Course 2 - Part 4 - Lesson 2 - Notebook (Cats v Dogs Augmentation).ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "view-in-github"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/lmoroney/dlaicourse/blob/master/Course%202%20-%20Part%204%20-%20Lesson%202%20-%20Notebook%20(Cats%20v%20Dogs%20Augmentation).ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "rX8mhOLljYeM"
      },
      "source": [
        "##### Copyright 2019 The TensorFlow Authors."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 0,
      "metadata": {
        "cellView": "form",
        "colab": {},
        "colab_type": "code",
        "id": "BZSlp3DAjdYf"
      },
      "outputs": [],
      "source": [
        "#@title Licensed under the Apache License, Version 2.0 (the \"License\");\n",
        "# you may not use this file except in compliance with the License.\n",
        "# You may obtain a copy of the License at\n",
        "#\n",
        "# https://www.apache.org/licenses/LICENSE-2.0\n",
        "#\n",
        "# Unless required by applicable law or agreed to in writing, software\n",
        "# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
        "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
        "# See the License for the specific language governing permissions and\n",
        "# limitations under the License."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "gGxCD4mGHHjG"
      },
      "source": [
        "Let's start with a model that's very effective at learning Cats v Dogs.\n",
        "\n",
        "It's similar to the previous models that you have used, but I have updated the layers definition. Note that there are now 4 convolutional layers with 32, 64, 128 and 128 convolutions respectively.\n",
        "\n",
        "Also, this will train for 100 epochs, because I want to plot the graph of loss and accuracy."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 0,
      "metadata": {
        "colab": {},
        "colab_type": "code",
        "id": "MJPyDEzOqrKB"
      },
      "outputs": [],
      "source": [
        "!wget --no-check-certificate \\\n",
        "    https://storage.googleapis.com/mledu-datasets/cats_and_dogs_filtered.zip \\\n",
        "    -O /tmp/cats_and_dogs_filtered.zip\n",
        "  \n",
        "import os\n",
        "import zipfile\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras.optimizers import RMSprop\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "\n",
        "local_zip = '/tmp/cats_and_dogs_filtered.zip'\n",
        "zip_ref = zipfile.ZipFile(local_zip, 'r')\n",
        "zip_ref.extractall('/tmp')\n",
        "zip_ref.close()\n",
        "\n",
        "base_dir = '/tmp/cats_and_dogs_filtered'\n",
        "train_dir = os.path.join(base_dir, 'train')\n",
        "validation_dir = os.path.join(base_dir, 'validation')\n",
        "\n",
        "# Directory with our training cat pictures\n",
        "train_cats_dir = os.path.join(train_dir, 'cats')\n",
        "\n",
        "# Directory with our training dog pictures\n",
        "train_dogs_dir = os.path.join(train_dir, 'dogs')\n",
        "\n",
        "# Directory with our validation cat pictures\n",
        "validation_cats_dir = os.path.join(validation_dir, 'cats')\n",
        "\n",
        "# Directory with our validation dog pictures\n",
        "validation_dogs_dir = os.path.join(validation_dir, 'dogs')\n",
        "\n",
        "model = tf.keras.models.Sequential([\n",
        "    tf.keras.layers.Conv2D(32, (3,3), activation='relu', input_shape=(150, 150, 3)),\n",
        "    tf.keras.layers.MaxPooling2D(2, 2),\n",
        "    tf.keras.layers.Conv2D(64, (3,3), activation='relu'),\n",
        "    tf.keras.layers.MaxPooling2D(2,2),\n",
        "    tf.keras.layers.Conv2D(128, (3,3), activation='relu'),\n",
        "    tf.keras.layers.MaxPooling2D(2,2),\n",
        "    tf.keras.layers.Conv2D(128, (3,3), activation='relu'),\n",
        "    tf.keras.layers.MaxPooling2D(2,2),\n",
        "    tf.keras.layers.Flatten(),\n",
        "    tf.keras.layers.Dense(512, activation='relu'),\n",
        "    tf.keras.layers.Dense(1, activation='sigmoid')\n",
        "])\n",
        "\n",
        "model.compile(loss='binary_crossentropy',\n",
        "              optimizer=RMSprop(lr=1e-4),\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "# All images will be rescaled by 1./255\n",
        "train_datagen = ImageDataGenerator(rescale=1./255)\n",
        "test_datagen = ImageDataGenerator(rescale=1./255)\n",
        "\n",
        "# Flow training images in batches of 20 using train_datagen generator\n",
        "train_generator = train_datagen.flow_from_directory(\n",
        "        train_dir,  # This is the source directory for training images\n",
        "        target_size=(150, 150),  # All images will be resized to 150x150\n",
        "        batch_size=20,\n",
        "        # Since we use binary_crossentropy loss, we need binary labels\n",
        "        class_mode='binary')\n",
        "\n",
        "# Flow validation images in batches of 20 using test_datagen generator\n",
        "validation_generator = test_datagen.flow_from_directory(\n",
        "        validation_dir,\n",
        "        target_size=(150, 150),\n",
        "        batch_size=20,\n",
        "        class_mode='binary')\n",
        "\n",
        "history = model.fit(\n",
        "      train_generator,\n",
        "      steps_per_epoch=100,  # 2000 images = batch_size * steps\n",
        "      epochs=100,\n",
        "      validation_data=validation_generator,\n",
        "      validation_steps=50,  # 1000 images = batch_size * steps\n",
        "      verbose=2)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 0,
      "metadata": {
        "colab": {
          "height": 561
        },
        "colab_type": "code",
        "id": "GZWPcmKWO303",
        "outputId": "9dc14530-e7af-4e53-d7c1-71051a60b154"
      },
      "outputs": [
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYMAAAEQCAYAAABSlhj/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f3da2643550>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYMAAAEQCAYAAABSlhj/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nfOi+cmtZqQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f3d8003d590>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "acc = history.history['accuracy']\n",
        "val_acc = history.history['val_accuracy']\n",
        "loss = history.history['loss']\n",
        "val_loss = history.history['val_loss']\n",
        "\n",
        "epochs = range(len(acc))\n",
        "\n",
        "plt.plot(epochs, acc, 'bo', label='Training accuracy')\n",
        "plt.plot(epochs, val_acc, 'b', label='Validation accuracy')\n",
        "plt.title('Training and validation accuracy')\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": {
        "colab_type": "text",
        "id": "zb81GvNov-Tg"
      },
      "source": [
        "The Training Accuracy is close to 100%, and the validation accuracy is in the 70%-80% range. This is a great example of overfitting -- which in short means that it can do very well with images it has seen before, but not so well with images it hasn't. Let's see if we can do better to avoid overfitting -- and one simple method is to augment the images a bit. If you think about it, most pictures of a cat are very similar -- the ears are at the top, then the eyes, then the mouth etc. Things like the distance between the eyes and ears will always be quite similar too. \n",
        "\n",
        "What if we tweak with the images to change this up a bit -- rotate the image, squash it, etc.  That's what image augementation is all about. And there's an API that makes it easy...\n",
        "\n",
        "Now take a look at the ImageGenerator. There are properties on it that you can use to augment the image. \n",
        "\n",
        "```\n",
        "# Updated to do image augmentation\n",
        "train_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",
        "These are just a few of the options available (for more, see the Keras documentation. Let's quickly go over what we just wrote:\n",
        "\n",
        "* rotation_range is a value in degrees (0–180), a range within which to randomly rotate pictures.\n",
        "* width_shift and height_shift are ranges (as a fraction of total width or height) within which to randomly translate pictures vertically or horizontally.\n",
        "* shear_range is for randomly applying shearing transformations.\n",
        "* zoom_range is for randomly zooming inside pictures.\n",
