{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this example we will take 2D images from a cardiac Left Ventricle short axis stack.\n",
    "These images have been segmented by an expert to give us our ground truth labels.\n",
    "The data is from the open-source Sunnybrook Cardiac Data set.\n",
    "\n",
    "We will read in the data pairs, train a 2D UNet to perform segmentation and then test the resulting network"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Set up TensorFlow MRi\n",
    "If you have not yet installed TensorFlow MRI in your environment you may need to do so now\n",
    "using `pip`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[33mWARNING: You are using pip version 20.2.4; however, version 24.3.1 is available.\n",
      "You should consider upgrading via the '/usr/local/bin/python -m pip install --upgrade pip' command.\u001b[0m\n",
      "Note: you may need to restart the kernel to use updated packages.\n",
      "\u001b[33mWARNING: You are using pip version 20.2.4; however, version 24.3.1 is available.\n",
      "You should consider upgrading via the '/usr/local/bin/python -m pip install --upgrade pip' command.\u001b[0m\n",
      "Note: you may need to restart the kernel to use updated packages.\n",
      "Defaulting to user installation because normal site-packages is not writeable\n",
      "Requirement already satisfied: pydicom in /usr/local/lib/python3.8/dist-packages (2.4.4)\n",
      "\u001b[33mWARNING: You are using pip version 20.2.4; however, version 24.3.1 is available.\n",
      "You should consider upgrading via the '/usr/local/bin/python -m pip install --upgrade pip' command.\u001b[0m\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "%pip install --quiet tensorflow-mri\n",
    "# Upgrade Matplotlib. Versions older than 3.5.x may cause an error below.\n",
    "%pip install --quiet --upgrade matplotlib\n",
    "\n",
    "%pip install pydicom"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, import all necessary python libraries."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2025-01-17 23:08:36.707715: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 AVX512F AVX512_VNNI FMA\n",
      "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
      "2025-01-17 23:08:36.803401: I tensorflow/core/util/util.cc:169] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n",
      "2025-01-17 23:08:36.828296: E tensorflow/stream_executor/cuda/cuda_blas.cc:2981] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n"
     ]
    }
   ],
   "source": [
    "import pydicom as dicom\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.image as mpimg\n",
    "import numpy as np\n",
    "from glob import glob\n",
    "import tensorflow as tf\n",
    "import os\n",
    "import random as rng\n",
    "\n",
    "import tensorflow_mri as tfmri\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Using a GPU \n",
    "TensorFlow MRI supports CPU and GPU computation. If there is a GPU available in your environment and it is visible to TensorFlow, it will be used automatically.\n",
    "\n",
    ":::{tip}\n",
    "In Google Colab, you can enable GPU computation by clicking on\n",
    "**Runtime > Change runtime type** and selecting **GPU** under\n",
    "**Hardware accelerator**.\n",
    ":::\n",
    "\n",
    ":::{tip}\n",
    "You can control whether CPU or GPU is used for a particular operation via\n",
    "the [`tf.device`](https://www.tensorflow.org/api_docs/python/tf/device)\n",
    "context manager.\n",
    ":::"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Prepare the Data\n",
    "First, we need to extract our data and store our image and mask pairs in lists, so that pairs have the same index. Our path names may differ\n",
    "so ensure you the given paths with the path where your data is stored. Depending on your data set, you may need different functions to read\n",
    "your images and masks. In this case, the images are stored as dicoms and the masks as pngs.\n",
    "\n",
    "In this example, we will be using a heart data set from Sunnybrook Cardiac Data, which you can download from https://www.cardiacatlas.org/sunnybrook-cardiac-data/\n",
    "This is Cardiac Left Ventricle Segmentation from Cine-MRI Images dataset\n",
    "\n",
    "This same data set in a easier formate can be downloaded using Matlab, by copying the following code.\n",
    "\n",
    "`zipFile = matlab.internal.examples.downloadSupportFile(\"medical\",\"CardiacMRI.zip\");`\n",
    "\n",
    "`filepath = fileparts(zipFile);`\n",
    "\n",
    "`unzip(zipFile,filepath)`\n",
    "\n",
    "We use the data in this format in the following example.\n",
    "\n",
    "The matlab subset of SunnyBrook data set contains of 45 cine short-axis stacks and their corresponding ground truth label images (LV blood pool segmented as peak systole and peak diastole). The MRI images were acquired from multiple patients with various cardiac pathologies. The ground truth label images were manually drawn by experts [ Radau, Perry, Yingli Lu, Kim Connelly, Gideon Paul, Alexander J Dick, and Graham A Wright. “Evaluation Framework for Algorithms Segmenting Short Axis Cardiac MRI.” The MIDAS Journal, July 9, 2009. https://doi.org/10.54294/g80ruo. ]\n",
    "\n",
    "As each cine stack consists of ~10 slices, and there are 2 time points segmented per stack (systole and diastole), this results in 805 dicom images of size 256x256, which each have a corresponding mask (stored as a png) which contains two classes: 0 (\"Background\") and 1 (\"Left Ventricle\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pathlib\n",
    "import gdown\n",
    "\n",
    "url = 'https://drive.google.com/uc?id=1y176qfBc_0-_kFetYsWpWrBf8vXpHb4Y'\n",
    "output = '/tools/docs/tutorials/segment/CardiacMRI.zip'\n",
    "gdown.download(url, output, quiet=False)\n",
    "\n",
    "import zipfile\n",
    "with zipfile.ZipFile(\"/tools/docs/tutorials/segment/exampleRawDataCine.hdf5.zip\",\"r\") as zip_ref:\n",
    "    zip_ref.extractall(\"/tools/docs/tutorials/segment/\")\n",
    "\n",
    "input_file =pathlib.Path('/tools/docs/tutorials/segment/exampleRawDataCine.hdf5')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Create empty lists to store the image and mask data\n",
    "images = []\n",
    "masks = []\n",
    "\n",
    "# This path will need updating for your system\n",
    "path_for_data = '/workspaces/Tutorials/2Dsegmentation_DONE/data/images/**/*.dcm'\n",
    "\n",
    "#Obtain all image and mask data, may need to check the image and mask pathnames to see how they differ\n",
    "for filename in glob(path_for_data)  :     # This will select all files ending with dcm - these should be your images\n",
    "    \n",
    "    #Now we must find the matching mask data\n",
    "    # The dcm data is stored like this: \n",
    "    #   SC-HF-I-08_rawdcm_009.dcm \n",
    "    # and the corresponding mask will have name: \n",
    "    #   SC-HF-I-08gtmask0009.png \n",
    "\n",
    "    mask_filename = filename.replace(\"images\",\"labels\")                #Replacing the different words in pathnames\n",
    "    mask_filename_mask = mask_filename.replace(\"_rawdcm_\",\"gtmask0\")\n",
    "    mask_filename_png = mask_filename_mask.replace(\"dcm\",\"png\")        #replacing file type \n",
    "    mask_exists = os.path.exists(mask_filename_png)                    #Checking that this path exists\n",
    "   \n",
    "    #If this path exists, we append the images and masks lists so that corresponding images and paths have the same index\n",
    "   \n",
    "    if mask_exists:                                                   \n",
    "        image = dicom.dcmread(filename).pixel_array\n",
    "        images.append(image)\n",
    "        mask = mpimg.imread(mask_filename_png)\n",
    "        masks.append(mask)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we will check if we have assigned the correct pairs of images and masks by plotting them on top of each other. Image cmap convention is 'gray' in MRI, and masks cmap is often 'viridis'. Alpha determines how in focus the mask is (range 0-1, 0.2 is a good choice)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Image With Mask')"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#testing blended images and masks\n",
    "plt.imshow(images[0],cmap='gray')               \n",
    "plt.imshow(masks[0], cmap='viridis',alpha=0.2)  \n",
    "plt.title(\"Image With Mask\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The shape of the masks is: (805, 256, 256)\n",
      "The shape of the images is: (805, 256, 256)\n"
     ]
    }
   ],
   "source": [
    "#converting to numpy arrays as they are easier to work with\n",
    "masks = np.array(masks)\n",
    "images = np.array(images)\n",
    "\n",
    "# You should check the shape of the data sets.\n",
    "print(f'The shape of the masks is: {masks.shape}')\n",
    "print(f'The shape of the images is: {images.shape}')\n",
    "\n",
    "\n",
    "#The shape of the masks is: (805, 256, 256)\n",
    "#The shape of the images is: (805, 256, 256)\n",
    "# This means that there are 805 data pairs, of images of size 256x256"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From this we can see we have 805 images and masks, with image size 256x256, as we expect.\n",
    "\n",
    "Next, we need to normalize the image data to so that each data point lies between 0-1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def normalize_images(images):\n",
    "    \"\"\"Normalisation function\n",
    "    Input: \n",
    "           - images, array of image data\n",
    "    Output: \n",
    "           - normalised array of image data\"\"\"\n",
    "    \n",
    "    return (images - np.min(images)) / (np.max(images) - np.min(images))\n",
    "\n",
    "#Creating new normalised array\n",
    "images = normalize_images(images)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can quickly check that the normalized images array will still give us the MRI Images."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fb3289010d0>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Testing normalised images\n",
    "plt.imshow(images[0],cmap='gray')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For a 2D Unet model, we need our mask data to be seperated in two classes, in this case background and bloodpool. All data points equal to 0 we can keep as zero, and all non zero points we will make equal to 1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "#simplifying masks\n",
    "def simplify_mask(masks):\n",
    "    \"\"\"This function take an inputed array of masks and keeps leaves all zero data points as zero \n",
    "    and converts all non-zero data points equal to one. Zero data point are equivalent to the background\n",
    "    and those equal to one represent the bloodpool\n",
    "    \n",
    "    Input: masks, array containing mask data\n",
    "     \n",
    "    Output: simp_masks, array containing the simplified masks\"\"\"\n",
    "\n",
    "    simp_masks = np.zeros_like(masks, dtype=np.float32)\n",
    "    simp_masks[masks ==0] = 0        #background\n",
    "    simp_masks[masks !=0] = 1        #bloodpool\n",
    "    \n",
    "    return simp_masks"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can now apply this function to our `masks` array, and do a quick check of the minimum and maximum values (which should be 0 and 1 respectively), and plot one of the simplified masks."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(min, max): (0.0, 1.0)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAa4AAAGiCAYAAAC/NyLhAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuNSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/xnp5ZAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAl70lEQVR4nO3de3TU9Z3/8dfkNhAgE0NIJpGLAeUmFxEwZkVKS5YEEVHoWaHUgqWw0sRfMag0/VmQtr/GYtd2tSg/d13QXfDCrshKld9iMKGUgBBlEZAUKBoQJig0MwRMyOXz+8Nl6ki4JCT58sk8H+fMOWS+35l5z+dkfDoz35m4jDFGAABYIsLpAQAAaArCBQCwCuECAFiFcAEArEK4AABWIVwAAKsQLgCAVQgXAMAqhAsAYBXCBQCwimPhWrp0qa677jp16NBB6enpeu+995waBQBgEUfC9eqrryovL0+LFi3S+++/r6FDhyorK0vHjx93YhwAgEVcTnzJbnp6ukaOHKnf/e53kqSGhgb16NFDDz74oH784x+39TgAAItEtfUNnj17VqWlpcrPzw+eFxERoczMTJWUlDR6mZqaGtXU1AR/bmho0MmTJ9W1a1e5XK5WnxkA0LKMMTp16pRSU1MVEdG0F//aPFyff/656uvrlZycHHJ+cnKy9u3b1+hlCgoKtHjx4rYYDwDQhg4fPqzu3bs36TJtHq7myM/PV15eXvBnv9+vnj17apTuUJSiHZwMANAcdarVZr2lLl26NPmybR6uxMRERUZGqqKiIuT8iooKeb3eRi/jdrvldrvPOz9K0YpyES4AsM7/HF3RnLd72vyowpiYGA0fPlyFhYXB8xoaGlRYWKiMjIy2HgcAYBlHXirMy8vTjBkzNGLECN1yyy367W9/q9OnT+v+++93YhwAgEUcCde9996rzz77TAsXLpTP59NNN92k9evXn3fABgAAX+fI57iuVCAQkMfj0RhN4j0uALBQnalVkdbK7/crLi6uSZfluwoBAFYhXAAAqxAuAIBVCBcAwCqECwBgFcIFALAK4QIAWIVwAQCsQrgAAFYhXAAAqxAuAIBVCBcAwCqECwBgFcIFALAK4QIAWIVwAQCsQrgAAFYhXAAAqxAuAIBVCBcAwCqECwBgFcIFALAK4QIAWIVwAQCsQrgAAFYhXAAAqxAuAIBVCBcAwCqECwBgFcIFALAK4QIAWIVwAQCsQrgAAFYhXAAAqxAuAIBVCBcAwCqECwBgFcIFALAK4QIAWIVwAQCsQrgAAFYhXAAAqxAuAIBVCBcAwCqECwBgFcIFALAK4QIAWIVwAQCsQrgAAFYhXAAAqxAuAIBVCBcAwCqECwBgFcIFALAK4QIAWIVwAQCsQrgAAFYhXAAAqxAuAIBVCBcAwCqECwBglRYP1+OPPy6XyxVy6t+/f3B7dXW1cnJy1LVrV3Xu3FlTpkxRRUVFS48BAGinWuUZ14033qhjx44FT5s3bw5ue+ihh/Tmm29q9erVKi4u1tGjRzV58uTWGAMA0A5FtcqVRkXJ6/Wed77f79cLL7ygVatW6Vvf+pYkafny5RowYIC2bt2qW2+9tTXGAQC0I63yjGv//v1KTU1V7969NX36dJWXl0uSSktLVVtbq8zMzOC+/fv3V8+ePVVSUtIaowAA2pkWf8aVnp6uFStWqF+/fjp27JgWL16s22+/Xbt375bP51NMTIzi4+NDLpOcnCyfz3fB66ypqVFNTU3w50Ag0NJjAwAs0eLhGj9+fPDfQ4YMUXp6unr16qXXXntNHTt2bNZ1FhQUaPHixS01IgDAYq1+OHx8fLz69u2rAwcOyOv16uzZs6qsrAzZp6KiotH3xM7Jz8+X3+8Png4fPtzKUwMArlatHq6qqiodPHhQKSkpGj58uKKjo1VYWBjcXlZWpvLycmVkZFzwOtxut+Li4kJOAIDw1OIvFT788MOaOHGievXqpaNHj2rRokWKjIzUtGnT5PF4NGvWLOXl5SkhIUFxcXF68MEHlZGRwRGFAIDL0uLhOnLkiKZNm6YTJ06oW7duGjVqlLZu3apu3bpJkn7zm98oIiJCU6ZMUU1NjbKysvTss8+29BgAgHbKZYwxTg/RVIFAQB6PR2M0SVGuaKfHAQA0UZ2pVZHWyu/3N/ntH76rEABgFcIFALAK4QIAWIVwAQCsQrgAAFYhXAAAqxAuAIBVCBcAwCqECwBgFcIFALAK4QIAWIVwAQCsQrgAAFYhXAAAqxAuAIBVCBcAwCqECwBgFcIFALAK4QIAWIVwAQCsQrgAAFYhXAAAqxAuAIBVCBcAwCqECwBgFcIFALAK4QIAWIVwAQCsQrgAAFYhXAAAqxAuAIBVCBcAwCqECwBgFcIFALAK4QIAWIVwAQCsQrgAAFYhXAAAqxAuAIBVCBcAwCqECwBgFcIFALAK4QIAWIVwAQCsQrgAAFYhXAAAqxAuAIBVopweALDdn5YP1wMjNl1w+5+/SNThzGjVBwJtOBXQfhEuoAkiOnRQxcxhaoh2Bc+beXORFnTdf8HLHKv7QOP+/lFFVkvRVUYJK7ZKxrTFuEC7RLiAy+SKilKEN0n/+ZMn1T2q82VfLiWqsz586FlJ0mtVHq34j6EyZ8/KGCNTU9Na4wLtFuECLtPhR2/R8z/4nVIiY5t9HZM6fS7vzk1qMBH6Px9PUMTYwy04IRAeCBdwESZjqA7d82Wobkw/qNs6XNnxTG5XtEZ3kKQGBXoWav6SGeq77Jjq/vzxFc8KhAvCBVxAZHKSykd30v7vPtsq139XpzO667vP6bbtfy+PP6D6Eydb5XaA9obD4YELuP73f9EfH/yHVr+dtU89pU9fSG712wHaC8IFfMXh//03OrjqJv151U36ftfN8kR0bPXbTIzspPwB63Xg34YpKsXb6rcH2I6XCgFJLrdbGnS9Rty5Wy/1OveZLHeb3f7ULn/RPd98XpkZuYrbHqW6w0fa7LYB2/CMC5Ckgdfr9bXLvxKttud2Ravomee0d2GKYzMANiBcCHuHH/sbXfvcJ4qNiHF6FEW6IvTrb7yqwNt9FNGli9PjAFclwoWw5YqOUfWdtyhlzBG90HOz0+METekc0OobX1TlxBsVeUNvp8cBrjqEC2ErMilRrzz7lAoH/qfTo5yne1Rnlfx6mQ5N42AN4OsIF3AV+8fv/ZMi302VIiKdHgW4ajQ5XJs2bdLEiROVmpoql8ulN954I2S7MUYLFy5USkqKOnbsqMzMTO3fH/oFpCdPntT06dMVFxen+Ph4zZo1S1VVVVd0R4CmMBlD9cl916mD6+r+f7dxsbV6/Lq1qshNV2S/650eB7gqNPlRe/r0aQ0dOlRLly5tdPuSJUv09NNPa9myZdq2bZs6deqkrKwsVVdXB/eZPn269uzZow0bNmjdunXatGmT5syZ0/x7ATSBy+3WnyfHavf/elbXXMH3DraVW9zR2vnjZ1UxpptcbveXh+67XJe+INBOuYxp/t9XcLlcWrNmje6++25JXz7bSk1N1fz58/Xwww9Lkvx+v5KTk7VixQpNnTpVH330kQYOHKjt27drxIgRkqT169frjjvu0JEjR5SamnrJ2w0EAvJ4PBqjSYpyRTd3fIQhl9utEdvO6P5rStQn+vK/4f1qsOtstXx1Xx5puOQH9ymy6H1nBwKuQJ2pVZHWyu/3Ky4urkmXbdHXSQ4dOiSfz6fMzMzgeR6PR+np6SopKZEklZSUKD4+PhgtScrMzFRERIS2bdvWkuMAjRrdeZ910ZKkITEdNC62VuNia3VoltHJ72c4PRLgiBYNl8/nkyQlJ4d+71pycnJwm8/nU1JSUsj2qKgoJSQkBPf5upqaGgUCgZAT0FQRHTooMtWrGFe906NcsYNjlyt++pEvvyKKAzcQZq7ud6b/R0FBgTweT/DUo0cPp0eChY7PGKZ/Ll6p2zvUOT1Ki1jX/3U9vfXfFdmPz3ohvLRouLzeLz9zUlFREXJ+RUVFcJvX69Xx48dDttfV1enkyZPBfb4uPz9ffr8/eDp8mD++h6ZriHIpJaqzIq/yIwkvl9sVreuiYvVpQZROzOZlQ4SPFn0Ep6Wlyev1qrCwMHheIBDQtm3blJHx5QMrIyNDlZWVKi0tDe6zceNGNTQ0KD09vdHrdbvdiouLCzkB+PIrov77lpf1RXZAEUMH8LIhwkKTvx2+qqpKBw4cCP586NAh7dy5UwkJCerZs6fmzZunX/ziF7rhhhuUlpamn/70p0pNTQ0eeThgwABlZ2dr9uzZWrZsmWpra5Wbm6upU6de1hGFAM63J2OlPlp7RvNHTFT95yecHgdoVU0O144dO/TNb34z+HNeXp4kacaMGVqxYoUeffRRnT59WnPmzFFlZaVGjRql9evXq0OHDsHLrFy5Urm5uRo7dqwiIiI0ZcoUPf300y1wd4DG7V8xXDOGFTk9RqvqFRWlmP+I0qcrMpSwvMTpcYBWc0Wf43IKn+NCU92+q1qPJe5zeow20XvD93Xdv7oU/U7ppXcGHHLVfI4LgPP+/Lf/ov/13KtyRTv/Z1qA1kC4gHYos+PnGlUa0OlvN37AE2AzwgW0Q50jOuixxH36dHy9Tt17q9PjAC2KcAHt2KHx/6zbFvBVamhfCBcAwCqEC2jnxsR9pEO/zFDk9WlOjwK0CMIFtHMTYqv1p5nPqXJ4siLjPU6PA1wxwgWEiTd+/Q+q+NfkS+8IXOUIFxAmkiI7qYv7rNNjAFeMcAFhJLFjlSKGDuDDybAa4QLCyL/3eUerf79cLv6GFyxGuIAwE+2K5JEPq/HrC4SZCEXo+K3XKGJQf6dHAZqFcAFhJtoVqdLHn9O++Z2cHgVoFsKFsPCHObfo+pVznR4DQAsgXAgPW3epy8cup6cA0AIIFwDAKoQLAGAVwoWw0XVPtdJ+P1vldVVOj3JVGHfjXpUv/BtFxMY6PQrQJIQLYSOy6H31y92lfWevcXqUq8L/7V6it2YtkasTRxfCLoQLAGAVwoWwYmrr9PCzs5X10Z1OjwKgmQgXwktDvVKe2qI/b+/h9CQAmolwAQCsQrgQlvqsDuimJ36ov9SfcXoUAE1EuBCWTOkepa4qU075BP2xusHpcRxR9EWEFhyeJNXVOT0K0CSEC2Gr/vMTOnHbXzSjZJbTozji/k33yz/qhOr/8henRwGahHABAKxCuBD2Ov53R921P9vpMQBcJsKFsJe6ZIvOzuuqWlPv9CgALgPhAgBYhXABkiI/q1Tf3z+g16o8To/S6upNg/pvvk+Jm2KcHgVoFsIFSKo78qn6ztmuFw7f7vQora5O9Ur7Ra0Slpc4PQrQLIQLAGAVwgV8ReWLPZT2/9rv57qePNlHw5b9SK6jnzk9CtBshAv4iviXStT736TXqjzyN3zh9Dgt7vdHB6vHz7eo/vMTTo8CNBvhAr4mqrBU/zLwBj1zcpjTowBoRJTTAwBXI1NXp3rTvv6/ru9Lc3VtUZ2kj50eBbgi7euRCbSg1QeH6aVAotNjXLEjdVWaf+xmXfefZxSzfrvT4wBXjHABF3Dt5D36zdN/5/QYV+wfP79du4c3yLXlv50eBWgRvFQItGN9X5qr69aekUtEC+0Hz7iAi4grr9PoD+/R8frTTo/SJMfqqjRq12RdW1QnVwnRQvtCuICL6LDuPcXedVRbqpOt+RLeWlOv4i96qPOd5bynhXaJcAGXYGpq9PyELPVdO9fpUS5Lv//4oV6a+C0Z/rIx2inCBVyG+j8dVHRlpNNjXJS/4Qv12Xi/um80qv/TQafHAVoNB2cAlymyxqVDtVVKi+7s9Cjn8Td8oW3Vceo3r5xvxUC7xzMu4DJd9+RO5Yz/vo7UVTk9ynluLvqhfnvraKKFsMAzLuAyNZw5o4iPj2jc84/qW5NK9btrtzk6z69O3KDlr/+tJOnaD+qIFsIG4QKaoOHMGfX4xRa9lZSuwgk7NLajM0cabqqW/mnXKPVZtMWR2wecxEuFQDPcMG+Hnrx3mmOHyC/Mna0+9+1y5LYBp/GMC2iOhnpFHDisjIW5kkvy95X2f/e5VrmprdX1ynkiV66Gv56X/MEh1TXY8bkyoKURLqCZ6iv96vpCiSTpmlE36ZHML/8Myk2dyjW9S/Pfb9pULb1Z+dc/qbL9815K/Of3pK+Eik9oIZwRLqAFRGzeqV03f/nvjX//d5q+qPnPvmZs/IH6/mBH8Gc3f4YECEG4gBbmXfeJvvnJ7GZffkC5X7wICFwY4QJaWN2nRxXz6dFmX55oARfHUYUAAKsQLgCAVQgXAMAqhAsAYBXCBQCwCuECAFiFcAEArNLkcG3atEkTJ05UamqqXC6X3njjjZDtM2fOlMvlCjllZ2eH7HPy5ElNnz5dcXFxio+P16xZs1RVdfX9jSMAwNWnyeE6ffq0hg4dqqVLl15wn+zsbB07dix4evnll0O2T58+XXv27NGGDRu0bt06bdq0SXPmzGn69ACAsNPkb84YP368xo8ff9F93G63vF5vo9s++ugjrV+/Xtu3b9eIESMkSc8884zuuOMO/frXv1ZqampTRwIAhJFWeY+rqKhISUlJ6tevn+bOnasTJ/76TdklJSWKj48PRkuSMjMzFRERoW3bGv+LsjU1NQoEAiEnAEB4avFwZWdn66WXXlJhYaF+9atfqbi4WOPHj1d9/ZffwObz+ZSUlBRymaioKCUkJMjn8zV6nQUFBfJ4PMFTjx49WnpsAIAlWvxLdqdOnRr89+DBgzVkyBD16dNHRUVFGjt2bLOuMz8/X3l5ecGfA4EA8QKAMNXqh8P37t1biYmJOnDggCTJ6/Xq+PHjIfvU1dXp5MmTF3xfzO12Ky4uLuQEAAhPrR6uI0eO6MSJE0pJSZEkZWRkqLKyUqWlpcF9Nm7cqIaGBqWnp7f2OAAAyzX5pcKqqqrgsydJOnTokHbu3KmEhAQlJCRo8eLFmjJlirxerw4ePKhHH31U119/vbKysiRJAwYMUHZ2tmbPnq1ly5aptrZWubm5mjp1KkcUAgAuqcnPuHbs2KFhw4Zp2LBhkqS8vDwNGzZMCxcuVGRkpHbt2qW77rpLffv21axZszR8+HD94Q9/kNvtDl7HypUr1b9/f40dO1Z33HGHRo0apeeff77l7hUAoN1yGWOM00M0VSAQkMfj0RhNUpQr2ulxAABNVGdqVaS18vv9TT5uge8qBABYhXABAKxCuAAAViFcAACrEC4AgFUIFwDAKoQLAGAVwgUAsArhAgBYhXABAKxCuAAAViFcAACrEC4AgFUIFwDAKoQLAGAVwgUAsArhAgBYhXABAKxCuAAAViFcAACrEC4AgFUIFwDAKoQLAGAVwgUAsArhAgBYhXABAKxCuAAAViFcAACrEC4AgFUIFwDAKoQLAGAVwgUAsArhAgBYhXABAKxCuAAAViFcAACrEC4AgFUIFwDAKoQLAGAVwgUAsArhAgBYhXABAKxCuAAAViFcAACrEC4AgFUIFwDAKoQLAGAVwgUAsArhAgBYhXABAKxCuAAAViFcAACrEC4AgFUIFwDAKoQLAGAVwgUAsArhAgBYhXABAKxCuAAAViFcAACrEC4AgFWaFK6CggKNHDlSXbp0UVJSku6++26VlZWF7FNdXa2cnBx17dpVnTt31pQpU1RRURGyT3l5uSZMmKDY2FglJSXpkUceUV1d3ZXfGwBAu9ekcBUXFysnJ0dbt27Vhg0bVFtbq3Hjxun06dPBfR566CG9+eabWr16tYqLi3X06FFNnjw5uL2+vl4TJkzQ2bNntWXLFr344otasWKFFi5c2HL3CgDQbrmMMaa5F/7ss8+UlJSk4uJijR49Wn6/X926ddOqVav07W9/W5K0b98+DRgwQCUlJbr11lv19ttv684779TRo0eVnJwsSVq2bJkWLFigzz77TDExMZe83UAgII/HozGapChXdHPHBwA4pM7Uqkhr5ff7FRcX16TLXtF7XH6/X5KUkJAgSSotLVVtba0yMzOD+/Tv3189e/ZUSUmJJKmkpESDBw8ORkuSsrKyFAgEtGfPnkZvp6amRoFAIOQEAAhPzQ5XQ0OD5s2bp9tuu02DBg2SJPl8PsXExCg+Pj5k3+TkZPl8vuA+X43Wue3ntjWmoKBAHo8neOrRo0dzxwYAWK7Z4crJydHu3bv1yiuvtOQ8jcrPz5ff7w+eDh8+3Oq3CQC4OkU150K5ublat26dNm3apO7duwfP93q9Onv2rCorK0OedVVUVMjr9Qb3ee+990Ku79xRh+f2+Tq32y23292cUQEA7UyTnnEZY5Sbm6s1a9Zo48aNSktLC9k+fPhwRUdHq7CwMHheWVmZysvLlZGRIUnKyMjQhx9+qOPHjwf32bBhg+Li4jRw4MAruS8AgDDQpGdcOTk5WrVqldauXasuXboE35PyeDzq2LGjPB6PZs2apby8PCUkJCguLk4PPvigMjIydOutt0qSxo0bp4EDB+q+++7TkiVL5PP59NhjjyknJ4dnVQCAS2rS4fAul6vR85cvX66ZM2dK+vIDyPPnz9fLL7+smpoaZWVl6dlnnw15GfCTTz7R3LlzVVRUpE6dOmnGjBl64oknFBV1eR3lcHgAsNuVHA5/RZ/jcgrhAgC7OfY5LgAA2hrhAgBYhXABAKxCuAAAViFcAACrEC4AgFUIFwDAKoQLAGAVwgUAsArhAgBYhXABAKxCuAAAViFcAACrEC4AgFUIFwDAKoQLAGAVwgUAsArhAgBYhXABAKxCuAAAViFcAACrEC4AgFUIFwDAKoQLAGAVwgUAsArhAgBYhXABAKxCuAAAViFcAACrEC4AgFUIFwDAKoQLAGAVwgUAsArhAgBYhXABAKxCuAAAViFcAACrEC4AgFUIFwDAKoQLAGAVwgUAsArhAgBYhXABAKxCuAAAViFcAACrEC4AgFUIFwDAKoQLAGAVwgUAsArhAgBYhXABAKxCuAAAViFcAACrEC4AgFUIFwDAKoQLAGAVwgUAsArhAgBYhXABAKxCuAAAViFcAACrNClcBQUFGjlypLp06aKkpCTdfffdKisrC9lnzJgxcrlcIacHHnggZJ/y8nJNmDBBsbGxSkpK0iOPPKK6urorvzcAgHYvqik7FxcXKycnRyNHjlRdXZ1+8pOfaNy4cdq7d686deoU3G/27Nn62c9+Fvw5NjY2+O/6+npNmDBBXq9XW7Zs0bFjx/S9731P0dHR+uUvf9kCdwkA0J41KVzr168P+XnFihVKSkpSaWmpRo8eHTw/NjZWXq+30ev4r//6L+3du1fvvPOOkpOTddNNN+nnP/+5FixYoMcff1wxMTHNuBsAgHBxRe9x+f1+SVJCQkLI+StXrlRiYqIGDRqk/Px8nTlzJritpKREgwcPVnJycvC8rKwsBQIB7dmzp9HbqampUSAQCDkBAMJTk55xfVVDQ4PmzZun2267TYMGDQqe/53vfEe9evVSamqqdu3apQULFqisrEyvv/66JMnn84VES1LwZ5/P1+htFRQUaPHixc0dFQDQjjQ7XDk5Odq9e7c2b94ccv6cOXOC/x48eLBSUlI0duxYHTx4UH369GnWbeXn5ysvLy/4cyAQUI8ePZo3OADAas16qTA3N1fr1q3Tu+++q+7du1903/T0dEnSgQMHJEler1cVFRUh+5z7+ULvi7ndbsXFxYWcAADhqUnhMsYoNzdXa9as0caNG5WWlnbJy+zcuVOSlJKSIknKyMjQhx9+qOPHjwf32bBhg+Li4jRw4MCmjAMACENNeqkwJydHq1at0tq1a9WlS5fge1Iej0cdO3bUwYMHtWrVKt1xxx3q2rWrdu3apYceekijR4/WkCFDJEnjxo3TwIEDdd9992nJkiXy+Xx67LHHlJOTI7fb3fL3EADQrriMMeayd3a5Gj1/+fLlmjlzpg4fPqzvfve72r17t06fPq0ePXronnvu0WOPPRby8t4nn3yiuXPnqqioSJ06ddKMGTP0xBNPKCrq8joaCATk8Xg0RpMU5Yq+3PEBAFeJOlOrIq2V3+9v8ts/TQrX1YJwAYDdriRczT6q0EnnWlunWsm67AIA6lQr6a//PW8KK8N16tQpSdJmveXwJACAK3Hq1Cl5PJ4mXcbKlwobGhpUVlamgQMH6vDhwxwe34hzn3VjfRrH+lwc63NprNHFXWp9jDE6deqUUlNTFRHRtE9mWfmMKyIiQtdee60k8bmuS2B9Lo71uTjW59JYo4u72Po09ZnWOfw9LgCAVQgXAMAq1obL7XZr0aJFfGj5Alifi2N9Lo71uTTW6OJac32sPDgDABC+rH3GBQAIT4QLAGAVwgUAsArhAgBYxcpwLV26VNddd506dOig9PR0vffee06P5IjHH39cLpcr5NS/f//g9urqauXk5Khr167q3LmzpkyZct4f8WxvNm3apIkTJyo1NVUul0tvvPFGyHZjjBYuXKiUlBR17NhRmZmZ2r9/f8g+J0+e1PTp0xUXF6f4+HjNmjVLVVVVbXgvWs+l1mfmzJnn/U5lZ2eH7NNe16egoEAjR45Uly5dlJSUpLvvvltlZWUh+1zOY6q8vFwTJkxQbGyskpKS9Mgjj6iurq4t70qruZw1GjNmzHm/Qw888EDIPle6RtaF69VXX1VeXp4WLVqk999/X0OHDlVWVlbIH6YMJzfeeKOOHTsWPG3evDm47aGHHtKbb76p1atXq7i4WEePHtXkyZMdnLb1nT59WkOHDtXSpUsb3b5kyRI9/fTTWrZsmbZt26ZOnTopKytL1dXVwX2mT5+uPXv2aMOGDVq3bp02bdqkOXPmtNVdaFWXWh9Jys7ODvmdevnll0O2t9f1KS4uVk5OjrZu3aoNGzaotrZW48aN0+nTp4P7XOoxVV9frwkTJujs2bPasmWLXnzxRa1YsUILFy504i61uMtZI0maPXt2yO/QkiVLgttaZI2MZW655RaTk5MT/Lm+vt6kpqaagoICB6dyxqJFi8zQoUMb3VZZWWmio6PN6tWrg+d99NFHRpIpKSlpowmdJcmsWbMm+HNDQ4Pxer3mySefDJ5XWVlp3G63efnll40xxuzdu9dIMtu3bw/u8/bbbxuXy2U+/fTTNpu9LXx9fYwxZsaMGWbSpEkXvEw4rc/x48eNJFNcXGyMubzH1FtvvWUiIiKMz+cL7vPcc8+ZuLg4U1NT07Z3oA18fY2MMeYb3/iG+dGPfnTBy7TEGln1jOvs2bMqLS1VZmZm8LyIiAhlZmaqpKTEwcmcs3//fqWmpqp3796aPn26ysvLJUmlpaWqra0NWav+/furZ8+eYbtWhw4dks/nC1kTj8ej9PT04JqUlJQoPj5eI0aMCO6TmZmpiIgIbdu2rc1ndkJRUZGSkpLUr18/zZ07VydOnAhuC6f18fv9kqSEhARJl/eYKikp0eDBg5WcnBzcJysrS4FAQHv27GnD6dvG19fonJUrVyoxMVGDBg1Sfn6+zpw5E9zWEmtk1Zfsfv7556qvrw+5w5KUnJysffv2OTSVc9LT07VixQr169dPx44d0+LFi3X77bdr9+7d8vl8iomJUXx8fMhlkpOT5fP5nBnYYefud2O/P+e2+Xw+JSUlhWyPiopSQkJCWKxbdna2Jk+erLS0NB08eFA/+clPNH78eJWUlCgyMjJs1qehoUHz5s3TbbfdpkGDBknSZT2mfD5fo79f57a1J42tkSR95zvfUa9evZSamqpdu3ZpwYIFKisr0+uvvy6pZdbIqnAh1Pjx44P/HjJkiNLT09WrVy+99tpr6tixo4OTwVZTp04N/nvw4MEaMmSI+vTpo6KiIo0dO9bBydpWTk6Odu/eHfKeMUJdaI2++n7n4MGDlZKSorFjx+rgwYPq06dPi9y2VS8VJiYmKjIy8ryjeCoqKuT1eh2a6uoRHx+vvn376sCBA/J6vTp79qwqKytD9gnntTp3vy/2++P1es870Keurk4nT54My3Xr3bu3EhMTdeDAAUnhsT65ublat26d3n33XXXv3j14/uU8prxeb6O/X+e2tRcXWqPGpKenS1LI79CVrpFV4YqJidHw4cNVWFgYPK+hoUGFhYXKyMhwcLKrQ1VVlQ4ePKiUlBQNHz5c0dHRIWtVVlam8vLysF2rtLQ0eb3ekDUJBALatm1bcE0yMjJUWVmp0tLS4D4bN25UQ0ND8AEYTo4cOaITJ04oJSVFUvteH2OMcnNztWbNGm3cuFFpaWkh2y/nMZWRkaEPP/wwJO4bNmxQXFycBg4c2DZ3pBVdao0as3PnTkkK+R264jVq5sEkjnnllVeM2+02K1asMHv37jVz5swx8fHxIUeohIv58+eboqIic+jQIfPHP/7RZGZmmsTERHP8+HFjjDEPPPCA6dmzp9m4caPZsWOHycjIMBkZGQ5P3bpOnTplPvjgA/PBBx8YSeapp54yH3zwgfnkk0+MMcY88cQTJj4+3qxdu9bs2rXLTJo0yaSlpZkvvvgieB3Z2dlm2LBhZtu2bWbz5s3mhhtuMNOmTXPqLrWoi63PqVOnzMMPP2xKSkrMoUOHzDvvvGNuvvlmc8MNN5jq6urgdbTX9Zk7d67xeDymqKjIHDt2LHg6c+ZMcJ9LPabq6urMoEGDzLhx48zOnTvN+vXrTbdu3Ux+fr4Td6nFXWqNDhw4YH72s5+ZHTt2mEOHDpm1a9ea3r17m9GjRwevoyXWyLpwGWPMM888Y3r27GliYmLMLbfcYrZu3er0SI649957TUpKiomJiTHXXnutuffee82BAweC27/44gvzwx/+0FxzzTUmNjbW3HPPPebYsWMOTtz63n33XSPpvNOMGTOMMV8eEv/Tn/7UJCcnG7fbbcaOHWvKyspCruPEiRNm2rRppnPnziYuLs7cf//95tSpUw7cm5Z3sfU5c+aMGTdunOnWrZuJjo42vXr1MrNnzz7vfwrb6/o0ti6SzPLly4P7XM5j6uOPPzbjx483HTt2NImJiWb+/Pmmtra2je9N67jUGpWXl5vRo0ebhIQE43a7zfXXX28eeeQR4/f7Q67nSteIP2sCALCKVe9xAQBAuAAAViFcAACrEC4AgFUIFwDAKoQLAGAVwgUAsArhAgBYhXABAKxCuAAAViFcAACrEC4AgFX+PyWE64GC4z+BAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Simplifying all masks\n",
