{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import keras"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Input',\n",
       " 'Model',\n",
       " 'Sequential',\n",
       " '__builtins__',\n",
       " '__cached__',\n",
       " '__doc__',\n",
       " '__file__',\n",
       " '__loader__',\n",
       " '__name__',\n",
       " '__package__',\n",
       " '__path__',\n",
       " '__spec__',\n",
       " '__version__',\n",
       " 'absolute_import',\n",
       " 'activations',\n",
       " 'applications',\n",
       " 'backend',\n",
       " 'callbacks',\n",
       " 'constraints',\n",
       " 'datasets',\n",
       " 'engine',\n",
       " 'initializers',\n",
       " 'layers',\n",
       " 'legacy',\n",
       " 'losses',\n",
       " 'metrics',\n",
       " 'models',\n",
       " 'optimizers',\n",
       " 'preprocessing',\n",
       " 'regularizers',\n",
       " 'utils',\n",
       " 'wrappers']"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dir(keras)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Convolutional Neural Network\n",
    "\n",
    "* The patterns they learn are translation invariant.\n",
    "* They can learn spatial hierarchies of patterns:\n",
    "    A first convolution layer will learn small local patterns such as edges, a second convolution layer will learn larger patterns made of the features of the first layers, and so on. This allows convnets to efficiently learn increasingly complex and abstract visual concepts (because the visual world is fundamentally spatially hierarchical).\n",
    "* Highly efficient on perceptual problems\n",
    "* Dense layers learn global patterns in their input feature space (for example, for a MNIST digit, patterns involving all pixels), whereas convolution layers learn local patterns: in the case of images, patterns found in small 2D windows of the inputs."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![alt text](cat.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create a small CNN for classifying the MNIST dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras import layers\n",
    "from keras import models\n",
    "\n",
    "# Create a sequential model\n",
    "model = models.Sequential() \n",
    "model.add(layers.Conv2D(32, (3, 3), activation='relu', input_shape=(28, 28, 1))) # no: of channels/output_depth, window_height, window_width\n",
    "# Output of conv layer above is a feature map of size output_height, output_width, output_depth. Output_width/height different from input_height and width due to border effects\n",
    "model.add(layers.MaxPooling2D((2, 2))) # This layer downsamples the feature maps input to it, using windows of size 2x2 and stride 2. This takes the maximum value of each window\n",
    "# The above layer downsampled by a factor of 2 in each direction. Max pooling layers are necessary for learning spatial hierarchies\n",
    "model.add(layers.Conv2D(64, (3, 3), activation='relu')) # Relu functions have sparse activations\n",
    "model.add(layers.MaxPooling2D((2, 2)))\n",
    "model.add(layers.Conv2D(64, (3, 3), activation='relu')) # Note that the depth of the feature maps gradually increase while the size of the feature maps gradually decrease\n",
    "\n",
    "model.add(layers.Flatten())\n",
    "model.add(layers.Dense(64, activation='relu'))\n",
    "model.add(layers.Dense(10, activation='softmax'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![alt text](conv.png)\n",
    "\n",
    "![alt text](features.png)\n",
    "\n",
    "![alt text](layers2.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_1 (Conv2D)            (None, 26, 26, 32)        320       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 13, 13, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 11, 11, 64)        18496     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2 (None, 5, 5, 64)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 3, 3, 64)          36928     \n",
      "_________________________________________________________________\n",
      "flatten_1 (Flatten)          (None, 576)               0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 64)                36928     \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 10)                650       \n",
      "=================================================================\n",
      "Total params: 93,322\n",
      "Trainable params: 93,322\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary() "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Because of the shape of the sigmoid function, extreme values as inputs to the sigmoid function tend to result in smaller partial derivatives for that cost function. This results in slower learning when the inputs are really far away from zero if the objective is to bring the cost function to zero.\n",
    "\n",
    "It turns out that we can solve the problem by replacing the quadratic cost with a different cost function, known as the cross-entropy.\n",
    "\n",
    "For a simple neural network output given by 'a'\n",
    "\n",
    "$a = sigma(z)$\n",
    "\n",
    "where \n",
    "\n",
    "$z = w*x + b$\n",
    "\n",
    "Let the desired output be y, then the Cross entropy cost is given by\n",
    "\n",
    "$cost = y*ln(a) + (1-y)*ln(1 -a)$\n",
    "\n",
    "This satisfies all the properties of a cost function: when y = 1 and a close to 1, cost = 0\n",
    "and when y = 0 and a is close to 0, cost = 0\n",
    "\n",
    "Partial derivative of this cost is given by  $x * (sigma(z) - y)$\n",
    "This is great because now the learning rate is proportional to (a - y), so the farther a is from y, the larger the gradient.\n",
    "\n",
    "See here for details http://neuralnetworksanddeeplearning.com/chap3.html"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load and train on the MNIST Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "60000/60000 [==============================] - 27s - loss: 0.1700 - acc: 0.9473    \n",
      "Epoch 2/5\n",
      "60000/60000 [==============================] - 32s - loss: 0.0488 - acc: 0.9856    \n",
      "Epoch 3/5\n",
      "60000/60000 [==============================] - 27s - loss: 0.0346 - acc: 0.9896    \n",
      "Epoch 4/5\n",
      "60000/60000 [==============================] - 27s - loss: 0.0267 - acc: 0.9919    \n",
      "Epoch 5/5\n",
      "60000/60000 [==============================] - 29s - loss: 0.0213 - acc: 0.9936    \n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1a1acaeba8>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from keras.datasets import mnist\n",
    "from keras.utils import to_categorical\n",
    "\n",
    "(train_images, train_labels), (test_images, test_labels) = mnist.load_data()\n",
    "train_images = train_images.reshape((60000, 28, 28, 1))\n",
    "train_images = train_images.astype('float32') / 255\n",
    "\n",
    "test_images = test_images.reshape((10000, 28, 28, 1))\n",
    "test_images = test_images.astype('float32') / 255\n",
    "\n",
    "train_labels = to_categorical(train_labels)\n",
    "test_labels = to_categorical(test_labels)\n",
    "\n",
    "model.compile(optimizer='rmsprop',\n",
    "              loss='categorical_crossentropy',\n",
    "              metrics=['accuracy'])\n",
    "model.fit(train_images, train_labels, epochs=5, batch_size=64)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "** Test the accuracy **"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "** Accuracy is measured as (true_positives + true_negatives) / Total_instances **"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " 9856/10000 [============================>.] - ETA: 0s0.0361112085581 0.9902\n"
     ]
    }
   ],
   "source": [
    "test_loss, test_acc = model.evaluate(test_images, test_labels)\n",
    "print(test_loss,test_acc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## What happens when you don't have a lot of data ??"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Having to train an image-classification model using very little data is a common situation\n",
    "* Fortunately there are a few techniques to deal with this\n",
    "* Data augmentation ( We will only cover this in our example today )\n",
    "* Fine-tuning a pretrained network (Not covered)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  Cats and Dogs example from Kaggle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "ename": "FileNotFoundError",
     "evalue": "[Errno 2] No such file or directory: '/home/mapped/dogs-vs-cats-redux-kernels-edition/train/cat.0.jpg'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mFileNotFoundError\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-8-7694800ac5bb>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     48\u001b[0m     \u001b[0msrc\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mos\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpath\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mjoin\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moriginal_dataset_dir\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfname\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     49\u001b[0m     \u001b[0mdst\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mos\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpath\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mjoin\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrain_cats_dir\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfname\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 50\u001b[0;31m     \u001b[0mshutil\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcopyfile\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msrc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdst\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     51\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     52\u001b[0m \u001b[0mfnames\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m'cat.{}.jpg'\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m1000\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m1500\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/anaconda3/lib/python3.6/shutil.py\u001b[0m in \u001b[0;36mcopyfile\u001b[0;34m(src, dst, follow_symlinks)\u001b[0m\n\u001b[1;32m    118\u001b[0m         \u001b[0mos\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msymlink\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mos\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreadlink\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msrc\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdst\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    119\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 120\u001b[0;31m         \u001b[0;32mwith\u001b[0m \u001b[0mopen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msrc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'rb'\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mfsrc\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    121\u001b[0m             \u001b[0;32mwith\u001b[0m \u001b[0mopen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdst\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'wb'\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mfdst\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    122\u001b[0m                 \u001b[0mcopyfileobj\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfsrc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfdst\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mFileNotFoundError\u001b[0m: [Errno 2] No such file or directory: '/home/mapped/dogs-vs-cats-redux-kernels-edition/train/cat.0.jpg'"
     ]
    }
   ],
   "source": [
    "# Download the data\n",
    "import os, shutil\n",
    "\n",
    "original_dataset_dir = '/home/mapped/dogs-vs-cats-redux-kernels-edition/train/'    \n",
    "\n",
    "base_dir = '/home/srijith/cats_and_dogs_small/'                 \n",
    "\n",
    "train_dir = os.path.join(base_dir, 'train')\n",
    "if(os.path.isdir(train_dir) == False):\n",
    "    os.mkdir(train_dir)\n",
    "validation_dir = os.path.join(base_dir, 'validation')\n",
    "\n",
    "if(os.path.isdir(validation_dir) == False):\n",
    "    os.mkdir(validation_dir)\n",
    "    \n",
    "test_dir = os.path.join(base_dir, 'test')\n",
    "\n",
    "if(os.path.isdir(test_dir) == False):\n",
    "    os.mkdir(test_dir)\n",
    "\n",
    "train_cats_dir = os.path.join(train_dir, 'cats')\n",
    "\n",
    "if(os.path.isdir(train_cats_dir) == False):\n",
    "    os.mkdir(train_cats_dir)                                                   \n",
    "\n",
    "train_dogs_dir = os.path.join(train_dir, 'dogs')\n",
    "if(os.path.isdir(train_dogs_dir) == False):\n",
    "    os.mkdir(train_dogs_dir)                                                   \n",
    "\n",
    "validation_cats_dir = os.path.join(validation_dir, 'cats')\n",
    "if(os.path.isdir(validation_cats_dir) == False):\n",
    "    os.mkdir(validation_cats_dir)                                              \n",
    "\n",
    "validation_dogs_dir = os.path.join(validation_dir, 'dogs')\n",
    "if(os.path.isdir(validation_dogs_dir) == False):\n",
    "    os.mkdir(validation_dogs_dir)                                              \n",
    "\n",
    "test_cats_dir = os.path.join(test_dir, 'cats')\n",
    "if(os.path.isdir(test_cats_dir) == False):\n",
    "    os.mkdir(test_cats_dir)                                                    \n",
    " \n",
    "test_dogs_dir = os.path.join(test_dir, 'dogs')\n",
    "if(os.path.isdir(test_dogs_dir) == False):\n",
    "    os.mkdir(test_dogs_dir)                                                  \n",
    "\n",
    "fnames = ['cat.{}.jpg'.format(i) for i in range(1000)]                     \n",
    "for fname in fnames:                                                       \n",
    "    src = os.path.join(original_dataset_dir, fname)                        \n",
    "    dst = os.path.join(train_cats_dir, fname)                              \n",
    "    shutil.copyfile(src, dst)                                              \n",
    "\n",
    "fnames = ['cat.{}.jpg'.format(i) for i in range(1000, 1500)]               \n",
    "for fname in fnames:                                                       \n",
    "    src = os.path.join(original_dataset_dir, fname)                        \n",
    "    dst = os.path.join(validation_cats_dir, fname)                         \n",
    "    shutil.copyfile(src, dst)                                              \n",
    "\n",
    "fnames = ['cat.{}.jpg'.format(i) for i in range(1500, 2000)]               \n",
    "for fname in fnames:                                                       \n",
    "    src = os.path.join(original_dataset_dir, fname)                        \n",
    "    dst = os.path.join(test_cats_dir, fname)                               \n",
    "    shutil.copyfile(src, dst)                                              \n",
    "\n",
    "fnames = ['dog.{}.jpg'.format(i) for i in range(1000)]                     \n",
    "for fname in fnames:                                                       \n",
    "    src = os.path.join(original_dataset_dir, fname)                        \n",
    "    dst = os.path.join(train_dogs_dir, fname)                              \n",
    "    shutil.copyfile(src, dst)                                              \n",
    "fnames = ['dog.{}.jpg'.format(i) for i in range(1000, 1500)]               \n",
    "for fname in fnames:                                                       \n",
