiERGRTCnRiIhIppRoREQkU0o0IiKSKSUaERHJlBKNiIhkSolGREQypUQjIiKZUqIREZFMKdGIiEimlGhERCRTSjQiIpIpJRoRyV3HqraK7CPrh6ovEyAiG576gQ3MOffgHvfZ9tu35xSNZE01GhERyZQSjYiIZEqJpob11sa9IbWB670QqR710dSw3trBN6Q28CK8Fx2r2qgf2NDvfUTWN0o0GdAHiqyLOsBlQ6VEkwF9oIiIrKE+GhHZYOXRd9e2sr3H8h3H79TvcxSdajQissHKo++uYdAAJp1yV7fld03/VK/H6G9Te9vKdhoGDehxnywTnhKNSA1J84Gy8/uaeixX/2G+8mhq7y3ZQe8Jrz9/FzWbaMxsHDAD2BRYBEx19+erG5VIttJ8oPzqwslVH4EnxZNl7a6W+2iuAq5w93HAFc1SbwYAAAvNSURBVMDVVY5HRGSDVJM1GjMbCewC7BM33QJcbmabufuCXp4+AKCtredOwGFDu2+eaG1tpX3IJj0+v7W1tZcw0ulPHJWIYeWqdgYN7D6G8TuO6/U8He0rqR8wqN+xVPu96G8MlYqjpxjSxKH3orIx9BZHrbwXic/Md5yorrOzs9cA1zdm1gzc4O47JbY9Cxzu7k/29NyWlpY9gIczDlFEpFZ9uLm5+ZHkhpqs0fTTE8CHgdeBnscliohIyQBgC8Jn6FpqtUYzEpgNbOru7WY2gDAgYIcUTWciIlJBNTkYwN3nA08DU+KmKcBTSjIiIvmryRoNgJntSBje/B7gn4ThzV7dqERENjw1m2hERKQYarLpTEREikOJRkREMqVEIyIimVKiERGRTCnRiIhIpjQzgNQUM9uop3J3fzuvWErMrIHE/1qVYtjE3ZfmfV4RUI2mIsys3sz2q3YcRWFmTWY2NPF4YzPLaxnBZcBb8fuyLo/fyikGAMzsQDP7B/CvasUQ46gDHsv7vOuI4zQz27QAcfzOzA4zsyFVjOGRNNtqhWo0FeDuHWY2Dfh1tWMxs9uArjdHLSV80PzU3TtyCGMGsHvicVvctmvWJ3b3Il08/QA4BHg8p/d9ndy908xeMbP3uPs/qxUHYR6sZ83sN4QlPP5YpTguBI4CLjazu4Dr3f3RnGNYq+ZtZvXA8DwDMLMtgEuBjxE+M34HnOzur1f6XEo0lfO0mX3A3f9U5TjmET7Qb4mPP09INIcAOwMn5RDDAHdfWXrg7m1mluvfWndNaDk3Wy2uwgdYd5YCT5nZrwk1KwDc/dS8AnD3k83sW8ARwDVm1kZYK+oWd1+RYxz3APfE2tUU4Idm9i533zHrc5vZN4BTgU3MbH6iaCPg5qzP38WNwEPAyfHx0XHbxyt9IiWaymkG/mBmz7P2P/IHco5jIrCXu7cCmNmPgd8SrlqezimGlWa2rbvPiTFsR/4zYS8jXKXVsXYNr+dFOSrrDjM7Dvg5sPqDtBp9NMBf41dVufvy+Dc5H7gEOB34HzP7mrvfmnM4pVpmHfl1I/wYuA24HDg+sf3NKtQ2t3D37yYeTzOzKd3u3Q9KNJWTR00hjVGEpqqSlcCIWKuozCpOvTuHkHTviY/3A47N6dzA2k1oZjYYOBQYkWcMwLnx+xWsnfTyTHYAuPs5eZ+zKzMbBXwJOBL4M2F9qIfM7L3Ag0AuicbMJsUY9gDuIjQX/SGPc8cBGUvN7Chgqbu3xZgaUi7MWEkvmNn27v5CjGE7wqz3FadEUyHu/iCEju/4eFnPz8jMg4RmgRvj48OAh2NcuSQad59pZnuypgp+fumPuRpis8x1ZvYE8P0cz1uk/iLM7BOE5tPBpW1drmiz9hTwU+Aj7v6PRAwvmtn1OcZxUozjcHf/V47nTZoJfJQ1F4WDgF+xdt9m1oYAzyQGIXyIcIF4K4C7H1KpEynRVEicLfpGYALQaWZ/IcwY/becQzke+DLw2fj4N8BVsc8kzz/ieYRO8B5XNM1Klz6aeuD9wLAqxrBalYY3n094D3YiXMVPJjSp5um9pSbdrtz9rLyCcPd9AMxsqJkNdffleZ07oTH5dxCbFAf39IQM3Mza/UI/y+pESjSV81PgMkKygVCT+Cn5frgTE8pl8asq4lDvqwn9MtuY2a7AWe4+Kccwkn007cDz5N+8WYR+opJPA/8OtLj7l8zsu8A1eQbg7q0FqFVhZtsSPlR3JlwUPkWo3czJOY7VTWVxscZca8DuPiOvcynRVM7G7n5D4vFNZvbNvIPoZnhzRavBKZxDuHq+N577z7H9N3Nm1hivmt/VpaiTdbwvWSpIP1HJCndfZWadZjbI3V81s7F5BlCQWhWEi6AfA6XmuiPjtn1yjOFSQjPVDYQLkSOA7+V4fsxsB8J7MMbd32tmuwCfcfezK32uQrUhr+dazGyP0gMz+xChwzNvM4F74tf/ERZ+q/i4+N64+7wum/IaiFC6MfEt4M3E11vAcjNbbma5DektcfcV7n4d8Lm8zx29FZvyHgVmmNmFhBtJ8/RpYF/gDXf/EmGkZq73jkSbuft17t4Zv64HNsszgPi3cCzh/3MYcEyMI08/AqYRhr5DGJWayd+najSVMxF40MxeIFyhbAf8xcz+BPkNc+5aHY6drPfnce6Et+IIo84Yw17AkjxO7O67xO/rvIiKTRSPkcOggCL0EyVMITQhfh34Wowj76RX9VpV1GFmVlpx18zGkfPw+9h8N53QfAfwQTPLu/luE3e/z8y+B6tvPG/r7UnlUKKpnJN736UqOoExOZ/zdEKz2XvN7AFgB+AzOcewTu4+38wOzul0RegnAsDd30g8nFaNGHhnrep18q9VAXyLMBLzacLv5n2Epqs8FaH5rt3MBrHmgnAMa+4tqiglmsp5GjiNULNJdnR+LM8guvTR1BP+iXJpB09cwf8FmATsEh8/CVRz6pO1uHumN64WqZ+oxMwM+DawPWtP8JnnDcVFqFURr+KbgNJrf9zdF+Ycxmax+azkejPL+2L1SuAOYISZnQ18gZCEK06JpnKuA54FxgHfIUzn0FKFOGYmfl4J/CDHOaVKV/AlRRhtVQ2PEZLsW7wzsdSZ2b+Ac9w9t3t6CHej30gYCZn3LA1AYWpVwOqa7f3Ez0Az2yjnYedVb75z9xvMbA7honAIcIS7ZzKxpxJN5Wzv7geb2WR3v8XMfgn8vgpx3AV8k0TNysxyqVmV+kXM7AxC5/+PCcnmGKAh6/MXRZH6iRJWufsPcjzfOxSkVoWZHUQY9bVF3FSNGRuSzXcQ/l9zbb4zs/8lzNTQAjwDHGtm57n7BZU+lxJN5ZRGVbWZ2XBCU1GuI1min1D9mtVBpQ/b6AIzawHOyzmOQsq5n6jkPjP7lLvfm/N5k6peq4q+T5Vn1Y7NdzsBu8VN1Wi+M3dfamafJYxQPQV4HFCiKbDZMcH8jPDLWkJ1ms6KULMaso45lHpckGxDk3U/0Tr8FrjLzDoIF0V1QKe7j8wxhqrXqqJCzKodb9ac2euO2RkUv+8J3Ovub8e/j4pToqkQdz88/nhRHNI8DLivCqEUoWb1beDxWIuBcEf6F3OOQdb2Y8IaLE9SvdpEEWpVUKxZtavpWTO7FxgPnGYZLgSnRJOBrDrUUqp6zcrdfxkn6ks2C+Q5K62802J3/0WVYyhCrQoKNKt2lX2BcAPtM3GutTGEkbMVV9fZWZXRlpKDOFPBMOA+d19V7XikeiwsOLaYMBV/Va7i483Mp9GlVuXuc/OKQapDiUZkA9Cl7X31Vby753YVb2Z/ynuEmRSDEo2I5KIItSqpDvXRiEheSjdpXsmG3TeywVGNRkREMqVlAkREJFNKNCIikiklGpEaFdd92b7acYhoMIBIDszsJWA0MDo5p1Vcr35n4L3u/lI/jv8AcJO7X9uvQEUyoBqNSH5eJKzJAoCZTUBzwMkGQDUakfzcCEwFLouPvwDcQBz2a2abxLJPAW8D1wDnxSV2jyQst/A48F+EqYW+4u73mtm5wIeB3c3sEuCn7n5CPMfH43xWmwE3Aye4u4aaSq5UoxHJz+PAu81svJkNAP4TuClRfhmwCbAtYUbdqYSJMEt2AxwYQZjq/idmVufu3wYeJiSRjRNJBmB/4P2ElVYPIcxtJZIrJRqRfJVqNfsAzwGvxu2lxHO6u78V+2suZO3FsOa6+zXu3g7MICzcNaqX853v7kvc/WXCchE7V+yViKSkRCOSrxuBQ4EjCc1mJSMI64MkJ5icC4xJPJ5X+iExbcvGvZxvXuLnt1PsL1JxSjQiOYozFb8I7Af8MlG0EFgJbJ3YthVrajy9Ub+LFJYSjUj+/gv4mLsvT2xrJ0w2ea6ZvcvMtga+xtp9OD15g9C3I1I4SjQiOXP3v7v7n9dRdCKwHJgDPEJYvO66lIf9IfBZM/unmV1amUhFKkOTaoqISKZUoxERkUwp0YiISKaUaEREJFNKNCIikiklGhERyZQSjYiIZEqJRkREMqVEIyIimVKiERGRTP1/zRRKl0hpfE4AAAAASUVORK5CYII=", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "pd.crosstab(data.month,data.y).plot(kind='bar')\n", "plt.title('Purchase Frequency for Month')\n", "plt.xlabel('Month')\n", "plt.ylabel('Frequency of Purchase')\n", "plt.savefig('pur_fre_month_bar')" ] }, { "cell_type": "markdown", "metadata": { "id": "ONUwszQUQWrI" }, "source": [ "El mes podría ser un buen predictor de la variable de resultado." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 301 }, "id": "lIf2GHKdQZ2w", "outputId": "49bca116-7afe-4c92-dd8c-eedfe0806617" }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "data.age.hist()\n", "plt.title('Histogram of Age')\n", "plt.xlabel('Age')\n", "plt.ylabel('Frequency')\n", "plt.savefig('hist_age')" ] }, { "cell_type": "markdown", "metadata": { "id": "Zd0AUtEYQcit" }, "source": [ "La mayoría de los clientes del banco en este conjunto de datos están en el rango de edad de 30-40 años." