code/Course_93.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAADIAAAAhAQAAAACzBsUpAAAAmklEQVR4nG2PMQrCUBBEXzYBG0EE\nC22tBFsbK/9JrFN6A3/vlQS/eI6QgOlNoSAojsVaiDjNYxh22MkEwNFw/eM2QgG3SQlI3VBKBteL\n59WmAaTz/aBkUIcEBlny/ASA1HuulAzipzdEZyqdj7FTnXPaAsarnQUwLC+8d1kFQMr3tVIB68UA\nkBrzv2Lm96M+UMBuDmTfO+OPB95h6jjBi4YE2AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<PIL.Image.Image image mode=1 size=50x33 at 0x8CAAB70>"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import PIL\n",
    "import numpy\n",
    "basewidth = 50\n",
    "pil_image = PIL.Image.open('image_sample/0/3.jpg').convert('1') \n",
    "wpercent = (basewidth/float(pil_image.size[0]))\n",
    "hsize = int((float(pil_image.size[1])*float(wpercent)))\n",
    "img = pil_image.resize((basewidth,hsize), PIL.Image.ANTIALIAS)\n",
    "img"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABbUAAAV3CAYAAABmHILzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nGzq1AQAAAADohk5tAAAAAAC68f8BLN3U2RKZg7AAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x8d39710>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%pylab inline\n",
    "import os\n",
    "digits = []\n",
    "labels = []\n",
    "basewidth = 50\n",
    "fig = plt.figure(figsize = (20,20))\n",
    "cnt = 0 \n",
    "fig.subplots_adjust(left=0, right=1, bottom=0, top=1, hspace=0.05, wspace=0.05)\n",
    "for i in range(0,10):\n",
    "    for img in os.listdir('image_sample/{}/'.format(i)):\n",
    "        pil_image = PIL.Image.open('image_sample/{}/{}'.format(i, img)).convert('1') \n",
    "    \n",
    "        wpercent = (basewidth/float(pil_image.size[0]))\n",
    "        hsize = int((float(pil_image.size[1])*float(wpercent)))\n",
    "        img = pil_image.resize((basewidth,hsize), PIL.Image.ANTIALIAS)\n",
    "        \n",
    "        ax = fig.add_subplot(10, 12, cnt+1, xticks=[], yticks=[])\n",
    "        ax.imshow(img,cmap=plt.cm.binary,interpolation='nearest')\n",
    "        ax.text(0, 7, str(i), color=\"red\", fontsize = 20)        \n",
    "        cnt = cnt + 1\n",
    "        \n",
    "        digits.append([pixel for pixel in iter(img.getdata())])\n",
    "        labels.append(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy\n",
    "digit_ary  = numpy.array(digits)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(119, 1650)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "digit_ary.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:429: DataConversionWarning: Data with input dtype int32 was converted to float64 by StandardScaler.\n",