    "    src = os.path.join(original_dataset_dir, fname)                        \n",
    "    dst = os.path.join(validation_dogs_dir, fname)                         \n",
    "    shutil.copyfile(src, dst)                                              \n",
    "\n",
    "fnames = ['dog.{}.jpg'.format(i) for i in range(1500, 2000)]               \n",
    "for fname in fnames:                                                       \n",
    "    src = os.path.join(original_dataset_dir, fname)                        \n",
    "    dst = os.path.join(test_dogs_dir, fname)                               \n",
    "    shutil.copyfile(src, dst)                                              "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Check the files**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "total training cat images: 1000\n",
      "total training dog images: 1000\n",
      "total validation cat images: 500\n",
      "total validation dog images: 500\n",
      "total test cat images: 500\n",
      "total test dog images: 500\n"
     ]
    }
   ],
   "source": [
    "print('total training cat images:', len(os.listdir(train_cats_dir)))\n",
    "print('total training dog images:', len(os.listdir(train_dogs_dir)))\n",
    "print('total validation cat images:', len(os.listdir(validation_cats_dir)))\n",
    "print('total validation dog images:', len(os.listdir(validation_dogs_dir)))\n",
    "print('total test cat images:', len(os.listdir(test_cats_dir)))\n",
    "print('total test dog images:', len(os.listdir(test_dogs_dir)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "** We will reuse the same general structure: the convnet will be a stack of alternated Conv2D (with relu activation) and MaxPooling2D layers. But because you’re dealing with bigger images and a more complex problem, you’ll make your network larger, accordingly: it will have one more Conv2D + MaxPooling2D stage.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_7 (Conv2D)            (None, 148, 148, 32)      896       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_5 (MaxPooling2 (None, 74, 74, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_8 (Conv2D)            (None, 72, 72, 64)        18496     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_6 (MaxPooling2 (None, 36, 36, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_9 (Conv2D)            (None, 34, 34, 128)       73856     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_7 (MaxPooling2 (None, 17, 17, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv2d_10 (Conv2D)           (None, 15, 15, 128)       147584    \n",
      "_________________________________________________________________\n",
      "max_pooling2d_8 (MaxPooling2 (None, 7, 7, 128)         0         \n",
      "_________________________________________________________________\n",
      "flatten_3 (Flatten)          (None, 6272)              0         \n",
      "_________________________________________________________________\n",
      "dense_5 (Dense)              (None, 512)               3211776   \n",
      "_________________________________________________________________\n",
      "dense_6 (Dense)              (None, 1)                 513       \n",
      "=================================================================\n",
      "Total params: 3,453,121\n",
      "Trainable params: 3,453,121\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "from keras import layers\n",
    "from keras import models\n",
    "\n",
    "model = models.Sequential()\n",
    "model.add(layers.Conv2D(32, (3, 3), activation='relu',\n",
    "                        input_shape=(150, 150, 3))) # image size is 150x150 and 3 rgb channels\n",
    "model.add(layers.MaxPooling2D((2, 2)))\n",
    "model.add(layers.Conv2D(64, (3, 3), activation='relu'))\n",
    "model.add(layers.MaxPooling2D((2, 2)))\n",
    "model.add(layers.Conv2D(128, (3, 3), activation='relu'))\n",
    "model.add(layers.MaxPooling2D((2, 2)))\n",
    "model.add(layers.Conv2D(128, (3, 3), activation='relu'))\n",
    "model.add(layers.MaxPooling2D((2, 2)))\n",
    "model.add(layers.Flatten())\n",
    "model.add(layers.Dense(512, activation='relu'))\n",
    "model.add(layers.Dense(1, activation='sigmoid'))\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Compile the model\n",
    "from keras import optimizers\n",
    "\n",
    "model.compile(loss='binary_crossentropy',\n",
    "              optimizer=optimizers.RMSprop(lr=1e-4), #lr is the learning rate\n",
    "              metrics=['acc'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 2000 images belonging to 2 classes.\n",
      "Found 1000 images belonging to 2 classes.\n"
     ]
    }
   ],
   "source": [
    "# Convert the data from JPEGs to floating point numbers\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "\n",
    "train_datagen = ImageDataGenerator(rescale=1./255)             \n",
    "test_datagen = ImageDataGenerator(rescale=1./255)              \n",
    "\n",
    "train_generator = train_datagen.flow_from_directory(\n",
    "        train_dir,                                             \n",
    "        target_size=(150, 150),                                 \n",
    "        batch_size=20,\n",
    "        class_mode='binary')                                   \n",
    "\n",
    "validation_generator = test_datagen.flow_from_directory(validation_dir,target_size=(150, 150),batch_size=20,class_mode='binary')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/30\n",
      "100/100 [==============================] - 46s - loss: 0.6915 - acc: 0.5195 - val_loss: 0.6773 - val_acc: 0.5130\n",
      "Epoch 2/30\n",
      "100/100 [==============================] - 41s - loss: 0.6485 - acc: 0.6155 - val_loss: 0.6307 - val_acc: 0.6350\n",
      "Epoch 3/30\n",
      "100/100 [==============================] - 43s - loss: 0.5909 - acc: 0.6770 - val_loss: 0.6060 - val_acc: 0.6710\n",
      "Epoch 4/30\n",
      "100/100 [==============================] - 39s - loss: 0.5532 - acc: 0.7030 - val_loss: 0.6019 - val_acc: 0.6640\n",
      "Epoch 5/30\n",
      "100/100 [==============================] - 39s - loss: 0.5342 - acc: 0.7195 - val_loss: 0.6286 - val_acc: 0.6500\n",
      "Epoch 6/30\n",
      "100/100 [==============================] - 39s - loss: 0.5082 - acc: 0.7580 - val_loss: 0.5848 - val_acc: 0.6870\n",
      "Epoch 7/30\n",
      "100/100 [==============================] - 39s - loss: 0.4860 - acc: 0.7665 - val_loss: 0.5766 - val_acc: 0.6970\n",
      "Epoch 8/30\n",
      "100/100 [==============================] - 39s - loss: 0.4551 - acc: 0.7810 - val_loss: 0.5862 - val_acc: 0.6980\n",
      "Epoch 9/30\n",
      "100/100 [==============================] - 39s - loss: 0.4291 - acc: 0.8025 - val_loss: 0.6139 - val_acc: 0.6970\n",
      "Epoch 10/30\n",
      "100/100 [==============================] - 39s - loss: 0.4043 - acc: 0.8125 - val_loss: 0.5708 - val_acc: 0.7140\n",
      "Epoch 11/30\n",
      "100/100 [==============================] - 39s - loss: 0.3770 - acc: 0.8355 - val_loss: 0.5513 - val_acc: 0.7200\n",
      "Epoch 12/30\n",
      "100/100 [==============================] - 39s - loss: 0.3559 - acc: 0.8415 - val_loss: 0.5498 - val_acc: 0.7300\n",
      "Epoch 13/30\n",
      "100/100 [==============================] - 39s - loss: 0.3390 - acc: 0.8565 - val_loss: 0.5636 - val_acc: 0.7290\n",
      "Epoch 14/30\n",
      "100/100 [==============================] - 39s - loss: 0.3087 - acc: 0.8700 - val_loss: 0.5520 - val_acc: 0.7420\n",
      "Epoch 15/30\n",
      "100/100 [==============================] - 39s - loss: 0.2837 - acc: 0.8955 - val_loss: 0.5692 - val_acc: 0.7370\n",
      "Epoch 16/30\n",
      "100/100 [==============================] - 39s - loss: 0.2661 - acc: 0.8915 - val_loss: 0.5848 - val_acc: 0.7250\n",
      "Epoch 17/30\n",
      "100/100 [==============================] - 39s - loss: 0.2355 - acc: 0.9085 - val_loss: 0.6028 - val_acc: 0.7240\n",
      "Epoch 18/30\n",
      "100/100 [==============================] - 39s - loss: 0.2166 - acc: 0.9170 - val_loss: 0.6191 - val_acc: 0.7320\n",
      "Epoch 19/30\n",
      "100/100 [==============================] - 39s - loss: 0.1917 - acc: 0.9305 - val_loss: 0.6555 - val_acc: 0.7320\n",
      "Epoch 20/30\n",
      "100/100 [==============================] - 39s - loss: 0.1784 - acc: 0.9360 - val_loss: 0.6353 - val_acc: 0.7320\n",
      "Epoch 21/30\n",
      "100/100 [==============================] - 39s - loss: 0.1500 - acc: 0.9480 - val_loss: 0.6903 - val_acc: 0.7380\n",
      "Epoch 22/30\n",
      "100/100 [==============================] - 39s - loss: 0.1413 - acc: 0.9505 - val_loss: 0.6936 - val_acc: 0.7350\n",
      "Epoch 23/30\n",
      "100/100 [==============================] - 39s - loss: 0.1283 - acc: 0.9535 - val_loss: 0.7399 - val_acc: 0.7300\n",
      "Epoch 24/30\n",
      "100/100 [==============================] - 39s - loss: 0.1070 - acc: 0.9670 - val_loss: 0.7423 - val_acc: 0.7330\n",
      "Epoch 25/30\n",
      "100/100 [==============================] - 39s - loss: 0.0902 - acc: 0.9715 - val_loss: 0.9947 - val_acc: 0.7030\n",
      "Epoch 26/30\n",
      "100/100 [==============================] - 39s - loss: 0.0809 - acc: 0.9775 - val_loss: 0.7635 - val_acc: 0.7360\n",
      "Epoch 27/30\n",
      "100/100 [==============================] - 39s - loss: 0.0661 - acc: 0.9810 - val_loss: 1.1880 - val_acc: 0.6940\n",
      "Epoch 28/30\n",
      "100/100 [==============================] - 39s - loss: 0.0589 - acc: 0.9815 - val_loss: 0.9458 - val_acc: 0.7210\n",
      "Epoch 29/30\n",
      "100/100 [==============================] - 39s - loss: 0.0460 - acc: 0.9875 - val_loss: 1.2532 - val_acc: 0.7060\n",
      "Epoch 30/30\n",
      "100/100 [==============================] - 39s - loss: 0.0396 - acc: 0.9885 - val_loss: 1.0535 - val_acc: 0.7260\n"
     ]
    }
   ],
   "source": [
    "history = model.fit_generator(\n",
    "      train_generator,\n",
    "      steps_per_epoch=100,   # batch size of 20 times steps per epoch of 100 = training data size 2000\n",
    "      epochs=30,\n",
    "      validation_data=validation_generator,\n",
    "      validation_steps=50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model.save('cats_and_dogs_small_1.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEICAYAAACktLTqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f02ef00a668>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEICAYAAACktLTqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3Xl8VPW5+PHPQwg7sisKQgA3trBFwCJCJApqlctSK+KC\nV4tYra3e65Wr1lotrfqjSrGUlrZSFZRyxQUVRSmpaFVIoBDWAELQsIOyGRRCnt8f3zNkCJnkTDLJ\nLHner9e8Zs6ZM+d8Tw48853nuxxRVYwxxiSWWtEugDHGmMiz4G6MMQnIgrsxxiQgC+7GGJOALLgb\nY0wCsuBujDEJyIK7KZWIJInIERFpF8lto0lEzhORiPf9FZEMEckLWs4VkYF+tq3Asf4iIg9V9PNl\n7PdXIvK3SO/XRE/taBfARIaIHAlabAB8B5zwlu9U1dnh7E9VTwCNIr1tTaCqF0ZiPyJyB3CTqg4O\n2vcdkdi3SXwW3BOEqp4Mrl7N8A5VXRRqexGpraqF1VE2Y0z1s7RMDeH97P67iLwiIoeBm0TkEhH5\nTEQOiMhOEZkqIsne9rVFREUkxVue5b3/rogcFpFPRaRDuNt6718lIhtF5KCIPCci/xKRcSHK7aeM\nd4rIZhH5WkSmBn02SUSeFZH9IrIFGFbG3+dhEZlTYt00EXnGe32HiKz3zudzr1Ydal/5IjLYe91A\nRF7yyrYW6FNi20dEZIu337Uicp23vjvwe2Cgl/LaF/S3fSzo8xO8c98vIm+IyNl+/jblEZERXnkO\niMhiEbkw6L2HRGSHiBwSkQ1B59pfRFZ463eLyP/zezxTBVTVHgn2APKAjBLrfgUcA67FfanXBy4G\n+uF+wXUENgL3eNvXBhRI8ZZnAfuANCAZ+DswqwLbngkcBoZ7790PHAfGhTgXP2V8E2gCpABfBc4d\nuAdYC7QFWgBL3D/5Uo/TETgCNAza9x4gzVu+1ttGgMuBo0Cq914GkBe0r3xgsPd6MvBPoBnQHlhX\nYtvrgbO9a3KjV4azvPfuAP5ZopyzgMe811d6ZewJ1AP+ACz287cp5fx/BfzNe93ZK8fl3jV6CMj1\nXncFtgGtvW07AB2911nAGO91Y6BftP8v1OSH1dxrlo9V9S1VLVLVo6qapapLVbVQVbcAM4BBZXz+\nVVXNVtXjwGxcUAl32+8DK1X1Te+9Z3FfBKXyWcbfqOpBVc3DBdLAsa4HnlXVfFXdDzxZxnG2AGtw\nXzoAVwBfq2q29/5bqrpFncXAP4BSG01LuB74lap+rarbcLXx4OPOVdWd3jV5GffFnOZjvwBjgb+o\n6kpV/RaYCAwSkbZB24T625TlBmC+qi72rtGTuC+IfkAh7oukq5fa2+r97cB9SZ8vIi1U9bCqLvV5\nHqYKWHCvWb4MXhCRi0TkHRHZJSKHgMeBlmV8flfQ6wLKbkQNte05weVQVcXVdEvls4y+joWrcZbl\nZWCM9/pGbzlQju+LyFIR+UpEDuBqzWX9rQLOLqsMIjJORFZ56Y8DwEU+9wvu/E7uT1UPAV8DbYK2\nCeeahdpvEe4atVHVXOC/cNdhj5fma+1tehvQBcgVkWUicrXP8zBVwIJ7zVKyG+CfcLXV81T1DOBR\nXNqhKu3EpUkAEBHh1GBUUmXKuBM4N2i5vK6ac4EMEWmDq8G/7JWxPvAq8BtcyqQp8L7PcuwKVQYR\n6QhMB+4CWnj73RC03/K6be7ApXoC+2uMS/9s91GucPZbC3fNtgOo6ixVHYBLySTh/i6oaq6q3oBL\nvf0WmCci9SpZFlNBFtxrtsbAQeAbEekM3FkNx3wb6C0i14pIbeCnQKsqKuNc4Gci0kZEWgAPlrWx\nqu4CPgb+BuSq6ibvrbpAHWAvcEJEvg8MCaMMD4lIU3HjAO4Jeq8RLoDvxX3P/QhXcw/YDbQNNCCX\n4hXgdhFJFZG6uCD7kaqG/CUURpmvE5HB3rEfwLWTLBWRziKS7h3vqPcowp3AzSLS0qvpH/TOraiS\nZTEVZMG9Zvsv4Fbcf9w/4Ro+q5Sq7gZ+CDwD7Ac6Af/G9cuPdBmn43Ljq3GNfa/6+MzLuAbSkykZ\nVT0A3Ae8jmuUHI37kvLjF7hfEHnAu8CLQfvNAZ4DlnnbXAgE56k/ADYBu0UkOL0S+Px7uPTI697n\n2+Hy8JWiqmtxf/PpuC+eYcB1Xv69LvA0rp1kF+6XwsPeR68G1ovrjTUZ+KGqHqtseUzFiEt5GhMd\nIpKESwOMVtWPol0eYxKF1dxNtRORYV6aoi7wc1wvi2VRLpYxCcWCu4mGS4EtuJ/8Q4ERqhoqLWOM\nqQBLyxhjTAKymrsxxiSgqE0c1rJlS01JSYnW4Y0xJi4tX758n6qW1X0YiGJwT0lJITs7O1qHN8aY\nuCQi5Y20BiwtY4wxCcmCuzHGJCAL7sYYk4Bi6k5Mx48fJz8/n2+//TbaRTE+1KtXj7Zt25KcHGrq\nE2NMtMRUcM/Pz6dx48akpKTgJgs0sUpV2b9/P/n5+XTo0KH8DxhjqlVMpWW+/fZbWrRoYYE9DogI\nLVq0sF9ZxsSomArugAX2OGLXypjYFXPB3Rhj4s3OnTBnTvnbVScL7kH2799Pz5496dmzJ61bt6ZN\nmzYnl48d8zct9W233UZubm6Z20ybNo3Zs2dHoshceumlrFy5MiL7MsZUzJQpMGYMrF8f7ZIUi+vg\nPns2pKRArVruubLxskWLFqxcuZKVK1cyYcIE7rvvvpPLderUAVxDYlFR6JvLzJw5kwsvvLDM49x9\n992MHVvpeyoYY2LEMm/C6tdei245gsVtcJ89G8aPh23bQNU9jx9f+QBfms2bN9OlSxfGjh1L165d\n2blzJ+PHjyctLY2uXbvy+OOPn9w2UJMuLCykadOmTJw4kR49enDJJZewZ88eAB555BGmTJlycvuJ\nEyfSt29fLrzwQj755BMAvvnmG0aNGkWXLl0YPXo0aWlp5dbQZ82aRffu3enWrRsPPfQQAIWFhdx8\n880n10+dOhWAZ599li5dupCamspNN90U8b+ZMTXFiRMQmEklloJ7THWFDMfDD0NBwanrCgrc+qqo\nFG/YsIEXX3yRtLQ0AJ588kmaN29OYWEh6enpjB49mi5dupzymYMHDzJo0CCefPJJ7r//fp5//nkm\nTpx42r5VlWXLljF//nwef/xx3nvvPZ577jlat27NvHnzWLVqFb179y6zfPn5+TzyyCNkZ2fTpEkT\nMjIyePvtt2nVqhX79u1j9erVABw4cACAp59+mm3btlGnTp2T64wx4duwAY4cgdRUWLECtm6FWOgd\nHLc19y++CG99ZXXq1OlkYAd45ZVX6N27N71792b9+vWsW7futM/Ur1+fq666CoA+ffqQl5dX6r5H\njhx52jYff/wxN9xwAwA9evSga9euZZZv6dKlXH755bRs2ZLk5GRuvPFGlixZwnnnnUdubi733nsv\nCxcupEmTJgB07dqVm266idmzZ9sgJGMqISvLPf/61+45VmrvcRvc27ULb31lNWzY8OTrTZs28bvf\n/Y7FixeTk5PDsGHDSu3vHcjTAyQlJVFYWFjqvuvWrVvuNhXVokULcnJyGDhwINOmTePOO+8EYOHC\nhUyYMIGsrCz69u3LiRMnInpcY2qKZcugcWO46iro2TOOgruIPC8ie0RkTYj3x4pIjoisFpFPRKRH\n5It5ukmToEGDU9c1aODWV7VDhw7RuHFjzjjjDHbu3MnChQsjfowBAwYwd+5cAFavXl3qL4Ng/fr1\nIzMzk/3791NYWMicOXMYNGgQe/fuRVX5wQ9+wOOPP86KFSs4ceIE+fn5XH755Tz99NPs27ePgpI5\nLmOML8uWQVqa69gxahR88gns2BHtUvnLuf8N+D3wYoj3twKDVPVrEbkKmAH0i0zxQgvk1R9+2KVi\n2rVzgb06OqH07t2bLl26cNFFF9G+fXsGDBgQ8WP85Cc/4ZZbbqFLly4nH4GUSmnatm3LE088weDB\ng1FVrr32Wq655hpWrFjB7bffjqoiIjz11FMUFhZy4403cvjwYYqKivjv//5vGjduHPFzMCbRffst\n5OTA/fe75ZEj4ec/hzfegB//OLpl83UPVRFJAd5W1W7lbNcMWKOqbcrbZ1pampa8Wcf69evp3Llz\nueWpCQoLCyksLKRevXps2rSJK6+8kk2bNlG7dmy1gds1MzXZ0qXQvz/Mm+cCO0DnznDOOfCPf1TN\nMUVkuaqmlbddpCPF7cC7od4UkfHAeIB2VZUcTxBHjhxhyJAhFBYWoqr86U9/irnAbkxNF+jffvHF\nxetGjoSnnoJ9+6Bly+iUCyLYoCoi6bjg/mCobVR1hqqmqWpaq1bl3gKwRmvatCnLly9n1apV5OTk\ncOWVV0a7SMYkhMxM1/jpc9B5mbKyoHVraNu2eN2oUa7v+/z5ld9/ZUQkuItIKvAXYLiq7o/EPo0x\npiq8+Sa8915xrbsyli2Dvn0heA69Xr3ciPl58yq//8qodHAXkXbAa8DNqrqx8kUyxpiqE5j66YMP\nKrefgwfdvoJTMuAC/ciRsGiR2yZa/HSFfAX4FLhQRPJF5HYRmSAiE7xNHgVaAH8QkZUikh1yZ8YY\nE2UbvSrookWV20+gP0jfvqe/N2qUS/u8807ljlEZ5bbQqeqYct6/A7gjYiUyxpgq8t13kJfnxsQs\nXQqHDsEZZ1RsX4G0Tlop/Vb694ezz3YDmm68scLFrZS4HaFaFdLT008bkDRlyhTuuuuuMj/XqFEj\nAHbs2MHo0aNL3Wbw4MGU7PpZ0pQpU04ZTHT11VdHZN6Xxx57jMmTJ1d6P8bEu82boajIjYc5cQI+\n/LDi+8rKgvPOg+bNT3+vVi0YMQLefff0ObCqiwX3IGPGjGFOiRn358yZw5gxZf54Oemcc87h1Vdf\nrfDxSwb3BQsW0LRp0wrvzxhzqkC+fdw4qF+/cqmZQGNqKCNHusD+3nsVP0ZlWHAPMnr0aN55552T\nN+bIy8tjx44dDBw48GS/8969e9O9e3fefPPN0z