knn.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "e6873ff1",
   "metadata": {},
   "source": [
    "### 0. Importar librerías"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "1b4341cc",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "# Se define la semilla para que cuando se vuelva a ejecutar el split train-test\n",
    "# los datos que contiene cada conjunto sean los mismos\n",
    "semilla = 1234"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6011d883",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "# 1. Cargar el dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "85764757",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>AF365</th>\n",
       "      <th>AF450</th>\n",
       "      <th>35 deg</th>\n",
       "      <th>42 deg</th>\n",
       "      <th>Grupo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2.320158</td>\n",
       "      <td>1.229614</td>\n",
       "      <td>13.01</td>\n",
       "      <td>19.39</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.529274</td>\n",
       "      <td>0.616514</td>\n",
       "      <td>8.18</td>\n",
       "      <td>24.78</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.908257</td>\n",
       "      <td>0.955847</td>\n",
       "      <td>7.75</td>\n",
       "      <td>16.90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.590625</td>\n",
       "      <td>1.018495</td>\n",
       "      <td>4.10</td>\n",
       "      <td>16.30</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.617071</td>\n",
       "      <td>1.242961</td>\n",
       "      <td>5.76</td>\n",
       "      <td>21.05</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>115</th>\n",
       "      <td>3.034884</td>\n",
       "      <td>2.051768</td>\n",
       "      <td>4.55</td>\n",
       "      <td>8.50</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>116</th>\n",
       "      <td>4.093525</td>\n",
       "      <td>3.037288</td>\n",
       "      <td>3.97</td>\n",
       "      <td>5.35</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>117</th>\n",
       "      <td>4.730000</td>\n",
       "      <td>3.040000</td>\n",
       "      <td>7.04</td>\n",
       "      <td>14.44</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>118</th>\n",
       "      <td>3.200000</td>\n",
       "      <td>2.360000</td>\n",
       "      <td>6.70</td>\n",
       "      <td>10.57</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>119</th>\n",
       "      <td>3.885714</td>\n",
       "      <td>2.403073</td>\n",
       "      <td>6.61</td>\n",
       "      <td>14.30</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>120 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        AF365     AF450  35 deg  42 deg  Grupo\n",
       "0    2.320158  1.229614   13.01   19.39      0\n",
       "1    1.529274  0.616514    8.18   24.78      0\n",
       "2    1.908257  0.955847    7.75   16.90      0\n",
       "3    1.590625  1.018495    4.10   16.30      0\n",
       "4    1.617071  1.242961    5.76   21.05      0\n",
       "..        ...       ...     ...     ...    ...\n",
       "115  3.034884  2.051768    4.55    8.50      2\n",
       "116  4.093525  3.037288    3.97    5.35      2\n",
       "117  4.730000  3.040000    7.04   14.44      2\n",
       "118  3.200000  2.360000    6.70   10.57      2\n",
       "119  3.885714  2.403073    6.61   14.30      2\n",
       "\n",
       "[120 rows x 5 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Ruta del fichero\n",
    "path = 'datos/datos_filtrados.csv'\n",
    "\n",
    "# Cargar el dataset\n",
    "datos = pd.read_csv(path)\n",
    "\n",
    "# Al ser un dataset pequeño se muestra completo\n",