      "  warnings.warn(msg, _DataConversionWarning)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "scaler = StandardScaler()\n",
    "scaler.fit(digit_ary)\n",
    "X_scaled = scaler.transform(digit_ary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MLPClassifier(activation='logistic', alpha=0.0001, batch_size='auto',\n",
       "       beta_1=0.9, beta_2=0.999, early_stopping=False, epsilon=1e-08,\n",
       "       hidden_layer_sizes=(30, 30, 30), learning_rate='constant',\n",
       "       learning_rate_init=0.001, max_iter=3000, momentum=0.9,\n",
       "       nesterovs_momentum=True, power_t=0.5, random_state=None,\n",
       "       shuffle=True, solver='adam', tol=0.0001, validation_fraction=0.1,\n",
       "       verbose=False, warm_start=False)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mlp = MLPClassifier(hidden_layer_sizes=(30,30,30), activation='logistic', max_iter = 3000)\n",
    "mlp.fit(X_scaled,labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "predicted = mlp.predict(X_scaled)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2,\n",
       "       2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 4,\n",
       "       4, 4, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 6, 6, 6,\n",
       "       6, 6, 6, 6, 6, 6, 6, 6, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 8, 8,\n",
       "       8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 9, 9, 9, 9, 9, 9, 9, 9,\n",
       "       9, 9, 9, 9])"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predicted"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "target = numpy.array(labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True], dtype=bool)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predicted == target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAuIAAABhCAYAAACTWMRgAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAF1ZJREFUeJztnU2S5DYORlUTPoK9nj6E738C+w72evoO6YVDYzaMX5KS\nksz3IhxVqaQohSCAHwFU++v1eh0AAAAAAHAv/3n6BgAAAAAAPhGEOAAAAADAAyDEAQAAAAAeACEO\nAAAAAPAACHEAAAAAgAdAiAMAAAAAPABCHAAAAADgARDiAAAAAAAPgBAHAAAAAHiAnyqDf/7559e3\nb98uuhW4iz/++OP4/v37lzyOfffh999///56vX6Rx7HxHuDD+4MP7w0+vD+WD0tKQvzbt2/Hb7/9\n1n9X8Bb8+uuv6nHsuw9fX19/asex8R7gw/uDD+8NPrw/lg9LaE0BAAAAAHgAhDgAAAAAwAMgxAEA\nAAAAHgAhDgAAAADwAAhxAAAAAIAHQIgDAAAAADwAQlzh6+tf/7Rnenz1XAAYZ9Tvvr6+8F2ARcBX\nYSeWFeLnwhmJYMth5fF2rtfrlXb0r6+v4/V6/f9z+zuL+xjSttnn6b0T3vuQHZv9HmIsn41sIT+3\nfuddw7LZ6/VKzwF5KjE5+703NuPDVT8Hm8hXs0mq8zst5mfuIXMNmENF10i7jvh29fuVWE6It2K5\nXTylIPbO1YgWYu886xq9c8LfyA2OtLeGtzHSPsv3yRvr3R/UaJ+7JHvMm1ee610vQ/TewY9U7Vv5\n3hub8eGqn8O/ycbNjO/JNbP93TrHWvcztsSH+/Bs7o3t5ZPW39L/WfMdkBnn8/Pp8JkgK3dnVgbc\nWtSz96eR2TBATDbgZqobVna1Pb+y4YOYyoKp+XZ7TAb9SIzLsVaMGNkQfDpV0aWN03w3W63ETtdi\nrcPa50qlKbJtdN3oetCPl3TUxlm2Hd0IzUisvBvLZcRbejMb1o49M9cnlUvejcxmSVsAZDa9IgKj\n1iOYiwyw1gZbE+GZ0rekkunG3uO0vhhVKGeAXa9npBIdjdP8XrZGeBvxnmQa5PA0UrVKUWHHVqTl\nMuISazcus5jt9/J8a97oml4G3tqxEQT68TKUlljTkFm5ap9pRdBDH1a1q6f/1Jq/xcq0ZapuMJcZ\n2S75bkQZW+w6jrfmZTfM1pzR99JPrWvDHCJ/uUss72LTpTPiJ1a5s905V7LlM4w7ax74ES+z2ePw\nu+yoV0fLhLd4YqniZ1bFJFMBqV4L/qGNxefvPZnUHn897autBdl7gRi55rXZ60wVZNY9RPd13hv0\nM2rL3pi9K0sL8dbB22PH0W88WfbSyLQn7Fg+eQe8Z9/zHljfV9pXYJxMZaK3XUwb64lva27s3Y9s\nDev1y+y1Wjwh+AmL/FPITXW1FcmzTSYb663N2H2czDOcEbM/gaWFeLY0Ij+PZqu1HjWZ8bmyRwr+\nRhPemm0q70l03Lo+jCOFkhTLXvayupBr8cFrNYvmgzyaTdvP7RiNHjtE704mAQM6UbuYV8XUzvOO\nRfeR2Vzjx/cxw6e0OXaz4dJC/DjirJlmsJ6Wld77iY5DH5kFW46FtbAWVavFwDo/ixT+MB+tl7ca\ngyORXm1f0yqrME6mcuydo33Ozuf9PQlcx0grX6V9eDd7Lv3HmtquNypvZ+bwxrU/e++PgD+Otoj3\nBPvWHprg81omsON8tD+08gSUHCvnaMdaGfcoI4rPXosU51ZyRdpOnivHaNeRv8uWFexcR6saVRMk\nng2imDzS1gZ9WDGxqr+8sZm4v4vfLp0Rr/4Bhubgld1V1bm1xWL1F+ZJZpWrrZK4taBrC01lwYEx\nRp6x15JgLeryuq2tsXcf0bP0KhHSPlq2M2MXr0+djPizREmP7AYL7uEOX9lFZGdYOiN+0hosctg2\nC3YS9ZL2BmmrhP4JL9YVyI1NtjqRyXhVWpzoG56PrEwch/0vp1jHMn3i8hqVDM6OJdG76PWTyO+8\nd0R7N6xNF4wRxeLILu047TyrnSm67ieJuSe48rl6ttV03MoslxHXFs5KeTHaeUsyTm6hCfhdXpw7\n8GxS+ZzNvMF9eBsfLetpBd5Ka0Lm+/ZeomuBTZTZPo5cRVHz3fZzpRVlpGwOP2LZ0kpaeXaMroHP\nvQcjbSbe+Z5fZ87ZwY+XE+IelXaBikCWC32l/43gfx3nQt0G7ChrmWktsRZtbHkP3rOWi3Nrb0uo\nz7oXBEE/Wdt4/cBWa2E7vtKigj2vJ1tJjN4FbWw2qUIlaw6ePaQwzlQn5XEttmfuZwc/Xk6Ie20e\nPRnqapZF9hl6aIFih5fmLjK2kp+tgG5laTRbWhlZLWAQ4PuY1apg2VAbb4l5b5xmd4jxWkayWezq\n963Ajqqi7X9Qx1t3tWyltqHS/C/zbmjZd0/kyQ0dNp+L9LfsuyGPycRaVBHfyY7LCfGWqhDK7Lqr\nAiGza0N8z6H3OXoZtUz23JsH287jDMKZ1rL292wM0DKo5/GeKhrUsNqLvPHtee2xFm3xzvr1jov6\n3WSSYZo4z2bBLdu38cKaS26qK+8f5Ki0Glnxt/0+052wm65aWoi39C6kMkDMcNSecinUqNi7x55W\nuU1bUGAOvZvgkc2zJ8Jnt7l8MpX2Hqv6FI2X1REryRJlbOFaNL+tJkSkaMsIeET4XLTnK/GSVlYV\nMhuPd2J5Id4rsrTzvV2WfDmirF3mGPQzwymjzIwWRMiGX0+0KFc3t16LmLc4e60sUMNarD0BlbFN\nNFdW7Fnnw/VYWW0rPnuVLWv+83vW4euwfE1LamX11KckQ5YX4j30BNxZQXrXF+lOskIsWsgzGy/6\nCt8Lq59QGyPxMqTZ73gP7sXqGY0SIVGMqAh8yJGtUmb804rdvdUMNlnziapUrX2shGfk25UE6Mps\nI8S9sof22SMrvnraI3Z9ke4keoZWyas9ZmW2reBuBXsW7jlUy5FeFizj+9b5XvuCdz+QI9ve1W6U\n29iZjblRDyktZvPoqU61WD6s+aj3u3cftIheQ2bT653b/pTHj0P/O5EdWVqIV4Kplz2Ldl3VDJs8\nd/eX6C6qGRethzCyh2dTTaQhzOaQFdzR8676OJWPe8i0m0RtX6OJkayggxrVJFN1nNU/XEmEwTV4\nm95I93jnWeN3ZWkh7hna23lnhZQWYKJdd3Schb4fq8+0/T4jxGRg13b1WcEAc4h8Ugq5ShD3sqPY\n8h4yzznq9T3taGXQKpu0qGcV8vQI4sqaeNo8sn/vvcE4lQpiZkMdVbSyc63C0kI8i9Wq0JNhsdoY\nMteOxoJPb0bM6k+T351zSBG+i7O/M9bz1uw1mr2OfB8fnY/V2yvpefaWyKtUrqJNPthkKgpeW1+m\nYqy1qvS2e2Lj+Uj7RPaO8DoY5HV3YHkhXgkC1gKfzZZZWVUPnH4ePSXN1qEr/WzaufAMVqWiNwj3\ntLfAGJk+/ajCKcdX7KRVuGZtBOAfvEpiO0YSVTOtLLgX0y17Y+M5yLXR0mKW/TSfbs+32kp3ZGkh\nXilRnVTEunT2XV+Cd8QSwFkRfZJZFAjQ70GUAfGqFN6CnM3G9t4jjJHZ7FZ6UNtWBu061meYS9XH\nNHu1bWjVXn6rsonda1ixVRPK2s+M/bRNtif0d2M5IW4tzFkjWc6oibH2WLUffWQc2ItpxRml/Syy\nQiCbTYeYTAbLO8fLoJxjrUxN74KM7fPc9aw8Ma79DnPQNrcZKut0pmXFm1ueT8JlHl5yI+rj9+K4\nFO/adzvG4eWEuFfyyDb4y/OzQaVHjLEg9DPjeVXaUXquv2NQuINZvuC1rWT7gistT/hwjigbPTJn\nNVNWqZBBDm1z24rnqE+4/awJLMs3ZZX6HBOJ9pEN+KcTZbS9SoYcY2XN2zHZuL0TywnxlkzfYXS+\n94J5L00WnP45Rp/9zJYGyFPZAHnBPjqn9euRygvEVJ9nJjlSzZi2c2HfeWjCK/ucI5+1xsvrZ1vS\nMvcENazuAi8B1lvZbq+5k7b66ekbGEUax3NsGbgzhmQX/f54C7LM0GgtLzKDY/0ux5IhvYaMr1nP\nXRPVmTmkvfH3uWRtGm2wRlvV5Oar8r6AjtdGGNnda2HouQe4H0uDZdpBtbkyPrmbzy6dEbd2YtF4\nbZz28nzay7AjPVWNESEIc+h5vtoGS85pvQfY81oym9bIR70eVDlH5juE2zzkutkjjHuypCfY8n2x\nEhuWvSMBv6OtlxbimT40r31FmydbzurJpuz4At2NZ7sM1iZMm7d3wYdxZvT0ey0oVl8jrQvXEfXw\nar9r51lxN4rHXhUFf+5HWzOvrjTI/vDzd7iHSsvJSOU409mwA0sL8eOwy8sVMv1j2h+GVK9DoBgj\n2nRpyCBt2U3+vcC5OGOzZ6j0iUffz+oxRLCN4QkmLXZHSRGtZUwe167h3RvU0WJr+zP7bHt8C398\nBq/yqB3v9T0ruZK5/kosL8Rb7gimZMyeo3dnnWk3yvzBT3ssWvRhjIzIjmxQzY5qC4dWcocxvMRH\nbwm70lII15DdYM26lkyKsdG6D8uH5XHP/tm1M2op28G2WwhxKaLkH9JVFlOt5NXO0ZMFZwF/jky1\no/3e+2MxbV7sOpdsJrxHeI0GcDZd/Yy2eMlKlTw3qlZcmZWFf/B8bNY6KIXfjOoZ1JAbH6uCfI6L\nKsyVNXa3bPhxbCDEpeg+Dr10qQXyzNwtGYe3XqhdXpgnmfUMvb4zq+wN95DxuUwWVZu3p7cQ28+h\nEjt7e8lPWqGuifbqNSGPJtC8sT3zS3oqYjCOTFxZ8TXbxuLF/igrvjpLC/FocR0pV0mBHwX09jzv\nM/TTW42IILv9vmTahbSA7bUUtRkai8wGAPJkxFLWD72/8Wi/r27i+JuQOcjNUu8aaNkx20Io56Iy\nPR8vW50Z3x73NsXyO5k4W52lhbiVxbTaEVrH9tpW5Ethjc9kuqXzI8xzaM+2Kpgj+8hyGbZ5nuyC\nqYnp1paeCMvO357DuzGGlcnWfHqGX1qlcG1tQKTNJdrcWO0q2vqbWbPlMes+qEz3kVlH25/e8ahV\n1PPZ3pbEFVhaiB9HPXMVZVO077zsWlYwQA2t1BUtzNamShtn2W92KRVqZCpPlqC2ArtXOctWuLD9\nGNFirFU5ss8+SsBYx4jLc+iJmZY4z4q6VlhnfZPNVj