6fl5dHt25unNfRo0e54YYb6Ny5MyNGjODo0aMn\nt7vrrrtOThf8i1/8AoCpU6eyY8cO0tPTSU9PByAlJYV9+/YB8Mwzz9CtWze6det2crrgvLw8Onfu\nzI9+9CO6du3KlVdeecpxSrNy5Ur69+9PamoqI0aM4Ouvvz55/MAUwIEJyz788MOTNyvp1asXhw8f\nrvDf1phYEAju3bvDZZdVPLjv2AHbt5cd3AcNghYtojfXTMyOivnZzyDSNxjq2dPdMSWU5s2b07dv\nX959912GDx/OnDlzuP766xER6tWrx+uvv84ZZ5zBvn376N+/P9ddd13I+4hOnz6dBg0asH79enJy\nck6ZsnfSpEk0b96cEydOMGTIEHJycrj33nt55plnyMzMpGWJkQ/Lly9n5syZLF26FFWlX79+DBo0\niGbNmrFp0yZeeeUV/vznP3P99dczb968Mudnv+WWW3juuecYNGgQjz76KL/85S+ZMmUKTz75JFu3\nbqVu3bonU0GTJ09m2rRpDBgwgCNHjlCvXr0w/trGxJ7cXDd6tHFjyMiABx5wgfqcc8LbT2AmyJI9\nZYLVrg3Dh8Orr7pcvzc/YLWxmnsJwamZ4JSMqvLQQw+RmppKRkYG27dvZ/fu3SH3s2TJkpNBNjU1\nldTU1JPvzZ07l969e9OrVy/Wrl1b7qRgH3/8MSNGjKBhw4Y0atSIkSNH8tFHHwHQoUMHevbsCZQ9\nrTC4+eUPHDjAoEGDALj11ltZsmTJyTKOHTuWWbNmnRwJO2DAAO6//36mTp3KgQMHbISsiXu5uRC4\nUVpGhnuuyDQBy5ZBUpLr016WkSNdo+3ixeEfo7Ji9n9rWTXsqjR8+HDuu+8+VqxYQUFBAX369AFg\n9uzZ7N27l+XLl5OcnExKSkqp0/yWZ+vWrUyePJmsrCyaNWvGuHHjKrSfgLpB1YGkpKRy0zKhvPPO\nOyxZsoS33nqLSZMmsXr1aiZOnMg111zDggULGDBgAAsXLuSiiy6qcFmNiSZVF9x/+EO3nJrqpgdY\ntAhuvjm8fS1b5j5fv37Z22VkuF8J8+a5UbHVyWruJTRq1Ij09HT+8z//85SG1IMHD3LmmWeSnJxM\nZmYm27ZtK3M/l112GS+//DIAa9asIScnB3DTBTds2JAmTZqwe/du3n23eCqexo0bl5rXHjhwIG+8\n8QYFBQV88803vP766wwcODDsc2vSpAnNmjU7Wet/6aWXGDRoEEVFRXz55Zekp6fz1FNPcfDgQY4c\nOcLnn39O9+7defDBB7n44ovZsGFD2Mc0Jlbs3w9ffw0XXOCWa9WCIUNccPcxf+JJRUWuj3tZKZmA\nunXh+993s0RG+FYN5YrZmns0jRkzhhEjRpzSc2bs2LFce+21dO/enbS0tHJrsHfddRe33XYbnTt3\npnPnzid/AfTo0YNevXpx0UUXce65554yXfD48eMZNmwY55xzDpmZmSfX9+7dm3HjxtHXa7254447\n6NWrV5kpmFBeeOEFJkyYQEFBAR07dmTmzJmcOHGCm266iYMHD6Kq3HvvvTRt2pSf//znZGZmUqtW\nLbp27XryrlLGxKNAY2rw/eszMuDvf4f166HEXTJD2rwZDhwouzE12KhR8Mor8NFH4PWVqB6qGpVH\nnz59tKR169adts7ENrtmJl789a+qoLp5c/G6vDy37ne/87+fl15yn8nJ8bf9kSOq9eur3n13eOUN\nBchWHzHW0jLGmBohNxfq1HGTegW0b+8GIoXTJTIry41w9Xsbg4YNYdgweP11l9KpLhbcjTE1Qm6u\nC+RJSaeuz8iAf/4Tjh/3t59ly6BPH9fV0a9Ro1yXy6VL/X+msmIuuGs4LRsmquxamXgS3A0yWEYG\nHD5c3He9LMeOwb//7T/fHnDNNZCcXL3TAMdUcK9Xrx779++3oBEHVJX9+/fbwCYTFwoL4fPPSw/u\n6eluml4/qZk1a9yAJD89ZYI1beq+RF57LbyeOZURU71l2rZtS35+Pnv37o12UYwP9erVo23wLWiM\niVF5eS7tEugGGax5c5dmWbQIHn207P0EZoIMt+YObkDTj37kRt6XN/gpEmIquCcnJ9OhQ4doF8MY\nk2BK6wYZLCMDJk+GI0fAm+S1VMuWuYFPwY2yfg0fDnfe6Wrv1RHcYyotY4wxVcFPcC8sBG82jpCy\nslxKJsSUUmVq1cpNVlZdeXcL7saYhJeb62ZobNGi9PcHDIB69cq+9d7hw7B2bcVSMgGjRrkBU+vX\nV3wffllwN8YkvFA9ZQLq1YOBA8tuVF2xwjWGVia4jxjhnqtjGmAL7saYhFdecAeXmlmzBnbtKv19\nP9P8lqdNG3jkEXcbvqpmwd0Yk9AOHXIB209wh9BTAC9b5hpSW7WqXHmeeMJNWFbVLLgbYxLaxo3u\nubRukMF69nTdIkOlZrKyKpeSqW4W3I0xCa28njIBZU0BvGeP6ytfmZRMdbPgboxJaLm5LnB36lT+\nthkZkJ9fXNsPCOTbE6rmLiLPi8geEVkT4n0RkakisllEckSkd2nbGWNMNOTmQocO/u5hGsi7l0zN\nZGW5L4jecRTd/NTc/wYMK+P9q4Dzvcd4YHrli2WMMZHhp6dMQMeO7ougZHBftszdzKOs0auxptzg\nrqpLgK/K2GQ48KI3j/xnQFMROTtSBTTGmIoqKnIpFr/BHVztffHi4tviqbrgHk8pGYhMzr0N8GXQ\ncr637jQiMl5EskUk2yYHM8aU59AhF2grKj8fjh4NL7hfcYU7bna2W87Lc/dfrYnB3TdVnaGqaaqa\n1qqynUWNMQlv8mTXg2Xr1op93m83yGAlpwAOzAQZTz1lIDLBfTtwbtByW2+dMcZUSmBA0cKFFfu8\n326QwVq2dLM2Bgf3unWhe/eKlSFaIhHc5wO3eL1m+gMHVXVnBPZrjKnBjhwprjVXJrg3agRnh9kK\nmJEBn3wC33zjesr07u3upBRP/HSFfAX4FLhQRPJF5HYRmSAiE7xNFgBbgM3An4EfV1lpjTE1xr/+\n5Ro1zz/f1eD93uM0WKCnTLhT9GZkuONlZsLy5fGXkgEfN+tQ1THlvK/A3RErkTHG4BpSk5Pd3ZFu\nvhk+/dTNhx6O3Fw3nW+4Lr3UpWKmToWCgvhrTAUboWqMiVGZmdCvH1x7LSQlhZ+aOXoUvvgivHx7\nQP367kshML+7BXdjjImAgwddOuTyy6FJE7jkEnjvvfD2sWmT66NekeAOxaNVmzaF886r2D6iyYK7\nMSbmLFniBiClp7vloUPdzTL27PG/j4p0gwx2xRXuuaK31Ys2C+7GmJiTmely3oGbWgzzJkAp6zZ4\nJQW6QVY0uPfq5SYbu+qqin0+2sptUDXGmOqWmVl8X1NwXRFbtnSpmbFj/e0jNxfatoWGDStWhqQk\nl9qJx1o7WM3dGBNj9u+HlSuLUzLgZmS84gp4/32XrvEjnAnDQonXwA4W3I0xMebDD91zcHAHl5rZ\nswdWrSp/H6qRCe7xzIK7MSamZGa6VErJgUNXXume/fSa2bPH9bix4G6MMTFi8WI3iKhOnVPXt27t\n7nPqp797ReaUSTQW3I0xMWP3bli37vSUTMDQoW5agkOHyt5PZbtBJgIL7saYmPHPf7rnyy8v/f2h\nQ918M5mZZe8nN9d1pWzXLqLFiysW3I0xMWPxYjjjDNfHvDQDBrh8fHmpmdxcN+FYUlLkyxgvLLgb\nY2JGZqabHKx2iBE4deq4Wv1777keMaHU9J4yYMHdGBMj8vPdoKFQKZmAoUPdnZk2by79/ePHYcsW\nC+4W3I0xMSGQRw/VmBoQmIogVGpmyxaXl7fgbowxMSAzE5o3h9TUsrfr1Mk9QvV3t26QjgV3Y0xM\nyMyEwYPdVAPlGTrUbf/dd6e/Z90gHQvuxpio27oV8vLKT8kEDBvm7pD0r3+d/l5uLrRqBc2aRbSI\ncceCuzEm6vzm2wPS090t+EpLzVhPGceCuzEm6jIz4cwzoUsXf9s3auT6vJfWqGrB3bHgboyJKlU3\neCk9PbwpdocNg5wc2LGjeN2BA27SMAvuFtyNMVG2aZML0H5TMgFDh7rn998vXmc9ZYpZcDfGRFUg\n317e4KWSUlPhrLNOTc1YcC9mwd0YE1WLF0ObNnDeeeF9rlYtV3t//304ccKty81188l07Bj5csYb\nX8FdRIaJSK6IbBaRiaW830RE3hKRVSKyVkRui3xRjTGJRtXNBBluvj1g6FD46itYvtwtb9zoAnty\nckSLGZfKDe4ikgRMA64CugBjRKRkm/bdwDpV7QEMBn4rIiWm2jfGmFOtW+caQMNNyQRccYX7Ugik\nZqynTDE/Nfe+wGZV3aKqx4A5wPAS2yjQWEQEaAR8BRRGtKTGmISzeLF7DrcxNaBVK+jTx/V3Lypy\njbMW3B0/wb0N8GXQcr63Ltjvgc7ADmA18FNVPe0e5SIyXkSyRSR77969FSyyMSZRZGZCSop7VNTQ\nobB0qesW+e23FtwDItWgOhRYCZwD9AR+LyJnlNxIVWeoapqqprVq1SpChzbGxKOiIpdvr2hKJmDY\nMNeg+oc/uGUL7o6f4L4dODdoua23LthtwGvqbAa2AhdFpojGmES0ahV8/XXFUzIB/fq5uzfNmuWW\nLbg7foJ7FnC+iHTwGklvAOaX2OYLYAiAiJwFXAhsiWRBjTGJJdz5ZEJJToYhQ+DoUWjSxE1jYHwE\nd1UtBO4BFgLrgbmqulZEJojIBG+zJ4Dvichq4B/Ag6q6r6oKbYyJf5mZblreNiVb8CogcAOPCy6o\nWJfKRBTiToWnUtUFwIIS6/4Y9HoHcGVki2aMSVSFhfDhh3DjjZHZX2AqAkvJFPMV3I0xJpJWrIDD\nhyufkglo3x7+53+Kg7yx4G6MiYJAvn3w4Mjt86mnIrevRGDB3RgTcapw/Lhr5CwocM/Br+fPh65d\n3cRfpmpYcDfGVFphIfz61/DHP7p0S0GB68delvvvr56y1VQW3I0xlbJtG9x0E3z8MVxzjeuxUr8+\nNGjgnkt73aAB9OoV7ZInNgvuxpgKmzsXxo93tfSXXnJB3sQGm8/dGBO2I0fg9tvhhz+Eiy6Cf//b\nAnusseBujAnL8uXQuzfMnAkPPwwffQSdOkW7VKYkC+7GGF+KimDyZLjkEtfjJTMTfvUruzFGrIqr\n4D57tpsatFYt9zx7drRLZEzNsHOnGyD0wANw7bVu0q9Bg6JdKlOWuGlQnT3bNdwUFLjlbdvcMsDY\nsdErlzGJQBW++851Yyz5+PJLl34pKIA//9nl2m3+ltgnqhqVA6elpWl2drbv7VNSXEAvqX17yMuL\nWLGMqRFeecWN6Pz66+IgXljGvdN69nSfucgm8o46EVmuqmnlbRc3NfcvvghvvTHmdKrw+OPw2GMu\nYKenQ+PGZT/OOMP1Xa8dN9HCQBwF93btSq+5t2tX/WUxJh59+61Lqbz8Mtx6K8yYAXXsNvYJK24a\nVCdNcqPagtWp41rrjTFl27sXMjJcYJ80yXVjtMCe2OImuI8d62oa7du75bp14dgxeOcdOHgwumUz\nJpZt2AD9+0N2Nvz97/DQQ9YgWhPETXAHF+Dz8lze8JtvXA3k//7P5Q4/++zUba3bpEkURUWwf3/F\nPrt4seuXfuSIuxn19ddHtGgmhsVVcA+WlORqIB9/7Gohl17qgv2JE8XdJrdtc18EgW6TFuBNvPn4\nY1frbtkSuneHn//c3ejCTye35593fdPPOQeWLnX7MTVH3Ab3gP793bwW118Pjzzi8ooPPljcHz6g\noMD11TUmHmzZAj/4AQwcCNu3u3+7LVq4aXX79HHpyXvvdTXz48dP/WxREUyc6BpP09Phk0/cr1dT\nw6hqVB59+vTRSCoqUv3b31QbNlR19ZrTHyIRPaQxEXfggOoDD6jWqaPaoIHqY4+pHjlS/P7evarP\nP6963XWq9eq5f9fNmqnecovqa6+590eNcuvvvFP12LHonYupGkC2+oixcTOIya9Nm6BbN9fYWpIN\neDKxqrDQjf589FGXX7/1VtcTrE2b0J/55ht4/3144w146y03IAlcmnLyZLjvPms4TUR+BzHFfVqm\npPPPd70z6/fnAAARlElEQVRqSg64qFfP5eSrw7598PTTbjrUFSuq55gmfr33HvToAT/+MXTp4nq1\nzJxZdmAHaNgQRoyAF16A3btdiubBB+Hdd91djiyw13B+qvdV8Yh0WqakWbNUzzrr1LTMxRerTp/u\nfvpGWlGR6qefqt58s2rduu54jRqp1q7tflrbz2MTrKhINTtbddgw92+lUyeXVikqinbJTKzDZ1rG\nV81dRIaJSK6IbBaRiSG2GSwiK0VkrYh8GNFvoAoYOxZ27XJhfe9emDLFjdC76y5o3RpuvtlNWVre\nfR7LU1AAf/mLa+S65BL3E/mOO2D1ajc1wg03uKHe/fvD2rURObWI2r0b/vEP18vIVB1V1998+nTX\n+H/WWZCWBp9+Cr/9Laxb52rhVts2EVNe9AeSgM+BjkAdYBXQpcQ2TYF1QDtv+czy9lvVNffSFBWp\nZmWp3nWXapMmrsbUsaPqE0+obtkSXu16wwbVn/2seD/durlfBYcOnb7tvHmqrVq5RrKnn1YtLIzc\nOVXUsmWqN92kmpzsyv+976nm5ka7VImjqEh140bVGTNUx4xRPfvs4l+Qbdq4v/1f/6q6b1+0S2ri\nDZFqUBWRS4DHVHWot/y/3pfCb4K2+TFwjqo+4vdLpaoaVP0qKIDXX3ddy9atK15fr57rU9y0aemP\nxo3hgw9g0SJ3k4LRo12udMCAsmtde/bAhAnumN/7nsuTnnde1Z9nsGPHYN48mDrVDfpq3BjGjYPO\nnV1Xu6NH3d/j3nvdOAITnl27XAPnBx+4X4Xbt7v1rVu7LomBR6dOVkM3FRfJWSHbAF8GLecD/Ups\ncwGQLCL/BBoDv1PVF32WNSoC89SU7D1TWOj+87VoAQcOwI4dLp1y4ICb5qCoCM491/VkuP129x/X\njzPPdIH15ZfhnntcA9rTT7s0Ua0qbtbevds1Mk+f7m66cP75LsDfequb8Q/gP/4D7rzTNcTNm+ca\n9M4/v2rLFe+OHXN9yBcudI2iK1e69a1auSA+eLB7vvBCC+YmCsqr2gOjgb8ELd8M/L7ENr8HPgMa\nAi2BTcAFpexrPJANZLdr167qf7+Uo3370vvDt29f+vYnTqgePOieKyM/v7ghbcgQ1W3bKre/ULKy\nXANvnTruWMOGqS5YELr8RUWqL7yg2rSpav36qs8+W/lzTTSbN6tOm+b6mTdq5P6utWurXnaZ6q9/\nrbp8uf3NTNWimtMyE4H6qvoLb/mvwHuq+n+h9hvttAy4GnNppy9S+YbW8qi6hthAl7Vbb4WOHaFD\nBzeasEMHaNLE337274eNG4sfubku1bRhAzRqBLfdBnff7WqQfuzY4aZreOcdN63DzJnVn0KKFTt3\nwpIl8OGHLt2yebNbn5ICw4a5R3p68S8gY6qa37SMn+BeG9gIDAG2A1nAjaq6Nmibzrja+1Bco+sy\n4AZVXRNqv7EQ3GPh7k55eS7wLlniJncK1rRpcbAPBPwWLWDr1uIgvnFj8eAVcP37O3Z0N1e44gqX\nU69I4FGFF1+En/7UpR9+8xv4yU+qPoUUbdu2FQfzJUvcoDhwfcoHDXLBfOhQl7KyVIuJhogFd29n\nVwNTcD1nnlfVSSIyAUBV/+ht8wBwG1CES+NMKWufsRDcS96XFVwufsaM6r8vqyp89ZUL3Hl5xc/B\nr48eLd6+bVsXwEs+UlIiezf67dvd32jBAjfPyeTJrgtfIgR5VVcTDwTyDz8svrNX06bufC+7zAX1\nXr3sTkQmNvgN7gk7iMmvWbNcjl3EPc+aFZltI62oSHXXLtW1a0+da6S6jj1zZnG3z7PPVh0/XvXt\nt1ULCqqnDN9847pqfvdd5fazc6e7buPGqbZtW9zOcuaZqqNHq06dqrpypeXNTeyips4tU1ViqZYf\nLV995fLw8+e73iFHjri/wRVXwHXXwTXXuME5lVVUBOvXu2lqly1zz6tXu4FWycmu62Zqqnv06OGe\nzzqr9DTJ4cOuVr5okXus8RKFzZvDkCFw+eWuV4v1aDHxIqJpmaoQb8E9FvLzseS771waY/589/jy\nSxcc+/WDa691Uy83aeJu5Rb8SE52z0lJxcF01y4XwAOPrCwXlMHto29ft99OnVw7Q06Oe+TnF5en\nVavigN+tmyvPokWuP39hoRu/MHCgK1dGhrvBSyKklkzNY8E9wsLpWTN7thsU9MUX7gbekyYldu1e\n1QXbQKD3c1lFXJCvXdvNbgjudWqqC+SBxwUXhA7C+/e7Gn1ODqxa5Z7XrHHTTIi4KSECwXzAABfg\njYl3FtwjzG/N3dI3rivlZ5+52v2xY6c+jh8/fV3bti6Q9+4N9etX7tgnTsDnn7tRxs2bR+Z8jIkl\nFtwjzG/QtvSNMaYq1dj53KvK2LEukLdv737yt29fem080JWupFDrjTGmKljP3TCMHVt+aqVdu9Jr\n7u3aVU2ZjDGmNFZzj7BJk4onJQto0CD0XaBmz3apnFq13PPs2VVdQmNMTWDBPcL8pm+gOI+/bZvr\ncbJtm1u2AG+MqSxrUI0ia3w1xoTLGlTjQDiNr5a+McaEw4J7FIVqZC253tI3xphwWXCPIr+Nrw8/\nfGr/enDLDz9cteUzxsQvC+5RZH3njTFVxfq5R5n1nTfGVAWruccB6ztvjAmXBfc4YH3njTHhsn7u\nCcb6zhuT2Kyfew1lja/GGLDgnnD89p0Hy80bk8gsuCcYv42vlps3JrFZcE8wfhtfbWCUMYnNGlRr\nqHDuCWuMiR3WoGrKFE5u3hgTf3wFdxEZJiK5IrJZRCaWsd3FIlIoIqMjV0RTFWxglDGJrdzgLiJJ\nwDTgKqALMEZEuoTY7ing/UgX0kSeDYwyJrGVm3MXkUuAx1R1qLf8vwCq+psS2/0MOA5cDLytqq+W\ntV/LuccPGxhlTOyIZM69DfBl0HK+ty74YG2AEcD0cgo1XkSyRSR77969Pg5tYoENjDIm/kSqQXUK\n8KCqltnPQlVnqGqaqqa1atUqQoc2Vc0GRhkTf/wE9+3AuUHLbb11wdKAOSKSB4wG/iAi/xGREpqo\ns4FRxsQfP8E9CzhfRDqISB3gBmB+8Aaq2kFVU1Q1BXgV+LGqvhHx0pqosIFRxsSfcoO7qhYC9wAL\ngfXAXFVdKyITRGRCVRfQxIaxY13jaVGRey6tV024uXlL4RhTdXzdiUlVFwALSqz7Y4htx1W+WCYe\nhXPHqEAKJ1DTD6RwoPw7UxljymcjVE3EhDMwylI4xlQtC+4mYsIZGBVOCsfSN8aEz26QbSLKzw2/\nwX8Kx9I3xlSM1dxNVPhN4Vj6xpiKseBuosJvCsdGxxpTMZaWMVHjJ4UTTg8cY0wxq7mbmGZTExtT\nMRbcTUyzqYmNqRi7zZ5JGDY1sakJ7DZ7psaxxldjillwNwnDpiY2ppgFd5MwbGpiY4pZcDcJw6Ym\nNqaYBXeTUGxqYmMcC+6mxgk3N28pHBOPLLibGsemJjY1gQV3U+NU1dTExsQSm1vG1EiRnprYmFhj\nNXdjyhBOCscaXk0sseBuTBn8pnCs4dXEGptbxpgIsHltTHWxuWWMqUbWd97EGgvuxkSA9Z03scaC\nuzERYH3nTazxFdxFZJiI5IrIZhGZWMr7Y0UkR0RWi8gnItIj8kU1JnZZ33kTa8oN7iKSBEwDrgK6\nAGNEpEuJzbYCg1S1O/AEMCPSBTUm1vmZ1wZsamJTPfzU3PsCm1V1i6oeA+YAw4M3UNVPVPVrb/Ez\noG1ki2lM4rCpiU118BPc2wBfBi3ne+tCuR14t7Q3RGS8iGSLSPbevXv9l9KYBFJVUxNbLd8Ei+j0\nAyKSjgvul5b2vqrOwEvZpKWlRaeDvTExwM/0B+Hk5gO1/MCXQaCWHziWqXn81Ny3A+cGLbf11p1C\nRFKBvwDDVXV/ZIpnTM0VTm7eeuCYkvwE9yzgfBHpICJ1gBuA+cEbiEg74DXgZlXdGPliGlPzhNO9\n0nrgmJLKDe6qWgjcAywE1gNzVXWtiEwQkQneZo8CLYA/iMhKEbF5BYyppHC6V4ZTyzc1g80tY0wC\nKJlzB1fLDzXJ2cMPu1p9u3bul4Dl5eOHzS1jTA1SVbNXWg+c+GU1d2NqkHBmrwzn14CpPlZzN8ac\nJpyG13B64FgNP/ZYcDemBgmn4dXvF4GNpI1NFtyNqUHC6V7p94vA+tjHJgvuxtQg4XSv9PtFYH3s\nY5MFd2NqGL+zV/r9Igi3j73l56uHBXdjTEh+vgjCSfVYfr76WHA3xlRKOKkey89XHwvuxphK85vq\nCXemS0vfVJwFd2NMtfGbn7eRtJVnwd0YU2385ufDHUBlefzTWXA3xlQbv/n5qhpJW5NYcDfGVCs/\n+fmqGEkLNSt9Y8HdGBNzqmIkbU3L41twN8bEnKoYSVvT8vgW3I0xMSnSI2lr2oyYFtyNMXEvWnn8\nWE71WHA3xtQI0Z4Rs7pTPRbcjTE1QrRnxKzuLpsW3I0xNUY0Z8Ss7qmRLbgbY0wpIj0jZrhTI1eW\nBXdjjKmgqkj1RIqv4C4iw0QkV0Q2i8jEUt4XEZnqvZ8jIr0jX1RjjIk9kU71RErt8jYQkSRgGnAF\nkA9kich8VV0XtNlVwPneox8w3Xs2xhjjGTu26oJ5SX5q7n2Bzaq6RVWPAXOA4SW2GQ68qM5nQFMR\nOTvCZTXGGOOTn+DeBvgyaDnfWxfuNojIeBHJFpHsvXv3hltWY4wxPlVrg6qqzlDVNFVNa9WqVXUe\n2hhjahQ/wX07cG7QcltvXbjbGGOMqSZ+gnsWcL6IdBCROsANwPwS28wHbvF6zfQHDqrqzgiX1Rhj\njE/l9pZR1UIRuQdYCCQBz6vqWhGZ4L3/R2ABcDWwGSgAbitvv8uXL98nIttKrG4J7AvvFGJaop0P\nJN45Jdr5QOKdU6KdD1TunNr72UhUtYL7jzwRyVbVtGiXI1IS7Xwg8c4p0c4HEu+cEu18oHrOyUao\nGmNMArLgbowxCSjWgvuMaBcgwhLtfCDxzinRzgcS75wS7XygGs4ppnLuxhhjIiPWau7GGGMiwIK7\nMcYkoJgI7uVNKRyPRCRPRFaLyEoRyY52eSpCRJ4XkT0isiZoXXMR+UBENnnPzaJZxnCEOJ/HRGS7\nd51WisjV0SxjOETkXBHJFJF1IrJWRH7qrY/naxTqnOLyOolIPRFZJiKrvPP5pbe+yq9R1HPu3pTC\nGwmaUhgYU2JK4bgjInlAmqrG7eALEbkMOIKb8bObt+5p4CtVfdL7Im6mqg9Gs5x+hTifx4Ajqjo5\nmmWrCG/m1bNVdYWINAaWA/8BjCN+r1Goc7qeOLxOIiJAQ1U9IiLJwMfAT4GRVPE1ioWau58phU0U\nqOoS4KsSq4cDL3ivX8D9x4sLIc4nbqnqTlVd4b0+DKzHzcYaz9co1DnFJW8a9CPeYrL3UKrhGsVC\ncPc1XXAcUmCRiCwXkfHRLkwEnRU0b9Au4KxoFiZCfuLdQez5eEphBBORFKAXsJQEuUYlzgni9DqJ\nSJKIrAT2AB+oarVco1gI7onqUlXtibtL1d1eSiChqMvpxXtf2ulAR6AnsBP4bXSLEz4RaQTMA36m\nqoeC34vXa1TKOcXtdVLVE14saAv0FZFuJd6vkmsUC8E9IacLVtXt3vMe4HVc+ikR7A7cZct73hPl\n8lSKqu72/vMVAX8mzq6Tl8edB8xW1de81XF9jUo7p3i/TgCqegDIBIZRDdcoFoK7nymF44qINPQa\ngxCRhsCVwJqyPxU35gO3eq9vBd6MYlkqrcTtIEcQR9fJa6z7K7BeVZ8Jeitur1Goc4rX6yQirUSk\nqfe6Pq7jyAaq4RpFvbcMgNetaQrFUwpPinKRKkVEOuJq6+CmVX45Hs9JRF4BBuOmJ90N/AJ4A5gL\ntAO2Aderalw0UoY4n8G4n/oK5AF3xsu9CETkUuAjYDVQ5K1+CJejjtdrFOqcxhCH10lEUnENpkm4\nyvRcVX1cRFpQxdcoJoK7McaYyIqFtIwxxpgIs+BujDEJyIK7McYkIAvuxhiTgCy4G2NMArLgbowx\nCciCuzHGJKD/D32OJbiPHAjTAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f02df8d4828>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the graphs - These plots are characteristic of overfitting.\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "acc = history.history['acc']\n",
    "val_acc = history.history['val_acc']\n",
    "loss = history.history['loss']\n",
    "val_loss = history.history['val_loss']\n",
    "\n",
    "epochs = range(1, len(acc) + 1)\n",
    "\n",
    "plt.plot(epochs, acc, 'bo', label='Training acc')\n",
    "plt.plot(epochs, val_acc, 'b', label='Validation acc')\n",
    "plt.title('Training and validation accuracy')\n",
    "plt.legend()\n",
    "\n",
    "plt.figure()\n",
    "\n",
    "plt.plot(epochs, loss, 'bo', label='Training loss')\n",
    "plt.plot(epochs, val_loss, 'b', label='Validation loss')\n",
    "plt.title('Training and validation loss')\n",
