notebooks/(testing) Learn ETL8.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "from PIL import Image\n",
    "from io import BytesIO\n",
    "from IPython import display\n",
    "import numpy as np\n",
    "\n",
    "def display_image(arr):\n",
    "    b = BytesIO()\n",
    "    Image.fromarray(np.uint8(arr)).save(b, format='png')\n",
    "    data = b.getvalue()\n",
    "    display.display(display.Image(data=data, format='png', embed=True))\n",
    "\n",
    "IMG_ROWS = 64\n",
    "IMG_COLS = 64\n",
    "\n",
    "data = np.load(\"../data/etl8.npz\")['data']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import sklearn.model_selection\n",
    "import keras\n",
    "from scipy import ndimage\n",
    "\n",
    "NUM_CLASSES = 956\n",
    "\n",
    "# 150 ~ 159 for hyper parameters\n",
    "alldata = data[0:150]\n",
    "dataset_len = alldata.shape[0]\n",
    "\n",
    "# flatten\n",
    "x_all = alldata.reshape((dataset_len * NUM_CLASSES, IMG_ROWS, IMG_COLS, 1))\n",
    "x_all = x_all.astype(np.float32)\n",
    "x_all /= 255\n",
    "y_all = np.tile(np.arange(NUM_CLASSES), dataset_len)\n",
    "y_all = keras.utils.to_categorical(y_all, NUM_CLASSES)\n",
    "\n",
    "# # distortion (gaussian noise)\n",
    "# hard_len = int(dataset_len / 10)\n",
    "# soft_len = int(dataset_len / 2)\n",
    "# keep_len = dataset_len - hard_len - soft_len\n",
    "# hard_seed = 0.2 * np.random.randn(hard_len * NUM_CLASSES, IMG_ROWS, IMG_COLS, 1)\n",
    "# soft_seed = 0.05 * np.random.randn(soft_len * NUM_CLASSES, IMG_ROWS, IMG_COLS, 1)\n",
    "# keep_seed = np.zeros((keep_len * NUM_CLASSES, IMG_ROWS, IMG_COLS, 1))\n",
    "# seed = np.absolute(np.r_[hard_seed, soft_seed])\n",
    "# seed = ndimage.gaussian_filter(seed, 1)\n",
    "# seed = np.r_[seed, keep_seed]\n",
    "# np.random.shuffle(seed)\n",
    "# x_all += seed\n",
    "# x_all /= (seed + 1)\n",
    "\n",
    "# #  distortion (inverse)\n",
    "# choice = np.random.rand(dataset_len * NUM_CLASSES)\n",
    "# x_all[np.where(choice > 0.5)] *= -1\n",
    "# x_all[np.where(choice > 0.5)] += 1\n",
    "\n",
    "x_train, x_test, y_train, y_test = sklearn.model_selection.train_test_split(x_all, y_all, test_size=0.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_1 (Conv2D)            (None, 62, 62, 32)        320       \n",
      "_________________________________________________________________\n",
      "batch_normalization_1 (Batch (None, 62, 62, 32)        128       \n",
      "_________________________________________________________________\n",
      "p_re_lu_1 (PReLU)            (None, 62, 62, 32)        123008    \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 31, 31, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 29, 29, 64)        18496     \n",
      "_________________________________________________________________\n",
      "batch_normalization_2 (Batch (None, 29, 29, 64)        256       \n",
      "_________________________________________________________________\n",
      "p_re_lu_2 (PReLU)            (None, 29, 29, 64)        53824     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2 (None, 14, 14, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 12, 12, 128)       73856     \n",
      "_________________________________________________________________\n",
      "batch_normalization_3 (Batch (None, 12, 12, 128)       512       \n",
      "_________________________________________________________________\n",
      "p_re_lu_3 (PReLU)            (None, 12, 12, 128)       18432     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_3 (MaxPooling2 (None, 6, 6, 128)         0         \n",
      "_________________________________________________________________\n",
      "flatten_1 (Flatten)          (None, 4608)              0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 1024)              4719616   \n",
      "_________________________________________________________________\n",
      "batch_normalization_4 (Batch (None, 1024)              4096      \n",
      "_________________________________________________________________\n",
      "p_re_lu_4 (PReLU)            (None, 1024)              1024      \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 1024)              0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 1024)              1049600   \n",
      "_________________________________________________________________\n",
      "batch_normalization_5 (Batch (None, 1024)              4096      \n",
      "_________________________________________________________________\n",
      "p_re_lu_5 (PReLU)            (None, 1024)              1024      \n",
      "_________________________________________________________________\n",
      "dropout_2 (Dropout)          (None, 1024)              0         \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 956)               979900    \n",
      "=================================================================\n",
      "Total params: 7,048,188\n",
      "Trainable params: 7,043,644\n",
      "Non-trainable params: 4,544\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "from keras.layers import Conv2D, MaxPooling2D, Input\n",
    "from keras.layers import Dense, Dropout, Flatten\n",
    "from keras.layers.merge import Concatenate\n",
    "from keras.models import Sequential, Model\n",
    "from keras.layers.normalization import BatchNormalization\n",
    "from keras.layers.advanced_activations import PReLU\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "\n",
    "input_shape = (IMG_ROWS, IMG_COLS, 1)\n",
    "\n",
    "model = Sequential()\n",
    "model.add(Conv2D(32, kernel_size=(3, 3), input_shape=input_shape))\n",
    "model.add(BatchNormalization())\n",
    "model.add(PReLU())\n",
    "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "\n",
    "model.add(Conv2D(64, kernel_size=(3, 3), input_shape=input_shape))\n",
    "model.add(BatchNormalization())\n",
    "model.add(PReLU())\n",
    "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "\n",
    "model.add(Conv2D(128, kernel_size=(3, 3), input_shape=input_shape))\n",
    "model.add(BatchNormalization())\n",
    "model.add(PReLU())\n",