    "datos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "fe5f16f9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    61\n",
       "0    46\n",
       "2    13\n",
       "Name: Grupo, dtype: int64"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Contar el número de instancias de cada grupo\n",
    "datos['Grupo'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "02f716f6",
   "metadata": {},
   "source": [
    "## Preparar conjuntos de entrenamiento (80%) y test (20%)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "9fc8f7bc",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Se separa la variable a predecir (Grupo)\n",
    "X = datos.drop('Grupo', axis=1)\n",
    "y = datos['Grupo']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "793c7098",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importar librería de scikit-learn para separar los datos\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# Separar datos en conjunto de entrenamiento (80%) y test (20%) de manera estratificada\n",
    "X_train, X_test, y_train, y_test = train_test_split(X,\n",
    "                                                    y,\n",
    "                                                    test_size=0.2,\n",
    "                                                    shuffle=True,\n",
    "                                                    stratify=y,\n",
    "                                                    random_state=semilla)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "34696ba5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Features Shape: (96, 4)\n",
      "Training Labels Shape: (96,)\n",
      "Testing Features Shape: (24, 4)\n",
      "Testing Labels Shape: (24,)\n"
     ]
    }
   ],
   "source": [
    "# Visualizar la dimensión de los datos separados\n",
    "print('Training Features Shape:', X_train.shape)\n",
    "print('Training Labels Shape:', y_train.shape)\n",
    "print('Testing Features Shape:', X_test.shape)\n",
    "print('Testing Labels Shape:', y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "4f900cab",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    49\n",
       "0    37\n",
       "2    10\n",
       "Name: Grupo, dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Distribución de los grupos en el conjunto de entrenamiento\n",
    "y_train.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "5c1cd37d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    0.510417\n",
       "0    0.385417\n",
       "2    0.104167\n",
       "Name: Grupo, dtype: float64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Valores normalizados\n",
    "y_train.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "09a26af2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    12\n",
       "0     9\n",
       "2     3\n",
       "Name: Grupo, dtype: int64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Distribución de los grupos en el conjunto de test\n",
    "y_test.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "2f3bd577",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    0.500\n",
       "0    0.375\n",
       "2    0.125\n",
       "Name: Grupo, dtype: float64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Valores normalizados\n",
    "y_test.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c55e6fd7",
   "metadata": {},
   "source": [
    "### Visualizar distribución de cada conjunto (entrenamiento y test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "69ee43a4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "fig = plt.figure(figsize=(15,5))\n",
    "\n",
    "#fig.suptitle(\"Distribución de los grupos en los conjuntos de entrenamiento y test\")\n",