    "plt.legend()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "** Create more data through random transformations of the data. The goal is that during training, the model never sees the same picture twice. This helps generalize better. Also, apply dropout. **"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQUAAAD8CAYAAAB+fLH0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzsvVmMXdl63/dba+3hDHVODawqkk2ymz3fvkPfe6UryYok\nW4mRwHGE6MlOIsdwBkRQkAQKnCCx85aHAAECBPFTACEDktiBE8SZHowYgQNZVgRourpT39t9bw/s\nZneTLJI1nKoz7GGtlYc17H2KVWRVkWwWlfMBBE+ds/dew17rW9/4/4S1lgUtaEELCiSfdQcWtKAF\nnS9aMIUFLWhBc7RgCgta0ILmaMEUFrSgBc3RgiksaEELmqMFU1jQghY0R0+NKQgh/oIQ4j0hxPtC\niL/xtNpZ0IIW9GRJPI04BSGEAn4M/NPAp8AfAv+StfaHT7yxBS1oQU+Unpak8LPA+9baD621JfB3\ngV99Sm0taEELeoKUPKXnXgFutv7+FPi5YzuRKpt20oc/8akFXp7viM6HCXJCnOA64X849LsxFmMe\nvEkgUCqcFZbQhBCCLMsAyIQB2ieKPeITSNvu1/EDsa7h2P6x15lDfx8zaCGF77Prd7g2fBbuBywq\n/i2lG42USbxOChn7VWuNlDK2aa1pzZ+NA9CmiiMw1iJbL8laC+EeIebeX3yWtRjbfm5rXubet8UY\n438jLgYhBEYfmih/0fb+7J61duPISWvR02IKjyQhxK8Dvw6Q5gmvfOPFE9/7NFQee3jFPWESZ3y8\nNvqYX5oFJ4Q8/jrhvrfWYrXbBNNpyWw2Q9fhIkWSuKWwtryErSv3rYJUuTZeevFFXkxKACSarlSg\n3d9GQCV9OwJ6M7fBjDGYI/olhPu99pMipJzbPEeRmam5915X1ZHXJX1FkrqxKKnivCRJSpa6g0cq\nRaVWEML93et1GQ5XAUiTDlmWx/vvHez4a3rUVc10NgOgLAq01r7/gtH+HfdsKShL17csaw66uq5h\nWiKVewfK/w9QFIX7HSjLEmvrOE9hXsL1lb/OWsNkMnFtChl/z/Kc/dHogXmRSvK3/58ffnzkpB2+\n9iQXnYE+A661/r7qv4tkrf0ta+23rLXfUqliQQta0PmgpyUp/CHwuhDiZRwz+BeBXzv26lMe/OKw\nSPYEKJxeT0tisC32e1Kp4VFSQujzw6SEKO7qh5/E55XMzB0Yj5ISkr67LkkTlHSftdEkiTutszSN\np3SQEnq9LgDD4Spp0nHXZXm8/97BDr1ez7fppISyKNyztY5qymj/DtJ/LsvqQQkBopRwWEII15Rl\n6cdZx/cqhZi7vqrruD4nk4lTb3BSRJY76eawlCC9KnikSnEMPRWmYK2thRD/NvAPAAX8N9badx5x\nT/wsHiFKtqmtMz4JCi/EPfPpMYiHMYaHMQMgMoSTqQxhLiXTqVt4jergN1KSsLa85O6pK8I6TJXg\npRedWvdiUiJxzw2qg/GPrqTG+s+9mYy67mHVoT23tTAI2Sz+541OozIARzKEk6oM0DCE06oMgRlo\nW554bE/NpmCt/fvA339az1/Qghb0dOiZGRofRuHUf5YSg3vm05MajlMnnoRhcV5lcA0F4yIQpYQz\nGRchSglnMS5CIyU8C+Oi60t6KuMiEFWHtnHxeaAgIXSSk9vtziVTCHQWleJp2Bvcc+UXYm/ghE0k\nSYL1bqzjGcnp6VEqAzTehrOoDPBob8PTtCNA4204qR0BGm/DWewI0KgCZ7EjuGvMU7UjzN1/prsW\ntKAF/amlcy0ptOk8SA1fhBFSJgpTt0/beeMiONHVGjvfhyOMi8ZA5U/YoiweiEs4yrj4vNBZVQZo\nvA0nVRmg8TY8C+MiNN6G0xoXg9qg6/qBa4+jc8MUwpY9yXZvR6edlIQQz429QfoXKVobXwgZRdfA\nEGLkm2iYiCCJG8RaSVk68bcqdBSflZJntiNAozqcxY4AD/c2nNSO8DzQUQzhLCqDu0aeWGVo2xEC\nM5D65ExhoT4saEELmqNzIykEap/lD5MFzoM60Tz3yUoNwrTHNq8yhDaMtXMSAtadNEZbQvxBWRZM\nxk78lbJ51avDPpmS1L6Z0xgXofE2PAvjIjSqw1mMi9B4G05qXHT9F6c2Lj6vdD6YwqHElUDPK4M4\nKWMICw7mRc1jnx0SfVCYE+iIUgpS73bUpmFYVVWDrul0XIJTVUxP1N9AgSF80SrDYW/DWewI0Hgb\nTmpHgMbbcBY7QrjmrHaEcM1Z7AhBbRCn4FML9WFBC1rQHJ0PSYFWMEjrkD0sNZzUCBmfeY69FG0p\nof13Iub59LFBModUB6c2gDEiirLFrCa4qo2xDPvupFwZ9JnOxgjrniGVQt/6EIDPUbz+0mXfqaND\nmZ8XelSA0klVBmi8DSeNR4CjvQ1nMS7C0d6GkxgXo4TQXzrxvJ0bphBobhOYs6kT8fpz5KVoM4fD\nDAEeZAbgFpg5gqnU2omkwY5gjWMGrh1LWbjn11UzjkEvp5c7daEopkijmU7cc7o1jGtvu8Dw0Uc3\nAHjppRdOHLX4RdkRoHFBPsqO8LzQWROdTqQyeGawtrZ24v6cE6Yg4kS0N4GQ4qGSQ3P38XQeJIdA\nR4GanJmsQvpjoNQ6nijjgyll4RaFNpqOT0vPE0ma+TkuC4SUiNr9VltNjpMiaqMZT9yp85PPPonN\nvbV57YkmOp3VjhDGdVY7AjSGxZPaEeAYw+IJ7QjQuCCPsiOcN1rYFBa0oAXN0TmRFBqSQj4oLQR6\nDu0NQSx8mJRwWHUIp0h7HqyxIA+f0O2+uWuVVKS406l9dV3XpIkTJYuyQAiBCEkytWBculNzKRFo\nnJpRj6dxyj/89NP4rBcvXTx2LPDoVOgnleh0UtdjuP9JoCidxY4Ajbfhi7YjBLUh2E9OQueOKQBR\nlYAj1In4Q/PRWnsmewM8/UzMwAxkq+/G+PGp0JcmU/KwHSHEJrQZghIZtdZUfiFLKSgLr1/OCurK\nfU5FhvLZVhcuDCimY3e9UoytJvezNkxz9NgtSvI+VgbbgUX5fo+LZuPe+OxzhJRcf+GF+F3bjvC8\n0BelMsCTA0w5rR0hMIPQv5PQ8/MGF7SgBX0hdC4lhTY9Smqwxj7gMTiN1PA4EkP7/uPoKIOSlAYj\nLLJ1a/ROaKA1tjmVwQRDm0HIBOslhbIw2KKMn4XwqMuZpNd1RreiKOLbrhKF2Rmz2h0CMJQ5KnM/\nbu/cjc31+13CDCZZr1ELpCaTNbfu3wNgY33txMZFN1Y38LMYF6HxNpzFuOjabVSH54WClHBalSFI\nCGl6co/MuWAKJ92PRzGIk8Q3xHYe8uwvEg7OiZTHYyAoH3NQH9MNNw8SG+DbMQ0suRRR/M3zjDT3\nLkVKhHc7qrJGVBJrD1w7aQmp+219Y4NEuefeu7+N9JsqTdKI+6Bljsx7HMycyiF29yMz2FhbPrLP\n5wWNGY4GTDnvaMxfJC3UhwUtaEFzdC4kBZg3ItoT+PMPxzWc1EsBz16lsMKgrCCaR7WJHghhG1VC\nKIU1bfWjAWE11kSDoqg1lQ9U0lqgtTvp0lQhZOq/tyjCyehe+8ifCbKq6Hqrp5EFswNnkNxcvxBb\nvvnZ52xubgJQ1poDiF6KcVkT7Iv67na859LGKQJmviA0Zng8FKVnicZ8FuPiadSGQGdmCkKIa8B/\nD1zE7bffstb+LSHEGvA/AdeBG8BfttbunOrZp2AQxwY9BfJfnweVwgYLvf86CYqgaL6z1mCSEKnY\n4CnMVxuSDh/BMwWrBU1+lCFJ3JwsDRKq2iU7JQj85Uz2DkgFDFO3qZPZPn1focvUGuvv37l/l27H\nbbwrL1zm9u2tOJY0S+PiX19fowxh2kmzpG7f3WZzsPHAXDxXGAnnCI35adoR5to9012OauDfs9Z+\nGfgzwL8lhPgy8DeAf2itfR34h/7vBS1oQc8JnVlSsNbeAm75z/tCiB/hakj+KvDL/rL/Dvht4D88\nazttL8PD6JGxDU9IpYCzBT4ZdJQGABQSYUNNRoHxPTGJehA3gXmrfl1PUVphvOWvKCqMCScV9Pq+\nSIwtUP4NaxLM1H+vZ2wOOnStO7HzPGG/dhcmSULtxVJbFwjpg150zYvXrvjvDaPpJEoF29s7TXDW\n6jK93PW1qCx7k4PY716SHTlPXyQasxv/KaDVngM6icpgTmEIfyI2BSHEdeCbwO8DFz3DALiNUy8e\nv42noFJAs9lPCwcXn3tSBnHCdyKFxMigFhx/XYml9IlRopUsb7SOzCLLupSVVx+yhKlx1w/XVuhU\nY7Tve0VOohqmGl2Gqout3D2VrlBelLXWsrK0FDdNnefRkn///jbJpvNAJHmHSe3E7yRJSFoTnyGf\nCRoznA0w5VmhMR92QR5lR3jS9NjeByHEEvD3gH/XWjs3OutW15HbQQjx60KIPxJC/FFdPh8ceUEL\n+v8DPZakIBzr/nvA37HW/q/+6ztCiMvW2ltCiMvA1lH3Wmt/C/gtgN6we6qkgpNKDScNl3666dkP\n9s9ai8AgA9aBtTEmQVgTxySExOOpYq2JJ5iuE+q6iP2umGB8bMGgp1hdGwAwme4jsz4A9/cqdOkb\nrCuqLItGzG4mjzSOSqmolEdnqiqWUq/iWE09GUcxf6nTY2/sPBadTof7O3uAOw0vbjgPRmVslBoA\nSDKqyr2fwTH2sKeBxgyPh6L0LNGYz2JcPI3aEOhxvA8C+K+BH1lr//PWT/8n8NeA/9T//3+c5Hlt\nnfk0A3kiNofHDHw6tk3Z3GmMjCnHQggqa6m9HSCVj34NQki01/vLssKYhhlJpaj8osiXLnCw7zZo\nJRWzkWsjSXN06dofdHK0Kcm9alHYZI6xhc+9Xicutk6aMw44kAa6pkIGeLfxAUveS2ElTIpG77n5\nmdMk8zxnY201fl+pnI73xoynNX13+5wd4XmhLwqN+bSux7MwBHg8SeEXgL8KfF8I8R3/3X+EYwb/\nsxDiXwc+Bv7yox/V1DSw1kQG8TSYA5w9E7Pp7elJSoFtQRdpreeYwaP6XxUiGvPC6RN0YqM1/b6T\nCISZIIIeXGoOfMxBIiqurblTdzzaY0XO4kIqiiLq8WmSxL6Mx2M6qT+NlYqhpzJJqGWKmDojorB1\nPJGLoiTtOulCSUV/3fVLCsnegTc6mhyxaghDrcqKNHP3lFrT6ThbwdMAXg3zdlYUpaPsCH/a6HG8\nD7/L8fvjz5/1uQta0IKeLZ2/iEbTcF/ZOrZPKjU8aXtDfNYx+A1H2xMCXmLrNDe2wUk4dOK3T5w2\nlPv8uB48lbTXPaVSsZ02mrRSioF/w0aX3NvZBWBl0KUrBMOhS4gajUZMp06VqOqmz71eL57GtdGk\nwj1MSoFQiip1J7pKErz0T6ZSdg+caNzpdOby+FeWLwGwt7fHaDRmMHC/CZWyu1/E8c907sd3h1df\nfR14cmjM8Hgp0c8Ujfkp2hHadG6YQqD5TXm+GESbMegYY3AYh3C+naPAVYzxMQtH8JPDUYxWO7Fa\nax1FXGOsw5Dw1yVJwrDrodmqitxv1qKqiTUgipIkcfdPygM2rr4Qx93v99neduHJUkru3r3rrysY\ndJpNnS/lfvCu3dzbEbTW6KD+CBntCwDTgNMA6MqJ8qsrKwgp43ishTK8Xw1Zi6nduHEDgLffbuwR\ngc6iMsDjAaY8S+DVx7EjnCYi90+nUrSgBS3ozHROJAVL7U/YpHV8PkxqOK2IdJxYfphO6qWQnp8K\nIUAKZEtlCKSUnJMugtQwX0B2vm/tKEZrLBZfqEU0KNBBSghpzauDrPES5F1mU3ci1rWhLq3vi8Lm\n7jSqCxisrSO8i9Aay/r6ehxPOPWFlBz4082Y5gTM0gyj6yZ3A2ImeFWVcydo6Q2dF9bW4twW4wOk\nUiwtr8TrJl596Xcl97YdTsOlzSuMvavzow8+jtd+/evfnDMuPg/0pFCUTqsynBbNHM4LUxAiiuB1\na3CHGUTbf/801YlAJ/VSzN0jxZzK0H4pbbOANZYYhHCI5hhEkKpb9om6djEEg27QaSXK6/tl0dR6\nqCuL8iLqRGl6/nV/67WX0XVB7t/+hJKlzC0+rST5ZBzHm11wcQb7oxGZ9xDM9sdgDEnXi+91fQi2\nvvmcL7mYiWoyjnOR9ZagrjnY243XrW24KEgpBOsXXDu3b9+JY8nT/Tj/337nu3S6Ha5euQo8ezTm\nk2Q/ngYO7VnTQn1Y0IIWNEfnQ1Kgsa7Llvx9WGp4FkbINh2VU2GtRR26Lk2baTWt8Wgfzi2FANWK\nzThkXDyKlFRUc2hNiok/0ZMkjSdRkuTxNBOiaUOJnNpbzt/9+GO+/OJLXF7t+ydB0fHwZKNWtapW\nHMX6xkYU12/Zz0g0jEb7blxaz53CwbjZngfd6ZH7UxOP2BTeYJ1UbG+7Njc2LhJApl977bVYGPf7\n3/9BNMxdSa4yGe1Fdebi5sUvDEXpWaIxP45xUZtjRNsj6NwwhUBt15s0jX5dY79we8Ph6kyHmQFA\nckgFOIzJKP149KH8jsOBL5FB6EYUNQasDYuoQYEWwoAsSXIn8s80KJX69tNYGUpKSxWatRXKg6+s\nLa+QKtDGg7HIhHrk7tnd2WHgN69JE2bTWZyjIEq/cuUl7t69y+5eE8489dd1u50Yuam1IPXFaLTW\nzGZTf03fifVjF8xUiZLMj//u3Tt0Oz4QS6RR/H/77a+RZe5Z3/6T72Gs5fr16wDcvnM7uk4vtIBh\n4Jisx2cImPI4aMxnsSMEZlBVJy8ivFAfFrSgBc3RuZEUjkrIEVI+VJ1orpuXGh5XnTiq1uNhOlz3\n9bgyYEFCmDNYHhEeG04xIQX+QHfl43Xok43XqESTJmk8neq6QvpTVEuLnrmbOkWHkiJ0JLZltKaX\nd2NIRV1qdnd2fPsyJjrpSYXQ4dQ3EbLt7t27ZFnGm2+8Gcf+3nvvAS40OnhF+v1+7HOe5xF/oaoq\nhBBkfW9cLHfjdUmiGB/42AYrYphzKhMO9tzEfP3tr/C7v/f7vP/++wBcf+l6PLWLaifOb57nh9Sa\n80tPKtHpsJQQJIRievKtfm6YQqDDcO2H1YlAD2UQj2lvaG/w4xhEO1LR3XOiRx9LR1WXlkIiIryZ\niZ4E0HQ6XcoqYBWkcTFoo1GJ2zwTA2OcWJ93BWifldhfohaSzCdkTSdNgJEQjR7uxuUGJsfblB0n\nFvd6yywvL8f3VJYlX/7yWwB8/PEncYMfHOzHZ127dm0Oqs0YTVm6vvW6F+I8jg62IyM0xpL5d3tr\ndhBVPGMMX3rjtdjOJ598TObitbh69SrSJ1rNZtOosvR6vSeS/XhWNOZHQbQ/YJh6hrRQHxa0oAXN\n0bmQFATMZUYeh4z8MKnhpEbI+PUJpIbjvBLOtugzBuXDVYegNjzMwyGUQfoMSq0t4dgQqPhZ6xqE\nO3WVUggponGtPUdKKiz+pO/OyP3pWFYp0htFb+7tslcU/PR15+cviiIGCQ18PkR4rjWhRqWMovhk\nskN/kCPxCElZFt/Zq6++GqWrjz76KHoc7ty5HTMxNy+uz42/LGfxnkF/JYr/9XS/8bfMDMq3v717\nl35vGI126xuD2M8QFg1OasCD0Jajccy+fBwUpWeNxnxa42JQG8rJyUWRc8EU2iSFmBv4oxiENS7V\n+qT2Bveck6VnCymQLabSJG216joewRDadoRHMYMwrnCZkAIbpXcRC+VoO41t9ft9ZtNZ/Fsp2Xgv\nZlM8bAKzaWNHyCpDilu4tdWsmiIG/IzH4+iea4vSeZ4ja+/2HPQ5OHDi/sbFy1gkpXANqVrHzd+2\nfXQ6nfh5bW0ttrG1tcXGxgahRJZqQcGVZTOusjV3vUxSt9SaMc3mU0pGpjIcLMXkqFu3bnHpokMD\nHJfV3FpqR0M+L2jMZ7EjBGZwa+snR957FJ0PpnAoFPM4wJWjGMRJ4hvi/Q/JxGy3M9e+v3Ruc8s2\nI5h/UW3pQAqJPqYSlFAmjsEdGB4vsTTR0Nfr9ihK57ZTUsWIwv2DfYbDIaZl71B+sRUaisIbJJXy\nz4YkyeIizpOMvSQh3dry89LMR1VVcTH2VE1v1S3W3d0DNi5e9iOW6NYrU6miqN1CLqdVTK6aTidU\nfoN2ux2kL4G3NOhx69atiKGwtrYW5zzL0ia2IM+bSEUhybNGMqqKCuMHp/O02cDattzFils3P3Pv\nosV46qpmOpnG99S2dTxPdBLpIDCDJN078XMXNoUFLWhBc3QuWKRFzJ1WtnXqP0pqOCwxwNFSw6NU\niiAxtPEI2u2bY9inCbaFo/FpT4UI5dpvu0YrlGrbQbxUow1G64inoFSToyGEjOJ4WczPZdLx12DI\nhGBUOFF4kHWjyNrtdsi9G7M3XOPO1n0Arlx5EevPkCAleFhItNVRZL9z73bslxCSlRXndizrgxhg\n1c86ZGlKxxe/3d7epfLSwcXLL5D450tL1NUtULTUtkQJ0sTr8Qf7UX3oDQfxnfW7PSY083/vnku0\nWur3Kasyjvny5U20fnZozE8TWu0sdC6YAoBuhdQqmui/oxjEw5gDHM0gTqNStKHhTvoygpiMNY+G\nmBcaa20U7UFEM4U2GuVLvQlVoqvG5Zl6MXcmhHdD+tDeqoz3z2YlZazA1NRQoCyxtbdBSEFdSxJf\nFELXEzLt7AXG5Biv8hhj51SUCA4j55mXEDLOudbmWDduMGaCM2iuX3BZklmaInK3KT698TGrqw47\noajrCAentY5MSQiBRlEcUZdhtj+mM+jH9tIsRHoq+tJ9b6wr09ftOPXlxo0bvPTiSw886zwDr57U\njhDUBnmKMOeF+rCgBS1ojs6NpNCmo6SGR6kUj/JSPMwQeZxof1IpQVdVA60mJSokArXyHYw1IBo8\nBF2DFMHd6F2OgKkg6zVqQlE48bPb7bM/cqdBni9R1xWp73dtZLyurCqsDu5N3aRxL8EVj6a8u3UH\nk0MWKjZNCnIfmFQoQd/Dpt269TkvXX/F9QU1p0IlNJGPoLl1y6E2t99D24BXT0qWVlwatTaai5ub\nUSJbWR0y1W4uXn3zDX7yw3cBd1KH09RaQ+772+v1fFCS9350+ojCzY0UIuZrSNX00RjDUj9ICg4a\nL5z6GxsbfPrZp+5ZSnHt2jX/WVLXjdH4PKAxn9a4GNb9N776i/zvf/B/P/C8o+ixmYJw5Yn+CPjM\nWvsrT6LAbJsCg3gaKgU4BiGVQOgHX8Bh5tOuYvQwCpmRIpFN9qJiztsghcLE2ISaYuZLuHVAyACh\npul23UKejAuEaF5XappFVBQHzfiNaoGxGCoPwXZ17VL0Ctzf3WdzkEcINJUpSl9sdmN5yPZd55V4\n6dpr0WbSZgjSgEZH1eLmzZtz8x/iAaSVTEauzUsb60wKpz5cuvYSwjR1LFRvmc0LzvaQ53lUM+q6\nRntm1+v1uOf7VVRTep3+3HtIe26zlfUBuXQbN83S+Kw8zyNkm7WWJBEkiZtbayzLy679qqy4efNm\nfO7qimOkxhrUeQo7PIYOqwzf+OovAlDb8mG3zdGTUB9+E/hR6+9FgdkFLeg5psetEHUV+OeA/wT4\n6/7rMxSYfbSY/ihD5MMkBni01GB9tN9REgP4SEHbtBXaCapDjII7lMMQUqclSetkC7EMKt4TpBCl\nmrGkSSMNKKkaQ98hdUebo4177plOAjHWoHJ3guedDp/e2+VLr74MQDKxrA2c0W/37hZXrjqjm8kF\nUz/Ooi5ZxkkTUkm0Ntz4+IabM9GAsGZpivTp3uVkxOVNV4p+Z7TN5ZdfA6AipZaCqfdyJKMRK6sr\nfvyKfX+6p0lC6tWaupxw9YorcNvpdvnxe++xPHT3GGOpaudJSZKEkG9W1LMYVCWEiOqMtTXWtBLM\nWjEg9+7ei33R2nDfS1fdTofZbEZ2BILSF4XGfBLj4pOgx1Uf/gvgPwAGre/OUGD2UBLUI3DlHmZz\nOGngkz3GGntciHTbTWmsjVGNLpqudf9cdeiaxIv8bTRm10ZTXboqLUkWKi3nPtQZpEqZHriFM9of\ns+ShzTqege15ODOtTQukpWa4vBTbXxu46Z/sTRhPHChKIhS11ty85Rb8a6tJZMvW2mj7KKuK1Ovh\nKssR+97bgUCIBnZOyvm5DAxKSjGnytljFvuN+xNu3/kDAIaDIesBAu7gIEZBlpMZs1mocr1NkiRs\n3b0TWmR9wzEfKcRcRGYVwpeB1Htb6toxjzbm5JYP5FpeXmbUshe0Ix+tsdFF6Z7vGMkXicZ8WjtC\nUBvkEZm5x9GZ1QchxK8AW9baPz7umpMXmK2OumRBC1rQM6DHLRv3zwsh/iLQAYZCiL/NGQrM9leX\n7FEnykmQaLVMUKaey4UI9LD6lIevl8doMA9ICDiV4bCacGhsACgao19d11EKkUKidR1PNKlMDPk1\nFnLvs9/b2WfqDZCdToeeT+5Js4z79+5H8bONGTAYDgl/ZlkOvohrIlLKiX9Wf4naapZ9unGS9tm+\n7wJ7Xrh6FePDiYVKmM28oU7nhKp3lbHcu7/D1EvSqaoaMd1KtDco5olkPHNh2i+++RaFDxCqEeyN\nFNte8njxlZ/i3Xe/B8BsWtPKyaIaNzbqqe9Lp9ujv9SN89npZNy5fTte94JXM5SU4Oe1Kkt2jRO3\nV5Iu1th4aidJEo2w/aU++wf78VkH+67/QaVox22EgKu6rlnqhZiPp4fGfNJ4hLZxMUgIp0F1fpyy\ncX8T+Ju+wV8G/n1r7b8shPjPOEuB2YhPMNdG/PywQZ0m8AkcEPNxouxpyfWreVaI5gOQj6g6Gcak\nVDdmHCapZPu+Uwv296d0u94OIIgRgJ98/DHGWrreyp8kSVzgUukIxwYS68XHvb09tM+JGNUTBkuG\nrbtuk3VRrA3cTvzBxze5sOE+39u+x+uvfQ2Aj+/d5vrFFwD46OZnFKNpXMi9wQWsZwR1sU8/2AFs\nxZWrzo5QaIn14vvBbg5Jl9c9BoO1lr7nZMVkm62R8xKsVDUH3q2bZmkM3spSxXA4iPr5YZpjEC+4\nPleDjH7lNtHueEpGSs+vuZs3b3L5ksvr2N7epuuL2ewf7EdmNx6PnRvT237aamanl0ZVLrg9n2d6\nGnEKZyh4tHx4AAAgAElEQVQw21D7BH5aDOKBNm1jvDtcgyH2xVpsPCUaXbnNBAJlPpz3sB0hcO3w\nffCTixYsvD4UpBe+T/yJB84Y183zuEkOU1OOrvHFdzodxr5aU95JmKJJPTO7uTVi6sFIDuqCHX8i\nF7pm9M4PAUg7PW6945xMlYG8slxccbr/7GA3gsqqTs6S79bV629G6UDmKf3Ol9397JP1V0l9PPP+\n9j1qr9+bAoZeJDmoC4QJhr1lOn6z9vt9Z8fRjdp59apLA7fGxI382aefcvPe54CzVZS+QtUwH5Ao\nyf07LoT74oXLjaTTek/j8Tie7HnuStW1IySz3BsMqpJ8xUkS21t3UH5cS5cuP1Hg1bPYEZ5Z3Qdr\n7W/jvAxYa++zKDC7oAU9t3QuIxoDPUxqOKm9IZCinpMW2raENsS61Q3nbrsebSs6UCnVcgHOqw/t\n8vJFWTTSQMsqXte1d515F5mWWO/N2N0dMZk48bnT6TD08flKyaj3htMnuNjc81wfJpOKzGOTVXXN\nZOzaHI+nsVy9oUJageo4fTdJE+5pJymYdAlpnE6d50N0KNJTl2ReXWEyY6IMe5U73atpST/3NppS\nc7NwJ/jN997l7deciiCBwcAXnEkE07pkfxRg0wr6PRc8tFeUTG0zh3uFx3DoLLG35/q1NFjiEND2\n3LsN3gvbOpQ/unGTxOeU5GTc39vnQoiW1FNqP//WKCrbiGthjpVyeSSpV/OEtKR+CRbGgIfG29i8\nwMivk/F4zKVLLjr0SQGmPC07QpvOBVNo+5DrI5JcYL7y0gM4jidkENI0UV3R0FjpB9WE4Cq0Orpn\n2lWjH4wJEC2joYgip1Jqrm/t+4VMMBGgtWJv98B/tiwtuc2zlIHW7llFoel1l1rPlS01oWYy8XEC\nWSe2eTCaRJVBKQUeuzBVCWmaoVM/B8Iy8TERVkrWlHN9GpU0thipOGjrNmnCQcA6SJJodDNSkwh3\n3Wwy4Xs3HKDrX/qLf5390qkl49GM8XiM9iAws70Dtu86N2CVzpgZXyBXa8SS27il6HLp2nUAagrI\nBKpsgdG2GHmgDe+mBMiSDks+OvH9937MtCyaxDfA+PJ6mlms3JXleZxjrTXWWArhYyvSblwHVdmK\nTTCa5WXXzmQyaVyyh9TXsxoWn5bK0KZFQtSCFrSgOToXkgI06cpzSTSHpIbHNkJ6A6C1BulPuUf5\nIBo49kfDvgOgxCM9G9podFGQJb5ku9ExKCZNUrrey6DLgxgYtTRYZjZ1J61SGdamlGXzd5a5edvf\nP2BvN4j/nWics5St6D5JkjZSgBCCvl8KShBzLCQSEYKHLMRaWJlAigydNHM9nbo2007G/th9vri2\nxl/5S/+xa99adkZe3ahnUBZMfT/390YcCCcpba72uX2wF5ph4iWdernLjR+9A8CVC8uMx2NevOY8\nBuPJhGteCtBak/fce56WBYlyY856/Ri1OpkVXLp2lem2a6fbz6m8EZJUknhYeaP2Kbx0lGufwu2x\n8kwBI5+ivpSn5F1/T7cfU7+Xl5ejNJG2DMXnnc4NU7AtkJBASZI8VJ0IdBIGESo1A8iibsEszodH\nW2MjDoISci46r/25XUS2rusYKzAej5vyaBAL0Vprm7iCtIvVgrGvkHT79j1WfAXmzYub4KHPk/4Q\nYwOzSKhUs7D2D7bJfBKTg7V34yiKovFkaNNAz4v5uX0o+Y2UKNkSy22cq5oSIRsMRWktpV9KdWFJ\nc8eI/plf/jdjnENZlo4ZAMvpErO+YLbtN3wKsnQdHVcV09JN2ljXGOtrTewcxMpTH36+RY3h7p6L\nYbh2cYM/2HJuyNc2NxhtuTavXL7GzsSJ30mW8uEHH7r+djLqgwkhoGNa1CS5ry+hFKV/z6WF1Nsh\nsCVCJEjvbZpOa/r+nk4vR3ecS3htdXVu3Zm4ZhpV5Sg6afbj07IjtGmhPixoQQuao3MhKVjbJKe0\n1Qch5EPViUDHSQ3h2e5/g/TBO+aQoVCIFOsNesYaVCueoF30pd1OGyY8O5SiO9e+Z7umTMi9X3v7\n/i737t2P0sqX3nqLiT/Rprv7rKw6g+Kdra0YDHOwP6GuSz8PFf1+N7ZVljX377tTU9dObQh9karJ\nqZhLBBPikBG0GVcIzJmbb5UQzjmlFAJNSOhSSpKFgKWiRPvEqdHBdpQOLqxtcmHNifs7e1tooxlu\nuiCpux/tIGnyCNLMPWtaNPUzS2MIkorAkiQZtXHt37x7P/b5d97/PEo3o90xUz/J65ub3PXemywZ\nIISk9L91E9D+HlNqSl8kx6ICWD6pSkiwhGyrLtDrO4mozrqs+NRrcOC3MG9crKtqLtbkMIV1aqx9\naNTiWY2L9SmC9c4FU4BmkwV3HTCXfQgPtzcEOizan7x9LyZqTZD5289qM4S25b/b7bC7uxcjDw+T\nLkL/Lbs7Tofe+vQeF69dYuiz/OqqZrLjxMJ7t7a51W/GH9CMq7qm13Ui6vLyAK1hNPJBRjND7Vdv\nmnbJc28fSAx53vfD0jETsDiU6ddGqXfjPSLjUteIoJZIgzJNaLisXEg0gEgzLnddmPF0MqOb+4zD\nOiFRbhOtr6Xk6QHvf/ypf3gKnimoZEYxCx3KyJSv3ISM78CiqIWNvaytRHvvwZJ3ewJ8PJk2DOaz\nz12RDqDUNXtIal+KqjYCU7o5WemmVH6chYK+H2OFBAzGg64YJUi8oJ0eKjx8XKDcsYjhre8fFaB0\nWpXhNMwg9u3UdyxoQQv6U03nRlIIpJSakwIOqxNHfd++vi0lPGhcfPAEDKf+XOJS4MBSAUdLJIfJ\ntEKjj6pFmWVpFBGX1vokiWQ2dWJif6lPXXuDVC6iQc0Yy2Tsw2w7Cb2e91Yg5+o06ho6vnx7nucY\nfOJQZ9CoP9Y2IdNJgrE21lxsn1SpOv7UidJS5QFMfa0HKdMYnz1kg40LLmBnc+MyndyJ1f3eOvve\n+5Cmfb777j9G+zDlnb27IJworGyXYdfN/2h6v/GQSIX176Xw86p9f3pGEZZDVVZRlUhbiWKztB+L\n9EihSZRFBwTvJCNUDL47LaDrcSOsQcf3aumbJudh0B+w7WMTXnthk5mXDpJJOScFxLk7InBJtmIY\nIirX0VP/hdK5YwowX5nnJAxCKdWgKfNoO8LcC/EMIeYoCDmPqeCjDg2gxIMbP6gOYfMnSUrlo9v2\nd2YxinB7e5vci7L9zU33DJ9cU1cNGEiSJK0gpzpCm/X7/eheu39/i8lk0hR/FRm0SuIFrwRoKs9s\nlGz0WRdNKakrr0ZYfeRCPo6CJyZsUlE3OQjDtSG37ji14IUrV1ld9n0cjSJOwXbxEUU5ZlY4dWpW\njFgPcGydhL3Jrm+nsc8kSlP4IQoMFhgE1OwWnkOe57HwbpuEkCR5MwdaCLS3kczQCC8051mO8N9r\nk6JtCESTTISNa6MsS3I/Z+99dJOZh3r70qsv0mshARxmBm0Ka05rTe1tP+UsoZ66Obvd8jZQa77x\n9i89NTtCmxbqw4IWtKA5OpeSQpuOkxraOQWHpYTjVIajchpCEZNGZXiI0SaoAmnC7q4PfOl25oyj\ns9mUvXtOfK/qmsyfFN1uL0oD1lqUSucyLEc+9Tgb9BC+GORgsMTQpzRro7l5053ASaJYGgxi35eH\na7EAjKUBeJVSYG1Iz25iQawVWF3OZXDG8bckJaP1kbEN1pq5dyGShGHXITzt7u2xuupRm3VFmgeU\n6+b6Dz96nyxN5xCStreddPDy6jUupN4rsbsXQ87rGpLUv1cs1oIMwVQtmXs6m8ZCtvZwgkSLhLVx\n8RulsF4VoDZgQoFfYsizAWpZUmonbWRCUPulcu/uFstDJxF+70fv8+YrLwIwKwuupPPp3UEiq+s6\nAvxOVU019t6ssWX7njMuCrUd0aS/9bVfeqrGxTadC6Zw0iHMVe6p6ge+OyupuSAdcWxceiCjBdRN\nroYQIkJ77d0fx0W5sbERg5rahVXSLIuuTICdnRFKNc/r9z3sWpby2ecOckxKw5qHaO90M5K0WRRW\nJ6gk2ERUU7W5ruJYyqqMG0wpB7XWKq8Y7QXt1HE4OqgsXBcWeC+9QO2ToF65/iIbfedxsEzZG7n+\nT8Z3+JN3/hCAg/GYtbU1trcdsIu2JcpvvuW+wBo3/q2tLYyfFyUl3lOIsIalY9LGZQtx+WHBWlKI\nyPwUIH3AFPLQPaZ5npEZiY/iLGtL6W0qWW+ZOuaFKN75sQuSshL0FfdeiqLg9YtX53J8QkCT0Rrr\n1VQhm+CpbqYwvl9VcAc/JZWhTeeCKYDL0Q+UPkKpaSckNXDmD9oR2ou7bUcw1sZTNkmTOReSkM2y\nardjTR2z7EY7B02lINMBDTPrdEohJBdClp61LUDXBqi1Kks6nU50N7apfQKMywrpX1GaQpo2GZNS\ngAg2DqEbpqZUyziXxTmYzaZRgiirCmsMKg12jPniq0ehSrWNseDsFss9F3ewe3+bi+ubAFxcXnOn\nLaC3J3yw5Qq8HhRVNIzWdc3tW7fY3XPSlkrg0gXXl9n+mKGHXrq8vsFtzziwGuXrPsxyS6FnJML9\nnbe0YBeR6n3+WkeMxza5+Wvbjixt9PawAtxJ3qwhx0T934mgrr3rsjTUrTXbCeHkwMe37sbvJ9Mf\n85UrDiy3qiqqnpcaig7FvnvWrTvvkiqPNqUT3n77F+P4XcHi1jgeGNmToYVNYUELWtAcnQtJQRyC\nLTtOaggnVTu60Bj7gB2hjZzUVgUO2xHCMw8XFQ3kcBOctTeRKaOdg/h97j0B5aTg3v37ZMvu742N\njSbefQ53oZFmpFK+wpCP9x9Po1iZp10m3lUpS0h9FOTycp+8665JEoE1NSF8p66bQqhtL0Ke5xGQ\nOs/zKCnMZjNQjZqhdaM+PAx78ihXKzjpJtg0Pv30M15/xVWV2p8VmHtOGupsDun1nTRVFAVG62hv\nmc1mSB88JtIOBz46VBcl6x56vqpr7lQuIlEVApIOsnJ9rQpDmjc2IUWwo8iY0g3zLkonJRwtahvv\nqkURq+lmcoZtr1Oh8GkZiEQGOAV0KkhaKkcQwuq6ZqYl3/nox24OkoRrQydpFeOaydgBoCfZbiw6\n85Wv/Hz0NiRCYUQjHWjTeH+eNJ0LpoA4HKrcivDy7+c4lcJa66L1gsjaAkw5zjaQKBPdUZJ0Lvy3\nTVqX1N4PNvNlyQCoGxxAISWDzWET3ixFrMpsWgVKlVRMppP4iNFoHKHce70exdRt6oPZXmQ4eS+h\n4wFBs1yipN+IApJMUZdBzC0iNFtV19HV2VaZZJI31aqSxPn5TTCctt1mSZwPo4+vKvTlV36Kd991\n5d26WYfr15xxbbq3z743Gn5y8ybb0m1KeWca269MzcF4FoFSNi+uMy5CIdcC4+MXkrSH9kAuKYKr\nHluh0BW7ZRFLtdWJxPq5UCSQH80I2qS1jozsoeR34cxCx1hsYJ5ocq/Oaa2ZhXejoU6b+hjWl8NL\nkilFbSjCeVTVfFQ71cpYy0buGKGtFT/9tT/n5nK231QtF2Ax89G6sQ7Jk2UOC/VhQQta0BydD0mB\neUtxu7hI4IxFpeMJpqQ5U35DIIvEhDqKyqCVRdVHm23ayDtBmqmLOp76m5sXsTQFRq21sRBqQcnY\nQ7QbDxUOsLe3T5536fV8UddpHU9tqTKEF4uyjqK/5KIY0wyst3BXVcG0zpsUb4gGtfa5aLRGJg0a\ncRtaTIgml2AORHbOuq04KsbujRe/zgcffMDBgRvTtTeu8Ad//EcADDs9qgM3N/fqKYOBR3Eypqmx\nKQRprmL1p2kFw777vLtvWBu6zzu7O/T7LnhrvL3NYM3lfnTzLvvVjCrkqOAQsQHKZKcxIBpFKgJ0\n/rx78jgt6bCrzxKiNhUlkLXqVwofcCTRdNK26zzoEoLKB6zl9LHGkkg3N9poipmJc7M3dMZpEskn\nn3wMwOYLmy2pzQeYtToey4daMyctHFUt7GGu9sP0uGXjVoD/CvgqTkH714D3eMwCsxEmyzbhn20R\nXyOxhMrMNVltos/fWhP1a9WyxBtrmYngMzfgRexKChCWOncTmReNHaGcVEzHjYfgzqdOZeh3+7zw\ngkv6kcOEVDVwZEI20GzT2Yzai8wDmTAtm2clybzXI4j8SIGVDRzc