    "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "\n",
    "model.add(Flatten())\n",
    "model.add(Dense(1024))\n",
    "model.add(BatchNormalization())\n",
    "model.add(PReLU())\n",
    "model.add(Dropout(0.25))\n",
    "model.add(Dense(1024))\n",
    "model.add(BatchNormalization())\n",
    "model.add(PReLU())\n",
    "model.add(Dropout(0.5))\n",
    "model.add(Dense(NUM_CLASSES, activation='softmax'))\n",
    "\n",
    "model.compile(loss=keras.losses.categorical_crossentropy,\n",
    "        optimizer=keras.optimizers.Adadelta(),\n",
    "        metrics=['accuracy'])\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/2\n",
      "1008/1008 [==============================] - 120s - loss: 5.5930 - acc: 0.0578 - val_loss: 3.7261 - val_acc: 0.2089\n",
      "Epoch 2/2\n",
      "1008/1008 [==============================] - 118s - loss: 2.2687 - acc: 0.4729 - val_loss: 1.2346 - val_acc: 0.6761\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x7eff792589b0>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from keras.callbacks import Callback, EarlyStopping\n",
    "datagen = ImageDataGenerator(\n",
    "    rotation_range=12,\n",
    "    zoom_range=[0.9, 1.5],\n",
    "    width_shift_range=0.2,\n",
    "    height_shift_range=0.2,\n",
    "    fill_mode=\"constant\",\n",
    "    cval=0.0\n",
    ")\n",
    "BATCH_SIZE = 128\n",
    "steps = int(x_train.shape[0] / BATCH_SIZE)\n",
    "model.fit_generator(datagen.flow(x_train, y_train, batch_size=BATCH_SIZE),\n",
    "        steps_per_epoch=steps, epochs=2, verbose=True,\n",
    "        validation_data=(x_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAIEklEQVR4nFVXa3McRxU9p6d3Zx/S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\nkeoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAHSElEQVR4nFVX2XIcxxHM7O3dxeIm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\nEIz3KU9Nia7WrGOpiRR820CA/B+gIslNTrxtewAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACj0lEQVR4nO2WS0/bQBCAZ3aX9SMh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\nMJeylP8tfwHvQwHo8nN91gAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAGI0lEQVR4nFVXXXMbxxHsnlt8HUkY\nosTEtuSUK5Uq//9/kjdXuaw4jiuJLIomEREkAdxt52Fm9k54OeBwuzvTM9Pdx3vrjBAgGFQFkCCr\nyAoSgiSYQZBAgRJIkhKqWAyVEAAQEvwjEASqQQAIQBApggApkgAASoWAIMB3IQWyxipUekBQBUj6\nXYCEP24q/gsQKP9GUBVGkRLilkCAAgCDhBZaoW/n8TNOhiAIgQ0zoDg/b0i+ge8IitD8kXb1fxs6\nCZT/UXwrz0kgoCmnjGy2BhNe/kypFt9FQNXkUEucZcQskt+JQCWZlfh/Ojd2d2zox8kTFUEBiohI\nAcW8slDUMEqOqbCo8tKApAdGKatnXcNXHgUTwAZlFiYbgQwAHURQoqQvl0XsFdbwin+nwyVCpTJa\nIIMnHWiPyfshoVTGMsFVlO2VvcXpQB8JTYuy22fbFK9JHA9hPlH+OMGoqjzfhhLz5BjauEhfADge\nQEBVMcbeAEBMlkVX16opL6KKpB5fANz/+LGSsTQaSVIeZrMatjLo8f0vJ/D00z9GCcPd6Mt9aVzy\nwEISsGhw32X88P727XdAVwRgc7VCbXVUnNX6rjjU9GfE+nyu+98evv5hAdjlWCEbx5aeX0TvS4BA\niVuxKcaf77Xth7evINl2XwlbDDFg2R8kIcLrU3y2ous1/u84XvyN7w+j+XSCZfNEepRt0iYCpQWe\nid6Hi82uv9gefn8CwOMBxLDpvAK0AHNGTzQfEf91/uP53V8vqsaH279/BEGwViQ+XjtE60e/1JwV\ngPr8yx/9bonzuH84vvmmI60ezqrn/dlBVFVrJKlWCaoloCWH+1/X320B/r5Z74abtYi+VONQL5fZ\n7N4wcSXJWosFHT7//HzzbgFqfOm+3aqrFDR2xqpNl+PQcFTwAq3QuYDj4tW3BE775e56acfT2AHE\nKHZ2ClbVpF0pMWRJDC5+EDQenxZbbnA+nsaOwOlIWbcC5ofPZp5AaeIncHiou7JflNr3xyrC1k+P\nV7rYOknOOJqAzCWxJD8Qqs/drqgbCXYrg0g+n1UXy8bDEGARhyCRKIGuYXga+iugDmvS+hVJLv7U\nraF+QTXJSiLIK0uooM73ZWcCobNEezwtCdy8ptaDQIhNXhNHAZYYCFi86QQCtixU3T/dXAGiCbf/\nvPm+03zZjPao4kzXOAEaLk0qq6ehiND9p9ux0OkikKDTSu54SEEJ+h73/brW5/+O7y6Fz6dHbS9X\nEGjyDUSaOyGApEpTyyBurpeCXXz14bE/3x2/eluavNb8kowkQSo1IG0a4I3P47+O/fL1yvsn3YR7\nLDJ0DoCKZMAkV4ARhpeHT4u3u2INrxB9xuSxCX3xgyW2fifOh9tjd3m9nJxQyhwM8GREGSQUJxtP\nSTSCw2HPP3fjajkh65Kan7nHqSWFwbwww90Jy9cbDquX8yKPZ2AULQwwUgJYmDaMIjj85+7Nri8Q\nVuw0MyhZ6XAfk7Es9PYIguguX/UdKdh6PC2mxoFqBQa6JZqSIEomGMO6S8O0eh4BaJCej+XQ1Tvb\nnO7F67/0TcFcWBTTWRk67xHS2NXheNh33aKU3Wqwoa/26eN+2zdR8KgPUZa0EnTarv++u7o6qF9c\nsutECjLo9Hm8WgOGOqO08H8KcfVuGA6/vqzser0IzRBBiuuVEjJKhGTFmaUhI2IcPw/nzTf8etO4\nfG50moFxA1XSIkafDnh5BK+Wix2MQf5f+CMvqjwsAgUURRoqgeNvL9erbSlm6AQxPScgOMjuyUMk\nQDCtrv8D7a7XaAdOHl0NZjIMIiGK5GNreBEYGd47oJJgVI0A/CCiBjkKZiW8SuhVl9n5YpctBY+r\n6RHcdUOQuVdjcoKmNg03XMPAZSZIF0gCqiEsGVwyJdv6mXnD9FF7qrTKxliKas86hzqXyec99a3x\neuEXu0Y9wog06xMMDNCcItPdhjLNCuYAxguPpvPyDUiTSPvC4mTGWY6NPgLdmkn5sXX2SkKwaJ6S\nmH7Dr+4IA7FWAya1IxnJk2K+YirboL0JiDTk0LmbyI6wVhLOjAsmW91qBAE1RwNu0sg2Cx5gesDQ\nTw8zd2rylwASJP8P62wDbGVAujIAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAHM0lEQVR4nFVX2XIbyRHMrGlgCJDg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\nbr33O+Pq0wAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAEV0lEQVR4nO2VyVMbRxSHf909qzRa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\naufpV/ni5H9BSOEw2YxOXAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAADq0lEQVR4nO2WWXMaRxDHe2b2YsHc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\nsIUt7L/Z38WSwo/mQVryAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACRklEQVR4nO2WS2/TQBDHZ/blOA+n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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAGi0lEQVR4nGVXa3MbxxHsXiwBkBRf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\n/gcJyjV+u5ZdkQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAD4ElEQVR4nO2W21PbRhTGP92RLfmG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\nsn8AP1OZ92v6/OAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_len = x_train.shape[0]\n",
    "\n",
    "# distortion (gaussian noise)\n",
    "hard_len = int(train_len / 10)\n",
    "soft_len = int(train_len / 2)\n",
    "keep_len = train_len - hard_len - soft_len\n",
    "hard_seed = 0.2 * np.random.randn(hard_len, IMG_ROWS, IMG_COLS, 1)\n",
    "soft_seed = 0.05 * np.random.randn(soft_len, IMG_ROWS, IMG_COLS, 1)\n",
    "keep_seed = np.zeros((keep_len, IMG_ROWS, IMG_COLS, 1))\n",
    "seed = np.absolute(np.r_[hard_seed, soft_seed])\n",
    "seed = ndimage.gaussian_filter(seed, 1)\n",
    "seed = np.r_[seed, keep_seed]\n",
    "np.random.shuffle(seed)\n",
    "x_train += seed\n",
    "x_train /= (seed + 1)\n",
    "\n",
    "#  distortion (inverse)\n",
    "choice = np.random.rand(train_len)\n",
    "x_train[np.where(choice > 0.5)] *= -1\n",
    "x_train[np.where(choice > 0.5)] += 1\n",
    "\n",
    "for i in range(10):\n",
