Ejemplo-Fashion-MNIST.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The autoreload extension is already loaded. To reload it, use:\n",
      "  %reload_ext autoreload\n"
     ]
    }
   ],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "import numpy as np\n",
    "from helper import visualize_input, plot_images\n",
    "from matplotlib import pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Cargo imagenes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2ea3e44c50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Cargo las sandalias de disco.\n",
    "sandalias_orig = np.load('sandalias.npy')\n",
    "# Las grafico con la función plot_images\n",
    "# plot_images(array con imageners, Cantidad a graficar, columnas)\n",
    "div = 1\n",
    "sandalias = (sandalias_orig/div).astype(int)\n",
    "plot_images(sandalias,150, columns=25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "256"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "niveles = np.max(sandalias) + 1\n",
    "niveles"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2ea3ea1e48>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Lo mismo para los buzos\n",
    "buzos_orig = np.load('buzos.npy')\n",
    "buzos = (buzos_orig/div).astype(int)\n",
    "plot_images(buzos,150, columns=25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dimensiones de sandalias:  (6000, 28, 28)\n",
      "Dimensiones de buzos:  (6000, 28, 28)\n"
     ]
    }
   ],
   "source": [
    "# Imprimo las dimensiones\n",
    "print('Dimensiones de sandalias: ', sandalias.shape)\n",
    "print('Dimensiones de buzos: ', buzos.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Como se interpreta? Hay 6000 imágenes de sandalias de 28x28 pixeles  \n",
    "Lo mismo para los buzos"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Representación de imágenes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Como hago para refererime a la primer sandalia?:  \n",
    "sandalias[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Notar el subinidice en cero\n",
    "primer_sandalia = sandalias[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2ea3e44e80>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Grafico primer sandalia (a esta sandalia la llamaremos sandalia 0)\n",
    "visualize_input(primer_sandalia)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Notar que cada pixel esta representado por un numero entre 0 y 255.\n",
    "- 0 indica negro\n",
    "- 255 indica blanco\n",
    "- Los valores intermedios son los grises entre negro y blanco"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Grafico imagen"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "from matplotlib import pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Podemos graficar cada imagen directamente con la función imshow de matplotlib\n",
    "\n",
    "Por ejemplo grafiquemos la sandalia 10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2eaffa40f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# cmap es lo que se denomina colormap. Le estamos indicando que \n",
    "# grafique los valores entre 0 y 255 como estala de grises\n",
    "plt.imshow(sandalias[10], cmap='gray')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "O el buzo 5:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2ea3e3e5c0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(buzos[5], cmap='gray')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ejercicio 1: \n",
    "Que valor tiene el pixel (20,5) del buzo 18? (Donde 20 es la fila y 5 es la columna)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "216\n"
     ]
    }
   ],
   "source": [
    "print(buzos[18, 20, 5])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ejercicio 2:\n",
    "Imprimir los valores de los 4 pixeles centrales de la sandalia 10:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[0 2]\n",
      " [2 0]]\n"
     ]
    }
   ],
   "source": [
    "print(sandalias[10, 13:15, 13:15])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ejercicio 3:\n",
    "Imprimir los valores de los 4 pixeles centrales del buzo 5:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[242 242]\n",
      " [242 242]]\n"
     ]
    }
   ],
   "source": [
    "print(buzos[5, 13:15, 13:15])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Histograma de una imagen\n",
    "** Nuestra estimación de la función de masa de probabilidad **"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Probabilidad de algún pixel = 0\n",
    "(Alguno de los 784 pixeles)\n",
    "\n",
    "Trabajaremos con la sandalia 10 y el buzo 5"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Que tan probable es que agarre un pixel al azar de la sandalia 10 y me de 0?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Cuantos pixeles tenemos en la sandalia 10?:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(28, 28)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sandalias[10].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "784\n"
     ]
    }
   ],
   "source": [
    "total_pixeles_imagen = sandalias[10].shape[0]*sandalias[10].shape[1]\n",
    "print(total_pixeles_imagen)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Cuantos pixels en cero tiene la sandalia 10?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "528\n"
     ]
    }
   ],
   "source": [
    "pixeles_en_cero_10 = (sandalias[10] == 0).sum()\n",
    "print(pixeles_en_cero_10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "La estimación de la probabilidad es:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.673469387755102\n"
     ]
    }
   ],
   "source": [
    "p_0_sand_10 = pixeles_en_cero_10/total_pixeles_imagen\n",
    "print(p_0_sand_10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Cual es la probabilidad de un cero en el buzo 5?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.341836734694\n"
     ]
    }
   ],
   "source": [
    "# Hacemos los mismo ahora con el buzo 5\n",
    "p_0_buzo_5 = (buzos[5] == 0).sum() / total_pixeles_imagen\n",
    "print(p_0_buzo_5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Histograma de una imagen\n",
    "Ahora no solo trabajemos con la probabilidad de que sea cero sino cualquier otro valor posible en el espacio muestral: [0, 255]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "hist_sandalia_10 = []\n",
    "pixeles = list(range(niveles))\n",
    "for pixel in pixeles:\n",
    "    # notar sandalias[10] == pixel, donde pixel va de 0 a 255\n",
    "    prob_sand_10_pixel = (sandalias[10] == pixel).sum() / total_pixeles_imagen\n",
    "    hist_sandalia_10.append(prob_sand_10_pixel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x114c84fd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Como es de esperar el cero es el mas probable\n",
    "plt.bar(pixeles, hist_sandalia_10)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYQAAAD8CAYAAAB3u9PLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11570a7b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Graficamos desde la segunda muestra para ver detalles del resto del histograma\n",
    "plt.bar(pixeles[1:], hist_sandalia_10[1:])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lo mismo para el buzo 5:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "hist_buzo_5 = []\n",
    "for pixel in pixeles:\n",
    "    prob_buzo_5_pixel = (buzos[5] == pixel).sum() / total_pixeles_imagen\n",
    "    hist_buzo_5.append(prob_buzo_5_pixel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAD8CAYAAACb4nSYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n0AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x115543780>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Notar que a pesar de que el cero es el mas probable, tambien ya se ven altas componentes \n",
    "# en grises claros\n",
    "plt.bar(pixeles, hist_buzo_5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAD8CAYAAACb4nSYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1156672b0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.bar(pixeles[1:], hist_buzo_5[1:])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Histograma de una Clase\n",
    "Mejoramos estimación\n",
    "### Probabilidad de algún pixel = 0\n",
    "En todo el dataset de sandalias o buzos."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Podemos mejorar la probabilidad de tener un pixel en cero para un buzo usando todos los buzos, y lo mismo para las sandalias"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Cual es la estimación de la probabilidad de obtener un cero un cualquier sandalia?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Cuantos pixeles vamos a analizar:\n",
    "\n",
    "Todos los pixeles de las sandalias: 6000x28x28"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4704000\n"
     ]
    }
   ],
   "source": [
    "# como hay la misma cantidad de imagenes para sandalias y buzos, habrá la misma cantidad de pixeles\n",
    "total_pixeles = sandalias.shape[0]*sandalias.shape[1]*sandalias.shape[2]\n",
    "print(total_pixeles)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Notar que a diferencia del anterior aca no indexo a sandalias, ya que quiero todas\n",
    "pixeles_en_cero = (sandalias == 0).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.678858630952\n"
     ]
    }
   ],
   "source": [
    "p_0_sandalia = pixeles_en_cero / total_pixeles\n",
    "print(p_0_sandalia)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Es interesante notar que la primer sandalia era representativa del total de sandalias ya que la estimación de la probabilidad da similar"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Veamos en los buzos:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.35128125\n"
     ]
    }
   ],
   "source": [