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 356 }, "id": "_zXUZLCeQcJy", "outputId": "86c8dfbb-c961-4284-c9ee-02ec6d2af5d1" }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "pd.crosstab(data.poutcome,data.y).plot(kind='bar')\n", "plt.title('Purchase Frequency for Poutcome')\n", "plt.xlabel('Poutcome')\n", "plt.ylabel('Frequency of Purchase')\n", "plt.savefig('pur_fre_pout_bar')" ] }, { "cell_type": "markdown", "metadata": { "id": "HmQEMIGkQjMa" }, "source": [ "El Poutcome parece ser un buen predictor de la variable de resultado." ] }, { "cell_type": "markdown", "metadata": { "id": "mYGdInqiQxxS" }, "source": [ "### Model fitting" ] }, { "cell_type": "markdown", "metadata": { "id": "OdKOges_RPBz" }, "source": [ "#### Crear variables dummy" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "- Transformar las siguientes variables en dummies\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "up8IfMMDRRIt" }, "outputs": [], "source": [ "cat_vars=['job','marital','education','default','housing','loan','contact','month','day_of_week','poutcome']\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "- Guardar las demás variables en un dataframe" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "metadata": {}, "source": [ "- Separar las variables en matriz **X** y vector **y**" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "metadata": {}, "source": [ "- Separar los datos en train y test" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "metadata": {}, "source": [ "- Ajustar la regresión logística" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "metadata": {}, "source": [ "- Predecir con los datos de test\n", "- Calcular accuracy" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "metadata": { "id": "yaQIsnFiT1Bi" }, "source": [ "### Confusion Matrix & Error Metrics" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "from sklearn.metrics import confusion_matrix, plot_confusion_matrix" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "- Graficar la matriz de confusión" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "metadata": { "id": "jsjUwPJnU0qs" }, "source": [ "- Calcular precision, recall y f1 usando classification_report" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "from sklearn.metrics import classification_report" ] }, { "cell_type": "markdown", "metadata": { "id": "EXCzx71CV-BY" }, "source": [ "#### ROC Curve" ] }, { "cell_type": "code", "execution_count": 38, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 301 }, "id": "DHA4czSVV9eJ", "outputId": "5704236e-c225-40c8-d528-a544dece513e" }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "from sklearn.metrics import roc_auc_score\n", "from sklearn.metrics import roc_curve\n", "logit_roc_auc = roc_auc_score(y_test, logreg.predict(X_test))\n", "fpr, tpr, thresholds = roc_curve(y_test, logreg.predict_proba(X_test)[:,1])\n", "plt.figure()\n", "plt.plot(fpr, tpr, label='Logistic Regression (area = %0.2f)' % logit_roc_auc)\n", "plt.plot([0, 1], [0, 1],'r--')\n", "plt.xlim([0.0, 1.0])\n", "plt.ylim([0.0, 1.05])\n", "plt.xlabel('False Positive Rate')\n", "plt.ylabel('True Positive Rate')\n", "plt.title('Receiver operating