Model_03-Adam-k=10-amsgrad-0001-15p-Copy4.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Mandatory imports\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import log_loss\n",
    "from sklearn.model_selection import StratifiedKFold, StratifiedShuffleSplit\n",
    "from os.path import join as opj\n",
    "from matplotlib import pyplot as plt\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "import pylab\n",
    "plt.rcParams['figure.figsize'] = 10, 10\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = pd.read_json(\"input/train.json\")\n",
    "target_train=train['is_iceberg']\n",
    "test = pd.read_json(\"input/test.json\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "target_train=train['is_iceberg']\n",
    "test['inc_angle']=pd.to_numeric(test['inc_angle'], errors='coerce')\n",
    "train['inc_angle']=pd.to_numeric(train['inc_angle'], errors='coerce')#We have only 133 NAs.\n",
    "train['inc_angle']=train['inc_angle'].fillna(method='pad')\n",
    "X_angle=train['inc_angle']\n",
    "test['inc_angle']=pd.to_numeric(test['inc_angle'], errors='coerce')\n",
    "X_test_angle=test['inc_angle']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Generate the training data\n",
    "X_band_1=np.array([np.array(band).astype(np.float32).reshape(75, 75) for band in train[\"band_1\"]])\n",
    "X_band_2=np.array([np.array(band).astype(np.float32).reshape(75, 75) for band in train[\"band_2\"]])\n",
    "X_band_3=(X_band_1+X_band_2)/2\n",
    "#X_band_3=np.array([np.full((75, 75), angel).astype(np.float32) for angel in train[\"inc_angle\"]])\n",
    "X_train = np.concatenate([X_band_1[:, :, :, np.newaxis]\n",
    "                          , X_band_2[:, :, :, np.newaxis]\n",
    "                         , X_band_3[:, :, :, np.newaxis]], axis=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_band_test_1=np.array([np.array(band).astype(np.float32).reshape(75, 75) for band in test[\"band_1\"]])\n",
    "X_band_test_2=np.array([np.array(band).astype(np.float32).reshape(75, 75) for band in test[\"band_2\"]])\n",
    "X_band_test_3=(X_band_test_1+X_band_test_2)/2\n",
    "#X_band_test_3=np.array([np.full((75, 75), angel).astype(np.float32) for angel in test[\"inc_angle\"]])\n",
    "X_test = np.concatenate([X_band_test_1[:, :, :, np.newaxis]\n",
    "                          , X_band_test_2[:, :, :, np.newaxis]\n",
    "                         , X_band_test_3[:, :, :, np.newaxis]], axis=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "#Import Keras.\n",
    "from matplotlib import pyplot\n",
    "from keras.optimizers import RMSprop\n",
    "# from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Conv2D, MaxPooling2D, Dense, Dropout, Input, Flatten, Activation\n",
    "from keras.layers import GlobalMaxPooling2D\n",
    "from keras.layers.normalization import BatchNormalization\n",
    "from keras.layers.merge import Concatenate\n",
    "from keras.models import Model\n",
    "from keras import initializers\n",
    "from keras.optimizers import Adam\n",
    "from keras.optimizers import rmsprop\n",
    "from keras.layers.advanced_activations import LeakyReLU, PReLU\n",
    "from keras.optimizers import SGD\n",
    "from keras.optimizers import Adadelta\n",
    "from keras.callbacks import ModelCheckpoint, Callback, EarlyStopping\n",
    "\n",
    "from keras.datasets import cifar10\n",
    "from keras.applications.inception_v3 import InceptionV3\n",
    "from keras.applications.vgg16 import VGG16\n",
    "from keras.applications.xception import Xception\n",
    "from keras.applications.mobilenet import MobileNet\n",
    "from keras.applications.vgg19 import VGG19\n",
    "from keras.layers import Concatenate, Dense, LSTM, Input, concatenate\n",
    "from keras.preprocessing import image\n",
    "from keras.applications.vgg16 import preprocess_input"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Data Aug for multi-input\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "batch_size=64\n",
    "# Define the image transformations here\n",
    "gen = ImageDataGenerator(horizontal_flip = True,\n",
    "                         vertical_flip = True,\n",
    "                         width_shift_range = 0.,\n",
    "                         height_shift_range = 0.,\n",
    "                         channel_shift_range=0,\n",
    "                         zoom_range = 0.2,\n",
    "                         rotation_range = 10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Here is the function that merges our two generators\n",
    "# We use the exact same generator with the same random seed for both the y and angle arrays\n",
    "def gen_flow_for_two_inputs(X1, X2, y):\n",
    "    genX1 = gen.flow(X1,y,  batch_size=batch_size,seed=55)\n",
    "    genX2 = gen.flow(X1,X2, batch_size=batch_size,seed=55)\n",
    "    while True:\n",
    "            X1i = genX1.next()\n",
    "            X2i = genX2.next()\n",
    "            #Assert arrays are equal - this was for peace of mind, but slows down training\n",
    "            #np.testing.assert_array_equal(X1i[0],X2i[0])\n",
    "            yield [X1i[0], X2i[1]], X1i[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Finally create generator\n",
    "def get_callbacks(filepath, patience=2):\n",
    "    es = EarlyStopping('val_loss', patience=15, mode=\"min\")\n",
    "    msave = ModelCheckpoint(filepath, save_best_only=True)\n",
    "    return [es, msave]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def getVggAngleModel():\n",
    "    input_2 = Input(shape=[1], name=\"angle\")\n",
    "    angle_layer = Dense(1, )(input_2)\n",
    "    base_model = VGG16(weights='imagenet', include_top=False, \n",
    "                 input_shape=X_train.shape[1:], classes=1)\n",
    "    x = base_model.get_layer('block5_pool').output\n",
    "    \n",
    "\n",
    "    x = GlobalMaxPooling2D()(x)\n",
    "    merge_one = concatenate([x, angle_layer])\n",
    "    merge_one = Dense(512, activation='relu', name='fc2')(merge_one)\n",
    "    merge_one = Dropout(0.3)(merge_one)\n",
    "    merge_one = Dense(512, activation='relu', name='fc3')(merge_one)\n",
    "    merge_one = Dropout(0.3)(merge_one)\n",
    "    \n",
    "    predictions = Dense(1, activation='sigmoid')(merge_one)\n",
    "    \n",
    "    model = Model(inputs=[base_model.input, input_2], outputs=predictions)\n",
    "    \n",
    "    adam = Adam(lr=0.0001, beta_1=0.9, beta_2=0.999, epsilon=1e-8, decay=1e-6, amsgrad=True)\n",
    "    model.compile(loss='binary_crossentropy',\n",
    "                    optimizer=adam,\n",
    "                    metrics=['accuracy'])\n",
    "    \n",
    "#     sgd = SGD(lr=1e-3, decay=1e-6, momentum=0.9, nesterov=True)\n",
    "#     model.compile(loss='binary_crossentropy',\n",
    "#                   optimizer=sgd,\n",
    "#                   metrics=['accuracy'])\n",
    "    return model\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "===================FOLD= 0\n",
      "Epoch 1/100\n",
      "24/24 [==============================] - 20s 815ms/step - loss: 0.6233 - acc: 0.6295 - val_loss: 0.3974 - val_acc: 0.8210\n",
      "Epoch 2/100\n",
      "24/24 [==============================] - 14s 594ms/step - loss: 0.4824 - acc: 0.7564 - val_loss: 0.3715 - val_acc: 0.8272\n",
      "Epoch 3/100\n",
      "24/24 [==============================] - 14s 596ms/step - loss: 0.3568 - acc: 0.8422 - val_loss: 0.2923 - val_acc: 0.8580\n",
      "Epoch 4/100\n",
      "24/24 [==============================] - 14s 598ms/step - loss: 0.3249 - acc: 0.8617 - val_loss: 0.2738 - val_acc: 0.8951\n",
      "Epoch 5/100\n",
      "24/24 [==============================] - 14s 597ms/step - loss: 0.2963 - acc: 0.8676 - val_loss: 0.2394 - val_acc: 0.8951\n",
      "Epoch 6/100\n",
      "24/24 [==============================] - 12s 520ms/step - loss: 0.2525 - acc: 0.8877 - val_loss: 0.2920 - val_acc: 0.8519\n",
      "Epoch 7/100\n",
      "24/24 [==============================] - 12s 520ms/step - loss: 0.2537 - acc: 0.8850 - val_loss: 0.3177 - val_acc: 0.8704\n",
      "Epoch 8/100\n",
      "24/24 [==============================] - 14s 600ms/step - loss: 0.2576 - acc: 0.8853 - val_loss: 0.2229 - val_acc: 0.9198\n",
      "Epoch 9/100\n",
      "24/24 [==============================] - 14s 601ms/step - loss: 0.2085 - acc: 0.9122 - val_loss: 0.2163 - val_acc: 0.9074\n",
      "Epoch 10/100\n",
      "24/24 [==============================] - 13s 522ms/step - loss: 0.2138 - acc: 0.9143 - val_loss: 0.2711 - val_acc: 0.8827\n",
      "Epoch 11/100\n",
      "24/24 [==============================] - 14s 601ms/step - loss: 0.2206 - acc: 0.9085 - val_loss: 0.2080 - val_acc: 0.9198\n",
      "Epoch 12/100\n",
      "24/24 [==============================] - 13s 521ms/step - loss: 0.2018 - acc: 0.9191 - val_loss: 0.2279 - val_acc: 0.9074\n",
      "Epoch 13/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.2742 - acc: 0.8805 - val_loss: 0.2702 - val_acc: 0.8951\n",
      "Epoch 14/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1982 - acc: 0.9238 - val_loss: 0.2200 - val_acc: 0.9136\n",
      "Epoch 15/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1738 - acc: 0.9330 - val_loss: 0.2254 - val_acc: 0.9259\n",
      "Epoch 16/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2111 - acc: 0.9138 - val_loss: 0.2084 - val_acc: 0.9259\n",
      "Epoch 17/100\n",
      "24/24 [==============================] - 14s 604ms/step - loss: 0.1845 - acc: 0.9216 - val_loss: 0.1919 - val_acc: 0.9198\n",
      "Epoch 18/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.1929 - acc: 0.9266 - val_loss: 0.2012 - val_acc: 0.9198\n",
      "Epoch 19/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1744 - acc: 0.9318 - val_loss: 0.2174 - val_acc: 0.9074\n",
      "Epoch 20/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2027 - acc: 0.9180 - val_loss: 0.2262 - val_acc: 0.9012\n",
      "Epoch 21/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1911 - acc: 0.9168 - val_loss: 0.1952 - val_acc: 0.9198\n",
      "Epoch 22/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2019 - acc: 0.9179 - val_loss: 0.2191 - val_acc: 0.9136\n",
      "Epoch 23/100\n",
      "24/24 [==============================] - 13s 542ms/step - loss: 0.1927 - acc: 0.9222 - val_loss: 0.1982 - val_acc: 0.9198\n",
      "Epoch 24/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1862 - acc: 0.9205 - val_loss: 0.2080 - val_acc: 0.9136\n",
      "Epoch 25/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1760 - acc: 0.9318 - val_loss: 0.1950 - val_acc: 0.9136\n",
      "Epoch 26/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1614 - acc: 0.9271 - val_loss: 0.2006 - val_acc: 0.9198\n",
      "Epoch 27/100\n",
      "24/24 [==============================] - 14s 604ms/step - loss: 0.1634 - acc: 0.9369 - val_loss: 0.1802 - val_acc: 0.9383\n",
      "Epoch 28/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1585 - acc: 0.9369 - val_loss: 0.2409 - val_acc: 0.9136\n",
      "Epoch 29/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1510 - acc: 0.9376 - val_loss: 0.1906 - val_acc: 0.9136\n",
      "Epoch 30/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1505 - acc: 0.9429 - val_loss: 0.3114 - val_acc: 0.8827\n",
      "Epoch 31/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.1622 - acc: 0.9292 - val_loss: 0.1884 - val_acc: 0.9383\n",
      "Epoch 32/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1744 - acc: 0.9311 - val_loss: 0.1966 - val_acc: 0.9383\n",
      "Epoch 33/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.1367 - acc: 0.9396 - val_loss: 0.2005 - val_acc: 0.9321\n",
      "Epoch 34/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1359 - acc: 0.9465 - val_loss: 0.2271 - val_acc: 0.9259\n",
      "Epoch 35/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1464 - acc: 0.9442 - val_loss: 0.2161 - val_acc: 0.9074\n",
      "Epoch 36/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1069 - acc: 0.9635 - val_loss: 0.2503 - val_acc: 0.9198\n",
      "Epoch 37/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1262 - acc: 0.9506 - val_loss: 0.2515 - val_acc: 0.9012\n",
      "Epoch 38/100\n",
      "24/24 [==============================] - 15s 604ms/step - loss: 0.1352 - acc: 0.9474 - val_loss: 0.1742 - val_acc: 0.9198\n",
      "Epoch 39/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1441 - acc: 0.9350 - val_loss: 0.1987 - val_acc: 0.9074\n",
      "Epoch 40/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1305 - acc: 0.9483 - val_loss: 0.1882 - val_acc: 0.9383\n",
      "Epoch 41/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1178 - acc: 0.9483 - val_loss: 0.1833 - val_acc: 0.9259\n",
      "Epoch 42/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.1117 - acc: 0.9598 - val_loss: 0.1988 - val_acc: 0.9321\n",
      "Epoch 43/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1282 - acc: 0.9495 - val_loss: 0.2361 - val_acc: 0.9198\n",
      "Epoch 44/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1266 - acc: 0.9493 - val_loss: 0.2045 - val_acc: 0.9383\n",
      "Epoch 45/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1352 - acc: 0.9523 - val_loss: 0.1953 - val_acc: 0.9259\n",