    "p_0_buzos = (buzos == 0).sum() / total_pixeles\n",
    "print(p_0_buzos)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Histograma de toda la clase\n",
    "La clase puede ser buzo o sandalia. Y consideramos cualquier valor de gris entre 0 y 255 (No solo el cero)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Para sandalias:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "hist_sandalias = []\n",
    "for pixel in pixeles:\n",
    "    # notar sandalias == pixel\n",
    "    prob_sand_pixel = (sandalias == pixel).sum() / total_pixeles\n",
    "    hist_sandalias.append(prob_sand_pixel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYoAAAD8CAYAAABpcuN4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nhSSpyaCQJDUZFJKkJoNCktT0/wEtM2I7cXIGDwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1154ba1d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Sacamos el cero por claridad\n",
    "plt.bar(pixeles[1:], hist_sandalias[1:])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Para buzos:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "hist_buzos = []\n",
    "for pixel in pixeles:\n",
    "    prob_buzos_pixel = (buzos == pixel).sum() / total_pixeles\n",
    "    hist_buzos.append(prob_buzos_pixel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYQAAAD8CAYAAAB3u9PLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n89gX3gAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x114709a58>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Sacamos el cero por claridad\n",
    "plt.bar(pixeles[1:], hist_buzos[1:])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "** La diferencia entonces de estos histogramas es que son la estimación de la función de masa de probabilidad de los buzos o sandalias hecha sobre todas las imagenes. Los primeros eran sobre la sandalia 10 y el buzo 5 **"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJcAAAJOCAYAAAAK6OY1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nABhNuAQAAADAaP8P8B8bTOeziyUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1157ab358>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "f, ax= plt.subplots(2,1, figsize=(20,10))\n",
    "ax[0].bar(pixeles[1:], hist_sandalias[1:])\n",
    "ax[0].set_title('Sobre todo el dataset de sandalias')\n",
    "ax[1].bar(pixeles[1:], hist_sandalia_10[1:])\n",
    "ax[1].set_title('Sobre una sandalias')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJEAAAJOCAYAAAAH9pZyAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAADAmRAAAAABj6P8W8apUQKA8rAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1155a9ac8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "f, ax= plt.subplots(2,1, figsize=(20,10))\n",
    "ax[0].bar(pixeles[1:], hist_buzos[1:])\n",
    "ax[0].set_title('Sobre todo el dataset de buzos')\n",
    "ax[1].bar(pixeles[1:], hist_buzo_5[1:])\n",
    "ax[1].set_title('Sobre un buzo')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Que podríamos hacer con esto?\n",
    "Cuando me dan una imagen nueva y tengo que clasificarla como buzo o sandalia el procedimiento sería:\n",
    "- Calculo el histograma de la imagen\n",
    "- Lo comparo contra el de las sandalias de todo el dataset\n",
    "- Lo comparo contra el de los buzos de todo el dataset\n",
    "- Si se \"parece más\" al de los buzos es buzo, si no es sandalia.\n",
    "\n",
    "Obviamente este método no es preciso y tampoco definimos que significa comparar dos histogramas.\n",
    "\n",
    "Una de las falencias es que si la imagen esta *invertida*, nosotros seguiriamos reconociendo la imagen pero nuestro algoritmo seguramente fallaría:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIcAAAEuCAYAAAAUWnIaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nFEVRFEVRFKUb8/8AyHBkJL7PX7sAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1160005f8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Sandalias invertidas\n",
    "plot_images(255-sandalias,150, columns=25)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Histograma de un pixel por clase\n",
    "Una posibilidad es hacer un histograma por cada pixel"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nota: hablaremos de probabilidades, pero siempre son estimaciones de las probabilidades"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Cual es la probabilidad de que el pixel (14, 14) sea negro dado que es una sandalia?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.457833333333\n"
     ]
    }
   ],
   "source": [
    "p_14_14_0_sand = (sandalias[:, 14, 14] == 0).sum()/sandalias.shape[0]\n",
    "print(p_14_14_0_sand)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Formalmente:  \n",
    "$P(pix_{14,14}=0 \\ |\\  Sandalia) = 0.457833$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Con los buzos:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.00533333333333\n"
     ]
    }
   ],
   "source": [
    "p_14_14_0_buzos = (buzos[:, 14, 14] == 0).sum()/buzos.shape[0]\n",
    "print(p_14_14_0_buzos)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Formalmente:  \n",
    "$P(pix_{14,14}=0 \\ |\\  Buzo) = 0.00533$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Como lo usamos:\n",
    "\n",
    "Supongamos que nos dan una imagen, observo el punto 14, 14 y es un cero. Cual elijo?\n",
    "\n",
    "Buzo o sandalia?\n",
    "\n",
    "En este caso es claro intuitivamente, pero formalmente queremos ver si la probabilidad de la sandalia dado que observe un cero en la posición 14, 14 es mas alta que la del buzo:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "¿$P(Sandalia \\ |\\  pix_{14,14}=0) > P(Buzo \\ |\\  pix_{14,14}=0)?$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Por bayes:\n",
    "\n",
    "$P(Sandalia \\ |\\  pix_{14,14}=0) = \\huge\\frac{P(pix_{14,14}=0 \\ |\\ Sandalia ) P(Sandalia)}{P(pix_{14,14}=0)}$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.23158333333333334"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Probabilidad total\n",
    "P_pix_14_14_0 = ((sandalias[:, 14, 14]==0).sum() + (buzos[:, 14, 14]==0).sum()) / (len(sandalias) + len(buzos))\n",
    "P_pix_14_14_0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Probabilidad a priori\n",
    "P_sandalia = len(sandalias)/(len(sandalias)+len(buzos)) #6000/12000\n",
    "P_sandalia"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.988485066571\n"
     ]
    }
   ],
   "source": [
    "P_sandialia_pix_14_14_0 = p_14_14_0_sand*P_sandalia/P_pix_14_14_0\n",
    "print(P_sandialia_pix_14_14_0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Para el buzo:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Probabilidad a priori\n",
    "P_buzo = len(buzos)/(len(sandalias)+len(buzos)) #6000/12000\n",
    "P_buzo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0115149334293\n"
     ]
    }
   ],
   "source": [
    "P_buzo_pix_14_14_0 = p_14_14_0_buzos*P_buzo/P_pix_14_14_0\n",
    "print(P_buzo_pix_14_14_0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "P_buzo_pix_14_14_0 + P_sandialia_pix_14_14_0"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Histograma del punto 14, 14\n",
    "Y si observo otro valor?\n",
    "No observamos un cero si no cualquier otro de los 255 posibles restantes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Sandalias"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAD8CAYAAACb4nSYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nhr4kNeT/AXewLHnDeB4zAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11558f940>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "p_14_14_sand = []\n",
    "for pixel in pixeles:\n",
    "    p_14_14_sand.append((sandalias[:, 14, 14] == pixel).sum()/sandalias.shape[0])\n",
    "plt.bar(pixeles[1:], p_14_14_sand[1:])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.array(p_14_14_sand).sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Buzos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYQAAAD8CAYAAAB3u9PLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n3MNDoerlAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1151f8080>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "p_14_14_buzos = []\n",
    "for pixel in pixeles:\n",
    "    p_14_14_buzos.append((buzos[:, 14, 14] == pixel).sum()/buzos.shape[0])\n",
    "plt.bar(pixeles[1:], p_14_14_buzos[1:])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.array(p_14_14_buzos).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "def prob_sand_pixel(valor, posx=14, posy=14):\n",
    "    # Probabilidad de sandalia dado que el punto x, y tiene el valor: valor\n",
    "    total_imagenes = len(sandalias) + len(buzos)\n",
    "    P_sandalia = len(sandalias)/total_imagenes\n",
    "    p_x_y_sand = (sandalias[:, posx, posy] == valor).sum()/len(sandalias)\n",
    "    # Probabilidad total\n",
    "    P_x_y = ((sandalias[:, posx, posy]==valor).sum() \n",
    "                         + (buzos[:, posx, posy]==valor).sum()) / total_imagenes\n",
    "    return p_x_y_sand*P_sandalia/P_x_y\n",
    "\n",
    "def prob_buzo_pixel(valor, posx=14, posy=14):\n",
    "    # Probabilidad de buzo dado que el punto x, y tiene el valor: valor\n",
    "    total_imagenes = len(sandalias) + len(buzos)\n",
    "    P_buzo = len(buzos)/total_imagenes\n",
    "    p_x_y_buzo = (buzos[:, posx, posy] == valor).sum()/len(buzos)\n",
    "    # Probabilidad total\n",
    "    P_x_y = ((sandalias[:, posx, posy]==valor).sum() \n",
    "                         + (buzos[:, posx, posy]==valor).sum()) / total_imagenes\n",
    "    return p_x_y_buzo*P_buzo/P_x_y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.5, 0.5)"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "valor = 251 # Probar con distintos valores\n",