        "* horizontal_flip is for randomly flipping half of the images horizontally. This is relevant when there are no assumptions of horizontal assymmetry (e.g. real-world pictures).\n",
        "* fill_mode is the strategy used for filling in newly created pixels, which can appear after a rotation or a width/height shift.\n",
        "\n",
        "\n",
        "Here's some code where we've added Image Augmentation. Run it to see the impact.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 0,
      "metadata": {
        "colab": {},
        "colab_type": "code",
        "id": "UK7_Fflgv8YC"
      },
      "outputs": [],
      "source": [
        "!wget --no-check-certificate \\\n",
        "    https://storage.googleapis.com/mledu-datasets/cats_and_dogs_filtered.zip \\\n",
        "    -O /tmp/cats_and_dogs_filtered.zip\n",
        "  \n",
        "import os\n",
        "import zipfile\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras.optimizers import RMSprop\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "\n",
        "local_zip = '/tmp/cats_and_dogs_filtered.zip'\n",
        "zip_ref = zipfile.ZipFile(local_zip, 'r')\n",
        "zip_ref.extractall('/tmp')\n",
        "zip_ref.close()\n",
        "\n",
        "base_dir = '/tmp/cats_and_dogs_filtered'\n",
        "train_dir = os.path.join(base_dir, 'train')\n",
        "validation_dir = os.path.join(base_dir, 'validation')\n",
        "\n",
        "# Directory with our training cat pictures\n",
        "train_cats_dir = os.path.join(train_dir, 'cats')\n",
        "\n",
        "# Directory with our training dog pictures\n",
        "train_dogs_dir = os.path.join(train_dir, 'dogs')\n",
        "\n",
        "# Directory with our validation cat pictures\n",
        "validation_cats_dir = os.path.join(validation_dir, 'cats')\n",
        "\n",
        "# Directory with our validation dog pictures\n",
        "validation_dogs_dir = os.path.join(validation_dir, 'dogs')\n",
        "\n",
        "model = tf.keras.models.Sequential([\n",
        "    tf.keras.layers.Conv2D(32, (3,3), activation='relu', input_shape=(150, 150, 3)),\n",
        "    tf.keras.layers.MaxPooling2D(2, 2),\n",
        "    tf.keras.layers.Conv2D(64, (3,3), activation='relu'),\n",
        "    tf.keras.layers.MaxPooling2D(2,2),\n",
        "    tf.keras.layers.Conv2D(128, (3,3), activation='relu'),\n",
        "    tf.keras.layers.MaxPooling2D(2,2),\n",
        "    tf.keras.layers.Conv2D(128, (3,3), activation='relu'),\n",
        "    tf.keras.layers.MaxPooling2D(2,2),\n",
        "    tf.keras.layers.Flatten(),\n",
        "    tf.keras.layers.Dense(512, activation='relu'),\n",
        "    tf.keras.layers.Dense(1, activation='sigmoid')\n",
        "])\n",
        "\n",
        "model.compile(loss='binary_crossentropy',\n",
        "              optimizer=RMSprop(lr=1e-4),\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "# This code has changed. Now instead of the ImageGenerator just rescaling\n",
        "# the image, we also rotate and do other operations\n",
        "# Updated to do image augmentation\n",
        "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",
        "      fill_mode='nearest')\n",
        "\n",
        "test_datagen = ImageDataGenerator(rescale=1./255)\n",
        "\n",
        "# Flow training images in batches of 20 using train_datagen generator\n",
        "train_generator = train_datagen.flow_from_directory(\n",
        "        train_dir,  # This is the source directory for training images\n",
        "        target_size=(150, 150),  # All images will be resized to 150x150\n",
        "        batch_size=20,\n",
        "        # Since we use binary_crossentropy loss, we need binary labels\n",
        "        class_mode='binary')\n",
        "\n",
        "# Flow validation images in batches of 20 using test_datagen generator\n",
        "validation_generator = test_datagen.flow_from_directory(\n",
        "        validation_dir,\n",
        "        target_size=(150, 150),\n",
        "        batch_size=20,\n",
        "        class_mode='binary')\n",
        "\n",
        "history = model.fit(\n",
        "      train_generator,\n",
        "      steps_per_epoch=100,  # 2000 images = batch_size * steps\n",
        "      epochs=100,\n",
        "      validation_data=validation_generator,\n",
        "      validation_steps=50,  # 1000 images = batch_size * steps\n",
        "      verbose=2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 0,
      "metadata": {
        "colab": {
          "height": 561
        },
        "colab_type": "code",
        "id": "bnyRnwopT5aW",
        "outputId": "8cdd3e7b-43f0-44de-ad69-30f08e1c5d4d"
      },
      "outputs": [
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYoAAAEQCAYAAACugzM1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzsnXl4VNXd+D93ZjKTdbIvkwRCwhYCwYQdxAVBQFmE1ioo\nSvuzdUGpvtZStFJQEBR9hVbqQqu+4I7VKmhBFNCqLLITICFAIOtkXybrJDNzf39M5mYmM9kgyHY+\nz5PnSe4999xzz2TO937XI8myLCMQCAQCQRuoLvYABAKBQHBpIwSFQCAQCNpFCAqBQCAQtIsQFAKB\nQCBoFyEoBAKBQNAuQlAIBAKBoF2EoBAAYLPZSE1NpbCwsFvbXkxycnJITEzs9n537drFTTfdpPw9\nefJk9u/f36m2XeXpp59m7dq153y9QNAdaC72AATnRmpqKpIkAVBfX49Wq0WlUiFJEs8++yxTp07t\nUn8qlYqDBw92e9uLjWOOLmS/W7Zs6ZYxfPzxx2zcuJF33nlHObZs2bJzG6BA0I0IQXGZ4rxQjx8/\nnueee45Ro0a12d5qtaJWq3+OoQnOgwsl2C41xP/j5YUwPV0ByLJM6wT71atX8z//8z/84Q9/YOjQ\noWzatIlDhw5x5513Mnz4cK677jqWLVuG1WoF7F/cxMRECgoKAPjjH//IsmXL+N3vfseQIUOYNWsW\n+fn5XW4L8N133zFp0iSGDx/OsmXLmD17Np999pnHZ+nMGD/66CMmTpzIyJEjXd64bTYby5cvZ+TI\nkUycOJHvv/++zTl77bXXePzxx12OPfPMM7zwwguA/e3+1ltvZciQIUycOJGPP/64zb5uuOEG9u7d\nC0BDQwN//OMfGTFiBNOmTePo0aNu950wYQJDhgxh2rRpbN++HYDMzEyWLl3K/v37SU1NZfTo0crc\nrlmzRrn+gw8+YOLEiYwaNYpHHnmEkpKSTs1NV+YZ4MSJE/zmN79h5MiRjB07ljfffFO5z9///ndu\nvvlmhg4dyu23305JSYlHM99dd92lfM4ff/wxc+bMYdmyZYwcOZLXXnuN7Oxs7r33XkaOHMno0aNZ\nsGABNTU1yvUFBQU8/PDDjB49mtGjR7N8+XIaGxsZPnw4WVlZSruSkhJSUlKoqqpq83kF54ksuOwZ\nN26cvHPnTpdjq1atkgcNGiR/++23sizLstlsltPS0uTDhw/LNptNzs3NlSdNmiS/++67sizLssVi\nkRMTE+X8/HxZlmX5iSeekEeNGiUfO3ZMtlgs8mOPPSb/8Y9/7HLb0tJSOTU1Vd6+fbtssVjkt99+\nWx44cKD873//2+OzdDTG/v37y/PmzZNramrkvLw8ecSIEcqzv/POO/LUqVPloqIiubKyUr777rvl\nxMREj/fJycmRU1NT5fr6eqXv0aNHy8eOHZNlWZZ37Ngh5+XlybIsy7t375YHDx4snzhxQpZlWd65\nc6d80003KX1df/318k8//STLsiw///zz8j333CNXV1fLBQUF8q233urSdvPmzXJpaaksy7L8xRdf\nyCkpKXJZWZksy7K8YcMG+Z577nEZ5xNPPCG/8sorsizL8vfffy+PGTNGzsjIkM1ms7xkyRL53nvv\n7dTcdGWeq6ur5TFjxsjvvPOO3NjYKNfU1MhHjhyRZVmWX3/9dfm2226Tc3JyZFmW5fT0dLmqqkrO\nzs52m+vZs2crn/OGDRvkpKQk+cMPP5RtNptsNpvlM2fOyLt27ZItFotcVlYmz549W37hhReU55k6\ndaq8cuVKub6+XjabzfKBAwdkWZblRYsWyatWrVLu89Zbb8mPPPKIx+cUdA9Co7iCGTp0KDfccAMA\nWq2WQYMGMXjwYCRJIjY2ljvuuEN5EwbctJJJkyaRlJSEWq1m2rRppKend7ntt99+S1JSEuPGjUOt\nVvPrX/+aoKCgNsfc0RgBHnjgAfz8/IiJiWHEiBFkZGQAdl/B3LlziYiIIDAwkN/97ndt3qdHjx70\n69ePbdu2AfDDDz+g1+tJSkoC4MYbbyQmJgZAeePdt29fm/052LJlC/PmzcPf3x+DwcDdd9/tcn7y\n5MmEhoYCMGXKFGJjY0lLS+uwX4AvvviC22+/nf79+6PVavnDH/7A3r17KSoq6nBuWtPePG/btg2D\nwcCcOXPw8vLCz8+P5ORkAP71r3/x+OOP06NHDwASExPR6/WdGn90dDR33nknkiSh1Wrp1asXo0aN\nQq1WExISwty5c5UxHDx4kIqKCp544gm8vb3RarWkpqYCMGPGDDZt2qT0+/nnn3Pbbbd1agyCc0P4\nKK5gDAaDy99ZWVm88MILHDt2jPr6emw2G4MHD27z+rCwMOV3Hx8f6urquty2uLiYqKgol7at/+7q\nGJ3v5e3tTW1trXIv52d2LPRtMWXKFL744gumTJnCl19+ybRp05RzO3bsUMwjNpuNhoYGZbFsj5KS\nEpfnaz2GTz/9lHXr1mE0GpFlmfr6eioqKjrs1/F8Q4YMUf729/dHr9dTVFSkzElbc9Oa9ua5sLCQ\nuLg4j9cZjUZFSHSV1p97aWkpy5Yt48CBA9TV1WG1WhUhWlhYSGxsrEefzZAhQ9BoNOzfvx+9Xo/R\naFReiAQXBqFRXEUsXryYfv368c0337B//37mz5/vphl0N+Hh4W5htM5vwN05xvDwcIxGo/K3s5/E\nE7feeiu7du2iqKiIbdu2KYLCbDbz6KOP8uCDD7Jr1y727t3Ltdde26lxhIWFtTmG3NxcnnnmGZ59\n9ll++ukn9u7dS3x8vHK+I0d2RESES381NTWYTKZ2BW9btDfPUVFRZGdne7wuOjqanJwct+M+Pj6A\nfe4clJaWurRp/XwvvfQSOp2OL7/8kn379vH888+7jCE/P7/NOZ8xYwaff/45n3/+ObfccgteXl6d\nfHLBuSAExVVEbW0tAQEBeHt7c/r0aT766KMLfs9x48Zx/Phxvv32W6xWK//3f//X7hv0+Yzxlltu\nYd26dRQVFVFRUcE///nPdtuHhoYyZMgQnnzySRISEujZsycAjY2NWCwWgoODkSSJHTt2sGvXrk6P\n4Y033qC6upqCggLef/995VxdXR0qlYrg4GCsVisff/yxi1M2LCyMoqIiLBaLx76nTJnCJ598QmZm\nJo2Njbz88ssMGzaMiIiITo3Nmfbmefz48RQWFvLee+/R1NRETU0NR44cAeD222/nr3/9K7m5uQBk\nZGRgMpkIDw8nLCyMjRs3YrPZ+Oijj5Rgh/bG4OPjg5+fH0ajkbfeeks5l5qaSlBQEC+//DINDQ2Y\nzWYOHDignJ8+fTpfffUVX375JTNmzOjy8wu6xmUhKEwmE6+88gomk+liD+Wi42kuOhtS+ac//YlP\nP/2UIUOGsGTJEm699VaX8879dNRnZ9uGhoayatUqVqxYwahRo8jLyyMpKQmtVnteY3TMg/OiOnv2\nbEaPHs306dO54447mDx5crvPADB16lR27drlYnYKCAjgySef5OGHH2bkyJFs3bqVcePGtdmH8/PP\nnz+fsLAwbrrpJh544AGXRax///7cc8893H777Vx33XWcOXOGa665Rjk/ZswY4uLiuPbaaxk7dqzb\nfa677jrmzZvHww8/zHXXXUdhYSEvvfQSJpOJNWvWuH0O7X0u7c2zv78/b731Fl999RVjxoxh8uTJ\nin/mvvvuY/z48cydO5ehQ4fyl7/8RdEili1bxmuvvcbo0aPJzc11eTZPzJ8/nyNHjjBs2DAefvhh\nJk2apJxTq9W88cYbnDp1ihtuuIFx48axdetW5XxMTAz9+vXDy8uLlJQU5bhYK1ro1rnojMf7zJkz\n8p133ilPmjRJvvPOO+Xs7Gy3NiUlJfJDDz0kT5s2Tb711lvlzz//XDlntVrlJUuWyBMmTJAnTpwo\nb9iwoUse