      "Epoch 46/100\n",
      "24/24 [==============================] - 12s 517ms/step - loss: 0.1178 - acc: 0.9442 - val_loss: 0.1929 - val_acc: 0.9259\n",
      "Epoch 47/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1059 - acc: 0.9604 - val_loss: 0.2123 - val_acc: 0.9321\n",
      "Epoch 48/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.0956 - acc: 0.9656 - val_loss: 0.2225 - val_acc: 0.9198\n",
      "Epoch 49/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.0960 - acc: 0.9630 - val_loss: 0.2046 - val_acc: 0.9198\n",
      "Epoch 50/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.1032 - acc: 0.9617 - val_loss: 0.1995 - val_acc: 0.9259\n",
      "Epoch 51/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.0776 - acc: 0.9658 - val_loss: 0.2982 - val_acc: 0.9259\n",
      "Epoch 52/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.0821 - acc: 0.9685 - val_loss: 0.2089 - val_acc: 0.9198\n",
      "Epoch 53/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.0978 - acc: 0.9611 - val_loss: 0.2400 - val_acc: 0.9136\n",
      "Train loss: 0.0857778240548\n",
      "Train accuracy: 0.966019417476\n",
      "Test loss: 0.17420318437\n",
      "Test accuracy: 0.91975308642\n",
      "\n",
      "===================FOLD= 1\n",
      "Epoch 1/100\n",
      "24/24 [==============================] - 18s 747ms/step - loss: 0.8961 - acc: 0.6042 - val_loss: 0.4056 - val_acc: 0.8075\n",
      "Epoch 2/100\n",
      "24/24 [==============================] - 15s 609ms/step - loss: 0.5166 - acc: 0.7816 - val_loss: 0.3111 - val_acc: 0.8509\n",
      "Epoch 3/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.4352 - acc: 0.8033 - val_loss: 0.2417 - val_acc: 0.8944\n",
      "Epoch 4/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.3561 - acc: 0.8473 - val_loss: 0.2920 - val_acc: 0.8509\n",
      "Epoch 5/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.3458 - acc: 0.8469 - val_loss: 0.2891 - val_acc: 0.8758\n",
      "Epoch 6/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.3163 - acc: 0.8612 - val_loss: 0.1986 - val_acc: 0.9130\n",
      "Epoch 7/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "24/24 [==============================] - 13s 528ms/step - loss: 0.3226 - acc: 0.8572 - val_loss: 0.2174 - val_acc: 0.8944\n",
      "Epoch 8/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.2977 - acc: 0.8698 - val_loss: 0.2311 - val_acc: 0.8696\n",
      "Epoch 9/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.2392 - acc: 0.8990 - val_loss: 0.1886 - val_acc: 0.9006\n",
      "Epoch 10/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.2824 - acc: 0.8855 - val_loss: 0.1782 - val_acc: 0.9193\n",
      "Epoch 11/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2860 - acc: 0.8848 - val_loss: 0.2118 - val_acc: 0.9068\n",
      "Epoch 12/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.2388 - acc: 0.9051 - val_loss: 0.1837 - val_acc: 0.9130\n",
      "Epoch 13/100\n",
      "24/24 [==============================] - 15s 608ms/step - loss: 0.2283 - acc: 0.9040 - val_loss: 0.1686 - val_acc: 0.9317\n",
      "Epoch 14/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2159 - acc: 0.9045 - val_loss: 0.1774 - val_acc: 0.9255\n",
      "Epoch 15/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.2143 - acc: 0.9061 - val_loss: 0.2027 - val_acc: 0.9130\n",
      "Epoch 16/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2112 - acc: 0.9089 - val_loss: 0.1979 - val_acc: 0.9379\n",
      "Epoch 17/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2026 - acc: 0.9131 - val_loss: 0.2480 - val_acc: 0.9006\n",
      "Epoch 18/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2351 - acc: 0.8951 - val_loss: 0.2217 - val_acc: 0.9193\n",
      "Epoch 19/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2076 - acc: 0.9092 - val_loss: 0.2155 - val_acc: 0.9317\n",
      "Epoch 20/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.1936 - acc: 0.9248 - val_loss: 0.1922 - val_acc: 0.9255\n",
      "Epoch 21/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1732 - acc: 0.9270 - val_loss: 0.2620 - val_acc: 0.8882\n",
      "Epoch 22/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1891 - acc: 0.9157 - val_loss: 0.2341 - val_acc: 0.9130\n",
      "Epoch 23/100\n",
      "24/24 [==============================] - 15s 619ms/step - loss: 0.1790 - acc: 0.9282 - val_loss: 0.1643 - val_acc: 0.9255\n",
      "Epoch 24/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.2091 - acc: 0.9148 - val_loss: 0.2075 - val_acc: 0.9441\n",
      "Epoch 25/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.1596 - acc: 0.9310 - val_loss: 0.1832 - val_acc: 0.9503\n",
      "Epoch 26/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1509 - acc: 0.9350 - val_loss: 0.2770 - val_acc: 0.9193\n",
      "Epoch 27/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.1617 - acc: 0.9334 - val_loss: 0.2951 - val_acc: 0.9193\n",
      "Epoch 28/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1441 - acc: 0.9384 - val_loss: 0.2232 - val_acc: 0.9193\n",
      "Epoch 29/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1667 - acc: 0.9363 - val_loss: 0.2574 - val_acc: 0.9255\n",
      "Epoch 30/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1412 - acc: 0.9358 - val_loss: 0.2794 - val_acc: 0.9255\n",
      "Epoch 31/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1678 - acc: 0.9213 - val_loss: 0.1990 - val_acc: 0.9379\n",
      "Epoch 32/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.1242 - acc: 0.9487 - val_loss: 0.3980 - val_acc: 0.9130\n",
      "Epoch 33/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1654 - acc: 0.9388 - val_loss: 0.2314 - val_acc: 0.8944\n",
      "Epoch 34/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1838 - acc: 0.9283 - val_loss: 0.2723 - val_acc: 0.8882\n",
      "Epoch 35/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1549 - acc: 0.9315 - val_loss: 0.2068 - val_acc: 0.9193\n",
      "Epoch 36/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1635 - acc: 0.9397 - val_loss: 0.1932 - val_acc: 0.9255\n",
      "Epoch 37/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.1420 - acc: 0.9446 - val_loss: 0.2307 - val_acc: 0.9130\n",
      "Epoch 38/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1494 - acc: 0.9449 - val_loss: 0.2133 - val_acc: 0.9255\n",
      "Train loss: 0.125369279627\n",
      "Train accuracy: 0.954261954262\n",
      "Test loss: 0.164256208228\n",
      "Test accuracy: 0.925465838509\n",
      "\n",
      "===================FOLD= 2\n",
      "Epoch 1/100\n",
      "24/24 [==============================] - 16s 658ms/step - loss: 0.6605 - acc: 0.6267 - val_loss: 0.4844 - val_acc: 0.7764\n",
      "Epoch 2/100\n",
      "24/24 [==============================] - 15s 608ms/step - loss: 0.4411 - acc: 0.7721 - val_loss: 0.2807 - val_acc: 0.8634\n",
      "Epoch 3/100\n",
      "24/24 [==============================] - 15s 609ms/step - loss: 0.3238 - acc: 0.8396 - val_loss: 0.2289 - val_acc: 0.9006\n",
      "Epoch 4/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.3256 - acc: 0.8452 - val_loss: 0.2760 - val_acc: 0.8634\n",
      "Epoch 5/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.2895 - acc: 0.8658 - val_loss: 0.2390 - val_acc: 0.8882\n",
      "Epoch 6/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.2797 - acc: 0.8827 - val_loss: 0.2062 - val_acc: 0.9193\n",
      "Epoch 7/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.2551 - acc: 0.8902 - val_loss: 0.2117 - val_acc: 0.9130\n",
      "Epoch 8/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.2345 - acc: 0.8950 - val_loss: 0.2149 - val_acc: 0.8944\n",
      "Epoch 9/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.2296 - acc: 0.9079 - val_loss: 0.1957 - val_acc: 0.9379\n",
      "Epoch 10/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2241 - acc: 0.9068 - val_loss: 0.1965 - val_acc: 0.9193\n",
      "Epoch 11/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.2227 - acc: 0.9096 - val_loss: 0.1865 - val_acc: 0.9193\n",
      "Epoch 12/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2320 - acc: 0.8938 - val_loss: 0.2106 - val_acc: 0.9006\n",
      "Epoch 13/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2134 - acc: 0.9149 - val_loss: 0.2160 - val_acc: 0.9130\n",
      "Epoch 14/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.1938 - acc: 0.9193 - val_loss: 0.1780 - val_acc: 0.9130\n",
      "Epoch 15/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.1967 - acc: 0.9152 - val_loss: 0.2195 - val_acc: 0.9255\n",
      "Epoch 16/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2103 - acc: 0.9124 - val_loss: 0.1825 - val_acc: 0.9068\n",
      "Epoch 17/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2221 - acc: 0.9102 - val_loss: 0.1826 - val_acc: 0.9255\n",
      "Epoch 18/100\n",
      "24/24 [==============================] - 15s 605ms/step - loss: 0.1785 - acc: 0.9201 - val_loss: 0.1618 - val_acc: 0.9193\n",
      "Epoch 19/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1804 - acc: 0.9265 - val_loss: 0.1746 - val_acc: 0.9130\n",
      "Epoch 20/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1973 - acc: 0.9239 - val_loss: 0.1640 - val_acc: 0.9193\n",
      "Epoch 21/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1953 - acc: 0.9246 - val_loss: 0.2310 - val_acc: 0.9068\n",
      "Epoch 22/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1920 - acc: 0.9149 - val_loss: 0.1835 - val_acc: 0.9068\n",
      "Epoch 23/100\n",
      "24/24 [==============================] - 13s 521ms/step - loss: 0.1704 - acc: 0.9329 - val_loss: 0.2042 - val_acc: 0.9317\n",
      "Epoch 24/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.1791 - acc: 0.9299 - val_loss: 0.1836 - val_acc: 0.9068\n",
      "Epoch 25/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1498 - acc: 0.9379 - val_loss: 0.1697 - val_acc: 0.9317\n",
      "Epoch 26/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.1871 - acc: 0.9248 - val_loss: 0.1519 - val_acc: 0.9193\n",
      "Epoch 27/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1669 - acc: 0.9343 - val_loss: 0.1608 - val_acc: 0.9255\n",
      "Epoch 28/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1577 - acc: 0.9345 - val_loss: 0.1737 - val_acc: 0.9193\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 29/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1426 - acc: 0.9455 - val_loss: 0.1774 - val_acc: 0.9006\n",
      "Epoch 30/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1219 - acc: 0.9513 - val_loss: 0.1771 - val_acc: 0.9193\n",
      "Epoch 31/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.1470 - acc: 0.9393 - val_loss: 0.1334 - val_acc: 0.9379\n",
      "Epoch 32/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1557 - acc: 0.9349 - val_loss: 0.1877 - val_acc: 0.8944\n",
      "Epoch 33/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1674 - acc: 0.9283 - val_loss: 0.1647 - val_acc: 0.9255\n",
      "Epoch 34/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1438 - acc: 0.9468 - val_loss: 0.2052 - val_acc: 0.9006\n",
      "Epoch 35/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1209 - acc: 0.9465 - val_loss: 0.2124 - val_acc: 0.9068\n",
      "Epoch 36/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1322 - acc: 0.9511 - val_loss: 0.1433 - val_acc: 0.9379\n",
      "Epoch 37/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1408 - acc: 0.9470 - val_loss: 0.1631 - val_acc: 0.9255\n",
      "Epoch 38/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1451 - acc: 0.9455 - val_loss: 0.2200 - val_acc: 0.9193\n",
      "Epoch 39/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1273 - acc: 0.9485 - val_loss: 0.2137 - val_acc: 0.9006\n",
      "Epoch 40/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1626 - acc: 0.9352 - val_loss: 0.2069 - val_acc: 0.8944\n",
      "Epoch 41/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1217 - acc: 0.9532 - val_loss: 0.2169 - val_acc: 0.9006\n",
      "Epoch 42/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.1219 - acc: 0.9524 - val_loss: 0.1742 - val_acc: 0.9317\n",
      "Epoch 43/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1081 - acc: 0.9502 - val_loss: 0.1971 - val_acc: 0.9193\n",
      "Epoch 44/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1162 - acc: 0.9531 - val_loss: 0.1998 - val_acc: 0.9068\n",
      "Epoch 45/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1076 - acc: 0.9590 - val_loss: 0.1895 - val_acc: 0.9193\n",
      "Epoch 46/100\n",
      "24/24 [==============================] - 12s 518ms/step - loss: 0.1053 - acc: 0.9567 - val_loss: 0.2384 - val_acc: 0.9068\n",
      "Train loss: 0.0867806449326\n",
      "Train accuracy: 0.96534996535\n",
      "Test loss: 0.133399037008\n",
      "Test accuracy: 0.937888198758\n",
      "\n",
      "===================FOLD= 3\n",
      "Epoch 1/100\n",
      "24/24 [==============================] - 18s 735ms/step - loss: 0.7572 - acc: 0.5497 - val_loss: 0.5769 - val_acc: 0.6438\n",
