Face -> Emotion recognition

face_emotion_keras2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Emotion Recognition from Face Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "import csv\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import numpy as np\n",
    "\n",
    "# size of face image - see below\n",
    "IMG_SIZE = 48\n",
    "\n",
    "# utilities\n",
    "from cStringIO import StringIO\n",
    "from IPython.display import clear_output, Image, display\n",
    "from scipy.misc import imresize\n",
    "import PIL.Image\n",
    "import random as rnd\n",
    "\n",
    "from IPython.display import clear_output\n",
    "from tqdm import tqdm\n",
    "\n",
    "# utility function to show image\n",
    "def show_img_array(a, fmt='jpeg'):\n",
    "    a = np.uint8(np.clip(a, 0, 255))\n",
    "    f = StringIO()\n",
    "    PIL.Image.fromarray(a).save(f, fmt)\n",
    "    display(Image(data=f.getvalue()))  \n",
    "    \n",
    "    \n",
    "EMOTIONS = ['Angry','Disgust', 'Fear', 'Happy', 'Sad', 'Surprise', 'Neutral']\n",
    "NB_EMOTION_CATEGORY = len(EMOTIONS)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### preprocess data set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "0it [00:00, ?it/s]/home/nao/anaconda2/envs/keras2/lib/python2.7/site-packages/ipykernel/__main__.py:42: DeprecationWarning: `imsave` is deprecated!\n",
      "`imsave` is deprecated in SciPy 1.0.0, and will be removed in 1.2.0.\n",
      "Use ``imageio.imwrite`` instead.\n",
      "/home/nao/anaconda2/envs/keras2/lib/python2.7/site-packages/ipykernel/__main__.py:40: DeprecationWarning: `imsave` is deprecated!\n",
      "`imsave` is deprecated in SciPy 1.0.0, and will be removed in 1.2.0.\n",
      "Use ``imageio.imwrite`` instead.\n",
      "35887it [00:23, 1520.34it/s]\n"
     ]
    }
   ],
   "source": [
    "# dataset from:\n",
    "# https://www.kaggle.com/c/challenges-in-representation-learning-facial-expression-recognition-challenge    \n",
    "import os\n",
    "import random\n",
    "from tqdm import tqdm\n",
    "from scipy.misc import imsave\n",
    "\n",
    "try:\n",
    "#    os.mkdir(\"./images/train\")\n",
    "    os.mkdir(\"./images/validation\")\n",
    "    \n",
    "    for i in range(len(EMOTIONS)):\n",
    "        os.mkdir(\"./images/train/%d\" % i)\n",
    "        os.mkdir(\"./images/validation/%d\" % i)\n",
    "except:\n",
    "    None\n",
    "\n",
    "val_split = 0.10\n",
    "indices = {}\n",
    "    \n",
    "with open('fer2013/fer2013.csv', 'r') as f:\n",
    "    reader = csv.reader(f)\n",
    "    header = next(reader)  # ヘッダーを読み飛ばしたい時\n",
    "\n",
    "    for index, row in tqdm(enumerate(reader)):\n",
    "        emotion = int(row[0])\n",
    "        pixels = row[1].split()\n",
    "\n",
    "        face = np.array(pixels, dtype=float).reshape(IMG_SIZE,IMG_SIZE)\n",
    "        face = np.stack((face,)*3, axis=-1)\n",
    "\n",
    "        if emotion not in indices:\n",
    "            indices[emotion] = 0\n",
    "            img_id = 0\n",
    "        else:\n",
    "            img_id = indices[emotion]\n",
    "            indices[emotion] = img_id + 1\n",
    "        \n",
    "        if random.random() < val_split:\n",
    "            imsave(\"./images/validation/%d/%06d.jpg\" % (emotion, img_id), face)\n",
    "        else:\n",
    "            imsave(\"./images/train/%d/%06d.jpg\" % (emotion, img_id), face)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### load from preprocessed numpy npz"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 32419 images belonging to 7 classes.\n",
      "Found 3461 images belonging to 7 classes.\n"
     ]
    }
   ],
   "source": [
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras.preprocessing.image import ImageDataGenerator, array_to_img, img_to_array, load_img\n",
    "\n",
    "batch_size = 16\n",
    "TARGET_SIZE = 224\n",
    "\n",
    "# this is the augmentation configuration we will use for training\n",
    "train_datagen = ImageDataGenerator(\n",
    "        rescale=1./255,\n",
    "        shear_range=0.2,\n",
    "        zoom_range=0.2,\n",
    "        width_shift_range=0.2,\n",
    "        height_shift_range=0.2,\n",
    "        horizontal_flip=True,\n",
    "        fill_mode='nearest')\n",
    "\n",
    "# this is the augmentation configuration we will use for testing:\n",
    "# only rescaling\n",
    "test_datagen = ImageDataGenerator(rescale=1./255)\n",
    "#test_datagen = ImageDataGenerator()\n",
    "\n",
    "\n",
    "# this is a generator that will read pictures found in\n",
    "# subfolers of 'data/train', and indefinitely generate\n",
    "# batches of augmented image data\n",
    "train_generator = train_datagen.flow_from_directory(\n",
    "        './images/train',  # this is the target directory\n",
    "        target_size=(TARGET_SIZE, TARGET_SIZE),  # all images will be resized to 150x150\n",
    "        batch_size=batch_size,\n",
    "        class_mode='categorical')  # since we use binary_crossentropy loss, we need binary labels\n",