    "masks = simplify_mask(masks)\n",
    "\n",
    "#Test\n",
    "print(f\"(min, max): ({masks.min()}, {masks.max()})\")\n",
    "\n",
    "plt.imshow(masks[0], cmap=\"viridis\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we need to seperate our data into training, testing and validation data. We will use the ratio of 80%, 10%, 10% respectively, although this is some what arbitrary. Ideally you should use as much of the data as possible for training, while keeping enough for testing and validation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "#seperating into training, testing and validation data\n",
    "\n",
    "#           train, test, val\n",
    "data_split = [0.8, 0.1, 0.1]\n",
    "#Finding the number of images/masks in category\n",
    "\n",
    "n_train = int(data_split[0]*masks.shape[0])                         #80%\n",
    "n_test =int(data_split[1]*masks.shape[0])                           #10%\n",
    "n_val = int(data_split[2]*masks.shape[0] )                           #10% \n",
    "\n",
    "#Note: rounded we will have 81 test data and 80 val (not exactly 10%) \n",
    "\n",
    "#seperating the data into categories\n",
    "\n",
    "train_images = images[:n_train]\n",
    "train_masks = masks[:n_train]\n",
    "\n",
    "test_images = images[n_train:n_train+n_test+1]\n",
    "test_masks = masks[n_train:n_train+n_test+1]\n",
    "\n",
    "val_images = images[n_train+n_test+1:]\n",
    "val_masks = masks[n_train+n_test+1:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We need to add a \"channel dimension\" to have our data in the correct format for tensorflow. We can do this by using the `np.expand_dims()` function. Set axis = 3 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_images = np.expand_dims(train_images,axis=3)\n",
    "test_images = np.expand_dims(test_images,axis=3)\n",
    "val_images = np.expand_dims(val_images,axis=3)\n",
    "\n",
    "train_masks = np.expand_dims(train_masks,axis=3)\n",
    "test_masks = np.expand_dims(test_masks,axis=3)\n",
    "val_masks = np.expand_dims(val_masks,axis=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(644, 256, 256, 1)\n",
      "(81, 256, 256, 1)\n",
      "(80, 256, 256, 1)\n",
      "(644, 256, 256, 1)\n",
      "(81, 256, 256, 1)\n",
      "(80, 256, 256, 1)\n"
     ]
    }
   ],
   "source": [
    "#Checking shape of data \n",
    "print(train_images.shape)\n",
    "print(test_images.shape)\n",
    "print(val_images.shape)\n",
    "\n",
    "#Checking shape of masks\n",
    "print(train_masks.shape)\n",
    "print(test_masks.shape)\n",
    "print(val_masks.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We now need to set our parameters for the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "IM_height = images.shape[1]     #Extracting the image height\n",
    "IM_length = images.shape[2]     #Extracting the image length\n",
    "channels = 1                    #Only need 1 input channel so set to 1\n",
    "batch_size = 16                 #Arbitrary choice, you can try setting to different values to optimize the model\n",
    "epochs = 100                     #The larger the number of epochs the better, however increases execution time"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We now need to convert our data into tensorflow (tf) datasets, and then seperate them into batches."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2025-01-17 23:08:46.835956: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 AVX512F AVX512_VNNI FMA\n",
      "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
      "2025-01-17 23:08:47.316374: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1616] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 22173 MB memory:  -> device: 0, name: NVIDIA GeForce RTX 3090, pci bus id: 0000:b3:00.0, compute capability: 8.6\n"
     ]
    }
   ],
   "source": [
    "train_dataset = tf.data.Dataset.from_tensor_slices((train_images, train_masks))\n",
    "train_dataset = train_dataset.batch(batch_size)\n",
    "\n",
    "val_dataset = tf.data.Dataset.from_tensor_slices((val_images, val_masks))\n",
    "val_dataset = val_dataset.batch(batch_size)\n",
    "\n",
    "test_dataset = tf.data.Dataset.from_tensor_slices((test_images))\n",
    "test_dataset = test_dataset.batch(batch_size)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The next cell is optional, however if you would like to save you model so that it can be used at a later date it is quite useful.\n",
    "The code below will allow the model automatically save the best model. While the model is being build, it will only save itself if the validation loss has decreased compared to the model at any previous epoch."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "#First create a pathname of where to save the model\n",
    "checkpointpath = \"/workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\"\n",
    "checkpoint_dir = os.path.dirname(checkpointpath)\n",
    "\n",
    "#\n",
    "cp_callback = tf.keras.callbacks.ModelCheckpoint(filepath=checkpointpath, save_best_only=True, verbose=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Building Our Model\n",
    "Now we are ready to make our model. TensorFlow MRI has a premade 2D Unet model function which can be called from\n",
    "`tfmri.models.Unet2D()`. Details of this model and its unputs can be found on the TensorFlow MRI website.\n",
    "The basic parameters we need to define are:\n",
    "\n",
    "`filters=` - should take a list of 3 values, normally each value should be double the last eg. `[32,64,128]`\n",
    "\n",
    "`kernel_size=` - define the dimensions of the kernel as a list, if you only input 1 value then it will assume dimensions are the same size\n",
    "\n",
    "`out_channels=` - should keep the same as the number of input channels, in this case =1\n",
    "\n",
    "`out_activation=` - set this = `\"sigmoid\"` \n",
    "\n",
    "\n",
    "We then need to compile our model. The basic parameters we need to define are:\n",
    "\n",
    "`optimizer=` - normally set our optimizer = `\"adam\"`, more options available on [GitHub](https://mrphys.github.io/tensorflow-mri/)\n",
    "\n",
    "`loss=` - many options which you can read about on [GitHub](https://mrphys.github.io/tensorflow-mri/), however we will use `\"binary_crossentropy\"`. You can test and\n",
    "compare some of the other loss types.\n",
    "\n",
    "`metrics=` - we will use `tfmri.metrics.Accuracy()`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "model = tfmri.models.UNet2D(filters= [32,64,128], kernel_size = 3, out_channels=1, out_activation = \"sigmoid\")\n",
    "model.compile(optimizer = 'adam', loss=\"binary_crossentropy\", metrics=tfmri.metrics.Accuracy())\n",
    "#Accurracy estimates how often predictions match labels \n",
    "\n",
    "#Other segmentation losses in tfmi include:\n",
    "#   tfmri.losses.ConfusionLoss\n",
    "#   tfmri.losses.F1Loss (aka Dice loss)\n",
    "#   tfmri.losses.FocalTverskyLoss\n",
    "#   tfmri.losses.IoULoss (aka Jaccard loss)\n",
    "# (https://mrphys.github.io/tensorflow-mri/api_docs/tfmri/losses/)\n",
    "\n",
    "#Other segmetnation metrics  in tfmi include:\n",
    "#   tfmri.metrics.ConfusionMetric\n",
    "#   tfmri.metrics.F1Score (aka Dice score)\n",
    "#   tfmri.metrics.FBetaScore (This is the weighted harmonic mean of precision and recall)\n",
    "#   tfmri.metrics.TverskyIndex\n",
    "#   tfmri.metrics.IoU (aka Jaccard index)\n",
    "# (https://mrphys.github.io/tensorflow-mri/api_docs/tfmri/metrics/)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we will build our model to our data using the `fit()` function. \n",
    "input:\n",
    "\n",
    "`training_data=` - we use our `train_dataset`\n",
    "\n",
    "`validation_data=` - we use our `val_dataset`\n",
    "\n",
    "`batch_size=` - we use our `batch_size`\n",
    "\n",
    "`verbose=` - we set = 1\n",
    "\n",
    "`epochs=` - we use our `epochs`\n",
    "\n",
    "`callbacks=` - this is what saves our model, we set to our variable `cp_callback`\n",
    "\n",
    "Building our model will take some time  so you may want to test that it is working by first using a small number of epochs (even just 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2025-01-17 23:08:50.009773: I tensorflow/stream_executor/cuda/cuda_dnn.cc:384] Loaded cuDNN version 8100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - ETA: 0s - loss: 0.1629 - acc: 0.9813\n",
      "Epoch 1: val_loss improved from inf to 0.09235, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 11s 145ms/step - loss: 0.1629 - acc: 0.9813 - val_loss: 0.0924 - val_acc: 0.9837\n",
      "Epoch 2/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0687 - acc: 0.9836\n",
      "Epoch 2: val_loss improved from 0.09235 to 0.07003, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 116ms/step - loss: 0.0688 - acc: 0.9836 - val_loss: 0.0700 - val_acc: 0.9837\n",
      "Epoch 3/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0629 - acc: 0.9836\n",
      "Epoch 3: val_loss did not improve from 0.07003\n",
      "41/41 [==============================] - 3s 78ms/step - loss: 0.0631 - acc: 0.9836 - val_loss: 0.0781 - val_acc: 0.9837\n",
      "Epoch 4/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0618 - acc: 0.9836\n",
      "Epoch 4: val_loss improved from 0.07003 to 0.06825, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 116ms/step - loss: 0.0619 - acc: 0.9836 - val_loss: 0.0683 - val_acc: 0.9837\n",
      "Epoch 5/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0593 - acc: 0.9836\n",
      "Epoch 5: val_loss improved from 0.06825 to 0.06565, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 122ms/step - loss: 0.0594 - acc: 0.9836 - val_loss: 0.0657 - val_acc: 0.9837\n",
      "Epoch 6/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0581 - acc: 0.9836\n",
      "Epoch 6: val_loss improved from 0.06565 to 0.05933, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 115ms/step - loss: 0.0582 - acc: 0.9836 - val_loss: 0.0593 - val_acc: 0.9837\n",
      "Epoch 7/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0559 - acc: 0.9836\n",
      "Epoch 7: val_loss improved from 0.05933 to 0.05912, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 121ms/step - loss: 0.0560 - acc: 0.9836 - val_loss: 0.0591 - val_acc: 0.9837\n",
      "Epoch 8/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0543 - acc: 0.9836\n",
      "Epoch 8: val_loss improved from 0.05912 to 0.05866, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 116ms/step - loss: 0.0544 - acc: 0.9836 - val_loss: 0.0587 - val_acc: 0.9837\n",
      "Epoch 9/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0533 - acc: 0.9836\n",
      "Epoch 9: val_loss improved from 0.05866 to 0.05606, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 122ms/step - loss: 0.0534 - acc: 0.9836 - val_loss: 0.0561 - val_acc: 0.9837\n",
      "Epoch 10/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0513 - acc: 0.9836\n",
      "Epoch 10: val_loss improved from 0.05606 to 0.05285, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 122ms/step - loss: 0.0514 - acc: 0.9836 - val_loss: 0.0529 - val_acc: 0.9837\n",
      "Epoch 11/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0496 - acc: 0.9836\n",
      "Epoch 11: val_loss did not improve from 0.05285\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0497 - acc: 0.9836 - val_loss: 0.0549 - val_acc: 0.9837\n",
      "Epoch 12/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0492 - acc: 0.9836\n",
      "Epoch 12: val_loss did not improve from 0.05285\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0494 - acc: 0.9836 - val_loss: 0.0601 - val_acc: 0.9837\n",
      "Epoch 13/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0477 - acc: 0.9836\n",
      "Epoch 13: val_loss improved from 0.05285 to 0.04611, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 116ms/step - loss: 0.0478 - acc: 0.9836 - val_loss: 0.0461 - val_acc: 0.9837\n",
      "Epoch 14/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0413 - acc: 0.9836\n",
      "Epoch 14: val_loss improved from 0.04611 to 0.04307, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 123ms/step - loss: 0.0414 - acc: 0.9836 - val_loss: 0.0431 - val_acc: 0.9846\n",
      "Epoch 15/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0389 - acc: 0.9857\n",
      "Epoch 15: val_loss improved from 0.04307 to 0.03771, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 117ms/step - loss: 0.0391 - acc: 0.9856 - val_loss: 0.0377 - val_acc: 0.9882\n",
      "Epoch 16/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0347 - acc: 0.9874\n",
      "Epoch 16: val_loss improved from 0.03771 to 0.03747, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 116ms/step - loss: 0.0348 - acc: 0.9874 - val_loss: 0.0375 - val_acc: 0.9890\n",
      "Epoch 17/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0320 - acc: 0.9882\n",
      "Epoch 17: val_loss improved from 0.03747 to 0.03494, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 124ms/step - loss: 0.0322 - acc: 0.9881 - val_loss: 0.0349 - val_acc: 0.9885\n",
      "Epoch 18/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0295 - acc: 0.9891\n",
      "Epoch 18: val_loss improved from 0.03494 to 0.03083, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 116ms/step - loss: 0.0297 - acc: 0.9890 - val_loss: 0.0308 - val_acc: 0.9899\n",
      "Epoch 19/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0284 - acc: 0.9894\n",
      "Epoch 19: val_loss did not improve from 0.03083\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0286 - acc: 0.9893 - val_loss: 0.0374 - val_acc: 0.9875\n",
      "Epoch 20/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0261 - acc: 0.9906\n",
      "Epoch 20: val_loss did not improve from 0.03083\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0262 - acc: 0.9906 - val_loss: 0.0336 - val_acc: 0.9881\n",
      "Epoch 21/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0241 - acc: 0.9911\n",
      "Epoch 21: val_loss improved from 0.03083 to 0.02764, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 122ms/step - loss: 0.0243 - acc: 0.9910 - val_loss: 0.0276 - val_acc: 0.9901\n",
      "Epoch 22/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0227 - acc: 0.9917\n",
      "Epoch 22: val_loss improved from 0.02764 to 0.02748, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 121ms/step - loss: 0.0228 - acc: 0.9916 - val_loss: 0.0275 - val_acc: 0.9903\n",
      "Epoch 23/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0208 - acc: 0.9924\n",
      "Epoch 23: val_loss did not improve from 0.02748\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0209 - acc: 0.9924 - val_loss: 0.0277 - val_acc: 0.9898\n",
      "Epoch 24/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0190 - acc: 0.9932\n",
      "Epoch 24: val_loss improved from 0.02748 to 0.02481, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 117ms/step - loss: 0.0191 - acc: 0.9931 - val_loss: 0.0248 - val_acc: 0.9909\n",
      "Epoch 25/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0189 - acc: 0.9932\n",
      "Epoch 25: val_loss improved from 0.02481 to 0.02388, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 117ms/step - loss: 0.0190 - acc: 0.9932 - val_loss: 0.0239 - val_acc: 0.9911\n",
      "Epoch 26/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0186 - acc: 0.9931\n",
      "Epoch 26: val_loss improved from 0.02388 to 0.02326, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 123ms/step - loss: 0.0186 - acc: 0.9930 - val_loss: 0.0233 - val_acc: 0.9919\n",
      "Epoch 27/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0206 - acc: 0.9927\n",
      "Epoch 27: val_loss improved from 0.02326 to 0.02211, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 121ms/step - loss: 0.0207 - acc: 0.9926 - val_loss: 0.0221 - val_acc: 0.9920\n",
      "Epoch 28/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0193 - acc: 0.9932\n",
      "Epoch 28: val_loss improved from 0.02211 to 0.01808, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 117ms/step - loss: 0.0193 - acc: 0.9932 - val_loss: 0.0181 - val_acc: 0.9936\n",
      "Epoch 29/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0140 - acc: 0.9949\n",
      "Epoch 29: val_loss improved from 0.01808 to 0.01688, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 118ms/step - loss: 0.0141 - acc: 0.9949 - val_loss: 0.0169 - val_acc: 0.9940\n",
      "Epoch 30/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0165 - acc: 0.9941\n",
      "Epoch 30: val_loss improved from 0.01688 to 0.01678, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 127ms/step - loss: 0.0165 - acc: 0.9941 - val_loss: 0.0168 - val_acc: 0.9940\n",
      "Epoch 31/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0145 - acc: 0.9947\n",
      "Epoch 31: val_loss improved from 0.01678 to 0.01523, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 124ms/step - loss: 0.0146 - acc: 0.9947 - val_loss: 0.0152 - val_acc: 0.9946\n",
      "Epoch 32/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0163 - acc: 0.9942\n",
      "Epoch 32: val_loss did not improve from 0.01523\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0163 - acc: 0.9941 - val_loss: 0.0165 - val_acc: 0.9940\n",
      "Epoch 33/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0159 - acc: 0.9941\n",
      "Epoch 33: val_loss did not improve from 0.01523\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0160 - acc: 0.9941 - val_loss: 0.0181 - val_acc: 0.9936\n",
      "Epoch 34/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0134 - acc: 0.9950\n",
      "Epoch 34: val_loss did not improve from 0.01523\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0134 - acc: 0.9950 - val_loss: 0.0167 - val_acc: 0.9941\n",
      "Epoch 35/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0128 - acc: 0.9952\n",
      "Epoch 35: val_loss did not improve from 0.01523\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0129 - acc: 0.9952 - val_loss: 0.0171 - val_acc: 0.9941\n",
      "Epoch 36/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0135 - acc: 0.9949\n",
      "Epoch 36: val_loss did not improve from 0.01523\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0136 - acc: 0.9949 - val_loss: 0.0180 - val_acc: 0.9940\n",
      "Epoch 37/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0133 - acc: 0.9951\n",
      "Epoch 37: val_loss did not improve from 0.01523\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0134 - acc: 0.9951 - val_loss: 0.0168 - val_acc: 0.9946\n",
      "Epoch 38/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0113 - acc: 0.9958\n",
      "Epoch 38: val_loss did not improve from 0.01523\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0113 - acc: 0.9957 - val_loss: 0.0156 - val_acc: 0.9947\n",
      "Epoch 39/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0097 - acc: 0.9963\n",
      "Epoch 39: val_loss did not improve from 0.01523\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0097 - acc: 0.9963 - val_loss: 0.0160 - val_acc: 0.9949\n",
      "Epoch 40/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0092 - acc: 0.9965\n",
      "Epoch 40: val_loss improved from 0.01523 to 0.01414, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 117ms/step - loss: 0.0092 - acc: 0.9965 - val_loss: 0.0141 - val_acc: 0.9954\n",
      "Epoch 41/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0079 - acc: 0.9969\n",
      "Epoch 41: val_loss improved from 0.01414 to 0.01313, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 123ms/step - loss: 0.0079 - acc: 0.9969 - val_loss: 0.0131 - val_acc: 0.9956\n",
      "Epoch 42/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0073 - acc: 0.9972\n",
      "Epoch 42: val_loss improved from 0.01313 to 0.01215, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 117ms/step - loss: 0.0073 - acc: 0.9972 - val_loss: 0.0121 - val_acc: 0.9959\n",