characteristic')\n", "plt.legend(loc=\"lower right\")\n", "plt.savefig('Log_ROC')\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": { "id": "Kdtt04s1WI24" }, "source": [ "La curva receiver operating characteristic (ROC) es otra herramienta común utilizada con los clasificadores binarios. La línea de puntos representa la curva ROC de un clasificador puramente aleatorio; un buen clasificador se mantiene lo más alejado posible de esa línea (hacia la esquina superior izquierda)." ] }, { "cell_type": "markdown", "metadata": { "id": "Gj7V25T4Suz_" }, "source": [ "### SMOTE Oversampling" ] }, { "cell_type": "markdown", "metadata": { "id": "yK3la9rES-6W" }, "source": [ "Una vez creados nuestros datos de entrenamiento, voy a realizar un muestreo ascendente de la no-suscripción utilizando el algoritmo SMOTE (Synthetic Minority Oversampling Technique). A alto nivel, \n", "\n", "1. SMOTE: Funciona creando muestras sintéticas de la clase menor (no-suscripción) en lugar de crear copias.\n", "2. Eligiendo aleatoriamente uno de los k-próximos y utilizándolo para crear una nueva observación similar, pero ajustada aleatoriamente." ] }, { "cell_type": "code", "execution_count": 44, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "-hOuCvPPTMrI", "outputId": "a01177d7-d613-4649-88c6-11779a0bd0f8" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Longitud de data oversampleada: 51134\n", "Numero de no subscripciones en data oversampleada: 25567\n", "Cantidad de subscripciones: 25567\n", "Proporción de no subscripciones en data oversampleada: 0.5\n", "Proporción de subscripciones en data oversampleada: 0.5\n" ] } ], "source": [ "from imblearn.over_sampling import SMOTE\n", "os = SMOTE(random_state=0)\n", "\n", "columns = X_train.columns\n", "os_data_X,os_data_y = os.fit_resample(X_train, y_train)\n", "os_data_X = pd.DataFrame(data=os_data_X,columns=columns )\n", "os_data_y= pd.DataFrame(data=os_data_y,columns=['y'])\n", "\n", "# revisamos los nros en nuestra data\n", "print(\"Longitud de data oversampleada: \",len(os_data_X))\n", "print(\"Numero de no subscripciones en data oversampleada: \",len(os_data_y[os_data_y['y']==0]))\n", "print(\"Cantidad de subscripciones: \",len(os_data_y[os_data_y['y']==1]))\n", "print(\"Proporción de no subscripciones en data oversampleada: \",len(os_data_y[os_data_y['y']==0])/len(os_data_X))\n", "print(\"Proporción de subscripciones en data oversampleada: \",len(os_data_y[os_data_y['y']==1])/len(os_data_X))" ] }, { "cell_type": "markdown", "metadata": { "id": "VXt3vmJhTSWW" }, "source": [ "Ahora tenemos unos datos perfectamente equilibrados. Puede que haya notado que he sobremuestreado sólo los datos de entrenamiento, porque al sobremuestrear sólo los datos de entrenamiento, no se utiliza ninguna información de los datos de prueba para crear observaciones sintéticas, por lo tanto, no se filtrará