    "\n",
    "# this is a similar generator, for validation data\n",
    "validation_generator = test_datagen.flow_from_directory(\n",
    "        './images/validation',\n",
    "        target_size=(TARGET_SIZE, TARGET_SIZE),\n",
    "        batch_size=batch_size,\n",
    "        class_mode='categorical')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fine tuning\n",
    "\n",
    "based on keras tutorial  \n",
    "https://keras.io/applications/#fine-tune-inceptionv3-on-a-new-set-of-classes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.applications.vgg16 import VGG16\n",
    "from keras.preprocessing import image\n",
    "from keras.applications.vgg16 import preprocess_input\n",
    "import numpy as np\n",
    "\n",
    "# we use VGG16 as a base model\n",
    "base_model = VGG16(weights='imagenet', include_top=False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensorboard --logdir=/tmp/tf_logs/20171110-164415/\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "from datetime import datetime\n",
    "now = datetime.now()\n",
    "logdir = \"/tmp/tf_logs/\" + now.strftime(\"%Y%m%d-%H%M%S\") + \"/\"\n",
    "cmd = \"tensorboard --logdir=\" + logdir\n",
    "print cmd\n",
    "#!{cmd}\n",
    "\n",
    "# setting callbacks\n",
    "from keras.callbacks import EarlyStopping, TensorBoard, ModelCheckpoint\n",
    "es = EarlyStopping(monitor='loss', patience=2, verbose=1)\n",
    "tb = TensorBoard(log_dir=logdir, histogram_freq=0, write_graph=True, write_images=False)\n",
    "tf.logging.set_verbosity(tf.logging.ERROR) # suppress warnings\n",
    "#cp = ModelCheckpoint(\"models/emoface-{epoch:02d}-{val_loss:.2f}.hdf5\", monitor='val_loss', verbose=1, period=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "3000/3000 [==============================] - 316s - loss: 1.7143 - acc: 0.3059   \n",
      "Epoch 2/50\n",
      "3000/3000 [==============================] - 315s - loss: 1.6453 - acc: 0.3469   \n",
      "Epoch 3/50\n",
      "3000/3000 [==============================] - 314s - loss: 1.6236 - acc: 0.3586   - ETA: 6s - loss:\n",
      "Epoch 4/50\n",
      "3000/3000 [==============================] - 316s - loss: 1.6068 - acc: 0.3689   \n",
      "Epoch 5/50\n",
      "3000/3000 [==============================] - 314s - loss: 1.5994 - acc: 0.3734   - ETA: 0s - loss: 1.5993 - acc: 0.373\n",
      "Epoch 6/50\n",
      "2291/3000 [=====================>........] - ETA: 74s - loss: 1.5911 - acc: 0.3774"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-9-c799bdf25527>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     17\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     18\u001b[0m \u001b[0;31m# train the model on the new data for a few epochs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 19\u001b[0;31m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit_generator\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrain_generator\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m3000\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mepochs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m50\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcallbacks\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtb\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m/home/nao/anaconda2/envs/keras2/lib/python2.7/site-packages/keras/legacy/interfaces.pyc\u001b[0m in \u001b[0;36mwrapper\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m     85\u001b[0m                 warnings.warn('Update your `' + object_name +\n\u001b[1;32m     86\u001b[0m                               '` call to the Keras 2 API: ' + signature, stacklevel=2)\n\u001b[0;32m---> 87\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     88\u001b[0m         \u001b[0mwrapper\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_original_function\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     89\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mwrapper\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/home/nao/anaconda2/envs/keras2/lib/python2.7/site-packages/keras/engine/training.pyc\u001b[0m in \u001b[0;36mfit_generator\u001b[0;34m(self, generator, steps_per_epoch, epochs, verbose, callbacks, validation_data, validation_steps, class_weight, max_queue_size, workers, use_multiprocessing, initial_epoch)\u001b[0m\n\u001b[1;32m   1807\u001b[0m                 \u001b[0mbatch_index\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1808\u001b[0m                 \u001b[0;32mwhile\u001b[0m \u001b[0msteps_done\u001b[0m \u001b[0;34m<\u001b[0m \u001b[0msteps_per_epoch\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1809\u001b[0;31m                     \u001b[0mgenerator_output\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnext\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutput_generator\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1810\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1811\u001b[0m                     \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mgenerator_output\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'__len__'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/home/nao/anaconda2/envs/keras2/lib/python2.7/site-packages/keras/utils/data_utils.pyc\u001b[0m in \u001b[0;36mget\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    634\u001b[0m                     \u001b[0;32myield\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    635\u001b[0m             \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 636\u001b[0;31m                 \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msleep\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwait_time\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "from keras.layers import GlobalAveragePooling2D, Dense\n",