      "Epoch 43/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0067 - acc: 0.9974\n",
      "Epoch 43: val_loss improved from 0.01215 to 0.01170, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 116ms/step - loss: 0.0066 - acc: 0.9974 - val_loss: 0.0117 - val_acc: 0.9961\n",
      "Epoch 44/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0061 - acc: 0.9976\n",
      "Epoch 44: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0061 - acc: 0.9976 - val_loss: 0.0132 - val_acc: 0.9956\n",
      "Epoch 45/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0058 - acc: 0.9977\n",
      "Epoch 45: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0057 - acc: 0.9977 - val_loss: 0.0135 - val_acc: 0.9955\n",
      "Epoch 46/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0059 - acc: 0.9976\n",
      "Epoch 46: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0059 - acc: 0.9976 - val_loss: 0.0157 - val_acc: 0.9953\n",
      "Epoch 47/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0055 - acc: 0.9978\n",
      "Epoch 47: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0055 - acc: 0.9978 - val_loss: 0.0166 - val_acc: 0.9954\n",
      "Epoch 48/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0055 - acc: 0.9978\n",
      "Epoch 48: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0055 - acc: 0.9978 - val_loss: 0.0149 - val_acc: 0.9957\n",
      "Epoch 49/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0055 - acc: 0.9978\n",
      "Epoch 49: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0055 - acc: 0.9978 - val_loss: 0.0147 - val_acc: 0.9957\n",
      "Epoch 50/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0048 - acc: 0.9980\n",
      "Epoch 50: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0048 - acc: 0.9980 - val_loss: 0.0126 - val_acc: 0.9961\n",
      "Epoch 51/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0045 - acc: 0.9982\n",
      "Epoch 51: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0045 - acc: 0.9982 - val_loss: 0.0133 - val_acc: 0.9961\n",
      "Epoch 52/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9983\n",
      "Epoch 52: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0042 - acc: 0.9983 - val_loss: 0.0142 - val_acc: 0.9960\n",
      "Epoch 53/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0041 - acc: 0.9983\n",
      "Epoch 53: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0041 - acc: 0.9983 - val_loss: 0.0161 - val_acc: 0.9958\n",
      "Epoch 54/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0041 - acc: 0.9983\n",
      "Epoch 54: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0041 - acc: 0.9983 - val_loss: 0.0171 - val_acc: 0.9957\n",
      "Epoch 55/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0041 - acc: 0.9983\n",
      "Epoch 55: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0040 - acc: 0.9983 - val_loss: 0.0181 - val_acc: 0.9956\n",
      "Epoch 56/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0041 - acc: 0.9983\n",
      "Epoch 56: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0041 - acc: 0.9983 - val_loss: 0.0169 - val_acc: 0.9957\n",
      "Epoch 57/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0041 - acc: 0.9983\n",
      "Epoch 57: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0041 - acc: 0.9983 - val_loss: 0.0147 - val_acc: 0.9959\n",
      "Epoch 58/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0037 - acc: 0.9985\n",
      "Epoch 58: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0037 - acc: 0.9985 - val_loss: 0.0147 - val_acc: 0.9958\n",
      "Epoch 59/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0037 - acc: 0.9985\n",
      "Epoch 59: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 80ms/step - loss: 0.0037 - acc: 0.9985 - val_loss: 0.0128 - val_acc: 0.9960\n",
      "Epoch 60/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0044 - acc: 0.9982\n",
      "Epoch 60: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 80ms/step - loss: 0.0044 - acc: 0.9982 - val_loss: 0.0135 - val_acc: 0.9958\n",
      "Epoch 61/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0054 - acc: 0.9979\n",
      "Epoch 61: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 80ms/step - loss: 0.0054 - acc: 0.9979 - val_loss: 0.0134 - val_acc: 0.9956\n",
      "Epoch 62/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0050 - acc: 0.9980\n",
      "Epoch 62: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0050 - acc: 0.9980 - val_loss: 0.0129 - val_acc: 0.9959\n",
      "Epoch 63/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0059 - acc: 0.9976\n",
      "Epoch 63: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 80ms/step - loss: 0.0059 - acc: 0.9976 - val_loss: 0.0144 - val_acc: 0.9954\n",
      "Epoch 64/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0063 - acc: 0.9976\n",
      "Epoch 64: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 80ms/step - loss: 0.0063 - acc: 0.9975 - val_loss: 0.0156 - val_acc: 0.9948\n",
      "Epoch 65/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0069 - acc: 0.9973\n",
      "Epoch 65: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 80ms/step - loss: 0.0069 - acc: 0.9973 - val_loss: 0.0136 - val_acc: 0.9961\n",
      "Epoch 66/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0089 - acc: 0.9966\n",
      "Epoch 66: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0089 - acc: 0.9966 - val_loss: 0.0124 - val_acc: 0.9959\n",
      "Epoch 67/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0054 - acc: 0.9978\n",
      "Epoch 67: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 80ms/step - loss: 0.0054 - acc: 0.9978 - val_loss: 0.0166 - val_acc: 0.9955\n",
      "Epoch 68/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0051 - acc: 0.9980\n",
      "Epoch 68: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0051 - acc: 0.9979 - val_loss: 0.0179 - val_acc: 0.9951\n",
      "Epoch 69/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0050 - acc: 0.9980\n",
      "Epoch 69: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 80ms/step - loss: 0.0050 - acc: 0.9980 - val_loss: 0.0122 - val_acc: 0.9962\n",
      "Epoch 70/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0038 - acc: 0.9984\n",
      "Epoch 70: val_loss did not improve from 0.01170\n",
      "41/41 [==============================] - 3s 80ms/step - loss: 0.0038 - acc: 0.9984 - val_loss: 0.0132 - val_acc: 0.9963\n",
      "Epoch 71/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0033 - acc: 0.9986\n",
      "Epoch 71: val_loss improved from 0.01170 to 0.01161, saving model to /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /workspaces/Tutorials/2Dsegmentation_DONE/models/unet_segmentation/model.ckpt/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 5s 124ms/step - loss: 0.0033 - acc: 0.9986 - val_loss: 0.0116 - val_acc: 0.9964\n",
      "Epoch 72/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0032 - acc: 0.9986\n",
      "Epoch 72: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0032 - acc: 0.9987 - val_loss: 0.0122 - val_acc: 0.9964\n",
      "Epoch 73/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0031 - acc: 0.9987\n",
      "Epoch 73: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0031 - acc: 0.9987 - val_loss: 0.0130 - val_acc: 0.9963\n",
      "Epoch 74/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0030 - acc: 0.9987\n",
      "Epoch 74: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0030 - acc: 0.9987 - val_loss: 0.0127 - val_acc: 0.9964\n",
      "Epoch 75/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0029 - acc: 0.9988\n",
      "Epoch 75: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 80ms/step - loss: 0.0029 - acc: 0.9988 - val_loss: 0.0131 - val_acc: 0.9964\n",
      "Epoch 76/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0029 - acc: 0.9988\n",
      "Epoch 76: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 80ms/step - loss: 0.0029 - acc: 0.9988 - val_loss: 0.0130 - val_acc: 0.9964\n",
      "Epoch 77/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0028 - acc: 0.9988\n",
      "Epoch 77: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0028 - acc: 0.9988 - val_loss: 0.0136 - val_acc: 0.9962\n",
      "Epoch 78/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0028 - acc: 0.9988\n",
      "Epoch 78: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0028 - acc: 0.9988 - val_loss: 0.0136 - val_acc: 0.9965\n",
      "Epoch 79/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0027 - acc: 0.9988\n",
      "Epoch 79: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0027 - acc: 0.9988 - val_loss: 0.0151 - val_acc: 0.9964\n",
      "Epoch 80/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0027 - acc: 0.9988\n",
      "Epoch 80: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 80ms/step - loss: 0.0027 - acc: 0.9988 - val_loss: 0.0142 - val_acc: 0.9963\n",
      "Epoch 81/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0027 - acc: 0.9988\n",
      "Epoch 81: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 80ms/step - loss: 0.0027 - acc: 0.9988 - val_loss: 0.0142 - val_acc: 0.9962\n",
      "Epoch 82/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0027 - acc: 0.9988\n",
      "Epoch 82: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 80ms/step - loss: 0.0027 - acc: 0.9988 - val_loss: 0.0155 - val_acc: 0.9963\n",
      "Epoch 83/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0026 - acc: 0.9989\n",
      "Epoch 83: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 80ms/step - loss: 0.0026 - acc: 0.9989 - val_loss: 0.0168 - val_acc: 0.9964\n",
      "Epoch 84/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0028 - acc: 0.9988\n",
      "Epoch 84: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0027 - acc: 0.9988 - val_loss: 0.0172 - val_acc: 0.9962\n",
      "Epoch 85/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0028 - acc: 0.9988\n",
      "Epoch 85: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0028 - acc: 0.9988 - val_loss: 0.0166 - val_acc: 0.9960\n",
      "Epoch 86/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0027 - acc: 0.9988\n",
      "Epoch 86: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0027 - acc: 0.9989 - val_loss: 0.0150 - val_acc: 0.9964\n",
      "Epoch 87/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0026 - acc: 0.9989\n",
      "Epoch 87: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0026 - acc: 0.9989 - val_loss: 0.0157 - val_acc: 0.9965\n",
      "Epoch 88/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0025 - acc: 0.9989\n",
      "Epoch 88: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0025 - acc: 0.9989 - val_loss: 0.0177 - val_acc: 0.9965\n",
      "Epoch 89/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0025 - acc: 0.9989\n",
      "Epoch 89: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0025 - acc: 0.9989 - val_loss: 0.0198 - val_acc: 0.9964\n",
      "Epoch 90/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0025 - acc: 0.9989\n",
      "Epoch 90: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0025 - acc: 0.9989 - val_loss: 0.0221 - val_acc: 0.9961\n",
      "Epoch 91/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0024 - acc: 0.9990\n",
      "Epoch 91: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0024 - acc: 0.9990 - val_loss: 0.0195 - val_acc: 0.9964\n",
      "Epoch 92/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0024 - acc: 0.9990\n",
      "Epoch 92: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0024 - acc: 0.9990 - val_loss: 0.0201 - val_acc: 0.9964\n",
      "Epoch 93/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0023 - acc: 0.9990\n",
      "Epoch 93: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0023 - acc: 0.9990 - val_loss: 0.0196 - val_acc: 0.9965\n",
      "Epoch 94/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0023 - acc: 0.9990\n",
      "Epoch 94: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 80ms/step - loss: 0.0023 - acc: 0.9990 - val_loss: 0.0206 - val_acc: 0.9963\n",
      "Epoch 95/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0024 - acc: 0.9990\n",
      "Epoch 95: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 80ms/step - loss: 0.0024 - acc: 0.9990 - val_loss: 0.0188 - val_acc: 0.9964\n",
      "Epoch 96/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0024 - acc: 0.9990\n",
      "Epoch 96: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0024 - acc: 0.9990 - val_loss: 0.0184 - val_acc: 0.9965\n",
      "Epoch 97/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0023 - acc: 0.9990\n",
      "Epoch 97: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0023 - acc: 0.9990 - val_loss: 0.0192 - val_acc: 0.9966\n",
      "Epoch 98/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0023 - acc: 0.9990\n",
      "Epoch 98: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0023 - acc: 0.9990 - val_loss: 0.0209 - val_acc: 0.9963\n",
      "Epoch 99/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0023 - acc: 0.9990\n",
      "Epoch 99: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0023 - acc: 0.9990 - val_loss: 0.0183 - val_acc: 0.9963\n",
      "Epoch 100/100\n",
      "40/41 [============================>.] - ETA: 0s - loss: 0.0023 - acc: 0.9990\n",
      "Epoch 100: val_loss did not improve from 0.01161\n",
      "41/41 [==============================] - 3s 79ms/step - loss: 0.0023 - acc: 0.9990 - val_loss: 0.0170 - val_acc: 0.9964\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(train_dataset,\n",
    "                    validation_data=val_dataset,\n",
    "                    batch_size=batch_size,\n",
    "                    verbose=1,\n",
    "                    epochs=epochs,\n",
    "                    callbacks=[cp_callback])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Results\n",
    "Our model should now be build, we can extract the loss and accuracy data and plot to see how they very accross epochs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Accuracy')"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x1500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Exctracting data from our model\n",
    "loss = history.history['loss']\n",
    "accuracy = history.history['acc']\n",
    "epochs = np.arange(1,len(loss)+1)\n",
    "\n",
    "loss_range = len(loss)\n",
    "val_loss = history.history['val_loss']\n",
    "\n",
    "#Creating figures and plotting our losses and accuracy against epochs.\n",
    "fig = plt.figure(figsize=(20,15))\n",
    "ax1, ax2 = plt.subplot(2,2,1), plt.subplot(2,2,2)\n",
    "ax1.plot(epochs,loss,'b',label= 'Loss')\n",
    "ax1.plot(epochs,val_loss,'g', label = 'Validation Loss')\n",
    "ax1.legend()\n",
    "ax1.set_title('Cross Entropy Loss', fontsize='16')\n",
    "ax1.set_xlabel('Epochs', fontsize = '14')\n",
    "ax1.set_ylabel('Loss', fontsize = '14')\n",
    "\n",
    "ax2.plot(epochs,accuracy,'r')\n",
    "ax2.set_title('Accuracy', fontsize = '16')\n",
    "ax2.set_xlabel('Epochs', fontsize = '14')\n",
    "ax2.set_ylabel('Accuracy', fontsize = '14')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that the losses decrease accross the epochs, and the accuracy increases.\n",
    "\n",
    "Now we can apply our model to the `test_dataset` to inspect if the model is actual able to predict the masks of our test images."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6/6 [==============================] - 0s 55ms/step\n"
     ]
    }
   ],
   "source": [
    "#Predict the masks of our test images\n",
    "out_ims = model.predict(test_dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Image and ML Generated Mask')"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Plotting a random image and its generated mask\n",
    "i = rng.randint(0,len(test_images)-1)               #picks a random integer as our image/mask index\n",
    "plt.imshow(test_images[i], cmap = 'gray')\n",
    "plt.imshow(out_ims[i], cmap = 'viridis', alpha=0.2)\n",
    "plt.title('Image and ML Generated Mask')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can now plot the ground truth test image and mask (the image with its premade mask) next to the image and generated mask to compare."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Generated Mask')"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABRcAAAFPCAYAAADX1/nMAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuNSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/xnp5ZAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOy9d7wlRZn//6mqDifcNHkYwkhSJAi7sCJJUNFhCQKLImEVRBEDuAroigFk5esIK2YJu/oDF0RcFIFFBZVgAAxrQGVVggRBGWaYufmcDlX1+6NCd59z7syd4Q6TnrcvnLl9+nRXV/d57vnUfJ7nYVprDYIgCIIgCIIgCIIgCIIgiLWEb+gBEARBEARBEARBEARBEASxaUKLiwRBEARBEARBEARBEARBrBO0uEgQBEEQBEEQBEEQBEEQxDpBi4sEQRAEQRAEQRAEQRAEQawTtLhIEARBEARBEARBEARBEMQ6QYuLBEEQBEEQBEEQBEEQBEGsE7S4SBAEQRAEQRAEQRAEQRDEOkGLiwRBEARBEARBEARBEARBrBO0uEgQBEEQBEEQBEEQBEEQxDpBi4ubCYwxfPSjH93Qw1gtp556Kvr6+jb0MAjLpvDMOBhjOPPMMzf0MAhig/PRj34UjLENPYwtisceewyMMVx99dUbeihrxI31k5/85IYeCkEQFvpcEgSxOfGCF7wAp5566oYexrR4wQtegCOPPHJDD2OLYYtaXHz00Udx5pln4oUvfCEajQYajQZ23XVXvOtd78Jvf/vbDT289cohhxwCxtga/3uui02Tk5P46Ec/irvvvntGxl3mkEMOwe677z7jxyWmxn0hZozhoosu6rnPySefDMYYLRxvQKbz2WaMzcjncm0/43fffTcYY/jGN77xnM9NrD0//vGPcfzxx2PrrbdGFEUYHBzEvvvui3/7t3/DsmXLNvTwZpTLLrtsgy4AumedMYZrr7225z4HHHAAGGP0u4zYrNgUfgfR55IgiC1pLeA73/nOBjeRuNj71re+tefrH/rQh/w+K1aseJ5HR6wPgg09gOeLW2+9FW94wxsQBAFOPvlk7LnnnuCc449//CNuvPFGXH755Xj00UexePHiDT3U9cKHPvShygf7F7/4BT73uc/hgx/8IF784hf77S95yUue03kmJydx4YUXAjCLgcTmQa1Ww9e+9jV8+MMfrmyfmJjAzTffjFqttoFGRgDANddcU/n5v/7rv/D973+/a3v5s76u0Gd80+H888/Hxz72Meywww449dRTscMOO6DdbuOXv/wlLr30UnzlK1/BI488sqGHOWNcdtllmDt37gb/1/RarYbrrrsO//zP/1zZ/thjj+Hee++leElsdmwKv4Poc0kQWzZb2lrAd77zHXzxi1/c4AuMtVoN3/zmN3HZZZchiqLKa1/72tdQq9XQbrc30OiImWaLWFx85JFHcMIJJ2Dx4sW44447sNVWW1Vev/jii3HZZZeB89UbOScmJtBsNtfnUNcbr371qys/12o1fO5zn8OrX/3q1X4525SvmZg5Dj/8cNx44424//77seeee/rtN998M9I0xWGHHYY777xzA45wy6ZTLP30pz/F97///a7txJbD17/+dXzsYx/D8ccfj2uuuabrC92nP/1pfPrTn95Ao1szWmu0223U6/UNPZS15vDDD8ctt9yCFStWYO7cuX77ddddhwULFmDnnXfGqlWrNuAICWJm2RR+B9HnkiC2XGZqLWBDsqlq8sMOOwy33HILvvvd7+Loo4/22++99148+uijOO644/DNb35zA46QmEk23k/QDHLJJZdgYmICV111VVcwAYAgCPDud78b2267rd/m6gM+8sgjOPzww9Hf34+TTz4ZgPlwn3POOdh2220RxzFe9KIX4ZOf/CS01v79q6vR1Jl+7GpoPfzwwzj11FMxNDSEwcFBvPnNb8bk5GTlvUmS4L3vfS/mzZuH/v5+vPa1r8WTTz75HGeoOo7/+7//w0knnYRZs2bhwAMPBGD+dbjXIuSpp56KF7zgBf6a582bBwC48MILp0y1fuqpp3DMMcegr68P8+bNw7nnngsp5TqN2dXiu+GGG7DrrruiXq9jv/32w+9+9zsAwJVXXomddtoJtVoNhxxyCB577LHK+3/84x/j9a9/PbbbbjvEcYxtt90W733ve9FqtbrO5c5Rq9Ww++6741vf+lbl+h1KKXzmM5/BbrvthlqthgULFuCMM86Y1pfW3/72t95hVKvVsHDhQpx22ml49tlnK/s938/Mfvvth+233x7XXXddZftXv/pVHHbYYZg9e3bXe26++WYcccQRWLRoEeI4xo477oiPfexjXff6oYcewnHHHYeFCxeiVqthm222wQknnICRkZHVjumiiy4C5xyf//zn1+patlSm+1z+7//+L5YsWYK5c+eiXq9j++23x2mnnQZg+p/xNeGe3wcffBD//M//jMHBQcybNw8f+chHoLXGX/7yFxx99NEYGBjAwoULcemll1ben6Ypzj//fOy9994YHBxEs9nEQQcdhLvuuqvrXM8++yze+MY3YmBgAENDQzjllFNw//3394zPf/zjH/G6170Os2fPRq1Wwz777INbbrllWtf0yU9+Evvvvz/mzJmDer2Ovffeu2cquItZN910E3bffXfEcYzddtsNt912W9e+P/nJT/AP//APqNVq2HHHHXHllVdOayyAcS3OnTsXX/7yl7sWFgFgcHCw53377ne/i4MOOgjNZhP9/f044ogj8MADD1T2cb8fpxPLp/vcuZo4t99+O/bZZx/U63V/vVdddRVe+cpXYv78+YjjGLvuuisuv/zyrvc/8MAD+OEPf+ify/LvrOHhYbznPe/xv7d32mknXHzxxVBKVY4zPDyMU089FYODg/55GR4eXtN0Vzj66KMRxzFuuOGGyvbrrrsOxx9/PIQQXe+ZzjUCq/98ToXWGm9729sQRRFuvPHGtboWgpgpNvTvIPpcEsSWy7qsBQDT+1549dVXgzGGe+65B2effTbmzZuHZrOJY489FsuXL+8619p8z+q1DjEd7Xrqqafii1/8IoBq2QrHdOOx1hoXXXQRttlmGzQaDbziFa/oGuua2HrrrfHyl7+8p4bcY489epajmK4+f/rpp/HmN78Z22yzDeI4xlZbbYWjjz66S+938pWvfAVBEOB973vfWl0LsWa2COfirbfeip122gn77rvvWr0vz3MsWbIEBx54ID75yU+i0WhAa43Xvva1uOuuu/CWt7wFe+21F26//Xa8733vw1NPPfWcnCDHH388tt9+eyxduhS/+tWv8KUvfQnz58/HxRdf7Pd561vfimuvvRYnnXQS9t9/f9x555044ogj1vmcvXj961+PnXfeGR//+McrC6ZrYt68ebj88svxjne8A8ceeyz+6Z/+CUA11VpKiSVLlmDffffFJz/5SfzgBz/ApZdeih133BHveMc71mm8P/7xj3HLLbfgXe96FwBg6dKlOPLII/H+978fl112Gd75zndi1apVuOSSS3DaaadVHHY33HADJicn8Y53vANz5szBz3/+c3z+85/Hk08+WfkC+u1vfxtveMMbsMcee2Dp0qVYtWoV3vKWt2DrrbfuGs8ZZ5yBq6++Gm9+85vx7ne/G48++ii+8IUv4Ne//jXuuecehGE45bV8//vfx5///Ge8+c1vxsKFC/HAAw/gP/7jP/DAAw/gpz/9aVcjh+fzmTnxxBNx7bXX4hOf+ISvjfG9730P11xzTc+Fkauvvhp9fX04++yz0dfXhzvvvBPnn38+RkdH8e///u8AzCLRkiVLkCQJzjrrLCxcuBBPPfUUbr31VgwPD2NwcLDnWD784Q/j4x//OK688kqcfvrpa30tWyLTeS6feeYZvOY1r8G8efPwgQ98AENDQ3jssce88JnOZ3xteMMb3oAXv/jF+MQnPoFvf/vbuOiiizB79mxceeWVeOUrX4mLL74YX/3qV3HuuefiH/7hH/Dyl78cADA6OoovfelLOPHEE3H66adjbGwMX/7yl7FkyRL8/Oc/x1577QXAfHk66qij8POf/xzveMc7sMsuu+Dmm2/GKaec0jWWBx54AAcccAC23nprfOADH0Cz2cR///d/45hjjsE3v/lNHHvssau9ls9+9rN47Wtfi5NPPhlpmuL666/H61//etx6661dn7ef/OQnuPHGG/HOd74T/f39+NznPofjjjsOTzzxBObMmQMA+N3vfufvxUc/+lHkeY4LLrgACxYsWOO8Pvjgg3jwwQfx1re+da1qoV5zzTU45ZRTsGTJElx88cWYnJzE5ZdfjgMPPBC//vWvK/+QMt1Yvjbx8E9/+hNOPPFEnHHGGTj99NPxohe9CABw+eWXY7fddsNrX/taBEGA//mf/8E73/lOKKV83P/MZz6Ds846C319ffjQhz4EAH6uJicncfDBB+Opp57CGWecge222w733nsvzjvvPPztb3/DZz7zGQDmS/TRRx+Nn/zkJ3j729+OF7/4xfjWt77V83lZHY1GA0cffTS+9rWv+bm4//778cADD+BLX/pSz7pO07nGNX0+eyGlxGmnnYavf/3r+Na3vjXj3xcIYrps6N9B9LkkiC2XdVkLWNvvhWeddRZmzZqFCy64AI899hg+85nP4Mwzz8TXv/51v8/afM/qtQ4BTE+7nnHGGfjrX//aszyFe306383OP/98XHTRRTj88MNx+OGH41e/+hVe85rXIE3Tac8jAJx00kn4l3/5F4yPj6Ovrw95nuOGG27A2Wef3TMlerr6/LjjjsMDDzyAs846Cy94wQvwzDPP4Pvf/z6eeOKJLvOP4z/+4z/w9re/HR/84Aen7CdAPAf0Zs7IyIgGoI855piu11atWqWXL1/u/5ucnPSvnXLKKRqA/sAHPlB5z0033aQB6Isuuqiy/XWve51mjOmHH35Ya631o48+qgHoq666quu8APQFF1zgf77gggs0AH3aaadV9jv22GP1nDlz/M+/+c1vNAD9zne+s7LfSSed1HXMNXHDDTdoAPquu+7qGseJJ57Ytf/BBx+sDz744K7tp5xyil68eLH/efny5VOOxc3pv/3bv1W2/93f/Z3ee++91zjmgw8+WO+2226VbQB0HMf60Ucf9duuvPJKDUAvXLhQj46O+u3nnXeeBlDZt3zPHUuXLtWMMf3444/7bXvssYfeZptt9NjYmN929913awCV6//xj3+sAeivfvWrlWPedtttPbd30ms8X/va1zQA/aMf/chve76eGfcc//u//7v+/e9/rwHoH//4x1prrb/4xS/qvr4+PTExoU855RTdbDbXeC1nnHGGbjQaut1ua621/vWvf60B6BtuuGG14wCg3/Wud2mttT7nnHM051xfffXVq33Plsy73vUuXQ7v030uv/Wtb2kA+he/+MWUx17dZ7wXd911V9c9ds/v2972Nr8tz3O9zTbbaMaY/sQnPuG3r1q1StfrdX3KKadU9k2SpHKeVatW6QULFlQ+E9/85jc1AP2Zz3zGb5NS6le+8pVd8flVr3qV3mOPPfyzqbXWSim9//7765133nmN19n5vKdpqnfffXf9yle+srIdgI6iyP+u0Frr+++/XwPQn//85/22Y445RtdqtUoc+r//+z8thKjc217cfPPNXdftrqf8O2/58uU6yzKttdZjY2N6aGhIn3766ZX3PP3003pwcLCyfbqxfG3i4eLFizUAfdttt3VdT69YsmTJEr3DDjtUtu222249f0997GMf081mUz/44IOV7R/4wAe0EEI/8cQTWuvi9/sll1zi98nzXB900EFT/j4vU37Wb731Vs0Y88d+3/ve58fb63fZdK5xOp/PcszOsky/4Q1v0PV6Xd9+++2rHTtBzCQb6+8g+lwSxJbHuq4FTPd74VVXXaUB6EMPPVQrpfz29773vVoIoYeHh7XW6/Y9q3MdQuvpa9fOOOyYbjx+5plndBRF+ogjjqhc1wc/+EENoPK9fCqcflu5cqWOokhfc801Wmutv/3