tests_AdultsDataset.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "ad173045",
   "metadata": {},
   "source": [
    "# Librerías necesarias\n",
    "pyplot nos ayudará con las pruebas recogidas en matrices a exponerlas en gráficos. En cambio, SequentialDecisionTreeAlgorithm es el archivo que contiene la clase SequentialDecisionTreeAlgorithm como propuesta de algoritmo secuencial del alumno."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "48e1f09e",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "# Para modelo de aprendizaje por árboles de decisión\n",
    "from lib.SequentialDecisionTreeAlgorithm import SequentialDecisionTreeAlgorithm\n",
    "# Generalizado para más modelos: árboles de decisión o knn\n",
    "from lib.SequentialModelAlgorithm import SequentialModelAlgorithm\n",
    "\n",
    "# Selección de columnas\n",
    "dataset = pd.read_csv('./datasets/adultDataset.csv', header = 0)\n",
    "attr_cols = dataset.loc[:, 'capital-gain':'native-country']\n",
    "obj_col = dataset['income']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6954e9ba",
   "metadata": {},
   "source": [
    "## Sin ensamble de modelos secuenciales"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "12047c52",
   "metadata": {},
   "outputs": [
    {
     "ename": "IndentationError",
     "evalue": "unexpected indent (<ipython-input-32-778c94ed2d89>, line 44)",
     "output_type": "error",
     "traceback": [
      "\u001b[1;36m  File \u001b[1;32m\"<ipython-input-32-778c94ed2d89>\"\u001b[1;36m, line \u001b[1;32m44\u001b[0m\n\u001b[1;33m    attr = attributes_preprocess(attr_cols)\u001b[0m\n\u001b[1;37m    ^\u001b[0m\n\u001b[1;31mIndentationError\u001b[0m\u001b[1;31m:\u001b[0m unexpected indent\n"
     ]
    }
   ],
   "source": [
    "from sklearn import tree, preprocessing, model_selection\n",
    "from sklearn.metrics import balanced_accuracy_score\n",
    "from sklearn.compose import ColumnTransformer\n",
    "import sklearn.utils as skl\n",
    "\n",
    "def attributes_preprocess(attributes_cols):\n",
    "    # Acumulamos las transformaciones necesarias en una matriz para realizarlas en la función ColumnTransformer\n",
    "    transforms = []\n",
    "    for i in attributes_cols: # Recorremos cada columna\n",
    "        # Primer valor de cada columna. Podríamos recorrer cada valor de la columna para asegurarnos de que sean\n",
    "        # numéricos o cadena de texto, pero en los Dataset de ejemplo nos vale con comprobar el primer valor.\n",
    "        fval = attributes_cols[i][0]\n",
    "\n",
    "        # Comprobamos que sea numérico, sino acumulamos una transformación sobre la columna\n",
    "        if isinstance(fval, (int, float, np.int64))==False:\n",
    "            transforms.append((\"encode-\"+i, preprocessing.OrdinalEncoder(), [i]))\n",
    "\n",
    "    # Devolvemos como respuesta un array de enteros aplicando las transformaciones acumuladas en transforms\n",
    "    # # remainder = 'passthrough' nos devolverá en la transformación también las columnas no afectadas\n",
    "    res = np.array(ColumnTransformer(transforms, remainder='passthrough').fit_transform(attributes_cols), dtype=float)\n",
    "\n",
    "    return res\n",
    "\n",
    "# Preprocesado de la columna objetivo. Codifica datos en formato de texto en numéricos  \n",
    "# @ objetive_col columnas objetivo\n",
    "def objetive_preproccess(objetive_col):\n",
    "    # Primer valor de cada columna. Podríamos recorrer cada valor de la columna para asegurarnos de que sean\n",
    "    # numéricos o cadena de texto, pero en los Dataset de ejemplo nos vale con comprobar el primer valor.\n",
    "    fval = objetive_col[0]\n",
    "\n",
    "    if isinstance(fval, (int, float, np.int64))==False:\n",
    "        objetive_col = preprocessing.LabelEncoder().fit_transform(objetive_col)\n",
    "\n",
    "    return objetive_col\n",
    "\n",
    "def classify_prediction(pred):\n",
    "    Xmax = np.max(pred)\n",
    "    Xmin = np.min(pred)\n",
    "    for i in range(len(pred)):\n",
    "        pred[i] = np.around((pred[i] - Xmin)/(Xmax-Xmin))\n",
    "    return pred\n",
    "\n",
    "\n",
    "    attr = attributes_preprocess(attr_cols)\n",
    "    obj = objetive_preproccess(obj_col)\n",
    "\n",
    "    (X_train, X_test, y_train, y_test) = model_selection.train_test_split(\n",
    "            attr, obj,\n",
    "            random_state=12345,\n",
    "            test_size=0.33,\n",
    "            stratify=obj)\n",
    "\n",
    "    Tree = tree.DecisionTreeRegressor(max_depth=10)\n",
    "    Tree = Tree.fit(X_train, y_train)\n",
    "    #Tree.plot_tree(clf)\n",
    "\n",
    "    cols_mean = attr.mean(axis=0)\n",
    "\n",
    "    predict = Tree.predict([cols_mean])\n",
    "\n",
    "print(Tree.score(X_test, y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "81354878",
   "metadata": {},
   "source": [
    "## 1. Pruebas en el Dataset adultsDataset\n",
    "La primera prueba la realizamos con parámetros predeterminados que usaremos también en las demás pruebas. Esto nos ayudará a fijar unas proporciones y poder comparar los resultados del algoritmo entre diferentes datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "e4e84ea6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "---- adultDataset.csv ----\n",
      "Hiperparámetros predeterminados: \n",
      "- ntree = 300\n",
      "- sample_size = 0.65\n",
      "- max_depth = 10\n",
      "- lr = 0.1\n",
      "Score: 0.9005186117456888\n"
     ]
    }
   ],
   "source": [
    "print('---- adultDataset.csv ----')\n",
    "print('Hiperparámetros predeterminados: \\n- ntree = 300\\n- sample_size = 0.65\\n- max_depth = 10\\n- lr = 0.1')\n",
    "SeqTree = SequentialDecisionTreeAlgorithm()\n",
    "trees, score = SeqTree.start(attributes_cols = attr_cols, objetive_col = obj_col)\n",
    "print('Score: '+str(score))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "27609b4b",
   "metadata": {},
   "source": [
    "### 1.1 Alteración del hiperparámetro ntree\n",
    "Con esto aumentaremos o disminuiremos la cantidad de árboles secuenciales en diferentes pruebas. Probaremos fijando los parámetros predeterminados y moviendo solo el parámetro ntrees entre 1 y 500."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "b1b50291",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--- Alterando sample_size [1 - 500] ---\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mejor puntuación: 0.9118628588468929 con valor = 350\n",
      "--------------------------\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "values = [1, 50, 100, 150, 250, 350, 500]\n",
    "print('\\n--- Alterando sample_size ['+str(np.min(values))+' - '+str(np.max(values))+'] ---')\n",
    "scores = []\n",
    "\n",
    "for v in range(len(values)):\n",
    "    SeqTree = SequentialDecisionTreeAlgorithm(ntree = values[v], sample_size = 0.65, max_depth = 10, lr = 0.1)\n",
    "    trees, score = SeqTree.start(attributes_cols = attr_cols, objetive_col = obj_col)\n",
    "    scores.append(score)\n",
    "    \n",
    "# Realizamos una gráfica para determinar cómo afecta ntree a la puntuación\n",
    "plt.plot(values, scores)\n",
    "plt.xlabel('- nTrees -')\n",
    "plt.ylabel('- Balanced Accuracy Score -')\n",
    "plt.title('Alteración del hiperparámetro ntrees y manteniendo el resto constante')\n",
    "plt.show()\n",
    "\n",
    "m = np.argmax(scores)\n",
    "print('Mejor puntuación: '+str(scores[m])+' con valor = '+str(values[m]))\n",
    "\n",
    "print('--------------------------')\n",
    "print('\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5f0e079b",
   "metadata": {},
   "source": [
    "Observamos que a medida que nos acercamos a los 350 árboles obtenemos el máximo valor de puntuación. Es una función logarítmica por lo que no esperamos encontrar mejores valores por más que aumentemos los árboles. Podemos concretar que lo más óptimo sería fijar el algoritmo en 350 árboles al menos para este ejemplo, ya que nos devolvería un resultado similar pero en menor ejecución"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9bd58d59",
   "metadata": {},
   "source": [
    "### 1.2 Alteración del hiperparámetro sample_size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "932b8828",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--- Alterando sample_size [0.1 - 0.9] ---\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mejor puntuación: 0.9020215788410728 con valor = 0.9\n"
     ]
    }
   ],
   "source": [
    "values = [0.1, 0.25, 0.5, 0.75, 0.9]\n",
    "print('\\n--- Alterando sample_size ['+str(np.min(values))+' - '+str(np.max(values))+'] ---')\n",
    "\n",
    "scores = []\n",
    "for v in range(len(values)):\n",
    "    SeqTree = SequentialDecisionTreeAlgorithm(ntree = 300, sample_size = values[v], max_depth = 10, lr = 0.1)\n",
    "    trees, score = SeqTree.start(attributes_cols = attr_cols, objetive_col = obj_col)\n",
    "    scores.append(score)\n",
    "    \n",
    "# Realizamos una gráfica para determinar cómo afecta ntree a la puntuación\n",
    "plt.plot(values, scores)\n",
    "plt.xlabel('- Proporción sample_size -')\n",
    "plt.ylabel('- Balanced Accuracy Score -')\n",
    "plt.title('Alteración del hiperparámetro sample_size')\n",
    "plt.show()\n",
    "\n",
    "m = np.argmax(scores)\n",