    "from keras.models import Model\n",
    "\n",
    "x = base_model.output\n",
    "x = GlobalAveragePooling2D()(x)\n",
    "x = Dense(1024, activation='relu')(x)\n",
    "predictions = Dense(NB_EMOTION_CATEGORY, activation='softmax')(x)\n",
    "\n",
    "model = Model(inputs=base_model.input, outputs=predictions)\n",
    "\n",
    "# freeze base layers\n",
    "for layer in base_model.layers:\n",
    "    layer.trainable = False\n",
    "\n",
    "# compile the model (should be done *after* setting layers to non-trainable)\n",
    "model.compile(optimizer='rmsprop', loss='categorical_crossentropy', metrics=['accuracy'])\n",
    "\n",
    "# train the model on the new data for a few epochs\n",
    "model.fit_generator(train_generator, 3000, epochs=50, verbose=1, callbacks=[es, tb])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(0, 'input_1')\n",
      "(1, 'block1_conv1')\n",
      "(2, 'block1_conv2')\n",
      "(3, 'block1_pool')\n",
      "(4, 'block2_conv1')\n",
      "(5, 'block2_conv2')\n",
      "(6, 'block2_pool')\n",
      "(7, 'block3_conv1')\n",
      "(8, 'block3_conv2')\n",
      "(9, 'block3_conv3')\n",
      "(10, 'block3_pool')\n",
      "(11, 'block4_conv1')\n",
      "(12, 'block4_conv2')\n",
      "(13, 'block4_conv3')\n",
      "(14, 'block4_pool')\n",
      "(15, 'block5_conv1')\n",
      "(16, 'block5_conv2')\n",
      "(17, 'block5_conv3')\n",
      "(18, 'block5_pool')\n"
     ]
    }
   ],
   "source": [
    "# # let's visualize layer names and layer indices to see how many layers\n",
    "# # we should freeze:\n",
    "for i, layer in enumerate(base_model.layers):\n",
    "    print(i, layer.name)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from keras.optimizers import SGD\n",
    "\n",
    "# we chose to train the top 2 blocks, i.e. we will freeze\n",
    "# the first 10 layers and unfreeze the rest:\n",
    "for layer in model.layers[:11]:\n",
    "    layer.trainable = False\n",
    "for layer in model.layers[11:]:\n",
    "    layer.trainable = True\n",
    "\n",
    "# # we need to recompile the model for these modifications to take effect\n",
    "# # we use SGD with a low learning rate\n",
    "# from keras.optimizers import SGD\n",
    "model.compile(optimizer=SGD(lr=0.0001, momentum=0.9), loss='categorical_crossentropy', metrics=[\"accuracy\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 102/200\n",
      "3000/3000 [==============================] - 401s - loss: 0.0631 - acc: 0.9790   \n",
      "Epoch 103/200\n",
      "3000/3000 [==============================] - 406s - loss: 0.0587 - acc: 0.9813   \n",
      "Epoch 104/200\n",
      "3000/3000 [==============================] - 406s - loss: 0.0601 - acc: 0.9805   \n",
      "Epoch 105/200\n",
      "3000/3000 [==============================] - 406s - loss: 0.0622 - acc: 0.9808   \n",
      "Epoch 106/200\n",
      "3000/3000 [==============================] - 406s - loss: 0.0559 - acc: 0.9815   \n",
      "Epoch 107/200\n",
      "3000/3000 [==============================] - 406s - loss: 0.0593 - acc: 0.9798   \n",
      "Epoch 108/200\n",
      "3000/3000 [==============================] - 406s - loss: 0.0572 - acc: 0.9816   \n",
      "Epoch 109/200\n",
      "3000/3000 [==============================] - 406s - loss: 0.0586 - acc: 0.9808   \n",
      "Epoch 110/200\n",
      "3000/3000 [==============================] - 406s - loss: 0.0579 - acc: 0.9812   \n",
      "Epoch 111/200\n",
      "3000/3000 [==============================] - 406s - loss: 0.0508 - acc: 0.9839   \n",
      "Epoch 112/200\n",
      "3000/3000 [==============================] - 406s - loss: 0.0567 - acc: 0.9815   \n",
      "Epoch 113/200\n",
      "3000/3000 [==============================] - 406s - loss: 0.0502 - acc: 0.9837   \n",
      "Epoch 114/200\n",
      "3000/3000 [==============================] - 406s - loss: 0.0534 - acc: 0.9825   \n",
      "Epoch 115/200\n",
      "3000/3000 [==============================] - 406s - loss: 0.0542 - acc: 0.9825   \n",
      "Epoch 116/200\n",
      "3000/3000 [==============================] - 406s - loss: 0.0462 - acc: 0.9851   \n",
      "Epoch 117/200\n",
      "3000/3000 [==============================] - 405s - loss: 0.0467 - acc: 0.9852   \n",
      "Epoch 118/200\n",