      "Epoch 2/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.6097 - acc: 0.6449 - val_loss: 0.4274 - val_acc: 0.7750\n",
      "Epoch 3/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.4685 - acc: 0.7697 - val_loss: 0.3065 - val_acc: 0.8313\n",
      "Epoch 4/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.3633 - acc: 0.8285 - val_loss: 0.2456 - val_acc: 0.9000\n",
      "Epoch 5/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.3370 - acc: 0.8451 - val_loss: 0.2803 - val_acc: 0.8500\n",
      "Epoch 6/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.3091 - acc: 0.8619 - val_loss: 0.2491 - val_acc: 0.9062\n",
      "Epoch 7/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.2774 - acc: 0.8792 - val_loss: 0.1891 - val_acc: 0.9125\n",
      "Epoch 8/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2917 - acc: 0.8818 - val_loss: 0.2107 - val_acc: 0.8938\n",
      "Epoch 9/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2674 - acc: 0.8879 - val_loss: 0.2375 - val_acc: 0.9187\n",
      "Epoch 10/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2296 - acc: 0.9031 - val_loss: 0.2448 - val_acc: 0.9000\n",
      "Epoch 11/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2274 - acc: 0.9000 - val_loss: 0.2411 - val_acc: 0.8750\n",
      "Epoch 12/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.2181 - acc: 0.9065 - val_loss: 0.2139 - val_acc: 0.9062\n",
      "Epoch 13/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2080 - acc: 0.9070 - val_loss: 0.1898 - val_acc: 0.9000\n",
      "Epoch 14/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.2109 - acc: 0.9174 - val_loss: 0.1803 - val_acc: 0.9125\n",
      "Epoch 15/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2298 - acc: 0.8941 - val_loss: 0.1934 - val_acc: 0.9187\n",
      "Epoch 16/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2225 - acc: 0.9129 - val_loss: 0.1918 - val_acc: 0.9062\n",
      "Epoch 17/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1981 - acc: 0.9238 - val_loss: 0.1845 - val_acc: 0.9250\n",
      "Epoch 18/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1864 - acc: 0.9210 - val_loss: 0.2139 - val_acc: 0.9313\n",
      "Epoch 19/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1790 - acc: 0.9215 - val_loss: 0.2137 - val_acc: 0.9125\n",
      "Epoch 20/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.1925 - acc: 0.9197 - val_loss: 0.2229 - val_acc: 0.9125\n",
      "Epoch 21/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1881 - acc: 0.9254 - val_loss: 0.1810 - val_acc: 0.9313\n",
      "Epoch 22/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.1773 - acc: 0.9256 - val_loss: 0.1667 - val_acc: 0.9125\n",
      "Epoch 23/100\n",
      "24/24 [==============================] - 13s 533ms/step - loss: 0.1689 - acc: 0.9317 - val_loss: 0.1746 - val_acc: 0.9313\n",
      "Epoch 24/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1593 - acc: 0.9308 - val_loss: 0.2089 - val_acc: 0.9187\n",
      "Epoch 25/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1606 - acc: 0.9366 - val_loss: 0.2289 - val_acc: 0.9187\n",
      "Epoch 26/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1700 - acc: 0.9285 - val_loss: 0.3008 - val_acc: 0.8812\n",
      "Epoch 27/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1557 - acc: 0.9378 - val_loss: 0.2673 - val_acc: 0.9000\n",
      "Epoch 28/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1377 - acc: 0.9487 - val_loss: 0.3031 - val_acc: 0.9187\n",
      "Epoch 29/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1557 - acc: 0.9375 - val_loss: 0.1992 - val_acc: 0.9187\n",
      "Epoch 30/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1572 - acc: 0.9334 - val_loss: 0.2014 - val_acc: 0.9125\n",
      "Epoch 31/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1308 - acc: 0.9469 - val_loss: 0.2194 - val_acc: 0.9313\n",
      "Epoch 32/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2194 - acc: 0.9151 - val_loss: 0.2386 - val_acc: 0.9062\n",
      "Epoch 33/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.1788 - acc: 0.9267 - val_loss: 0.1617 - val_acc: 0.9125\n",
      "Epoch 34/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.1678 - acc: 0.9362 - val_loss: 0.1575 - val_acc: 0.9313\n",
      "Epoch 35/100\n",
      "24/24 [==============================] - 15s 605ms/step - loss: 0.1346 - acc: 0.9388 - val_loss: 0.1493 - val_acc: 0.9313\n",
      "Epoch 36/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1482 - acc: 0.9443 - val_loss: 0.2706 - val_acc: 0.9062\n",
      "Epoch 37/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.1301 - acc: 0.9513 - val_loss: 0.2810 - val_acc: 0.8938\n",
      "Epoch 38/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.1472 - acc: 0.9345 - val_loss: 0.2154 - val_acc: 0.9062\n",
      "Epoch 39/100\n",
      "24/24 [==============================] - 13s 523ms/step - loss: 0.1164 - acc: 0.9575 - val_loss: 0.1839 - val_acc: 0.9125\n",
      "Epoch 40/100\n",
      "24/24 [==============================] - 13s 523ms/step - loss: 0.1441 - acc: 0.9445 - val_loss: 0.1760 - val_acc: 0.9250\n",
      "Epoch 41/100\n",
      "24/24 [==============================] - 13s 523ms/step - loss: 0.1222 - acc: 0.9482 - val_loss: 0.2005 - val_acc: 0.9250\n",
      "Epoch 42/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.1140 - acc: 0.9578 - val_loss: 0.3652 - val_acc: 0.8938\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 43/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1529 - acc: 0.9370 - val_loss: 0.2472 - val_acc: 0.9000\n",
      "Epoch 44/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1239 - acc: 0.9557 - val_loss: 0.1673 - val_acc: 0.9375\n",
      "Epoch 45/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1011 - acc: 0.9608 - val_loss: 0.3699 - val_acc: 0.8875\n",
      "Epoch 46/100\n",
      "24/24 [==============================] - 12s 517ms/step - loss: 0.1281 - acc: 0.9495 - val_loss: 0.3426 - val_acc: 0.8938\n",
      "Epoch 47/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1193 - acc: 0.9502 - val_loss: 0.2789 - val_acc: 0.9000\n",
      "Epoch 48/100\n",
      "24/24 [==============================] - 13s 523ms/step - loss: 0.1061 - acc: 0.9534 - val_loss: 0.2196 - val_acc: 0.9375\n",
      "Epoch 49/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1088 - acc: 0.9559 - val_loss: 0.3102 - val_acc: 0.9062\n",
      "Epoch 50/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.0821 - acc: 0.9689 - val_loss: 0.2677 - val_acc: 0.9062\n",
      "Train loss: 0.0848054240491\n",
      "Train accuracy: 0.971606648199\n",
      "Test loss: 0.149287250638\n",
      "Test accuracy: 0.93125\n",
      "\n",
      "===================FOLD= 4\n",
      "Epoch 1/100\n",
      "24/24 [==============================] - 16s 677ms/step - loss: 0.8245 - acc: 0.5506 - val_loss: 0.4820 - val_acc: 0.7688\n",
      "Epoch 2/100\n",
      "24/24 [==============================] - 15s 605ms/step - loss: 0.4796 - acc: 0.7604 - val_loss: 0.2776 - val_acc: 0.8688\n",
      "Epoch 3/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.4288 - acc: 0.8064 - val_loss: 0.2055 - val_acc: 0.9000\n",
      "Epoch 4/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.3759 - acc: 0.8309 - val_loss: 0.2640 - val_acc: 0.9125\n",
      "Epoch 5/100\n",
      "24/24 [==============================] - 15s 608ms/step - loss: 0.3573 - acc: 0.8331 - val_loss: 0.1784 - val_acc: 0.9437\n",
      "Epoch 6/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.2985 - acc: 0.8697 - val_loss: 0.1448 - val_acc: 0.9563\n",
      "Epoch 7/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2923 - acc: 0.8778 - val_loss: 0.1454 - val_acc: 0.9500\n",
      "Epoch 8/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.3034 - acc: 0.8507 - val_loss: 0.1769 - val_acc: 0.9125\n",
      "Epoch 9/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2842 - acc: 0.8726 - val_loss: 0.1538 - val_acc: 0.9500\n",
      "Epoch 10/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2755 - acc: 0.8796 - val_loss: 0.2515 - val_acc: 0.9062\n",
      "Epoch 11/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2445 - acc: 0.8963 - val_loss: 0.2091 - val_acc: 0.9375\n",
      "Epoch 12/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2596 - acc: 0.8919 - val_loss: 0.1655 - val_acc: 0.9313\n",
      "Epoch 13/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2543 - acc: 0.8963 - val_loss: 0.1962 - val_acc: 0.9375\n",
      "Epoch 14/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.2413 - acc: 0.8967 - val_loss: 0.1491 - val_acc: 0.9500\n",
      "Epoch 15/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.2150 - acc: 0.9059 - val_loss: 0.2332 - val_acc: 0.9375\n",
      "Epoch 16/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2500 - acc: 0.8899 - val_loss: 0.1793 - val_acc: 0.9500\n",
      "Epoch 17/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.2624 - acc: 0.8907 - val_loss: 0.1786 - val_acc: 0.9375\n",
      "Epoch 18/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.2048 - acc: 0.9114 - val_loss: 0.1894 - val_acc: 0.9437\n",
      "Epoch 19/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.2118 - acc: 0.9138 - val_loss: 0.1496 - val_acc: 0.9375\n",
      "Epoch 20/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.2146 - acc: 0.9037 - val_loss: 0.1589 - val_acc: 0.9625\n",
      "Epoch 21/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1954 - acc: 0.9220 - val_loss: 0.2143 - val_acc: 0.9250\n",
      "Train loss: 0.222088955941\n",
      "Train accuracy: 0.910664819945\n",
      "Test loss: 0.144797550142\n",
      "Test accuracy: 0.95625\n",
      "\n",
      "===================FOLD= 5\n",
      "Epoch 1/100\n",
      "24/24 [==============================] - 17s 691ms/step - loss: 0.7047 - acc: 0.5933 - val_loss: 0.4140 - val_acc: 0.7750\n",
      "Epoch 2/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.5264 - acc: 0.7171 - val_loss: 0.3256 - val_acc: 0.8688\n",
      "Epoch 3/100\n",
      "24/24 [==============================] - 15s 605ms/step - loss: 0.3761 - acc: 0.8187 - val_loss: 0.2430 - val_acc: 0.9062\n",
      "Epoch 4/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.3215 - acc: 0.8603 - val_loss: 0.2800 - val_acc: 0.8750\n",
      "Epoch 5/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.3262 - acc: 0.8502 - val_loss: 0.2148 - val_acc: 0.9062\n",
      "Epoch 6/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.2934 - acc: 0.8630 - val_loss: 0.2687 - val_acc: 0.9062\n",
      "Epoch 7/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2436 - acc: 0.8845 - val_loss: 0.2178 - val_acc: 0.9437\n",
      "Epoch 8/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2744 - acc: 0.8842 - val_loss: 0.2418 - val_acc: 0.9187\n",
      "Epoch 9/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2600 - acc: 0.8860 - val_loss: 0.3265 - val_acc: 0.8875\n",
      "Epoch 10/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.2334 - acc: 0.9065 - val_loss: 0.2182 - val_acc: 0.9250\n",
      "Epoch 11/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.2109 - acc: 0.9057 - val_loss: 0.1733 - val_acc: 0.9437\n",
      "Epoch 12/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2219 - acc: 0.9039 - val_loss: 0.1949 - val_acc: 0.9250\n",
      "Epoch 13/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.2015 - acc: 0.9241 - val_loss: 0.2163 - val_acc: 0.9187\n",
      "Epoch 14/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2118 - acc: 0.9096 - val_loss: 0.2486 - val_acc: 0.9375\n",
      "Epoch 15/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2171 - acc: 0.9093 - val_loss: 0.1779 - val_acc: 0.9375\n",
      "Epoch 16/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1880 - acc: 0.9171 - val_loss: 0.1910 - val_acc: 0.9313\n",
      "Epoch 17/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2177 - acc: 0.9148 - val_loss: 0.2424 - val_acc: 0.9125\n",
      "Epoch 18/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2072 - acc: 0.9096 - val_loss: 0.1869 - val_acc: 0.9375\n",
      "Epoch 19/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.1937 - acc: 0.9073 - val_loss: 0.2265 - val_acc: 0.9313\n",
      "Epoch 20/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1822 - acc: 0.9202 - val_loss: 0.1906 - val_acc: 0.9437\n",
      "Epoch 21/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1822 - acc: 0.9171 - val_loss: 0.2543 - val_acc: 0.9250\n",
      "Epoch 22/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1746 - acc: 0.9355 - val_loss: 0.1924 - val_acc: 0.9563\n",
      "Epoch 23/100\n",
      "24/24 [==============================] - 12s 520ms/step - loss: 0.1548 - acc: 0.9351 - val_loss: 0.2132 - val_acc: 0.9063\n",
      "Epoch 24/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2104 - acc: 0.9096 - val_loss: 0.2242 - val_acc: 0.9187\n",
      "Epoch 25/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1614 - acc: 0.9326 - val_loss: 0.1989 - val_acc: 0.9375\n",
      "Epoch 26/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1884 - acc: 0.9142 - val_loss: 0.1918 - val_acc: 0.9563\n",
      "Train loss: 0.161459999195\n",