    "print('Mejor puntuación: '+str(scores[m])+' con valor = '+str(values[m]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a228a7dd",
   "metadata": {},
   "source": [
    "Debido a la aletoriedad de la muestra, este parámetro no muestra un cambio directo. Comprobamos también realizando el proceso 5 veces y realizando la media de los resultados obtenidos para cada valor (puede tardar bastante):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "0e161271",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--- Alterando 5 veces sample_size [0.1 - 0.9] ---\n",
      "[0.89862987 0.89665521 0.9020798  0.89282503 0.89349684]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mejor puntuación: 0.9020797991597934 con valor = 0.5\n"
     ]
    }
   ],
   "source": [
    "values = [0.1, 0.25, 0.5, 0.75, 0.9]\n",
    "print('\\n--- Alterando 5 veces sample_size ['+str(np.min(values))+' - '+str(np.max(values))+'] ---')\n",
    "\n",
    "tests = []\n",
    "for j in range(5):\n",
    "    for v in range(len(values)):\n",
    "        SeqTree = SequentialDecisionTreeAlgorithm(ntree = 300, sample_size = values[v], max_depth = 10, lr = 0.1)\n",
    "        trees, score = SeqTree.start(attributes_cols = attr_cols, objetive_col = obj_col)\n",
    "        if len(tests)<= v:\n",
    "            tests.append([])\n",
    "        tests[v].append(score)\n",
    "\n",
    "print(np.mean(tests, axis=0))\n",
    "# Realizamos una gráfica para determinar cómo afecta ntree a la puntuación\n",
    "plt.plot(values, np.mean(tests, axis=0))\n",
    "plt.xlabel('- Proporción sample_size -')\n",
    "plt.ylabel('- Balanced Accuracy Score -')\n",
    "plt.title('Alteración del hiperparámetro sample_size')\n",
    "plt.show()\n",
    "\n",
    "m = np.argmax(np.mean(tests, axis=0))\n",
    "print('Mejor puntuación: '+str(np.mean(tests, axis=0)[m])+' con valor = '+str(values[m]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4ca74d2f",
   "metadata": {},
   "source": [
    "Podemos decir que cuanta mayor proporción de ejemplos del conjunto de evaluación mayor puntuación obtiene el algoritmo. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3b9c2211",
   "metadata": {},
   "source": [
    "## 1.3 Alteración del hiperparámetro max_depth"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "18bdf175",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--- Alterando max_depth [1 - 20] ---\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mejor puntuación: 0.9119241483786409 con valor = 10\n"
     ]
    }
   ],
   "source": [
    "values = [1, 5, 10, 15, 20]\n",
    "print('\\n--- Alterando max_depth ['+str(np.min(values))+' - '+str(np.max(values))+'] ---')\n",
    "\n",
    "scores = []\n",
    "for v in range(len(values)):\n",
    "    SeqTree = SequentialDecisionTreeAlgorithm(ntree = 300, sample_size = 0.65, max_depth = values[v], lr = 0.1)\n",
    "    trees, score = SeqTree.start(attributes_cols = attr_cols, objetive_col = obj_col)\n",
    "    scores.append(score)\n",
    "    \n",
    "# Realizamos una gráfica para determinar cómo afecta ntree a la puntuación\n",
    "plt.plot(values, scores)\n",
    "plt.xlabel('- max_depth -')\n",
    "plt.ylabel('- Balanced Accuracy Score -')\n",
    "plt.title('Alteración del hiperparámetro max_depth')\n",
    "plt.show()\n",
    "\n",
    "m = np.argmax(scores)\n",
    "print('Mejor puntuación: '+str(scores[m])+' con valor = '+str(values[m]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7ca0b87e",
   "metadata": {},
   "source": [
    "### 1.4 Alteración del hiperparámetro de aprendizaje lr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "8f55157e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--- Alterando lr [0.1 - 0.9] ---\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mejor puntuación: 0.9210153184043275 con valor = 0.25\n"
     ]
    }
   ],
   "source": [
    "values = [0.1, 0.25, 0.5, 0.75, 0.9]\n",
    "print('\\n--- Alterando lr ['+str(np.min(values))+' - '+str(np.max(values))+'] ---')\n",
    "\n",
    "scores = []\n",
    "for v in range(len(values)):\n",
    "    SeqTree = SequentialDecisionTreeAlgorithm(ntree = 300, sample_size = 0.65, max_depth = 10, lr = values[v])\n",
    "    trees, score = SeqTree.start(attributes_cols = attr_cols, objetive_col = obj_col)\n",
    "    scores.append(score)\n",
    "    \n",
    "# Realizamos una gráfica para determinar cómo afecta ntree a la puntuación\n",
    "plt.plot(values, scores)\n",
    "plt.xlabel('- lr -')\n",
    "plt.ylabel('- Balanced Accuracy Score -')\n",
    "plt.title('Alteración del hiperparámetro de aprendizaje lr')\n",
    "plt.show()\n",
    "\n",
    "m = np.argmax(scores)\n",
    "print('Mejor puntuación: '+str(scores[m])+' con valor = '+str(values[m]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "6f2df236",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--- Alterando 5 veces lr [0.1 - 0.9] ---\n",
      "[0.90935954 0.90461854 0.89973575 0.90668713 0.90593361]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mejor puntuación: 0.9093595367079773 con valor = 0.1\n"
     ]
    }
   ],
   "source": [
    "values = [0.1, 0.25, 0.5, 0.75, 0.9]\n",
    "print('\\n--- Alterando 5 veces lr ['+str(np.min(values))+' - '+str(np.max(values))+'] ---')\n",
    "\n",
    "tests = []\n",
    "for j in range(5):\n",
    "    for v in range(len(values)):\n",
    "        SeqTree = SequentialDecisionTreeAlgorithm(ntree = 300, sample_size = 0.65, max_depth = 10, lr = values[v])\n",
    "        trees, score = SeqTree.start(attributes_cols = attr_cols, objetive_col = obj_col)\n",
    "        if len(tests)<= v:\n",
    "            tests.append([])\n",
    "        tests[v].append(score)\n",
    "\n",
    "print(np.mean(tests, axis=0))\n",
    "# Realizamos una gráfica para determinar cómo afecta ntree a la puntuación\n",
    "plt.plot(values, np.mean(tests, axis=0))\n",
    "plt.xlabel('- Proporción lr -')\n",
    "plt.ylabel('- Balanced Accuracy Score -')\n",
    "plt.title('Alteración del hiperparámetro lr')\n",
    "plt.show()\n",
    "\n",
    "m = np.argmax(np.mean(tests, axis=0))\n",
    "print('Mejor puntuación: '+str(np.mean(tests, axis=0)[m])+' con valor = '+str(values[m]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "54007ad5",
   "metadata": {},
   "source": [
    "## Exploración y evaluación de otros hiperparámetros\n",
    "Evaluaremos hiperparámetros opcionales como min_weight_fraction_leaf, min_samples_leaf y max_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "f6f7e4e6",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--- Alterando min_weight_fraction_leaf [0.1 - 0.4] ---\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mejor puntuación: 0.7644799597440504 con valor = 0.2\n"
     ]
    }
   ],
   "source": [
    "values = [0.1, 0.2, 0.3, 0.4]\n",
    "print('\\n--- Alterando min_weight_fraction_leaf ['+str(np.min(values))+' - '+str(np.max(values))+'] ---')\n",
    "\n",
    "scores = []\n",
    "for v in range(len(values)):\n",
    "    SeqTree = SequentialDecisionTreeAlgorithm(ntree = 300, sample_size = 0.65, max_depth = 10, lr = 0.1, min_weight_fraction_leaf=values[v])\n",
    "    trees, score = SeqTree.start(attributes_cols = attr_cols, objetive_col = obj_col)\n",
    "    scores.append(score)\n",
    "    \n",
    "# Realizamos una gráfica para determinar cómo afecta ntree a la puntuación\n",
    "plt.plot(values, scores)\n",
    "plt.xlabel('- lr -')\n",
    "plt.ylabel('- Balanced Accuracy Score -')\n",
    "plt.title('Alteración del hiperparámetro de aprendizaje min_weight_fraction_leaf')\n",
    "plt.show()\n",
    "\n",
    "m = np.argmax(scores)\n",
    "print('Mejor puntuación: '+str(scores[m])+' con valor = '+str(values[m]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "42c5ce0e",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--- Alterando max_features ---\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZwAAAEWCAYAAABSaiGHAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAA75UlEQVR4nO3dd3hUZfbA8e9JCCWU0EINHaS3JIIFEMUGKiDYQHddOwoi6lp2dW1Y164grqvurmJBsSG6dgXRnyWB0EEQBELvvSWc3x/vjY7ZlBvIzJ1Jzud55knmtjn3zp05c9/73nNFVTHGGGPCLS7oAIwxxpQPlnCMMcZEhCUcY4wxEWEJxxhjTERYwjHGGBMRlnCMMcZERJlMOCLybxG5NwriaCoiu0QkPt/wKiLyjYicXoqv9ZWIXO5zWhWR1iVdTmHrEwtE5L8icnXQcRwuEWnuvW8Vgo6lNInIn0RkRsjzXSLS8giXOV9E+h5pbEEQkbYiMktEdorI6KDjKW0xnXC8L8etIlKpiGn6ikh2JOPKo6orVbWaqubmG/UP4BFV/SiIuA5XEesT1UTkj8BmVZ0QxtcoNIkb/7z9a9kRLqOjqn5VSiFF2s3AV6paXVWfOpIFleRHaKTEbMIRkeZAb0CBgWF8nVL/Ramqf1TVd0p7ubFKnFLdF/MdhVUDrirN5ZdUWTkyKSvrEcWaAfODDgLC9F6rakw+gDuAb4DHgKn5xv0buBeoCuwFDgG7vEcjXKK9FfgZ2Ay8AdT25m2OS2KXASuB6d7wN4F1wHZgOtAx5PWqAI8CK7zxM7xhecuq4E3XCJgCbAGWAleELOMuL46XgJ24nS69iPU/BVjkvd44YBpwecj4S4GFwFbgY6BZyDgFWhey3K+Asd623Ql8AtTNt20qhEz7APCDF8d7edvRG38M8C2wDZgN9M33Ovd5r7MXaO1jeUW9B/8GJgAfAruBk4EzgFnADmAVcFfI9Hnrcok3biswAjgamOPFPC7ftilwm3qxqPe6u4Dzgb5ANnCLF/PLQCXgCWCN93gCqFTI+xAPPAJsApYBI/Nt+yTgBWAtsBq3v8cXsqwewP9567QWt79UzLc/jPZeZxPwMBDnjfuT9x49jttv7/XW4xHc52M98CxQxZs+b71vBDZ4r3dJyGvVwX0Gdnjv81hgRv59E/dZ2RXy2AOoN00r4AvcZ3cT8ApQM2QZvwAne/8X+lkvYDvlxX5zSOyDgQHAT976/9XPdgWO82Jr4j3v6k3XrojP9BdALrDPW+ejitnWtYCpwEbcPjkVSPHG3ZdvWePI9/kN+RxefpjvdV3vNbd503+Nt98Uuo5BJ47DfeC+sK8B0oCDQP18Xz73hu5E+eYdA3wHpHgb9B/Aa/m+iF7CJay8jXspUJ3fvjSyQpY33nvjGuO+KI7zpvvdG4xLCs8AlYFu3o7Szxt3l7dzDPCW8QDwXSHrXhf3gT0HSACuB3JCdpzB3vZpD1QAbge+zf+hLmTZX+E+nEfhkuZXwIP5tk1owlkNdPK21VvARG9cY9wHfADuQ3+K9zw5ZN6VQEcvxoSilufjPfg3LhEd771eZeAkoIv3vAvuS2RwvnV51pv2VG/7vwvU8+LfAJxwONsUt9/lAA958VYB7sHtd/WAZFwyHlvI+zAC94OiCVAb+DLftn8Xt99W9Zb3A3BVIctKwyX/Ct56LwTG5Iv9S+91muK+XEO/hHKAa735q3jbfoo3fXXgfeCBfOt9j/eeDsAli1re+NdxX/pVvfd5NQUknALW4RV++4y2xu1PlbztOB14ImTaX/gt4YyhkM96Aa+RF/sdXuxX4D6jr3rr2dHbR1r63K734ZJIFdyPmFE+vte+4vc/HIva1nWAoUCiN+5N4N0iltWc4hNOSd7rB3CfnwTv0RuQItcv0omiNB5AL1ySyfvlvQi4Pt+XT1EJZyHeF733vKG3vLwdR/N2qkJev6Y3TRLuy2wv0LWA6X59g3FfHLlA9ZDxDwD/9v6/C/gsZFwHYG8hr/9HQpIRILhfZnk7zn+By0LGx+E+9M2K+lCH7IC3hzy/BviooB2WkGQUEvMBXMK8BXg537I/Bi4OmfeeAl67wOUV9R6EvOcvFbPfPAE8nm9dGoeM3wycH/L8LbwvkJJuU9x+dwCoHDLsZ2BAyPPTgF8KifULYETI81ND9qX6wH68H0Pe+GHAlz4/P2OAd0KeK3B6vvf8c+//PwEr8+1ru4FWIcOOBZaHrPdefv+ltgH3xRyP+5y1Cxl3P8UkHG9fygxd33zjBwOzQp7/wm8Jp9DPegHLyYs93nte3YunZ8g0mXg/Wnxs1wRv+rnARxTzZRzyGcj7HBe5rQuYtxuwtaBl5dvni0o4JXmv78G1QhT4XVLQI1bbYy8GPlHVTd7zV71hj/ucvxnwjogcChmWi/sg51mV9493PuA+4FzcL6q8+erifjVVxn2ZFKURsEVVd4YMWwGkhzxfF/L/HqCyiFRQ1ZwClvVrfKqqIrIqZHwz4EkReTRkmOB+ta8oJs6C4qhWxLShr7sC9yGr68VwroicFTI+AfdLuqB5i1yeiGyi8Pdge0HLE5FUb572uPVPwv1CC7U+5P+9BTzPW/fD2aYbVXVfyPNG+aZd4Q0ryO/e43zzNcNtl7UikjcsjoK3JyJyFK7pOR33a7gC7oswVP7XalTIuGRvGZkhry24ZJJnc759Nm8fSvZeu7D1Kij2/sB1uC/9vd6wesBTuF/U1XHrvrWQRRT1WV9dwPSb9bdOMXu9vwXuE8VtV1U9KCL/9mK9Qb1v6RIocluLSCLuO+90XPMaQHURidfD79hTkvf6YdwP5U+88c+p6oNFLTzmOg2ISBXgPOAEEVknIutwTUpdRaRrAbMU9CavAvqras2QR2VVXV3IfMOBQbjzAkm4XwrgNv4m3GF2q2JCXwPUFpHqIcOaUvBOX5y1uCMmF4R7t5uEjF+Fa14JXb8qqvrtYbxWcUJftynu1+MmL4aX88VQNd8OWdB7U9jyinoPClveJFwbc2tVbQb8J9/0JXE42zR/PGtwX4B5mnrDCvK799ibNjSW/bgj/LxYaqhqx0KWNQHXCtBGVWsAf+V/t0P+1wqNK3Q9NuG+dDuGvHaSqhb1oyTPRlyTTWHr9Tsi0hb3np2nqqFfhA94MXXx1ueiAtYnj5/P+uEqcruKSGPgTuBfwKNF9aYtRHHb+kagLS4Z1wD65L209zf//rfb+5sYMqxBvml8v9equlNVb1TVlsBZwA0i0q+oFYq5hIM7fM7FNbd08x7tcSes/ljA9OuBOiKSFDLsWeA+EWkGICLJIjKoiNesjvuAb8a9WffnjVDVQ8CLwGMi0khE4kXk2Pw7l/eB+RZ4QEQqi0gXXMeEV3yud6gPgI4iMsTrSTKa3+84zwJ/EZGO3volici5h/E6flwkIh28X1v3AJO9X1cTgbNE5DRvm1T2uqinHObyCn0PilAT1yyZIyI9cM1Oh6u4bboeKO76kdeA2739rS7uXMHEQqZ9AxgtIikiUgt34hsAVV2L68zxqIjUEJE4EWklIicUsqzquHN+u0SkHVDQ9Ug3iUgtEWmCO6KYVNCCvP39n8Dj3pEGItJYRE4rZt3x3se3gbtEJFFEOuBaJv6HiNTANdfcrqoz8o2ujjsRvs37Ur+piJct6We9JArdrt6PwH/jOnZchvsBMbYkC/exravjEsI2EamNS26hfrdPqupG3A/ci7zP5KUU8UO5uNcXkTNFpLW3rjtw38tFHlnFYsK5GPiXumtC1uU9cL0wLszflU9VF+E+6MtEZJuINAKexJ0I+0REduJOKvYs4jVfwh36rwYWeNOH+jOunfZHXG+Nhyh42w7D/TJfA7wD3Kmqn/pe89/WaROuaelB3BdwG1zvkrzx73gxvC4iO4B5QP+Svo5PL+M+WOtwTYujvRhW4Y5I/or7ZbsK98VQ3D5X4PIo/j0oyNXAnd57fAfuS/yw+NimdwH/8fax8wpZzL1ABu4E8lxgpjesIP/EnfOa7U33dr7xfwQq4rbFVmAy7vxEQf6MO0Lc6S23oGTyHq45KAv3g+aFQpYF7pzKUuA7b1t8hvul7ccoXJPUOtz7/K9Cpkv1lvmYuItBd4nILm/c3d747V6s+bdNqJJ+1kuiqO06Gtds9zevKe0S4BIR6V3C1yhqWz+BO7G/Cbde+a/rexI4R9y1innX9FyB+xxuxnWCKK7Vo6jXb+M934XrrfeMFnP9k5S8WdEYR0S+wvUiez4al2f8ERHFNQstDTqW0iAiK4GLVHV60LGY34vFIxxjjCmQiCTjTnb/EnAopgCWcIwxZYKIHA0sAZ5W1ZVBx1MQ+a0eYUGPQjtQlBXWpGaMMSYi7AjHGGNMRMTqhZ8lUrduXW3evHnQYRhjTEzJzMzcpKrJpbW8cpFwmjdvTkZGRtBhGGNMTBERP5VJfLMmNWOMMRFhCccYY0xEWMIxxhgTEZZwjDHGRIQlHGOMMRFhCccYY0xEWMIxxhgTEZZwjAmAqvLD8i288v0Kcg9ZeSlTPpSLCz+NiRbZW/fw9szVTM7MZuWWPQBs3LmfMScfFXBkxoSfJRxjwmzPgRw+mreOyZnZfPvzZgCOa1WHMSe3YfpPG3ny8yV0b1qLE44qtQoixkSlsCYcETkdd9e5eOD5fPezx7t17ou425zuAy5V1XlFzevd1vcu3G2le6iq1awxUUdVyVixlckZ2Xwwdy279ufQtHYiN5xyFENSG5NSy91Wvn+nhixcu5Mxr89i6ujeNK5ZJeDIjQmfsCUcEYkHxgOnANnAjyIyRVUXhEz2VyBLVc/27gk+HuhXzLzzgCHAP8IVuzGHa/W2vbydmc3kmdms2LyHqhXjOaNLQ85Ja8LRzWvhbv/+myoV45lwUSoDx33DyFdm8sZVx1Kxgp1aNWVTOI9wegBLVXUZgIi8jrvHfWjC6QA8AKCqi0SkuYjUB1oWNq+qLvSGhTF0Y/zbeyCXj+av/bXJTBWObVmH0Se1oX/nBiRWLPpj1jK5Gn8/pwvXvDKT+z5YwN2DOkUocmMiK5wJpzGwKuR5NtAz3zSzcUcrM0SkB9AMSPE5b5FE5ErgSoCmTcv8jfRMhKkqmSu2Mjkzm6lzXJNZk9pVGNPPNZk1qZ1YouUN6NyQy3q14IUZy0lrXpuBXRuFKXJjghPOhFPQIUj+/p8PAk+KSBYwF5gF5Pict0iq+hzwHEB6err1OzWlYs22vbw9M5vJmdn8snkPiRXjGdC5IeempXB089rExR3+kfet/dsxe9U2bn1rDu0bVKdN/eqlGLkxwQtnwskGmoQ8TwHWhE6gqjuASwDEtZEt9x6Jxc1rTKTsPZDLJwtcL7MZSzehCse0rM2ok9rQv1MDqlYqnY9RQnwc44ancubTX3P1KzN5b+TxpbZsY6JBOPfmH4E2ItICWA1cAAwPnUBEagJ7VPUAcDkwXVV3iEix8xoTTqrKzJVek9nstezcn0NKrSpc168NQ1NTStxk5leDpMo8dUF3Lnrhe259ey5PXdDNzleaMiNsCUdVc0RkFPAxrmvzi6o6X0RGeOOfxXVtfklEcnGdCS4ral4AETkbeBpIBj4QkSxVPS1c62HKl7Xb9/56YebyTbupkuCazM5JS6FniyNrMvPruNZ1ufHUtjz88WLSmtbkT8e3CPtrGhMJolr2T2+kp6er3WLaFGbfwVw+nv/7JrOeLWpzTloK/Ts3pFoAzVqHDilXvJTB9CUbmXTVsaQ2rRXxGIwRkUxVTS+15VnCMeWRazLb5jWZrWHn/hwa16zCOWkpDE1NoWmd8DSZlcT2PQc5c9zX5OQqU6/tRZ1qlYIOyZQzpZ1w7IykKVfWbd/HWzOzeSszm2Vek1n/zg04Jy2FY1rUiUiTmV9JiQlMuDCNIRO+ZcykLP59SQ/ioyg+Y0rKEo4p8/YdzOWTBetdk9mSjRxS6NGiNiP6tmJAQE1mfnVqnMTdAzvyl7fn8uTnS7jhFCvyaWJX9H7SjDkCqsqsVa7J7P3Za9i5zzWZjTqpDUNTG9OsTtWgQ/TtgqObkPHLVp7+YgmpTWvSt229oEMy5rBYwjFlyrrt+3hn1momZ67i5427qZwQx4BOrpfZMS2jq8nMLxHh3sGdmL9mO2MmZTH12l6/Fv80JpZYwjExb9/BXD71msy+zmsya16bq/q0on/nBlSvnBB0iEfMFflMY+DTM1yRzxHHUqlCfNBhGVMilnBMTFJVZmdv582MVbw/ew079uXQKKkyI09szdDUFJrXjZ0mM79a1K3Kw+d2YcTEmdw7dSFjB1uRTxNbLOGYmLJ+R16TWTZLN+yickIc/b0ms2NjtMmsJE7v1JArerfgn18vJ715LQZ1axx0SMb4ZgnHRL19B3P5bKFrMpv+k2syO7p5LR4a2pkBnRuWiSazkrj59HZkrdrGrW/NpX3DGhxlRT5NjLALP01UUlXmZG9ncmY2U2avYfvegzRMqszQ1BSGpqXQogw2mZXE+h37OOOpGdSoUoEpo3pFddduE7vswk9Tpm0IaTJbsmEXlSrE0b9TA85Ja8KxrerYhY+e+jUq8/Sw7lz4/HfcMnkO44Z3tyKfJupZwjGB25+Ty2cLNjA5cxXTvCaztGa1eGBIZ87o0pAa5azJzK9jW9XhptPa8dBHi0j7phaX9rIinya6WcIxgVBV5q52TWbvZf3WZHZ131YMTU2hZXK1oEOMCSNOaEnmiq3c/+FCujZJIq1Z7aBDMqZQdg7HRNSGnft412sy+2m9azI7vZOrZXZcq7rWZHYYtu89yFlPz+BAziGmju5FXSvyaUqJncMxMWd/Ti5fLNzAm5nZTPtpI7mHlNSmNbn/bNdkllTFmsyORFKVBJ65MJUhE77lutdn8dKlPS1xm6hkCceEhaoyb/UOJmeu4r3Za9i25yANalTmqj4tGZqWQitrMitVnRonMXZQR255ay5PfPYTN57aNuiQjPkflnBMqdqwcx/vzVrD5MxsFq/fScUKcZzWsQHnpqVwfGtrMgun849u6hX5XEpq01qc2M6KfJroYgnHHLEDOYf4YpG7MPPLxa7JrHvTmtx3difO7NLImswiaOzgTsxbs+PXIp9NaluRTxM9LOGYw6KqzF+zw+tltpqtew5Sv0YlruzTkqGpKbSuZ01mQaicEM+EC1M56+kZjHx1Jm9akU8TRSzhmBLZtGv/r73MFq1zTWandqjPOWkp9G6TbE1mUaB53ao8cl5Xrno5k3veX8B9Z3cOOiRjgBImHBFpoKrrwhWMiU6uyWwDkzOz+WrxBnIOKd2a1OTewZ04q0sjkhKtySzanNaxAVf1ack/pi8jvXktzu6eEnRIxpT4COdDIDUcgZjoM+/XCzNdk1m96pW4rHcLzk1LoXU9KxgZ7W46rS2zVm3jL2/PpUPDJNo2sPfMBKukCcfaS8q4Tbv2816W62W2cO0OKsbHcUpHr8msdV0qxMcFHaLxqUJ8HOOGd+eMp2Zw9cRM3ht1fLmrrG2iS0kTzj/DEoUJ1IGcQ3y52DWZfbnINZl1bVKTsYM7cVaXhtRMrBh0iOYw1atemXHDujP8+e+5efIcnrkw1Yp8msCUKOGo6jPhCsRE3oI1O3gzcxXvZa1hy+4DJFevxGW9WjA0LcXusVKG9GxZh5tPa8sD/13ECzOWc3nvlkGHZMop66VWzmwOaTJbkNdk9msvM2syK6uu7OOKfD7430V0a1KT9OZW5NNEnhXvLAcO5h7iS6+X2Rdek1mXlCTOTUvhrK6NrMmsnNi+9yADx81g38FcPhjd24p8mmIFUrxTRJoBbVT1MxGpAlRQ1Z2lFYQJj4Vr3YWZ785azebdB6hbrRKX9mrB0NQU67FUDiVVSWDChWmc/cw3jH5tFi9fZkU+TWQVm3BE5ArgSqA20ApIAZ4F+oU3NHM4tuw+wHtZ7sLM+Wt2kBAvvzaZ9WmTbE1m5VyHRjUYO7gTN0+ew2OfLuam09oFHZIpR/wc4YwEegDfA6jqEhGxqoBR5GDuIaYt3sibmav4YtEGDuYqnRsncffAjgzs2ohaVa3JzPzmvPQmZP6ylfFf/kxq01r0a18/6JBMOeEn4exX1QN5XSlFpALg68SPiJwOPAnEA8+r6oP5xtcCXsQdOe0DLlXVeUXNKyK1gUlAc+AX4DxV3eonnrJm0bodTM7I5t2s1WzadYC61Sryp+OaMzQthXYNagQdnolidw/qyNzV27l+UhYfjO5tRT5NRBTbaUBE/g5sA/4IXAtcAyxQ1duKmS8e+Ak4BcgGfgSGqeqCkGkeBnap6t0i0g4Yr6r9iprXi2eLqj4oIrcCtVT1lqJiKUudBrbsPsCUrNVMnpnNvNWuyaxfu/qcm55Cn6OSSbAmM+PTys17OOPpr2lWJ5HJI46jcoIV+TS/F0SngVuAy4G5wFW48jbP+5ivB7BUVZcBiMjrwCBgQcg0HYAHAFR1kYg0F5H6QMsi5h0E9PXm/w/wlRdjmXUw9xDTf9rI5MxsPlu4noO5SqfGNbjrrA4M7NaY2tZkZg5D0zqJPHZeN654KYO731/AA0OsyKcJryITjojEAXNUtRMlrzLQGFgV8jwb6JlvmtnAEGCGiPQAmuE6JRQ1b31VXQugqmsLO58kIlfiOjvQtGnTEoYeHRav28nkzFW8M2sNm3btp261ilx8rGsya9/QmszMkTulQ32u7tuKCV/9THqzWgxNsyKfJnyKTDiqekhEZotIU1VdWcJlF9TfMn/73YPAkyKShTuCmgXk+Jy3SKr6HPAcuCa1kswbpK27DzBltrswc+7q7VSIE/q1r8e5aU04oa01mZnSd+MpR5G1chu3vTuXDo1q2I8ZEzZ+mtQaAvNF5Adgd95AVR1YzHzZQJOQ5ynAmtAJVHUHcAmAuF4Jy71HYhHzrheRht7RTUNgg491iGo5uYeYvsRrMluwgQO5h+jYqAZ3ntWBgV0bUccu0DNhVCE+jqeGdeeMp77m6omZTLm2FzWsyKcJAz8J5+7DXPaPQBsRaQGsBi4AhodOICI1gT2qegB3nmi6qu4QkaLmnQJcjDs6uhh47zDjC9xP63cyOTObt2euZtOu/dSpWpE/HNuMoakpdGhkvzJN5CRXr8S44akM++d33PzmHCZcZEU+TekrNuGo6jTvRP7R3qAfVLXYowpVzRGRUcDHuK7NL6rqfBEZ4Y1/FmgPvCQiubgOAZcVNa+36AeBN0TkMmAlcK7/1Q3etj0HeN9rMpud7ZrMTmpXj3PSUjixXT1rMjOB6dGiNree3o77PlzI818v54o+VuTTlC4/3aLPAx7G9QYToDdwk6pODnt0pSTobtE5uYf4eskmJmdm8+mC9RzIPUSHhjU4Jy2FQd2sycxED1Xl6okz+XThel674hh6tLAin+VZaXeL9pNwZgOn5B3ViEgy8Jmqdi2tIMItqISzJK/JbNZqNu7cT+2qFRnUrRHnpKXQsVFSxOMxxo8d+w4y8OkZ7DmQy9TRvahXvXLQIZmABHEdTly+JrTNgLX7FGL7noNMmeM1ma3aRoU44cS8JrO29ahYwTadiW41Kicw4aLfinxOvKyn1eAzpcJPwvlIRD4GXvOenw/8N3whxZ6c3EN8vdRrMpvvmszaNajO387swKBujawMvIk57RvW4N7Bnfnzm7N59NOfuOV0K/JpjpyfTgM3icgQoBfuHM5zqvpO2COLAUs37OTNzGzembmaDTv3UysxgeE9m3JuujWZmdh3TloKmSu2MOErV+TzlA5W5NMcGT+3J2gBfKiqb3vPq4hIc1X9JdzBRaPtew7yvtdklrVqG/FxwoltXZPZSe2sycyULXee5Yp83vBGFh9c25umdazIpzl8fjoNZADHedfKICIVgW9U9egiZ4wiR9ppIPeQ8rV3YeYnC9ZzIMc1mbleZo1Jrm5NZqbsWrVlD2c89TVNaify1tVW5LM8CaLTQIW8ZAPg3aqgXFSL/HnjLu/CzGzW79hPzcQEhvdo6vUyq2EXxplyoUntRB4/vxuX/SeDu6bM58GhXYIOycQoPwlno4gMVNUpACIyCNgU3rCiw/NfL+eNjFWc2DaZuwe6CzMrVbBfd6b86de+PiNPbMX4L38mrVktzk1vUvxMxuTjp0mtFfAK0AjXaWAV8EdVXRr+8ErH4Taprd2+l/g4sesQjME1Lf/hhe/JXLGVd6453sovlQOl3aRW7BluVf1ZVY/B3bumg6oeF0vJ5kg0TKpiycYYT3yc8NSw7tRMTODqVzLZvvdg0CGZGFNowhGRs0SkWcigG3D3rZni9VwzxpQzdatVYvzwVFZv3ctNb86muBYSY0IVdYRzH7ARQETOBC4CLsVVa342/KEZY6JRevPa3Nq/HZ8sWM9z05cFHY6JIUUlHFXVPd7/Q4AXVDVTVZ8HksMfmjEmWl3WqwUDOjfg7x8v5vtlm4MOx8SIohKOiEg17zbT/YDPQ8bZiQ1jyjER4aGhXWhWO5FRr81iw459QYdkYkBRCecJIAvIABaqagaAiHQH1oY9MmNMVKteOYFnLkpl576DjHptFjm5h4IOyUS5QhOOqr4InIC7KdqAkFHr8G4LbYwp39o1qMH9Z3fmh+VbePiTxUGHY6JckRd+qupq3C2eQ4fZ0Y0x5ldDUlPIWLGVf0xbRlrTWpzasUHQIZkoZZUmjTFH7I4zO9C5cRI3vjmbFZt3Bx2OiVKWcIwxR6xyQjzPXJhKnAgjJs5k38HcoEMyUajYhCMij4hIx0gEY4yJXU1qJ/LE+d1YuHYHd7w3L+hwTBTyc4SzCHhORL4XkREiYncWM8YU6MR29bj2pNa8kZHNpB9XBh2OiTJ+aqk9r6rHA38EmgNzRORVETkx3MEZY2LPmJOPolfruvztvfnMW7096HBMFPF1DkdE4oF23mMTMBu4QUReD2NsxpgYFB8nPHlBN2onVuSaV2ZakU/zKz/ncB4DFuOuxblfVdNU9SFVPQvoHu4AjTGxp061Soy/MJU12/Zy4xuzOXTIinwaf0c484AuqnqVqv6Qb1yPMMRkjCkD0prV4q8D2vPZwvX8w4p8GvwlnK1AQt4TEakpIoMBVNUaaI0xhbrk+Oac0aUhD3+8iP/72Yp8lnd+Es6doYlFVbcBd4YtImNMmZFX5LN53apca0U+yz0/CaegaYosiWOMMXmqVarAsxelsXt/DqNencVBK/JZbvlJOBki8piItBKRliLyOJAZ7sCMMWXHUfWr88CQzvzwyxYe/tiKfJZXfhLOtcABYBLwJrAPGBnOoIwxZc/g7o35wzHNeG76Mj6aty7ocEwA/Fz4uVtVb1XVdK9L9F9U1Vd1PhE5XUQWi8hSEbm1gPFJIvK+iMwWkfkicknIuOtEZJ43fEzI8K4i8n8iMtebt4bPdTXGBOz2M9vTtUlNbnpzNss3WZHP8sbPdTjJIvKwiHwoIl/kPXzMFw+MB/oDHYBhItIh32QjgQWq2hXoCzwqIhVFpBNwBa7bdVfgTBFp483zPHCrqnYG3gFu8rWmxpjAVaoQz/jh3YmPF66emMneA1bkszzx06T2Cq6eWgvgbuAX4Ecf8/UAlqrqMlU9ALwODMo3jQLVRUSAasAWIAdoD3ynqntUNQeYBpztzdMWmO79/ykw1EcsxpgokVLLFflcvH4nf3tvHqp2UWh54Sfh1FHVF4CDqjpNVS8FjvExX2NgVcjzbG9YqHG45LIGmAtcp6qHcBeb9hGROiKSiKty0MSbZx4w0Pv/3JDhvyMiV4pIhohkbNy40Ue4xphI6du2Htee1IbJmdlM+nFV8TOYMsFPwskrhLRWRM4Qke5Aio/5pIBh+X/KnAZkAY2AbsA4EamhqguBh3BHMB/harflePNcCowUkUygOq5Dw/++kOpz3nmn9OTkZB/hGmMi6bp+bejdpi53TLEin+WFn4Rzr3dLghuBP+POoVzvY75sfn/0kYI7kgl1CfC2OkuB5bgCoajqC6qaqqp9cE1tS7zhi1T1VFVNA14DfvYRizEmyrgin92pU7UiIyZmsn2PFfks64pMON6J/zaqul1V56nqiV5PtSk+lv0j0EZEWohIReACIP98K4F+3mvVx52fWeY9r+f9bQoMwSWX0OFxwO3As77W1BgTdWpXrcj4C1NZv2MfN7yRZUU+y7giE46q5vLb+ZIS8U72jwI+BhYCb6jqfO8mbiO8ycYCx4nIXOBz4BZV3eSNe0tEFgDvAyNVdas3fJiI/ITryLAG+NfhxGeMiQ6pTWtx24D2fL5oAxOmWYNFWSbF9RARkfuAJNyFn792nFfVmeENrfSkp6drRkZG0GEYYwqhqox+PYsP5qxh4uU9Oa5V3aBDMoCIZKpqemktz09NtOO8v/eEDFPgpNIKwhhTvokIDw7pzII12xn92iymXtubBkmVgw7LlDI/lQZOLOBhycYYU6qqekU+9xzIZdSrM63IZxlU7BGOiNxR0HBVvaeg4cYYc7jaeEU+r3s9i4f+u4jbz8xfnMTEMj/doneHPHJxpWqahzEmY0w5NqhbYy4+thnPz1jOf+euDTocU4qKPcJR1UdDn4vII/xv92ZjjCk1t53RgdnZ27lp8hzaNqhOy+RqQYdkSoGfI5z8EoGWpR2IMcbkqVghjvEXppIQL1zzykwr8llG+KkWPVdE5niP+cBi4Mnwh2aMKc8a16zCkxd0Z/H6ndz2zlwr8lkG+OkWfWbI/znAeu+iTmOMCas+RyVzXb82PPHZEtKa1+LCns2CDskcAT9Nag2BLaq6QlVXA5VFpGeY4zLGGABGn9SGPkclc/eUBczJ3hZ0OOYI+Ek4E4BdIc/3eMOMMSbs4uKEJ87vRt1qFbl64ky27SmwQLyJAX4SjmhI46l3vxo/TXHGGFMqaletyDMXpbFh5z6un2RFPmOVn4SzTERGi0iC97gOr6KzMcZESrcmNfnbmR34cvFGnvlqadDhmMPgJ+GMwNVTW427x01P4MpwBmWMMQX5wzHNGNi1EY99+hPfLN1U/AwmqvippbZBVS9Q1XqqWl9Vh6vqhkgEZ4wxoUSEB4Z0pmVyNUa/Not12/cFHZIpAT/X4fxHRGqGPK8lIi+GNSpjjClEXpHPfQdzGWlFPmOKnya1Lqq6Le+JdyO07mGLyBhjitG6XjUeOqcLmSu28sCHi4IOx/jkJ+HEiUitvCciUhvrpWaMCdiZXRrxp+Oa8+I3y/lgjhX5jAV+EsejwLciMtl7fi5wf/hCMsYYf/46oD1zsrdx8+TZtGtYnVZW5DOq+ek08BIwFFgPbACGeMOMMSZQeUU+KyXEc/XETPYcsKpb0cxXtWhVXaCq44APgSEiMi+8YRljjD8Nk6rw5AXdWLJhF39924p8RjM/vdQaisgYEfkBmA/EA8PCHpkxxvjUu00y1598FO9mrWHi9yuDDscUotCEIyJXiMgXwDSgLnA5sFZV71bVuZEK0Bhj/Bh1Ymv6tk1m7PsLmL1qW9DhmAIUdYQzHnc0M1xVb1fVOYAdqxpjolJcnPD4ed1Irl6Ja16ZydbdVuQz2hSVcBoBrwOPichiERkLJEQmLGOMKblaVSvyzIWpbNy5n+vfsCKf0abQhKOqm1R1gqr2AfoB24ENIrJQRKxbtDEmKnVtUpO/ndWBrxZvZNyXVuQzmvjtpZatqo+oahowGNgf1qiMMeYIXNSzKWd3b8zjn/3E10s2Bh2O8fhKOKFUdbGq3h2OYIwxpjSICPed3Yk29apx3etZrNm2N+iQDIeRcIwxJhYkVqzAhIvSOJBziJGvzuRAjhX5DJolHGNMmdUquRp/P6cLs1Zu4/4PFwYdTrlXaC01EUktakZVnVn64RhjTOka0Lkhlx7fghe/WU5as1qc1bVR0CGVW0Ud4TzqPcYD3wPPAf/0/n/Kz8JF5HSvS/VSEbm1gPFJIvK+iMwWkfkicknIuOtEZJ43fEzI8G4i8p2IZIlIhoj08LWmxphy6y8D2pHWrBa3vjWHpRt2BR1OuVVUt+gTVfVEYAWQqqrpXi+17kCxfQ1FJB6XrPoDHYBhItIh32QjgQWq2hXoCzwqIhVFpBNwBdAD6AqcKSJtvHn+Dtytqt2AO7znxhhTqIT4OMYPT6WyV+Rz934r8hkEP+dw2oWWslHVeUA3H/P1AJaq6jJVPYC7iHRQvmkUqC4iAlQDtgA5QHvgO1Xdo6o5uPI6Z4fMU8P7PwlY4yMWY0w51yCpMk8N687PG3fxFyvyGQg/CWehiDwvIn1F5AQR+Sfg5+xbY2BVyPNsb1iocbjksgaYC1ynqoeAeUAfEakjIonAAKCJN88Y4GERWQU8AvyloBcXkSu9JreMjRutH74xBo5vXZcbTjmKKbPX8PJ3K4IOp9zxk3AuwVWJvg73Zb/AG1YcKWBY/p8UpwFZuDI63YBxIlJDVRcCDwGfAh8Bs3FHPgBXA9erahPgeuCFgl5cVZ/zmgHTk5OTfYRrjCkPrunbmpPa1WPs1AXMWrk16HDKFT83YNsHPAvcqqpnq+rj3rDiZPPbUQlACv/b/HUJ8LY6S4HlQDvvdV9Q1VSvtM4WYIk3z8XA297/b+Ka7owxxpe4OOGx87pSv0ZlRr4yky1W5DNi/NwPZyDuKOQj73k3EZniY9k/Am1EpIWIVAQuAPLPtxJXpw0RqQ+0BZZ5z+t5f5sCQ4DXvHnWACd4/5/Eb4nIGGN8qZnoinxu2nWAMZOyyLUinxHhp0ntTtxRxDYAVc0Cmhc3k3eyfxTwMe6czxuqOl9ERojICG+yscBxIjIX+By4RVU3eePeEpEFwPvASFXNO/a9AtebbTZwP3Clj3Uwxpjf6ZJSk7sGdmT6Txt5+gv73RoJhV74GSJHVbe7jmQlo6of4m5LHTrs2ZD/1wCnFjJv70KGzwDSShyMMcbkM6xHEzJWbOHJz5fQvWktTjjKzveGk58jnHkiMhyIF5E2IvI08G2Y4zLGmLATEe4b3Jm29asz5vVZrLYin2HlJ+FcC3TE3ZLgVdx9ccaEMSZjjImYKhXjeebCVA7mKiNfsSKf4eSnl9oeVb1NVY/2Hrf77KVmjDExoWVyNR45twtZq7Zx3wcLgg6nzPLTS+1TEakZ8ryWiHwc1qiMMSbCTu/UkMt7teA//7eCKbOtgEk4+GlSq6uq2/KeeL3F6oUtImOMCcgt/dtxdHNX5HPJ+p1Bh1Pm+Ek4h7xrYQAQkWb8b8UAY4yJeQnxcYwbnkpixXhGTMxklxX5LFV+Es5twAwReVlEXgamU0j9MmOMiXX1a7gin8s37ebWt+ZYkc9S5KfTwEdAKjAJeANIU1U7h2OMKbOOa1WXG09ty9Q5a/nPt78EHU6Z4fcW05Vw9cy2Ax1EpE/4QjLGmOBdfUIr+rWrx30fLmSmFfksFX56qT0EfINrWrvJe/w5zHEZY0ygXJHPbjRIckU+N+/aH3RIMc/PEc5goK2qnqGqZ3mPgWGOyxhjApeUmMCEC9PYvNuKfJYGPwlnGZAQ7kCMMSYadWqcxD0DO/L1kk08+bkV+TwSfop37gGyRORzXHkbAFR1dNiiMsaYKHL+0U3IWLGVp79YQmrTmvRta5ciHg4/CWcK/3sfG2OMKTdEhLGDOjFv9XbGTMpi6rW9SKmVGHRYMUfKQx/z9PR0zcjICDoMY0yM+2XTbs56egYtk6vyxohjqVQhPuiQwkpEMlU1vbSW56eXWhsRmSwiC0RkWd6jtAIwxphY0bxuVR4+tyuzs7dz79SFQYcTc/x0GvgXMAHIAU4EXgJeDmdQxhgTrU7v1IAr+7Tk5e9W8F7W6qDDiSl+Ek4VVf0c1/y2QlXvAk4Kb1jGGBO9bj6tLT2a1+bWt+bykxX59M1PwtknInHAEhEZJSJnY9WijTHlWIX4OMYN707VShWsyGcJ+Ek4Y4BEYDSQBvwBuDiMMRljTNSrV6MyTw/rzi+bdnPLZCvy6Yef4p0/quouVc1W1UtUdYiqfheJ4IwxJpod26oON53Wjg/mruVf3/wSdDhRr9DrcETkfYq4742VtzHGGBhxQksyV2zl/g8X0rVJEmnNagcdUtQq9DocETmhqBlVdVpYIgoDuw7HGBNO2/ce5KynZ3Ag5xBTR/eibrVKQYdUKkr7OpxCj3BiKaEYY0yQkqokMOGiVIY88y3XvT6Lly7tSXycBB1W1LELP40xphR0bJTE2EGd+GbpZp747Kegw4lKduGnMcaUkvOObsJ56Sk8/cVSvly0Iehwoo5d+GmMMaXonkGd6NCwBmMmZbFqy56gw4kqduGnMcaUosoJ8Uy4KJVDqox8dSb7c3KDDilq2IWfxhhTyprVqcqj53ZlTvZ27nl/QdDhRI1i74ejqj96/+4CLglvOMYYUzac2rEBV53Qkn9MW0Zas1oMSU0JOqTAFXqEIyJ1ReROERktItVEZIKIzBOR90SktZ+Fi8jpIrJYRJaKyK0FjE8SkfdFZLaIzBeRS0LGXee93nwRGRMyfJKIZHmPX0Qkq2SrbIwxkXHTqW3p2aI2f31nLovW7Qg6nMAV1aT2KlAJaAP8ACwDzgGmAs8Xt2ARiQfGA/2BDsAwEemQb7KRwAJV7Qr0BR4VkYoi0gm4AugBdAXOFJE2AKp6vqp2U9VuwFvA2/5W1RhjIqtCfBxPD+9O9coJXD1xJjv3HQw6pEAVlXDqq+pfceduqqnqw6q6SFX/CdT0sewewFJVXaaqB4DXgUH5plGguogIUA3Ygut+3R74TlX3qGoOMA04O3RGb57zgNd8xGKMMYGoV70y44Z1Z+WWPdxczot8FpVwcgHUbZ1N+cYd8rHsxsCqkOfZ3rBQ43DJZQ0wF7hOVQ8B84A+IlJHRBKBAUCTfPP2Btar6pKCXlxErhSRDBHJ2Lhxo49wjTEmPHq2rMMtp7flv/PW8cKM5UGHE5iiOg20FJEpgIT8j/e8hY9lF1TXIX9qPw3Iwl3X0wr4VES+VtWFIvIQ8Cmus8Js3JFPqGEUcXSjqs8Bz4GrpeYjXmOMCZsrersinw/+dxHdmtQkvXn5K/JZ1BHOIOBR4JGQ//OeD/ax7Gx+f1SSgjuSCXUJ8LY6S4HlQDsAVX1BVVNVtQ+uqe3XIxkRqQAMASb5iMMYYwInIjx8bldSalVh5Ksz2bRrf9AhRVyhCUdVpxX18LHsH4E2ItJCRCoCFwBT8k2zEugHICL1gba4zgmISD3vb1Nccgk9mjkZWKSq2f5W0xhjglejcgLPXJjGtj0HGf3aLHIPla/GFz8Xfh4W72T/KOBjYCHwhqrOF5ERIjLCm2wscJyIzAU+B25R1bzzRW+JyALgfWCkqm4NWfwFWGcBY0wM6tCoBvcO7sS3P2/msU8XBx1ORBV74eeRUNUPgQ/zDXs25P81wKmFzNu7iOX+qZRCNMaYiDs3vQmZK7Yy/sufSW1ai37t6wcdUkSU6AhHRBqEKxBjjClP7hrYkY6NanB9OSryWdImtQ+Ln8QYY0xxKifEM+HCNACufiWTfQfLfpHPkiYcu4WdMcaUkqZ1EnnsvG7MW72Du8tBkc+SJpx/hiUKY4wpp07uUJ+r+7bitR9WMjmzbHe8LVHCUdVnwhWIMcaUVzeechTHtqzDbe/MZeHaslvkM2zdoo0xxvhTIT6Op4Z1J6lKAldPzGRHGS3yaQnHGGOiQHL1Soy/MJVVW/dy85tls8inJRxjjIkSRzevzV/6t+Oj+et4/uuyV+TTEo4xxkSRy3q1oH+nBjz40SJ+WL4l6HBKlSUcY4yJIiLC38/pQtPaiYx6dSYbdu4LOqRSYwnHGGOiTPXKCUy4KJUd+1yRz5xcP7cgi36WcIwxJgq1a1CD+wZ35rtlW3j005+CDqdUWMIxxpgoNTQthWE9mjLhq5/5dMH6oMM5YpZwjDEmit15Vgc6Na7BDW9ksXJzbBf5tIRjjDFRLK/IZ5xIzBf5tIRjjDFRrkntRB4/vyvz1+zgrinzgw7nsFnCMcaYGHBSu/qMPLEVr/+4ijcyVgUdzmGxhGOMMTHihlPaclyrOvzt3XnMX7M96HBKzBKOMcbEiPg44alh3amZmMA1r8xk+97YKvJpCccYY2JI3WqVeObCVFZv3ctNb86OqSKflnCMMSbGpDWrzV8GtOeTBet5bvqyoMPxzRKOMcbEoEuPb84ZnRvy948X8/2yzUGH44slHGOMiUEiwoNDO9OsdiKjXpvFhh3RX+TTEo4xxsQoV+QzjV37chgVA0U+LeEYY0wMa9ugOvcP6cQPy7fw8CeLgw6nSJZwjDEmxp3dPYULezblH9OW8cn8dUGHUyhLOMYYUwbccVYHuqQkceObs1mxeXfQ4RTIEo4xxpQBlSrEM354KnEijJg4MyqLfFrCMcaYMqJJ7USeOL8bC9fu4I735gUdzv+whGOMMWXIie3qce1JrXkjI5tJP64MOpzfCWvCEZHTRWSxiCwVkVsLGJ8kIu+LyGwRmS8il4SMu05E5nnDx+Sb71pvufNF5O/hXAdjjIk1Y04+il6t6/K39+Yzb3X0FPkMW8IRkXhgPNAf6AAME5EO+SYbCSxQ1a5AX+BREakoIp2AK4AeQFfgTBFp4y33RGAQ0EVVOwKPhGsdjDEmFsXHCU9e0I2eLWpTqUL0NGSFM5IewFJVXaaqB4DXcYkilALVRUSAasAWIAdoD3ynqntUNQeYBpztzXM18KCq7gdQ1Q1hXAdjjIlJdapV4uXLetKmfvWgQ/lVOBNOYyD0LkHZ3rBQ43DJZQ0wF7hOVQ8B84A+IlJHRBKBAUATb56jgN4i8r2ITBORowt6cRG5UkQyRCRj48aNpbdWxhhjDks4E44UMCx/He3TgCygEdANGCciNVR1IfAQ8CnwETAbd+QDUAGoBRwD3AS84R0h/f6FVJ9T1XRVTU9OTj7ytTHGGHNEwplwsvntqAQgBXckE+oS4G11lgLLgXYAqvqCqqaqah9cU9uSkOXmzfMDcAioG8b1MMYYUwrCmXB+BNqISAsRqQhcAEzJN81KoB+AiNQH2gLLvOf1vL9NgSHAa9487wIneeOOAioCm8K4HsYYY0pBhXAtWFVzRGQU8DEQD7yoqvNFZIQ3/llgLPBvEZmLa4K7RVXzksdbIlIHOAiMVNWt3vAXgRdFZB5wALhYY+mWd8YYU05JefiuTk9P14yMjKDDMMaYmCIimaqaXlrLi54O2sYYY8o0SzjGGGMiolw0qYnIRmDFYc5eF+uUYMLH9i8TbkeyjzVT1VK7rqRcJJwjISIZpdmGaUwo279MuEXTPmZNasYYYyLCEo4xxpiIsIRTvOeCDsCUabZ/mXCLmn3MzuEYY4yJCDvCMcYYExGWcIwxxkSEJRyfRGRwAXcsNaZQIrLrCOZ9xbuN+jwReVFEEkozNlO+iUg3ERkQ6de1hOPfYNytso2JhFdwt+roDFQBLg82HFNWiEgF3P3HIp5wynWnARF5F3fPnsrAk6r6nIjsUtVq3vhzgDNxvTymAtu9x1CgOvAskAj8DFwaUtHaGPL2Je8GgX8H+uNuQnivqk4SkTjcXW9PwN0LKg5XVX1yvuVcD9RV1dsiuwYmmolIVeAN3L3G4nHV97cDT+AqC8wEWqrqmSJyF+5Gl829cb1wP2RWAw+o6qRIxBy22xPEiEtVdYuIVAF+FJG3CppIVb8VkSnA1LwvAxGZA1yrqtNE5B7gTmBMpAI3MWUI7hdlV1yZkR9FZDpwPO4LoDNQD1iIu/3Gr7ymtD8A10UuXBMjTgfWqOoZACKSBMzD3S9sKZA/iaQBvVR1r4j8CUhX1VERjLfcN6mNFpHZwHe4I502fmby3tiaqjrNG/QfoE94QjRlQC/gNVXNVdX1wDTgaG/4m6p6SFXXAV8WMO8zwHRV/Tpy4ZoYMRc4WUQeEpHeQAtguaou8e4RNjHf9FNUdW/EowxRbo9wRKQvcDJwrKruEZGvcE1roW2MlSMfmSmDpITD3UiRO4Fk4KpSj8jEPFX9SUTScOdiHgA+4fffX/ntjkhgRSjPRzhJwFYv2bQDjvGGrxeR9l77+tkh0+/EnbdBVbcDW71fFeCaPKZhTMGmA+eLSLyIJOOOhn8AZgBDRSTOu8V637wZRORy4DRgmKoeCiBmE+VEpBGwR1UnAo8AxwEtRKSVN8mwImb/9fsskspzwvkIqOCdixmLa1YDuBXXQeALYG3I9K8DN4nILO8NvRh42Ju/G3BPpAI3MecdYA4wG7df3ew1ob0FZOPa3f8BfI876QuuQ0p94P9EJEtE7oh41CbadQZ+EJEs4DbgduBK4AMRmUHRt2T5Eujg7Vvnhz1ST7nupWZM0ESkmqruEpE6uKOe471kZMwR8U4b/FlVzww4lF+V23M4xkSJqSJSE6gIjLVkY8oyO8IxxhgTEeX5HI4xxpgIsoRjjDEmIizhGGOMiQhLOMZEmIj0FZGpQcdhTKRZwjEmSnhVfI0ps2wHNyZABVTxHR5kPMaEkyUcY4L3axXfoAMxJpysSc2Y4AVexdeYSLCEY0wYiMhIr05VlldksSgFVvEVkfvyllH6ERoTedakZkwYqOp4YPwRLuM2XFFGY8oEO8IxxhgTEVZLzRhjTETYEY4xxpiIsIRjjDEmIizhGGOMiQhLOMYYYyLCEo4xxpiIsIRjjDEmIizhGGOMiYj/BzDARoJr6BoHAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mejor puntuación: 0.9011208974378893 con valor = log2\n"