    "\n",
    "# Representación del conjunto de entrenamiento\n",
    "plt.subplot(121)\n",
    "y_train.value_counts().plot(kind='bar')\n",
    "plt.title('Distribución del conjunto de entrenamiento')\n",
    "plt.xlabel('Grupo')\n",
    "plt.xticks(rotation=0)\n",
    "\n",
    "# Representación del conjunto de test\n",
    "plt.subplot(122)\n",
    "y_test.value_counts().plot(kind='bar', color='seagreen')\n",
    "plt.title('Distribución del conjunto de test')\n",
    "plt.xlabel('Grupo')\n",
    "plt.xticks(rotation=0)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9c9bd050",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "# 2. Crear el modelo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "f073e52f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=StratifiedKFold(n_splits=8, random_state=1234, shuffle=True),\n",
       "             estimator=KNeighborsClassifier(), n_jobs=7,\n",
       "             param_grid={'algorithm': ['auto', 'ball_tree', 'kd_tree', 'brute'],\n",
       "                         'leaf_size': array([10, 20, 30, 40, 50]),\n",
       "                         'n_neighbors': array([1, 3, 5, 7, 9]),\n",
       "                         'weights': ['uniform', 'distance']},\n",
       "             return_train_score=True, scoring='accuracy')"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.model_selection import GridSearchCV, StratifiedKFold\n",
    "import multiprocessing\n",
    "\n",
    "# Definición del modelo\n",
    "modelo = KNeighborsClassifier()\n",
    "\n",
    "# Definición de parámetros de búsqueda\n",
    "parametros = {'n_neighbors': np.arange(1,10,2),\n",
    "              'weights'    : ['uniform', 'distance'],\n",
    "              'algorithm'  : ['auto', 'ball_tree', 'kd_tree', 'brute'],\n",
    "              'leaf_size'  : np.arange(10,51,10)}\n",
    "\n",
    "# Definición de los folds estratificados\n",
    "folds = StratifiedKFold(n_splits=8, shuffle=True, random_state=semilla)\n",
    "\n",
    "# Búsqueda de parámetros\n",
    "grid = GridSearchCV(\n",
    "    estimator  = modelo,\n",
    "    param_grid = parametros,\n",
    "    scoring    = 'accuracy',\n",
    "    n_jobs     = multiprocessing.cpu_count() - 1,\n",
    "    cv         = folds,\n",
    "    verbose    = 0,\n",
    "    return_train_score = True\n",
    "    )\n",
    "\n",
    "# Entrenamiento\n",
    "grid.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "9eb770f3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>params</th>\n",
       "      <th>rank_test_score</th>\n",
       "      <th>mean_test_score</th>\n",
       "      <th>std_test_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>{'algorithm': 'ball_tree', 'leaf_size': 50, 'n...</td>\n",
       "      <td>1</td>\n",
       "      <td>0.760417</td>\n",
       "      <td>0.065052</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>{'algorithm': 'auto', 'leaf_size': 40, 'n_neig...</td>\n",
       "      <td>1</td>\n",
       "      <td>0.760417</td>\n",
       "      <td>0.065052</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>{'algorithm': 'auto', 'leaf_size': 50, 'n_neig...</td>\n",
       "      <td>1</td>\n",
       "      <td>0.760417</td>\n",
       "      <td>0.065052</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59</th>\n",
       "      <td>{'algorithm': 'ball_tree', 'leaf_size': 10, 'n...</td>\n",
       "      <td>1</td>\n",
       "      <td>0.760417</td>\n",
       "      <td>0.065052</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>69</th>\n",
       "      <td>{'algorithm': 'ball_tree', 'leaf_size': 20, 'n...</td>\n",
       "      <td>1</td>\n",
       "      <td>0.760417</td>\n",
