notebooks/2-interpretable-representations.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "ECML-PKDD 2020 Hands-on Tutorial\n",
    "\n",
    "*What and How of Machine Learning Transparency*:  \n",
    "Building Bespoke Explainability Tools with Interoperable Algorithmic Components\n",
    "\n",
    "<https://events.fat-forensics.org/2020_ecml-pkdd/>\n",
    "\n",
    "Authors: Kacper Sokol <k.sokol@bristol.ac.uk> and\n",
    "         Alex Hepburn <ah13558@bristol.ac.uk>  \n",
    "Licence:\n",
    "[![Text Licence](https://img.shields.io/badge/licence--text-CC%20BY--NC--SA%204.0-red)](https://github.com/fat-forensics/Surrogates-Tutorial/blob/master/LICENCE)\n",
    "[![Code Licence](https://img.shields.io/badge/licence--code-new%20BSD-red)](https://github.com/fat-forensics/Surrogates-Tutorial/blob/master/LICENCE-code)\n",
    "\n",
    "---\n",
    "\n",
    "---\n",
    "\n",
    "---\n",
    "\n",
    "[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/fat-forensics/Surrogates-Tutorial/blob/master/notebooks/2-interpretable-representations.ipynb)\n",
    "[![Open in Binder](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/gh/fat-forensics/Surrogates-Tutorial/master?filepath=notebooks%2F2-interpretable-representations.ipynb)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "rqaSG8NJRbzk"
   },
   "source": [
    "First, we need to set up the environment."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "6dPXBDCflrq0"
   },
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "q0RGSb4smTTw"
   },
   "outputs": [],
   "source": [
    "LIBRARY_URL = ('https://raw.githubusercontent.com/fat-forensics/'\n",
    "               'Surrogates-Tutorial/master/notebooks/fatf_ecml.py')\n",
    "\n",
    "try:\n",
    "    import google.colab\n",
    "    !wget $LIBRARY_URL -O fatf_ecml.py\n",
    "except ImportError:\n",
    "    pass\n",
    "\n",
    "import fatf_ecml\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "from IPython.display import display\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 306
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 4983,
     "status": "ok",
     "timestamp": 1598921381134,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "KoZdMV0Jo2LT",
    "outputId": "bd7c1ad8-5990-49c0-db59-73be3368777c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Not in Colab; nothing to do.\n"
     ]
    }
   ],
   "source": [
    "fatf_ecml.initialise_colab()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 7786,
     "status": "ok",
     "timestamp": 1598798302172,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "FVJs7aZHRAe9",
    "outputId": "5019a6d2-e461-47f9-b209-8a52dbcf7cb7"
   },
   "outputs": [],
   "source": [
    "import fatf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1237,
     "status": "ok",
     "timestamp": 1598798356202,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "L5iOSJQkW8EB",
    "outputId": "f735ab93-83ba-4047-cff7-8309ec9acb40"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "22-Sep-08 12:50:54 fatf.utils.array.tools INFO     Using numpy's numpy.lib.recfunctions.structured_to_unstructured as fatf.utils.array.tools.structured_to_unstructured and fatf.utils.array.tools.structured_to_unstructured_row.\n"
     ]
    }
   ],
   "source": [
    "import fatf.utils.data.discretisation as fatf_discretisation\n",
    "import fatf.utils.data.transformation as fatf_transformation\n",
    "\n",
    "import fatf.utils.data.augmentation as fatf_augmentation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sklearn.tree"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "mUOv1Axm5okO"
   },
   "source": [
    "# Binary Interpretable Representations for Tabular Data #\n",
    "\n",
    "In search of a binary interpretable representation for tabular data."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "7Ov_k-vhR-rR"
   },
   "source": [
    "## Toy Example -- Two Moons ##"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "irkpt149UZkj"
   },
   "source": [
    "### Loading the Data ###"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 68
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 7914,
     "status": "ok",
     "timestamp": 1598703173541,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "xZej7VRrSqzf",
    "outputId": "c5c4a26e-7158-4e64-a759-7ba09dc9abb2"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "22-Sep-08 12:50:55 fatf         INFO     Seeding RNGs using the input parameter.\n",
      "22-Sep-08 12:50:55 fatf         INFO     Seeding RNGs with 42.\n"
     ]
    }
   ],
   "source": [
    "moons_X, moons_X_test, moons_y, moons_y_test = fatf_ecml.generate_2d_moons(\n",
    "    random_seed=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 423
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 8908,
     "status": "ok",
     "timestamp": 1598703174688,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "9GXLX60aSq93",
    "outputId": "1d43e30b-1dbe-4e99-9231-1ee38f0a167b"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f5691e2c650>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 480x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot data\n",
    "plt.figure(figsize=(8, 8), dpi=60)\n",
    "plt.scatter(moons_X[:, 0], moons_X[:, 1], cmap=plt.cm.Set1, c=moons_y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "K3ebV1HUU674"
   },
   "source": [
    "### Selecting an Instance to be Explained ###"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 68
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 8511,
     "status": "ok",
     "timestamp": 1598703174689,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "k_Xy8HkoVBIq",
    "outputId": "ac92a629-57b7-408c-aef4-db50417a2a73"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Instance *m*:\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([0.30886033, 0.73748125])"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "instance_m_index = 42\n",
    "instance_m = moons_X_test[instance_m_index, :]\n",
    "instance_m_label = moons_y_test[instance_m_index]\n",
    "\n",
    "print('Instance *m*:')\n",
    "display(instance_m)\n",
    "display(instance_m_label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 423
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 8388,
     "status": "ok",
     "timestamp": 1598703174690,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "cke44IJ5Vjue",
    "outputId": "78d38534-d26c-46be-d4a3-befe566752d0"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f568e00bad0>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 480x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot data\n",
    "plt.figure(figsize=(8, 8), dpi=60)\n",
    "plt.scatter(moons_X[:, 0], moons_X[:, 1], cmap=plt.cm.Set1, c=moons_y)\n",
    "\n",
    "# Plot explained instance\n",
    "plt.scatter(instance_m[0], instance_m[1], c='green', marker='*', s=500)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "bdvvS3T9T73f"
   },
   "source": [
    "### Quartile-based Interpretable Representation ###\n",
    "\n",
    "This interpretable representation is built in two steps:\n",
    "* discretising each *numerical* feature into 4 bins based on\n",
    "  data quartiles; and\n",
    "* encoding the hyper-rectangles built in the previous step\n",
    "  in a binary in/out representation that depends on the\n",
    "  location of the explained instance.\n",
    "\n",
    "Let's go through these steps one by one."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "stVbe0UNY8yd"
   },
   "source": [
    "Before building a binary interpretable representation, we generate a sample of\n",
    "250 data points in the neighbourhood of the explained instance (*m*), which are\n",
    "required for building a surrogate -- see the *data sampling* notebook for more\n",
    "details."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "n31_FnUpZs49"
   },
   "outputs": [],
   "source": [
    "augmenter_moons = fatf_augmentation.NormalSampling(moons_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "N_31H5FdZfV7"
   },
   "outputs": [],
   "source": [
    "sample_moons_m = augmenter_moons.sample(\n",
    "    instance_m, samples_number=250)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 423
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 7763,
     "status": "ok",
     "timestamp": 1598703175275,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "bARQjOnjaE83",
    "outputId": "098112fe-5d88-4538-db40-d45556578420"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f568dfa0b10>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 480x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot data\n",
    "plt.figure(figsize=(8, 8), dpi=60)\n",
    "plt.scatter(moons_X[:, 0], moons_X[:, 1], cmap=plt.cm.Set1, c=moons_y)\n",
    "\n",
    "# Plot explained instance\n",
    "plt.scatter(instance_m[0], instance_m[1], c='green', marker='*', s=500)\n",
    "\n",
    "# Plot augmented data\n",
    "plt.scatter(sample_moons_m[:, 0], sample_moons_m[:, 1], c='blue', alpha=.4)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "J6Xs1IncUe2t"
   },
   "source": [
    "#### Step 1: Quartile Discretisation ####"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "2szia6TVSrBH"
   },
   "outputs": [],
   "source": [
    "q_discretiser_moons = fatf_discretisation.QuartileDiscretiser(moons_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "lgfu4G07zG-O"
   },
   "outputs": [],
   "source": [
    "moons_X_discrete = q_discretiser_moons.discretise(moons_X)\n",
    "moons_X_test_discrete = q_discretiser_moons.discretise(moons_X_test)\n",
    "\n",
    "sample_moons_m_discrete = q_discretiser_moons.discretise(sample_moons_m)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "W9MbUkx2cKd4"
   },
   "source": [
    "Let's print out and visualise our quartiles."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 51
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1644,
     "status": "ok",
     "timestamp": 1598703179858,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "kMP_zgQnSq87",
    "outputId": "87282a01-678f-4525-e76f-d278b9cbca32"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{0: array([0.35216058, 0.48959027, 0.64267622]),\n",
       " 1: array([0.36564816, 0.52390789, 0.67212217])}"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "q_discretiser_moons.feature_bin_boundaries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 406
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2490,
     "status": "ok",
     "timestamp": 1598703181382,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "H_mLf6CcSq7P",