tb2vGmH7ssce8Jli+fPlaXeOqVat8XF0dixcv1kcccYTWWuvPfvazmjGmP/axj61x7MS6sdmnRY+OjgJAT/fGIYccgnnz5vn/nH24TKeb7jvf+Q6EEHj3u99d2X7OOedAa43vfve76zzWt7/97ZWfDzroIDz77LP+Gr7zne8AQNe53/Oe96zzOaczjpmm13X++c9/XufjvepVr6r864T7V6njjjsO/f39XdvL5yrX85qYmMCKFSuw//77Q2uNX//61wCAv/71r/jd736HN73pTZXn6OCDD8Yee+xRGcsNN9yAwcFBvPrVr8aKFSv8f3vvvTf6+vp6pm2WKY+n3W5jxYoVeNnLXgYA+NWvftW1//P5zOy22254yUtegq997WsATCrR0Ucf7f8lbXXXMjY2hhUrVuCggw7C5OQk/vjHPwKAdybefvvtXencnWitceaZZ+Kzn/0srr322rV2E23JTPe5HBoaAmD+hTfLsvU+rnKTKSEE9tlnH2it8Za3vMVvHxoawote9KLK51YI4VN9lVJYuXIl8jzHPvvsU/mc3HbbbQjDsOJu5Zx7x4lj5cqVuPPOO3H88cf7Z3XFihV49tlnsWTJEjz00EN46qmnVnst5ed91apVGBkZwUEHHdTzc3vooYdixx139D+/5CUvwcDAgL9GKSVuv/12HHPMMdhuu+38fi9+8YuxZMmS1Y4DmPr33sjISOV33rx58/Cb3/wGgHFNDw8P48QTT6w8I0II7Lvvvj1j15pi+drGw+23377n9ZXndmRkBCtWrMDBBx+MP//5z2ssn+DGcdBBB2HWrFmVcRx66KGQUuJHP/oRABMvgyCo/N4XQuCss85a4zk6ec1rXoPZs2fj+uuvh9Ya119/PU488cQp95/ONa7N5zNNU++c/c53voPXvOY1a30NBDFTbCy/g+hzSRBbHuuyFrAu3wvf9ra3VTLMDjroIEgp8fjjjwNYt+9ZvbL6pqNdV8d04/EPfvADpGmKs846q3Jd66IhZ82ahcMOO6yiIffff/8pm+dM5xrr9TqiKMLdd989rdJjl1xyCf7lX/4FF198cVeDUmLm2OzTot3i0vj4eNdrV155JcbGxrBs2bKeRaeDIMA222xT2fb4449j0aJFlUUroOiA5wLIulAWkYD5IAJGqA4MDODxxx8H57wiSgH41LGZYvvtt5/R45Wp1Wq+Xo5j1qxZz6mIdue8uQWrzroZbnv5XE888QTOP/983HLLLV1jcF8c3T3daaedus690047VRYPHnroIYyMjGD+/Pk9x/rMM8+s9lpWrlyJCy+8ENdff33Xvr1E9PP9zJx00km49NJL8d73vhf33nsvPvjBD0657wMPPIAPf/jDuPPOO/0v9s5r2X777XH22WfjU5/6FL761a/ioIMOwmtf+1pfh6/Mf/3Xf2F8fByXX375asUA0c10n8uDDz4Yxx13HC688EJ8+tOfxiGHHIJjjjkGJ510EuI4nvFx9frs1mq1SrF9t72z7uhXvvIVXHrppfjjH/9YEXTl+PX4449jq6226loA7/wsP/zww9Ba4yMf+Qg+8pGP9BzrM88807MMguPWW2/FRRddhN/85jdIksRv7yxlAHRfN1CNg8uXL0er1cLOO+/ctd+LXvQi/48GUzHV772+vj58//vfBwB873vf8+UJAPOMAMArX/nKnsccGBio/DydWL628XCq3z333HMPLrjgAtx3331d/wgxMjIyZfmE8jh++9vfdo23cxzueekUIOsSL8MwxOtf/3pcd911eOlLX4q//OUvOOmkk6bcfzrXuDafz6VLl2J8fBzf/e53qas7scHZWH4H0eeSILY81mUtYF2+F65OkwFr/z2r1zoEMD3tujqmG4+d/u38Ljpv3jx/bWvDSSedhDe+8Y144okncNNNN+GSSy6Zct/pXGMcx7j44otxzjnnYMGCBXjZy16GI488Em9605uwcOHCynt++MMf4tvf/jb+9V//leosrmc2+8XFwcFBbLXVVvj973/f9Zpzsk1V9DOO43XuGtVLUAJYbeOSXsWkAaxV3cOZoFd3TsZYz3GsbSOWqa7xuTDVMdc0n1JKvPrVr8bKlSvxr//6r9hll13QbDbx1FNP4dRTT+0q9D8dlFKYP38+vvrVr/Z8fSpx6zj++ONx77334n3vex/22msv9PX1QSmFww47rOd4nu9n5sQTT8R5552H008/HXPmzJnyX92Hh4dx8MEHY2BgAP/2b/+GHXfcEbVaDb/61a/wr//6r5VrufTSS3Hqqafi5ptvxve+9z28+93vxtKlS/HTn/608kv1gAMOwG9+8xt84QtfwPHHH9+ziQzRm+k+l4wxfOMb38BPf/pT/M///A9uv/12nHbaabj00kvx05/+dK3q902HXs/vdJ7pa6+9FqeeeiqOOeYYvO9978P8+fMhhMDSpUvxyCOPrPU43PN47rnnTukM7PWPC44f//jHeO1rX4uXv/zluOyyy7DVVlshDENcddVVXQWsgfX/ud1ll10AoOv3XhAEOPTQQwGgq6mTm4Nrrrmm60uZe2+Z6cTytY2HvX73PPLII3jVq16FXXbZBZ/61Kew7bbbIooifOc738GnP/3pacVppRRe/epX4/3vf3/P11/4wheu8RjrwkknnYQrrrgCH/3oR7Hnnnti11137bnfdK9xbT6fS5YswW233YZLLrkEhxxyCGq12nq5RoKYDhvT7yD6XBLElsW6rAWsy/fCNX23W9vvWb3WIWZCuz5XrbquvPa1r0UcxzjllFOQJAmOP/74nvutzTW+5z3vwVFHHYWbbroJt99+Oz7ykY9g6dKluPPOO/F3f/d3fr/ddtsNw8PDuOaaa3DGGWesVyPVls5mv7gIAEcccQS+9KUv4ec//zle+tKXPqdjLV68GD/4wQ8wNjZWcS+6NE9n73Ur+p1dJp+Ls3Hx4sVQSuGRRx6pOCn+9Kc/rfMxp8usWbN6pi53Xs9Ui6obI7/73e/w4IMP4itf+Qre9KY3+e3O2eNw9/Thhx/uOkbnth133BE/+MEPcMABB/QUyqtj1apVuOOOO3DhhRfi/PPP99vdv3StCzP9zGy33XY44IADcPfdd+Md73hH1y9Cx913341nn30WN954o2/CAQCPPvpoz/332GMP7LHHHvjwhz+Me++9FwcccACuuOKKSqHdnXbayX8hP+yww3DHHXd0OYiJ3qztc/myl70ML3vZy/D//t//w3XXXYeTTz4Z119/Pd761rduFJ/xb3zjG9hhhx1w4403VsZzwQUXVPZbvHgx7rrrLkxOTlbci52f2x122AGAcbW4xbe14Zvf/CZqtRpuv/32ilPlqquuWutjAeaLXb1e7/nZn85n90UvehF23nln3HTTTfjMZz6DZrO5xvc4d/P8+fPXaQ6mOua6xkPH//zP/yBJEtxyyy0VV0Cv9KGpns0dd9wR4+Pja7yuxYsX44477vAFxx3rGi8PPPBAbLfddrj77rsrTbY6WZtrBFb/+Szv8/a3vx1HHnkkXv/61+Nb3/rWlPGaINY3G9PvIPpcEsSWx9quBTzX74W9mInvWdPVrsDqvxNNJx47/fvQQw/5+QBMds26ZBzW63Ucc8wxuPbaa/GP//iPXVlKjrW5Rnc955xzDs455xw89NBD2GuvvXDppZfi2muv9fvMnTsX3/jGN3DggQfiVa96FX7yk59g0aJFa30NxJrZ7GsuAsD73/9+NBoNnHbaaVi2bFnX62vjFjn88MMhpcQXvvCFyvZPf/rTYIzhH//xHwEYa/PcuXN9LSfHZZddtg5XYHDH/tznPlfZ7jpdrk923HFH/PGPf8Ty5cv9tvvvvx/33HNPZT8n4DsXVTdG3L8wle+/1hqf/exnK/stWrQIu+++u0/Ldfzwhz/E7373u8q+xx9/PKSU+NjHPtZ1vjzPVzsvvcYDPLf7uz6emYsuuggXXHDBauuQ9bqWNE27nv/R0VHkeV7Ztscee4BzXkktdbzkJS/Bd77zHfzhD3/AUUcdhVartc7XsSUx3edy1apVXc+f67zs7sfG8Bnv9Xz97Gc/w3333VfZb8mSJciyDP/5n//ptymluurrzp8/H4cccgiuvPJK/O1vf+s6XznuTTUexljFyf3YY4/hpptumvY1dR5vyZIluOmmm/DEE0/47X/4wx9w++23T+sYH/3oR7FixQqcfvrpPeuAdd7nJUuWYGBgAB//+Md77r+mOejFc4mHjl73emRkpOfCbbPZ7HnM448/Hvfdd1/PuRseHvYx6PDDD0ee57j88sv961JKfP7zn1/jOHvBGMPnPvc5XHDBBXjjG9845X7TvcbpfD7LHHroobj++utx22234Y1vfOM6ufEJYibYmH4H0eeSILY81nYt4Ll+L+zFTHzPmq52BeD/YbkzVk43Hh966KEIwxCf//znK+d7Lhry3HPPxQUXXDBlqjkw/WucnJzs6jS94447or+/v2fs3WabbfCDH/wArVYLr371q7vKLREzwxbxz2U777wzrrvuOpx44ol40YtehJNPPhl77rkntNZ49NFHcd1114Fz3rOuQSdHHXUUXvGKV+BDH/oQHnvsMey555743ve+h5tvvhnvec97KrXt3vrWt+ITn/gE3vrWt2KfffbBj370Izz44IPrfB177bUXTjzxRFx22WUYGRnB/vvvjzvuuKOno26mOe200/CpT30KS5YswVve8hY888wzuOKKK7DbbrtV6unV63Xsuuuu+PrXv44XvvCFmD17NnbffXfsvvvu632Ma8suu+yCHXfcEeeeey6eeuopDAwM4Jvf/GbPf435+Mc/jqOPPhoHHHAA3vzmN2PVqlX4whe+gN13372y4HjwwQfjjDPOwNKlS/Gb3/wGr3nNaxCGIR566CHccMMN+OxnP4vXve51PcczMDCAl7/85bjkkkuQZRm23nprfO9735vS7Tcd1sczc/DBB+Pggw9e7T77778/Zs2ahVNOOQXvfve7wRjDNddc0/XL+84778SZZ56J17/+9XjhC1+IPM9xzTXXQAiB4447ruexX/ayl+Hmm2/G4Ycfjte97nW46aabEIbhOl/PlsB0n8uvfOUruOyyy3Dsscdixx13xNjYGP7zP/8TAwMDOPzwwwFsHJ/xI488EjfeeCOOPfZYHHHEEXj00UdxxRVXYNddd618Ho855hi89KUvxTnnnIOHH34Yu+yyC2655RasXLkSQPVfdb/4xS/iwAMPxB577IHTTz8dO+ywA5YtW4b77rsPTz75JO6///4px3PEEUfgU5/6FA477DCcdNJJeOaZZ/DFL34RO+20E37729+u0zVeeOGFuO2223DQQQfhne98J/I8x+c//3nstttu0zrmSSedhN///vdYunQpfv7zn+OEE07A9ttvj4mJCfz+97/H1772NfT393uX/cDAAC6//HK88Y1vxN///d/jhBNOwLx58/DEE0/g29/+Ng444ICuf1RbE88lHjpe85rXIIoiHHXUUTjjjDMwPj6O//zP/8T8+fO7vvDvvffeuPzyy3HRRRdhp512wvz58/HKV74S73vf+3DLLbfgyCOPxKmnnoq9994bExMT+N3vfodvfOMbeOyxxzB37lwcddRROOCAA/CBD3wAjz32GHbddVfceOON06phNBVHH300jj766Bm5xul8Pjs55phjcNVVV+FNb3oTBgYGcOWVV67ztRDEurKx/Q6izyVBbFmsy1rAc/le2IuZ+J61Ntp17733BmCaei5ZsgRCCJxwwgnTjsfz5s3Dueeei6VLl+LII4/E4Ycfjl//+tf47ne/O6XrcE3sueee2HPPPWfkGh988EG86lWvwvHHH49dd90VQRDgW9/6FpYtW4YTTjih57F32mknfO9738MhhxyCJUuW4M477+yqdUk8R9ZfI+qNj4cffli/4x3v0DvttJOu1Wq6Xq/rXXbZRb/97W/Xv/nNbyr7nnLKKbrZbPY8ztjYmH7ve9+rFy1apMMw1DvvvLP+93//90qbdq1NG/W3vOUtenBwUPf39+vjjz9eP/PMMxqAvuCCC/x+vVqwa120tn/00Uf9tlarpd/97nfrOXPm6GazqY866ij9l7/8peuYa+KGG27QAPRdd921xnE4rr32Wr3DDjvoKIr0XnvtpW+//XZ9yimn6MWLF1f2u/fee/Xee++toyiqjGuqOXXnXRMHH3yw3m233SrbYFvcl3n00Ud7tqa/6667NAB9ww03+G3/93//pw899FDd19en586dq08//XR9//33awD6qquuqrz/+uuv17vssouO41jvvvvu+pZbbtHHHXec3mWXXbrG+h//8R9677331vV6Xff39+s99thDv//979d//etfV3uNTz75pD722GP10NCQHhwc1K9//ev1X//61w32zEw1l530urf33HOPftnLXqbr9bpetGiRfv/7369vv/32ynP35z//WZ922ml6xx131LVaTc+ePVu/4hWv0D/4wQ8qx+p1n2+++WYdBIF+wxveoKWUqx3flsa73vWunp+pNT2Xv/rVr/SJJ56ot9tuOx3HsZ4/f74+8sgj9f/+7/9WjjPVZ7wXvT53Uz2/U8WIzs++Ukp//OMf14sXL9ZxHOu/+7u/07feemvPeLR8+XJ90kkn6f7+fj04OKhPPfVUfc8992gA+vrrr6/s+8gjj+g3velNeuHChToMQ7311lvrI488Un/jG9+Y8vocX/7yl/XOO++s4zjWu+yyi77qqqt6xrZez7LWWi9evFifcsoplW0//OEP/TzvsMMO+oorrph2vHTcfffd+nWve53eaqutdBiGemBgQO+zzz76ggsu0H/729+69r/rrrv0kiVL9ODgoK7VanrHHXfUp556auUZWNtYPp14uHjxYn3EEUf0vIZbbrlFv+QlL9G1Wk2/4AUv0BdffLH+//6//68r1j399NP6iCOO0P39/RqAPvjgg/1rY2Nj+rzzztM77bSTjqJIz507V++///76k5/8pE7T1O/37LPP6je+8Y16YGBADw4O6je+8Y3617/+dc/fCb3mrvNZ70Wv32XTucbpfD6nitmXXXaZBqDPPffc1Y6NIGaCjf13UC/oc0kQmz9rsxag9fS+Fzrt9Ytf/KLyXhd7ylrbbV/X71laT1+75nmuzzrrLD1v3jzNGOuKydP5bial1BdeeKHeaqutdL1e14cccoj+/e9/3/M7ay+m+s5bppcmmM41rlixQr/rXe/Su+yyi242m3pwcFDvu++++r//+78rx+/1/fJnP/uZ7u/v1y9/+cv15OTkGq+DmD5M6+e5WwhBbEbstddemDdv3pR1IAiC2Pi46aabcOyxx+InP/kJDjjggA09HIIgCIIgCIIgiE2aLaLmIkE8V7Is66oNePfdd+P+++/HIYccsmEGRRDEGumsy+lq6A0MDODv//7vN9CoCIIgCIIgCIIgNh+2iJqLBPFceeqpp3DooYfin//5n7Fo0SL88Y9/xBVXXIGFCxfi7W9/+4YeHkEQU3DWWWeh1Wphv/32Q5IkuPHGG3Hvvffi4x//+Dp3MCYIgiAIgiAIgiAKKC2aIKbByMgI3va2t+Gee+7B8uXL0Ww28apXvQqf+MQnKk18CILYuLjuuutw6aWX4uGHH0a73cZOO+2Ed7zjHTjzzDM39NAIgiAIgiAIgiA2CzZoWvQXv/hFvOAFL0CtVsO+++6Ln//85xtyOAQxJYODg/j617+OJ598EkmSYOXKlbjhhhtoYZGYESgWrj9OOukk/PKXv8TIyAiSJMEDDzxAC4sEsZFCsZAgCMJA8ZAgiE2NDba4+PWvfx1nn302LrjgAvzqV7/CnnvuiSVLluCZZ57ZUEMiCIJ43qFYSBAEQbGQIAjCQfGQIIhNkQ2WFr3vvvviH/7hH/CFL3wBAKCUwrbbbouzzjoLH/jABzbEkAiCIJ53KBYSBEFQLCQIgnBQPCQIYlNkgzR0SdMUv/zlL3Heeef5bZxzHHroobjvvvu69k+SBEmS+J+VUli5ciXmzJkDxtjzMmaCIDZttNYYGxvDokWLwPkGrQjhWdtYCFA8JAjiuUGxkCAIwkDxkCAIYuZi4QZZXFyxYgWklFiwYEFl+4IFC/DHP/6xa/+lS5fiwgsvfL6GRxDEZsxf/vIXbLPNNht6GADWPhYCFA8JgpgZKBYSBEEYKB4SBEE891i4QRYX15bzzjsPZ599tv95ZGQE2223HZoYAAP9awxBEGtGQ2MCo+jv79/QQ3lOUDwkCOK5sLnHwgNxOAKEG3BkBEFsKuTI8BN8h+IhQRBbNDMVCzfI4uLcuXMhhMCyZcsq25ctW4aFCxd27R/HMeI47trO7P8IgiCmy8aUHrK2sRCgeEgQxMywucbCACECRmKaIIhpYDsPUDwkCGKLZoZi4QYpLhFFEfbee2/ccccdfptSCnfccQf222+/DTEkgiCI5x2KhQRBEBQLCYIgHBQPCYLYVNlgadFnn302TjnlFOyzzz546Utfis985jOYmJjAm9/85g01JIIgiOcdioUEQRAUCwmCIBwUDwmC2BTZYIuLb3jDG7B8+XKcf/75ePrpp7HXXnvhtttu6ypeSxAEsTlDsZAgCIJiIUEQhIPiIUEQmyJMa6039CDWltHRUQwODqIPg1RjjCCIaaGhMY4RjIyMYGBgYEMPZ8ageEgQxNqwucfCQ3A01RgjCGJa5DrD3biZ4iFBEFs0MxULN0jNRYIgCIIgCIIgCIIgCIIgNn1ocZEgCIIgCIIgCIIgCIIgiHWCFhcJgiAIgiAIgiAIgiAIglgnaHGRIAiCIAiCIAiCIAiCIIh1ghYXCYIgCIIgCIIgCIIgCIJYJ2hxkSAIgiAIgiAIgiAIgiCIdYIWFwmCIAiCIAiCIAiCIAiCWCdocZEgCIIgCIIgCIIgCIIgiHWCFhcJgiAIgiAIgiAIgiAIglgnaHGRIAiCIAiCIAiCIAiCIIh1ghYXCYIgCIIgCIIgCIIgCIJYJ2hxkSAIgiAIgiAIgiAIgiCIdYIWF4mNEsYYGGNd29b2/b3e81yPTRAE8XxC8ZAgCIIgCIIgiI2ZYEMPgCDKQlZrXfmzvE/nttUdZ3Xv7XXsNR13TecmCIKYCSgeEgRBEARBEASxqUGLi8TzhhOlvcSrE6uMMTAwaOjVCuDpHNv9zMAANvVxVnds4+rhYNBuAwAgEAJ5nkNptdpxEQRB9ILiIUEQBEEQBEEQmwu0uEg8r6xOSPu/szUL3KledwK64viZxvFW+34oCCEghEAURQiDEEEQgAsOzjhEIMA5x+TkJJYvX77a6ycIgnBQPCQIgiAIgiAIYnOAFheJGYMzU8JTo7fDZrqpdFPt50Svc96sTlCv7uc1CXoA4FwgEAIiEKjX6gjCEEJwCCEK9453/5j/arUaFm21CACQ5zmeWf7Mmi+WIIjNEoqHFA8JgiAIgiAIYkuBFheJ9ca61OUqu2R6OXKYzedbm2M78VxOE3THAWNeFHPOEQSB/69WqyEMI4RhAM45lFLQSkMqaf6uq+NgjEEEwhxLcMyfvwDQGhqaHDwEsYVD8ZDiIUEQBEEQBEFsrtDiIrHWTCV4NXQhUlez31Svd9KzHhi6t60uLbAspCt/QpsUPsERhhFqtRqiKEQUxQiEAOMMQgQIhIDSCkqaWmJKKyukuxstaK2hVfGzEEUz9qGhIS/mh4eHe46VIIhND4qHFA8JgiAIgiAIYkuHFheJtaLcEMC7ZnoI3PL+nT9P1QF1qm2d71vde6baz43bieQwDBBHMcIoRK1WQxzXvIgGAK00cimRSwkpc2ilobSClMal4/Zx116+ps7TMwZEUWxyBRnDwMCgf218bIyaIBDEJgrFQ4qHBEEQBEEQBEHQ4iIxDbrcLuUaYtDV1zsMN5XUux5uHLd9Og0GpuP86fyZgYHb2mBhGCIMQ0RRhHq97l05IjDNCaSUkLn0zpw0S/01OHeOksqL34pwtul+PcekS4sOWqMWx77LKrQR6a1WC1LKKeeAIIiNA4qHFA8JgiAIgiAIgqhCi4vEaikLVCdgvZAtaddO983qOo5O97xlIT7VeMrnLP+93IAgjmNEUYw4jhCGEYIgQBSFEMLUA1NKIc9zKKmQ5RmkVFBKenHraos5ka206roWreEFdTkVslzDzL2Fc+b3qdVq/lpkLpFmKfI8n/Y8EQTx/EHxkOIhQRAEQRAEQRDd0OIiMSW9nDW9RPF0RPK6NDPo9T7nvnFuIQZWjJMxBAEH5xxRZOqGGeEcIYoiCCEKAS2VF61KaWhXO8w2KJBS+pphPhXQCuaymC4LejfUsnjmnFX2YYxBKQ7GqtcVxzF4jaOdJEjTBFJKZFm2TnNGEMTMQ/GQ4iFBEARBEARBEL2hxUXC06uLqKPXts73rovI7jx3+X2dx/T1zcr7cI5ACN/RNIxCn+5Xr9XLJ4BSGowZIZxlWSXtzonhskMHAKRUyLIU0jYwcG4cV2esc+xdaO7PbzeAMWWupTTfnAuAK8RxhCgKkaYZGFrQ0CSqCWIDQPGQ4iFBEARBEARBENODFheJnil7ZTeM26dzW+f7ewni8vbVNSAop/2tKdUPjEFwAcEFgjBELY6NKycMEQQBhOC2yQKrNB8A4J06eZ4jz/NKN1OlFdI0A7QGt51NnWPGpzmqai2x1XV15YxDlRw53rWjzTwy+7MR8gpKFscKggADgwOQUmJsbMyPmSCI9QvFQwPFQ4IgCIIgCIIgpgstLm7BTCWa/d/ta17s9hLSHfus6XzA6ruidqb4db6XC25S+8IIcS1Gvd5AGAYQXEAqCa0BmUufyqdk4aipNF4oNSNwY5dSImkn0NAQQvi6X1KVHT1dI/cvdI5b8+oiASDAuXEXuWty1+U7rpoGquCcQ0kGxjgGBwehlcbwyIjdlxodEMRMQ/GQ4iFBEARBEARBEOsGLS5ugXhnTWcr09K2skCeqpmA28/V+ep8T89zg/nGB71S/NzxNLQv/s+56W4aBAFi58oJjIBm3BysnSTIs8w0GSjXDOvoXlrexli18YAbj5QKMpdgnHlnTue8AMaJszq8q8e+XykNzhmEMO9XqjvV0s2FUgqMcTBWiO6hwUFoaIwMW1GtFQiCeG5QPHTnpHhIEARBEARBEMS6QYuLWyjl7qZdKW2MgWmgh9Ze7XE6t/eqU6bRu+NpxfHDOThj4JwjDCOEYYA4riGOTRMCzjiUNp1M8zwvupuqogmBF842zc+5aKZKT1RKIxACfX19yKXExMQ4tDLpgF3jBQMXHJxxaGhIqaCt48c3Muiw9DgXEeesyz3EGAfnVVeSWXRQ4JyBcwHGikWKoaEhcMHx7LPPdl0TQRBrD8VDiocEQRAEQRAEQaw7tLi4hbCmFD2gEMZMV1Pm3Ps7nTuV93UeZ5pjcG6XsuvEdDQN0Wg0EYaBf69SyjcecMJZ5tKn8mkN6NJxpFSFqLWuH3esQrAyhGGAWhyjnSSI4giBVGi324AAwjDoUfesEL9KKnAuqwK+5P7xY7HphFprKKiOOVKAFgBj4JwViw5wnVQVAA6Xcujmf/bs2dAaWLVqJQlqglgLKB5SPCQIgiAIgiAIYuagxcUtgF5pfFM1DACqwrZX11O33aX09epmOh3x7lIHGZjvbhpFMer1mq8Z1m63faMB54pxArmzZhiz7h5ua3iFQeH+caLdNQ3gjINxs38URejv74eSCiuefRZ5liEMA0RRhCAIwMAgAtE1fplLSCERqMAI5U5BXbnWHMhLc1LaQWlT64yBmSYGLk2TaXAOSAmUO6oqpiHA/XUPDQ1heHiYBDVBTAOKhxQPCYIgCIIgCIKYWWhxcQugs/toma66XqiK4k5xXD5Gp5unfJ6p0gCdiPXnZybNLQgC735pTbYq7ptynS3f+dS82btvXL0xzrl30nTWAHOvA7ZuGRcQgUAYhIjiCAAwF6ZeGWBcOkIE/ljl+mJKaZt6aN031jXUJaat60YIjkAI31ihMBAxQDGfRugWMrwjSmkwXt3GGIdShduIcZMayMAwMjoCKanBAUFMBcVDuzvFQ4IgCIIgCIIgZghaXNwCcKKyVxdTJ946hXEvOgW2Y3WuHCf2AFPQX2kFrjiCMIQQpjGBqaFVFYtCFELYCd9yvS9X3N8hRABmmx04Me06nHam8JljmtfLxy2LdK0V6vUGoiiEUtqL5nKqYRSFvl5Yr8YJSkm/rVwDLU1Tu0+x+OA6tSqlASuqXcdUpq2YZ4BZC1BQipu/MwC6SB3s7+/H+Pg48jxf4/0kiC0RiocUDwmCIAiCIAiCmFlocXEzY6oOpu61souml5OmM+2v/J7O83SlFjrXjxO8toaX7xJacupwblL/6rU6avU64jhCGITgVkQ7keqFbqf7piKQrVOHFZ1XASOsezmD3Ni8kNVFowNX30xJhcA2S9BMgnPhU/eUZvb4JjVQKennryy2Xf0zpTSkzK1DR6NWq/maY0opSKm868eJb6ZY5Vga2jSVgJ13aChVpAq69EHGgGazCcYYJidbSNOk63kgiC0FiofFGCkeUjwkCIIgCIIgiPUFLS5uJqypppdzoZQdOaykPCsiuJTyN93z9Er7M9uVF9FhGKJWq6FWq2FoaAiNZgPNRhNhFPr3csZ9qlwnnHEvtsvj63XecmdUxpmr/V+4hrT2ziElFaSSpt5YaDqwuvcqbWp7Cc69q4mX6ntpXYzZOWvcud2xNey5Sk4eN26zSGDSBBkDONPQ4HAdW53Atif08yIh/f1znVqdmObcpErW4hgASFATWxwUD6tQPKR4SBAEQRAEQRDrE1pc3EzoJSyd2FJaVRwsfv819xjoLahLjpxKl1Ron57nnEBBECCOYy+i+/v70ag30Gg2TF0xKwJLRbcQaNsQoFSni5VcOt49BO2FpjsnAL9Ns2qDA601tCxEshPMABCIoGhuoLkXw0wZIS5LaYAAIGWOLM16OnF61RvzolgXrqXK2Oz5yvfRiGsOLooUzrKwdkgpkesi7a8srBkDarUalFRIs3TNN5wgNgMoHlI8dFA8JAiCIAiCIIj1Dy0ubkZ0dicti7bODqZTNTQo79vrmKxDgTuByys1wQTiOEKtVkOz2cTg4CAajQbiOEa9Xjd1xWxDAC8wrcgEAKVMWh8Y800NnIB3YlIIbp0vuS3yX3KrQFX0ObSGhAQUisYISpcEuQI0fEpguaaYqwvm0vWgNXIpvZh2XVHdtqmEc3EfijG5VEY/1h4LF25OVndPup4BaGilKtudW6rc+ZYgNmcoHlI8pHhIEARBEARBEM8PtLi4idKZqteZwtf5Wi8BPVW6X3l7Z8dTX+Pf1g+zO5pUOSFQq9XQqDfQ19+P/v4+9PX1oa/ZZxoWcN41PiNuAa2UF8ucldL3/L4AlIKwbhonmpUqRLaC9MX/tU2ZU9L+3aYWupS/QiAX25zTRtvaXmUh7RoO+NQ8qZBL6fdzTpti3ipl0Py8KlUWxd33y72vXAOul1uq+p6q68fPV4/3kaAmNkcoHlI8LN+jzntO8ZAgCIIgCIIg1i+0uLiJMZVY9q9bR0uZNYmxKd04bDXphaxoQiCEQLPZh3q9hmajaVL9mg00m03Ecc2k+3FWccgorQBljsVNYSyTdgfVuw4aA8Ctu8cW/JcyR5IkvhmAlNI2AjC1vLIsg5QSMpfIpfQiuCymC6Fc1EPrFKQ90/bKjhuLcx/ZF8B54V7yc8bM8cqdWM1hlH9f533rXADxzRdQvTfV+9xDzbtxKhLUxOYBxUOKhxQPCYIgCIIgCGLDQ4uLmxC9UvZW17ig06FTEWUdorszxa9TqDkxyGyB/CAIEAQB+vv6Kw0J6vU6arUa4jhGFEY+PVAr4/JRzDpecmXEJUrdSyWgOUweIGBS++xrMjdiOc1SJO0EaZYiyzK0JltI0sSIZiumfVdSK5xdd1ago7mATdUrX2+ngJ2qO2zn3PtmBF3bqvPXfQxUnEHm3KUdtK7USuu6P6XGFJ3jmQrNNJhmFXcRQWxqUDykeNh1fygeEgRBEARBEMQGgRYXNxGcWNMl8dOzPhi63R299tUdjhBXK6yXS8Zt45whCiPEtRj1Wh1RHGFoaAjNRhMDgwOo1WqIwsgLbd+JVJUEoQSkliYdDQqcCzDGIJWE60KqOfPpdlkqkeUZknaCdruNVruFpJ0gSRNkaeb/dKLZ/enG3cuB1Olkst6X6jwx+MUDM0duHsv3o/dCQ/nnXumancK8l0vIn7PkFCrXLVNKVdw6nddYvs7OunAmjRLQXIMp1rMTLUFszFA8pHhI8ZAgCIIgCIIgNh5ocXEjp5paZujVeKBXDbFu0dZdc8y/1iv1zuyMQAgEQYAwjNDXbJq6Yf39qMUx+gcG0GgYd44QAkIECIQAUDQLYNyIPK0BSOkbDgTM1BxTUhlxzRWgGWSeY2JiHK3JFiYmJ5AmKdpJgiRpI0kSn95XrvPVWQ8NsMLWCcySuHULB+Vr7PyxK+3OdU7lonIct2DgxHWnmC2nGHY5cXS1aUNnXbBy8wNTe6yocwaU6pkpk/InRCm1UFeFdFV02y6z2jSK0No0XSCIjR2KhxQPKR4SBEEQBEEQxMYHLS5u7JSEUUVouTS1UjpfWTz3coT0PnypvlhJdLrXwjBErVYzjQkaDcyaNQvNpqkjFkexcefEETgXgKsdxpiv51VGaZOWB8Cn95VFsMwl2u02JsYnsGLFCoyNj2FyctLWDVM+xc/PQcmJwjj3ItcJYSEEOBfgvEjnKwvgcnqgE6nObVMcl0Fw4f/kgvtjmuNzL06Le2M6qWZZCplLK6qL1Muy6C93ey2Lay+ibfqkuV7u38OtmwlaI9ey8kz4e9vh0PHbdLX+mDs3pQQSGz0UDykeUjwkCIIgCIIgiI0OvuZd1o6PfvSjhWvB/rfLLrv419vtNt71rndhzpw56Ovrw3HHHYdly5bN9DA2G5wLBaimtfnXS7XApqpB5p0rrOhO6h0osI4TxsEFRxgGiKIItVoN/f39mDU0C/Pmzcf8+fOxYP4CLFiwEPPmzcOc2XMwNGsW+vr6UKvVKml/Uua+K6nWGnme+5pgWgPMp/mZZgRKK6RZiuXLl+PJJ5/E4088jqf++pQR1GPGsdNut5FlmXH/oBCdnXPAGQfnHMK6i6IoRBTFCMOokqIoRAARGDHsaqaZfSOEQWj+i0LEUYwwCs17AudYChCGIYQIEAahcTIJYUQp4IV/lmVFzTOlbF0zVbkHnDMrzM0Y/LlDO5YoLP6z1xDHEeI4Ri0uxuYaREglO4qUwS6QmOtk3LmOqnMnAtHtYCKeExQLZx6KhxQPKR5umlA8JAiCoFhIEMTmzXpxLu622274wQ9+UJwkKE7z3ve+F9/+9rdxww03YHBwEGeeeSb+6Z/+Cffcc8/6GMomS6fTprOgfpnO7pnlY3Rts24RX1OLs0pDgjCMvABt2g6nzWYTtZpJ9RscGEAYRRCcVwRYrlJfq8rV+OKCF91QpW1Y4LukGlGptMLExARGR0bx+BOPY3x8HGmSIs/zrjmpiD2vBl0NLgUpAcU0hF0z11pDStnTseKOYcSsQGCFdXluOeMIwtAc23ZgBeDdNYwBqR2HUhpS5tBKI5fS/70s+l0aYPn+FPeIe0cRA/PunyAortmnCVr3loaGlApAG3lePX45hdAtGJS7tWqtAV518/CII0szcuzMIBQLZwaKhxQPAYqHmzoUDwmCICgWEgSx+bJeFheDIMDChQu7to+MjODLX/4yrrvuOrzyla8EAFx11VV48YtfjJ/+9Kd42ctetj6Gs0niUtacsOnVwGC6x/E1x6yILtfMCkPjDGk0G2g0mqjFMaLYOHWajSbqjTqazT7EUYTQOngAU5NKCOFrfTlnDmPm2FrDbIPZ5ruk2tQ2xjm0Umi12vjLX/6Clc+uxLMrn/WCXAjhxaNPjfP1uIo5Kf9pxKOs1MuqNDDoqiXGfTofdAwRCEipvDuKM44gz6G0cd44YVpOvXRitNw4wR684txx/5UFMeelemdMVdIY3T3yDRHsMd3cwu4jhK0TxjXCMARnHFJJZFkOpfLKfJWz/9zfBReVuY1rMZIkpZpjMwTFwpmB4iHFQ4qHmz4UDwmCICgWEgSx+bJeFhcfeughLFq0CLVaDfvttx+WLl2K7bbbDr/85S+RZRkOPfRQv+8uu+yC7bbbDvfdd9+UQTNJEiRJ4n8eHR1dH8PeKOjVXMAXxtfWYcOKGlSV11E4NDrT/fyxrIiO4xhRFKHRaKBWq6Gvr8+I53odcRx74Vyr1RBHkU2ZM+luJp0NkFJaB4sRa4IL27TApAI6YcyZEYRSGrdLEATI0gwrnl2BZcuW4cknn0SapL7hAefcu1wY4+BeLJtrKNfEMnXByg0LrDi1OyuoDrFdzI1SOfIcCITZJ9ABlDKuH6VUMQYw79Ipj6NTmJabGnDOIaV5USqTBiil8jsbs43pEOt+9vfLLpwwBQgRFMcGhwarpIY6B1QQhgjDwN8DxtpIk9JAtfYC2b3XiXDnDHLnaTTqmJxskaCeAWY6FgIUDykeUjykeLhpQt8NCYIg6LshQRCbLzO+uLjvvvvi6quvxote9CL87W9/w4UXXoiDDjoIv//97/H0008jiiIMDQ1V3rNgwQI8/fTTUx5z6dKluPDCC2d6qBsdnAtorbygdi4XBmYFUlUYOzo7ovYS0K4Iv0n1C9Hf3496vWFqhMUxmn1N1Gt1NJoNxFEMLjii0NS4cm4S51hRPi2uu0GBksU+rtC+VNI6XWDEuFR47PHH8Oc//xmJTfnzaWtKQ0F5YezqZzHGKgIXACAZGCscMG7BoZLmWJqX8s9mW2nsyqQMll0tSiu/IFCeT+XvUbXxQdllUxbIgEmNLN9Ds+BQdef4+wUGMGVSJpX2CyB+gcSOxS0+BMI0Vig3bgiCwKdSBkEArbV170jvFpKQ4Lpad87NU7PZwOTkpL9+Yu1ZH7EQoHhI8ZDiIcXDTQ/6bkgQBEHfDQmC2LyZ8cXFf/zHf/R/f8lLXoJ9990Xixcvxn//93+jXq+v0zHPO+88nH322f7n0dFRbLvtts95rBsTzLs3ip+BoiaYwwuykvCppPmV/g6gVMg/9s6bWq2GocFBNJoNn+JXr9d9p1NX+6PSSVQDsM0GtHXJuMHKHp1QgaIDqhPSgnOMT0zgiScex+OPP4Hx8XEEQthz2BRHVtQ8K9fdctepGazrRHnnkqsNVmnwwFCk7GnYWdQ+vU9Dg3NzjVxwKFXUBSvX2Co6lFbvD6CrYlkDWqnKvSq7X9yxzNDsnx0iuTiyqcGmNas4ecouriJ9sKhxBjt/rsaaOy5nHEy4ewEAEkqa65JSQnNt64+JyjiazT5MTk70rPdGrJn1EQsBiocUDykeUjzc9KDvhgRBEPTdkCCIzZv1khZdZmhoCC984Qvx8MMP49WvfjXSNMXw8HDlX2WWLVvWs/aEI45jxHG8voe6wXDiTilVaTBQ7gJaFkaeDtNO2dFjOoQGaDYbppOmTfVr1Buo1+toNpuoN4xTp+ggGvi0M6207fCZIxDCptAVTQGcwFRWSLs6YkppMK3AJLPXpE0dK8YwMTGOhx56CA8//DA4ZwiEgFQSHEb0OmHt0/lMMa3KPHEB5BLQSnrR60RqGIb+up34dM6hPM8LFw6rLk5oWYhwL55LrzPGwDS8M6kQx933suKkcnPGSj9DQzHjYOKKQzOAMV24szrqqjnXj2u0AG2cTBI2xZKbRQhXk8ychtt0Rm3dUhKCmRRBBOY9OTNdW2Uuobgyc1Y5hqnXVq/X0Wq1SFDPADMRCwGKhwDFQ3ftFA8pHm6q0HdDgiAI+m5IEMTmBV/zLs+N8fFxPPLII9hqq62w9957IwxD3HHHHf71P/3pT3jiiSew3377re+hbJQ4IQ1UXTllx0cZpa1bxtbfqtQWgwbnAnEco7+/H7NmDWHevPmYP38BFixYgAULFmLevHmYPWc25sydY9w6DevQiUIIEVScMSYNLkcupRWV8M0KymlujBtHj8wl8izzYtS4R4wof+aZZfjDH/6AJ5543LyHMe/g0dp0FAXK7hpVWkgomgb4zqMwjhvT2TTA0NAQ5syZg1mzZqHRqCMIAkgp0U4SZHnmxbRWutRMQNvFAZOyqKSqNiGASc0spxQWorf3/el0VZl0SAnt/q5NgwRlu6jmWebFvpQmXTJ3tdukgrQOosztJ+12+59SJiXTHcMsgLjr1d7VpJQ0Lh4uIIRAGIQIwtA0RLALJ2mWIkvte2WR0hlFcaWTHbFuUCxcMxQPKR5SPNwyoHhIEARBsZAgiM2LGf+GfO655+Koo47C4sWL8de//hUXXHABhBA48cQTMTg4iLe85S04++yzMXv2bAwMDOCss87Cfvvtt0V2wGKsJJ7L7oxSuldnql8lZawj9S0IAtRqNTQaTePKadQxa9Ys1Gt19Pf3o1arIQhDBEGAZqNhmhLYNDitjZjMy0IYpoC+EYPGRZLlGQCAs6IzqhF8ElmWmZS70hCllHhm2TP486N/xsqVK5HnOYQw4rjRbCAIAusEScBCk/pXLtBvzqu860SVu5AyhlocY+HChRCBQKvVQpqkmJiYQJqmkFIWc9hZd61DGJfP6V8vLVQUU977752pfJ3bOsV3pcutQnWZ3+3rnDtwjSlysFJ6o7lPxkmklAbyjsnX2u/TOQYuOEJuHECutpqSxgHEGIcW2jc+4MJ0XgVAjp21gGLh2kHxkOKhmWxQPNwMoXhIEARBsZAgiM2bGV9cfPLJJ3HiiSfi2Wefxbx583DggQfipz/9KebNmwcA+PSnPw3OOY477jgkSYIlS5bgsssum+lhbNS4VL6ykAaqtah8h0zornQ3DzPuHi64FchN9A/0o9lootFsoFFvYPac2ajVahjoH0AYGneGsDW8ykXvZS6hmS5Ela3h5ZoSSFv43qXJpUnqXR5OiHvhZ1MEk6SNVStX4am/PoXx8XFEUYQwDCGlRBhGqNdr0BpotVo+dRHWORMIU+/KuXkA0xU1DEPowJ6fAbVaDWEUYnx8HCMjI8jzHO122y8GMKZ904NigaK7dlv5HnghD90z16+XSHa1yzrva9HQAF3v6WymYFIXAW0bMXhhrHVlIcCJYwBgurtBQ9cijAJkabGg2mSBeXGumLL3USGX7lrtn5xBlO4JsWYoFk4PiocUD7uPQ/Fwc4PiIUEQBMVCgiA2b5ieKp9pI2Z0dBSDg4Pow2C1ptMmQi9XB1AIPK1NOh9jVZeF76xZEoGFO6eBgYEBzBqahWZfs2hUMDRk0vzCyLgzAuO4YNwKP3d8Ww+sSEkrivlnWWbqiHHTUEBJhXa7bcbEua9JJpVN5WMcUko8/fTTePKpJzE2Nm6Op4sxu66nANBut5GmmT+nuz7nQuGMe4dRGAZgjGFiYgKNRgNRFCFNU7Tbbd/9U0nl67TBL1CorsWITjdOucusm/Pyvr1EcCe93Dvl+1p2YZXfU+l8yoqGBe51p611rxpkdvzlemSd4tr83TU2qF5T+Zg+VbLUybXcaIEx0+xhUxTUGhrjGMHIyAgGBgY29HBmDIqHFA8pHlI8XBs291h4CI5GwMINPRyCIDYBcp3hbtxM8ZAgiC2amYqFVDhoA9OZalZ2dbh0N7etnCrIGUcYhWjYJgR9fX0YGBjA0NAs9DWbqNvaYc2maVAAmHpZTmi5bqZZnvm/a60hlakr5lL8nFNHaQWugNzW4srzvNIV1QltzgUmWpMYXrUKf3v6aQwPDyPLcu88iaIQtTj27qMoiiGEgNbjyLPMC3In3tx1RlFk6mIFAaSSiKIIggtMTExicnLCi8Jyep0XmB1z2H0TTIOCclpmz3tVEre9fu6k4vgpUW5C0enqcWP24hWmQ2qhva2qdpfphLK216uBkpHH7A8nnN19Lgn8DreS1uaYHNy/7htUcA5jptrk/j2C2ESgeEjx0O1L8ZAgCIIgCIIgNh1ocfF5ptPlUREzZdGlS/XDSm9xAiuMQvT392OgfwB9/f3o7+vDwOAABvoHUKvXfddTJ0AB2wnUCmkpTd0wmUtIpSp1xLRSRRH7kjh1dcRcsXtpC+1raC/UZS6x4tlnsWLFCoyMDKPVagOw7hxhapq5lELAuHQCIRBHsd+X80Js9vX1odnXBACMjY1heGQEgRAIoxBj42PIsszUEgOruEumI4rdnDMrIP28T+XQcTeiJMp77dtLZFdcN6g6dcqOIbOPgpSFawdwKYKFyHY1yYpxoBDYKFw9LnWTsdJ+gF9U6WzYUJ6Xrm26OB9nvPJsEMS6QPGQ4iHFQ4IgCIIgCILY9KHFxeeRNaWA+f1KYrDXa0EYol6vo9lsoq+/HwMDRlT39/ej0WyYFMA4BuOm/pjDpcdprX0al3PmaA2fQqds907n0gHgO3OmaYI0zZCmia3/pSG48MecbE1ixYoVGB0ZRTtJAK0RhCHiOPLXlaYpOOcQXEAqacR0LUYzbwAAoigy7h4AzWYDjHEkSds2SFDIcmUcRjZlrbzg0DmfPnXOC9ZuAegFdcd9cpQ72Lrj2HJo3u8yXXqlGfa6x6Wd7D0rXivXhqv8yZ3bxh23cphCSaMQ2sUEVeejlxvHdHU1aZ+MM3AtTNdVglgHKB5SPKR4SBAEQRAEQRCbB7S4+DzRKX6ANadTMWYUm3NEMMZ8TbH+/n6T9jc4iIHBAfT19XmBHUexF9FKKqiSgHR1xNIs9a/neW6aEWj49D6gqBuWpRnaSeLrj2VpiizPkGWZP26WZWi12hgfG8P4+DiSNAFjQGibFnDOzfGkcQVxxiFqAkIEYJwhiiLMGpplaokJgYnJCeR5jiRJkSRtpGlqa5+5ufCTZDLjyumTHfW3OpmyVphNB7RvrqTxuYUApq2Q5LajrCpS8MrHnOredqZ9+vtiXUa9xtrpPnLOpPJEcG5SJuEdPbYZhOsmy1iXcC4vNFTGqOHnsrKwU5pPxhgE51CcbXL1xogND8VDioflMVE8JAiCIAiCIIhNG1pcXM+srtj9VK97YdWRRsYYR61eR39/H4aGhjB79mwMDc3CwEA/4jhGHMWI4xoEr9awUkp7R4WSCrlNAXRi2dQPM2ldZXeOlBJpZpoDTE5OIs9y5Lb+mE8ttM0MJluTmJyYRKvdQp7n4Nw0SwjC0Ig2ey5tXUBSZ1AthSAIEMcR6vU6OBf2dSPwW602kqRtUhKV2Q4vcLvnrbz4sLp57+WM4YKDcwEpzbl0xdWiC3FtzT6ql+On5BLqdPtU9utwwvRya5XH2mtRwItxbYS+khrKrgRwziAEg1PEiimf4ukUtD+91l2phloXwrlXuqJbXOCCI2CmGUW73e45foIoQ/GQ4mHXfhQPCYIgCIIgCGKThxYXNxCrE1B+n1Ixe845giBAf38fZs+e7f8bGBhEo1H3HUeDIPAdTwEjdqXtZunqSTnnj9LKb8ulRG7rh+VS+iYGaZohSdpoTbaQZqmvPRbFEQDj0JmYnMDk5CTa7TakVBBCgHOBIAgghHW0QBU1qmzKYJZlaDSaCIMQQghkaYYkTZAkCcbGxpEkbUg5RZpZRQQagcetEHTzCwZAdwhn5otzdbllnJic6n6V74tWurfInUJQlxsUrE0DAHcdPRtdlH7W0F5Ya23uO5hNGdTM15SrpPx1/t3VL/PzYRZwyvXNynPBuYAQHAKCxDTxnKB4SPFwOlA8JAiCIAiCIIiNE1pcfB4ouxucmJqqrlg5nc3hhPTgwCDmzJmLOXNmY2hoyAvpOI7BGQfjHFopL7CU0silNMLROmUAo5t8B1SlwRiQZybVL0nayNLMp/zleY4kTdBut5EmqW8+kOUZGBiSNMH4+Diy1KQEhmEAwUWlDpZxCinjwrE1zBgYwjBEvV4DGDOpg0mCVquFJDHnc44hM0HdKXudc+frcHUW13KuJ8b82yo60rqZAGn/nOI+dqRwlhsRTFcgr21nUX/8KeqYlZ8pt7/WZnGEcwbGhF2METDXB19gzD9vqz2eOaZJMWSVTELzmhHbtVqNBDUxLSgeUjwsv29toHhIEARBEARBEBsntLi4HlldapdDa22ECjoEtduXMwgu0Gw0sWDhAixYMB8DA4Po6+vzQto1EFBKFmlyNh1OyryUggdfd8rU6zLdTRljaCcJ2q0WJlvGceO7oCapd/EkaQLOORjjyDKTWpYmKdLU1CsTQpj/uAAYg+nyacbgnEIuRS+MTBOGMAzRbhkXUJqkvpaZtp1cWUn9dtbVKk2oTW0zIr1cF6vSnECb2e3l0nEdSTvv03RS9FYndMu1ylZHZ92zcm0xd33lWnPdLqPqAoJZIOBgTPljcMahmcRUw6l0cNVuDApac2hWzL/rrMqsuBZcoFFvQGuNJEnWeK3ElgnFQ4qHFA8JgiAIgiAIYvOEFhfXE2UhtjqxNWUqoN0eBAHq9TrmL5iPrRZuhVmzZ6PRqKNWqyGOY5P2p+HFZ+66VXpRDZ/WxxjzaYFaachcIkkS5FKi3W6jNTnpa4UltvtplmZeiEuZQ4gAYRh4gS6VtAKb+UUBl16olBHTpkaY9I4R7+ZhHK1Wq2hOkNt0RTtf5a6k5Q6oThwXzqYe9cZKItThhPVUbqnO+zfVPesSsqsRze4elxscrE5c93ytQ/z756okwMvX668P2jd84LxITWSicFCVLrjj2gtBrZVJ4zSl6zgAJ6SNK8g4ljiajSaJaaInFA8pHrrXKB4SBEEQBEEQxOYHLS6uJ7oK4ANd6X1mW9U5UhSON+l/zWYfFiyYj6233hqzZs1CX18farUawtDUE/OiHQzKunXK7h+f8qU0pJZQ0ohuKaWt5TWGJE0wMTGByYlJ79RJk9R3SnWZdEIYEe02OIEchiGUMk4ZJ6KdK0dJk5ZoXDwBAmFqj+VSYrI16R1DJk2wNC1WHDIv1krupg6B3enOqYjrHoK6876sqdZb+b64Mbkfy8frFMpd9cB0WYRW31ceTxnnzuk5ps5xl58jJ+Dds6GKDqouRdN1qq0cAubZUVB+Gl2DC6kVGFMQXJQEuWlKwRiHECZdlbqlEp1QPKR4SPGQIAiCIAiCIDZfaHFxPbA6cVRO8XNip3M7GIMQARqNOrbddhvMnz8fs2fNRqPZQK1WQxSFxsWiAa1MzS6XBuiEjIRx37TbbX9sJ8y00mjZlL/RkVG0kwQTE+NoTbaQ5RnSNDUpgtLVJCucLZIBkAzCCjLO7Th0XnF2mA6s5nymA6pxFXHGTe2yLEOapEZwWyFuLp0X9a90h1jtmLtyimBPN9QUIrrsntEd81+uBefuJQMD4+VzFwK9Wtur2FYRzKvR6qt7VtaUPlh2G1UWGDqO3eucjHEEwtxPlx7qar917ude08pdFxDykttHK4ApQAs0Gk2Mj493iXRiy4XiIcVDiocUDwmCIAiCIIjNG1pcnGHKqWud24EeYq1DILpmBc1mH+bOmYNFixZhaGgIfX19iKIYYRgU4ljZgvtW7CiloaTCZGsSI8MjGB4eRpIkYJwhz3MvZt25kyRFu9Xy9b+CMIQIBGq1GhjjkDI3rh2bpufEcyBMDTFXX8qIUm7dNm4sEpxxhFGIOK4hjiMIISBtyqGGcft454+fhKoA7lVrq3PeHF0pgYx5ke5eL7t5mC4Jzg5HjxlKIYYZmBedYIUzqJdg7bVN695itzOtcDouns4adas7L1CkPyoFaCaBHGBcgXPu75/MS+KXMbghCWGf21zbRQ8NQCLPjXPL7GrmTlrnVrPZwNjYWM+xEFsWFA8pHlI8pHhIEARBEARBbP7Q4uIMU657ZUo19XBTlIV0ubYTGKIoQhhGmDVrCAsWLsCcOXNQr5uaYkKYulyMMSitbKF/e0zrjmm1Wli5chVWrVyJlatWQimNRqOO0dExL5zrjTrq9TqU0r65AACfzhWGAaIowuTkJFqtthfGjHFwbrpkupQyADadTEHm0gvMKIwQxZF1FsUIhPCNEwoXTJGu1iVcS9t8fTHdMbdu3souGVs3i5WcJKxjju1Bi23cLBK4c7m0SuM8ci4oVYxRTy1c/XOgewvtrv3KaXtao9Od05k62CmuO7d3vV5edNDm6VRagSmzXXBhJ8EKaDu35bG7BR7TzMDca/PsGReWd0ppBkD7OZ9O4wZi84biIcVDdxyKhxQPCYIgCIIgiM0XWlycQcqimVn3Qtlt0tNV4VLemHHAxHGMRr2BefPmYe7cuRgcGDRiz4poxphJsVPGgeI6nJpOqEas1us1hAvmgzGG0bFRDPQPWIFjNGQtjpGkCZIkQavVQpZmfpyuYUKSJEjTDHEcIQgCJEmKWhxXhKyGEZntdtt3YQWMwIriCHEcm+6tIoDWCmmaIk0zKFW4Zzjj0FyDKVZNQTOq2k6RGRtnvJKO1ylWGWPQzDiYuBbFOO38l7uLls/jmjG4Ywsu/N+daHT11MoC37meet3Xqdw0ZaYrNjtrofVKPeykMjcdAtktgriFEbewwHkxZ8w1o1Cufpjw7i2tjSNMMQmlikWLQDDv2qnXG2i1JklQb8FQPKR4SPGQ4iFBEARBEASxZUCLizMEZ9w4ILzAKeo1FWls1feU078CIRDFEfr7+zE0NIT58+ZjYGDQdj+1DQACgcnJlk9FS9MEkxOmCUBZcMdRjFqtBs44ojgyf0Yh6vW50EpjYnICIyMjSJMUSWq6WQoufGH7PM+RZRnSNEOaJgjDEGEYgQsnsDSyLLN/N2l9fh44AxdGfDn3i5Q50jRFYpsiGAeMhuACEhICwtYqc6lm1aL9nHEvCMvumTK+HpabW60AXW0WUHbD+HtSvg/aprNBQrAAQhhnkuswWj7X6n6eTmreutKr6cLqzuHrhmmz0MAYbHMC1/hC+8UN10G1cCRpKK0B5GAsAOcMWvOu2nNa2UUjcHBmBDXnDIODgxgZGSFBvQVC8dDOA8VDiocUDwmCIAiCIIgtAFpcnCE6u1i6dDWtlE91A3oLLSEE4lqMRqOBOXPmYM7sOZgzdw4a9YZxhAiOJE2RjieIwghgDONjYxgZHbF1vCIwxiCVgpRFZ0ouOObMmWPqhGUpJicnMTw8jImJCSilEcURmn1NBEEAmUukWWocGMrUn6rZpgPtdhsyb0Ep6cV9krR96qFLEwyCAGAMURSjXq/79L88z63jw9Ss0to2KeAaAkZQa1u/qiyUeYc4diLcveYdUWYyoUtOHG0dUEaIl+5L2UHTUQvOuXGgGfIsg5RFemCv97v7XGmEUHLqlNP6utxBPRw+U9UW61VnzB3HGGN6O8HKPyutfPqiWXTQlWOb6+e+MYXWGlLa1E5h9hNCgIH57qlmTO583Lueqi4nEtJbIhQPKR6WX6d4SPGQIAiCIAiC2LyhxcUZoJwOZoSK8kXjy44Qn0LnBKHg3gXTaNQxe9ZsbL31Nujv70Oj3kAYhWCMIcsyKCXRbDYxMjKCVatWIUkSNBpNNBsNMM6tiJbQyrppbDdLrTWyLEeapBgeHsbY2Di0VoiiEGEYol5vAADGxkbRarUgpRHMURghjEI04jqE4BgfH0e73QbnAoEQyLLci2NobWqgBUZsxXHkhbTSCrmURbF8DS/KmLINDCSD1jkEL5w1zv3hKNfaKhxQvJzF1lO49momUabTrVN2WWmljWgsiVXnyOo8T9d5O4RqxS3UMabVCelex+q8tqmcOp0upWIs5vqUAhgrC3FlHTncz021VptyB/bbpbSNLZjwnwGuOJgwr/f19WFiYoJE9RYExUOKh13npXgIgOIhQRAEQRAEsflCi4szgBM1ZWeJw4k1Llyam6nZFAiBIAwR21pcs2bNwoKFCzBv3lzUajWEUQgAvnNpX18fVqxYgV//+tdo1BvYaqutMDDQD6U08izz3VGBQkhLJSGlxOTkBFY8uwJJkiAMAy9UZS4xOjqCdruNJEmglHHpZFmGLMvQRNPU2wJDGEa+jpiGqUUWBIGvOxYI4Z0zQghAa2S5OU6e51BSQSpZuGk4A9MM4IAAB7TwcylziVxLMKu/XHplZydZpmFqWsGkt3W6W1ab8ufujXPW6A5B7U6lWWlfe17Wfazyebuejw5B2yksVyek13SM6mJN16mnpOxyKkS7qVmnlXYTWjQ4AHy6IGMmbdTcT3fdClpXUxOdq6uzAyyxeUPxkOJh+bxdzwfFQ4qHBEEQBEEQxGYHLS4+R7gVG1qrop4TSsKKFU4HWGERBIEX0fV6Hf19/Vi09SLMmTMXzWafESE2xY5zhiAM8ac//gkP/N8DWDB/AbbffnvUG3VkWV4R0k7sSiWteFUYGx3FsyuexejoKEx3U5PGJYRAu93G+Pi4LT4vEEURtNbI89yI8NakF0FBEHghDgBRxBCGoRHODgWIQEBKibZKoLVJI5PKCDQv+LX2otSLQ2HT4pRZeAhgFwW0BteisONYurp+asDVGNNaVsRoRRx30sNB4lP5YFIL3bxJKbtEbjntr3ObHycrid/On6cQz73O0XmMrnlAddvqBKy7Np/G54/DfMpgZXHIzbFzU5VPW1okEqJ4D4OpN9fX12ees45UWWLzg+IhxcPyn11/p3hI8ZAgCIIgCILYLKHFxZnAimTGmHeVuKL75mVTs0uIACIQCMMIcRyh2Wz6hgVzZs/BwEA/wtDckiw3tbuyVGJkZATLVyzHNttsg5133hmCGyGcZUVXUylz3ynVdC3VaLda+NvTf8P4+LgXvVqb1LbxkXEkaVKk0JVqTnHGoZgR1RPjEwijEHEU++uRSiIOTYME59xxLp88z/2xXDpfZarghBeriDVXC0xzDa45FFNefCsloRS8CAdKgrEk0LwrqnRcdz/86zZv0P2v9+2s1vVisEKydI+BIs2uV90xd63uZz8npUWEcppfJ1Ol9unSHLjr7KxXVtbX5TH2HEfHWCuLEExXXte61MWWmfprSrnab7DinIHxIhVVoLTYQmwZUDykeFiC4iHFQ4IgCIIgCGLzhxYXnwOsJMyq27ntJGlEmBPSzu0SxzXU6zU06g00m000Gk00mg3EcQzOTPdJzhmgOZ4dfhZj42PYdptt0T8wYBw0E5Ne0PnC/8p2FrWiJ0naWLVqFcbHx6G1RhiGYIwjy1K0Wq2iK6rgXmRJKb1QCoIAUsLUEstNOpirK2UEmkKeW/eNNHXEpMyrxhdjn/FdN4v32sYCohCuwi4iuJphxi2UgzFlM9KMWHcdTP3xprDgOEcJpEvd62hEUHpflxgtbfc1wTSrbCuntvUS1GUR74/TY7/VpQS6+Sqn+VXEbjmlsST+e17L6txKjJmUy5LId+l9xhRUXG+xECJgarxpMzUltw63ItylvQJArVZDu90mt85mDMVDiofmMikeUjwkCIIgCIIgtjRocXEGcOJQw/zp9JLWRqyGYYRACIRRiCgy6X+NesML6Eajjka9AcEF0iyFlBJCCEy2JiGVwqyhWYhrMfI8R7vVglSyOLdNr/MOHakwPj6OlatWYnR0FIBpGBCGESYnJzA2NoY8zytF6ivOGqvDzGJAcY48zxEEARg3QjvPc9/1VEpZctRolOt9lYWnW2RgjEHzoosmZxxhGFTqWeVB4N1HSiov5KSU3pFUdsx4sewWN4zC67xR5g/oLhFedvO49EM3NnO8bsfRWovWzv16nL/ztbKg7dy/p0uo5ADqPG752BWh71IBWTWt1e9r7xnTrNJYwu2vde7dOkpJSAYICLOvTUslthwoHlI8LD8LFA8pHhIEQRAEQRCbP7S4uI6UHRRdgkUZ94IT0vV6zdYVqyEMA9TrdS+mB/oHMNA/gDAKTV0waYTH2Ng4kqSNRqOOOI6RJAnSNEMuTXogg0m3ckJaKtPFc7I1iWXPLMPY2Bi00uCcg3Pj0JmcmESe550XYtIYrUtDSmVFbPGy0gqMm3pRppaZRNJOjJi2Th1/7VbIehdLSbgyrqC1GY/WZmwMRliHYYgojPx4cimRZ8X1OqGvbFOGPM/9AoJzF2lt3DzGvaSK1D10O6p6Cmp7HOcy8S4dd12oitry62U3DdC7mYU/9xQifKrUv8r5KymNvY8xlTAvj9uNV2nlHUBci9JCiPDvFVz4FFBwXVlIMC6sALnO/fncYoc9mXHxaIUojpAmKbl1NkMoHlI8pHhI8ZAgCIIgCILYcqHFxXWkV/pZWdQFQqBeb/hGBUIIxHENtZpNAWw0MDg4iKGhITQbTX8MzjiSNMHExDjiKAbnAlmaIU1S2wjAiOcgCABtOolKaVwr7XYbzz67EiMjI5BS+RS/JEnRbrWQ5Zl36CilILgA5wxKcS9YAVQEmdZFJ1QAkDJHlimkaVrUD2OuGL77uXAA+bmxrhwplX8t1alJjeShnaPicRSBQCBMB05Tx0qXOqzCXLt1gLhaV1maQSqJLMu9a8jhxueFZNlOVabDJdPpiCmLd5fqWX6r77LakSI6lajtcuVM8Xrl3vRaBJhi3nsxlZvIpHbmNm1V2O6mHWPmzLqg7HyCQQgOpVjRndc6uxgvpY9KVBZdiM0LiocUDykeUjwkCIIgCIIgtlxocXEdKAsroFtYh0GIMArR12wiiiOEYYggCFCvN9Bo1FGr1dBs9mFwYBB9fX0QwqTVaaXRbrcx2ZpEEAQQQiBLU+PCsTinjXfoWLdKmmYYHR3F6OgItNYQwrhhsjTzx+RWPDuxzIUT1rIiHIUQ3p2htfLdWjOZ+9pfTuA6ceecI16o6m4njMMJ4oAZkV6r1UyDhArC1jlT0FpBSgXlxDMrCuRzbtwlWmmkWYo8z5GmmRHTUiHLM/9eZR08nXSK5c77POW+JWdM2fnia3SVnw9WFcRd6XhTpAUW4r/Yp5fLaCoBPVUa4FTC3TmdGOMIAldvTHWJee8UsveaMw7wovttLiUEOBhjCFgAzjhyna9R6BObHhQPKR6W/07xkOIhQRAEQRAEseVBi4vrQJdItIICMKl/jWYDURSj2df0wjoMQ9uswNQVa9qmBUIYx41WJoVvYnwCAFCv141gllbocu5rbXEukOc5siw3wiXPMTY2htHRUaRJ6tP5zD4p0jQFYFK2nBBlnJtmCVZ0wgtpjjAw4t+l+GlotJOkSMdTVSFYSbGbIpVNK+1f00qDC46+ppmDer0OxllXJ9XyfAvBEQgBLnhFjJbT7YIwhNbGRaSsoynLs8qfzvkjpeqZGugoC/bqrWa+Lpf26YI2pZFxBMIKxzxHmsmKmPX1yjquzTU4MFPZUcus5Kqp7FOa5rJo7yVW3fPamUpYqUWGotYcmHNk2eOXHFmwKaGCmwYG7r0iEEAOKCuYXSqlso01wJjv0OueaWLzgOIhxUOKhxQPCYIgCIIgiC0bWlxcS7yYQlHriTEjrITgqNfqGBgYRC2OUW/UIbhAo9lAEARoNvtQq9UQRSFqtZoXxUpJZFnmu556IW3rNDHOkaUpsjwzIkWZ+lmcM7TbCUZHR7Fq1TDGxkaN48eLGIk0Sb0A51zYlC3T4ZJxBpWb1DohAiNYgwBRHIEzbgSoFZ95lsE1IShfv604tdoFBv93bUR0GEao1WoYGhpCEAS2dlmpTpkdv/m7snPMfVqjuw/+VDbFMICARpHmqAMNkQsopRCktumCrU/GkNm0ym7nTXnMvV7rTCPUGlBQEJxBCFNLrlwfrWfqXg+3V1lUrwudQrqzTtpUuLEVCy2meUWWZkXzgtL9LKdBllMtUZ4fe35oDSlz3x3YpRi6VE1i04biIcVDiocUDwmCIAiCIAiCFhefA4WoMCKv3qijr68PAwP9qNVqvjZXs9ln0wBriMLIuBoYM+l0UiHNUrTbbWitUa/XwblAkrRtIX4rijPjtimLIyUVxsbGsGrlKoyMjqA12arU3dJaV4Qy54VQczXDnIhtNOqIotgLHKkkkiRFlhnXixPMjMHWmCrmQATOgeG6lRY62r1PBMbZUYtj9PX1oVavoxbHxg2kNAJROG6qqXjcXg+z16P8z51pfe5aXBMIABVnkWmuYAUhZ+BWeGtbs8ynMRYnNz/b1ypiu0dKnnG6mOsPoxBRFiJNSucuPzdlnAjV3cdbYzOCksNmqrRFc4re3VS7D8j8vhoaUAATrMsF1LHmAMC41KAD5Hb+dfl6dOHWcbXL0pTE9OYExUOKhxQPKR4SBEEQBEEQWya0uLgWdDkurKLgnCMMQwwMDKC/rx/9/f2IwghccERRjDiKTKH+MAQXJv3OOBiMayRNUmRZjlocmxSyLIPWRS0u7Zw1YD41L8szjI+PY9WqVRgdG0WWZn6M5VpRQgQIhPCOC2VdPlEYIc1ME4Jmo4n58+YDjGFsbNR3Ps0zkzrnFgyAqkByx4zj2HbETCrzYv7KfR0x17yh2Wj6+maQABfw6X1KKXBV1KtyHU/z3JzcpeW5Ol6ulhgALyrda7C1rpQy82xcUbpyHHddQLEY4FxFgHWtgJUWErrFsUNrjTzL0GbwCylKKiMutfYLAx2Pj73uIiWxWLjgVdHekb5XqefWQyR3buv1sxfNbpsGnGmqa7wll5RbNGCMARyANvPDAuav16WW+oUVba6EW9cV54LcOpswFA8pHlI8pHhIEARBEARBEAAtLq4V5fQv51BwqXONRtMI6YF+9Pf1QwQCnHHTwCAIIQKBMDBiWogAjAFpmiG3hZzq9RoAIM1SL/jKcC5MZ1QrwMfHxzE8PIyJiQmkSWqK+QsOSCBX0qTPMe6FtGtIoFReuGsyI2IXLFyARrOBlStXod1u2yYAqRecjAGCC+8CcjXPnDjL89zXIisLsCAIEEYh6nXjYKrX696pIaVxAmmb7qi0MosMZesH4J08TlRX0v9KqXRpmvqaaxraLxq44vtOgLpUQg3tz8sYM2MNmE8VdMdxriQhTA0xNw443e3+Z91AEhK5lF5MM87AFABt0kSBwlHkr7HH/Xbz7hW3Krleyo4irbtcNOXndaoUwM7XtNZFvTYrhN2tUEoDvLg/RVMD7hcjvIHJuaEY8000/MKEE992AUZrjXabxPSmCsVDiocUDykeEgRBEARBEARAi4trRSVFywrpMIzQaDTQ39+HwaFB9PeZFEAnmuMo8n8PhDCdRwOBdruNJGlDcIEwDKG0QuYcOjL352QM0DBuHamMsHRiutVqATCF+11NK9khpJ3LRgTCF/YPwxBSKrTbbURRjEajgeHhYYyMDCNLMyOOnYi11+zqcRWTob2zJc9zcMYr6V4m/bGB/v4BNJsNRFHkHSl5nmNyYhLLVyz3oi7LMnDGEddiv/jg3EBm8YGDc0AIUysNMC4dJ8qTpA2trUNGuNetALeLIEKY+lbOEZRlOZKk7fer1+sQgYCcnDSXAdMd1sDBGBCI0Kdw5nkOKJQWHdyzYZpHlAWym0fT0dUI0HLaJBjzgrnoVFt+BortZYeOhmmeAF0sQng3E1af9lcW5k7k+ocbzrCk7FxyMGEaNyjrBHOppG5fdx3OhRUEtutvJX2QeVHNuVlw6mwSQWwaUDykeEjxkOIhQRAEQRAEQQC0uDhtjAgo6nRxxn1NsWazD7OGhjA0NIQ4jiG4caPEcezFDOdGZCitkLZSKKlQq9XAGEOapKaoPjdOEQAIQ5NCJpXyXVJzKZGmCSYnJjE6Oorx8XHf6TOXElLmVuSb8/uC+6zUtMA6LfIsg5IKSkk8/fTTaLXaRpwq6RslOHRJVIOZ2lxlAeSaIbgOqLVaDf39/Wg2m6jX6wiCwHfGTJMW/vq3v5r5q9UBAEmaIE1S1Go1xHFsxL7taBoGIXhYdNbknCPLMiRJgqSdGGeTVKjV64iiEJxxn1IIwKdDOpcRY0WHWOemybPMpwnGcQyZm5puUsqqq4crIADiuGZqnikNCQmuuKlZxrkX5gC848cIaFGaL+5FqvmTF84goEvYdpl4SgK5EN49apl1COmuGmPltFZdrdfGehh83HU54SxzCQTmepybyzmupKp2QHWupyJ9UyGKQgDA5ORE98mIjRqKhxQPKR5SPCQIgiAIgiAIBy0uTgMnpAEjSMLQdDft6+tDf38/BvoHMDg4iEa9gXqjDq1MF1AuuOm+yYt0OSeWncDMMvOzEMI0KdDaOyCkdd6kSQqpFJKkjfHxcYyOjmFsdAztdtum9hnHitamC2oQBMY5w4uaUAC8SyYIAohAeCfF8PCw38cJSPdelwZX1Jcy/8dU53aGMA4RRTH6+/sqnWABYHJiEiOjIxgfH4cQxsXDGMfk5ATyPEdff7+vQ6a1hhYaIUKEUeiFtFYa7bSNyclJTE5OGocQ5xgYGEAcx77WWHnhQwhh5kXyirgDAAEBFhkHD0sZWpMthGGIoaEhtNtttNtttNotf41KKWSpcVM5kcwYAw84wiBEEIbIshRJkhrjjSrGo7WC0ib9LcuK58A+VH4u3TEL6wuKeXYZgbbjqjt2pUMpukV0r+cYpfQ8a52Bds0hrHuGM+6FsmuIAZiFofIiQxAwkxbKGbjiftFIKe0XGKRU4FwDEGDMLiCgNAZik4HiIcVDM38UDwGKhwRBEARBEAQB0OLitCg7EYwjQ5guqP0DmDVrFvr6+9HX14e+vr5KB1LXnVRlqkgFDAJfuD1JEi9QjBguupZmWQ6tNbI0846V1mQLoyPGoaOhEcURACDPnaBSCITwtaDKBe+VFeOzZs2C1hrNZhNpmqLVakNrVS3ez5i/1rLzxL3s5wXVrqeNZgNRFKPZbKBebyAMAuQyx9joGJ5d+SzyPIcQAQYG+gEAY2NmQSBLMyAC8ixDkhSuIs44pJT+/GWHjpQmDS2KIkS2QQQC+Gt3hfE541BQtiOn9qlt1WuyIlFwTE5OYvbs2ajV62i3WshWZGZuSk6YPMu8I4Zzk1oYxRGiKAa3bitX18wtavT1mcUFzjlGRkaLOe9MFXT3DCblriKodbGPT0eF9ml1ADMLCk6MA5V0wTLlnzmrLrpU3T7w3XUrzhutvYNLluYazCxguA61QgTQKrNzzsGYglKA4EaAE5seFA8pHlI8pHhIEARBEARBEGVocXENuJpUgBUZChCREVD9/f0YGhpCrVZDo1FHvV7zqVDtdtsU9VdG/IRhiDAIjRizHVFd91MpTQMAxrkXJtAaWZ4hSRNoDbTaLYyNjSFJTQdSxjhknvn0PwDegYOSiAaANE2Rpin6+vqQJimCMISU0rpOlF0ssK4S68jQMEIwCAJfm8y7crhxfAhu6qXNGpqFMAoRRzGiOEKtZpoxrFq1CitXrUSWZYiiCP39/eBcIE1SAPBCOZcSIUxzhzC0CwRZZpxLGXxKY5pmaLfbxh1i3VKNRh2NeqNaH4sxSIlKTTTOhamLpiQAU69MQYEpszgihEYYmvumlfYdXfv7B5AkbbQmWyg3Z3DC1hleTM0s5t0tgFkQcOlvQpjuuFmWAVr7OdfoFrvVeS5S/IwwLuqYMfc/K2A555BMGoErVTElYEVaXyXdkPs0w85FoPLzrrmC1ubays0hXDqju49KK0RRbGvC5dYpxZFnxrXEuem8ysAAbubM1eMbHx+f1ueR2LBQPKR4SPGQ4iFBEARBEARBdEKLi9OAM44gDNFo1MG5QJ5lYIyh3qhj/oL54IyjVq8jDAMkSYJWq2XcFJyjFobefaNsYXuZ21peaebrTzHOoZWCBnx6VZ7nEFwgSROMjo6ZxgJKoZ0kaLdaxgVkU6wCK6bKOCEmpWla4JoTNATH8PAw0iy1qYPdqWNFcX3uxZpDa22EccmFEgQB6o06GGMYHRnFquFVvsFCFJmmBOa4uV8AyLIMSTtBGIUYHBiECEQh7q0jCnDpi5lvNhBGIZqNJmr1OuI4su/TRTqcS6vjtglAKW2v0KUcjJm6acaFwpFnpk6YVBIBM2l9fX19qMUx8jxHlma+OL82xih/PplLZNwsbjAGQBepctAa4+PjSJI2ACDLjfunXNPLCdjyfTBC3byfi2oaHiB8R1qX/uhcVW4RQevCdVN29vTCdKC1LqDyGKww991TbUqj+7mYd+1TZI2DyizYeAeYP09pkQHuXvROWyQ2TigeUjykeEjxkCAIgiAIgiDK0OLiNHDibXBoEP39A8jzHK3WpHfcMG5qcrVbxiUShRHiWow0SU2tMcatEDbCw6X8gZlmAEVJKeYFt9ZFJ9Z2u408z9FOEoyPj2NifMLXomJgCMPApmqZdC1HlmVIE+PQUVphdHQMQnCkaQYhuFkYyPOKqwNASYxpMOvYKXfnrNfr2GqrraC1RpqkiGKThjc2No7xsTGMjo1CSeU7nbp6ZU70ORdOu91GXItRr9V9SmOapFA680IRMLXZkiSFUhJCmNS/MAoRhqZraiWFEYVgdPWtOBdgDOCKQ7MirdAhZdEEwt2HMAxMgwKZgwuOwYFBjE9MeEEMwLuDlFbI8swcy7qv3Di01gCD76bq51kX8wnAN1Uoj79I++M+7c6l2blzmGdKAhIoPUj2FEVn2PL8FCfSRbphifJ7GDPPg5Iayu1o36egrJvIHFops6gghOn4m6YZODc19gQL/HvLzSU6F4CIjR+KhxQPKR5SPCQIgiAIgiCIMrS4uBrq9TrCIES9UcfAgGlSMDg4ZFPjTDreihUr0D/Q75saNOoNcMG904Yzbrqa2i6eLs1PcA4IYcSXdUkorZBluXmNGUdHu91Gq9VCmhoh3W63wThDJGIoJcEZ93W+OHfdJ01qV5ZlqNVqyKVElqXQWiGKatAaVrAXorKc5gfAOjI0tHWzyFwCjGFocADNvqZvwBDHMZRSWLlyFVaufLbSqdQct1pDS0qzWJBlmRfGURz5umu5lL5bKZgRatI6exhjFUeSr2NlHT1TuVC8gO2su2bvh1LSN5fg1iHEGEMmc4yOjkLm0teQE4JjYmLSzx0Ds0KyuLeuwYB36sC6ppRJ++PcpMFBwu/T5aixItz+AKWL182963GhVkgXmtos1DjHlhf5JUFbvuecG6dUr/pm7vjlRZVCtBeiO8szI55FYMYtdeU4LoXUpS8CQBhGGOgfwOjYaM/7R2wcUDykeEjxsHp8iocEQRAEQRAEYaDFxdUQxzEa9QYGhwYxMDCAvj7TwbNer6HZWACpJBr1Bvr6+gDYWlicQymJsfExTE5MIghDKCWLzp3WtSKt+HICRyplHBdaQyqFNE0wOTGJVruNVquN8bExTE5OmO6ljCPPMojA1Pjy3UOtYHKukHq9jiiMsOLZZwGtEYYROOdI2olPVyt0TtEFs9yl09X80tBoNBoYGBxAFBkBLYRAmqb461//Cq004jhGGIbedWJElXV62NQxpSSy3IjlOK5BCNPJNU0TTE62MDExbmtWaQjBS9dmFiC44KZTp3U8+dcE73Lg+JS4DqeKqwEGOBeKmXulJGpxA4EtwJ/nuZmrPIfSCo1GE/VaHUk7QZarkvCtnquckuhEtbIdc/19UkVzAygUXVHt+BgK15F3/ADQWvr51aZ6mE+rc+4uo8OLsWmGrjpm9kQd81JKPWTWKVRyCxWLNdW3OzUdBIFduDFin3EGKSU4eLWumT990YihS7gTGx0UDykeUjykeEgQBEEQBEEQvaDFxdXQqDcwNDSEWbNm+c6f9XoNtVoNcRz7FDbAiCENjaTdRrvdxujIqBdPgO16at0amc4gZW5FmLQ1xwoh2G61kKSmVplx6kyahghK2xQwV3See9eKqw3lHCdOmLWTxDt6nDvHpf0JLroElVNUQgg0Gg3EcYyxsTHTDXZgAFEYmZpcSmN4eBgrn12JLM/Q39ePMAp9x0wv8lzKmHXFOJFqusSaelWtdgutVsuP2zQDgHEHoSTAyi4brYC8uFdOCBa2Ee3vC3SRZseZSQX0l2tTOAEgDEPEtRgiEJBSIctSKK0Q2AWCiYlxNBtNxLUYclJWhLrrhAvAz4Hbzji342HWuZWX0vucWGU+/a40fLgmA2CoCs8Oq065c2ln6l/ZkeWfFRTn8WMt1xnrrEmmNVSpppqpnVbaRwGa2zlX0jSQcKmKpfMze31mP+VdVqSlN34oHlI8pHhI8ZAgCIIgCIIgekGLi1Pwghe8AIODg+hr9qHZbKLZ10Sj3kAUR4jCCH39/ajFMQCTbqekqTNlXBwZ4tiIMgBgzPwppfT7ArACxaROKam82DZ1v4yjJcszpKnpJhpFoW+gAG7EcDmtzRWMj6MYKAnrKIy8QASAQBS1zLitNeabANh9OOfIMlNov6+vD82GSf1zHVhXrVqFlStXQQiOOXPmADA1xLxwYi7V0HTTzGxqn3PXeCdMkvhUOMZMuqFvPFDGunzy3IptxsEDXrn2cp0upZVxx3BWcfvAiUXGwYV1EFkRWavVTAdbbWqLydw4hqIoQBxHSJIUrXYLjUYDMpdoJwlcHS/n1umElcSx1hpSS++q0VoW+5TS8cpinJXEbmU6Sk6e6vl61+wqp/759+lqKmDlWM411HF+xgAuROU9LvVUSgXOGZRU1ZppWnsHmLs+CePgAcxzLERAqYAbMRQPKR5SPKR4SBAEQRAEQRBTQYuLUzA4OIihoSE0Gg006g30D/SjXjfdUWu1mumUykwnzXLTAS44IhYh59I7ZKQ2nVBzKaFVITjLoo25ulnKpMslSWqcK2CIIpO+xwX3DROEME4X54Yxwld6R5Crm6WkAhfcCmH4TqpBGFpxazquuhpicS1GFEW+o6vgAvVG3c9LGIVI0xQjIyOoxTH6+vvBmK1RxplpEgAjoI2AVL7eWpZl0BqmUypjtu6ZttesEIYRGs3Yi/Iiza+YL9MAQoFzCa45mCpZPFwD15LpqCzqygpd23vWarWQpaYullsAMYsSpoMtY8X7anFsasUpZeq6pcUhe4npSi0vznxjCMDU86rXTb23JGkXr3U6cEo1vZwDqtN5Y661VMfL1vuaioqwLjm0ysLZNz8oHZQxQIgAYRj45hCOLHPPUnG8IAjM8y1l4QCyuaduvpz455z5Z47Y+KB4SPGQ4iHFQ4IgCIIgCIKYirVuTfijH/0IRx11FBYtWgTGGG666abK61prnH/++dhqq61Qr9dx6KGH4qGHHqrss3LlSpx88skYGBjA0NAQ3vKWt2B8fPw5XchM09/fj8GBQQz0D2BgcADNZh/iODZujjAwDpssQ6vVRpblXrAKEZTEa2CaEqSZ74wpAmFS0TQqghFwtcEyW2NLIooiNJoN9Pf1o15v+G6mLtON266ZWZb5ml+cCy+QtN2xKGCvfboVY0CjUUezaQrz9/X3o7+/D/V6A3FkBHWz0US9UYfWxlmjodFqtbFixQpfb8x0KGUVh4hLU9SurpgVf0FgHC9BGIIxk+antUYQBGg0GnZuQ5MiaIvcK6WsMC/qYBU60NTq6tyHM9OR1nWblVIiTVPvfMqyDEmSYnKyhdZky6clhmEIIYRJy5TSij1ma8HZhZIoRrvdhpQmhY1z7tMItdJexFacN9plARYNDeI4wrx58zB//jw0Gg0zJ7wkiNFd6N+lVPrX0bE/mE37s+ftcPG4Z6A8MN88oXTsXueCdgskyj4PxuHkBHJgO6J6Ta+tS4oL/5yWa5X5cZaf1U2MLSUWAhQPKR5SPKR4uHq2pHhIEAQxFRQLCYLYklnrxcWJiQnsueee+OIXv9jz9UsuuQSf+9zncMUVV+BnP/sZms0mlixZgna77fc5+eST8cADD+D73/8+br31VvzoRz/C2972tnW/ihlmt912w9y5c1FvmOYERszWfPOCMDSdPydbk5DSdDMNggBaG8cBt10pk7ZpQpCkCWBT15QtvO9cNAwMjHNImSNLM+8QqdfrqNfrCIIAURyBMZiuoSXxndu0QiklGJipN2YFNOcCggsEwnROLcSoEXQuRRDaCJwwNGKScytAc9Op1Al7zkzH1xXLl2NychLNZp+pd6aMe0a5Lq+unphtouAcTEZIuzQ75d0utVoNzWbDuqAY0jRFmhnhK22XUSNajXD1Ata5PVwan9JeUOeyGLtzCLk/jZBOkCRtZFlqhDJsCqUV91meeeHpFz5UkSKpbKqmCIR3NfGSEK6kzcHnAJqFACuA3eKDEMIsvIgipbOM6yZa2ca4cWNZNw3nHBpVYdw5Fid6vVsGhQuoc7/yn+XjuFS/LEvNf7YRhZLmXoZh6GuGSdu0g3FmFpB4x2JLaZzm2AAX3DcD2RTYEmIhQPGQ4iHFQ4qHa2ZLiYcEQRCrg2IhQRBbMkw/h38iZ4zhW9/6Fo455hgA5kvyokWLcM455+Dcc88FAIyMjGDBggW4+uqrccIJJ+APf/gDdt11V/ziF7/APvvsAwC47bbbcPjhh+PJJ5/EokWL1nje0dFRU/8Lg9VUpRniyCOPxNDQEDjjGBwaRL1WN+lxYeQ7PrZaLUxMTCC2dca00qbJgBWiUimMjY5CKmUcPJxD+vphxm3jUgKVVJhsTRZiTRvXTZZmRlhmGVatGsbo6AjSxNQb48IJyqITpxN0rutmnpsK/70Ekqn7JLz4NwK82I8xDiGMUySKTYfUdruN0dExNBp1DPQPeAdOlmVeSDsB7dwzZly2o6l1FWVpBg3TTbVer4OB+evMUiOitdI+3TEQwqQYcu7dSH7RopQK6a7P13CzDhINbeqFWeeQVmaczrUDAFEYIYojNOoN04hBm0YSK1etBGMMtVrNi/g8z5GmGaLIOHuUUsYJlOXe8eLrppXdNJz5OYmiCLWa6Q5rhH2Rkun3B/P11pzrqeyKcoscJq2u9Dp6C3t338vOH7d/xZXT432V/V33WwZ/T6IoBrRGmqVI0wzQGmEUIooiMDCzOCLNZwHWvcXtYo+rcxcIAaUVVqxY0XXumUBDYxwjGBkZwcDAwIwee0PFQoDiIUDxkOIhxcO1YX3GQmDDfzc8BEcjYOGMXxdBEJsfuc5wN27ebL8bUjwkCGI6zFQsXGvn4up49NFH8fTTT+PQQw/12wYHB7HvvvvivvvuAwDcd999GBoa8gETAA499FBwzvGzn/2s53GTJMHo6Gjlv/VJs9mHMAyNW6ZWR61WQxSa9DUA1vGR+7Q/15jACck0S5GlKbjgiKMIgXANDIzYEiURoTW86OXC1BELw9ALPiWNUJMyB+cCURx5YR8GoU1d4zatqqhF5VLYvABl3KZgcS/AnZvGOS6c60dKVaR9weybJinGx8fRaBgHkXMMuXQ5V2ctSVKkaVoS8qaTatJOkCYptIYXWXEcm7Q7JZGmWVVIevFcpFSa+RYQgnsRCRjxnGW57STbwmRrEpOtSbRbLS/ipDLNCMq1t4zLRfhjtVot5FJacV50c3U12MzfdcVZI7iwTp3CJcXAij9L+BRB26l2bGwcIyPDSJLEL0D4fX16nPu/UqMDraG1Koli5V8HejiFeuDSC7teZ73HXjk/ilRMKZV/ljS0Xexgts6d8s91EAQm9VPmRZOLztTDknNoU2d9xUKA4iHFQ4qH5n0UDzcVNpfvhgRBEM+Fzem7IUEQRC9mdHHx6aefBgAsWLCgsn3BggX+taeffhrz58+vvB4EAWbPnu336WTp0qUYHBz0/2277bYzOewunBNlcHAQ9UbD157SytQLS5IEWpkOkEYEWVHATGdI1wXUdSUF4AUE4859YdMBM1P3Soig0u00zYwgzW19LFPPCd41UqvVbOqh9mLR6RDnkvGOGy6so8W9rn0qnpLmz3LanrKpcU485VmGVrtl3ClhZI9vUuxMB9FCSOe2SYFz5wDwwjsIQzSbDTQbTcRxDYwxn6oHFOlt5u/OLWQ6txZCOqjsB5iUs7JDKMsytNttL66TpF00a7CiP8+N+CuEn72OLEOapGi3275zK+fcu3OkLBY+ZC69gGSMF6ltrJjr8jhdjS1mGwwoJ9alQmdTAudCKqcXumM4x5H7uZcA7eW06XzdOXTc372A7jieTyEtve52cQI5TW1jBy58GqyUytQis6mm2jYt6DV21+hidY0XNiXWVywEKB5SPKR4WD4GxcONn83luyFBEMRzYXP6bkgQBNGLGV1cXF+cd955GBkZ8f/95S9/WW/neuELX4TZs2ZjcHAQzb5m4aixKW5SSV8XTGkjKjln1gVSpPUB8OlnWmsvsJ14dEK0XDMMKLp5OoHpuks6N40Trxq2llYprRBaF6+7Iva66MYqpfTNFJzQUbYgvRHU0jsvnBh3Y3HpciZ1Lvfi0v3shaZNvXP1x/IsA2NAvVFHo2FqpuV2ESFNM59C6NxG7lo4t44cbo7l6owxVtGnxbxx5ut1mc6d2i5EZF4YZ6kVytYZUxZyaWrqirWTBKNjo5gYn0Ce54hrpmmFhrkWk65n7llu65IBKGq5WUFdFoUV50yPsfcSvi4FUtgUyJ6CudPB07Hd1xwrz5NtitB1HMbABe/5vl71xsruIa20dzJJJX1qpnF7uedRmW6+1iHknWO20UQZ6pK6eigeUjykeEjxkHh+YyFBEMTGDMVDgiA2Bmb0W+vChQsBAMuWLcNWW23lty9btgx77bWX3+eZZ56pvC/Pc6xcudK/v5M4jn0tr/XN61/3OjSaDeNg0cqk5jlHjk29CoIAyHNwziBziVa7ZTpmWueGq/+lSmLVpQpa3eIxTgzj5nH1pJQTtNqkACZJ4lP0TJ0u02m1aBAAX4dLKVfU3ggXJ+6lkr5+F4BiHNqKVDD0yhhTWtlUOmncF7kEIP1YnZj3jQYEL1xIeQ7GbHMHV4PNpjcaccUBBKYrqGaQyjiSipQ/W+fMpaaxUqqb1t4x4hxBWpmLCoPQu4fcdmX/51IkXaH9skjV2ghm2HprnAvEsamRlfIUEtI7nBhjEIIjzVLjqBEccRx5V5V3vzAAuqjhVUlDZN0iuUizY3YxgZeu2d+VQhCzws2jpPIOq65jWnHrH76S+8aJY6VQSSftNa7y8cpphwzMP59mMUP5enyccf+ZgLtmZp79coMGd8pGvYHRsU07nWN9xUKA4iHFQ4qH7p7Zu0LxcCNnc/huSBAE8VzZXL4bEgRBTMWMOhe33357LFy4EHfccYffNjo6ip/97GfYb7/9AAD77bcfhoeH8ctf/tLvc+edd0IphX333Xcmh7NOuPSu3IqAMAiNECg1B3DOncmJSSxbtgzLnl5mhWPhFimLIidQXOqXVkYEltOgAlsHTCmJPMttUfsMk5OTaLfbcN06jZvGNQgwNa+cKAFs3Sxh64rZPxlnxmGiZLVeFxe2jlbua3pxXohW5zBKktRfX15yDbmUQaDo8skZh5QSSWLSwqC1qQMyNoqxsTG0WpNQqtrNlZXS5hhnvoaXqVdlH1ErnAt3Crz68seyP4tAII5rNtUQvt6VS59jzHaAtc4pwHa15dXUOzMP3KYLcp/S6Wq/uRppaZpCa+1TM2GdM775QMn1okrOqdXh3DeC2w6yjSaiKEQgBMIgNB1H7WIDL3W/nRJXz0vrsir3otg8i1UxXnH/lNw73o1UOo57fsypCkFumnnY+m66lH5aachQNN/onK9Nlc0hFgIUDykeUjykePjc2VziIUEQxHOBYiFBEJs7a+1cHB8fx8MPP+x/fvTRR/Gb3/wGs2fPxnbbbYf3vOc9uOiii7Dzzjtj++23x0c+8hEsWrTId8p68YtfjMMOOwynn346rrjiCmRZhjPPPBMnnHDCtDtgrU9cp9EwDBBHMbjgSJMUSiovOkZGRvDggw9heHgV4ijGNttsg1qt5t/rxIqGhspt8XypkEsJwbl3s7j9XdqTK5Zv0u6MyySzqXKudhmAimjhNuXK1RVzdco4N+lePDCpdJntWOlcM77GVF6uF8X8f06glWt3uVpb7txuPxEK2xgBXuAzW0cr1yZN0v1rWtkB4oSWEMVjGASBEa1WzGulIbWygsukjrkxFvdMAaowowCwLp8ISQJfF6zSFAAAtLJiskhpA+BTBKWUxTzaRhBZZhotaAHUreA3Nee0XwjI8xysh7B059C6EKZuwUAp5e+Le59UEkwx39iC5xwK1SYHfpGm1NCgfK3uOSz/3aWaKq3AwYtaY2VxXBLU7jX3/LvjMNiutFZ0Cy68M8w92xo2JZKXjo3CoVS+l665Rq/mCRsjm3ssBCgeUjykeFieC4qHU7MlxEOCIIg1QbGQIIgtmbVeXPzf//1fvOIVr/A/n3322QCAU045BVdffTXe//73Y2JiAm9729swPDyMAw88ELfddhtqtZp/z1e/+lWceeaZeNWrXgXOOY477jh87nOfm4HLee4E1nEShqFJY7I1whgz3TOfeWY5nvzLX7DsmWVeSM+dOxftdhthKIzTwApiJU1RdleXSnCTEseFEaXGneCEdI48N/W2clnUF2u320iTFIAReSIQvkaYcc+YelwAIFxaFYpUK8Zsza9A+EYJmhdF5EMWVZwbRjjaulZKm2uwdb86zSAuxS8Qwtdb01p7ISus2BQi8GMwHVqZrxfGOcfE+ARWDa+CS7F0qW+uMYTgRTfUrnpana6Okng1CwAMnIcQXCCXJo2v2NW8VymbBgczL0oakS2lQtJOSosfJqXTjNG4cITtZpu0E2R5VkppK7qDOgeXO4abuzAMEMc1hGFg3U2JTyM0dd5yn06pbcolXCqfFeSF26VIj/SLIagK+dKNqyz4lN9XLDjoqmundAwGBhGYe6uUhkTuFwI4qwp+n2aI7lpo/nlj3LuX1ug42ojY3GMhQPGQ4iHFQ/czxcPVsyXEQ4IgiDVBsZAgiC0Zpjelb6+W0dFRDA4Oog+DM/qv+kEQ4EMf+jD6mk0EQWBcGdp0RB0bG8Po2Cja7TaCwIggYd0YSZqCMUCIwHfidA0LknZiumzagvSuLplzvACmE6q0xfCllJiYnMDkxCTaSYKnnnrSiyOppBdmrsuk+7ko7s/8NkcYhpBSYWxsFPV6A0JwL+o4594JBCtmwzAwKVw2nS9J2oVwtGmR5XRCxpg/hitez53QDsOKkIZ1oDiECCBljrGxcSRJ22yzYr4swgH4Gm7unjux6B09LsUSRcpkWSQWNd/KdcXMooWrT+bEqhPMeZ5XnS2MI67Ffs45N7Xk2u022kliGjEA3nkDm+YWxzUoZRoqQJs6dFEUo16vIYoiyFyabq7tlm+KUKbspHFC2Yy/6jLSunDi9EqnK89HWTyXtwHdDp3ynLtFlCgKzfNrm1e4+yWV9E6zcmMJbevLuXkUgUAYRvYZKsYopcTw8HDX2J8rGhrjMIWuBwYGZvz4GwqKhxQPAYqHZvwUD6fD5h4LD8HRCFi4oYdDEMQmQK4z3I2bKR4SBLFFM1OxkNoQlrjg/AsQxyb1L5dGuLXbbYyOjCLNUvT392PBgoWATY9rtVpYNbyqdITEC+Lc1gJLkgRKK4QsBGPGFSKVgrRiFlb0AaZOk8xMo4M8zzExMW5T0QQUpBfLRu9xI0B8OqEsBHUpzco0O2BoNBpot1pIkjYajYY7tamFxZhJx7Pvd24jKZWv7QUU9cuiKEYtjr2bx9TgKmqMOeHsmiq4VLrCXQLvFJEwoiuKwqKxgXW/BEGAMAh9/bSM5abBAKrjUUpBapMCKWFSLitpZmDGtcQZmCrcPk5Yc86tS8eILlPfjPsFCdfttnJeJ7BtupsIBIJc2Lkwr0nvPim63TIGaDDb0MI4sLLULtrYFEMnwB1aGQHLYNL9ysK6LKLdz2t6HSgEecU9Uz4ntH82yqLci+pSF2CzOMPMvXFOIrsPY4FdsJBVx5D9u5TGfSQCYVMtTe27uXPnYsWKFSA2HBQPKR5SPLTnpHhIEARBEARBEKuFFhdLhFEIZp0rWgPtdhvjY2MQgcCc/jm+EUCWpkjSFO122ztoXFdPxZR1jhhxBsYQCiMMNTSUlNDKiVTt058448hVjjTNjIAfG8XY6JhP8QPKwsh2xzTq2otszouUKpkagS5ziTA0jpoojjA5MWlTELkVdtqnmXHfjdOMmyl4Vw239bOckI7iyKeAQQFxbNL+inQ9M2bfhbQkyJybRUrlHUBRGCENMuRZBs6Kul5grOi2aQVxZ4dRI/qMjnP7mjHzyr4mlVEXjRFyQOkcrlEAAOQSPs1SCO5FdxRGyKWEUqaBAwDfrADMdGRldY4sS5HZayjOq5CmSdH8wt77LDP144y7ivvFBuOSKaXdlRYknFA2DQFKzwGqDqSKC8ctJpTcNn672bmS4+lTAe0iiS3BVtxDbTqp5nkOMOabKOTaONSCMEQUhdbBo8CYgpRFLT1/M9wz7Z9rO52cQ2bdbiXi+YXiIcVDiocUDwmCIAiCIAhiOtDiYgnBBfIsgxAC4xPjWL5iOfr7+tHX1wepTDqgVNK4JZRCGBZpfc7tkmZp4ZJJU0RRaOpscYF2kiBJ00qdKJeyZ5wbxqEzMjKCVatKDiC7v1IagRWizr2itUtlU15EupQ4J6iyLIfgAlEUI02M2HPCptVqQwiORqPut5kUQOVro5nzaARhiHq95vfTWpnUxpD5a6xaTLQXYK4Bgvt7+brc9QjBkedGMLp0QmgNmZs5d6mVAAqnjFRd6W6mzpcp6C91sXDh5ts1RXDncdcMAAIaea78ed3reZ6X0gsVhFBFaiPMZQgloJTtDKtt04TS2MxiirILBMYV5Zo3uDpubj+my+KVQdnr8Q0mOKCkddF0CGHux2TFd8cYyvXDnIMKKItrVHGLI6pw8JSbU7jFF7fwo5QEYzZFUmmTHhgEEK7zsE0FRF6qMaYUXCNczrhf3CA2HBQPKR5SPATFQ4IgCIIgCIKYBrS4WCLPc3DBkSQJVg2vsjWg6sjyDK1Wy3fANLW0OGC7QHJhHCFZlltnh/ZuDpdKJpVEnmXQyoowMEiTd+ZFVrvdxvj4OEZGRgAY0WneX6r/VHI1lJ0dsOmBAABWNAIwbhiFLM+Ms4MzqFShlbXM+LhAvd5AzaY/Kmn2zbIUaZoYES0EwqhTSBe1phiDF9KuW6u0ItcLZwBVV0mxIAAYV00QBNaRwwrhrZXvGmvS4Yq0Nefa8a4eXozLNVNwThvn/PBuF3sewUXF/VK+Z8YNY5xEUknrBCo7o3jl