       "      <td>0.065052</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               params  rank_test_score  \\\n",
       "99  {'algorithm': 'ball_tree', 'leaf_size': 50, 'n...                1   \n",
       "39  {'algorithm': 'auto', 'leaf_size': 40, 'n_neig...                1   \n",
       "49  {'algorithm': 'auto', 'leaf_size': 50, 'n_neig...                1   \n",
       "59  {'algorithm': 'ball_tree', 'leaf_size': 10, 'n...                1   \n",
       "69  {'algorithm': 'ball_tree', 'leaf_size': 20, 'n...                1   \n",
       "\n",
       "    mean_test_score  std_test_score  \n",
       "99         0.760417        0.065052  \n",
       "39         0.760417        0.065052  \n",
       "49         0.760417        0.065052  \n",
       "59         0.760417        0.065052  \n",
       "69         0.760417        0.065052  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Visualización de los resultados de la búsqueda ordenados de mejor a peor\n",
    "df_resultados = pd.DataFrame(grid.cv_results_)\n",
    "df_resultados = df_resultados.sort_values(by=['rank_test_score'])\n",
    "df_resultados[['params', 'rank_test_score', 'mean_test_score', 'std_test_score']].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "7368512f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mejores parámetros:  {'algorithm': 'auto', 'leaf_size': 10, 'n_neighbors': 9, 'weights': 'distance'}\n"
     ]
    }
   ],
   "source": [
    "# Se muestran los mejores parámetros\n",
    "print('Mejores parámetros: ', grid.best_params_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "10056ced",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.7604166666666667\n"
     ]
    }
   ],
   "source": [
    "# Se visualiza la exactitud (accuracy) del entrenamiento para compararlo con el que se obtenga en la evaluación\n",
    "print(grid.best_score_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "579a9751",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Se toma el modelo que ha obtenido el mejor resultado\n",
    "modelo_knn = grid.best_estimator_"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "43e38a29",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "# 3. Evaluar el modelo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "b93ca4fa",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Se lleva a cabo la predicción para evaluar el modelo\n",
    "y_pred = modelo_knn.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "39f54714",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1 1 1 0 1 0 1 1 1 1 1 1 1 1 0 1 0 0 1 0 0 1 1 1]\n"
     ]
    }
   ],
   "source": [
    "# Visualización de las predicciones\n",
    "print(y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "88b10bc9",
   "metadata": {},
   "source": [
    "## Matriz de confusión"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "1e7a81e5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATIAAAEGCAYAAADmLRl+AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAAAaZUlEQVR4nO3de5zd873v8dc7mZCpaIRcCCGCIpTIxaVI49qyW3Ure5+wm960WjsHBwelbnv3EFvZiqL2EcemSnF6XNpSgqpq5IZEW200oZFGMHWJkJk1n/PH7zeyjDVr1iQzs37feD8fj3lkrd/18/tl5Z3v77t+v+8oIjAzS1mfehdgZra2HGRmljwHmZklz0FmZslzkJlZ8hrqXcC6ov+Gg2KDIcPrXUZhDf5Yv3qXUHgD1vc/x87MmTP71YgY0n66z1w32WDIcA698NZ6l1FYX52wRb1LKLw9t9mk3iUUXmM/La403ZeWZpY8B5mZJc9BZmbJc5CZWfIcZGaWPAeZmSXPQWZmyXOQmVnyHGRmljwHmZklz0FmZslzkJlZ8hxkZpY8B5mZJc9BZmbJc5CZWfIcZGaWPAeZmSXPQWZmyXOQmVnyHGRmljwHmZklz0FmZslzkJlZ8hxkZpY8B5mZJc9BZmbJc5CZWfIcZGaWPAeZmSWvod4FWPdr7NeHL+0+guED+0PA9Jkv8cJr79S7rEJYtaqZ0y+4kebmFkqtreyzx2iO/+L+9S6rcH71xHOcddlPKbW2cvwXPsUpUw6ud0lVJRFkkjYFrgAmAO8Bi4CTI+L5Lm7nZOD6iOjSv2pJb0fEgK6sU0//OHZz5i99i2t/s