    "outputId": "817d9b16-8e11-4525-bae8-592495aea31c"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZMAAAGMCAYAAAAfnE8vAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy86wFpkAAAACXBIWXMAAAk6AAAJOgHwZJJKAADysklEQVR4nOyddXhc1daH3zMS9zRaSVI3qtRLBQpcoBd3a7Hibhf5Woq7u1RwuMjFvVSAulD3pEmapknjNpnMzP7+WBk/k6ZGWzjv8/SBnDlzZp+RvfZe8luaUkphYGBgYGCwD5gO9gAMDAwMDA5/DGNiYGBgYLDPGMbEwMDAwGCfMYyJgYGBgcE+YxgTAwMDA4N9xvJXvIjL5cLpdP4VL7XfMJvNh92Y95S/+z0a93f483e/x8Px/qxWq+7xv8SYOJ1OSktL/4qX2m8kJiZSUVFxsIdxQPm736Nxf4c/f/d7PBzvLzMzU/e44eYyMDAwMNhnDGNiYGBgYLDPGMbEwMDAwGCfMYyJgYGBgcE+YxgTAwMDA4N9xjAmBgYGBgb7jGFMDAwMDAz2GcOYGBgYGBjsM4YxMTAwMDDYZwxjYmBgYGCwzxjGxMDAwMBgnzGMiYGBgYHBPmMYEwMDAwODfcYwJgYGBgYG+4xhTAwOaWbPDmf27PCDPQxdDuWxBXI4jdXg8OQv6WdiYLC3jBnTeLCHEJJDeWyBHE5jNTg8MXYmBgYGBgb7jGFMDA5pxo5NYezYlIM9DF0O5bEFcjiN1eDwxDAmBgYGBgb7jGFMDAwMDAz2GcOYGBgYGBjsM4YxMTAwMDDYZwxjYmBgYGCwzxh1JgaHNNOmlR/sIYTkUB5bIIfTWA0OTwxjYnBIk5PjPNhDCMmhPLZADqexGhyeGG4ug0Oa3Fwzubnmgz0MXQ7lsQVyOI3V4PDE2JkYHNJcemkSAL/+WnqQRxLMoTy2QA6nsRocnhg7EwMDAwODfcYwJgYGBgYG+4xhTAwMDAwM9hnDmBgYGBgY7DOGMTEwMDAw2GeMbC6DQ5pDOfvoUB5bIIfTWA0OT4ydiYGBgYHBPmMYE4NDmkO5d/mhPLZADqexGhyeGG4ug0OaqVPjABgz5tBz0xzKYwvkcBqrweGJsTMxMDAwMNhnDGNiYGBgYLDPGMbEwMDAwGCfaTFmUl9fzwMPPEBhYSEPPfQQHTp08Dzmcrl49dVXKS4upmPHjkycOPFAj9XA4ICglIIlS6G4GI4eixYdfbCHZGBw2NHiziQsLIy77rqLoUOHBj22dOlSkpKSuP/++7HZbGzcuPGADdLgn0tWlpOsrAPXi0M5HHDjzXDTzXD/A3DxBNTSZYfE2PYnh9NYDQ5PWtyZWCwW4uLidB/bsGEDAwYMAKBfv36sX7+erl277v8RGvxjULvKoKIcOnZEM0vvjRkzDnCHwC++hAULQCn5O78Apk2DgQN2+9QDPrb9yOE0VoPDk71ODa6rqyMqKgqAqKgoamtr/R6fP38+8+fPB+DKK68kMTFxH4b512O1Wg+7Me8ph8o9KqWomzyFpp9+RlVVYe7alejHHsHSvfs+Xbc191dbUIDdbUiaMVXXkHAIvC+741D5/A4kf/d7/Dvd314bk+joaOrr6wGJrcTExPg9PmzYMIYNGwZAU1MTpaWHV357YmIiFRUVB3sYB5RD5R7Vb7/BZ59DYyMAzjVrqH7gQbQXX2DGDFmwTJxYv8fXbc39qWFD5bUbGjzHXG0zW/W+7MvY9gd78vkd7LHuLYfKd/RAcTjeX2RkpO7xvc7m6tq1K6tWrQLgzz//pPs+riIN/sEsXOQxJB52lgAwc2Y0M2ceuIC4NmQInH8edOgAKSkwdCj85z+teu6BHtv+5HAaq8HhyW53Jo888gh5eXkUFRVx7LHHsnHjRiZNmsTAgQNZvHgxkydPJjs724iXGOw9QwbD5//zNyhpqX/Zy2tXXYm6/DKw29GaXbcGBgZ7xm6NyV133eX395gxYwAwm81ce+21B2RQBn9/1JdfwU8/yR/jxsFJJ8KcOVBdA506wo03/KXj0SwWsBjqQgYGe4vx6zH4y1E//ARPPwN1dXJg1Wq46z9wxRWSzdWpE5rJqKc1MDicMH6xBn89P/3kNSQg///Tz2htktG6dDEMiYHBYYixMzH467HqfO1CuJhGjWrUPX4ocCiPLZDDaawGhyeGMTH46zn7LFjxJ+zaJX+3aQNnnal76tSp1X/hwPxRSsH27RAbixYfH/T4wRzbnnI4jdXg8MQwJgZ/OVr//qinn4KPP5YDZ5+F1qPHXl9PORzw3Xcy8Z96Glp62j6PURUVwb2TYetWiIpCjRoFd96Opmn7fG0Dg78jhjExOChoPbrDlMm7PW/KFJHzCbWyVo2NcP0N8OdKcLngq69Rt92GNnbMvg3wyaehuY6Kujr48ksYdCQcc3Srx3YocTiN1eDwxDAmBoc0c+fuptXsZ5/D8hXev0tK4cMPYeyYPX4tVV0NH34ETXYoLPR/sKkJFi/xMya7HdshxOE0VoPDE8OYGBzeBE76ADU1e3wZtWMH3HAT5OXJgbAw/xPMZjii9x5f18Dgn4KRg2lweHPCCRCobN2p055fZ9oMryEBsNshOhpMJoiNlR3JCf/al5EaGPytMXYmBoc1Wu9eqEmT4OuvoL4BunaRAsg9JUD1GoBeveCKyyA5Ga19+30frIHB3xjDmBgc9mjnnAXnnLVvFxk5AubN8+qDmc0woB9av377PD4Dg38ChjExOKSZMKFu9yftB7STTkSVlMDs2eB0wpGD4JJLDomx7Q8Op7EaHJ5oSgV0BjoAGP1MDk3+7vdo3N/hz9/9Hg/H+8vMzNQ9bgTgDf6WKJuNhhkzUc+/gNq+/WAPx8Dgb4/h5jI4pJk4MQnYsx7mqqEBrr6WhjVr5MB336PuuH3fCxkBVbwTli2DQUdyye3d9nhsB4u9eR8NDPYEw5gYHNJs22be8yd98CG4DQmIBtjHH+9VIaMv6o034ZNPobwc2rRhW/FPkBCs2XUoslfvo4HBHmC4uQz+fpSUBB/TS/3dA1RlFfzvCzEkIAaqulokXAwMDIydicHfkJNOhB9/8q+E79Z9365ZWBBspJwOcDj8DimbDZRCi4zUvYxSCt54E/74Qwoixx6NdtEF+zY2A4NDAMOYGPzt0I44AnXtNZi/+w5nTS307AG337pvF+3cGbI6wLZ87zGrVf7RrFz84EOwbDkoherXByZPRmt+3MP778OMmaL1BbBpMyo5Ee3EE/dtfAYGBxnDmBj8LdHOOJ34yy/bb2mXWkQE6uqr4a1pUFoKaWmwKQnckvQzZsK334E70/77YkhLh+uu9b/QosVeQwJgs8GceWAYE4PDHMOYGBzSTJlykJtjff4/WLIEkpLgsstgzGiorITERKbMdQDN41u/3mtI3GzaFHzRCB33V5S+S2x/cjDfR4N/BoYxMTikGTPmILabff4F+Ohj707izz/hrTfRkpODx9Z8zA+9YxddAGvXQnGx/N2uHVwycf+OW4eD+j4a/CMwjInBAUeVlMCzz0kAOz0Dbr0ZLTHxYA+rRZRS8Pvv/i6p9RvElXXqKcFPuOoq2LAB1q2Xv7t2hauvCjpN690b9fqr0jfFbIELzvMYJwODwxnDmBgcUJRS8J+7YNVqOfDnSigpQb32Sqta4I4dmwLAr78eBDmegEwtQJSJm/Edm5aYgHrzDfj9D9H2GnUUmkX/56VlZMDNNx2IEYfkoL6PBv8IDGNisFuUzQYvvwJ52yClDVx3bet3Fhs3yYrelw0b4PsfUBs3wtAhMHjwIddbXdM0VN++UODTfCssDD75L2rpErj3HiDF/zkWC4we9dcO1MDgEMEwJga7597JMGeO9++tuai33kAztaLmNToKwsP83UVOBzz8iGQyffKJZDLtTQ+SA809d0Niguyq1q0XefqCQvl33/3Auwd7hAYGhwxGBbxBi6jKSli92v/g+vWiT9UKtHbtYOCRgUfFkADYGuHX2ahWqkqrWb+irrsedfkk1Cuvog5gBbpmsaDdcAMcdZS3z4mbP/80qt8NDHwwdiYGLaNp3loK32Mh4gG6PPKQ6GVt3gwpKfDRR/6PV1VBSak81gJq/QZ49DFw146sWgUuJ1x7bYvPay1q61aor4eePf13XdnZUpzou7uqq4OanZCRsV9e28DgcMcwJgYtosXHo/r1g59/9h7s1Qv69m39NaxWuPgiANlJLFrkH0fp2BG6dd39hT77zGtIQHYGS1u3Q2oJZbfDnf+R3VaTA7p2QT32KFpampwwcgT06QNLl/o/0dYoveINDAwMY2LQCu6/T6RENm+Wyu+rr9rrgLlmMqHuvQeefR52FkNKKlx/LRQUop5/AXaVymr/tlvRUlOZNs1HMj0mOviCYeG7fU3lcknl+pIlcv6Zp6ONHu094Z334LffvX+vWQvPvQAPP+gz5rvh3PP93F1vpV8FDz0CtNvTt+Evx+99NDA4ABjGxGC3aFYrXHXlbs9TTU3w6WeQmwtjx6INHaJ/vW7d4JWXvM9zOuHiibBxoxxYvwEqKlCvv0ZOjtP7xAsvhHm/Q16e/B0fByf/e/c38PrrMH2mN8axapWkJneTfiTk5QY/p7RE9LaeehpWroKICMjMlHtrJqdvNIxuu/vXPwTwex8NDA4AhjEx2C8opeCmW8SFBdKQ6sLz4YwzpIrcbIZzz0GL9+//oZqa4MUXvYbEzcpVcOd/2HrevWgJCeTkONGSklCvvARvvyMxi3+PR+vXb/eDW7TYP1heVwdT74f335O/u3eHH370l0PJbCu7p08+9R6LiYExY2R3kp5O7sk3oeVZ9niiVuXlsHMndO4cLAR5gMjNlX4mhlExOFAYxsTAg6quhuUroEd3tNRUOeZywfc/SEbXyBEwbJi+i+uP+bB8uffvhgb46hv45jsoKpJjP/2EevYZyfAC1KZNcOPNIpwYiMsFv87msv89CJlJzPq+AMLD0dq0gVtu3rMbq6sLPla8E+V0ojUbOdavg8VLwN4E3brBLTfBTQGvU1srzbDOPAPeeZfLzgqDKCu/LnVKlf8nn0J8PJxxOlpEhO5Q1PMvwHffS1+U7GzUbbegDRq0Z/ezF1x6qXRaNIoWDQ4UhjH5h6HmzIHf/oBOHVGXXeo9/sWXElfYsQOSk1CnnIJ29VVw9z0w61dZtX/6KRxzDDz8UPCFS3b6ZzuBBMt9U2q35cPV16KeehKtaxd48SV9Q+JLYyMUFsLpZ0JqCuqqK9GGDt2zm+7ZA7YGuLIiI8Vgmc1iUB54QHYMjY0SWJ88BTZvCb6W1Qq33ynvU/0d0NCAeuhZWLDIq7f13feol19Ei4vze6rasAE+/xxqm43b1q3w8qsw/cAbEwODA41hTP5BqOdfgI//65ngaxctRj31hMiGvPe+TJAAZeXw9juozZtFHsTt/nEp+GUWamsuWscc/4sfeyy8/a5M/G4iI4PrM4qLYcJE1JlnyuvsftTQZBejU1oKzzyHenegxz2kmprg1tth3TqIioIJF6Odfpr/JW6/HRYvFdeSmyN6B7mYtKQk2YlNuERqaQLp2lXeC/f7BKBc8PMs/06OGzaIKy5Qfn7+Aq8hcVNainI4QkqvGBgcLhhFi/8QlN0Ov872m9ybFi2SdNiyMmlD64vDAXPnic6ULy6X9FMPQIuJEYmRIYMlzffYY+GsM/XrURwO+PprqS4PxLybXuV5eZJV5mbCRJg/X2Thi4rg8SdQc+b5jy0qCp58HEaNgr594LRT4b4p+tffskX+BXLmGfDaK2DRiXHoFS+WlQUfGzkCYmP9j6WlGobE4G+B8S3+p2C3g63B/1hjoxiR7j0gPd2/zW1LpLTRPawNHAADB3j+VkqJq+uzz4Mn3NpaGDtGXEpr1siKPyEBrr1GdMBCub+SkyEjA1VXB/83GTZt9n/c6YTHH0c9+SQVdjsqORlGj0K7chI89cTu7y0xUQLtgU21dpagxcSgzj9XpGXcuxPNJF0YV670nhsVJe7AwPenc2fUGaeL8nBFBWRnyf0aGPwNMIzJPwQtJgbVpQvs8lkxd+gAX3wpfTtMZplEfd01AHFxYmTcrq6ICP0GT3qvqWlw5x2ovG1S4+FLRgYceyza6adL8DoqSsb46Wc6hkTzjuWkE9ESElD33AvzftN/4eZe7Qok0L1tGyozE+3f43c/5jZtUF27wMJF/g9s2ICqq0PLyEA9/qjUpnwWLYbjtVfgscclAy08DI49Dm3kCP3rX3sN6sIL5B6zs41dicHfBk2pwPZw+5+mpiZKW6m9dKiQmJi431q+Hiqoqip47AkoyIekJCz2Jhy+k3xMjHQUzG/ucx4eDhdfKLPyzLe9AfaICLjpRjn33ffExZTVAe65Gy05WeIYv/wiq/ajx4pEyb2TYeWfUmGemQmXTkQ78UQJSn/6mVzzoouks+Ebb/gPPC0Nzj8Phg5B69hR7uXsc/1qPnbL0KHQpRN07gLHHydB91Dv02+/w623+e+m0tLg4w/FZXaI8Hf8jgbyd7/Hw/H+MjMzdY8by6J/EFp8vKeqG8B19rn+J9TWwoknwDFHy+r+qJFoxxyDundycN/y99+HBpt3F5GfD/fdj5p8L9x8i0jPA7zTFZ55Gu2F58TtVVUNsTFoZrNklj38qOweQKrQH3oAvv4KdhR7X6+yEsrLPYYEEMMXSFSUGC49li6FBQskJvPDD6hnnvbobymHQ1xrv/8hhu7ee6BnT3+BywEDPIZElZVJ6nPbtoecdL6BwcHCMCZ7iMrLgxlvQ0O9TLbjd+86OVTRkpP8V/cWC/Q5Au344/1P1HPF5BcEH1u5EiZd5Z/RtWEDvPYaatIV8PgTsL0I2rRBXXO17EjKfTK6Cgrgtjtkl2KxgMPBbNtosMGYTz9FjRuH1r25av3MM6Fou2SEaRr06wfPPi2vvyGgf4qmeY2h0yn1JL/9DqOOkmNXXwsrVsj/b90KZ50NH38oqdKlpVJ3cslEqdSfer+4wBobmZ14NlxyCWNP3r2ky8Fm9mwZo9G+1+BAYRiTPUCV7oKbb/VOln/8gaqoRLvowoM7sL0k8vIrqN1eJMHksDDJxBo3LvjEC86TmIdvaq0e9fX6O4Pycrj3/6QgEiQbq7QU4mKDz22Od7i5v/r/ABgTMQ5++w2ajYl24r9QvXvC/76A9u1g/Hg0qxX1zNNwxSTYvt17kUBPblOTGNFRR6Hq64Ml9mtq4JNP0O68w++w+u8nUinf7P66P/cC2BDG2JNbflsOBaZOlZqXMWMOL3ezweGDkRq8J3zwgf+q29YIv/6615dTgWm3fzFh446Gd2bCHbfDM0/DU0/qxhK0Ll3k8RHD9/xFNA06ZMGWrf7Ht26VDLLWBqBjYoJeX+vQAe2G69FOO81TM6KltBHJ+JZITYWTTpL/V0pqRQLJyw8+tmZtcFZaYyPq2+/EVWZg8A/GMCZ7gt7k79rz/AX1w4+oiyfCKaehrrkW5dsa9i9Gi49HO+tMtMGDWvT/a106w2OPSjqrL23aQKeOwU+wWiX+MP4kuHSixDN8iYqSgPuECVLcqEdYGKCB2QKnnIzWo0frbqrPEcHHopszr9q2hZtuRGuTLPcVHQ0JAS2INQ3i41BLl+GXn9K1S/B1HU3i+rriSlRgJlwrUYXbUYUH7ztgYLA/MIzJnnDuObKadmO1wvBhe3QJVVkJL7woFds7d4r//sEHd/u8vxJVWYXSqTnRwsMlVuG7m9A0uPxy/wJEkwnOOxc++Rht8v+JrMiYMZId5n58yGAxUD266RcqWiyiVNw2E9q1RbvpxtbfQOdO8hq+1NWJC27nzuA05bfegIx0MV5hYfLan30u2lyPPOo97+yzpDYmUArf5RJX2dvvAKBqalBz5qCKi2kJZbOhbrxJjOqFF6OuvV7qZwwMDkOMmMkeoGVkoB55GN59VzKZhgyWSXNP+GVWcOwhv0BqGKJ1+nX8hSibTaRJNm4EkwkVEwPHjoMLzkdzV27/uVIq2N2Ulkq9xx13SFpvY6MULl45yb9b4S03wZEDRVKkV0848QQRenz40eDalugouPdetHHHwLS9CG6vXhO6pa7DAXPmoC44H61DBwC0tm3hyy8kdfriCVDUXJBos8GsX1ETJ6BlZkpNyOOPyS7i9juhJODaxTtRX38Dr78hcajERNS/x6Ndf53+WF59TQQy3SxaBC+9LG5HA4PDDMOY7CFa717w6CN7f4GuXcSt0+BTjR4fLxlMBxFls8FZ53jFCkEC529NgzlzUa++LKnFDToB9vo6mfjHHYP68SeJIz34EOqSiWjt2wPNBYyjR8k/92sGZnOBVMFPe9PzvKysvYgrnfAv2VlUVek/XlYOl1+BGjMG7vqP171XUyupy75UVcFzz6Pi4+H009G6d0Nr1w7Vty9ZC33iKmFhsrt56mmvcayogC+/Qp12Glo7nb4nBToZcduL9vh2W8NevY8GBnuAYUz+YrQjjkCNGglz5koAPzERTjutxSK6v4Q33vQ3JL5s3iy7sWuvhSOPlNRYd6ptWBg0N8FSH34k9RpuQ7l8Ber11yQorkeUzk6sY0ePIQGYMWPPOwRqOTmoSyZi+uYbXOXlUN/gb7wBKirhm28lqO/uutg2U4ov167znmexiGoywOw5qLv+gzZ2DNx8I9NrHoQ1qyEsW1oPf/Jp8C6rslLO0TMm7XQ6NGYemJ7ye/M+GhjsCbs1Ju+++y4bN24kJSWFq6++Gkuzv9xut/P000/T0NCA2WzmhhtuICEh4UCP9+/BAw/AkqWwehUcd5y4WQ4gassW8efbbDBqFNpJJwafpFc34ktl8yr//PNktT5/gcRLjhoJp58uj/38s/+kXVgIN9yASkyUCvIbbkDzja1cdAHMmwt52+Tv2Fj490nB46+skky6+gZREC4rg/btYdIVaBERNDQ0sGHDBux2OwkJCXTp0gXtgvOJv+pKKnbsgNJd4j5aMB8afXq22+0Ss2o2Jpqmoe65G556RtxUjY3+O6eKCulDP3aMxI8eekAy8kwmqZivDtjVyEVh2gxUt25ogVlmV10p0vgr/xSVgd69DK0ug8OWFo1JXl4e5eXl3H///Xz22WcsWLCAkSNHArB8+XI6dOjA+eefz9y5c5k1axanuycVgxbRNA0GHSn/DjCquFgmOrf75I/5qKoqtPPP8z8xq0Poi8TFQXNxpqZpcM3V8i8Qp06cwrcnSEEB6o3XPW4lLTER9crLYuhqa+Gkk0Qs0ofpzzbBJ98xsXG67rWdzzzF4sWLqW6eyEtKSrDZbPTp0wfNYhE145gYePJx1PU3wIKF3udbrTCgP2rGTKmBiYmByy5Be+0V1I4dokgciI/q8vTpUZCfz8Sx64J7ubhRStKgL56IuuVmtFNP8TykRUbCC8+h8vPBpdACM+X2IzNmSDbdxIkhFAIMDPaRFrO5Nm7cSN++fQHo168fG3wqi9PT07HZbADU1tYSF9AIyOAQ4f0P/P3wNhvMmhV83uWXwciRMqGGhUnsoksX6NcXbrwBrW+f3b/WsKEyQbsJTDVetx42bfI7pLVpAzffJOnFb7yBuvY61Nffeh5/+80w3i4I3q0AsHIl21ev9hgSN7t27UJXcu6mG6Xo0WIR9+KJJ8KKP5sD4X/Ajz/CpZej7robHnpYXGF+g9UkiQBQ9fW8/Wg1b0+PgNvvkB1GeAvJAg0NEsfRQevQ4YAaEoCZM6OZOfPgJngY/L1pcWdSW1vrcV1FRUVR6+MPzsjIYPv27dxyyy0opXjkEf+g9Pz585k/XzJVrrzyShITA3L5D3GsVuthN2Y96qwWAgU0TOUVhM2YiXbxRSQmJ3sfmDENZ2EhOJyY92JyU3fcji0piaY//gBNw7F8hb+svdVKfEoK5oD31fb229S/+roYOoB164lu3w5TRjo0JbXwgooIvVhTSQnh02Zg+c8d/p/hkUeivvwC54YNmJKTMaWmUnnSv3H51g/V1cHPv+i+nKl7N+Jvvx3NZKL+jbfAdrY84HJBSQnmXr3QEhJwbtiACuwPA5jq6kjYj9+pPfmOmpvfp8PtO/13+R2G4u90fy0ak+joaBqafeD19fXE+IjrzZ49m27dunH22WezYMECPvnkEy680CsrMmzYMIYNkxoMQzX44KFOPQ1+/AmKvenIroICbC+9TONnn6PuuN1fLt2dnry3937mGfIP4I47pSGXmz59qE5MDLq2+uEnryEBqKmh9qab5Zjt+9Cv1b07Sa+8Smy/ftRkNNf/uFwkb9hI43ffUx0ZTuP55wc/z10rVFGBcrUyy8lkwnXKKVQ2Z4ip3K1BpzgjwtGefVrcVjfdDAHFqK6c7P36ndqT76jTmQJw2H2n/y6/w1AcjvcXGaLIuEVj0q1bN77++mtGjx7NihUr6Natm9/jbtdWbGws9aHUWg0OKlr7dlIbM2WqV1q+GbVjh6j/hui9sc889CDMmCmdCzt0gEsv8b62zQb//UTUgfXSjfWqycOskJIqux2TBuVlmPO2MWjdejaOHY09KpqEoiI6z/sNlMKxcBHoGRNfwneTkm02i4EdPBhOORn18X9FXbhER6csS3ZzWocOqA/eh6eflnbBIG68e+9p+bVCoKqrYdp0aWQ2ZAiMP8lQKzY45GjRmGRnZxMfH8/kyZNp06YNJ598Mq+//jqTJk1i5MiRPPPMMyxYsACXy8VVV131V435H4uqrpYmUnvaUKlzZ0lR1aO13RX3ELVsGXzwETTaYMhQOP9czwSobDaYdKXEUPYEa5hI2rhjJM0ZZpE1NfT98