    "    display_image(x_train[i, :, :, 0] * 255)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "1008/1008 [==============================] - 118s - loss: 2.3413 - acc: 0.4795 - val_loss: 0.3342 - val_acc: 0.9356\n",
      "Epoch 2/50\n",
      "1008/1008 [==============================] - 118s - loss: 1.1052 - acc: 0.7154 - val_loss: 0.1943 - val_acc: 0.9613\n",
      "Epoch 3/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.8139 - acc: 0.7859 - val_loss: 0.1714 - val_acc: 0.9636\n",
      "Epoch 4/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.6479 - acc: 0.8268 - val_loss: 0.1502 - val_acc: 0.9625\n",
      "Epoch 5/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.5441 - acc: 0.8535 - val_loss: 0.0932 - val_acc: 0.9781\n",
      "Epoch 6/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.4752 - acc: 0.8725 - val_loss: 0.0945 - val_acc: 0.9759\n",
      "Epoch 7/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.4277 - acc: 0.8844 - val_loss: 0.0959 - val_acc: 0.9738\n",
      "Epoch 8/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.3934 - acc: 0.8931 - val_loss: 0.0778 - val_acc: 0.9794\n",
      "Epoch 9/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.3644 - acc: 0.9024 - val_loss: 0.0750 - val_acc: 0.9801\n",
      "Epoch 10/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.3429 - acc: 0.9082 - val_loss: 0.0603 - val_acc: 0.9853\n",
      "Epoch 11/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.3210 - acc: 0.9142 - val_loss: 0.0535 - val_acc: 0.9861\n",
      "Epoch 12/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.3067 - acc: 0.9174 - val_loss: 0.0545 - val_acc: 0.9867\n",
      "Epoch 13/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2951 - acc: 0.9217 - val_loss: 0.0544 - val_acc: 0.9862\n",
      "Epoch 14/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2841 - acc: 0.9247 - val_loss: 0.0482 - val_acc: 0.9880\n",
      "Epoch 15/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2715 - acc: 0.9281 - val_loss: 0.0502 - val_acc: 0.9872\n",
      "Epoch 16/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2664 - acc: 0.9295 - val_loss: 0.0468 - val_acc: 0.9877\n",
      "Epoch 17/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2546 - acc: 0.9334 - val_loss: 0.0418 - val_acc: 0.9895\n",
      "Epoch 18/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2496 - acc: 0.9351 - val_loss: 0.0450 - val_acc: 0.9884\n",
      "Epoch 19/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2484 - acc: 0.9354 - val_loss: 0.0399 - val_acc: 0.9904\n",
      "Epoch 20/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2441 - acc: 0.9360 - val_loss: 0.0456 - val_acc: 0.9873\n",
      "Epoch 21/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2402 - acc: 0.9385 - val_loss: 0.0387 - val_acc: 0.9904\n",
      "Epoch 22/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2334 - acc: 0.9395 - val_loss: 0.0370 - val_acc: 0.9911\n",
      "Epoch 23/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2284 - acc: 0.9415 - val_loss: 0.0439 - val_acc: 0.9895\n",
      "Epoch 24/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2288 - acc: 0.9425 - val_loss: 0.0374 - val_acc: 0.9907\n",
      "Epoch 25/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2226 - acc: 0.9440 - val_loss: 0.0440 - val_acc: 0.9895\n",
      "Epoch 26/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2232 - acc: 0.9443 - val_loss: 0.0365 - val_acc: 0.9904\n",
      "Epoch 27/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2206 - acc: 0.9445 - val_loss: 0.0386 - val_acc: 0.9902\n",
      "Epoch 28/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2138 - acc: 0.9471 - val_loss: 0.0359 - val_acc: 0.9925\n",
      "Epoch 29/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2160 - acc: 0.9469 - val_loss: 0.0405 - val_acc: 0.9904\n",
      "Epoch 30/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2166 - acc: 0.9467 - val_loss: 0.0418 - val_acc: 0.9904\n",
      "Epoch 31/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2115 - acc: 0.9484 - val_loss: 0.0390 - val_acc: 0.9911\n",
      "Epoch 32/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2092 - acc: 0.9496 - val_loss: 0.0381 - val_acc: 0.9909\n",
      "Epoch 33/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2119 - acc: 0.9496 - val_loss: 0.0419 - val_acc: 0.9894\n",
      "Epoch 34/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2104 - acc: 0.9504 - val_loss: 0.0373 - val_acc: 0.9918\n",
      "Epoch 35/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2055 - acc: 0.9515 - val_loss: 0.0403 - val_acc: 0.9909\n",
      "Epoch 36/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2056 - acc: 0.9522 - val_loss: 0.0437 - val_acc: 0.9898\n",
      "Epoch 37/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2031 - acc: 0.9539 - val_loss: 0.0403 - val_acc: 0.9898\n",
      "Epoch 38/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.2035 - acc: 0.9542 - val_loss: 0.0397 - val_acc: 0.9911\n",
      "Epoch 39/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.1984 - acc: 0.9560 - val_loss: 0.0385 - val_acc: 0.9904\n",
      "Epoch 40/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.1981 - acc: 0.9569 - val_loss: 0.0368 - val_acc: 0.9918\n",
      "Epoch 41/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.1971 - acc: 0.9572 - val_loss: 0.0375 - val_acc: 0.9920\n",
      "Epoch 42/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.1970 - acc: 0.9571 - val_loss: 0.0483 - val_acc: 0.9900\n",
      "Epoch 43/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.1911 - acc: 0.9590 - val_loss: 0.0383 - val_acc: 0.9916\n",
      "Epoch 44/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.1918 - acc: 0.9587 - val_loss: 0.0411 - val_acc: 0.9911\n",
      "Epoch 45/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.1885 - acc: 0.9592 - val_loss: 0.0523 - val_acc: 0.9903\n",
      "Epoch 46/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.1881 - acc: 0.9590 - val_loss: 0.0382 - val_acc: 0.9919\n",
      "Epoch 47/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.1870 - acc: 0.9601 - val_loss: 0.0385 - val_acc: 0.9926\n",