    "prob_sand_pixel(valor), prob_buzo_pixel(valor)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nhAh3AEiIcAeAhP4fY0lV+lGOcOkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x114d21f28>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# La probabilidad del punto (14,14) para todos los valores posibles\n",
    "p_sand_final = [prob_sand_pixel(valor) for valor in pixeles]\n",
    "plt.bar(pixeles, p_sand_final)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Algo bastante inesperado? Si obtengo valores observados altos en la posición (14,14) tambien hay alta probabilidad de que sea sandalia."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x115c50a58>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# La probabilidad del punto (5,5) para todos los valores posibles\n",
    "p_sand_final = [prob_sand_pixel(valor, 5, 5) for valor in pixeles]\n",
    "plt.bar(pixeles, p_sand_final)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Estaría indicando que para el punto (5,5) casi siempre decido por buzo, salvo si el valor observado es 0"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Probabilidad Conjunta: Más de un pixel\n",
    "Como deberíamos hacer para tener en cuenta mas de un pixel. Idealmente todos los pixeles observados"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Analicemos el caso de que obtengo una imagen y veo dos pixeles de la imagen. Con eso quiero determinar si fue una sandalia o un buzo.\n",
    "\n",
    "$X = Pix_{13, 14}$\n",
    "\n",
    "$Y = Pix_{14, 14}$\n",
    "\n",
    "$P(Sandalia\\ |\\ X, Y)\\ P(X, Y) = P(Sandalia, X, Y) = P(X,Y\\ |\\ Sandalia)\\ P(Sandalia)$\n",
    "\n",
    "$P(Sandalia\\ |\\ X, Y)\\ = \\large \\frac{P(X,Y\\ |\\ Sandalia)\\ P(Sandalia)}{P(X, Y)}$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Que es P(X,Y)? Es la probabilidad por ejemplo, de que uno de los puntos sea 0 y el otro 1. O cualquiera de las combinaciones posibles de valores que pueden tomar los puntos X e Y"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$P(X=0, Y=0)\\ =\\ ?$  \n",
    "$P(X=1, Y=0)\\ =\\ ?$  \n",
    "$P(X=2, Y=0)\\ =\\ ?$  \n",
    "...  \n",
    "$P(X=255, Y=0)\\ =\\ ?$  \n",
    "$P(X=0, Y=1)\\ =\\ ?$  \n",
    "$P(X=1, Y=1)\\ =\\ ?$  \n",
    "$P(X=2, Y=1)\\ =\\ ?$  \n",
    "...  \n",
    "$P(X=255, Y=1)\\ =\\ ?$  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Cual es el espacio muestral para 2 pixeles?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "espacio_muestral = np.stack(np.meshgrid(pixeles, pixeles)).T.reshape(-1,2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(65536, 2)"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Notar que la cantidad de posibilidades es la cantidad de intensidades de gris al cuadrado\n",
    "espacio_muestral.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0,  0],\n",
       "       [ 0,  1],\n",
       "       [ 0,  2],\n",
       "       [ 0,  3],\n",
       "       [ 0,  4],\n",
       "       [ 0,  5],\n",
       "       [ 0,  6],\n",
       "       [ 0,  7],\n",
       "       [ 0,  8],\n",
       "       [ 0,  9],\n",
       "       [ 0, 10],\n",
       "       [ 0, 11],\n",
       "       [ 0, 12],\n",
       "       [ 0, 13],\n",
       "       [ 0, 14],\n",
       "       [ 0, 15],\n",
       "       [ 0, 16],\n",
       "       [ 0, 17],\n",
       "       [ 0, 18],\n",
       "       [ 0, 19],\n",
       "       [ 0, 20],\n",
       "       [ 0, 21],\n",
       "       [ 0, 22],\n",
       "       [ 0, 23],\n",
       "       [ 0, 24],\n",
       "       [ 0, 25],\n",
       "       [ 0, 26],\n",
       "       [ 0, 27],\n",
       "       [ 0, 28],\n",
       "       [ 0, 29]])"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "espacio_muestral[:30]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[13 14] [14 14]\n",
      "[0 0]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([ True,  True], dtype=bool)"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# La sandalia cero tiene dos ceros en la posición:\n",
    "# filas 13:14, columna 14\n",
    "X = [13, 14]\n",
    "Y = [14, 14]\n",
    "Rows = np.array([X[0], Y[0]])\n",
    "Cols = np.array([X[1], Y[1]])\n",
    "print(Rows, Cols)\n",
    "print(sandalias[0,Rows,Cols])\n",
    "sandalias[0,Rows,Cols] == [0, 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ True,  True],\n",
       "       [False,  True],\n",
       "       [False,  True],\n",
       "       ..., \n",
       "       [False, False],\n",
       "       [ True,  True],\n",
       "       [ True,  True]], dtype=bool)"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Todas las sandalias y su coincidencia con un [0, 0]\n",
    "sandalias[:,Rows,Cols] == [0, 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2, 1, 1, ..., 0, 2, 2])"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Sumo los true en el eje 1\n",
    "(sandalias[:,Rows,Cols] == [0, 0]).sum(axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1926"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Los comparo con 2 por que los dos tienen que coincidir (Es la conjunta)\n",
    "((sandalias[:,Rows,Cols] == [0,0]).sum(axis=1) == 2).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.32100000000000001"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "P_0_0_sand = ((sandalias[:,Rows,Cols] \n",
    "               == [0,0]).sum(axis=1) == 2).sum()/sandalias.shape[0]\n",
    "P_0_0_sand"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "def prob_2_pixels(clase, pixels, X = [13, 14], Y = [14, 14]):\n",
    "    # Clase puede ser sandalias o buzos. Es el listado de imagenes\n",
    "    # Pixels son los valores de los pixeles adyacentes\n",
    "    Rows = np.array([X[0], Y[0]])\n",
    "    Cols = np.array([X[1], Y[1]])\n",
    "    return ((clase[:,Rows,Cols] == pixels).sum(axis=1) == 2).sum()/clase.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.32100000000000001"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Probabilidad de observar los pixeles 0, 0 en las posiciones \n",
    "# 13, 14 vertical, 14 horizontal dado que es sandalia\n",
    "prob_2_pixels(sandalias, [0,0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Probabilidad de que el punto X sea igual a 140 y el Y sea igual a 120\n",
    "prob_2_pixels(sandalias, [140,120], X = [10, 10], Y = [10, 11])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0018333333333333333"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Ahora para los buzos\n",
    "# Probabilidad de observar los pixeles 0, 0 en las posiciones \n",
    "# 13, 14 vertical, 14 horizontal dado que es buzo\n",
    "prob_2_pixels(buzos, [0,0], X = [13, 14], Y = [14, 14])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculo de toda la conjunta\n",
    "def get_joint(clase, X = [13, 14], Y = [14, 14]):\n",
    "    conjunta_clase = []\n",
    "    # Recorro todo el espacio muestral\n",
    "    for pixs in espacio_muestral:\n",
    "        conjunta_clase.append(prob_2_pixels(clase, pixs, X = X, Y = Y))\n",
    "    return np.array(conjunta_clase)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 8.32 s, sys: 47.8 ms, total: 8.36 s\n",
      "Wall time: 8.52 s\n"
     ]
    }
   ],
   "source": [
    "%time conjunta_buzos = get_joint(buzos, X = [13, 14], Y = [14, 14])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATsAAAEyCAYAAACF03cPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1146c4ba8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATUAAAEyCAYAAACbGke8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11510bef0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(5,5))\n",
    "plt.imshow(conjunta_buzos.reshape(256, 256), cmap='gray')\n",
    "plt.show()\n",
    "plt.figure(figsize=(5,5))\n",
    "plt.imshow(conjunta_buzos.reshape(256, 256)[210:250, 210:250], cmap='gray')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 7.97 s, sys: 34.9 ms, total: 8.01 s\n",
      "Wall time: 8.09 s\n"
     ]
    }
   ],
   "source": [
    "%time conjunta_sand = get_joint(sandalias, X = [13, 14], Y = [14, 14])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAT8AAAEyCAYAAACMONd1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nDD9JTTL8JDXp/wDEIiuwlfOdogAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x115124f98>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAT8AAAEyCAYAAACMONd1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1158abda0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(5,5))\n",
    "conjunta_sand_mod = conjunta_sand.copy()\n",
    "conjunta_sand_mod[0] = 0\n",
    "plt.imshow(conjunta_sand.reshape(256, 256)[:20, :20], cmap='gray')\n",
    "plt.show()\n",
    "plt.figure(figsize=(5,5))\n",
    "plt.imshow(conjunta_sand_mod.reshape(256, 256)[:20, :20], cmap='gray')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Problemas"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Ya empieza a ser significativo el tiempo de calculo de la conjunta con solo 2 pixeles\n",
    "- Que tan buena es la estimación?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "62916"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# De cuantas combinaciones de puntos no tengo información?\n",
    "prob_cero_n = (np.array(conjunta_sand) == 0).sum()\n",
    "prob_cero_n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.96002197265625"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prob_cero_n/(len(pixeles)*len(pixeles))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "El 96% da probabilidad cero"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2620"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dist_cero = len(pixeles)*len(pixeles) - prob_cero_n\n",
    "dist_cero"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Tenemos solo 2620 salidas de nuestro espacio muestral de 65536 y tenemos que hacer estadistica con 6000 muestras!!!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2.2900763358778624"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(sandalias)/dist_cero"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "en promedio tengo menos de 3 ocurrencias por punto"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 0]\t0.321\n",