uug07XdNPFSjY2hJevdDB0Kd94uBasz35FeK+4gvGaSTK42bUSy3qdKXwsPh7vuav29hRqj3Q7X3SCKCSBpyvnbJE3bwOAQYrez0kUXXeT396RJkwCJodxzz96ttAz2DLV1Kzz8iKTaxiegzjwD7awzW38Bmy30pNnZX29KzZ4jVfox0XDJREhJ2eNVsMrLk5a6Jc2uzUWL4b33UJGRUrTZvUerDcmocJ9+7pGRoWthAklNJfySCYSSX1S//wHPPutNSw7FrlK0jObaj99/88vmGhU2B3KypWlWYG/3FlAOhyQipKai+cas9Pj+B1jv8141NcGceahrWk4hVtOmS+zH0QR9+jDqqEfE+BkYHCCMosXDgUcfFxkTkOrtN99CHTUSzVcnrAW0hASRRgncnbRtC3d6pTvURx/BK69CXbPh+epriI9HZaTDNVejDfBP2w3JRx97DQnIbsK3idbGTfrPS0iQMWZmQnISrN/AffFT5bHICDhmLHz8SbCkvC8DBkCXznD2WYT17UvtokXSJdLWCEeNQDvlFMn0ev2N3RsS8C+sDDCq98VPhYZIeGQEaup9HuVi5XSK/tqy5eKaO/FEtNNPk8dW/CktfrduFbmVY49Fu/Xm0K9fWxt8v44QacjNqJ9/genT5Z4BcvO47/xotJtv2v39GhjsJYYxOcRRdjtsD1CULS+XVeeFF7T+QpPvhbvu8a6sU1OJf2cm1b4yLj/86DUkIOeWlMi/x59Evft261xsuztHKXEP+epnmUxw3xRIiBeD8/4H/hpgDTYpMAwP9w/W+77mkMHw5BOeMTqLiqTZVlFzavTChaiqajjnbNilkxAyYIC4/dzpy8lJcNYZ3sePPRbWb/Df5TU0wE8/Q36+GBFrmMRYli3z3t+Wrai2mWhDhkjx5ObNcrysDL78EjXumNCp1zk5we9VekbLu8W5c72GxI1vVb+BwQHAMCaHOlYrJCb5r/SjouAIHZn1lrDZ4IjeImLYqRP83z2YO3Twz6xqaqEnx5YtomfVvTtcf11IUUqVlyeTpKaF3kEkJ0sNy+zZ3kly6BBRYF60WCrQm7Om7quaIv+Nnyq1HIGYzVI5fs010L+f3yRrmzbda0jc78GcuXDRhZCW7v+exsXBXf+B9DT476dibE49FS0n23OKduYZqPg4ePoZ2FXmP7YNG73XCpz8a2vFXTVkSLDIZ329BPRDGZOvvw4WrQwl2e8mKTid+r7NV8OUOKZO1anSNzDYDxjG5BBH0zTUBeeL5lVxsax6x45tXRFhM2rOXHjkMW+wvapKJm3fnuogBXkbN4ZW3F2zVv5tL4IXngt+nfx8uOEmkSIJhcWCSklGe+ghaZX7yy/QvTscdxw89hh894Pfyn9u41Et31xmJtqbb+g/pmvLlLyn446BDevB3iSTf8eO3sLBi0Lv+LRjj0UtXgKf/y/02PTeP3dMJT3N//2JioJRI72j21UmbsLwMNlB6fXQacnNBxLrWrLU2744M5O5JUfB3EO/vbDB4YthTA4H+hwBt98qK/4ePb190FvLt9/5Z23V1Untybnn+J93/XWiXrx8hbS1DVT0dbN2Lap4J1p6mv/xd99r2ZAAOBxcbPmId17rgHbVVdC/PwDqnXfh8y9CG7JQ9Oge8iHLmNE0fvSR1y0WHg5HHSUxk59+EkMC8ppbt6K2bEHr1Gn3r3n11bBpM/zavAuKiQlOZY6I8L5uxxyYcLH8//XXw6OPQm6eVOEfdyxa794AqDVr4O57vbupH3+S1OZFi72fX3Q0jDu6xeFp8fGoN1+X97OmGs46E04/uKrUBn9/DGNyCKOUgqkPwLx5Mpl03MtaBbNOFo9ee16TCa6UbD119TWhjYnZBBadynO7TmA4wOWzMcHO8pRGNn8xnc4JiWhug7ZyVQuGJCA+EB4uyQNdu4hrClB//AEz3haDm5EBY8fQ4GtIAHr2RJs4QVKTdwY0I6muFvHKVhgTLTEB9dYbMKxZQujRh+HBh7yFoV27wp13iDxLTDScd57I9wPaEb1Rb88UY90mBS0h3nvhmW/7u+Vyc2HdWrnWt99Je+Fjj0U7UUeoM3CMERFw3jm7Pc/AYH9hGJNDmZ9+lgnJLSK4ZQu8/DK8/JIE5j//Agq2SbZQz56hr3PWmZJZVFYmfycmwmmntvzagZLtvrgU3HMvKikJTjsNjhwohui0U+GP372aVlYroPyMxLvd66iMULzTuYKpM2dK8LlNG4mjhCI8XHZnxTulf/tVkySY3Yx6+ln4+GOvSyg/H5Yvx2W3+1+nssJ7veRkb4aZm93tqnzQTCaIktiFNmQI6s03pAgzNg7OOgMtKkq/fTBIu4HOnYMf8E1+cFNTg3b8cXD8ca0em4HBwcAwJocgqroaJk+RjKBANdrSUqlTuP5GeRzg2+9RV05CO+fs4GvZbFKnMHwY1NaJL/6009AG9Pc/z+GQlNUVf8quoyVjUlUlxgmkJqXPEahnnkbr20ck3D//n4y7R0/46CO/wP6qFLmflSlNsKAMHn8CVd+gHxsAMJkhMQHtrTdRDodfNplSSnYEX38TvKsJNCTgqbPQNA1lD1QsA/77CSojA21PsuTcl05Lg+v2sZCwb19pKey+l/BwGHGA1AlaQCklcZs/5sv35dxz0AYO/MvHYXB4YRiTg4SqrobZc6Bzp+BdxWNPSIaPHukZsltZscJ7rKZGakICjImqroZrrvMGYjPSYcqUIEMCSID/f1/sxY0oqYG5737Uk4+jjR7tbTYFqLVrJKUX2BbbxMZEMSYbE5vIT4IOc+a2GCe5+JTtaP2kuVZQWvL8+fD9D9giI1FmM5F6/UR8SZYsJ+V0Bulmee7lw4+k1W58fPDjAUyY4N+vXSkFzz0vk7DTIY3E7rzDv31xS1x+KdTVSPDcZJa+KXq9Z/aCwLG2yNvvwuuvew3ymrWoF59HC0zYMDDwwTAmBwH1yy/w/AuSphsdhRp5FDww1ZvWmq9TTGexiHz6rTdLYDZwAq7TmSxmzPQaEhAdrHfelR7tgHPXLtTk+6B4h/7kuifMmwc33ox6+kn/Sf/xx+D0M6Gignd71FEVIZlIVRGKd3vZuHtey4HhiWuugykfolYWwCuviT5VRgbcfhuuRYtZ8e+TKMvOxmU2E1+0g4H//QRrY6PI6AfuTkzNcZ5Zs0L3H9m5U9yJrSjQDOoN8uGH8OFHPu62AtiyFd58fbfXgmbX2c0tFDDuA3vUx+S33/zfu9JS2ak0x6cMDPQwjMnB4O13xZCA+Ml/+QXGn+Rpf4tex8qLL4SrrhIXzamnwBdfSjGhm65dgp+jF0BvzjpSSlF37XWwdNm+3YsbpWDBAn7++EE+TtlOpMVbC6HGlEFFBWuT/SfwWWnVFI8JNoI2i+LsDdGMK4iUivinnpaq+dzmOpO8PJhyH1svvogdNdWeyvSyTh1Ze/yx9J01G/r3g3m/+V84rLnXyo7dSLI8+zxq5vQWCwOV0wnTZ0iqdHy8JC4sWx7srlu5EjV3Htqo0CnOymaTVsgdOnjqWlRNjcRglIKzzkT7K/sFtTJhw8DAF8OY/MUopbyBcDcOh8hfuI3J5ZeLscnPlyB2//5w2WXeDoWpqajbb4X3P/Rmed11p1zf5ZLJd/ESiXuYzf4TXM8e8t9t+Tj2RGgxsBBP/+YYkx/O3Ox0Pt/8ORWNzQHvlOZ/AaxPsrM+oL4uscHEaZsjGVMoO5ZLyqbBF1FMjzjX/8RNm6g2m4IkTuoTEuHYcXD+eZhy83AVNu+4kpO80vgnnyz92kMF3Netg7emoS6+SFx/hYVw8slonX0yvR5/gkve/DeooUxPvlTiUnppykrJSj+EMVHzF8CTT8lnHRuLGj1K6kRuuQ22Ne9Qv/se9X/3wG+/y/fhnLNb5YbzZeJEeaNb1Qv+2OOk0t+9283M3DO1BYN/JIYx+YvRNA2lt8or91aia337oN6ZKZIpqakwZHDQKlkbMwbGjJEV8h9/wPLlqOHDYcYMmSjdE7/JBCkpUgvR5wi44Xo5HhmBFh6O8pUGiY6GgQNl8vM1HFZraLeQLxERWEaN5oHhwzmq7VE8PP8BNtUE9/0IRddyC3cvTmDcNq/ra5uzAziigt1WNhsxP/yItc8RNEVFeYdQ3wC33IwWEUHM2zOofvkVGfuZZ3rrc5YtE+PSUC/3VhrcyIo//pDdgjs29d33qOuuRTvlZDHYi5ewzeETcN+yBdq11b+x1atRpbvQUtoEP/bmW97WADU1ksG3s8RrSED+//obvUkRP/2Mev5ZycprJdu2tX5noZ1xGioxXsYSHgETLkLLzGz18w3+mRjG5GAwfJhM+L74trsFSS09+d8tXkaVl8PNt8qqWNMkplJT628IXC7o1RPticf9r5+WhmX4MJq+/c5bUZ2W6q8pZbXKrufUk0XixKEjt9KtqxjCyEg45hi04cMBOC7rOI58/QduMBXxW6aNpha+aVYHHLU9gudmJ5HYGOBiMZvBYtV97S4//Ej27DnUpKay4pR/E1HfQI9Vq6TGArB06IB25x3+79mcudKS1x2sj431LzB002ATA+GmslKy1E452X2l4BtZuEj/BjdthrvvQb3+qt+iQDmdojjgS2Oj/o7JN7suL0/6mzz5hP7r7Qe0o4+Go1sujjQw8MUwJgeDSVfAggVQuF3+NpslvfaWW6UyOj4Ozjpr95k8L78Ka9d6/16zVt+37bNy9yXmqSep6NZNjFGnjvDl1/6V3GaT9Gcf0B9+ny87lkDsTfD8s5CV5VHNdZO4o5K3VyXz2KAqXu1Ti0NnaBalceWqaO5cHI8WWJwI4HRBdRVEBLvYNCCssZHkggJGvf4mFrsdzTPZh+Dbb72GBPT7uSQmStW6W5DRTbPbRzOZUF27wtKA5+kJULrZtElkaHx3L9//EPwck0lcYh//V994e8at0zzMwOAgYjQ4OAhoiYnw6CMwbpw0X5p0hehuzftN/PNr1sKLL0knxGbU5i2oWbM8bimllGhLBaJXrzFunP44LBa0885Fm3ofnHlmcEaYrREKC2U1/fSTMHJ48EVyc+HNaUGGBJB+HWhENJl0DQmAQ1NEOUz6hgQIIbAVhNVuF0n+/9y5mzN305tF06QKftIVwYKJvkkOd97hzQ5rDWFhMH0G6qmnUe4EgMWLg8+LjZUmW+edEzroHRGh/1nsJ5TNhvriC9S336Fa4940MMDYmRw0tG7d4JGHPH+rU0/zP2HXLvjyK9SkK6Tl7rx5smvo0AF16y2ys9kUsHIGUb/1XXm3zRQBx92NJyoK1bGjvK6bdu3gGHF1aJqGGjIMftOpfwmsJHdz802ws4TFmbNbfO1FPaJgfaS4csLD/RpQ7RHt2+9eIv/0U2H58tA97pWSOMnmLXDLTaIgXFPjJ90CzQuCeIv3vW4pQUHT5LP78kv5e+48VE62fiZd3z7yXl91lXSqDCQiAi6ZiDZ+fMv3uZeoggKR7d+6Vcb94Ueo555FS0w4IK9n8PfBMCaHCnFx4gZxY7GI5Mb8+VIX4Z5g8/OlsVN5efAuZOhgOO5f8PbbUjuSlg4TL0YLIVmulEK99DIsWiSr7KFDICIcCgpE9n7iBLSYGO8T9OpfQAyWDlpaGuUvP8GGj8eCXdw5WbFZjGo7ijnb55BfI4HnDWkalS8+RkKFTVxMzz4n2Wx2O5NtD+z+vXOzdCnqyafhums8cRPfe2XhIolR3H4r/Dpbikb1Vt4OB6xbh3bdtXD88SFfbvIjdnj3U6hOl+sEZul5X9z/dYqK/DW4QHraDxoEdze3+rVaZTcT6Or61/Fol14SckyhmDKlldLzr74mhsQ97nXr4M034fbb9vg1Df5ZGMbkUOGMM0R3q7xCDMngQTB2jFSmB67UfQsRfUlNR/v3Sajjj4Xt20WePTy07HjDq6+K0q97wtqyBe65W7Sg9Dj+OBEc9HWHhYVBeDiqtBQtJTj/98ONH1Js34WGxpD0ITw7+lnax7Unvzqfm+bcxKLiReyo28GHahlXj74aAPXWmxJPmjyZMbVzdy+57sblEvmWwgJ49hnPYaUU3PEfMcyNjdKz/dqrpYfIM88FG+X4eGhF5fmYn+6AjT5thWNjITNDdjWh5GFCjl2hvfiC509N01D9+8Pvv/uf14KGmdq+XbLDaqqhX3+44HxPwH/MmFbu9pr7yPixSyfbzcAgACNmcoignXIyvPWW1BjcPxWeeVoqoo8d5+2F4cbh0A/ONsdYtLAwtJycFg0JgGPBooBuhg2yWg81xr594dJLJKYQFyc+fbtdajGuvgalszKfv2M+ieGJXNb7Mj4+6WPax7UHoENcBz4+6WMu7XUpieGJzN8x3/s6mibV+6vXtN6Q+L3oApRvweL8+ZLq6zbKu3ZJdlp9A5xzjn+titksGVvrN6DmzpOMK6CpqYnq6mpcza4s1dAQbNRraiQ+pRc/8iVEMaR66mn/A3qf35q1wcdobkp20cWiUzZnnnR0fO75lsehR9eu/n+bzdCv755fx+Afh2FMDiG0dm3Rrrka7dhxHj0nrVs3uOA8qQHZHQGunZZQc+bidLen9SWm5dfRLr4I3n1bald8V9/b8kWqxYcaew1bKrfw3JjnuG/Q/2H65VfUt9+imid1i8nC/cPv59kxz7K5cjM1dp/MquaOhEeX/MzRJT+3+r4A2aFMn+79e936YGmV2lpx6Xz3nb/BcjrFOE6eAnf+ByZdxZZVq5g7dy7z5s1j7ty5FBcXg8XC0Rv+6z82q1X6z7sLQwPRNMjOhuOO1X/8p59lF+UmNib4nGj/zDxVUYGaPx8mXemf4eVwiKR+M2PHpjB2rE7laCBXXwUnniiuy/bt4fTT4Nxzd/88g388hpvrMEC77DJUdY30RffFt5AvOdm/X3kLqJ9+hkcfEyFIXzp0kJ1RS8+tqICbb/GvwXDjWwAJWE1WvjvtO+JswCWXSlU1wKuvoV5+Ca1dOwDGdRjHoLRBhJnDfMbSHhYu9L9+bIzUf/jupkIVVFZWev//hH9Jqm2gvIzLFdzUCrxBdYcDVq5Emz6DhqPHAlBbW8vGjRtJO+ooiIqQtGU3ffrAsGHQu7fsCtyxEZcSufpjjkG7cpIoNP/4U/Cuq7ISzjgT1bOn9K257FL480/Ia45Vtc2ECRM8p9c//Yzc1350Q2lWK0ydImPUNJHLNzBoBYYxOVy45BKZXLc0B0eTkuCKyyXzqL4errgcrUeIFXEg337nn/EFogf2+qtoLfUVAYnh6LlaEhPh9NP9DkVYIoiwRKCefNhrSEC0sa6+FvXO2xARjhYRQXx4gDzIdddK3OdLs2TzHj1WJtiSEqmvKSwUY+Byyb34Gg+ANG8XSC0zU7Kjpk8PLgiMCEdZzGgNzfUeOgKRYQGZX7b6ehrvuRcaJktB5bBhInlz3jmyo4yPh8n/5zlfORxgMnl3mxYL6rxzgxcHbjXjgkKIikK7+y6JH330sRjMc872fD6qcDuNH3wYOisNYPBgz7k0Jeze/ebDbrPiDAwCML4xhyiqogI2bJTq9dhYtIR41OuvwXvvi2/+5JPh1VclQ8nhgNVrUC++iNZVp+mS+5q//wE//+zN1vGlshLeew9uuKHlgYVaBXfoEOSC8bAksLoPqas5+xxRTR53DNq1/r1AtKgoeO5ZWJYIaGiPPSoPxMbCU0+giorg0sv9M6jck2VOTtB9aKedghp/Iky8REQj3bfTpQvbO3Ukbe06zDExJMfFYfrJ67pSFgvVOf7S6+Fl5YT9/AvYm1OFtxfCow8HZZB5XltnYtZuvgmVkSGfZ1VVcP+Y5oJJLS5OFg2BLF0q35FQ9OkDV1+JuvEmWLkStn8liRJ1jWi7cZkqu1203ZKT97xF9CGA0+mksrKSqKgoIkNkMhrsfwxjcgiips8QxdjSUsjIQF0yEe20U2ViufoqOefd9/x7nlRUwEUXofocITpUARlZ6v0PpEeFXjc/N3/M370xCcWff8JV16AemIo2YABOp5O8vDxsNhudIiPQTQWoqJB/H36MGjAAbdiw4HNCrZC//iY4FTcpSYorO3XSdc9oVivqqSfhiadg5Z80OZ04q2soT02jsLmnTKd27ejudMruy2JBGz4M66mnEJGfj81mIzoykq7ffofm66LKL4DLrkA9/5y+/lYItHPPgXPPQb36Grw1zf/BuDhRE374UVEoiIyEk/+Ndkbz7m/oELTkZP+kB4sZjjhC6oouvxxefFE+U5AdXEMDvPiKFFzqoJqaJKngwYdlwREZiRp0JDz6yGGzU9m1axerV6+mtrYWq9VKZmYmRxyh3/HSYP9yeHxD/kGoigqvIQFxy7z3HurEE/yzs3ybY7lxuaRTYn4Bqkd3tA4d5JoOhwTHWzIkAFu2os44C555yvNcv7HV1YncSyhKSuCDj3D27cv8+fOpbHY9FZ11FmMfexxzqHRZmw3mzhN3UWvp3DlYETkxES0wGwmoq6sjLy8Pq9VKdlYW1tISqKzCCqTVbCa8vp7fL7+UDkuW0Pbd90V+5qQT4dJL0MLD6awUOT/8iFq4CJPVisktzOjL5s0w8RLUSSdIjGNXGaSnwy03SVvilrjoQtlhrmnOXmvbFiZOgMefkAQBN4WFqO7d0Xr1REtLI3ziBGzPv+CNGTmckpZ8800i+bKtIPi1AutbaFaafvQxqTfaVebNequvl8/liy8kdf0wYMOGDdQ2x8GampooLCykbdu2JAWqGRjsdwxjcqixcWNwRXnpLslu8p3gu3aDOXP1r1FeLhlJboXg+6bquqe0+DhUVUDsJD8fbrsd9fxzkJbmr1a8fr3EMVqioYG8vDyPIQFotFrIG3cMnX7+RQyeyeRvBEwmEakMQJWV8dbY98HpQu0Yh5aRLsddLklh9h1bbCwMG4ZyOv12JTt37mTlypU0Nk+QO/LyGFq4HZ9QP7E7d5K1cCHdfp2D1R0vWb9eMtsuugimTcf83vtBtSNvJQW4n0pKYPpM79+rVkFJCeqN11rsjaJFR6PeeA2+/x7KKuDUk9Hi41FPPOV/YnU1fPMN9JJdlKV37+AU8ZoauG8q6v33oH274LFm6KghvPue9MfRq+BXStythwFKKWwBWmdOp5PS0lLDmPwFGKnBhxq9eknhmy8Z6dJTwpfLL5Xiu1BkZKC2bkU9/qTX1eFLYiIRPkFiP3Lz4JzzYMIlqOXLvcezsnaffjxkUNAPOik3j3bzF8hkrJRM/BnN9xgRASNHwEkn+T1H5efD5ZPI+eo5cr59Aa68ErV6jTz4zbfSuth3Iq2vl8r/iyeifAQac3NzPYYEoMZuZ8uokX6v1RQRQZu8bV5DArLaX9isnbV0mW4RYo4ljxxLXsvvx6aN4gbbDZrFgjZ+PNqEi7y9SnzVB9ykp3ufkxCvH1TP2yY7lCsnwbChEB1NTuxOcoa1geuuCT5/1erQUjBhYTBkyG7HfyigaVpQjMRisZDmk4xhcOAwjMkhhhYTI5lbWVnibunYEa6+KshnrZlM8NkncMbpkJPtX+DWvbus9q+6Bv77X31l3AvOI+LEE0Mbh/p6kdJ4+lnZCYBkgbWkjBsdDRdcQPv27QkL8679sxctJtw3e6yyEnr3gkcehldeQnvqSXC5UG9NR/3fZNR774vEemEhuY5sch3ZkgH27jvy/J9+DJ7cnU75t3EjPP+iz+FgI2DPzvbGYiwWzKOOIrlTp6DziGh+T0P0cPeMrSXCIyQteG8452z/BUNCPHzxJeriiaiP/4u5a1fJHAskMRHaJKNFRqI9/xy8M5Pc8+4lN/IIeP99cVf6ohfnsVrF4J92qmTSHSb07NmT+Ph4zGYzkZGRZGdnk6DXudRgv2O4uQ5BtFNPQZ3wL3GbZGSEDH5qkZEelVz12OPw1dfi7965UwK6obJ9evWECy+UmoLExNAdB0F87Lt2SZMuX7l7Pbp1RTOZiIuLo2fPnmzbtg2Hw0GUVWf8ThfauGO8f9/xH690yA8/QqoU2F1W/iYAs1LHQYMN9dIrsFgnO8wXHzdhQkICFT7vg8Vioe2JJ0JGJixfAQP7Yx03TlKNV66S/4Lc73nnym4oUIq+mcvK34SwMGb1v1g/lqRpMHiQrsxMa9COHovq1hU++x8sWyqKAJXNdS2FhdQtWx7sErVa4eix/ineP//MZY+dDWqIvI/zF0pvFfeu5oor5LNdt152jp07wT13Q3b2bjO/DjUSEhIYOXIkdXV1hIeHY92DdGiDfUNTam/0KvaMpqYmSkMpy7aSUNW706aVk5PjJDfXzKWX6vtFf/1VXnv27HCmTg3upZ2V5fS0M50xI4qZM6Mxm81+q9pRoxqZOlVW11OmxDF3bnB+0oQJdUycKEHuiROTdLvbTZlS7dFJ2m/3NCVGJkEft0+WJZ/pSZfKPdVN4O26iwANwsPEVWI2c9xxGv/ZNQHmzuW+qinMbQxuLXtxh2+YOOsMNKuViaMqg+9JMzE59THG9CiAu//D0VeNDLoG1TW8ZT2LHOdGch3ZXFY5DdqkeKvt7XYoKmJWG1Eonm0bzf21U0Ap8pqyAMi2biOrs8b0+AmQX+BzT/6MCp/HfR1fhCN688Dma5hVPpBGexNOpxNNA4vFyhVX2PU/J4cTqipBweR7yxl7TgLq9js5+uN7A+5ZAwV2SyRhqYm8Na2My860iiF3OEC55L0OszJrNWjh4f7fvYYGqVbXNLJ6hTPjI9ntub97HhoboaqKUakruS/6/2DnTv/PyWQGl3xHL45