2KMsr2TdmPMvEtCliCqxzRSF\nT+7JlToSjTrP8zmbjFJybk6bxC8iarUQAuEjIaUQvPXy1wE4ODig0+mwftHBnd24+Qkd7zqcFDO2\nC8cs0lYswmHGs7Tcj+Of6hk+t4vVlT6fbbnIx8Fgld3bTsReXl6JqoCV1lV/8sVnK2GRuHsSmxDC\nP6RsPmtjSWLdjfn1ZI1p4Pla+J9SWHSwDwinltYtV2jibVNGQIsnRLdhUVuMx5gk7yBLjfH6SF3K\nWFXKaNjd9bD2yz1u+fc6/eSTuH6vX7/u1EET1GRN7sFgEA3TC9Wx4WjmcBJ6XPXhbwH/l7X2S8DX\ncYVmFwVmF7Sg55jOLCkIIZaBPwv8KwDWVR0phRBnKDB7NIUTLXyO3+vGL69MjZZg68a42LbAB6lh\nJmyTOyGyxnshdDQeAYynhsob/XQxZTRy8fn1rOK6Bw4VUsKSjJ/roooTWZZ1lA7quiSbOW59UM4o\nCQE6Q4yAYtaqeRiGVxUkXn1QqWS47JOglPM4AFjRAYooXVhro/hYWB1ZfU9259Kk27kjWhvCMdo2\n8upWFKOURHu7sZaqdhJAp6sYDAaMPOxabTSzMP5pk5NR13WU9PI8b5J+tCbrdqijpKSQqjkB8apA\n2hlS5E4C6PRW0D7Rywjo9ZaY4eIc0pkE1TwrDb02MgYCKZVBz0uKs4CyFSTHo8/GstIxrVyS+Mlo\nTt9aNnUmjXZqgbUG5YPXWmh1yFntPWUhX0UQgkJ1aRwYLTAS0wjCm1xaie/ixo0bpFnGFV/cJjWN\nqtN+/21SSj2gNp2EHkd9eBm4C/y3QoivA3+MK0t/hgKzzIn58bvWYpVzupSMYlQg750iRVCFQKaq\nBuOG2MkV0of8aurojJJWodGonVAAZJ+tWz6KUClWhk4nXrkwiK7GWa5Jg1g6npFKQTFxL3VaV+CT\nkBJtGzwEoci6HogECySkHbdAppXG+mzM/lJGp+NDo3sK4VeWVKDLJkHoOGAZiYwFZca6IG/ZOtrU\nVh/qqj42JDhsnM3ljYgMPRqNSFTOng8+Ko2e01l13IjqSJRuIxywTUhc6vW61B4PIUu7LK+75KgL\nq2t0X3Buu0RK3v/xdwDIhwnS1mx4XMbJTsmsDGxJEEM1rUK1hh36UEmLqSq6Yh4Q5yhKvE0mMOS5\nCMfWnEnhGYQ9iC8klU3xYUPtwsv9/VVdo33EkzYVeDWtmJXUXn28vz2m6zE0xlYiteayX2cGogFJ\nSdXYpI6xLZzGpvA46kMC/BTwX1prvwmMOaQqnLjAbLEoMLugBZ0XehxJ4VPgU2vt7/u//xccUzh9\ngdm1QWQc8zUfZfxbGz2nQiTB4AIUpY4cfb+YUvlTp68MaXAflwajQ4iqmTPTKxQygHXOithmN89J\nfS1xK4mSozGGma85MNsdkSUZdtr49MPp2u12o6SQJ025d2Ud/Fco2JqlhqWhTx3upuTBfiQaT4yp\nEw5LiMG42va367pRMwAC8JM1hlw2p5wQIvYzz/OY7n24psZa353Go/0DpHDiuyBhMh5HfAJbaCwN\nGneI6e/3+/F0Hg6HfOYlsI31DTrdDnnmUas12NS1eeXqNb785pf8PUv84Ps/AOCHf/R79HvOE1Hu\nTRgOh4x94lieixgCXteGmR9LrXVE1DJZjfABJLrWZKIbPTazSUnedZ/zrBNP3TRJWqXn3byEqXZT\n31JpQ+q66BGWaWqK1jXCQ7kFg25zXkqRNgZRMmr/x8H+jKkvCnzp8jpJkvL7n7mkuJ+7crUJwkub\nkP52Gnmb7Cm8dY9TYPa2EOKmEOJNa+17uFJxP/T//hqnKDArEMdWjQ7qhHOh+d+KCu1z+IupAaUY\njz2uXV0TBKBkliB9os105wDpYdCrWUH3goPtuv3p56wN19EeV49pxYqHXhciieLY5zc/5cKSE+Wq\nezMmwjORgxmpEHS8VVm08gimQOYrSNfaxBejkX59uJfa7XcjrHuaSZI05PPXMc+/TVJIqrqJ3HRQ\na2mcpyQ9ZDsI/fF6hbWGjkwjM9FGRzHfWBut50LIuEHqqsaaEEWqqarK6enA2nqfcup0/0xB2h36\ndizr6+sAzErL9Zdeje1dv36dXQ9B96U3v0JROtXkq19+m8nE2Sq+/73vMvXZo2p1DV8flnw0oxAS\nG4rPqgzlGYbWGuk9RhMzo+dLa++MtqOruK5rOt1OVMGG62v0pGPKZbbHxU2Xx3FnayuqjKPRHtpq\njA1z86BKdpgq29hRMIXLH/Gvo66IEamVyLCeQYhaYD2w53iqIxPaH03JhtPoxv3J3g65P1ReXl2P\nbZ5m8x9Hjxun8O8Af0cIkQEfAv8qbkeeucDsgha0oGdLj8UUrLXfAb51xE+nKjBrBMy8n7vTqvnY\nNjpaa1DGnYalrmNYcD3TTMppYygTkqE35hhpKSfeAV5rpPdQDJOMcscZpq6sXGK0v8+qR+NNhSLx\nYuXW1ha3t1yWWydJ2NpyVYo7nRTl+9ZNEhLdiPCilR9gjYyZgUVR0PFArdQ15JbNS5u+yxI/fKyy\nVMF6rhTSOw/G4/1YIl2lriS6JtQ/tFFlAh2t/9aauRDmqUcHSjs5lTbRS9L2zVtrY8h2R+SMps7j\nYLR2KeNAXVuf7+BxG5Ri2c9fXcxIfFDPrG7qHYCOkHO9XpfrL73M54lTJ9566y1+8hOHIfDSiy/y\nwx/9IPY5AOLmeR5Vkd20Rz6BzIcWGyvo+XDwSR1g9CDr5kxGPk4kG7K/69ZCvztgtDuJ2aH95TXq\nxD37zTe+Ru3Vr2K54IMPP3Dv6Aiko0chH7VBXI1NsMY2sSIWSh9YZWWMWPbo3/65WgWbNZ+NbvHa\n4HL8aW9vj4FHkqqGZQTltU8Az+3cRDQGKkUSU9qFfjC1GNxkh4o+u/e36Q8GhCCypJcx8gAo/X4/\nutSEEBFvTyAR3iYgVcZbr73B+qqLq3/33R+xu+/E180LG/RaMGkBTsxaC0F9KEvy4YDSx+7nCcx8\nwsZkNnVyJs4lVntdd3lzlU6/E5lekjhm4DokCDUfE2sYh5yLpEFhdnH7DZZlovI578HMBzwpqVB5\nw6TytoitJMYbL7TWqKpZTV3lrptOZpR+nvKscb8Joel2+1x+0QVw9RN4953vu76kKdLjRQ7WVsH0\n/LAmfOOb3wTgZ7/187zxxpdi/5eXl7l25SUA/t/f/V0+v3XT9aPX45ObjnEURcH29l3fvsOobOD0\n62jlx0oKP366KbOJ6//e3pSVobOPjMdj1pbXYiWn2X7J1atuLNbYGKk5nZRzKfYP4wFBzXyUCzB4\nkyokAeN/biO3g6owFN5VmeqE0WRCx6tAy3keVY47u7sMvcrW7zZo1kqcDF/0MJ0bphDMBbKtEokG\n7FQXJXXt9M7d7T2kZwKr3TXKaQV1wOIuI3BoVUxQXlfuDbpov8GWl1eY+k185epVMqljFN8Lly9H\nvWyp32+BpRJPwzzPuXHjhr/+Gjc+/oA8cxvBigxrvM/amLkovhBXkCQJWZoxK5uT+jgKuv7+wYTc\n2yfSLHXPss2GD5vCWhONbnOkmuMoVM0OC1hIQeUXa5b3KGaNN2hj3c3laDSO0PGDwTIrmxuMt53n\nOU0Vf+6XXN5/knf4nd//NgDXli7ytbffBuBnfuZnuH79ums/yVleXouMrCoqPv/sjwHo9DI2N52O\n/MGH70fwmq2trRhpKoSkKIpYa2NpacCd+37zIOkO3cYY745Z9zgPWMNSz22cletDet1etJcMXxpy\n4zMnEVTbFdPSHQr3t7eiB1KjsNqSqmbTB3tDkqSNy0+3sD1Mk6UrW5iQAGkLrLVoRUPalg0IKSP0\npq4T7u7MyJST9vJLaxEsVxQFVoQMXknHM3BtkzMxhkVC1IIWtKA5Oh+Sgp2XEIQJeQwWPfMic2E5\n8OJ7VqUUvqKSsgkX1zb40le/AsDdu3dZ9Xh5b735JT654aCxlrrdWIWpt7QcrfX9fp+9vT2+923n\nWR3vjWIc/+f37jcJOeNxLBG/tzeKomdRFLz6ypvxRC+Kgt0dx/oHw2V2D9xptrl5EbHk2pSJoDaa\nrk8C0kJHMR1r6ASTs2mKzqRJQhXcoCJFyvnIzWhHMAITIh+tbQrXqmwuJl4eqr4VpKMkSZAeYt0c\nTGK/Op1OFKt7vQF5nrPrT6rRdMz6qjuF/7d/8Du8cv1rAPzmb/4mV65ejc8NyVFJkpDnOXs7PjlI\n1HzlK+79ra4NmPr6kx9/ciN6CDqdlPG4jO+iriuWff1Naw3f+IZTTfa2x7z0kkvjLssZAy9O3717\nl9x7lTq9lEkxjkjbP3j3uxQeWi3J4e59nxafqpj6rL2MPysbb0LMkVAqYiCoVu5JqfUDQXnzQqG3\nfdjmPRaycW2LwpJ6DAwMiMR67xqMy5Jhz/W/nuyx73MslgYDEtNKeZfHV6U6js4FU2iraqIlVk32\nRtRT9zJ27k25csGFeG5urCF8xeVXr7/M9mTKxoYLnOznHfo9r8daxS/8mV9w3/f73PNlwwbDIR9+\n4MTF8XjMhx9+xNa2Exm1NpR1gFYT1D7+WMksRvA5o5frV6fj4N3v3nX6rjU2VimqyorXX74OuFDg\nIgsuQAPUsaKxpQFTUQYCMoqxBhGSXrARjKYoCpLM0u1ksZ+z/aMDwEIU5VwBU2+TSP2zhRT0l9wG\n65Iwuufmaam/zNgz5f3dbbp+g/UvrDAZbaEnbj5+7qe+wQd+PqtiEm0vL7/8SkxiUkpFiPsgand7\n3qBXOJUE4PXBG2z793Tt2rVoqLt16xN6/r0KIdnb2+XgwL2z5eVlpj5r9c0334jrJ00zcm+ovGAu\nctzvHVkAACAASURBVLDn+tUf9sjKhO+88ycArKwts3/X5ezt3B81yCi2KZVX1/Wci1yKNCoDVVlR\nyyZkPYyvXSDZGhsLxwZq3olAe/U3sXkwKWGxaJoNXpR1rNWxvTPGeiP22lIe3cj3R3ukaxeaRszx\nmBjH0UJ9WNCCFjRH50JSABFz7duoBlVZUcwcp+t2Oly85Ixel9YuUPtgl+ULa9BpUJE63T7Ly06U\nHQyGHPiqRInM2fRBKUJIXn7ZQYb94Affp7/Uj/dr3cCXJ0kS4/PrumZ5uYHrDiXWp14lCZbwuq6j\nAcway/2dbf9czZIXn+VyghYmIi8hND3c/QYVkYGVSl2at5shHwUH2kyxJo2BNHmWo6WHMDOaXHlJ\nKWnPZhP7L4V0J1eAc6ugH1CKJxPWlt1Js703iuKq7PUY+qjLlZ7i9vs3+dLLzmL/wQcf8u6PHczY\n9bf+SX7jN34DgKIs4tymaRoNm6lIwEyjuzRJ9ul0XDv7Y83v/d4/BuBHP3yXnb2QOr0c59UYTafT\nJfVBWv1+n9FdJzXwYmMQLoqCTicEoglyv9zDaZvnDXJSkC6UVBHWXaCie1IKgdam8fK0XM+lTaJB\nUIqS1LsH0yRpjI6xGldIQpOxEpg7m2PqGZXyKdXWkvg+F7pASRkjNOu6ZjQLampG5YO83Iu7Hz9e\naksNJ6RzwhRs9JnP6pqxdykKregk7gX/3M/8Gbpe119bWeLi+usA5KniGs7SCpCphL5PIukkWYyC\nLHTRiH6YqE8uLQ1Yv3AhWuy1qegl7n6lZIOXKMSR2IuiBbwSKOjqWZZRt0A5Qshz90CRGJgt+4Ug\nZdT9rZ7HQmyeRWSc1azwGXyNaBgWi5BNIddaz7Be/UnSBvMhAMLg1YeXLl7ivoeSX15aJfFVqzu1\nBh81mBeWpSX3LqZjy2DjNT5637kh/4lf+mXWXvo5AN54/XX+qT/vwlS2t7dRLZyC4H1JE8NkWjUh\n4HkDFffh++/w3T9xUPDTYjtWXiqKIn6WUtHtNku3Oqjprbt3tvX5rZgE1R0OYvj29vY2ouPG/+73\n3iWxkr19X31J1dSh6rgCdMhktC3AFldgNmQtGq1jRWzV8khMKxlrXAghospU14WLQo0RuiLGcGjd\nhD+bRM/h2lQt8d/WsHfg640MewSP7P3RjAvD5mDbmR7NIE5KC/VhQQta0BydC0lB1zoal2xZ0c+c\nQevyxeu8/abzc68vryJxnH7Y67ZEwBylVDz5hRDU48YfX+gAVppGS3BZVnPViYSQIEO6r4yRf/2s\n4b6dTpfS50fYYorRDUxbkiaxfqNUMj671A3aUNaSLsbjMVIIuj4fvypnVMutmIOWPcrqkEDzIP8O\ndQeUIhagKavSV1oETDIHgRYDa7Tr57QFJHvt2jUA7ny+hU+9oN/vk/swjcl+zeqKQ1qSQjM+2OOF\n15wR94MPfsLN3X68Jxhd24a5PEnmalC0U6n3Dw4oPIpSURS8cM2lS3//3R+11LJxIylVJcPhkK52\nv33l62/T9TU3x60xzfbHfHrHxVJM9Q633/OFaab7gKaqZn7+FIWPgUjTXixW69SKFjpWK/GsrkWM\nM9LaNLgfWjMrfWEXIcnyMrbRxpQophpjWjkm6sFAPRfb0JDWhk7iVOOqLikDQnBl2Zs287ncPUZq\nOCGdC6ZgW0Ed7Sy98XgcmcVKL2fd2wr6vXxukwghor4JxGIsxd4ByuvV7YpMbSyC4XBIp5fyyc0X\nY5u377jFM51M6flEpbBoAaazMm72sNEyv3iN1k32Y5LOib9dbz3PpPQFXqf+GRZ9z33OLg4jFLs1\nMhpZyrLBPOh1lpBS+sXtnh02j0z6TL2aoCzk3kNhasPAZzxOZiNWVtfp7/s2rWTmk4iWlldYXnO2\nEysUvczZDS6sLfOLP/fzAFzcGPLuO9+J7+l//B9G/PKfdQzir/4bvxaZ73A4jJu1qmvKmRN9tdYR\nUdqNs9HVX3n1Vbb3nMh76bvfZWfPh1mTolQIFkqiBwWgnhV0vL2m39+I76nf73P1omMw7968F9//\naFSTZQ0GgbU2cuKaeg7LMgSFaaOxxiJkCO2mRXqOyYX7td1nfxJK+Ek6SYdgO5joklgcyD6YYQsP\nopFb0+Bq5rrDxAeZuf/7HEWRQRQnZw4L9WFBC1rQHJ0LSSEjIfP8qdtfJRNOfbiycZUL3uJdlWWU\nBvq9RgzLsoyyLCN3ttZy/75PXEIRMa8EschIEzMPw+GAsk5j8MzNmzejmjCbztj3vvAkSSiOCEtO\nkxwl05hv0On3yL31ejKZRF90p9OJJ2hRFK5QiX+G1joiO5v706bm5OYFAt+2tpxLg67rIoY9G+oo\nSpZ10ZygWsdQ2ETICBzb6y4hCs3bb34ZgN37e1gPPDEcDlnbCF6aNd541aU7f+unfpZvfuUt1y+V\ncvXSJW5+5ub5V38V/oW/8mtuzmYTOrk34g4GMfiprg6YTme+7xV5nsfgo+3t7ShFvfeTH/Deu+/E\nOb+3dSu+s0p7w17SYy1d4rXXXgPg0qVL0VBZzwryvnvW+z/5MVsTd//tO3fYH7lYhDRVc6ew1jWo\nB4N8hBSIUKCXJgEN5sFe24ZnY23sixCgvNXTGEOhC3KfV9LtdhmPD3w76ZGZTFIITGjDOLDh4LEw\nxkEDBNqlLQk8KDW0VYpH0blgCv3BEl990S24jZU1SutE9vXhMrkvpHp5Yz0WJ5lMJnS8WFqWpUt2\nakFP9Vqx/yGHvisESd8xm7KlroynI7I04/U33gDgjTff5B/99j8C4MOPPoyBSFtbWxGnIIi/4Nx7\nUklisj8NGnWWZbFfZVU1QTVZxmw6JWm5DIsiuCFbMFpb9zE+6jHJuwhvRwmLbjr1m7zXQfocfJFl\nUWSfFg2DsFpz8ZIL8CrGM3ppn97ypu9nP2IZzsqaay87BrnUXebLb7wJwLVrL8Rci1lR0emvcmHd\njecv/Mo/y61bn7vrXnqBzAdi7e3uUniVAebF4TzP+dBnIKZpxtatjwD49re/zc1PXELUpzdvkHq3\n5XQ6jcFLvXzIT//CLzQFWus62oH29nYZ+yjSH3z2gwikomRTM3NWjOn1etEzomWD2dnOG7GmwT40\n0iBt0opMtOiqwado7jEIFSDcVEScBuj2FdZ7w5IuUGRhYuI1QsqYh2NaWZbtQKpA7UNCzRqhPzCI\nlWGfg/+PvTeNsexMz8Oebznn3L2qblV1Va/sbjbZHA6H1JAcWUNlJEoayA4gW4gdAzEQJLHsCEIc\nOEgQjKD4h2EjBrLIRhLAkRVJlqVIckawRh7NWCONRlI01HAbDreZZpPNJrvJ3qtrvVV3Oed8S358\n67lV1VXdXFIy7vunb92+95xzz/J+7/K8zzO4e1azSfowsYlNrGIHIlJIkwRPPGaw6/NTMxhZgtBa\nkhm5ZQCMaHBuVkDGmF8B/WivTQkYKKgDiDAGZld3ouFX+oTQQC5qFY5dmD8cDPDUU08BqIpsxL30\nfr/vQ3wFiUFv6OG0Ccs8jp4Q4sEzQojKrIEDMpl99n3BUhZhNSioAO07Qk7iVzZKCZSiSGwURXMF\nh56ZnpmK5ig0pHCEsGGkNms2UAwFmrZwm9MUzCl1k00PUz5+9Jj/jpISQplVJ0k4cjHy4+PrG33A\nSr1tbKwiG1k+hmYDyhVg8xItC7n+5jf/DK+//jq+9z3Dm/BTP/V3cP2KAT/dd+Qw/vAP/xCAUWZ2\nYTWFNFoXAB589Ky9JuZ8vnX5AogVVlldXcXt8rY9zoBTKYoRspotIPIGijJHWjPnhBHq4dgx8anS\ngS1LUAYGFRigCYHeIeQnqENrx1CloQqz3SQtkY8kOLcALqSoW3m7shDI3XfGCXapnY/xWhzRaHUU\nocQSDy5q2Cul2M0OhFNglGJh1lSNW406Mk+RHR7+XISKPyHEtyABU9ltWcBSmmVeV3HY7/vvZGkW\nOgZCVqriWisIe/OUvASz7aEnn3jCP9S3l2/7CvPK8gpu3jJDM1mWYWpm2j/wg8GgotnIXFiapsiH\n5iaYmekgLwa+jZhlteCkUhoNgSmUNkQcjlY8AlAyiuZUF9LOa4AH/cdyMPRTNzOtlq+PgGis90wO\nm/EannzkM3jgjOFCHG72kSTmfFy7cRMn7EBRs9nEiROmep/WmQc93Vp6D+vrW1hdNvSbnUYT3Vnj\nYEabPWRZ156nm5i2g2NMlPjCz/w35vzrAnRq1kvO/9Iv/Dz+0mdN9+LixYsoLTFOjTNPHT+SCol9\noM6dex1PPvEE6vXQabi+bgnEaeguaR2UxIQQHtFICEGjXvfT9oQS7wwIJbtSmgkSWsMqosIHzGiz\ns6p6maW5kxRlodCpO+RqMMoYMriUQcGO+4Ai2aaEViV22e4gCKGBu3OHlGI/dkCcAkPbPnzTnWkM\nE9sqYwlqtg7Qjx5wQpS/OEmSVAqHshT+7yQJk4Gj/haEZV6qNxphW7mBr8JyPmZZhrrlUKCMeQcz\nPzfve9Gf+9zn8Oyz3wIAXLt+HTxJkduWj3MiQBVmK0ugM9WK3g+kprVaLdCl9yKdhox7BeV60oRw\nLTlCIFfWfZs0abUxGkYCuNayLEPb7lPkAsSeywx1HD12Aix1HItt354702zh0Jxp6R49fMTjL8Qw\nx/U1k/f3NjawtdHDnOUt6HY6UBb5WGMMuX2oW0mGG5fNd772b38X718xcmjdhS62li5hdPMt8/ex\nR/An3/i62Y8QHiZcFgW0LY7WsiZ6mz37mRzf/vYLHlsyMz2DoS30bo6WA4sUI16huNGMitNZAk1S\nz93LEx4IgqXyUZtWGsJpa1huBOWeSc5ARVR4pOG1imjgnYMyqEWKjX7gPXBTjoRQrG0sm+9KBW4j\nMIkQ6ShlOEpjJ7GTg9gretiPTWoKE5vYxCp2ICIFQgimp017yrSqLN+f0JVWo2NBKkvhh2GUEmbw\nx3rQpJOEllxe+OqzlPWo7hBAR4QzQKrwfyWDVTVHt9uNOAiCB37j/BtYWzdy64cWFrG+vhbYkJX0\nVXLGmMfes4hSPslYZXtKUk935qjqAVPfmLItWQkJB1oTQoBx7mcHtmBYosxxBiANWethy66uaVZD\nvWN+y6Fjx5GmKWoWl0/B0bRtxFqS4MiiSRkyysFs3ry+fBUX3zMh+rX3L+P7v+9xTNmBIhVpZPaW\nlnDohK1FaIGVmwYI9qP/4V/Gi6+amQZdTzBf9JFbFOX1pfdxaN6kLOsbG9DSrPqcUdRaJsWYn5/D\nrSWTroxGIxw7fshHN5fefQeShSq7dgNNhIK4GYl6hjrccFJqU4aQx5cR+CiuI+zEF+pM8aCEFZF4\ng/rWZXi8KCUo5AhMBlWvga39NJtNMCeEixGkbtjPjEcGEaM5sEvUsHfNYS/7oAKz/y2AvwvTK/ku\nDJtzA3cpMMsY8xTn7XYbnFnyiLLwF2N6OjxVlDIfllGaoVS5p11jEdQsqYU6Qpqk/iTlee6dBYdT\nYbZ9fmpEWgEgR8SLqCTeesuEu/1+Hw37EB47dhRClL4+kCQcy8vL4be4+sZwBOWQa4oiL/r+hjty\n+LhHPq6trvp9NiJYrCgEapaifrVfQGnh+/xSCo/8TJIExA6OHVqYh4NXnjn1AB5/wiASFxePoiTE\nKyS1utOeQGU0GnkItyaB0PXNdy/i5WcNEc2lty/i1NFj3imkjPpWXsoAYVOZ9lwXJ87cb/d5BP/l\nz/xXAIDbq6v4nV/7ZXQPOZh3D9Q+/OVQgFh9CcYYCuGuWeGp11dXl7G2vooVe55VIgIcPEt8ykcT\nDs2CHoV09R2HUpTVyUVn0p4/HckT7mSh+Ee8qhNX8AXYuIMolEFgOn0MhaAeNRoVkA4yresgxH0x\nOIGyDKLJzkJqceeUAqg6iL3sntMHQshRAH8fwJNa60dgFuD/BBOB2YlN7C+0fdD0gQOoE0JKmAjh\nOoCfw10KzBJCKsVC/z5lnl9BxiPFPFSBzQhqPWgWxjhyqX0obb4bVg2/jyQ1rLo2HWnwwJIc4+uv\n31jCq68YLcOp6RlMWR1DQqgZtrKr0NZWPyp01iBlaOMNLLPwwsIhbGyuhm6ChKePlzJ0WUZCApZ5\nhyccpQUoLS4uoigKH10QqrF42LBSNRo1tNum+n/k6FGcOGpC9O/71BOgFrU3Pz2L9f4AaJrorF6r\nQ7mxahBPoksIwRd/+9cBAM9+68/x7nlDw764cAjobfqx6lKXGPTN8XcX55FMm5SneegQmocMB0aW\nZXjqcz8KAPjmV/8IX/gH/zN+44u/ao7tBz6Fty+bluRgawvSXvVOXaIhTUTW7U7h+HGTlrz3foo3\n3zqHrG5+zzAK/Skh/voSxnykAMBHQKV9j+4RUlODJDPf3UP4xa3QJmKwyEspIRydGjUKYaTC1hwQ\nkQ3LarU6CGCvNNn+TMQRQxAC3iulAO5m/f8gClHXCCE/D+B9GDGkr2utv04IuXuBWUI8bRhH2z8s\nBUaBB88hBwFkLPOKQm76zHUGhsNIcRq5D2u1kGFbtQxOv5IxBk0JeLod5mrk1CzlVbeLltVtuHHj\nBo4eNTc75xzr62v+AU2SMJmY58Kj6KSUXgOCUoKTJ077kL0oAkIySRJoHUhe3K+RhCP1bc8cSRK0\nFtI0Ra1uztlnn3raO6X7Tp70N1urPYOuZTMWI4m0O+WdTz3LMLR7okr7UHO1v45LV4xM2drWEF07\nKJUQhjfffBMnF82w1I3BKj5x0pDWtBYW0LG8jCOpvexenudoWJm4H/z8X8by6jpyZf5+960LGIws\nGU2+ibmuaY8qTdDtmt8yPz+LZ5835CvD4RCcExSWayDhHKTu8ABJEJjlFMw+OI4sx+zE/COih4m7\n90gkIHuHRFxrFZTLd2lhSkY9kY6jS3PuSysF2HuLJxyMuM5IgEzLSP7P8TI4c84B2DmlGD+uvTQq\nYvsg6cMMgJ+EUZ8+AqBJCPlP48/sV2B2fe2OJYeJTWxiH6N9kPTh8wAuaa1vAwAh5EsAnsI9CMw+\n/PDDGqVVP8poJJISSa8T4VFnSkpfraeEgJBQNErTUFCkkoK7ohGXvkNBCPGYekqpL+zZ4wqhoBB4\n7z3TWz9//jyuXjWr5szMDI4fPwnAFB2z7AY2N7f8til1AjKJjxqOHFlEu22QeFNTU1heXsLamukM\nrKys+HTIzNwHfQiXfmRZA3k+8MdYlhKJDS+np6c9mCtJamjagaCEZWg27bh5q4l+aZmOWjW0sjpK\nESIXR5u2fnvFay0gAf7G3zSqf5944zx+9Rd/EQAw3+jgpVdfxicefwwAcPbsWUwvmMhJJsSkJgCG\nwxJTdQdqWsX6lvm9vc0h/u/f+nUsr5jzCdlDBlOcbDUYmF2rWp0mrt80ILFz59/0Q2eEJKi3mz46\nTOp1j1ZlCYeIhV+d1kVUgJZKglEGGq2JpWVPImrnCGGclZmQwLRNKal0k5wJGVZn6b7nQGqMQTps\nRDlEs+N4EgRGdnAMJPwOqQQY5dsiBuDOUcNuUcyd7IM4hfcB/AAhpAGTPvwYgJdgJOnvSmAWAGAB\nS1Jx8Djv8srIdQgHH6YUnDjKLEOT5fgIR3nfh3UGiuoGSoiPixjn/nwLIZAkiU8TAPjXt27dwrlz\nZmLvtddew4N2aIox5qv9V65cwYUL7/g6RHdm0V+Y7vQcpmYMnPmBBx7Au+8auvlLl97DrVs3/H5a\nrZZPH7Is86kIAHQszdZgkMPN39dqdUxNtbwzW1w8ikc+9X1mn90umraSX6810O0aYRXOOZKW2UeR\nS2QsgbTdA04ZiFVwzusNP2UqpcTMvKmdaCG9TNyV1WW0Gk1P2Z7WMgjbuhwOCjRcHl3k8CTTpIcS\nxinkQoIxiraDXhcb8BdHGEZuAEjSzKdYeZ77TkzW4NiUEh0vdbczSzKlxDuYuOXIaaBNB1ChzNNU\nBnVvWb0Px+sKLqePuRQAQCpX7wrvea7G6BiZtChD3vDXUivh4dt5RPFHqL5jOuFs3EHQXVSo72Qf\npKbwAiHk3wB4GSZVegVm5W9hIjA7sYn9hbUPKjD7DwH8w7G3c9ylwGxsEkM4+cRa2giycUrBrZSE\nMFASe23qRTaajY4n4TQrrls1WCDN1MwKLAAQ1mvb1X1rawvXr5sx4KtXr+LSJQPTffjhhyuycRcv\nXgQAXLt2DUmSYGHRVPm7Mwtot8xKd+zYMY/vv3z5speaGww2K5FJPPodRwlKCZ9iZFnDg2rOnPoU\nzpw9itlZU/g8fPSkDzkbjQ5mpm1BkNdQ5nYVVBRtl2JwCVCJzEZktTrH+m1bKAXBvNXVZAnHd22k\n9N577+G//1nTRPqDr3wVf/Ov/w08YDkoWMIxtFGc1BpLlixUAVhee9NeF4abSya6+uIXv4TBZs9r\nhcrhKuY7Zp+b+dCnQs995yUIu1in9cxHADnP0KhnYI6qjYSwXIN5fYzYkqjj5KKGQMq67eNmWyRE\nEISGTsS4xUU8rXWYj4hYlGJGMX8MqRuKKyBt0dTdLwDQ6/W2RQvO9lOEjLtnd2MHAtGoNeBAcYwC\nxKLSR7lAoxZAS6TCUxjRsUUnaxTRTnHOK7x61FFpUVppVRZF4fPooij8AzoYDHz4OhgMcPbsWf8Z\n7esehhrMzSFwzgFi0YH1WXBuPtfpdJAlpm5S8KHXc3TmJhuTJCAyy7Lw1GwAPIfEmTNncf/J+zE9\n78A/BWbnzHDQwvxRP2/BOY9CaxVAUc0MSZpFYWruBXAEStQTi8hMOebmTPrx9OEf8w/r3/uv/zvM\nzk5D2G2vrC6hYW/MzeEAvZ4d1EolGjYVuPD22/izP/5jAIAsB5hpcFy5aFqcs/Up/9skJXjzknG4\nkpHAfJ5wZNZxMC2RJimkfZoLIVDae6Mhpaddq/ANMOoBZoB5KMuxsN/ZTgNRWm33HHIHJ6ErKEa6\n62eNkpTZ5ohkIKX7oZv+M51Ox0+sAtvTCb/tfXYp9muT2YeJTWxiFTsQkYJUCutrBrRRq9UMYSlM\n90FapenhYIi5+QB5iHUR0zSFtErPMSkrpdUZA1cYyvNAk1aWJYbDIYYWmru0tITnnnsOAPDNb34T\nDzxg9CVOnTrli4G3bt3yqUBZljhx8gFPinrixBnM2SnD++47iovvngcAvPHWOeSliTq01uh0Okgt\nNqLRaODGDQftAIZD87kkqflIYX7uMI4fvd9/vlareQxEf6vwFGiE0CicZb4TIUQe6QwIJOBeZl1K\nhcUFc25XV1cxtOe8lXb8TMrKygoGrsOTtbDa2wLVdkSccGzlIe3hDQsky0usrZkZkUsXL2LKaUlS\nDSiBT5y23ZjuNNozZj9f/Orvemas5swU6lPmO2VBwCzMm0sJoZWHQBNKQd3wAclQ2vsnYVXWIWG7\nLZwnKKUKgLcd5h68RR2EWLtzW5RhNRtJFFForbdFCK5YOR7Ue7LikqOZmG0Phpvb0gln41GDtuOb\nUoldi5D7tQPhFJSU+MY3/ggAcPbsQzh65BQAYGP9CmZn5+xnFNbXTK7a7/dx6JAZ2pFKQsi+P/lp\nmkaIxXBBjHMI7L0uXB+NRijL0rcen3nmGe8UHn74Yc/dePr0aazbIahXXnnF30gPnH0Ex4+eQrdr\nWnIzMzOQdqDn0nsXcP4NQySyvHwNxKpATU1Nodls+nD+1q1b3kH0emtes7HTmUG7aR6WBx/8FB57\n7HEApqbRmqLY3Oz7fR49ch8Ao8noRsRNS9QhKhOPyWcshVYCxFHV1WoYDqxQTaOG1VXzO3MxRMPS\nwZVlBwN7K/f7fZRyCGqLP1KPkPKoZWd5CLWmqFvymVarA22dULNZw/Gjh5Fv2joIUTh6/0kAwC/9\n1q/5gbR2d8YD1hijfiFgjEHko8pDGnMbeOOpnyPQUQ1HiBKMhdmDCnFJTKout9cm/EMd5epSR1R8\ncedMBsCccxAkqjf42kNUa6AwjiFYNZ1wNu4g9ptO7McOhFPobW7imWcMWu3ixYv40R/7PADgT7/x\nDfzET/yEf//pp38EgLkplyyhJ2UU9UbiB6oGg0HgT6TUo/tkdIHGNRTKsvSFv+FwiLk5s2q221P+\nO0opvPTSSwBMDj53yKD5Tp44g/vvf8gjJ2dmmjj3xmsAgDfPn8PquumzT3UWvShtmjXB09S3/ggh\nHoMwP7/oCU8e/dTjeOjBR+12Z4McmtrAYDDCrM33u9Pz6Fi0ZZqmaNlC52jU8wK1lClTLINRPqIs\n9Q8VoQStdt2f22nLyNTvD6B9TYRhyn6+N2JopQSJnQYtc4Vpy63AG4f8xODKyhKE1VY4ffwETv7Q\n5wAAq8u3UPSHWHzUnGclJH7hl/8v89uIRtMVdDtNUDiOysCKVCgBUOpXW0UA7tqHzPxWZ54dKdFA\n6VZTXZmS5Jx7bIK+A8mKOXfbGZJ0JAJLEJCxsQqUskSvNHIScSHSDegprSs5/Z0chHMMWcb3VW/Y\nr01qChOb2MQqdiAihc3NTV8Jv3z5HfzSL5oRZSEEXn/9dQDAfSdO+HD7+PHjaLfNaqaVRL3dxPzs\njN+eC7OSJPHfYQzQOlTbnfX7fbz++ut44QUzFtzr9fGZzzwJADh9+n7MzITtuvx6OBxixqYLo9EI\n/X4fc3Mudw/MT1tbWygKm4dz4MgRM7TU7XbR729gxbIeSwCjkfHo9Xod09NByLbddqlEx283ZW3k\nebkrNl8oU5PgadXnu3OslYaQORInIMMoYFfk6ZnAHVmr1dBpZ/48ORToPKVotVq+3sPQhWTmPFGl\nILWJuo4fOYbSCvEeOTTvdFvRTA14zHFxPvf6d7BqPzd7+DjQMecykwQkCbWj3NYE8rKEUsqzIHFJ\n/YrMeLxSqgoqEDZdYgCKUVCSGjfi0gFa7TrEqQFVCtLWPuIWqNZFJVrwnx+LTuO/1XhnI0YlRm+P\nRw37rTfcrR0Ip8AYRW/LzD+MhsNAi14U0HYysDPd9Ai4xx9/HO9eMlN1w4HA4NbtMFDEE/8goklI\nMgAAIABJREFUlWWJJHHFtVBHYIx5zoNnn30Wly5d8hJeTz/9w76OMTc361s63/3u69iwakWHj57E\nJ85+CgDQanXR7XYxGpmLcuHiG/j2S8/74z+8eNL8SEUxP2ccSW9zFW9eOIdWyxKcJifQtbJxDz74\nIM6cNt85cvg+HD5shosIpVhaNXUPRjkefOABEHv5Op0Zf2MlSQ2gTl8hnh5kcJebUAGKUCzLGEOt\n7ijQ6j6UzbLSpwJTU1P+HDunmtXMTcmRY6Vn6gudVhOJJQzJRz3UG/ZaFH1IERCct28tedm6N9+9\niCu3TZpVn5sJk4lUe7ShlgK5TfGcQ3ApA6UU1N/JEr7WHDsEnYUQXUqwJAsKVSLUGxg4VPy9HeT6\nzLEFZ0AVDaQ9hEC7SV0KL1arlKrgFOKaglGg3gUsMeYgXDprHMT+6g13a5P0YWITm1jFDkSkQAA0\nLBvzVKuDW1bLMU1TX0DsdFp44jOGBl6JAU6dOAkA+M6rL2MwGGKqb0L7kua45oZguER3yoS140NP\nL7/8MgBT+T927AROnTKjv/Pz836f9XrdRxScJzh50rAfLy4uYsG28DqdOUg5xNVrlwEAV65chrDy\n7/Nzx/zg1PHjx/HW26YTcfX6JUhZIuMGBfnoo49iwQ4UdbtdnD5tui/1rOPD1ys33sa8VW46fPgw\n6rUWKLUDYkohtWPJUm8h+PoISENCAY1RhSxresQeZdy3cluduq/GF8Odw1wXMbgVqdFooFEPg0LK\nohszzj3y1KQ6Zn9CGMo6F8l8/kd+FEdsdPTbX/9qpavg7R6QebHJiEVJK1UFjzHulbSAalvbjT77\nv3eAPxJKQFV0nuAYnViIOhiFioBEFNW04Y7phDPG9l2EdOau0d1EDAfCKSRphvssgcZwGDAE9913\nn+c7/PznP4+ZGcsHkBeec+CRT3wSQgg8/+0XAZh23f2nzeCSLEusKAOtbTWaFRTjqVP2wavX0W5P\n+3yZMer5HwF4DoEjR46g7zQYaMhhtc497gAw2IpPPmxah+32FCh1bbwehgOTfpRFgb/+k/+ZrxGc\nPHnS7//QoUMe6jzMe9jcNBd7cfGwF7HN0gaSJPMIPSkBntgQWEStMhl0ChSlyBKnxp1YtKNt8fG4\ntZd4BmmkgQ1ba11pw/V6Pf+bR6PAe6GRw4MwhcSwb35Lf7AB6iDrhUAxCk46h8L33jZw6FipWkQh\ndyEFlE0Fy1IhE+EaUA6vQs4YtqUNgMmthVMN17rSPVBagztaeISePwB/Xcx+yx2xDWb7EZ7BOQjC\nELfFkXCvKjXeZdjLQVBKjWPdZ73BmbtGcUqxl03Sh4lNbGIVOxCRAiUUjJnwf2HhBJ56yqzOp06d\nQrcbBD/yzbiAYgVFNQEhFJ9+1Mz2F2WJG3YGv9VsYrprVzoC8Mx4/Wat4T1+s9kxuoKOoosyD7hJ\nktSvCFNTC7Dtcyil0LJFT0IIlpeXcendK/7Y3OpWlgLNhhUUbWRoW699ePEYjhw54nEHSikfEcXF\nqNXV1cB2xGto1AxqcTgcIsvqEXFoEIMhlKK0BVUlpeeg0EpH0Y2G1jqKjlhlv86yLPPf4ZxHnRxW\nGfeOj5sQ4tGhUkqIkSkgc0qg7Hg2ihIrK8tY6RmQ1OXVW7i2alSdiqLwq6HWGqUtNAqoAFDTxGg7\n2iooJSosoGNRQpwy+HM0dsycAMoRvzIGHkVONH5EGPGUbpTQij5EHCk4LcntCU+g/QPgowazn2A7\nRQ07phR7RA3NRGAwNFFDnFLsZQfCKQAEDz1kQDoPPng/hJ0zrzcTuPCLSwFl2yv9Qe5vEDP0hCD0\nUmtium0eiunZGR+iUygklme9lAKt1pT9/siSrjj1IBoEXKSElOY73W4XTSumopT0acGtWzcxGo2C\nk6lPY9pyFBLU8ImHDUz68nsXcPLkSQDAf/DU05iamvIoRqWUv+i9Xs9v6+TJk74NKqVCZvkPhBAV\n8oysViCmFIiRfy6s54nwk6R3ouYar4o7c3By95l6ve4f/izLPL9EGtHaKaXQbJlOTp7ngDROnfAE\niwuLrjCPtbfewKYNbxlj/mYvpPCEKVJKT5mmtaH+d90EEA6QkI64lCEWaN2JY9HxrmgV6iSxQhRj\nDMrWGrQi0NCeVj62iqoUTSozmjEc39wz1XTC2U4OghBSSR+AO9cbxr+/W0qxl03Sh4lNbGIVOxCR\ngtYa1AI+GMnQscSdQoyQSyuzpbPgNYkCT90qwZCkCTIp/XeaLasryRLvmGmSQdm+drtZj+jbkkoY\nTGlYXdO0A0ptLz/LIq8fKNy01lhbW/P/12o1cfq06WQsLi769KPVbuLkicfsPmhFzn4wGPjvdzod\nD5gyQ0+OkYlVKNuEENBwxxDT1hHQ6Pdw5oaewvnmnNsIK4TplYKWfS2EqKxQ7nVpwUOe3yDPfYRA\nCPGQcaYEcuEGgAiELWwWRODWyhJee8NwNTzzxitYt7J57akpf24JD/oezISD2M1kdA5261N4cdhI\nDxIwKYNbfxmjIHCfI07f1gClRKDqMxiEaOM0FCRjJijXydmZ1+DOUcNOKcVeXQrKmI+Kdi9E3tkO\nhFMghPnc2YT7llsgkyhHtnpPC4z6Jlyt86Af6VCLcU5csxGsVNFNLbinptqJT4/S7RctfhCSJKl0\nL9xD0Gq1sLi46LsEWZZ5mrIkSf1xHl44XhnaAkIFu9Fo+M9lWYbUDrEQQvxnY/IOSqlFK1p6tTL3\neTCjmb/pGSkqKUN8Y8Z1hMpwjlL+uOIpv/gznvgm8jTuBi3LEilcx0DBAeqKYgBlQ28pcly7/D5e\nfM22aK9dw7S9/kpro9oFoNQy1E0K6anMKKWgjO3YupSS+EdNa4Aibon6g4WkgcIPjMEF/YRQn1eY\nVME5TgkJBe0mMwmpKFTL7dmJ/V4QfjW72s1lufOsvVL6TinFXbUxd6k57GWT9GFiE5tYxQ5EpDAq\nc7S6pqDGU2BocekaIxRDW0DSCtS641yqyqrNmPJeu9TK94mV7EPb9xXnKG3RqKEzD1/OssxPODor\ntfm7xlmFUNWFtf1+30Om8zxHnue+INhsNiPattRHEJ1Oxx9zURT2uANrcxz5hE5IzBpFqys7I/C1\nrRiYlAl/zMVQ7gilTdMUo9HI738cl++suiolKK2Cs9YEgK58z9PLFTkKHVIGKMdIxTDomxRh+eYS\nvvbC8/jeBYNNyJqdEJEQEyEAlk7Mph8KYZXdqVAaMyfDnycNaWXz4vVZMg3GuNdiIJT69EErBaJc\nx4JAunF7d7KZwzMA1OIZSr23psJ4xAB8uCnF3RQi97ID4RSkVnj2VcNh8MQjn0C/NJXsbq0GaVWI\noBWEBbyIvPQUi82FLmY7NT83n2rAke4JGRK/QUTTVhQBhFKWpSFpcdTbWSeSMqe++t1fXfUXMcuy\nCohlZmbGpwztdttTuzHGfNuPkMAH0Gg0IKX0FGSMMU91lqZpJUyPHwQP5BFDMFDfBkuzBEnixoIl\nhCU8oTRmNqYVsd4kSSo3UFxHcBanZPHoufmc8iQvscNhNChMaaXBWQi/vfhMo1EBBfGEQ9l2sYIG\nIon3+PhJ9BADgcKO1VuApcI3HQfbHqUEnLiuQurplQlGlc6M1hqIuhQOvES09vsyfRBd6SzsxxmM\n2071BmD/DuIDpxT7sD2dAiHkXwL4CQBLVjMShJAudhGRJYT8HIC/A/OL/r7W+g/3cyB/8Od/AgBg\nXGEgzer62IMPwmkkZVL6B/Tq1atecXmhGCA9fgSDGwaOfOyTp327bjgMbSqhla8ljEZD1C2s2j0s\nipk2pNIKjDD/WjinxJnXkKCUVghW5+bm/AM0GAx8S5MQEuWS1KtmuyJfaB3WPH06ifrs4xEMYeYm\nqfEMSirfUkuziK9SSI98dPyTACBl6YfLGGOVQqEQwq/0jnPSve+Yn7QWlYdIKe3bwINBjsTe7Mqc\nOHMsqvQ8jv3NdQy2jGP+lS/9G/zZc99Cc9EwVBVEgbqHRSpQZW9LzaG1dXBRDcEdn3sAclGAhSko\nv39zzuw5JtQPVxFSg5IA4+EB85FCdL7jR4pzDqlUpTZRIWe5CwHX8e/s7RwA7yAi57BTQfHDcBD7\nqSn8KwB/Zey9HUVkCSEPw4jMftJ+5/8khOw/bpnYxCb2/7vtGSlorb9JCDk59vZPYmcR2Z8E8P9o\nrXMAlwghFwF8P4Dn7rQPqSWmjhrE1Z+/+ZJvyZ06eRLlwKwUHRAs3TSDUjevXsGjZx8BALz8rRdw\nToeV5iemW1BtCyRqdLA+cIg86tWWhBB+pXdhpJMy11H4OW7Cz8+rSj7OeYKtLRPdEEI8FyBjmQcc\nxTUBEwqHcNhENm7VSKKwPISnCgLMERKAAVCeYUhKgOq4E2BnB0QedQ+0TysJMZGKSxViDog4giCE\nV1IIn75YNmv3fS5UmBfQFJD23HKGQd+0lPu9LXz7XcPS/LU//QayuWkI2zEhnEGXrvenIODGjTWk\nZfZuUlS6IpxzCMuPEMdTlFFYgS6QSGB4pMN4NLcRU5mH6MJFYVoTk8KgOrTk83d7EgW4p24jUvvV\n/oNEDMBdpBT3iIzcj91rTWE3EdmjAJ6PPnfVvndHY4yitMdMKUPf8r2zRhMrPXNTCSHx/pqpNWTI\n8PJL3wEArN64ASYFjh83u2FZioWuCUt7/QG6TVt0IzV/8re2+uh0zAlP0wT9HOBJ0AAIrcOQ9yoV\n2nMiFyGETROsr6/5+oBpKab+tbswMc7BqWwz5tqNKlIQ1kjSaEpPuHPEvFMDsK0d5x7qwWCAMhZT\n3cEcLiHGZ8QWoM3EtxGB0MpljKEsC2TOEVBA2jRPFrlv9S2trqK3ZCDnb91ews//H/8bAKBzZAGo\nJz6cZzLk9IUC7NwTtA6yZ7lOwWyrNlE5hpQisc5TCAFt6xAJTVG4EF9qlJaOjEXVAEGVqXdE7+V2\nkIxHWg8SCtotBC7VivkYXb0kyv2J/HBSCrPPqoMIqEhzdN7u4CBiOsH92gduSd5JRPZOFgvMiuLO\nN/HEJjaxj8/uNVLYTUT2GoDj0eeO2fe2WSwwW5vpaIccLPoSfSvW+rU//iMMrahoVmhsXTO7OX3m\nJE4eMWzOTzz5aaxevoyzZ88AMIW2cmR81HSniyVLMS55D8NhAAt59qBm17A7R3wLrsBXpYcPSD0R\nMfVs9ftIsgSpLSJmvIqQdCs4EMJ6pYTtYDji0MxTp1ECDEdOV1Aii2NjFliq49WJRcNCcYeg0hVg\nxKMaXRriQmulVAWI5LgZtFaQ0jFXyUr6BQTRVO3jhO0j1q4ce+7cOT83wBOOUkWfk9rTzQsNJFYA\nKOO8slq6botmDVAlfTrDGENuj4lKGdI/At9CjHkRiH0tXHAYpUhxJMCgwj6VRqmZn6Zi0TJIot/7\nUUYNexYhgW1Rw72s+vfqFH4PO4vI/h6A3yKE/DMYefoHALy4nw26qi5JJABz861urGHKcjHmYgR2\nyPT/3169jZlpy62wcRvHFubx7ZdfAQA8/vhjWNsyt+J9Z06hbqcPBQj63EzsLfWWMW2pxDY2NtBs\nNreRaQCm+xA7i6GVZtva2ooQiCmazRZq3LUeg/Oo1WqVSrkQJiJqtmrIahzcxckQcJdCiBJp4rgU\nBRCFuNrm6mAESmlouz1FKNwckuE2CJONDvNQlqrSBozVtcfVquLug7sxpVS+xqG1AgWF5vb/BDAa\nbNpzVmLDUrNdvvwufulf/goA4LU3zqF7yqR4I0hoDXA3EaUUlD2fDco84Q5lFP2+bSVLHe59Ztqz\norChPQt1EAXt49+4PhI969BSg+idh6RiEyp8iVNZefZiWMR+HcS9OAa/rSjFBO7cpXCLD4v4G+7G\n9tOS/NcwRcU5QshVGO3I/wk7iMhqrc8RQn4bwBswd/rf0zqqgE1sYhM78Laf7sPf2uW/dhSR1Vr/\nEwD/5F4PSGuNuu3zJynHpk0lknYT2YwZXS4GPVxcNlnJwydO4rkXXgCsR7y11cN//lN/GwCgCIeb\naCGU4+ayKVTWmg1s2QxgsWbGfh2HAo1WTarDGPXm1qYPfweDwA5FmRlu0h5nELj+46EnIUT02uhM\n+MIjlSiLiJHHFrsI1M5VY6nR39zwhVCtNUY2zeI8ICIBYDg00UStVtu2usTpQ2Xz0mEbQoivtfDp\nFCG0sgoqCNxcsfoWzTb++b/45wCA5158ESs9E53N3HcU1GpDyHIIlqbIbGehSDjSCAU5tBEQRrFE\nfEiNxjtEWlJQHyhLlDsU1ViUVmlmogVXfNwtYojZm4vdhhtw56hBO3JYRj7SIuTYEYWXyd0nAwcC\n0QiEefYYpKKVRtK2eSg2sVVa5SFZotkxHQYqNbaKIaTlOljozKBvORLXV1Y8/PjGjRtewbmRN3Ds\nqAlRVzY2UW9mqKPmj8VV3LXW6NuW2u3by95ZZFnmSVGSNEESQTGEEBVOgZh8hFDHAZGhFCOkLqaN\nKvwaCsxP3IWagJSy0h6jlFR4J+M6RtzlcE4lRio6gFKMcNxpuIlzgrI0+6C0eqsopTy/hQYwM23a\nyOsb67h12xCmXF9dwbFHDDVe0q5D23QpK40QS98+JDzav4jSmoo/LLXvBDiLiU3CgQGpdRBFBD8K\no032vERf2ZETElU+hpgm7k427iA+rnqD3+edHMQ+bTIQNbGJTaxiByJSIGNeOKwAZSggSQrmRGCv\n3MCD95vI4M13v4uzP/BZrCyZ1WlLSqzbFmd7bh43Vk108K0XX8LCEdOhKJVEzxYNm506+qMcWcPi\n6HmCOYt+2djYwIad89/cDOIbaZr442rXGpVC3Thbjo8UIKFtuVunAlpJFIXD8mu/anOS+G3Fq3fM\neaBUWYFQJwn3/WtKiS80FUVZGYjyXQMpK5HC+OrivuN0Ns17AptWO7KepBiORh4wJWSO828bAZ//\n4Wd/FqJrCsKnHvkkUisu0xtsgHMTjfEk8SSqAFBI6VdxRpmnOYv4U03PPYoMtNKVdMJZ/Jk0Ylgu\noKr0B6ga3aFwp8b4G+KIYr9Rw8EqQu7PDoRTAKoEGG7icRxYGE8Vvm9Zgkc8wdZrr8CpgXTn5vDM\nt43m4w99/keRWWGYpNPCtZsGb3X+4gV8/oefBgAc58eRF2XE5achFw0Wq9ZsIF8ybdDRaBRRv9fQ\naTiV5xiVCCRJladASSdAQyHslCFhBl1YOiEiSpBxJ8qqd+wEmPcL+3m6bcoy7FP7in1cR6hU4i1H\n406Ixvj9uNYwEAVgawrDfg+jjS1c3TR1hJXbt/GP//E/Mvs8cxoNqxQ9KAYYEQsearYh7Y1fDvsg\nPIODtzDAX3NOqZ9yHWnhKdgQ1UkopZWxR6UUqKdVD+/HDiIDQ27TtETsH1ZTEYyNEYL7dBD77VJ8\nfOnE3jZJHyY2sYlV7MBECjHZKJibHWdwCly1tI5rN64DAObnFsFcuNxO0JcUh4hZxQWlKO0Y7m/8\nzu96yOwayUE2zUpRbzbwqp3lz5VEs9kCs9/RIse8NGSjS8s3sWrHoAnjfrw5SVJf5OOcWzGWwIxc\n5eo3+y+LIoxnJwY3QOwxU6qRk0Dv5vAEcXHR/Z8zQqjHDZiVIcz5xaPXYY6B+O06Pgf3G2LNTSFE\nhbU5tymCLgVS+1vyYYnbvTX8i1/9dQDAtctvofOYoZoDlehbaLGOgFtCC1AHk05qFa0FgPlOAI0W\nyQZ4KGaS6u+PjSJewaPJwLEVt+5mKpiuTDuOpw6epm1bWrt71HAvRUi/XcY+liLkfu1AOAVCItZb\nLvwJI4p61Nzq6gayzFT8WZYgVxZdpzU0AdYa7oyPMLhu0oTRaAQ5Hx6ces12DFSGnjIP+/XVZUyp\nEvW6eWCOLCxiaGnflm7chMzNfroL8xja151G03cfOOdI07TSZfB0ZWUOFyITSsGZG6/esg7CAZsS\njwnMeFpxBCGVUJ5x2nEsuge+LMuoXhAenrIsfcrjvufeBwLaMu5iGHw99e9r7sbIq0zGlBBPJX95\nMMJx671lSUFs90WiDKhDaYa4djIGUnEGTklJj48ms+oN7gRg9FitYdwZAAAnrFJVIJTs+MBorVAV\ndomO6w4O4qOsN8THdre203j2XnYgnILSVXiu86jxPaAIENWNMLKQW8oYSJpA6NAuzFP7gDQYpM1p\n0yTF0HI/5qWEslf70sp7mC/n8JCVasuFwKqVilteWUEzs7l+IYApi1psNlHYh7hpHUKFDUi5Ilp0\ns0R3gNYMjBlH52xuzkQnMg0w3VqaoVZzU5ZhpXfwZfdwG8fh6gvERzSMsQrOIa4VmEIlCecscgzO\nci1BrdyYGuYoLebh5voK/tE/+1/QtzwOx88GDgtoAWa/U6YcxEHFZVVwjcfna+wZcjewcVBRzUY5\nJ0oqN0fl3FME5q2xh8g9yEYZO9QF7vQQ79dBfJT1Bv/+ByhI3s33JjWFiU1sYhU7EJECCNC3qysD\nhSMSKvPAt1jkEjUblm6Vg8pKESP4ChVES5jWINYdq0yiLOwKzoLaUY3UcG1jGecvmVn/ZrOJR04Z\nwM3tm0u+kr+5tQlVM6erVasjt/uoZxlqPPXaloSQIHmuA+eBSQncABQx7NJuroEGVaWarvmwvtXJ\nfNTAWRK1B1llPFrrQLWWJEmFbt39Timl/z4hBJzzSqThooiyLDGyNQFOqOcqWF3fwhvvXgAA/I//\n+z/F9OnjKBxHgVSVlchFDVmuIDywK0o93OdsmsBp6J4YKviwLc/ZEIGzCKHIQA3MEYYy3p+LKH2g\nhPpzrCMgEgBfkzKfIz7SlEqB78I9MB417Lfe4OyDRA1ayg8FFbkfOxhOITJzGd3EYImNdYMobDeb\n/uITyqBdKJjwygml0JU+v+czGGUgFhfQI0Eth+UNCCpw4er7AICjR4/ghpUw6/W3fNFpceqkLyaW\nWgFOmk1p9PMhuD3mRi3i3Y+wBFVqLIIZy+kIGCLYfDD0n/NIQxXyeK1H/rVSoRC57fyNqTrFw03O\nhBAVqratra2QWnAG5UbZFTDcNLWXK0s38E9/6RcAALUj8+jr0t+UsVoVpaSClEyJO+Yq7kALhRoP\nyE/ieB0r3BLKc05U8AcVpkQg1RwjSxCrdCDxpQj1Ha22pxJBfBdQTp+C0Uo2Q3Z5jgkl+643OPsg\nBcmPgrthN5ukDxOb2MQqdmAihcyOHieEobQh7+3bt9G1mPoSEiRWcnJVcQCaEL+SxBRcMWrPUYgB\nQFvX/Sw9YFqM/dKs1M1aDedsmJxJjbRpOx4Jh4oquE4sViiJGg0U7QoyiJByd4SASR2Ml2+mDRBK\nMGPHwrc2t1BYiqWaUBhYyfo0DUjJrMZROD6HchNZWgfnmf1c6ouQnHMfAcTtxU6n41MhzjmEEJ5+\nnnOO69dNuzdZaKO0BKvTrIH+ljmWF198ETN2jmQVpeVacCjOuAAWMRONzRQ4YRhKGZCw7ei0MZMy\nYowmxMcHEtUQWyuNRLvoSkGxmFpv5xVUq9DN0FL6dKKiCwkzi+J/z3hB9C66FPEMxU6f283kWJ16\nry7FhxExHAinEKPtKCWVk+1vPM7gxIALBJYBBQBag4yFlEDAEADV4aBCq5B3pjkKnSODueFffOVl\nNDLTxptrdtC2weRQlJ6bIU0Sz+PXbrWhBJDb8JVJgdQSexBBQOzNJmiJJrMOhjEIJX06kHEOF9xr\nCjDqWofC1w6EEJB2H4RSS0vv6hiZh2PXarVKP991IlZXV73jev/99/HWW2/h5ZdfBgAcOXIE60Pz\n/bNnz2Jh2gybvXX9Mr7znVcBAF/78/8XomNxFa0aoKq8hJw7nAaN0glZoXNzUGzP9ehIV1R4WIQI\n05hZlkapBKkiPaOW3rZOgP8DcGpjO5m7B2JHQAn1nSnzofAbNblzOuF/p9r5uIB7qzcA250DsLOD\n+DBSikn6MLGJTaxiByJSYJQiczLxeY6NdbNqNesNaLvqMkJ8WpBEYqO5/f+6W2orgiWiUnjzr3fQ\nksy5pX1La+AW3bjR3/QszQu3Fvyq3Wm20bDfG8gCmeaAY+khQG4/lyQAs9W1Og2j2UKUYJR5+XhC\nCdb6pqDXzWqgdnR6NCoioJH27zdqLeR50GpQSiNNIxouu6Kur6/71zdv3vQpwpe//GV0Oh3cvOlm\nF25BuQGljSHy++4DALx7/Qr+8NVnAQC9QxlaPIjXxCAtxhIfKZj9uSJsNUooioACjQV5FCfQhbLn\nLIUjiDKRXlCO8h0naeI0X5Dj1Bd6cy3HdB8COS5UNODEKBAj/xweIipUAghRg4sY7pBOjG9rp5Ri\nry7FfiMGYJcuxYdQiDwQTgEanoyi1+uhsByFzfm5gMKz0Nxxa0gNSShopFbknEesKgTYoR5r7sKn\nWQZKoop5InGbmWGrw6OgzDwajjx348ZWD4cPHQIAFEOAJdIrjGo18mkOJdw/rJJI31XRSkJFd9RM\nZwpDxysyGvkLSClDkgQ6tXrdbCvPBRgLYbhSQVU6z7eQJObhLYoCq6uGWOZLX/oSXnnFUNatr6/j\n8ccf905lc33dpxkXz7+FBg8cifUT5nfqzRGIddyqEJU7UkoFQlzHp9jG3eCuhRPONVJ9gRGbMQaa\nEb+tDNS/jjsWsXIXAGjphqA0NIl4D6L4t0LXb9MyYo+dRh+MgU6VlqYbzhtzDsDeDmK/KYU5zruv\nOew3pQCqDmIvm6QPE5vYxCp2ICIFQgkKG3L3e5s4dNQwNRtpModN2Nl/CQYQoqFsqU7LajThVlNB\nwqqVJmkFDKMjXL/gCbgVJlmrEaS2fz3V6aBdd4VCitwOV+mkQEoTOAH0lFIwxzCU1HwxnmkGUZiq\nfjEagBMNkZvfvKFSvyr0NrZ8KN7u1KGt2C1j8HqRtZoJ32MOh16v53/P669/GwDwy7/8y7h27Zo9\nl0GUt9Pp4Pz586jZlay/vIxux3AgUEJxw4rudO87gumkYfchMXJy6VRCCurJpaWUEcZUBLZ0AAAg\nAElEQVRejw1qBYtTuTjqi0VvAKCwy64hpjLfIWPM2ub/HR4h4BzM3xHe3+4yjl7g07Yovdkhaoij\nh20Rg/nDv3RRw72kFOb/770Qebdw6b3sQDgFIJx0w5Fop+kY9ZX4IiLliM3ltsI+IAmDJ/CgLLAX\n05T7hy02qQyaL/U3Xwl3r4hSIIUJebd6A4hDZruztQU/DKSVQJ1Tj1xknMHdiQORo+ZZmnNo6wSU\nlChkCWJvujRfg7Ctw9Ewx9x8236OwtIYgoCDEgv2URS5GPl2o/nXHPRv/uZv4vd///cBGJIY14ac\nm5vzylt5nkPKAnkeqSbZY25OtbFmmZlJbwNnj5wAANyureDCtatm98oMOyG3grs8gSBOFYuBsSCq\n6506IZWHnxBScQzOwbl2KQCQQgbmxTRqdXquSPugafhukNvvuMXb9RaxNTsHoZWO+Dx2Tin8e/YJ\n13En5iNIKcx+P5yaw37sXgVm/1cAfxWmovQOgL+ttV63/3fXArMEBJsbZqVrz0wFwhUeevOM0YoO\nQ+kKULlpFbnikhLSt95qPEVuc/qEhJ5/kqZRDmcRhFbLIdEI+gqawPUK33j/Hc+8BACN2mnzfcbQ\nHw2Q2u8kdYbU1cMEMBTmwS1FiQQBaZnQBEK6wluKxB5/O8uDpB2l4Lxjz1GQqivKEmUhkY/MwW1u\nreHXfu1XAQDf+ta3vCOYnZ31atgAPC6hLEsMiwEyiyjsjwb+pu/lA3QPG5KZrc11zHWOAQA2GEeS\n2ElWTZCRFJltMW7SAu5WIpp4BzOOU5BjBd64pRkIYhWYFa5VCCgPVkgPHy9thOA+B1SjjXL84cdY\nbWEnU4FE12/zQ4oegDs7iLtBRu7XOQDbJzH3a/upKfwrbBeY/SMAj2itHwVwAcDPAZgIzE5sYv8e\n2D0JzGqtvx79+TyA/9i+vjeBWSkhrFdNKfMe0kihh6q0x6qr0HainAGMgGe2yg3qZxRU5kjEzT6c\nZ447GUorUCW8liWjNGiO8BR53ypXFVtYF1aMZauHbtcoT2UiRSvpopYE3+dYhFVZoixMqzPlzNcX\neJJByBLCDvRIaFimNrQ7c1izKEJJmJ+9KKTE1mbf78Mh/QCAktR3EvI897l7v9+vKFTF4XPKMgi7\njCgCjCwQSkDj6nvvAQAOHTmExcOmvrN4aBZL66aTcXPYA80YhsJ1SYinWKcqDBtxCSiX90cDSGUZ\nRsud+ei2rM4+ODO08sYSCYBTEyFaC6I18q5W0222Q0phftfdpRT2j3B8d6g5fFzj2fu1D6Om8FMA\nvmhf35PArFIKrZYJcynCj9SIlYYV3JOTNWqQDr+QcigaWIiYAoTTYIgGiuIHQiuFRDvNgQJcEh/a\nap6A2SKk1ArZlGnVrecKF65fAQAc6ndx+rDJtaUqsFYboM27AICRKAErdpox7h0UpxTE3tYCHIwD\nNcfVICSattCnBNCzLdm1Xs/zOLLIWSZJguFIePm6r3z1d/D88+a0O9IX/1vtuRwOh2OThgTUUuGf\nPnbKQ6tPHzuOQhjMxNGFBZ9+vHb5AjJLRJOpzOABXGsP1N+k4+l8jZjviOh9B8l2NRFKQ2rIAP+w\nx0TscSqitQYBQCLH6NI8Ej11MSMTWBhO21Zb2M12cBB3cg7AnR2EQ0UC+0dGfpiTmPu1D+QUCCH/\nAOZy/OY9fPenAfw0ALBausenJzaxiX1cds9OgRDyX8AUIH9MB7D9PQnMZlMtvWNxQ0jEIw0k8o6Z\njQYEbHvLpRmQvjgZjy4D8CKmQPD8vAAEV579R2sBpZ3yksS6XfVrKcdAOLBN4mf+aQEQwiFgmYtk\njjQxMwIEEgl3ylEKQlrwTi1BRmuQlkOCcKC0oXhejjyQKx8MfFwtyhLcHmOr0YSUEq++9gIA4Lnn\nQnbWarX8vEez2fSoxTQNNG9JkqDdbld4Fz796ScAAJ959AlImKjh6c/9EHIbjZBagn5u0qc/ee1F\nMMGAzA2blT4SoCokBZoAyl4nKQJYKU2Z0cyM9h/Pa7jUTgxzn/7IuPsUMVQ7c7xO1Y4Gxjoe21mo\n9m2KAVRGxcE7A5+AHaKGuyhCAh/tePad7J6cAiHkrwD4AoAf1loPov+6J4FZFrehohuEU+bJPUcZ\nwXTLVOIVoxhZgg/FGTSU70YkKvG06pTzCrwz9M8JmG9sEySSgNhKfEJSFMJxGCggdXEpUFj47jtX\nL2G2aY7l0VP349bSMrpHFux3KIjtKlBCkLAQBTn4tBASPEvBmNnPsD/0x6lVgdkpM5xVFAUuXTE8\nDxcuX8KTTz4JAFjQBuPQ7ZqUpd/vY2HB7L8sS/8g9fv9ACVWocLf7XY9TTwAfOELXwCzY6Oz8zOe\nRLWeMN9J+MFHHsP5Nw3Z7Ww2BcklVoSpl3SaLfRH5jYoofztLhmvPCTuWKTUSJI6XBs1loRDJM/G\n65mXgEuSxG/L3R9e/aoUFVUs//2oDRpzPuwmn7en7VFvALY7CN9q32XQCvhwJjE/zHTiXgVmfw5A\nBuCP7EV4Xmv9MxOB2YlN7C++3avA7K/c4fN3LTA7LnAaZN4zbFlMfY0xlJ6NuADsyi6UGc91suZS\nh352nuc+ZGSMInWVeBUk1j3pqh2WKSGQWl3JAgJ1S7OGvEBivXZCGG7cMECeQ50ZHJ+fQs9qTjay\nGhp2cIhQCmoBUyv9HjoW6aiUQNHfBLSlSiuG0BabLkuF0h7z4lwXL3zXjC4LKDz/+ncAAE995rO4\n2VvCrcsmimi2ppBwh3YMg1f9fr9C6+5HlosCaZr6SKPRaKCdGdEcEInCgpqGSkFaluYmTfHpBx8G\nAFzr93FraQnlTQtmEhrapj+EkcBNoIhfRUkaNEI5Y6CMoShNOlIhWGVRvhiJtkglK3T1cbpBGa38\nHW/LbTlWzhoHTsX2YUcN+y1COtNQdzWebfaj91WE3K8dDESjDqGlkBLUvs4TAmZFWIUoK6mAq/Cn\nAApBPaKNRtOUWZZVADLuwVdK+huEUneDhW0XGNjD0oAlX0kpAyxqL2UZtkZmknN1cxmUSMxPmQes\nLCVKq1sx6g9ww6YcGU8wSG3eK0tAF2AOGs0Ykqztj39geRMkpfj+H/x+AMCfvfAihla27WvP/zHW\n1teRDW1dpFkDy822BSNgtr2YpmlFAds5CMYYZmdn8eM//uPm927mENZ5lcMc3Y45FqkLTDdMVwRK\noi/NQ/yphx7G448+ht/7xh8AAC7euOpZr3v5AIV9RhhjYLZDwCkLilRSIo1hzrtpNTDi2xlkfHpx\nF7r43UyRan0hXoc+bAdR4WbYZ71Be4dF911vCPvZnz7Ffm0yEDWxiU2sYgciUqCMemwAbzeQ2lVr\nIKvzDuMEoc5KmSNhJjUghEZhZtXneXLSCMhEaWq3vZ0hiEW97VIKEG1W6pEkyG3Id2PlNhbnD4HY\nYx72R0jc7H2DYYqaVTcFMF03mIfh5gaS5hTUyEC7m60ZD1KSSqJDTSi/lG9iqm5W4E8+8hBeu2gK\nfSvXLyHPcyxZ3om0nUEIS0arAGlX51o0Ek4I8QI2DzzwAB5++GGctFiL7twUaok5f9OzcyhtoTSh\ndUjbfSioALXh/IOLC5BS4emnnzbH+ZUve2h1AQrGQpjvLJeiUvQbyRzwlfyI/kxrn36UUhigErZj\n96milcLjOKR6J6sI1t4hathpDmPfpti+i5D+Mx9xEfJu7UA4BRACNK3AKYIz0FpVHEGVRtz82JEo\nkDCOgGORSGyrSysdkG4iiJ3wSKXZtQxdt5KxKPdNDBmIe98hEgUR2JLmAW8WI7zyzts4PGPEXDLC\noG1q06kHNCFPEhR2IIpzw+kos1l7DIC2Pb3BQKDTsnWBhHqntNbbwNljJwEA586dAxIO1nSf46Dz\npmOh8wK1hnl/dTBAzSpEZYMBWi3jbDqdDrrNaczOmQGpRlpD3ZG0UAlmYXBSlCA2FdFKoVUzHZdc\nFACjOJmY1OLIwiLeWDatz3q/esMXjqKdBPCVZtJ3OADzDLhQOFadTmiYciRjUbCOBqBiUNZODNc7\n2Z0cxMdVb4hff5T1hru1SfowsYlNrGIHIlLQJEw9YiwycCtllTE4ONCUcdDo/4jUoFEXNEcIBZ3F\n4qzOk8ZTesr6yqIoK9Vbt5qVpUaqTeh/e0Vj/cYSrt13BgAwPzWDWTv6nPEU1EnJD3JQC/bJ2nVk\njRqoHb8uhcAoN+F3I2sgt9RsajTy+5+bPYTe0EY7jEEQoLBMTHWaQGT23DQTP7sxc/ywX93W+33c\nsqv5Z6c+i3feeRef+ORZACZlGdmJS6UZ6g5+LeEp1FCWoBZX0chqyIWAapho6UdOPYJO20Qqf/72\n9zx8OWZJVlqDutmVxGDSfA9fKpTumhEK5lbNsevtVkCv5eDwLLx6G+83WvDbJoiYpslHWoR0Fqc/\nwF1Ape+CDs7sx21//xHDgXAKQKQERNlYuzB8ptJJsEYJhVISxIFeIkQdAD/eG9Oz0zFx0XH+upgm\nzJnSGqms2V3o6EYcokgElpaNgMyhBODEPCBUSrQtDXt/sIYbPTtfMHUaKeXYFH37G4hjFEGuc5+m\nNGo1vLtihpD6xQgrdiBJJxSqLFG39PObwwGonbGI7jsUDe5rJvV8Bo1DpkPy3Mvn8dN/6z/Cxqod\n6gKwdNuI8j71lz6L3sYaAECIwp+b2dkFP2sBTaFZBsHNzlazKPzOi+BwwfyDzQiDcvMDxHQV/Fmn\nZobFnHztEwsNWkn4QyRNKovHTvRvzj6MdMLZPYOe/MaYv0AGrbh/VKRywrf3gIoEqinFXnYgnILW\nGpyGlTog1nb20poFOTBAQynt1QZ2k+GOZ+91uvN2d7LqSQ0tIHexirJAmqR45oKBGj9y+K9hYCXg\n0nodUzPmwb1y7R3MLxi+w1rWAijFlG1jrqwsQdhtTzWn/Up76/ayVa4GrvVW8eVn/hgAUDJAUoBZ\nhqh4dkSx+GGR4PY4s4UZ0KGpL6ys3QDhDXz5y182n8s3sTh/yO6/iYVjBh1JGcF810xJSiHQSExk\nMBqWkIMBChtRcVB874IpgioaJvM4Y3AertSFLwYqQhFfgXjakRECeJJTYhyD/3HhGuYMHu7smLLM\nbxbbHIOzA+Eg7qHeAHx43A37sUlNYWITm1jFDkSkYBBqIWXYPUKwrS6pfSirtQJTiAaawvdNmhBh\n4R0dmwoRhdhhV/Hq4AVeFfOETJRy30JsNpsQpYBeMZ+7tD5CMmN87dFWYHhq1BtYWzPh+pHDJyr8\nAO3WlOdYHIkSac0OAWmCwcj8zm6j5Vd9SQBaCqRuFDhCAebRgsCj7kdRCDisY+3wAv71v/sK7j9x\nBADw3rdfxFMPnbH/yXDztkmFzpw8E2j1kwxXr5maRFmUmG5NY7hh0plvfedFbBYG2CR5iOEoCEpb\nhyCaVToIOh5UUtTn9ErLCksSj6+FY3kG/Bg6AIiEbYsW/Pc/5Kjh40RFfhTcDfuxA+EUoPU+UgYC\nafuGqhS+cMJBjENwbcQxzgBnMT9jnGLYtNj39sfrC4kfgA/bEjL3E36DQR9t1gWrmzD/+s2ryPsG\nM3D0sz+Ci+fMkOhUU6AzY0J0AYKEEq+CXQgNmjhodw3n3jIkJxoS76wtAwC++CdfgbQQ5pIBNW2G\nj+LfAACIYMZDSH9eCe9gZElimlmCWxvr6OQmfUEjQdkw7cbNa7dQt1T2F3EJZ06eAgDQJAlhPiPQ\nQkFYOPTCVBfv3jK/k6bMT0xKd22sORSi0hoZiTgfqAqDUCxofWlVpRuljobN/u24MgglEBHJzV4O\nQkpZwSPs1z5OVCSwfzo4APfEF7mbTdKHiU1sYhU7GJHCXVrcXtFaQykgBm7tJ0KITUrp1VRUxObT\nYFFIpgVAHNIxaFQOOYcWI2jr3W+vrODcy68BAJrgOGbpzJB0wYTtFmxtIk9SrK2ZKn+7nnmhme9d\nvIbXLxqB2/duvo/vXHsLAFDWOJQdehJaYdAkcGtm0g8rXo0kPmVJAchaM/yGjgVqaY1Wp44rt8xA\n09nTpzEcmhVta3ULF24aJalGrYkpO8ZdzwokTkRXAFeWb+K7tw0T1bkr76K07dGCalC7aqthHq2s\nxOtlxhRxAMAlrVCQ++AsoZVRamdSKjBWXc9k9P29ooadUoqPoggZCxzfVafiDiSye3Yp9lGI3MsO\njFPYq44gSwFRBNizV+G1v9U5AsZYRWnawZd32777jjcR3ZxURB2MuHor4U5ds10HSqPaBABXN5dB\np81N/+pb53Br1YT/Tz7+/bDBOjbKOtjmbVy+fBkA8NBDD+FPnzXqTd998zW8ffVdAMAazUHtw4aM\nQHptBQka8eGWzUAg0xTKV++RlYA0mAdOU6BmcRJpitGqALdKUjUkuG4Fat9/+zIKW994+LFPQ9jz\ndm1jHY2RPdkZxys3LuHrzz8DANhiCgV3iESCmnsQOYfQoWPj0hoohzuwHScCKDsNq8euh3LU+aD+\nN7pK+rhjcLaTg9ivcwA+PAexU0pxL84BuHNKEb/eT81hL5ukDxOb2MQqdjAihV2GWeLi4niUEFey\n6/Va5XthLDoMR8X4hzi9cEUnv+14lSLM/81UGKIihEJbFiUmCxQKqFlS02anif6GKTTmnOK7y4bz\nQL5KMJQGv/D1Z/4UWkjUrLT7352awkuvGd6E82+dQ25HrGWLIXWqTBqgtv2RF9XVhjHmV6oezdGW\nlmtCCK8qlVAGZaOdPO+jPtWEtNv53nAVj9XN7MaAM3DLKnXmU48hVSbqEYNNDLQ53i994+tY6t3E\nwK7iAtrPMqRj4T6PWYCi97WAX5KoBrgdaFOjAiKJug82RVJS7hoZmHNw56hhp5Riry7F3RQiXdRA\nNaLBr/j4Ppyo4V5SCmAsatjDyI7kFB+z1edn9OmffNr/vd+UAdjuEIDw0HPO/YnJi3ybM9jJGGMg\nnPnXvg1KGbi9iCk4CnuRKSEQFJ7YJZat4zIBtyjh/mgI2TdOoYRAQhmylnkQerTnb6TZzQb63NLJ\nNamhpwagICBpyMXL+LwwWuGHUFGomDG7LXBQSw1HCYFUQJFZLsuCYXrDOk+l8dce+0EAwEtvnsfi\nvBna2mwyfO6hxwAAv/EH/xZDXYJakpgCOrQYd1EfkVLCE3FSBkLDBCJTDuhknIL/mAKEE8WSctvE\n315cAbH4jIymYH2budz5Hoi7FPHx363RuPOiqtDseJt3jYwEKtDVuIWrdqmbUULx5tdf/I7W+sk9\nN333RzOxiU3s32c7GOnDR2RSKs/yS6OZip161J74k4eUQTMCEgGhUHc0bUDdLtRSSSia+Co5TwDh\nmJlZDvn/tXc2MZJVVRz/nfs+unsaBHGQjIIwGCFhhcSwQjYmCERB4wbjAiNLYyTqAmXDFo26MZFo\nJKIhGI0S2WgUY3RnhMkMAyIyg/hBBhA0AjN0V733jot3v97rquqqmu6qwtx/0umqW+/VO3Xffeee\n72ONe9sMeNt5barxtg4oB8LrtMa9jJCu/doFFa6FjSnrsNvJAUy0yxRlEdKNs8wHc8VSgjFCxYY9\nxuDCpartIflGRWF7MqyXOafPsR6L/AA/Pm6rQxcFr55pv7c6rTx7ysZPlMJWtLuJZGQuoUm7Ycu+\nCa4x5K4KlBoyQp9QDNQuRV2Mz5doNkKQs6rZWTNhlwpD86gUsDdeChhtiHRkqurZqRRzhktPg5Vj\nCvPaERykV9bdLYy6DhmXbXz87q7KSXjTn55htkNcv4rgPBUGpTJW5VgvOaNt8NBWPWRLQ5enfp9D\nLV3kniG3pd0GVcg4BNC6K4b2mQG0to/4wRhauppC2KjhPKuvZ8AZm1C1JTVc1GZ5VsMz/KeyjXQZ\n+sa9eS6oqTAaVCj3TBVlidhzMhN1xcpz5/VtGwdnBnG1LLQmixheZT0ReRMekNacYjta1TuLuDgG\nMU6lGMUgpvVSOPp3nD+PzWFKe0P/+pO/fLpMzGmxq/ogIveLyMsi8uSIz74oIioiB6OxL4vICRF5\nRkQ+PDNFCQkJS8U0ksL3gW8BP4gHReQS4Abg79FY3GD2XcCjInLFNGXe98q42JcSnPhdVcMORw65\nE20adFbk9pyM2lX7MdnIPIj2s8hoWQTVpKzV70A21MgeX/s6Dbk05KIE83uQYoo871ynLtpqSVlT\ndYxlfXSlnW7Kt0NcG+IN1nBhTXnTsG6pzQ3814ZDFyiVT8hvJQSwc2wEHD3FGthdbXNzk+1tG7Cz\ndSbq6xhVURKhlminFjA2dbrKhdpltA7D8oylBpN1JQb3ne2Huxct3RH4FHmWplUpYq/V2UgMsL9S\ng5khPsFhrgazFt+kbQjz82hsrgazwL6pDO7hz7LMW36bJrgn19bWkDwEPGkm/pHqF3bppGXbo1zJ\nN3f9gRmGuoJakdlF2jQ16hiElqjplpoLDW81PLxGfMBOXLuwv0BU44cjlHCLK/62F3EPU42K8kpm\nbRdFxvnbNjgGwwHnPFTIbL3GGvUp0dv1G22laG29IXmeYzZtPYi68vRna+vBK0JwSTYGX30b2gcl\n1r/F1oCoi0EYHMEgJjIHmMvmsFtk5KICn1r69rZe5LSYt0PUrcALqnqsZ/yZq8HsshHvoP1ir07S\nKMrCZ0aaLPMPm2MIMcPYdtknkrFpF7CiNNal6MvR+8WQdc6vcYk/ga5Y6nG7mvpdIHR/MtFOY0R8\nIJsRDVFvln7f3VkMr284Jlezbvmy1BW5ZZ5npKaxxVRKs0Fer/kFW6NoJMW4Klo5wrAcEYNSNdRx\nRSsDmevbUTdR/ZRo1yyixW0ZRJ85wHwMYtrIyLd6Jua0mJkpiMgB4Cu0qsPciBvM5udsnM1XJSQk\n7CHmkRTeCxwGnJRwMXBERK5lzgaz6wfP13nsCDBadRhlR2gTp0LNBpeUkxV5u7vagCVMN3XX7SLa\nqPdYVNVwh0QxDqed242czEkGtVNjXM3HAZubm/b3GLaHtuqz5OQ2IrGJdp2Q/GJ1/MilGasfjSpq\nd9N4z6rrmo2N9Y5dJN7pt6xnZXNY09jrV2s5A5ujXZ7OGVJT2h3SGMPAJZRVoftWvTZ6eVU0ZATb\nTT7IQ1CPaMjdiKNL4y8oarKhUpmuxACTpQYf0drrvRhjktQwS3q2P2+V0rOnxMxMQVWPA+9070Xk\neeADqvqKiMzVYLbTcXgGO0L734raM9oRnGHRMwRXdk1M50bGD38dqQ8eTa+gbM+9GKNyMQfaUNd1\nRz/1peTz3Jedbw8NDCSEbCuNdsXyWQpzZlnGsApFWhrTjHRDnT6wHtkntttMUYKtYmDLv6+bUBuh\nrYdgGfG465uspdcaFLWKOhwJvtRebHfYwSBGqBMwmUHMYm9oLxk2mMxkU2diOsbQf4hXohzcFJjG\nJfkQraHwShH5p4jcMe5YVX0KcA1mf0lqMJuQ8JbDvA1m488v672fucHsrNgzlQG8lOBE8kk5Eb4j\nUaNBaogMQ6MwLtY9z/OOyjAcwekl8j50OmJN2BTivoLajD8wiyI8C1N2OmQ5NI3S1DZ3IstZ98vF\nqgf2nK3hwGeWG209EACFRFWw4grattKSe1+XUXTkdmQojcT/Bo2az2Y71AmPCVLDfhoh/TWH9b4a\nIf14E9bWXksNKxPROIsdYVlwjKPS4Y7P3APSiTqMGIIv0U27UPI89+pE3z7hIy+HodXasKowMvp2\nxd6T1qbQXncQLbzYvbW2tkZe5L6CdttkN5SjC6X0w+9UorZ72pCJ6fxWf1wmlJnrdB3OL8si2Brq\npktzJLDqWqS69RjERHXCoccgprU3hIPmszc4NW8WL8U89oZA5h73p4iQEqISEhI6WAlJQZjPuAhB\ndZjLuAhedZikNjiEpCPtNEutqqpznPusjvITdsttFy8F1LvmYvRbj7fXCjEMbp5ccVmA7e1tv1OV\nZdn2rvB0NiODZERM6J9ZVd3qPiLeOEujXh2Ld6ayLDxdg8GQ0uZX9HfcToGgntTgpAXREeqEQ09q\nMFa0rwuZ2gjpICLzRUXOUA4OFu+lmEViWAmmEN/feaMW57EjwE5vQ4w+Q+jWLLDiYlX5xe7oGVau\nE1I3s88xKOddaEYGlmT+BhvTeFtDrDq0rkYNobmRxbvRypeGK8syYqTij6/rGjNG1+3X1xjHoDpd\ntrJwC+PF3jLLcM+cOlHkeccdOrY5KmaiOhGO6zKIRdkbwiXnYxDLaFozDZL6kJCQ0MFqSAp7gHlU\nBhjNladVGQAvJTgxOW4rLpGIvqNB7oREFXfOYNCgdqcyEHkVtLu7ZhnO3NbUddQ8t8FYvl+WJYXr\nN2nnq7ISTetlsIVTI0NjnHabZcZHQIVKT7WfL29E7ORz7JTssL8jlgj6UoMz2hmR6Y2QTh2odWoj\npEPeVP93odLQjW2ZRWJYGaYwjx0BggtyFcrKTUKWZby5Zd17kgNN5zeNZBhRY5RGAlOIRUWAyvab\ndJ/FGZcOeV744iWOQY1jiNWYheiTxrQhL/JO/scot2xVN+RZqGERI37Y+wxikjrh0GcQ09obPGxU\nJEBlpvdSzBsV6TIx57E3wN5mYu6GpD4kJCR0sBKSgjGjedMsKdHLNi7Glnkn3RgjUfHQYZT6vLOZ\niUOjSmGDdDJCbIFqEAWruvYxBn2YLPMSQl03XmWYFlVdd36zQ6z+NI3STAiM8qqcNr403c5jIskI\n4+MW+qrRRCOkCyaLvBQwvZdiUaHSMWZtWjNvqDSMN0TuhpWo5iwi/wJOA68sm5YIB1ktemD1aEr0\nTMaq0XOpql6420ErwRQAROSxacpPLwqrRg+sHk2JnslYNXqmRbIpJCQkdJCYQkJCQgerxBS+s2wC\nelg1emD1aEr0TMaq0TMVVsamkJCQsBpYJUkhISFhBbB0piAiN9rGMSdE5K4l0XCJiPxWRP4kIk+J\nyOft+D0i8oKIHLV/Ny+QpudF5Li97mN27AIR+bWIPGv/v31BtFwZzcFREXlNRAhZK1wAAAMOSURB\nVO5c9PyMakw0aU72uzHRGHq+JiJ/FpEnRORhETnfjl8mIm9Gc3XfXtOzZ1DVpf3RJuaeBC4HSuAY\ncNUS6DgEXGNfnwv8BbgKuAf40pLm5nngYG/sq8Bd9vVdwL1LumcvApcuen6A64FrgCd3mxN7/44B\na7SFhk8C2QLouQHI7et7I3oui49b5b9lSwrXAidU9TlVHQA/om0os1Co6ilVPWJfvw48zWr2q7gV\neMC+fgD42BJo+BBwUlX/tugLq+rvgX/3hsfNiW9MpKp/BVxjon2lR1V/pa7CbdsD5eK9vOYisGym\n8G7gH9H7pTePsd2w3g/8wQ59zoqC9y9KXLdQ2rZ7j9seGQAXqeop+/pF4KIF0uNwG/BQ9H5Z8+Mw\nbk5WYW19BvhF9P6wVR1+JyIfXDAtU2PZTGGlICLnAD8F7lTV14Bv06o2VwOngK8vkJzrVPVq4Cbg\nsyJyffyhtjLpQl1HIlICtwA/sUPLnJ8dWMacjIOI3E3bSPRBO3QKeI+9p1+gbYXwtmXRNwnLZgpT\nN4/Zb4hIQcsQHlTVnwGo6kuqWmubefNd9lj8nARVfcH+fxl42F77JRE5ZOk9BLy8KHosbgKOqOpL\nlralzU+EcXOytLUlIp8GPgJ8yjIqrBrzqn39OK2N44pF0DMrls0U/gi8T0QO213oNuCRRRMhberb\n94CnVfUb0fih6LCPA0/2z90nejZF5Fz3mtZ49STt3NxuD7udbnPfReCTRKrDsuanh3Fz8ghwm4is\nichhpmxMdLYQkRtpGy/foqpnovELRSSzry+39Dy33/TMhWVbOoGbaa39J4G7l0TDdbRi5xPAUft3\nM/BD4LgdfwQ4tCB6Lqe1nB8DnnLzArwD+A3wLPAocMEC52gTeBU4Lxpb6PzQMqRTwJDWRnDHpDkB\n7rbr6hngpgXRc4LWluHW0X322E/Ye3kUOAJ8dNHrfNq/FNGYkJDQwbLVh4SEhBVDYgoJCQkdJKaQ\nkJDQQWIKCQkJHSSmkJCQ0EFiCgkJCR0kppCQkNBBYgoJCQkd/A91xNwU/yLXjQAAAABJRU5ErkJg\ngg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f02df84c1d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQUAAAD8CAYAAAB+fLH0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nKIhIySfIomq0//KH6QAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f02df90d5c0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQUAAAD8CAYAAAB+fLH0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nA08A0WnXAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f02df7215c0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQUAAAD8CAYAAAB+fLH0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f02df705198>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Use data augmentation\n",
    "\n",
    "datagen = ImageDataGenerator(\n",
    "      rotation_range=40,\n",
    "      width_shift_range=0.2,\n",
    "      height_shift_range=0.2,\n",
    "      shear_range=0.2,\n",
    "      zoom_range=0.2,\n",
    "      horizontal_flip=True,\n",
    "      fill_mode='nearest')\n",
    "\n",
    "from keras.preprocessing import image                           \n",
    "\n",
    "fnames = [os.path.join(train_cats_dir, fname) for\n",
    "     fname in os.listdir(train_cats_dir)]\n",
    "\n",
    "img_path = fnames[3]                                            \n",
    "\n",
    "img = image.load_img(img_path, target_size=(150, 150))          \n",
    "\n",
    "x = image.img_to_array(img)                                     \n",
    "x = x.reshape((1,) + x.shape)                                   \n",
    "\n",
    "i = 0                                                           \n",
    "for batch in datagen.flow(x, batch_size=1):                     \n",
    "    plt.figure(i)                                               \n",
    "    imgplot = plt.imshow(image.array_to_img(batch[0]))          \n",
    "    i += 1                                                      \n",
    "    if i % 4 == 0:                                              \n",
    "        break                                                   \n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "** Let us train a new network with data augmentation and dropout **"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model = models.Sequential()\n",
    "model.add(layers.Conv2D(32, (3, 3), activation='relu',\n",
    "                        input_shape=(150, 150, 3)))\n",
    "model.add(layers.MaxPooling2D((2, 2)))\n",
    "model.add(layers.Conv2D(64, (3, 3), activation='relu'))\n",
    "model.add(layers.MaxPooling2D((2, 2)))\n",
    "model.add(layers.Conv2D(128, (3, 3), activation='relu'))\n",
    "model.add(layers.MaxPooling2D((2, 2)))\n",
    "model.add(layers.Conv2D(128, (3, 3), activation='relu'))\n",
    "model.add(layers.MaxPooling2D((2, 2)))\n",
    "model.add(layers.Flatten())\n",
    "model.add(layers.Dropout(0.5))\n",
    "model.add(layers.Dense(512, activation='relu'))\n",
    "model.add(layers.Dense(1, activation='sigmoid'))\n",
    "\n",
    "model.compile(loss='binary_crossentropy',\n",
    "              optimizer=optimizers.RMSprop(lr=1e-4),\n",
    "              metrics=['acc'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 2000 images belonging to 2 classes.\n",
      "Found 1000 images belonging to 2 classes.\n",
      "Epoch 1/100\n",
      "100/100 [==============================] - 81s - loss: 0.6881 - acc: 0.5406 - val_loss: 0.6796 - val_acc: 0.5920\n",
      "Epoch 2/100\n",
      "100/100 [==============================] - 77s - loss: 0.6771 - acc: 0.5700 - val_loss: 0.6557 - val_acc: 0.6166\n",
      "Epoch 3/100\n",
      "100/100 [==============================] - 77s - loss: 0.6598 - acc: 0.6228 - val_loss: 0.7063 - val_acc: 0.5238\n",
      "Epoch 4/100\n",
      "100/100 [==============================] - 74s - loss: 0.6453 - acc: 0.6106 - val_loss: 0.6175 - val_acc: 0.6720\n",
      "Epoch 5/100\n",
      "100/100 [==============================] - 73s - loss: 0.6335 - acc: 0.6422 - val_loss: 0.5944 - val_acc: 0.6875\n",
      "Epoch 6/100\n",
      "100/100 [==============================] - 73s - loss: 0.6116 - acc: 0.6578 - val_loss: 0.5887 - val_acc: 0.6910\n",
      "Epoch 7/100\n",
      "100/100 [==============================] - 70s - loss: 0.5949 - acc: 0.6781 - val_loss: 0.5894 - val_acc: 0.6669\n",
      "Epoch 8/100\n",
      "100/100 [==============================] - 71s - loss: 0.5970 - acc: 0.6806 - val_loss: 0.5877 - val_acc: 0.6745\n",
      "Epoch 9/100\n",
      "100/100 [==============================] - 70s - loss: 0.5868 - acc: 0.6825 - val_loss: 0.5677 - val_acc: 0.6872\n",
      "Epoch 10/100\n",
      "100/100 [==============================] - 71s - loss: 0.5758 - acc: 0.6944 - val_loss: 0.5476 - val_acc: 0.7126\n",
      "Epoch 11/100\n",
      "100/100 [==============================] - 71s - loss: 0.5821 - acc: 0.6944 - val_loss: 0.5451 - val_acc: 0.7191\n",
      "Epoch 12/100\n",
      "100/100 [==============================] - 72s - loss: 0.5596 - acc: 0.7138 - val_loss: 0.5722 - val_acc: 0.6823\n",
      "Epoch 13/100\n",
      "100/100 [==============================] - 72s - loss: 0.5597 - acc: 0.7119 - val_loss: 0.5660 - val_acc: 0.6933\n",
      "Epoch 14/100\n",
      "100/100 [==============================] - 72s - loss: 0.5621 - acc: 0.7116 - val_loss: 0.5298 - val_acc: 0.7214\n",
      "Epoch 15/100\n",
      "100/100 [==============================] - 72s - loss: 0.5608 - acc: 0.7094 - val_loss: 0.5325 - val_acc: 0.7329\n",
      "Epoch 16/100\n",
      "100/100 [==============================] - 73s - loss: 0.5453 - acc: 0.7219 - val_loss: 0.5108 - val_acc: 0.7430\n",
      "Epoch 17/100\n",
      "100/100 [==============================] - 73s - loss: 0.5478 - acc: 0.7219 - val_loss: 0.5211 - val_acc: 0.7265\n",
      "Epoch 18/100\n",
      "100/100 [==============================] - 72s - loss: 0.5365 - acc: 0.7316 - val_loss: 0.5016 - val_acc: 0.7610\n",
      "Epoch 19/100\n",
      "100/100 [==============================] - 71s - loss: 0.5397 - acc: 0.7206 - val_loss: 0.5031 - val_acc: 0.7436\n",
      "Epoch 20/100\n",
      "100/100 [==============================] - 72s - loss: 0.5315 - acc: 0.7303 - val_loss: 0.5399 - val_acc: 0.7223\n",
      "Epoch 21/100\n",
      "100/100 [==============================] - 71s - loss: 0.5257 - acc: 0.7312 - val_loss: 0.4969 - val_acc: 0.7461\n",
      "Epoch 22/100\n",
      "100/100 [==============================] - 71s - loss: 0.5350 - acc: 0.7356 - val_loss: 0.5028 - val_acc: 0.7544\n",
      "Epoch 23/100\n",
      "100/100 [==============================] - 72s - loss: 0.5160 - acc: 0.7387 - val_loss: 0.4817 - val_acc: 0.7633\n",
      "Epoch 24/100\n",
      "100/100 [==============================] - 71s - loss: 0.5180 - acc: 0.7344 - val_loss: 0.4739 - val_acc: 0.7640\n",
      "Epoch 25/100\n",
      "100/100 [==============================] - 71s - loss: 0.5080 - acc: 0.7456 - val_loss: 0.4666 - val_acc: 0.7741\n",
      "Epoch 26/100\n",