      "3000/3000 [==============================] - 406s - loss: 0.0492 - acc: 0.9844   \n",
      "Epoch 119/200\n",
      "3000/3000 [==============================] - 405s - loss: 0.0520 - acc: 0.9830   \n",
      "Epoch 120/200\n",
      "3000/3000 [==============================] - 406s - loss: 0.0502 - acc: 0.9843   \n",
      "Epoch 121/200\n",
      "3000/3000 [==============================] - 405s - loss: 0.0523 - acc: 0.9831   \n",
      "Epoch 00120: early stopping\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x7f25babc1d90>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# train the model on the new data for a few epochs\n",
    "\n",
    "es = EarlyStopping(monitor='loss', patience=4, verbose=1)\n",
    "\n",
    "model.fit_generator(train_generator, 3000, epochs=200, verbose=1, initial_epoch=101, callbacks=[es, tb])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model.save(\"model_keras2.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/jpeg": "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAgGBgcGBQgHBwcJCQgKDBQNDAsLDBkSEw8UHRofHh0a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\nS1RoHYls7TXLanEsdyoBP3AefqaAP//Z\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "emotion: Fear\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/jpeg": "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAgGBgcGBQgHBwcJCQgKDBQNDAsLDBkSEw8UHRofHh0a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\nzFs9ODQB/9k=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "emotion: Happy\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/jpeg": "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAgGBgcGBQgHBwcJCQgKDBQNDAsLDBkSEw8UHRofHh0a\nHBwgJC4nICIsIxwcKDcpLDAxNDQ0Hyc5PTgyPC4zNDL/2wBDAQkJCQwLDBgNDRgyIRwhMjIyMjIy\nMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjL/wAARCADgAOADASIA\nAhEBAxEB/8QAHwAAAQUBAQEBAQEAAAAAAAAAAAECAwQFBgcICQoL/8QAtRAAAgEDAwIEAwUFBAQA\nAAF9AQIDAAQRBRIhMUEGE1FhByJxFDKBkaEII0KxwRVS0fAkM2JyggkKFhcYGRolJicoKSo0NTY3\nODk6Q0RFRkdISUpTVFVWV1hZWmNkZWZnaGlqc3R1dnd4eXqDhIWGh4iJipKTlJWWl5iZmqKjpKWm\np6ipqrKztLW2t7i5usLDxMXGx8jJytLT1NXW19jZ2uHi4+Tl5ufo6erx8vP09fb3+Pn6/8QAHwEA\nAwEBAQEBAQEBAQAAAAAAAAECAwQFBgcICQoL/8QAtREAAgECBAQDBAcFBAQAAQJ3AAECAxEEBSEx\nBhJBUQdhcRMiMoEIFEKRobHBCSMzUvAVYnLRChYkNOEl8RcYGRomJygpKjU2Nzg5OkNERUZHSElK\nU1RVVldYWVpjZGVmZ2hpanN0dXZ3eHl6goOEhYaHiImKkpOUlZaXmJmaoqOkpaanqKmqsrO0tba3\nuLm6wsPExcbHyMnK0tPU1dbX2Nna4uPk5ebn6Onq8vP09fb3+Pn6/9oADAMBAAIRAxEAPwD34UHk\nUHpSDOaAE2HIPFOAxS00sPWgBGYJyf0qCX5A9yfuIN5HfA//AFVPjd7iuI8WeIrayOp2P26SKdYC\nFRQ4wSmRyBjvQBU8U+LbCDU41aG5JMIPCr/eb3rzRbGXUz5MLIrL85LkgY6dvrVu1stS1uM3KB7k\nKfL3ySDIxzj5jnv+tdxpmjC1uWeWziRShGdqnnI9KAOa0rw7dw3EDNJAQq84Y/3fpXV2dpJBEVYq\nSWzwa0UhiRgfLQAeiipCqscqox9KAKqxn2qQRn2qwEH90U4J7CgCqYz7U0xn2q7tHoKaUH90UAQl\nDjtURjPPSrpUegppX2FAFFoz6iomhbjkVpeX/sikaIf3RQBjvC3qOtMMLbTyOlarxDH3B1qMxDaf\nlFAGJLE24cjpVSWFio5HWtyWH5h8g6VUeLj7goA4nUrSRo58FeW9feuXubd7PUoL2QqY7fbK4XqQ\npyce/Fel3VqGWT90pyfQetcvremySF9kClfJII496AOp+HnxA0o/2l/o97/yy/gX/b/2q9bt51vL\nOG5jBCTIsihuoBGefzr5Gvhd6R5fkO9r5uc+S+3djHXaff8AWvUfhv43iW7SDUdZunSOxC7JGkcK\nwKDpz70Ae15wwXvTiwU4NVLS9g1LTVubSTzEkDbHwQSQSO/PUVLarIkRE+S2eMnPFAE/WgCkXrTi\ncCgBrHtTME9jT85NKTigCjqV8NOtUmLxrlwuZDgdCf6V414pmn1XxjcbI/MineJN8SkggooODzXo\nfjy7xo0Q2dLkDr/stXI6VZfaLm0uvM25lU7duejev4UAafhrRo9O06SFvOQmYtiTg9APT2rbUlzj\n+VPmT5xz2pyReWc5z26UALJCggyCS2Bxmo40O3kHrSpNvn8vbjkjOashcd6AIdmPWnbfrUmKMUAR\n7frSbfrUuKAKAIivsaNvPepiKbigBgT60FPrUg6UE0AVZU+Xv1qLbx3q0w3cVEy4zzQBUkjy3fpV\nN48L0PWtFutVpV+Uc96AMidMh+vWsnUYV8iVsneIjgevBrelT73Pesq/hzuO7+D0+tAHmmvWr3P2\nfzI5F27sYGPSsa2v20K4eeDyy5BjIl6Yzntjniu41i38zyfmxjd2+lea+Ih5HmH7378j09aAPevA\nHj7zNG0m0nuNOjMkpRl34YAyn1b3r1dZ45xugkSVBwWRtwz6cV8Z+FtX8vVtJh8jP+lxjO/1ce1f\nVvg668zSJTsx+/I6/wCytAHSAih/uGjFDfdoARAcde9R3BI24JHWpBkIe1Yeuak1n5H+kJHu3feI\n5xj1+tAHF+I9Zi1OefToxN5kFwxYyY2/KSpxz7+lXdEtSNMtzhNwJ5/4Ea5fSyL7xbqfmHzFLSuM\nd/3g54+tdjbBoI0ijBVQeBjPegC6SYziQlj1z1qVQc8nNQ4Z+XBz9KfvY/dOT7UAOj2C4ztGcnnF\nWCcniqyAq4dgQO5PFS+YCRtYEe1AEhU0wnBpd5PemkUAOGSM0E4OKQEgYppyTQA9XEhwM+vNDfKC\nTTQCvKDmomaYkghsf7tAEofPrSO2MdaauQpJ4PvUbyA4yw/OgByyAsetI/OSKJAoiVo/vHrjmmru\nKjIP5UAQsCD1qJxkVZdDnoaqvuA7/lQBTlAAbjvWVfjOSOmz/GtWVlw25h19azbwbgxQbl2ckc0A\ncrqiZ8rGO/8ASuA16KMbzJGrDzj1GfWvSL+MHy8qe/8ASuE8RQp5bYXnz/X60Acrp22LxFYyIoVF\nuYmwox0YV9O+AtWiOhT5EpP2luoH91fevloySQ6xGUO3bIhHHTpXrfg7X76HSJVW7VQZycFV/ur7\nUAfRtFIRkU1hhTQASf6tvoa81+JV5Ja/2ZsVTu83O4f7leiyf6iQ+in+VeQfFpg39j4/6bf+yUAV\nfCHOszzfxPAxI7csprtklYSKMDqK5XwlZSSJCQyc2qnkn/Zru4pVsdNJlBPlKzNt9OTxQA1GLrk+\ntRr+7OR9Oa5XW/H+lWd6kclveEmMN8qL6n/arPi+JuisxAtb/p/zzT/4ugD0IDz0CNwGHOKPs6RA\nhSx781yth460y4niVILwFhkZRfTP96umsdWgv4WeJJAN235gOuB7+9ADulSDoKeBv6dqcIm9RQBA\netKvSpjCx7ik8ph3FADIfmcg+lNnkKIxGOD3qVFINDRMcnIoAqRuZuGwMnHFStYRH+J/zH+FR3eo\nRabbymZXbahkOwA8Y9z7VwfiDx7pafZ8295zu6Ivt/tUAd28YiGFzwcc1C9w6ZwF49a8PvvFlhdS\nybIrkZkLfMq+/vTIEOp7HhwomO1d/GDnHOKAPbDfSk/dT8jQWLjB/SvKrbw3eRRlWkgznPDH/Cuj\ntNThtZS7rIQVx8oH+NAG/dINsnXr/WqnSF17EGnpfRXirHGrguMjcB9aSSJk4JHSgDHvoVfy8k8Z\n/pXEeIbSPyict/rvX616HOp+WuZ1uBniIBH+tzz+NAHj+pHyNUkK87SpGfoK29D1+6gsnVY4SDIT\nyD6D3pNb0+Vr+5IZOg7n+6PaqFnYSiE/Mn3vU/4UAfbm4etIxG2k2/Whh8lAEcuPskp/2T/KvI/i\navm/2XwGx5vX/gFetzcWU5/2G/lXkfxEcn+zen/LX/2SgDY8LosNnasVC5tUGQPZas6lBfXMk4ty\n5jddqgSYB4x0zUXhsLJY2oc4H2ZP5CugWOJVyHyRyBkUAcQPCd1P81zpkMrjgNJsY4/E1l6v4B1S\nS0UWOjwrLvBJRokOMHvke1eizXskLhQEwRnkUjaw4HBhJ/z70AeCT+HfFGlXMs88U0MMLkErcqdv\nOBwG963/AA54gexCW95qFwsjzghS7tkHA7fQ16Lq2nWWp2M6zXDKZiGYI6jB3A8ZFcfceC7FLyKa\nCS7fy8MCGUjIOey0AdppGrw3HnYuXfbt67uOtbyXMZRTvPT0NcLp9vNZeZ5MUjb8Z3KT0z6fWuph\ncC2iJIDlRuHocUAbCSoQPm/SlZkz/wDWqhC7ELgZGeuKmd2B6UAEkqheGxzWTeXjrHJtnkBB4wx9\navTlQgwR19aybxIjFIVfL56Aj1oA5jxDLqFysiwzzMGgK480jJ5964j+wNWuv9ZbmXb03yqcfma9\nKe0Ex3OHC9CR0AqvOljY7cXSDf8A35F7f/roAqeF/CdvFKr6jo9myG3HMkUb5b5frz15rsYNI0SA\nIE0yxQKcjbbKMc/SsGPXTFGqxvbsAAAc54/OqNx4tvorwxiK28kEZcq3Axyc7sUAdz5WlD/l2t/+\n/I/wqhc2GmtGAllbZz2hA/pWBb+KbaSMmW7sUbOMeYBx+dXoNXjncoJoGIGcK2f60AULy0ljEpt4\nwhDfLsIXAz2/Cq8AuVXE7OWz3bPFbsipJGSGyW5wDWfPGVlAwcY70AQgA/eAP1qpf20bwjESE7s9\nB71pxwo+csRj3qreKQmFBOGoA808SWbC5vikSjEfGMD+AV5/MbyNwFkkUYzgP/8AXr1PxGjZv2Kn\nIiJ6f7FeXXs5WYD5fu96APubNDcrSEGlP3aAIZ1/0KceqN/KvKPHtrv/ALP+fGPM7f7terzMBbSj\n/YP8q858aBW+w5UH/WdR/u0AX/Den77G1Hm4/wBGT+H2HvWhdL9leQZ3bBn0zxmjSUJ0uzEWEbyE\n5HH8Iqtq1yttBdtLuYxxMxI5z8uaAM661PdKD5P8P97/AOtXNXfiX7JEJPsm/LbceZj+lcd4rl1f\nV9UiuNI1Ge0t1hCNGZ3jy25iThcjoR+Vdpb+DdR8PyG71ua1vbZx5axqzSYc8g4ZQOgPPvQBXg8V\n/apFi+xbd3fzc44z6Vr2up5t3Pk9z/F7fSuR1DxFo9hLcEWMimJyv7uJBjnHHNXdEv01+ze4sfMi\njEhiIk+U7sA54J9RQB2OmXfn+b8m3bjvn1q1HNvmZNuMZ5z71z0MNzYbt8x+fpsY9v8A9dbNmCAs\njHO5PxoA3rVsQJx/nNSTS5ccdvWsqOdt6oGYZOOtXQG7tn8aAIZm+Qcd6xZXxM/H8RqzqE7CBdrM\nDu7H2NZ9uryXC5bIOTyfagBt3qn2W0mj8ndmNjndjt9K5OaT+19vHleV/wACzn8vSuo1G1L3KrhN\npUAg9Dya1fDekWn+lebaWz/cxmMHHX2oAxrfwttsLeb7Z9+NTjyumRn1rJvtKxcy2nnfeAXdt9R6\nZ966Xx5qcenaHCtuJIWW5WPMXy8BW44PTivDtY8Qah/bksyaheiMFDt85uyj3oA6zUPDv2OdY/tW\n/K7s+Xjuff2pfCev/btUli+zbMQFs+Zn+Jfb3rovhtf2mreHbie/gN3Kt2yB7hBIwXYhxkk8cn86\n4690i50uET28kcDs2wtCxUkcnGQBxxQB6nDc7YkbZn5Rxmld/tAL424GMdapaJfwSaZYQyIzzfZ0\nDuwB3EKMnNajKrKSihRjpjFAFSPjNMnXK9e9WY1HOQKhvBiMY4+agDh/Ew+XUB/0xP8A6BXkGoRb\np1O7+H09zXrXiZiH1Dk48o/+gV5jO0e8bkycelAH265wKDylNl+6PrTh9wfSgCKZcwuMH7prg/F9\nvu+x/u2P3/X/AGa79z+7b6GuO8WH/jz/AOB/+y0AXtLj8vTbQhSP3CDn6CsfxGpktdRRAWlaBgqr\nySdnAArctZCNLteB/qk/9BqndWyM8lyS28DdjtwP/rUAeU2elzrCRd2c6SbuA6Mpx9Pzr0X4hXe7\nQIBp8qyzfalyIiHO3a/bnviq11GLuUSSEggY+Wsn7Q9x8jhQBzxQB45q5vJ57uPZI8jStlQnOd3P\nGK7P4brNb6d5NwjxFr3O112kghB3rcPg/T2u3vTNdea7FyNy4y3X+H3rW0zQrWBkZZJiVkDDJHt7\nUAX9Uhj/AHXlrnrnBz6VZt0/cRgKc7Bx+FTSWkcmMluPQ1NZxL5pTJwFoALeFTJGWU53DP51cuSI\n5AEOBjNQO5juNoxgEdalcecdzcEccUAZtxDG0YEi8Z7nFUQFjlymAQTjmrmoTMIF4H3v6GoUtUMa\nyZbcQD+dAFW4WWVt6ozEDghaksbu8tPM+Zo92PvIOcZ9RU4laGZIVAKsRknrzxVxrOO5++WG3pg0\nAc74qgOqaVEnltO/nB2VASfutzgfWuHk8H288++bSpyWI3HEg4/OvV1sYo2OGf05IprQKZdmTgkC\ngDgtMsp9BtmtdLt5YIHcyMuwvliAM5bJ6AVt6bpNlqdw0OpQb4VQuoZ2T5sgdQR2JrpH0yEH70n5\nj/CmW+mwrISGk6eo/wAKAKNpplvb3ypHAVgQlUOSQFAIHNaU0Mao2xeNp6HNWDEqJgE8cVGR+7Ye\nxoAzOV6cVHKN6gHnmrTxjjk1Vc7ScetAHEeKLcbdSYRnPkNzz/crxnUZZYrhVUkDZnp7mvd/EKh4\nL/PeFun+5Xjmq2ETXSks/wBwdx6mgD7ZA5pgU781JTe9ACMdoJPYZrlPFp+0fY9nG3fnP/Aa3tTn\nEMExLlcRE8Z9DXIvdLeYxIZNn97PGfr9KANwIV0ezzj7if8AoNViOtWpXH9l2yg9FXj/AIDVQEet\nAEbISe1MMbeoqyAPSl2j0FAFPym9RU0UZC9utTbR6Cl4FAEEvGKiSQBzwasSDOOKr7cOcgUASbgw\n4oUcVLCqlFyB+VTkxLwQo/CgDGu7KUxD5k+96mqbRmJctg7eDite4nj8sfN39KyrhwyuFOST/WgC\na3kDW7qAckkfpUttEy7skdqpWp2Lhjj5q07eSM7ufTtQBZjlU/Jg5AqXqKrBSp34wD0NToflHNAD\nXU5qMIfarBIzURZaAGEcVVnOAR/s1YZxg81Xm+ZWPXigDOlGcVTDBJWJ9T0q7KyrjNUpBuJx60AY\n2ryB0ukGctGQM/7tcvbaLc3EZdHiABxyT/hXVXsTGaTK5GP6Vb0j7LFaOsiRht5P3M9hQB7EaTFO\nooAx9cVjaXGFJ/cN0Hsa4/SonbztyMPu9vrXe38Pm2sw3YzEw6exrm7e2+z7vn3bsdsUAWZciyiG\nOgH8qrA9KtXLf6Mgx3H8qphvagCYHinZqINx0p4OTQA7NLjPNApwOARQBEc+lUJ52QnAH3sc1qrH\nuzzjFZ15bdTv/j9KAFguZCi4Vevp70lzdzJIBsXp3BqW0t/3Ufz9/T3qPUI9twozn5f6mgCu6+Yu\nBz34rLuZDF5hOAFOOfrWlp0/n3DLt24TOc57iqGswYtblt38Xp/tCgCtDe7pEQNGSzAYB5ratkk+\nb5G7dq4+xO3U7RfWZP8A0IV6HbR53c+nagCv5mVCHHHapUPyiqkv7qeQ9fmI/WpYpcqvH60ATu2D\n2qsZDjtUznJqmx4oAcz9elQSzEArxyKUt7VXl+aVT9KAK1yW+XA9e1VgzbjkYq9OudvPrVd4sDOe\nvtQBRuIt5cgEkjt9KzmkntjsWIkHn5lNbgTLhc9TimXNjukB8zHH92gD1cnZyaCflzQ4yv40HiP8\nKAGyDdBIPVSOfpXOamy2vlZGN2fu+2K6PrE30Nc34gVj9n+U/wAXb6UARSlntoyCcHB5PtVUsQcZ\nNTqxNvGpPQDj8KrSnEhoAlVjjqakR8nvVZXOOtSRt83XtQBbBwoJphmHmKvPNNZyI+tNjXewYjOD\n1oA0IHVd24ZzVW9kQpwv8Xp9afux0NVbnJT/AIFQAsUwEagZBrJ1iSX7WmJGH7sfxH1NatvGGMYK\nnk/1pNRsI3uFJhY/L7+poAxfD7k38mST+6PU+4rS1AxS20sfljcSOSo9aXSrGKG6ZhEVyhGST6im\n3oA87H97+tAHOS2hj1CCVQiqhViBweDmun0y5Nx5u1n+XHU/WsSZdzZAzx2q7pUhh87Lbc46/jQB\nbvWypC5Db+T+dV47jy9qsWJB7VbuE3RhgpJJzkVnyIVkPykYoA0UmEgyM+nNRSgqoJ9arwylUI3A\nc0+aXcgG4HmgCMvkkDNNblgO5pgb5+tDMdwOelADLlHG35vXvRt3Io74pJ3Lbec1YhiBVSVPIoAz\nZJBFc4OeCDxVjzVn+ZQQBxzT5rVGuCTGTkj1qeG1jRCPLI59TQB6NTX+4adSN92gBqDKEVl6xbo/\nk5Lcbv6VqrwKgu1LbMe9AHMugTgZ4OKqS8yGtG9Ugt/v/wCNZr8OaABRxToz834VGWFJCwLn6UAX\nioMf4VLAoWJsetOiUlE+gqbaQjUAQHmlliUxKcnnFLU0rDyE/D+VAFWH5JUA7MP51buHJkHTpVMc\nzr9RU0ynePpQBUtp28w8DpVa+GYZW7k5/Wmaawa5YD+4f5ir+oKRp8h+n8xQBj20KyJ8xPXHFPnh\nWDbtJO7rmp9OcAqOeZB/Srmpjd5WPf8ApQA6BBJbxg5+4Dx9KrXNugMhy3T+lTWUyh9uDkLVib50\ncjuDQBzr/IcCkDk+lS3qlZgD/dqlG43d+lAFn3oz8ppuflpvWgAA3da0M+XbxkegHP0qlEMZrWFs\n81tGFKjgHn6UAZpuH+0gYXqKsPMwPQVINKna5Uh4/vDuf8K0xo9wR9+L8z/hQB1WaSijFABimuu7\nHGafSZoAxr6AnJ2D7/t71h3SFZH4xgf0rrLiKNly77Rn1ArAv4oPOlAlzx/eHpQBiFjnqasWyEyH\njtUEqhXAXkYq7GBG2Qe2OaANONcRpx2FJI4GRntSxvmNckdBVW6l23CICpBA/nQA/OelSy58hfw/\nlUAOKmzvQKe3pQBXX/XL9RU1xIokHPamFAHyM8c1Wu5T5o6fdoAxdHuAbt/3h/1Z9fUV0F1mSwIH\nOQP5is230iCzkMkbSkkbfmI/w9qvLIXxEcbenvxQBShjdJU4x8wPBq7cBn298Z60GFVO7J45o37u\npFAGbHIY7mTLEdR1961IH326nJOc9aypk2yu3PLGrNrOQI4xtxnHv1oAi1FD9oXj+H+prGizuP0r\noryIPMC2Qdtc8isrZKkcdxQBdVSUHHakYBQcjHFSJny1z0wKhu32RSFcFghIHqaAK810kW3MhXPp\nmuq0CWO4KLnf+5BwR9PWuMsreTUt/wBojdPLxt2DGc/X6V1fh1RbXRU/Kqw7QW47igDrI7eLAbyU\n+u0VL+6Xjao/CmwvvhBGCD3FO8sNyc0ASUUUUABppNKaaaAKGrS+XaqdufnA6+xrlrm53XTDZjOB\n19q6fVU32yjj746/Q1yt1HtvG6cEfyFACMm45zirO2q/mqOxrQCAnGBQBLu224PXAFUJn3XcfGOn\n86v3A22h9gKoRxmSZG44YdfrQBbIp6t2xT5E246VXDfOetAE/UZrMvzidf8Ad/qa1ohujHvWfqMR\n+0L0+4P5mgCeY7UB96qwvm5Ax3NTzHcgA9apQki8HPc0AX5mxG5x0U1Rjnzn5f1q3PzbyH/ZP8qy\nYt3PNAFm5XMYbPU1Bbvi6iXH8a/zrTMHmW0fC9AefpWXMhhvc8YUg8fhQBqXYzKP92si8t/JhDbs\n5bHSti1YXMZfrg4+aqmrqPsiYAHzj+RoAq9LRT/sis+aXddRxYxvwM56ZNSyT7bfbluABVJcy6jb\nuDwHUEH60AdBptn5fm/vM5x2+tWfM+zOWxu7elJCwXdjjPpVOR2mldFYghieTQB2ukyebpUT4xnd\nxn3NXV6VnaECujQKxyfm/wDQjWh0oAdRSZpaAENNNPppoAoanxbqenzj+Rrlb7/WysPTOfwrqdY4\ntE/66D+RrnLmMNFKxznaf5UAYcly6tjeBx7V0sZ+Y59K5G6UCUf7tdeqgGgAuGzAy5Hbj8aZZxg4\nJHRqjdiWZe2asWvCf8CoAmucDb+NUh/rGqzdMfk/Gq6D5qALcJwijNVr4Azr/u/1NSKxGKr3bEyj\n/doAjRmkOM578VWGUuCemCavWEStOwJP3f6io7iFVeQgn7x/nQAhctbyc54P8qz1BXPGK0IUBhbr\n3/lUEkSjHJoAv2rB0VSQcIOKo6hCPOmYIeBkH8KnsjtlOP7tTXCB0kJzyp/lQBT0ubZbMHcKd54P\nHYUagDJAoIz82f0NUJpDA4RMEEZ5rVlUFR9aAOY1EtBaSv8AcCkcnoORUGkSpOyOzqzCUAEH6Vc8\nWIIfDV5IuSw2df8AfWuX8M3chaJcLg3A7fSgD0Ve9M0+IS6hMGUkYJ/UUI556VY0zi9kPqp/mKAO\no09BHaRIowBnj8TVtutV7Pm2jP1/nVhutAH/2Q==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "emotion: Surprise\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/jpeg": "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAgGBgcGBQgHBwcJCQgKDBQNDAsLDBkSEw8UHRofHh0a\nHBwgJC4nICIsIxwcKDcpLDAxNDQ0Hyc5PTgyPC4zNDL/2wBDAQkJCQwLDBgNDRgyIRwhMjIyMjIy\nMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjL/wAARCADgAOADASIA\nAhEBAxEB/8QAHwAAAQUBAQEBAQEAAAAAAAAAAAECAwQFBgcICQoL/8QAtRAAAgEDAwIEAwUFBAQA\nAAF9AQIDAAQRBRIhMUEGE1FhByJxFDKBkaEII0KxwRVS0fAkM2JyggkKFhcYGRolJicoKSo0NTY3\nODk6Q0RFRkdISUpTVFVWV1hZWmNkZWZnaGlqc3R1dnd4eXqDhIWGh4iJipKTlJWWl5iZmqKjpKWm\np6ipqrKztLW2t7i5usLDxMXGx8jJytLT1NXW19jZ2uHi4+Tl5ufo6erx8vP09fb3+Pn6/8QAHwEA\nAwEBAQEBAQEBAQAAAAAAAAECAwQFBgcICQoL/8QAtREAAgECBAQDBAcFBAQAAQJ3AAECAxEEBSEx\nBhJBUQdhcRMiMoEIFEKRobHBCSMzUvAVYnLRChYkNOEl8RcYGRomJygpKjU2Nzg5OkNERUZHSElK\nU1RVVldYWVpjZGVmZ2hpanN0dXZ3eHl6goOEhYaHiImKkpOUlZaXmJmaoqOkpaanqKmqsrO0tba3\nuLm6wsPExcbHyMnK0tPU1dbX2Nna4uPk5ebn6Onq8vP09fb3+Pn6/9oADAMBAAIRAxEAPwCr438c\naZquiwwQQXastwrkuigY2sOzH1rzn7QjSmQBsMSaq3EzTxhRIzYOcEmltoZWlQEEjHTPtQBo28qv\nKgAPLAVozQsdvIqraRLFhpEUYbOcZ4q+1xDJjawOPY0AV4Z1WQoQcqMGql5+8lkx3Hf6VI4McjSH\nhWJwalgVJHjyobLAcjrzQBStrZ/LPK9a2xMsZyQfTitOzsEaEkW8Z+b+6KZd2DiIYhUfN2xQBiT3\nkZDjDdfT3qhcXkaowIb7p7VrjTpmlI8hTknuKjn0S5kPyWqkYx1X/GgDmvtkf91vyq5FexFVG1+n\noK0k8M3zZxYof+BJ/jV6Hwtf4H+gJ0/vJ/jQBjJdIcHDflUpvY14Kv8AkK2z4bvEiJNigwCeqf41\nh3+nXMU6qYQvy5wCPU0AVZL+Lb91+voKj+2Rk/db8qy50mCDr19agUy7vvN/31QB0ENwhdeG+8K0\nVcPnGePWuQ82WNgTI4xz941Yh1PZu3XMgz7tQBr3iFAXOMFu1ZFwcs/0/pT5NQSdQizsxBzg5qrI\nxIbk9KAK7DmomjJHUU2VyGHzHp60wSHPLGgCRYyCOlKPku4pD0VgTj61GZOPvGml89zmgDqLHWba\nLzNyS846Ae/vWnpbi6vHZMgMhYbvTIrjbKOWXfsBOMZ5+tejaLYtBbQSvCq7oF+bjJ4FAGrZWziO\nJsrgHP61otIEOCDVeKREjC5wR2qdAJFyBu5xk0AedRaXtcmRJlGOpGP6VditI42Ugtkepq9qk/kW\nytt3ZcDGcdjWet3wG2dvWgCw6gKRntVckx/dGc+tQT6ltmVPKzkdd3/1qsWg+17/AODbj3zmgBls\n1xfXDwNCdigkFFOeDj+tbVlpb7of3U33h/D7/Srlho32JVu/tG/zUA27MYzg9c+1dFp8G9YTuxlv\nT3oAWw08iBspKPm9PYe1WJ9IDoBtm69h/wDWrbgg2IRuzz6VaVNxxmgDlI9BUSA7bj8v/rVch0BG\n/huOvp/9aujEHAO79Kv2VpviLb8Yb0+lAHO2/hyL5si57en+FaCeH4FUcz9Pb/CuhitMZ+f9KlaD\nCj5v0oA5S50KLyJceefkPp6fSuG1nQv9MTEdwR5Y7e59q9fkg3RsN3UEdKwb/SvMnU+dj5cfd9z7\n0AeB3GiuYx+5uOv93/61Qx+H3Zx+5uuf9n/61eqSeHcr/wAfXf8A55//AF6dH4e2kH7VnH/TP/69\nAHlM/httrHybv7v93/61ZU+hyx7cQXPPqh/wr2qfRcAj7R/D/c/+vWPeaDnZ/pPr/B/9egDx8W0l\ntKxaORRkrllIqTIYcnrXXa5oW2Mn7T/y1/ufX3rmLiy8h3Hmbtoz93GeKAKUlurtnLdO1VzBgdGq\n4r4HSmTNhBx3oApFQM80gUetDnLH60gOBigDT0txH5vIGcdfxr0mxu0bTrZQ8ZIiTgH2FeWWhxv/\nAArtdFm8wIm3GIRzn6UAdUh3qCOc+lW4JHRCAvfuKoWrYSNcd/61qRR7lJz3oAi8R29pd6fHHb20\nUbiUEkxgcYPpXnN9HMrTRpJtKuQMEgDBrstR1NRbr/pUX3x/Evoa5J3We9kBYPudjwetAEel6dcT\nTRTyyI6JKNwYkkgYJHSuqL2lv923Vd391AKoaeiRWz8bfmJ5PsKdNKr7fnU496ANawmluLhkEj+W\nFJVWY4AyMcV22i2pNvbMQh+fv/vVxuhxq0+QM5i/qK9A0dALa2GP4v8A2agDaS3XH3E/KrSaeUbL\nCMjH+e1PiiBU/KetaIQHtQBVjtkdlTy0z7ir0NqIkK7U654FEMQEobaR71ZI5oAYkajOVX8qV4gy\n4AWnCnigCq1uefu1VltAzA7U6dxWk3eq7daAMc2EWP8AUw/98j/ContYQpAhjz/uitJhxVZ168d6\nAMmezRjkRx4x6VSlsI2xmKI/Vf8A61bciD07VXeMcfLQByWoaFHIpJhtiC+eU+vtXEa74d/fXeyO\n2UbOMLjHy/SvWLqJTGPlP3q5rVLJHe4zETlPf+7QB4headLbTBGMeSuflJ/wqg67xgfrXf6zpSG8\nTFs/+rH971NcVPbtEgYRMvOMkGgDJZcOcgdaNoKEgDihsmZgfU0E7TtHQ0ATWNrLc+Z5bqu3Gckj\n1rp/Dl2kF+8coZikRU45GQRXO2Mwt/Mw6pux1I5610GmWrCX7QIn/eJnfg4OcGgDuLYrNCkka4U9\nMj3qx5jR8bm9eDVHS3228EbMB82CD9a2DbI/IQn6ZoA81v5maBQQPvf0NGmwq13ESTyCf0NSNGZB\ngY/GrlspjZGPQDt9KALrRhIHAz0PX6VTjjD5znippZ15GD0otXB39e1AHTaEgR1Iz/qf8K7nSWO2\n2H+3/wCzVxWjjLL/ANcv8K7rRYWMdocj7/8A7NQB1tugMZ69auRjLH6UyBCsZBx1qeNSG/CgB5G1\nMikDE06TmMimIMCgB4pc02igBHPWq7Mc1K56/SqrnmgBrDioHHWpSeKjbnNAFaT+lV3PSrrISD06\nVA8TccigDPmOR+NYl/8A66Ue39K6KeJgvUdax7uBmlkII5H9KAOP1CFXuFJJ+72+prh9TsIhbL8z\n/fHceh9q9Nvbd1mAJX7tcfqaH7MvT74/kaAPLbqMR3E2CeHI5+tVicuK1dTQia56f6w/+hVlbTnP\npQAP2rv9MmYabaDA4hT+Qrh4YzcbtmBt65rptG0yaSQYaP8A1WeSfb2oA6qz+aSEnu4/nXUW52xk\nD1rK0i2e3s7dXKkqxJx/vE1rTMJHBHp3oA8/tvLaQjCnj0qeQxqh4Ax7VUizExZOTjHNQ3N9sjfL\nRgg9CfegCSaRBk57elLZXEZ8z5vTsfess3bS8DYQeOKtWEbnzPkbt2+tAHpWiRq0UTKgOYQc4+le\ni6Db/wChWreWv3jzgf3jXEeErT7SkMeHJFqpwvX+GvSdLtjb2sCbXG0/xD3oA0mGw46U8ECmSglh\ngHpQnJ+f5R69KAHK258ZzUpXHaoi9rF8/wBoQEerimSajaKpP2qDdjgeYOf1oAnx7UEe1UodUtm3\nebdW6emZAM/rUj6lZY+S8t2PoJVP9aAJJB976VUcc08XsLsB50W0nGQ4pk8sQceVKjrjqGB5oAho\nxSOxUZ4qT5PLDbhnA4zQA3Ax0pjAccClL/MBkbe5qG8uba32b7iJd2cbnAoAhuQNnQferGu3RHky\ncYHp7VQ1zxWLWMi0ms5pBLtK7txA55wD9K5W48U6vcSP5dnC4cYBSJznjHHNAGrqNzH9oXDn7nof\nU1zN5GZIQAoPzZpJ7zWpXDNprg4xxA9VZ7u8VATb457oaAOF1uMo14doGJT/AOhVhIC/yDlmOAK6\njXos2d1MwYMWyR2yWFYWl2vn3EJIfHmqPl+ooA09A02c/aN8Cn7uMkH1r0zRrS2hiiMlvEp8kAnY\nOvFc/pVgkPnY8znHX8fauitJolwjyIoVccsBQBfkQEt5KgL2xxUSloxiQnPXrmplmg8v5ZkJ7DcO\naryGSVtyoSMY4BoA4eHlz9Kz7jT/ALRJIvm7dzE/dzjnPrXdaP4Tuobt2uGtZE8sgDJPOR6iuivP\nDttFpjSC0sw4VfmEYz1HtQB5HFpH2aFpPP3bMtjZjOPxq3pk+fN+X07/AFrsLjSVZGVY4ACpGNv/\nANasz+zFtOscI3f3R6fhQB6j4G0/ylgufNz5lmp27emdp65ru0+UD2rnvB8IWwszhf8AjzTt7LXT\nMFUHgce1ACGXHb9apTXuEH7vv61M7jdxmoGtyRztNAGbMfP3j7u459ar/wBj/aCJPP244xsz/Wr8\nt1bWoZpIiQhwcKPpWRqvi3T7KCZRDcqfKZgUVR2Pv7UAM1HTfI8v99uzn+HHp71kvJ9kYtjfztx0\nrmtR+IdmPL3f2gevp7f7VVLL4iaRfXL2/wBmvS6Akl0THBx/e96AOyj1zbtj+z98Z3//AFq6HSH+\n22jSY2Ycrjr2H+NczpGo6ffpbMlqcyvtBeNf72Oea6fymt/kiIjU84TgZoAvsfMGOnemyHy4S3XG\nKmWMk8YrP1JmitZmLHAI6H3oAbJfbAR5eeP71c7rtx9o+z/Lt27u+fSor7VVgR9xlzsJ4/H3riPE\nPiQL9m2yXS/ezhsenvQBYk0bzLuaX7RjezHGzpk/Wr9rb/ZFjG7fsOemM85rzi71u8uy0dte3cbh\nyxJlI4/A1PFb69LppuRq0u3ax5uJM8Z/woA9VS485d2zGDjrXMaidtup/wBv+hrkNP1m/s4Gjub6\n7kctuBErHjA9T7V2QnhuTs2E45+YCgDg9efzbK6jxjLDn/gQrO0OPy9hznEwP8q6HU4Y2luV2Ljz\nDwRx96rGi6WstszLHCMSY5X2HtQBYhv/AC937vOf9qmC882Vh5eOp61qLokrdPJGPr/hVi30kQPu\nkSFuMcDPP5UAVbRspG2O/wDWtm3u/LjI2Z5z1p0cMCxBfJQH1CioZrdy42FVGOnSgD1G+0XTYYQ1\nlADKWwdjsxxg9sn2rn9UWRLKZHVlQEDkYxyK3NIvpbi7ZHVAAhPAPqKo60PNt7qNuAX7f71AHFz7\nNjYI3bTgZrJmt7mfbthlfH91CcV0D6fF5inc/HuP8K1dK0+GTzss/G3oR7+1AHXeH41g0TTyg2y/\nZYww7j5Rnitbc7DnJz7Vl6SfnEP8KR4B78YFaZlaNtoAwPWgBrRvnhG/KkYOB0P5VaikMikkDr2p\n8qAL360AcnqUNy8c4SGRiW4whOea47WdMvpmbdZXLIYiDiJunPtXqR+8R71Wuow4ZCThlwcUAeD3\nnh1G2eZp83fGQ4qLw34Wt21u53abNjy2/v8A94e9exXehWs+zdJMMZ6Ef4VlWlnHp1/NJCzMfmT5\nznjPt9KAMK1tZNOvIYYIZIo45FIBUnHOe9d7pSteWrSTKXYOVBxjjA9PrWULVLu7SSQsC7AHb+Vd\nTptulnbtHGWILlvm+goAlgiG87lIGO9ZOvxoul3JQfNkd/8AaFbUTF2wfSqOt2qHSp2y2TtP/jwo\nA8p1YXD3KoiOysgBAXOeTWXL4dF5j7Rp877Pu/K4xn6fSurvFEepQKvQ7Tz9a2LZRJuz2x0oA81u\n/CFlDbrJbaZL5xIDYMhOMc8Z9a1LPRraPQfKltWRtjgqxYHqfeuwliVWYgnrVOWFZZCjEgN8px70\nAefSeH7FmyLRjx2Zv8a6z+yrGD5lh2k8ZLn/ABrUXQrXH+sm/Mf4UX1sggXlvvf0NAHm+qWaCe6K\nxNjzGwef71SaQTDDtHygyZwfoK1NehW3066nQksGHB6csKxtMkM9uzsACHxx9BQB1Fkyyb8sDjHe\nrFxHiIFFOSe1ZmnMR5n4f1rWklYQJwO38qAM/wDfB8bWxn+7UhJ/i60/zWL9ByaJFBb8KAPatPtJ\nIbhmYqRtxwfcVPqEDSWcigjJx1+oqeJkLnHp6VI67kIxnPagDkpbCUSqdyYGO5pWhZepFdFNbZVi\nI16e1ZdxbS/LhPXuKAKVkpS6dj0Knp9RWgDmqwgkj+YrjPGcipkzgZoAsL0qUMKhBoz70ASHnNMZ\nCQenSnKRkZpXYbT9KAKM0LNtwRVT7DKzkhk59zV9nXjmpomj4zjp6UAV7e1dETJXg9vrV0HaMGl3\nx4wMflUTEk8GgBEGDTNR+bTpAPb+YqfiorkBrdlxnpx+NAHnutjZqcLHoEU8f7xrQ0e8j/ffK38P\nb60eI7ZvNZxGPlg68cda5/Tp2i83dIwzjufegDr/ACyzFwRhuRVqNSIB9DRbFHsLcgAkxqScdeKe\nxAjI6cGgDMvP9cP92sG8YCEf71bt0cyj6VzuoHEC8/xf0NAHJ6kc/aP98/zrIVT58Z9CP51sXKl5\nJRjOWP8AOqjRbZF+UCgDRsJFTzMg84/rWoJ1KLwelY1urfNj2q+qvsX6etAEjuCSeagbk0OcEgmo\nt/uaAPctPvvOnZcx8Lng+4rXByoPtXGaFcZvX+T/AJZnv7iuxjOYUPqooAVsHj1qGS2VsctUpHzC\nn0AZt3bqsQxu+9VLbt/CtS/bEI4/i/oazGbOaAFDU/FRCnhsdqAHbsU1nyD06UhOajbhgKAGMOlI\nJtndfTmnSNtxxWfLN8zDb39aANAXBJHK1YRwRyR+dZkPzBT0yatAbeKALSuSegpshJUjFKgwaey8\nUAc9rVuJYZiQ3+pI4/GuJaIwYwG+b1r0HVm2QyjGf3J/ka4W9fOzj1oA6bSrsywQxEp8sQ6degq/\nIflb6Vz2iy4m6f8ALL1+lb2/dGTjsaAM64/1g+lc3qTgWy8j74/ka6O6bEo4/hrktUf/AEZeP4x/\nI0AYTyAzv8w+8e9QysDKpBFV5ZsTP8v8R70iybxuxjFAGlasvz5Ydu9aO5REvzDoO9YUEuN3Hp3r\nR35iXjsKACaUeawytJHHvXOD17VUlfE549Ku2kv7o/L/ABetAHV6VdzW90ztNKQUI4Y+orsrXWwY\nolLz52jv7fWuMhiw5+U9K2Lc7dnOMD+lAHTDVQed035//XpraoeMPN+f/wBesZZePvineZn+IUAa\njah5gwWkPfk//XqWLMkYcHg+tY+/HO4VoWk/7pF3jr049aALLfKcUnPrTyNxzjNNNACo2XC0sgAb\nkdqYvDg9KbO55wf4aAIbqZV2de9Za7prhwCepPJ96tTF5MZBOPap7SzXfu8pslevNAFqxs2NtG52\nEZJP51LcTQwSBWTJIzwBUse+OEIoIAzgYqvPA8zhmjZiBjODQBQTWoJDhVlB68gf404azATs2y5+\ng/xpsmnW8K7khIJOOprnpXlS7kCZwHYDj3oAl8R61FFHLxNxbk8Y9/euFj1eO9zs84bOu73/AB9q\n3tajnuIpcxu+YSvC/WuZsLB4vM/0eRc46qfegDq9LlK7Xy3MY6fhW9FcAwA5bvXL2LshCsdoCY5/\nCtaKZvJG1hjmgCzOTK4ZTgYxzXHalKDbr1++P5GurWXj5mGfeuH1WYi1Xa4zvH8jQBz08wa4kUZz\nuP8AOnxShUKnJJNU0LvfsGyQWbt9asONsgAGKALkLgbutakcqGJBg9B2rFjJ5q/EzbV+npQA25ce\nc5GR/wDqot3cxnDsOfWoLhm81v8APan2zfuzkjrQB6SowanVyMdKoCQZ71KrjjrQBeEh9qkWQ89K\nohx71IHA9aALjyHYOlTWszeZEMD7w/nWfvB45p8LhZkJzww