+h4nRMyE2lktK32CbI0ve37o6iWPZO5BiKmFh+nDQNI52roA4H5kepaCunreu/Ymc8weRtz1yv/6egj6n5t+TUoobbnCycKHssCwWM+HNu+wJE+q4+eZwKioqmDgxiU2bXDgcTSilMJnMREREcN99B+D3tI/3BK2fIwoLw4J2zwdkjmi+J/exfSEzRNdNY2dyqOF0oR5/HLVrIcrUDWqf1Ped+9LkkInQF90lgoJGu9R4JCXRtGIjNGwSl49JkwI8s1niBS6nHB8w0LuC7dABdpZDo02uHxYmK/hb7oDTo0MHmeNi4fQLYM0XUN8W1iT7y7a4lI4GlyZNwsqar5mQAOlRUN6Cm81NWRnMnoOzagyaqR0RGQExKOyozz6Hb75FLbsH5cpBa5MsqbXNK3otzQo0BlX1A5Kmm5AAWjiEa1Jfkpok72uTO+6ipHHYF9/B2Wd5n1tbJzu9ZiPAqp2o/DBv9pzdLu+FySztex1N0LQdEkqCx2Eyea/jpqZGsvacddKWWG+tqFSwm1DTICYabfQotLA9FKncB3Jzc6mqisflku+43e5E0zQ/NylAU5Mdu93pkZtxuVx73anA4MBw2OxM/moSExP93CP7C1VQID79sjKRx7j1Vgmkuh+/7Xb/LK02bWD6NLT00EFEZbfDBRdJO1c3sbHiwtJLYwXRXaqo8JcTGTwI7aUXZYxr1sDQoWgB/mblcklAutEmj4dwwanPPpfAf1SkSNl37Kh7nueal17mX1l/5JFor7zkWZm5V1Tqhhv9NKd0W+76EhEB099C6+wt5lSLl8Add3ruXQFFY0aTMGUyMQGGW73xpijxuuNG4eFw939g1CiOPikLTCbv2C6aIFlgvpx6Cto9d3uvd/2N0ovG75xT4dab4fY7JLXb6RQhzcA6GotFdj6aBkf0JnL8eBqef0FfEiYAt4bYrNRxkswxc3rQZ3swWLx4MTubddjcpKamMri5et/9O1y5ciX5Ad/luLg4Ro0atU+v73Q6WbduHbW1tYSHh9O9e/cWCx0bGhrIzc3FZDLRsWPHIKO3pxyoeeZAYuxMDgGUwwF33e1NtVy1CsrK4WVvwJjNAZpXu3ZJnv+Vk0JeVwsLQ914Pbz6OuRulZ1KTU1wVbUv27cHZ/AUNasOt28vxkbvtUwmGHRk6OsCatp0CaC7l46rVqHeeENW/yGuqR57VPSsSnfJZHfLTfoXf+B+eOxxyViyWkWq/+dfxM1nsQSvupuaZDfgy5df+E3AGhC5ZQtr1671TGIeLrtUzl2yRHZtRx0lHTFfegUKP4aICJTDKUa1baa/MbFaPZImHrbp9IRpssO0GbDAR4tMrzdK587ScyYtDU7+N5FpaTR0aA933qUvHR+IyQzDh8PFFx0ShgTQjWlYdBYoUTqtp/dHPGT58uUi3NlMTU0NI0eOxKSTdFFRUcGyZctoaP5d7dixg8GDBxN9mMWVDhRGNtdfgHI6UTU1sGx5cHfEjRtRvru2QKVYs1mUZneDNnIkPPQAhIV7XRst9RCPjxc3lS8+QVtls6G++x71228ohwP1/geoO/6DeuyJlmU8QArdfH0Qhdvhgw9Cn4800tIefgjtjdfQpk7xpscGnhcfDzffLGNfvx7efkfqajp1gjtuCzZ0Tic89BBq/XqU3Y668y7d+hxHWJhnkvB7PZMJ7eab0N57F+3tmSI0+fMvkhzhcIihmdZcvX7H7VJ9npAgcanTTwWfJAO1ahWUBdT7WCyyMynQ2UH6TpYREXDcsWjXXoN25hlozZ+dNnCgGJhAzGb/TD2TWTK8nnsGbeCA4PPdY6ytRT33POrue1AffODN5DtAdOnSxW83GB0dTVedwtPArKyIiAhycnL26bWbmpr86qEAqqurQ3pRtmzZ4vcdqaurY3OI5Ix/IsbOZD+jbDYR8Nu0WSYWqxU+/FCaXiUmyN++E214uH9a72mnieZWRaX4xAcO8JuQPK/jcEhh4voN0L8fnHiC+OxbcnlomhiaxAQ4+yysO4po+uVXCdK2bw+XSQc/tXUr3HWPSGpYLGJ4Kiu9K/7Vq1HT3tTv+w6h/fR7gW7A8IUXxA3npr5eVIw/+5/s8q6/wd9llrcNXn9TUqhnzQq6XENMDNsGDfIEflvERy15Vuo4+Z8NEhDXkpLglZdF4iQiIjgT6pdZEkfxxWJB69cXtXSJGClf0tIkbbeyEo4/Hu2iC/XHdM7ZMGcO2Jq/V1Yr3HAddO8hqeEOJ7NOWIQ2ZkyLt6acTrjxZtHyco83N8/b/fEAEB0dzYgRI9i2bRtKKbKzs0PuVoYNG0ZBQQE2m42srKwW3VFKKbZv305dXR0dOnTQPVcvxqdpmu6uBPTTzB0tLdj+YRjGZD+ili0Tl4N7tfPzz/4S6WVlEkx3GxOzGYYNFc2tZrTTT0P16ysVyRs3QlW1KOyefiqMHQsz3xZXyZat3ontm29EXuW2W0XkMDdXf4CdOkn3xrQ0OO5YYjMzKV+0SHqRDB8m4ooAr7zqFYN0OIJdLhs2iIR6qG6EI0fApk3e+87MhHPPad2b2BpKQsTfNm2SuFFOTrCycUWFrkGrT0hg4QXnY+mYQ18f1WXlcsH0GZI+HBkJEy5G691Lt1C0ISOTzStX4nA4yMzMJN2nsNCP7KzgYw6HGJ8JE2QBsmixLAiUEtfd9u3y/x99hLJY0CZcHHyN/37iNSQg7/uq1Wjnnrtnja3mz/dP/3a5YOEiVGPjbhut7QtWq5XOPjGtUJjNZrJbsUt3uVwsXLiQsubvbX5+Pr169Qry9VssFpKTk9nuo+qQkJBAmxDFwElJSezatcuTBGA2m42CSB8MY7I/ef3NYCn0wF4bFjNcdpnsIvr2ka5+AWgdO0pQ3VdtNy9XJo2Vq4Jf1+EQ0cdrr4X/3AGvvd6cMaQkO6i2VqQ1MjKk/0VVFbz9Dva770Yb2B+6dPG/XuA9BKJUcLMpXy6/TDK4FiyEyCiYcBFaamrL1wzB7Nkyifm1nW3XDpYG15o4wsKwmc3EDB0ixs53jDnZuqKR1lFH0f+Uk4mPj/dfkb74kuww3avRDRtQb74u7X63bIb1G5htG01Txxys/TtR1xwc3rlzJ927d9d3wXTvHnzM6YTly9HGjYNHH0HdfCv85tMl0m0AKyqlYPGcs4PViet1YmMBWWi672MgDTb9YtAD7Ora3xQUFHgMCUBjYyO5ublkZGQE7Ub69etHbGwslZWVREZG0rVr15BZie3ataO2tpaamhpMJhPp6em0a9dO99x/IoYx2Z8ENl/SIzEJLr909yqsq9f4/11RKf9C0dAA9XVoAwbAa696Dqv6ejEsFZVww43eSSY/n4aXX0a9+Xrwj2d3q62OHaXhVAg0TZN2uOe0bjeimvuS+L4ndrudrVu3ctddgwkLszJnjs8keNMNsLMYli7D5XBgUgoFlGZns7qwkIyMDFJGjyJ57VosDoe4e66/Du6+x+vqA+jaFetJJ5LwxluQ0kYmavfubPFi/4l1xw746GO066+THiN//MH9k0Zg22zhnoYvPac5nU6Kior0jUlmJqSlSsqv5+YVvPQKatAgiQfpCDF6qKyUosX0AGMybKjsKtzG02oNkkBx1xmMGdNCVuXoUbJ79Y0D9OopxbGHEbU6rt6Kigq+++47wsPD6dGjh2eXomlaq3ZFa9asoaioiMbGRmJjY+nRowfJu+v98w/DMCb7k/btgl1MYWHy466rExfJ2We1Ts57T3/AnTt79a580KKioEMH1Kxfg1arzi1b4IorUeHhIi54bHM72cCe8yD1JB1zJH5y5ZUhe3fsKeqVV2HWr9BoQ/XoAfdNoVHTWLBgAbW1tTQ19cPhcFBSUkJq8+5Gi4mBF57HtWMH62e+TczWrVRlpFMwcCDKbmf7unUU9e1Dr6IdJNbVEQXisgrsq6Jp/m7JX2ejXntFJk+TTodBd9DbbKZh+Qrqw8bgtAT7991uEGWzycTcrp0YiclT/A2Jm8JCuPEm1AvPt9zsqn17MUaBnH6a3MO8eZLnPGK4xFH2EC0sDPX4Y+LmrCiHnI4SeznMaNeuHYWFhTQFeAVcLhcNDQ2sXLmSlJSUVmeDlZeXk5+f74mZ1NTUsH79ekaMGLHfx344YxiT/cl/7hQX0tp1suJskyy9I9IzJKg5alTI9NggTjgBfEUWWyImBq69JmiHoex2qWmYPUeKFWNi/AP0dXXShwQkqywtDa1PH381Xjfp6WgvPO+99o4d8OY0SUEeOgROO7VFZVw91KzZ0hzKHUPaUQwxMWw+7VS/1aVSLvLy8jzGxI2Wns7OQUeS27uX33GX2cygDz+mjTsNNy9P4lOBFBT4G9h16+DY41EXXgBjxsDWLd5YRFaWX9xnrdWCU28ycjpJ3rJFEiRefFniWykpkllV0ILg45q18PAj6FbiJcRDVjbceovnPVbvvU/V7NkoWyMMGgjXXYd22aV+T1N/zIeZM2VXuvP9loVBm9Hat4OHH9zteYcy8fHxdO3alYKCAmpra3EFuOkcDgfFxcW0D5H+HkhpaWlQ8N1ms6GU2uPv/N8Zw5jsR7SUFHjzDVRlJURG+gctO+sICbZE3e4L0TzU1sqk3BxsVS6XaHN99rm/9HxsrLiw6urEhdPQgELqLKislKB/nz4yac6ZK+mvICvyo7wyKaqmRlxm7t7ks2dL8DtE06WQ/DYvaPKsW7WKpvEnBZ2qlzWjaRqJFgv1SvkZQGtDA3EBhXB6qr+6BqaxUXYxd9wGd/1HakpiYqS3iU+6ckNM8O4trKaGrMVL6LJwkUzcbpdVaan+awWydBmkp8FWn92tpsHZ56BdcZnnkPrlF3jtdZzuNNWNGyEyCnXuObKrKN4pNS9z5np11+rqJIbG4dnTvba2lrVr19LQ0EBkZCS9evXy1HcopVi/fj27du1C0zQyMzPp2LEjOTk5LFu2jCId12GcT9LL7khNTSU3N9fvOxgZGWkYkgAMY3IA2NOCMLVpk4gb9u8vWVWapq/02xLN9Qtq61b4vykywQRSUwOjjoIrJ1H0xptsTUulMTqa6PIKen/zDTG5eajrbhAX2yUTJW5ga4Thw/wlQT7+xGtIQHZhX3yJuuJySEjw9Dp3426jGhMTIxXDpaVSi6FTSVtrsRAWFhakjabXj0IpRWKbNuwqLKTRZ6KPrKnBGcqVaLFIdXm3rhK/+vbb4HNcLkkeuPxycLpk5xUZiVq6DLp0RouLIyLQ3aQUnX6fT8eFzYWHgRlwoRqD+VJREZwkoBRMn44aNBCtXz/J/Hr5Ff+CVJcLli2TTDC9pmlumuwom7bfXJR/FUopli9fTlVzYWZNTQ2NjY2MHCkLnGXLlrHDR6y0pqaGiIgIj1EpKSnxMwSJiYnEh6hj0iMxMZEOHTpQVFSE3W4nJiaGHj6ZfwaCYUwOMmr6dHjnPZnoP/hQ0n8fvB/OPEMk332qcz2dBZOSglNd3RPzs8/pGxI3TQ6akpNZP6A/7unIlpDAytNOY/iMmX61JAwfJt0Px5/kvwpzNAVdlqYmOONMSEmVnu6TrgCguLiYdevWiYqryUzbNWvp8f33cg+nnExVTjaxedswKUVtUhKbjzqKOKeT7t27U1RURGZmA1ZrGN26SfMsp9NJbm4utbW1nlhK+tZcGuPjsEdFEW6zkWSzU3VEbyIWLPRvrwvQrRs8+zRaQoK4AVetCnI/Oc1mzGvXwcUTZKLWNIl72e2Qno46/zx6nnIymen1NDkcWBobSd66lezFzf1PTCa5v5bUmEOhVzDX1ASvvIZ66gm45jr9/igOh7hXQ5BlzpfMOmtwB81DnZqaGqp9Wxg0H6utrWXjxo1+hgTkO1JcXExmZiYJCQkMHTqUTZs2UVdXR7t27eik125gN/Ts2ZPOnTtjs9mIjY01diU6GMbkIKLsdvj6W+8uxG6H335DbdwohiMnW4T8YmJFzbasOVssOgqO6C2S4TabrLJvu00e0wvwuomJgRP+RXFxMQ0BD0VWlONyubySCLt2wZdfSa/1H39CvfCct4Ph2WdJvUtg2nNtHdTmwjvvonr2gBEjPD9igEaXk/ys9rSzWojNy4N33yPvphtx/Pkn4XW1bO/TB2dkJNlJSbRr146cnBy+/BKgCdBQSrHyq69I/uprUhsbWfXv8biioynq3w9rfT0JhYW0GTWKTkceiVKKqC+/ov7Z57xxougoOOFfnp2jFhZGzZOPs+v1N0hZvISo6mqq09JY9e/xHPnRR0S4fe2+qdDFxfDhh0T+ezxffKVRVVWPJb8L0Z98KufFxMDQoVKD8+KLUvS3p6m1bg0uX8rL4KOPvPU/vqSmwnHHBtfWWK3NKsaK6Vk3SK2MecKejeUQwGq1YrFY/ALqFouF+vp6Skr0v+++kiwJCQkMGjRon8cRFha2z1pcf2cMY3IwqasLdmfV1TVrT+VBdfNjVdX+sYWCQujUGT54D2w2/xqOtFT9CcdshmPHoY0cQVxlpZ8bydLQQLeffsGkV6XudqH8MksmLEBLTEQdfzx8/bX+fdlsMHsuzqFDg7ouOqKiKMvpSGxZOZSV0b2sjCUjhlNaU4PFYiEjNZW2bdt6zlcVFTKJmsyUDx5E51deI6Y5BXvT6Bqamv3mTVFRlHbtSkJzJpqmaURMnED9sKEwY6ZIvo87Bm20Vxhw3bp15Ofn0zRyBNrgwcSWllKblorLYsEeGUVETYi4VdEOyC9A69lDJD4SElBvz5SECVuj1InMnSs6YgsWwvMv6F9Hj4QE0f/6+mv/nWdGpqgB65GdDWeeCd//6K1eB9lVjh0tWnDHH4fWu3frx3EIERkZSVpaGtu3b/cEvdPS0mhqatKtSo+KiqJLYO2UwQHHMCYHES0xEZXVwT8dVNOCCxP1MnyqqqRyPjCQeOMNcr2Nm/xXxE4nrF2HUoqEhATS09MpLi7G6XTSbsWfRLUUo3E6gwUKJ98r6cjLlop7JbDHfPu2nm53vm1zrbW1pGza5Pk7/I/5jDj/POrq6rBarX6SJmrTJmZMWAyVLiZGTyf+ww8x+0zwKVu3UJOa4gluR0VFBVVIa6mpopkVgM1mo6CgwLPaVWFWqtt6Yzi2mBjiQqx6ycyAnGxmzJCalIkT69GsVlREBEx9wJu48PsfcOstzVLxLexOLBbZhUZFicrw+PGo+DiRV6mrE2Nx+22y0/jmG5Hm8WXjRnn/n34SXnpZdqc52dKhMzwcxksPDpbg6aVxOFFRUYGmaSQnJ2O1WklPTyczMxOn00lUVBT1Phl5ERERjBw50thBHAQMY3Kwmfx/shMpKBA3VqgmDb7FdgBd9AuttE6dUDOmi8z9++/7P1hRLu6a8HD69etHY2Mj+fn5tHv9zZbHmJISVKmvaRpccB5ccJ70ib//QamXMJulsv+cc9A0je7du7NmzRqJmTgctF+0iGhfhdt166BoBzFtg4PxTJ/J29ulk+TE6Jl+hgSg+8+zUGHh1J54AhaLhc6dOwfpa9XW1qJpWpCya21tLfYWqvjLjx5L6rZt+oHz+gbYsJGZM0WbyzNBf/CR15CA/P9330lmVyjDBJCSgvb+ewCo3DzUlPskwD7hIhgxQkQj6+tFL82m8/2oroabbxUXXggdLXczp8PNmOTn57N48WLPZxUVFUWPHj3QNM0jh2Kz2XC5XFgsFnr37m0YkoPEbo3Ju+++y8aNG0lJSeHqq6/280X+/vvv/PzzzyilOP/883XVPg30UUpJEd2SJXDTjaInddrpUmsRSEQ49O0rK84mO/TuDTdcH/LamtmMOu9c+OEH/6yitu086cqappGRkUH4rjL9oG94uKzA4+Lh3HNalEPR+vdHvT0DvvlWhAnHjPHEV9q0acOoUaPEmJSUYnniKf8n22xQG2JXFNirHHBFRGBqdp0ps5luKSlYAqq9QVKJf/rpJ48CbGJiIkceeaTn+5uYmEh0dLQnngNgMpmIiIggMTGRriecICm2eq68ykqp32Cc/3E946QQV9PnX4SWoDlWrqNKS+GWW0RlGaSqvcEmSQNT7gve/XneFJfopeXno9LT0I4JFgY9FKmqqmLLli04HA5SUlLIzs4OCmxv2rTJz+jX19ezcuVK+vfvj8PhYOfOnZ46EofDwZYtW0hPT8fpdLJhwwZqamqIjIyke/fuhpE5wLRoTPLy8igvL+f+++/ns88+Y8GCBZ50vPLychYvXszkyZONzIa94cEHRT/K1iiaW+efJxpZvsYkLAx69YJjj0E76ywxQC6XNxDeAlp6GuqyS+HTT0WCo11b0e0K5Oef9aXqhw9He/zRVt+OFhsbUsxR0zRiYmJQ0dEyMboLJQG6dgnWBnMzcAB84bMjs1oxnX+eTLb1dZiOHATnn6v71PXr1/v1qdi1axcbNmygVy8pcDSbzXTv3p1NmzbR2NhIREQEPXr08BP5U11bkNnYuFHG5fvdHzta0qndabtRUSKseeIJcOSRsGCRNCz7b7M+mskEA/rDFZdLkePb73oNCch3Y9p02e21pl9JQwPMmg2HgDFRSlFcXExFRYUnq8qXhoYGli5d6nFRlZaWYrfbPVl7bvRiIrt27WLevHnExcUF7S7r6+txOp0sW7bMr+lWTU0Nw4cPN+aqA0iLxmTjxo307SuFcP369ePXX3/1GJMVK1ZgtVp58MEHSUhI4IorriDiMMtf3xOUyyUCisuWyar99NPRjh67d9fK2wa/+kiG19RIvcNrr4oBycsT2ZILL0Qb6ZVs0DQNzGZUUZFIkPTtg3bEESFfRzvrTNRpp0JDg0z2ehzRO7hbYXgYXH3lXt1bKJTLJYFws1myj+LioEN7uOnGoLoUDxddCE9ZxK3UrZsUTk66olUTQp3OrqY+QE4mIyOD9PR0HA4HFosl+LpnnAG//S47yMCYR0kp1JZIkWEz2vjxqIYG+OVXOXDMWDS3svKYMajRo+Gqa7yGweWCNilSNzJ3rv6utBVdFP2Iid79OX8By5Yto7i4GKUU+fn5dOrUibS0NJRSxMXFkZub6/d5KKUoKSkJMiYZGRmUlJQQ2BDWZrPpFrJGRETQ1NQU1L2wsrKSiooK3Xolg/1Di8aktrbWs6KIioryk7ioqqqipqaGe++9lx9//JHvv/+eU0891fP4/PnzmT9/PgBXXnkliYmJ+3/0BxCr1eo35vonn8Y2822PD13bvJnYbt2wBEh5tAb7smXUBga8y8oxP/gQWmMjpsFDiL7nLl3Zb9vMt2l45VXUrl0QHY31hH8R8+gjLb6eY8MGnGvWEHbUUaJr5XuPxxxD3fHHY//5J6irR0tMJOru/xA+IHQDpb2h7sGHaXznHc/7Z4qKJPbRR2j66muafv8DLSaaiKuuwtLVf5diTpH3IOmbr1q8vsvlYu3atVRUVBAfH09iYmJQk6OkpCQSExMpLi4mLy+P6OhoevToodvZz416712a5s2j4fU3cS5c6B+3arRhcjgIe+11VFUV4aefhnXSJJik3xWzafFSalav9j/4xx9eRYK9wV17BJg6diT2ppswB/zWnDuK0SoqwWwmISLigAs3lpWVUVpa6jEADoeDzZs3s6k58SI5OVl3UrdYLEHzREpKCkoptm3bFtTIKtCYxMbG0rdvXxISEoJ6kmiaRlxcHImJiZ46lNjY2D2qhD8QBM4zhzMtGpPo6GhPZ7H6+vqgjmi9evVC0zSOOOIIPv30U7/nDhs2jGHDhgF/jx7w6o/f/X7wqqyc6pkzQwY8W0L16CHif4WF3oP19Tj/EOPL0mXYy8vQHrjf/3kOB7z7ntd3XleH/fsfKD/jdL8e557zlYL7pkrnw9pa6tq1g1tvlq6MzfdYWVmJOnqM1E/U16Ouv5b6/v2p3899qdVvv/m9f668bVRddY0UDTZnVNmXLYe33pAmU82MGCE/9ooK/6K1QJYuXepXvNamTRvPqlbTNBISEsjKymLp0qVs3LjRMxFt27aNYcOGhWyIBMARR6BuuA4uWe6XIDEq8neoj6Dxzbdk/N//IDutk/+t/x6U7wo2GntqSHxqR4iMhH8dL+608HBc551HdWSEX3agWrAAHnyYo+qvAqDirA/h5ZdC71T3Azt27Aia6H3dVaWlpVitVr9MLJPJRJs2bYJ2FImJibRv357U1FR+++033W6Y7uf36dOHuLg43XNiYmIwm81s2rSJVatWUVdXh8ViIT09nb59+7Zqt2uz2dixYwfx8fH7bYdzOPaAD9WUrEVj0q1bN77++mtGjx7NihUr/Lag3bp140upKCMvL+/v3yRGL3in05e6NWiRkaiuXf2NSSCr1wQLydXWBfvOa2ul26KejPbCRVIf4p4ACwul1/hIH52t77+HJ56SjCCAyfehnn8OLSd7r+4tJCadH2tBgX/hY2GhZERde7Xn0NSpLRsRkB/5roDgdEVFBePHj6eqqsovm6uwsNBvoquoqKCgoICsLJ3GVc04nU4Kw8KwHHM06d9976nHua/zy/6S8TU1ErAPYUwYOFAyu3x1w/akoDEnB6a9KbUrK/4k9qQTqN2drMf7H8DOndwXP1X+Xg+88y5cc3WLT9sX0tLSglJ2A7Hb7Rx55JGeAHxqamrIz2DTpk0UFBTQ2NjoMfqB4o0ul4tVq1YxePBgysrK/NLRQQoONU1jw4YNHheow+FoVlnIDBIRDaSgoID169fT2NjoaYo1YD/v3g93WjQm2dnZxMfHM3nyZNq0acPJJ5/M66+/zqRJk8jKyiI5OZn77rsPq9XK9deHzi76