      "100/100 [==============================] - 71s - loss: 0.5076 - acc: 0.7600 - val_loss: 0.4784 - val_acc: 0.7680\n",
      "Epoch 27/100\n",
      "100/100 [==============================] - 71s - loss: 0.5067 - acc: 0.7409 - val_loss: 0.4644 - val_acc: 0.7687\n",
      "Epoch 28/100\n",
      "100/100 [==============================] - 70s - loss: 0.4999 - acc: 0.7509 - val_loss: 0.5012 - val_acc: 0.7552\n",
      "Epoch 29/100\n",
      "100/100 [==============================] - 71s - loss: 0.4899 - acc: 0.7566 - val_loss: 0.4739 - val_acc: 0.7771\n",
      "Epoch 30/100\n",
      "100/100 [==============================] - 70s - loss: 0.4846 - acc: 0.7694 - val_loss: 0.5274 - val_acc: 0.7341\n",
      "Epoch 31/100\n",
      "100/100 [==============================] - 70s - loss: 0.4855 - acc: 0.7656 - val_loss: 0.4927 - val_acc: 0.7570\n",
      "Epoch 32/100\n",
      "100/100 [==============================] - 70s - loss: 0.4822 - acc: 0.7684 - val_loss: 0.4751 - val_acc: 0.7722\n",
      "Epoch 33/100\n",
      "100/100 [==============================] - 70s - loss: 0.4803 - acc: 0.7744 - val_loss: 0.4747 - val_acc: 0.7874\n",
      "Epoch 34/100\n",
      "100/100 [==============================] - 69s - loss: 0.4713 - acc: 0.7850 - val_loss: 0.4576 - val_acc: 0.7841\n",
      "Epoch 35/100\n",
      "100/100 [==============================] - 69s - loss: 0.4697 - acc: 0.7784 - val_loss: 0.5418 - val_acc: 0.7384\n",
      "Epoch 36/100\n",
      "100/100 [==============================] - 69s - loss: 0.4749 - acc: 0.7709 - val_loss: 0.4507 - val_acc: 0.8015\n",
      "Epoch 37/100\n",
      "100/100 [==============================] - 71s - loss: 0.4634 - acc: 0.7797 - val_loss: 0.5817 - val_acc: 0.7055\n",
      "Epoch 38/100\n",
      "100/100 [==============================] - 69s - loss: 0.4636 - acc: 0.7722 - val_loss: 0.4292 - val_acc: 0.7995\n",
      "Epoch 39/100\n",
      "100/100 [==============================] - 71s - loss: 0.4662 - acc: 0.7747 - val_loss: 0.4393 - val_acc: 0.7931\n",
      "Epoch 40/100\n",
      "100/100 [==============================] - 69s - loss: 0.4552 - acc: 0.7750 - val_loss: 0.4548 - val_acc: 0.7817\n",
      "Epoch 41/100\n",
      "100/100 [==============================] - 70s - loss: 0.4488 - acc: 0.7856 - val_loss: 0.4968 - val_acc: 0.7798\n",
      "Epoch 42/100\n",
      "100/100 [==============================] - 70s - loss: 0.4640 - acc: 0.7759 - val_loss: 0.4729 - val_acc: 0.7719\n",
      "Epoch 43/100\n",
      "100/100 [==============================] - 70s - loss: 0.4516 - acc: 0.7872 - val_loss: 0.4761 - val_acc: 0.7945\n",
      "Epoch 44/100\n",
      "100/100 [==============================] - 70s - loss: 0.4557 - acc: 0.7778 - val_loss: 0.4228 - val_acc: 0.8138\n",
      "Epoch 45/100\n",
      "100/100 [==============================] - 70s - loss: 0.4484 - acc: 0.7859 - val_loss: 0.4704 - val_acc: 0.7899\n",
      "Epoch 46/100\n",
      "100/100 [==============================] - 70s - loss: 0.4279 - acc: 0.8056 - val_loss: 0.4453 - val_acc: 0.7893\n",
      "Epoch 47/100\n",
      "100/100 [==============================] - 70s - loss: 0.4404 - acc: 0.7931 - val_loss: 0.4657 - val_acc: 0.7697\n",
      "Epoch 48/100\n",
      "100/100 [==============================] - 69s - loss: 0.4193 - acc: 0.8059 - val_loss: 0.4396 - val_acc: 0.7938\n",
      "Epoch 49/100\n",
      "100/100 [==============================] - 70s - loss: 0.4398 - acc: 0.7925 - val_loss: 0.4315 - val_acc: 0.8039\n",
      "Epoch 50/100\n",
      "100/100 [==============================] - 69s - loss: 0.4374 - acc: 0.7966 - val_loss: 0.4223 - val_acc: 0.8080\n",
      "Epoch 51/100\n",
      "100/100 [==============================] - 68s - loss: 0.4290 - acc: 0.8022 - val_loss: 0.4469 - val_acc: 0.7854\n",
      "Epoch 52/100\n",
      "100/100 [==============================] - 70s - loss: 0.4337 - acc: 0.8013 - val_loss: 0.5178 - val_acc: 0.7597\n",
      "Epoch 53/100\n",
      "100/100 [==============================] - 69s - loss: 0.4213 - acc: 0.8044 - val_loss: 0.4096 - val_acc: 0.8093\n",
      "Epoch 54/100\n",
      "100/100 [==============================] - 70s - loss: 0.4327 - acc: 0.8003 - val_loss: 0.4836 - val_acc: 0.7773\n",
      "Epoch 55/100\n",
      "100/100 [==============================] - 69s - loss: 0.4192 - acc: 0.8069 - val_loss: 0.4165 - val_acc: 0.8109\n",
      "Epoch 56/100\n",
      "100/100 [==============================] - 69s - loss: 0.4243 - acc: 0.8031 - val_loss: 0.4996 - val_acc: 0.7582\n",
      "Epoch 57/100\n",
      "100/100 [==============================] - 69s - loss: 0.4143 - acc: 0.8053 - val_loss: 0.4039 - val_acc: 0.8249\n",
      "Epoch 58/100\n",
      "100/100 [==============================] - 70s - loss: 0.4102 - acc: 0.8141 - val_loss: 0.4206 - val_acc: 0.8061\n",
      "Epoch 59/100\n",
      "100/100 [==============================] - 70s - loss: 0.4103 - acc: 0.8103 - val_loss: 0.4269 - val_acc: 0.8093\n",
      "Epoch 60/100\n",
      "100/100 [==============================] - 68s - loss: 0.4092 - acc: 0.8125 - val_loss: 0.4321 - val_acc: 0.7932\n",
      "Epoch 61/100\n",
      "100/100 [==============================] - 69s - loss: 0.4127 - acc: 0.8050 - val_loss: 0.4090 - val_acc: 0.8093\n",
      "Epoch 62/100\n",
      "100/100 [==============================] - 68s - loss: 0.4049 - acc: 0.8163 - val_loss: 0.4052 - val_acc: 0.8090\n",
      "Epoch 63/100\n",
      "100/100 [==============================] - 69s - loss: 0.4071 - acc: 0.8094 - val_loss: 0.3928 - val_acc: 0.8198\n",
      "Epoch 64/100\n",
      "100/100 [==============================] - 70s - loss: 0.4090 - acc: 0.8109 - val_loss: 0.4338 - val_acc: 0.7976\n",
      "Epoch 65/100\n",
      "100/100 [==============================] - 69s - loss: 0.3966 - acc: 0.8178 - val_loss: 0.3948 - val_acc: 0.8274\n",
      "Epoch 66/100\n",
      "100/100 [==============================] - 70s - loss: 0.3991 - acc: 0.8178 - val_loss: 0.4456 - val_acc: 0.7957\n",
      "Epoch 67/100\n",
      "100/100 [==============================] - 70s - loss: 0.3945 - acc: 0.8219 - val_loss: 0.4097 - val_acc: 0.8151\n",
      "Epoch 68/100\n",
      "100/100 [==============================] - 69s - loss: 0.4073 - acc: 0.8069 - val_loss: 0.4103 - val_acc: 0.8209\n",
      "Epoch 69/100\n",
      "100/100 [==============================] - 69s - loss: 0.4041 - acc: 0.8135 - val_loss: 0.3886 - val_acc: 0.8267\n",
      "Epoch 70/100\n",
      "100/100 [==============================] - 69s - loss: 0.3915 - acc: 0.8266 - val_loss: 0.4639 - val_acc: 0.7925\n",
      "Epoch 71/100\n",
      "100/100 [==============================] - 68s - loss: 0.3808 - acc: 0.8241 - val_loss: 0.4070 - val_acc: 0.8128\n",
      "Epoch 72/100\n",
      "100/100 [==============================] - 69s - loss: 0.3794 - acc: 0.8363 - val_loss: 0.4053 - val_acc: 0.8166\n",
      "Epoch 73/100\n",
      "100/100 [==============================] - 70s - loss: 0.3916 - acc: 0.8288 - val_loss: 0.4388 - val_acc: 0.8071\n",
      "Epoch 74/100\n",
      "100/100 [==============================] - 69s - loss: 0.3760 - acc: 0.8291 - val_loss: 0.4162 - val_acc: 0.8189\n",
      "Epoch 75/100\n",
      "100/100 [==============================] - 69s - loss: 0.3946 - acc: 0.8222 - val_loss: 0.3787 - val_acc: 0.8222\n",
      "Epoch 76/100\n",
      "100/100 [==============================] - 71s - loss: 0.3784 - acc: 0.8262 - val_loss: 0.4084 - val_acc: 0.8189\n",
      "Epoch 77/100\n",
      "100/100 [==============================] - 69s - loss: 0.3831 - acc: 0.8297 - val_loss: 0.4074 - val_acc: 0.8267\n",
      "Epoch 78/100\n",
      "100/100 [==============================] - 70s - loss: 0.3671 - acc: 0.8356 - val_loss: 0.3981 - val_acc: 0.8223\n",
      "Epoch 79/100\n",
      "100/100 [==============================] - 70s - loss: 0.3741 - acc: 0.8338 - val_loss: 0.4118 - val_acc: 0.8135\n",
      "Epoch 80/100\n",
      "100/100 [==============================] - 69s - loss: 0.3671 - acc: 0.8356 - val_loss: 0.3989 - val_acc: 0.8230\n",
      "Epoch 81/100\n",
      "100/100 [==============================] - 69s - loss: 0.3733 - acc: 0.8338 - val_loss: 0.4026 - val_acc: 0.8299\n",
      "Epoch 82/100\n",
      "100/100 [==============================] - 70s - loss: 0.3665 - acc: 0.8400 - val_loss: 0.4222 - val_acc: 0.8157\n",
      "Epoch 83/100\n",
      "100/100 [==============================] - 70s - loss: 0.3820 - acc: 0.8209 - val_loss: 0.3911 - val_acc: 0.8222\n",
      "Epoch 84/100\n",
      "100/100 [==============================] - 69s - loss: 0.3606 - acc: 0.8338 - val_loss: 0.4338 - val_acc: 0.7970\n",
      "Epoch 85/100\n",
      "100/100 [==============================] - 69s - loss: 0.3711 - acc: 0.8359 - val_loss: 0.4009 - val_acc: 0.8273\n",
      "Epoch 86/100\n",
      "100/100 [==============================] - 70s - loss: 0.3569 - acc: 0.8397 - val_loss: 0.3903 - val_acc: 0.8306\n",
      "Epoch 87/100\n",
      "100/100 [==============================] - 70s - loss: 0.3527 - acc: 0.8509 - val_loss: 0.3967 - val_acc: 0.8198\n",
      "Epoch 88/100\n",
      "100/100 [==============================] - 70s - loss: 0.3532 - acc: 0.8469 - val_loss: 0.3891 - val_acc: 0.8299\n",
      "Epoch 89/100\n",
      " 99/100 [============================>.] - ETA: 0s - loss: 0.3479 - acc: 0.8491"
     ]
    }
   ],
   "source": [
    "train_datagen = ImageDataGenerator(\n",
    "    rescale=1./255,\n",
    "    rotation_range=40,\n",
    "    width_shift_range=0.2,\n",
    "    height_shift_range=0.2,\n",
    "    shear_range=0.2,\n",
    "    zoom_range=0.2,\n",
    "    horizontal_flip=True,)\n",
    "\n",
    "test_datagen = ImageDataGenerator(rescale=1./255)               \n",
    "\n",
    "train_generator = train_datagen.flow_from_directory(\n",
    "        train_dir,                                              \n",
    "        target_size=(150, 150),                                 \n",
    "        batch_size=32,\n",
    "        class_mode='binary')                                    \n",
    "\n",
    "validation_generator = test_datagen.flow_from_directory(\n",
    "        validation_dir,\n",
    "        target_size=(150, 150),\n",
    "        batch_size=32,\n",
    "        class_mode='binary')\n",
    "\n",
    "history = model.fit_generator(\n",
    "      train_generator,\n",
    "      steps_per_epoch=100,\n",
    "      epochs=100,\n",
    "      validation_data=validation_generator,\n",
    "      validation_steps=50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model.save('cats_and_dogs_small_2.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEICAYAAABYoZ8gAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n4zhOjvH/AeCWIkGowRHSAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9adef9a278>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEICAYAAABYoZ8gAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\njlNguPA7juMUGC78juM4Bcb/B/AQiZE+NdG+AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9adf8ec048>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the graphs - These plots are much better\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "acc = history.history['acc']\n",
    "val_acc = history.history['val_acc']\n",
    "loss = history.history['loss']\n",
    "val_loss = history.history['val_loss']\n",
    "\n",
    "epochs = range(1, len(acc) + 1)\n",
    "\n",
    "plt.plot(epochs, acc, 'bo', label='Training acc')\n",
    "plt.plot(epochs, val_acc, 'b', label='Validation acc')\n",
    "plt.title('Training and validation accuracy')\n",
    "plt.legend()\n",
    "\n",
    "plt.figure()\n",
    "\n",
    "plt.plot(epochs, loss, 'bo', label='Training loss')\n",
    "plt.plot(epochs, val_loss, 'b', label='Validation loss')\n",
    "plt.title('Training and validation loss')\n",
    "plt.legend()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "** Reloading a model and verifying accuracy **"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_19 (Conv2D)           (None, 148, 148, 32)      896       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_17 (MaxPooling (None, 74, 74, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_20 (Conv2D)           (None, 72, 72, 64)        18496     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_18 (MaxPooling (None, 36, 36, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_21 (Conv2D)           (None, 34, 34, 128)       73856     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_19 (MaxPooling (None, 17, 17, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv2d_22 (Conv2D)           (None, 15, 15, 128)       147584    \n",
      "_________________________________________________________________\n",
      "max_pooling2d_20 (MaxPooling (None, 7, 7, 128)         0         \n",
      "_________________________________________________________________\n",
      "flatten_5 (Flatten)          (None, 6272)              0         \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 6272)              0         \n",
      "_________________________________________________________________\n",
      "dense_9 (Dense)              (None, 512)               3211776   \n",
      "_________________________________________________________________\n",
      "dense_10 (Dense)             (None, 1)                 513       \n",
      "=================================================================\n",
      "Total params: 3,453,121\n",
      "Trainable params: 3,453,121\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "from keras.models import load_model\n",
    "m = load_model('/home/srijith/cats_and_dogs_small_2.h5')\n",
    "m.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 1000 images belonging to 2 classes.\n",
      "1/4 [======>.......................] - ETA: 0s[[ 0.99640971]\n",
      " [ 0.12612586]\n",
      " [ 0.9985494 ]\n",
      " [ 0.07121485]]\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "test_datagen = ImageDataGenerator(rescale=1./255)               \n",
    "\n",
    "test_generator = test_datagen.flow_from_directory(\n",
    "        test_dir,                                              \n",
    "        target_size=(150, 150),                                 \n",
    "        batch_size=1,\n",
    "        class_mode='binary')\n",
    "\n",
    "predicted = m.predict_generator(test_generator,steps=4,verbose=True)\n",
    "print(predicted)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'cats': 0, 'dogs': 1}"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_generator.class_indices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 1000 images belonging to 2 classes.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.47597323375518941, 0.80600000000000005]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_datagen = ImageDataGenerator(rescale=1./255)               \n",
    "\n",
    "test_generator = test_datagen.flow_from_directory(\n",
    "        test_dir,                                              \n",
    "        target_size=(150, 150),                                 \n",
    "        batch_size=1,\n",
    "        class_mode='binary')\n",
    "\n",
    "\n",
    "m.evaluate_generator(test_generator,1000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['loss', 'acc']"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m.metrics_names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'2.0.8'"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "keras.__version__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}