      "Train accuracy: 0.930747922438\n",
      "Test loss: 0.17331725657\n",
      "Test accuracy: 0.94375\n",
      "\n",
      "===================FOLD= 6\n",
      "Epoch 1/100\n",
      "24/24 [==============================] - 17s 705ms/step - loss: 0.7574 - acc: 0.6588 - val_loss: 0.4672 - val_acc: 0.8000\n",
      "Epoch 2/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.5443 - acc: 0.7656 - val_loss: 0.3690 - val_acc: 0.8187\n",
      "Epoch 3/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.3982 - acc: 0.8349 - val_loss: 0.4190 - val_acc: 0.8187\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 4/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.3551 - acc: 0.8474 - val_loss: 0.4669 - val_acc: 0.7937\n",
      "Epoch 5/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.3248 - acc: 0.8559 - val_loss: 0.3458 - val_acc: 0.8625\n",
      "Epoch 6/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.3079 - acc: 0.8765 - val_loss: 0.3013 - val_acc: 0.8875\n",
      "Epoch 7/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.3021 - acc: 0.8739 - val_loss: 0.3069 - val_acc: 0.8750\n",
      "Epoch 8/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2503 - acc: 0.8914 - val_loss: 0.3226 - val_acc: 0.8875\n",
      "Epoch 9/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2476 - acc: 0.9018 - val_loss: 0.3220 - val_acc: 0.8812\n",
      "Epoch 10/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2330 - acc: 0.8966 - val_loss: 0.3246 - val_acc: 0.8938\n",
      "Epoch 11/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2402 - acc: 0.9008 - val_loss: 0.3241 - val_acc: 0.8875\n",
      "Epoch 12/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2193 - acc: 0.9068 - val_loss: 0.3148 - val_acc: 0.9000\n",
      "Epoch 13/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.2221 - acc: 0.9104 - val_loss: 0.2618 - val_acc: 0.8875\n",
      "Epoch 14/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2143 - acc: 0.9138 - val_loss: 0.3033 - val_acc: 0.9000\n",
      "Epoch 15/100\n",
      "24/24 [==============================] - 15s 608ms/step - loss: 0.2359 - acc: 0.9062 - val_loss: 0.2617 - val_acc: 0.9125\n",
      "Epoch 16/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1992 - acc: 0.9210 - val_loss: 0.3312 - val_acc: 0.8500\n",
      "Epoch 17/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2027 - acc: 0.9130 - val_loss: 0.3142 - val_acc: 0.8375\n",
      "Epoch 18/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.1942 - acc: 0.9213 - val_loss: 0.2605 - val_acc: 0.9062\n",
      "Epoch 19/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2142 - acc: 0.9122 - val_loss: 0.3764 - val_acc: 0.8375\n",
      "Epoch 20/100\n",
      "24/24 [==============================] - 15s 608ms/step - loss: 0.1920 - acc: 0.9173 - val_loss: 0.2575 - val_acc: 0.8812\n",
      "Epoch 21/100\n",
      "24/24 [==============================] - 15s 604ms/step - loss: 0.1625 - acc: 0.9311 - val_loss: 0.2466 - val_acc: 0.9000\n",
      "Epoch 22/100\n",
      "24/24 [==============================] - 15s 604ms/step - loss: 0.1757 - acc: 0.9274 - val_loss: 0.2448 - val_acc: 0.8938\n",
      "Epoch 23/100\n",
      "24/24 [==============================] - 12s 518ms/step - loss: 0.1877 - acc: 0.9185 - val_loss: 0.2723 - val_acc: 0.9000\n",
      "Epoch 24/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1577 - acc: 0.9360 - val_loss: 0.2665 - val_acc: 0.8875\n",
      "Epoch 25/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1528 - acc: 0.9396 - val_loss: 0.2752 - val_acc: 0.9000\n",
      "Epoch 26/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1588 - acc: 0.9379 - val_loss: 0.2685 - val_acc: 0.9000\n",
      "Epoch 27/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.1740 - acc: 0.9282 - val_loss: 0.2931 - val_acc: 0.8750\n",
      "Epoch 28/100\n",
      "24/24 [==============================] - 15s 608ms/step - loss: 0.2084 - acc: 0.9207 - val_loss: 0.2319 - val_acc: 0.8938\n",
      "Epoch 29/100\n",
      "24/24 [==============================] - 13s 523ms/step - loss: 0.1492 - acc: 0.9485 - val_loss: 0.2413 - val_acc: 0.8938\n",
      "Epoch 30/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.1274 - acc: 0.9469 - val_loss: 0.2832 - val_acc: 0.9000\n",
      "Epoch 31/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1467 - acc: 0.9453 - val_loss: 0.3365 - val_acc: 0.9000\n",
      "Epoch 32/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1355 - acc: 0.9435 - val_loss: 0.2604 - val_acc: 0.9000\n",
      "Epoch 33/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1210 - acc: 0.9555 - val_loss: 0.3189 - val_acc: 0.8875\n",
      "Epoch 34/100\n",
      "24/24 [==============================] - 15s 604ms/step - loss: 0.1431 - acc: 0.9445 - val_loss: 0.2314 - val_acc: 0.9062\n",
      "Epoch 35/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.1539 - acc: 0.9355 - val_loss: 0.2584 - val_acc: 0.8938\n",
      "Epoch 36/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1154 - acc: 0.9546 - val_loss: 0.3345 - val_acc: 0.8938\n",
      "Epoch 37/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1344 - acc: 0.9446 - val_loss: 0.2708 - val_acc: 0.8938\n",
      "Epoch 38/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.1150 - acc: 0.9515 - val_loss: 0.3565 - val_acc: 0.8625\n",
      "Epoch 39/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1251 - acc: 0.9533 - val_loss: 0.3126 - val_acc: 0.8812\n",
      "Epoch 40/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1232 - acc: 0.9502 - val_loss: 0.3335 - val_acc: 0.8875\n",
      "Epoch 41/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.1246 - acc: 0.9448 - val_loss: 0.2979 - val_acc: 0.8812\n",
      "Epoch 42/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1317 - acc: 0.9446 - val_loss: 0.3566 - val_acc: 0.8750\n",
      "Epoch 43/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1130 - acc: 0.9591 - val_loss: 0.2977 - val_acc: 0.9125\n",
      "Epoch 44/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.0884 - acc: 0.9637 - val_loss: 0.3498 - val_acc: 0.8812\n",
      "Epoch 45/100\n",
      "24/24 [==============================] - 13s 523ms/step - loss: 0.0885 - acc: 0.9702 - val_loss: 0.3989 - val_acc: 0.8875\n",
      "Epoch 46/100\n",
      "24/24 [==============================] - 12s 520ms/step - loss: 0.1066 - acc: 0.9624 - val_loss: 0.2913 - val_acc: 0.9000\n",
      "Epoch 47/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.0982 - acc: 0.9630 - val_loss: 0.2968 - val_acc: 0.8875\n",
      "Epoch 48/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1226 - acc: 0.9524 - val_loss: 0.2703 - val_acc: 0.8938\n",
      "Epoch 49/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.0977 - acc: 0.9573 - val_loss: 0.4354 - val_acc: 0.8438\n",
      "Train loss: 0.105215750697\n",
      "Train accuracy: 0.957756232687\n",
      "Test loss: 0.231410250068\n",
      "Test accuracy: 0.90625\n",
      "\n",
      "===================FOLD= 7\n",
      "Epoch 1/100\n",
      "24/24 [==============================] - 17s 715ms/step - loss: 0.8013 - acc: 0.6041 - val_loss: 0.3975 - val_acc: 0.7937\n",
      "Epoch 2/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.5161 - acc: 0.7598 - val_loss: 0.2933 - val_acc: 0.8625\n",
      "Epoch 3/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.4737 - acc: 0.7964 - val_loss: 0.3568 - val_acc: 0.8562\n",
      "Epoch 4/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.3947 - acc: 0.8349 - val_loss: 0.3219 - val_acc: 0.8313\n",
      "Epoch 5/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.3376 - acc: 0.8480 - val_loss: 0.2493 - val_acc: 0.8875\n",
      "Epoch 6/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.3706 - acc: 0.8384 - val_loss: 0.2652 - val_acc: 0.8625\n",
      "Epoch 7/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2834 - acc: 0.8783 - val_loss: 0.2862 - val_acc: 0.8875\n",
      "Epoch 8/100\n",
      "24/24 [==============================] - 15s 608ms/step - loss: 0.2974 - acc: 0.8736 - val_loss: 0.2376 - val_acc: 0.8812\n",
      "Epoch 9/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.2699 - acc: 0.8857 - val_loss: 0.2229 - val_acc: 0.9000\n",
      "Epoch 10/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.2478 - acc: 0.8917 - val_loss: 0.2063 - val_acc: 0.9000\n",
      "Epoch 11/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2337 - acc: 0.9094 - val_loss: 0.2563 - val_acc: 0.8750\n",
      "Epoch 12/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.2726 - acc: 0.8906 - val_loss: 0.1919 - val_acc: 0.9187\n",
      "Epoch 13/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.2402 - acc: 0.8963 - val_loss: 0.2332 - val_acc: 0.9062\n",
      "Epoch 14/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2351 - acc: 0.9076 - val_loss: 0.2152 - val_acc: 0.9187\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 15/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1993 - acc: 0.9171 - val_loss: 0.2065 - val_acc: 0.9125\n",
      "Epoch 16/100\n",
      "24/24 [==============================] - 15s 608ms/step - loss: 0.2458 - acc: 0.8870 - val_loss: 0.1849 - val_acc: 0.9187\n",
      "Epoch 17/100\n",
      "24/24 [==============================] - 14s 603ms/step - loss: 0.1629 - acc: 0.9264 - val_loss: 0.1539 - val_acc: 0.9313\n",
      "Epoch 18/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2141 - acc: 0.9114 - val_loss: 0.1694 - val_acc: 0.9250\n",
      "Epoch 19/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2110 - acc: 0.9106 - val_loss: 0.2260 - val_acc: 0.9125\n",
      "Epoch 20/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2148 - acc: 0.9147 - val_loss: 0.1962 - val_acc: 0.9000\n",
      "Epoch 21/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1918 - acc: 0.9145 - val_loss: 0.2079 - val_acc: 0.9000\n",
      "Epoch 22/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2006 - acc: 0.9207 - val_loss: 0.2230 - val_acc: 0.9000\n",
      "Epoch 23/100\n",
      "24/24 [==============================] - 12s 520ms/step - loss: 0.1834 - acc: 0.9240 - val_loss: 0.1653 - val_acc: 0.9250\n",
      "Epoch 24/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1843 - acc: 0.9228 - val_loss: 0.1794 - val_acc: 0.9437\n",
      "Epoch 25/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1709 - acc: 0.9236 - val_loss: 0.2166 - val_acc: 0.8938\n",
      "Epoch 26/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1930 - acc: 0.9254 - val_loss: 0.1921 - val_acc: 0.9062\n",
      "Epoch 27/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1510 - acc: 0.9350 - val_loss: 0.2318 - val_acc: 0.8938\n",
      "Epoch 28/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1846 - acc: 0.9270 - val_loss: 0.1919 - val_acc: 0.9250\n",
      "Epoch 29/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1769 - acc: 0.9262 - val_loss: 0.1855 - val_acc: 0.9125\n",
      "Epoch 30/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1716 - acc: 0.9278 - val_loss: 0.1728 - val_acc: 0.9313\n",
      "Epoch 31/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1385 - acc: 0.9425 - val_loss: 0.1963 - val_acc: 0.9062\n",
      "Epoch 32/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1609 - acc: 0.9407 - val_loss: 0.2614 - val_acc: 0.9000\n",
      "Train loss: 0.152429132725\n",
      "Train accuracy: 0.934210526316\n",
      "Test loss: 0.153939536214\n",
      "Test accuracy: 0.93125\n",
      "\n",
      "===================FOLD= 8\n",
      "Epoch 1/100\n",
      "24/24 [==============================] - 18s 731ms/step - loss: 0.8794 - acc: 0.5915 - val_loss: 0.5174 - val_acc: 0.7312\n",
      "Epoch 2/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.5075 - acc: 0.7772 - val_loss: 0.5214 - val_acc: 0.7688\n",
      "Epoch 3/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.3939 - acc: 0.8337 - val_loss: 0.4298 - val_acc: 0.8000\n",
      "Epoch 4/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.3572 - acc: 0.8494 - val_loss: 0.3910 - val_acc: 0.8375\n",
      "Epoch 5/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.3080 - acc: 0.8770 - val_loss: 0.4409 - val_acc: 0.8063\n",
      "Epoch 6/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2822 - acc: 0.8849 - val_loss: 0.4641 - val_acc: 0.8313\n",
      "Epoch 7/100\n",
      "24/24 [==============================] - 15s 608ms/step - loss: 0.2622 - acc: 0.8949 - val_loss: 0.3772 - val_acc: 0.8250\n",
      "Epoch 8/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2654 - acc: 0.8793 - val_loss: 0.4716 - val_acc: 0.8375\n",
      "Epoch 9/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.2640 - acc: 0.8910 - val_loss: 0.4893 - val_acc: 0.8250\n",
      "Epoch 10/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2361 - acc: 0.9070 - val_loss: 0.4382 - val_acc: 0.8187\n",
      "Epoch 11/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.2699 - acc: 0.8974 - val_loss: 0.3849 - val_acc: 0.8187\n",
      "Epoch 12/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.2353 - acc: 0.9047 - val_loss: 0.3397 - val_acc: 0.8438\n",