información de los datos de prueba al entrenamiento del modelo." ] }, { "cell_type": "markdown", "metadata": { "id": "kXL7Nka_SwsJ" }, "source": [ "### Feature Selection" ] }, { "cell_type": "markdown", "metadata": { "id": "J1WobhXRTXC6" }, "source": [ "La eliminación recursiva de características (RFE - [Recursive Feature Elimination](https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.RFE.html)) se basa en la idea de construir repetidamente un modelo y elegir la característica con mejor o peor rendimiento, apartando la característica y repitiendo el proceso con el resto de las características. Este proceso se aplica hasta que se agotan todas las características del conjunto de datos. El objetivo de la RFE es seleccionar características considerando recursivamente conjuntos de características cada vez más pequeños." ] }, { "cell_type": "code", "execution_count": 53, "metadata": { "id": "rc1bAf5KYqoP" }, "outputs": [], "source": [ "import warnings\n", "warnings.filterwarnings('ignore')" ] }, { "cell_type": "code", "execution_count": 86, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "SVy5oFkIRLJV", "outputId": "ee9608e6-fb54-45b9-b585-99f0b31dd525" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[False False False False False False False False False False True True\n", " True False False True True False False True True False False False\n", " True False True True True False False False False False False False\n", " False False False False False False False False False True True True\n", " False False True]\n", "[32 31 26 35 23 24 27 28 34 29 1 1 1 3 33 1 1 20 4 1 1 11 12 30\n", " 1 36 1 1 1 10 37 14 18 17 15 5 7 16 8 13 25 9 6 21 19 1 1 1\n", " 2 22 1]\n" ] } ], "source": [ "data_final_vars=df.columns.values.tolist()\n", "y=['y']\n", "X=[i for i in data_final_vars if i not in y]\n", "from sklearn.feature_selection import RFE\n", "from sklearn.linear_model import LogisticRegression\n", "logreg = LogisticRegression()\n", "rfe = RFE(logreg, n_features_to_select=15)\n", "rfe = rfe.fit(os_data_X, os_data_y.values.ravel())\n", "print(rfe.support_)\n", "print(rfe.ranking_)" ] }, { "cell_type": "code", "execution_count": 87, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "MNWvZtCLZG7w", "outputId": "064528b6-4831-41fe-f479-f5c368252432" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "RFE nos deja con las siguientes variables: ['job_blue-collar' 'job_entrepreneur' 'job_housemaid' 'job_self-employed'\n", " 'job_services' 'job_unemployed' 'job_unknown' 'education_high.school'\n", " 'education_professional.course' 'education_university.degree'\n", " 'education_unknown' 'day_of_week_mon' 'day_of_week_thu' 'day_of_week_tue'\n", " 'poutcome_success']\n" ] } ], "source": [ "print(\"RFE nos deja con las siguientes variables: \", rfe.feature_names_in_[rfe.support_])" ] }, { "cell_type": "code", "execution_count": 