/nQB19v8ANGSfWmbRVayuU8k8N96r\nycmgBjqBHmoGG4HPpV2TmIiq0ikQufRTQBUZAvTNWYLl1OAF4GKyblgdv40Q2zscgryM0AdEkrOo\nJAyak81l4AFc86GNWBxwO1UZPmbI9KAOnmiV0AJPWsCWwi+0yNufO4nqP8Kx5mCoCfWsi4O4yAdz\n/WgDor+3SOGTBb/Vk8/jXNO5jxjHPrWXf2zyhtpX7hHP41hvp03HzJ+Z/wAKAOi+0v5zjC9TW9p6\niWziZupz0+prjtC06Zb+Qlk/1Z7n1HtXcWY8mzRG5K5zj6mgCnfyG3nVEAIK55+prhbqQyxBWAA3\nZ4rrNclVr1CAf9WP5muLuP3cYJ9e1AFZYlSbeCc5PWmzMTKv0oZwwIGeajKmgCxGx5rTtxuC5/u1\ngn5etadjKu4DB+5/hQBLdIA0h56f0qkkrIMACtOQb1bHcVVFrIe6/nQB3mfenq3Tk1VErk8gflUo\nY8cUAWlf3NP3+5qqHYdqcHY9hQBbjbLHk9KeWwc5IxVVXZeQMmnmRip3AD1oA1LS6xEf3rfe9TXU\nRN8x57VwkUxRSEwRmu3Xepyqkn6UATb8tjND4IKdyMYqJN3mZZcfhU6orENnkGgCt9hZusKn64qz\nGkMSgMiAgYPy1MHI6CoXUsScHrQBHLHE+4hFII/u1TaGJTgxp/3yKuEMAQFP5VA6MxyVP5UAYl2k\nPlDCJ97+7WcYIiT+6Q5/2RWu1pJKNrRSAdeFNVJLSSPdiOTg91oAyrq3iCP+6T7h/hFYNzADt2xr\n36AV1U1s7o25HHykdKyp7ILt4fvQBS06L7NMZJFCqUwD+VX5btBG+JCAFPTNQXASKBMsByB8x9q5\n/VNZFslyqSQEpGSAx/2c+tAFXXNTjS9QG4cHyx/e9TWJNcJKgVXJOc96ozXr6o4nkCAqNn7vp6+/\nrVuW18lQyByScc0ABG1Nx6etNEqZA3daQyMU2uAq9z0qu3E6BPmXIyRz3oAuFQ3QA1PAdjZ6cYqJ\nM88U9BIWPyHH0oAti4QcFz+tWYZ4ih+bv6VmsoAJbg+lMFwYxhdpFAHoA4NSBunFQBW9akVGJHNA\nEob2p6nrTFiY9xUgiYdxQA8PsGcZpzNugZsfwmkW3km+VWAI55NXobB/s4VihODQBTsIvNgZs4+b\nHT2FegQfM5HtWJpGlgWj5SE/Oe3sPatONyG6npQBZmXbGzVHFLhenf1qdlLwfUDrVV0KNjI/CgCy\nsmc8U8HNU1YjuakWXHc0AWCuc1GU96FkBA608MPSgCIjAqpNHuVuepqWXeFHzHr61A5Ow5JoApXE\neARn+Gse8jxs59e1aV9NsY8t9zPH41gX96q+X9/v/SgDnPEuofZLQHyt/wC+2/ex2PtXnWq6p59z\nOPJ271x97OPl+ldP4iuGuw8as/E5PzHjvXI3VqweRjtOBn9KAJ9Ij32jnOP3h7ewrpZo8oOe/pXM\n6Y+22YDI+c9PoK7K402fyxh06+p/woAw7m2zE/z/AKe9VEi8rvnnPSluGlS4ljMjfK5H3jjg06AF\noznk570ATRy9eP1rShXcqnPVc1k7GHQ1pWjngEnhaAIrv5XkHt/Sqca71znHNaFzGWMjcdP6VRV1\niG0j34oA9UWzGf8AVN+tWY7FSV/ct0960kiXPU1cihXK8npQBkrYLj/Ut+tI1oox+6YfnW/5ShDy\naqXCgbfxoAraRaRSXjrJGSAhPUjuKnufIgvTArKoBAC7ueQKZYytDdOygE7SOfqKxL++lfxUEKph\npYgeD6LQB2mnELbsAcfOf5CkT734U6xQeS3X739BTygXkZoAso3yKMjpUcw+b8KSI5kUVLIgz36U\nAU2OMYqFpWX+IDmrUkY461RuFAB/3qAHfaiv/LRR+VL9ux/y3T8xWZMxBb6VReZgegoA2pL8Ff8A\nj4jPPqKrSX3yn9+n5isISsewqJ5W54FAFzUbwkMRKpxGfT3rltQvXPl/vV7+ntWjduTDIePuGuZu\n2J2fjQBlXwdndsE5cnOPrWVPFu3gqSSMY/Cuoa1SSFSS3ODxWBeDytTaJfuhl69egoArWVoFhI8p\nh83vXeXE1sIx++j6/wB8VztrCrREknrUzRLKNrEgdeKAOfvJof7RuP3iY81v4h6mmpN86iJwVJGc\nc0l9pkKvPIGkzvJ6j1+lWNI0+KSLcWfIk7Ee3tQBKAx6g/lVm1OH54+WrpsYl6M/5iqzII3OM9cc\n0ATuFaJuhOD3rIuEbzBtU4x6Ve81vu4GKckSyDJJ9OKAP//Z\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "emotion: Neutral\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/jpeg": "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAgGBgcGBQgHBwcJCQgKDBQNDAsLDBkSEw8UHRofHh0a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\npnOZmH+0agl4lUUAWkc880jEmo4+c1LjigADEAc0vmN/epNoppHNAH//2Q==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "emotion: Angry\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/jpeg": "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAgGBgcGBQgHBwcJCQgKDBQNDAsLDBkSEw8UHRofHh0a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\n5HoarmZj2FAH/9k=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "emotion: Happy\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/jpeg": "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAgGBgcGBQgHBwcJCQgKDBQNDAsLDBkSEw8UHRofHh0a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\nQB//2Q==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "emotion: Happy\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/jpeg": "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAgGBgcGBQgHBwcJCQgKDBQNDAsLDBkSEw8UHRofHh0a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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "emotion: Happy\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/jpeg": "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAgGBgcGBQgHBwcJCQgKDBQNDAsLDBkSEw8UHRofHh0a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\nwoYDNADEchlGe9NumHmjkfdpWG2TjtVW8kIlHT7tAH//2Q==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "emotion: Fear\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/jpeg": "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAgGBgcGBQgHBwcJCQgKDBQNDAsLDBkSEw8UHRofHh0a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\nOhjzhn59SP8ACpVjCcAnjigCcNx1FIZD/eFR44o2igD/2Q==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "emotion: Happy\n",
      "\n"
     ]
    }
   ],
   "source": [
    "import random \n",
    "from glob import glob\n",
    "from keras.preprocessing import image\n",
    "\n",
    "paths = glob(\"./images/validation/*/*.jpg\")\n",
    "random.shuffle(paths)\n",
    "\n",
    "for path in paths[:10]:\n",
    "#    face = mat_faces[random.randint(0, mat_faces.shape[0])]\n",
    "    img = image.load_img(path, target_size=(TARGET_SIZE, TARGET_SIZE))\n",
    "    img = image.img_to_array(img)\n",
    "    show_img_array(img)\n",
    "\n",
    "    result = model.predict(np.array([img])/255.)\n",
    "    print \"emotion:\", EMOTIONS[np.argmax(result)]\n",
    "    print"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [default]",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}