      "Epoch 13/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2140 - acc: 0.9267 - val_loss: 0.3575 - val_acc: 0.8438\n",
      "Epoch 14/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2111 - acc: 0.9153 - val_loss: 0.3551 - val_acc: 0.8562\n",
      "Epoch 15/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1829 - acc: 0.9202 - val_loss: 0.4123 - val_acc: 0.8625\n",
      "Epoch 16/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.1993 - acc: 0.9264 - val_loss: 0.5030 - val_acc: 0.8313\n",
      "Epoch 17/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2386 - acc: 0.9033 - val_loss: 0.3906 - val_acc: 0.8688\n",
      "Epoch 18/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2457 - acc: 0.9094 - val_loss: 0.3674 - val_acc: 0.8313\n",
      "Epoch 19/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1942 - acc: 0.9262 - val_loss: 0.4014 - val_acc: 0.8187\n",
      "Epoch 20/100\n",
      "24/24 [==============================] - 15s 608ms/step - loss: 0.1966 - acc: 0.9259 - val_loss: 0.3313 - val_acc: 0.8562\n",
      "Epoch 21/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.1850 - acc: 0.9261 - val_loss: 0.2827 - val_acc: 0.8500\n",
      "Epoch 22/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1628 - acc: 0.9296 - val_loss: 0.3134 - val_acc: 0.8562\n",
      "Epoch 23/100\n",
      "24/24 [==============================] - 12s 520ms/step - loss: 0.1534 - acc: 0.9371 - val_loss: 0.2995 - val_acc: 0.8813\n",
      "Epoch 24/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1646 - acc: 0.9347 - val_loss: 0.3062 - val_acc: 0.8625\n",
      "Epoch 25/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1728 - acc: 0.9339 - val_loss: 0.3374 - val_acc: 0.8438\n",
      "Epoch 26/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1404 - acc: 0.9521 - val_loss: 0.3548 - val_acc: 0.8625\n",
      "Epoch 27/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1623 - acc: 0.9432 - val_loss: 0.4824 - val_acc: 0.8375\n",
      "Epoch 28/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1652 - acc: 0.9272 - val_loss: 0.3652 - val_acc: 0.8688\n",
      "Epoch 29/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1499 - acc: 0.9423 - val_loss: 0.3366 - val_acc: 0.8750\n",
      "Epoch 30/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1379 - acc: 0.9427 - val_loss: 0.3395 - val_acc: 0.8750\n",
      "Epoch 31/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1590 - acc: 0.9450 - val_loss: 0.3372 - val_acc: 0.8438\n",
      "Epoch 32/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1298 - acc: 0.9500 - val_loss: 0.5790 - val_acc: 0.8562\n",
      "Epoch 33/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1226 - acc: 0.9505 - val_loss: 0.3785 - val_acc: 0.8500\n",
      "Epoch 34/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.1362 - acc: 0.9539 - val_loss: 0.3540 - val_acc: 0.8875\n",
      "Epoch 35/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.1382 - acc: 0.9432 - val_loss: 0.4359 - val_acc: 0.8562\n",
      "Epoch 36/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1494 - acc: 0.9349 - val_loss: 0.3314 - val_acc: 0.8750\n",
      "Train loss: 0.126577656138\n",
      "Train accuracy: 0.954986149584\n",
      "Test loss: 0.282740962505\n",
      "Test accuracy: 0.85\n",
      "\n",
      "===================FOLD= 9\n",
      "Epoch 1/100\n",
      "24/24 [==============================] - 18s 744ms/step - loss: 0.7262 - acc: 0.5401 - val_loss: 0.6366 - val_acc: 0.5312\n",
      "Epoch 2/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.5600 - acc: 0.6712 - val_loss: 0.5180 - val_acc: 0.7000\n",
      "Epoch 3/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.4360 - acc: 0.7847 - val_loss: 0.3146 - val_acc: 0.8375\n",
      "Epoch 4/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.3587 - acc: 0.8363 - val_loss: 0.2637 - val_acc: 0.8875\n",
      "Epoch 5/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.3164 - acc: 0.8479 - val_loss: 0.2535 - val_acc: 0.8750\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 6/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.2682 - acc: 0.8819 - val_loss: 0.2132 - val_acc: 0.9250\n",
      "Epoch 7/100\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.2642 - acc: 0.8901 - val_loss: 0.1874 - val_acc: 0.9000\n",
      "Epoch 8/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2391 - acc: 0.8998 - val_loss: 0.2151 - val_acc: 0.9250\n",
      "Epoch 9/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2434 - acc: 0.8922 - val_loss: 0.1970 - val_acc: 0.9187\n",
      "Epoch 10/100\n",
      "24/24 [==============================] - 15s 607ms/step - loss: 0.2404 - acc: 0.8971 - val_loss: 0.1866 - val_acc: 0.9125\n",
      "Epoch 11/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.2048 - acc: 0.9158 - val_loss: 0.2327 - val_acc: 0.9000\n",
      "Epoch 12/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2397 - acc: 0.8984 - val_loss: 0.1916 - val_acc: 0.9187\n",
      "Epoch 13/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2494 - acc: 0.8915 - val_loss: 0.1912 - val_acc: 0.9313\n",
      "Epoch 14/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1990 - acc: 0.9179 - val_loss: 0.2505 - val_acc: 0.8750\n",
      "Epoch 15/100\n",
      "24/24 [==============================] - 13s 529ms/step - loss: 0.2271 - acc: 0.9086 - val_loss: 0.2420 - val_acc: 0.8812\n",
      "Epoch 16/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2110 - acc: 0.9158 - val_loss: 0.2228 - val_acc: 0.9187\n",
      "Epoch 17/100\n",
      "24/24 [==============================] - 13s 529ms/step - loss: 0.2060 - acc: 0.9184 - val_loss: 0.2042 - val_acc: 0.9187\n",
      "Epoch 18/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1831 - acc: 0.9226 - val_loss: 0.3407 - val_acc: 0.8688\n",
      "Epoch 19/100\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.1867 - acc: 0.9238 - val_loss: 0.2382 - val_acc: 0.9187\n",
      "Epoch 20/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1972 - acc: 0.9202 - val_loss: 0.2150 - val_acc: 0.9187\n",
      "Epoch 21/100\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2452 - acc: 0.8945 - val_loss: 0.2058 - val_acc: 0.9062\n",
      "Epoch 22/100\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.1719 - acc: 0.9306 - val_loss: 0.1895 - val_acc: 0.9125\n",
      "Epoch 23/100\n",
      "24/24 [==============================] - 12s 521ms/step - loss: 0.1601 - acc: 0.9338 - val_loss: 0.1922 - val_acc: 0.9313\n",
      "Epoch 24/100\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1738 - acc: 0.9257 - val_loss: 0.2244 - val_acc: 0.9062\n",
      "Epoch 25/100\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1645 - acc: 0.9375 - val_loss: 0.2301 - val_acc: 0.9125\n",
      "Train loss: 0.188236133007\n",
      "Train accuracy: 0.926592797784\n",
      "Test loss: 0.186617267132\n",
      "Test accuracy: 0.9125\n",
      "\n",
      " Train Log Loss Validation=  0.115472399724\n",
      " Test Log Loss Validation=  0.179352245512\n"
     ]
    }
   ],
   "source": [
    "#def myAngleCV(X_train, X_angle, X_test):\n",
    "K=10\n",
    "folds = list(StratifiedKFold(n_splits=K, shuffle=True, random_state=16).split(X_train, target_train))\n",
    "y_test_pred_log = 0\n",
    "y_train_pred_log=0\n",
    "y_valid_pred_log = 0.0*target_train\n",
    "for j, (train_idx, test_idx) in enumerate(folds):\n",
    "    print('\\n===================FOLD=',j)\n",
    "    X_train_cv = X_train[train_idx]\n",
    "    y_train_cv = target_train[train_idx]\n",
    "    X_holdout = X_train[test_idx]\n",
    "    Y_holdout= target_train[test_idx]\n",
    "    \n",
    "    #Angle\n",
    "    X_angle_cv=X_angle[train_idx]\n",
    "    X_angle_hold=X_angle[test_idx]\n",
    "    #define file path and get callbacks\n",
    "    file_path = \"%s_aug_model_weights.hdf5\"%j\n",
    "    callbacks = get_callbacks(filepath=file_path, patience=5)\n",
    "    gen_flow = gen_flow_for_two_inputs(X_train_cv, X_angle_cv, y_train_cv)\n",
    "    galaxyModel= getVggAngleModel()\n",
    "    histr = galaxyModel.fit_generator(\n",
    "            gen_flow,\n",
    "            steps_per_epoch=24,\n",
    "            epochs=100,\n",
    "            shuffle=True,\n",
    "            verbose=1,\n",
    "            validation_data=([X_holdout,X_angle_hold], Y_holdout),\n",
    "            callbacks=callbacks)\n",
    "     #Getting the Best Model\n",
    "    galaxyModel.load_weights(filepath=file_path)\n",
    "    #Getting Training Score\n",
    "    score = galaxyModel.evaluate([X_train_cv,X_angle_cv], y_train_cv, verbose=0)\n",
    "    print('Train loss:', score[0])\n",
    "    print('Train accuracy:', score[1])\n",
    "    #Getting Test Score\n",
    "    score = galaxyModel.evaluate([X_holdout,X_angle_hold], Y_holdout, verbose=0)\n",
    "    print('Test loss:', score[0])\n",
    "    print('Test accuracy:', score[1])\n",
    "\n",
    "    #Getting validation Score.\n",
    "    pred_valid=galaxyModel.predict([X_holdout,X_angle_hold])\n",
    "    y_valid_pred_log[test_idx] = pred_valid.reshape(pred_valid.shape[0])\n",
    "\n",
    "    #Getting Test Scores\n",
    "    temp_test=galaxyModel.predict([X_test, X_test_angle])\n",
    "    y_test_pred_log+=temp_test.reshape(temp_test.shape[0])\n",
    "\n",
    "    #Getting Train Scores\n",
    "    temp_train=galaxyModel.predict([X_train, X_angle])\n",
    "    y_train_pred_log+=temp_train.reshape(temp_train.shape[0])\n",
    "\n",
    "y_test_pred_log=y_test_pred_log/K\n",
    "y_train_pred_log=y_train_pred_log/K\n",
    "\n",
    "print('\\n Train Log Loss Validation= ',log_loss(target_train, y_train_pred_log))\n",
    "print(' Test Log Loss Validation= ',log_loss(target_train, y_valid_pred_log))\n",
    "#return y_test_pred_log"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "#preds=myAngleCV(X_train, X_angle, X_test)\n",
    "preds = y_test_pred_log"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Submission to kaggle\n",
    "submission = pd.DataFrame()\n",
    "submission['id']=test['id']\n",
    "submission['is_iceberg']=preds\n",
    "submission.to_csv('sub_Adam_10folds_amsgrad_0001_15p_4nd.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['loss', 'val_acc', 'val_loss', 'acc'])\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYwAAAEWCAYAAAB1xKBvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f86d4035668>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEWCAYAAACJ0YulAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f85f8ad9240>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# list all data in history\n",
    "print(histr.history.keys())\n",
    "# summarize history for accuracy\n",
    "plt.plot(histr.history['acc'])\n",
    "plt.plot(histr.history['val_acc'])\n",
    "plt.title('model accuracy')\n",
    "plt.ylabel('accuracy')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'test'], loc='upper left')\n",
    "plt.show()\n",
    "# summarize history for loss\n",
    "plt.plot(histr.history['loss'])\n",
    "plt.plot(histr.history['val_loss'])\n",
    "plt.title('model loss')\n",
    "plt.ylabel('loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'test'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.image as mpimg\n",
    "\n",
    "min1 = np.min(X_holdout)\n",
    "max1 = np.max(X_holdout)\n",
    "diff1 = max1 - min1\n",
    "my_sum = lambda x: (x - min1)/diff1\n",
    "\n",
    "X_holdout2 = my_sum(X_holdout)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "13      1\n",
       "14      0\n",
       "38      1\n",
       "43      0\n",
       "44      0\n",
       "80      0\n",
       "104     1\n",
       "127     0\n",
       "137     0\n",
       "138     1\n",
       "141     1\n",
       "167     1\n",
       "195     1\n",
       "199     0\n",
       "207     1\n",
       "212     0\n",
       "213     1\n",
       "221     0\n",
       "240     0\n",
       "256     0\n",
       "267     1\n",
       "274     1\n",
       "276     1\n",
       "285     1\n",
       "290     0\n",
       "294     0\n",
       "300     1\n",
       "307     1\n",
       "311     1\n",
       "318     1\n",
       "       ..\n",
       "1363    1\n",
       "1364    0\n",
       "1373    1\n",
       "1380    1\n",
       "1394    0\n",
       "1411    0\n",
       "1435    0\n",
       "1440    1\n",
       "1445    1\n",
       "1447    1\n",
       "1460    1\n",
       "1470    0\n",
       "1488    0\n",
       "1495    0\n",
       "1504    1\n",
       "1516    0\n",
       "1525    0\n",
       "1533    0\n",
       "1537    0\n",
       "1539    0\n",
       "1541    0\n",
       "1542    0\n",
       "1549    0\n",
       "1552    0\n",
       "1559    0\n",
       "1566    0\n",
       "1567    0\n",
       "1577    0\n",
       "1590    0\n",
       "1602    0\n",
       "Name: is_iceberg, Length: 160, dtype: int64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_holdout"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7f85f898c6a0>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUUAAAEICAYAAADIsubvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f865c5509e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_holdout2[0], cmap='gray', interpolation='nearest');\n",