5i+fcR6fVXvkgqjX78GLj73SzT2X5+WlhKnnfefjB+zHTtuN6LepRVGqdTK6dNu5+6rTmL4sI3Y/0uXcsjET7LDqM3qXVqHCn9pKUnA3cAjEbFNRIwGzgaGrcHmTgY+1sF++q5xkQXSv6EPnxiyAY+/8DoApdZgZXNrnasqDkk09l8fgJZSiZZSK8JBX272gkWMGjGYkVsMZr1+DRx50Fjuf/SZepdVVQotsv2A5oi4tm1CRMxT5lLgECCAf42In0iaBJwPvArsDMwGjgP+BRgOzJD0akTsJ+lt4PvAZ4D/IWl34Cv5bm6IiCt64fi61ZAB6/HWeyW+vMcIttiokcWvv8Ntc15mVclh1qbU2srUs67j5b+9zucOnsAO221R75IKZenyN9h82KD33w8fNojZ8xfVr6AaFL5Fxuowau9IYAywK3AgcKmktrbvbmStr9HAKGDviLgSeBnYLyL2y5fbAJgfEXsAK4EvA3sAewJfl7RbTxxQT+ojseWgRh7502tc9Mvnea+llUNGD613WYXSt08frr7kRG6+5lSeX7iERS8tq3dJhRIRH5qmgjdaUwiyjuwD/DgiShGxDHiUrA8NYGZE/DUiWoF5wMgOtlEC7izb3t0RsSIi3gbuAvatVoCkEyTNkjTrvTeb1u5ouknTymaaVjbzl9ezbsA5f32DLQc11rmqYhqwQSO7jB7JrHl/rncphTJ86EYsWbb68/zysiY2HTywjhV1LoUgWwCMqzC92v8R75W9LtHxJfS7EVGqYXsVRcT1ETE+Isav//FBna/QC958t4Wmd1YxbMOsH2iHYQNY+sa7da6qOP7+5greXrESgPdWNTP32RcYMXxwnasqlrGjt2Lhi8tZvORVVjW3cNeDczhk4i71LquqFPrIHga+J+nrEfEjAEkTgCbgWEk3ARsDE4HTgR2qbOstYEOy/rP2HgOmS7qYLNSOAI7vtqPoRT+evYSv7bUlDX3E8rdXMf13L9W7pMJoanqLf//h3bS2BtEa7LvXTuwxbvt6l1UoDQ19mXbGMRw19WpKpWDyYXuy4zbF/cYSEgiyiAhJRwBXSDoTeJf89gtgAPA0WWf/GRHxN0nVgux64OeSlpb1k7XtZ46k6cDMfNINETG3Ww+ml7z093f5twf+VO8yCmnrrTbl6otPrHcZhXfw3jtx8N471buMmqlSx5513SajdopDL7y13mUU1lcn+JvBzuy5zSb1LqHwGvtpdkSMbz89hT4yM7OqHGRmljwHmZklz0FmZslzkJlZ8hxkZpY8B5mZJc9BZmbJc5CZWfIcZGaWPAeZmSXPQWZmyXOQmVnyHGRmljwHmZklz0FmZslzkJlZ8hxkZpY8B5mZJc9BZmbJc5CZWfIcZGaWPAeZmSXPQWZmyXOQmVnyHGRmljwHmZklz0FmZslzkJlZ8hxkZpY8B5mZJa+h3gWsK7YY2J+LP7djvcsorG33O7XeJRRe01NX1buEZLlFZmbJc5CZWfIcZGaWPAeZmSXPQWZmyXOQmVnyHGRmljwHmZklz0FmZslzkJlZ8hxkZpY8B5mZJc9BZmbJ63D0C0k/AKKj+RExtUcqMjPromrD+MzqtSrMzNZCh0EWETeVv5e0QUSs6PmSzMy6ptM+Mkl7SXoO+H3+fldJ1/R4ZWZmNaqls/8K4DPAawAR8TQwsQdrMjPrkpq+tYyIl9pNKvVALWZma6SWMftfkvQpICStB0wlv8w0MyuCWlpk3wS+DWwOLAHG5O/NzAqh0xZZRLwKTO6FWszM1kgt31qOknSPpOWSXpH0M0mjeqM4M7Na1HJpeStwO7AZMBy4A/hxTxZlZtYVtQSZIuLmiGjJf/6LKo8umZn1tmrPWm6cv5wh6UzgNrIAOxa4rxdqMzOrSbXO/tlkwaX8/TfK5gVwUU8VZWbWFdWetdy6NwsxM1tTtdwQi6SdgdFA/7ZpEfF/eqooM7Ou6DTIJJ0HTCILsvuBQ4DHAQeZmRVCLd9aHg0cAPwtIr4M7Aqs36NVmZl1QS1BtjIiWoEWSR8HXgF8Q2xBnXnJbex+xHc55MvT6l1Kofzg3Mk8/8v/xRO3nf3+tC8csBtP/OQ7vPa7Kxmz45Z1rK54fvXEc0w46kLGHnE+l09/oN7ldKqWIJslaSPgR2TfZM4BZna2kqSSpHmSFkh6WtKpkvrk88ZLurKT9adIuqqG+srXObvzpWrazkRJcyS1SDq6O7bZW4787AT+9yUn1LuMwvnxvU9y9NSrPzDt9wtf5p/P+BFPzF1Yp6qKqVRq5fRpt3PHf3yLJ28/hzsfmM0fXlha77KqquVZy2/lL6+V9Avg4xHxTA3bXhkRYwAkDSV7QmAgcF5EzKJnhtI+G/heN2znRWAKcFo3bKtX7b7rNvz1b6/Xu4zCeWLuQkZstvEHpj2/aFmdqim22QsWMWrEYEZuMRiAIw8ay/2PPsMOozarc2Ud67BFJmls+x9gY6Ahf12ziHgFOAE4SZlJku7N97O7pCckzc3/3L5s1RGSfiHpj/mXDm21HSdpZt7iu05SX0kXA435tFuqLNdX0nRJ8yU9K+mUCvUuysO6tSvHabYuWLr8DTYfNuj998OHDWLp8jfqWFHnqrXILqsyL4D9u7KjiHghv7Qc2m7WH4CJEdEi6UCyFtVR+bzdgZ2Bd4CnJN0HrCB7umDviGjOh92eHBFnSjqprBW4Y6XlgAXA5hGxc77cRl05jnKSTiALaDbfwn0stm6I+PATiFKFBQuk2g2x+/XA/iqdjoHATZK2IwvIfmXzHoyI1wAk3QXsA7QA48iCDaCR7AuI9g7oYLl7gFH5r7u7D1jjnsyIuB64HmDX3cb5+VNbJwwfuhFLljW9//7lZU1sOnhgHSvqXK/9gt586J8SHw6di4AZeQvp85TddMuHH05ve2TqpogYk/9sHxHnV9plpeUioonsFpJHyAaIvGFtj81sXTJ29FYsfHE5i5e8yqrmFu56cA6HTNyl3mVVVdOd/WtL0hDgWuCqiAh9sJ06kGzkWcg62MsdlD+8vhI4HPgK2WXmzyRdHhGv5PM3jIjFQLOkfhHRDDxUaTmyS9NVEXGnpIXA9B445Lo5+aKb+d28P9P0xgr2/uIF/Pcpn+GYf9iz3mXV3Q3/OoW9x23HJhsNYP69F3Hx9ffT9OYKLjntiwweNICfXP5Nnn1+yYe+2fwoamjoy7QzjuGoqVdTKgWTD9uTHbcpbkc/9GyQNUqaR3ap2ALcDHy/wnLTyC4tTwUebjfv8Xy9bYFb8287kXQO8EDe59ZM1rJaTHaZ94ykORExuYPlVgI3tt0KApzVviBJE4C7gUHA5yVdEBE7reF56FVXnHt8vUsopK+dM73i9PseqeUL+I+eg/feiYP3TuIjD2RjjVVfIGs+TQZGRcSFkrYENo2ITu8l+yjZdbdxcf+M39a7jMLadr9T611C4TU91aXbJj+SGvtpdkSMbz+9lj6ya4C9gH/K378FuP1tZoVRy6XlHhExVtJcgIhoyn8tnJlZIdTSImuW1Jf8G8S84943ippZYdQSZFeSdXwPlfRvZB3w3fEYkJlZt6jlWctbJM0mu8FUwOER4d80bmaFUcvAiluS3bt1T/m0iHixJwszM6tVLZ3997H6jvr+wNbAH4F0bjIxs3VaLZeWnyx/n4988Y0OFjcz63VdftYyIuYAE3qgFjOzNVJLH1n5Ldl9gLHA8h6ryMysi2rpI9uw7HULWZ/ZnT1TjplZ11UNsvxG2AERcXov1WNm1mXVhrpuiIgS2aWkmVlhVWuRzSQLsXmS/h9wB9lYXgBExF09XJuZWU1q6SPbGHiNbIz+tvvJAnCQmVkhVAuyofk3lvNZHWBtPD69mRVGtSDrCwyg8i8McZCZWWFUC7KlEXFhr1ViZraGqt3ZX/DfZGdmlqkWZAf0WhVmZmuhwyCLiNd7sxAzszXVa7+g18yspzjIzCx5DjIzS56DzMyS5yAzs+Q5yMwseQ4yM0ueg8zMkucgM7Pk1TIemdWgoY/YZMB69S6jsH5+2wX1LsHWYW6RmVnyHGRmljwHmZklz0FmZslzkJlZ8hxkZpY8B5mZJc9BZmbJc5CZWfIcZGaWPAeZmSXPQWZmyXOQmVnyHGRmljwHmZklz0FmZslzkJlZ8hxkZpY8B5mZJc9BZmbJc5CZWfIcZGaWPAeZmSXPQWZmyXOQmVnyHGRmljwHmZklz0FmZslzkJlZ8hxkZpY8B5mZJa+h3gVY9/vVE89x1mU/pdTayvFf+BSnTDm43iUVxqpVzZx+wY00N7dQam1lnz1Gc/wX9693WYWT2meox1pkkkqS5klaIOlpSadK6pPPGy/pyk7WnyLpqi7u8+y1qblsO6dKek7SM5IekrRVd2y3N5RKrZw+7Xbu+I9v8eTt53DnA7P5wwtL611WYfTr18DF536Ja6Z9i6svPpHZ8/7M7//0Ur3LKpQUP0M9eWm5MiLGRMROwEHAocB5ABExKyKm9sA+uyXIgLnA+IjYBfgpMK2bttvjZi9YxKgRgxm5xWDW69fAkQeN5f5Hn6l3WYUhicb+6wPQUirRUmpFqM5VFUuKn6Fe6SOLiFeAE4CTlJkk6V4ASbtLekLS3PzP7ctWHSHpF5L+KOm8tomSjpM0M2/xXSepr6SLgcZ82i1Vlusrabqk+ZKelXRKhXpnRMQ7+dsngS166tx0t6XL32DzYYPefz982CCWLn+jjhUVT6m1lW//zx/yTydcym6fHMUO2yXz19srUvwM9VofWUS8kF9aDm036w/AxIhokXQg8D3gqHze7sDOwDvAU5LuA1YAxwJ7R0SzpGuAyRFxpqSTImIMgKQdKy0HLAA2j4id8+U26qT0rwI/X5tj700R8aFpcoPjA/r26cPVl5zI2ytWctFlt7HopWWMHDGs3mUVRoqfod7u7K90OgYCN0naDgigX9m8ByPiNQBJdwH7AC3AOLJgA2gEXqmw3QM6WO4eYJSkHwD3AQ90WKx0HDAe+HQH808ga2kyYsstO9pMrxo+dCOWLGt6//3Ly5rYdPDAOlZUXAM2aGSX0SOZNe/PDrIyKX6Geu32C0mjgBIfDp2LgBl5C+nzQP+yee3/awiyMLwp738bExHbR8T5lXZZabmIaAJ2BR4Bvg3c0EG9BwLfAQ6LiPcqLRMR10fE+IgYP2TwkI4PvheNHb0VC19czuIlr7KquYW7HpzDIRN3qXdZhfH3N1fw9oqVALy3qpm5z77AiOGD61xVsaT4GeqVFpmkIcC1wFUREfpgO3UgsCR/PaXdqgdJ2hhYCRwOfIXsMvNnki6PiFfy+RtGxGKgWVK/iGgGHqq0HNml6aqIuFPSQmB6hXp3A64DPpv37yWjoaEv0844hqOmXk2pFEw+bE923GazepdVGE1Nb/HvP7yb1tYgWoN999qJPcZt3/mKHyEpfoZU6Xq4WzYslYBnyS4VW4Cbge9HRKukScBpEfE5SXsBNwHLgYeB4yNipKQpZN90bgBsC9waERfk2z4WOIusRdkMfDsinpR0CXAYMCciJldajiwUb2R1a/SsiPhAH5ikXwGfBNq+c34xIg6rdrzjxo2P3/xuVtdP1EfEkwtfq3cJhbfnNpvUu4TCa+yn2RExvv30HguyjxoHWXUOss45yDrXUZD5ESUzS56DzMyS5yAzs+Q5yMwseQ4yM0ueg8zMkucgM7PkOcjMLHkOMjNLnoPMzJLnIDOz5DnIzCx5DjIzS56DzMyS5yAzs+Q5yMwseQ4yM0ueg8zMkucgM7PkOcjMLHkOMjNLnoPMzJLnIDOz5DnIzCx5DjIzS56DzMyS5yAzs+Q5yMwseQ4yM0ueg8zMkucgM7PkKSLqXcM6QdJyYHG96ygzGHi13kUUnM9RdUU8P1tFxJD2Ex1k6yhJsyJifL3rKDKfo+pSOj++tDSz5DnIzCx5DrJ11/X1LiABPkfVJXN+3EdmZslzi8zMkucgM7PkOcgKStKmkm6TtFDSc5Lul/SJNdjOyZI+tgbrvd3VdbqLpJKkeZIWSHpa0qmS+uTzxku6spP1p0i6qov7PHttai7bzkRJcyS1SDq6O7ZZYR8pn59T88/zM5IekrRVd2zXQVZAkgTcDTwSEdtExGjgbGDYGmzuZKBikEnqu8ZF9qyVETEmInYCDgIOBc4DiIhZETG1B/bZLf9QgReBKcCt3bS9SlI+P3OB8RGxC/BTYFp3bNRBVkz7Ac0RcW3bhIiYBzwu6VJJ8yU9K+lYAEmTJD0i6aeS/iDpFmWmAsOBGZJm5Mu+LelCSb8D9sr/h5yf/5zc60faiYh4BTgBOCk/pkmS7gWQtLukJyTNzf/cvmzVEZJ+IemPks5rmyjpOEkz8xbNdZL6SroYaMyn3VJlub6Spped/1Mq1LsoIp4BWnv0xKzeX2rnZ0ZEvJO/fRLYortOhH8K9gNMBS6vMP0o4EGgL1nr7EVgM2AS8Eb+oegD/BbYJ19nETC4bBsBHJO/Hgc8C2wADAAWALvl896u4/F/aN9AU37Mk4B782kfBxry1wcCd+avpwBLgU2ARmA+MB7YEbgH6Jcvdw3wz+332dFy+fl6sGy5jaocw3TgaJ+fqsdxFXBOd5yTBiwl+wA/jogSsEzSo8AE4E1gZkT8FUDSPGAk8HiFbZSAO8u2d3dErMjXuwvYl6z5XzSqMG0gcJOk7cgCul/ZvAcj4jV4/7j2AVrI/rE9lV290wi8UmG7B3Sw3D3AKEk/AO4DHlj7w+o2SZ0fSceRheenaz/EjjnIimkBUKmjuNKHtc17Za9LdPx3+24ehJ1trzAkjSI7plfIWgNtLgJmRMQRkkYCj5TNa3+DZJAd700RcVZnu+xoOUm7Ap8Bvg0cA3yl9iPpGamdH0kHAt8BPh0R77WfvybcR1ZMDwPrS/p62wRJE8guH47N+yKGABOBmZ1s6y1gww7mPQYcLuljkjYAjgB+vdbVd6P8OK8Fror8eqTMQGBJ/npKu3kHSdpYUiNwOPAb4CHgaElD821vXPatWbOkthZLxeUkDQb6RMSdwLnA2O46zjWV2vmRtBtwHXBYZP173cItsgKKiJB0BHCFpDOBd8n6uk4m68t6mux/0DMi4m+SdqiyueuBn0taGhH7tdvPHEnTWR2GN0REES4rG/PL435klzs3A9+vsNw0skunU8nCv9zj+XrbArdGxCwASecADyi7XaGZrOWwmOw8PSNpTkRM7mC5lcCN+TSASi2SCWTfOA8CPi/pgsi+XexOyZ4f4FKyz/Ad+WXpixFx2Bqcgw/wI0pmljxfWppZ8hxkZpY8B5mZJc9BZmbJc5CZWfIcZNZrtHrUhvmS7tAajMpRtq3pykeXkHSDpNFVlp0k6VNrsI9F+b1RNU1vt0yXRg+RdL6k07pao2UcZNab2kZt2BlYBXyzfKbWcDSOiPhaRDxXZZFJQJeDzNLhILN6+TWwbd5amiHpVuDZ/KmFSyU9pWzMqm9ANrSRpKuUjWV1HzC0bUPKRv4Yn7/+rLLxwJ5WNt7VSLLAPCVvDe4raYikO/N9PCVp73zdTSQ9oGy0iOuo4REuSf9X0mxlY4Od0G7eZXktD+V34CNpG2WjTsyW9OtObma2WvXE0/n+8U+lH/IRFMieKPkZcCJZa2kFsHU+7wTyERGA9YFZwNbAkawe+WM48Hfy0SXIniEcDwwBXirb1sb5n+cDp5XVcSurRwfZEvh9/vpK4Lv5638ge3picIXjWNQ2vWwfbaNIbJK/D2By/vq7ZI8QQfZ4z3b56z2AhyvV6J+u/fgRJetNbY/WQNYi+0+yS76ZEfGXfPrBwC5aPbrqQGA7sudK20b+eFlS+0duAPYEHmvbVkS83kEdBwKj80dkAD4uacN8H0fm694nqamGY5qaP04GMCKv9TWy8ch+kk//L+AuSQPy472jbN/r17AP64SDzHrTyogYUz4h/we9onwS8C8R8ct2yx3Kh0dsaE81LANZl8peEbGyQi01P7MnaRJZKO4VEe9IegTo38Hike/37+3Pga0995FZ0fwSOLFtpAVJn1A2MsdjwD/mfWibkY2i295vgU9L2jpfd+N8evsRQB4ATmp7I2lM/vIxYHI+7RCyB7+rGQg05SG2A1mLsE0fVg/F9N+AxyPiTeAvkr6Y70PKhr2xteQgs6K5AXgOmCNpPtmQLw1kI0r8iWxE2x8Cj7ZfMSKWk/Wx3SXpaVZf2t0DHNHW2U82Au/4/MuE51j97ekFwERJc8gucV/spNZfAA2SniEb++vJsnkrgJ0kzQb2By7Mp08GvprXtwD4Qg3nxDrh0S/MLHlukZlZ8hxkZpY8B5mZJc9BZmbJc5CZWfIcZGaWPAeZmSXv/wPmsL0STEf3OwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay, classification_report\n",
    "\n",
    "cm = confusion_matrix(y_test, y_pred)\n",
    "# Filas: valores reales\n",
    "# Columnas: predicción\n",
    "\n",
    "clases = ['Control', 'Diabetes 1', 'Diabetes 2']\n",
    "\n",
    "# Visualización de la matriz de confusión\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=clases)\n",
    "\n",
    "disp.plot(cmap='Blues', colorbar=False)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "38845ca4",
   "metadata": {},
   "source": [
    "## Classification report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "34d743ca",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "     Control     0.8571    0.6667    0.7500         9\n",
      "  Diabetes 1     0.6471    0.9167    0.7586        12\n",
      "  Diabetes 2     0.0000    0.0000    0.0000         3\n",
      "\n",
      "    accuracy                         0.7083        24\n",
      "   macro avg     0.5014    0.5278    0.5029        24\n",
      "weighted avg     0.6450    0.7083    0.6606        24\n",
      "\n"
     ]
    }
   ],
   "source": [
    "clases = ['Control', 'Diabetes 1', 'Diabetes 2']\n",
    "\n",
    "print(classification_report(y_test,\n",
    "                            y_pred,\n",
    "                            target_names=clases,\n",
    "                            digits=4,\n",
    "                            zero_division=0))\n",
    "\n",
    "# El parámetro zero_division no muestra warnings cuando se vaya a realizar una división por 0 al calcular alguna métrica"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f26709f6",
   "metadata": {},
   "source": [
    "Se calcula la __especificidad__ a partir de la clase _multilabel_confusion_matrix_. Se van a obtener también el resto de métricas con las obtenidas en el _classification report_ para comprobar que son correctas.\n",
    "\n",
    "Código basado en la documentación de scikit-learn: https://scikit-learn.org/stable/modules/model_evaluation.html#multilabel-confusion-matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "55a59809",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------\n",
      "Exactitud: \t0.7083\n",
      "----------------------\n",
      "\n",
      "----------------------\n",
      "      Precisión       \n",
      "----------------------\n",
      "Control: \t0.8571\n",
      "Diabetes 1: \t0.6471\n",
      "Diabetes 2: \tnan\n",
      "MACRO AVG: \tnan\n",
      "\n",
      "----------------------\n",
      "     Sensibilidad     \n",
      "----------------------\n",
      "Control: \t0.6667\n",
      "Diabetes 1: \t0.9167\n",
      "Diabetes 2: \t0.0000\n",
      "MACRO AVG: \t0.5278\n",
      "\n",