      "[0 1]\t0.0388333333333\n",
      "[0 2]\t0.0176666666667\n",
      "[0 3]\t0.00883333333333\n",
      "[0 4]\t0.0045\n",
      "[0 5]\t0.0045\n",
      "[0 6]\t0.00183333333333\n",
      "[0 7]\t0.0\n",
      "[0 8]\t0.00133333333333\n",
      "[0 9]\t0.0005\n",
      "[ 0 10]\t0.0005\n",
      "[ 0 11]\t0.000666666666667\n",
      "[ 0 12]\t0.000666666666667\n",
      "[ 0 13]\t0.000666666666667\n",
      "[ 0 14]\t0.000333333333333\n",
      "[ 0 15]\t0.001\n",
      "[ 0 16]\t0.000333333333333\n",
      "[ 0 17]\t0.0\n",
      "[ 0 18]\t0.0005\n",
      "[ 0 19]\t0.000666666666667\n",
      "[ 0 20]\t0.000333333333333\n",
      "[ 0 21]\t0.000333333333333\n",
      "[ 0 22]\t0.0\n",
      "[ 0 23]\t0.0\n",
      "[ 0 24]\t0.000833333333333\n",
      "[ 0 25]\t0.0005\n",
      "[ 0 26]\t0.000166666666667\n",
      "[ 0 27]\t0.000166666666667\n",
      "[ 0 28]\t0.0\n",
      "[ 0 29]\t0.000666666666667\n",
      "[ 0 30]\t0.000333333333333\n",
      "[ 0 31]\t0.000333333333333\n",
      "[ 0 32]\t0.000666666666667\n",
      "[ 0 33]\t0.000166666666667\n",
      "[ 0 34]\t0.0\n",
      "[ 0 35]\t0.000166666666667\n",
      "[ 0 36]\t0.0\n",
      "[ 0 37]\t0.000833333333333\n",
      "[ 0 38]\t0.000666666666667\n",
      "[ 0 39]\t0.0005\n",
      "[ 0 40]\t0.000333333333333\n",
      "[ 0 41]\t0.000333333333333\n",
      "[ 0 42]\t0.000333333333333\n",
      "[ 0 43]\t0.0\n",
      "[ 0 44]\t0.000166666666667\n",
      "[ 0 45]\t0.000166666666667\n",
      "[ 0 46]\t0.000333333333333\n",
      "[ 0 47]\t0.000166666666667\n",
      "[ 0 48]\t0.000333333333333\n",
      "[ 0 49]\t0.0005\n"
     ]
    }
   ],
   "source": [
    "for i in range(50):\n",
    "    print(str(espacio_muestral[i]) + '\\t' + str(conjunta_sand[i]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Condiciones ideales\n",
    "- Velocidad de procesamiento infinita\n",
    "- Tamaño del dataset tan grande como quiera"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "En estas condiciones puedo calcular la conjunta. Y por que es tan importante?\n",
    "\n",
    "** Si tengo la conjunta tengo todo **"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Que puedo sacar de la conjunta?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Supongamos que nos dan una imagen y observamos que los pixeles 13, 14 son [0, 1]. Por eso lo que quiero saber es que es mas probable: Una sanadalia o un buzo?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$P(Sandalia \\ |\\  pix_{[13,14], 14}=[0, 1]) = \\large\\frac{P(pix_{[13,14], 14}=[0, 1] \\ |\\ Sandalia ) P(Sandalia)}{P(pix_{[13,14], 14}=[0, 1])}$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Si X es pixel 13 e Y el 14 de la columna 14 entonces:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$P(Sandalia \\ |\\  X=0, Y=1) = \\large\\frac{P(X=0, Y=1 \\ |\\ Sandalia ) P(Sandalia)}{P(X=0, Y=1)}$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Tenemos todo!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.038833333333333331, 0.00016666666666666666)"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Como P(Sandalia) y P(Buzo) son iguales y el denominador tambien, basta con calcular el lilelihood\n",
    "pixels = [0, 1]\n",
    "prob_2_pixels(sandalias,pixels), prob_2_pixels(buzos,pixels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pero en este caso, si le sacamos las idealizaciones, que problema tenemos?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.0, 0.0)"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Tomo una imagen y obervo los valores 10, 56\n",
    "pixels = [10, 56]\n",
    "prob_2_pixels(sandalias,pixels), prob_2_pixels(buzos,pixels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Las probabilidades para la mayoría de las combinaciones del espacio muestral son cero tanto para la sandalia como para el buzo. Es esto razonable?\n",
    "\n",
    "Si vinieran imagenes nuevas con alguna de estas combinaciones, no podría decidir. Ambas son iguales de probables"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Conclusiones:\n",
    "\n",
    "Intenté construir un modelo mejor pero al final solo me sirve para un porcentaje muy bajo de combinaciones... \n",
    "\n",
    "(Vale aclarar que uno podría argumentar que es posible que las combinaciones que no aparecieron en el dataset probablemente no aparezcan en las imagenes nueva. Pero... es esto razonable en esta aplicación?)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Sería interesante imaginar que pasa si ahora quiero armar una conjunta con 3 puntos!!! Todo los problemas se potenciarían"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Independencia -> Naive Bayes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Una de las formas de encarar el problema en casos donde la cantidad de ejemplos (imagenes) no es suficiente para estimar la conjunta puede ser suponer independencia:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$P(Sandalia \\ |\\  pix_{[13,14], 14}=[0, 1]) = \\large\\frac{P(pix_{[13,14], 14}=[0, 1] \\ |\\ Sandalia ) P(Sandalia)}{P(pix_{[13,14], 14}=[0, 1])}$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Suponiendo independencia:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$P(Sandalia \\ |\\  pix_{[13,14], 14}=[0, 1]) = \\large\\frac{P(pix_{13, 14}=0 \\ |\\ Sandalia )P(pix_{14, 14}=1 \\ |\\ Sandalia ) P(Sandalia)}{P(pix_{[13,14], 14}=[0, 1])}$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$P(Sandalia \\ |\\  X=0, Y=1) = \\large\\frac{P(X=0 \\ |\\ Sandalia )P(Y=1 \\ |\\ Sandalia ) P(Sandalia)}{P(X=0, Y=1)}$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Para buzo"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$P(Buzo \\ |\\  X=0, Y=1) = \\large\\frac{P(X=0 \\ |\\ Buzo )P(Y=1 \\ |\\ Buzo ) P(Buzo)}{P(X=0, Y=1)}$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Notar que no voy a tener necesidad de calcular el denominador y que el numerador ya lo tenía calculado de la primer parte"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "def prob_pixel_clase(clase, valor, posx=14, posy=14):\n",
    "    p_x_y_sand = (clase[:, posx, posy] == valor).sum()/len(clase)\n",
    "    return p_x_y_sand"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "prob que la posición 13, 14 tenga un valor de 10 (para sanadalia): 0.000833333333333\n",
      "prob que la posición 14, 14 tenga un valor de 10 (para sanadalia): 0.0015\n",
      "prob conjunta suponiendo independencia (para sanadalia): 1.25e-06\n"
     ]
    }
   ],
   "source": [
    "# Usamos [10, 56] que eran uno de los muchos valores de pixeles que daban cero para ambos (Sandalia, buzo)\n",
    "# Para las sandalias\n",
    "print('prob que la posición 13, 14 tenga un valor de 10 (para sanadalia):', \n",
    "      prob_pixel_clase(sandalias, 10, 13, 14))\n",
    "print('prob que la posición 14, 14 tenga un valor de 10 (para sanadalia):', \n",
    "      prob_pixel_clase(sandalias, 56, 14, 14))\n",
    "print('prob conjunta suponiendo independencia (para sanadalia):', \n",
    "      prob_pixel_clase(sandalias, 10, 13, 14)*prob_pixel_clase(sandalias, 56, 14, 14))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "prob que la posición 13, 14 tenga un valor de 10 (para buzo): 0.000666666666667\n",
      "prob que la posición 14, 14 tenga un valor de 10 (para buzo): 0.0025\n",
      "prob conjunta suponiendo independencia (para buzo): 1.66666666667e-06\n"
     ]
    }
   ],
   "source": [
    "print('prob que la posición 13, 14 tenga un valor de 10 (para buzo):', \n",
    "      prob_pixel_clase(buzos, 10, 13, 14))\n",
    "print('prob que la posición 14, 14 tenga un valor de 10 (para buzo):',\n",
    "      prob_pixel_clase(buzos, 56, 14, 14))\n",
    "print('prob conjunta suponiendo independencia (para buzo):', \n",
    "      prob_pixel_clase(buzos, 10, 13, 14)*prob_pixel_clase(buzos, 56, 14, 14))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "En este caso entonces me quedo con que dado los valores [10, 56] de los puntos 13, 14, decido por un buzo ya que 1.66666666667e-06>1.25e-06\n",
    "\n",
    "Notar que no estoy considerando la P(Buzo) ni la P(Sandalia), ni el denominador"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Escalamos a toda la imagen\n",
    "\n",
    "Recien solo consideramos 2 pixeles, consideremos todos los pixeles"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Podemos hacerlo a toda la imagen, 28x28 pixeles. Lo unico que tenemos que hacer es multiplicarlos y comparar."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Que potenciales problemas aparecen?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.power(0.1, 784)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Voy a terminar multiplicando probabilidades que seguramente sean bajas 784 veces entre si.\n",
    "\n",
    "Problemas numericos -> El logaritmo al rescate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-1805.2267129073316"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.log(0.1)*784"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Cual puede ser el problema de hacer logaritmo?\n",
    "\n",
    "Si una probabilidad es cero, log(0) no existe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/julianganzabal/anaconda3/envs/mllab/lib/python3.6/site-packages/ipykernel_launcher.py:1: RuntimeWarning: divide by zero encountered in log\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "-inf"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.log(0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Laplace smoothing\n",