88, "metadata": { "id": "8ucN3dsnTiFc" }, "outputs": [], "source": [ "cols=rfe.feature_names_in_[rfe.support_]\n", "X_train_2=os_data_X[cols]\n", "X_test_2=X_test[cols]\n", "y_train_2=os_data_y['y']" ] }, { "cell_type": "markdown", "metadata": { "id": "HykOXfZNaX-e" }, "source": [ "#### Train the new model:" ] }, { "cell_type": "code", "execution_count": 89, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "D3oAwJ7BZ88Z", "outputId": "8d254ae9-6d17-40e3-c004-db250b4b4fa9" }, "outputs": [ { "data": { "text/plain": [ "LogisticRegression()" ] }, "execution_count": 89, "metadata": {}, "output_type": "execute_result" } ], "source": [ "logreg = LogisticRegression()\n", "logreg.fit(X_train_2, y_train_2)" ] }, { "cell_type": "code", "execution_count": 90, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 319 }, "id": "f71xev3BalLf", "outputId": "15558e99-4b5a-4b14-ec9e-4bc7092df9b1" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[[8534 2447]\n", " [ 841 535]]\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "class_names = ['no subscripción', 'si subscripción']\n", "\n", "disp = plot_confusion_matrix(logreg, X_test_2, y_test,\n", " display_labels=class_names,\n", " cmap=plt.cm.Blues)\n", "\n", "print(disp.confusion_matrix)" ] }, { "cell_type": "code", "execution_count": 91, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "ww0VfgDnabbq", "outputId": "0b7f102c-5d04-49ea-f7e4-563bb56792d4" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ " precision recall f1-score support\n", "\n", " 0 0.93 0.98 0.95 10981\n", " 1 0.67 0.41 0.51 1376\n", "\n", " accuracy 0.91 12357\n", " macro avg 0.80 0.69 0.73 12357\n", "weighted avg 0.90 0.91 0.90 12357\n", "\n" ] } ], "source": [ "print(classification_report(y_test, y_pred))" ] }, { "cell_type": "code", "execution_count": 92, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 301 }, "id": "0-SER3zra2GZ", "outputId": "27d31935-3eab-4426-aa85-eb27ca303387" }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "logit_roc_auc = roc_auc_score(y_test, logreg.predict(X_test_2))\n", "fpr, tpr, thresholds = roc_curve(y_test, logreg.predict_proba(X_test_2)[:,1])\n", "plt.figure()\n", "plt.plot(fpr, tpr, label='Logistic Regression (area = %0.2f)' % logit_roc_auc)\n", "plt.plot([0, 1], [0, 1],'r--')\n", "plt.xlim([0.0, 1.0])\n", "plt.ylim([0.0, 1.05])\n", "plt.xlabel('False Positive Rate')\n", "plt.ylabel('True Positive Rate')\n", "plt.title('Receiver operating characteristic')\n", "plt.legend(loc=\"lower right\")\n", "plt.savefig('Log_ROC')\n", "plt.show()" ] } ], "metadata": { "colab": { "collapsed_sections": [ "yaQIsnFiT1Bi", "Gj7V25T4Suz_", "kXL7Nka_SwsJ" ], "name": "CLASE 4 - Clasificación - Sincrónica.ipynb", "provenance": [] }, "kernelspec": { "display_name": "Python 3", "name": "python3" }, "language_info": { "name": "python" } }, "nbformat": 4, "nbformat_minor": 0 }