99zcXLlfv35ybm5ul667Ernc58JqtcpjxoyR9+3bd179XO7z0J1crXOxYMECJSLMwdU6\nF57ozrnolEaxePFi5syZw5YtW7jrrrtYtGiRW5sVK1aQnJysZJauWrVKsUVv3LiR3Nxcvv76az74\n4APWrFnToVoquHL4/vvvqampobGxkb///e9oNJp2negCQUfk5uayfft2br/99os9lKuCDgVFeXk5\n6enpTJkyBbCr6sePH3ezM584cYLrrrsOgJCQEBITE9m8eTMAmzdv5o477lDOTZgwocOyB4Irh/37\n9zN+/HhGjx7Njz/+yKuvviqcj4Jz5uWXX2bGjBk8+OCD5+TIF3SdDsNjjUYjkZGRir1TpVIRERFB\nYWEhwcHBSrtBgwbx5ZdfMmjQIHJzczl48CCxsbGAPcMyOjpaaWswGFwiQwRXNo899hiPPfbYxR6G\n4Arh8ccfd8uqF1xYui2PYsGCBaxYsYIZM2ZgMBgYPXo0Gk3XuzeZTG7Ol8LCQoYMGSJqw2B38sXE\nxFz1cyHmoQUxFy2IuWhBrVYzZMgQj1We9Xp9pxMloROCwmAwUFRUhCzLSJKEzWbzmEQVEhLCiy++\nqPx9//3307t3b8Aee11QUMCgQYMAu5bSVjLUunXrXGrbgD3B5oMPPuj0Q13JGAwGpT7Q1YyYhxbE\nXLQg5qIFg8HABx98wOzZs11CiwEeeeQR5s+f3+m+OhQUDn/Dpk2bmD59Ops2bSIpKcnF7ARQWVlJ\nQEAAarWaXbt2cfLkSV555RXAXrZgw4YN3HzzzVRUVLBt2zbeffddj/ebO3cuM2fOdDnmeDuoqKjF\nZruwCWKXA6Gh/pSV1XTc8ApHzEMLYi5aEHNhR6WSCA724+WXX3Yp+Ah0SZuATpqelixZwsKFC3n1\n1VcJDAxk5cqVgF1rePTRRxk4cCBHjhzhueeeQ61WExwczOuvv45OpwPgtttu4/Dhw0ycOBFJknj4\n4YcV/0Vr2lOJbDZZCIpmxDzYEfPQgpiLFsRctNC6lM+5IMnyBa7h0I2UldWIfwAgPDyAkpLqiz2M\ni46YhxbEXLQg5sKOSiURGurfPX11Sy8CgUAguGIRgkIgEAgE7SIEhUAgEAjaRQgKgUAgELSLEBQC\ngUAgaBchKAQCgUDQLkJQCAQCgaBdhKAQCAQCQbsIQSEQCASCdhGCQiAQCATtIgSFQCAQCNpFCAqB\nQCAQtIsQFAKBQCBoFyEoBAKBQNAu3bYVqkAgEAguHnn5eby5/n0qTA0E67156LdzCQ3t0y19C41C\nIBAILnPy8vNYuHQ1WdZkTIFjybIm88Lf3uq2/oVGIRAIBJc5b65/H3XseNQa+66iao0OVciYbutf\naBQCgUBwmVNhalCEhAO1Wttt/QtBIRAIBJc5wXpvrBazyzGrtbHb+u+UoDh79iyzZs1i8uTJzJo1\ni5ycHLc25eXlPPDAA0yfPp1bb72VZ599FpvNBsCaNWsYM2YMM2fOZObMmSxdurTbHkAgEAiudu67\n9y6sedsUYWG1mLEZd3Zb/53yUSxevJg5c+YwdepUNm7cyKJFi1i3bp1Lm9dff53evXvzxhtvYLVa\nmT17Nlu3bmXy5MkAzJgxgwULFnTbwAUCgeBqo3Vk03333kVsTCyxMbE8v+ix5nNmggN0PPT7/9dt\n9+1QUJSXl5Oens6UKVMAmDp1KkuXLqWiooLg4GClnSRJ1NbWIssyDQ0NWCwWIiMjlfOyLHfboAUC\ngeBKxpNAAFi4dLXdaR2oo8JiZuHS1Ty/6DFFWCx+suVlXKWSum08HQoKo9FIZGQkkiQ131xFREQE\nhYWFLoJi3rx5zJ8/n7Fjx1JfX8+cOXNITU1Vzm/evJmdO3cSFhbG/PnzSUlJ8Xg/k8mEyWRyOaZW\nqzEYDOf0gAKBQHCp4ywYtGoLOYUmdL0muQiEmHB/t8gmYsfz5vr3XQREa4xGI1ar1eWYXq9Hr9d3\nenzdFh67ZcsWEhMTWb9+PTU1Nfz2t79l69atTJw4kdmzZ/PQQw+hVqvZuXMn8+bNY/PmzQQGBrr1\ns27dOtasWeNyLCYmhu3btxMa6t9dw73sCQ8PuNhDuCQQ89CCmIsWLqe5yMnJ5c/L/waGcagDdeQf\n2UhU0iRFIFjMNZSaLBQWnqTnqOtdrlVrdNTWW9p93rvvvpv8/HyXY4888gjz58/v9Bg7FBQGg4Gi\noiJkWUaSJGw2G8XFxURFRbm0e/fdd1m+fDkA/v7+jB8/nj179jBx4kRCQ0OVdmPGjCEqKoqTJ08y\nbNgwt/vNnTuXmTNnuhxTq9UAlJXVYLMJE1Z4eAAlJdUXexgXHTEPLYi5aOFym4uVq9fahUSzYJAk\nlfK7ubaMksxvMSRPpTjjG6wWs0sYrNVixs9b4/F5VSqJ0FB/3nvvPY8aRVfoMOopJCSExMRENm3a\nBMCmTZtISkpyMTsBxMbG8v333wPQ2NjIrl276Nu3LwBFRUVKu/T0dAoKCoiPj/d4P71eT2xsrMuP\nMDsJBIIrldY5EJIkKdFL5Wd2Y0ieilqjIyR+FMa0L1wim6x52xT/RVsYDAa3NbWrgqJTpqclS5aw\ncOFCXn31VQIDA1m5ciUA999/P48++igDBw7kqaeeYvHixUyfPh2bzcbIkSO54447AFi1ahXHjh1D\npVKh1Wp58cUXXbQMgUAgaIu2In0uJc5njMF6byqcNIWQ+FEYj2zCMHgasiwrx3V+oYT3u5HijG9Q\nN1UwPCWR+5od2RcaSb6MwpGE6cnO5aZaXyjEPLRwpc6Fo4aRw4nreIt+vp0F8kLNRVvC4FzG2Lrf\n1tfXn95EfGwExzOzCEme7WZuSlCntevAhhbTU3cgBMVlyJW6KHQVMQ8tXKlz8cyKlWRZk7u0UF6I\nuWi9mNdXGSlL30TvhN6UFOXjN+DOc1rMnft/c/37VFTbcyC6Qwh1p6AQRQEFAsElS4WpAXVgqxpG\nGh0VJnMbV1wYnIvumWvLKMvaiWHYXMwaHdWFG9G3rrPUiTF2xlzlKZHO2dz0c5nlhKAQCASXBJ4W\nvdb2e7C/rQcH6Nrpqev3aWtxdbTdeyiDiFR7lKazgxlApVK7RCOZa8soPf0j5VINz6xY6bF/F03B\nQ/KcM60T6c6lj/NFFAUUCAQXHU/7KSxcupopk25yq2HUmUifrt4nLz+v3bZWr2BlDM4OZsAlGslc\nW0bJiR1EJU0iJPkuMkwx/L+Hn+Chx//MMytWKvfxVBZc3Zw811m6o4/OIjQKgUBw0fG06BE7ni+/\n2u5meply/6w2ncpvrn+f2gYLft4aj2/ybd3HU3azc1uHMDAkT1XCV52jkUITxlCb/hFmiw3D4Nke\nTVRZFjOPP72c+NgIjmXmKBqKg66a1H5Os5zQKAQCwUXH434KGh0V1WbF9LJ6+SLuu/cuVq/90E0j\n2Hdgn/L2X+ozqk1Nob37tDcm59BUuc5ISdonLlqOuuIgq1cuo3dCH+Wa1iYqi7mGyloo1I100VAc\ndNWk5rG0+Hma5dpCCAqBQHDR6eyi11ojsJhrKDFZWPCXFe2aYfLy83hmxUpOnz7pdp/6KiN52ad5\n7MlnWfD0X/jToiU89uSz5OVkubTV+YUSkTiB0cNTWPvyEhLUaehNP5KgTlP8As7P0dpEVX5mN4bB\n084rec4ZT6XFz8cs1x7C9CQQCH422nIk33fvXSxcuhpahYHet+gxl+udzS3O5S2K0r/2rCmYzC5O\nX33/eCWZzRHmWpr5DdGpd1JirqHkxA77OT8dXj17U3BwA9Gpd7iNqS0Hs/NztDZRdXfyXEcRUd2J\nyKO4DLlSY+a7ipiHFi7luXAIh4KiMnKNZYQn/9JjTkBbuQTOOOdVGI9+SUTiBLffHdRXGWnK+Rpz\nk5WQwXe5RSX5SDVosCg5EO31ERvXu80xtfm8xWXkFrQ8b+tif9D1fIuuIBLurnIu5UXh50TMQwuX\n6lw4v80XZ3zjthCfS2Kao7+i9K+JTp4KuGoXrprCHS7tnNGbfgRZxhQ4FoCCtC/abLd6+aJzeXwX\n4adVNbWUDz+HDO6uIhLuBALBJYuzeSkvJ0t5Y29ts4euR+k4m1vKLaWKacdhyik8/hV+6lpUtiai\nU+33bW0CAlf/hyNPo6N250JrE1XL3HSfqcgmy6ik7tukyBPCmS0QCLqN1nkK1XKQU/lsqVuidByL\n72svL3dx5mp0/oTrNby39kVieya4Ftlrw3Hs7BB2FOO7kM5h5wiuxU8uOC8hYZNltu7N5eFV/+WH\nI8ZuG6Mn1EuWLFlyQe/QjdTXN3L5GMouHH5+OurqGi/2MC46Yh5auBhzkZefx6q/v85n/9nGnr17\n6JMQx5vr36cqYJSySNeWnMI3JA6VSoPWL5Si9K34hfdGpdJQX2Wk+MgGtDpf9h/YT5+EuC6Vv9br\n9YxITeJs2jbk6jNEe5ex4Pe/I7F/b3Z8+z3l1lBUKg0arS86fSTFmTuwFO8nKcrKgt//jtiYWJc+\nVA1FxIR5E2AtQNOQp/R3vm/8u48X8vZ/0hkxIBIvTfe8mxeV17Hm0zT+e7gAtSRxtrCam4bGuGgW\nkiTh66vtlvsJH8VlyKVqj/65EfPQwsWumOookqfx8iZq6L3K9Z58B2Xpm4iJjqKorKZjx/Y51DAK\nDw/g4KH086ro2p288skRDp4sZfLIntwxrs9595eZW8nLHx1Co1Yxe0JffHQa1nyaxgPTBzIyKVJp\n150+CmF6EggEHmmv3EVbRfJkXahb7oEja1lv+pFEfT5v/f0levWIVoQEuOY9dKXMRls4fBmech0u\nBLIsY6p11+hkWeZ0gQmVJPH13lwKSmvP+15b9uTgo9Ow9LcjuTbZQErfMKJCfNm8J5sL9d4vnNkC\nwRVEd1YTba/chXM+g6dd2Bx/O7KWX1y5zGUc7ZWf6EqZjfZoK9ehu5BlmZyiGvadKGZvejHFlfU8\nMSuFpF4hSpvSqgZMtY3MGBvP1r25vPd1Jk/MSkE6R+dzVW0jR06XMWlED8W3o5IkJo/syf9tzuB4\ndgUDne7fXQhBIRBcBnRGAHSlmmh7/XmqmOrAsZg7V3U9l0Sy9qrC/hw1jBqbrOxJL+LaZMM5Rwxt\nP5DPe19nopIkBsQFUVlj5uDJUhdBcTq/CoCUvmH4+3rx7tZM9mYUM2JAZFvdtsueY4XYZJkxya7b\nQ48eGMW/v89iy+7sCyIohOlJIPgZcZSSeOzJZ12qiXZ0TWdMMW2Vt3hkwRKXe7XXX1sVUx04FnPn\naKHW0UyOUhfDUxLbjOxpr/zEz1HD6Ku9ubz9nwyy8k3n3MehU6UYQn1ZNf9a/jArlf49gzl2ptyl\nzel8EzovNTHhftyYEkPPSH8+3HaSerOly/eTZZkf0ozEG/TEhPm5nPPSqJg4rAfHzlaQXdj9vioh\nKASCn4lztb13tpy0cxE7hxPZUe66Lf9C6/48VUx1LNr1VUaM+9ZRWFLFm+vf57H7Z5GgTqNnuMat\nSJ5zWGljk5U1n6bx5a6zygLZng/hQtcwamyysm1fLgCVNeempciyTHZhNb1jAglojiwaGB9CYXkd\npVX1SrtTBVXEGwJQq1SoVBJ3TehHZU0j+04Ud/meOUU15JXUMjY5yuP5G1Ji8Naq2Xag876cztIp\n09PZs2dZuHAhlZWVBAUFsXLlSnr27OnSpry8nCeffBKj0YjFYmHUqFE8/fTTqFQqbDYbS5cu5Ycf\nfkClUvHb3/6WX/3qV93+MALBpUxHtve2zEGdNcU4m3NaVy513Osf6z+kqlV/5toyys/spqypAo1G\nRUiy3dzUumKqysvXpWT26rUfsuRP8+kTH9duIllWgYkDmSUcyCzhq59ymTSiB0P7R+DlG8KDDz6C\nj06D3imM80LXMNp5tBBTXRNgt/l3RL3Zgo/OdamsqDZTU99EXGSAcmxgr2AAjp+t4PprfDA3Wckr\nrmHyyJa1sk9sIN5aNWcLq7lucNfG/WOaEY1aYkSSZ7OVr7eG/3frAGwXwKHdKUGxePFi5syZw9Sp\nU9m4cSOLFi1i3bp1Lm1ef/11evfuzRtvvIHVamX27Nls3bqVyZMns3HjRnJzc/n6668pLy9n5syZ\nXHvttURHR3f7AwkEF4PO+BA8LfgWcw37Mo/ywGN/aqmD1Mq/0Nld3pwL0rXOgjbXlmGuLiYnMoWG\n6j1o/cxKxJJz+Gr+kY1uey1EJE6gNv0j/Ab8spXgmcDyD09y72QNN6a27Tg+22wKefT2wew4mM8n\n32XxyXdZynkvjYoV948iRO+tHLtQjmibTWbLTznERQaQU1TtMVLJwcm8Sj7/4QzpZyt4eu4w4g0t\nOR7ZRfZnchYU0WF+BPlrOXamnOuvieas0YTVJtM7OlBpo5Ik4iIDumweslht7D5eRGrfcPy8vdps\nNywxokv9dpYOTU/l5eWkp6czZcoUAKZOncrx48epqKhwaSdJErW1tciyTENDAxaLhagou4q0efNm\n7rjjDgBCQkKYMGECW7Zs6e5nEQguCp01KbW2vTt2QwtJnk1OqcUlXNTZv1BbW4P57FcdmmKczTle\nzeUtHPepryxAHzUASVLhHTOKstPfY7WY3TSPsN7XumUnS00mghOu91B+QwuSio+2n6Kksp62OFto\nIkSv45o+YTz2q2tY/Ovh/G5qEr+bmsTcyf2xWG3sOJh/LlPfZQ5kllBcUc+U0XH4+3ph8pCkWFxZ\nz0sfHmTFuwfIK64BCY6cLnNpk1NUgwT0iGjJU5AkiYHxIRw/W47NJpNVYPd/JMS4JhHGRQWQW1yD\n1Wbr9LgPnyqjpr6Ja1s5sX8uOtQojEYjkZGRSjiXSqUiIiKCwsJCgoODlXbz5s1j/vz5jB07lvr6\neubMmUNKSgoABQUFLtqDwWDAaPSccm4ymTCZXB1MarUag+HiTJDgyqG7N6J39LfvYJpLdVK1Rkdj\ncCqPLXia2J4JaNUWJElFZXU9JcZPFIFQevpHpdy1swbQ+i2/yGLGZt1ElHkP5jq1iynG065uDjOW\nQ7swVxcTFJOsjFuSJML73UTtyU2om6pdBIDOL5Tw/uMoT/uA3r37og/wI0+TgiwHudVBkm02NFIT\nkuTL2/9J54nZqR4jiLILq13evOOiAoiLavn7yOkyvjtUwLQxvdB6qc/58+gIWZbZvCebiGAfhvQL\n5/Mfz3ikQ+BiAAAgAElEQVTUKL7ceZZTeVXcMa4P41JjeP79A5zIqQDiXZ4pKtQXndZ1vAPjQ/gx\nrZDsompO5VcREezjYlYD6BUVQJPFRkFpnYugaY+dR40E+msZGB/cceNWGI1GrFaryzG9Xt+lLPhu\nC4/dsmULiYmJrF+/npqaGn7729+ydetWJk6c2KV+1q1bx5o1a1yOxcTEsH379m7LMrwSCA8P6LjR\nVYCnedhz1Mi+jGLm/XKw8oKTk5PLn5f/DQzjFNPOn5f/jddfeoqePXt0+n7VdY28/ukRxgwI4LkX\n7P01eRW6mXnKsnZiSHbf4yDQx0jh/vX079cXP3WtWx2ktvwLPr2nEeKbwUsr/qLcx+WZfOwCxfFM\nqSkD+Meqp1m65jMKowa4PYckqdD3mUyiNo/shka7dtCMRufP2JGpvLTiL/x4uIDn1+/FJnlB2R6s\noSlKfoRK7cWNQ3swoHc0az4+xL6TZUy5Nt7lPrX1TRRV1HPzyLg2/2dvn9CPP7+2k/S8KiaMiOv0\nZ9Eenu6VdqqUM8Zq5t1+DZGResKDfKkzW9zaVtdbiI8O5J6pAwFI7R/Blz+eITDIVxFkeaW1DIwP\ndbv2+qFa1m48zpmiGs4UVpPaL9ytTWoSsOk4ZTWNDBnY8fe4rqGJo2fKuWV0L6IiAzts35q7776b\n/HxXje2RRx5h/vz5ne6jQ0FhMBgoKipClmUkScJms1FcXKyYlRy8++67LF++HAB/f3/Gjx/Pnj17\nmDhxItHR0RQUFDBo0CDALuFiYmI83m/u3LnMnDnT5Zhabf9wRAkPO6J0hZ225uH9rzLIKjAxMC5I\niSlfuXqtfUF1WnythnGsXL3Woy3cYrWhVkluiVH/2HSMXceK2H2gRumvddVR54W+OOMbRWsA8Ak0\nEDX0XkLVaYQG9SWr+TrnRLW2qqwaS2tcnrejZ/LxCUIfPZiynAIarOpWPo5GJLWGPEtv1E6rgMOs\ndfeixygpqeanowVovVTYbDKjb5xEwfFvqDCZ0QVEUiL1YUi/aBLjghkYH8Lbm44RH+FHeJCP0l9G\ntt1EHRaga/N/NkqvIybcj3/vOMXgXsHnnIzmoK3/i0+2ZRLg68XguCBKSqrx1qrIK653a1tUXkt0\nqJ9yvGe4H00WGz8dyad/z2Cq6xoprawnMsjb433iIgPYsussldVmYkN93dp4IeOtVZN2qoSUhJac\nh7ziGnYfL+IXNyS4aGZ7M4ppstgY0COwS997RwmP9957z6NG0RU69FGEhISQmJjIpk2bANi0aRNJ\nSUkuZieA2NhYvv/+ewAaGxvZtWsXffv2BWDy5Mls2LABWZYpLy9n27ZtbWoaer2e2NhYlx9hdhJ0\nltKqesU2vGV3tnK8K3slA7z80SGee2c/tQ1NyrGDJ0vYdayIeEMAjZJ/m9VJbTarx13NWt/XOQzU\nUerCuG8danNRu3kEjlyMvYcy2n2mJouNrAITw5NiPISbfsP//KIPf5iVwtwJMQTL9nDRKHWOS4Le\niZxK+sUGkdQrhJMFDfxl4R9ZvXwRA1OvR6dV07dHEJIk8evJiUgSfLzjlMt4HI7sXlFtvzlLksT4\nobHkFNdwMq9KOS7LMpU1Zo6fLefrfbls3p19zi+KDY0WjmSVMWJApKIV6H21mGob3cpeVNaYCXIK\nFOgXG4gkQUZOJWD3TwDERXq2cCTFB1Na1QBAQrS7BqCSJHp6cGh/uTub/+zOJj3b1f97ILMEfx8v\n+sYGdeWRFQwGg9ua2u2CAmDJkiW8++67TJ48mffff59nn30WgPvvv59jx44B8NRTT7Fv3z6mT5/O\nL37xCxISEhQH9m233UZsbCwTJ05k1qxZPPzww8TG/ryFuQRXB/sySgC4ISXaJfmoK0lcFdVmMnIq\nySow8dKHh6htaKKmvon1W07QI8KfP901BC+5Hlm2OyOd90IoT3ufAKlSuVd7pbVb5xI46iC98dcX\n2swj6GxCHMAZowmL1caQAZ5zFgb1j2dgrxBuGNaf5/94D95aNYnXXK8ICVNdI/mltfTvGcSQfuGU\nVjWQW1yDLMukZZWRFBeMRm1fQkIDvbk22cChU2UuyWQOR7ber/0qpqOTovDz1vDN/jxssszejGIW\nv/UTj6/5kZc+PMQH35zk429PczKvst1+2uLwqTKaLDaGO0UFBfprabTYaGhsedtuaLRQb7a6/F/4\nenvRMzKg2U/REvHUI9Kz8BvUrMXqvNTERvh5bNOrlUO7scnKoZOlAHx3qEBpZ7HaOHK6lJS+YahU\nF3bPifbolI8iISGBDRs2uB1fu3at8nuPHj146623PF6vUqm4jKqZC34G8kpqCAv0xlvbvVVk9p0o\nJi4ygF/d2Ic9x4vY8lMOD0wfyH333sWfX90GAXZ/hGyzYs3bzn2LHnXrIy3LHuFy+429+ez7LF76\n8BDhgd7U1Ddx141RrHjpf7FVqpCCh2CzNqFSeyl7ITy/aAmA4kgOiR9F+ZmfCOszVhEazntBtxUG\n2lYewTMrVrolxDnXVXLu27Go9o21J4W1F27qpVExuHcoh06VYrPJqFQSmc1v0Ik9gwkP8kGS7G+3\narWKMpOZKWN6ufQxPDGCbfvzOHy6lFFJdtN0a0d2W+i0aq67JpqtP+Wy+M2fyC+txRDqy6yb+hAb\n4U9ooDeL/rmHQ6dK6d+z6w7dvRnFBPpr6RPb8obvcDKbahuVPAmHNhbk7yrYEnsGsW1/Pk0WKzlF\n1YQFeuPv4zlMtU9sEFqNSkm080Rrh3ZaVjnmJitxUQEczCyhqraRQD8tGdkV1JutDOkX3uVn7k5E\nZrbgZ6exycrSdfv4cld2x427gMPsNCwxHF9vDTemxrA3vYinV6xm+RtfoA7ogW9jNt71WUgqNddc\n9yuiotxzedKyyggO0HHLyJ7ceb2BbGMV+06UoG04w//+dY39bT54CFaLGdnahLb0W5fMYmdNIZhC\nwvuMQS3bF6BwtbFTVUxjY2J5ZN7vSRr9CxY+8QelvbMJzTkhruTwe24VUk/kVhId5qdkDnfEkH7h\nVNc1caq5PtGJnEp0XmriogLQ+2npGxPIgcxS0ppDRQcnhLpc3yc2kEB/LXvT7VnHdQ0Wiirq2zU7\nOXNTagxqtYRNlrl/ehJL7xvJxBE9SeoVQmSwL/17BnPoVFnHHbWiodFCWlYZw/pHuNj+A5u1HOek\nu8pmQRHs76pp9u8ZjMVqN+V1JPy8NCrm3pLIjOsS2mzjiPo6W2g3k+47UYy/jxf3TRmA1SbzY5o9\nKvRAZgk6L7WSzHexEIJC8LNjLKujyWIjI6ei48ZOZOZWthtv7zA77fr2cx578ln2bP8Qq81GvjyQ\nBt9+yDYblQXHeOr+ydw1oS/puTWs3XjMxUZtsdo4frac5IQQ8gvyefvtf2K12U0pp06kKfsdQ3Pp\nC60vPqF93WoaxcbE8stZ99EUmEyPyEBWPTaBuKgAvEKTiDZ4DuRozbGz5ew+VkROUYstu7UJzZEQ\nN3b4IJcx2Gwyp/Kq6Nej83bt5IRQNGqJA5n2eczIraBvbKBiXkrtF05eSQ3fHS4gJszPJUEO7Lb3\nYf0jSMsqp95sUcYdF9U5e3hYkA//+/C1LL1vJKOSotxMLSl9wigqr6OwvK7TzwSezU6AYg5zDpGt\nrLH/HtTKJOnwUxw8WUpRRT092/BPOBg9MKrduY8M8VUytBubrBw6VcqQfuHEhvvTr0cQ/z1UgNVm\n4+DJUpITQvDSXLiw4c4gBIXgZye/1O4MzC6spsli7aB1C1v25PDBN5lYrJ4TlXam5WEzV5Fr7Ycp\ncCxHT5fao/VU9i+ZpFIpNY0SDRKBcj77TpTw1POvKclxp/OrqDdbST/0PQ/9z5N2M4/avqBISG7O\nY4Bqm/tCWG+2sObTNKJCfHliVgoBvlpuGdmTovI6DjbbojuiujkZrMzUoBxzOMAd/hFZlrEV7+P3\nD/3G5drc4hoaGq30i+18OKWPTkNSrxAOZJbY/RMldv+EA4f5o6i8juRW2oSD4YkRWKw2Dp8u7ZQj\nuzX+Pl5t2uKv6WO/5+FTnZs/B/s8mJ3As0ZR0Vz7qbXvyuGn+L55y9G4LjyTJ5wd2mlZ5ZgbrYog\nuzElmuLKer7clU1VbeNFNzuBEBSCC8i/vj3NP7847nY8v8S+eYvFKnPG2Plwv7ySGixWWbnemdKq\nevJKG0Dj55SboEKlcn0TU2t0FBSXsXDpasqtYQAUyn2VTOqdh88iyzYKrT1o8gp3EQyeHNM2mxVU\nrm/WAIXldTQ22bhtbLxi+hnaP5zwIO9ObzBjqrVHXJVVtQiK2JhYVjz9KCpkvBqL0NBIeN8bCAlz\nrf+TmWv3L3RFowAUp/U3++yCM9HJHxAe5KMkiCUneC5l7Wx+6qwju7OEBfoQG+7XJUHhiHZqbXYC\n8Pf1QsJVo6ioNuOjU3v0nSX2DFIc9T074XfpCIdDe096Ef4+XiTG2T+rof3D8fPWsPGHs6hVEoN7\nh533vc4XISgEFwRZltl1rJB9GcVupQryS2uVNzaHPbwj6hosSshhdpG7cHGYnVROSQFtRRyVFRe5\nFOeTVGrUsRN4c/377D6aDzIu+REOQuJHuZW3sFVmYpO8qGtwLRtdXGEvaREZ3JJToFapmDyiJ1kF\nJpcw0LaorndoFK2K/4VGIUtqfjn5WhbOHUN1vZW/fnTQRfhk5lUSFujtZh7qiJQ+YUgSfPVTjuKf\ncObaZAPBATr6tiGAnM1Pp/KrOuXI7grX9AkjM7fKJWzZGavNxrGsMhqb7JpqW2YnsH8erct4VFab\nCfJ31xoBxYke6Kdts01XcDi092cUM6RfuOL49tKouTbZgE2WGRAXjK/3xd82SAgKwQWhpKqBimoz\njRabmwaQX1JD/x5BRIb4cqoTCybYtQkHR08ZeWbFSn4z70lln4V9J4rRyrUdLuzWvG2ER8Z4rFtU\nViPTJPkiNX9hW+dHaHT+BPlBlHmPEmZ694yb7M/bqtZRcYXdju6cfAb2hTbA14v/7O7YkV/d/KZb\n7mR6ghYNIzTQm97Rgfzqxt7sPlqoaAGyLJOZW3lOcfcOp3WTxebin3Bw87BYXpw3xu24Mw7zU7nJ\n3CWzU2e4pk8YtubwXE/8+79nWPj3H3j0bz/w+udH+WZfrkezkwO9n9ZVo6hpe98Lh5+iO7QJaDFf\nyeAmyG5IiUatks55g6Pu5uKLKsEVycnclnj3s4XVyper3myhzGTmxnA/NGqVPRxTljvcZSy32C4o\nwvRe/HQ0G1mdrJStePwvL+HfZxqqGiMlp35Saik5L+zONZLeXP++khHtQJZlyuq06ALsWctqjdYl\nP8JHqiF1UD/uW/aUi9M6p6ga/mukuLLe5e27uKKe4ACdW+0irZea8UNj+ez7MxzILGnX/lxd5256\nghafRWigXVu4eXgPzhTVsGHHKXrHBOKjU1Nd1+TiX+gKQ/qFk5lX5fF6SZLoKJrfYX6qqmnstCO7\nsyQY9AT4enHkVJkSgusgq8DE5j3ZjBoUhbdGxb4TJdTUNzFhWGyb/1+BflpXH0W1maQ4zxFGvt5e\nTBvTy6WK7PngcGhr1CrF7OTAEOrHS/PGdJvZ7nwRgkLQIY6dta7pE+ZW4KwtTuRW4uetwSbbE7+u\nv8YehurYXD4mzJ8AXy0/pBkpLKsjOsxzYpKD3OIa/Lw1NFblImkjlS++xVyDWR2OvyRhC+hDYIIf\nxn3r6N27N1FhgW4LO7iW41ZrdDSYitAFhKONGIxss2I8slEpu+GcH+EppNWhMTg0CAfFlfVEtNIm\nHEwe0ZMjp8tYu/EYf7p7SJsLj8MkUmpqUErogJNG0WxWkiSJR2elMv/FHbz++VHGDbFHVfXtgiPb\nmRFJkRzILDnnktUqSWJ4f3tORXdrFCqVZM/3OFmK1WZTzDVNFitv/SedIH8dj80aQl1NA3dP7MdZ\nYzUx4W3/b+n9tBRX2LVam02mqqbRLeLJmfZCXrv8LJLEjakx6H21HvMtArvBvNVdCNOToENyimp4\n+z8ZbNmT0+lrTjabPnpFBXDG2FINON8hKML9lIWsM36KvJIaekT401Rf6VILqPzMboJ7jVD+9gk0\nYBg2l6iwwDa34WydEd2Y/ZWyEEsqNeH9xylZ1q1zE1rjo9MQ4OvlwfRUT0SwZ0Gh9VIz/5eD0ftp\n+du/jrhpDGAXztV1TWg1KsyNVuqcsp3LTA1oNSqXhK8AXy0PzhhIRbWZT77NQu/rRVSIbxuz2T5B\n/joWzhlKZPC5XQ9w23XxPH5nygV5I76mdxi1DRYOZpYqfpnPfzhLQWktcycn4tc8L2qVit4xge0m\ndTqX8TDVNWKT5W7dcrUj7hjXx2Vjo0sVISgEHXKk2R58ILOkU9E6VTVmiirq6dcjiHiDnvySWiUM\nNq+kBq1G4u+vvcLzL/0vKrmJw5kF7fZns8nkldQQG+FPsI+rE1OWZY+RTW3VcHLgyIhevXwRsT0T\nXPrQ+YUSM3g6veLi2hQ2zkQE+SjOa7BH2lTVNrYpKMBu8nj0V9fQaLGx+l+H3fZQrjdbsNpkxWTn\nLEzKqhoIDfR2K57n8FfYZFmpwXSx8PP2YmC858io82VgfAh6Xy9e/ewoT/1jD+9/ncnmPdmMHWxg\ncG/PYbtt4VzGo6KNZDuBEBSCTuBwHBZX1Cumo/bIbHZQ9+0RSLwhAKtNVgqpZeWV01BbwRlrMtWB\nY2myWPnpaA4PPf5nxTHdetEsqaynsclGjwh/HrjnF8i2JntYKiBp3N9Y26rh1BZdqQPlifBgHxeN\nwiE0Ijp4I48J82PezEHkl9S61PcBlK06exk8CApTg2J2as3Nw3sw87p4Jo+49N9SzxUfnYZnfzuS\neyf1JyRAx7YDeQQH6Jh1U98u9+VcxsORld2e6elqRfgoBO1S29DE6fwqxg428MMRIwcyS4gJbz8r\nNTO3Eq2XirjIAOXt7IzRRO+YQM4aq8ArALVKrey8FhSTjNlnPFkWM39+ZTNewb35873D8LKZeHP9\n+xTX+YHfAHTU0SO2DwmGPPILS9HXp+Ef34MaWhzQresddYbWPouu9hER5MOeY0U0WWx4aVSK0GjL\nR+HMwF4h+HlrKKlyNV05InEc/otSk6ugaCvhS5IkprXaE+JKRO+r5cbUGG5MjaG6rhFJks4pjNQ5\n6a6tZDuBEBSCDjh2phxZhusHR2MsrWX3sQIO/PfjdneJy8ytJDbUm+dWvkS5qQGVfiTHsgoZkRSJ\nTfLCkXhbfmY3EYk3K9epNToI6o1Nhs07T7Lz6w/s+Q5+9h3g/vbaWp5/+vf07xVObmkDa//2HC+u\n38vhUyWEyUeobFVAr7M4fBaeivB1hvAgH2TsSX+GUD8njaJjQQEQHOBNRatcCUfEU3SoH14alRIi\na26yUl3X1OX8iCuZztay8oRzGY+KajMqSep0wMbVhBAUgnZJyyrDz1tDQrSehCgtXx8wYbUmK7vE\nLVy62sXZW9fQRF5xDU0VmUgB9nayzcqhDCOHe9ud11ZrI2q1tnmvBtcvZW1ZNmrJxt6MGNekOElC\nHTOON9e/zy0z5mKxyuQUmsjIqSApPpR5M9qujNoZ2qri2hkcAqGk0i4oiirqCfD1UiqSdkSIXkd5\ntatD21G+Q++nJUTvrZieHAIjTAiKbsFZo6isNhPor72o5bwvVYSguIro6p7RNlnmaFY58ZE+LH3h\nRQ4ezyZk4C9diuLRXDvJscieyq9CBiR9r5YaSSo1spcf/966G6RYrAXfg2Gs285wNSVZ+AS3JMM5\nu6itFjPFGd9Q1lRBEx+ClMx/D+a3G/f+c+EwMZVU2hfx4oq6TmsTACEBOmWzJQeO0NgAXy/C9Dol\nd6J1DoXg/HAu49Fest3VjhAUVwmODW/UseM9agOehIhVHUhVbSNlZ/ciBSfTYCt061et0bmYTU7k\nVoJsU4SEA0mSMFlD8PPT8MyTD/LWOx/gHa4hN+0TwgffTm3xKfzCeytCwtJYh1rjrWRJ1xSfJCJx\nAmqNjlyLGZWqif/szAJgwEUuwaz306L1Uikmp5LKevr16PyYgvX2vS6aLFalSmh1XRO+Og0atYrQ\nQG9ym8trOzSLEL1Y0LoD5zIeFdVmokPbz+e5WhFRT1cJb65/38WUo9bolEqqzrummQLHkmVNZuHS\n1fxwyL4QS8GJSu0jm801Iql1dNDJ3Cp01LlEEZlrywGwqX2x1FcgSRKLn1zAG6teYO3LS0hQHaGh\n8IB7ZVZJwma12+r9I/u7ajKoaWi0EarXdcppfCGRJInwIHvkU5PFSrnJ3GWNAqDcKaS3uq6RgGaz\nSKjeG1NtI00WK2WmBlSSJN58uxFHGY/WW6AKWhCC4iqhvT2jWwsRi7mGEpOFLT+mY6krc9kbuiTz\nO+X6+iojxn3rKCyp4pkVK8k6m8MZo4nhg3ooW3maa8soObFdyb+ol32VSq3Q4hsYnpLoIlw0Wl97\nHoDF/gatVrvuJubQNBLjgi9qvoCDiCAfiivrKalsQKbzjmxoibJx1sxMtY0E+Nqf2eG4LjOZKatq\nIDjAcyav4NwI9NNSXFFPvdnqtrOdwI74b7vE+e/hAl755EiH7X5MM/LShwdpaLR4PO+vD3RLlrNa\nGgkO0LkIEXNtGSWZ3xKVNAkvvwjqqkuVBVznF0pQjxQsjXWYK3OoytqBYdhczBHjybIm8+f/fQer\nTebAnv8SHeZLlHkPpsxNGAZPUxZzlUqtaDLOOPZZcKnM2lCOSheASra4l/du1jSS4i5MUldXcWgU\nRc2lPLqkUTQLAuckwer6JiX6JizQISgaKDOZ28yhEJwbej8txjL75yY0Nc90SlCcPXuWWbNmMXny\nZGbNmkVOjnsphz/96U/MmDGDmTNnMmPGDAYMGMCOHTsAWLNmDWPGjGHmzJnMnDmTpUuXdu9TXKHY\nbDKbfjzDwZOlbmWsnamoNvPu15kcP1vBJ99muZ1vslixhQwDW5Oy4MqyjNRYyX333uWScFZ+ZjeG\n5GmK4PDWR7lUYNXo/JHqC9EFxirF98Cuhdh8opFlmTr/ZIq8R5NfUkNsTI82NRlnWpfVSFCn8Ytx\nAwB7lc3WQsSSt41po6MZ2v/ib+oCdsHQZLFxMteebNgVc1iwYnpqiXyqdtIoHIKhrKqBsqoGQoQj\nu1vR+2qxNb9Eiaxsz3TKmb148WLmzJnD1KlT2bhxI4sWLWLdunUubV544QXl94yMDH79618zduxY\n5diMGTNYsOD8QhivNo6eKVP2Iigoq6VPjOcibxt2nMJqlRmWGMG2A3kM7R9OYnMkkNVmY/2WExRW\nmJkzoRc/bN9IhcmMKqAnlb49qLb42RPOnnsVKfZGIvqPdwlZ1foEEt5/HOVpH9C7d1/7XtKzf8Hr\n/8lxEQDlZ/cQlXSLojk4IqJK0j/CL8S1UmtbWc+tQ1TrzRa+OlBKYnw4v5nYKs/h6d+TmjKAkpLO\nb3x0IXEIhqNnyvDRaVzqMHWEzkuNn7dG8VHYZJnq+iYlPyAoQIck2Z3kFdVCo+huAp3MTcJH4ZkO\nBUV5eTnp6elMmTIFgKlTp7J06VIqKioIDvYc2fGvf/2LadOm4eXV8mXpTI0ggSvfHSpA66WisclG\nQalnQZGRXcGe40VMG9OLW0fFkVNYzT82HcW/eg+VJjNN+iQapECmX9uL8SMTGD8yEbBrGU+t3c2G\n7acY0CsYv/ibabLYMNcWIAUYUDn5BDQ6f4alDFIWcVmW+eeXJ2hChyTZlVKtb4ibr0Ct0REaEUlt\n3rZzynr20WlYet9I/Hy80HmpzznP4ecgvNnUlFdSS1xUQJf9Js5Jd7X1Tcgy6Js1Co1aRZC/jtP5\nVdhkWYTGdjPOCXbC9OSZDk1PRqORyMhIJxuzioiICAoL3UMlAZqamvjiiy/45S9/6XJ88+bN3Hbb\nbdx3330cOnSozfuZTCby8vJcfoxGY1ee6YqgotrM4VNljB8Si1aj8lhjyWK18e7XmYQFenPr6Dh0\nWjVTR4RSXt1ItnwN1YGjqUePpTSNYQmuTjovjZobBgWTXVTNlt3ZaC2lzL8tnhfmT8CS+7XbZj/3\n3XuXcq0kSVyfEgu07ACnC4j04AMxEx0R6mZSaq8aa2tC9N7ovC7uxvKdIVTvrZQ+j+yCf8KBc9Kd\no86Tc8ZxaKC3kmshNIruxZF019YWqJc7RqPRbU01mUwdX+hEt8/K119/TXR0NImJicqx2bNn89BD\nD6FWq9m5cyfz5s1j8+bNBAa6vyGvW7eONWvWuByLiYlh+/bthIa2X2PoSmLboQJssswvxvcjM7+K\nkqoGwsNb6vuEhwfw2XenKCit5enfjCA22r7xyY/ffoksD0Slsn+0kiQhBfXjvY8+5qUVf1Guz8nJ\n5eOP/g8pejwqtRcmiz+r//46r7/0FP9Y9TR/e+1tyqrqCQ304fernqZnzx4u47vz1hS2H95GlK4M\nW00OqtBUmuoKQReiaA4Yd7Dgpafo2bMHa1IujF/KeU4uNuHBPhSV1xEXHdjlcUVHBHC2sJrw8AAK\nmzWLHk79xIQHKLsB9u0V6rH/S2kuLjZdmYu4RvtWvWFBPlfkHN59993k5+e7HHvkkUeYP39+p/vo\nUFAYDAaKioqUev02m43i4mKioqI8tv/000/dtInQ0JbSv2PGjCEqKoqTJ08ybNgwt+vnzp3LzJkz\nXY6p1fY3yrKyGmy2K9+EZbPJbNl5hoG9glHbbEQE+pCRU8HBQ+m8uf59ahss+HprKPcdycD4EOIj\n/BRbvbGkGlVgy8dqri2j/MxuypoqeOTxRUo29srVa8EwTjExqTU6rIZxrFy9lsVPLuBPj/+Py5ha\n+wK8gIRoPRaLP/fcNY3n3tnPHRNT2ffDFy1+hKd+j49P0AXzI4SHB1wyPgqAUL2OovI6/HXqLo/L\nRyNhqm2kwFhJbkHzRjpNFqUff+8WrUqyWN36v9Tm4mLS1bmwNTZrcD5eV9QcqlQSoaH+vPfee1it\nVp9C6SgAACAASURBVJdzen3XdunrUFCEhISQmJjIpk2bmD59Ops2bSIpKcmjf6KwsJD9+/fz8ssv\nuxwvKioiMtK+92t6ejoFBQXEx3uucKnX67v8EFcaDif2nc1lk2PC/dh1rJCFy9agjrkBtY/9jV1d\nbyUhQuNiDw/We1PRXBbDEepqSJ6KWqMjyykbu8LUgDrQQzSSqf19HJwZmRTJB9+c5Mtd2ahVEuOG\n92XKdZeuH+FCY9/truKcNvxxDpF1VI51Nj05zvv7eKHTXvqmuMsJRxmPKzXiyWAwnHcfnTI9LVmy\nhIULF/Lqq68SGBjIypUrAbj//vt59NFHGThwIACfffYZN910k9tCv2rVKo4dO4ZKpUKr1fLiiy+6\naBlXO3UNFr7Zn0uD2S71j58tR++nJaVvGIBSVkAdc6MSkWSzNqHW6Hj3nbc5sjtS0RScS2bbQ12n\neqzN5CxQHHR1H4cRiRF8uO0kh06VMjA+BF/vzkf6XIkYQnyRODcfhRIiazLby2YD/j4tX0+HX0L4\nJ7oftUrFDSnRJHdx06OriU4JioSEBDZs2OB2fO3atS5/P/jggx6vf/75589haFcHFquNv/87jYzs\nCry8VMg2GYvFgrYhh6cWb0WSVNRbfUA/XBES5toymupNaLR+BPab4qIpOJfMLmuq8JzDYDLzxMO/\nOa89GMC+p29SrxCOnSlnSLNQu5q5ISWGuKiAc9rr2FmjqK5rws/HyyX72hHpJCKeLgz3Tk7suNFV\njMjMvojIsszr/z5AenYFurpM9BXbqTrxORarBZNNz6HMUgp1I6nRD8dmsyh1lsrP7MYvNE4xOTlK\nbjyyYAnPrLBre57KYkCL1uApwa0r0UgObkyJwUenZki/SyPx7WKi06rp3/PcChQ6zB7l1Q2Y6lqS\n7RyECY1CcBG58mLBLmE+3HYSmywzIjGShBg9H21N48ApEzabhQbffpw+kkFU0iTUGh3FGd9gGNyS\nIY0MjbWlePkGg0qj5C+054foaOe289mDwcHQ/uGk9rteCQ0VnBs6bUvSXXVdk9vmOTqtmrsm9CWp\n16VRskRwdSEERTvYbHK3bWJSVFHH1r25AHyzL48gfy2VNY3YbFanUFaVIhjsm/q0mDBUag06/3BK\nD7+Df2BLqGp7fojFTy44r53bOosQEt2DI+muuq6RmDD3ctcThvXwcJVAcOERgqINahuaePKN3cwa\n34cxg84/auDwyVIAlvxmOPmltezLKCbteAYWrxbbvvNGPq039QH7BkBDUq5hwPDJfLXXiM3a5CZQ\nwDV6qTu0BsHPgyPpzlTbqJRgEQguBYSPog1+Si+mpr6J42cr3M6dyq9i4eu7yC7sfMz1oVOlxIT5\n0TMygNEDo5j/y8HEaM66+BBC4kcpBficfwewNldLnXTLdPLLLcSG+ZKgTsPLUtqmH0JweRESoKOs\nqoHaBovYt1lwSSEERRv8mGYvG5JT5C4Mjpwupbiynr/+67Cyh3F71DU0cTKvimv6uEYGtS6trdH5\nE+QHUeY9hFszSOkXRpR5D3rTj/RQpQPQYPPhZE4FAxPCWfzkAl57eblbZdXWJTcElwfBATpqm6sE\nt3ZmCwQXE2F68kBBaS1ZBSb0floKSutobLKidao3dMZYTXCAjnqzhb/+6wgL7x6Cj851Kp23FtXq\nDVil3qS0EhTOoayKD2HZUx59CDZZZt7L37HrWCGNFptSINBjHxfADyG48IQ4RTQJjUJwKSEEhQd+\nPGpEJUncNjaed746QV5JLQnR9iRCWZY5azQxtH84w/pHsPrjI7yx8Rjzf5msxL233p/aZrMi2cxo\nZRPgWt+qsz4ElSRhCPXjdL69mFff2JZ+hB/iysDZXCg0CsGlhDA9tcJmk9l1tJDBvUNJjreHIjqb\nn0oq66ltsNArSs+ghFDuntiPI6fL2H6gpehW661FVSo1sqTh7Xc/OK+xOTK0o0J9zympS3Bp4yoo\nhEYhuHQQgqIVx86WU1nTyLXJUYQGeuPnrSHbSVCcMdp/jzfYNYxxqTFEBvuQkd3i9Pa0P7W1sY59\nh47y2JPP8syKlcqe0V0hJtwuKAaIWPorkpAAJ9OTnxAUgksHISha8WOaEX8fL67pE4YkSfSMDHDR\nKM4YTWjUKmXRBrvQOFtYTV5+Hs+sWMnp0yddIpFkWaY4YxshybMxBY4ly5rMwqWruywsoptj6wfE\ni5o0VyKOpDuVJOHrLazCgkuHq05QWKw23vnqhMdoptqGJg5kljIyKRKN2j41cZEB5BbXYrHaa9af\nNZroGemvnAfoZdBTUW1m4fLXyLImo+8/3SW01VxTQtSgW1yS4tTNSXFdYUBcMDcP68HY/9/e/Qc1\neef7An8nTyAgECAIJBARqLtSf3StB8c91u6sUoQKCtw9bUFZuVtctlqp3XbHjdtDpbWzVTx37bYM\nXb3TPdJT7dbuDws9xa5XnWmP9qz31tqlSLf+QhBCoBAMEURInvtHNOGRAAExGPJ+zTgD33zz5Pt8\nxzwfvr+/FzOue6d7X3hIAIKn+XERI91TfC5Q1F824fgXzdhTVYf+Aeke7R+fasSA1YaH5jvP2ojT\nBGPAakNrRw9sNhGXjRYkaKS74yZo7YedCDE/gKBQQhkUgcjZy9B+/r8AANbuRteL4rrd39IbsJ+t\nnPfId9h/PYVFhwciQsXxJ7q3+Fz79v/Wt0EhyGHo6MGhTy/hsWWzANjHJv7z5GU8NF+D+EGBYGa0\nPQhcNnYDMqCv34p4rfQUrLioEEAUIQw6Z1oZFAFNUgoA4DtafzTd4Zbe5BvWpH4X/Tdbr0T3iinb\noth/5BscPH5ekjZgteGLc+1YlBSFH3wvBodPNeJC81V0Wfrwv6vqEDM9CPkrZkveEx0+DUo/AZeN\n3bhksE9NvTWQfYvSX4AfeiHapF9w0WaDQuzFU+v+RbIorveqAYb/V4nW9qvjHtimqSk8RImosLGf\nZ0F0N03JQNE/YMOnX7bgr6ea8O3VXkd6/WUTrl0fwKKkKDyxfBbUIUr8/qN67K2qw/V+K57Kngel\nn/T0MLlchhlRwWhs7UaDoRsB/gI0EUNPMJs/SwOIA5IV0pCJWHS/VrKlt5/xY1y9eBza5AL0RaWM\ne2CbiMhTpmSguNhyFTcGbLCJomPHVsDe7RSoFDA3QY1ApQL/89H7YejowdeNXfjxitkud+wE7N1P\nl9ssuGgwI14T4nKgce4sLWSCP2YI/4DKfAIaoREymYB//t5MAM5FcTFREYic/6M7HtgmIvKUKRko\nzjaYIJMBD35nOj75sgWW3n5Ht9OCWZHwU9hve26CGv/jB4lY/VA8Hpo//A6xcdHB6LthxeXWbsRr\nXZ/nfWtAe3V2Hl77dQnm/9MP4e8nx+wZYZJ8rtZYjGdgm4jIU6ZkoKi/bEK8RoWchxNxo9+G46ev\nSLqdBstcEo/shxNHvN5MjXPw+vbxiVt0kcFQCDJcMpghiiJqL3bg/rhw+CmkXVnhqgDu9kpEXsWt\nQNHQ0IDc3Fykp6cjNzcXjY2NQ/L88pe/RHZ2NnJycpCdnY37778fx48fBwDYbDa89NJLSE1NRVpa\nGt5///2JvYtBevsGcMlgxpz4cOiigvHAfRH4P59fwYlag6PbaaxipgdBIdi7mxI0IS7zKAQ5ZkSF\n4JKhG0ZTL9q7rrs8rP32HWO52ysR3evcChTbtm1Dfn4+Dh8+jDVr1qCkpGRInp07d+LQoUP4y1/+\ngh07diA0NBRLly4FAFRVVaGpqQlHjhzBu+++i/LycrS0tEzsndx07koXrDYR9988+OXRxXHo7unH\nqfo2SbfTWCgEOWKnByM40G/Ew+0TtCG43NqNL8/bDymanzg0UEzUWdVERJ4y6jqKzs5O1NfXIyMj\nAwCQmZmJ7du3w2QyITzc9Slcf/zjH7Fq1Sr4+dnXFdTU1ODxxx8HAKjVajzyyCM4fPgwnnzyyYm6\nD4ezDSYoBLljG+7vzghDglaFSwbzkG6nsVj5zzNxrbcfzS3Nju3Dw1UBKFy3xvGQT9CqcOx0M45+\nfgUa9TREDjPNkbu9EpE3GTVQGAwGREdHQ3Zzpo9cLkdUVBRaW1tdBor+/n58+OGH2LdvnyOtpaUF\nMTHObSe0Wi0MBoPLzzObzTCbzZI0QRCg1bp3HGn9ZRNmxaoc50fIZDI8vuw+fHyqaUi30+AzI25/\n6N9uUVLUkO3DTQN90G9/zdEiuDXQ/e3V60jl+cZEdA8wGAywWqW7UKhUKqhUrsdbXZnwldlHjhxB\nTEwMkpKSxvX+yspKlJeXS9JiY2Nx7NgxREQEj/jeq5Y+NLVZkP9oEiIjnWMJkZEhWPpPcZK8jY1N\neOHXrwPaZY6H/gu/fh2/+7dfIS7O9UN+52/el2wfLiiUuBH+IJ7b+iLiE2ZBrQqE0u9B9PXb8PBC\nnaQME+1uXtubsB6cWBdOrAuntWvXorm5WZK2adMmFBcXu32NUQOFVquF0WiEKIqQyWSw2Wxoa2uD\nRqNxmf/Pf/4zfvSjH0nSYmJi0NLSgnnz5gGwR7jY2FiX7y8oKEBOTo4kTRDsrYOODgtsNnHYsp6q\nNwIAZk4PQnv7yOdZl7221x4kBj30rdplKHtt77DdQob2bgihztlJfdc60HHxJLTzH8O3CiWMvX2Q\nDbRDGTQd0SrlqGUYr8jIkLt2bW/CenBiXTixLuzkchkiIoKxf/9+ly2KsRg1UKjVaiQlJaG6uhqr\nV69GdXU15syZ47LbqbW1FZ9//jl+85vfSNLT09Nx8OBBpKamwmQy4ejRo3jnnXdcft5Ym0SDfX3Z\nhAB/YcheTK6YzNclD33g5noG8/DrGcJVATAN2rOp89J/Qzs/UxpsIINq4AL8FMvGdQ9ERBPJ3W77\nkbg1Bai0tBTvvPMO0tPTceDAAbz88ssAgKKiItTV1TnyHTp0CMuXLx/yoM/KyoJOp8OKFSuQm5uL\np59+GjrdxM/yOXvZhNkzwhxHko5kLOsZbp0z0WLsQHvtnxzvs9msLhbP+aO/2/X4CxGRN3JrjCIx\nMREHDx4ckr53717J70899ZTL98vlcpSWlo69dGNg6u5Dm6kXyxe6F4AK162BfvtrwM0xB8d6hpJn\nJfkkA9gaJUID7Rv63XfffQiRdcHKXWGJaIqbMiuzL7fa+yQTY9zrtnJ3PcPt518HhmqhTS6AZnoo\nXit7hYvniGjKmzLnUTS2dUMGQBfpemM/VwavZxhuquxIYxm3go39ffaWRCEXzxHRFDNlAkWT0YIo\n9TQE+I/9llytj3juX3+NBF0ULly4APUDi4btXuLiOSKa6qZM11NjWzfiokZeZzGc27uXBvos6LoG\ntCoXDzn/mt1LRORrpkSLouf6ANq7ruMH34sZPbMLt3cvdV76b2gfWGU/K0KhROTsZWg9+zECZRY8\nOO+77F4iIp8yJQLFlXYLAGBG1PhWY96+PkIURUlXkzIoArEPrIbKfILdTETkc6ZE11Oj0T7jKS56\nfF1Pt2/9LYo2nhlBRHTTlGhRNLZZoJrmh9Ag/1HzDje7afDspemJKjQ2fAxlfNqIayyIiHzBlAgU\nTUYLZkSHOHa4Hc5ou78O7lZyBhROeyUi3+b1gWLAakPzt5Zht/Ue3IK40ngRQfc/IdmbCboUvPX2\ngSFjD5z2SkRk5/WBorWjBwNWETNcjE/c3oLoFjuhGrI308gbARIR+TqvH8xubLs5kO1ixtPt6yPk\ncoGD1EREY+T9gcJogb9CDo162pDXTObrkmmu6oTvw1D7IRfPERGNgdd3PTW1WRAbGQy5fOhA9u3r\nI5RBEYhIXIJr9e9BN/M+DlITEbnBqwOFKIpoNHYjOSnK5euuthIXTF9gV9krDA5ERG7y6q4nU3cf\nrl0fGHaPJ3e3EiciouF5dYui0Xhz647o4bfu4DRXIqI749UtiuZv7YFiLGdQEBHR2Hh1oLh67QYC\n/IVxnUFBRETucesJ29DQAL1ej66uLoSFhaGsrAxxcXFD8n300Ud48803AQAymQz79u2DWq1GeXk5\nDhw4gOjoaADAwoULUVJScseFt/T2IzjQ746vQ0REw3MrUGzbtg35+fnIzMxEVVUVSkpKUFlZKclT\nW1uLiooKvP3221Cr1bBYLPD3d27Sl52djS1bJnaswNLTj5Bpo28ESERE4zdq11NnZyfq6+uRkZEB\nAMjMzMTZs2dhMpkk+SorK/Hkk09CrVYDAIKDgyWBQhTFiSw3AKC7tx8h09iiICK6m0ZtURgMBkRH\nRzt2ZpXL5YiKikJrayvCw8Md+S5cuACdTof8/Hz09PQgNTUVGzZscLxeU1ODkydPYvr06SguLsaC\nBQtcfp7ZbIbZbJakCYIArVY7JK+lpx+x0zmQTUQ0HIPBAKvVKklTqVRQqVRuX2PCRoEHBgbwzTff\nYN++fejr68P69esRExODrKws5OXlYcOGDRAEASdPnsTGjRtRU1OD0NDQIdeprKxEeXm5JC02NhbH\njh1DRIR0vYTlej+iIoIQGTm+k+28mS/esyusByfWhRPrwmnt2rVobm6WpG3atAnFxcVuX2PUQKHV\namE0GiGKImQyGWw2G9ra2qDRaCT5YmNjkZaWBoVCAYVCgZSUFNTW1iIrKwsRERGOfEuWLIFGo8G5\nc+eQnJw85PMKCgqQk5MjSRMEAQDQ0WGBzWbvwurrt6LvhhUCRLS3d0vyD3c40VQRGRky5J59EevB\niXXhxLqwk8tliIgIxv79+122KMZ0rdEyqNVqJCUlobq6GgBQXV2NOXPmSLqdAPvYxYkTJwAA/f39\n+OyzzzB79mwAgNFodOSrr69HS0sLEhISXH6eSqWCTqeT/HPV7XSttx8Ahgxm39pa/KJ1PsyhS3HR\nOh/67a/hSvOV0W6ViGjK0Wq1Q56pYw0UbnU9lZaWQq/Xo6KiAqGhoSgrKwMAFBUVYfPmzZg7dy4y\nMjLw1VdfYeXKlRAEAUuXLsVjjz0GANi9ezfq6uogl8vh7++PXbt2SVoZ49HdYw8Ut0+PvX1r8ZEO\nJyIiotG5FSgSExNx8ODBIel79+51/CyTyaDX66HX64fk27Fjxx0U0TVLr+tAYTJfhxDKw4mIiCaK\n167M7u65AQBDpseGqwJ4OBER0QTy3kAxTIuicN0aWK8c5eFEREQTxGs3SbL09EMmA4IC7IFi8Eyn\nmOnTIOv7G/p6BB5ORER0h7w3UPT2IyjAD3K5zDHTSdClQAhVOloRPHuCiOjOeW/XU88Nx/iEq5lO\nws2ZTkREdGe8NlAM3jnWZL7uCBK3CAolTN2c6UREdKe8NlB0DwoUnOlERHT3eG2gGLzFOGc6ERHd\nPV4ZKERRhGXQFuO6WB12lDyLRKEWKvMJJAq1HMgmIpogXjnrqbdvAFabKFlDoYvVcYsOIqK7wCtb\nFMMttiMioonnlYHC0nNr51gGCiKiu80rA0X3MFuMExHRxPPKQGEZZotxIiKaeF4ZKLp77TvHMlAQ\nEd19XhkoLD39UAgyBPgLk10UIqIpzysDxa1V2TKZbLKLQkQ05XnlOgpLTz8C/GR46dUymMzXEa4K\nQOG6NVxgR0R0F3hli6LjqgUtLc24aJ0Pc+hSXLTOh377a7jSfGWyi0ZENOW4FSgaGhqQm5uL9PR0\n5ObmorGx0WW+jz76CKtWrcKqVauwevVqdHZ2AgBsNhteeuklpKamIi0tDe+///4dFdrQ1gUow7mt\nOBGRB7jV9bRt2zbk5+cjMzMTVVVVKCkpQWVlpSRPbW0tKioq8Pbbb0OtVsNiscDf377OoaqqCk1N\nTThy5Ag6OzuRk5ODhx56CDExMeMq9IAoQC6XDmQLCiVMZm4rTkQ00UZtUXR2dqK+vh4ZGRkAgMzM\nTJw9exYmk0mSr7KyEk8++STUajUAIDg42BEoampq8PjjjwMA1Go1HnnkERw+fHhcBbbabIDcDzbr\ngDSd24oTEd0Vo7YoDAYDoqOjHTOM5HI5oqKi0NraivDwcEe+CxcuQKfTIT8/Hz09PUhNTcWGDRsA\nAC0tLZLWg1arhcFgcPl5ZrMZZrNZkiYIArRaLQDgWq89QNi6/gExdBYEhfPo08KSZ8dy70REU57B\nYIDVapWkqVQqqFQqt68xYbOeBgYG8M0332Dfvn3o6+vD+vXrERMTg6ysrDFdp7KyEuXl5ZK02NhY\nHDt2DBERwbjWbwMAFOVn4MTRD9BxtRcRoYF4Zve/Ii5uxkTdzj0vMjJksotwT2A9OLEunFgXTmvX\nrkVzc7MkbdOmTSguLnb7GqMGCq1WC6PRCFEUIZPJYLPZ0NbWBo1GI8kXGxuLtLQ0KBQKKBQKpKSk\noLa2FllZWYiJiUFLSwvmzZsHwB7hYmNjXX5eQUEBcnJyJGmCYB+P6OiwoLG5CwCgDg3DL5/7uSRf\ne3u3m7ft3SIjQ3zmXkfCenBiXTixLuzkchkiIoKxf/9+ly2KMV1rtAxqtRpJSUmorq4GAFRXV2PO\nnDmSbifAPnZx4sQJAEB/fz8+++wzzJ49GwCQnp6OgwcPQhRFdHZ24ujRo1ixYoXLz1OpVNDpdJJ/\nt7qdAKCb+zwREblNq9UOeaaONVC41fVUWloKvV6PiooKhIaGoqysDABQVFSEzZs3Y+7cucjIyMBX\nX32FlStXQhAELF26FI899hgAICsrC19++SVWrFgBmUyGp59+Gjrd+BbHWbhzLBGRR8lEURQnuxDu\n6uiw4IP/uoS/fHIRe37xQ/gpvHK94B1j09qO9eDEunBiXdjd6nqakGtNyFU8yNLTjwB/wWeDBBGR\np3nd07a79wbHJ4iIPMjrAoWlp59HoBIReZDXBQr7FuMcyCYi8hSvCxRsURAReZb3BYqbhxYREZFn\neFWg6B+woa/fyhYFEZEHeVWguHadq7KJiDzNqwJFz3X7zrEMFEREnuNVgaLJYAQA/Pt/HMBLr5bx\n6FMiIg/wqkDxXtUxAEBP8AM8J5uIyEO8KlDIw77r+JnnZBMReYZ3BQq5dLNbQaGEqZvnZBMR3U1e\nFShsNpvkd56TTUR093lVoBD7voUo2oOF45zsdWsmuVRERFPbhJ2Z7QkJcTNh+UcjlOYvER6iRGHJ\ns9DFju8AJCIico9XBYp+mxz3fycez/zL6skuChGRz/CqrqeevgEutiMi8jC3WhQNDQ3Q6/Xo6upC\nWFgYysrKEBcXJ8lTXl6OAwcOIDo6GgCwcOFClJSUAAC2bt2KkydPQq1WAwDS09Pxs5/9bMyFvXa9\nH8HTxnYoOBER3Rm3AsW2bduQn5+PzMxMVFVVoaSkBJWVlUPyZWdnY8uWLS6vUVRUhLVr195RYQcG\nbAhhi4KIyKNG7Xrq7OxEfX09MjIyAACZmZk4e/YsTCbTkLyiKE58CW/DriciIs8aNVAYDAZER0dD\nJpPZ3yCXIyoqCq2trUPy1tTUICsrC4WFhThz5ozktX379mH16tXYtGkTLly4MO4CB3OLcSIij5qw\nWU95eXnYsGEDBEHAyZMnsXHjRtTU1CA0NBQ///nPERUVBQA4dOgQfvrTn+Lo0aOO4DOY2WyG2WyW\npAmCAK1WCwAI4TGoRERuMxgMsFqtkjSVSgWVyv3x3lEDhVarhdFohCiKkMlksNlsaGtrg0ajkeSL\niIhw/LxkyRJoNBqcO3cOycnJjiAB2McxXn31VbS2tjoe/oNVVlaivLxckhYbG4tjx+wbAsbpwhAZ\nGez2DU5VkZEhk12EewLrwYl14cS6cFq7di2am5slaZs2bUJxcbHb1xg1UKjVaiQlJaG6uhqrV69G\ndXU15syZg/DwcEk+o9HomPFUX1+PlpYWJCQkDHnt008/hUKhcPx+u4KCAuTk5EjSBEFw/Hyj9wba\n27vdvsGpKDIyxOfrAGA9DMa6cGJd2MnlMkREBGP//v0uWxRj4VbXU2lpKfR6PSoqKhAaGoqysjIA\n9plMmzdvxty5c7F7927U1dVBLpfD398fu3btcrQy9Ho9Ojo6IJPJEBISgjfffBNyuevhkZGaRDIZ\nMC3Aq9YIEhFNKlc9N2MlEz0xVWmCPPO/jqNkXfJkF2PS8S8mO9aDE+vCiXVhd6tFMSHXmpCreEgQ\nWxNERB7nVYFiWgCnxhIReZpXBQq2KIiIPI+BgoiIRuRlgYKL7YiIPM2rAgWnxhIReZ5XBYrgQAYK\nIiJP86pAwVlPRESe51WB4sP//BBXmq9MdjGIiHyKVwWKVtsM6Le/xmBBRORBXhUoBMEfgi4Fb719\nYLKLQkTkM7wqUACAoFDC1N032cUgIvIZXhcorAN9CA9RTnYxiIh8hlcFCqv1BqxXjqJw3ZrJLgoR\nkc/wqoUJcfJ/YHPJs9DF6ia7KEREPsOrAsXmjT+DzeY1x2cQEU0JXtX1REREnsdAQUREI2KgICKi\nEbk1RtHQ0AC9Xo+uri6EhYWhrKwMcXFxkjzl5eU4cOAAoqOjAQALFy5ESUkJAOD69evYunUr6urq\noFAosGXLFvzwhz+c2DshIqK7wq1AsW3bNuTn5yMzMxNVVVUoKSlBZWXlkHzZ2dnYsmXLkPS33noL\nwcHB+Otf/4rLly9j7dq1OHLkCAIDA+/8DoiI6K4ateups7MT9fX1yMjIAABkZmbi7NmzMJlMQ/KK\nousZSTU1NcjNzQUAzJw5E/PmzcMnn3xyJ+UmIiIPGTVQGAwGREdHQyaT2d8glyMqKgqtra1D8tbU\n1CArKwuFhYU4c+aMI72lpQUxMTGO37VaLQwGw0SUn4iI7rIJW0eRl5eHDRs2QBAEnDx5Ehs3bkRN\nTQ1CQ0MdQcYdZrMZZrNZkiYIArRaLeRy968z1bEu7FgPTqwLJ9aFsw4MBgOsVqvkNZVKBZVK5fa1\nRg0UWq0WRqMRoihCJpPBZrOhra0NGo1Gki8iIsLx85IlS6DRaHDu3DkkJycjJiYGLS0tCA8PdxT8\n+9//vsvPq6ysRHl5uSRt4cKFePfddxEeHuT2jU11ERHBk12EewLrwYl14cS6cHruuedw+vRpODNY\nlgAABdpJREFUSdqmTZtQXFzs9jVG7XpSq9VISkpCdXU1AKC6uhpz5sxxPPRvMRqNjp/r6+vR0tKC\nhIQEAEBaWhree+89APYZVF999RUefvhhl59XUFCAo0ePSv49//zzyMvLY3cV7EF2+fLlPl8XrAcn\n1oUT68LJYDAgLy8Pzz///JBnakFBwZiu5VbXU2lpKfR6PSoqKhAaGoqysjIAQFFRETZv3oy5c+di\n9+7dqKurg1wuh7+/P3bt2uVoZRQWFkKv12PFihUQBAHbt2/HtGnTXH7WcE2i06dPD2k++SKr1Yrm\n5mafrwvWgxPrwol14WS1WnH69GloNBrodHe2P55bgSIxMREHDx4ckr53717Hzzt27Bj2/YGBgfjt\nb387juIREdFk48psIiIaEQMFERGNSCgtLS2d7EK4Q6lUYvHixVAqebod68KO9eDEunBiXThNVF3I\nxOGWUxMREYFdT0RENAoGCiIiGtE9HygaGhqQm5uL9PR05ObmorGxcbKL5BFdXV0oKirCo48+iqys\nLDzzzDOOjRjPnDmDrKwspKeno7CwEJ2dnZNcWs8pLy9HUlISzp8/D8A36+LGjRsoLS1FWloaVq9e\njRdffBGAb35Xjh8/jpycHGRnZyMrKwtHjhwB4Bt1sXPnTqSkpEi+D8DI9z7uehHvcevWrROrq6tF\nURTFDz74QFy3bt0kl8gzurq6xFOnTjl+37lzp/jCCy+IoiiKqamp4unTp0VRFMWKigpx69atk1JG\nT6urqxPXr18vLlu2TDx37pwoir5ZF9u3bxdfffVVx+8dHR2iKPrmd2XRokXi+fPnRVEUxa+//lp8\n8MEHRVH0jbr4/PPPxdbWVnH58uWO74Mojnzv462XezpQdHR0iIsWLRJtNpsoiqJotVrF5ORksbOz\nc5JL5nkff/yx+JOf/ET8+9//LmZmZjrSOzs7xQULFkxiyTyjr69PfOKJJ8QrV644AoUv1sW1a9fE\n5ORksaenR5Luq9+VxYsXO/5QOHXqlJiWliZ2dHSIycnJPlMXg/9wGun/wZ38H5mw3WPvhpG2OL99\nr6mpTBRFvPvuu0hJSYHBYEBsbKzjtVv1YDabx7QbpLd5/fXXkZWVJbl3X6yLxsZGhIWF4Y033sDf\n/vY3BAUFYfPmzQgICPDJ78ru3buxYcMGTJs2DdeuXcPevXthMBig0Wh8ri6AkZ+ZNptt3P9H7vkx\nCgJefvllBAUFIT8/3+Xr4hSf4XzmzBnU1tYiLy/PkTbcPU/1urBarWhqasK8efPwpz/9Cb/4xS9Q\nXFyMnp6eKX/vt7Nardi7dy9+97vf4dixY3jzzTfx7LPPoqenZ7KLNuXc0y0Kd7c4n8p27tyJxsZG\n7NmzB4C9Tpqbmx2vd3Z2QiaTTdm/oAHg1KlTuHTpElJSUiCKIoxGI9avX48f//jHPlcXMTExUCgU\nWLlyJQDggQcegFqthlKpRFtbm099V+rr69He3o4FCxYAsB9HEBgYCKVS6bPPjZGembe+O+Opl3u6\nReHuFudT1e7du3H27FlUVFRAobDH9Hnz5qGvr8+xv/wf/vAHPProo5NZzLuuqKgIn3zyCY4ePYpj\nx44hOjoav//971FYWOhzdREeHo7FixfjxIkTAIBLly6ho6MDiYmJPvdd0Wg0aG1txaVLlwAAFy5c\nQEdHB+Lj432uLm61Jkd6Zt7J8/SeX5l98eJF6PV6mM1mhIaGYufOnYiPj5/sYt1158+fx6pVqxAf\nH+9Yfj9jxgy88cYb+OKLL/Diiy/ixo0b0Ol02LVrF9Rq9SSX2HNSUlKwZ88ezJo1C2fOnEFJSYlP\n1UVTUxN+9atfoaurC35+fnjuueewdOlSn/yufPjhh9izZw8EQQAAPPPMM1i+fLlP1MUrr7yCI0eO\noKOjA2FhYQgPD0d1dfWI9z7eernnAwUREU2ue7rriYiIJh8DBRERjYiBgoiIRsRAQUREI2KgICKi\nETFQEBHRiBgoiIhoRAwUREQ0ov8PGezJ0mW3p2QAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f3d7f79fbd0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYoAAAEQCAYAAACugzM1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f3d7fad3510>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "acc = history.history['accuracy']\n",
        "val_acc = history.history['val_accuracy']\n",
        "loss = history.history['loss']\n",
        "val_loss = history.history['val_loss']\n",
        "\n",
        "epochs = range(len(acc))\n",
        "\n",
        "plt.plot(epochs, acc, 'bo', label='Training accuracy')\n",
        "plt.plot(epochs, val_acc, 'b', label='Validation accuracy')\n",
        "plt.title('Training and validation accuracy')\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": "code",
      "execution_count": 0,
      "metadata": {
        "colab": {
          "height": 3731
        },
        "colab_type": "code",
        "id": "TmYgnrcobqw1",
        "outputId": "556cc15b-fe3d-40c0-a5e7-33c787ce1031"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "--2019-02-12 07:59:45--  https://storage.googleapis.com/mledu-datasets/cats_and_dogs_filtered.zip\r\n",
            "Resolving storage.googleapis.com... 2607:f8b0:4001:c1c::80, 173.194.197.128\r\n",
            "Connecting to storage.googleapis.com|2607:f8b0:4001:c1c::80|:443... connected.\r\n",
            "WARNING: cannot verify storage.googleapis.com's certificate, issued by 'CN=Google Internet Authority G3,O=Google Trust Services,C=US':\r\n",
            "  Unable to locally verify the issuer's authority.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 68606236 (65M) [application/zip]\n",
            "Saving to: '/tmp/cats_and_dogs_filtered.zip'\n",
            "\n",
            "/tmp/cats_and_dogs_ 100%[=====================>]  65.43M   243MB/s   in 0.3s   \n",
            "\n",
            "2019-02-12 07:59:46 (243 MB/s) - '/tmp/cats_and_dogs_filtered.zip' saved [68606236/68606236]\n",
            "\n",
            "Found 2000 images belonging to 2 classes.\n",
            "Found 1000 images belonging to 2 classes.\n",
            "Epoch 1/100\n",
            "100/100 - 14s - loss: 0.6931 - acc: 0.5350 - val_loss: 0.6907 - val_acc: 0.5080\n",
            "Epoch 2/100\n",
            "100/100 - 14s - loss: 0.6855 - acc: 0.5400 - val_loss: 0.6660 - val_acc: 0.6200\n",
            "Epoch 3/100\n",
            "100/100 - 13s - loss: 0.6702 - acc: 0.5810 - val_loss: 0.6665 - val_acc: 0.5650\n",
            "Epoch 4/100\n",
            "100/100 - 13s - loss: 0.6541 - acc: 0.6000 - val_loss: 0.6342 - val_acc: 0.6300\n",
            "Epoch 5/100\n",
            "100/100 - 14s - loss: 0.6415 - acc: 0.6180 - val_loss: 0.6457 - val_acc: 0.5920\n",
            "Epoch 6/100\n",
            "100/100 - 13s - loss: 0.6248 - acc: 0.6495 - val_loss: 0.5875 - val_acc: 0.6840\n",