    "tl = \"Actual label : \" +str(Y_holdout[13])+ \",\"+\" iceberg_probability : \"+str(pred_valid[0])\n",
    "plt.title(tl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7f85f8970940>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUQAAAEICAYAAAAncI3RAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f85f89468d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_holdout2[1], cmap='gray', interpolation='nearest');\n",
    "tl = \"Actual label : \" +str(Y_holdout[14])+ \",\"+\" iceberg_probability : \"+str(pred_valid[1])\n",
    "plt.title(tl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7f85f88cdf28>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUUAAAEICAYAAADIsubvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzsvXm0bdtd1/n9rbX75nS3e23yEsAGZNCoKCoSCYJgkyhU\nhkrBCwZTtoAdCZSUiBSVwgYZQwVfEeApYIhRDIVCGYGUppTog0Q0DZDmxbz23ndPu/u915r1x5r7\n/D7n3H3uu++muefuzO8Yd9x11llrdmueteZ3/n6/789CCEpISEhIqJDd6QYkJCQknCekl2JCQkIC\nkF6KCQkJCUB6KSYkJCQA6aWYkJCQAKSXYkJCQgJw17wUzexlZvbETX4fzOzTb6Gch+K1tdtow23f\n+7HCzH7GzB7+GMv4DjP70Y9Xm84LbvXZn3Hv42b2pWf87ovM7FdXXWtm32ZmP3h7LX7BbfwRM5uZ\n2eOfjPrWEWb2wTiGzzv/b/mlaGZvN7M9M2ve4vV37AVyXmFmf8vM/puZLczsO17IvSGErwghPPoJ\nalrCCoQQ/kMI4Tee8bvvDiF8g/RJm+vfE0J46KxfmtmOmf2kmQ3N7CNm9idvcu13mNnczAb499L4\nuy86dX4Q+/ZVuP+lZvbTZnZkZs+Z2ffE800ze2Os/8jM3m1mX3Gq7peb2fvNbGRmv2BmL8bv/o6Z\n/Xq89/1m9nWn7n3EzH7VzEoze/Wp373azIpT7X7Z8vchhE+T9N03H+IKt/RSNLOHJH2RpCDpj9zK\nPQkr8QFJ3yLpX9/phnys+GR97NJH9ZbxDyXNJF2R9DWSvt/MPusm1/9ECKGHfx+Sjj8Ex+cl/SFJ\nA0k/K0lm1pD0Nkk/L+keSQ9IWq6+apI+KumLJW1K+uuS3hzfHzKzi5L+paRvl7Qj6TFJP4E2DSX9\n4Xjvw5K+z8x+F37/XyX9OUm/fEaf/tOpPr39Jv0/E7e6Uvw6Sb8o6UdiY49hZm0z+7vx63BgZu8w\ns7akfx8v2Y9v7S88Td9Of2HN7OvN7H3xS/EhM/tfbqdTZvYHzexdZnZoZh89Y1X2p8zsKTN72sz+\nKu7NzOz1cbl93czebGY7t9OO0wghPBpC+BlJRy/03rhS/wb8/KcxVu81s8+P5+8zs39hZtfM7MNm\n9o2nimqZ2U/E+37ZzD4HZZ55b3x2bzGzHzWzQ0mvjs/+UasYxPvM7FvsJlscKOtxM/vW2O49M/th\nM2vF373MzJ4ws9eZ2TOSfhj9/YCZ7ZrZT5nZfaeK/co4Z54zs79tZlm879PM7Ofjs3zOzH7MzLZO\n3fvbb9aWM/rAuXx6rn9xbOdn4/rLVq2OLj3f+LxQmFlX0ldJ+vYQwiCE8A5JPyXpaz8OxT8s6S0h\nhGH8+dWSngoh/L0QwjCEMAkh/IokxZ+/I4TweAihDCH8tKQPS/qt8d4/Juk9IYR/HkKYSPoOSZ9j\nZr8p3v83Qgjvj/e+U9J/kPSFy4aEEP5hCOHnJE0+Dv06Ey/kpfhj8d+Xm9kV/O7vqOr071L19v8W\nSaWk3xt/vxXf2v/pFuq5qurLtCHp6yV97/KP/QViGNu8JekPSvqzZvbKU9f8PkmfIenLJL3OfF/p\nL0p6paqv3X2S9lR9hZ8XZvaPzOwf3UZ7XxDM7H9SNaG+TtVY/RFJ1+OL4P9W9UW9X9LLJX2zmX05\nbn+FpH+u6ln9uKR/ZWb1F3DvW1SN649J+huSHpL0Ukm/X9L//AK68TWSvlzSp0n6DapWFUvcE9v3\nYkmvNbMvkfR/SHqVpHslfUTSm06V90cl/TZJnx/b+afieYv33ifpN0t6UNXY3WpbbgWn5/r/G9vH\n8fgTkn4uhHDt9M1m9iIz2zezF73Aepf4DZIWIYRfw7n/KulmK8U/HF/c7zGzP7vqgviy/WpJ3Lb5\nnZIet2qP+7n4sf7sM+6/Etv2nnjqs2K7JFUvUUkfXNXOuLD67bj3VvB5sU2/ZmbfbrfLMkIIN/0n\n6fdImku6GH9+v6S/FI8zSWNJn7PivodU0e0azn2HpB+92TWnyvhXkr4pHr9M0hM3aWeQ9Oln/O7v\nS/reU3X+Jvz+eyS9MR6/T9LL8bt7Y/9rz9feW/2nim58xwu85+2SviEe/z/LcTl1ze+Q9D9OnftW\nST+M8f9F/C6T9LSqrZFbufffn/r9hyR9OX7+hps9I1z3uKQ/g5+/UtIH8Zxnklr4/RtV7aktf+7F\nZ/IQnv0fwO//nKoX0Kq6XynpXS+gLU+cuvZLT8/lVfNiOZ6SLP78mKRX3eZ8+RFJ33WT33+RpGdO\nnfvTkt5+xvWfqeojkatazDwt6U+suO5rVa30DOf+bRz7r5DUkPTX4jxonLq3LunfSfrHp57jG05d\n9/9JevWKuh9VRdltxe/ecfoeVR/ml8Q5/dmS3ivpW09dc/zMbvbvVlaKD0v6tyGE5+LPPy6n0Bcl\ntVS97T9mmNlXmNkvxi/YvqoJevE2yvkdVm3iXjOzA0l/ZkU5H8XxR1RNEqlanfxk/HLvq3pJFqr2\nas4LHtTqMX+xpPuWbY/t/zadbPtxv0MIpaQnVPX9Bd0bcd+pc6d/fzOcNf6SdC1U9Ir1fATtHki6\nrmpFe9PyzOyKmb3JzJ6MtP9Hdetz4bYRKvo3kvSySA8/XRWl/URgoIoxEBs6Y5smhPDeEMJTIYQi\nhPAfJX2fqhXhaTws6Z+E+EaJGEt6RwjhZ0IIM1VM8YKqVbikagtK0j9V9XH7Cy+0nWb2tyX9FlUf\nkVtSrAkhfCiE8OFQUe//Juk7z+jT8+KmL8W4hH2VpC82s2fiHs9fUrUP8DmSnlPF7z9tVTtXnBtK\n6uDne1BXU9K/UDXIV0IIW5L+jSr680Lx46om4IMhhE1JP7CinAdx/CJJT8Xjj0r6ihDCFv61QghP\n3kY7PlH4qFaP+UclffhU2/shhK/ENcf9jpP3AVV9v5V7Tz/Tp+P9N5R9Czhr/FfV85Sql/ay3V1V\nf4h8JmeV992xvM8OIWyoorS3OhduFWf94T4a6/taVftyn6i9sF+TVDOzz8C5z9GtU8+gU2NiZg+q\nWin/k1PX/orO7q/MzFStCK9I+qoQwhy/fk9s1/Larqp5/B6c+5uqVqFfFkI4vMX2r8INfbpVPN9K\n8ZWqVkmfKelz47/frGoD9OviSuOHJP29uEmfW2VQaUq6pmpv8aUo792Sfm/cQ9lURc+WaEha3rew\nypT/ZbfTKUl9SbshhImZfYGkVe4J325mHassdF8vt4L9gKT/3aKrgJldMrNX3GY7TiDu3bVUjXvN\nzFpmlsffLY1OD91CUT8o6a+a2W+1Cp8e2/ufJR1FI0U7Po/fYma/Hff+VjP7Y3G/5ZslTVUZ0W7l\n3tN4s6RvNbNtM7tfJ1cFz4c/b2YPWGXE+l910gp5Gv9M0teb2efGufXdkt4ZQngc1/y12I4HJX0T\nyuurWqEcxDb+tY+xLauwaq5L1ar0j6p6MZ5+uXzcEKq9uX8p6TvNrGtmv1vVvuo/XXW9mb0ijpXF\nv49vlPTWU5d9raT/GEI4zUh+VNLvNLMvjXP3m1Utjt4Xf//9qt4RfziEMD51709K+i1m9lXx7+B/\nk/QrIYT3x3Z9q6q/1S8NIVxf0e5GvM8k1ePfz9Kg9hVxD1NxZf7tK/p0a3ievYyflfR3V5x/laRn\nVO2ztVXt2T0p6UCVJa4dr/tOVRNmX9LvjOf+Yfz5A6r2PY73YiT9eUnPxt//U1Wb1d8VVuzvrGjT\n8Z6iqmXzR1Qty39a0j/Qjfs/r1W1InhG0regnEzSX5b0q/H+D0r67lP3nrUH+gOSfuAmbfyReD//\nvTr4vtDjkupn3Pt2xT3F+POfiW0cSPrvkj4vnr9P1UvkGVVGol/UyX2wt6j6oz+S9C5Jn48yn+/e\nHz3Vpm58Tstthr+uuB/3PPPqcVUfxPfGex+V1LnZc479/aCk3fhMHzj17L9R1d7WdUl/V1Ief/dZ\nkn4pjtO7Jf0V3bhPeEtt0Rl7imfN9Xj+38X7btgbwzUviu170U3mzZl7ivGaHVV78ENVe5l/Er/7\nIkkD/PzP4jgNVNkIvnFFee+X9Joz6vpjqv5+D1XNy8+K518cn8Uklr389zW490tj2eN470OnnuP0\n1L3fdupv4PTfz8vi7/6OqnfHMM6D79Spv6XTz+ysf8tN4IQ7DDP766r20v7xnW7L7SJaMf94COGL\nn+e6x1W94P/dJ6VhdxBm9kOqXFheqEWbZfxfqqzXz4bKCTnhBcKqyKT7Jb05hPCnbnpteikm3C7M\n7F5VlPE/qXJv+teS/kEI4e8/z32P61PgpRi3Qt6tahX/4TvbmoRbxV0T+5xwLtGQ9I9VUfGfV7WH\n84/invHpULHlv9v1xburYGZ/S9W2xt9OL8S7C2mlmJCQkADcNStFM/sDVgWDf8DMXn+n25OQkLCe\nuCtWitH0/2uqQsmekPRfVHngv/esexr1Rug0WzecL9HfWubfhCy4S1OxIjgoK080aOVxaV52UfoN\nOdylyvxEY/yYh0W8N/P7AsqLXgjxBz/MddZ5tpHVhxsut4XXU6J+1fyY9Qc2vMAg5atdxIpFcXyc\nLfx8VsuWha+8zzgWqKdA/SWOc1v9vbfCrwnsXxwLy1ffx+ePqaIi8LmwPNTJcnTjfDlxruREQxmc\nqzg+UX9x4/OUpMA+4zkez1F0Oc98ghbZ6neDxdsmo7Fms9lt+QKeZ9wtCiRfIOkDISp5mNmbVPlh\nnflS7DRb+j2f+wU3nJ9OpsfHO113rm/PfWYc7eQ6jS49rpoYthrua/hf+eF4eHy8UTa8/h5eKFO/\nPsz8ZTHfH1Xnul5PMfR2N5qu3lY0fE5uh/bxMV++G7l/HGbeFI0n7ktciy8Ru+7n5m38se54Gax/\nGrwPtu/3lhuoCDjaPTg+7l73e9s7/XhQ94vx0s6afn4x8HoO4A89k4/hZs3Hgqgdui/xtOuDlE2q\ne+vbnRvukaT20Nsyq/nx/mJ0fNzEuNjMX1b14M+8mXk/pnlVZwPnxiMvj2i0ffxbLa/nqMS8OPDx\nrOHNvTjCGG35nDqaVZM6NH0cNju94+O9Fr5aQHNc9f8/v+NW5AzuPtwt9Pl+nQzFekInQ7wkSWb2\nWjN7zMwemy3mp3+dkJCQ8Ly4W1aKt4QQwiOSHpGk7Y3NUG9XX9RJ6S/I3o6rgLUGoGRYWbWfqr6g\nwxf7quHAfBXSqZ/1LfHypnVfTUyO/GueHXlF3bp//Y/kS1HrV+1uHHm7m5tQu2p4GbXC69ydDY6P\nN+UrpafqvvrozfzegXn526PqfGcDK4XmzKucYGtg4f0ZmR8L8sOLKerESvVC1vXLd7CamVRtKbnA\nA5XN5qCVPV+FlvLznMwHB17/VuHXZ/c6O2iM0PZedfd8iJUytkOyts+FcuErqEa4kVVI0hRjcWIb\nou7X93TjajaQDs+9/qb5anLSwriVoNLYG8mwUm0Y+o+5Wytjn8H1ZxN/5hsNf27casji39MZ7Pqu\nx92yUnxSJ+NTH9DJuNeEhISEjwvulpfif5H0GWb2EquUf/+4PnGKIwkJCZ/CuCvocwhhYWZ/QZWO\nYC7ph0IIN1UACWYKS6pQd/owwWa8Nf2b0JXTk9qDFVXtjJ1KVNVWOCqcmtFyOMicjrYNm/ibMDqM\nvMzFkVPmbtf51jyLG/CbN1rPJWk0goGk6RRss9v3tsy87Hrpj5m0vkTbF3l1zXgOwwkoGA1K2Z7X\nv933dpNNlVMfi3kTBqUcdLPADZ2q/sNDF0bJpzBW9L2f+YGf797jtLYgHS79eWY9GKZAD+vbTuUX\nRdWY8sAb1c29DNJaYrlFI520Su80vF3zKeYRjGvL8V/WLUmzQxhuut7n6dzLsBGMToX3uY4G1Drg\n7zAMidS/VpVTh8WbxqIh2w20GrH+MzwF7nbcFS9FSQoh/BtVUmIJCQkJnzDcLfQ5ISEh4ZOCu2al\n+LHArjuVHDlj0hyW23rTqacWFW2gpY6O0RvXnVY8swnXH9DBmZymFHWvZwsO5WBnmgydNlmr+sU+\n8vPs1Nwq3AcFLWlN9OK0OffCi6k3bLzj5zczL3NJvRbyvsH4qRHoeK3p/Zld2z0+3rrklv0sP+Gl\n7vXXvS1TjMuwrOrdDE7fZlOvs8BWR73p2xH1Gaz8Ha/TQBPnh26Vr+fYSsCuxtJaS+ftCbYawoK0\n2m9cgKW2zrBEt4LXuYDJdlZWbaTTdYOU+Rn36cx3nI4b/Ffr+6sp7vQyrim97TU4Zy8p9mK22n2t\nv0C74Q/bC1X/89vTcD33SCvFhISEBCC9FBMSEhKAtaXPWSm1RhUtqdWc44xGHn6nhlPJQ4SLLSlG\nGMHpu4nwry2nT/Pd/ePjna1tvwYOswXYCSm5YPUrOv4oLDp792HxHpVOAYk2rIV0DJ6XTh8nNGLD\n+p1twaIb2zhCJJAhzO4CUuscwNG8f9n7PBmCYtfRH1hF6xiLWQf0KxpRDzew1UAPaLDE/eBj0QTd\n3ph4/5+D5bSz5Xsm2QRW5IGP0dVG9fw3W35tHdZvWoivT53W9pqbx8eGUM3WAs+CSw/Eu4+jRZlW\n5vEMzuOXfX5eqeEhjkHrt3wLpBi7JXr+jOeCmmNLYI4Q1U6rOs5qiHc+8GfImOkunn+IW0N3g27C\n7SCtFBMSEhKA9FJMSEhIANaWPpcqNSkqehJa3k2qrdQaiAlF3OhomZURjKWYOO1u1/y+zr2IpZ7B\nnAwM5ZxxFEDrQIM2oKRytVbRsBqUURpjpziLJ52yTy95uw/L1VbE7tD73wAln7Xg7Hy1ok05FGgW\nV5wyDdHuXtsp5hAW4iYcmUtYa1sTb+M4hzLO1NvSb1c0tJxBmWXLx6e18PGxFrYv5t7nvRLO8Jlf\nY2NvC9WQDDHU90ZXgBMjCPauERRo6v6cy333GjiAVbgmL3tv7HS/1fexszjnZjN/Djms9nnHafWz\ne3vHx52ul1Hr+DPKQPGbDVjZBz4uGZzwD46iozzGoXURWyrPet8YE92J1nRb7c9+1yOtFBMSEhKA\ntV0pqpTCuPq6T/tYkZQIp4PySgP+buOl+ilCsjZbvqGeYwXXxl7zfu5f1hoWPNbGqgXBcMM+ykfa\n722rvv51+DRmMBwU2Fzn0qbE4ywCxFz7Xv8Ym/Ha9RVMvlGtEIo2/AGhwJJhNbE7gaEDK8+Dwlck\nF2tsozeyoEYk/BezFeKqGQwdJZSGBP/B4qK3a4SlS6/hK54AA5SNVhsHslrVrr55+xja96EOni3m\nSmfi41zH+FOLsMQ46hD6hzFEkqKxS39N6aQOY3vb/WgnBSbXAAY4GOs6CO3MwTLG0L/Moj8sRyRg\npR62/G+lZgx5jOWdIch7t2M9e5WQkJBwm0gvxYSEhARgbemz1TPV7qkERfN9+M9tO63K4EA4QUhd\nEcVFGbRFWtMbg+5AGadAaNkUYVGbc4SFrbbFqGjg+xRp2xSKKczdUUOoWNnxAusoo4QyCsVoZguQ\npTHSJ0Ta3MVG/xT0FkGQaoHWkwJvwDIxpsEE/ogMo6OI6yLS+uYZKT9mF2jE8bKhg6vtHEYCqBRR\nVH+cw+hSR5lRsWjBwQLub8PokYFKQ73nRNqBBWcP0lHk8AN9rqK+/XsvHJ/r0UAy9jnZyb2tTE3Q\n9ku0GECZCM8/px8k0icsp1FtgHxCTMsDn84cIZnhQuzPmi6p1rRbCQkJCbeH9FJMSEhIANaWPpch\naBAqSrb9gIeizaEIgigmNeCzqLz6xRy+eTlCAoegYORmsxqsqSDfc1DM5i58FidO68db7m/WXFRU\nZWkdlKSjhV+7C//CFyPMjc0yKKOEuR/3EOU4uux1dmKzjmrI27IH/0LkeTGEqm3AZ3GWw5qPjHNz\nZFAUBGqbCDlb3mlQmgmbzFrn/W/vgTMjd0hrw69f1LzPBSzEEH5R4xApGmO12QacU49g8QcbvjSA\n3ycsxxNOBmTIY1rRRgOUPVr8wzW3IJdQA2pCZHgBS3gdnSixlZIhE2IDc3uK/EJHUx//zSjxUw/w\nu0Q/Qw8W5wyKOder9obFejoqppViQkJCApBeigkJCQnA2tJny+yYEjPRfAuWO+a0GI/gyBwtbTM4\nw5IZBYRKMRl6jvCzEnUyPSdzegj0eTJ3itmaxHpB+zY23ILYKaDug1SiXeRrYdiigfqo78dNCI4O\no/jr/QOEjaGpAS6+IyRgH9W9bx04+I7mTtOojJMjRHKUef+Wz2rY83ZDy1Z9WJYP+15gC3VOQQ2H\ncHC+lLsj+QK2+Nmm39uMXWK+8Lzlz7mBrYnrm96wiy5Goyasz6TMzZq3fQbn+dyq60dbsJSDDo/N\n58cOtkmuy595u+Z9mCNHTQ3Pv465u1mH2lOcowuMVW3u95X9DJdiXJbbJ0klJyEhIWH9ca5eimb2\nQ2Z21cz+O87tmNnbzOzX4//bNysjISEh4WPBeaPPPyLpH0j6Jzj3ekk/F0J4g5m9Pv78uucryILH\nH+ewxE7GzndqcFQO27A6DiINGbj1cbKBnBeIpe3KqRHTp7Zs9dDOIP66uOR19uZ+/SSakcMElAWG\n0jaccZugo0fXXT0ngyNvA9b3Gr6DE8RHd6NFk4K0CPfVBCk2wwYEdGHNn+ZOp2oTP55uYLymyHXT\nQj6SKFCbw1BdwIRaFohr7jKuGvlHIGw7nHkw+UHwQjdqbi2fz7xP4xjnXkOcOmO/5+jbzhGc5M/w\n9mY8M0Vc1aaSTnVNC/NwBkWbGlRvhluIiT6EkzZSnAqO/AUs63Wo7fRA9w8GlStCccXj+ktYlHME\nDDTqXsZgWI1tqUSfP+EIIfx7SbunTr9C0qPx+FFJr/ykNiohIeFTCufqpXgGroQQno7Hz0i6ctaF\nZvZaM3vMzB6bzlZnOUtISEi4Gc4bfb4pQgjBzM5cs4cQHpH0iCRd2N4JSzHY+Z57LPMrMId1jYKZ\nR9Fct5M51eojBepeoAOy880mqRyox0d3n1nZ3oub2B5FTHQ7yltNSbsQ18z43Rqsjycchg1OzYXf\ny5DcPgKx59HUWx97u3cPfNHOPCIawjG6QWuxTyc6snfgHDzr+/UL+Ms3Ytz0DI7GNlgtzquFj/+F\nDR/D+tSnxoP5lpeDcTxCKlkhnncQ91ruGXlbFzA5T9BPNW50ZJakLuKTNUbbz6DY4+ioTXHYWhci\nyPjzrEEoF+HOJ6S+coxXRrk6eD/kPW/jTu+SpJMCwiWc4duYN9Ox/w3Ve9VcsOxuWFO9cNwNvXrW\nzO6VpPj/1TvcnoSEhDXG3fBS/ClJD8fjhyW99Q62JSEhYc1xruizmf0zSS+TdNHMnpD0NyS9QdKb\nzew1kj4i6VW3UtYiFLoeKqq03YPUFXp8Im4UpLy/dHCFxbMYuJWvvQkpsLFTD+YiofDY/TW37j1d\nuvkvY6wy4qM/2Kko2f1zp+85JL8MzrgNBHC3Ol7P3sxzelAirEClBWhV+2Ils7aAMnN/c+P4+IRT\nM2hTK3O6R5rWhNQWZc/yqdc/B1UbxdSeJbWrWqByT0ARGlsJ15796PHxlXvv0fOBElxHoITbsa+z\ncvXuTIClfjJyy651/DlncJ4nsue8HlL56byqqwC9nkNVfLvuz1+IE8+hZK4Lfk3nhFyZozwjRnkZ\nQl3LEXSAZ3Vw5FsNDeSl6R1V9Vuxntbnc/VSDCH8iTN+9fJPakMSEhI+ZXE30OeEhISETxrO1Urx\n44mQmRYxdnU2gIwWHHxNiA+mhW5Y0YNJ5udI2bKnYMGEkjdjTOnY2gKVfChH7CkwaHldl4vKOXkO\n62gLlmKwepVwKp4iPvVCy6nvAnQrTLyRVMGeXKsccmvYaihBJQskjip6HENHBoftAMo4hQr0rO13\n1CaIPY6O0oMmngOtm7DKTgtIusGa+uTCnde7PXfwLhFjDnUzteE8vpwhTF86hGM0Y5KtvpqmDkrG\nHvv5BgIDuE2zrJ396UMW7KgNlXBY9lv4s4USnA56SKiFKTpFWtl8jgbEaZHTYZ2G8i2n5o3nIKMW\ngyFCct5OSEhIWH+kl2JCQkICsLb0OSuCevsVh2FypQbS5B7J6Y7BKtqKztMtKB+TKJSI/bUDKGLD\nEtk8EfsMyrjwY8bTZpAJ6/dupNglnJpzOPKOydkh5bTbhOwY5KDmkA5j3PAyT3EfVIrtbiHGuwBN\nbyD2ugHL6aLrtNbe9/Txceul7mxtUOeuRYtmd+x1TmB8LbZRHuLA63Beb7bg7EwZL+Sanh/5zbSo\nN6L3/DiAd2LJMJzDUt3B1gQcr3vItVzWuLGA9sLSm0XrLedBiZzS3SOc9yrVrHsZc3gL1AP6WbiL\nd4bnmHGMYl2GsSqQLKuGcVYL8majikqvJ3lOK8WEhISEE0gvxYSEhARgbemzMlPRrpb8TeS1Paw7\nZS4hzXUSFd0oEWZ90HSaMoM1dxNW0YzOyKVfU9ah1N1yK97WDMMPg2Yt0pMZrLbZod/XBGVswVoI\nY65yqIrTsk3r42AG2alIj54qXXLrgZrHD886TJDkll3sRmgL1G8+9/EqP/Pi8TEt3oLjeyMmBmsN\nEb/b8EGpm993BCkyG2H7AFsQQgKoHO0qqEjN/GNxS2DeJCn0a3fmTt87hz6eAzjDjyEd1y9862WK\nuZB3nPqO4/if9Uc4EHKNj7AFgiRrVCRfwFvgCDQ5O0KyNqhsL5YUHznNMwQmBMzbBkLPB3Fr5oSj\n/RohrRQTEhISgPRSTEhISADWlj6XZdA0Jglqw6K5AYvaNVClDMmojqWRmHBp6kP1TAZnaFCIDImg\nDE7iS1kqSeoIcdNQmmriUSyTBOVNWIIvwal4CioNK3d95rSOVtF5GzQITuobiG0uYwKue8ZO+0Yt\niFSBKU0XXv9lJoVqrpY3A2OVgVaP4RzdOKzOZ1tOB3M6SSN+t0EncXzXm9gNgN+9Aih20UWZ8ILP\nu9VYNyitNocidgux15Bl6+G5FZDxGh0hEdqm92k48wm4dH6edLwPVEYvoVL+1MLLe0g73i54JYQh\nk0vBW6FGtW/HPOaSZr5oqoRz26fEVOxMY17yLNHnhISEhLXH2q4UTf4lK3oIkUPoXI2b9NgA34tf\n2Z2BfzVZ9JJ0AAAgAElEQVRb+wihu+B+hEeFf/nbdSz9sLLpwooyqvkXvDPH6pRLseiHxrC5HF/w\n7Zar4XDlQePCAkaHDN8+2/CV4GKBFeyxworX2a55f+hLd+nI29rYoDquH27twRiBHClziLXWIVZb\n7lQrTq7wWvCBnDegwAN/zNmh+xXOqckLQ4NG8Ec8gojtfS4cPI3+eRmEaocHEFZFatwS4zwdQXA4\nh88kFtk5DFqNe914dRDD+PqT1R5/WPjqUt1X5B8orh8fbyB5x+WtC8fH3Ym3ca/p/a+NKJxb1buz\nQGIe2q3wzKnAs2hW8yKJzCYkJCR8CiC9FBMSEhKAtaXPhcJx7onOJlJijqBIsu00lDR4abw4bCOE\nCukuW9e9jBl80PaaLuy60UCoHowxU/iETaE8c6nn4W8hboAXFGcNTl8moGO25zxtAMe7Lgw6NVC5\noofvIKjSfjQkdGcwFsFYsyFsDdAfEnQsgw9iAb/CKdK6khO2NnyMlgarBcaq2D3wa/se87cABS9A\nZQu0dzHzTtsO8q4gpC0HPZ7HS+qD1SpCDBWloaXZ9HFZwAC2teXU3La9n0wJ2wvVM5pNvB01jNuF\nls/bLtSRO6Dy1wbXvA8fclqdv/TS8fEG/FqHLd9u6MccLLsFtHbA5Hee9LnQ3/S/lWWSmBoVddYI\naaWYkJCQAKSXYkJCQgKwtvSZmI+ds13oum+eIfxJDadBs+tVHpX2NnzwQAHHW0iZeeTnL9SdMg13\nPVyuhfSgvRa+QzVYghEilkdLNC3lV+deXj6Bzx6stdvBKSZz0RTgRPt7TvG7bad1W8t8o/BNo030\ncOR+cpsNr2eAdK+9Eko28PtsZE7l4b6o+tBrqEUHwWcLz2Gzjek5fdapdGPHn0sfqjOjkvZax3zK\n54z+QSWoH/lz1vO+nUj9CZUYWs0HC6+zx0HHY87giTCFQO9kWo3djP6yoNedM/482xTyueRjUfR8\nvoRdv6hE6F4P4ruD56oxDX2kxp36fH6u78dNCPUW7apd5Zrq5KSVYkJCQgJwrl6KZvagmf2Cmb3X\nzN5jZt8Uz++Y2dvM7Nfj/9vPV1ZCQkLC7eC80eeFpL8SQvhlM+tL+iUze5ukV0v6uRDCG8zs9ZJe\nL+l1NysoM3MxToSLLfadVtThbGtIA3p8Dpbd9v1ufaOAaKtk7gqnExcuuiPt3sipbwGRWSG8qg6L\nZhmPaxB8bUAoduM6zMkEQrEKhNONoLaykSHd65NOVRXziIQ2wg0nTg03kM9krBvHSpIGoJu1KdLD\nIidrh8pAujFMbLvpWxD7M6fM9StOzZvDsVah23dqSDHZHKGNQygfFaDEtWHVlnyMrYlLvr1wbQ5a\n34BKEVKm1pAvZrYPKr8L5Z+nsd/xQDXmZ6UKHY7cKpwXUK/BM7oIx+smlJHGLa9/AseBgPmaX+7F\nc3Cuh/Dug22fw2OI2VIxZx1xrlaKIYSnQwi/HI+PJL1P0v2SXiHp0XjZo5JeeWdamJCQsO44byvF\nY5jZQ5I+T9I7JV0JISw17Z+RdOWMe14r6bWS1IQvV0JCQsKt4ly+FM2sJ+lfSPrmEMKhgVqGEIKZ\nreQbIYRHJD0iSRtbm8HKG6+hWGwTtOqoDefcTsU3aMu8SAFTWBA3mbKUyjSwJm/ACXeOHJs54nDL\nJ587Ps4++54bGw6n5rDlL/x2ga2BAMFTqKfQ8Zyob8NavVOVyTSp447X2QEdrpnXOZ04lW8h9ntx\niJhg1DnaXK0I04jXo2i1exBQFS24flE28PYOpqDMcPBWA4KrEFltwwt9EVOvTmlRPfQZcE/fvRZa\noP0FvBImUKaZYyujCbrJlKzt3ereWd/Lm7VOJI09Pjo4whYI/mwZY21Q/Sno4D7yfjC/zzOb1XFj\n38t+sOuCwESv9LEaDKrtoLJc8Qe2BjhX9FmSzKyu6oX4YyGEfxlPP2tm98bf3yvp6p1qX0JCwnrj\nXL0UrVoSvlHS+0IIfw+/+ilJD8fjhyW99ZPdtoSEhE8NnDf6/Lslfa2k/2Zm747nvk3SGyS92cxe\nI+kjkl71fAVlWaZup6K2o6lzjC04GC8QB0yx0GOrK+S69gU6OAU3goO17SH/C+TKAuhbMwOZ3IAE\n1pNuaR0cVpbO+gZTncIqiv1Shp8GWrbhsF42vR+UK5vBolh7snLOrsNSX4Mgaw4B2wlirEn2JmOn\nrxug5gvkwqEQ7aDtd3cPqvJrsKCWcK4eU6j3wMdtCsdjg6RYVodnwcTvzSB3NZ3fmKOnhFAsH/P2\nHAHfmCsj5L/JQJMXEI7N4a0wQEra2TJWGhQ4QFi4tuvPsAWqX0Dy7QjPuQtLOLdS6lD5DXAkvzyp\n6tqHd8Lw0J30t2fIC8Pu92Ib11Q67Fy9FEMI79AqP40KL/9ktiUhIeFTE+v5qk9ISEi4TZyrleLH\nFSHIonW5g1hRwVOnCXrAONeldXXO+FXkXxnCytkZIF/IhlMfGnwXY+dHiw23qAZY71q/8fLxcS9a\nQOegsiUs5SPULzj+PgeNsAYck1uQw8qaXj+U0TTcrK6HspZGY0hKGVSwYaCl03mr5fRtAtP/PGNF\nfqg2YqX71XGB/jQRhz1c8Ea/pjECfe04x6uB7tdgxaYlfG/iVLEcRyqNfDY10MOjqU+iKUzkh4WP\neR31NHJsx1BOHPsdR8tUoaDs2xO/zzZ9stJdvb3ntB9OBhrAX2KB7YNp0+vfxPbRNDbmxZvu4dYY\nYMti4H3OIX+2VNy2szjdXY60UkxISEgA0ksxISEhAVhf+gxQqXk2dMpE+rS9c6PGRKvudKgYOk3q\nXofTNaypQnKr2SYsziOneyXSowrJlYrrTg8tJsZiTGwTTtqC1fgAfaCBcNEnt/HjCdKTBljOR8v4\n5EMog0P5uoBVtHefx8T2USsd1gtKdD3rtK4J2a+s9D4Vg+qaRU767OO/g6xYMxeV1gKdHl71+OQ5\n9gEuIgHWUcPbuA2n+qO9pWq1U03bccpIybPmAbYjEAe9t+fPonHRE1Q1oHw9hTRX/2o11jWkQBVi\n1pkIjIEIM4zRxmw1h513/YZDbKtwzNsxVjwPfA0gZWrfx22852MbojcDLdnrhLRSTEhISADSSzEh\nISEBWFv6nJemXpSpGsKR9TlYQhkTPEfc8oUYz2zIATzJnXbML0OF+QmnTK2O063WLuJQSY8AWhRz\nXHNwVFGVztStj51Acoz6t53KXt/zJMANULxJg98+JLeCCnOIDsYFwsq7dd9SWPQR+wr5qU4fFBMG\n4hKW4/62lzOWbx+0QP2WBLsNKmfP+QjRkRqGZeUTL2PzIrZADpyyDwtvb3uMbYiaF9S/VN1r2D7I\npn7tjLSyi2cBB/CdzMeiQRVuskzI1dUuVXJ082s+h2jlnWfIUc71C5zxxwhVnsEUfbRwyny5icRt\nHsJ9jGfl9ffwnDOo0Q8x5ztPrJZuWxeklWJCQkICkF6KCQkJCcDa0mdl5kmKmGt44vSw1XFaUYc0\n0nx6YwKkRR0WZDgPd160c3ycHziVykFxcxQ3bfi9m9CX6iz8+9QZVeUcgDJmDHLecFoTsDWw0e5q\nFRYL3AyL4gAxrxbbW+tA2gpSVDVQ43nfrykGTtPGoJgNeMnPEJ/cA32dI7lVnlXjUoIytmDxno6h\nmA7n6S0k0Sqhnj5C7PkMTshzbIkwH/UyB/UFWKqJgGc+mvsDrdX9T2jRxjNHUrQCdDtDPP2xOjiU\ntEt4BOQwRAd4UNTrCBJAHHQD8+xC7uNihY/zYfCLOjE4gEy/fkj5MT+/AcX0o0vVL6hAv05IK8WE\nhIQEYG1XipNyrvdPn5V0cvN+ewOrw2d8w9gu+Gex6FTXT5GjZI6QvO0LvjpktpISxprDZ93o0b3i\nBoBAxZZ89aqktai+4HnLv/YTiLmWZ+QoqSPkDTYatZq+spuj/g2sZq93qvMXSxiFoJgSIHKqJ3D8\nEjf00HASYNDKhlDYwapIMx+vWataiTWg6DIc+vK0tuUGiC7DCff8Gqb1LJpeThsrqzyHSg5Wisu2\nTBCSR+OKsKqdQ71nDl+9OhwLhxDTpYlM6F/9QpyLEKddHHgfwqb3M8NzOxqckaMG8zygvYYQ0RZ8\nY8ulTyRsT6Out6+PcELmbsnyeFwkP8WEhISEtUd6KSYkJCQAa0ufm0Wmzzis6OdBDrqH8LfiPqen\nDUTf1WJYXtZyDtrDBvnsACF5XRgDsIk9amFj/sCp9EWk8BweeOhUseW0frBd0ZLGBPQeoXrj0hu7\nHcCTh0i3CmPMCcMQ/B3n2B5oxNCxjGFmfRhLuj5V8rr3QVOnfptdd4ILFIilRybC3GYbXmY9crjp\nglYxUMAx/AeRCyZHGthh5nR/hpw2JVKsBvhPNun72azaMnvS09Hqks+PfHijIK10Mg2uQViXqV72\nS+//lvmYLrcbhuZlDzt+43ax2rgyHkG9Bkansu/j2SAnBtqQtlnaX6AZrG6JfDbGzSHHLBr3Vidm\nvfuRVooJCQkJQHopJiQkJABrS5+JLsK1FlsIkYNwbIC/V71V0aoC6UhruJa0eooYrqzldCx3xizr\n0TfQqdLGJqzLEKJdfqqGV/e93ZchlErfOFgW821vVzalsCxDt/ze8cJpXTuGy1E4tANrZr2DbQKk\nQR0gZSx1hji2BazFVCaqDZBWNv7fRWjZDNsE7YDcMVDAmSENaa3lbZzMvW8DpETdgoW8X/PxOoqW\n5vEDSKt63bc3tqCos4GwuQlobcCYUwGpA0raQErWcbyVSkP0DVxgC4LpS3stt8SP4VO7gPcDx7GJ\nVLozpOedz6otiVlztb8hQytH8Idtxf5kZ2YOubuRVooJCQkJwLl6KZpZy8z+s5n9VzN7j5n9zXj+\nJWb2TjP7gJn9hJmtdvBLSEhI+Bhx3ujzVNKXhBAGZlaX9A4z+xlJf1nS94YQ3mRmPyDpNZK+/2YF\nhRCOHZ6LS06JxhO33PUNFANpQ+eRzI3g3Nsl1QaVaWY+hHQSb3XdQnsEi/cJix2M4gzd09KBFyF5\n7QwOyKDAU4QfdmYImyuRPhXOwXMc13ve58NINxtI0zrtgB6NqLLqh72ubwHM4OC9D2WaQ1hfaRQt\nES7Xr1eUcArK3J/QgkpzLtLNQunoAPdmXW/k9gzpZlFOQG6UZtw3mMMdv19frW5UQ6jgiWUFTpPi\nF6DBhw1saxxUY1Rs+XOoo2yqBJXbyO2DOrcGSEMKijs7dOrP7R6Dh4LilGrC4D9o+O8XM1BzlF2L\nOX9Cct7+xCNUWG461eO/IOlLJL0lnn9U0ivvQPMSEhI+BXCuXoqSZGa5mb1b0lVJb5P0QUn7IRxr\n8D8h6f4z7n2tmT1mZo9NF7NVlyQkJCTcFOeNPitUsi+fa2Zbkn5S0m96Afc+IukRSdre3Aq6sLTS\nIe9Hw2loa+zfhPHUKVktWkvrcJ7dPdo7PqbD8E7plkDBwTs0/LgJSrqbez29vsdQj47cKrtU5GmC\nse5NnA71BDHbJpy3keIzz/x4cuhW7BYfOahnb1odF+hOmPuHhYo69S7qhzV1ADWczKvUJhRzfs3c\nLP+STU+2crjMnUMn+gzpWOGwvYC1dDGl9dvrGYDunVCS6XnbS6jaHMQUqlSmacAxegan9t0LPm4b\n2Wplomm2mlq2IUlTiylxM1rNEdfegmjuLp5hA1s9tsX8Lv6MSsQ+j3uI90Z+n80PVNs0+RW3ps+z\n1S7ZdYxLFh3ZrVxP9+1zt1JcIoSwL+kXJH2hpC0zW86mByQ9eccalpCQsNY4Vy9FM7sUV4gys7ak\n3y/pfapejl8dL3tY0lvvTAsTEhLWHeeNPt8r6VEzy1W9sN8cQvhpM3uvpDeZ2XdJepekNz5fQUUo\ndRQtqqReNvMl/zjA8Rq+09fKiqpsQrqrc2Hz+Hg6coqzW3dqvDmExRH0KYez9SU5VakNqBzrCJGW\n7Hacs3QEigWHXTqVzxGfvJgiFwopNuhmDsfvSadq+wCxx53gNLW58HqOnnYK3LrgSUKKPR+XTge0\nEnJhL4I31ajl5f+G/D5J0nOHvk3R2PIydq/7eeFZ9eHIXEBMN7uOmOTLXs4cFLPedOq5GWOSZwOP\na58hKPioDTo88rIniM+uYbwMFvLsoreRhLOIzs+Lmp9twNFdu17GVunn3X9CJ9ynM2zrCOlxC8wF\n/sHnL4n0fARB3PIMizKc0Yfb1Vwo8/V03j5XL8UQwq9I+rwV5z8k6Qs++S1KSEj4VMO5os8JCQkJ\ndxrnaqX4cUVm7hC9D8dTSnBtInfIzI8vxmHZC5DiQl7NVgNWXliol4rdkqSJ30uK2wGVH0GaqQVr\nbSt+q2qQ7logxrpX9/MF6BuVpNvmbTGkLR23vb0ZLb2Renbm+E6e+GR6Gf0LW8fHQxTS3XCaOB15\n/5ugz23E+X5euHJ8/EC/kh37KKTVaGV9T+F9ozP0LhLwNBYgp5tOJTNsE9QRBx6edYt/0Y2U8CKs\nuci5c7Hm2x77mddZQtLs6Dn3ECC2kLZ00fRBvZpV2w07Qx+f3X2fCBc2EU2O+OnO2Ps/gYN9ed2p\nf971sWtcA5W/1+XdBtHBvtaBnBpeCdfh8dCZwuJ+oSojz1fLk93tSCvFhISEBCC9FBMSEhKAtaXP\nmZlaVlGl/qbTl8MCTtow/s7hEGxXK/sefHtPSDeVdVjoELd6feH0ZYtWViRxKger1YwXHa+stVdd\n04NK9K4hDhXJkpqgg/PcO1TA+lii7Q3ETZdIzNTNq37Mhz4+C6QJbYOOTc3vO4tA5UiDWkfiqs/M\nPdGVDM7jUYLr0zL//fsm17x9UMnea3h5HThsF3DAz5BitLYJaS7Q+gLbJ0un7RYc2ptw9M5AWS+G\nE6mojrFfg0QbHOnHiAMfgsovn8Qcngo1xGyXAwQlI/VrGWjD9uc5uuz3dp9yT4DxRbR3gly1y2tB\nzcdNxNJv+VZGHamBl+rlYT2Nz2mlmJCQkECkl2JCQkICsLb02UqpPqloyRgWyg2oLY8RaFuAki5m\nNyYpMsQyH7aRrGnkdHgHeZwPc5QBmrFAbHEbXrjNEeTFejfSs82a3zhnu/eRJxh0e7APR+orbi2e\nzmAVRnz0ciuhqK/+Tg5gZa5Du8qgNt5CUqbBFO2C9Xl/5hT3s5v3eP2xWUd7bhG+F+3bh2W51oby\nNqTT+IxCA3JpY3/+VB5vQs27bN3Y74IyZ4wJxtZEjuffAt0OoOligjBQ2VrMAU0f6AKeCiPMlQ4c\n1ok6ghEu1Zw+F6hnWvrNORTUe1GRvoVc1ELMOJOaZ03EgS+H0NaTP6eVYkJCQgKQXooJCQkJwNrS\nZy1K6VpF4Q5byPuLeOYpHJlJcdsx5nZac0o9BqMNMFvvQu35nqkPZx/W593CqWQPCZAyWHpJz2rR\nKlvC0TeH9ZH5nYnZGPmoYRXfhRNuo42cyTXIWEUn9C3Q9I82vN3NOpJ/IZZWYz/utFZPpwIx5g/2\nXC5s0kCfo+U2zH2gf/26iyE9teFjtVMiHzU4Hq3Fc9DEEvG8DYO1voG8yrEbAfHT3EYpEDM9gvdB\nr8utDlDsHXcCL9H/8tDHtNG5UXaszCFzN/LjA1iwm9i+6ENVXJvIzQ3H6lbmzzwgT/d0ULUFNm7V\nF17GoPQ6w7aXMY39LJWUtxMSEhLWHumlmJCQkACsLX0uG5lmL66oanbglGUIy+3m3ClO0YBScaMi\nFMxri/TGmmDUqGRtm1BERlvuvY4f4Lw7vQDz4nWYomNbmK+ZqIEmUil6AuXrXu5tKaEaPYLD9gTU\nc+ngPTpw62/ZhwMyqNQGEnRdaHks7QzxyZtNp4b3QC7tanCKf2GBMY8W0sc/+uHjc4M6iZ1jCkf7\n7cJp3UEG5WnER08PvM5s2x2SM8Rh12I8e4BHcj6DBbvu13aOnMpPL8Lpn47ZzCNeeD/rdSh/x+tr\niKtvI7EYvQm6SFy2YBw4km+VsLLXQJ9r2IYRtkxmMYlWEzmi6ZCd57gPTDmLVntLeZ8TEhIS1h9r\nu1K0Iig/rFaFxTbEN5FXojX1889C3HSJRse/oA0EtG3gy9/N/ZoJfAAbyAVTtJCvpY7z8LE7gjjL\nOFRf/D4eT478L23kSBkOfGXXL2FEgK5sgXwdjbF/8g2+mRSuXWLjw+7ruH+fl10yfSeMFTlWXsNd\nb9cIjniX2i7WO8eK56MHV6v2dSGm24NBY+IrqKvlgfcBbnUbbV+d9hCKx54VUC/KsfrZ3auEczvb\nvvK1PV9h1pBHpdzxZ96BUO04ICUpbC7zOtKQmle6uajKHF530d4TuOL9CXhW9WveB0MIq2FlyxjV\nbOr3IkWMGjH16dhWh57WkNb20LzOrXr1DLO0UkxISEhYf6SXYkJCQgKwtvR5ngU906o2xNvIb5Fj\nyb9Xc2LV6zpVWURfsfoIeVZgiGGayjlCwTrwTTwaueGkwC71BiMIwXYMfoCNqEhSQMC03nGaNqVv\nHh7hbIF0ozNQuSedbtYvOT1s5RDLjX59sy6lgfywB1+/Ev54w2suirpp3sZtbE3UYADamzitbqMf\nB4tqvH79AnwqM+/bZfjacdwo2rt5zX9B48EhfDOpL1Mf35ii83Dhz20ToW2DDRpUoBI0W50Lh2F+\n85J5U30eHdWqe5t9H7dsB4YYjjl8I8tLaDdcVutNH3PLoYY08TZm2PpoTatypts0qCC3C2h/rcRY\nxC2bslidY+huR1opJiQkJADn8qVoZrmZvcvMfjr+/BIze6eZfcDMfsIMS7KEhISEjyPOK33+JlX5\nnpdc7/+U9L0hhDeZ2Q9Ieo2k77/VwvpQoGlsuc8cl/+z0qn0ZHSjJMkO/RFhzW2Bpkwgclqn/Cos\nh1OkymR+lxOpKg+qttQ6TsfmEJk9rDkdamdIsdn3b8UmLMF7cvo8hkpNhhSj8xhGGKg0g5C37tz7\neVBzzjaGNXUD1+/mFGV1+jpECtVfnV/1cuLYXUA85ZFHBOoAfb5S+jMczr08hvbRr3Sr6dfPJt5/\n0trpxaqNPagoBYSBNkErR03vWwOpVIcD37Lo7biVnWpEB9jWGGVVOT2ER3YbnAmYQ/AaGMNsPt3y\nezNY83tTUGz4KRrmy/6g2qqoj73PGcZQ2F4JmNt5XEuZncs11ceMc9crM3tA0h+U9IPxZ5P0JZLe\nEi95VNIr70zrEhIS1h3n7qUo6e9L+ha5D/0FSfshhOVn8AlJ96+60cxea2aPmdlji9lq0YSEhISE\nm+Fc0Wcz+0OSroYQfsnMXvZC7w8hPCLpEUna2tgMl0fV8r9cIEdG0x1y2w2nDS2op1i8frrv1167\n12lNH6FV/RJe10AD7+R5DZQI1LvkedDd8aXqezCBuooZLYh+1xRUymAtLJDWtP2Q89AFnJenoL6Z\nlqFbjisjp7010NFWe/W02e/DSRh23tnI6fsMluhcXv54Xo11Dyo52wM4iSPkrdCNVmNJGkEI10BD\nM1ixBefwE/llwo25SyZj5DmB9bcNNaTDTa9np+MpSRd4dlPm6GmiLVGxZgFH690BHNPhsN9ceD1M\ndxug3tMGTV/0YH2G8zyFlRuXK0s3mqo6wgMnUOapo/5JdPYOYfVzuNtxrl6Kkn63pD9iZl8pqaVq\nT/H7JG2ZWS2uFh+Q9ORNykhISEi4bZwr+hxC+NYQwgMhhIck/XFJPx9C+BpJvyDpq+NlD0t66x1q\nYkJCwprjvK0Uz8LrJL3JzL5L0rskvfFWbnLrGIRDoXAyaSOdIxyyw6SiB42W07smRGwEcZsGaCVz\nl0xh2e7C8TVDulE6Sk/GXkErxlYH0B7mEwlQVZnAmjxFvpA6ncRp0QTFJiWqLx2IIWZbbOCbeYjU\noHBYppJLDelBR3O0iw7ucOomlX1w50pV3pOH/vt9iLxeQhw66qnBqbuJcR4fupM4ES75w3vS3FH8\nYreyFlNRyDYQQP6kxyd3OnCAxxZIF+pBU3gwTF2YR2Hq5V/sV7lzDLlOdo8oqeTX1nqwoM8x/iiP\n48m0K5yjAfHMe82qnG0IC3M7hk7y4x49JSJNp/DxGuHcvhRDCG+X9PZ4/CFJX3An25OQkPCpgXNF\nnxMSEhLuNM7tSvFjRmbHqUJbY6dVfUNay+ec4hTbPhStzSoO2ibOO9olhmrmZRSgOBkcY4cw19IS\nneOa1hCW4xqs2NGqV5yRA+Ms0sK8HEJ8bughDhmO5PS9HR9V1tdDeD3fL6eJk9ydpKc9vzFDus8e\nxrZbd+rZQD8Ze26ge8udh9llOKzDGbl74MczOFVTfJWwy043J4hDXwR/GPdmfo3G1XgVsKwzR0t/\ny8cigzhuhkc0h/N2xryl17z+LQYBRMfz4oL3ub3pbWpe9fkZCgigYVy4qgmoJwNlzyE1RyHgftzX\n2Ct8q6G18PGcIdyc+WK6G1XhnMvrhPXsVUJCQsJtIr0UExISEoD1pc+S8phwouh4NykX1dxxqjKC\nFVcx/neZdlOSaMw1WJYNFrg56FbZp8Osc6wF6fYcKT7heGuRBocxzIaQsVogDWYN6uATpHSp43u3\nD2XnnXsRUDyEI2+0Sm5C/qp9RIks789WAUdzWCUnhZe3bz6el+GkLeT9KPec7s2ihPakxUv92mub\n3pbLSOs5AB2mOvlw5FSW8b4Gy/20gf5HB+9w4O02UGA62k8w/kPMm/amt+sQ+WIuZ+gUPAQUt2dq\nUMbeROz14cXVztFhDwrvmKMZ5kitxTH3tm92fc4vnfc7iMFnCtTuBA7j8FoIe9XYUg18nZBWigkJ\nCQlAeikmJCQkAGtNn49BNWEkkcpBWxYBnqqRQuQtpEAduvUv3/LzTJB0AkgVegSHWFqopzBdzmp0\n9s5i/Ui+BFXt+RBWyQ1aX0ETG5Sd8sPnBq6UfSl3KhW6VV21Pe/nAIrVF+R9ZirVFuJfQ8vr36p5\nu3anHs/bl5/fbnj99Tj+XbDLfUiUbWVe/3ABakwVbsSv1+kJkOEiyJs1xqDEulEurgfH8GUSNEma\nbb0Jk6AAACAASURBVEKuC7JcU61Woj7o+Rhtoppn47hcniGZGeLqW12v/+jQHc2X6VglqdVaHXs/\ngvN6r4FnRwHzGB8/RBBDt+X0ndtHNvJ5USy3jNbTdzutFBMSEhKI9FJMSEhIANaXPodwHKNLG15Z\nc8raQPf7sBDu9ioatFtzmnYRFIRO1dMm4odhxZtB1brWQTIitKWNXMYaOG2ab8UDOEPX4bxMS/gQ\nNOnChQvHxwF0ezMHxUIM9/5lr78XlbNafU/gVYAy7WY+FjsI5p0eIUFWH9sU2BqghVxwAp7BWr20\nos6YF7lYvTWQ9eCMfOAdGiEmvbVAnuy200BayKeIc27FYqg8PoDkXHN79TZJD5nLSiSlqoNWM0Z4\nfwd7GVF4fG/qz3C7j7zYiHGeL/jUHTms7A3kFM8x5vM5nOdB32d51f8cWyAZ4uQDE3SVa8qVVyCt\nFBMSEhKA9FJMSEhIANaWPlspNaKj9hSxsrUxZMTg1DotnZ50ojWwOXBqNG0g9hRyWa2BlzGa3mjB\nlKQMklJZTorl50dw/B7OK0q00XA+GJDcqNVx+prDan4wAzfG524HuZHHG96nizXIcTVudMQtYLWl\ndf6wBGWGvFgGJ3lBtboDxhgQKz3ad5mwyUWn7V6gj20LVusJtkAOemhj6XT3AfMxGg5cVbuOLQmq\nZGXNih7XqdINA35B+TXMmwas7IIEWH7kz2LS9fMbsKKPL1ZjR5V0g0N0E+bd0GzjGr+eMmrYvVHA\nptEcz+6EF0N0Hu/l3KbAw9r2+ZdphSP7mjLqtFJMSEhIANJLMSEhIQFYW/osSRbX9y1Y5TI4b4+e\nc/rWhYV4/9kqVnhDiFntwRIMa2YBh90SlAVGVHUmdBKGEzDa2kQbLTqKH5VOwUj12pSyppWb10Dq\nbNSgkzpoFaTRtqKFOsxAR6EenfV9LKZ1L2MDDs4FYrkn2BrIkIO4fg0SZJh9RYwnHyHJUw8XMBe3\nTX3ksvt83K7MvI2PF65gvTmnlwG9CBy1brzm0NvH2PTQ9TLovM/c4VnNr6nVkYBrH9dA+boR5+fu\nhtdzZN637TmUtKlejnzMpPh0sGZMeIntns6mt33Z2gw5pTug0uOBl8dc061ZvHM981allWJCQkIC\nsb4rxTIojKqv5QR7+F347O1sbh0fDxbI3RIXHC2qgGQwumCzul73IZy3/RtzaYTN65l/UhfYVJ9j\nOTmHj1sRU16y/jnymo7hj9e87l/zbWy6C8KlWO/qKlJothDSZdFnb4ZUnroA/8brCC3c8pXKyKi0\ng3A2GE5G8MPThq8sDYKnrWj0KWFEWKDPbaxUF1gpbj6HlS3CDy9ueLpRg9pQbc7n5c2axXXjbAOK\nMtdXr/yHR96fZuF1bjaR6wThhHOEPzaxUlyK7JZ4JjMo8EywOq3BTXGQQyUHz/x/ND0UcHOEJSRs\nhLW+G6CKsmpLCytPqgHVYESaIufOct6GNV0qppViQkJCAnDuVopm9rikI1VbPosQwm8zsx1JPyHp\nIUmPS3pVCGHvTrUxISFhfXHuXooRvy+E8Bx+fr2knwshvMHMXh9/ft2tFpYhnC4Ep1ITpO3sNZ3W\nlXlFg/5Hw+nIvW2qh/iwDWEAmIFOjCB4Wgdl0jbCr5h3Y0U+linoZa2DnCe5118iF8ei5XSnDeMO\nBV9fFDyMrBzfSH8ypGk9kab0HvjJQXVoLoZNQpR0H0YiUObxZLXay2xePYsMvn519JPhf/sXT2Qm\nOT7qICyuAyPNoo5QPAxz/4TvY1VOw/zcAULyZjCLNUA3d/a9LQsKFbe9zmXfJGlxBSGPMYxus/Ct\nBuyuaO/QDYFt5Lypib6JfsO9KGcKZZ5uy2n9HFsPS2GmCUxOLQoCoy1tKBB1Yl6cfE2J5t3Sq1dI\nejQePyrplXewLQkJCWuM8/hSDJL+rZn9kpm9Np67EkJ4Oh4/I+nKqhvN7LVm9piZPTY9I4A+ISEh\n4WY4j/T594QQnjSzy5LeZmbv5y9DCMHMVpq9QgiPSHpEknY2t0K+U9EJuIxpARFPBjeNCqcVs071\nrXhw7lRzgJC3bVgwGwi/0pGXUdA3cIPxYvDDQxjfACbCg2lFPS+ghcNDqKFsO2eu9b3+DsMJEfLW\nhoWS1uKOfGCWeV+Yq2axAyoF0B+vnHp5A9DtnL50K2i6JI0fv+ZtfPHFqk2wiM9A9TCCajG0rYCP\nHXxNucUxMGxTwA9zPPa270R1mhL8uoZ0n1eRsvY6/EdbPRfKpRV3Pva208vA9j3ksNGt+hrQuZF5\nPQVS1pY1hFNiO8D2YBXGhL7Q8pSsC/6VY1xm02pc2rCg149QP9wWMjuP66dPDM5dT0MIT8b/r0r6\nSUlfIOlZM7tXkuL/V+9cCxMSEtYZ5+qlaGZdsyqS38y6kr5M0n+X9FOSHo6XPSzprXemhQkJCeuO\n80afr0j6SasskDVJPx5C+Fkz+y+S3mxmr5H0EUmver6Cgk7SkuPzUJspFouV57uRnizgyHzpGaiH\nICxrWDp9aYHkHRy55bCxDUfmhVO5Eso7W2O/t2s3PpYurI+TAag0LdFbTpnmoOylN0XZAbYJdvxe\ni0Kk3dwp8/6z7qTc7kIo95K3pb6Akgt2EooAGgplmgx5Pwoq48TQSeY5YZ6ZvZm3pcHMrxtexhzm\n0joVdsiewep7U8i8jCKVhtW4zlA98/73oDrDFKP0H6jD2Xw2drq9gCjx4tnKqwyRjcqvOB2HP/8J\nR/YCWz22jWc49IL2g8/LrYnz6lm4MW3uAApR1sEAwcugRP2zWVV/COvpvH2uXoohhA9J+pwV569L\nevknv0UJCQmfajhX9DkhISHhTuNcrRQ/ngiZtIjCoIv6ajXMOWVSQCEaMcXpDFTu6LKXsU0LXce/\nK0eId+13nMpeX7jFkWKd3fHqdi3p1ARtWsxAR+GkXUz8msMDj6ENoI85iN28iZjcuVPSy63K+kq7\n/o68DyNQYxtA3wdWyYDu1PZgiW+Bbm4gcFs+LoNB1Zas7/R9VPf7mrCUM3VLf9f7n0EUdQ9CuPXh\nait2AWWicdwqac+8P4egoK0jjOGWl7IY+DXNDgRk991JvUSsOJ/4sneGGOcGxi27CId9bKnsZagT\n49La8v43IKtz1bwtfXhOZFHwl38G8zFS+SLFan2CvETdquywpkuqNe1WQkJCwu0hvRQTEhISgLWl\nz7YIyncrKhAgo0WraJuCmhD3XMYwl7Dm9SHaenTJj+twJO6DYk1BWXcablE84SSOmNg6xFrzSLL2\nEb/agfpWceT0qYH8KxmsrAs4m8/hkJxB6qyL2ONJvJ6OvNMO8nIcOB1dUl1Jat+/c3xsEPBlXhIw\ndk3H3naOfzMK0e63kZoVztOMtw4Qx2Vcs+Tnu3heLYjFNjFG+1ew9RGf/2KGrQGIsy5glO087YXM\nkf/nqQLh+uDJzQEGAHJwWZTpyjrezyNYtjfR1hKeEJchpisEI+SQTsNjVH/mYzHBM20UWfwfzxnW\n9w6c7pnhtBY1A6xYT+tzWikmJCQkAOmlmJCQkACsL302U9aoutdADgqmp6Q2UpeR0FGxO0c+jyms\nr62Zl1FgBAct5LEo/HszbMHxlmphcPw9gkN2b6eyBHdg3utDvXmQOwUdIOdLk3JVC+YOQaww6NsM\nKVkb4+qaadfbNDxwmmwbcNiGvNTswC3IzS04HtOpGbG/xZaPcw5Jt6XVuz3xAeojThfDqd0GxT6g\n3j2HajTUzjsb3i6msu3Bj3m3Xj3zvRqc66d+wU7L480NseecTW1scZSIyWZa1Q62L5aG4wmeVRcd\nvSr3ur8nd5X4AhSbmwf7iMkWxvbaruerue/yPV5OfBTM4bIBZfoS8znbQ76WWP+NYnfrgbRSTEhI\nSADSSzEhISEBWFv6XJalxsPRDefpYFuKMlFOhJbxn6QHVJUuh07BSqhrGej41bpT0y7oZh0JsBZz\np2cbmdPTcaSQpGbPIVlRC+zxEIrQ9y+8by3QoCks0V1YfPd3d4+PazFx1KJcbVHsPANaeZ/TUSo5\nNzGGJcyVI7gH8ytch9q5dqtn1Ybxd7yDJwDJr1aG5FMzH4yDum9B3NP3xFVHVJYundZPZ9jLiLHP\neRNbEAxKBuZQ+F5glrTgMF00V/9pTTG+i7itUKNpdw6vBaQcG068bzVQ8Cl84fcKv2bz8Ay5tqf8\nmTevRGk8tGncPkOHtInyJqsvWReklWJCQkICkF6KCQkJCcDa0udg7nC7KJw+TaZuLa3DeZlmvNaw\nur6A8nQNsZ9jUNYWKE4JmnYBOlqzA+cb2RSJnpDQibJbG5GGDttOWXpw0h7DQtqTS2cNRjg/9O9d\n66LTXUqHUQ5rmbyoCct6d8Mds+ddyFINfAxPoCxWns7aTtnncI4Ou4jJjVsMBeTKFqC3dSbuApOb\n4xFuYi/jiaE7UneRjKpvfk0Dfd2Kk+VaExbnBmTJIDNnWEsYnMdzJCuDULdGOX4AVZ1aNV50SOhD\nCq62A8k5SKe1EGS+OnpeqmHn4eIe4uA3MWCRqrOMwZ7HzzcbCAzoNnAcy6it55pqPXuVkJCQcJtI\nL8WEhIQEYG3ps9Uy1XcibYS1sN9wnjoZwqLXBj2NyZsC6PCUSXxgIT3qgKdARquNeFvmRmYqqCFi\niOlgfFhU9bcWuBoOvr2u96FOYyESR01qcIw+dLprDcQTg4Y1r1cFzSBrPUC8bRPSVTX0zUAZOUZN\n5BqmWNhi4oPXnULBPKqULV4Ex+gOrMyId86R0KkBa74gl3b/ro/dBPHJczi7N+CwX4unmfxrAR21\nOr0P0J/Wwp95Dgt5gZjk/gUojAN5tCg3M8h/tZFTHNsRGzV/5oMD33agkvtOy+dQvoNxAQ3XVffI\nOIiTp475nGM7IEN/NIPa+4XqYa3rimpd+5WQkJBwW1jblWJZlhrGr9sF8y9lNvavX30I/7ncv/+9\nuII4MBhocgieIlSts8CXFUuIAjkthLwXc3zZ4daogxl8KmORPYrZbkCoFMaaOXwd1cZKlX5qJ3J9\noM8NGmOqFdrs0FevB0NfkbTbvlI5IY7bwxQ68lXYFPlVOhOvp7GP9l7y59KIg9dg3g/zegaIwpyW\nXsYM57vI5Vkg5LGB1SxcCYVoSU1iTps2lJGOfOGlAg8rh1/hnAwCK755FyvIq75SD1e8zxv1uIZG\nn+ld2MaUYOrZBvw0c+QWWgx9btX6vj4vsLadPugNni6qJeL8RC4jKArBp5OhitMnqtwy5Wy1Ye1u\nR1opJiQkJADn7qVoZltm9hYze7+Zvc/MvtDMdszsbWb26/H/7ecvKSEhIeGF4zzS5++T9LMhhK82\ns4aqffpvk/RzIYQ3mNnrJb1e0uueryBbVFRglNOvz/nWFAorHYTuLTfya8i/0oC6TDix1Q76nPn1\n9X1QGfhJjnpOSbrIt3mtcK50T6goVr0HH8gJvdm8vBIpM1sDb5eBkh32+e3DNgEUWWb16vocRoFt\nKKt24feoJvKlYCiudpALZQ8pXmte5gK0UhDcPciq8d8aoU0t0PQz0mly+6I2vlF1RzoZRjfLIZwK\nUVxFId4JUsNSsYZKQzPMmym2WKbl6pDHIdRmekd+zSQqH/HaAbZXDhdu3OgHH3M+zRPbBEjJKhim\nsr7P+Tae12a92hLZnWGbBP6wjT6fMxRzOtW+gv3auVtTfVxwrnplZpuSfq+kN0pSCGEWQtiX9ApJ\nj8bLHpX0yjvTwoSEhHXHuXopSnqJpGuSftjM3mVmP2hmXUlXQghPx2uekXRl1c1m9loze8zMHpvN\nzghsT0hISLgJzht9rkn6fEl/MYTwTjP7PlVU+RghhGBmK7lUCOERSY9IUq/XC8NhZfVrwEQ57jkN\noPVsgtC5bl5Z7tqgZuMSqR9Bk+nfSO3TaR88Zd/r6SB3yBgW522kqlym0KRPXQ76Modlu11zK2NZ\nIPyOOULAvMdwGgy4qDiKyjwUQaV4C2jvidHHWOyMaI2ExXff20XL7egeWPFjalHsKKgGP8ZaDzls\nYH22EfKywAd1OHXqOccWQyfzOkcBCjsxjK4BP8U6fAMJ5j9hnp0xtgP6BToCivvkxLcV8njNxQFj\nTOERUPO27kIliZbgOrd9WgiFnNBP1sH+L9u7WfftjencJwvVhfql3zcI1TXlGVsadzvO20rxCUlP\nhBDeGX9+i6qX5LNmdq8kxf+v3qH2JSQkrDnO1UsxhPCMpI+a2W+Mp14u6b2SfkrSw/Hcw5Leegea\nl5CQ8CmA80afJekvSvqxaHn+kKSvV/XyfrOZvUbSRyS96nlLyUy1qILThdpJAXoyH7hD7OgKFEGi\nUkmGELLuLkLiQF8KY4yUU4wF9jQXcAy3J1zks70U+ZSORU4laRYtlAG0K2w6NWrByhqQR+Q4rFHS\nDGo0UxCoKdoFn2YtiSKtqc2696cEZZ+JNNn71tlybk7x17zjY3v43L7f+hzEeqNiUAsO1e0ZUqZu\nIB1p8L5t4Ls+OnKaPt+BMg8o7rS+mvJNYhMbVLdBDpvuxJ9/G4oxY2xrLJC+9ckt79t9BYIH6jf+\nyZWb2IIZw2rMNUvXy+gYcvsYPLzRdjqbU+FpuvAxasXn20DYZg3Pajh0R/4jwzzrVnPR7CyNnrsb\n5+6lGEJ4t6TftuJXL/9ktyUhIeFTD+eKPickJCTcaZy7leLHC5llajerZT7dnltQiVm8dOP4uIcA\n0FBEOkE/Y9BuFliOnZodgW60+xAozZxKtXB+BofsxiZUZaKFdnfuFtR2Ayo1cOqeFYgDRuxrgWMq\n84xdhEaTqV8zulRd1SggSDplbhXvfxPKMDM4uGcQ7Z0PnNbR+tm75Kk6qbCTRytyCQHfKdRkbc+p\naeeSW4UfD9iOgMJMA54FVPjJBtjW6MLxOdLGxuREILBfy3w+8FhvQhQ2a3n/HwhQPYIMDUVui2jG\nH6HszT3QcaoODbBlAafqizV/oLs1ny8ktjmCB5ivp7haCcrm97mY8Kjm49PEdpCQBjWPXgZ2Rj6f\nux1ppZiQkJAApJdiQkJCArC+9DlIzcha6l3IbiGF5AYMx8gOquY40gJQ4/bAaeqBs+4TMMTq0u23\nnzsPmsspToC8VklB03Z1/TZibwsI2M5PaJT54QAxtl3IdXVqTtnayNEyhkP21Si424A47FGAQ/sz\nsIpCrqqZ+/U5PrG9Tad1U1JptL3DVJlRIHaGGdkGTZ8it0gTjuEPblw+Pn7uAFQaTsr88ueggUPk\ngFkKB8/BOzOj9RfPauKDnoOoXoaOWB3ybktZNkkKEAsOeXUvRYiHW15nD89wAqt5c0ja6s9lEylj\nR5hnOsS2St/bO40eFxlitts5PAW6XsY2ZNnGoTpPBYB1QlopJiQkJADppZiQkJAArC19Lk2aRgY1\nhRV3O3O6l4EGUSZqEkclP4SjMxy2a0OnG0/trKYvU1rm6khJifDkkmrOdAiPh8VZxj04eu/XEBML\nJ9xOCwSeVl5YSJl343JZXV8g3rieexmHl/zarhvZpcVqEnU4cDmqnL7eULZ+dtvbtR1TxW5MYfEE\n1aw/630ePgir8Z73/zJStmagtYxVXwRvbxvbCpNI8RtQY7cLCBRf+Lgw5wtzpNw79D+nGei2TWB/\nZ1rQ6HhfA6Uv4Qw+2aHfgGPCeGcEGIzqzGMDTwAY1Okcvt+syunUnPYzHbDhWQ3hfWHL7ahsPZ23\n00oxISEhAUgvxYSEhARgbelzZqZOVBE+gvP0FD2eMm73Wad7ulI53h7B0dlIWfpONx4447tSgjJl\nsHgHWMIzpLCkBNUypLQxo0o0KCCo+daelzFEfPDcvM5GQHzw0C3BTCGqaAmtoa3s2lLaS5LKDb9v\niq2BsOtjWBSMvYZqMyh+98DpeX+pgt3DNgJyyY4Re9t4Gk7KvdV0u2ysVtuewFxfQ4rXi0c3/ink\nSP7VRlrZASzBI6hjP93z9u6ASjOHWRPllKNqXObYXyjhwTAbI0oAyc9aMzh417xwxm1nW15PMfW5\nOCi8zEWtKnMXknMX5MEFTHdaQ9lqVnNhPclzWikmJCQknEB6KSYkJCQAa0ufi1DqYFrRggAr2S4o\nSRfWx4MLcJQOFcXMQFMECqaRU6Y6pJbmR36e1uQaiQYoc1hhiZSkIt7bQhKhKetHs4ZbXs8mEk0V\nyOk8f8bNxbA9///tnUuoJFcZx/9fd1f1896+j3ncmMlLDJHZaETUkCCSIPgI0YVIgkKQQFxkkYCi\niXshbnwsRBgSJQtRwygoLgRJguhmUBNBklGMozET5pWZ++h3d1V/Lur0/f537s3ExMl0d833g2Gq\nq6urz6mqe/r8z/dCShbSSWwv1wjut60/XYpgLieUBZusnGWyMm9V7MWBPjsV2/VfIut28q/M8VrL\nFIO+ZlKuImQhpfholnCybOeTFtVJJs8CoXRoIEuv7ss+y6nLYvImUHKubwztW9dgMc5cuGy8SOm4\nxpcWmjG1qTi052DERbYojdmATlejWOn0gPWfPSf46zkmvxwc+Qu0jJLw8gI9wyMlTwDsHR+eF3ym\n6DiOQ/ig6DiOQ+RWPr8RPfJpHp23IkKcAmoQLHo1EptCqahAWbC5iBNLVs58rCU7d5cdptWkTI8s\nkL1RkEr0k8VO34Oa6aFoTPHLFZOmq2Pr6KBp318nK3JapFjt7fOTAzLFPjepiFOfnpqFsV2jiCyn\n11N7i5T2bCUmkz45lfcOZTK0eI6WIEgCKmWnTkjK8dKIcBbyKuXd6nKQOy0PUEb0KBSJGjTtfAnd\n3BoluN6iJYAdKbqqJINpf4fkfnlEtcFDqrURSeYhPZ+UBB1Kz9YipzdrkmSnWuPsUl9Ysevfp9Rh\n1XBPhXoh5KReolRwHEiwFVLkcS3oPOEzRcdxHMIHRcdxHCK38rkIQQOZnNnsmfYhhYmY6uSmI6of\nHKzSJZJv0bJZGXdYjYmUZNLkuwGgQE7YQnGmHbL01Qu2f5IdOSX5NiTnZWxSRuwFk5UVSinVpmLP\nZUr11aqx4CP5U836xLG3pZr1udc362enaG0pUe3kDjmML++zz45oKSFuUREvrg1dytpVfJfJ6z7V\nd06pWBVnuC5sWLsmFmQASMgTgDNE8zWNXicZXM3aVVm17x9Sf4bkVN7Z5FzuRoXyjnFN5FLRPttK\nrb3LE+t2Yv1MyOl/c4HShVHG+EVhHwIj6lFaNFuZ2ZEpvV+2PvVK2TatImCs1pZGh9KfUR9qoY46\nO7/niZmaKYrILSLyF/q3JSKPiMiKiPxWRP4R/l+edlsdx8knovpGqVimi4gUAbwG4MMAHgJwQVUf\nF5FHASyr6tcv9fnmUlNvu+P2XfvbXbMA1Mc0US5wVpXsV7FQplAt8kEb0ccSmmHyDyfXOmFaRZsJ\nrFRsVlKg7CXr/TYuhuuMnKcEoms9Ko9JxwwX9vYli6hdYw6LC6Fb/XS06zMAdoSZ0SQUbZodlyic\nkGcf9aplm6mSOBmTgaETEp22Y06pYyxQklOuF6N16vOIZvYFOv48+ZVWbMY1JMNQ8ULWqTEZn4r7\nabrFGXQ3bbbXJhWSUls4sTCzRTPFSVIbTprbofDIUUL9aVB4IBldDnapDC3dz1bEcXkGf1aDKuDQ\ny7Q/3PUZANv1jgCgGAyHz/7hd1jf3MjddHGmZooXcReAf6rqKwA+A+CpsP8pAJ+dWqscx8k1s7ym\neC+An4Ttg6p6KmyfBnBwrw+IyIMAHgSASnXvX2rHcZxLMZODoojEAO4B8NjF76mqisieml9VjwA4\nAgCLy01NQwaXmOQJlwcFSaykbZpwcrRQEtoSZbcp9mw/KQ8kZKQ4XaeylkqL5Ilts3+YVsnfb6LP\nqaxnp2nnPlizMqEokEzqkgyicMIC+c+NSeMXZbdfXUr+mn0yrjRP2fVpr9oxHPLFhoZk3WRlr0k+\nm6kdX67sTqLaHNp16FNoYb9u526U9k6+mpJkjFiFkzGqQBK/QksJg0b2vQMq+yoUzhlTOGO/SUsT\nVAqFkxZLTGVQ6c+sR8sw7WBIqXS5lq5t7qjLwvV8yDBzNrJ7VCYDYESxgK0GZSYq0T1v7V4qKZFR\nEPTMt8kfclKaN937z3DumVX5/EkAz6vqmfD6jIhcAwDh/7NTa5njOLlmVgfF+2DSGQB+BeD+sH0/\ngF9e8RY5jnNVMHPyWUTqAD4O4Mu0+3EAT4vIAwBeAfD5Nz3PGCgMMi3CmT+ascnnPpWHjGucESWT\nHlsD00Y1kqYjSghbq5uFskNpYlYLlKxzRyggbffs/H3yD6uE8LfqfpPJSjVPQKVDohr5WkacYcW2\n+xSWxj+D11BMWW8he6NIWWLa1P82ReetpPbYcGhbKzHLPqk0gDITVWJr/IBKjBbrcTgHZebpUmwd\nydT4wP7t7WHXGtw6v2HnXrbrzx4Cq0qZdMQk5mRZpUjStEJlUhPytazS5dxasmNKVAY3paTAXfIH\nLVPZ0nZ4LttiMjZesuuzFjXt2JH1c1impQSyyte6tHxBXgSbVPp2E3aPrm9kvqTDlBMic2Yc6085\ntedsb/t0fpi5QVFVOwBWL9p3Hpk12nEc5x1lVuWz4zjOVJi5meLlIsEY65pJjrVocXv/gFKPjLk8\nJ1kOk1DfI6WQq1OJyddDsQXUcFjcEoXqKTljpw2z4rGU4wqRleIeoVtkWI5jk/eUGAdKzsZxhZyk\nxQ460Tu3vb1GknlcJem1vtsSuW9IGpis45OQvIvbUm2QZE1N71YpnJKtqyD5utXPZJ2QBXU5tlBB\nUuwYb9i5W0NydKcEtexIvlA17X9maJmR1gq2fyM4zC9QHyih0I4Qyj7VualRCCXLyi2qL9OktELx\n2Nq4LzxSmx17tgr0vlSplCuVMo36JOUTksw16j8tNxw4R8sUFK7XK2b9iCIqh0rhjKAwv/7Zze3t\n/zSyezXUvR3E5x2fKTqO4xA+KDqO4xC5lc8lFLA8ziRnOyGTKs/4ySGWfx1qQXr0SCY2FkyC9ONs\nRAAAA9dJREFUt7smNVNK2hl1TKbE5MgcUalISliCYkRxwCmVKg1WV262sNZu2bHlJcoYM6CaIwOT\nb6tkoexUyOJOx1dCIthki5yBGybHeZmAJXCH6rjUSeMu0PXiWiNKSxZCNXAmTuDKdU7Ight39o6J\nrtdM7rLErlLyW7bsL0TWpz7F+ZbDIT2ljEpkQm9R8teIEuhy+dg2lRsdde1ZWKcMP5yVdjXdHXU1\nyVwDAAu0NFLq2f5OYucbUundxS4t31BhFnr8QOocUThNQn8U0rftIkn2itrFvUEzq/iZnNZq8Zmi\n4zgO4YOi4zgOMbOpw/5fROQcgA6A16fdlivAPng/88S89PMGVd3/5ofNF7kdFAFARP6kqh+cdjve\nabyf+eJq6ees4vLZcRyH8EHRcRyHyPugeGTaDbhCeD/zxdXSz5kk12uKjuM4b5W8zxQdx3HeEj4o\nOo7jELkdFEXkEyLydxF5OZRFzQUicp2IPCciL4nIiyLycNifu9rYIlIUkRdE5Nfh9U0icizc05+F\nWj5zj4gsichREfmbiBwXkdvyeD/nhVwOiqFm9PeR1Xo5DOA+ETk83VZdNhIAX1HVwwA+AuCh0LdH\nATyjqjcDeCa8nnceBnCcXn8LwHdU9T0A1gE8MJVWXX6+B+A3qvpeAO9D1uc83s+5IJeDIoAPAXhZ\nVU+o6hDAT5HVjp57VPWUqj4ftlvI/oCuRc5qY4vIIQCfBvBEeC0A7gRwNBwy930EABFpAvgogCcB\nQFWHqrqBnN3PeSKvg+K1AF6l1yfDvlwhIjcCuBXAMfyPtbHniO8C+BqwXeBkFcCG6nbKm7zc05sA\nnAPwo7BU8ESoU5S3+zk35HVQzD0i0gDwcwCPqOoWv6eZn9Xc+lqJyN0Azqrqn6fdlitACcAHAPxA\nVW9FFq+/QyrP+/2cN/I6KL4G4Dp6fSjsywUiEiEbEH+sqr8Iu/NUG/t2APeIyL+RLX3ciWzdbUlE\nJon98nJPTwI4qarHwuujyAbJPN3PuSKvg+IfAdwcrJUxgHuR1Y6ee8La2pMAjqvqt+mt3NTGVtXH\nVPWQqt6I7N49q6pfAPAcgM+Fw+a6jxNU9TSAV0XklrDrLgAvIUf3c97IbUSLiHwK2bpUEcAPVfWb\nU27SZUFE7gDwewB/ha23fQPZuuLTAK5HqI2tqhem0sjLiIh8DMBXVfVuEXk3spnjCoAXAHxRVQeX\n+vw8ICLvR2ZQigGcAPAlZBOW3N3PeSC3g6LjOM7bIa/y2XEc523hg6LjOA7hg6LjOA7hg6LjOA7h\ng6LjOA7hg6LjOA7hg6LjOA7xX7H03WNrFdeIAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f85f896be10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_holdout2[2], cmap='gray', interpolation='nearest');\n",
    "tl = \"Actual label : \" +str(Y_holdout[38])+ \",\"+\" iceberg_probability : \"+str(pred_valid[2])\n",
    "plt.title(tl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7f85f88af320>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUQAAAEICAYAAAAncI3RAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f85f88cd4a8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_holdout2[3], cmap='gray', interpolation='nearest');\n",
    "tl = \"Actual label : \" +str(Y_holdout[43])+ \",\"+\" iceberg_probability : \"+str(pred_valid[3])\n",
    "plt.title(tl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7f85f880fdd8>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUUAAAEICAYAAADIsubvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n4ziED4qO4ziED4qO4ziED4qO4zjE/wA4Jsme5lE5gwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f85f886dba8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_holdout2[6], cmap='gray', interpolation='nearest');\n",
    "tl = \"Actual label : \" +str(Y_holdout[104])+ \",\"+\" iceberg_probability : \"+str(pred_valid[6])\n",
    "plt.title(tl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7f85f87f57f0>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUUAAAEICAYAAADIsubvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ncRzBJ0XHcRzhv1FqPa0N5gMcAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f85f87d17f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_holdout2[9], cmap='gray', interpolation='nearest');\n",
    "tl = \"Actual label : \" +str(Y_holdout[138])+ \",\"+\" iceberg_probability : \"+str(pred_valid[9])\n",
    "plt.title(tl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7f85f8758ef0>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQsAAAEICAYAAABIwdH8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f85f87b15c0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_holdout2[10], cmap='gray', interpolation='nearest');\n",
    "tl = \"Actual label : \" +str(Y_holdout[141])+ \",\"+\" iceberg_probability : \"+str(pred_valid[10])\n",
    "plt.title(tl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7f85f873e7f0>"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUUAAAEICAYAAADIsubvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f85f8717f98>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_holdout2[11], cmap='gray', interpolation='nearest');\n",
    "tl = \"Actual label : \" +str(Y_holdout[167])+ \",\"+\" iceberg_probability : \"+str(pred_valid[11])\n",
    "plt.title(tl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7f85f86a60b8>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUUAAAEICAYAAADIsubvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f85f86f6828>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_holdout2[12], cmap='gray', interpolation='nearest');\n",
    "tl = \"Actual label : \" +str(Y_holdout[195])+ \",\"+\" iceberg_probability : \"+str(pred_valid[12])\n",
    "plt.title(tl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7f85f8608978>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUUAAAEICAYAAADIsubvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f85f8665c18>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_holdout2[13], cmap='gray', interpolation='nearest');\n",
    "tl = \"Actual label : \" +str(Y_holdout[199])+ \",\"+\" iceberg_probability : \"+str(pred_valid[13])\n",
    "plt.title(tl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "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
}