    "https://en.wikipedia.org/wiki/Additive_smoothing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$P(pix_{14,14} = 0\\ |\\ Sandalia) = \\large\\frac{N_{sandalias\\ con\\ 0\\ en\\ 14\\ 14}}{N_{sandalias}}$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$P(pix_{14,14} = 0\\ |\\ Sandalia) = \\huge\\frac{N_{sandalias\\ con\\ 0\\ en\\ 14\\ 14} \\ + \\ \\alpha}{N_{sandalias}\\ +\\ 256\\alpha}$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Si alpha es 1, es como suponer que a priori (antes de estimar) tenía uno de cada clase con distribución uniforme. Es decir cualquier intensidad de 0 a 255 era igual de probable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Modificamos levente la funcion para evitar ceros de probabilidad\n",
    "# Es la misma función definida mas arriba, la copio para tenerla a mano\n",
    "\n",
    "def prob_pixel_clase(clase, valor, posx=14, posy=14, alpha=0):\n",
    "    p_x_y_sand = ((clase[:, posx, posy] == valor).sum()+alpha)/(len(clase)+256*alpha)\n",
    "    return p_x_y_sand\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Sumamos alpha en el numerador y alpha*256  al denominador en la estimación de la probabilidad"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.00095907928388746806, 0.00083333333333333339)"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prob_pixel_clase(sandalias, 10, 13, 14, 1), prob_pixel_clase(sandalias, 10, 13, 14, 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Modelo generador\n",
    "\n",
    "Supongamos que vamos a generar 4 puntos de la imagen:  \n",
    "X = (13, 13), Y = (13, 14), Z = (14, 13), W = (14, 14)\n",
    "\n",
    "Todas las probabilidades tienen la condición de la clase, en nuestro ejemplo sería el buzo o la sandalia. No la ponemos para simplificar la notación."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Suponiendo independencia"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$P(X, Y, Z, W)= P(X)\\ P(Y)\\ P(Z)\\ P(W)$\n",
    "\n",
    "Cada punto lo genero de manera independiente\n",
    "\n",
    "Estrictamente si quiero generar un buzo:\n",
    "\n",
    "$P(X, Y, Z, W\\ |\\ buzo)= P(X\\ |\\ buzo)\\ P(Y\\ |\\ buzo)\\ P(Z\\ |\\ buzo)\\ P(W\\ |\\ buzo)$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "En este caso mi modelo generador solo depende de la clase y no de las muestras generadas anteriormente. En nuestro caso, cualquier punto de la imagen se genera solamente teniendo en cuenta que es un buzo o una sandalia y no en que puntos se generaron anteriormente"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_hist(clase, posx, posy, colors=256):\n",
    "    hist = []\n",
    "    for val in range(colors):\n",
    "        hist.append((clase[:, posx, posy] == val).sum()/clase.shape[0])\n",
    "    return np.array(hist)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_image(clase, sample_type='argmax', saltear = 0):\n",
    "    imagen_gen = []\n",
    "    for i in range(28*28):\n",
    "        y = i%28\n",
    "        x = i // 28\n",
    "        histog = get_hist(clase, x,y)[saltear:]\n",
    "        if sample_type=='argmax':\n",
    "            imagen_gen.append(np.argmax(histog))\n",
    "        else:\n",
    "            imagen_gen.append(np.random.multinomial(1, histog))\n",
    "    return np.array(imagen_gen).reshape(28,28)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAADVBJREFUeJzt3V+oZeV5x/Hv45k//okXSlDnbyeK\nhAYvTBmkkFAsxWBLQHMRiVdTUjq5iJBALyreRCiBUJq0vQpMyJAJNCYBYx2k1ARpay6KOEqJJlON\nxhnnOMNMxUAMyujMeXpx1oST8ey19uy91177zPP9wLD/vHvt9Zx15nfW2vtd630jM5FUzxVDFyBp\nGIZfKsrwS0UZfqkowy8VZfilogy/VJThl4oy/FJRm+a5sojwdEKpZ5kZ47xuqj1/RNwdES9FxCsR\n8eA07yVpvmLSc/sjYgl4GbgLWAaeBe7PzF+0LOOeX+rZPPb8dwCvZOavMvM94PvAPVO8n6Q5mib8\nO4ATax4vN8/9nojYHxFHIuLIFOuSNGPTfOG33qHFBw7rM/MAcAA87JcWyTR7/mVg15rHO4GT05Uj\naV6mCf+zwK0R8ZGI2AJ8Djg8m7Ik9W3iw/7MPBcRDwBPAkvAwcz8+cwq09g26mhMEWN9Ka2eTNzV\nN9HK/MzfC8OvteZyko+kjcvwS0UZfqkowy8VZfilogy/VNRcr+dXP+wy0yTc80tFGX6pKMMvFWX4\npaIMv1SU4ZeKsqvvMnDs2LGRbXv27JlbHdpY3PNLRRl+qSjDLxVl+KWiDL9UlOGXijL8UlGO3rsA\n2vrpAZaWllrbV1ZWRrZdcUX73/ddu3a1tk/rxIkTI9vOnz/fumzXz9137RuVo/dKamX4paIMv1SU\n4ZeKMvxSUYZfKsrwS0VN1c8fEceAt4HzwLnM3Nvx+pL9/G193dDeTw/dQ3O/9957I9u2bt3aumyX\nafvi284zOHfuXOuy026XqmMZjNvPP4vBPP40M9+cwftImiMP+6Wipg1/Aj+OiOciYv8sCpI0H9Me\n9n8iM09GxA3ATyLifzPz6bUvaP4o+IdBWjBT7fkz82RzewZ4DLhjndccyMy9XV8GSpqvicMfEddE\nxLUX7gOfAl6cVWGS+jXNYf+NwGNNd8sm4HuZ+e8zqUpS77yefwZef/311vau/uiuvvQtW7a0trf1\npb///vuty3bpuma+ayyCtr721157rXXZTZva901dP1vb/+1bbrmlddmNzOv5JbUy/FJRhl8qyvBL\nRRl+qSjDLxVlV98MnDx5cqrlu7r6urq02rr6Nm/e3Lps12WzXUN/dzl79uzItptvvnmq9z5+/Hhr\n+5VXXjmyrWubb9++faKaFoFdfZJaGX6pKMMvFWX4paIMv1SU4ZeKMvxSUbMYvVcdtm3b1tr+6quv\ntrZfffXVre1t5wHs2LGjddkuXcOOd13y23XZbpuu8yfahiwHeOedd0a2dV0uXIF7fqkowy8VZfil\nogy/VJThl4oy/FJRhl8qyuv5xzTP7XSxl19+ubX92muvHdnWdV1617Dju3fvbm3vU9c27xoSvW35\nrnMIpj0/Ykhezy+pleGXijL8UlGGXyrK8EtFGX6pKMMvFdXZzx8RB4FPA2cy87bmueuBHwB7gGPA\nfZn5686VFe3nn6Y/ehynT5+e+L3bxtUfR9e4/m3Ti990002ty/bZzz+trnUPaZb9/N8B7r7ouQeB\npzLzVuCp5rGkDaQz/Jn5NPDWRU/fAxxq7h8C7p1xXZJ6Nuln/hsz8xRAc3vD7EqSNA+9D2QWEfuB\n/X2vR9KlmXTPfzoitgE0t2dGvTAzD2Tm3szcO+G6JPVg0vAfBvY19/cBj8+mHEnz0hn+iHgE+G/g\noxGxHBF/BXwNuCsifgnc1TyWtIF0fubPzPtHNP3ZjGvRhNrmml9ZWWldtqu/uqsf/9y5c63tbbX1\nPUZC28825PgMi8Iz/KSiDL9UlOGXijL8UlGGXyrK8EtFOU/xZaCtS2vz5s2ty3Z1BbZN/w2wdevW\n1va2LrXl5eXWZXfu3Nnarum455eKMvxSUYZfKsrwS0UZfqkowy8VZfilouznb/Q5NPeQun6utktu\nx9G1/NLS0si2rnMI1C/3/FJRhl8qyvBLRRl+qSjDLxVl+KWiDL9UlP38l4G2vvZNm9p/xV3X83cN\n3T1NP3/XutUv9/xSUYZfKsrwS0UZfqkowy8VZfilogy/VFRnP39EHAQ+DZzJzNua5x4G/hr4v+Zl\nD2Xmv/VVpNpN05feNa5/Vz9/V3vbeAJd03urX+Ps+b8D3L3O8/+Ymbc3/wy+tMF0hj8znwbemkMt\nkuZoms/8D0TEzyLiYERcN7OKJM3FpOH/JnALcDtwCvj6qBdGxP6IOBIRRyZcl6QeTBT+zDydmecz\ncwX4FnBHy2sPZObezNw7aZGSZm+i8EfEtjUPPwO8OJtyJM3LOF19jwB3Ah+OiGXgK8CdEXE7kMAx\n4As91iipBzHNePWXvLKI+a3sEg05bv+0v4O2ee63bNky1bq7xtZ/9913W9vb1t91jsH27dtb27u0\n/V6m3eYLPlfDWMV5hp9UlOGXijL8UlGGXyrK8EtFGX6pKIfu3gCOHz/e2n7VVVdN/N5dXV5dQ39P\ns26H7h6We36pKMMvFWX4paIMv1SU4ZeKMvxSUYZfKsp+/g2g67Lctv7yrr70rn78PnVdDqx+ueeX\nijL8UlGGXyrK8EtFGX6pKMMvFWX4paLs598Azp4929redk39NFNoQ/fQ3V3vv2PHjpFtb7zxRuuy\nfeoaenueQ9oPxT2/VJThl4oy/FJRhl8qyvBLRRl+qSjDLxXVOUV3ROwCvgvcBKwABzLznyPieuAH\nwB7gGHBfZv66470WtvN0kafoXuTpoPs05DTaG/l3Msspus8Bf5OZfwj8MfDFiPgY8CDwVGbeCjzV\nPJa0QXSGPzNPZebzzf23gaPADuAe4FDzskPAvX0VKWn2Lukzf0TsAT4OPAPcmJmnYPUPBHDDrIuT\n1J+xz+2PiA8BjwJfzszfjPuZJyL2A/snK09SXzq/8AOIiM3AE8CTmfmN5rmXgDsz81REbAP+MzM/\n2vE+fuE3wboX+culPvmF32Rm9oVfrP6U3waOXgh+4zCwr7m/D3j8UouUNJxxuvo+CfwUeIHVrj6A\nh1j93P9DYDfwOvDZzHyr473c80+w7kXeywypz+22kX8n4+75xzrsnxXDP9m6F/k/2pAM//pm2c8v\n6TJk+KWiDL9UlOGXijL8UlGGXyrKobsbi9x1I/XBPb9UlOGXijL8UlGGXyrK8EtFGX6pKMMvFWU/\n/wLwHIPJ9LndKvxO3PNLRRl+qSjDLxVl+KWiDL9UlOGXijL8UlGGXyrK8EtFGX6pKMMvFWX4paIM\nv1SU4ZeKMvxSUZ3hj4hdEfEfEXE0In4eEV9qnn84It6IiP9p/v1F/+VKmpUYYx7ybcC2zHw+Iq4F\nngPuBe4DfpuZ/zD2yiLaVyZpapk51kgknSP5ZOYp4FRz/+2IOArsmK48SUO7pM/8EbEH+DjwTPPU\nAxHxs4g4GBHXjVhmf0QciYgjU1UqaaY6D/t/98KIDwH/BXw1M38UETcCbwIJ/B2rHw0+3/EeHvZL\nPRv3sH+s8EfEZuAJ4MnM/MY67XuAJzLzto73MfxSz8YN/zjf9gfwbeDo2uA3XwRe8BngxUstUtJw\nxvm2/5PAT4EXgJXm6YeA+4HbWT3sPwZ8oflysO293PNLPZvpYf+sGH6pfzM77Jd0eTL8UlGGXyrK\n8EtFGX6pKMMvFWX4paIMv1SU4ZeKMvxSUYZfKsrwS0UZfqkowy8V1TmA54y9CRxf8/jDzXOLaFFr\nW9S6wNomNcva/mDcF871ev4PrDziSGbuHayAFota26LWBdY2qaFq87BfKsrwS0UNHf4DA6+/zaLW\ntqh1gbVNapDaBv3ML2k4Q+/5JQ1kkPBHxN0R8VJEvBIRDw5RwygRcSwiXmhmHh50irFmGrQzEfHi\nmueuj4ifRMQvm9t1p0kbqLaFmLm5ZWbpQbfdos14PffD/ohYAl4G7gKWgWeB+zPzF3MtZISIOAbs\nzczB+4Qj4k+A3wLfvTAbUkT8PfBWZn6t+cN5XWb+7YLU9jCXOHNzT7WNmln6Lxlw281yxutZGGLP\nfwfwSmb+KjPfA74P3DNAHQsvM58G3rro6XuAQ839Q6z+55m7EbUthMw8lZnPN/ffBi7MLD3otmup\naxBDhH8HcGLN42UWa8rvBH4cEc9FxP6hi1nHjRdmRmpubxi4not1ztw8TxfNLL0w226SGa9nbYjw\nrzebyCJ1OXwiM/8I+HPgi83hrcbzTeAWVqdxOwV8fchimpmlHwW+nJm/GbKWtdapa5DtNkT4l4Fd\nax7vBE4OUMe6MvNkc3sGeIzVjymL5PSFSVKb2zMD1/M7mXk6M89n5grwLQbcds3M0o8C/5KZP2qe\nHnzbrVfXUNttiPA/C9waER+JiC3A54DDA9TxARFxTfNFDBFxDfApFm/24cPAvub+PuDxAWv5PYsy\nc/OomaUZeNst2ozXg5zk03Rl/BOwBBzMzK/OvYh1RMTNrO7tYfWKx+8NWVtEPALcyepVX6eBrwD/\nCvwQ2A28Dnw2M+f+xduI2u7kEmdu7qm2UTNLP8OA226WM17PpB7P8JNq8gw/qSjDLxVl+KWiDL9U\nlOGXijL8UlGGXyrK8EtF/T/tDoZCtmWREQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1146e6940>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(generate_image(buzos, saltear= 0), cmap = 'gray')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/w9ZKR/uDqVb2QAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1155a8240>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(generate_image(sandalias, saltear= 0), cmap = 'gray')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x114628400>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(generate_image(sandalias, saltear= 1), cmap = 'gray')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conjunta sin independencia"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$P(X, Y, Z, W)= P(X)\\ P(Y\\ |\\ X)\\ P(Z\\ |\\ X,Y)\\ P(W\\ |\\ X,Y,Z)$\n",
    "\n",
    "El primer punto lo genero independiente pero luego cada punto a generar dependera de los valores de los puntos generados previamente"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Estrictamente si es un buzo la ecuación quedará:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$P(X, Y, Z, W\\ |\\ buzo)= P(X\\ |\\ buzo)\\ P(Y\\ |\\ X, buzo)\\ P(Z\\ |\\ X,Y, buzo)\\ P(W\\ |\\ X,Y,Z, buzo)$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "En este ejemplo, el modelo generador depende de la clase (buzo o sandalia) y ademas cada punto depende del anterior"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Para pensar\n",
    "\n",
    "Tengo que si o si siempre plantear un modelo generador para hacer clasificación?\n",
    "\n",