            "Epoch 7/100\n",
            "100/100 - 13s - loss: 0.6115 - acc: 0.6575 - val_loss: 0.5864 - val_acc: 0.6810\n",
            "Epoch 8/100\n",
            "100/100 - 13s - loss: 0.6010 - acc: 0.6780 - val_loss: 0.5550 - val_acc: 0.7130\n",
            "Epoch 9/100\n",
            "100/100 - 14s - loss: 0.5972 - acc: 0.6670 - val_loss: 0.5640 - val_acc: 0.7020\n",
            "Epoch 10/100\n",
            "100/100 - 14s - loss: 0.5877 - acc: 0.6920 - val_loss: 0.5830 - val_acc: 0.6900\n",
            "Epoch 11/100\n",
            "100/100 - 14s - loss: 0.5761 - acc: 0.7055 - val_loss: 0.5663 - val_acc: 0.7030\n",
            "Epoch 12/100\n",
            "100/100 - 14s - loss: 0.5708 - acc: 0.7100 - val_loss: 0.5662 - val_acc: 0.7030\n",
            "Epoch 13/100\n",
            "100/100 - 14s - loss: 0.5810 - acc: 0.6935 - val_loss: 0.5600 - val_acc: 0.6980\n",
            "Epoch 14/100\n",
            "100/100 - 14s - loss: 0.5734 - acc: 0.7025 - val_loss: 0.5253 - val_acc: 0.7220\n",
            "Epoch 15/100\n",
            "100/100 - 13s - loss: 0.5616 - acc: 0.7150 - val_loss: 0.6329 - val_acc: 0.6470\n",
            "Epoch 16/100\n",
            "100/100 - 14s - loss: 0.5487 - acc: 0.7150 - val_loss: 0.5577 - val_acc: 0.7160\n",
            "Epoch 17/100\n",
            "100/100 - 13s - loss: 0.5575 - acc: 0.7180 - val_loss: 0.5160 - val_acc: 0.7390\n",
            "Epoch 18/100\n",
            "100/100 - 13s - loss: 0.5481 - acc: 0.7250 - val_loss: 0.5057 - val_acc: 0.7360\n",
            "Epoch 19/100\n",
            "100/100 - 14s - loss: 0.5398 - acc: 0.7285 - val_loss: 0.5052 - val_acc: 0.7320\n",
            "Epoch 20/100\n",
            "100/100 - 13s - loss: 0.5448 - acc: 0.7240 - val_loss: 0.4988 - val_acc: 0.7560\n",
            "Epoch 21/100\n",
            "100/100 - 13s - loss: 0.5321 - acc: 0.7345 - val_loss: 0.5014 - val_acc: 0.7500\n",
            "Epoch 22/100\n",
            "100/100 - 13s - loss: 0.5379 - acc: 0.7295 - val_loss: 0.4910 - val_acc: 0.7500\n",
            "Epoch 23/100\n",
            "100/100 - 13s - loss: 0.5211 - acc: 0.7395 - val_loss: 0.4985 - val_acc: 0.7400\n",
            "Epoch 24/100\n",
            "100/100 - 14s - loss: 0.5236 - acc: 0.7420 - val_loss: 0.5055 - val_acc: 0.7410\n",
            "Epoch 25/100\n",
            "100/100 - 13s - loss: 0.5206 - acc: 0.7360 - val_loss: 0.4907 - val_acc: 0.7550\n",
            "Epoch 26/100\n",
            "100/100 - 14s - loss: 0.5234 - acc: 0.7310 - val_loss: 0.4880 - val_acc: 0.7430\n",
            "Epoch 27/100\n",
            "100/100 - 13s - loss: 0.5126 - acc: 0.7470 - val_loss: 0.4863 - val_acc: 0.7510\n",
            "Epoch 28/100\n",
            "100/100 - 13s - loss: 0.5086 - acc: 0.7545 - val_loss: 0.5446 - val_acc: 0.7160\n",
            "Epoch 29/100\n",
            "100/100 - 13s - loss: 0.5237 - acc: 0.7330 - val_loss: 0.5041 - val_acc: 0.7470\n",
            "Epoch 30/100\n",
            "100/100 - 13s - loss: 0.5077 - acc: 0.7475 - val_loss: 0.4819 - val_acc: 0.7510\n",
            "Epoch 31/100\n",
            "100/100 - 13s - loss: 0.5134 - acc: 0.7425 - val_loss: 0.4766 - val_acc: 0.7480\n",
            "Epoch 32/100\n",
            "100/100 - 13s - loss: 0.5066 - acc: 0.7535 - val_loss: 0.5445 - val_acc: 0.7220\n",
            "Epoch 33/100\n",
            "100/100 - 13s - loss: 0.5005 - acc: 0.7535 - val_loss: 0.4800 - val_acc: 0.7490\n",
            "Epoch 34/100\n",
            "100/100 - 13s - loss: 0.5027 - acc: 0.7625 - val_loss: 0.4992 - val_acc: 0.7570\n",
            "Epoch 35/100\n",
            "100/100 - 13s - loss: 0.5069 - acc: 0.7565 - val_loss: 0.5262 - val_acc: 0.7460\n",
            "Epoch 36/100\n",
            "100/100 - 13s - loss: 0.5000 - acc: 0.7560 - val_loss: 0.4814 - val_acc: 0.7500\n",
            "Epoch 37/100\n",
            "100/100 - 13s - loss: 0.4965 - acc: 0.7595 - val_loss: 0.4773 - val_acc: 0.7650\n",
            "Epoch 38/100\n",
            "100/100 - 13s - loss: 0.4830 - acc: 0.7665 - val_loss: 0.4946 - val_acc: 0.7370\n",
            "Epoch 39/100\n",
            "100/100 - 13s - loss: 0.4884 - acc: 0.7635 - val_loss: 0.4844 - val_acc: 0.7500\n",
            "Epoch 40/100\n",
            "100/100 - 13s - loss: 0.4742 - acc: 0.7745 - val_loss: 0.4790 - val_acc: 0.7500\n",
            "Epoch 41/100\n",
            "100/100 - 13s - loss: 0.4752 - acc: 0.7760 - val_loss: 0.4774 - val_acc: 0.7550\n",
            "Epoch 42/100\n",
            "100/100 - 13s - loss: 0.4743 - acc: 0.7655 - val_loss: 0.4838 - val_acc: 0.7700\n",
            "Epoch 43/100\n",
            "100/100 - 13s - loss: 0.4805 - acc: 0.7725 - val_loss: 0.4712 - val_acc: 0.7780\n",
            "Epoch 44/100\n",
            "100/100 - 13s - loss: 0.4706 - acc: 0.7790 - val_loss: 0.4564 - val_acc: 0.7840\n",
            "Epoch 45/100\n",
            "100/100 - 13s - loss: 0.4730 - acc: 0.7785 - val_loss: 0.4546 - val_acc: 0.7890\n",
            "Epoch 46/100\n",
            "100/100 - 13s - loss: 0.4762 - acc: 0.7835 - val_loss: 0.4599 - val_acc: 0.7570\n",
            "Epoch 47/100\n",
            "100/100 - 13s - loss: 0.4656 - acc: 0.7785 - val_loss: 0.4936 - val_acc: 0.7450\n",
            "Epoch 48/100\n",
            "100/100 - 13s - loss: 0.4719 - acc: 0.7685 - val_loss: 0.4557 - val_acc: 0.7910\n",
            "Epoch 49/100\n",
            "100/100 - 13s - loss: 0.4671 - acc: 0.7755 - val_loss: 0.4599 - val_acc: 0.7930\n",
            "Epoch 50/100\n",
            "100/100 - 13s - loss: 0.4559 - acc: 0.7815 - val_loss: 0.4702 - val_acc: 0.7620\n",
            "Epoch 51/100\n",
            "100/100 - 13s - loss: 0.4663 - acc: 0.7725 - val_loss: 0.5053 - val_acc: 0.7500\n",
            "Epoch 52/100\n",
            "100/100 - 13s - loss: 0.4641 - acc: 0.7715 - val_loss: 0.4698 - val_acc: 0.7700\n",
            "Epoch 53/100\n",
            "100/100 - 13s - loss: 0.4549 - acc: 0.7900 - val_loss: 0.4361 - val_acc: 0.7950\n",
            "Epoch 54/100\n",
            "100/100 - 13s - loss: 0.4506 - acc: 0.7835 - val_loss: 0.4389 - val_acc: 0.8000\n",
            "Epoch 55/100\n",
            "100/100 - 13s - loss: 0.4462 - acc: 0.7890 - val_loss: 0.4503 - val_acc: 0.7730\n",
            "Epoch 56/100\n",
            "100/100 - 13s - loss: 0.4399 - acc: 0.7975 - val_loss: 0.4550 - val_acc: 0.7860\n",
            "Epoch 57/100\n",
            "100/100 - 13s - loss: 0.4473 - acc: 0.7905 - val_loss: 0.4416 - val_acc: 0.7980\n",
            "Epoch 58/100\n",
            "100/100 - 13s - loss: 0.4471 - acc: 0.7880 - val_loss: 0.4301 - val_acc: 0.8070\n",
            "Epoch 59/100\n",
            "100/100 - 13s - loss: 0.4310 - acc: 0.8005 - val_loss: 0.4300 - val_acc: 0.7980\n",
            "Epoch 60/100\n",
            "100/100 - 13s - loss: 0.4435 - acc: 0.7960 - val_loss: 0.4549 - val_acc: 0.7950\n",
            "Epoch 61/100\n",
            "100/100 - 13s - loss: 0.4436 - acc: 0.7910 - val_loss: 0.4562 - val_acc: 0.7710\n",
            "Epoch 62/100\n",
            "100/100 - 13s - loss: 0.4402 - acc: 0.7975 - val_loss: 0.4704 - val_acc: 0.7740\n",
            "Epoch 63/100\n",
            "100/100 - 13s - loss: 0.4225 - acc: 0.8125 - val_loss: 0.4337 - val_acc: 0.8020\n",
            "Epoch 64/100\n",
            "100/100 - 13s - loss: 0.4312 - acc: 0.7975 - val_loss: 0.4532 - val_acc: 0.7800\n",
            "Epoch 65/100\n",
            "100/100 - 13s - loss: 0.4314 - acc: 0.8000 - val_loss: 0.4433 - val_acc: 0.7990\n",
            "Epoch 66/100\n",
            "100/100 - 12s - loss: 0.4273 - acc: 0.8055 - val_loss: 0.4204 - val_acc: 0.8030\n",
            "Epoch 67/100\n",
            "100/100 - 12s - loss: 0.4427 - acc: 0.7950 - val_loss: 0.4254 - val_acc: 0.8040\n",
            "Epoch 68/100\n",
            "100/100 - 12s - loss: 0.4229 - acc: 0.8055 - val_loss: 0.4383 - val_acc: 0.7950\n",
            "Epoch 69/100\n",
            "100/100 - 12s - loss: 0.4116 - acc: 0.8150 - val_loss: 0.4749 - val_acc: 0.7670\n",
            "Epoch 70/100\n",
            "100/100 - 13s - loss: 0.4167 - acc: 0.8125 - val_loss: 0.4388 - val_acc: 0.7950\n",
            "Epoch 71/100\n",
            "100/100 - 13s - loss: 0.4216 - acc: 0.8015 - val_loss: 0.4123 - val_acc: 0.8240\n",
            "Epoch 72/100\n",
            "100/100 - 13s - loss: 0.4165 - acc: 0.8075 - val_loss: 0.4087 - val_acc: 0.8060\n",
            "Epoch 73/100\n",
            "100/100 - 13s - loss: 0.4048 - acc: 0.8105 - val_loss: 0.4676 - val_acc: 0.7890\n",
            "Epoch 74/100\n",
            "100/100 - 13s - loss: 0.4071 - acc: 0.8185 - val_loss: 0.4029 - val_acc: 0.8140\n",
            "Epoch 75/100\n",