/YAVgoL7xQNfF40V95Ax+G6rjXoDjDPTqCJN/cKB0biljqQKvjGAu37n8vLz6xYy3J9a++vtpEiPrC5+lI9f3i6VqQvGuXNzFfXinBvKNHXQPg3TrZUEQqBWq5nUQWXSPKEYtDYdg8EYAhb4IZQXOYgNB8VDiofle0bxsNhO8ZAgCIIgCIIgqtDiYolWq4XJ1iTGxsbRbDRQt7W48jxHlmbQWiOMIi9atO2qmeUmPYxzhjAMkU5OQAiBKIwglfTNCYAiFcoJHKkU8ixDOzG1zIaHh82+VhCXa0NxK2AA62wJTHMA7ovmay9YBBeVTqoT4xNQtg6XChS4NmIutkX0yyJZKeUbEoRR6Dt4FmKX+8L0rIfmWVOPINe0wOGEJys5NJwDRinjmJFKesHtzsEYA7h1/ij4FDujhYuC+bmUYCzvdq1o7UWrc/J0F/aHr7UFGBdMFEUIg6I4clmwFt1pbUoiV4A2whEwAlhAeNGZ53llUaBMuZlAeczlul8u7bLa9bUqsv320rVXhHTJMeUWBjgXFQeQP6e2iwIo3lNuSOHmwKf0MYYsz1BDvXBzAWBMQeZFWqHWCloXCw1S9r4G4vmD4iHFQ4qHFA8JgiAIgiAIYjrQ4mKJiYkJjIyOII5rqDcaYMwIBSeEGeeQMjedH7VGbhsVQJv6Yu12G5MTk75rqLJOBNfUwIm9cifUJE3QarUwNjaOFStWoJ0kCMLQd+ZUsnD2AKYOWhSZel6uK6gTQEJwcB5AiABxHIExjlZrEq1WC2mWgluB7VLshBAIw9AK2UJAamVcK8wuDvhURtvd1dfoKlNS1U54OnwDg9I+5VpgACp/B2AXBazjBuY6wd3xTZqfGzcAn/ZYLpLPBLfpZnnhcnLC2Z+nWKzQdtjOFWQcNuYYbjGj3BlW8GIhAxLQXPu6aNCuOysq5yyn8mloJGlSdHa1w+rsburdWEBFSJfnvVyjrHMxo7MLqj9mJdVQd4j2kiuIMb+vcUAxKKYrWrw8nCiKbG2xwlGWpglqtRriWow8z5GmKRRX/r1ufK5xhUv3JDYcFA8pHlI8LOYfoHhIEARBEARBEFNBi4uWM854O5I0QWjraLlultp2MuWuPlJJQEslvctBKY1Wq2VrZQm/D2PGQaJssXnGGDKZIcszJO0E4xMTmJgwqX+TrUnjnolrSJK2r58VhpERcMKI4SiKEIb/P3t/GrVZdlYHgvsM9973fb8xxoyIzFQqNaERCSMM2dAUtrAGA8tektuLKuwGm2WvpgRr2ax2uXGxTAnaS1VeHmrZ3YCrixK4Ctl4aJuCMrJlY4mWkQTINkIoNSeZSqUiMjKGb3iHO5xz+sdznnPOfb8vUplSZGYMz/YKIr53uPfcc+/3OPejvZ9dJVUNg0iuSQEBq9UqhjGEaN/KpIo+Z9csbz4luyIEIovxTyKQJlvAAByxkXkPaI04X0rDsJ3rGEkPEy0ottbd+P7kcIUYsAAdrXhlwEMkocGnz2mlUwAAABBH49TSbCsk4kxKnKBCUgCRgCfAmEgqC6sdrSmf30eVVAg+PRMI1JSY1lNUtoLzLs2q4zXwXpQKmxjLMLJ5MqEdNwMy6T06a628tmyf5L+Sugi50cDfy+fwI5JdBkao4hr55rG9USkd56aRJXC1WkFrg+l0khQ7Xue5eNys4GdYK43d3V1cv379xg+F4DmD1EOph1IPpR4KBAKBQCAQCATPFNJcjGArnI5zs3zgOVgaGgFDH2ddBR9DC3wepg+NwTnUVQ1jDc0jC9nyx0QhgGxpbdditVphPp/j+vXrmM8XWC2XsNaiqio4N6TkSRVoIDxgYasKTVMTuTbjsAATyWbwpCBq2xXatkPX9ekzPD/MDQ6T6TTNIQOIlLnBoWtptpgPPr6vaMZYVWUizVAqDalnlciR+VRPx5Dj+5mf5tlhSulExBRUItMMPZpPlhsCZCOLiayRyBpjQbY8Sur03iUVFJRKKahGG3iVyV05d4zPqeK8N1bfeO/oOMgzyRgpvRTU6KjqCugVejVkBc8NmgjrapyC6+Z9WiPQ62R53T6YSPS6TVNhrNoprj/ZLdeOx5ZMtjM6T3PgrK3TtZfwPmDoe7i6SqopnkOWjq1U2hdWRwleGEg9lHoo9VDqoUAgEAgEAoFA8EwhzcWIEydOZCVNTCnVViMEoOtcUmaooOCcgyrIBAsVmgmlj3rnUTd1GuzOJKTrerTtimaZLRbY39/H/v4+2raFVhqz2QaGvkdfEGD+HitpmPQaa6I1j+15RDZ716PrWrRti6Gn43jvRjO9AiKxqSzcYGCrCiGmj7ZdS3OvlErkns+l4nD+EBU5dOys2KEggdKKt0ayS4seMonmuVOlVZKJdCjILJFAD1LpIM9ci80KaFqUjtY9IDc0EAIGB5o/FgkipdwGADYpT+hSVHoWtMGaxXFMRr3zsfESEttli2MISASRnwvH96JQt7DlkPelVOMk216h0vlKJDOT6mN8ejdAmi0Wmz587pAZfPFhUoh571HXNVStsVqtEMKAqqpHirDgicQrRcq2YRhGc+L4+Qj8rASPfqBncd0aKnj+IPVQ6qHUQ6mHAoFAIBAIBALBM4U0FyMmkwnZ/IYB/dAnMuXcQAmoWgFBY4j2LSKNhsgUSFXA368bsu0xKVbJgkZBCEykD/YPsFxQSulkg0ICVstlUgklMMnoezhHSiEmuvR3Dedo5lXXd1itVkSmo5VxZNcCka3BOXRdl+ZkhUAKH+doVhcT9cpWqGyVB88fYzcrVTOl9QzAmEiXBBKZePJ7ZDkk9QvPdePvBx+gDJNdOhcTZk5mNTBJTcVrAWJggnFQ3qXZaTyfjJsUAO0BX0uaKxbvXQqQKF4rSbDWGh4+WTDzrfNpv5O1tFQ0hTDaz5LIpv1mCylUUtuM9mZtr9cVU/ws8z6zLZBOv2Yd1Arwa+tYI9I8863vB2hNgQ51XaGqLOqoxPHO0yy+uDe8Z2xLDXF/mcR7T00JTocdhuGZ9AAEzxGkHko9lHoo9VAgEAgEAoFAIHimkOZihHdxvtYwJBtZ3/fo+yGqBoiI+SIJMqkn4sB/tgUabYhoJbUFqVDatsV8Mcf+/j729vawWq0AIAYWmBEBLhUvTCz5pH0/wDmHrutRVRZV1UclDYUeEOl26dxkqYvqGmYoIaDve7L6MZnu+9g40HGemE2BB7ogiSkplC2ABdZtgEyk6bPIM8WgcrBBIniIKg8iWZTESvvAoQtMgks1Du1XDkRoXzxLNkJG11ZQX1wiHBwmFRKpTbqofim3mO6nMeqIWsWYOENLqzhTLKQ5YiEAzuVGSFLdYNzIACLxDeNz8rnomnV+yPLGp+8kAl76A4s953OO7sea+qf8d1LppMMWSbzM+/2Y+CMEdF0HBYXJZAJtNOqqTvfFuYGeofj7xA2FwA0NRQ0IVmRxWqwrrJWCFwZSD++cegiM1Ys6/u6RUC5IPZR6KBDcFfjsz3/jsa9/3f+4hP/Pn3yeVyMQCAQCwZ0HaS5GpP+IDx7WkOKFyS1bvRiccMokK5E2o6F8VnqkOWUKiUQf7B+QQme5hNY6zfoa+h5d39FxtMkEsLCEeecBk21z3jt0nY8NgEiYSxKiVLK3UbInUjgBkdUA53y6vqFoBmhtaLaYrchmGMlpVqYctaKVyZuO1xGJdLmmpDwpCGNJhGn4vU+WMRLi0OeTYqaAthXaC5NEbL9+50n4gmBqpdD1PT7mZugXm9BKoz4EzOVrad8QAs0xS2moREpToEF5Pm0yyVekiyEiPURllxvx2xtZ90pieuR9ttGtK53482tKnHWMiPwx4DVxEyLP/RqnoiLaXNePy+97nwmwsTF916tR2qsPPlk2fTxepS01SGITgtc6OAcOgBC8cJB6eBvXw9EsRhN/78aNRprZqOA9W6gVjHFSD6UeCgR3DMz2Nj7zE69OP3/hzT977OdetfFn0T/2LQCAe/+9x+RXf+t5WZ9AIBAIBHcapLkY0fc9EWBHc6faVYuu7VKKZakOcY4G/ZfpmForOEfswXk3mo3Vti2uXr2Ga9euYn44x2K5gFIq2qfqlLoKIClSAIzVKpHEeOehrEoEyjmfPlPOpdIpACBb7ZQCdCDFTPAe5PfKRJ1VSDT0n9Q6tqooLGENZJH0o5+TCuUY/xYRtxgyUKh3+Dha63Q8slySckgpjKx1bMVTCjDVBO3JCqG2eN3JywgB8f7kBoSJRNsYgzfs7qPfHBBCwGc3d6CGLahLPbwfyMKnoirJk4Ikr10f4XZaaRibgxLK/YsvHrH08T0t1To3wpH31uyCRxJqQ2EBLN97GrA98DhrZ+oCIVsZy++lz8S1DsMAo03cK1LeaK3SvLGg6Hj83PEz6lRuSnEgBCX3Hv8cCZ4fSD28/eqhDmakzuNGmfO+qIe0dhNDacr16fj7K/VQ6qFAcDvD3ncv/uD/+gD6zYDP/Zc/8xU///C3/q/At9K//9gbvgcr/GFpMAoEAoFA8FVAmosRznsEVh04h7braB4SAkwkkzR8PlqZSmWcNui6bOFj5YsxFvP5Ia5dvYanLl/GweEBuq4jpUJloZRC13Xo+yFa4EyyGxLCiDAZrcmuOLiRkoE+konbaD5VtCny5zOpJbISEIqZTmzFy0T6CJHF05PAZ4RiLcUF5HOw/dE7qEGlgAb+uJ42cNs1hirgNSefjMcp16ngkooq74u1NP9qGAa8fLKHT5/fQN9twl/ZAyeBjhQ2xyhpSmLNc9gAkDXTO5ot19NcuuDzrDOVlCoeUJmQrhPr8t4dt46k+insnEfAVstQHLc4T/nsjghrCEk9FYIfNSrW55atv8ZWTYCe/wCyFfIzDXCTZJyuS9dCZJ3XYo1BP4gN8IWE1MPbpx7yGr33CCCVHDcR+Rgu0O9gOX/RWgsFlZR2OjZRvVdSD+P3pB4KBLcPzCteikt/5CwOXgx85vt/+qs6xvtf9Sv47v/72zDv34j6X//OzV2gQCAQCAR3OKS5GOEcDUznAIO+62hOl1JQ8Eklk4bmW5PIpfcuDmaPx/JkI+u6FpefvIzLly/j+t4e2nYFrTUqS4qRru3S+UOphCisewBS+maIJCkET7OmVKGECCzYKFV7ZmTb4rlnPE+Kj+ucI+Iaz8nBBSlZtIDSNJx+fU4VrS3/zKqcdH0hHJlHxq+5UuESVTh0nYgKkD6qPjT0ZILlaYvXn3oq7g0Tb5/tgbHxwK+PGh/GQMc9eMX0AJ+6bwLfb8LvH6brIuKbmw0lYUzriJbBdFx+XXOCLCulmCQjHunG9r00X+wGRLr8XNrrQmWjkFU37Fw81iJYNmLKe4QAHfe9PM/oGGwN5HXFY9FzYxGi/VVB0e9InC2XbIdFc6bcn4BA9tPYTPHej+ydgucXUg9vj3qolCLFIsZzFqlZWNbD3BAu+2faaOigU9PVGBPVn07qodRDgeC2gXn5S/Dp//oMPv+nv7JS8SvhV1/xa/jev/5HcWXxDdD/v/90E1YnEAgEAsHdAfmv1Yi+IwLd9R26tkM/9GnWFCtZeFaSMSbZ/0oSBhApV4pI4JWnruDxLz2Op65cQR/nh4UQkk2QgwHI+pfFNyUppCRPmmkVeFg+8oeTuiSESJb5q5FkqTUitIYQQgpIsNbG4AKaFbVu/1NaFWENhWUPmaAROfcj4uz9USLNr68thq6jeJmVUc45oKqwOmvxupNP5blu6vhHWGkNpTVMkcyZ3uP1KoWXTefYfHCA2dpMBNEURFlBFddIyhO2z/EaQjhqWWNiOwojKN47sl5khRXPchsphUpFTEG4n1Y1xQ+Vyp9dX2f5jOSGyzhdlo9B389z5lRqJvE8PZ8bApqum5s3PtpV+bj0TPjUsKBkWmpS0Jw2f2TfBM8fpB7e+vWwbNTx79aNGlA8z8/E+ZWj97ieKNpbvmaph1IPBYLbAfbBB/Cpv3QGn//Tx89U/Grwjx/8dXzd3/kk1De+5qYdUyAQCASCOx2iXIw4ODhICpuReiHw/4lQKiWPMklwzkWCSWqf4D2eeuopPPHEl3Ht6rVkDwOQ1Bwje1VUujjnkoWKSAnPNzs6TypJc+JLSTGhdUFyHBRCIpz0FZUUR0zEXbJpGRijUVUVqYnUmLBxoAJ9VqV1lwP6OVSFVUT8mePINC8qDdEvSC/b0LSKzYa6xupChT906iq0MlCKFFEllAI0xvPB0n4hklQmdFpDx3W/uFni0QcbHPyeiuRZw2sP7XVUDHFiabzucp0+xJTUrKZK5LJQ+awrYo7Y65ggR3XMcaEH5Qw5YEzK10k1Hzc9LsU9LOfYceBFSmddJ/BKlVs4XlfxTIZAabtKRaVS3Bcm/yE2QPiY1ihw8ANfEwdltH0L5zzqGoIXCFIPb/F6GJtvFHZD+zdSPAK5lvn1eoi4Np/urdIKOtDvujEaCJTwLfVQ6qFAcCvD3nsBn/zxM3jkbf/gph/7/3XvR/FX/j8dPv6HbvqhBQKBQCC4IyHNxYi260Y/M4Hwrk8vkKUpzwDjmWSsJCEli8f+wT6++MUv4urVq2i7FnVdo6rq44QygM+BB875RKqMITWDUmyzo4+HgBScUCpQjiVeQSdLXAApcpK1LRFhD4AIkDWkWqmqajScPy11TZ2zrvooE1GBrCp0Pn+vJNtPB5ofFlDZCs1sBvfiCd6wexWIFkCoHChxnHqR1+a9T3+C9/CBDJdGawSfFVZaBeimhmJb2tMIYLghQgoqAI4VSi42VkrbJpIlLzdJsgKG38v3Mt4XH44Q5PXZcen10v7HVsNE/sfKTbZ6kiLGoQtdIvpMfvmY5XdCUogVexwbNfydYRhQVRW0tnDeoe966NicMUbBDSE1lmxVpWseevods8aktFRj9GhunOD5hdTDW7ge1g3NS1xrsnIj8Tj1In+M75ePe8y/80bHZmKxDlU0DqUeSj0UCG41mFMn8dm/fRqPfPv//JydY9O0sOfux3Dx0nN2DoFAIBAI7hRIczFi6Ps4byqqVoweERSEkOb7lXORAkL6jnce8/kcT3zpCVy+/BScI3JhrU2zwIBs5UoWKB8wOB6obxI5YeJVkmoexG+tHStgSoIaiDQ65ZAG7nvk5E/F6yACbwxgTDWyAZLlLatvEoFGSK+Vipz1QfjlLCnE6/Q+ByCUBLNECiaI12WsQXjpBK/f3svWy+BhEOeMxetz3sPEZiPvMb2v4AH4aCNkLujj2o0xCCHg/qbDU6+xuP7w5MizQXPDDNYbFt67mEJL6aBd18XGij/CxrN1Llv4SEQUSe9onlz+Hj+D5Ryx8vXUYIjdH36trqu0Zp7b5Rw1fgbvYBEbFqVtND2fOh8rUNgFL3g9FCEEagQYbTA4l54hVv2wcoktfgDNvrPGxIAMg1WhjmOrq9Z6pIgSPL+Qenjr1kP+k9fjU9BMqWI02hTW23g8peGRFaGl9bish7CADjn9e31NUg8h9VAgeIFx+RdO4zN/6B8+p+f4iTOfxM9/4Cn8o1deeE7PIxAIBALBnQBpLkZww4r/A55VGGwD9IFSSZUuggWcRx9VBoNzODw8xN7edVy8eBFd16KuG9g4r0tBwWMcLgKQSodTVTnN2PusJiHLlo+EziY1jdJqpAhZV3Yoz5Y5lciehoaLc9P4GksCrzQPnTd0zT7O29LZPsakhwfT8+cAZFLHDQATbXWGUl1DHGzP180olSTeR1slNwkLu1wIAR6A9h5DaYkr+gnrUFon4uidg9I6NRr5+1opBK2hvE5NCzMatH9MsEAgMj2EIT0HXUzUZSKaPrvWMEjEt0itzcQzpOdt/bvGGBht8t6FosFR2DGttairGj7QukpbJBN954b0zCilE0lOllFkFRHPHMvL12kTtFao6xp1U2OYL2Di8zmdTikhtvguPW/0bPVDn+yIWpOVfRgGesa4gSV4wSD18Bavh4HrIdmVh0B7Vvb7jtQdUEPVwuTZmMW+0ffp95FEhkHqodRDgUAgEAgEAoHgGUGaixHGGFLneJUG33NogXf0H/pQCpWmLXPOoe1aeOehjcbiYI4rV57ClStXsFguUFVVIZwJiSzxz955OE+Wsbqpk5piRFwioWZSbeKw/aom0j0MQzqm90T2mWgRGeekSTey/ZVWMApjyH8nZQ/y2r0PQJy59bTzwoovaq2yLTBiNPsrHf/o697zTC9eY1RNhVBcEwe6KJqfWFikc7DA2CoHIM1czK8XC0EYfY/VSt5TM8IYA72WgaQUNSBUn+etpXlerPopXqNj56YCvccEn9eBqO6imW8cmKFjMu0wDOj6Dr5oSKQ1g56hfuiTRZXf4xlxI4tf/DzPcCsJNpN+CiLI4QTFxaNpGkwmE7L/GXo+tNbY2dmBGxwWy8XoO1or9P2AMGQLYfmcUDPFx3NC8AJB6uGtWw/59yYEjK4JyE3P0iLNxxspALmJt7b2cn4g1wqph1IPBQKBQCAQCASCrwRpLgL4M3/mz8AYm8IIWFnC9j7nXbKZBR/gkBM7vQ+YLxbY39vH9evX8eSTT9Igdq0KgpyJA1vcnHcIAcl2x3OwAMQ5TQZVZdN8MICIj7EmWQN55te6XSr4QCIO7eEHn9NYoyqCyTMrP/gYTLLYssWppiF4eK+hTCbSTPrJkmfz/LCI0p5IFsAxYWbi6CLRL5VAFAQR0h6qgjBDZXUHKVoC4ByCIltfGeZC5JDmLDo3xICKtblsWkMbE4+j0t6ShTpkpVYk6qx6YkWK1obSPLnhkq6Pvj/qTPCaWPlTBDzQe2NbYFVZbMw2sLO7A6U0rl+/jvn8EIMr5tyVCs54EO892rYjIgw1JsCIyh6f/12S5PRzaT1cI7VMvo01MMamZ0grjb7roWZ0rLZrqdnrA4wZr8ENDr0aKOwgZJtg1/Wx4UNJw9tb29g/2Ifg+YPUw1u7HnLTU+tylqDJCj+HpAIsz6Oi3ZZVp3yu8n2Aju+1J+u41EOphwKBQCAQCAQCwTOANBcB/OIv/iL+q//yv0o/lyQ1zbrSJs1I6vseXU+BB/v7e7h69SoO9g9w/fp1IlrawLsAbXUkfqS84xTNFDASQrRk0fnY4jeZTEYBApmMx/mCzlEYAVvEfCRVsfHGtjDvyZ7IxJcJYLL9KQotALLdLMQ1JjVHYX+jYAV9RIFDH1NAPC9/L/39DCQXPnjoYIo9CukeMNzg4EBWTBp2r6G9h/MejhVLzkeSZ+K9GhAQcohCIo3xT0G+WY2SricSU+89ygDqpm5IPcMBDc6h62m+GL+WZmz5cShAQEiKGU5jLdUzIRL4MrTCOQ/nOhweHmK1WqX7qaBgKxvXHtN149yz4saMlELleUb3rrhPrP4sCXb6XHFcVvn44OF7j8E5LFdLzPoZrly5AoBmxFHjCaBbYhCi5dB7lxoVKpJuvgchAH3XC5F+ASD18DaphzEwhdfNMySdd3Ahzr30PjVOAaAfeqoVhdIRAIJiO7kfrUEFnoco9VDqoUAgEAgEAoFAcGNIcxHR6ha4+aSPvueIMGijMTiH1WoFF+f2XblyFfv7e7h27RqWiyUAInAAkbO6bsApmd6RYoTthVpp6IoUORx0wMPfyyH+Yytcfs27sUoPGNvQEEIiKwAnZeY5T8ZoVHVFa2XSCCSFEn8HQGoGlDgyu4oWcmR/FbJNL1ndVEFa0/HKdFEf55IVx9Hj73ADwWhN6sTBJWLH++QcqWdMVCjyOo6bu5jveWwoqKyGGiKRDz5gNpvB2vyrk0ICsEZSQZbt8nqD94CKpJaJbnw+yvWw9fBwPsdytYQbyHZa3hNWDk2nUygorNo22v78+PiFGocbRPxzUukUG82Ef/16xvfeYQCAtkvXGoJH2w44PDzEfL7AdDpBCHGdPt4f79M1hxBiUwPxOXWFos0de17Bcw+ph7dJPUyNu3x5VA+pKcnfLWcquoGuhZvDWuf5i2XTTGudrlvqodRDgUAgEAgEAoHgK0GaixEcAuCCS0QXAM1ych7WWvToM4FVCgcHZP1r2xUWiwXcQOmQUApD30e7XVRzDFnNQV9XsFWF6XSCuq6jpUwnImWMhq7YljcOU7DWkr3N+qTaIcKqEILONsPCbgiQio8JsGUboLGjY6d5hT6kYf9M3lLCaJpfGNJcNFU0IdYtgTfY8Bu+ldRM4WjKqNY6NwJZrRiJmbEGyhNZ5PtkDD3i2ujUhEwJqkBSP/qocmKCWTYXODiC7aAlnHPo+x6DG1sMR1Aq2t1IPZVDGYoGjlIwRiWFEamSBiL+a9ZLxDln3Jxo2y4GTdB5hoH3LjcrOICBGxBaGYT4nAeEpMzhMIe4bFKBFYSaybfmBNUYkqAVqdEQAtq2BYl/qMFjtIEDPY/cIOLmER2zpzXE54sSew2qqrrhMyJ4biH1ELdJPcyNvFwPqZnIKdfcHASQFKBa69SE9OkWqpygDR9HUkg9lHooEAgEAoFAIBB8ZUhzMaJru0SuACICTHyruoZSIIVNJNOr5RLXrl3DarnEcrVMza26qQEgpU92XYcQ6LtDT4mQZMWzBZmICZWKEk3ryiQCxmmWQEwULex1GhoOY3LHljfnXApJAJAG0ztHx8yqoKIBFwfkd32HwTlYRe9772FiaiVjXaWjlE/HGRHpwoIXlw0a2O+S3UuvzUhThZXOeQe4IVsTx0Kq1CwMPsRkV50+c2TgPmIADNSRY5BKJqa6lsQzBAQd02mVTWSb7+lyucB8vkDbro7cByiaz2WtTfOz1qHiLLG6rikIwFak0Dk8xKptk3qH1S1KFSEXge7J0PdwLqbpYkyI+btZFVSoeOINUQFJUTRK2Q3HqK7i3ihN+5OaLSqTbn6+gPzccQKq9x5VRdc6aRoYa9CuWhzO5+m55ZCN4+ymgucHUg9v3XroBgdtuLaZ0WepWUi/5xR8kj9zfD3UR17zIasYpR5KPRQIBAKBQCAQCJ4JpLkY4YOH0XkYO6dKWmuhlU5E2nmP1WqFvf097O8fYL6Y09D2SDLqugGQZz61bQvvaRh/VVeR15CqRsX5Y8YYNE0NHtK/nibJTS8OMyAVCZPRbPMqQw200gg6kNKiIEj03Q59T+oOYw0mE43KVjQPSg3Qg4ZHnFcVVULrChUAx1r+0vyrkXIn//u4hFQisDnhFSET2r4f4PsB3jnYUrkRz63pBIDO5NkYk9Q2fE6lFFzcL1b2BJAtz3tP/w4ezg10Tyw1NJwb4FwcsK9VslEqBbTtCvPDeSS95ezGzGR5Zhw/Uwg5BZdmx5FFsWkabG1uYTKZ4HA+x+F8PjqmUhoaOt3rNMusUE2VjRVOVWVSOgzD2NKXGqo+KWzynLlI2m80l0zxMxpyMIbROU04BkWwfZaVZxR0Qe/3fYeqsrCqOvIckdoqYBi6I8+K4PmB1MNbtx5qrVGhukE91ID2oFl+ZT1Estby8V1wqSHGexZSaE2skz5IPZR6KBAIBAKBQCAQfEVIczGC7XBQCn3XoR96stsFIpKchrpqWxzs72M/Bha0UU1htMFkOsX29hYNstcKi8USwABrzGiWl9YaVVUldQaHCgBZCaGjbSwrHUgBkeZ0ITvkeAg+Aln3DAyCDlA+px8z+fbeJ/XPcrmk4fiDw8bmRjqWtTaRIMAlmxqtb3xufo3Tm/kaAT8i0fn1/GWyISOS2Uimo9VOKw2vaN32cY9PVbt4zdYhoElpaDTPwIohLSGk5iEfWysi4wCgTLb3pVCB4iJSMyLub3md3jsEb2CsSeqmruuwWCwoyGLtOAA1M4w1sFWVkma5wZFtdlHtEojsdn0HKIW2XaVmjioWwkS6xIgg+/KfrFzikIW8zHK22voxSMVzlECXnyG7oUn3i1VfdI0+NS+C4vOg+K5G3w+U3gqypvJcNAQi4BwUclzjRfD8QOrhrVsP2W5OzTYTG8EcREMWX6qH5VzZmF4c519Sw4prW0i/9wyuh9T0k3oo9VAguPVw8icbfNtPvh0f+vr/7wu9FIFAIBAIBJDmYkZkUM4NiYCxGqLvezhP9rj5/BDX9/awt3cdy+UC3gdYY9BMGmxubESSEVBXNVxDlqY+zhurbJXsfdZaVFUd/66y8kIXw/6jnY6TUDmwhJZbNNGMBhwAQ6TH+XFYCZgQeR4cT+oM5xxWyyUp9rxLxJ7mGpJaB0CaVwUgzX0q7YlQKhGfkkCzjWtEwNeI5xo3TMfQRkPFoIFh7wD9chd+00N5trkdtSSOFE5Kw6kwGtjPf6fz8z6GgKd6jUuPb0ANBwDcaNFKUYNhMp3CGoPVaoXVaoXlcpmTdGNDI8300go874ztf0SG1/eAiPRqtYr720brqB9dT0mARyEVEeN01UBz05xPn+c0XD53OSuu/DklAkc1WRn0kI+jkurGDTEMwpAaDbFBQmvVtCsqFFbS3JTJyrRsm/TBwyhzpGkgeJ4h9fDWrYdqgO451EbH2ZL586xQHCs+NVycBckNQ1YLjs4fm7IhNmCpvkk9lHooENyC+K3fwxMXvxH4+ufuFP9w/zT+p//u7djCR567kwgEAoFAcIdAmosRzjsEl4kZBxH0XQfns7Ll8uXL2Nvbw2KxjLO16LPTyRTGmpSc6hypezY2NlBXdbIAsrXLjAIEClITQwcG59LsrdJ+xaoUD3+EcKzbAlMiZZEcmlUqecaUGxzmc7IzNpNmlPwZQg5VALLSiKyBcQD+iNj61Awor6mUa7AteV2JEaJih+ZkGRjtMQTaC/vkgE9OZnjt9iqRM4ZWCn5tL5wnaQrPYCwbEbxXpDTx8CFg7iqopw7SIP9EOiOBnG3M0DQNhmHAwd4h+r4jRYpWI4LMBFErTSojN9DsL82kFkfXGpN2eZ85EILJeIl1hdFo/9YUOM47YEAkxgZaBQwxAVaF4vNQ8Xw5yfQ4lCnd1ECawRiDvu+BnoMsckOFZ+T5mCobNDUuEAK0MVnh4/KzzKSf7YWCFwZSD2/teqj6PlqSNUKgxOcoNYW2Fn5wQNFw4+Ynz2Dk60/nAikVHVt3Yzq11EOphwLBrYwX/VONP3b/9+D9r/qV5+T4v7e4H1v/WBqLAoHgNoJSgNJA8FDGQFkLVBX8fAEcN9ZHILiJkOZiBNv8tNKoalLOsB3Q+4Cua3H1ylVcuXIFXUspp0oBdV1jOpvCWou27dC2KxhNtr9Tp05ha2sLVVWjbVfoO0rRRLTrsXXu2EHtIYBZU7YIxs97hzK3IBHoQplHMwZdMeD/KEFSitJb2XbFwQVVRQELxuik3mCFBhE/+v4opbRQaGQlIV/GUVK0TpT4Gqge6vRgciNgeGoP852TcBsDAJtST0uk2YoFOeXXaX98cS4gRIv0k53G9ctTKHQjwshgcth3PVarFdp2RbPI4r6lPQfPjYsWxWizVkpBeZVmza0rZPizwzCk/aY0VQ03rKlyFAUOlN8d7Wt57Ggx1MqALa4GOaW1VFrxPUmNFv4MCptpca+00dQoqmvM53N0bZeCC5SiBgGFQPi4B3qkFFJKJzLd9306LodQDIPLz5fgeYfUw1u7Hvb9AGP62PTleYI6zp9VgHPwLqTfxXE9zM1WIKv7SMkYm4purAaXeij1UCC4FdH8H7+Nz//RbwFedfOP/fP7Z/Gv/7eHcB6/efMPLhAIBM8FlMoNRa2hmgZqYwYACKs4E1vcEILnENJcjOi7HlAKtiYLXN/3NJDdBywXCzz11FO4/NRltG2bvmO0IYWOsWTlalt477C5uYmNjU2cOLGLpm7Qti1Wy3FKKKt5yvldAKkbgEicC7BShhMkmTCu231TsmckP1ppKMPsi4bclxxKgVQlPNOMmgoDzT3TOaXVGJPIDc+QKgkxz/x6puBEWLbmhUAKonQcA1ilUpPDuQFmr8dndib4uq0OOhShCPF4ZVADp9ymOYshQGlNDUUA3rmkWrzWTWAvHcIBiczydzhNdbEguySn3bLyJql0imYCK20o9IAaEpy0e5zMhok/NVX6qIQaE1vEtZU3r1TbpPfXQI2EobBQRiWO9nA3+B+v0inD+NilEohmWtLxtrd3cHh4iPn8EE1Tp8AEVmQpKNhIkr2jRlLTxPl6rPQB8n7FYx/XQBY8P5B6eOvXw77XMNZCWwNojWANfG0o7bhX6ThADsWhehibjcXICO98mmFZpkVLPZR6KBDc6jj7W8Cfeeg78L+9+AM39bj/8tI34PzfkcaiQCC4jUD/UQZV18B0gnDmJOYPbsPOHZr9A/iFR3AuKRvTdwSCmwRpLkb44FHbOhEf5yjhsetaXLx0CU8+eQlt245IBQ/6Xy2XNNvLeUyaBhuzDVSVRdu2GIYBy8USbddS8jGnmsZG2LoSxnsT1Qo0dJ4DC8qkT+eGpIZhEs1ke33+lNIKCApKeTqbopljIUQliaL5YqQOIUti3/fo+wFO0x4MgyYyZDSMN4kgMtnhNNK4gLSfSWnB7EwxORzXsfU0TiA3AawxcC6GOVzew7XJLtq6pVCWOAuNGwEliDBi1GhI99qRtdB7j6uDwvKwhgl0Dymt1iKEAQpE/LzzaNsVEW/vUyOB52+xQofneTHKdFLej8D7sL7eeJykcoIa2biP+2x5nGNnjiUrI+K8tRz8wAERCHnmnPfIjZdSSVWQ+XKO2mq1Qtu1OHXqNACg7zvMZjMslysMfT6PsrRWShd2yTbbNHWyjTLZ5+fBWHOkoSR4/iD18Navh8MwUHPXWAzTGn4DwIaDhoGxE6ilo7ujFOA8lPMIw3DkvyG987EeumIOZZB6KPVQILgtsPVLH8F/eun/CfjhD9y0Y/7zw2188j+8BA/i0k07pkAgEDwv0ApqZxuLV5/Dk3+oRve6BYaFxekHXoPdzy6hBw83saiuLqAuXUU4OIRftWKZFtwUSHMxwloLpVWaJwZF88UuXSQivVwuI8mioe3GErlcLpdJ0RKCB5TC4eFhQU6Btm1JbRLnfZWEK6ketE7qGGNNDjGIqhL+XvDhyDFKIkvhBYg2rqzQ0VrHmWhMkrJyiFOSrfWoLKW2tm2HruvTvCyelcVrM4YeHR7UP/IlRjAZ1BjbBfO6x6EHrDbMBwAljHo61zAM8Ic9Hp0rPLBBihajNQxsUqmkAfxA2iuAGoo+kDV6cA5PdgpXuw0s9mtUX9yHixY0TvlkOyItg2az8dwyUvKMlT0puIAVNoh2OW2g1NqcsGjJ43WuWyJTAu7oHofYRI2vr+3lOkrlEe99CMAQXEo3tTAYnCO7uAIpfwKT5TUlaEHe87xKakZorVBVFhsbG9iYbaDvetgYdEENIroenm/W1A2MNTngAioTaR+gDO2hXn8eBM8bpB7eHvWw6zqgrqH1gGrS4sHmOoyxuFjvYD6fQBmNoDxUZ2AXBmoBqBDguLHrae6gc0Pey9hglHoo9VAguF2w+cWA/+dTr8SPn/7U13Scn98/i3956RuosfjXPnyTVicQCATPI3xAmDa4/tIaZ/7IE/ifvu4X4YLCu1/zVnzot16NUHmYnR7mD07g7O/sYOuTV6CfuCQzGQU3BdJcjOA5XkNPdkDnBly9chVPPfUUhmHAZDIBAHRdB2MNqqoGQkBTN9BGo+t6hBDQtjSDCsjKDVZ4JHVOMeSeZ42xqoNnTfFsMIBIZ/n9cn7WOAUVhYIHIzJG88RI/UOciAg5E2rnKAW2t30OMAghzXpSWsEpB+MNjDYIlggdz9nyntVGesTx0rqS9Pro3jORMmlmV2wIghQkTNxDAHDlAFewjeqBOe6pOgxao4r7YYyB9roIcIgNCMQZcsOAy73G3FnsPbUBc/EANiwxRFLXNA2UArzrk3KE7Xu+aLKkdRcKqxKcJquVRtA+3QZW2CTRklaAL1Q2bCvEMcobVXxPZVLNih6tdD5+OQ+sIMMhSqS814ngr9s5uZGQ3gxhRMhp/SHNmrNVRXuvDTY2NqCNJuWXYs4f0rr5mV+39/G6ymuFUqMgDcHzC6mHt0c9dG7AMF+gCsBqaPDomSk2G2DlHHTdY6dpMVEdFk2FK2YKrTTs3MK1HYauj7/HHChDqj0n9TCvXeqhQHBbYPd//TDet/gvcPqnDvB/2/3Ss/7+P9w/jd9b3E8zFv/Ob4piUSAQ3LYIfQf/yBdx4V/1uDjci5/8ge/CyzeexFOrTZx66VW85b6H8arpE/j9V9+Lf3z/G7Fz4SzO/McN2E89Bre3Lw1GwdcE+a/VCCaxAHB4cIArV6/g6pWrWK1WmEwmUEqhbTtUtkIzaVDZKn236zu4GExQqmtGdi+loE228TLxZIKR1oGAvu+TxYxJXwk6bCgPfQSsBimTTcfqB5MVF1GB4kFhAV3XwxpK9NSB0liDz4msTjmadeZdDDpwo2sZW7joe0CpAuE1j21eKQ02cOIq7aExGtY0MNqg7QB39QCX3Cbci/ZRGYezISRVTVA+q09ASaRXemDVO3jn8eSXNmGvL6Cxh8F5eO+gtUFdVxTYoDT0hEg0zZhzMZDAp43OpFaNlTkIOYFWKwQV01e9y82PFKKgEqFmgl3eUJ4flq13Op2blTbU7BiKPdVrpDQ3Jzj1lex+LjUsSpUMPb/j4IBybllajw/ouh6TicN0OoVzHtpQ8Md8PkffD+l4HBShogWyaipMJhPUdU3qHVZfKZU+Rw0Ujbqqjz7YgucFUg9vo3roHLr5IXTX4nA5w+FOBTUBTm4tsYNDaOfQeI+mMbiiZtjzNfSBg+86eJfrh5N6KPVQILiNsfHPP4pfmv9xVH/3V7BllvjTm3tP+/l/criDAzcFAPzcf/cnsfVLH5HwFoFAcEcg9B2GR7+Ie/75IR679Ar89htfhf5FLd70yk/j+3Y/igtW4S2zx/Cmb/19/OOv+xb8+xd/PV72S/dCf7KFXy5H//0kEDwbSHMxYm9/D7PpDMvVEg8//DD29vdQVZRw2tQNDg8PAQDT2RQ7O7tomhp9T2mZAJECVjAoRfO30tD/SAjJxksqhdLKx0QqFH+YiPOsLf4sq2YS+WJrVqHgOariKXlaVvSw/ZA/WxL0wTkojzz7KWT7V7ImRutgCJYG4weaW8WzuhjJ5hdVPOW60vUXoGsg9Q+rapTWyRLonMNw5TqevKqhZ1OYVx7QnncDKYpiuAMCMLgBT3x6B2HlAGgoHCKk8AEPYwzqukZV1agqmrG2XC7Rtl0ilkyGee2khtJA3HO+Pm5K8DXR/cjWwJH6hvc6Hre4eDoPxoS6eBta6djgAZaLJT17CCk1tfy81oruC0edptMXc92O3oDRjTnOCoiiMdK2K2xvb6OyFa5cuQIXGxapiaN0ajI0dYPJZILKVli1LRaLOZaLZX6m47M3nU0heOEg9fD2qod930M7B933sIcThFmNy0uL+dYEFZZoXYWDRQO176AP9uHaPv0PMBTSLfVQ6qFAcPujft9v45+87xzMq16Oc//HP33az/789/wxuE9/DgCwhY88H8sTCASC5w8hwF29ho33fwIv+9gJLF99Hh/8o1+Pyw9t4rU7T+AL89P48O++HKc+ZvCKj1xF+IPHpbEo+JohzcWID33oQ3j961+PL3/5IvYP9mGMxckTJ3Hy5EnMD+dw3mEymWBnewe7Oztw3mGxWBChblu0bZuGsTMxI4I5tjcFZPVGqcBRCjAw6XPEU4iM++BJ+cMfBNvINHiGE4IC4HMyssrpnYo5ELKqRGsmRxh9tiworDxRKqeEAkTEXHDAkJUeCgou+LiGrGBhmyMFHpQqnFLtksl82qeCsCtuGIRsf2YCOcznePQ/mvjdTBi1MvEaNIBltlmGbP+z1iZixzPmVqsVlotlIpD90JPCKaqsgoqztwLvOeJ1IzU1cnhAvFcluY3hM8kSWhJqRYTyOLsnAlkurbUwxmDSNLCWGgtQKqmKyhAFBQXvjtpF2TLIr5dkX6+pfUqFD68jPR+BVGXWWtR1DQ6/4DCd4oP0uxHVPLPZDEopHM7nWCyW8MGj0lWaq2e0QVVVePzxxyF4YSD18DashyFg6Hu4wQHLJcJhhStTg2AMql5Bd0ugH+AHB6NzEJbUQ6mHAsGdBvfwZ/E3XvKGr/Cpzz0fSxEIBIIXDiHALxYIXYfJcokHr9+LL33mJfjC7kux84jD1/3BIcyT1+CvXadQF2ksCr5GSHOxwGc/+zm07QrGWGxubuLkqZNwAyULz2Yz1HWNZtLAB4/FYoHFYoF21WKxXMA5R6SsqtA0dVJ0OHd0bsGY5ObZXNYYIjlOpfcAILg8y4tREiAeOA9omtM1Uu4gkRT6Hh8hH4/VQ3l9GB2DyV/6Jg/2D2RBpgTWkGZuhVAQ6XgerTUFOGg/umZWMvlivoP3RJLZlsm2NCLXIYU9sFXSJwUOzyrjhFmd5rtxkwGgsIqmqVHX9AcA3OCwv78/Ul7xPC0mqC5kO9/RDStUNYU0itNwoRSCL+bGFZ8N/ukLOQc7OEcNGGs8WtsC4Llo9H7f8z0+ukYm0CPVDf0jrg/jdfG5+Rksvqe0Sk0B56jJVFVVbHAghXcghNgIou9ub+/gzJkz2NnZwZNPPokrV57C0Peo6ipuI93buqnRdf3T7onguYfUQ14fbrN66ADvoL1DtYyqUW2gtYJXGt6w4pKOL/VQ6qFAIBAIBII7F2EY4K5cBa5ew6n/aAClEeLc7QGQpqLgpkGaiwUWizl2dnawMdvA5tYWFosllssFQqDkzOAD3OBw2B9if28f8/kCfd8BIIJmrYU1Bm5wKRggeCJSPLOrJDpMJBOzDkRkrGHVXbZRqcJClkguwuh4TJwBIoeskEkJpgVB1skaHJJ6ozwnwJa3PNCe7caJ/Ckii30X4IyDtXY0Qy2EcaCC1ioFrhhrUtOPz5Vmfw0OXd/lVNi4h0C2tdV1A+eGbIGO98BoSjhl9Qsl0/qU7qqNxnQ6wWw2A4CYAtuia7tE3vlv3l+yysX7pPnS1ciOOSKha2oWAPRcqGi/xJjslt8dkVYmskXwhVJkk2vbFsGHGCBA8924uVKqb5iIBz8+B5+T5osVik+dlVS6sJiPZqmF8rMBdd2k5kZdV/Q7EdgOSorR3c1dnL9wHru7J9C2K1y/fh1d18Fai74f6NnQBtaaON9tBcELC6mHUg8BqYdSDwUCgUAgENwRCAEh/nedQPBcQJqLBU6cOIHt7W2sVis8+eQlrFYrKKVQ2SqpCfqhx3K5xMHBAUKgRE8mkNZasM0PoPdY1bA+6D79HThx0o9Iy4gURQsfExRGHlzPCpnCKlckUBKhC5kEFXPKeJB9aQtLn1FqNN/MWprBxeQUHkSog0dw+XPG2GhX0wB03A+fwxqKBsBxaqBhGDBEhVOZmFpaBYHcOAieFCHDMCDoEC17dP1VxUETFZqmQVM3aW8WiwUODg4xDETmytAInr+WGge898w7VZ6JVd6TEt6HpGAaYsBFvqfRbscHKex+I+XMGpFlrFYreOfRpIaLSmmr5O5UaY3ZYphDC1gdxWqkvGZffDcfm1E2hJiIb25uoqkbUqvF/Wbyj6i+uXDvBdxzzzkcHhzg8ccfx8HBQaE01anBg0C/P8cqogTPK6QeSj2Ueij1UCAQCAQCgUAgeCbQX/kjY/zGb/wGvud7vgcXLlyAUgr/8l/+y9H7P/ADP1DYvOjPW9/61tFnrl69iu/7vu/D9vY2dnd38YM/+IMpIOCFhFIaly5dwt7eHvq+pzACZFvZarXC1atXcXBAASKUqGlgrEkkQimNuq4waZpkC9RxgLvzDj5kZUpJYkq72eBcnENWfKZQgygQKVXF/yuPwxZg/jcdB3R+z4QxXjP/P6XjnzHhX1+f1mRXNMbCVjSbS2sDBZVsj33fYbVaoV216Lo2Wfq6juexrbBcLbFq27TPXddhvpjT3KnlAqvVCkNPVrBRGASIPDs3wA2OxDNGJ0UIr1trBRPXFULA9tY2tra2YKsKqxUpRebzOUIgkl/Oe+O95X0ICFH5o0ekl/+sfzZ9JymjYjNgrYly3D6zNZKPOVLchMICrnR6TrTWpCQLAOI9KhsW9DwURBq0Ht5P/j1N9xoBIYyt5CGErJoC4gxLOt50MknPeU7FpXN0fYe6rnHq5CkYrfHk5Sfx5JOX0fc9qqpKah+lFOqqRt3UaJpmFIBxq+JOroWA1EOphwSph1IPnwnu9HooEAgEzwRSCwUCwd2MZ61cnM/neP3rX48//+f/PN7+9rcf+5m3vvWteM973pN+bppm9P73fd/34ctf/jLe//73o+97/Lk/9+fwF//iX8R73/veZ7ucm4rDg4NEzEKcZcUKECKDHUJAItFAHgiPaONjux+ARDzYTlaqaYCsikExTywhETE9IonripobKRpKop6dhkfPwR/Ib2VbWEn0iBCGwpKWPx+3AmU4AFn0hrQ/vCfptCEUdkH6Hn+e1zk4AKsWrvKwhux9taGZYH0k6N77ZLukawxYzBekFLLUAKmqGv1ASbartoVzQ1L3jG2NKgUK8DoVFA3xR7YiclMjXQ/C6H6sq2oQsoWvnDlXfg/wIyJKx9Lp+6V9kMMX8tw4lcgtWz99ofzyXielEy+tTLzV2vDFJpVYSebpun2xLmpU9P1As9qamu5PRQqt8hjeB5w7dw5QCo998TF8+ctfxtD3RL61Sfd8Y2MDOzs7CD5guVqiazvc6riTayEg9ZAg9VDqodTDZ4I7vR4KBALBM4HUQoFAcDfjWTcX3/a2t+Ftb3vb036maRr6D+hj8PDDD+N973sffvu3fxtvfOMbAQB//+//ffzxP/7H8bf+1t/ChQsXnu2Sbhq6vsOs2gCANHyfFSYm2vwqW41sfWX6J32xIFIqzw1TbFkLRKaIgK8T1rF6Jx+GrIRMVPm18vMlWS+/N7LNrStN0nlj0ufI3oWsDlIKLJzgtMxkwQuZ8ANkfaOUTp3UHoN3UD4GJUTizsmuvC5tFYwhy2CyRCLQrLYecG6ADx416pRqmpoc6BAGTnJV8CBVFFwktiHg8PAwhRgkFYrPtkpSK5Hdkvc4v17Y4BQwEvyuNTVGz0F8zwcPa+zoc8fdkxCyBW8dpX2QFTIlKLBBRwtdbliEI7w+HJ0nF/IecEOhfC5V8UhaazGZTmGtRde12NraxMbGRk7PDXTPgqc5cufPn8PZM2fxxBNfwmOPPYau6+LvjEoWwa2tLZw8cRJVXaHrOrRdi4PDgyN7cKvhTq6FgNRDqYdSD6UePnPc6fVQIBAIngmkFgoEgrsZz4nX5gMf+ADOnj2Lr/u6r8MP/dAP4cqVK+m9D3/4w9jd3U0FEwC+8zu/E1prfPSjHz32eG3bYn9/f/TnuYbWbIsj21tlK1RVHWdnGXovEmkTVSREsvm7Y2sYCx4027y0yoqeSIpHRDoSpxCQSCIdIxP3NAdMlcfJPHn0cwgj8uQLWxqvkz+sVMm5x2Q9L08nolkSa7p+Q7PXtCYr3poSqaoqzKYzTKczNE2Dqq7SvlYVNSxMEfjA9jPvfAqGqKoKG7MN7Ozs4syZs7jnnntw4sQJbGxspLROgCyCKg7311rn4IaCSKf7FGeCje4FNwlCJPk+71N5n0o7ZmkB5L1jSyfbEtftfwCK5szaH6OTMowaDUiE1XtHVkrv0lw3ev5oHZTOmhswKt/Y9BoKJRivf/16+Bqttdje3sLm5iastdiYbaBpJvQ5n4/ZD2ThfOXXvRJd3+GRRx7BckmhBMZYsK1wNpvhzJkz2N3dxaSZoKpqVFV1bFPpdsTNroWA1EOph1IPpR7enrgT/ttQIBAIvlbcKf9tKBAIBOu46YEub33rW/H2t78dDz74ID7/+c/jr/21v4a3ve1t+PCHPwxjDC5evIizZ8+OF2EtTp48iYsXLx57zHe/+91417vedbOXeixo5lTuuRoTCXVMPi1VOqVdim1XAFKAAQpSNrIEokjVLD8T0iLo76h8CAFwcRaWUoD3WT1RWr7GKZpRpVMqd5DVI1rpkUolf+iokoTtaiVpB3wi+Uzay68rpWG0AjRgYBBCgHMObnDx/RgMUVVZIcI8z2e1x+BcJIR5Hzm9lO8FQA2HruvR9x3atqM5aNagjsETXdfDOzom2eHGipR15UtSsxSqHT4nBxOU1xsC0vy4I6odZBJcHvvIAdJfrI4ypGiJe86WR+8dtDY5tCHuf2nB431OFsc18HOyHmqANRLLqi0OkphMJtjc3AQA9H2H2cYMVWVJdWM0mkmT7B3nz1/A4Bz+4A/+AMMwJHUVEJshGxs4efIkdnZ2aEZaR0q00f7cxnguaiEg9VDqodRDqYe3H273/zYUCASCm4E74b8NBQKB4Ea46c3F7/3e703/ft3rXoev//qvx0tf+lJ84AMfwJve9Kav6pg/9mM/hh/90R9NP+/v7+P+++//mtd6IxhtktoiEWlrI5lVI7sUVFZ+lHYuBpNAX8xoggKCL4IMSsJTktmS2ISslgjK4bi0VLA1seBnWS2SrX1JAcOksPjCyBLIBDcolDOmjp2TFo9bXkZp9eN5WPRvh7btMAxD2ltS59hsq2S7W2EvHPoeg3MY+h6L5WKk3OHPGWMwmUzSd/0xyhoemu993uNyHzntFsgzwbIFjyyR/N0QkPZKK5PuSdlooVtOTQ9uTjDpxdo+cqOFVVBK5bldzjs441JjIgS6Pq0CgvYIIafiKqWhnIoBFv5IMyels/KDUijL0n1eu4bKVpjOptBKY7FcwBiD6XSalGdak2Jta2sLZ86cwenTp/HZz34We9f3MJlMUkiENQa7u7s4deoUNmYbqKoq7U1lq9FzdDvjuaiFgNRD/rfUQ6mHUg9vH9wJ/20oEAgEXyvulP82FAgEguNw05uL63jJS16C06dP43Of+xze9KY34dy5c3jyySdHnxmGAVevXr3h/ImmaY4Mu32uoFS2tHGYAdurRvOjir+TQiaS1zQPCpm0skXuaIgBHYqVE+m4YY38aCaBcaFriop1pQyfD9DjzyqV1CDj72elEVS5MJVI+rp1jW1l3vt03LHwJKt7+LPJGqgVnHNoVy1a1VGabE1pmdZa1HUdGxg6q0gmEzg3YLVaAQugVz3CEGKSrEtJsOtrY6sb7/1x1zLaC1Cggoq2S74wpdZUWNyoCMiWzkINta7CeTYKFFbbKKVRWUVJtMHCBw/vYqiGpsaCDz7tg412QQ50YGijU6LqqLkQz1Uqv4B4PdxsiPd/Mp2iqRsslgvs7x+gaepRYi1AzYzd3V00TYPDw0McHh7CWot+oMCJqqqxs7ON06dPY2dnN/eOHJ2zqqmpcifiZtRCQOqh1EOph1IPb3/cbv9tKBAIBM8Fbsf/NhQIBIIb4Tn32zz++OO4cuUKzp8/DwB46KGHcP36dXzsYx9Ln/n1X/91eO/xzd/8zc/1cr4iuq4lwgEiBtbkeVlKq0SQVBaz0M8FQUtznBKBVEnlA+AIkebvJxRvMJFkgmnY+lYoekq1zJg3R6VNae+KRCrN00LAsTOo+NisJilIMdsky4ZBuaRsXUMMDcjztUxsTlRVhaZpYsqnS2Sr6zp0XYe+60dJoqwiGoYBbdtisVygbTuyu0WLHH/OOUcqnkh8jyOx6xbJ4xACRkSafw5r89nWbhl4Np0qmxLFA6NiaMFIyaNiimz8w3vvvcMwDCnAgS2MxtI8NqVzs4bv5+Dyd0pFFTVsaC0BIdkW+b2RSif+0wcPhJDO1w89FvMFuq6FG1yy9QFIM8fqqsZ8PsfVq9dQVRYBAX3fQ0Fhe3sLZ07zTLEGxlgKtkBsSvmAS5dubPu4nXG71UJA6qHUw7XbIPVQ6uFNwu1YDwUCgeBmQ2qhQCC4k/Cs/yfxw8NDfO5zn0s/P/LII/jP//k/4+TJkzh58iTe9a534R3veAfOnTuHz3/+8/hv/pv/Bi972cvwlre8BQDwqle9Cm9961vxF/7CX8DP/uzPou97/PAP/zC+93u/95ZIwBqGAXXd0BD+qCihwfeaCFskMyMeVpDVkQrkiJrmBuRtTdFRWveUIltiVVXjFNaScBNrGxFsPl8ihOvrUeMU0PW183l88CPlznhuVgAKq15O8CSr3LpV0FgKfuC5baxc0kanpNOqquC9Rz/0kcBVaa1t22G5XGA+n2O1WmUyr+l+GEMk1TsPGEBDJ2shXLEe5KbD+myxkWKKr1lxD2K8X6PvlNxcqfLrhfUO4/uxtvfH3QvvA/qhz5a8eJCycZMshFrBGJOaHTx/jBfjytRXpbFu7Rw1acKYXANA3w8IcT1sZ0xBGEqhqii1dm9/D3t7e6kp0rYdqqrCdDrD6dOnceLkyWjVVLAABudSYIUPHnt7e8fu862GO70WAlIPpR5KPeTPSz18etwN9VAgEAi+EqQWCgSCuxnPurn4O7/zO/gjf+SPpJ95vsP3f//342d+5mfw8Y9/HL/wC7+A69ev48KFC3jzm9+Mn/qpnxpJtX/xF38RP/zDP4w3velN0FrjHe94B/7e3/t7N+Fybg6UojlIed7VOFHzuMH3/HNpAyxDDUrBBv1NBHidV5UERilKWuV5V0w4vTuqFClJ1fiAxYnXF1F+/xhrXLqmkO18/PpxBDqvRUPrsfJHQaXAgZJkK6VQVRW6toN3HsZYWEuBBzywX2makdV1LRaLBbquS8csL9F7n0i8hoaxFDgRorrGF3PdUBBQPoZS+ZiqaJgk+2Cp+HkalU8I/uh+8glKInwckS9INwdFkGBpSA2I4McqG1IqAVVVpKj6QOoe749twLBSyJcEG6QoYyslXztZNj3adgUfU2eVpgCKuqrR93367HK5xGK+iEohh8VygaqqsLOzg+0tsv9tbW2B56QxjLFomgk1Qm4T3A21EJB6uL4WqYdSD6UeHsXdUg8FAoHg6SC1UCAQ3M1Q4YbykVsX+/v72NnZwSZ2jhCSm4GtrS1UVX0ksKAkQkkdEckiE7UjNrqnIdPrqaKJrCMnr5YD/b13ZHE75lxjAlQcPwYrlFyaF8REqkxKLa9rnfSuE3Zef2lxZCWOMWT3YwteCk2I87FIheLRti0AYLlcIgRg0jSYTCY0UysmnBpj4H3A4cEB9g/24ZwbqWn42Ly2ckaZ9x7OOXRtl+ydfB+J9KvRdZTXu36d3Nw4Et4w3laUiaNJYRNvzPp7689RIsxrx4FSND9MKZqndkxDZTqdwVqbVE08f+y4Z7g8f24WRIticV2klqrh3AClFOq6Scc4f+48Xvqyl2I224BSwN7eHr74xS9iPp+jXbVYrpbw3uPkyZO45557sLW1hd3dXdRVja7v0Pd9agxopeGDx2KxwAc/+MEj+3ozEBBwCFIRbW9vPyfneCEg9VDqodRDqYfPBnd6LfwO/AlYVb3QyxEIBLcBhtDjA/hlqYcCgeCuxs2qhXfmpPCvEW5wiIKENABfYRw8sD4vbB3HkdKSZ5ffS+QJY6Lqg4cKai1Z9CiRXp+XFVCQXiZQoThR8T2lxoqSseJGHXmtJJpJ3RJ/1ppsd1Vl0xwxHedHcSNgfY4VKzrYupYIvqZ1ucFFAusxOEdpoDEEgckgn4PVOFVVwVp6tPueZm05746oZPiyQqD/411W03j4RCzzmjWc86NrLkk47z7CMQEJ5b6yddGPnyX+WE5vHSuDAkK8l/kamCzzZ4a+j+EFY8ULf3ddWVY2UkY3NTYpNjY2MJvOsLe/F5sYtKdKK2xubiYi7Z3HwcEBuo5mxXF67anTp3D27Fncc885TKeTFE6gFQUqBBOgAqmihn7AlStXILi1IPVw/G+ph1IPpR4KBAKBQCAQCARjSHPxGCyWizgDKSelritusE6W+NWkfMARsrZOaGkmlI5D/ot5T+t2vOL8x52PDjZi6kmJwqdzPhPWpCApGgJJzcFIRLq4BgBGZzJmjErnoPdMEbRgiOTGpoBzDn3fJ/IMUGPAOyLA/B0+NqtxrDGFqomIrDE27RuANP+N1UrGkI1wGAb0fZcbEDpfY6k8UoouNPh4jvj/vHfjfeEGR/EaKY6ePhcpqXX4GYqEm9QzxZaDAh600bQvcKPjkM0xzruDT2Q8hACtNPqhzz9rA6U9lF9Tsil1JMUXGJNqpaghYzQlnU6nUyyWCwSfmy9N3WBzc5NCDfoB+wf7mM/nsMZgsVig73tsbGzi7NmzOHfPOexsb8PY2FhxeVZf33Up5KJdtfjEJz7xtHspeP4h9RBSD6UeSj0UCAQCgUAgEAieBtJcvAFKi1Z67Rjeyhjb+kIkXiER2HTMUmkTVThkVfM0gH/teF8JSsWZUMXiRmoYnwlgSZ5KIpfIOrvVkJUaJQEvmXVpUzxOjRRCQN9TeqmPqabDMMDH2Ve8dqUVVKDzVFUF5/JnK0tWQABwzlESKpACEIzJSpoQyAqpFJFOPgbNKFMpgRMqJ9Xy3o/uT7kPxzQYtDL5Z97z9fsUCavSdI/L99PssbRfOhF4bXTaTz5uee+014COVkutMASXFWQISUWktc7Nm6fB6FlAtnDyszg4R5ZYW1FzJM6IU0pjOplia3sbWmkcHh7i+vXrcIPDfDFH27aYzSis4Nw957C9sw2lFKmu4sw3KFJKMcEGxv8W3FqQeij1UOqh1EOBQCAQCAQCgeBGkObiDeCcT+EFDJ5HxWA1zmiuWCCayYT26QgNf40JHSkW8jFDYBXNUXJ9xEYItvv5Qi2U/z1SlqypTeirxbwpJuMe0Q5HxLuclaXS+eJRoxqJCDOd1zu27rnx2ou5Z0YbIvxMbENI1kuAFEZ936OP9rZ87XmAP8/bYvVOu2oxDAOgVJpVViqPRvPTfBg1H8aWPD0itJqVTXH9HCzBTZQ0EyzkY9H5dLT2ke1x1IxQINWOVqSWCfF54OM7jFRV5f2ml8Ypp2mPC7vj+I2QmgZam5ESS8XnACHbBXnW3mQySc2JyWSCs/ecxebGBnzwODw4wDAMaLsWTz31FOq6xsmTJ3H+3Dns7u5CGw03kAWUZ8wpRcEVTd1AG42u7fDUU08duQ7BrQGph5B6KPVQ6qFAIBAIBAKBQHADSHPxBtjf38Pu7i6qqsrEK2TSGJkHkVisEbNQEJmwPtdKJbUID/hv6pwQxsP9iXAcJc/HkfN4yESo14lyaQdUAYkYM6kq56ElRU5xIlICeQBu9F6yt6295gYX51zlNWqtks2PGZz3Ac6TkgcDMOgBRpuYkGpHqhtOzQwBo1RN3hutSOXiPRHryWQCgMIMmPQyifaemw6A0h7a61FKaB7mz3sa771PLJnUQogJuD4cS169J4Jf2SomjLqohCHiTcqt3LgYghupkLRW8IoaEiObqA+jv1PoADdy4sbT7LOs5BqtUSE9x3ytZD2k2WqDc9jd2UHXtei6HpPJBG3bYrFY4szZM7hw/gKMMZgvFgiBbJvXrl1H13U4f/48zt1zDqfPnKFZaiGgquvUmOD7Ujc1tNJw3uPqlav4zQ//5tGHW3BLQOqh1EOph1IPBQKBQCAQCASCG0Gai08DsqyxbS/zS8UD65OaplBpFCT1SMgAfy1+x/sAbcj+xHY7Y81YuRMJ5zpKhUXIHA/G2LQG53w8l48EMsBok61oYDVJJtFZeUQkjIliaZfj7wBI1kZWy3AaqVIqzcOi7+hEZtNeRoLqfYCHgxoUdE3ks66rEfn1kfwi5Huio7rFWBNTQcl2Wdc1qaygRtayrEbycA5Qyqdr4CbHqNnAjRMgkfVyNlkZyHBEEVOodJRWyZZIIQ8FoY6NgFFDIpBSjAmw1greqXwNoO9Ya1PTgkhrvNcup7KOVEjlPQbPVCsaNfzZQBbQ6WyKy5cvY7FYYHNzE94HzGZTnLvnHE6cOIG2a7FardD1HS5dvIT5/BDb29u498K9OHfuXFLyWGtRWUoEOTw8hHMOdVOjqiq6t0M/UnIJbk1IPZR6KPVQ6qFAIBAIBAKBQHAcpLn4dEgKFqS/M3EtbHnpc5E4hvj+DexlbIdyzmGIRGp87CwGQkFc0yyxZGkjlVBSbhQ2PbasMSkjwpYPXhLddSVS/tmPVTx8XibrcX4YBwfwBRhjiLRz2EHBk/I+RJVTVC0xcXfOo6roGNZasvW1XVLqcOOBZ6sZozGZTFKIgQJZ/5hocwqpTw0Ftlr6cVNgnVDmBWdSGlRKQuXmQWmNTPc3eATvk80u+GyPLOeH0Yy28Qw6momWGwd87GwrjXPr4r6xOind67VrKAl1QIjhBSovt3zo4ud98Kh0hfk8zwtTUJhOJzh58iROnTqFEAK6tsMwDHjs0cewt7+HppngwQcfJCK9uQltNKyydA+QFWGcnMunZbWO4BaH1EOph1IPpR4KBAKBQCAQCATHQJqLT4P9g31sb22jbupENEsyCyATaTApG5OC0iZXEgpGspFFlcXIVqfysH2e/QTk05fBBUkx4x20NjFVVEPrAK2qpDLh7xC5HJO2cn2sNOLUTFa5eB/gvUt/8+eMMVBaQyt9ZPYWrzk3BfJ56foMkVutoqrIpbljnKgKpVDXVdo35xwQVSB1XSMEwNpo+9NkpbPGpGRWHXQkwBrAAO9Vvj7Nc82IMK8T6tJGOQxD2iMm6/SZkpRrKIRo6VMYom0xq5ZIQcR7yns0nmkXn4lIgPlZYkLOHNx7n4IdeK38rAFIIQr8XCZCHZCeM1b5MLSmhsXh4SG2t3dwz9l7MDiH2WyKey/ci9nGDF3fwTmH69ev4XA+h1IKFy6cx8te9jJsbW3BWpv2g5Nxgw9oJg1ms400r89ogy9+8XH863/9PghubUg9lHpIa5d6KPVQIBAIBAKBQCAYQ5qLXwFkbQvH8OdMgkrFRpo7FXhOVQ4SKFUvJVhxQ6mXNGzfuSGpZhLhUllhwWR0nQADgPcOAwrSDQ8UTkLvA4a+Tz9z6uU60Q8hYPAOzg1JDVLaG2kOGJNJJorj61u71PVNHKl/jBk/jm5wSY1jTJ32tms79L5HM2mwubmZLIB1VSdiqFQkpz6TXE7oTLPewE2IqDDSBkGPmyO+UCuV6poyzGJd8aS1gtY2pd2G4JItsXw28vVzMyXvX3pW1pQ25fZpnVVEvP95XyOp9uN7ytcEANZYsN0zXxsFC9RVnQj03v4etre3cf78+RRIEHzAlatX8LnPfQ4heJw7dw4PPPBibGxsoq5qCjxwDqvVCpWtUNmK7KE63w/nHNk4C7WS4NaG1EOph1IPpR4KBAKBQCAQCATrkObiV0AiqmukJJEgZBWOClntwmSHbWTk3CrmREUyzMSGf3ag+Vx5KL4eE/liiD5UnonFayotfAj0meD9mBQCx8zL8kfOUyab8mfZ4lceY3RNIPUI2QzL+V75/Exa05yu1BQgpVI5z4seUJMIf9d1gFJomgm2tjZRVVWa0UUqHTUi85xySxa5AQ5upDxyMcGVFSvJGhhVLNxICQGjlNR1WyQwJrvls8OBBAoK/dCP7IB8HJ4nR8GouXmCMCb3JVKKaqHw8W6sJONGD6ezEmMfr49VPFpTM0cpIv1bW5sYhgHXrl3DmTNnsLOzQwocRTPTvvzlL2O5XOH8+XN4xctfgfPnz6GqKrJlGg3fd3BugNE1tra3Ya1F264oZTXe936ISizBbQGph1IPpR5KPRQIBAKBQCAQCNYhzcWvgMPDQ2xuIs1EqmyVLE5d343IJql64iwso6OSJSso1m2A5dyo9Fok0SrOyNLx+0yGeY7YSLGDNUUPAGMK8qUUtDLpC0zgU0BAnJeFkEmhc2THA0iRwwqi8rwlkSayVipJAkJwkSySLS8El9ZD5NCPVDPr87pCJLjGEElmS2DT1JjNZhTWoJDWZk0m+Xx9pI7J9jljTbLy+XiNyboZ+w/W5D1HsefGmNE1q3QdtGdpjlu8BL530IUFD0x+ySqYb9Fx9sP4f/i+BG4MsK3QAMXeex9GBDztqSrCGGIDpnxWWe0TAigsw1Fjou8HODdAKYXZdJZ+Bw4PD/HZz30OTz75JOq6wkte8hLce9+92NzcpGfB6JRo29QN2UJDgDUGZjrDarVCr3oYY/DpT38a/+bf/BsIbg9IPZR6KPVQ6qFAIBAIBAKBQLAOaS4+Q9C8LoWqrlBVZGlCj0xIIsliYsfKDARSvKyHAYzIL5OfSFattWSJi0oG/g4rUkLwKZggBQgUBAqs2FGAViqtiT8fXEwYDX4cQABk1RGi8iXODCtJJKtX1u1qiUgmcsbNg/EejQi9VlA+Eu5o4fPejRJWq8rCOYflYom2a7GzvYPZbEaztYq9LpNQ+Z4ppYA1IYgPPqWVlteu9HivlFJpXtoREq2ygioERMVUngnHz0sIAcqr9D1tdD5nqcQp1EGj48RNLWebAXG+XPR2GmOoCaNcbq6guC/F8zXan0S081w87z3qmvaz61p450mls7uDum7g3IBLly7iC1/4PCaTCc6fO48X3f8i7O7uAgC6rocbHIwx2NjYgLEmXUPf91A6Npl03j9uDghuH0g9lHrI1y/1UOqhQCAQCAQCgUAgzcVnCP4Pfx6qTyoGNyJZ9EEisCEEuBgoUJJPPgaQLX3ljLIytdR5h2EY0uwnrTVUQLRRrZFpIBNcH+AcEft6VidFi/cecHm9w0CklVUsTN7YtlauJatoyKZ25LqRiWYicgV5y/O3WDGU95bOz5/zaeB9VVHjQmuN1WqFwTlMJhNMZ9NEBHnOVr5Puvj3uMdAjQA+PpHpknTma6LGh45k3Bdkk19nAk/WwLgx8Zp537hZQmtiq52BViErrxLP1TDm6H6W88WSPTQEKAR4l62L0ICGhhvcmEgXBD0dNzY6mNBS34WTX+n+cEDBzs4O7r/vfpw8cRLWGOztXceTT17GMAw4f/48Xv6KV2C2MUPf92lmmPMexho0kyZZOrXSyfLHYRdD36MvZt0Jbh9IPZR6yMeQeij1UCAQCAQCgUAgkObiM8BisQQANE2TFBNMQjNpYmKU50qFEdHN87yANWENxqST5n35NIeJSVTvengfKNwg2gWTvS3O/iILn4NzHmZi8wyueE5rFPobBCqUBJXnTQHRglgQZCbYaf0FgT4O5cB++vxY5QOMk0G9JyVNUxMZW61W6LoOk6bBZDpNQQdMaFlJxESaFTKJqOrxcYn4DZGg5msJIcBFqyKT6LLJwKqgROCVgoY+QrZDwGh/S5KarttHxVUk4/S6LmbL5eciIMBaC+99ut+8dq2z8qck32s3KP0VfBg1Do5+NM87c85je3sHp06fwmQ6xcHBAR555BFcvHgRm5ubeNH9L8KFC+dR2Qpt1yUlm3c+Ng00bFWl81IjheyBAPCJ3/99/NZHf+uGz43g1oTUQ6mHUg+lHgoEAoFAIBAIBCWkufgMwLa0EJAVMj4TqHXVCpHsY1QXJZFet2YRA0s2OADwzidSV5LoMXkjG1hd11BKj+xi1pg85wpE6rzPSqE0rD7OGGPFUDmcv5yDRoKPMHp/TJLVyALI38vKlbhmH0b7QWsrrIaFNW9wDvP5As4N2JhtoGnqlDrKCqZ1211eF6+ZjusGSuschgFGm1Ej5Lh7uX69NDMOo896j5HqSEXpi3f5/vIepMTS0Wyxte9GUU9SZhX7qXW8v57n1XkoZOujVhpehaQ0onXn5yRd5trribSHAGMMJpMJpZpqi5MnT+DkiZNwbsDjjz+OL33pS7DG4EUvegDnL5zHZDKBdx6TyQR1XUFBwWmHqrKw1qKpawCk/KlNRb9LAz3fq+US+wf7ENxe+OrrIT3X67//qR4qxI4PpB5KPZR6KBAIBALBnQ5toOsKvm3H/0szQylAaSDOfRYIBLc2pLn4DNF1HbTWqOt6RNhKFUpJjBLZZlKJo59JCJnIssoDIVvFvPeJTLngE3kjhQ6RPFZyIK6nqgypJHRBdj0RTIDIKzTgez8iv+VQfT8iZSopSkoFCa+jVBuVyPO46Ps8ZJ8/TxbJrKRxg0NVV6gjCVsuF2jbFaqqonlVSicSHfxYAVQS+HUlinMOy9USbdvS+/H7oxljx18CrdGE8dwvJs0oQxrS7QQnvfKe0et8P4+Sd54fh0AJuXQvohoo7j/bEjkcICDAeZdIN6usvM+2w3yOQE2KeI6AvCbeT+89Njc30zmm0ylOnTyFzc1NXLx0CV964ktYzBc4c/YMHnzxi7G7u5vOy0R6sVzAGIuJmZKCKqq6OATDO48ePT7xiU/g4U996vgNF9zy+OrqITe7AkK4QT2Makeph1IPpR4KBAKBQHCHI3gE57/iZwQCwe0BaS4+QzhH8744GXWs6FhX4KgbErORahH536zIUKxmAZEushKS3ZDOPcD7PNzfaIOqspHMcZqoifO5bCL8fA4AYzsZW9UimSPyd9Tmx6SJyWEIYaSoSZezRrDTDCwwyeVPRsIXVUhahRQqMJ1NUdkKzjssl0sEH2KoQ1alKKWgjEr7UGxltN0hndM7j+VyiXbVQmsDpRBnxB39/6zG6iE+Djc6SktnfDXuF68hz0vTR56B/JxEAlva8fiZUdFa6IksG63JUhfndA3OQSkKblCBVE1sJUz2zJBn1sXHMTdi8mLKGxyDIiqcPHkSV65chdYaJ0+ewtb2FnzwuHr1Cg72D6CNxulTp3Hu3DnMpjP6uqfr8XE+nLUWVWVRVzWUot+d4AMGP8TwAoXLT13Gk09eOrL/gtsDX109LORiReMpS/jS/5F6KPVQ6qFAIBAIBHc6QkDou6d9XyAQ3D6Q5uKzwDAM6LsetqrAJDhxZSbLkRCraDHjN6NAolAyFmqNwurFg+r54KTSwVixE4fCG22gjU7WQVa9GGNoSLwqLH6R9Pd9h3bVAkqRTTBa9EqynH6mxaVrSGmfSZkUCWOhvOTvJRtbmm/mR0SVXiOlifcOwWg479I8LW00Vm2bBt6bNUsjnT+rgIBsn+R5Xrxvi+UCbdsSIbcmWiWjqkmP7Yxl2ESaWxYZqY+kEQbQnueZ6YIkh0Re8zUiPStp31jpA+TnApn4Ejmnz2lt0iy1ALaK5n1MDZjY4PDrDY4Q4D3oeHxLtU5zvvgeVrbCqdOnsLOzk8j07s4OmrrBYr7AwcEBBucwm85w8tRJbO9sk+WPfKXwwaOuatS7NbTRaOoGk8kEfd9jCESiffCobIVPfOIT+NKXnoDg9obUQ6mHUg+lHgoEAoFAIBAIBIA0F58VnHPohx7W2mSjWmdPaY4UK1kKEoowTsW8kboRIFI3OEekD4BzNHNMG53UDjTjKsC7nDDJg+1DIOVLmq/lPJwb0HUdhmFIJ7fGjNQ0I0tf2QSIa0IkwKVqx3vEBoIakVC2zRHJ0zBagUMXyj3yvrQhEgn3zqPvO+T5XERAjdEja2O67qSi0eleqRh+sL9/AKUUmkmTVEpp2H9Q0Mccj5VL9F6pNBp/lojxWLF1HEbvrR9DoUjK5edFj5ovPngiwErFIAAKqSBLo87nOG4NiVDn+8E31hiNqqpx8uQJPPDAAzDawBqDuq4w25hBaYXDw0Msl0tYa7G7u4vd3V3UVU12xNgMqWyF6XSKjY0N2vuoGBucI2I9nSLE1NRPf+bTeOyxR2+4V4LbA1IPpR5KPZR6KBAIBAKBQCAQANJcfNbw3mNwDrawpAFjTh2iPctHu5f3nJTqRwEEo+H/o7lVNGtrcC4RILYCVlWFcp7ViFRFIkxkDOi6PilvhoGINBMdWqdHgB5Z3EYzxZJNLRRD80tLoQbAs7d0ItBJUZOUKTSjKytJWD1U7KvzyaqWLWW+OFcGry9dQ8jhCEqN7Xbz+RzeO9R1Q7vl+phsG458nn9WSo0UPGUaLKujyrloCFFHo3KwhdZjAs5ku+CxCMFHYkxbWdoqKfWUFDBwKjVitKZ5XsOgYhPCpe+kyzgqFaJjxgaBcwP9bDQ2NjawsbGBB170AO699z48/sUvIiBge3sHW1tbMMbi4OAAy8USTVPj1OlT2N7e4Y0Chx5MZzMKMWjquM8OznvUdZWIdggBn/nsZ3Dt2nUI7gxIPZR6KPVQ6qFAIBAIBAKBQCDNxWcJ5xy6roWdbaTXxmoWVoIg2eNKjElVSeJothQTVhctbVVFg+GHYYA2OqlWmFR775Kli94ncjb0Pbq+o8RMZHXQeE6WPkJUeY3xauIw/BxmUNre6PoUzcEyGlrpRDTHqpVssQthSOdGcNRwUHR8YwwR7uABr5JKCQV55vWVoQHl+TKpJpVOCAFbW1sAgMViga4jMo0QwBbCNGetsC0aJrlKwRibmgR8nQCpotzg0j0jC2E8ltLx3uf7zc9K2rv4J6mkCgLPjQwi/S6dkwmxtUDf9/TdqNYp70s62XFysBDIqtc0OHXqFCaTCbZ3tuHcgKvXrqJtO5w9ewYnT56C1grXr1/HarXC7u4u7rnnHmxvb+VDeTpWVVkKzFAKla3QoYN3PbY2t6OyLUBrg1/7tV/DZz7z6aNrEtyWkHoo9VDqodRDgUAgEAgEAoFAmotfBYg4uTRQPwTAuSGpL0oyTWRxTPyS+qUg3AoKXoWk8GDSWtc1Eb5o73OOzktD4utEppXOKakhAKvlMrH8EJU2xmgMPUlFaDD+OGDguOtEIuLjpkBp8+PUUmN02pPjAh60CnGfsiXSx3UppYmYxWssCa5JTQSMVUlry2Z1i1IaQ9/jypWn0NQNNjY2MJ/P0bYtuq5Ndrry+3yPrDFEVuPfAFIQhLV2RKQBxDlbLjZHfGqssLUvWUAxVnKxWmvc4Mj7np6T4OFdgLIqNhvitSoNYyyC79MXjdFxf/y4SRK/ROEHgLEGVVXjxIkTOHHiBK5du4bPf/4L0Frh4sWL6LoWJ0+cxObmJg4PDnB9bw8+eJw5cwbnzp3DxgY1koym++2cQ9/1wCykZ7PrO2ijyUqoFPq+x+XLT6Ft22OfNcHtC6mHUg+lHko9FAgEAoFAIBDc3ZDm4lcB5yi1c2NjM5FntpaNyBRIfQMgKjMA6GMINSK5CgCCHik2OI2VrWjOkWqD7X5KmZSeyamo9F0iTX3fo+97OBUTOW80T4tWc0Q9wutY/7y1Wb1ibA5CSPPAbrBvqlfJCsnpr84B7KqsbJWsdLze8tppoQZeZStlmY7K6qWDQ7KuVVWF1WqF5XKJvu9J+aLy/iqVFTjGGFLoRDLNqiNrLRRUuk5en/ceyirooNO18P4Nw5DUNiGqd8omS9rXNQkPr4nJujZ87ACjDamDuHliTPo+3zutVVJXMaFParEQgKBQ1RV2drZx7tw5AMC1q9cwDAOaSUPz5wBUdQ2tFVarFQ4PDlDXDc6dP4/Tp0+n+WKmMphOp9RM8FmhRPPdFOqqhjEW3lNi6k//9P8bn/3sZ499NgS3L6QeSj0EpB5KPRQIBAKBQCAQ3M2Q5uJXieApaRRg5U45H0qNCCpbAwEAPqo8SrJapqOqAHgiXtaSQqS07gGA8w5d38Uh8TYRaaNNWocxFnVN5BKIJNzoRMjK6/DKp/OMbIJliicy6bPGoKosjDZH1kDXH62Dad5WnJdV7BGrdYhw0s9VVZHSyIdEQGldSAQthEBPrVMw4HADmuHW9xTksLe3h73re0BUiFAYQk/7Y2h2mdYqNSF0HNrPiiOlQEmqx1wfX0sIOZGW7Yp8n52Lf4cc1FBCaQUdDG8EWeRiU6Dcc7qPlHA7OIe2iwmvUbkUisYAK7ZI0VXYDrVKRFdBwVYVdnZ3cP78BZw5cwaf/9zn4bzD5uYmJtMpEXXFc8wGXN/bQ0DAuXvuwYtedD+2trbQrlqyocbnfHNzE4vlAsvlEtPZFMETqW6aJq2LGhn+mN8kwZ0AqYdSD6UeSj0UCAQCgUAgENy9kObiVwkfPFarFaqqTq+xLa4E27ESYeYQgIJflYSZCQdZyEil4aPNDNEuOAwDFvMFQgiYTCbQuuIDpTWk4f+KiA0RR5oHxTPJ0geApyXSyaoYbXJVXaGuayKXipRFPNifmwg8M6zchx5kAWRbYkAgshdo76aTaSTAHZHUYpO883DKpfXRugycyvu1XC6xXK7Q9x3tSxE0YK0lQj4QeWcbodZmRFD5Ouq6SUSb7Y1pX5CVMc65tNfeu6R08Z6IeWl35H+PbH9KAZ63OYDCVss1RrVSiHPm4hqNNvDWJwsiX+MwDIW9dPws2qrC1tYm7rnnHtx3771QWmNvfw/GWBhrsFwusFqtcPbsGUwmExweHuKJJ76Erc0tvOxlL8OpU6eS+qhuamhtMPQ9jDHY3NzEcrmktUXF2GQygdIKbnD4q/+Pv4ovf/nLENyZkHoo9VDqodRDgUAgEAgEAsHdC2kufo1g0gMQMUvMJoJIqCL7H4g4KmRSxSS5PB7DOY+ua+PnKIUTIOLYDi2MNWmu2I3Xp+LweJXIIZN3HeeCAQCHEzB8MaeKbYHWWjRNjaqqk1IHSiUiWn5f+Wxr887DY5wIO0QSqjXNO6tqIl/Ou0imC+IcCbgrmgDeBzhNlkjeo4ODQxwc7OPUqVOo6wZ936Hv+3QPymAIpYiU0qw2O9pHSqG1MMYmBY8xZqTKYXUQWSJDnO2l83ywmPDK++dcJNoOGLyLlsHcZKF7y1Y+gJNn6fvZEqq1hg8ebdcetZ+qUYdm9BwYYzCbTXHq1Gncc/YeTGczXLz4ZfQd7U/bdmT3a2p8/eu+Hk3T4JFHHsHFixfxmte8Bg8++CCqqqLE00iktVaYTCaoqgpVXWE228CkoRRaelYaKB1tgccEZQjuPEg9lHoo9VDqoUAgEAgEAoHgqEZSoAAAVMxJREFU7oM0F78GhBDQdR2m09n4jULhUs6ZIoUGWbd0MOD5ZPl4yDOoFKWDeuej4kTn2VSa1B1uoMHxPNfLGLqdTDqNNomwjZemoaJKZh2J3CHO0IozuKy1mDQNJpMJbFXRfKu4FrbiFZeNglfTmryD9y7NPPPexeuJJL1u0jlTEEGhiiF7mqf5Yt7DaNoXAMlidnh4iKqq0TQNQqDXl8tlIqLlPhujUTc16qpGVVeoqopmYllKaDXGHLFYDnHGGYdMUPgCkVlrPdzgUqCBM4Vqx/movDKpuZHGpcVngJsapRoo/aw4eZXusVJA3w/o+x5d26X7OOrjpEF3pLCq6wbb2zu4994LOHfuHPq+x9WrVzHbmME7j/2DfQQEvPjFL8br3/B6PPzww3j44Yexv7+Pey/ci7P3nKWk3ZRiS4TZGkPPQwzO2NzagtZkv6TU1Arf//3/Na5evXLkWRPcWZB6KPVQ6qHUQ4FAIBAIBALB3QlpLt4EGJPnTh0LlQfV5/ACIIQ8l4vtfuUcLv6u1gZakbpnGIZIJg188FhE69ZkOsXWVkBTN0TmIltzzqWEViDP/yrnXpWBBSn1ky1oilQrdV1jMp2irqs0aytff76GEiEEeEeqkq7r0LVdtKllYtn3PZqmwcbmBgUtDI5SP8MAwMENLhFsrTWCpiaAU2TpA4Cu69D3A+q6ws7ODoyxWC2XWCyWWK1WqOuKFENp5plCZSs0TYO6rqOl0aKuK9R1M1JfMbTRaGy2PQafbYCDG2L6LKl63JCJNAB4HdJcObJ0hkSYBzduPCQrYDx9XrNJKiFOXTWGngEdaEYaq5DoOETUSXVUYdI02NraxJnTZ6CNxqXHL+HKU1cwmUzgrYe/7vHAAw/gDW94A7zz+NznPodr167jxQ88gHPnzwNAUj15R2qd6XSSGi11U8M5lwg2hRcYKJ2bLIK7A1IPpR5KPZR6KBAIBAKBQCC4uyDNxa8RIQQcHh5iNiO1jirmjJXz2lmxU6IksESoDJQie6D3pE6pbLTwKYN+oGH8IYSkUvE+oHct+qHH0PekpLE2zRIbnBtZ6o6S9xxeMFoTW/9Mthpaa4lMqlJJootUWFKd8MB8H8im1vcD2rZD13cIUXHivYsEmeZRTSdTIs1GYxgGODfQ56M9UWtNSZvGH5ndNcTggt3dXVRVhb7vsGpb9H0X1xTgBiLmAEiV09SJSDd1A210sgRykICPSikTSbS1du3+eminYZxJCaHeeviKwi042ZRmg6m4F33a6zJNd/RMsS0wqm1IGRSSGsr5uC5jk/WUZp7FmXJaw4OUTk3dYDKdYjabwlqL69ev49FHH8Xnv/B5KKUwHA5YrpZ4+Stejje+8Ztw/vw5/Ptf//d45JFH8NKXvgTf/n/+dpw6eRJd2+X1aSLUwzBgY7YR962C0Qar1QpGa+zs7qCuarzlrW/F/v7es/iNEtzOkHoo9VDqodRDgUAgEAgEAsHdB2ku3iQQwQrJNpdSKlm9ElmTj3OpeHA/qy/yccii50FzqJx3ach+P/RErkDJmC64RHq99zicz7FqW0ynE9Q1BSvQ7Cuf/g1w0ML62jM45dMYkxJVR/PHnE+k0vshXSPb9ULghFCXZ2vFddJnaN3DQOqara2tUXOgVPQwOQ+GyKL3ReCCD0mFVFU1tNHpfEoBTcPWQgenqNFhlEVVRftfRWS6qqti3lhU9TgPo0x6HQqwxqZ1strIGksWR+fTdfFMN2MMpdJqBe8tnBvSfg/DQOQ+jEnwjZ4tfqZCoECKoEO8bktzwqKHNEClBkldN5g0Dawha+PVq1fx5S9fBABMJhN0bYfFcoETJ07goYcewoXzF/Doo4/iP/6n/4jd3V28+Y+9GS97+cuT6oiuI8AaQ6qtqo7psYZCMaLdsq6r1NDhaxbcXZB6KPVQ6qHUQ4FAIBAIBALB3QNpLt4EhBAwn8+xsbFB1rDg4uvxAwVZ1UqPZpAhfYRmSQFAAIUChEDqDlaiuCESMxdG1r4chAAMw4D5fI62bZNdjy1neaC+SsohtvrlayEFRvAhDsxnZY/H0PcIHGyQFCRIaauc1MnBAUyIU3Kq1+lYPBeN7ICkKiIS79I6y3lbZDF0CCFfM+8np7WGENDHvWIVkIKKhNak0IW6JqVO09QxGdWiikokCmVAWh+vkcMfyvli6Z56DW8CTLT4sSWQlTkcfjD0NA+u73soqCOBEWNrZlRXQdH90lGdFJsrrK4yxsIZF+2DWUFU1zUmk0kitk3TwDmHw8NDbGzMsLW1hUuLJ/Hggw/i277t27C7u4vPfu6z+OhHPgpjDN7whjfgZS9/OawxaNsV7QtIuTWdTbG1tZXSWPk5NFpjc3MTk8kEAPDd3/3dWCwWENxdkHoo9VDqodRDgUAgEAgEAsHdBWku3iSEELCYLzCZTgFwair9Y2QLjIoKXxLpIjSgfE1pDe8d2q7FMAwYouqDlCNZ+cNkOilknEcf+jhgXyeynDl8SCSb1B+ZILs4bB+RyIbg4YNCGEKy0RlrMhmOQ+2BOM/Me/Rdj7ZrMyGvdFITUSJqXpMxGovFHG27imQ6ZKviEFVMKhNYpXLSqdGcXEqPcd+TQif4EIk9KYroHiDOEiPbIf2pk/UvEf6oxOGmh1LI72mT9jAuMamHiFzb9KLzpBiqbJXmvHVdh2qo0HdE+Llp4YOHH3xSu7A1UCmMbJesWmIy7QsVFt9/Ciuosb29ja2tbcxm00RuDw8PUdcVnPPYu76He++9gIe+5SG86pWvwoc+9CF85KMfxZe//AS+/dv/C7zxjW+ENQartqX7bCgEoalrTGcz1DFwQseZb9PiuUcI+FP/lz+Fy5cvP9NfH8EdBqmHUg+lHko9FAgEAoFAIBDcPZDm4k1EVphkax2rTZJKBpySmsGvMYEq504RcUQih95r2FpTQqXPc7acG8gipkl5o5ROqpJsNYxEkdUmhSLGhzwbzFZVIo1uiOqTuD6tFYwfK4D4GtgC2HVdUseQEobWSsfysNaSJTEO8l8ul0ldUlrl2O52HLTW0YpIChY3OPjQp3UaQ6SyshUWywWRaFvBWFLvkFIJsJFc8/nYosk/QykYTSSb3hunwAbviXAX5JpuWEzFtdmuWNkKwzCgqzoEULIugGQj9PCJRadzIERlU7720lrHzxITaWstNjc3cWL3BLZ3tmGtxXK5xMH+AQ4OD8h6WWns7u7ida99He6//348+uij+PjvfRxPPPElnD59Gn/4D38T7rvvPoRAlj9Ea18dk2frukrPUFPXqJsas+ksWVy10XjiiSeOvW+CuwdSD6UeSj2UeigQCAQCgUAguDsgzcWbjNVqhdmMVQukKCE1DUa2vzIwgP/wvDAFBehIUAeyfvH7ITh4Zygtk4l5YQNkMpPsfYU9jM5Hn+GB+2m2l/PwwaOum5S0GQKSioSPHQLQdz3NlooWQV4HhyI4R+u11sJYA++JYA/OJaIPRFLtHSlXlELfD4kUAnEmGMyxtkD+GSDFCyuWnHNomgZKURCC0QbT6RQKKqp66A9ik0Ipug4m0OV1lUTaGAsT1UlJTeMDgtYjlZUbHJz3kWzG48S/vHUwA6mLAFYO2fQsuBgoUYYXAKzCKZ+deK+NBnrEMAO6rrqusbO9g82tLRhjcPXqNVy58lTaCwCYzqZ4/Rtej5e/4hVYtS0++BsfxJOXnsR9992Hb/zGN+L8ufMIIWC5XGEYBlhD9sc6znEzhgIdAErGnUwm6V5Za/HOd77zSMNIcHdC6qHUQ6mHUg8FAoFAIBAIBHc+pLl4k+G9w2q1SrYo78MREl0S6XLmmEJWaTB5dorIrNKZWDnvoFy2hvkYw0rWwTw3TCmaV1Zw6TTXipdEBJiUH3XdEDFSAHqyADJJpi97wJHdz5g854rPz2TW+4CqMmgast2tVissF8sYxqDg3JBSV9uuhfOOkk9TiivS/kAjzSbjaytnqrH1ESFEVUtIc8RWq1VSALGyiEMKrDFREUX3g0k6k2VKpNUpNZWtiWkflYayuVmS99dAuwHe6dHrAOAdrYGP1dQNqqpK9yX4kOyTDBXnzVEKrUvEPSmpnIOK1lBrLaaTKTY2N1BZi2tXr+GpK1fg3ACtqblgrcXO9g5e9KIHoBTwhS98Hg8//DB2tnfwbd/2bXjta1+LuqnR95RQ671D0DqRdaVoZhuAUegDXZ/Hf/vX/lv8zu/8DgQCQOqh1EOph1IPBQKBQCAQCAR3A6S5+BxgGIbR3DDKrFQFEVJZdVKqPhCgAqKKJCersvKGlDyRUDsOScjEEirPM8tWxHxOBiW3IqljjNGYTCaYTqYw1pBdzwTQSDEPgBQ4HB7gPc0kc4osguPZWHRsSsms0350fUeNA03XapsqkrVIOEOAdzSviu2BSQFTWirjPvC1q0h8Q4hLTVuhIvGL6aqqDIvIs8R4FpgFqYq00qTI0fR3VvfYqIgKI8INADroJKwJgQb5e+tTYyHEa9S2vI80l6xu6mS34+vqhxzCQNdHCbWuaDLQvDZ6BjigoGkm2NzchNYah/NDXLt+DV3XoqporlhAwGxK4QVD3+MPLl3EJz/5SRht8NKXvRTf+Ie+ETs7O0TQawor6HqyKtYV2f040ILVQ8aaNEcNBvh3v/7vnv6XQ3DXQeqh1EOphwKBQCAQCAQCwZ0NaS4+R1i1LQUNFCoQRkmkgUhMCitZiMPofbSTGVONFTPIRLIkzOn4x1jlShJavl7XFeq6wXQ6SYqLEAIwkOqI55cNA6l5KD1UkU1wiPY152GMBqDBwQJVVaX0TEo+DVAmAEHFMACFtiUyrRQHKQRor+FQ2h4zSaXrD7FRkK9Xa0NzxuBgoNM1aE3rCQjJtkbv+WSdCz4k26I2MbTgGCLNzQc+zljZs3bzIzEeFNkTOT1WaQ3tKE1U97oUcKW9V0phtVrFoIdI1hWthdk0Px9t28F7j6qqsTHbwMbmBmazGZbLFQ4O9tG2HV1jPNHW5hbOnD2DpmnwxS9+EV945At49NFH8ZKXvgSvfvVrcPrMGbTtKp1TV2QHLW2dVVWNbIDWWiit4foO7/lf3gOB4DhIPZR6KPVQIBAIBAKBQCC4cyHNxecIXdcixKH1KipJVCgUM/x3SXAxZomk0jCJTCo1tpUxofM4fgYXv1b+mxUwLs772t7eQl03ibgBZOfqhx5t2yYljg++UI3oZP/j4fx0fJqlVtVVIoec6ppSWkOA1jYS7Hw9TKp98FB+fB0B471ia+NR9VFOjuXGA6taAop5asm2SDPUnHdQXsE4UqGwUkdrItKkBAopqZTtiEyS1y2C63tPs9XibDdHzQgiuUjn47RRpem9ru2ogREVUAqkDBp6ag4MfQ/nBtR1jel0gs2tLWxvb2EymWCxWGC1WtFex7Vtbsxw//334/SZM+i6Fl945At45JFH0Pc9XvPq1+AlL3kQzg3ouz7ZOcn+SMmxVV3FvTDJcmi0QV3VWC4W+Gf/7J/h5/6Xnzt2HwQCqYdSD6UeCgQCgUAgEAgEdy6kufgcou97VBVZ4VjpEcLahxTN8ip+jKAZYZnkBnC6KZNrerlQoajxPLHRadhWpxRcPyCEgMmkxmQyGSWt8sysvu/R9308RYAxFtaYpF5hstq5NipqPIaBFDtVVSVrGCeAsmLER5vg4Fy21KVrzzZDngkWEI7YAlVggs12QJcUQmwjZIthsiy6qJiJT7wPPibKEmlOs8JCnuHF+6+VTmRdGx2TVRWUiWqdmJjKqiKeHcdBBTYpdWg+FzUK6mTjmzQNJnGvWeW0v7eP+WKe7IJ8bwbn4KN1saoqbG1uYWubiPTW1hbqqsZ0OksJtM55TKcTnL9wHi9+8EHMZlM8/vjjuHz5Mvqux7333oeXvvSl2N7ewWq1whDXNDiHyWSCSZyVt7Gxiaau4XxUOdVMsA2uXL2Cv/f3/97xD55AECH1UOohIPVQIBAIBAKBQCC4EyHNxecYzg3ZRgaMSBeQSTETy0zIMrkMBcEbIYQjBFqtfSgdN9rsnCP1x2y6gc3NzawyUQoaGh6syDHpPa01KlvBWAOlyMLnPRHoct4XDconguWDx2q1Qtd1kZSaZCX03tMcs/hdrfXIppeuO645zRKLpJZIb0xrLcVN0RYZEKK9zyMEn1JBh2FIQRB0bxwqW6Gqok3SB7K9sZLHkPXQVCrNL+MZW6yeKfebFD7ZKqjZkmiyaoifBb4W3oOJn8Tjq0iEHbq+i4qqkIMwonJKG43pdIrpbIqNjU0YbVLzgxJq6dqV0tjZ2cWDDz6Is2fP4ODgAJcvX0bXdbj3vnvxhje8ASdOnqQ5aPFZMtpgY0azyGxVoW1XqKoYVND3SamjlMJytcTHPvYxCATPBFIPpR5KPRQIBAKBQCAQCO48SHPxOUbbtqhrwBozYtDrqaXHIRPVY4g0cKwyp0xeLV8rye/Ozg42ZhujxFU+noZGVdWYTj2tGUgJoeNQBcB5wBiblEbGWNRVnZI4M5nWmQh6UgH5OFCf11qqXZiEZnthEcyA3Bjw8NDQeS8DhSBAxXCBeL5hGKCNxhA4QdWnxFNeW6lQMtoAVYUqNgiMLu1/SGvS2pAlkuejFeqcMliinBV23D1jxc5G2IifJ2I8ny+wWi7TfroYFOG8Q9NM0DQNvPeYzw8xnx+m4+zt7WF+OIc2GhsbGzh79gwunL8ApRQuXbqERx99FCdPnsQ3f/M345WvfGVKkqVkXAVbVdjY3IAt5p55R3ujjUYzaWCrCsvFAr/9O7+Nn/iJnzju8RUIjkDqodRDqYcCgUAgEAgEAsGdB2kuPg/ouhaqaZJihK1tCGqk1uAUVWA0emyE9ddvRLTLgAOyg5FtbHtrG9PZNJHgcs4XgERE67pOAQA8xN+5IQ2zV1CwsPH4ROK01mjqBkpptKsluraDcx7WxOsEEUxWhbC9T0f1Tzw9825SwiCx13TxpNAp9izQ+z7EoAQF8KMdQk5A1dpgcPTvuq7RNBOyDHoH7XS6DibuvaYU1RCApq5hVJ4lxkRaaU0rLG4Eq3nyZ/OsOKMNtNHUTIj3iBVMAOAGUgrNNmaYTidYLZdo2w5d18I7n6yE1lp4H7BarTCfL9LDQLPhOiitsLm5iQsX7sW9994HpRQuXryIxx57DFprvOIVr8A3fMM3oGkaLJcr9EMPM1DSaV0Tie66Nu13P/QYhgHNpIFWGm27wm9++Dfx4z/+48c/qALBDSD1UOqh1EOBQCAQCAQCgeDOgv7KH8l497vfjW/6pm/C1tYWzp49iz/5J/8kPv3pT48+s1qt8M53vhOnTp3C5uYm3vGOd+DSpUujzzz22GP4ru/6LsxmM5w9exZ/5a/8lRGhuBPRti0G52LaJM1+onlRYysdCoUHkC2BCcw0+c8NEAJZ2NxA4QE72zs4deoUprMptNZpBld5Dv6jtSFrXrTnqagu0Vqn1FNbVajqCpubm9jZ2Yl/djGdTeG9Q9u18IFmjgXQOvh8PMtLKZ1CBdjuVypclCJbXVLIRLWMNpoSOjWFIrDtzzueG5ZnuhlD10GBCY6G9Pc9+n7A0Pfo+g59R0RxGOi9ru2wWq3Qdh3atiXFUd+NEmK1zuohrXPiK+0XWQKtjXPZjEFVkYrJxv2rmxrNpKG/G/q7rmpUFZHZpm5QVRUCApbLBdq2hdIKVVXhxIkT2NiYQSmgbTv0fQfnBlhr0TQNNjZm2N7expkzZ/DiBx7A1tYmHvviY/i93/s9XLlyBa99zWvx2te+FhsbG1gulhSEUMXz13Vsbpg0e42e05D2fdW2+I0P/sZdT6SlHn71kHoo9VDq4Z0DqYUCgUBAkHooEAjuZjyr5uIHP/hBvPOd78RHPvIRvP/970ff93jzm9+M+XyePvOX//Jfxq/8yq/gn/7Tf4oPfvCDeOKJJ/D2t789ve+cw3d913eh6zr85m/+Jn7hF34BP//zP4+//tf/+s27qlsUXdei7/pIZF205eX3R/PBCmLNn0l/I7HN8qMjeO8w9D2MNdjd3cXW9hYlhwKRwN94nSba/pq6SbY2Si8lMlXZClqrNGyfrH+Umqm1Rtf1KWUzpagin3BEQnU+PqetJkWOVpnUa51/VvGPzgofVgLxhoSQv6/izC9Sx9DfTKTbtsViuUDXduAk2K7vsFwusVouo5WRCPViuUDf9SlMoLwGay3NYTNmdH3W2rRHVV3DGiKppIhp0DQNptMp7anRoxloACl3FBSapiESXtXY3t7Gzs4OJpNJ2itjbEzhpSZFZWO4wdYWDg8P8ZnPfAYXL13C7u4u3vAN34Dz585jMV9gGGgG3tb2Fja3ttDUNamzYjOgrijkYmM2w872Doyx+LV/9a/w13/izv99/UqQevi1Qerh+DqlHko9vF0htVAgEAgIUg8FAsHdDBXC09Gqp8fly5dx9uxZfPCDH8S3f/u3Y29vD2fOnMF73/te/Kk/9acAAJ/61Kfwqle9Ch/+8IfxLd/yLfi1X/s1fPd3fzeeeOIJ3HPPPQCAn/3Zn8Vf/at/FZcvX0Zd11/xvPv7+9jZ2cEmdrJN7DYBqV50VMWo9HOp/Fi/ptIeWBxo/HNhJQwhUErmbIrJZJJINCtYtNZJNRMKXs6JpGlGF0jtM/Q9BucScTQmf59UP6SCqesGwzDg2rWrWCwW6fgheBhj4KIFj8mtiUmjQJ4nVs7h0roINSjtgGvX7LzLL/mQ9kutkelxCq2GNZSmaq2lMIDpFE1DBDcEoKosrLGoonJGG43phPbUWCLQdd1khU68h/wrVVVErrUihVOpSuK0WU5bbbsOh4eHuH7tGq7v7WFv7zouX76MixcvYrlYwlbVyEbIabCDc+j7LqXwDsMAawzOnD2Dc/ecw2QywVNXnsKlS5cwm87w6te8Gn/4m/4wprNZ+mzd1NjY2IBWGs57KEVr5+doMpmgbmpYa7GYL/Arv/or+Jmf+RncbggIOMQe9vb2sL29fdOPL/Xw2UPqodRDqYfPP+70Wvgd+BOwqrrp1yUQCO48DKHHB/DLUg8FAsFdjZtVC5+VcnEde3t7AICTJ08CAD72sY+h73t853d+Z/rMK1/5SrzoRS/Chz/8YQDAhz/8Ybzuda9LxRIA3vKWt2B/fx+///u/f+x52rbF/v7+6M/tClKv+GxbGxy8d+nfznk475LqxMc5YD74PKcLIc4Hc+mP8w790MM5j6ZpsL2zjc3NzRg8gBQ8AIyVOmz9MzoO81cqWeW8D6TaqaqUjmnj0H9jbEz6pMYAJ4L2fRe/y2Q7q1WU0ih72XytpfKlRJ63FlU4BZFmmx8UKYaMsYloAkizxPq+B6WEZqshp3s679C2LZaLJfb39nH16lXs7e1huVjGwIC4r12frHBt12G+WGCxWKJdRYtg26HtWvT9AO8yYS+vieef0dqpWZBCFDQ3U2jdy8UC8/k8zUILCFgs5gg+QGmFEydOwGiD5WqJrS2yYXLYRNPU2NzcxNbWFnzw+NITX8JTTz2F06dP4/VveD1e+tKXwVhDDZLYJFGgOWeL5QJDDJeg9FoKrwgIySZ5OxLp5wtSD589pB5KPZR6eOdBaqFAIBAQpB4KBIK7CV91oIv3Hn/pL/0lfOu3fite+9rXAgAuXryIuq6xu7s7+uw999yDixcvps+UxZLf5/eOw7vf/W68613v+mqXessh0HT+lPTpnYdXYSSb4Xn9x4UTJIUPaGi/1gaTyQTT6SQFDxhtstXvGKLKKhEmeBxOMAxDSi5VUDDWJCURQHYzVhXlpFIN5x26rkW7ajEMA3g4vzGlPbDjywMnboaQbY1BAdCeSHRQ8IrVRJGEFlbC0SVFZVEIgPYaQYfR91xBcLU2yVLXdT3adoXgI1l0RJz7vkddNzhxYhcAMGCAiWQVISDExFXnBjTOkRrHkBpHNQ2stlGxE9fJ9zD+QOEHlJYbPKWyekcNlqHv0Q+0huADNjc3UdkKe/t7sNZiZ3sH586dg7UWjz76GJHtENC1XVLd1E2NrqNZaYvFErPZDA+86AG87GUvw8bGBrwP8GFIaac+eLRtS+qfhkqCsQbTyTQGNCgYY/FzP/dz+Ef/6B/d6LG+qyH18KuH1EOph1IP7xxILRQIBAKC1EOBQHC34atuLr7zne/EJz7xCXzoQx+6mes5Fj/2Yz+GH/3RH00/7+/v4/7773/Oz/tcIoQApTUqY5MFjlmX9yEpa4h/qZFNkNUmTdPkQIGo+gCyagaI9j5jip+jkqXv0a/6mMbp0kB/593ILsi2QCZqk8kk2duYUCulYSzge58sebxWYyzqus7KnajyKfchqYhCgPeAiWmqnHTKzYMRjusypLei1ZJJevBwAw3gz4ECVVyLTnsQAs1mOzg4QF23qCqyB/K562FAVVdRdaSiEmggu6C1o3tmjUmzugLbJ2PaKimLDF1vIFXM4EixZK3FbLZBc8XqBs47NGca3BvuhYJC13d47LHHsFgsMJtN02sAUNVVegZWyyUAYHNjA+fPn8d9992HE7snACClwG7vbJOl0VgMfY+qqTCdTqGVTqScElk1/s7f+dt43/ved8M9v9sh9fBrg9TDvA9SD6Ue3s6QWigQCAQEqYcCgeBuw1fVXPzhH/5h/Oqv/ip+4zd+A/fdd196/dy5c+i6DtevXx/9LzKXLl3CuXPn0md+67d+a3Q8Tsjiz6yD5z7dafDewSmgqeO1RXnOcfO2mDAzsWNrHc/hIlUIqWOY/PLxKJE12+K89+hi8ucwDEmtU84VG/kEI6y1NG+qpoH8k+kUk6aJFjRDig6dB/hnMqwRIsnmFFHvA7TSCCrA4/hABZoJpo/nzWsypnQuraC9hkcxdyw2LnRUx6RZZ1CobIXKVqmh4D2975zHcrFE8AHGGviozqmHGq5xqZFAiaEWTV1jGAa0XYum7lA3dWo4BISY5ppnnrhInn20QnZdCzc4aKMxm01hNM372tu7jmEY0jOyv7ePJ598EtZanDt3DnVD57VVVVgz6ZzWWpw5exYveelLcObsGWoAxPNWtsJsOkNV0bVzOqvWRPK7jmyg1lr8zf/+3fgP/+E3j7VqCqQe3ixIPZR6KPXw9obUQoFAICBIPRQIBHcjnlVzMYSAH/mRH8G/+Bf/Ah/4wAfw4IMPjt7/xm/8RlRVhX/37/4d3vGOdwAAPv3pT+Oxxx7DQw89BAB46KGH8Df+xt/Ak08+ibNnzwIA3v/+92N7exuvfvWrb8Y13VZwzqHrO2xubiaCBhApZKtYObg+hRyoTDJ5ZhUP0QfIWtd3ZAPrujaRtr7riRRG8lQzkQ95pplzbpygGvJcs1VMDQVAs6y2tyjds6rSugGyAhjNVris4NGRbGsdTxkUlAq04oCo1vGFMsfDO5WI81HBjhpdN3+IGgP03uAcJpMJJk1uWrDFUWtNvNzFffMGgyayvVwtMTiHpqlTcivfL61Naj4ASEmiVVWhbmrMpjMYY5KSp+97TCYTNHWDQQ90ZdFu6X1A3/doW7JRzg/naNsWq+WSUlpXq0j0HZbLJabTGabTCYwxZEdM6qjcEGmahiyD99yDE7snEmkGAGPINqo0WQBp/2j+2zAM+VnwDj/1Uz+Fj3/8d+GLkAgBQerhzYfUQ6mHUg9vP0gtFAgEAoLUQ4FAcDfjWTUX3/nOd+K9730vfvmXfxlbW1tp7sPOzg6m0yl2dnbwgz/4g/jRH/1RnDx5Etvb2/iRH/kRPPTQQ/iWb/kWAMCb3/xmvPrVr8af/bN/Fn/zb/5NXLx4ET/+4z+Od77znXft/+IyDAMW8wVOnDwxIodZhVMOvs8kOtn1CjLpnEukjGdV9X0PAKjrOpKosa2uJPEBAVobjJJEiwRTJuXDMODg8AD90GN3dxfb29vxuzpaA1VWCg0u2QBVnK2mlEJQIYYKpCFcKWwhvYSxFbAUiygoBAUoQ/PQmIzzNQGUnNo0NNukqiy6roMbHIw1qKo67SUpoWhIP8hVB+cGtO0qJ88GQHVIiasmppRy0APPU5tMJlitVqirGsvVEnVF6aN912NZrRJhLe/DYr7A4eEh9g/2sZgvsGpXRKhXNAPNWou+pwZHVZEaZ7lcRqWVT/Y/pRSqqsapU6dx4cJ5nDl7BtpoDMOQnpOySeIdPURhGICqQm0NAJoN9653vQuf+tSnaL2CI5B6+NxA6qHUQ6mHtxekFgoEAgFB6qFAILib8ayai5yK+B3f8R2j19/znvfgB37gBwAAf/fv/l1orfGOd7wDbdviLW95C376p386fdYYg1/91V/FD/3QD+Ghhx7CxsYGvv/7vx8/+ZM/+bVdyW2Oru9w7dp1nI12rXXk+WJkjeM0TYBI7tCSxa/veqzaFl3XQimFuq4pQCASqaz0UWl+WDwwvQ4FYzTWg8RTWIENkUD6pCpJyg8ohECzvCpQGiqAOL8rRFVMthdqxUqeOGdMAUGHpIphhEDLCQWpjYeg6wkBXvs8Ew1Itj3vPTZmG9jc2IDSCl3XoR/6ZLWsKtqfuq4QQkDf9Tj0h0nVEnxA27Vw3qGqenAoAxH7kK6LQx6MMejajtRQVQ2zJIK9Wq0S4V0sF+l9nlXWdi36ro8pqz36fkihEqA7Tk0Ak2e0MdGmxgDZGbd3trG9tY2z95zF2bP3YGNjk+7B4DBoBwuafea8x+DcyHbKzwmUwk/+5E/hU5/6VGrECI5C6uFzB6mHUg+lHt4+kFooEAgEBKmHAoHgboYKt+HgoP39fezs7GATO4k83QlQSqFpGpw5c+ZpP0fkjeZ0dV2LriWCWKaSlja30fciKQZwLJk+ZlHHvhyCR9t2WK1WmE4nOHnyJIwxWCzIthaCT8fXhYrIOU/vRfLJRDolpEblC3+f7Idh7dwhK3/iOZTKya1sZyOCrnD61CmcOXsGwQccHB6g67pkqwwhoKoqbG5uQmuDrmsxn8/RrtrUlCDrYr5HdVWT4iWum69FK51SUqfTGeq6GgVLOOfQ9z3m8zm6luaQaW2SEqaqKEWWmwnlrLm2I9VOUzeo6gp9R2ETdV2jrms0dYOtrS1cuPcCdnd3sbOzg43ZBs2li5ZSDl/Q2qQGAv/+VFWFyXQKrRX+h//+f8DHf+/jd5xCJyDgEHvY29vD9vb2C72cmwaph1IPpR5KPXw2uNNr4XfgT8Cq6it/QSAQ3PUYQo8P4JelHgoEgrsaN6sWftVp0YKbjxACVqsVrjx1BadOn0qvl9Y/IM4l63r0fZdsfsHH9E9raQg9kBJVk42waDyEQMoYBCKDaTZZYaVjkpq/U3JrShrlVNB21aJuagAxGRTRIud8TDyltSk4DC6rbEK0w7HvTyFAGyBA56aB82RBjCQ7hJCvReV1h0DzyOizRNjL5NLBOWitMZ1OUVc1nCfLJJF8B+88VqsV2rbD4BwdzwcALquJgJgYqtJ+sHUTQFQDUWOjqmpUFYUKdF2Ltm1TcIR3HsaaRLStpSRWY0y6TsYQBrRtS5+riEh3fRdnxEUivb2FkydP4uyZs9ja3qKZb3F/KQXWwBibn49oz7TWUgOgrlBVFn/7b/9t/O7HfzcphASCFwpSD6UeSj0UCAQCgUAgEAhuD0hz8RbEYrmAuqpw4sQJcDgB2/f6vsdyscSqbeHcAK3JFqZsDjlYR2krLJUvKU01UHpo/kQMRFibcQaE4phEHIk4BxrqH5NPrbUpCMF5DkOwSTWimTgrDQovQFLYoCD8HBLgdYBzA7wfsu3O6FGzgK+FUleJoPrCSjgMQ7IF1nWNuqnjPK4qKYiGvsdyuUTbkr2OzjXeOwBEykNuUmitETSti5oHPu1bCB5aD1itVjg8PEzqF6UUfE9qJLbf8ffcQAmmrEhyg0NAwHQyRQhk31NKwVYVmrrBdDbF6dNncPrUKWzvbKNpmpHdkwMKtFZxvhsQWAWFAKOJaP+Df/AP8NGPfDQphASCWwFSDxFfk3oo9VAgEAgEAoFAILg1Ic3FWxTz+Rxaa+zu7kYyGtB1fVJ8UPqoSQR1fS4Zk8tEmJHJIKMMLvA+FCmopO4IqvhuCKPvAUQijTXo+x7O5TliSitor+FBdjYmsQCphbznUAIkhUupCor8PquCokpGKaRZWsaapGAqLXO8jmFQScG0iiqX2CaAM6TMaZoG08kUAHB4eIj5/BB91yf7IM9MWw9NYBhtRs0Jmqnm4ZyHMQ4h2GTN5D3gY9N9oYvVhuZ60WeRZovxPYFSmEwmaa8BoLIVJpMJtne2sbGxidOnTuHUqVNomial6pYLN1rTPUW+nvKe/qP3vhf/9t/+2yOWS4HgVoDUQ6mHUg8FAoFAIBAIBIJbF9JcvIVxcHCQZlQNw5BJtLGoa5tsciVKMrROgJkk+0CkrkxgRQhwzsO5IVkKK1SjGWUllKLvrq/hyMw3NQ4UUCCbnlb6yFiz0lLHtj8fk1gBYDKZYDKZRCtbtN0pHe16dE390MfrcKnZ0HU0K4yTS51zaFctjDHY3NxEVVUUUuAcDfUH2RadQw4QUPlaeR+ttQggNY2OShjvcnPCe5/shQEhJ5hGlYxCDF9whSon3gM+r7UWs+kGZrPpyD5YNzW2tjZx6tRpbG5uYmdnB1tbW0TcnUuNiBRcoUtbZiB7YrSLvu9978P//iv/+7H3WSC4VSD1UOqh1EOBQCAQCAQCgeDWhDQXb3Fcu3YNdd0k1QUPnj8OWTFyVF2Tgg20QhhIOQLFSahE8AbnMPSUHGqDjYP7mbRmQl4em0MRlNKJnIUQVUHx+M5pBE9kPQlItIf3LquJoMCpr3x8HxNKK1uhbmpsbGygaSbpmviz3vmYwAqY3qDrOiwWAdZaNJOGElG7HlqbOGvLwHmXZoE1zSQljvqY5Kp8XofRhkIAjE3KIGarCiqlo+b9J7XNapVDGLTWOX0UYzvm4BwwhGSZdM4DgRoaxpiU7DoMA1Qk8VtbW9jdPYHTp05htjHD5uYmrLVYLpfo+g6c4trUpNxRCnA+p6DytXzkIx/BL/2TX3rGz6NA8EJC6qHUQ6mHAoFAIBAIBALBrQdpLt4G4ITOuq5QRSL9TOxao9liIURyrOAK9YtzKqljnKdZVnngvV53k43Ar6s4V8xWFUIgVYobsjXOGIshDAjBA0El2yIHAayrjXzw6PsBxmhMpzNsbmxgMpmgqqsRoed/sxqGXqMZXZxUSlZBFcmwjgmqBs7RzLHlcgmtNPq+i+oaIvEcVsBWR62LZoGODYVIiGlvy9wHTq7tAJBdEBYwxqRUVW4sQBGBZzUPKbFo/4kQ0752XUtE2hhUVY1Tp07hxIldbO9sYzqZwliDru+SlVDH+WkVJ7RqQ2sFXd/Dn/wk+qHHe97znq/4HAkEtxKkHko9lHooEAgEAoFAIBDcWpDm4m0C7x1WKwetDaqKbtto9pW6wReRSbVSZE8bhiErW2IKqQsuzvEiJUhVEQkrwwH4WJn80nvGWDQNJYHSbCwPp3xaoFKkquHATZ6BFYIHPKCMiooXOq5zDn3fYWfnNE6fPp1UKN75tHZGmqVmYnJp8OhsB6WJrDuXAwPYYmeNScS473ochjnm8wVWqxXZIKOSiNUuHBLBm66UHquEAgUgrKfRhpg0G3SIyiOaK8az3tjKyOmvvDdVXcEYC+8dBUTE95qG0lA3NjZx4sQJ7OzsYnNzk66j71NDgZsP1to8u01R2ETbtvjUpz+Nn/u5//lpnjaB4NaG1EOph1IPBQKBQCAQCASCWwfSXLzNsFjMMZttUHBB5LQhz7lfI7tjhBBVIXEmFoBo4YvEED4fDKWtMB+DZ1bRWzQbyxqDpmlQVVUk5CqRRh9J+2gOWZxtxmqeEAI8PFRQNP+r7TCZTLCzs4OdnR1szDYAgJJLu3ZE7kmVoqFMnmVW2QpGGwx+KK49oO879P1ACpaqIuseAlarFZZLItNsTdS6JgLvo+pHk/rGBx/TRElJ43VA6El5w+og76ONTyt4j0i+TZEASzPMrLUpCRVAapRYa2nOm9KkTgKR+slkQvPEtnfiTLFNaG3QdW2aG1ZVFWYbM9R1k9bM97PtOnzy4U+KOkdwx0DqodRDqYcCgUAgEAgEAsELD2ku3oZgQl1VNpLK8WwxBocF+Ghp865HQEBd11BKpeROnnvlVYBzA7qe7GshhBQWMJ5XhoJwIyl7tNJRGaJGKhQe1s/hBsEHBJXndNHPgFb0uel0EpUoRKTrpqZk2L5DG5NO0/G9hzEGVkUSGgDn9+PQfpVSWVmZozUR08pWRHydR9e1WCwWcG7IyaIRTJrLfS732nsHNzg476C1gkYOfLCWfr2YjGttSLkTEH/WGDCAwwtMVBAFT3a9pmkwmUzS9W5vbePM2bPY3dnByRMnMZlM0HdkXzRaYzqdxvtl4v0hdQ6UQt91+MQnPoFf+IWff8bPmUBwO0DqodRDqYcCgUAgEAgEAsELC2ku3qZYLOaYTWewVZUUOgzvfbIIcnIoz70CEGeHaRgT0PdDTtOMX2KroDbZ/jYi6mtDx6q6Ql3XNMi/i3O5YhLo4BwQAmwVyasvCWk+nFLAMAyo6gqnT5/Gzs4utre30TQNnHNYLhaYz+dYrVakUmJiHgkuE8++7zGfz9F3/Wjov7GRTMeZZ3VTwxiD1WqFriUFj1I0K41mk6kYIkAhDmzFSwEEcbZZ13ZpXltpxVNaoa4b2BiWEJIFMKenhjDADfReVZHFj69pMplga3ML1loMw4C6rnH69Gnce+ECtra3MZ1O0Pd0Xlb35ORYAxOTXPueQhp+93d/F7/43l98dg+ZQHCbQOqh1EOphwKBQCAQCAQCwQsHaS7exlgsF8ASmEwmmEwmhXokJPLH6Z7aaDhHwQV9n1NJQ/Cj+Vhs4VsPFlD///buPMbOqv4f+PvZ7521y3RWoJZFSG0BRWgaYiVp08VqQPwDgSgaAwGniSw2pEQBibEGE+NXw1cS/wD/AFQSsJGf+PtWupBKqYI0hSLzpZNiETp0Y5bOXZ7lfL5/nOc5M9OFttNl5j7zfiU3zsy9c+9z5t686Tl+zvnAGqmskbTrpyiIKHieh8APAACe56XXoLcIjprV6+exYCarSZLASrcVAkAYhfADH9OnT0djYyN8X0+kBwYHcOjgIVQqFV0JhMRMogE9ec/O3jp48CBK6Xlhnq+3A7qjGgfESQIRBdt2TCVNNuG3LN3JVP8NsuYNApG0Y2p6Vld27Ul6aFp2lpeIPitMd2F1dVWQ6wAxkCD9m6dbBLMKKiUKfuDD9wOzoOA4LorFOhSKRcRRhKAQoK2tDW3tbZjZMhPFQhGJUqhWq7pxRHrzPb11MU4SRLGeRP/P//8fvLThpbPx8SOaVJiHGvOQeUhEREREdK5xcTEHKpUKRAR1dXX6B+m2MkBvf3M9L63OsZDEeqKVnRXmuR583zJb+7JJnuO4adXHSEWMBcs8t+O4aUMAZSZxKp1Yq0TprXFJYs4TA2CqWDJJnMCy9ZldSdoh1PcD1NfXo1goIo5jDA4O4MCBAxgaGjJnbulzyyTdbqi36RUKBXiei+HhktnGaNs2CsUinPTcL6WU6YgaBHryH4URoig041NKX7vrpY0dkvS8sPTvlzV0EJU2b3BHKnmUSv/e6eQ6W7zIzh4TEdN8QVcC6a1/vudDN5fQFVZ+2tFUKf33mzlzJi666GLTrCBJ31vP9eC4ekujZVnpQoEgiiMoJVi37o949dVXz94Hj2gSYh4yD5mHRERERETnFhcXc6JaraJarcJxHBSLRb3tzNbVNdm2PwBwXAd2bMP1PDTU16eTU4VqpYowChGHCZRKAOizxWzLhhI10jAAeiJpuUCSxFAqQVAI9GRT9PldsCwkSk8kRQkcz4Ft2Wa7nOkeimxSCsRRBMd1UFdXhOfpSeJQ/xD27duHjw99bLYIKlGm+sjzdVWQbg7gIKyGqJTLiOMYQaAn5YVCwUziBYIwDFEul2FZFoqFIoZLw6hWqnqrZNbdFLrCKNtqOFpWzZSIgm3pKiAASNJGBZ7nmsWBKIqhlILv++Ycsez8MEkn6KMn3ipRCAr6ul3HgUoUGhsb0dLSgva2Nli2hdJwCeW4ZKqMsvc0jmOUSsMIggC/+c1v0Nvbe3Y+aEQ1gHnIPGQeEhERERGdO1xczJkkSTA8PAzf9+FY6flgsMx5XhYs2I6NQhCgUCxCRKFcKqNULpnzv7JtZVmFzegmB5IdCJa+lm075mwsc0ZX2j01SSexjuPo88fSqh/btuG5ukNpkigMDx+G7dhoampCY2OjOddsYGAQhw4dQqlUMr8v2RllPuBKdr6anowfrlT01sh0DFk1DjCy9VBvg4xQLpcRRRHCMDKLBZZlme2OWQMEB7apesoqlPSWSb1N0HSZVSPbKLOFBKVGOqmqtLJHIPpcsbSRhG07uglCksCCrjjyfR+O66C+vh6ts1oxa9as9Lw3B57vIwpDJErp1xcbFnTFFQD88r9+iQ/3fngmP1JENYt5yDxkHhIRERERnX1cXMwhEUG1WoVlhWhqajI/V0ohTifAruchjiIMl/Rh/yICJ52wZZNpCyNnbqUHbumJddoYwLYd1NfXwXXckW151sjh/6PP5LJg6c6f6WRSKaVfJ+0IWigUMGPGDDQ0NCBJYgwNVTA4OIBqpaobHCCtZlEKlmWbibRjOxABBgYH8PHHH6eH/QcI0vPJsqqebAukmRgnCpFESJLYLBQgPV/NtvUWwOy6syolve1Qf+84o7q9Qkz3VSVKv64SswUzO8vNXIOMdELNrgUAgkIAx3GQJAr19QG6OrvQ1dWFadOmmYYHjm3DLgTpNkGB6+i/3cM/+hGGhw+bc8+ISGMeMg+JiIiIiOjs4uJijokIhgaHELQEZkKYJLGexKVnUenzyXRlSzaZdhxnZIIJmMmxGtUhtFKtwLYtfa6Zpat2TGfRtBrIsZ2R33Fs+IGvJ7FxZCa2WRdVWPp8rzAMoZTC0OAQKpWKPqNM9KQ+m0w7jr7WICggCHzYtoWPP/4YBw4cQBhG8H3PnOnl2A78wNfVORW9RdB1XV31I+m2Plu/tiWAnVUVWbbZrji6kYPezqcXHrLKI8hIm9es2ml0h9U4neBmDRRcx9ENH9LKIUB3dw2CIN3GWcCsWbNw3nnnoaWlRV9foifjjq2bMWTbAF3XxcMPP4zBocEz/fEhyhXmIfOQiIiIiIjODi4u5pwShQMHDgCWhaamRl3BohJUqtW0EUC6qy89UyvbypZV5hx5zlY2matUq6irK6K+rl6/jhqp1NGH+lvpRFhPzj3Pg51O4j3lIYpixFFkJuFKJRgcHESSJPD9AIODAyiXy+a5kZ77lTVJCIIAdXVFOI4+X2tocAjDw8PpBFgf6O95nq7aCXyE1dA8n97Cl23jc+DYKu3Uqrf36UodS1fHpF1Ys4m1/n0cUcGD9DGACwfJqO+TRJ+zllXwuI6DIK20CaMQcRyjWCygWCzCtm3U1dWhs7MTs2fPxrTp02FlVVFpwwmzDREWfvzjHwMAJ9JEJ4l5yDwkIiIiIqIzj4uLU4ASBQgwODgEiKCxqVF3BY0jvaXOdXXTAMcxE+IjZZPKrALFdRw0NDSgWFfUW89GTbz12V2hmSjryhrHVPDYtoMoClEqlUw3UKUk/R2FQhCiVCojjiLTgEEpBVgWPNtLz+EKYFs2wlA/z+HhYcRxbCp/LOhqlqzTqPlbJEp3QxWlu4ralqkmclwHjjOyBRDIKnVsACr9Ov37HFG9kzV7EFuAxNKT6PR5YVlQiX4uP/DNAoDuBush8AN4no9CEKClpQXnn3c+2lrb4I56nFiCOI5hWRH++9f/DYCTaKLxYB4yD4mIiIiI6Mzi4uIUorueAocPH4YoQbGuqM/ByrbvmS4HoyaKGKlQGd3EoKGxEU2NTXBsZ6Tradp9tFQqo1KpmMm0LvyxzGvYtgXbtuE4+r76+npzXVEYoSwKYViFZVnwPQ+2rbfkqUTBcfWZWgBweHgY5XIJpVIJURTqa7T1ZF2PVxBFkam4sS0bCRJEUaQP/fd8WJZlGjToxYSRs8WOrFQy1Tnp32dM9Q50ZVI654YoMQ0clNKLDZ7rpQsJukLJSc8HKxQKaGpqxMyZLejq7MTMlpkIgkCfRQa9wDE4OITf/e4ZAMD+/ftP96NANOUxD5mHRERERER0ZnBxcQrKDrmvVqsAgMbGRjOpHH2eVibrGGrbFqIogmM7aGhs1GeGiTIH8osIKpUKKpUKlFKmekVPnB3zfCK66kREwXU8BIGeHFcqFURhdv6YwPNGtg8qlUBZKu2+6qBSqaBcLqFaqSKKswYMjmmKAMtCGFYRRSEsWAjDcKTaKK1Ocl0XSKtsLNsyk/Cxg087y0raNXV0p9lRjQ0AjHRuTbugiiizCGA7ttnuGEUxHMeG7/soFAqoq69DR0cHurq60NLSgrpiXXqemIuPDvThz//vzwijEB988MFpv/dENBbzkHlIRERERESnh4uLU1hWYTN8eBiWbaXb63yYg8eOmDxmW/U8z0exWIDjOFBKV7lkHVLLZV2lk23Dc1zdEMD3A7iOY6p5lFK682kg6ZY5fT1mS166HU8pgVIxlNJngOnfryIMI1TKZSRKdyK1LRu2OzJBtWwLpVIJYRiOVPoopa/L8/Q5X46DMApN1VB2rljGNChIv7btke1/tm2bKp7scdlZYqOrm0xDA8uGEpVuVdTXGQQBmpqaMWPGdHR1daGzoxN19XWwLRvv7noXr7/2Og4fPoz/ffd/z9IngIgyzEPmIRERERERjQ8XFwlhuoUu606qJ3q6eibtYwClEoRhBAAoFgsIgkJ64D/gOC5cx0WlWkG5VEYYVmHbDnzfh+f5aSMBD77nI04SxHGMJE5QDauIkwS+NwSBoFyu6BeDrg4SkfQ8sJEJqYiejMdxrCfSAlNNY1kWfN83555FUYRyuaI7mVq2OfPLdV14rgclCmEYwbIAz3MhSiBWYrb66cl82t3V1pN7/Texxm4PzDrPSmI6uGaTd4iYyXaSJBARFItFs/2vva0N7e3taJ3VivqGerzzzjvo3dWLDz78AL29vWfzbSeiY2AeMg+JiIiIiOjUcHGRjCiKEEWRnqiOOgMrjnUnUwBoaGjA9OnT4XkuILpKJfADXaVTKqNSraYH+ltwHAe+78HzPABAnCQIwyrK5bLu+GnrCe7h4WG9zU+JqeYBRrYrZtsQxUqgROlOoVkXVwgc29EVMumWvsAPUCqX9MH/opAkgGUpU4DkplsSw2qIJIkR+IGpArJGVSeJjC5aGmnqMHqrpGXr61BKdz9N4kRfq8ioib6FRCWwbRue56G+vh719fVoaGhA13ld6Orqwq5duzA4MIi3//U23nvvvbP/ZhPRJ2IeMg+JiIiIiOjkcHGRjpJNql1XV6/ESQLf15PA6dOno6GhwZxJlp3VValUMHR4CGFYhWM7I9U16bld1WoVURShUqmgNFxCnCQoFusAwEzeAUAwMmnVE9L0DDABBBZcV587phIFlSjA1tsNoyhEEuvKnThJUC6X9cTbzpou6Od0HBeWbSGMQoRRqBs4eLo7q55Bj/wdzDY/a2RybVkYU6WTPX+SJEgSvQXQEr0l0M62B6ZVO57noVgsorm5GS0zW9DS0oL9+/djaGgIW7ZsQV9f35l+K4noNDEPNeYhEREREREdDxcX6bjiOMbg0CBsy4bve7AsC2GoGxgkKtHneqVnZ1XDKoaHS0iSBJalt8/pSXKMMASiMDLNDaphFcViEQ1pV9RSuYQ4ipAoPSF13exjqcykXImC67gI/ACe7yEKI1i2Zbq7xlEEgUCpBKXSMMqlctrUwDXb/0QEQeADgNnS6PmeOVss20o4mm1b6fbDkZ9llUTZVkWV6LFmzRCUUqYyx4KFOImhlEKhUEBDQwOq1apeVCiXsHHjRgwPD5+tt5CIzhDmIfOQiIiIiIiOjYuLdEJKFPr7+9Hf3w/Lssx5XLZjY/r06QirIUrDpbQRgUAkgUoUQiuC61YRRTGiKESlUjFNE3zPR6FY1L9TEiQqMa9n2+kkVaXb8iw7PcvMSbujjjrry7bGnEuWJImuBEpfR8+NbT1hThQ819MPF4Hv+boaSQSOM7aix7IsiAXYsDG6TCdrhpA9RiUKURzpqqHsrLFEmecRSLotcOTv2dPTY56DiGoL85B5SEREREREY3FxkU6JiGDnzp3m+yuvvBLVahUD/QOQdPKbJMp0Aq1UKrAt2zQtsCwbxWKAoBCkVTwhojCCUgqO45iqmaxjquu6cBw9gbUsS58lFouewCqBbcN0VEX62nEcQzBSNZNV4Diuk55Rprf+uY6Tng8muuoGulpnZLIukHTLn0Bgp+eP6a/1Y7JxZeI4Ts9Ps9JJta48KpWq6Ovr41Y/ohxhHjIPiYiIiIiIi4t0mrZv326+LhR0R9WsLEWpBGFVdwe1bducR1YoFmHBwvDwMKrVKuI4Nh1LLVtX28RxDCc9nyw7ayyOY5SGS1CizxfLtv2J6AYHWWMC/TMxW/qy/3UcJ21qIPB9H7YzUnWjRI1sBUwnzNmZYiL6Z47t6LEoG5Ztma2O+iwzB0oliMJsO6KNUonb+4imEuYh85CIiIiIaCri4iKdMZVKxXztOA5EACV6oux4DnxPT2Adx0YYhaaRgeu6cL2Rs77iWFfpjJ4sZ9sCJeuCatmmC6lt2ebcL5VuJxzpamrBth1zzlgcx3ri7brmGgHohgjppFg/j66y0d8LLNuG7eiurAoKogRJnJiurHrhIESpXDp3f3AimrSYh8xDIiIiIqKpgouLdFZk3U4BIIxCwLJQV1cPEYUwDFGtVhFGIWzLhuM6+twyAUTiURUyohsiwDIlMxYsU3mjVAKIDbHFnBumt/1ZpjOpnZ5PJqK7vgJIzxWD7ogqupLoyAYFRx4BNvKaSjcoSBT6BwYQhtWz90ckolxgHhIRERERUZ5xcZHOiTCsYt++j476ue8HAABRevY6usLGdCodtcUvUQlESTrhFSDdYpid8yUCOOkZYaObByilq2tcz9NbAG3LVO9A6dewYUNEIYnFVP5k2wn37t17tv40RDTFMA+JiIiIiChPuLhIEyoMq8esdgkCPcnWlTp6cmuNaiYAAI6jzywDAFgq7ZyqO5YmKjGNCPTWQV2h43l6+581qhmBEpV2VlU4dOjgORg1EdHRmIdERERERFSLuLhIk1K1esQEOzr240ZPxE+0JS8EgPLpXRcR0bnGPCQiIiIiosnMnugLICIiIiIiIiIiotrExUUiIiIiIiIiIiIaFy4uEhERERERERER0bhwcZGIiIiIiIiIiIjGhYuLRERERERERERENC5cXCQiIiIiIiIiIqJx4eIiERERERERERERjQsXF4mIiIiIiIiIiGhcuLhIRERERERERERE43JKi4tr167F1VdfjcbGRrS2tuKGG25AT0/PmMdcd911sCxrzO3OO+8c85g9e/Zg5cqVqKurQ2trK1avXo04jk9/NERE5wjzkIiIWUhElGEeEtFU5p7Kgzdv3ozu7m5cffXViOMYDzzwAJYuXYq3334b9fX15nG33347HnnkEfN9XV2d+TpJEqxcuRLt7e145ZVXsHfvXnzzm9+E53n4yU9+cgaGRER09jEPiYiYhUREGeYhEU1llojIeH95//79aG1txebNm7Fo0SIA+v+NufLKK/GLX/zimL/z4osv4stf/jI+/PBDtLW1AQAef/xx3H///di/fz983z/qd6rVKqrVqvl+cHAQ559/PhrQDAvWeC+fiKYQgeAwBjAwMICmpqYz/vzMQyKqBXnPwutwPVzLO+PjIqL8iSXCJqxjHhLRlHamsvC0zlwcGBgAAMyYMWPMz5966im0tLRg3rx5WLNmDUqlkrlv69atmD9/vglLAFi2bBkGBwexc+fOY77O2rVr0dzcbG7nn3/+6Vw2EdEZxzwkImIWEhFlmIdENJWc0rbo0ZRSuPvuu3Httddi3rx55ue33HILZs+ejc7OTuzYsQP3338/enp68NxzzwEA+vr6xoQlAPN9X1/fMV9rzZo1uPfee8332f8bQ0Q0GTAPiYiYhUREGeYhEU01415c7O7uxltvvYUtW7aM+fkdd9xhvp4/fz46OjqwePFi9Pb24qKLLhrXawVBgCAIxnupRERnFfOQiIhZSESUYR4S0VQzrm3Rq1atwgsvvICNGzfivPPO+8THLliwAACwa9cuAEB7ezs++uijMY/Jvm9vbx/P5RARTRjmIRERs5CIKMM8JKKp6JQWF0UEq1atwvPPP48NGzZgzpw5J/yd7du3AwA6OjoAAAsXLsSbb76Jffv2mcesX78eTU1NmDt37qlcDhHRhGEeEhExC4mIMsxDIprKTmlbdHd3N55++mmsW7cOjY2N5tyH5uZmFItF9Pb24umnn8aXvvQlzJw5Ezt27MA999yDRYsW4fLLLwcALF26FHPnzsU3vvENPProo+jr68MPfvADdHd3n3Q5d9bgWjDuRtdENMVkeZHlx+liHhJRLcp7FsaIwDgkopMRIwLAPCSiqe2MZaGcAuh4Our2xBNPiIjInj17ZNGiRTJjxgwJgkAuvvhiWb16tQwMDIx5nvfee09WrFghxWJRWlpa5L777pMoik76Ot5///3jXgtvvPHG2yfd3n///VOJvUmfh729vRP+N+WNN95q75a3LOS/DXnjjbfx3vKWh/y3IW+88Tae2+lmoZUGYU1RSqGnpwdz587F+++/j6ampom+pNOWdfXKy3gAjqkW5G08wPHHJCIYGhpCZ2cnbHtcx81OSv39/Zg+fTr27NmD5ubmib6cMyJvn8u8jQfgmGrBVMtC/tuwNnBMk1/exgNMvTzkvw0nv7yNB+CYasHZzsJxd4ueSLZto6urCwDQ1NSUizc6k7fxABxTLcjbeIBjjykv/8AaLfsPQHNz85R4D2tZ3sYDcEy1YCplIf9tWDs4pskvb+MBplYeAvy3YS3I23gAjqkWnK0szM//RUNERERERERERETnFBcXiYiIiIiIiIiIaFxqdnExCAI89NBDJ901a7LL23gAjqkW5G08QD7H9EnyON68jSlv4wE4plqQt/GcjLyNOW/jATimWpC38QD5HNMnyeN48zamvI0H4JhqwdkeT002dCEiIiIiIiIiIqKJV7OVi0RERERERERERDSxuLhIRERERERERERE48LFRSIiIiIiIiIiIhoXLi4SERERERERERHRuHBxkYiIiIiIiIiIiMalJhcXH3vsMXzqU59CoVDAggUL8Pe//32iL+mkPfzww7Asa8ztsssuM/dXKhV0d3dj5syZaGhowNe+9jV89NFHE3jFY7388sv4yle+gs7OTliWhT/+8Y9j7hcRPPjgg+jo6ECxWMSSJUvw7rvvjnnMoUOHcOutt6KpqQnTpk3Dd77zHRw+fPgcjmKsE43pW9/61lHv2fLly8c8ZjKNae3atbj66qvR2NiI1tZW3HDDDejp6RnzmJP5nO3ZswcrV65EXV0dWltbsXr1asRxfC6HYpzMmK677rqj3qc777xzzGMm05jOlFrNw1rPQiB/eZi3LATyl4fMwuOr1SwEaj8P85aFQP7yMG9ZCDAPP0mt5mGtZyGQvzzMWxYC+cvDyZSFNbe4+Pvf/x733nsvHnroIfzzn//EFVdcgWXLlmHfvn0TfWkn7TOf+Qz27t1rblu2bDH33XPPPfjTn/6EZ599Fps3b8aHH36IG2+8cQKvdqzh4WFcccUVeOyxx455/6OPPopf/vKXePzxx7Ft2zbU19dj2bJlqFQq5jG33nordu7cifXr1+OFF17Ayy+/jDvuuONcDeEoJxoTACxfvnzMe/bMM8+MuX8yjWnz5s3o7u7Gq6++ivXr1yOKIixduhTDw8PmMSf6nCVJgpUrVyIMQ7zyyiv47W9/iyeffBIPPvjgRAzppMYEALfffvuY9+nRRx819022MZ0JtZ6HtZyFQP7yMG9ZCOQvD5mFx1brWQjUdh7mLQuB/OVh3rIQYB4eT63nYS1nIZC/PMxbFgL5y8NJlYVSY6655hrp7u423ydJIp2dnbJ27doJvKqT99BDD8kVV1xxzPv6+/vF8zx59tlnzc/+9a9/CQDZunXrObrCkwdAnn/+efO9Ukra29vlZz/7mflZf3+/BEEgzzzzjIiIvP322wJA/vGPf5jHvPjii2JZlnzwwQfn7NqP58gxiYjcdtttcv311x/3dyb7mPbt2ycAZPPmzSJycp+zP//5z2LbtvT19ZnH/PrXv5ampiapVqvndgDHcOSYRES++MUvyve+973j/s5kH9N41HIe5ikLRfKXh3nMQpH85SGzUKvlLBTJVx7mLQtF8pmHectCEeZhppbzME9ZKJK/PMxjForkLw8nMgtrqnIxDEO8/vrrWLJkifmZbdtYsmQJtm7dOoFXdmreffdddHZ24sILL8Stt96KPXv2AABef/11RFE0ZnyXXXYZLrjggpoY3+7du9HX1zfm+pubm7FgwQJz/Vu3bsW0adPw+c9/3jxmyZIlsG0b27ZtO+fXfLI2bdqE1tZWXHrppbjrrrtw8OBBc99kH9PAwAAAYMaMGQBO7nO2detWzJ8/H21tbeYxy5Ytw+DgIHbu3HkOr/7YjhxT5qmnnkJLSwvmzZuHNWvWoFQqmfsm+5hOVR7yMK9ZCOQ3D2s5C4H85SGzMB9ZCOQ3D/OahUBt52HeshBgHgL5yMO8ZiGQ3zys5SwE8peHE5mF7mle+zl14MABJEkyZtAA0NbWhnfeeWeCrurULFiwAE8++SQuvfRS7N27Fz/60Y/whS98AW+99Rb6+vrg+z6mTZs25nfa2trQ19c3MRd8CrJrPNb7k93X19eH1tbWMfe7rosZM2ZM2jEuX74cN954I+bMmYPe3l488MADWLFiBbZu3QrHcSb1mJRSuPvuu3Httddi3rx5AHBSn7O+vr5jvo/ZfRPpWGMCgFtuuQWzZ89GZ2cnduzYgfvvvx89PT147rnnAEzuMY1HredhnrMQyGce1nIWAvnLQ2ahVutZCOQ7D/OYhUBt52HeshBgHmZqPQ/znIVAPvOwlrMQyF8eTnQW1tTiYh6sWLHCfH355ZdjwYIFmD17Nv7whz+gWCxO4JXR8Xz96183X8+fPx+XX345LrroImzatAmLFy+ewCs7se7ubrz11ltjziupdccb0+izO+bPn4+Ojg4sXrwYvb29uOiii871ZdIJMAtrTy1nIZC/PGQW5gfzsPbUch7mLQsB5mFeMAtrTy1nIZC/PJzoLKypbdEtLS1wHOeoTj0fffQR2tvbJ+iqTs+0adPw6U9/Grt27UJ7ezvCMER/f/+Yx9TK+LJr/KT3p729/agDheM4xqFDh2pijABw4YUXoqWlBbt27QIwece0atUqvPDCC9i4cSPOO+888/OT+Zy1t7cf833M7psoxxvTsSxYsAAAxrxPk3FM45W3PMxTFgJTIw9rJQuB/OUhs3BE3rIQyFceToUsBGonD/OWhQDzcLS85WGeshCYGnlYK1kI5C8PJ0MW1tTiou/7uOqqq/DSSy+Znyml8NJLL2HhwoUTeGXjd/jwYfT29qKjowNXXXUVPM8bM76enh7s2bOnJsY3Z84ctLe3j7n+wcFBbNu2zVz/woUL0d/fj9dff908ZsOGDVBKmQ/5ZPef//wHBw8eREdHB4DJNyYRwapVq/D8889jw4YNmDNnzpj7T+ZztnDhQrz55ptj/mOwfv16NDU1Ye7cuedmIKOcaEzHsn37dgAY8z5NpjGdrrzlYZ6yEJgaeTjZsxDIXx4yC4+WtywE8pWHUyELgcmfh3nLQoB5eCx5y8M8ZSEwNfJwsmchkL88nFRZeEqtZyaB3/3udxIEgTz55JPy9ttvyx133CHTpk0b09lmMrvvvvtk06ZNsnv3bvnb3/4mS5YskZaWFtm3b5+IiNx5551ywQUXyIYNG+S1116ThQsXysKFCyf4qkcMDQ3JG2+8IW+88YYAkJ///OfyxhtvyL///W8REfnpT38q06ZNk3Xr1smOHTvk+uuvlzlz5ki5XDbPsXz5cvnsZz8r27Ztky1btsgll1wiN99880QN6RPHNDQ0JN///vdl69atsnv3bvnrX/8qn/vc5+SSSy6RSqUyKcd01113SXNzs2zatEn27t1rbqVSyTzmRJ+zOI5l3rx5snTpUtm+fbv85S9/kVmzZsmaNWsmYkgnHNOuXbvkkUcekddee012794t69atkwsvvFAWLVo0acd0JtRyHtZ6ForkLw/zloUi+ctDZuGx1XIWitR+HuYtC0Xyl4d5y0IR5uHx1HIe1noWiuQvD/OWhSL5y8PJlIU1t7goIvKrX/1KLrjgAvF9X6655hp59dVXJ/qSTtpNN90kHR0d4vu+dHV1yU033SS7du0y95fLZfnud78r06dPl7q6OvnqV78qe/funcArHmvjxo0C4KjbbbfdJiIiSin54Q9/KG1tbRIEgSxevFh6enrGPMfBgwfl5ptvloaGBmlqapJvf/vbMjQ0NAGj0T5pTKVSSZYuXSqzZs0Sz/Nk9uzZcvvttx/1H+jJNKZjjQWAPPHEE+YxJ/M5e++992TFihVSLBalpaVF7rvvPomi6ByPRjvRmPbs2SOLFi2SGTNmSBAEcvHFF8vq1atlYGBgzPNMpjGdKbWah7WehSL5y8O8ZaFI/vKQWXh8tZqFIrWfh3nLQpH85WHeslCEefhJajUPaz0LRfKXh3nLQpH85eFkykIrvSAiIiIiIiIiIiKiU1JTZy4SERERERERERHR5MHFRSIiIiIiIiIiIhoXLi4SERERERERERHRuHBxkYiIiIiIiIiIiMaFi4tEREREREREREQ0LlxcJCIiIiIiIiIionHh4iIRERERERERERGNCxcXiYiIiIiIiIiIaFy4uEhERERERERERETjwsVFIiIiIiIiIiIiGhcuLhIREREREREREdG4/B+vdUsMw0dqKgAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 1600x1200 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Once again choosing a random test image and mask\n",
    "i = rng.randint(0,len(test_images)-1)\n",
    "\n",
    "\n",
    "figcompare = plt.figure(figsize=(16,12))\n",
    "ax1, ax2, ax3, ax4 = plt.subplot(2,4,1), plt.subplot(2,4,2), plt.subplot(2,4,3), plt.subplot(2,4,4)\n",
    "\n",
    "ax1.imshow(test_images[i],cmap='gray')\n",
    "ax1.imshow(test_masks[i],alpha=0.2)\n",
    "ax1.set_title(\"Ground Truth Image and Mask\")\n",
    "\n",
    "ax2.imshow(test_images[i], cmap='gray')\n",
    "ax2.imshow(out_ims[i],alpha=0.2)\n",
    "ax2.set_title(\"Test Image and Generated Mask\")\n",
    "\n",
    "ax3.imshow(test_masks[i])\n",
    "ax3.set_title('Test Mask')\n",
    "\n",
    "ax4.imshow(out_ims[i])\n",
    "ax4.set_title('Generated Mask')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Conclusion\n",
    "Congratulations! You've performed image segmentation using TensorFlow Mri. The code in this notebook will work with any 2D dataset, so feel free to use your own. You can also continue to test this model using different parameters, which can be found on [GitHub](https://github.com/mrphys/tensorflow-mri/issues/new)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Let us know!\n",
    "Please tell us what you think about this tutorial and about TensorFlow MRI.\n",
    "We would like to hear what you liked and how we can improve. You will find us\n",
    "on [GitHub](https://mrphys.github.io/tensorflow-mri/)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Copyright 2022 University College London. All rights reserved.\n",
    "#\n",
    "# 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."
   ]
  }
 ],
 "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": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}