    "Se les ocurre alguna manera de estimar la $P(Clase|x_1, x_2, ..., x_N)$ de otra forma. No necesariamente generativa?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Modelo discriminativo**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Naive bayes gaussiano"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nlSL8QKUIP1Apwg9U6v8BxSRPTGh9u9cAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x114f98cf8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n4QcqRfiBShF+oFL/C+TDdEFZ5RaoAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x115523b38>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(buzos.mean(axis=0), cmap='gray')\n",
    "plt.show()\n",
    "plt.imshow(sandalias.mean(axis=0), cmap='gray')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x114680860>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\npeQXqdT/A0Q8roNtyjKcAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x115838080>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(buzos.mean(axis=0)+buzos.std(axis=0), cmap='gray')\n",
    "plt.show()\n",
    "plt.imshow(buzos.mean(axis=0)-buzos.std(axis=0), cmap='gray')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n4eMAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x114c14518>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSsUvUqn/B5lgGdCdhU3/AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x115834630>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(sandalias.mean(axis=0)+sandalias.std(axis=0), cmap='gray')\n",
    "plt.show()\n",
    "plt.imshow(sandalias.mean(axis=0)-sandalias.std(axis=0), cmap='gray')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 234,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.naive_bayes import GaussianNB"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Armo training set y cargo testing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 396,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_images = np.vstack([sandalias, buzos])\n",
    "train_labels = np.hstack([np.zeros(sandalias.shape[0]), np.ones(buzos.shape[0])]).astype(int)\n",
    "buzos_test = np.load('buzos_test.npy')\n",
    "sandalias_test = np.load('sandalias_test.npy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 237,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((12000, 28, 28), (12000,))"
      ]
     },
     "execution_count": 237,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_images.shape, train_labels.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 556,
   "metadata": {},
   "outputs": [],
   "source": [
    "from HistogramBasedNaiveBayes import HistogramBasedNaiveBayes as HistogramBasedNaiveBayes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 557,
   "metadata": {},
   "outputs": [],
   "source": [
    "histogramBasedNaiveBayes_clf = HistogramBasedNaiveBayes()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 558,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_reshaped = train_images.reshape(train_images.shape[0],-1)\n",
    "%time histogramBasedNaiveBayes_clf.fit(data_reshaped, train_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 552,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYQAAAD8CAYAAAB3u9PLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCTAQJEmNgSBJAuD/A7ICyEzEqVlsAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1146ed1d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYQAAAD8CAYAAAB3u9PLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1146ed550>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.bar(histogramBasedNaiveBayes_clf.sample_space_[1:],histogramBasedNaiveBayes_clf.histograms_matrix_[0, 183][1:])\n",
    "plt.show()\n",
    "plt.bar(histogramBasedNaiveBayes_clf.sample_space_[1:],histogramBasedNaiveBayes_clf.histograms_matrix_[1, 183][1:])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 553,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[6000 6000]\n",
      "[ 0.5  0.5]\n",
      "(256,)\n",
      "(2, 784, 256)\n"
     ]
    }
   ],
   "source": [
    "print(histogramBasedNaiveBayes_clf.class_count_)\n",
    "print(histogramBasedNaiveBayes_clf.class_prior_)\n",
    "print(histogramBasedNaiveBayes_clf.sample_space_.shape)\n",
    "print(histogramBasedNaiveBayes_clf.histograms_matrix_.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 564,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_predicted = histogramBasedNaiveBayes_clf.predict(data_reshaped)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 566,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.99750000000000005"
      ]
     },
     "execution_count": 566,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(y_predicted == train_labels).sum()/train_labels.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Naive Bayes Multinomial con Sklearn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 313,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(12000, 784)"
      ]
     },
     "execution_count": 313,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_images.reshape(train_images.shape[0],-1 ).shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 339,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.99033333333333329"
      ]
     },
     "execution_count": 339,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf = MultinomialNB(alpha=1)\n",
    "clf.fit(train_images.reshape(train_images.shape[0],-1 ), train_labels)\n",
    "(clf.predict(train_images.reshape(train_images.shape[0],-1 )) == train_labels).sum()/train_images.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 348,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(784,)"
      ]
     },
     "execution_count": 348,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.feature_log_prob_[0].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 353,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAEYlJREFUeJzt3V2o3dWZx/HfY4yaxKh5T0wzxilh\nGIlv40EGHAaHYrVDQXtRaS5KCqXpRYUp9GLEm3ozIMO0HS+GQjqGRmhtC62jFzKpyIAtDCWJaJPq\nTOtLrBnjiSYa8/76zMX5R0717OfZ2Wv/93/nrO8HQs7Z66z9X/vld/be5/mvtczdBaA+l3Q9AADd\nIPxApQg/UCnCD1SK8AOVIvxApQg/UCnCD1SK8AOVunSUBzOzWXk6oZmF7ZdcEv+OvfTS+GGYP39+\n2H7ZZZf1bDt37lzYN2vPZLc9um1nz54N+x49ejRsP3XqVNgeXf9sPrPV3eMHpVEUfjO7R9KjkuZI\n+nd3f6Tk+sZZ9CQvDe/KlSvD9htvvDFsv+6663q2HTt2LOybtWchmTdvXti+ePHinm0ffPBB2Hfn\nzp1h+549e8L2Q4cO9WzLfnHM5l8O5w38tt/M5kj6N0mfk3SDpA1mdsOwBgagXSWf+W+X9Kq7v+7u\npyT9RNK9wxkWgLaVhH+1pLemfb+3uexPmNkmM9thZjsKjgVgyEo+88/0IfgTH5TcfbOkzdLs/YMf\ncDEqeeXfK2nNtO8/JentsuEAGJWS8G+XtM7MrjezyyR9SdLTwxkWgLYN/Lbf3c+Y2QOStmmq1LfF\n3X83tJENWVaPnjNnTti+ZMmSnm033XRT2Pe2224ral+3bl3Yvnr1J/7U8pFFixaFfU+fPh22Z+cB\nZGXM6Pr37t0b9n3ttdfC9pdeeils37Gj95+Ztm/fHvbNxjYbSoVFdX53f0bSM0MaC4AR4vReoFKE\nH6gU4QcqRfiBShF+oFKEH6iUjbIe2eXpvVkd/9prrw3b77777p5tt956a9h36dKlYfuKFSvC9oUL\nF4bt0Xz+BQsWhH2z6cjZ+RHZ8yeak5/N1z9+/HjYfuDAgbD9vffe69mWnUOwbdu2sP3ll18O20+e\nPBm2t5m7fufz88oPVIrwA5Ui/EClCD9QKcIPVIrwA5Ua6dLdbcpKUlnJa/369WH7zTff3LNt+fLl\nYd+rrroqbM+mxV5xxRVhe3TbT5w4MXDffpw5cyZsj0paUYlSyseWTauNpiNn152VEScnJ8P2d955\nJ2zPli0fBV75gUoRfqBShB+oFOEHKkX4gUoRfqBShB+o1Kyp82fbYF999dVh+9q1a8P2aHnsZcuW\nhX2zOn3WnmlzK+qsHl6yBXjp1uUlU50vv/zysO+aNWvC9mwadnaeAHV+AJ0h/EClCD9QKcIPVIrw\nA5Ui/EClCD9QqaI6v5ntkXRY0llJZ9x9YhiDGkS2NHdWE86W7o5q+ddcc03Yt1RWE45q6W3X+dus\nV2fnAWTrAUS1/KzvypUrw/ZsOfa5c+eG7dHS3qNaTn8YJ/n8nbv3XiAdwFjibT9QqdLwu6RfmtlO\nM9s0jAEBGI3St/13uPvbZrZc0rNm9j/u/vz0H2h+KfCLARgzRa/87v528/9+SU9Kun2Gn9ns7hNd\n/jEQwCcNHH4zW2BmC89/LemzknYPa2AA2lXytn+FpCebUtClkn7s7v85lFEBaN3A4Xf31yX1Xsy+\nBVHNubTOn9Vto/UAsuvO1s7P1p/P1sbvss6fXX/J2vml8/2jWnt23UuWLAnbs3M7sjp/dNtHVeen\n1AdUivADlSL8QKUIP1Apwg9UivADlZo1S3dnpb4rr7wybM+W9o620Z43b17YN1veOiv1tSkrK5Vu\n4V3Sv3Rs2fLckax8m227XlLqGxVe+YFKEX6gUoQfqBThBypF+IFKEX6gUoQfqNRFVeePaqPZ9M6o\nTi/l5wFEdduszp9NyT1+/HjYntWEo+svmXLbz7GzqbHR8bO+2diy/tFjnj0m2bbp2fMlWxqcOj+A\nzhB+oFKEH6gU4QcqRfiBShF+oFKEH6jURVXnj2Tz+bM6f1arj84jyI6d1auz/iXLZ7e5hXY/orFn\ndfpMdtui+zW7T7Pnw4IFC8L2rM5fetuHofsRAOgE4QcqRfiBShF+oFKEH6gU4QcqRfiBSqV1fjPb\nIunzkva7+/rmssWSfippraQ9ku539/fbG+ZHY+nZls3nz+q22RrvWS0+cvLkybC9dM581D+77tJz\nDNqU3e6Src2zxzur02d1/mzd/ui2lW6L3q9+Xvl/KOmej132oKTn3H2dpOea7wFcRNLwu/vzkg5+\n7OJ7JW1tvt4q6b4hjwtAywb9zL/C3fdJUvP/8uENCcAotH5uv5ltkrSp7eMAuDCDvvJPmtkqSWr+\n39/rB919s7tPuPvEgMcC0IJBw/+0pI3N1xslPTWc4QAYlTT8ZvaEpP+W9BdmttfMvirpEUl3mdkf\nJN3VfA/gIpJ+5nf3DT2aPjPksaTarPNnddmoHl66Ln/JPvKZknX1+2lvsyadjf306dNh+4kTJ3q2\nZXX67DHJ1ofI1v1nPj+AzhB+oFKEH6gU4QcqRfiBShF+oFIX1dLdUVkpK9VlpZmsf1RKzKbsHj16\nNGzPpo9morJRtmx426KxlU5Nze73I0eO9GxbunRp2Dd7PpROEWeLbgCdIfxApQg/UCnCD1SK8AOV\nIvxApQg/UKlZU+fPpvRmUyxLpvQeOnQo7JvV+RcvXhy2Z0q2wW77PICSenbplN7Dhw8PfOzs+ZA9\nn7L2ku3DR7l0N4BZiPADlSL8QKUIP1Apwg9UivADlSL8QKXGqs5fslV1Nie+zaWUs3pytIR0P8cu\nXT675NhtngdQujV5NrZoPn/WN9u6PHs+tbnl+7Dwyg9UivADlSL8QKUIP1Apwg9UivADlSL8QKXS\nOr+ZbZH0eUn73X19c9nDkr4m6d3mxx5y92faGuS0sfRsK11nPau7RrX2bAvubN55Vs8+e/bswP2z\ncwSyenebW0lnx87OA8jao3UUsm3Vs9udnVeS1fmj6x/Vmv79PLI/lHTPDJd/z91vaf61HnwAw5WG\n392fl3RwBGMBMEIl7+keMLPfmtkWM1s0tBEBGIlBw/99SZ+WdIukfZK+0+sHzWyTme0wsx0DHgtA\nCwYKv7tPuvtZdz8n6QeSbg9+drO7T7j7xKCDBDB8A4XfzFZN+/YLknYPZzgARqWfUt8Tku6UtNTM\n9kr6tqQ7zewWSS5pj6SvtzhGAC1Iw+/uG2a4+LEWxlI0vzur82d12ZK549m6/FmtPTvHIKtJl9SF\nS2vpJWvIZ/d5duxsr4ZoHYXS+zR7zEqeb+NU5wcwCxF+oFKEH6gU4QcqRfiBShF+oFJjtXR3JiqB\nZKWXrCyUlVeiUt+xY8fCvqUlrTbLbaMqK82kdKvprLx78uTJnm2nTp0K+2ZLc2fPp6ydUh+AzhB+\noFKEH6gU4QcqRfiBShF+oFKEH6jUWNX5S+rdWZ0/a8+OHS2fndX558+fX3TsNrfozpTW4ts8dlZL\nj2r52bbppVu6l7RT5wfQKsIPVIrwA5Ui/EClCD9QKcIPVIrwA5Uaqzp/Jqp/ls6Zz0TbbGdbdC9c\nuLDo2F3W2ruU3e5sPn+0PHdW58+ULive5ToK5/HKD1SK8AOVIvxApQg/UCnCD1SK8AOVIvxApdI6\nv5mtkfS4pJWSzkna7O6PmtliST+VtFbSHkn3u/v77Q11fOv82Xz+rOab1bPHuc7f5p4CpfP5S/Za\nKL3PS56PJetaXMi4+3nlPyPpW+7+l5L+WtI3zOwGSQ9Kes7d10l6rvkewEUiDb+773P3F5qvD0t6\nRdJqSfdK2tr82FZJ97U1SADDd0Gf+c1sraRbJf1G0gp33ydN/YKQtHzYgwPQnr7P7TezKyX9XNI3\n3f3Dfj9Dm9kmSZsGGx6AtvT1ym9mczUV/B+5+y+aiyfNbFXTvkrS/pn6uvtmd59w94lhDBjAcKTh\nt6mX+MckveLu353W9LSkjc3XGyU9NfzhAWhLP2/775D0ZUm7zOzF5rKHJD0i6Wdm9lVJf5T0xXaG\n2J+2t7mOtnvOpvRmZZ9oWXApLln1c/1tKilTZrcrU7Lc+pEjR8K+2dhKn2/jMKU3Db+7/1pSr5F+\nZrjDATAqnOEHVIrwA5Ui/EClCD9QKcIPVIrwA5W6qJbuLpHVbbP2ku2eszp+dN1SPJ1Ymr11/uy6\ns/7R/Xb06NGwb7Tst9Tutuls0Q2gVYQfqBThBypF+IFKEX6gUoQfqBThByp1UdX5S5ZTzmrCWV03\nmrOf1YyzueMffvhh2J6dB1CypHmpkjp/dv5DaZ0/esyyxyQ7t6J0LYLIqJZq55UfqBThBypF+IFK\nEX6gUoQfqBThBypF+IFKXVR1/kjp2vdZ/6iWn9XpJycnw/Zs/nZWc26zzl9acy6Zz1/6mB06dKhn\n2+HDh8O+0T4NUvl8/3HYdp1XfqBShB+oFOEHKkX4gUoRfqBShB+oFOEHKpXW+c1sjaTHJa2UdE7S\nZnd/1MwelvQ1Se82P/qQuz9TMpiS+dtZLTyr22b9o7rwwYMHw75vvPFG2B7Vo6V87FGdf86cOWHf\nUiWPWWktPKu1R+dXLFu2LOwbrQUg5ecYZI9Z1H9U5wD0c5LPGUnfcvcXzGyhpJ1m9mzT9j13/5f2\nhgegLWn43X2fpH3N14fN7BVJq9seGIB2XdBnfjNbK+lWSb9pLnrAzH5rZlvMbFGPPpvMbIeZ7Sga\nKYCh6jv8ZnalpJ9L+qa7fyjp+5I+LekWTb0z+M5M/dx9s7tPuPvEEMYLYEj6Cr+ZzdVU8H/k7r+Q\nJHefdPez7n5O0g8k3d7eMAEMWxp+m/pT8mOSXnH37067fNW0H/uCpN3DHx6AtvTz1/47JH1Z0i4z\ne7G57CFJG8zsFkkuaY+kr5cOpqRslJV9slJeSenm/fffD/seOHAgbF+4cGHYno0tmrabTRcu3Q66\nzS26S7ZNl+Ly7PXXXx/2LZ0iXrL9eOnW5f3q56/9v5Y00zOkqKYPoFuc4QdUivADlSL8QKUIP1Ap\nwg9UivADlRqrpbuz+mVUa3/rrbfCvtu2bQvb33zzzbB9//79Pdt27doV9j127FjYPnfu3LA9W367\nyy26M10u3V1y3dkW3tk07t2743PeonMQSm7XheCVH6gU4QcqRfiBShF+oFKEH6gU4QcqRfiBStko\ntwo2s3clTS+oL5X03sgGcGHGdWzjOi6JsQ1qmGO7zt3jdckbIw3/Jw5utmNc1/Yb17GN67gkxjao\nrsbG236gUoQfqFTX4d/c8fEj4zq2cR2XxNgG1cnYOv3MD6A7Xb/yA+hIJ+E3s3vM7H/N7FUze7CL\nMfRiZnvMbJeZvdj1FmPNNmj7zWz3tMsWm9mzZvaH5v8Zt0nraGwPm9n/Nffdi2b29x2NbY2Z/ZeZ\nvWJmvzOzf2gu7/S+C8bVyf028rf9ZjZH0u8l3SVpr6Ttkja4+8sjHUgPZrZH0oS7d14TNrO/lXRE\n0uPuvr657J8lHXT3R5pfnIvc/R/HZGwPSzrS9c7NzYYyq6bvLC3pPklfUYf3XTCu+9XB/dbFK//t\nkl5199fd/ZSkn0i6t4NxjD13f17Sx1eNuFfS1ubrrZp68oxcj7GNBXff5+4vNF8flnR+Z+lO77tg\nXJ3oIvyrJU1fdmevxmvLb5f0SzPbaWabuh7MDFY026af3z59ecfj+bh05+ZR+tjO0mNz3w2y4/Ww\ndRH+mdacGqeSwx3u/leSPifpG83bW/Snr52bR2WGnaXHwqA7Xg9bF+HfK2nNtO8/JentDsYxI3d/\nu/l/v6QnNX67D0+e3yS1+b/34oIjNk47N8+0s7TG4L4bpx2vuwj/dknrzOx6M7tM0pckPd3BOD7B\nzBY0f4iRmS2Q9FmN3+7DT0va2Hy9UdJTHY7lT4zLzs29dpZWx/fduO143clJPk0p418lzZG0xd3/\naeSDmIGZ/bmmXu2lqZWNf9zl2MzsCUl3amrW16Skb0v6D0k/k/Rnkv4o6YvuPvI/vPUY252aeuv6\n0c7N5z9jj3hsfyPpV5J2STq/TO9Dmvp83dl9F4xrgzq43zjDD6gUZ/gBlSL8QKUIP1Apwg9UivAD\nlSL8QKUIP1Apwg9U6v8BxSRPTGh9u9cAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1223172b0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nUlGGXyrK8EtFGX6pqP8Hnxm/jUj7HGkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1158524a8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sel_class = 1\n",
    "plt.imshow(clf.feature_count_.astype(int)[sel_class].reshape(28,28), cmap='gray')\n",
    "plt.show()\n",
    "plt.imshow(clf.feature_log_prob_.astype(int)[sel_class].reshape(28,28), cmap='gray')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 331,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-19.23573844, -16.21531355, -14.37592603, -13.51871074,\n",
       "        -12.52338995, -11.98050033, -11.1578257 ,  -9.91506493,\n",
       "         -8.78496798,  -7.86264992,  -7.10536503,  -6.79454426,\n",
       "         -6.87481393,  -6.97952178,  -6.9936232 ,  -6.90579687,\n",
       "         -6.74558896,  -6.88631465,  -7.53468332,  -8.42854906,\n",
       "         -9.57066672, -10.86368663, -11.79383171, -12.44339401,\n",
       "        -13.98346501, -15.3039128 , -16.98444664, -19.92888562,\n",
       "        -17.28982829, -15.3039128 , -14.24871301, -12.4016297 ,\n",
       "        -11.51905494, -10.14830043,  -8.40495098,  -7.30173373,\n",
       "         -6.74914052,  -6.39622085,  -6.20083781,  -6.07911954,\n",
       "         -6.00839452,  -6.02821623,  -6.03979751,  -5.99636121,\n",
       "         -6.01667312,  -6.13823349,  -6.30766758,  -6.59538164,\n",
       "         -7.06135772,  -7.97161413,  -9.51987648, -11.18758934,\n",
       "        -12.68394407, -14.33046366, -15.92155243, -17.09567227,\n",
       "        -16.37353756, -14.7989869 , -13.38497377, -11.98439345,\n",
       "        -10.31836072,  -8.14639114,  -6.93094099,  -6.53239512,\n",
       "         -6.40605493,  -6.28121991,  -6.21640829,  -6.18335342,\n",
       "         -6.13510546,  -6.11814341,  -6.11632942,  -6.11778577,\n",
       "         -6.16754618,  -6.21968603,  -6.25872343,  -6.36797703,\n",
       "         -6.48630134,  -6.74006806,  -7.62032859,  -9.41952593,\n",
       "        -11.40531282, -13.39800799, -14.84748125, -16.71000979,\n",
       "        -15.77000253, -14.25213182, -12.90780165, -11.29390939,\n",
       "         -8.88753301,  -7.24566617,  -6.54439031,  -6.39762038,\n",
       "         -6.31896852,  -6.22836064,  -6.17593793,  -6.17261882,\n",
       "         -6.17194105,  -6.17068007,  -6.16669989,  -6.1828484 ,\n",
       "         -6.19075083,  -6.19230139,  -6.21432089,  -6.2975311 ,\n",
       "         -6.37999256,  -6.48372306,  -6.91637282,  -8.22045916,\n",
       "        -10.24020149, -12.64406471, -14.21515281, -16.34536668,\n",
       "        -15.63842618, -14.3012645 , -12.53685405, -10.27751996,\n",
       "         -8.15727187,  -6.87502595,  -6.38211239,  -6.26223708,\n",
       "         -6.22800844,  -6.18925229,  -6.17618494,  -6.18731012,\n",
       "         -6.19228299,  -6.19008657,  -6.18489577,  -6.19845485,\n",
       "         -6.20690912,  -6.19577997,  -6.19542712,  -6.22834942,\n",
       "         -6.25940712,  -6.34156633,  -6.65673937,  -7.61128949,\n",
       "         -9.43886218, -11.79559176, -14.12979296, -15.53443646,\n",
       "        -15.11670126, -14.07668314, -11.91587351,  -9.67210383,\n",
       "         -7.79822102,  -6.65775574,  -6.26169991,  -6.18033678,\n",
       "         -6.18079667,  -6.18560388,  -6.19223647,  -6.20007194,\n",
       "         -6.20732536,  -6.20785498,  -6.20242488,  -6.2146672 ,\n",
       "         -6.21811022,  -6.20502116,  -6.1979378 ,  -6.19970885,\n",
       "         -6.19838735,  -6.23680903,  -6.49316804,  -7.28941822,\n",
       "         -8.94852027, -11.09600616, -13.78684821, -15.33376577,\n",
       "        -15.9585937 , -14.39549613, -11.46298572,  -9.28935903,\n",
       "         -7.52710228,  -6.51977911,  -6.19108798,  -6.1285486 ,\n",
       "         -6.1625506 ,  -6.1943487 ,  -6.21124922,  -6.22014212,\n",
       "         -6.22729544,  -6.22696938,  -6.22315979,  -6.22683159,\n",
       "         -6.23343757,  -6.22856484,  -6.2200431 ,  -6.20330086,\n",
       "         -6.17248726,  -6.18139697,  -6.38208884,  -7.08570718,\n",
       "         -8.6155703 , -10.64010382, -13.13766415, -16.1676855 ,\n",
       "        -15.78575089, -13.33447216, -11.12546141,  -9.00393684,\n",
       "         -7.33534213,  -6.42531157,  -6.14671056,  -6.08734383,\n",
       "         -6.13789831,  -6.19245506,  -6.20913074,  -6.22273133,\n",
       "         -6.22612056,  -6.22973473,  -6.21984619,  -6.2237671 ,\n",
       "         -6.22890152,  -6.22960218,  -6.22815648,  -6.19898524,\n",
       "         -6.14052223,  -6.13934231,  -6.30841475,  -6.93613219,\n",
       "         -8.36547619, -10.30451691, -12.60505505, -15.10057188,\n",
       "        -14.84748125, -13.0161428 , -10.82757917,  -8.76347716,\n",
       "         -7.18520911,  -6.35844474,  -6.12533631,  -6.05037397,\n",
       "         -6.10653054,  -6.17676544,  -6.18856639,  -6.20604533,\n",
       "         -6.21085339,  -6.22302141,  -6.21817129,  -6.21036184,\n",
       "         -6.21515865,  -6.20611116,  -6.2164105 ,  -6.18714863,\n",
       "         -6.09868351,  -6.11104079,  -6.26010402,  -6.81664935,\n",
       "         -8.14260269,  -9.97946913, -12.18258496, -14.37592603,\n",
       "        -15.11670126, -12.56117705, -10.54602153,  -8.57171378,\n",
       "         -7.07071698,  -6.30569604,  -6.1055696 ,  -6.03621311,\n",
       "         -6.10938702,  -6.16377297,  -6.17553455,  -6.1910307 ,\n",
       "         -6.20797259,  -6.21893893,  -6.21564591,  -6.21354242,\n",
       "         -6.21031116,  -6.19660451,  -6.20455438,  -6.1995279 ,\n",
       "         -6.08421051,  -6.09129659,  -6.22176906,  -6.72223624,\n",
       "         -7.96342064,  -9.68856875, -11.76153363, -13.95762378,\n",
       "        -14.81689783, -12.20443897, -10.27597705,  -8.3833768 ,\n",
       "         -6.9701618 ,  -6.27039034,  -6.09296379,  -6.0383418 ,\n",
       "         -6.14388272,  -6.15970281,  -6.1646852 ,  -6.1869667 ,\n",
       "         -6.1993023 ,  -6.2054443 ,  -6.20387543,  -6.20370685,\n",
       "         -6.20259323,  -6.19526539,  -6.19736229,  -6.2250488 ,\n",
       "         -6.10212953,  -6.07228686,  -6.18982207,  -6.64292489,\n",
       "         -7.81060647,  -9.43384074, -11.4077004 , -13.62809982,\n",
       "        -14.43582417, -12.02899029, -10.03106409,  -8.21051632,\n",
       "         -6.88328666,  -6.2448423 ,  -6.07773189,  -6.05365438,\n",
       "         -6.1889934 ,  -6.15444396,  -6.15717335,  -6.18183914,\n",
       "         -6.19150418,  -6.19763096,  -6.19995962,  -6.1981903 ,\n",
       "         -6.19967505,  -6.19667624,  -6.18878634,  -6.2531223 ,\n",
       "         -6.13609931,  -6.05607454,  -6.16073295,  -6.57038191,\n",
       "         -7.66017626,  -9.23292605, -11.14640935, -13.38929966,\n",
       "        -14.16369451, -11.84555701,  -9.84199376,  -8.05644824,\n",
       "         -6.79617199,  -6.2197472 ,  -6.06169574,  -6.08302434,\n",
       "         -6.2419951 ,  -6.15708978,  -6.15820612,  -6.1796762 ,\n",
       "         -6.18475077,  -6.19167287,  -6.19494744,  -6.19642086,\n",
       "         -6.19413294,  -6.19516337,  -6.18666752,  -6.28836794,\n",
       "         -6.19229923,  -6.04977698,  -6.13742495,  -6.51049558,\n",
       "         -7.52539757,  -9.06357963, -10.92237649, -13.4596353 ,\n",
       "        -13.7304069 , -11.64564418,  -9.62984862,  -7.93225704,\n",
       "         -6.72681867,  -6.19805534,  -6.05218213,  -6.12864607,\n",
       "         -6.29485617,  -6.15391793,  -6.15511623,  -6.17291808,\n",
       "         -6.17858794,  -6.18740488,  -6.19305483,  -6.19234143,\n",
       "         -6.18851248,  -6.19112041,  -6.17906518,  -6.31279232,\n",