            "100/100 - 13s - loss: 0.4111 - acc: 0.8205 - val_loss: 0.4629 - val_acc: 0.7800\n",
            "Epoch 76/100\n",
            "100/100 - 13s - loss: 0.4103 - acc: 0.8145 - val_loss: 0.4440 - val_acc: 0.8040\n",
            "Epoch 77/100\n",
            "100/100 - 13s - loss: 0.4101 - acc: 0.8110 - val_loss: 0.4412 - val_acc: 0.7920\n",
            "Epoch 78/100\n",
            "100/100 - 13s - loss: 0.4050 - acc: 0.8095 - val_loss: 0.4516 - val_acc: 0.7890\n",
            "Epoch 79/100\n",
            "100/100 - 13s - loss: 0.4211 - acc: 0.8060 - val_loss: 0.4109 - val_acc: 0.8020\n",
            "Epoch 80/100\n",
            "100/100 - 13s - loss: 0.4120 - acc: 0.8070 - val_loss: 0.3999 - val_acc: 0.8190\n",
            "Epoch 81/100\n",
            "100/100 - 13s - loss: 0.4064 - acc: 0.8215 - val_loss: 0.4246 - val_acc: 0.8010\n",
            "Epoch 82/100\n",
            "100/100 - 13s - loss: 0.4150 - acc: 0.8100 - val_loss: 0.4260 - val_acc: 0.8120\n",
            "Epoch 83/100\n",
            "100/100 - 13s - loss: 0.3988 - acc: 0.8195 - val_loss: 0.4198 - val_acc: 0.8090\n",
            "Epoch 84/100\n",
            "100/100 - 13s - loss: 0.3964 - acc: 0.8265 - val_loss: 0.4204 - val_acc: 0.8080\n",
            "Epoch 85/100\n",
            "100/100 - 13s - loss: 0.3946 - acc: 0.8200 - val_loss: 0.4179 - val_acc: 0.8060\n",
            "Epoch 86/100\n",
            "100/100 - 13s - loss: 0.4067 - acc: 0.8170 - val_loss: 0.4001 - val_acc: 0.8210\n",
            "Epoch 87/100\n",
            "100/100 - 13s - loss: 0.4072 - acc: 0.8135 - val_loss: 0.4360 - val_acc: 0.8010\n",
            "Epoch 88/100\n",
            "100/100 - 13s - loss: 0.4049 - acc: 0.8090 - val_loss: 0.4009 - val_acc: 0.8220\n",
            "Epoch 89/100\n",
            "100/100 - 13s - loss: 0.3992 - acc: 0.8170 - val_loss: 0.4432 - val_acc: 0.7840\n",
            "Epoch 90/100\n",
            "100/100 - 13s - loss: 0.3868 - acc: 0.8280 - val_loss: 0.4135 - val_acc: 0.8140\n",
            "Epoch 91/100\n",
            "100/100 - 13s - loss: 0.3855 - acc: 0.8310 - val_loss: 0.4134 - val_acc: 0.8100\n",
            "Epoch 92/100\n",
            "100/100 - 13s - loss: 0.3789 - acc: 0.8340 - val_loss: 0.4227 - val_acc: 0.8170\n",
            "Epoch 93/100\n",
            "100/100 - 13s - loss: 0.3828 - acc: 0.8295 - val_loss: 0.4429 - val_acc: 0.8120\n",
            "Epoch 94/100\n",
            "100/100 - 13s - loss: 0.3903 - acc: 0.8260 - val_loss: 0.4364 - val_acc: 0.8030\n",
            "Epoch 95/100\n",
            "100/100 - 13s - loss: 0.3718 - acc: 0.8385 - val_loss: 0.4082 - val_acc: 0.8280\n",
            "Epoch 96/100\n",
            "100/100 - 13s - loss: 0.3831 - acc: 0.8305 - val_loss: 0.3946 - val_acc: 0.8240\n",
            "Epoch 97/100\n",
            "100/100 - 13s - loss: 0.3824 - acc: 0.8270 - val_loss: 0.4335 - val_acc: 0.8160\n",
            "Epoch 98/100\n",
            "100/100 - 13s - loss: 0.3835 - acc: 0.8330 - val_loss: 0.3861 - val_acc: 0.8320\n",
            "Epoch 99/100\n",
            "100/100 - 13s - loss: 0.3789 - acc: 0.8330 - val_loss: 0.4047 - val_acc: 0.8250\n",
            "Epoch 100/100\n",
            "100/100 - 13s - loss: 0.3822 - acc: 0.8275 - val_loss: 0.3997 - val_acc: 0.8260\n"
          ]
        }
      ],
      "source": [
        "!wget --no-check-certificate \\\n",
        "    https://storage.googleapis.com/mledu-datasets/cats_and_dogs_filtered.zip \\\n",
        "    -O /tmp/cats_and_dogs_filtered.zip\n",
        "  \n",
        "import os\n",
        "import zipfile\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras.optimizers import RMSprop\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "\n",
        "local_zip = '/tmp/cats_and_dogs_filtered.zip'\n",
        "zip_ref = zipfile.ZipFile(local_zip, 'r')\n",
        "zip_ref.extractall('/tmp')\n",
        "zip_ref.close()\n",
        "\n",
        "base_dir = '/tmp/cats_and_dogs_filtered'\n",
        "train_dir = os.path.join(base_dir, 'train')\n",
        "validation_dir = os.path.join(base_dir, 'validation')\n",
        "\n",
        "# Directory with our training cat pictures\n",
        "train_cats_dir = os.path.join(train_dir, 'cats')\n",
        "\n",
        "# Directory with our training dog pictures\n",
        "train_dogs_dir = os.path.join(train_dir, 'dogs')\n",
        "\n",
        "# Directory with our validation cat pictures\n",
        "validation_cats_dir = os.path.join(validation_dir, 'cats')\n",
        "\n",
        "# Directory with our validation dog pictures\n",
        "validation_dogs_dir = os.path.join(validation_dir, 'dogs')\n",
        "\n",
        "model = tf.keras.models.Sequential([\n",
        "    tf.keras.layers.Conv2D(32, (3,3), activation='relu', input_shape=(150, 150, 3)),\n",
        "    tf.keras.layers.MaxPooling2D(2, 2),\n",
        "    tf.keras.layers.Conv2D(64, (3,3), activation='relu'),\n",
        "    tf.keras.layers.MaxPooling2D(2,2),\n",
        "    tf.keras.layers.Conv2D(128, (3,3), activation='relu'),\n",
        "    tf.keras.layers.MaxPooling2D(2,2),\n",
        "    tf.keras.layers.Conv2D(128, (3,3), activation='relu'),\n",
        "    tf.keras.layers.MaxPooling2D(2,2),\n",
        "    tf.keras.layers.Dropout(0.5),\n",
        "    tf.keras.layers.Flatten(),\n",
        "    tf.keras.layers.Dense(512, activation='relu'),\n",
        "    tf.keras.layers.Dense(1, activation='sigmoid')\n",
        "])\n",
        "\n",
        "model.compile(loss='binary_crossentropy',\n",
        "              optimizer=RMSprop(lr=1e-4),\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "# This code has changed. Now instead of the ImageGenerator just rescaling\n",
        "# the image, we also rotate and do other operations\n",
        "# Updated to do image augmentation\n",
        "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",
        "      fill_mode='nearest')\n",
        "\n",
        "test_datagen = ImageDataGenerator(rescale=1./255)\n",
        "\n",
        "# Flow training images in batches of 20 using train_datagen generator\n",
        "train_generator = train_datagen.flow_from_directory(\n",
        "        train_dir,  # This is the source directory for training images\n",
        "        target_size=(150, 150),  # All images will be resized to 150x150\n",
        "        batch_size=20,\n",
        "        # Since we use binary_crossentropy loss, we need binary labels\n",
        "        class_mode='binary')\n",
        "\n",
        "# Flow validation images in batches of 20 using test_datagen generator\n",
        "validation_generator = test_datagen.flow_from_directory(\n",
        "        validation_dir,\n",
        "        target_size=(150, 150),\n",
        "        batch_size=20,\n",
        "        class_mode='binary')\n",
        "\n",
        "history = model.fit(\n",
        "      train_generator,\n",
        "      steps_per_epoch=100,  # 2000 images = batch_size * steps\n",
        "      epochs=100,\n",
        "      validation_data=validation_generator,\n",
        "      validation_steps=50,  # 1000 images = batch_size * steps\n",
        "      verbose=2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 0,
      "metadata": {
        "colab": {
          "height": 561
        },
        "colab_type": "code",
        "id": "I3yFv2Jpb2Dl",
        "outputId": "fc13860a-d65f-4836-ad41-79b8c45588eb"
      },
      "outputs": [
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYoAAAEQCAYAAACugzM1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f3d81a57bd0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYoAAAEQCAYAAACugzM1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nijAUAoFAIOiU/w+MwYsGLuyCGgAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f3d7edc5d50>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "acc = history.history['accuracy']\n",
        "val_acc = history.history['val_accuracy']\n",
        "loss = history.history['loss']\n",
        "val_loss = history.history['val_loss']\n",
        "\n",
        "epochs = range(len(acc))\n",
        "\n",
        "plt.plot(epochs, acc, 'bo', label='Training accuracy')\n",
        "plt.plot(epochs, val_acc, 'b', label='Validation accuracy')\n",
        "plt.title('Training and validation accuracy')\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()"
      ]
    }
  ],
  "metadata": {
    "colab": {
      "collapsed_sections": [],
      "include_colab_link": true,
      "name": "Course 2 - Part 4 - Lesson 2 - Notebook (Cats v Dogs Augmentation).ipynb",
      "provenance": [],
      "toc_visible": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}