     ]
    }
   ],
   "source": [
    "values = [\"auto\", \"log2\", \"sqrt\"]\n",
    "print('\\n--- Alterando max_features ---')\n",
    "\n",
    "scores = []\n",
    "for v in range(len(values)):\n",
    "    SeqTree = SequentialDecisionTreeAlgorithm(ntree = 300, sample_size = 0.65, max_depth = 10, lr = 0.1, max_features=values[v])\n",
    "    trees, score = SeqTree.start(attributes_cols = attr_cols, objetive_col = obj_col)\n",
    "    scores.append(score)\n",
    "    \n",
    "# Realizamos una gráfica para determinar cómo afecta ntree a la puntuación\n",
    "plt.plot(values, scores)\n",
    "plt.xlabel('- lr -')\n",
    "plt.ylabel('- Balanced Accuracy Score -')\n",
    "plt.title('Alteración del hiperparámetro de aprendizaje max_features')\n",
    "plt.show()\n",
    "\n",
    "m = np.argmax(scores)\n",
    "print('Mejor puntuación: '+str(scores[m])+' con valor = '+str(values[m]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "fcc2e46d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--- Alterando min_samples_leaf [1 - 20] ---\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mejor puntuación: 0.9009756308274788 con valor = 1\n"
     ]
    }
   ],
   "source": [
    "values = [1, 5, 10, 20]\n",
    "print('\\n--- Alterando min_samples_leaf ['+str(np.min(values))+' - '+str(np.max(values))+'] ---')\n",
    "\n",
    "scores = []\n",
    "for v in range(len(values)):\n",
    "    SeqTree = SequentialDecisionTreeAlgorithm(ntree = 300, sample_size = 0.65, max_depth = 10, lr = 0.1, min_samples_leaf=values[v])\n",
    "    trees, score = SeqTree.start(attributes_cols = attr_cols, objetive_col = obj_col)\n",
    "    scores.append(score)\n",
    "    \n",
    "# Realizamos una gráfica para determinar cómo afecta ntree a la puntuación\n",
    "plt.plot(values, scores)\n",
    "plt.xlabel('- lr -')\n",
    "plt.ylabel('- Balanced Accuracy Score -')\n",
    "plt.title('Alteración del hiperparámetro de aprendizaje min_samples_leaf')\n",
    "plt.show()\n",
    "\n",
    "m = np.argmax(scores)\n",
    "print('Mejor puntuación: '+str(scores[m])+' con valor = '+str(values[m]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7c8f8ad8",
   "metadata": {},
   "source": [
    "## 2. Ejemplo con mejor configuración\n",
    "Ajustamos con:\n",
    "- ntree = 350\n",
    "- sample_size = 0.25\n",
    "- max_depth = 10\n",
    "- lr = 0.25\n",
    "\n",
    "La alteración de ntree supone una función logarítmica, y después de muchas pruebas para aborrecer la aletoriedad, supone que en 250 ya puede dar el resultado más óptimo con la menor cantidad de árboles que suponen más ejecución.\n",
    "\n",
    "sample_size no nos ha ofrecido mejores valores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "1d490171",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Score final: 0.9105984662270998\n"
     ]
    }
   ],
   "source": [
    "SeqTree = SequentialDecisionTreeAlgorithm(ntree = 350, sample_size = 0.25, max_depth = 10, lr = 0.25)\n",
    "trees, score = SeqTree.start(attributes_cols = attr_cols, objetive_col = obj_col)\n",
    "\n",
    "print('Score final: '+str(score))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "08a6f4b2",
   "metadata": {},
   "source": [
    "## 3. Ejemplo con knn\n",
    "### Alterando nmodels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "018fcea0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--- Alterando nmodels [1 - 25] ---\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mejor puntuación: 0.8560309846527981 con valor = 20\n"
     ]
    }
   ],
   "source": [
    "values = [1, 5, 10, 15, 20, 25]\n",
    "print('\\n--- Alterando nmodels ['+str(np.min(values))+' - '+str(np.max(values))+'] ---')\n",
    "scores = []\n",
    "\n",
    "for v in range(len(values)):\n",
    "    SeqModel = SequentialModelAlgorithm(nmodels = values[v], method = \"knn\")\n",
    "    models, score = SeqModel.start(attributes_cols = attr_cols, objetive_col = obj_col)\n",
    "    scores.append(score)\n",
    "    \n",
    "# Realizamos una gráfica para determinar cómo afecta ntree a la puntuación\n",
    "plt.plot(values, scores)\n",
    "plt.xlabel('- nmodels -')\n",
    "plt.ylabel('- Balanced Accuracy Score -')\n",
    "plt.title('Alteración del hiperparámetro nmodels')\n",
    "plt.show()\n",
    "m = np.argmax(scores)\n",
    "print('Mejor puntuación: '+str(scores[m])+' con valor = '+str(values[m]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0e1c51bb",
   "metadata": {},
   "source": [
    "### Alterando n_neighbors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "d43fb8dd",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--- Alterando n_neighbors [1 - 5] ---\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEWCAYAAAB8LwAVAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAAwxElEQVR4nO3dd5wV1fnH8c93G7tLl94RpIhKc4OAxlgTu8YExRIVYyExEUuiJvGnxhI1EVs0KtYYC2BL1NiiRo2K4qIUEVAEKdJ7Z1n2+f0xs3q93N2dhb07W57363Vfe2fmzMwzc8uzc+bcc2RmOOecc8ky4g7AOedczeQJwjnnXEqeIJxzzqXkCcI551xKniCcc86l5AnCOedcSp4gaiFJj0i6vgbE0VnSBkmZSfPzJL0n6Ygq3Ndbks6JWNYk7VHZ7ZR1PLWBpJcl/SLuOGoDSddIeixi2cjvu7rIE0QNFr45V0tqUE6ZgyQtrM64SpnZfDNrZGbbkxbdB9xiZq/EEdfOKud4ajRJZwArzeyeNO6jzKTr6i5PEDWUpK7A9wEDjkvjfrKqeptmdoaZPVfV262tFKjSz1rSVU4j4Pyq3H5lpeN95OLnCaLmOgP4AHgEODNVAUkNgZeB9mHVyAZJ7SVlSLpC0peSVkoaL2m3cJ2u4X+DP5c0H3gznP+UpCWS1kp6R9JeCfvJkzRa0rxw+bvhvNJtZYXl2kt6XtIqSbMlnZuwjWvCOB6VtF7SdEkFZR28pMMlzQz3dxegpOVnS5oRXmG9KqlLJc5tl7AKbL2k1yS1TDo3pcfzlqQbJU0M4/hX6XkMlw+W9L6kNZKmSDooYdlbkm6Q9B6wCegWYXvlvQaPSLpH0kuSNgIHSzpa0ifATcBMSdcklC89lhGSFoTnaaSk70maGsZ8V5RzKumdsMiU8D12ssIrV0mXS1oCPCypgaTbJS0KH7erjKtfSWeF76Nbwv3NlXRk0vm7Pjy/GyS9IKmFpMclrZP0kYJ/okrLDw3nrQ3/Dk1Ytrukt8PX+z9Ay6RYynwdk8rtEW5nraQVksalKlenmJk/auADmA38EtgX2Aa0SVj2CHB9+PwgYGHSuhcRJJeOQAOCKp8nw2VdCa5KHgUaAnnh/LOBxmH524HJCdu7G3gL6ABkAkPDcqXbygrLvQ38DcgF+gPLgUPDZdcAW4Cjwm3cCHxQxrG3BNYBPwWygYuBYuCccPkJ4fnZE8gCrgTeT1jfgD3K2PZbwJdATyAvnL4p6dxkJZT9Gtg7PFfPAI+FyzoAK8PjyQAOD6dbJaw7H9grjDG7vO1FeA0eAdYC+4f7ywUOAfqG032BZcAJScdyb1j2h+H5/yfQOox/GfCDnTmnBO+7YuDmMN484FqC911roBXwPnBdGa/DWQTv63MJ3g+/ABYBSjh/s4HuQFPgM+Bz4LAwvkeBh8OyuwGrgZ+Fy04Jp1uEyycAt4ZxHgisr+TrWPq+exL4Q8L5PyDu74m0fw/FHYA/UrwocED44WkZTs8ELk5Y/gjlJ4gZhF/M4XS7cHtZCV8c3crZf7OwTNPww7AZ6JeiXOm2soBOwHagccLyG4FHwufXAK8nLOsDbC5j/2eQkDwIrh4WJnxQXwZ+nrA8g+C/9C7hdEUJ4sqE6V8CryQfT0LZm5JiLiL4Qrsc+EfStl8FzkxY99oU+065vfJeg4TX/NEK3je3A7clHUuHhOUrgZMTpp8BLtqZc0rwvisCchPmfQkclTD9I+CrMmI9C5idMJ0f7qNtwrn6Q8Ly0cDLCdPHEiZQgsQwMWn7E8J9dCZIZA0Tlj3BtwkiyutY+r57FBgDdNyVz3dtengVU810JvCama0Ip5+gjGqmMnQBngsvmdcQJIztQJuEMgtKn0jKlHSTgiqpdcBX4aKW4SOX4MNfnvbAKjNbnzBvHsF/aKWWJDzfBOQqdd11+8T4LPh0LkhY3gW4I+H4VhEkkcR9lSc5jkbllE3c7zyCK4GWYQzDSmMI4ziAIBmnWrfc7VXwGqTcnqSBClovfSVpHsEX4neqT4ClCc83p5guPfadOafLzWxLwnT78JgSj699Oet/8zqY2abwaeJrETX25P2W7rtDuGy1mW1MWlYqyutY6jKCczJRQRXp2eUcW53gN5ZqGEl5wElAZli3C8GlcTNJ/cxsStIqqbrjXQCcbWbvpdh+1xTrnQocT3D5/hXBlcNqgg/DCoKqie5A8r4TLQJ2k9Q4IUl0JqhSqazFBFckpTErcZrg+G4ws8d3YtuVlbjfzgRXYivCGP5hZuemXCuQ6rUpa3vlvQZlbW8cwVXDsWZWLOkOoEX5h1OmnTmnyfEsIvjCnR5Odw7npVvpfhN1Bl4heC81l9QwIUl05tvYo7yOAJjZEoIqMSQdALwu6R0zm10Fx1Aj+RVEzXMCwX/7fQjq8fsT1Av/j6DqJdlSoIWkpgnz7gVuSLjJ2ErS8eXsszGwlaAKIh/4U+kCMysBHgJuVXATOlPSkOSbj2a2gKDO+UZJuZL6Aj8HduZL/N/AXpJODK8wLgTaJh3f7xTexJXUVNKwndhPFKdL6iMpn6CO/WkLmsE+Bhwr6UfhOckNb9x23MntlfkalKMZQTVdsaRBBHXvO6uic7oU6FbBNp4Ergzfby2BqwjOU7q9BPSUdKqkLEknE3x+XjSzeUAh8EdJOeEX+7EJ60Z+HSUNS5i/miDJ1Kom0ZXlCaLmOZPg5tt8M1tS+gDuAk5LrpIxs5kEH8w54SVye+AO4HngNUnrCW4c7lfOPh8luOz+muBm4AdJy38DTAM+Iqh6uJnU751TCOq+FwHPAVeb2X8iH/m3x7QCGEbQOmcl0AN4L2H5c2EMY8PqmE+BI1Nsqir8g6D+fwlBVduFYQwLCP7j/z3BzfgFwG+p+DOVcntU/Bqk8gvg6vA1vgoYH+2QdhThnF4D/D18j51UxmauJ/gynkrwfvk4nJdWZrYSOAa4lOD9chlwTEIV7akE7/9VwNUE57p03cq8jt8DPpS0geDzNcrM5qbjmGqK0hYDzrkkkt4iuJn5QE3cnnPp5lcQzjnnUvIE4ZxzLiWvYnLOOZeSX0E455xLqU79DqJly5bWtWvXuMNwzrlaY9KkSSvMrFWqZXUqQXTt2pXCwsK4w3DOuVoj/BV+Sl7F5JxzLiVPEM4551LyBOGccy4lTxDOOedS8gThnHMuJU8QzjnnUvIE4ZxzLqV6nyC2lxh3/3c2UxasiTsU55yrUep9gtiwtZjHP5jHReMms3FrcdzhOOdcjVHvE0TTvGxuPbk/X63cyHUvfhZ3OM45V2PU+wQBMLhbC37xg+6M/WgBL09bHHc4zjlXI3iCCF18eE/6dmzKFc9OY/HazXGH45xzsfMEEcrOzOCO4QMoKi7h0vFTKCnxcTKcc/WbJ4gEu7dsyDXH9eH9L1dy///mxB2Oc87FyhNEkpMKOnHk3m255bVZfPr12rjDcc652HiCSCKJG0/chxYNG3Dh2E/YVORNX51z9ZMniBSa5edw60n9mLtiI9f/e0bc4TjnXCw8QZRh6B4tOe/Abjzx4Xxem74k7nCcc67aeYIox6WH92LvDk24/JmpLF23Je5wnHOuWnmCKEdOVtD0dfO27fzmKW/66pyrXzxBVKB7q0Zcdcxe/O+LFTz03ty4w3HOuWrjCSKCUwZ14vA+bfjzK7OYvsibvjrn6gdPEBFI4uaf9KVZfjajxk5mc9H2uENyzrm08wQR0W4Ncxh9Uj9mL9vAjS9701fnXN3nCaISvt+jFeccsDuPTpjHGzOWxh2Oc86llSeISvrtEb3Ys10TLnt6KsvWe9NX51zd5QmikhpkZXLn8P5s2FrMb5+a6k1fnXN1lieIndCjTWOuPKYPb3++nL9P+CrucJxzLi08Qeyk0/frzKG9W3PjyzOZuWRd3OE451yV8wSxkyRx80/70iQ3m1FPTmbLNm/66pyrWzxB7IKWjRpwy7C+zFq6nptenhl3OM45V6U8Qeyig3q1ZsT+XXnk/a/476xlcYfjnHNVxhNEFbj8iN70btuY3z41lRUbtsYdjnPOVQlPEFUgNzuTO4YPYN2WbVz29FTMvOmrc6728wRRRXq1bczvj+zNmzOX8Y8P5sUdjnPO7bK0JghJR0iaJWm2pCtSLG8q6QVJUyRNlzQiYdnF4bxPJT0pKTedsVaFM4d25aBerbjh3zP4fOn6uMNxzrldkrYEISkTuBs4EugDnCKpT1KxC4DPzKwfcBAwWlKOpA7AhUCBme0NZALD0xVrVZHEX37aj0YNsrjwyU/YWuxNX51ztVc6ryAGAbPNbI6ZFQFjgeOTyhjQWJKARsAqoDhclgXkScoC8oFFaYy1yrRq3IC/DOvLzCXr+fMrs+IOxznndlo6E0QHYEHC9MJwXqK7gD0JvvynAaPMrMTMvgZuAeYDi4G1ZvZaGmOtUof0bsOZQ7rw4Ltzeefz5XGH45xzO6VSCUJS28oUTzEvuXnPj4DJQHugP3CXpCaSmhNcbeweLmso6fQyYjpPUqGkwuXLa86X8e+O2pMerRtx6VNTWLWxKO5wnHOu0ip7BfFSJcouBDolTHdkx2qiEcCzFpgNzAV6A4cBc81suZltA54FhqbaiZmNMbMCMyto1apVJcJLr9zsTO48ZQBrN3nTV+dc7VTZBJHqqqAsHwE9JO0uKYfgJvPzSWXmA4cCSGoD9ALmhPMHS8oP708cCtS6Ydz2bNeEy4/szeszlvLExPlxh+Occ5VS2QRxf9SCZlYM/Ap4leDLfbyZTZc0UtLIsNh1wFBJ04A3gMvNbIWZfQg8DXxMcG8iAxhTyVhrhBFDu/L9Hi257sXPmL3Mm74652oP1aWqj4KCAissLIw7jB0sW7eFI+74H22b5PLcBUNpkJUZd0jOOQeApElmVpBqmf+Suhq0bpLLzT/py2eL13Hra5/HHY5zzkXiCaKaHN6nDaft15n73pnDe7NXxB2Oc85VKFKCkNRF0mHh8zxJjdMbVt105dF96N6qIZeMn8xqb/rqnKvhKkwQks4luGF8XzirI/DPNMZUZ+XlBL2+rtpYxBXPetNX51zNFuUK4gJgf2AdgJl9AbROZ1B12d4dmvLbH/Xi1elLGffRgopXcM65mERJEFvDvpQACPtG8n99d8E5B3Rj/z1a8McXPmPO8g1xh+OccylFSRBvS/o9Qcd5hwNPAS+kN6y6LSNDjB7WnwbZGYwaO5mi4pK4Q3LOuR1ESRCXA8sJfrB2PkF3G1emM6j6oG3TXG46sS/Tvl7Lba9701fnXM2TVd5CSRnA1HBMhsi/onbRHLF3W04Z1Il73/6SA3u0Ykj3FnGH5Jxz3yj3CsLMSoApkjpXUzz1zv8d04fdWwRNX9ds8qavzrmaI0oVUztguqQ3JD1f+kh3YPVFfk4WdwwfwPL1W/n9c9O86atzrsYot4op9Me0R1HP7dOxKZf+sBc3vzKTpyctZFhBp4pXcs65NKvwCsLM3gZmAo3Dx4xwnqtC5x3YjcHdduPq56fz1YqNcYfjnHORfkl9EjARGAacBHwo6afpDqy+ycwQt57Un+zMDEaNm8y27d701TkXryj3IP4AfM/MzjSzM4BBwP+lN6z6qX2zPP70432YsmANd77xRdzhOOfquSgJIsPMliVMr4y4ntsJR/dtx7B9O3L3f2czce6quMNxztVjUb7oX5H0qqSzJJ0F/Bt4Ob1h1W/XHLcXnXfL5+Jxk1m7eVvc4Tjn6qkoN6l/S9CTa1+gHzDGzC5Ld2D1WcMGWdw+fABL123hyn9+6k1fnXOxiHKTenfgJTO7xMwuJrii6Jr2yOq5/p2acfHhPXlhyiKe++TruMNxztVDUaqYngISm9RsD+e5NBv5g+4M6robV/1rOvNXboo7HOdcPRMlQWQldvcdPs9JX0iuVGaGuG14fyS4aNwnFHvTV+dcNYqSIJZLOq50QtLxgA+qXE06hE1fP56/hr++OTvucJxz9UiUBDES+L2k+ZIWEHT/fX56w3KJju3XnhMHduCvb37BpHne9NU5Vz2itGL60swGA32APmY21Mz8X9lq9sfj9qJj83xGjZ3Mui3e9NU5l35lJghJx0rqkjDrEuDdsDfX3dMfmkvUODeb207uz+K1W7j6X9PjDsc5Vw+UdwVxA8FIckg6BjgdOBt4Hrg3/aG5ZPt2ac6Fh/TguU++5l+Tvemrcy69yksQZmalbStPBB40s0lm9gDQKv2huVQuOLg7BV2ac+Vzn7JglTd9dc6lT3kJQpIahcOOHgq8kbAsN71hubJkZWZw28n9Abh43GRv+uqcS5vyEsTtwGSgkGAMiEIASQOAxWmPzJWp0275XHfC3hTOW83f3voy7nCcc3VUmSPKmdlDkl4FWgNTEhYtAUakOzBXvhMGdOCtWcu4440vOKBHSwZ2bh53SM65OqbcZq5m9rWZfWJmJQnzFpvZ/PSH5ipy7Ql707ZJLheNncyGrcVxh+Ocq2N8XIdarEluNncM78/C1Zu86atzrsp5gqjlCrruxq8O6cEzHy/khSmL4g7HOVeHROnu+xZJe1VHMG7nXHjIHgzo3IzfPzeNr9dsjjsc51wdEeUKYiYwRtKHkkZKapruoFzlZGVmcMfJAygpMS4eN5ntJT7AkHNu10Xpi+kBM9sfOAPoCkyV9ISkg9MdnIuuc4t8rj1+bybOXcW9b3vTV+fcrot0D0JSJtA7fKwgaPZ6iaSxFax3hKRZkmZLuiLF8qaSXpA0RdJ0SSMSljWT9LSkmZJmSBpSqSOrh04c2IFj+rbjtv98zpQFa+IOxzlXy0W5B3ErMAs4CviTme1rZjeb2bHAgHLWywTuBo4k6An2FEl9kopdAHxmZv2Ag4DRkkoHI7oDeMXMehOMhT2jUkdWD0nihh/vQ5smuYwa+wkbvemrc24XRLmC+BToa2bnm9nEpGWDyllvEDDbzOaEo9CNBY5PKmNAY0kCGgGrgGJJTYADgQchGMXOzNZEiLXea5qXza0n9WPeqk1c+8JncYfjnKvFoiSI1UB26URY9XMCgJmtLWe9DsCChOmF4bxEdwF7AouAacCo8Ed53Qh6kn1Y0ieSHpDUMNVOJJ0nqVBS4fLlyyMcTt23X7cW/PKg7owrXMDL07xXFOfczomSIK5OTAThf/JXR1hPKeYlN6/5EUF/T+2B/sBd4dVDFjAQuMfMBgAbgR3uYYTxjDGzAjMraNXKO5ktddFhPenXsSlXPDuNxWu96atzrvKiJIhUZcrswynBQqBTwnRHgiuFRCOAZy0wG5hLcCN8IbDQzD4Myz1NkDBcRNmZGdwxfADbtpdwybgp3vTVOVdpURJEoaRbJXWX1E3SbcCkCOt9BPSQtHt443k4wWBDieYTdCWOpDZAL2COmS0BFkjqFZY7FPAK9Urq2rIh1xy3FxPmrOT+/82JOxznXC0TJUH8GigCxgFPAVsIWh+Vy8yKgV8BrxK0QBpvZtPDH9uNDItdBwyVNI1gvInLzWxFwn4flzSVoPrpT5GPyn1j2L4dOWqftox+bRbTFpZ3y8g5575LZnWn6qGgoMAKCwvjDqPGWbOpiCPv+B952Zm8eOEB5OdEqSF0ztUHkiaZWUGqZVF+B9FK0l8kvSTpzdJH1Yfp0qVZfg6jT+rH3JUbue5F/zmJcy6aKFVMjxP0x7Q78EfgK4L7C64WGdq9Jecf2J0nJ87n1elL4g7HOVcLREkQLczsQWCbmb1tZmcDg9Mcl0uDSw7vyT4dmnLFM1NZum5L3OE452q4KAliW/h3saSjwzGpO6YxJpcmOVkZ3D68P1u2lXDp+CmUeNNX51w5oiSI68Muvi8FfgM8AFyc1qhc2nRv1Yirju3Du7NX8NB7c+MOxzlXg5XbnCXscK+Hmb0IrAW8i+86YPj3OvHWrGX8+ZVZDOnegr3a+xAfzrkdlXsFYWbbgeOqKRZXTSRx04l9ad4wm1FjJ7O5aHvcITnnaqAoVUzvS7pL0vclDSx9pD0yl1bNG+Ywelh/Zi/bwA0v+Y/UnXM7ivKLqaHh32sT5hlwSNWH46rTAT1act6B3RjzzhwO6tmaw/q0iTsk51wNUmGCMDO/71CHXfrDnrz7xQoue2Yqr3T6Pq0b58YdknOuhqgwQUi6KtV8M7s21XxXuzTIyuTOU/pzzF/f5TdPTeWRs75HRkaqntqdc/VNlHsQGxMe2wmGEO2axphcNdujdWOuPLoP73y+nEfe/yrucJxzNUSUKqbRidOSbmHHbrtdLXfafp15a9Zybnp5JkO6t2DPdk3iDsk5F7MoVxDJ8gmGBHV1iCRu/sk+NM3PZtTYT9iyzZu+OlffRenNdZqkqeFjOjALuCP9obnq1qJRA24Z1o/Pl27gppdnxh2Ocy5mUZq5HpPwvBhYGg4G5OqgH/Rsxdn7785D783lBz1bcXDv1nGH5JyLSZQqpnbAKjObZ2ZfA7mS9ktzXC5Glx3Ri95tG/Pbp6ewfP3WuMNxzsUkSoK4B9iQML0pnOfqqNzsTO48ZQDrtxRz2dNTqEujDjrnoouSIGQJ3xBmVkK0qilXi/Vs05jfH7Un/521nEcnzIs7HOdcDKIkiDmSLpSUHT5GAXPSHZiL3xlDunBwr1bc8NIMPl+6Pu5wnHPVLEqCGEnQH9PXwEJgP+C8dAblagZJ/GVYP5rkZnHhk9701bn6psIEYWbLzGy4mbU2szZmdqqZLauO4Fz8WjZqwF+G9WPmkvX8+ZVZcYfjnKtGUX4H8XdJzRKmm0t6KK1RuRrl4F6tOWtoVx56by5vf7487nCcc9UkShVTXzNbUzphZquBAWmLyNVIVxzZm15tGvObp6awcoM3fXWuPoiSIDIkNS+dkLQb3oqp3snNzuSOU/qzdvM2Ln9mqjd9da4eiJIgRhOMKnedpOuA94G/pDcsVxP1btuEK47ozeszlvHgu3PjDsc5l2ZRenN9VFIhwQhyAk40Mx+jsp46a2hXPpizkuv/PQOAc77v/TY6V1dF6s3VzD4zs7uAl4ATJX2a3rBcTZWRIe46dSBH7dOW6/89g7++8YVXNzlXR0VpxdRO0kWSJgLTgUzglLRH5mqsnKwM7hw+gBMHdmD0fz7n5ldmeZJwrg4qs4pJ0rkEiaAjMB44B/iXmf2xmmJzNVhWZga3/LQfedmZ3Pv2l2wuKubqY/fy4Uqdq0PKuwdxNzABONXMCgEk+b+J7hsZGeL6E/amYYMsxrwzh41F27n5J33J9CThXJ1QXoJoDwwDbpXUhuAqIrtaonK1hiR+d2Rv8nMyuf31L9i8bTu3n9yf7MydGazQOVeTlPkpNrMVZnaPmR0IHAqsBZZJmiHpT9UWoavxJHHRYT35w1F78u+pi/nFY5O83ybn6oCorZgWmtktZrYvcALgP6V1Ozj3wG5cf8LevD5jGT//+0dsKvKBB52rzSpdD2Bms/xGtSvL6YO7MHpYPyZ8uZIzHpzIui3b4g7JObeTvKLYVbmf7NuRu04dyJSFazjt/g9ZvbEo7pCcczshrQlC0hGSZkmaLemKFMubSnpB0hRJ0yWNSFqeKekTSS+mM05X9Y7apx1jflbArKXrGT7mA5at3xJ3SM65SiozQUgaWN6jog1LyiRoKnsk0Ac4RVKfpGIXAJ+ZWT/gIGC0pJyE5aOAGZU8JldDHNy7NY+c9T0WrN7ESfdO4Os1m+MOyTlXCeVdQYwOH3cDHwJjgPvD53dG2PYgYLaZzTGzImAscHxSGQMaSxLQCFgFFANI6ggcDTwQ+WhcjTN0j5b84+f7sXJjESfdO4GvVmyMOyTnXETlNXM92MwOBuYBA82sIGzFNACYHWHbHYAFCdMLw3mJ7gL2BBYB04BRZlYSLrsduAwooRySzpNUKKlw+XIfzKYm2rdLc548dzCbt21n2H0TfHxr52qJKPcgepvZtNIJM/sU6B9hvVQ/p03+JfaPgMkEP8rrD9wlqYmkY4BlZjapop2Y2ZgweRW0atUqQlguDnt3aMq48wYj4OT7JvDp12vjDsk5V4EoCWKGpAckHSTpB5LuJ9p9gYVAp4TpjgRXColGAM9aYDYwF+gN7A8cJ+krgqqpQyQ9FmGfrgbr0aYx488fQn5OFqfc/wGT5q2KOyTnXDmiJIgRBL24jgIuAj4L51XkI6CHpN3DG8/DgeeTyswn+JU2YXcevYA5ZvY7M+toZl3D9d40s9Mj7NPVcF1bNmT8yCG0bNSAnz04kfdnr4g7JOdcGSpMEGa2BbgXuMLMfmxmt4XzKlqvGPgV8CrBFcd4M5suaaSkkWGx64ChkqYBbwCXm5l/Y9RxHZrlMe78wXRqns9Zj3zEmzOXxh2Scy4FVdSPv6TjCIYYzTGz3SX1B641s+OqIb5KKSgosMLCwrjDcBGt3ljEGQ9NZOaSddwxfABH7dMu7pCcq3ckTTKzglTLolQxXU3QZHUNgJlNBrpWUWyuHmveMIfHz92Pfh2b8asnPuaZSQvjDsk5lyBKgig2M29y4tKiSW42j/58EEO7t+TSp6bwjw/mxR2Scy4UJUF8KulUIFNSD0l/Bd5Pc1yuHsnPyeKBMws4bM/W/N8/P2XMO1/GHZJzjmgJ4tfAXgRdfD9BMC7ERWmMydVDudmZ3HP6vhzTtx1/emkmt/3ncx/n2rmYlTeiHABmtgn4Q/hwLm2yMzO4Y/gAcrMzueONYHS63x3Zm6AnFudcdavwCkLSfyQ1S5huLunVtEbl6q3MDPHnn/TlzCFdGPPOHK7856eUlPiVhHNxqPAKAmhpZmtKJ8xstaTW6QvJ1XcZGeKa4/YiLyeLe9/+ks1F2/nzT/uS5eNcO1etoiSIEkmdzWw+gKQu7NinknNVShKXH9GLhjmZjP7P52wp3s7tJw8gJ8uThHPVJUqC+APwrqS3w+kDgfPSF5JzAUn8+tAe5OVkcv2/Z7C5qJB7Tt+X3OzMuENzrl6I0tXGK8BAYBwwHtjXzPwehKs253y/G3/68T689flyRjz8ERu3FscdknP1QtTr9QYEg/msBfpIOjB9ITm3o1P368ytJ/Vj4ler+NmDH7J287a4Q3KuzquwiknSzcDJBD26lg7eY8A7aYzLuR38eEBH8rKz+PWTH3PKmA/4x88H0aJRg7jDcq7OinIFcQLQy8yONrNjw0eN66jP1Q9H7N2W+88o4MvlGxg+5gOWrquwY2Hn3E6KkiDmANnpDsS5qA7q1Zq/nz2IRWs2c9J9E1i4elPcITlXJ0VJEJuAyZLuk3Rn6SPdgTlXnsHdWvDYOfuxemMRJ907gTnLN8QdknN1TpQE8TzBwD7vA5MSHs7FakDn5ow9bwhbi0s46b4PmLVkfdwhOVenVDhgUG3iAwbVT7OXree0Bz5ka3EJj549iL4dm8UdknO1xi4NGBR28f20pM8kzSl9VH2Yzu2cPVo35qnzh9KoQRan3v8hH321Ku6QnKsTolQxPQzcAxQDBwOPAv9IZ1DOVVbnFvk8NXIIrRs34IwHJ/LuFz60uXO7KkqCyDOzNwiqo+aZ2TXAIekNy7nKa9c0j3HnD6FLi3zOfuQjXv9sadwhOVerRUkQWyRlAF9I+pWkHwPem6urkVo1bsDY8wazZ7vGjHxsEi9MWRR3SM7VWlESxEVAPnAhsC/wM+DMNMbk3C5plp/DY+fsx8DOzRk19hPGFy6IOyTnaqUoI8p9FD7dAIxIbzjOVY3Gudn8/exBnPePQi57eiqbi7Zz5tCucYflXK1SZoKQ9ALljPvg3W24mi4vJ5MHzizgV098wtXPT2dT0XZ+cVD3uMNyrtYo7wrilmqLwrk0aZCVyd9OG8il46dw8ysz2VRUzCWH9/Rxrp2LoMwEYWZvl7XMudokOzOD207uT35OJn99czabirZz5dF7epJwrgJRuvvuAdwI9AFyS+ebWbc0xuVclcrMEDeeuA95OZk8+O5cNhUVc/0J+5CZ4UnCubJEGXL0YeBq4DaCH8qNAPxT5WodSVx1TB/yczK5+79fsrloO7cM60dWpo9z7Vwq/kM5V69I4rc/6s1vf9SLf05exAVPfMzW4u1xh+VcjeQ/lHP10gUH78HVx/bh1elLOe/RSWwu8iThXDL/oZyrt0bsvzs3/2Qf3vliOWc9PJENW4vjDsm5GsV/KOfqtZO/15nc7EwuGT+F0x74kEdHDKJpvg+g6ByUcwUhqaWkqyVdKKmRpHskfSrpX5L2qM4gnUun4/t34J7TBjJj0TqG3/8BKzZsjTsk52qE8qqYngAaAD2AiQRjU/8UeBF4IP2hOVd9frhXWx44s4C5KzZw8n0TWLJ2S9whORe78hJEGzP7PcG9h0Zm9hczm2lm9wPNqiU656rRgT1b8ejZ+7F03VaG3fc+C1Ztijsk52JVXoLYDmDBmKTJo6+URNm4pCMkzZI0W9IVKZY3lfSCpCmSpksaEc7vJOm/kmaE80dFPB7ndsmg3Xfj8XP2Y93mYobdO4Evl2+IOyTnYlNegugm6fmw077S56XTu1e0YUmZwN3AkQS/wj5FUp+kYhcAn5lZP+AgYLSkHILR6y41sz2BwcAFKdZ1Li36dWrG2PMGU1xSwsn3TWDG4nVxh+RcLMpLEMcDowk67St9Xjp9QoRtDwJmm9kcMysCxobbSWRAYwWd4jQCVgHFZrbYzD4GMLP1wAygQ9SDcm5X7dmuCePOH0J2ZgbDx3zA5AVr4g7JuWpXZoIws7fLe0TYdgcgcaSWhez4JX8XsCewCJgGjDKz71RfSeoKDAA+jLBP56pM91aNGH/+EJrmZXP6Ax/y4ZyVcYfkXLVKZyc0qfprSh5f4kfAZKA90B+4S1KTbzYgNQKeAS4ys5TX+ZLOk1QoqXD58uVVEbdz3+i0Wz7jzx9CmyYNOPPhibz9ub/HXP2RzgSxEOiUMN2R4Eoh0QjgWQvMBuYCvQEkZRMkh8fN7NmydmJmY8yswMwKWrVqVaUH4BxA26a5jDt/CN1aNuLcvxfy6vQlcYfkXLWoVIKQ1LYSxT8CekjaPbzxPBx4PqnMfODQcNttgF7AnPCexIPADDO7tTIxOpcOLRs14MlzB9OnfRN++fjH/Gvy13GH5FzaVfYK4qWoBc2sGPgV8CrBTebxZjZd0khJI8Ni1wFDJU0D3gAuN7MVwP4EfT4dImly+DiqkrE6V6Wa5mfz2Dn7UdClOReNm8zYifPjDsm5tFLwM4eIhaVPzGxAGuPZJQUFBVZYWBh3GK6O21y0nZGPTeLtz5dz1TF9OPuAClt9O1djSZpkZgWpllX2CuL+KojHuVotLyeTMWfsyxF7teXaFz/j7v/Ojjsk59KiUgnCzP6WrkCcq00aZGVy16kD+PGADvzl1Vn8+ZWZVOZq3LnaIMqQo865FLIyMxg9rB+52Zn87a0v2VS0nauO6UOGj3Pt6ghPEM7tgowM8acf701+TiYPvjuXTUXF3HhiXzI9Sbg6wBOEc7tIElcevScNczK5883ZbN5Wwq0n9SM7M50/M3Iu/TxBOFcFJHHJD3uR3yCLm16eyeai7dx16gByszPjDs25neb/4jhXhUb+oDvXHr8Xr89YyrmPFrKpyMe5drWXJwjnqtgZQ7ryl5/25b3ZKzjzoYms37It7pCc2ymeIJxLg2EFnbjzlAF8Mn8Npz3wIcvW+xCmrvbxexDOpckxfduTl53JLx7/mEE3vEGjBlm0a5pL26a5tG+aR9umubRrmku7ZnnB36a5NM7Njjts577hCcK5NDp0zzY898uhvPvFChav3cLitZtZsnYLM5esZ8WGrST/ts6TiKtJPEE4l2Z7tW/KXu2b7jC/qLiEpeu2sGTdFhatCRJHZZNIkDTyPIm4tPAE4VxMcrIy6LRbPp12yy+zTFFxCcvWlyaOLSxes9mTiKs2niCcq8FysjLo2Dyfjs13PonMWrKe5WUkkW8TiCcRtyNPEM7VcjuTRJas3cyiNVvCaq3NnkRcSp4gnKsHKpNElqzdwqIqTCJtm+bSxJNIreQJwjkHREsi27aHN9Y9idQLniCcc5FlZ+5iElm3hVlLlleYRDo0y6Nj87xvbuJ3ap5Py0Y5BMPVu+riCcI5V6V2NokEN9iDK5EZi4PWWYnysjO/TRrh347N8+m0W/Dcr0CqnicI51y1i5JENhdtZ+HqTSxYvYn5KzexYPVmFqwK/n40dxXrt363I8SmedlBsmie/00S6RhefXRsnuc96+4ETxDOuRopLyeTHm0a06NN4x2WmRlrN29jwarNLFi9KUwcm1iwajOzlq7njZnLKCou+c46rRs3+M7VR2nVVafd8mjXNM8HeUrBE4RzrtaRRLP8HJrl57BPxx1/pV5SYizfsPU7iaP0+Udfreb5KYsoSbgHkpUh2jfL+84VyLfVWfX3/ocnCOdcnZORIdo0yaVNk1wKuu62w/Jt20tYtGZz0hVIkERen7GUFRuKvlO+vt7/8AThnKt3sjMz6NKiIV1aNEy5fFNRMQtL73ms2on7HwmJpEOz2nv/wxOEc84lyc/JomebxvSs7P2PJet5Y8YyirZ/9/5HmyYNUt48r+n3PzxBOOdcJUS5/7Fs/dZvk0dCIpk4dxX/mry51tz/8AThnHNVKCNDtA1/Hf69FPc/iopLWLx21+5/JLbASmdfWJ4gnHOuGuVkVf7+x/xV316BbEi6/9EsP5uerRszfuSQKo/VE4RzztUgFd3/WLNp27dNd8MrkO0llmJLu84ThHPO1RKSaN4wh+YNc+jbsVna95eR9j0455yrlTxBOOecS8kThHPOuZQ8QTjnnEvJE4RzzrmUPEE455xLyROEc865lDxBOOecS0mWPHJ4LSZpOTBvJ1dvCayownCqisdVOR5X5XhclVMX4+piZq1SLahTCWJXSCo0s4K440jmcVWOx1U5Hlfl1Le4vIrJOedcSp4gnHPOpeQJ4ltj4g6gDB5X5XhcleNxVU69isvvQTjnnEvJryCcc86l5AnCOedcSvUqQUh6SNIySZ+WsVyS7pQ0W9JUSQNrSFwHSVoraXL4uKqa4uok6b+SZkiaLmlUijLVfs4ixlXt50xSrqSJkqaEcf0xRZk4zleUuGJ5j4X7zpT0iaQXUyyL5TMZIa64PpNfSZoW7rMwxfKqPV9mVm8ewIHAQODTMpYfBbwMCBgMfFhD4joIeDGG89UOGBg+bwx8DvSJ+5xFjKvaz1l4DhqFz7OBD4HBNeB8RYkrlvdYuO9LgCdS7T+uz2SEuOL6TH4FtCxneZWer3p1BWFm7wCryilyPPCoBT4AmklqVwPiioWZLTazj8Pn64EZQIekYtV+ziLGVe3Cc7AhnMwOH8mtQOI4X1HiioWkjsDRwANlFInlMxkhrpqqSs9XvUoQEXQAFiRML6QGfPGEhoRVBC9L2qu6dy6pKzCA4L/PRLGes3LighjOWVgtMRlYBvzHzGrE+YoQF8TzHrsduAwoKWN5XO+v2yk/LojnfBnwmqRJks5LsbxKz5cniO9Sink14T+tjwn6S+kH/BX4Z3XuXFIj4BngIjNbl7w4xSrVcs4qiCuWc2Zm282sP9ARGCRp76QisZyvCHFV+/mSdAywzMwmlVcsxby0nq+IccX1mdzfzAYCRwIXSDowaXmVni9PEN+1EOiUMN0RWBRTLN8ws3WlVQRm9hKQLalldexbUjbBl/DjZvZsiiKxnLOK4orznIX7XAO8BRyRtCjW91hZccV0vvYHjpP0FTAWOETSY0ll4jhfFcYV1/vLzBaFf5cBzwGDkopU6fnyBPFdzwNnhC0BBgNrzWxx3EFJaitJ4fNBBK/bymrYr4AHgRlmdmsZxar9nEWJK45zJqmVpGbh8zzgMGBmUrE4zleFccVxvszsd2bW0cy6AsOBN83s9KRi1X6+osQV0/uroaTGpc+BHwLJLR+r9Hxl7XS0tZCkJwlaH7SUtBC4muCGHWZ2L/ASQSuA2cAmYEQNieunwC8kFQObgeEWNllIs/2BnwHTwvprgN8DnRNii+OcRYkrjnPWDvi7pEyCL4zxZvaipJEJccVxvqLEFdd7bAc14HxFiSuO89UGeC7MS1nAE2b2SjrPl3e14ZxzLiWvYnLOOZeSJwjnnHMpeYJwzjmXkicI55xzKXmCcM45l5InCOdiIuktSeUONB+ljHPp4gnCOedcSp4gnEsi6SxJz0p6RdIXkv6csGyDpJvDztJelzQo/C9/jqTjwjK5kh5W0G//J5IODufnSRqroJ/+cUBewnZ/KGmCpI8lPRX2M5UYU6akRyR9Gm734mo6Ha4e8wThXGr9gZOBfYCTJZX2b9MQeMvM9gXWA9cDhwM/Bq4Ny1wAYGb7AKcQ/Io5F/gFsMnM+gI3APsChH34XAkcFnbEVkgwFkFyPB3MbO9wuw9X9QE7l6xedbXhXCW8YWZrASR9BnQh6Ea5CHglLDMN2Gpm2yRNA7qG8w8g6OETM5spaR7Qk2BgqDvD+VMlTQ3LDwb6AO+F3SjkABOS4pkDdJP0V+DfwGtVerTOpeBXEK5eknSBvh0usn2KIlsTnm/n23+mtiX0uVNSWs7MShLKpOpyuVSqvm1EMEZD//DRx8x+/p2VzFYD/Qh6Yr2ApIFswiqo0uO5FueqgCcIVy+Z2d0JX8hV3X30O8BpAJJ6EnQiOCtp/t5A37D8B8D+kvYIl+WH630jrIbKMLNngP8jGKI28Xi2JxxPtY0n7eo2r2Jyrur9Dbg3rHYqBs4ys62S7gEeDquWJgMTAcxsuaSzgCclNQi3cSXBWNulOoTrlv5T97v0H4ar77w3V+eccyl5FZNzzrmUPEE455xLyROEc865lDxBOOecS8kThHPOuZQ8QTjnnEvJE4RzzrmU/h+jrAp+dGbn4QAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mejor puntuación: 0.8935434935904601 con valor = 1\n"
     ]
    }
   ],
   "source": [
    "values = [1, 2, 3, 4, 5]\n",
    "print('\\n--- Alterando n_neighbors ['+str(np.min(values))+' - '+str(np.max(values))+'] ---')\n",
    "scores = []\n",
    "\n",
    "for v in range(len(values)):\n",
    "    SeqModel = SequentialModelAlgorithm(neighbors = values[v], nmodels=10,  method = \"knn\")\n",
    "    models, score = SeqModel.start(attributes_cols = attr_cols, objetive_col = obj_col)\n",
    "    scores.append(score)\n",
    "    \n",
    "# Realizamos una gráfica para determinar cómo afecta ntree a la puntuación\n",
    "plt.plot(values, scores)\n",
    "plt.xlabel('- nmodels -')\n",
    "plt.ylabel('- Balanced Accuracy Score -')\n",
    "plt.title('Alteración del hiperparámetro nmodels')\n",
    "plt.show()\n",
    "m = np.argmax(scores)\n",
    "print('Mejor puntuación: '+str(scores[m])+' con valor = '+str(values[m]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6eb69d62",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "# Generalizado para más modelos: árboles de decisión o knn\n",
    "from lib.SequentialModelAlgorithm import SequentialModelAlgorithm\n",
    "\n",
    "# Selección de columnas\n",
    "dataset = pd.read_csv('./datasets/adultDataset.csv', header = 0)\n",
    "attr_cols = dataset.loc[:, 'capital-gain':'native-country']\n",
    "obj_col = dataset['income']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "ecfa9456",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--- Alterando nmodels [1 - 600] ---\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mejor puntuación: 0.9291888032077444 con valor = 300\n",
      "--------------------------\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "values = [1, 50, 150, 300, 600]\n",
    "print('\\n--- Alterando nmodels ['+str(np.min(values))+' - '+str(np.max(values))+'] ---')\n",
    "scores = []\n",
    "\n",
    "for v in range(len(values)):\n",
    "    SeqTree = SequentialModelAlgorithm(nmodels = values[v], sample_size = 0.65, max_depth = 10, lr = 0.1, random = True)\n",
    "    trees, score = SeqTree.start(attributes_cols = attr_cols, objetive_col = obj_col)\n",
    "    scores.append(score)\n",
    "    \n",
    "# Realizamos una gráfica para determinar cómo afecta ntree a la puntuación\n",
    "plt.plot(values, scores)\n",
    "plt.xlabel('- nTrees -')\n",
    "plt.ylabel('- Balanced Accuracy Score -')\n",
    "plt.title('Alteración del hiperparámetro ntrees y manteniendo el resto constante')\n",
    "plt.show()\n",
    "\n",
    "m = np.argmax(scores)\n",
    "print('Mejor puntuación: '+str(scores[m])+' con valor = '+str(values[m]))\n",
    "\n",
    "print('--------------------------')\n",
    "print('\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "19041ef1",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), DecisionTreeRegressor(max_depth=10), KNeighborsRegressor(n_neighbors=2), DecisionTreeRegressor(max_depth=10)]\n"
     ]
    }
   ],
   "source": [
    "print(trees)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "86df608f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--- Alterando nmodels [1 - 300] ---\n"
     ]
    },
    {
     "ename": "ValueError",
     "evalue": "Unknown label type: (array([-0.1,  0. ,  0.9,  1. ]),)",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-2-de4d97685cad>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m      5\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mv\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mvalues\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      6\u001b[0m     \u001b[0mSeqTree\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mSequentialModelAlgorithm\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mnmodels\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mvalues\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mv\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0msample_size\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;36m0.65\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmethod\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;34m\"gradient\"\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mrandom\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;32mFalse\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m----> 7\u001b[1;33m     \u001b[0mtrees\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mscore\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mSeqTree\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mstart\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mattributes_cols\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mattr_cols\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mobjetive_col\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mobj_col\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      8\u001b[0m     \u001b[0mscores\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mscore\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      9\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Desktop\\ia\\entrega\\Jupyter\\lib\\SequentialModelAlgorithm.py\u001b[0m in \u001b[0;36mstart\u001b[1;34m(self, attributes_cols, objetive_col, random_state, ftest_size)\u001b[0m\n\u001b[0;32m     71\u001b[0m         \u001b[0mmodels\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m[\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     72\u001b[0m         \u001b[1;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mnmodels\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 73\u001b[1;33m             \u001b[0msubmodel\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpred\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mmeta_algorithm\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mX_test\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my_test\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpred\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     74\u001b[0m             \u001b[0mmodels\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0msubmodel\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     75\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Desktop\\ia\\entrega\\Jupyter\\lib\\SequentialModelAlgorithm.py\u001b[0m in \u001b[0;36mmeta_algorithm\u001b[1;34m(self, X, y, prediction)\u001b[0m\n\u001b[0;32m    127\u001b[0m             \u001b[0msubmodel\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mrandom\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mchoice\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mtree\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mDecisionTreeRegressor\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mmax_depth\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mmax_depth\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmin_samples_leaf\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mmin_samples_leaf\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmax_features\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mmax_features\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmin_weight_fraction_leaf\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mmin_weight_fraction_leaf\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mKNeighborsRegressor\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mn_neighbors\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mneighbors\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    128\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 129\u001b[1;33m         \u001b[0msubmodel\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0msubmodel\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mXm\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mi_resm\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    130\u001b[0m         \u001b[0mi_prediction\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mprediction\u001b[0m \u001b[1;33m+\u001b[0m \u001b[0msubmodel\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpredict\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mX\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mlr\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    131\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mD:\\windows10_folders\\Archivos_De_Programas\\Anaconda3\\lib\\site-packages\\sklearn\\pipeline.py\u001b[0m in \u001b[0;36mfit\u001b[1;34m(self, X, y, **fit_params)\u001b[0m\n\u001b[0;32m    344\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_final_estimator\u001b[0m \u001b[1;33m!=\u001b[0m \u001b[1;34m'passthrough'\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    345\u001b[0m                 \u001b[0mfit_params_last_step\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mfit_params_steps\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0msteps\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;33m-\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 346\u001b[1;33m                 \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_final_estimator\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mXt\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mfit_params_last_step\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    347\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    348\u001b[0m         \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mD:\\windows10_folders\\Archivos_De_Programas\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_stochastic_gradient.py\u001b[0m in \u001b[0;36mfit\u001b[1;34m(self, X, y, coef_init, intercept_init, sample_weight)\u001b[0m\n\u001b[0;32m    727\u001b[0m             \u001b[0mReturns\u001b[0m \u001b[0man\u001b[0m \u001b[0minstance\u001b[0m \u001b[0mof\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    728\u001b[0m         \"\"\"\n\u001b[1;32m--> 729\u001b[1;33m         return self._fit(X, y, alpha=self.alpha, C=1.0,\n\u001b[0m\u001b[0;32m    730\u001b[0m                          \u001b[0mloss\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mloss\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mlearning_rate\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mlearning_rate\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    731\u001b[0m                          \u001b[0mcoef_init\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mcoef_init\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mintercept_init\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mintercept_init\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mD:\\windows10_folders\\Archivos_De_Programas\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_stochastic_gradient.py\u001b[0m in \u001b[0;36m_fit\u001b[1;34m(self, X, y, alpha, C, loss, learning_rate, coef_init, intercept_init, sample_weight)\u001b[0m\n\u001b[0;32m    567\u001b[0m         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mt_\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;36m1.0\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    568\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 569\u001b[1;33m         self._partial_fit(X, y, alpha, C, loss, learning_rate, self.max_iter,\n\u001b[0m\u001b[0;32m    570\u001b[0m                           classes, sample_weight, coef_init, intercept_init)\n\u001b[0;32m    571\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mD:\\windows10_folders\\Archivos_De_Programas\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_stochastic_gradient.py\u001b[0m in \u001b[0;36m_partial_fit\u001b[1;34m(self, X, y, alpha, C, loss, learning_rate, max_iter, classes, sample_weight, coef_init, intercept_init)\u001b[0m\n\u001b[0;32m    497\u001b[0m         \u001b[0mn_samples\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mn_features\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mX\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    498\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 499\u001b[1;33m         \u001b[0m_check_partial_fit_first_call\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mclasses\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    500\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    501\u001b[0m         \u001b[0mn_classes\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mclasses_\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mD:\\windows10_folders\\Archivos_De_Programas\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\multiclass.py\u001b[0m in \u001b[0;36m_check_partial_fit_first_call\u001b[1;34m(clf, classes)\u001b[0m\n\u001b[0;32m    337\u001b[0m         \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    338\u001b[0m             \u001b[1;31m# This is the first call to partial_fit\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 339\u001b[1;33m             \u001b[0mclf\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mclasses_\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0munique_labels\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mclasses\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    340\u001b[0m             \u001b[1;32mreturn\u001b[0m \u001b[1;32mTrue\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    341\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mD:\\windows10_folders\\Archivos_De_Programas\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\multiclass.py\u001b[0m in \u001b[0;36munique_labels\u001b[1;34m(*ys)\u001b[0m\n\u001b[0;32m     96\u001b[0m     \u001b[0m_unique_labels\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0m_FN_UNIQUE_LABELS\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mlabel_type\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     97\u001b[0m     \u001b[1;32mif\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[0m_unique_labels\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 98\u001b[1;33m         \u001b[1;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"Unknown label type: %s\"\u001b[0m \u001b[1;33m%\u001b[0m \u001b[0mrepr\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mys\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     99\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    100\u001b[0m     \u001b[0mys_labels\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mset\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mchain\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfrom_iterable\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0m_unique_labels\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0my\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0my\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mys\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mValueError\u001b[0m: Unknown label type: (array([-0.1,  0. ,  0.9,  1. ]),)"
     ]
    }
   ],
   "source": [
    "values = [1, 50, 150, 300]\n",
    "print('\\n--- Alterando nmodels ['+str(np.min(values))+' - '+str(np.max(values))+'] ---')\n",
    "scores = []\n",
    "\n",
    "for v in range(len(values)):\n",
    "    SeqTree = SequentialModelAlgorithm(nmodels = values[v], sample_size = 0.65, method=\"gradient\", random = False)\n",
    "    trees, score = SeqTree.start(attributes_cols = attr_cols, objetive_col = obj_col)\n",
    "    scores.append(score)\n",
    "    \n",
    "# Realizamos una gráfica para determinar cómo afecta ntree a la puntuación\n",
    "plt.plot(values, scores)\n",
    "plt.xlabel('- nTrees -')\n",
    "plt.ylabel('- Balanced Accuracy Score -')\n",
    "plt.title('Alteración del hiperparámetro ntrees y manteniendo el resto constante')\n",
    "plt.show()\n",
    "\n",
    "m = np.argmax(scores)\n",
    "print('Mejor puntuación: '+str(scores[m])+' con valor = '+str(values[m]))\n",
    "\n",
    "print('--------------------------')\n",
    "print('\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "44f904c3",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}