9a/JT+acXYSuIsSqC0Y3fSVssL8ZZoV6wR\ntZLIcrb8rpIR36mUcheZBVUT7ZFdqxsybQ6YT7aUbdlJ+qRXzfI2aplrQZ7qprodY2EJ9eg4zCXr\ns9m5WjSxJ8dl28ywfx9e1cEae+LZRla9vJiuXUdb91dlaSHuGfgk8/J446OMG7vx+6lmNr3NlByn\nLeIjdoccVvkxGi/psUsk2DJZevBpNz7e4llJprSLuNWWlEmUYNdxMi0DmdhbqUBVM63E7Hlk19Es\n1S6GyvcrsLQQz/QYaX1ico5M8O8JLnAdmcVWy6ZEDq9lTzzhj/3nUMm6eM/e8//M+dp9RaVXiKmI\npvOnFNht/LaqHu082U0ddr0Hb8Mb+X+2Gj2yiYc8li0zG1/5+6g9drDn0kJcYrWVnGR3xZGYjxg5\nF+pk+hCvevY77MbfmWzP4HHEfxgmj2cWgp36EN+FTBuBZmOrzeics1opk+dh5zGqm17pf9XKo5dA\n86ojxOw5aKK6klDR5jmJWkvb4zvYcwshLkta0uF6AmymvSWze8+U7KCGzGB5AVZ7L+R3MgsHzyFt\noPlS22rUjsn2FFvX0pCtFPjwON4zlLY9j1XmjDKt1mYN245R2TRLolis2SwT9637I85fR+Tf1hhP\nfFutpbv47PJCXPYeHocdaLPB3St1tudWyjAwl0w25PxdK1NXyqIWBPP5eFUr6b9V4Z1tMclk2mGM\nan+vdtyaU/Nzy96I8Hlks5gSbe2OWkYrGzX8+V4yCQ4vkdZqupZsR8OKLC/EKyK7uouWmfboHjwQ\nbXOJ2gnOn5ZTH0f8R3qRXektnUtPhrtdVDW797YrVI5DjmqM9sZme4m1sdjxGry12LN9xt+0ZEqG\nbOUU+qlWryQVX9aO7+DPSwtxq2zdft/+552vzdF+zgg/bS7rM4xReZ5auXN080QmbT5RNsxCLtJR\nL3HP/WhzQR5vk1ttGcjE8/Yd0K7dthztsJC/E56w0kRUVKnU1vQKCPBrsCrQHhlfl5XLrE5bmaWF\nuKSSObG+0zJslYWY7MtzWPaZ3eO7g+O/K9qmKUMU4Ktzte8MPjyOV5GqtAmdxyL7VN4f/HkcL0vp\nZTa9hJWsaPbYaed2hnehNwue1VOZdX11lhbiWiDXytRehqxFivDMQuHdV3Q9qBMtvhaRLSs23sX5\n35EzuPa0glnla9mmElWrtA04Nh9jxKbt71pVq6fFbKds2tNI/7Lso/mdt2GSIqyysfIqIjAHza53\nPevd/Penp29gBu3iPZoNs3rKooDhzbfLy3I3UelLirZIZMtNVu9OnsB+LVYWNOvjme+9jKom2vHh\nMSIRLmOttHNvX6g2d3Q/UENuWOV6qflZjz0135fHtESalVzjHZhLdl2Vwj1jBzkuk7BZiaUz4iez\nDJLJypzjstfb5UV5N7Qgm0FWSqrX28n5n2Bk82MtxNrPkfsYqYZ9OpnnJZMa3ibJosfO2HVtKkk2\nrxUK+tB81vPlk5GKlLaJbuettKC9M1sI8VnBtWdXPmPhB53qhkdrQfLKZplWFjknwXyMbAvCOTYS\n2nJjpZ2bvQ/tHaICUiPjHzKzJf8baTEbvTcY4w5/sWKyFeuz4hB8tMq09fxbrOcfiWgtC67Ns4Mt\nlxbivQ4/Yjgpwq3ytizN7fCyvDtaObSlYgNvHm0cjKGVHY8jfr6yLN7+HgV57Zh2XcT4fLRnKm3m\nZdKq1/I25KPzfzpW1cFjdA32Kh3apty6XxhDs0MkntvvqskSec1dWFqI9/QanWNn34O8NsJ7Pp44\nk0FAy5z1iHNLhNFrOher7CgX2OxGuILMhJ/H4Fo0H+1pF8ucJ6+hCfMdF/g7sDKl73Iv3hgYI9os\nWz7VkySxrrmD3y4txK2M80hg8AR171w4/lwy4vn8Ge26PXGuZVsz50MdLbhqrSaagKoE5ujdsAQa\nWfF5RH29kV3kPO2mTfPd9qeWhfeuC/1IH5PxtBLHvXl7wJevIftcvZhttaC0Y3s37+/K0kL8xOtd\nskrNFl4fsRc8rFI42ZZ5SDGUKWNHLQpWQM9WOHYIAu9IJKBau40Ef80/I6EIY0SbH6/yoZ3fjrFi\nNLH5fiybWc/dEmfSptZ5nv0l+PIcWl/teabyXBmHLf22W2Jki3++8Dh+zFpFIioS1JUXKnoJyYrP\npXXYKLNSLXN7i3hWmEMf2mLrVbm0fsRoTnnMCuZWNhX66RVIVhbVE+PyeprNs9UV6MOydzYLHs1X\njcmZ9QL6sOL0SXaNzsT89ved7Lh0RtzKhFtj5DGZeenZYVk7ud75wEcumJYtrfPk79qc8rudHP5d\nkX7kZcDk99Gc3rFIbFsZGqjRk9yw5rDiqyXU22y5dS+jrQ6gZ0cr/uJVlaOqSLSOs6m+hihOyzEa\n2UpG66O72XHpjPisRdFrR+mdxxPju71Ed5HZ4IyK8eqcx4E9R8nY4kkBjPgeo+f5Red4Ynz0XvDn\nGpENqvbvbSOqJtWwcz/W8x2JlSPnrm7LpYX4HQ+/p1Vl9ZfiXeG57klvbyF+uQbYd29Yhz8Pnu1c\nlm5NOY45u21vbI/zk0F7D67IxM24BvybK59j1S9pLZsP9t0b1uHPY0bb0Yy5d2B5IV7dmVX/gO84\n8i/Frv1L70LVmXvsUA3g2HoOd/hx9hz8eD7Yd29Yhz+PHhtW+KS/yVm6NeUucOj34C47YG8AgPeC\nuLw2Vwv3lVk+Iw4AAAAAsCIIcQAAAACAB0CIAwAAAAA8AEIcAAAAAOABEOIAAAAAAA/wVfz3Of93\nHMef190O3MR/X6/XL/Ig9t0KbLw32Hd/sPHeYN/9UW0sKQlxAAAAAACYA60pAAAAAAAPgBAHAAAA\nAHgAhDgAAAAAwAMgxAEAAAAAHgAhDgAAAADwAAhxAAAAAIAHQIgDAAAAADwAQhwAAAAA4AEQ4gAA\nAAAAD/AXFX8dagEKfI4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xdc35a20>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%pylab inline\n",
    "import os\n",
    "import PIL\n",
    "\n",
    "fig = plt.figure(figsize = (20,20))\n",
    "fig.subplots_adjust(left=0, right=1, bottom=0, top=1, hspace=0.05, wspace=0.05)\n",
    "\n",
    "for idx, img in enumerate(os.listdir('image_sample/prediction/')):\n",
    "    pil_image = PIL.Image.open('image_sample/prediction/{}'.format(img)).convert('1') \n",
    "    ax = fig.add_subplot(10, 12, idx+1, xticks=[], yticks=[])\n",
    "    ax.imshow(pil_image,cmap=plt.cm.binary,interpolation='nearest')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0xdc21668>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%pylab inline\n",
    "import os\n",
    "data = []\n",
    "basewidth = 50\n",
    "fig = plt.figure(figsize = (20,20))\n",
    "cnt = 0 \n",
    "fig.subplots_adjust(left=0, right=1, bottom=0, top=1, hspace=0.05, wspace=0.05)\n",
    "for idx, img in enumerate(os.listdir('image_sample/prediction/')):\n",
    "    pil_image = PIL.Image.open('image_sample/prediction/{}'.format(img)).convert('1') \n",
    "    \n",
    "    wpercent = (basewidth/float(pil_image.size[0]))\n",
    "    hsize = int((float(pil_image.size[1])*float(wpercent)))\n",
    "    img = pil_image.resize((basewidth,hsize), PIL.Image.ANTIALIAS)\n",
    "    data.append([pixel for pixel in iter(img.getdata())])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "scaler = StandardScaler()\n",
    "scaler.fit(data)\n",
    "data_scaled = scaler.transform(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([9, 6, 0, 0, 9, 7])"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mlp.predict(data_scaled)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}