       "         -6.25496973,  -6.04968596,  -6.11532696,  -6.45549924,\n",
       "         -7.39701175,  -8.88171681, -10.75541672, -13.20866546,\n",
       "        -13.4831658 , -11.3513502 ,  -9.42790862,  -7.7928707 ,\n",
       "         -6.65555032,  -6.18063408,  -6.05154176,  -6.18147512,\n",
       "         -6.34225629,  -6.14643352,  -6.14983886,  -6.16666508,\n",
       "         -6.17262201,  -6.17891889,  -6.18395633,  -6.18349607,\n",
       "         -6.18212617,  -6.18489469,  -6.16983113,  -6.32238521,\n",
       "         -6.3198385 ,  -6.05996037,  -6.10094069,  -6.40993913,\n",
       "         -7.28285304,  -8.70793141, -10.47795478, -12.93495264,\n",
       "        -13.09262634, -11.07522019,  -9.27223821,  -7.68384229,\n",
       "         -6.59783954,  -6.17148098,  -6.0558064 ,  -6.23281422,\n",
       "         -6.36655136,  -6.13845905,  -6.15079659,  -6.15979918,\n",
       "         -6.16607671,  -6.1735158 ,  -6.17811733,  -6.17880998,\n",
       "         -6.17963987,  -6.18277873,  -6.1658206 ,  -6.3182748 ,\n",
       "         -6.37998603,  -6.07350456,  -6.09203048,  -6.37544267,\n",
       "         -7.18518574,  -8.56662026, -10.22837141, -12.71364564,\n",
       "        -13.20265222, -10.82212956,  -9.12998374,  -7.588986  ,\n",
       "         -6.55708735,  -6.16328615,  -6.06335708,  -6.28424523,\n",
       "         -6.37459438,  -6.12501049,  -6.1524933 ,  -6.15615313,\n",
       "         -6.16155846,  -6.16636558,  -6.16975284,  -6.17333315,\n",
       "         -6.17400762,  -6.18017427,  -6.15935931,  -6.29873402,\n",
       "         -6.42911236,  -6.09452877,  -6.08724056,  -6.35518551,\n",
       "         -7.12171255,  -8.45905556, -10.01688441, -12.41299653,\n",
       "        -12.38561227, -10.60298691,  -9.0302407 ,  -7.52596822,\n",
       "         -6.53007605,  -6.15805448,  -6.07259019,  -6.34143819,\n",
       "         -6.36279329,  -6.11739212,  -6.15101151,  -6.15505471,\n",
       "         -6.15714828,  -6.1619237 ,  -6.16532963,  -6.1702534 ,\n",
       "         -6.16897868,  -6.17728871,  -6.15963892,  -6.27201564,\n",
       "         -6.46872802,  -6.1199003 ,  -6.08271104,  -6.33737022,\n",
       "         -7.0711234 ,  -8.37719146,  -9.82654721, -11.99222546,\n",
       "        -12.73219905, -10.4148914 ,  -8.93911143,  -7.47705115,\n",
       "         -6.51065476,  -6.15588702,  -6.08543115,  -6.39680485,\n",
       "         -6.3424562 ,  -6.1150104 ,  -6.15346607,  -6.15206807,\n",
       "         -6.15391168,  -6.15962216,  -6.16339338,  -6.16806675,\n",
       "         -6.16934876,  -6.17658966,  -6.16538756,  -6.24912491,\n",
       "         -6.50561353,  -6.15449397,  -6.08316695,  -6.32604738,\n",
       "         -7.03262266,  -8.3051525 ,  -9.68333435, -11.9268608 ,\n",
       "        -12.22949621, -10.33267079,  -8.88894422,  -7.43426378,\n",
       "         -6.49925668,  -6.15556674,  -6.10517577,  -6.45275472,\n",
       "         -6.32166325,  -6.11531995,  -6.15748056,  -6.15489209,\n",
       "         -6.15533729,  -6.15951429,  -6.16265775,  -6.16683596,\n",
       "         -6.16895754,  -6.17835635,  -6.16961425,  -6.22706461,\n",
       "         -6.53175521,  -6.19323245,  -6.08325331,  -6.31730317,\n",
       "         -6.99974783,  -8.27367965,  -9.60451903, -11.66529519,\n",
       "        -11.96021992, -10.21190838,  -8.8304761 ,  -7.41015052,\n",
       "         -6.49298677,  -6.15503595,  -6.12871512,  -6.51271596,\n",
       "         -6.30400544,  -6.11978456,  -6.16268611,  -6.15775233,\n",
       "         -6.15706366,  -6.15863187,  -6.16244241,  -6.16520323,\n",
       "         -6.16841008,  -6.17947533,  -6.1744359 ,  -6.20786048,\n",
       "         -6.54937618,  -6.24197577,  -6.0870847 ,  -6.31586463,\n",
       "         -6.98329045,  -8.2249597 ,  -9.50603896, -11.47016936,\n",
       "        -12.3018282 , -10.20171767,  -8.82775525,  -7.39487707,\n",
       "         -6.49020898,  -6.15737293,  -6.15925357,  -6.57401942,\n",
       "         -6.29420054,  -6.13209316,  -6.17127751,  -6.16345015,\n",
       "         -6.16288363,  -6.16206438,  -6.16607461,  -6.17020576,\n",
       "         -6.173946  ,  -6.1849527 ,  -6.18430302,  -6.20339058,\n",
       "         -6.57029032,  -6.30158955,  -6.09552925,  -6.3141971 ,\n",
       "         -6.96709718,  -8.20001847,  -9.47695879, -11.31058092,\n",
       "        -12.06600358, -10.22336132,  -8.81189581,  -7.39463576,\n",
       "         -6.50043558,  -6.16659969,  -6.19449293,  -6.63156607,\n",
       "         -6.30730696,  -6.15942947,  -6.19493984,  -6.18719384,\n",
       "         -6.18430194,  -6.18346496,  -6.18735858,  -6.19285776,\n",
       "         -6.19620356,  -6.20764728,  -6.21338878,  -6.22139132,\n",
       "         -6.60030025,  -6.36103161,  -6.11341306,  -6.31719281,\n",
       "         -6.95946598,  -8.18495955,  -9.44466399, -11.3579621 ,\n",
       "        -12.18561592, -10.35320793,  -8.83869859,  -7.39734752,\n",
       "         -6.51302163,  -6.18183807,  -6.23994786,  -6.71907929,\n",
       "         -6.37590414,  -6.22677895,  -6.25542492,  -6.24836924,\n",
       "         -6.24280516,  -6.23963354,  -6.24249221,  -6.24822732,\n",
       "         -6.25323501,  -6.26492147,  -6.27336777,  -6.28225062,\n",
       "         -6.67257019,  -6.43130193,  -6.13710951,  -6.33130348,\n",
       "         -6.96209593,  -8.19619831,  -9.48753137, -11.39700088,\n",
       "        -11.89945278, -10.43344149,  -8.88936155,  -7.42132785,\n",
       "         -6.52798485,  -6.20225545,  -6.29719093,  -6.83634042,\n",
       "         -6.51577986,  -6.3670023 ,  -6.39919753,  -6.38899682,\n",
       "         -6.38092123,  -6.3804718 ,  -6.3834399 ,  -6.38708848,\n",
       "         -6.39174595,  -6.40562086,  -6.41590955,  -6.43173355,\n",
       "         -6.79435062,  -6.5161463 ,  -6.16908864,  -6.34361876,\n",
       "         -6.97220145,  -8.22403486,  -9.5607211 , -11.3354724 ,\n",
       "        -12.1735468 , -10.62706056,  -9.02574092,  -7.52370805,\n",
       "         -6.57670501,  -6.22661319,  -6.36908829,  -7.01418622,\n",
       "         -6.75942494,  -6.62112209,  -6.66698293,  -6.65137491,\n",
       "         -6.64113154,  -6.64161646,  -6.64258362,  -6.64804266,\n",
       "         -6.65628492,  -6.67111236,  -6.67775267,  -6.70695749,\n",
       "         -7.01299306,  -6.62465078,  -6.21213187,  -6.36798477,\n",
       "         -7.03921878,  -8.3318672 ,  -9.71651728, -11.41952301,\n",
       "        -12.54800658, -10.91395204,  -9.17678678,  -7.59385264,\n",
       "         -6.61246647,  -6.23913672,  -6.41108188,  -7.21574239,\n",
       "         -7.25070843,  -7.08555119,  -7.12026907,  -7.09409232,\n",
       "         -7.07580659,  -7.07119116,  -7.06995928,  -7.07826272,\n",
       "         -7.09183911,  -7.11688083,  -7.14115896,  -7.22039963,\n",
       "         -7.40413381,  -6.70707336,  -6.22696266,  -6.38005522,\n",
       "         -7.0707274 ,  -8.38468438,  -9.93871591, -11.81216999,\n",
       "        -14.93167334, -12.45354638, -10.26283072,  -8.42687955,\n",
       "         -7.33017984,  -6.86731525,  -7.09370844,  -8.11049956,\n",
       "         -8.49753904,  -8.32120467,  -8.2447164 ,  -8.18681067,\n",
       "         -8.12821308,  -8.11407936,  -8.11527795,  -8.12693149,\n",
       "         -8.15515094,  -8.21377061,  -8.28501961,  -8.41003772,\n",
       "         -8.42137479,  -7.42624797,  -6.80874471,  -6.98099468,\n",
       "         -7.77888609,  -9.23371863, -11.08066355, -13.56585751]])"
      ]
     },
     "execution_count": 331,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.coef_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 287,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.0 0.985 0.9925\n"
     ]
    }
   ],
   "source": [
    "acc_buzos_mult = (clf.predict(buzos_test.reshape(buzos_test.shape[0],-1 )) == 1).sum()/buzos_test.shape[0]\n",
    "acc_sand_mult = (clf.predict(sandalias_test.reshape(sandalias_test.shape[0],-1 )) == 0).sum()/sandalias_test.shape[0]\n",
    "print(acc_buzos_mult, acc_sand_mult, (acc_buzos_mult+acc_sand_mult)/2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Naive Bayes Gaussiano con Sklearn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 292,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.99516666666666664"
      ]
     },
     "execution_count": 292,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_gaussian = GaussianNB()\n",
    "clf_gaussian.fit(train_images.reshape(train_images.shape[0],-1 ), train_labels)\n",
    "(clf_gaussian.predict(train_images.reshape(train_images.shape[0],-1 )) == train_labels).sum()/train_images.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 304,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2, 784)\n",
      "(2, 784)\n",
      "cantidad de parametros aprendidos: 3136\n"
     ]
    }
   ],
   "source": [
    "# Medias\n",
    "print(clf_gaussian.theta_.shape)\n",
    "# varianzas (desvios)\n",
    "print(clf_gaussian.sigma_.shape)\n",
    "print('cantidad de parametros aprendidos:', np.product(clf_gaussian.theta_.shape) \n",
    "      + np.product(clf_gaussian.sigma_.shape))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 288,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.0 0.991 0.9955\n"
     ]
    }
   ],
   "source": [
    "acc_buzos_gauss = (clf_gaussian.predict(buzos_test.reshape(buzos_test.shape[0],-1 )) == 1).sum()/buzos_test.shape[0]\n",
    "acc_sand_gauss = (clf_gaussian.predict(sandalias_test.reshape(sandalias_test.shape[0],-1 )) == 0).sum()/sandalias_test.shape[0]\n",
    "print(acc_buzos_gauss, acc_sand_gauss, (acc_buzos_gauss+acc_sand_gauss)/2)"
   ]
  }
 ],
 "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}