      "Epoch 48/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.1820 - acc: 0.9615 - val_loss: 0.0373 - val_acc: 0.9934\n",
      "Epoch 49/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.1786 - acc: 0.9618 - val_loss: 0.0373 - val_acc: 0.9918\n",
      "Epoch 50/50\n",
      "1008/1008 [==============================] - 118s - loss: 0.1774 - acc: 0.9626 - val_loss: 0.0382 - val_acc: 0.9922\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x7f02983d84e0>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "BATCH_SIZE = 128\n",
    "steps = int(x_train.shape[0] / BATCH_SIZE)\n",
    "model.fit_generator(datagen.flow(x_train, y_train, batch_size=BATCH_SIZE),\n",
    "        steps_per_epoch=steps, epochs=50, verbose=True,\n",
    "        validation_data=(x_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model.save(\"../etl8_noise_smoose.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "9504/9560 [============================>.] - ETA: 0s"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.021832448980774903, 0.99529288702928875]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "evaldata = data[150:]\n",
    "dataset_len = evaldata.shape[0]\n",
    "\n",
    "# flatten\n",
    "x_eval = evaldata.reshape((dataset_len * NUM_CLASSES, IMG_ROWS, IMG_COLS, 1))\n",
    "x_eval = x_eval.astype(np.float32)\n",
    "x_eval /= 255\n",
    "y_eval = np.tile(np.arange(NUM_CLASSES), dataset_len)\n",
    "y_eval = keras.utils.to_categorical(y_eval, NUM_CLASSES)\n",
    "\n",
    "model.evaluate(x_eval, y_eval)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "score = model.predict(x_eval)\n",
    "# predicted = np.array([np.argmax(r) for r in score])\n",
    "predicted = np.argmax(score, axis=1)\n",
    "\n",
    "idx, val = np.where(y_eval == 1)\n",
    "ans = val[idx]\n",
    "\n",
    "correct = np.where(predicted == ans)[0]\n",
    "errors = np.where(predicted != ans)[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "45\n",
      "9515\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACvElEQVR4nO2WW2sTQRiGnz1kT91s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\n0Hg8jQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAADR0lEQVR4nO2W31PbRhDHvzpJlpB/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\nZNFzentFF88/QEgk0VeykpUskn8BGUpBz07uiK0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABeUlEQVR4nO2Uu2/CMBDGz3FePBva\nJhO0hQGpEkKd+/8PHak6oIqBh0pB4hFIIODg2B2oOsUOj6WDv9G+++nuPp8BlJSUlJSUlE6VYZwQ\npImvcOs1fxUAFapeNkAXX7EN0RFH2DTZhl0ECKirOWXTgdXb8hIAbJJ6O2S7mK2IOEgKiMPpOEhI\nQg6XAZKIdnwuCQCQugB0TQ5Z+VIAIM3OypcCWGyU0DUAHiaZBchb2BqubMgAIHMBY2xhy5A4KAFo\n5fvyLcpb3nre38uNSBsSKjZdjIhPco3iNB4OJO8wFYAbz9ZoFO0ooOajX6l8d2eSIlJaKLb098Hx\nBQXx+OOh2u5OztpGtFv2f0fHyTbqLeovxlBYQ4qN4Zf1d0p1SOaf5E7sdkoFycwrHOeGS3oMwIMO\nOu9D8Ve1gAIAlLxVBAB8AeIpptUW9+wnCwDsWm5yAADgEhfSl8WpM3+fd9FiSMWpMgCYN45Ng3Cb\nvU3CddU0zjJ/EyWlf6MfhIqL+esa6PgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABn0lEQVR4nO2VS0/CQBSF7/RNIaEQ\nCMESTEDjRk0wRn+8C1euXZFoTAhawrsgQl902nEhuimdEWbbs+zc+XLm3jNTgEyZMjEl1jVmjUBb\nzN+bfAAkczoIAp26zgQQgggXAEBCfADC3E8HxNuQD0B+9yM5l+ZFogICEQGSZMW4kB+dIwBaITak\nSkuWFLWLj3HQPBXPBY0sBvOxndYNKsAnCgwn9nwVpNdQARPbena9iFZCB2BnPGNFkR4kzBkkQDJn\nlBWFzwESw5gLIGk+nwMI6CNkAgjzACwA2wCjiZHL6eAfQUpGWS4I2zD6uXxx2iWmAPTbNnZCPFu4\njito7C4mAPXc00YQSvlGUZyu8+rhAOX9jQBYSKmYRrMxG2wPBcQ5RAAI8Qcj3axcii8z+iwTU7DL\njd23aG1ZFr6+q1NvVMLBcnyV7+2eMDydDEvtm2V/nn6QBCDqFzrt3sgNAFBB//LHn8ZJp/eaOo49\nSVFbZ+VobXug1qIHDwCkarBMfdn2RQ2pRtWsadhbdT+YQUjJKpKKeow3DvPvnilTpkx/+gZLR5SZ\nJ7/bhwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACwUlEQVR4nO2Wy2/aQBDGv901GGMM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\nX9hWC4jyiCieAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAChklEQVR4nO2W21PTQBjFz2Y3aSlt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\nKNNpXHA8mgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABdElEQVR4nO2UT0/CQBDFZ5cW2hSB\nBqJQqIKxgCEx4eD3/wIa4z88iAUUMYCpUIJYLcWL0UindOFk4r7jzrxf3uxOFoCLi4uLi+vfiWxc\nBAAaV7YUuzfbDBCVMiltHhPJc3P4tlgbIG7vJaUXu++5kiG7JxZOCAIQuZRLv3fvrQ8AACFVi9z2\n5msAlH097ra61rdHKFUaLSyDgNkj2druwrwZeD9HrhmvWiM2gFQw8vTpvO/8Op03pB3b83f7AJHi\nYdodmO3pcl6no4nOcjcCEOuZh47fDkDSaowF4Jwl2mMkKsg6emG+s4UJ6IOTUuEOCYBAA/ZFP5qZ\nr0iBou1+v3asNHsYmxGQrKvNK2wCRkC0nGidTtASuom+pmK2c4n7Q/8DAADBOOhfTwOKDCOIRtkK\n9DMAaL46vAj0h49AcpVJA1uAL4VdIknpw/YKf2iCpOY8ou/PmEBQRwP0J2NMQFXPXu0PeYUoHYf4\nQ0SZFpWL6w/oEzn5dXoHbAxsAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABN0lEQVR4nO2Ty1KDMBSGQyDcUsRx\nLKMsXDm+/xN1HMa6QBjoDQohiYtuLObU42XhIt8OEr6cc/5AiMVisVgslv+Fc3ExZJ4WQkr9E4Eb\nJWniSncrdbsdQAUgYOHdInDedt1ElRckrGy/JQiWSxk29XGQp2c/i156vMDLcn9fdOJD1f5t+Ho0\nC7x57Wnq+9VmP501LTruAhWcC9hjvqif28N8ZHpUvFcIgRrFetUYJj7Rq3ZECOSqUNK0TTvMwwiI\nFsZdxCGj+XtCza/nRAy6SfMUANTOPEJsBTTWG+NssAIWTx3QA0pAOel+1QK/HqECUAL3JmiAeFEC\nmoT9AK4iYozSQwVEgBE4SSZK4BZiBC7PVQ0O4EuBEz94svr0d+MF/CkpCzDB0xkXBf79VMEBYKCX\nD7BY/op390pwln34aiUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAB4UlEQVR4nO2UXXPSQBSG3yUJkIZC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\nEv0f+gE+epwVBNg6qgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACbUlEQVR4nO2V21PaQBTGT7Ih3IYI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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABU0lEQVR4nO2UzU7CQBhF73RKGSlQ\nqPwEXQAxBjfy/m/hSnBJjMEghdSCDfRnxoXgAjK0kxpXc1Zf8vWe3ElmCmg0Go1G8x8YzXK+72QL\na3xLCgl4VCsVEwSlep4KpnQTbhlNCjRIw9htmNK1pIHVWqSHUQi304uC2VqoNGAj5zhWbu6apdW+\nbys1SPb0Z7D7LvGpN8WwtlURcMEAgHTGdLqsPl4h9rperNJgwwHQ/mAzDcRu3rH9bY+pCEQkAHZ3\nPXuLAB6aDx9w6hsFQRqViTMyJmsOADxYul06u5g/vUix0743n73DLp2/Vs1ASfA5HPhPh85GJeU8\nI34m2AVicswYxMtxlU8EyQvZH2dmhtn5s8cU/U6k/pVxhwBceEywumGOE8gFxE2z3tFlQaP1vsuR\nlwos11/lKSAV1JIFz5OXYbdZkThowyqUh5H9L9RoNH/MN+9mbMY7PJ6SAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAADR0lEQVR4nO2W31PbRhDHvzpJlpB/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\nZNFzentFF88/QEgk0VeykpUskn8BGUpBz07uiK0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABi0lEQVR4nO2VS08CMRSFT6d1mEEE\nJYioRGJUCLJw6f//CxKMD4hC4oP4GifiPBhaF5C4sO0os7Rn1Uzu+ebcm9sUMDIyMvofsjLWuO18\nNoDdKGcEhNP0JnQVq84HzwRYoSLVrwVQZAREAZkfbFtdxDQATgQAOLVt0ntfBjDjAMhaS/TjYKkE\nzCYg1UNvEGqKdDNgrk22jp8udX5tgpgVrObjYKbzSwHuRpXF5NWzp7QR3Oj9IJJPnZMXn7PQ2SmM\nojPl+NUJxE08CgGU6qVGd5Lil7YwuQIAUsszGqVeBvUQy60J9ytFLwUgmcFchVN617xO3IG/XAKn\ng264699XtnNviQ6gWCR6tHn+GIlcMh7l6hv0zwnocb03Fhw2RHCXr5a8D2UKaQKyd9AfciSfjABi\nMhzb645q56UJqvsXtxyYeUIAAA9CYjEuDyEDF9sP11MA4nm26F7wJFY0IQG4zdfFBfJ/8bT8bMHa\nC/rx4seWck00ABYNo8WRx6l+icg30yr+5nU0MjIymusL8MGC7JcU0tgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAChElEQVR4nO2W2W/TQBCHZ31hO2dT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\npAP+JKAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACvklEQVR4nO2WS2/aQBSFz3j8wLZM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\nCJ0aWCUBAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAC6UlEQVR4nO2WS3PbNhDH/wuAT8VW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\nrkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACOklEQVR4nO2US2/aQBDHZ9aPtY2h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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAC8klEQVR4nO2Wy28URxDGv+ruea5h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\nLQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAA80lEQVR4nO2TzQ6CMAyAu61jKMSf\naDTx4Ps/mVFjMBECKIx5MDEiq6Dx2O9W2n7r2AbAMAzDMAzzZ8TntBZKKtXk1U+CcGlcGoAeYbK3\n3wuC8SY4ZRkIiUJDfv1SgLOoLIraPSI1d+eGKPSvHmt7uLpnbKsoKIcLMJzmSe1eP8koHy4Yx+Wx\n3Q6AQPR7BMZkheuUSeoY3gTCWHvp1uIEyb/d7hfwPvyjClNqgpZYav9pq63b3QiBfA0aomoapeRd\nbgnANz5AvDid/ZmOwEu4rRPiGkLvawQAXMOhptO9Ar2oqGcwCLWaD9jlJ4Hp3yTDMP/gDlNqRzUi\nLf6tAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABN0lEQVR4nO2Ty1KDMBSGQyDcUsRx\nLKMsXDm+/xN1HMa6QBjoDQohiYtuLObU42XhIt8OEr6cc/5AiMVisVgslv+Fc3ExZJ4WQkr9E4Eb\nJWniSncrdbsdQAUgYOHdInDedt1ElRckrGy/JQiWSxk29XGQp2c/i156vMDLcn9fdOJD1f5t+Ho0\nC7x57Wnq+9VmP501LTruAhWcC9hjvqif28N8ZHpUvFcIgRrFetUYJj7Rq3ZECOSqUNK0TTvMwwiI\nFsZdxCGj+XtCza/nRAy6SfMUANTOPEJsBTTWG+NssAIWTx3QA0pAOel+1QK/HqECUAL3JmiAeFEC\nmoT9AK4iYozSQwVEgBE4SSZK4BZiBC7PVQ0O4EuBEz94svr0d+MF/CkpCzDB0xkXBf79VMEBYKCX\nD7BY/op390pwln34aiUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAB7UlEQVR4nO2UTU/bQBCGZ22vQ0KM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\nXSdkgGJIyMY5QqJEiZ6M/gLVmKkJIgS9XgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACbUlEQVR4nO2V21PaQBTGT7Ih3IYI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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACJklEQVR4nO2V32/aMBDHz05IQgKB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\n/NawseZl85ekSOudIv0P+g341rCdvOjytgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACUUlEQVR4nO2W32/SUBTHz723LRQE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\nhf4Cxu276wa0rocAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABf0lEQVR4nO2TS0/CQBSF77SdYmuN\nSiAWQhc2wfgKCxMX/n7duNDEYExEIOUhBCEgQgN9TMcFibroTLVNXM3ZTe6ZL/feOQMgJCQkJCT0\nX5KL+USPxCsaFxa3ngjAaEfOBAgiQJkAMlI3HSg606fwAP5SoQCAC8fqjZsGAIqKAO3VTL/hpepA\nwlsYjivEaU9oKoDnGlU1P2x9EE6TPACNckanMQ95Hv4OorXT4RoSnpEGhDn77wAUZUuihHG2JEZy\nLrEDzhIRwrqWHoDtQ0kyNXPocl+R+dvU2pkRSIY2XfQHs3dmkNkA6+qtjbTTyiLEeug0+2sWgDGC\ntBs5XZAV/No9KJes0uBhzIgzA4BkWQUg7mrU69eLFcsuvNRXfwGQqb+PA5SjhNJ1f9is2ufkMXYT\nrBxMZuUCktQwAAAaju9uR2Yx1sgCrJ6iS9tQveXm6LWun+NHYEZVPqptz2BwH3w7Yz8WM0ik6Z/o\n897X/fjrnA4AkK75y4hdFxIS+qlPBOF+MHun7VYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABn0lEQVR4nO2VS0/CQBSF7/RNIaEQ\nCMESTEDjRk0wRn+8C1euXZFoTAhawrsgQl902nEhuimdEWbbs+zc+XLm3jNTgEyZMjEl1jVmjUBb\nzN+bfAAkczoIAp26zgQQgggXAEBCfADC3E8HxNuQD0B+9yM5l+ZFogICEQGSZMW4kB+dIwBaITak\nSkuWFLWLj3HQPBXPBY0sBvOxndYNKsAnCgwn9nwVpNdQARPbena9iFZCB2BnPGNFkR4kzBkkQDJn\nlBWFzwESw5gLIGk+nwMI6CNkAgjzACwA2wCjiZHL6eAfQUpGWS4I2zD6uXxx2iWmAPTbNnZCPFu4\njito7C4mAPXc00YQSvlGUZyu8+rhAOX9jQBYSKmYRrMxG2wPBcQ5RAAI8Qcj3axcii8z+iwTU7DL\njd23aG1ZFr6+q1NvVMLBcnyV7+2eMDydDEvtm2V/nn6QBCDqFzrt3sgNAFBB//LHn8ZJp/eaOo49\nSVFbZ+VobXug1qIHDwCkarBMfdn2RQ2pRtWsadhbdT+YQUjJKpKKeow3DvPvnilTpkx/+gZLR5SZ\nJ7/bhwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACJklEQVR4nO2WXW/aMBSGXyc2SRgr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\n3CbquLoEt/yg8a/ZSS9TfwBIet4aBlQgTQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAC/klEQVR4nO2WSW/bSBCFXxd30dQa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\nnvUb+gngKUK89NxlWAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAC00lEQVR4nO2WS2/TQBDH/7t+JHGc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\nEplGMgTjX9TUv5A0/7ml/K/yG+cmFOc9yZi8AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAADV0lEQVR4nO2Wy2/bRhCHf7vLtyRL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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACdElEQVR4nO2W20/bMBSHj+M4l7Zp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\ngg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACxUlEQVR4nO2W2U/bQBDGZ9dnbMeJ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\ntvGK4w/MOQ+8hvpfSgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACx0lEQVR4nO2VS3PTMBDH/yvJjyaN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\nsLe9/Rv7CVGtSRjhjZNmAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACoUlEQVR4nO1WyY7TQBCtqt7sOEPE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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACaElEQVR4nO2W227TQBCG/107dmKn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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAADR0lEQVR4nO2W31PbRhDHvzpJlpB/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\nZNFzentFF88/QEgk0VeykpUskn8BGUpBz07uiK0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAADNUlEQVR4nO2Wy28bNxDGv+FjV1pL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\nDfsXwyd1jSQnVZAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAADSUlEQVR4nO1WS2/kRBCuqnbbYzvJ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\n4z6FLV1M6df6Lbv/zPCtf7YP9v+2vwApE364BQNzmQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAADYElEQVR4nO2W227bRhCG/93l8mBJ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\nAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAD50lEQVR4nO2W308bRxDHv7t3u3f+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\n/3S5l3u5l7vLP8jnqgXdZS3cAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABx0lEQVR4nO2WSU/CYBCGpx+l1FIK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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABx0lEQVR4nO2VzU/iQBiHfx37AQVs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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAADlElEQVR4nO2W3VPbRhDA904flgRY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\nrkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAD7klEQVR4nO2W21MbNxTGP612vV4b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\niBgYwABNqvibdGVX9v+wfwFwbbpCkU87kgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAB1ElEQVR4nO2VS0vDQBDHZ/NqUhvX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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAB1ElEQVR4nO2X30/CMBDHrz+2wX4A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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABSklEQVR4nO2US0vDQBRG7+T9amxs\nEUugKa58UHEjrlz513WnKIIIootIX5pqrEmb1DaJGxfC3NBJceec5Td3DjNzZwaAw+FwOJw/hLAV\nkaJYV0AUy7C2pUnQW6whEO1mw23EoQyaNLh/xVZRJiBgbliq0xHePvrjOAOlfuhc+jmLgJiyWNs0\nNNeMJu/DIM5+cvWocz6g10ALWmcizPPPIIySJPs9oJzoF1MGgeZk2WyOLBaap7dPVChRSTpE5gIA\nQByrdCiUVSPkC6QNVQSSjVyFKgJ5GbEKRAMJSbtgFuQpEureCInpLgAAYE+HtKCPNJf9DOrewxiJ\nmQV29+sZu134Fugy72B6hR0M24cC+t5O7w5pAaNA3erqN/4SH1wlkPWa27aC6xG2fwaBdLxfJPGj\nPyutWCEQd/WXMM3KflQGhCqPhcPh/G++AdmFYhu7cQOlAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABN0lEQVR4nO2Ty1KDMBSGQyDcUsRx\nLKMsXDm+/xN1HMa6QBjoDQohiYtuLObU42XhIt8OEr6cc/5AiMVisVgslv+Fc3ExZJ4WQkr9E4Eb\nJWniSncrdbsdQAUgYOHdInDedt1ElRckrGy/JQiWSxk29XGQp2c/i156vMDLcn9fdOJD1f5t+Ho0\nC7x57Wnq+9VmP501LTruAhWcC9hjvqif28N8ZHpUvFcIgRrFetUYJj7Rq3ZECOSqUNK0TTvMwwiI\nFsZdxCGj+XtCza/nRAy6SfMUANTOPEJsBTTWG+NssAIWTx3QA0pAOel+1QK/HqECUAL3JmiAeFEC\nmoT9AK4iYozSQwVEgBE4SSZK4BZiBC7PVQ0O4EuBEz94svr0d+MF/CkpCzDB0xkXBf79VMEBYKCX\nD7BY/op390pwln34aiUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACGklEQVR4nO2WTZPSQBCG357JZEgI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\nsX9rPwBB9LpVZWWhyAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAEcElEQVR4nO2XTXMTRxCG3+6Z3dXK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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACXklEQVR4nO2V23PSQBTGv7OkkKQU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\n9sSkn2JUSdVbMNdc/49+A4Ra4XxN0dSpAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACyUlEQVR4nO2Xy3baMBCGZyRfMQRI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\nbPHP8l7eC8BfhgUYjwdOd0EAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAB70lEQVR4nO2Wy2sTURTGv3Pn1cmk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\nycjQluZxQX5MuePHWKlSpXP9BhF0qZt6gy8/AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABn0lEQVR4nO2VS0/CQBSF7/RNIaEQ\nCMESTEDjRk0wRn+8C1euXZFoTAhawrsgQl902nEhuimdEWbbs+zc+XLm3jNTgEyZMjEl1jVmjUBb\nzN+bfAAkczoIAp26zgQQgggXAEBCfADC3E8HxNuQD0B+9yM5l+ZFogICEQGSZMW4kB+dIwBaITak\nSkuWFLWLj3HQPBXPBY0sBvOxndYNKsAnCgwn9nwVpNdQARPbena9iFZCB2BnPGNFkR4kzBkkQDJn\nlBWFzwESw5gLIGk+nwMI6CNkAgjzACwA2wCjiZHL6eAfQUpGWS4I2zD6uXxx2iWmAPTbNnZCPFu4\njito7C4mAPXc00YQSvlGUZyu8+rhAOX9jQBYSKmYRrMxG2wPBcQ5RAAI8Qcj3axcii8z+iwTU7DL\njd23aG1ZFr6+q1NvVMLBcnyV7+2eMDydDEvtm2V/nn6QBCDqFzrt3sgNAFBB//LHn8ZJp/eaOo49\nSVFbZ+VobXug1qIHDwCkarBMfdn2RQ2pRtWsadhbdT+YQUjJKpKKeow3DvPvnilTpkx/+gZLR5SZ\nJ7/bhwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABHUlEQVR4nGNgGAWjYBSMglEwCkbB\n4AKMeKRY2YSF/zN///T10x/SDWBkkVDlZxRlev6dWZDt8aXX/0kzgFlSTor91/unH/58/s3MJarL\nduwxLhOwGcCtLi/84/HN9z9hAvw2rIfeEW8Ao6H1wyuvviJbKWb+4jyOcGDBIvby5JVvqCJvn0sI\nv8ThBGyACcNdgu7aTDjUYhP8hxFiXz8IsZJgACZg5f+NIxqINEBC6OdfSgxgl///lhIDmGSkXr3B\nJUeEfkYpg293vuGQJMIAJilzjitPccliS0hoKuSsOI/f+4fTeQS0M/KpKP27dR13fsZvAJOAhDbP\ng6u4YoCBgZAXWPWUn519+hOfEgIuEOZ8/QNnWTIKRsEoGF4AAI5GUGm4WC6uAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABOElEQVR4nO2XXUvDMBSGzzlpU9Mq\nWx2bZahj///XeCuC4GDMWVhb+23WeDlcM5a6C73Ic/m270ObhIQADAMBAQAPAQ3sIyIA4vk3LRaL\nxWKx/BVnDmrXURI6VN2vBHw0DXjTSUZ5nH0NFZAfjUJW1p/EWQey2aXtAAHnk7EgxevVrgUXFHkB\n894zQwFdhxPPc7JtUdXy4JzyTWUkwOhBtIms8lz+yJnwd5qBcPrRXpSbtGnVcdwunLWJQMVPsjhu\nAwB0DVF/OjVf0CWaNgAA3jj9mdAITuK5uiE37/ui2V8kEJTIfmouCOZOoYmNBexunF4ioNl9+ab5\nA1MB3i5plWvNZv1wcRV/aDcFk3XA3NmjG7/qdwQDAUZLn9Yvpf6pgUDlAp7X9Sn9eQHQfL/VrMEB\n2DuS5f/zDZhWYLz82W1NAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAB7klEQVR4nO2WwW/TMBTGPztN0jbZ\nujXdCoRB2SiHreJQtMGFG+LPRpwQEkgUoXal1cTQRNWKqaxrsyWpHXOYJiE1sUtzQSjf6Unv+adn\nP3+WgUz/t8hSVVQCWI6QKdO/Lm2zsESVxAtrr1/mUwH8sH4nHaDL3VwaAE6HtVIqgN9HlYIQQqhu\nWwmvQ3yPRM8VigW+zh/OQ1vA3Fq3wjejpQHEflwv27pg/GLjUUAv55QH1z6L7yCuse1X99j0fDRg\nRmOn1Y7CKBIifnVCB7P2u3EQACB39/gsSlybCLj6fJss8bLhywGyKRDXDamsQAUwnxg5Q+UH2VWr\n1rwZVHaQdKC5ZvdMKM5QBqjWznu/iubKgHzDbl96wlH4KRFA93ZOe9F0YmkrAu4/Z198cGYq5piU\ntp7qnZ8AG5v2SgBScwfHDJiPub4SoHzgtTwACA3FGOIB+ebW14EAgCkq8lOMBWj79R9dDgDgUIwh\nDkCq++P305uYTTT5GOKyay/IxxFAKQARJL8lCQBiNyvtbwLQdzcJqGHIO1i4qPRB0+l0OIDw+qj/\nHcKX22kBUHpWOW75ACCGxYbj6aF8CwuA8Kx3cnUT8pPZUZ315d/FBYD3KeK3cTR86x86fwnA/I9Y\nTD7oWroPK7EchZ8zZcoEAPgNC22bARSEdr0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAADGUlEQVR4nO2X22/TSBTGvzNjT53Y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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACYUlEQVR4nO2VTVPaUBSG35PkJoGE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\nxJLpZjPIepZIpDkLCy200L+l32bH29x+MDHHAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAADUklEQVR4nO2Wy2/bRhCHZx+kaImi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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACz0lEQVR4nO2W227TQBCGZ3e9PseJ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\nyiWdr8EN+juXtxf9X/oBsw0dBvZ0b2YAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAC7UlEQVR4nO2VW3PiNhTH/5LlK+ay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\nSUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABuklEQVR4nO2WzU/CQBDFp7sLpB9A\nDRTRoGiIGjXxoFEjiUf/ci9GSTSoIVETNfGjCFqkLQVKqQfjjZ027cFL33Vnf30783ZTgESJEoWS\nQPD1gGWA8k46FoBtHxRjASRtMIoFKClP33EAqXnzeRIHIBY/DbwiACATL8OIgFQwdL8wxypVvWtb\nlu1OOSUoIF0vflTzU6D9futuNgF3wGT7uqnJkizJy69mBEBWfLfHPQCayjCH8w0UoLDeBADA84a8\nEnwK/tjk9C4kwGW4w0CA6Uj4mIIAVkehsQCTTlaJBQDdqwRYCAAM2uVsLIDfoRrexqAHpdcvpSIC\nfs/utnMqCuAFRdjccAa9tuV8Ea2LpZGbNLHGTB+Mly7kaBSA/7iZOUtn1cUlVbzvRnFgNA5p0xfV\ndW1l95RzlQGQJvpPen2d2G9vtltZRcLEn8Lwyjw5Vgs7xBA3CnwAkhJSO5qzPOljVHIfGuMIACD5\nWk0xmmZFVm8feZPAc0pzyvBzSujW8oUeCfAncV86/+LYDAVwWnRPmz2KcA5AWNiVb1qz+hDOAfj6\n5XRNClfLESks4f86wQp52kSJEv2TfgA1noLMoME9pwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABn0lEQVR4nO2VS0/CQBSF7/RNIaEQ\nCMESTEDjRk0wRn+8C1euXZFoTAhawrsgQl902nEhuimdEWbbs+zc+XLm3jNTgEyZMjEl1jVmjUBb\nzN+bfAAkczoIAp26zgQQgggXAEBCfADC3E8HxNuQD0B+9yM5l+ZFogICEQGSZMW4kB+dIwBaITak\nSkuWFLWLj3HQPBXPBY0sBvOxndYNKsAnCgwn9nwVpNdQARPbena9iFZCB2BnPGNFkR4kzBkkQDJn\nlBWFzwESw5gLIGk+nwMI6CNkAgjzACwA2wCjiZHL6eAfQUpGWS4I2zD6uXxx2iWmAPTbNnZCPFu4\njito7C4mAPXc00YQSvlGUZyu8+rhAOX9jQBYSKmYRrMxG2wPBcQ5RAAI8Qcj3axcii8z+iwTU7DL\njd23aG1ZFr6+q1NvVMLBcnyV7+2eMDydDEvtm2V/nn6QBCDqFzrt3sgNAFBB//LHn8ZJp/eaOo49\nSVFbZ+VobXug1qIHDwCkarBMfdn2RQ2pRtWsadhbdT+YQUjJKpKKeow3DvPvnilTpkx/+gZLR5SZ\nJ7/bhwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAADqElEQVR4nO2Wy1PbVhTGv/vQ0/IL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\n7dvW/5lhnXCB/pNjAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAD/ElEQVR4nO2Wy28cRRCHf9U90zOz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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAADJElEQVR4nO2XTXPbNhCG312AFGWZ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\nTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACxUlEQVR4nO2W2U/bQBDGZ9dnbMeJ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\ntvGK4w/MOQ+8hvpfSgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABpElEQVR4nO2USU/DMBCFx4mzdQGS\nNiqUooJA4oIqISTO/HsOcEdcaBEtpagt3eI4i21OIBCJ0xJxQMo7JvM+PY9nDFCoUKG/l6Zn1yiS\nf6h9UckFUO2WlQuAy7NVLoBljYJcgG0x5nkAihmQTD9gGYBSQADilwBUtsO6KJt0/CZDoJ+frBIX\numZUnXpsetSqPl2HmyVwL3GgVjD3/B6d45PFfSTxJyXQXVcnNAwO67cj40ibzn22WYJwMAAA0NtR\noDWr2FXuBpsBPoKYQcN2eDRcTSX+VADeOnBYM37uLT3ZARJ7AMiouK22Net3J34sdackaJ3bKplg\n2n2Uz1AqYDl8Hcz1DiZr+BMBsxsAqJmRbH4+lbpMuhJn77IMgAWVtz8DgCwjyn4MZAm4CHIBBIuV\npBlZP0HMLdlzlw0Q0Vo9TAfM/UrpW+HucXUjACF7h1+nTIjOVS2hLnWd2cP+qXghEecMABQVl3hi\ncXqnkXPWQOGSMY9psYnVkt/tJ+ym7Kp02ykjVfXUHY/RBfFJ0mDI7xohQCAAgQCxzmoWKlSo0P/W\nO77OlR25i605AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAB4ElEQVR4nO2Vz27TQBDGZ2ftXbtx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\noP+ie7CWpdmE47QNAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACAUlEQVR4nO3VzW/TMBgG8MeO67Zp\n15IqGqLdKiEqTUPlsiFx4a/nwGUMOCDBBNuhjHYj6wfpsnzVMYe1wCGOUyIuKO/JkvP+9CZ54gBl\nlVVWWf++CMlzFVVv2Qc8B8DUW49e4EwWmWDGDmv6CTKA649NqxAQf0r6RhEAP5bdRiEgvGatQkAy\nQSdrXwXQXwGaJnvaKKQA+/2NEC+a5l8A1rC6AZxKW5fnlCSu6puXJ33ep1EQ+KutgFuz5d2vqrZ1\ndGBEvvPlq5JIAVzRmQAA3R0+ZvLDmPcG/VfnSX4guGsSCbDBs+nrYXc5Sj6/f7q/cPIDoCYVqB12\n341E26YJ5Pzs+OFUMUIKkPhVJurPH5xOJG4EJxJwXZsH6UDKa4wdXq8d2W8nEpgvJAEQz7gqUWkT\neLQ2sN+MJQDfq9AEkIKqMp0CyNA8bpxeSgCEG+2qIKyxx+L8ADzSG31bH2ZB5yU4t8ITfwvADfjF\n/SOTt56V0Hh1df59ixwgHPXm62UShScuAFV7OiBuVsv10tgxiLpZBWAyDtcrgkhzsqcCvzNDuMz4\nEoHsIw0AYRm3r57gjxJCA2gmqNfvRCGgZTiqCOYCyE40y+7XAHw3WmZeoAMMeqk4BnICplB+A7kA\n9sSbavqzgWbnKioCVHrjsa4/M4mN8EI7ADJ+fYRBE6Ky/pv6CXqnsGV6WDudAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAADR0lEQVR4nO2W31PbRhDHvzpJlpB/\nWzZ2TBjMuDUUkuk0M2TaP6BP/aMz03amD6V0kjiBAAWHhEz8I8ZIsiVb0l0fLAks2RaZPtb7tLd3\n+9He3e6tgJWs5P8ggszHrCDLZ3d+2YshCEtn2fixeN3/DxGwT5fZ7eUhLAfA+iR/k/oaAFdsFLl7\nY9o11I2lIYTOgN87OPyN3eFTKsn9mGy79sh0HwZQFTvw55XNWjVNMnubIrE6p/25iBBAztJbD8AX\ntktp4ziz+/mFJRVL1fLZuRkPyKzpX6YaV35ebTdbZmOb6CN0+PzTg8LhMBZAmGlMlfx+5vUrjWE8\n3ZLb+8N6Ovw7uosQIMEGLgCQzSfp5psxgMnQ25LVrDQ+tCOA2WvkZH5gA0DuWfnd2zEAcETw1hgX\n+Vo0bWYtfE4xHACJWvG8aU0XJERvDbsaNPIxANCJDUDY2e2/sT0T4/xEGjTlnUhSzQKoPqYAt/4k\nceiXkGMEAHatb2diInBcANlnuctrP50cUQq+OtTl4nIAURwX3FbNPnZ8Ey9C9HXGBA4hmQW4mgHk\nGomXvcDEC7IU6BIfuYZZA6Mi4TbrN8f0DkCcIHvWJNtBSEJbSCti6lv68l7KctAMX1XTw8jzFIrA\ndrhSpfvPXUFDFvqWpwoloo1iAIMva9+pV8adhVMSjh+2mOX7NkISqoWJvFHST++VjJghE3+YT9FB\npJpCp2roY9zc36dcUoaeE1+R2tdh/zBg3KVacwQAcjYBAJkkHXlXomzkOtEHIVwLPUptBoCr/Hyw\nBiCZ8m9O2n1ktSK3GD4DppnKBADYR2WfnmhckZgaAHDpH/b53z9H/COdSW8pScEB4Jzc1p+PJjXJ\nYIIgJrfqZe3PkznPaiS313/CrzfTKTlTrxbXR50RSLrAPhxdzHuWI73xpl+vH1EAYKbZ31PRHtqm\n27notoYs6j4HYJ+q1Xe6N3AgG0dnlDGAzXWf1xt7JzQoerksdj9OHJdSusB/Tnt331tBJuWLkjVZ\n4LkQAPO9/zWiqtZZDGBeew+iVR5Ll2cLmuoyQDC3UbYvI/X7FYDC95VOa9HhPQBAtgq9v25j/Jf9\nZNFzentFF88/QEgk0VeykpUskn8BGUpBz07uiK0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABdElEQVR4nO2VTU/CQBCG320LS4tW\nUBTUeBANGL8OevKg/95480SMURFjkCoF+bRA2+1689RugSae9rnOzNPZnZ0UkEgkEsm/QLQ5khRB\nTK+kkwnyRTWZwFB4MgEJ4uvFAiQ8glK6Lq/FTUIUd9SjPefT6jI+nkUmEYGgcNUclgqUe4OavVQH\nrtN7q6epyt1pdJJQAEMJXFeQAfElclcXheMFbJxJJghYPpNMMPOTdYAJi62PnALVVN9jqtFfTqAU\nq5S6jrmybsUtZLhg8/S7rqVXdzbO2Wv0KxYIcsHTiEBp3WQvi42m8CZCBWp2OgMH6/9YdrXyUBsI\nzhHeAZkyAOAT+9E6O9i+bS0oIBmFAAAJRkHvrn0oWtnwDrwUAQCdeoD3/CFYxvCHxIbZFAAUyBgA\nG3mLCnhb2wKgmh3RtwUCDDtlAzBzX/6SAr9Bj6l+4rZj66N2oftykffM+0m8IGpC2v6u827PsY3R\nZm2Ov4pEIvnjF12YeToDwJReAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABn0lEQVR4nO2VS0/CQBSF7/RNIaEQ\nCMESTEDjRk0wRn+8C1euXZFoTAhawrsgQl902nEhuimdEWbbs+zc+XLm3jNTgEyZMjEl1jVmjUBb\nzN+bfAAkczoIAp26zgQQgggXAEBCfADC3E8HxNuQD0B+9yM5l+ZFogICEQGSZMW4kB+dIwBaITak\nSkuWFLWLj3HQPBXPBY0sBvOxndYNKsAnCgwn9nwVpNdQARPbena9iFZCB2BnPGNFkR4kzBkkQDJn\nlBWFzwESw5gLIGk+nwMI6CNkAgjzACwA2wCjiZHL6eAfQUpGWS4I2zD6uXxx2iWmAPTbNnZCPFu4\njito7C4mAPXc00YQSvlGUZyu8+rhAOX9jQBYSKmYRrMxG2wPBcQ5RAAI8Qcj3axcii8z+iwTU7DL\njd23aG1ZFr6+q1NvVMLBcnyV7+2eMDydDEvtm2V/nn6QBCDqFzrt3sgNAFBB//LHn8ZJp/eaOo49\nSVFbZ+VobXug1qIHDwCkarBMfdn2RQ2pRtWsadhbdT+YQUjJKpKKeow3DvPvnilTpkx/+gZLR5SZ\nJ7/bhwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACGUlEQVR4nO2VS3ObMBSFjwTm6YCx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\nQQCAN+OLyef/xFhQAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACYElEQVR4nO2Vy48SQRDGv+l58Bhe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\nSPfHB0/gbyJhnvdIkSL9I/0BATrPeEp+/iwAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABTUlEQVR4nO2SW0sCURDHZ4+7q7mi\naxLegqAoKaGSXvoG0afuE1RQBAlpBpmX1lX3pns9vdRDcC5rj3F+r8P8mJn/AAgEAoHgB4layVST\nzxgAQMIsgUytKB3txpDkUjnz5v1JEE6PD/VKTlPMd9YE9BVAv9aWQ8P0wpAloE8AzsfBQy9iNQMA\nIHopGvkBt58lAMvJMjZMIfBjYCbIFaRo5wgCly8gp6CW5GLgbCn8E5AFqHVqSWomypVkbgxEgVrz\n7tx8/aR4kX8xOYcgr4DiubcYo/Nob6c3WG0uiGyt4EHiju5x+6x+u2QJiCnEBt4GkDV7Nno0ss0K\n65bECfBaKyPcaDz7eBI2a7v9Af2W5BtYi6Np0p4PMeCZE7Q60E82WgHsLrq6XD25AAD+a3ddVagT\nUNZT9wvjufedoKKvnTRf/Vuc4gkFAoFA8P/4AlfCcrT+B0KEAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABn0lEQVR4nO2VS0/CQBSF7/RNIaEQ\nCMESTEDjRk0wRn+8C1euXZFoTAhawrsgQl902nEhuimdEWbbs+zc+XLm3jNTgEyZMjEl1jVmjUBb\nzN+bfAAkczoIAp26zgQQgggXAEBCfADC3E8HxNuQD0B+9yM5l+ZFogICEQGSZMW4kB+dIwBaITak\nSkuWFLWLj3HQPBXPBY0sBvOxndYNKsAnCgwn9nwVpNdQARPbena9iFZCB2BnPGNFkR4kzBkkQDJn\nlBWFzwESw5gLIGk+nwMI6CNkAgjzACwA2wCjiZHL6eAfQUpGWS4I2zD6uXxx2iWmAPTbNnZCPFu4\njito7C4mAPXc00YQSvlGUZyu8+rhAOX9jQBYSKmYRrMxG2wPBcQ5RAAI8Qcj3axcii8z+iwTU7DL\njd23aG1ZFr6+q1NvVMLBcnyV7+2eMDydDEvtm2V/nn6QBCDqFzrt3sgNAFBB//LHn8ZJp/eaOo49\nSVFbZ+VobXug1qIHDwCkarBMfdn2RQ2pRtWsadhbdT+YQUjJKpKKeow3DvPvnilTpkx/+gZLR5SZ\nJ7/bhwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABOUlEQVR4nO2Uz0+DMBSAW6CMssDQ\nKcFFYpa4gwczNTH6//8BJi6LBwkxmUGZW2QbbGMg4MmYrG/auYuHfsf34+P1tQEhgUAgEAgE/wvM\nUaKpxEyTefEXAdHM1p6hYzXPRt5LvqWAHLg2JeUinMUryWrXX/sxr0DSkXbYUBw9moTvSVYhhBA5\nOpv0lkypAgrUW7tGwqC3HH935M9Vx2IF8ATYcrJVMl87s3Y99JlNwhNUUQREi5RKjEACBTASrSo2\nuIXANKpyFwFuKtOdBFY7HrNRfkH9nPjsLfIL6MXxU8DukFtAuy3f/wASnAJ6dTq4X0AZ+CGtY9w4\nXj8FUxz/A4T3LxuPHtzPI5DdbnH3Bp2fS4DNjjt8mAL75xPQExsPgmxzwS9LlJs1f5Rv/DzHBDIq\nf2oXCARffALAU186l30FkgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABN0lEQVR4nO2Ty1KDMBSGQyDcUsRx\nLKMsXDm+/xN1HMa6QBjoDQohiYtuLObU42XhIt8OEr6cc/5AiMVisVgslv+Fc3ExZJ4WQkr9E4Eb\nJWniSncrdbsdQAUgYOHdInDedt1ElRckrGy/JQiWSxk29XGQp2c/i156vMDLcn9fdOJD1f5t+Ho0\nC7x57Wnq+9VmP501LTruAhWcC9hjvqif28N8ZHpUvFcIgRrFetUYJj7Rq3ZECOSqUNK0TTvMwwiI\nFsZdxCGj+XtCza/nRAy6SfMUANTOPEJsBTTWG+NssAIWTx3QA0pAOel+1QK/HqECUAL3JmiAeFEC\nmoT9AK4iYozSQwVEgBE4SSZK4BZiBC7PVQ0O4EuBEz94svr0d+MF/CkpCzDB0xkXBf79VMEBYKCX\nD7BY/op390pwln34aiUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABu0lEQVR4nO2VUU/CMBDHr2s3NgEh\nwwhq4oskmvj9v41PiIlGQXDMsW5dfYC6QbsewosPu6c1bX/7392/LUATTTTRBBaOe9iy2pnwsXca\nIAke2gcASP1U7yqdiENE1EYwHlj426hPASBZ9Dsogdkm52koY3kCoIhbF7CyAxCJpINJsNUAAGTG\nqH0FAmCUI520A1pDb40U0Q6QDAr7fgRAGbYfAfD4GAA5a6vmsoBiVjQA2PjBUyyRHaNAZsqfTpIg\nTTAqkEJlTrYJkKDWTgYAlVKZR32493etPyjgkVIg+EZC/jm8qTl2OsDp+FwBMrYpZzGddG/Nl6wO\noF3IfivnbYsgXt5GI6NnDAry6FsBnEClnr9/XPdNntABAaS8HHlqgXiOxpcGggZwQ1iW/qVl+5Kn\n2SDUCTqg+FqX+0XFSHy6CvVW6ID1rCKAuJV/pq9LHOD5aV6O9hTzudRy2EeKFa+436PZznnOhXY0\nNEBcHVHBd6f1o2W9UByZpLZ5k4Ld//EUfV3N9tza3iE5dh2YAdJjAADEd9D95hRk6ucFcc/ZAr3R\n6t5G13ddypYReinXP66EAPqoNNHE/4kf4KWHM2FpldsAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACMklEQVR4nO2W2XPTMBCHJSuW44ME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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAC/klEQVR4nO2W22/TMBTGP7u59Zo2\nXYF2bAzEBhsaMJCGkBD/vwDB2MTYDdaxeyltl2RtEic2DxMPgO2Oicedx/h8v3NyfHxs4Nqu7dou\nacQ0dKtj9bnbD47XM+UyHau/87I+4FfPIDezlL49EmoH3e8BoK3nZ++7Gv04QP1p+KavD6FddZ9g\ndaAPoc2gOBuzqaRz5QycmeTd9+lJfZKaVTrBt0bdwLWuCKBeuhMhiPIWAFCVoxJAiqLDgIzlPYC6\nr5ZtuZ+yiLk4AAA+MrzzUqta6JuxPJAKAACE2pWFe6w7jPzDkVyvzoDQcnniRqloGZ8/nTN1ECmA\nEKfSaNZYPGJhnh8MdFlKALTZcl3W762FUfn1ZHc00ullAMGT/m7PzwA66WwJ29QCdEUkjafx1qPG\nqR8ZhTRNw7ArmQu6s1BcyK9lFom6PnNrXjk76f0bwFmsr522LLbRA4KDnG1EqcRLDaB3bn5r8zQd\nXhQxGyrcVHpy+2H/cwLOZWEvA/AWzldCIOGpbqCpAc59se4DiCIrfxWAPet8PBYARFapagHyItLp\n5mZHAAA7GPpagLyR7Ln4SwYAxBpXAzmAWFkKAObczY2jXx9zQnZByX9BXBx+0prrDy6+WIWG19mX\nbKl2rDfm4/UhQEolb6JSOPFlV6wGQLxn9koXttds8sRvj4JIVg01gNSWq5vhVL0+lWx/DROuqKXy\nOJOJ5el4QHIRKjsrkTpPeQamXao99n4Ew1Pfme23Nfq/AcS07Fu3Ck6Bv2uf8YzOWzvfNfrfAcR2\nXLfkMkoMJ/iwnwKoFA921e8b/FGDxkK+w3wWxa2lYPWYA3AWyx9/6PS/ZzDcZT0mhHF3nrTPLC64\nOdfaGwAg6naW7UL1RSPqpWlqpRUvd3KaMKO8r3zmyHYhXPWCJMfNmXvGYT9mwjPrdKC6nOR9QIUA\n8s+nttdjIQgl+VQxEXWNVJ29sbet64ALU7Zybamw+XXMQAXUM7G46G1cRq8EkPPVvcvolUaMsa/o\na/tv9hNVwTA1rMxgXQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAADM0lEQVR4nO2W228bRRTGv5nZi9fr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\nA6emTqn7nQpqAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAAByUlEQVR4nO2Vy0/iUBSHf215pK3V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\ngg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAACbUlEQVR4nO2V21PaQBTGT7Ih3IYI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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABCUlEQVR4nO2TQWvCMBTHX9patWVO\n6+Y2WAcORTz5/T/DLjtYdhkiSEM3RyfDrtZo4kHYqM2j1fP7HZN/fnkv5AEQBEEQBEHkYWWBmmXI\nTF4oYI57P3RYGs8+kwsEptd/6Npit9rfQPQ+350paLbHQ8Xj32i9Ma7uHkXwpe9DK7AbXtd3zeVy\nHisFAADe6Hu21Qqs4lJr4Lk1JjlfpOqvHUcglWoEe8vMQp4mQv2n/NtEFZOIIHlhoPLx3gA48oiG\nZk1JmT9f7zvhB1KBTlDIPPvRa4ZtVhC0R2K6xjY1b3DK9SQLONJAlQrcMQtCfBhKK7Cftm8/6P3l\n08g61kr/BavCSgeeIAjiyAHSeFSBLzMumQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABOElEQVR4nO2XXUvDMBSGzzlpU9Mq\nWx2bZahj///XeCuC4GDMWVhb+23WeDlcM5a6C73Ic/m270ObhIQADAMBAQAPAQ3sIyIA4vk3LRaL\nxWKx/BVnDmrXURI6VN2vBHw0DXjTSUZ5nH0NFZAfjUJW1p/EWQey2aXtAAHnk7EgxevVrgUXFHkB\n894zQwFdhxPPc7JtUdXy4JzyTWUkwOhBtIms8lz+yJnwd5qBcPrRXpSbtGnVcdwunLWJQMVPsjhu\nAwB0DVF/OjVf0CWaNgAA3jj9mdAITuK5uiE37/ui2V8kEJTIfmouCOZOoYmNBexunF4ioNl9+ab5\nA1MB3i5plWvNZv1wcRV/aDcFk3XA3NmjG7/qdwQDAUZLn9Yvpf6pgUDlAp7X9Sn9eQHQfL/VrMEB\n2DuS5f/zDZhWYLz82W1NAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABJElEQVR4nO2UUWvCMBCAL22qtIpO\n1mYbMsocjOHYH/D//4GxBxnCEEHGdG1FS9rFttmTMOoF4/qwl3yPyeXLXcgdgMFgMBgMBoOhDjkV\nYLktAmWRl38R2J3u4Np3rDwt4/dYnimwOyxkcpcnQnwXvtd+W6FJKAWDB0Z49JmICkACcViwnWMG\nqjjvjMPpnItD2lIs4alYIFWoBMV0E21+x8uP3u2aq/LFIPXqvMkQCbOUAlnPN4u6SLRacGzc991G\nAqC0mcBxd1kjQdvL8iYCMnTTqomAjTjWDtqC/iOfCWRd9RNrkOAZXmNsR0/QCu/h5Qvt55MDBQBs\nf3SznCXnzoMD1sXdZbpYYfVrCZyroFpHAr9eQ0B7dLtXzUMdiKXzSoZ/5gfgNlxAJ2JkeQAAAABJ\nRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEAAAABACAAAAACPAi4CAAABN0lEQVR4nO2Ty1KDMBSGQyDcUsRx\nLKMsXDm+/xN1HMa6QBjoDQohiYtuLObU42XhIt8OEr6cc/5AiMVisVgslv+Fc3ExZJ4WQkr9E4Eb\nJWniSncrdbsdQAUgYOHdInDedt1ElRckrGy/JQiWSxk29XGQp2c/i156vMDLcn9fdOJD1f5t+Ho0\nC7x57Wnq+9VmP501LTruAhWcC9hjvqif28N8ZHpUvFcIgRrFetUYJj7Rq3ZECOSqUNK0TTvMwwiI\nFsZdxCGj+XtCza/nRAy6SfMUANTOPEJsBTTWG+NssAIWTx3QA0pAOel+1QK/HqECUAL3JmiAeFEC\nmoT9AK4iYozSQwVEgBE4SSZK4BZiBC7PVQ0O4EuBEz94svr0d+MF/CkpCzDB0xkXBf79VMEBYKCX\nD7BY/op390pwln34aiUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(len(errors))\n",
    "print(len(correct))\n",
    "for e in errors:\n",
    "    display_image((x_eval[e] * 255).reshape(IMG_ROWS, IMG_COLS))\n",
    "    display_image(data[0][predicted[e]])"
   ]
  },
  {
   "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}