      "----------------------\n",
      "    Especificidad     \n",
      "----------------------\n",
      "Control: \t0.9333\n",
      "Diabetes 1: \t0.5000\n",
      "Diabetes 2: \t1.0000\n",
      "MACRO AVG: \t0.8111\n",
      "\n",
      "----------------------\n",
      "       F1-score       \n",
      "----------------------\n",
      "Control: \t0.7500\n",
      "Diabetes 1: \t0.7586\n",
      "Diabetes 2: \tnan\n",
      "MACRO AVG: \tnan\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Alberto\\AppData\\Local\\Temp/ipykernel_12308/1740227734.py:14: RuntimeWarning: invalid value encountered in true_divide\n",
      "  precision   = tp / (tp + fp)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import multilabel_confusion_matrix, accuracy_score\n",
    "\n",
    "# Se obtiene la matriz de confusión de cada clase\n",
    "cm_multi = multilabel_confusion_matrix(y_test, y_pred)\n",
    "\n",
    "# Se obtienen los valores de TN, TP, FN y FP de cada clase\n",
    "tn = cm_multi[:, 0, 0]\n",
    "tp = cm_multi[:, 1, 1]\n",
    "fn = cm_multi[:, 1, 0]\n",
    "fp = cm_multi[:, 0, 1]\n",
    "\n",
    "# Se calculan las métricas (array de dimensión 3 con la métrica para cada clase)\n",
    "accuracy    = accuracy_score(y_test, y_pred)\n",
    "precision   = tp / (tp + fp)\n",
    "recall      = tp / (tp + fn)\n",
    "specificity = tn / (tn + fp)\n",
    "f1          = (2 * precision * recall) / (precision + recall)\n",
    "\n",
    "# Visualizar la exactitud del modelo\n",
    "print(\"----------------------\")\n",
    "print(\"Exactitud: \\t{:.4f}\".format(accuracy))\n",
    "print(\"----------------------\")\n",
    "print(\"\")\n",
    "\n",
    "# Visualizar la métrica precision para cada clase y la global (macro avg)\n",
    "print(\"----------------------\")\n",
    "print(\"      Precisión       \")\n",
    "print(\"----------------------\")\n",
    "print(\"Control: \\t{:.4f}\".format(precision[0]))\n",
    "print(\"Diabetes 1: \\t{:.4f}\".format(precision[1]))\n",
    "print(\"Diabetes 2: \\t{:.4f}\".format(precision[2]))\n",
    "print(\"MACRO AVG: \\t{:.4f}\".format(precision.mean()))\n",
    "print(\"\")\n",
    "\n",
    "# Visualizar la sensibilidad para cada clase y la global (macro avg)\n",
    "print(\"----------------------\")\n",
    "print(\"     Sensibilidad     \")\n",
    "print(\"----------------------\")\n",
    "print(\"Control: \\t{:.4f}\".format(recall[0]))\n",
    "print(\"Diabetes 1: \\t{:.4f}\".format(recall[1]))\n",
    "print(\"Diabetes 2: \\t{:.4f}\".format(recall[2]))\n",
    "print(\"MACRO AVG: \\t{:.4f}\".format(recall.mean()))\n",
    "print(\"\")\n",
    "\n",
    "# Visualizar la especificidad para cada clase y la global (macro avg)\n",
    "print(\"----------------------\")\n",
    "print(\"    Especificidad     \")\n",
    "print(\"----------------------\")\n",
    "print(\"Control: \\t{:.4f}\".format(specificity[0]))\n",
    "print(\"Diabetes 1: \\t{:.4f}\".format(specificity[1]))\n",
    "print(\"Diabetes 2: \\t{:.4f}\".format(specificity[2]))\n",
    "print(\"MACRO AVG: \\t{:.4f}\".format(specificity.mean()))\n",
    "print(\"\")\n",
    "\n",
    "# Visualizar la mátrica F1 para cada clase y la global (macro avg)\n",
    "print(\"----------------------\")\n",
    "print(\"       F1-score       \")\n",
    "print(\"----------------------\")\n",
    "print(\"Control: \\t{:.4f}\".format(f1[0]))\n",
    "print(\"Diabetes 1: \\t{:.4f}\".format(f1[1]))\n",
    "print(\"Diabetes 2: \\t{:.4f}\".format(f1[2]))\n",
    "print(\"MACRO AVG: \\t{:.4f}\".format(f1.mean()))\n",
    "print(\"\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.8.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}