W/Dvf4uYYVXz5JaVBRdftMeXUY2N0m43L6/lE/UWSvFxEkj37ZuSkeEnwgiI337RYvhjfnCqsU92l6uqCj793GtIQHYo770P997duhtqLSNGSDW423jk5KCsVvJysinLziKsvoEuc+YSuRfxUZfLFdSr3GW3U5iXx67KSuLi4oiMjMRkMukGdGtqali9ejVNTU2eOIqb6upqli1bJi7eQUeS0LMHgxcswpqRLjuEz//nfzGf1F214k/J+qqphU4d5TsTuChRSmJwgZ+T1SoSOQUFck5sDFx/rbgpxx0D447BmpjYsmQ9yGsHEtiHfj/jTtF1JzdAcLwqKiqKuLg4+vfvj1KKiooKampqiA3YMVVVVbF161a/6ve4uDgsFgvlAf2Dqqur+f333z2uMV/c4wg0Mi6Xi02bNrFu3TqcTicJCQn07dsXc3OSS25uLtu3b6e6utpjwNxusvLyciMG48NuU4Mvush/wpzk4w8+//zz9/+IDlG0E09AZWbAV19BdAxcdJGfO6Y1KJsNrrkWVq3e/ckNtqBsIU3TUHffJXUp24sgKREuvMBPzVb99hs8/axMQtHR/m2DAdq1lR/ak09RNXeefv3Dgehrfc7Z8N33koZstcKRA1nXri15UVGo5phFZU42w084AV9h9ylTxM3V0g4lKiqK+KoqynzjQXV1LF++AlfzjqikpIRBgwYRHx/vF5wPDw+ntLTUc6y4uJhu3brRsWNHQJJQfGOFldHRbL7iMnr27MmUmxQ0Dua+8GbDq2mQk4V6+hlITpKe9O6di28GWyA52bK79KVDBzEwbiNZXQPvfSDV8XtCjx6wejX3VU0B4L7MJ2HcsXt2jQCUUuzYsQO73U67du10Y06pqame1b47u6q8vFzEORMT6dmzJyBZXUuWLKG6uhqTyURycjJHHnmkZwei15eksbGRzMxMampqgh5raGggJiZGfis+BsXtmgncMZlMJnH1Np9bX1+PxWKhT58+lJSUsGHDBt1Av8vlorKy0jAmPhhFi3uA1q8f9Ou39xf46KNgQxIVJa4P9wrUTVUVrFuP6tbVr5mW1qkTvP6a6HqtWAEffIj6/AsY0E9Ubt+c5hUurKuT6urkZHGHRUXB+JPg999F5VevQNJshs2bUB99jBbQYEk1NMj17HaZHDt2QkvfvXtTNTaKaKK7d7vTCZ9+Rtj4k1D9+3nOq0lKoqCmho4pKZ5jc+eG7j1eW1vL5s2baWpqos327YQ32LAlxBNeU0t5u7YeQwLip6+YNo1+i5bQ0dFE3vDh2Hr2ICwszE+i3N0p0W1M9AobGxoaKCoq4tdlvTFH/BsGfS7ZcJFR8Pv81qXxusnMFKn7hx8R2ROQ3UtsDGzZ4n9uXh6qpmbP4h033wj2RubOPEaM3aQqtGFDAVC//Q7Tp8tON1NSx7Xd9PhwOBwsWrTIsyvIzc2lX79+LQaRzWYzgwYNoqGhQTTTfNzDa9eu9agBO51OSkpK2Lp1q6f7YXp6OhaLxW9CdzqdrF+/PuTrpaWlUVVV5VELjouL8xivXr168eeff1JbW+vR/KoJSIZxi0oWFRXpGhKQRUiGTjO6fzKGMfkrKd4ZfKx9O3j2GTj/Qn+XhVJw8y3ScyQmBo45Gm68Ac3dkGnnTph6v3eCdutCBUqfOxwyeTc2yr8nnoLu3fTdKk6n/NuwEfJfRkVGoZ08XvqX/N9k2LRZVsqNjRIfSExEnXgC2k03tnzfX3/jHacbl4uETZvAx5jIYf0YQl1dHXl5eVgsFk9L3MWLF3t2FCVdutBx3m/0/+JLGqOimBOQ2uxyOrHNmk3Chg3EA32Ld8K/jmedTr973zHExcUFuVPKysrYsWMH9fWdsFjMqBeex2QyoSZc0npDkpkJI0fABeejZWaiXnpRUqd/mSW7txU6O5mYGEnj3gM0qxXuvQd+EQOtnX8eAKquDp5+WnTeQOpb7n8A3ni9xett2bLF7/2oq6tj06ZNDB48eLdjce8OysvLKSwsJDIyUjdY7ju5Jycn065dO4/8e6A8TyBms5nw8HD69+9PQ0MDdruduLg4T+wxNjaWESNG0NDQQFhYGFu2bAkyJu7iRr1OjO6aqezs7JCB6H8qhjH5KznpJClU9I1RdOmK1qYNathQ+P4Hr1vD4fAaF5sNPvoYli8XwxIeITsI3wna5ZJ6iMwMf9n6sDBvW1oQI1BSGqxMGx3tf15DA8yZDSePh1tvg+Urgu+nogK++hp1ysloej3P3QSIPbqxBPSUiImJ0Q3COhxO5s+f7xGN3LFjBykpKX7uKmUyUdqnD93/mE+4y0WcrdHP7RVTXU2Kb2/znTspfPoZSsafFLTy9e3t3aNHD+x2u0cfyuVy+YhXKhwOB3l5ebKT0alt0cVkElWB87wFl1p8PKp7d8nWC/WcTp28i4l95bc/vIbEzfr1qIoKtBZ2GXqTf0uyNIHk5eWxfv16z/ut5yILzKzq3bs3HTt2pK6uDpPJxIIFC0Je3+l0snTpUoYMGUJkZKTuhK9pmmd31KlTJ0pLSz1px1FRUR6J+i5durBr1y6PsQkPD+eII44gLS3NKH7UwTAmfyFa716oa66WoG1+vhiAFSukD/qUyeJCW7MG/lypH6Df6BVI1F2hWq1w263w0CPi6oqPl+LANWv8z4uKFAmPP+aLccnOlgk/cJXeaEc99LCMJxTV1bBshWQaheKUk8UY+vadiIoi4cYb6OpyUVFRgdVqpWvXrrqrwaYmu5/6cG1trSdA6kdWB3jjNbBY6Ne+PRs3bqSyspLw8HC6/vIr5oDge71SHu0ud1A3KSnJL2vRbDbTv2tXXI89Dlu38vsJ/8IWUM3tial06eIviJmYKOm7gZ3+OnTQVwv47bfgY25cLli8GLVgIdrQYPmYPSa7Q7AAZaMddpXJuEOQkpJCUVGR3+4tOjqaP//8k5qaGqxWK506dfJTEvAlMJtOz40UqDQNMslHRUWhlCI2Njaov4kvFRUV1NfX+7nTQmGxWBg+fDiFhYU0NjaSlZXl2ZmEhYUxfPhwT5fFrKysII03Ay+GMfmL0c44HbV4sWgpgaTDzpgJP/4oro9LL5Xg++6yvWw2/57u0VFw/HGy63BXlx95pMjGX3+DtxeKpkH/fmg330xsdQ3VGzbCyy8Fy9bHxEiDrIULaYiLw2y3E6a3w0hOhhHDW77nmBjsTz6O+dXXMW3eLDGixx5BS0mhNWpugZk5IC6TpqYmTzBV0zRSUlLQmlNlI4ExY8Z4cvjV1lxYutQjdVLTJpm85klZKUVMTExQqqfT6ZRA8AMPYpolVe3RvXtTHWBMKisrKS8vJ/Heu8FsEndgbCycdaYoGwQak44d9Ve2LRlkEMP/zTfS+TEEyumEzz6DteuhV0847VR9+Z327YONiVLyHvmIiCqHA7V+PU0JCYS1bUtmZibV1dUUFxfjcrmIj4/H5XKxfbtXBqa2tpajjjpKVwsrVAzCl1CuTpDPecCAAaxdu5bGxkasVitlZWW63xG5JYXL5WLHjh3s2rWL6OhoOnbs6LcYMZlMdOjQAaUUTU1N0tenOQHAvcgx2D2GMTkY5Of7/+1yic+6cLtIajzysKQh5wb0Ag/8wRw7TtxVzb06aN8Orr3eW9To7rx3553wySdigPr0geuuBcCS1QHmzQvu0BcbC8nJ2EpLWT7hYqrTUrHYm0jZvJkjvv5Gspajo6Wb4BmntxiEr62tZeXKldTW1hJ+yr/p0KGDJ+axO3Jzcxk5sjjIjWI2m2nXrh1RUVFs3rwZh8NBcnKyJ2iuh3b+eaj0dJj1Kzsb6lk1cABNPi4Q3x1RQ0MDK1asoLa2lrCwMDqYTGQ3P9b72+9oioykrGMOo0dLM6rq6mpWrFgh0ucPPuD3umpXGSxe4s2oi4iA0aNQ334nrsPePWH8eJnwL7xA6kJairuEhU5IAODeyfDLL/Jd+fZbSdJ48AEmTAhwwb37fnDWXmysNPZyj33NGhon34e1oAAVFUVxnyOIun8qPXr0oHv37iilMJlMzJo1y+8yDQ0NbNu2TbenSHx8vF92nNVqxel0+hkQXzejHjExMX4xmgULFvjVGSUlJREZGcmaNWsoKSnBZrP5pYTv2rWLwYMHs3PnTmw2G+3bt6e+vp5Vq1ZRX19PWFgYWVlZrf6eGgiGMTkYJLaQTlhQAGvXwDsz4YsvYf4CmbijokSU0R0s7NBBdKp8hQkfe9xfWbbZNaJde40IDupRURFcOBcXC3Y7a044nvJsiWE4IiMp7NuHxG35tC8qgrfehOSk3fYTX7VqlSdga7fb2bRpk6eorSWKi4tZv349Q4Z4J4GwsDCioqJo27atpxZkTwrHtKPHwtFjCSsvR1u2zBPLiYmJ8WQPAaxcuZKy5kSGxsZGNg4cQJtly4kpLyfMZmPwBx/y6//dw9ix3s6G9fX15Ofn+10HECHK+jpxKZpMcPRYWf1/860kO/zvfzBnHjz9JFpYGOqF5+C5F7yf4/bt/pP+ksWon35COzY4vVcVF8OiRd5Fh8sFCxehSkqYONFblKecTvjpp+A3aOgQtGxvzMr53POENy98wuvqaLN4CRs+/ZT4yy5D0zTP7krP5RjYCsBNnz59POm4drudiIgIIiMjsdvtKKVITk7eo52Ae3dUV1eHUoqUlBR69+5Nfn4+eXl5ujuWsrIyZs+e7Yn/5OXlYTabPbGR3X1P8/Ly2LFjB5qmkZmZSYcOh18r5AOBYUz2E0opaWrV0ABHDvRL5w3iskskSB64QwGJe6SliS7X2WfJP/drjDta+sJHRUkL1kC/tN4EvbtA4cn/hk8/86/G79ULbDYa4hP879FioTy7A+1NJjS9/iMBuFyuIMFFu91OUVFR8KQbwI4dO4IKDGNjYz0SPftCUlISI0aM8EwiOTk5fjuT2gBxxaaICLYPHEi35glY69xZJtCA8YWZzSin08+tpGkaXHG5/ANUeTmcc57/c3/7DfXzL2jjjhE33asvowoKpSYpcPdQtIP6Z54jPyWFoaNH+z9WWxec7NDQANXVqIgIaaYG8to2HVmS/v09/6scjqDaF4vDQczqNTidTj8DkpmZyebNmz2fV2JiIu3atQu+PmJ4+vTpw++//05jYyONjY1UVVXRtm1b+vu8fmtZuXIlhT7fXbfiQUuuL/BPJKivrw9yO9rtdnbu3Bm0OykoKPAUOILEZ8xms1930H8qhjHZDyibDW65VYxJUxN07YJ6/DG0EHno2oABqHdmwqxZkoE161f50WuaSJKHmDC1wYOhpRTMCy+AOXP9g8BbtqLuuBMefABNx4etxcWhpk6B6TPFvdKlM9x8EzQ2Ev7Z5/4nu1zEV1XDqafs7i2Ra2sa4eHhfgFVs9kcJEeuhzvL56WXjgfg2mt/CFoB707m3I3T6UTTND/9rcjISHqEkCIJCwvzm2xMJhMJJ48XY710KeTlkfHrbG5eeT9KM3Hd1d8x4MefSHtrGmgaatAguPMONJNJUnAfeFDSraMiYdjw4FbNSsHH/xVXpZuPPvLPyvMhYtcuShcvZlFkpCdZQFVUwLy5kJriL2OflAR33cMlqx6AyCqmfwla+/aoLl38U9XbtYMT/uX9+9XXMAVkbjnNZiqPOIKsAB2zLl26EB8f71mtu6vKO3bsqJtQUVJSEiTaWF5e7lFvbi0ul8uzg3RTXV1NUVHRHiuYBxY5hvqeuruTunHXJRnGxDAm+4cZM8Qv7mb9BnjhRXj4oZBP0aKiYPx4GD8edcK/4NfZkNMRTj/Vs0pSTU0SOE+IB6t1t31MtMRE1Msvwo03gzsNtqFBrv3WNLj6qqDnqPp6SUl2NEHHHLjmarSICNRvv9Pjt9+xDRtCdWoqFqeT5IYGOlxxOVr37q16WzRNIycnh/Xr12Oz2TCbzSQnJ5OXl8fGjRuJiYmhV69eum6Szp07N6vPSoFeeHh40CoxUOY8ISGBqKgo0tPTSU1NpampicWLF1NVVYXZbCYjI4PurRh7p06dWLt2rWfMKSkppA0cKJpYzavgrr/OprpKwxkdxaBVq2mzaDGaezLaXiQyNxMniFrBLz4xhaIdEjcJTCMOfA9a2FE2xMdTn5jokWxX8+fDw4+K8TGZJIsvMxPsjbA1F5RiW0MGNACP/EcSLqbeB48+LokeCQlw6UQ/RWlWrwlS9LElJtLmuGN1kwdSU1Ox2+2sXbvWE+PauXMnw4YNCzIoejuGlnYRe0OXLl0oLy/3GC2LxeKpUdFLZU5ISKChoYGGhgYsFgsZGRm6hZh6gqAtioT+gzCMyf6gYHvwscDiQZp7ehQWQlKS3w9XGzIEAroFqs8/F42s7c2ZQMnJqPEnoV3Vcp8RLTUV1SbZa0zcbM3Vf8Jdd4s/301eHurll+CNN4jZmsuIVauoaNeOsOhoYl96wesqaSXt2rUjNTWVHTt2EBMTw+rVqz2+6fLychobGxk0aFDQ8yIjIxk+fDhhYWEoBTk5OZ5AqtlsprKy0q/PhcPh8ARht2/fTqdOnTyd+DxvwdatxFVUkPHV12BvgnFHo/3rX0GvnZ6ezq5duygpKSEsLIyMjAxZuRbt8DvP2tiI1WQmpaTEPznC5YLVq+T/A6vYa2uhX1/ZxbpjVZomx3w5/3xRe27OBGsKC8Nqt9MQF0ve4EE4IiK8Rvid97y7GJdLdpjjx0tjtMBJuqAQ5XJJFf1D/skCfkQHu0wjxowO6b4CabHrO1FXV1eTm5tL586dPe0BLBYLWVlZxMfHeyrfQVxje7IrAa/ScIFb8QEpMs3MzMRkMjF8+HCPgnHbtm2xWCxBwXqQRc/gwYNxuVyUlJQQFxcXUpo+Ozvb870FWeS0lPjxT8IwJvuD3j0lOO4byA6QpVDr1olkxrZ8SEhAnXgCXDkJXn9DJg2nEwYdKS6m6hp4a7q/uuzOnfDOu6jevTwS8iHRU3DODHa5uXYUw5q1/gfXrZd6h+ZGTSalSHb/WNetCzJ6rcGdHVNUVBRUbVxVVRXSvREWFobVKi4nt3xGbm4ugwYNoq6uTle0EcT1sH379iBXh8vlomTeb2T89LOct2gRm1b8SeHAAZ6q5oyMDNauXUt+czzLZrOxZs0aYmJiiO3aBe3XX/1fzGqFBJ26jITmJAu9SSkxEa65SppoOZ1w5ABvTOWnn6QhV2oKPDBVunBu3oIWHs6qHt0pzs7CHh2N2WwmOztbrletk/21YIF+sWhSEuzYgZo2XWIsw4ehnXJy8HkXXSQpzu7aoM6dMTePMRR6ab+NjY2sWbOGbT6u15qaGvr160dRURGNjY3ExsYGNcVqLX369CEyMtJbT9S1q2enYDKZgtxP8fHxQcYkPj7e8/1ryViCZJr169ePNWskdpSent4qt+0/AcOY7A/OPlsKChctloBp9+5iFHx56hlvQLOhAabP8MY33Gmjubky+eTk+BsSN3a7PGd3xuS660Shd906qXvo3VsMVyBmk7hFfDGZIDVNDJJvrUtqKnQL7SJSlVUiv56R7tcD3hd3N0Nfl4ZvVlDQNZWisdGGy+U1GtXV1WzcuJFevXq1KHPudDp1M4qit3trPswNDSQvW86W3r1obGz09AsP7C9ht9tZuHAh5uwsIm6+kS6zfiWqugZqo0XQ8aorpY2vezfYtYv3/e7cSXYhvq6VOXNh/XowW6TmZtRoia+89rrIw7vlQn6dLanc447Bcv55dDebMTf3RE8vKaHDjJnY4+Mks8+t6wWS/TdwYPCuSNPgzNPFDeqe3H//HVVainb5Zf6n9uuLmvYWfPyxXO/MM9B2U7CXmJjoV0zoXkSsWLEi6P1cunQpcXFx9OjRY5/EEjVN26Psr06dOlFSUuJZ1ISFhe3RzsLlcrF+/XpPkkZubi4ul4sjWuh4+k/BMCb7Ac1shimTUbW10NQUJEehnE6ZPHxxuYJdUS4XrFwJJ50oldN6jYB85EZU3jYRbRwyBK1zJ+85jiYYfyKceza07wDduupO2KbUVKm6911tH9Fbit0mXCS7o8JCaU179jloCfFB14DmBkyPPCYuGYsF+vVFPfdsUMA/JSUlqHrZarXqxkxsNhtLlizB4RgX9JjdbvfInG/evJmGhgYaGxuDNLX69OlDVVUVdXV1aJpGYnU1OYsW+V1LU4qe331P7M4SmqIi2elyYdbZTbjdNw1xcSw+7VRMmkb91FjCXQotIx310APw9DMi9njXnWhJSagPPoLPv/A3JCCfs7tFb2Eh3HMvauQIicn46k5ty5d/AH/Mx/raK/Tq1Qv1+hvwzjvY3XL33bvD2LFi0CIjJZB+ztmwYb1XvcBkgqRkeV3fBI3GRmkZHGBMALQ2ySH7niiHQ9KX27TxZC66m+W5M5yysrKIi4vT/XydTicVFRUsWLCAlJQUunbtSnyzvE5hYSEFBQUopWjfvj3t2rXbb/Il7qr2rVu3YrfbPWN0Op1s27aNuro62rZtG9LAFRUV+bnnQEREfQsd/6kYxmQ/4hfAbEbV1oq2VQvidH5ER4vwX+9e/kF9kL4XZ54h1339TVk1VlVBbCzq1JPRbrgB9dFHUlG/q0x2OeNPROveDfX11yKJXlcnro6EBGoTE+CqSSKBvjUX2mZ6e6yPH486+mjYvAWys1qOlbz9jrfK2+GQDLXPPg+SDNE0LWhiqa+vp7q6OshHvX79eiorKznrrGAdJvek4ytzXlRUxJ9//ulxfdXX1xMZGclRRx3F9u3bCQ8PJ7WqSgx9c9fPprAwzI2N5Cxa7Ll24iuvEfP0k7ry5m6UUjiV4vTTfycuLg41fyc88JDnuhQWol55GWb/2rrPvaxMaopaYvVqUUk48UTJ/vPpm8KGDfK9eOwRv0lXvfyS9M4p2sHkjCa0NAVrdSrQXXsW/FbffS99WnaWQEYG6spJaKNHYTKZ6N27d9D57dq1o6amJqSU+86dO6mpqWHkyJFUVVWxevVqz7kVFRXYbDbdAsi9xWq1+rnVXC4XCxcu9NRDbd++nS5dutCxY0fKysooLy8nMzOTmJgY3er8/Z08cLhiGJMDzetvwLLlrTs3I10ygACeeBwumuCVk9c0OOlEybSqrZW+Ku4VUk0NfPs96qyzxMe+qzn4X10txwcPhedfgIpKOd4snWIHCb6/8jJahrchlBstKgr6tGL7rlex7bv6bUYpFSQU6HA4KC4uDjIm7nTiXr28NQQmk4l27drpTixlZWV+MZTa2lpWrFhBr169vOKR6emohx+Ejz6mbOdOCjrm0GX2XL/rRBQVkblmLVFHjaSgoACTyeRRrA2kVy9RvuWDr7yGBGTHOXMm+h3O9hKXC0qbayfsAQZKKZg2XQoUr7/O81lqYWGeOqWxADSijjhbjJLb3Wc2w+DgBIhQqPp6+U67azs2bYKXX6FpyGDWrF9PdXW1x3XkNvRZWVnExsaybdu2oNRaN+6iz0Cjo5Ri69at1NXV0b179z1O+W0NRUVFfkrIDoeD3Nxcj2yM0+n0JBJkZ2eTm5vrF/tLSkr6x+9KADT1F5jV/dEDfuzYFN3j06aVk5PjJDfXzKWX6m9Nf/1VXnv27HCmTg1eYWdlOZkxQ75MM2ZEMXOmBDh9v/SjRjV6GjRNmRKn22djwoQ6Jk4UH/7EiUls22aW2IdPGujkuAcYEzEHgKNLfva/QFw8tEn23NPWP0q47IpUcDllV5KQKPOT3c4sdQQ0NTHbNpr7q/9Pnp+WBqW7wOUky5zP9ORL5Z7SpvL2qhFB4x0VPk/auZ5zNvfVTPXeU1OTGCLNxMVXOrnkGuV/TwFMznqBMWue8d6TyQRp6RDp/eG77+mTT1bw6KNeyQ5N04iMjMRsNvt9TnfdFU5Tk/8Enp3t5JNPZLfg/pzcNDQ04HA46NlzO2efLdlp33wzjvXrg/tzTJhQx3nnVbBsyRKeu64DO8sTfB7VIDWVKU84GTNGJu3hw+NoamoKWoFec82PdOtmJuPFH7nsj9v8XyQ2FsLCmWXpD0r5f05mCygXuFz+n1PdBN6uu0gWDiYzOB3ez6nHG3D3Xdx3m5m5hd2DVAsujn6HidEzoUd3LtG+ZFt+8DpxypRqxoxpRK1fz9Gnd5NrREZIanAz06aVk5hYxh9/lPDAAwOD3ZANNmY5pTeI5540Ew2pqTiaY1spKTXcdttcRo4cyQcfJPp9To2NjTQ1NdGjR6Hnc/r442GsXduW8PAIXC6nZ0c4evRaj8rASy8dT1lZfJDQovueYO/nCLvdzl13iVrzmjXt+O9/5fvpG99LSanhzjt/Z9SoUbzxRhjTp0eilMJsNnlic3szRxQWhgUZ1/1xT6A/77mP7QuZmfoFy4Y5PdDo9Y5PS5PJJhDlP0FoGRmS1ZOeLhlA7oVuXX2wBIo1TPz11oBJJDMTdldQ5bvqbmiQDJ6qKqisgPc/QOklA/hyztkwerTUVoSFyeQUqb+CzM7O9qziNE3DYtGPmYSHh2GxWCgtjae0NB6LxUp0tLgRnU4nBQWF1NXVUVdXj91u172G3yrW4QSnvGc2WyMLFiygrKKC6rQ06hMSUG73UHh4UFpseHg4kZGRhIWFeZIIAF544URuueU4/hw4gPqEBGyxsSg0QJNMKpdTmpH5jk0zSQZYbJzUg1hC1A6Fh0NMrLQbyEiHu++Cp56WxYknpdhE0O5n4yaoqgy6XF6ehTvuEPeg1r07pKXKdXVEKxcvXkxxcTFNTU00NPjrWhFmlWQMXyxmXPgb2oaGBk9GXPB7GUVERISfS85kMmO1WgkLCwu5yhdFhfqgbOd9xWq16sZkAhcPdrsdh8NBeHgYkZERREVFhpSN+Sdy2OxM/moSExODsnr2BuVwwOQpIurncolUycMPwmuvSx2J++2Pi4MnH0fbjaSEamiA8y4QzSY3ZjPceYekiK7fAHnbpHtecjJcfCGMGSMCkCt1pOSTk+G+KaJGW7oLthcGN/E64wy0/9yx+3t1OsFk2m2w1B18jY6OxmQyUVNTQ0JCgm568JgxIhkze7Y3nXPVqlV+qaZms5l+/fpRWlrqSftMTExkzJgxbFu7Fu3ue4jalo8WFoZ56BCWjj+J4oB2xV3yC+ja1ASXXaob+/LFnap8xRWiVHDffZ8A0HnOXLrMmYPJ/YsKC4Pbb5MFwYcfwbJlIvMOUrh47TXQrZuoJwRIuADQuzfa9LdQ69bDfVODlZ01TXaBvpN9RDi8+gpar15+p7pXuLtbmS5dupQdvq0CgIyMDAYOHOj5W02b3uxO3SWG5eILmZueHpT23a9fv6BUW3eBqbtVrjvrrmvXrh6V4by8PFavDt3aumfPnvu9tmP27NlBMjpWq9UvbmYymWjfvj29e/febwkB+2ue+SsJtTMxYiYHGM1igYcfktRZl9PTN15dd61MAitWyCp0/PjdGhJAfsAV/p3/cDolCO72YyclwfXXwQn/8qToqpdflGDsjh0yCW3chDUmmqaT/w2vvAprA+pNfKnTmej07nU3FfpuzGYzbdq0Yf369RQUFNDY2EhUVBTdu3cP+qLq/WgDe1k4nU6Ki4vp37+/J0DqFhOsffAh2rpVkevqUN//QJLVSnFAh8fy/v3QWqn7FREREZTRA5CYX+A1JCA7vgUL0B59BPXzL15DArJzef0NmDABvvtGDEpgwsW6dbIrbO6lEoRSQfpg9OkLzS1qW0IpJYuZJUtEJuaSiWhduugGmAOPaZdegjr535IU0LcvWmIibTdvZtOmTX76XHqTzvLly/3aJLdt29YvaG+329kSmNIcgO/nb7fb0TRNV7ZlT+jYsSPr1q3zGI+4uDg6duzoUW8AeR/ci5iGhgZsNhuRkZH07t3b6LqIYUx2i6qokJV+925o+/CFCUyr1cxmuOXmPb9QZia0ay9poG7i4/2FGsvL4Ztv0Maf5H298HBxR/kQm5hI+S+/eHur6BEdBcccE/rxPaCiooL169fT0NBAeHi4X7C1vr6eTZs2earNW0JvB+N2afm6SNatW0dKaUC1s1Ik7NgR1C64pqaG3Nxczw66Q4cOHmViX1RVFewqw6WTmeSI0HF5uN2Zeoa2pgZee03cYSeeEGxMnM5mFQQdhQU9xo6B+6e2btX88ivS1dF9H2vXod56g8zMTMrKyjyfi9ls1jUKWps2suNtpnPnzh59rqioKHJycoLcVQ0NDUEeitLSUmw2m+fzKy4u1u3m6Ivdbvd0VHQLOsbExDBixAiPu9PtcGnNLtldN5KcnOyRXXHriu3YsSOoWVdhYaHHaFZXV2O32xk+fPg/vvuiYUxaQL3+hqzmS0uhfXvU1VeiHXusyKK8NU2ytKIi4fzz0Qa2Xgp9X9DMZtTtt8Hzz0vNQJs2YkwWLvQ/sbUtZMMjpDZEr5o8Lg4mXY42ZnTwYy2gVq6E198UefsO7eE/d6JiY1m5cqXHFaJXbNjY2IjdZiMswJ/uizvLxpf4+HhdFWKlFI06LqvoHj2CiifdulLuY5WVlVitVr/eGur5F+CHH6GighFRUVhsx/oZkM0jR5JUWkqE24Dl5EizM4ALzpd2AoHxp6YmqVZ/6UV44ikIfF+ysiSuodOrPgiTebctATzMX+CvSLx9O3z4MW2vvdqv2VXbtm1bLWKYkpJCSop+wBjE3R2YHuxwOGhqavIYk5iYmKDkl0A0TWPdunWU+Lgqq6ur+eGHH+jRowcOh4Pt27d75On79u0bUqpl6dKlftdJSkqib9++nu+fXoOvwLHV1NRQX1//j+/CaBiTEDiLiuCTT7192AsKYNoMqb146WVZMbq3/xs2ol5/FS2EL3FvUC4XfPU1rFoJ3Xv4dczT+vVFvfWmBMsjI6Ut7/p1UOUzyfbsFeLKAfTsIem/gatiECPTq5XXcY+7oQHuf8BbbLdxI9TVUTX1viCfui8mh4O+X35N2EuvQHgYavQYtGv9C+bsdjtbt2716CKB/NiHDh2q6+bo0qUL80/8F9EV5cQW70SZzTT16IF5wsVY5s8PqiMJNC75+fkeY6JWrRKp/ubJPqyqioiaGhxOJ9HR0URERNCmWzfCTz9NvjdWK5x9lif+omVno55+Eu6+Nzht2mpFs1hQV04Sl6N7Jdy3r7Q8djhh+nSpR9FrkubGZkPdN1Uq32PjYMJFovumh56xNstOon379rQPkAPaH8TGxhIXF+fnIoyLiyPGx+AnJSV5tNxCERMT45fK68blcrFmzRq/hUJ9fT0Wi4W+ffsGnd/Q0BAUr6isrKS6utpTy9S1a1cqKys9392oqChsNpuf689sNu+zm+3vgGFMQuBYvMRrSNzs2iUrxCVL/LOpdu6UHueBEir7wgMPwXffyY7hq2+kEPDRhz0Pa5rm7V/Suzfq+uvhy6+kT0WvXnBr61xomqahnnoS7rpHqul9KS+H665HjRuHdu89rRv3r796DYmbTZsJdzqDApoWiwWr1UpjYyN9f51N6nKfepzt76M65fDyy8dgMplQSmIggS4Qp9OJ3W7X/TGnpaXRffRoNmdlEb15M0lt25IwZAgLFi0KMiQmkyk4NuA74c6dF7RreCNlEmtOHk+EuR1t2rShc+fO8pwA6RqllIgcpqcT+/yzcM11XtdVXBwcfTTq4UekCPCokdIyuX0HOOsMibmdcxZqzCipVP/hR29VeyB5ef4usaVLUX37wN13MW1agJtt1FFSD+PeKUREeBqoKaU8Lj+TyUROTk7Inu57gm/LXbdry10178uAAQNYtmxZkEFx7xS7du3KkiU6i59mAnOKWlrE7I7IyEhGjBhBfn4+LpeLrKwsVq5c6RmbpmmkpaXp7mBAvp+rV6+msrISs9lM+/btvXVPfzMMYxIC6/Bh0CbZWwAIkrmSmCh1AoHsx2IqVV4O8+d7XU8uFyxahCrcjtZO3+WgnXKyrGL1rldXJ/75tDRd95EWGYl68nGp1PdtMQuShvzLLNS55/pLtoQiNVWymHzTjaMiiYyLIz093eN+cGfGdOvWDZvNRtRn//O/jt1O6ZdfkTve1Vy4ZiIqKgqLxeLnKomMjAwZ/HQ6nTQ2NpLUpg1hmZlU1ddTtG6d7uTicrn8rm0ymfwlNUYMl4wsH/95crsGLAPiKKuuprymBqVUkE5UQ0MDS5Ysobq6Gk3TSE5OZuDzz2J+9z2pjj/pJAnE//mn90mjjkK7+y6/62hpaXDWWahf5+jeK716BSdmuFySRXj/g2S/9Yb/Z3/uOfDf/3oLWW02ePpZeO0VtmzZwsaNGz3GtbKykqFDhxKrl86O7OJyc3NxOp3k5OQQGRnpcWkFpgBHR0frqkT7YrPZgvqUREVFYTKZqKioYMmSJXTt2pWysrIW+8W7CbVriIyMDBJ+TExMDCqgtVgsftljAwYMoKCggMrKSpKTk0NmNwGsWbPGT9W4traWuLg4XXn7wx3DmITAlJICF14ovdN37YK27eDaa0SQ77hx4kpwr5Kzs+G88/bfi9fWQkOA77y+Xmo/QhiTUKgXXoSffpZrZmeh7puCptNmVLNYUM8+Az/+BA897K8LVlsr7qrWGJOBA2HgANGZUkpSVY8+BqfJhNls9vxQu3bt6qmQjomJQcUHF5NWhYdTXCwukNTUampqajCbzR6fenR0NJ07d9atS7Db7Xz//fdB7pCWgqQpKSmUlZVht9txuVye3hydO3dG69cPNf4k+Px/4HTSEBPDrIHnUF0SR2pqNUopSktLg4zJunXrPG4d9zlbEhLo1mws1OrVwZl0q1ajinfCtGmwapUY5xNOQDv3HKjVWWWnpMCbr8NFF+vfWH4+uctr0BITyclpXqDM+81rSNxsy0NVVbFz506/SbqxsZG8vDxdMcOGhgYWLlzoSavdsWOHJ921qamJmJgY+vfv7xdPcO98qqqqiImJoWPHjpjNZurq6ti8ebMnqO2Lb4yttLSUuro6Ro8ezZIlS6itrfXsRmQXqzx/W61WOnXS/94WFRX5pQNHRkYycODA3QbSNU2jQ4cOrWrXG5j153A4KCgoMIzJPw3tgvNRp50qbqz27T2Cdtp556HS0kQjKSZG0ipDiCDuFe3bQ5cu/u6Mrl2hR+uaUrlRK1aI/979Q1y1Gh54CDV4EPToIZOj+9zqaqll2LQpuCAyIwNGBlfQ6+Fxm33+PzG4o0bB8GEsmj/fb2IvLCz0GBNAhAbz8iC/AEwmatq3Z/PIEbz8yHGAt5bDHfw0m80kJSWFlAzftGmTrl89VFmVxWIhOjraz7XicDgoLCykU6dOom585x2oEcPhpVfY2qsHj/5vot/Y9CahRh1tLr9WxpoWHL/QNHjzTX+9rg0bUfPmeQUifamtFcP9r+PhjbeC9cDi4rjslo6gad46k6wscZP6uu5iYiEqSn/3GmKC3bRpk9+E7G4w5aayspLVq1czZMgQlFKsXbuWgoICv91lRUUFffv2ZdGiRUFtnkPhzv4bPXo0a9eupbCw0PPZ+rowfRcwgWzdutUvU8u9I8oI0SF1b2gp8/DvhmFMdoMWFSVZOYHHjz4ajj76wLympqEefgiefFqaHqWmwI03oPmswFVzjYFer3n108+iv7Q1Nzg7aOVKqW0xm6n55lvUww/KRPHwI7JadWMyQXycVOtfeOEeNcXSmoPPbsrKyoLatLp3AG5fs9ajB2rmDIkPxceyOSUFZwuV906nkx07dpCTk6M7WehN4p7xNdclOBwOlFJERESQmJhIbm5wAzGn00l9fb3HjZPVuzcJH7xHx9paTF964ywWi0XXsEVFRfm7bFwu4tauhQHN2X89e4qLyjde1KdPcDMzp1NaHOj55uPjwWxGmzAB1auXdF0sKpLnxMbCSSfAu/7GQOvVEzVsqMSCmppkUXTiCWhWK5mZmVRVVXkMd6TZTE6ImEAoMUxf3MZl48aNuu9xWVkZ69ata7UhcVNYWIjJZGL79u0hs79sNhvV1dV+MZ/i4mIKCwuDsgKVUlRWVu5XY5KVlUVtba3n+xgfH/+3baZlGJNDFC01FR5/VJpqffsdLFmCysgQd9T338N7H0BlBapde7j7LrT2MpGpb7+Fx58MnRrs3nU4nTh++QU+PxJOP036nwSeN2gQKOCbb1D5+XDpJa0uTPTF1+3Q0jEtJgbOE6Xh+C1bKNqNjIvD4aCiokLXmIQKiLpfOzw8nGOOOcZT87Js2TLdCSkqKoqFCxd63Cw7d+6kT58+pKenExkZid1u96TPpgbKjAA9srNpmD+fqsRETC4XyXnbyJn3G+rEE9GSkmTh8PST8PKrsnDomAOTrpD4lR52u39MymqFM073LDS0I49EffpfWLhIKu7HHYPWtSu8q3Othx+C2XOkK+Sxx6L16AGI5E1YWBg7tmzBtGIFnb7/gaj6etSgwfDwg34LmLS0NEpKSlpM5XVLjujtFEHiVS09vyVaMiQgrivf78fOnTv5888/dY1gWFjYfu/lnpmZSXx8PHl5eYSHh5Odnb3HHSUPF/6ed3UQUEpJdkxTEzTXMez2/Dffgj/+kF3A2LFoF17gf86HH8Ebb4rooskEP/+Cun8qvPiyt16heCc8+BC89or8/f0PwYbEnU4amFaqFHzwoRiT+Pjg5yxd5m0/vHARlFfAnbfv6VtDcnJyUEpoQkJCi7pGeiq9gZhMJt3CQgiuBQikpqaG2bNne2oc9AK57uC+787Cbrezbds20tPTMZlMRERE0L8F5QJreTlD3n2fhrAwTE4n4e6dYn6+KBXQbETv8Dce6vzzZXcSIPsCSNV8ZARhpaXYjzkaLSDtVdM0GDpE/rlxOKC2DrVgIQwZLG47kwmOHiv/AsjMzCTjqWckg8zN7Nnw+edwlnfX2a5dO2w2Gzt27EApRZs2bbDZbJSUlOBwOIiJifHIvYeaRGNjY+nevbtHbn5PaCkAHxERQceOHf0WFgUFBbqZfNHR0bRv3z6kS2xfiI6OptceptgfjhjGZD+gGhrg9jskxuFyQffuqCefQEtMCP2k9z+UbovuL/bGTaikRLQTT5RrKiWpvu6tuMsl6cHXXh9c+Jafj6qvF5ecTu95IiPh/qmiERbo9nILQ154vsQsKiqkviQjHQp8qurdGWW1teJeCzQ+LaBpGoMGDfJIU8TExNCjeRUcig4dOvhlwejRpk2bkAapNas/t/slUJMJxNfeu3dvP+kPNy6XyyOnoZRi8+bNnrhKEO3bQ/t2RG72kQjp0EEaWrWANmwo6q034M67pGOmexHQoQOccxZafDwxrdR1Uo89AQVXyzVuvhmOGoUK6H2iy64AHS+ldNUSOnfuHFQ0WltbS0NDA0lJSZ6q9M6dO3saSfnSvn17oqOjGTBgAFu2bMHlcpGcnEx8fDxLly7V7YPixp2QoefWDAsLIz09ndraWqKjo0Peb0pKCkceeST19fXs3LnTUwlvsGcY79j+4I23ZOXuZuVK6R8y5f9CP2fRQv8UXJsN5syT5kcgP9zAHYbLFdyKFaRC2j2pZmWJcqwv4eGyC9JrcZudIwKNn/1P0odB2rT26etvTEAKN8efDBERqJ494aEHJK1YKcne+mO+9LE/amTQDzciIoJ+/fqFfj8CiI6OpmfPnpjNZpRSxMbGYrfb/XzPA5rjDk6nky1btlBXV0d8fDw5OTmt8uUHEhMT42nglZmZSVpaGiCukcB05IULF9LU9C8Aj5HUawylWSyoG24Q+ZKdOyXz6rJLW1WprqWno2ZMk1a+y5ZDTDRcdlmrDLnL5aKsrIyo/2/vzaPjSss7/++9t/ZSLdpK+75ali15391ttgCBMBACMwmEhgl0gMDvN4SwzCS9hgC/wwATBqaHyXS6k07mJIEOyQCBQOi23W7vsrXv+1ZSlUqqfb/398f1fV237q1SyZItyf1+zulz2rU+t6r0Pu/7LN/n1m2YXn4ZEH5fvCORBC5eFL+vE8ezv0hpGSDpmgGiJMx+ZfOf1+vF6OgoEokE8vPz0dzcjLy8PFkzIiCW3ZpMJoXzlv5dUFCgmHD45je/GT09PUgkEvB4PIoTZ15eHjo6OjAxMYGFhQXZ/T6fD+fPn4cgCLBarTh48CCqq6vh8XjIyVen06G6uhp9fX1YXFxELBaDyWTCnj17tjR38kaAOpOtYG5OeZtaeCIVtcUkpV+CYVkIjQ13JximI4WsCguA9733bi7jC38EXL8OrKWUJJpNyrHBGg00R48i8YU/An75S1HSQ/pD9HrFrnqDQdZXQRxcMCguSP/tO8CXviBKo//TP4nT/15+GXj724E/ybHJMQuVlZW4elVyqI8gHA5jenoaBoMBVVVVZMd748YNovk0Pz8Pr9er6FMAxHCKVLaq1mtSVVUlKyMdGBjA4uIieJ4Hx3FgGAbJZBJzd75vqYoLEMtVBUFQr4Q6cRzCsaOAzw9Y8jaUd2IYBvjdD4v/5YjP58OtW7fg9/tx8B9+ABPP41eOlPHHiQT4/n6wFy+KfSgaDfDmN4GRBrNJfP5zgGcFGB0Tf5snT4iS+ilI89ylnJLb7UY8Hld1rABUncny8jKGh4fR3KwcL+1wOIhi8eXLlxXfq16vh8ViQXt7O9xut+rwNUCsGOvv78fhw4fR2dkpGwscCAQwMzMj65ofGxuDzWZDb28vAoEAGfi11TmVhwnqTLaCxnr5HHUAGBuD8N73if0pX/wCSZATPvwhoH/gbsiqshJ4LK1P4MkngK98Vey4T6s8QVkp8N73Am97q0zGhbHbITz9tLib9XrFhG4yCcyliQXu2QPrX72A1dVVCMMjSm2umRlliXA6U1OiGnLqGNlYDLhwAYLz98CUlmR//gbRaDRoaWmRLThra2uKBSbTNL/CwkL4fD5VIUGDwYDa2lryb5fLhampqZya4jIhzM2LKs0d+8Wph1tZPp6FoaEh4ix5RtmDwzMM4v/wA+g8HjINRZiaAiorwbzlrqgnU1wM4fv/U5zKmGcGkzb7BBDl4tN11tQcuURrayvC4bCsNyQcDmNsbAwsy2Ydz1tZWal4ba/Xi3g8Dq1Wi/z8/KwikdJ90rhnQRBw/fp1mTaXRCQSQXd3N3m/cDiMwcHBrKHVNzrUmWwFH/0oMD4h7vISCXFB9XgAD8RF/Jlngf/1P2VPYdrbIXz/OeD//B2g5YD/8NtgiuWSFYzVCnz9qxBevwx84Yvy/oHOTuVOUnreyRPiLvIOwk9/KpaApia121MSgu98p5ifSW2wymURLSoUB2il7/J9PjHevklnkkwm8Vd/tYRgMIjm5klShVVbW0sW/kQioVjw1RyAwWBAIBBQVBRpNBrY7XYcOXIELMtiZWUFbrcbwWBwXUfS3y9uEPbunUNRURFxcoIgiN/5+QviZ1NYCOGZp8AcPXqPn4QS4eZN+F78Kwgulyi98uUvivPtIS+Lnj5yCIUTE7iyKr73WcN5QBCgT2/mjETEBHuKMwHunIyyNMqqNYwyTieEz30eOHYE+MAHyOcyOTmJ6elpRCIRcBynGM/rdrvR1NSE+fl5kqtqa2sjDY96vV4h0JlKZ2cnjEYjfD4f4vG4ohw9vb9jYWEh45wlo9GoKFWORCKk74iihDqTLYDRaoGvfw2C2w3826+Ab/xX+QOmpyGsrSl2dkx5+boaWkJ/P/DVr911JBoNcPw48KUv5m7gO94h6mWdPw/E4kDHfnHeiWRHYwOEj/9HsUnO578zbzzDECWpya65Cfj0p8WJffV18th6Q8O6CWY1kskkeJ4n8he3bt3Cd797GgDw1FNi4jcej2NkZAQOhwMmk4kkalMrxfR6vaIqqLCwULG4AGJoS6q06e7uJnIvapMb0/nhD0+CZVm873235AnoS6+LVXXSYrmyAnz+CxD++Z+2pLlVCAaBr3wVCalAYWxcVEz4zp8DEBdC6fNYranBzf/wQfzJt34PAPAvZe9GXoakfcxkwkb33LW1tZibmyOhKyaRgGF2FjdKS6CZmED9X/xv2D7+ewiHwxgdHc1apcdxHObn59Hb20sczdraGjo7O4kisc1mk32PBQUF5PfCsiwp7BAEAV1dXXC5XEgmk7BarYqiD5/Pp+qYbDYb2tra0NvbK/sdsSx7X6q9HhaoM9lCmKIiCB0dys5iq/WuKONG+euXxP4DiUQCsFo2NFuFYRjgk78v/pfpMR/8IIQPfEAMd42PA//pD+UORaMRmzTf914xj3LsKOk3EJ58AvjWt0U7i4uBz/yBajNlNgYHB+F0OpFIJGCxWLBv3z7VxR8Q4/QLCwtEWLG4uFi2MJhMJrKgSjLkZrMZS2lVcAzDEFmLQCBA8iOA6NgkhyJJtxiNRvj9fvKapjvd4orQzI0bcnl3QJSn+cu/3Box0H/7lVgMkcr4BKno27t3L2KxGHw+HzQaDQxHjiBqs0IQBPWZKwB8JQ5wH/7Qhp2JRqPBsWPHREcxOARMTmK5qRH8nVDQqj+AE+EwnE5nVkcinTinpqZkJ5ZoNIqZmRkUFxeDYRg0NjZiaGgIgiCIQp4ZNi0Mw+DQoUMIhcSxzjabTZGPqaysxMzMjKxYo7CwEMePHwfDMKipqcHIyAgZwFVcXLwlgpcPK9SZbDFMawuEM6eAV86LYSWLBfj1d4ox8zSEhQVR1ryoSJSYV0vKq9XdrzM86F5hGEZ0Gi0tEL72Z6IS8uoakG8HPvQ7pKlN8bz6erIrvheWl5cxMTFBnEE0GkVfX1/GUk5JSgUQQ1oLCwuyHebq6iqOHDkCq9UKn8+HSCSC/v5+RR6lpKSEVOwEAgFFCaq0IPE8j6KiIlKCyvM8jEZj5tLaRx8R+3fSw2SvXYLw+CfEEu57RIhEAJNZVUxT6o43Go04efIkwuEwUWY2m/MQj8eRPHYMwo9/AubOtUYNBkwfOQzfe34DRzLIzicSCSwvLyMvL091Z240GrF//34Ir13CJauFOBIACFnyMDExgcrKSoVIZypS0+d0ujw/7kq5jI+Pk6oxAIrNxtLSEkmkl5WVoaqqCiaTCaYMn7fFYkFzczNxKBaLRTbLpLa2FsXFxZibm4P1jlDpG30AVjaoM7kfPPusGOq4dQv4tbeJHchpCBcuAl/7+t3d/89/DuF73wVjMomluvG46FyOHRNlNKTdk0YDeFYh/N0/AL/53g2fAHKF2b9flPV4ADidTkW4YXV1VVWNVWpULCgoQDAYRFdXlyIBLAgCenp6cOTIETgcDly/fj1jE6O0OBQVFcFsNsvi5Hl5ebJcCIB1k6+CIIjFGByndCYzM8AXv3TPjlf43v8QhTj9frHcOx4XK+xMJrEQI+23kKqmzDCATqdF4X/+MtDUBP7KFfiSPObe+mZw9fU4eKexMJ3l5WX09/cjGAxCo9HA4XDgwIED6ovqv/+g2NSYbrcgwGazoaKiAgsLC6pl21KYTEqySw5Dq9USyfa5uTmZM/J4PJifn0dVVRXcbje6u7vJ6UeawLieGGNdXR3Jv6ldk9lsJk2XlOxQZ3IfYBhGFEbMJo743/+7PIzUPwA886cQ2vYAP/4JEAxAqKsHnnpCDJm9dkksQV5bE+XKe3pEuYyvfzVnu4TRMeCHPwD0RuBDvyPK6e8A1OLWyWQSra2t0Ov1SCbFORJmsxn5+fkkNDUwMKA6ix0Qk6VdXV04deqUapIYAEm0sywLhmHQ1taG0dFRRKNRMtubYRjwPC92jOeyK/3Zz4F/+KG8hyiVvn4Ii4tgNtjDINy+LZ4UUxwn29wEvqNDVE84cjin12EYBvjAb4H7wG8hH8B6v4DR0VHiYBOJBJxOJ5aXl0kPjuy1CwrgaG+H1+mEcCdEaDAYUHdH227fvn2oq6tDd3e3otlSCina7XbZCYZhGPL/ahsCqUJrdnZWFkZLJpOYn59HKBQCy7JEIibjZ0LZNNSZbAPCwoKojpvOxYuivIoUxlp2AV/9Gpj/+g0I/+49wO98KOVFhHVnnMje89LrwLN/erdD/uJFJF/6q3vP5Wwh5eXlim53nU53p1SUA8Cpyp+vJ70RDAbhdDpRX18Pj8ej6JJOJpO4desWgsEgGIZBQUEBmeWdTCYxNjZGTjVSN3VqjL6mRuW0c+1aZkcCiLI4ac5tbGxMVCteXUWR242WmlowbzonX+R+9Yqy6TSRBPOF3ORtVG3NgfTPjOd5eDweVWcCAE1Hj0I7NQWXywWO41BfXy+Tn8/Ly4PZbJY5E5ZlSYXU2NiY7HuNxWKYmppCSUkJbDab7BRqMBiyDppaXV0lpb3j4+MoLCzEnj17Ms5loWwO6ky2g//7Y/WZ62oJSsnphEJAOG3xjERyn3HyDz+QS63MziL8P7+/tdMh7xGLxQKTySRbKEpKSqDRaPDCC+rigIC4mGQ6mUiYTCbk5+fj2LFjuHLlimz3qtVqZXF3p9OJ0dFRNDQ04Pz587KeBWlksM1mI3kWVdvWm3HR2SEOuoJ4IhscHMTkpFj2DJaFr6AAmn/5KRpv3RabBiX27xe/w5QwD1tcjFxdRKbPMRQKYXx8XFRErqlRzNlI/140Gk3WYVAMw6Curo6cRtKJx+OKXhGWZaHRaMDzvOr3KZ1IOjo6wLIs/H4/mVMihR3r6urgdrtlzi+1tDuZTGJ5eRnBYBCnTp3KKgRKuTeoM9kOGhrE3Wl6TJ1jAV6QizFK+l6VlXdmnKRM5Gtuyn3GSVRlFx9UkVd5wMzNzRE5EkCM8zc0NJAdZ3q5cCptbW3w+/2KnEkqt2/fRl1dHRoaGnD27FkMDAwgFArBYDAgEokoKoykDnq15jee57G0tJRdZuM//HvxdNLdI55QamqAtj3iZqC2Fvj475GH9vX1KRLOgkaDlcoqNP7br5D8xO9haHYWwWAQRkcxWt7yZmil8cE1NTB89jPYmGi7nGAwKFNEXl5exr59+2TX197eju7ubtIFXlFRQeajCysrwMgIFgsLMbe2BkEQUF5eLpsf7/F4MDo6ShLcDQ0NipxJIpFAKBRCb2+vIqnOMAypoNJoNBlFNe12O9rb29HV1ZWxD0W65unp6azNkVtNMpnEzMwMYrEYamtrH9qmR+pMtoOREbnDAMQM6ZEjYuL2xk2xr6S8HPi1X4PwrW8B9nzgiT++I1W+KJbgps04ycrhw8Dt7rs727w86N7zbmxcwSo7gtMpljPHYsD7fxNMluSlOI53QhHWkMIi/f39eOklG3iex/vet4iDBw/KnIqUQ8nmTCKRCCYmJlBRUQGDwYCDBw+SBU5tfkYoFMr6eqlJ7RdeEEOEjz129/GMwQDhe98VtdpWV4Fzj6pW6UWjUThTS75T4GIxIBxGz5WrmI/d3Wn73/F2HP/wh8AsLADHj0NXVoZgDkKPmWwdHx8n18rwPIpu3kTk5i0If/ApMHdCQXl5eTh16hSi0SgRPxwdHUXe//k7FL/2GrjVVdjsdiw9cgbznZ1YXV0Fy7KoqKhAPB7H7du3yXusra0Rp5Ia5rJYLAiHw4oTC8dx2LNnT84z06X5NDuJWCyGK1eukNkpc3NzOHDggEKD7GGAOpPt4MIFpTN5168D/+U/iyeW69fF8JbRCPz374pjgwGxv+C574mS5RvlYx8VHcm1a4BGC7zz7dCdPZvzYpQLwuwc8NnP3pVuuXgRwh//MZgshQjpJ4NkMom1tTUSK//FL34TAHDmTA/6+voUO9PUeHwmotEo3G43KisrFQvcRsjPz5d1P7/4ovje0gIdjUbBMIwYQkmVf1chFoupVjXpfX40XHodybo6LAvyk+va2hr8e/fCplIduB7ptgIpUytjMRx96W9gn5sHK+XinnoSwYZ69Pf3k5kvra2t4hCrqSmcfuVX4AKiMzatraH+0mUs7NuHBMTO8oqKCszOzio+Z6fTCbPZTEQgDQYDWlpaZBMuyWeh1+PAgQM5KSMDovqvXq/POhgtLy9PJptzv5HGEEtI0jFHt1ANYadAncl2EFepta+puSsAePQocPQohE99+q4jAUT577/5W+DxT2z4LRmGEZ93D8/Nmb/6a7kG2IpHHBucwZkwDAOz2Sw7mej1elRUVGBkZESxy1Q7STQ0NGBlZQUejyfjrlSn05GdoFThs1E4jsORI0dUpckTiQS6urqwtrYGlmVRWFhI4vuZkFR1UxcaXTSKk6+8AlNLMyKf+QPlbPgNIoRCYjPp1BSw9Jyieq+qqgrLy8uoOX8BBakK0XNzwF/+JW795vtIDiMQCODmzZuIRCIoWViEISD/Lgx+P3TBIKJWK7zLyxg+fx7mDHkT6XusqKggmwOtVovp6WlZxVY0Gt3Q9EWDwYDGxkZMTk7K1H8DgQDcbje0Wi2amppUQ6b3CzXHdi+K1rsB6ky2g479Ys+BRGUl8N5/p3ycWk4ji4jetqOWl1lnyNX+/fvR09NDYvI1NTUwmUyqFTdqSVOO43D8+HG43W4sLS3B5XIpFqDi4mJoNBoMDQ1l7KpfD0EQEI1GVW0YGBiQiQXOz8/DYrEoZnykwjAMOjs7MTAwgEgkAr1eL+pQ/aZ4EjMCME9Py+y12Wwbk/N44ilRQgcQlZ7jcQhRgLkTsy8qKsLevXuh/fFPFE9NelYV6r6S0/dUVyFks8LkvesIQ3Y7ohYLaq9cQc2NLugCAUQKC1H1/vdhNkMJ+uLiIvbs2QODwYC8vDwYjUbZeyaTSfT3928ov1FXV4fq6mqiaLC8vIza2lpZQYAgCBgZGYHb7QbHcaiqqrpvasClpaUK4VHbBmYB7SaoM9kOvvwlwGIFhgZFqZWPfVR9xvreNvnu1GIBfv3XlY/bKbzj7eIceWlB0GrFpsssmM1mnDhxAolEgsi8A2L3sSjCxwAQZBP70pHKem/fvq26E4xEInj99ddVh2Dlil6vx+DgIOLxOPLy8mST89R2z+tVmQGA1WrF8ePqM0Xi8bgiBKjT6XLuiRBCIXGoViqxGPDqZeDX3kZuqqyshPC+9wG3b99VfgaAhgZotVrZIih104cBjJ86hborV2EMBiGUl2PuPb8Bvc+HhkuXYbjzOesWFlD7o3+G4SvPYmp6WrEj53kePp+PCDCqOepsg7EyEY/HcevWLfJ9Ly4uoqOjA8XFxQDEWfSjo3dn/vh8PlL1t9WUlZURx5ZMJlFQUIC2trYtf5+dwLrO5KWXXsLIyAiKi4vxyU9+UnHM/9GPfoQrV67ga1/72n0z8mGD0WqB//T/rP/A//T/iiXEvX2A0QC8611gOh5MV/q9wJw4AeEPPwf85KdIxqIQjh+H5nc/tP4ToZyMyHEcjh07BpPJCJ4XcObMmazii319fRlj5X6/P6cxwOmwLEtG80qSIoDYvyC+nlj1lK5GCyCjhEeuqOUbJP2xnBwKx4nOPBWGASx3821+v18sSy7IR81HH4PtZ/8qFn7s2QPui3+E0rExzM7OIplMgmEYlJaWorq6GqOjo/C99S2YfPe70FZUBE1lJSr9fuDPv0MciYTOs4JCgwGrNhvcqSHbO0hVThUVFSgqKhJHItwJV0q9PV1dXUgkEigsLER9ff261z82NibbOEQiEUxOThJnkm5HLBbDzMzMfXEmANDU1PRAq8e2i6zOZGpqCh6PB8888wxefvllXLlyBadPnyb3h8NhzKSGayhbCqPViqeYXUT0LW9GV0E+ERl09Paira2NNP5tpNuYYRicOyfuTLM5EmmqYCa0Wm1GZyLthnmeV+yCpSZFt9uNnp4e2X1erxdnzkTAMAxaW1sRCATg8/mIeORmFw+18tFclIwlGL0ewqmTwMv/CCQSOKu/CDiKRcVpiMn8GzdukNDVclUVOr/7HRQVFAAsi0gkgpaWFjgcDiwvL6OwsJBoUx1TOW3a7XZM7G1D7Jf/Bl1KDiyal4e4VqvqcAGQjvpAIEBOnm63GyzLory8HD09PWQ2i8vlQiwWyzryOZFIqJ4UU79btc9xI58tRZ2szmRkZAQdHeKYzs7OTrzyyisyZ/LTn/4Ub3/72/H888/fXyspu4a+vj4yMySRSGBmZgZLS0sQBAFmsxkdHR2Kca5qSHHtt7xlGQzDYGysNGsOQi35rtFoUFJSgpKSEnR3d6vKcWg0Gmg0GkSjUYUziUajMJlMqnM0WJbFU0957yTZjTh9+jRWVlbAcRzsdntGpyn1HCQSCdTU1GRsnisrK8PU1BSpZOI4DmVlZes642AwiKmpKXAch7rPfga6piagqwtPlc8CHzlHSsnTS7Kj0Simp6fBcRwGBgYQCASg1WpRXl6ecWpiKgzDYP973oOF1y6h5NZt6EMhBPPtWHjLm9FSWop5lWotCZ7n4XQ60dLSgubmZjTfqVYbHh6WTcQUBAEulyujM5menpaVO6eSWopbXV0Nn89HNhhms5nOKNkCsjqTQCAA+50ZHOnjNkOhEGZmZvCbdxKG6Vy+fBmXL18GADz++OP37Qh5v5Amtz3M3I9rVDsBSOGnWCyGoaEhvPWtb5XdL3WCz8/Pg2VZNDc3w+/3Y3x8nHQx+/1+5OfnZ3QoRUVFMkkWrVaLuro6tLS0wGAwIJlMore3VzHwKltlVzgchtVqhd1ux/T0NAlzsSyLmpoaFBYWyh6/Xu9ANBrFL37xC+IgFhYWcObMGcXrAADv9+NtR45g0OlEMBhEZWWlQrQw/ftzOp2KJsQ3vePtYHr7kLxxE+z0DIyf+xw0zY2qVWkrKyvE8QPiZmB6ehqNjY1wOByyxwqCQBo86+vrSf9Nwf/+C9z+l39BYmgIif370XHyJEZHRzP21EhIJ7pUZ2lUGbPAcZzqbzYejyumPrIsC5vNhtLSUhw8eJBU1+Xn58PhcGB0dBRarRZ79+7ddFjyXnmY1pmszsRsNpNO4FAoJNtR/uQnP8E73vGOjM89ceIETpwQp/3F4/GME812KtKs8IeZ+3GNaotUKl6vFx6PhywagUAAXV1dshLZxcVFWCwW8DyPv/978Tf0gQ9cxuTkpOrCu7i4SE5D0utKQ7RGRkbAMAxYloXFYkEkEsnah5AKx3Hwer1gGAYHDx7ExMQEQqEQ8vPzUVxcjD/4gzgCgQC+8IXFnOTJ+/r6ZJ+33+/HzZs3ceTIEXKbwPPAV78OXL4MxGKoamkB/uxPwaQ1+gHK76+7u1u2mPp8PnT9xf/Gvn/+ZzzlfRIA8NTUZ4G/egFFRUVYWFiQncbUSlYTiQRZdImNd8bdut1u8DyPgYEBtLW1EZmVhtOngdOnwfM8ZmdnMT4+nvVzAcRihPRKu9LSUkxOTpLTCcMwKCwsVP3Ner1eRYEFz/PYu3cv7Ha7oiBCo9GQE040Gs35N7HV7MZ1Rs3JA0DW9umWlhb09vYCEGUpUqtpnE4nfvjDH+IrX/kKFhcX8fLLL2+huW9MhE3MG98ptLa2kvJVtT4LKWQEiH/E165dkzkSQFyspAVkYKACAwNi2WYymcTw8LDsj08QBFk3uyAIipCXIAhIJpNkXnguMAyD6upqYqtGo0FzczP0ej3Gxsbw6quv4l/+JYmLFw24efMmLl++vG7lkdqpTWHPj/4J+L//F1haEjvor1wBvvHNnGxWC+Ml7mwGL0TP4EL0jDj47MoVlJeXk+Fi2ZD6ZlJZXFyEy+UipzxJZSCVaDSKS5cukd4UNfR6PfR6PQoLC1WFPHU6Hd70pjehsrISJSUlaGtry1jRl5eXp8jLGI3GnEKqlK0h6zaytrYWNpsNTzzxBIqKivAbv/Eb+P73v49PfOIT+MxnPkMe96UvfQnve9/77ruxDyvC6Jg46ndhAUJRoVgqfObMdpt1T1itVpw+fRoejwc6nQ7j4+NYWlpCIpGAyWSShamknb4agiDI8gkajQYrKytwuVwYGxuDVqtFQUEBmpqa1lUPTmW9ue4SqdLpEisrK5iamkpZtO86LY/HgwsXLkCn08FisWDv3r2KU5rD4YDT6ZTZIIWRpXyA5do1GNKdwpyKwnQagiAoHBPHcShLHzal15HmRaPRmFV+hGVZlJWVKUJcXq9X8bx0Rzk0NJSxPJrjOFgsFvh8PvA8j2g0iq6uLhw5ckTh3KxWKzo7OzPamPqaWq1Wpqmm0+nWPSlTto51P+kPf/jDsn9/4hPKDmpaFnzvCIIAfOXPgP5+8QanE/jmtyEcOrSpiXzbCcuyRJzvwIED8Pl88Pv9cDgcinBJNvbv3w+dTgdBEBcLaecvCAJisRicTidCoRCMRuM9lf4C4gmkvLwc8/PzstvVuqTTm8/SkXS91tbWkEgkcOjQIdn9lZWVRBaf53kUFBSgtbUV4XAYN27cgNfrRYMgQCHdqRLaS2diYkJxwrNarSitqAQ0N+/eeOAgcKdHprCwcF35kZqampzG3UonAGkwWfrnCdz9rBsaGnDz5k2ZU11ZWcHq6irJO8XjcfT29iIajYLjOLS0tGRt9gsEAoqNSTgcRiQSyVhJRtlaclQJpNw3lpeV87zn5kR9rocEq9WKiooKxQJdW1ubMf5aXFyMkpIS6PX6rKW9Pp8P0WgUer0eLMtuWCpDEAR4vV5Zxz3LsqrzOjbSubx2R0U3lWg0Cr/fD0EQoNfrUVJSApZlZbv4yRPHsdxQj6RWK+q0NTcDf/Bpuc3uFQg//1ckFxZl75cOy7LAF/9I1HyzWICCQuAb/x8Z+KXRaFBXV0dGEKc7DZ7nMTc3p3hdadytxWKBwWBAcXEx9t+Zyjk5OYnZ2VnVE6AUgjQajYqQYDKZlDmDrq4uLCwsYGVlBcvLy7h161ZWR55IJBT38zyf9TmUrYWeAbcbm038L3VXabWKcuUPOSaTCYcOHcL4+Dji8Tg0Gg0JgdTV1ZHFjeNY5OXlycpEU4lEItDpdDh06BCZA56KNHxpcXFRtQs+EAiguroahYWFiMVicDgcqKyslD0mGAzKuqbXQ61vobu7Wya7IiXzZTM4NBpc/53fRn0kij2VFcCRI7JxvMJLfyPqs7nd8BUVQfjN94L5vd9T3X2T/NS7fh34r2LDHqOLYXh4WDY+12AwoKCgAOFwmBQySGQqXZbG3QqCIMuNrSdXs7a2Bq1WC4vFIrtus9mM0tJSAOKpJP2UFQgE4HQ6M8qeSFIzqaE1q9W6bVVab0SoM9lmGIMBwrvfBbz0N6JDMRhE2fIcZbd3O3a7XREOSuUjHxET6/v27cPg4CC8Xq/qrjcWi2F1dRWlpaWqzoRhGJw4cQIXLlxQDeskEgkcPHhQcbsgCJidncXo6Khixskjj4hSNxzHIS8vjyxkHMehvLxcttPneV7hDCORCObm5pCXlyfvymYYaDs7wKQ1PgqhkCiceeexgtsN/OM/QXj/+9Hc3Iy1tTVyIrJarbJktfQ5ulwuTExMyHbssVgMPp8PdrtdFvayWq1Z+y/UTjPrzeqIRqMIBALo7OxEb28vQqEQdDodmpqaSH6DZVlVZ+zz+TI6E4ZhcPjwYaJ1ZjQa0dbWRkfyPkAY4QEMAKClwesjTEyIo1kPHQSTYQDQVrMbyxJ5nsfMzAxGRkZkoS+GYUh+ZmxsTPW5VqsVzc3N6OnpUYTNNBoNrFYr2tvbZWKKQ0NDGB8fz5rfYRgGNpuNnGxKS0sVYTJBEHDhwgWFQ5GaKp1OJylDLiwsRHNzM2KxmKwBUpiYAD78EaV45l/8LzAd+0kSP5lMkhBaOj09PVlVKw4fPgy32w2dTof6+vqcEtiTk5OYn59HNBpFLBaTOar0Zk8AJDRWV1eXUbhSzU6tVovDhw+rlofvVnbj32CmSZu7xpmcO1esevvzz3tQV5fE5CSHj31MvWnslVfE9371VT2eflr5462pSZKxpi+8YMKLL5rBcZzsj+Ls2Sieflo8ej/5pBUXLih3YB/5SJDMi3jssQJMTyt3V08+6cOjj0a35ZrSedvbGHz5y8u78pq+9a0Y/vZv7RAEHgADjYaD0WhER8cy3vIWUQX37//+BCkrltDpdPiP/zGGzs5LWFtbwze+cRYu1918CcdxMJlM5JpeffVVfP7zb1e191Of+lc4HD4sL1vxve+9DVqtFnq9HrFYDIIgQKvV4sIFD7mm//JfDOQ+EQbFxT58+tM/h8PhQHf3afzN31gRjUYRjycgVosxOHzYg+99Tw+G5/HUiSu4sNxx1witFjGHA+9+5zT+n7eNwbC3DY99qgJjYwIEAdDptMQZPfmkD7W1w+jr68NTT71f9Zo+97kLOHGiGF1dXnz966fAcSz0egNSN/ip39Of/IkJkUhE5jCKi/349Kd/Do7j8Ktf7cErryingba1zeN3f/cWDh48iP/232oVv71EIoGjR2/i3Dnx9Pfd7/4aXC4LGIYhn7N0TTvl7+le1oi5OZ0ir3M/r0m6bTNkciY0AU/Z0czPc5ifVzqw/Hw7TCYjdDo9jEYDSeRbrdasiXJBEMBxLA4dOqTaTc/zvGxhzLTXYhgGL710Bt/97q/JnhsKhcjgq3A4jIWFBXK/TqeD0WhMUUe++9putxvhcPiOI4mn3CcgFApheHhYzJ10dgI6PcBygE6PqMmMaCwG1+oqXh8ewuKX/zM8y647p4QoQqEQkkke8/McvvENC6qrq4ngoRqRSOROONEHnk8iHo8jGhWdhVp4MR7PPN1QGrmc7b1u376NhYUFhMNh2aLKcRx0OmUxhVQC/aAmKgqCdB00kb8eu+Zk8qDZjcfPjbIbrlHamWXaUUllwlLyHhB/b+Pj43C5XLKErEajwdGjR0n5qdfrxaVLl2QLXl5eHs6ePUtCRN3d3TKZFomCggJ89rNvAgA89dQPMtpfVFREZOaTySSuXLmS8TM3m80Zh0HZbDacudN7JCSTwOoqRl0ujKR1lxvW1hC507si4XA48MUviqMLXnnFBUEQsLCwgFu3bmW0OxWtVguO4xCLxWCxWLB//37isHt7exVz7CX0er1qlVUmTCYTTp48SZ47NTWFyclJ1V6kEydO3Pdwl8vlQn9/PwKBAPR6PcrLy2WjB7aC3fA3mA49mVB2HZFIBJFIBOFwGP39/YpFyefz4dKlS3j11Vdx4cIFItuh1WrR2tqKd73rXaipqYHVakV+fj7a2tpk+lmplVWAuBuuqamR5RpaW1sVuQeDwYBDhw5Bo9GAYZiseYVUm2dmZrIuHNmmCvp8PtJNPjk9jaVEAlGVjvuESgJcnL0uqun+/Oc/x8WLF1FUVIRHH31UUe2kVlodj8cRiUTA8zy8Xi/6+vrIfY2NjbIuc4ZhSEFCc3Ozaol1JkKhEC5evIjz58/j4sWLMBqNeNe73qUYzZyXl0eaPe8nw8PDpPovGo2u+/290aHVXJQdCc/zuH79OuJxURRycnISkUhEVvnV19dHSlHj8TgGBwexuLhIOuNZllWV6QDECqbp6WnZqUSr1aKqqkr2uMnJSUWoxmg0Qq/Xkw7ympqajPpTer0ely5dSsuVbBxBEIiMSSKRIEl/lmHAS6/L89BEokik9e4kEok7KgFiiCgej+P8+fN485vfjHPnzmFqagoul4v0nfT395PPlWVZxfWvrq6iv78fe/fuhdFoxMmTJzE2Ngae51FbWwuWZaHX6xEKheB0OknZclVVFfR6PZn3HggEFJV10r/j8TiGhoZQXl4uewzLstizZ09WyXin04mZmRkkk0k4HI6cZqCofd7pygrJZBJut/uhEWbcaqgzoexI0kNUgLiIxeNxaLVa+Hw+RS8CAFIeu7q6ine+850ZXz8QCCgWi1gshnA4LGtgVGuWTF1cGYZBWVkZJiYmVJ2F2+1eV7NLreIpE6kqAGtraygpKUF4YRGCdw0Fk1PIi8XQf0I+vVG0Vyl/0tPTg9bWVtTV1cmkY06cOIGpqSlEIhGYzWb09/cr7JuZmUFZWRlsNht6enqwtrZGHI/kwG/duiX7jlwuF06dOoXaOz1Uo6OjGBsbyxgGCwQC+NnPfib7/Hieh8vlynjiWVtbk1XreTweCIKQdXyBGgzDwGg0yn4j0kiDrcDlcmFmZgY6nQ5lZWVEMWI3Q50JZUei1sMg3d7X14e5ubmsi/Tq6irm5+czCv3ZbDaYTCZZPN5kMpGQirR4VlZWKtR100Msdrsd1dXVqrmDdBvVdvrZOvzXw2g04sh7fgNCJAJ+YQG3PR6wy8s5aZDNz8/D5XKhvr5etthyHEf6S0ZHR1UdXTKZxODgIPLy8mTy8jMzM9BqtSgrK1M4e7/fj2AwSL6TpqYm5OXlYWlpCRqNBk6nU7Z4Z/pc0sOTqUxPT8ueIwgClpeXN+xMAKCtrQ29vb0kZ1JRUZGxlHkjLC8v4/bt28TOxcVFHDhwIGthxG6AOhPKjqS4uBg2mw2/9VtXyG2FhYWIRqOYn5/PaTZ4tgVa0nsaGxtDOByGyWRCc3MzWJbF+Pg4GZvLsix0Oh0MBgOZuSElYZ988u5iuW/fPoRCoXULTdRCZnv27MHY2Jhi8TUYDNBoNAgGgyQXkaqHpdVqSRMfYzBgOBLBYoa5IamfYyrSyNra2lrV3E+2jvbV1VXVHML4+DjKy8sV5fWSzamUlZWhrEwcf6zX6zE5OYlYLJZ1Kme2rna13hq123IhPz8fZ86cQSAQgMFg2LBUTyakUcUS0ndAnQmFch9gGAZHjx6FzTaMSCQCm60ZjY2NcDqdOcnIS8n0TBIsAFBRUYGysjJEo1HiLFZXVzE6OkqclTTOV6fTwWazkUR4a2srHn1UfB23243h4WGy6G8kN2Kz2VBeXo6ysjL09/eTmetS/qK+vh6rq6tgWRZGoxHd3d1kJHJzczNMJhMGBwfB83zW5PDevUqNLYlIJIJQKKS6616voz0Ti4uLCmeQTCbR09ODffv2qcq/NDU1obKyEh6PBwUFBbhw4YLqax/I0tTb0NAAl8tFTpwajUYhjbMRGIaRhT23ArXfx4Mqdb6fUGdC2bHodDpFAr2oqAhGo1EmbaK2gEvjcbMlSwVBQCKRkM1YyXTqicVislNHKBTCmTNnIAgC+vr6ZJpfGo0mp5MTcLeCS1JBlnby0sjj2tpaWQXa0aNHwfM8qSL7xS9+QRbOe92Bp4b30mlublaE+XIh0+OXlpYQi8Vw8uRJhbMRBAHBYJCcBLVarWLjICkdZ7uWo0ePkimdlZWVO27HLzlM6do26/B2CtSZUHY06X0mOp0OLS0tGB8fRyQSydoY53K5YLfbsba2hvn5efA8j+rqatjtdvj9fvT09JDFq7q6GvX19TnvQv1+Px55pBAMw+ALX5CLRyYSCdXciBpS6MTr9SoKAqTS6PSFXnIafX19spwPz/OqoSUApOP9qad+QIZShUIhCIKAaDSK119/HQ0NDYoeAoNBbAjNdsJLh+M4NDY2YnV1VXWmidfrhc/nkzWXxuNxXLt2jYTVrFYrioqKZJIqDMOgJoNmXTgcRm9vL8LhMPR6PVpaWnZs1VVZWRkZe6zValFSUkJELncz1JlQdh2VlZWoqKjAjRs3sLS0pPoYKTxx8eJFWS7C6XSira0Ns7OzJCwUi8UwOjoKh8OBqqoqLC4uyoUXVZDEI6WcSnp+JhdHwnEcIpEILl++jPLycsWJy2g0ZpToB6AQngTEXW51dTUmJyczPq+kpATt7e147bXX4PP5yBTKwcFBFBcXK3IDmZSDJRvr6uowMjICnudhMplw4MABGAwGWQI7FZZlFfmZ9Ama0liA5uZmLC4ukkFdmRrmbt26RRSP/X4/IpEIzpw5k7WEeDspLy9HeXn5rmxazAR1JpT7wujoKBYXF8kQqPb29nsOw6jBMEzGklK9Xg+Hw4H5+XlFUlvqL0lfiKWuea1Wi4aGBtTU1GB+fl5WqZRKQUEBWajKy8tJriOXnElBQQFZwKWBWn6/H1VVVXA6nWR3XVpaiqGhISSTSdJ8KSH1hqQTjUZhsVhQVFREHCLLsmBZltglOeD0zyAcDqO7uxuHDh2ShaDMZjNWVlZUr6WyshL19fWoq6vDxMQElpeXMTQ0hJqaGpSWluKRRx7BtWvXZLYWFBQoTltqjjEajeLcuXPrLraRSEThsAKBANxu95aV8lLWhzoTypazuLgo6x8IBALQaDRoa2vb0vexWq2yEwTDMGhoaEBdXR30ej3Onz+v+jye56HX6xVhJUk2ZWpqCoWFhRnHATc0NKC5uZn8u729HQ6HAy6XC263O2tISAoBDQ8Py26PxWJIJBI4efIkrl+/jrW1NZn6sdPpxL59+1BaWgpBEDA1NZWxEGFtbQ1Hjx7F7OwsPB4PHA4HGIYFz4t5jGg0irm5Oej1esVrOJ1ODA0NYc+ePeS2lpYW+Hw+1XG94+Pj8Hq9MBqNmJmZIfd7vV7o9Xrk5+fjyJEjGBwchMvlQl5enuoYXqvVqjhl5jq/PVVKR4LjODph8QFDnQlly1Ebb7ve0KSNwvM8WJaFwWAg1VYlJSVoaWkhu2qTyaS6sIfDYSLxrrYjlhrj1GBZFoWFhbLFq6enhySWM51K9Ho9NBoNysvL4XA4VGXytVotxsfHVT+raDSK6elplJaWwuVyqQ75kpByBW63G263G0tLS0gm5X0W0ux1NVwuF8xmM8LhMGpqamAwGHDq1CksLi6it7dX5oB4nsfy8rLs5AOIznFqagp5eXno6+vDwsICBEFAIBDA66+/jjNnzshOP01NTVhbWyM5FmmkcS5oNBqUlZVhenqa/O6k0nLKg4M6E8qWoxZjz2UuxkaQ1GYlLBaLYhjSgQMHcOnSJYVDicVi8Hg8OHv2LC5dupR1YU5HGmUrhtkKwPM8Zmdn1w1t7du3Dw6Hg4T6qqur4ff7ycJssVhQX1+Pa9euZXyNQCCARCKBiYkJ1fs1Gg1MJlES/saNG2nNfUr7MuV1gsEgenp6AIintf3798PhcKC4uFjR65LttRYXF0moMxWfz4fe3l4y6jeRSODGjRtYXV2FIAiw2+3o6OjIOd8hCAKqqqpgt9uxsrICu93+UFRH7TaoM6FsOU1NTVhZWSH5CqPReE8dyIA4tyGdWCymiOF7vV54PB6ZkmxxcTHOnDmDq1evKh4fDofhcrmySsxnus/lcmFhYQEf+9gwtFptTj0Ct27dgtVqRWdnJ8xmMyorK2E2mzE7O0sGUel0uqw9NOFwGK+99lrGslue5zPKzHzqU/+a8XVTrzU9FxWJREiZ7erqKiwWS8bwX3olWbYihNSQ1tDQkCxc6fF4MDQ0lFFXLRWPx4Pe3l4Eg0Ho9XpUVlYq9NUoDwbqTChbjk6nw6lTpzA9PY14PI7a2tp7bn6rq8tNvlwQBNVFnWVZOBwOhTPx+/3o6upSfS1pdrg0tTAdacEvKlrLyTZA7HuRBBKPHj0KQAxHpZevZpOhB5D1FJVt8XY4lA5GQpoBk0gkEIvFFJVsXq8XN2/ehCAI0Gg0KCwsRCgUkoUI8/Pz0djYiNnZ2YxFC6nEYjEsLCygvLxcNdSoJjuvxsDAADl5hsNhTE5OoqysbEtkTygbgzoTyn2B4zjU19dnvF8QBLIjZVkW5eXlMrFBiclJMdSR6lR0Oh3sdrtsdyuNzVWjvr4ePp8PbrebNPyll/Lq9XqYTCaysx0aGoLRaEQwGMx48lheFhes6uoYotFoTieU9RbJiooKeDyeDTcJrodkq8Phk51E9Hq9rL/E7XZjbW1N9v7JZJI8XlIgfvTRR+F0OuFyuWAwGNDY2AiNRrNuSbWEIAiYnp5GeXm5qjxKtpJoCZ7nFaekRCKB5eVl6ky2AepMKA8UnucRj8cxNTUlk233+XwwGo2K5i1pJGn6cKyDBw9icHCQ6Ca1trbenZUuCBgfH0cwGATLsmhpacGBAwfIfPIbN24onAnHcTh16hQAsRlwampK1f7UnpLvfe9tABi8/PIwbDYbFhcXyWksE+tVGFVUVEAQBPT29uY8VCoXRFvFpsWCggJyIqiurpYt5kVFRWhubiZCmlqtVtF4GI/HkUwmUVRUhPLycsTjcfh8Plit1g0JViYSCUSjUVitVkVYMVMoLRWGYWAwGBTKvjut4/2NAnUmlAfG7OwsxsfHEY1GFSGZZDKJhYWFnDuBOY5De3u76n09PT2y6Yher5eE3ebm5lTDKpFIBF6vFzabjTS/pWI2m1FWVobq6mr09PTA6/WCYcTGxeXlZcTjcbS0tKClpQV+vx+3bt1SJP51Ol1OFUqlpaUYHBxUOBOdTgeWZXNaaCWbq6ur74wK1kAQBJSUlCi0sXieRywWI7Iy9fX15FQZjUbxy1/+UrbQJxIJXL58GZFIhFRxxWIxmEwmkqRXc4TptwuCgAsXLqhWleVSFMEwDJqbmzEwMCCbhkiruLYH6kwoD4RYLIbh4eGsC2E2VdapqSmsrq7CbDajoaEhY6WPNMAolbW1Ndy+fVu1skhCmiJos9lUX7uiooL0lhw/fhzhcBgsKyarnU4nnE4nAoEAOjo6MDMzo1qSnEwmc1KelaqyUhdZvV6Ps2fPgud54szWOwWwLEuk5E0mMWx05MgR2WPm5uaIg5e61lNnaywtLSnCdzzPq15fKBSC2+1GQ0MDFhYWiFyLVLZtt9tJc6fZbCaDvtRIrQh0u92Ynp4mVVupjYgOhwOFhYWYmJiAy+WCz+fDxMQE6urqNjwQi7I5qDOhPBDcbndWR2I2mzNWfEUiEdmoWKfTidOnT4NlWSKdUVhYSHbJarmL+fn5rPYZDAZyKqqoqIDf7ycLncViUeRz/H4/kskC2W1utxvJZDJjj4okPpnaEJiJffv2kSoljuNgNpvh9XpRXFyMY8eOkeFYIyMjEAQBhYWFmJqakjmg9UbbxuNxDA8Pk5NaLBZDf38/zp49S8JOmcqQMxGJRFBRUYGmpiYMDQ2RZLwgCETxWafTQavVZnUkkgbX0tISbt26Ra7L7Xajo6ODyNYD4mZhYmKChBeljvlsOTvK1kOdCeWBYLPZFBpWRqMRDocDWq2WlMamIyr7Kme/nz9/HiaTCaurq0gkErBYLOjo6IDdbkdBQYGsByUX8vLyyPvX1tYiLy8Pc3Nz0Ol0aGxsVJwoxIVQ7rQkR5ZNNiZXIUmr1YpTp06hu7sb8/PzCIfD8Hg8qKqqwr59+0g4Kj8/H4IgYHBwEIB4GuE4DmVlZRnDgBKrq6uqkiqhUIhUlakVDGQrm5b0xKampmTTJ9Mr1ILBoEIV2GQykVCiJLcyMjIic5CJRAJzc3MyZzIzM6NopFxaWqLO5AFDnQnlgWA2m1FRUYG5uTnE43Ho9Xo0NjZmVIGVo1y4gsGgbIHy+/0YGhrC8ePH0dHRAYPBQHpJcqmMSg+JFBUVZR2lWlpaCpblwPN3HZ3dbodGo4HD4VANA7Esu6HkcCQSwdLSEgnN8TwPp9OJ5uZmIoUyOTmJ1dVVmfOU1AEyObWVlRXMz89Do9EoHLxeryf5FK1WC61Wq8hpsCwry31I76PX68FxHK5cuYJIJJK1ui0ajaK4uJjMgTebzdi7d6/iM1d7DZfLhV/96lfgOI6cSNPZSh04Sm5QZ0LZcnieRyAQgNFolO3o9+7di5qaGqytraG4uDin3pNXX3Xj6tWrWGeAIQCQRY/jOLS1tSE/Px8/+tGPFAt7egUQkFspaiosy+KVV5YxPDyMaLQQFouFhK9aW1uh1+sxPDysaOK7du0abDYbkcJf73rSK8NisRgpYHjttdcyytSkNy5K1XCTk5MYHh4mDlav15MTgl6vR21tLckZcRynmndIT67zPA+j0UjyTrnAMAyZ1SLJ7DMMA5/PB61WS76PpqYmLC4uyhwez/PkxOT3+0mRgXTK0mq1tHFxG6DOhLKlLC8vY2BgAMFgEAaDATU1NbJcSF5eXs4CfhIHDx7E1atX19X3Si+75XleNeEeiUTAcRzpObHb7ap5jHA4jP7+fkQiERgMBuzdu1fmdMxmMw4ePKh4nhSCmpubUyzqXq8XXq8XTqcTHR0dsmTy6uoqbt++jWg0SmTa008BFosFPM+jq6sr6+chOSHp+qX5KiMjI7KTWjQaRWNjI1EaTnXwvb29OVeOqVXIpYbDtFqtbPYMwzBYWFhAUVERVldX0d3dfScPJU6ZLC0tJbmR/fv3o6urK6v8y9GjRzExMUFm1mTqOaLcP6gzoWwZUiOiVNYpdSRnakzLhVdf1QPQ45FHxPyBFLZSKz1Nn3URjUYzVjwlk0kiDGmxWBQ78Hg8jsuXL8tyBpFIBKdOnSKL5D/+oygd/+5362TTECWkcJdaqEYSQpScidfrxZUrV2TXlWq7VquF1WoFx3G4fPnyuj0o4XAYN27cgM/ngyAImJxshlarRWGhsgdGEAQySz6VjWiWqcFxHARBAM/zyMvLQ0FBAekt4nmeTLVcWVmRObhEIoH5+XkUFxejoKAAJSUlMJlMGe3RarUkZ0bZPqgzoWwZaiq80WgUbrcb1dXV9/SaTz8tdjK/8koUnZ2diEajRMgw9b00Go0ibCRNCczURBiPx6HVahWOxO12o7u7W3EtPp8PgUAAFosFt27dwp/92SkAhXA4/hF1dXWKHpKWlhbo9XoiS5+ezE51CJOTk1kdhNVqRWNjI65fv57T4K1EIiGTNXnuuSYwDIsnn7yheGxqMjuVXMqY17NBYnV1VTWZn6nbXxAE0u/DMAzKy8sxPj6u+IxMJtM9675RthaapaJsGTqdTpF70Ol0WRPZG0Wv18Nms6GsrIwkWRmGgcPhUFRKMQyD1tbWjBVUHo8HFy5cwNWrV2UL2ujoaMYphhqNBj6f744ir3jikBou0xc6hmFQV1eHI0eOoLa2VvF6kpMLhUKqjZKpsCwLj8eTkyPJjHpCvKurS3UAVWNj44ZOlDqdDnV1dSgvL1d1RGpOPZuadOppr7m5GQcPHkRVVRWamprQ3NyMpqYmnDp1at3cE+XBQE8mlC1D6kgeHBwkOZOqqqp7DnFlo62tDcXFxVhaWkJBQUHG3bXD4UBRURFu3bqFxcVFxf3xeBwulwuDg4NEpTZTaKyoqAhGoxFer1exm45Go7h9+zYSiQTy8vLQ2toqa37MJN4YCARw7dq1rJpdUmIaEBffXKvT0sNrLMuqCkmGQiGMjY0pGhqLiopw+vRpzMzMYHZ2NqsAJQCSV/J4PGny93dtslqt5FpMJhP27NmDvr4+1Yqx9NBhSUkJnZy4g6HOhLKllJaWori4GH6/HyaTKev88M1SXFycU6kty7KwWCyqzkQidaE0m82yCjCp8kiaFFlcXKwYOwuAvL7L5UIoFCKLsyAIqtVi0jAsNUdiMBhgsVjgdrshCAKi0SiZVJhIJMDzPARBUOz2JZ2ykZERxeuyLIvTp0/j9ddfV1S4ZWsgbGxsRGVlJQYGBogjqKqqwujoqOx5VqsV4+PjitslpOR6RUUFOI5Da2srdDodvF6vYliYTqe7r78dytZDnQlly+E4bseFHurr67G0tJSxdDW1Eqy9vR2JRAJ+v59MR2xpaSH3cxyHffv2kQSzzWZTvO7q6irm5uZIV7rRaEReXh5JIktluJnk2i0WC4xGo+J0IT3fZDKhqqpKMf63pKQElZWVGBkZUbwmx3Fkxn1PT48sZLbe92UwGBSVawzDEEFIu92OvXv34vXXX896cpLCeZLG2OHDh1V7ctRG8VJ2NtSZUHY0NTXKpHQwGEQikSBqs7mg0Whw4sQJTE9PIxwOY21tjVQ6SaXB0o7fYDDg+PHjiMfj4DhOtQGuqKgIe/aIZbR79uzB1atXZQt/ulRJOByG3W5HW1sbIpEIampqYDabwfO8are+TqfLqjAcCoUwOTkpu43jOLS0tGBiYkKR8ykvDyM/XzwdVVZWIhwOY3FxEYIgoKCgQOYsJQRBwOLiItbW1lBRUaEQUKyvrycOLRKJZJ1Ln4407nd2dlb1O8xVKYCyc6DOhLKjeeGFu4lpqb9CmktitVpx6NChdRsOxdLYSbjdbmi1WjQ1NcFsNsPn84HnedhsNgwMDJBuc4vFgoMHD64bZnnhBQ8EQUAkYoZWq5XlWgRBUCzoXq8XNTU1MpkPtdyCXq9HS0sLdDodlpeXM/aTpC/ckuikmjjj00/fxoEDB8i/GxsbieKx3+/H9PS0TH8sFovhwoULpM9kamoKLS0tRDhSusbU/h+n07nhIWherxdVVVXweDzk89PpdPdc/UfZPqgzoewaJicnZWGhtbU1DAwM4NChQ1mfNzw8jPHxcbLArq2t4dSpU2SnPTc3R1RpATGZ3t/fL1t8UwkGg5iYmEA0GkUgEEAsFlPdkacnwQVBQH9/PzQaDSkYUOudKC4uJkULx48fx/Xr17G6uqqo5EpvaARAGh3TYVkWq6urGB0dJYu25EwBcVHX6XRE8qavr08WruJ5HhMTE6ivrycnCZfLpQjvMQwDvV4vS6hrNBpyW+prShpixcXFOHDgAGZmZgAA1dXVdCbJLoQ6E8qO5oUXxEX1scdCqrPNc+nQXl5eli3qwWAQU1NTRFJeSnKn4nQ6ce3aNTQ2NsqqirxeL+lxeeWVNgAFOHduQPV9zWazwlkkEgn09vaiqKgIWq1WdSefOiXQ5XIpHAnDMMjPz4fdbsfk5CSx3W63o7S0lCgKSw7OaDTi0qVO/N3fzeHkSeVJCLh7qikrK8PY2Jhq3kMaiiWV80pFAKnEYjE0NjYimUzC7/eTU5bBYEA0GiVzXiSHKjmNXIspKDsX6kwoO5oXXxSrph57LASbzaaQkt+oppYaavH5ZDKJ5eVlBAIBnD59moS8xsfHSfjq/HmxuivdmUgNlJlKjGOxGHp6enDo0CE0NzcjHA4jEAiQctjUnhSXy6U4keTl5eHkyZOw2+0wmUxYWVmB0WhEU1MTWJZFRUUFdDod5ubmwLIs6uvr8cQTNkSjepw8eSvj5+Dz+RCJRDI6aK1WK+sLcTgcslJfAESyRZoLE4lESNm1Xq9fNxdF2b1QZ0LZNdTW1mJtbY0ssDabjZTrZkMqVZZ20WazWbZg19bWwu12Y2VlRbFwh0IhTE9Po6mpCQBy7j73+/2qEwQlpBNLfn4+zpw5g8XFRRgMBhQWFsoS0mqzzKXEvFSyrNYQmb7TZ9n1CxUkAUmj0aiosGIYBp2dnbLbWJbFkSNHMDg4CKfTKftsUsNfUjhNCkeu11nP8zzm5uYQCARy+n4pOwPqTCi7BpZlcfDgQUQiESQSCaI0ux6Siq/L5QLLsqS/QYLjOBw9ehSrq6vo6upS7MxTQ1FlZWVwuVzramNlcySAfEHlOA6VlZWqj6uurobL5SKDtywWS06jf9NJL7WVPrf0cbzhcBgtLS0YGhpCKBQCy7KwWq3o7OxUbT41Go1ob2+H2+3OOvnR6/VCEIR1vy9BEHDt2jUyLXN+fh5NTU2qDpOys6DOhLLryFYyqwbDMGAYBqFQCJFIBF1dXaSDPvUxBQUFqK6ulmlA2Ww2mQhiRUUFYrEYFhcXwXEcOE6DAwcOYGBgYF0Hkvpe6ZMbMyHt/ldWVjA7O0umSfr9fiwtLSEvL0+1gVINo9GE5uZmhEIhOBwOjI+Py04QRqMRBQUF4DgODocDwWAQJpNp3X4PnU4Hm82WccIkkFnOPp2FhQXZ2OVoNIqZmZkNORNpvkt+fj7Kysro+N4HBHUmlIeeaDSK8fFxcuLw+/0YHBxEUVGRYqFpbm6GxWLB0tISGeCVvpjW1dWhrq6O7NQrKiowPz8vK/PVarXIz88nZcypFBYWZpR/USOZTGJoaIjoZ83NzQEQw0FarRZ1dXWkmAAQczJ9fX0IBoMkAQ4Ug2Ege5zJZML169eJE4zFYhgbG0NLSwtRDUglHA6T6Yl1dXUyJ3bgwAHynjqdDtFolORSOI5TVSVWQ626LRf5GImRkRFMTEwgkUhgamoKCwsLOHz4cM7Pp9w71JlQdjRnz+a228+G2vz5cDiMaDSqOOWsra2B4zjs379/3QRxqm0dHR3o7u5GIBAgM8yrqqrQ3d2N2dlZ2fM2Os9lZmZGJsSY6pzi8ThmZ2dRV1dHQme3bt2SnRJCoRDOnKlUOM704oVkMom5uTnU19cr8hp+vx/Xr18nEi1OpxMHDx4klW46nU7WIZ9MJjE1NYVgMIjy8vKcxT6rqqowMzMjO+Xl2sCYKmsPiCEzqRouPz8/p9eg3DvUmVB2NE8/rSwH3ij5+fmK8bQGg0GWN+F5Hjdu3CAnCalxMdtClmqbXq/H0aNHFY9pa2tDKBQizqCgoGDDOQ81BeNUJFl+aWJiegl1IBDA448PKU4HoVBIEZpLfa1U0jXEIpEIJiYmVOe4AOJpJLXBMVdMJhP27t2LyclJxOPxDX1eiURCtZEzGAxSZ/IAWNeZvPTSSxgZGUFxcTE++clPktLAGzdu4OWXXwbHcaivr8dHP/rR+24shXIvmEwmVFdXY3Z2FtFoFCaTCQ0NDbKTx+TkpCxMJYXC1BzERtBqtTh+/DipjlKrzlqPmpoazM/PZ8zJmM1mEnJiWVYRlmNZVjXPZLVaZXph0mupnZzUCg7WK0K4V8rLy8mgs/z8fFV5fDV0Oh0sFgtWVlbIbSaTCaWlpffFToqcrOf4qakpeDwePPPMM6ioqMCVK1fIfbW1tXj22Wfx7LPPwuv1kglqFMpW8uSTVjz55MYX4HRaW1tx5swZHD16FGfPnlVUT6mJDWarTtqIbZL0+r04EkBc4Pfu3YuCggJYLBYUFhbCarXCYDAQgUXJMXIch5KSEpmjLCwsxJ//eZ3CVo7j0NbWBrvdDr1eD7vdjvb2dtWEdUlJiaIabCc2GXZ2dhJV54KCArS3t2edmULZOrJ+yiMjI2QUZmdnJ1555RWcPn0aAGQxUI1GQysmKPeFCxc2pvWUDYPBkLESrLCwEPPz87JS2fXmsORqm9vtxuLiIqxWK6qqqhS5GJ/Ph6mpKWg0GjQ0NKh2xafu1gExHyCdKtL/9tra2pCfn08k62tra/GlL6nb6nA4UFxcDJ7nwbJsxr9ju90Os9mMUCgEjuNQXV2dc0WaZG8ikbjva4XRaMSxY8fu6blra2tkjnxFRcWGiiQo6ziTQCBApKkzzWAeGxuDz+eTidcBwOXLl3H58mUAwOOPP77rYpZSNc7DzG64Rmk3fC92buT67HY7QqEQFhYWkEgkkJ+fj9OnT2cVLszFtu7ubgwMDJCksMfjwblz58iCOj8/j2vXrpECAZfLhba2NuII2traMjb5abXajGKU6bkMluUQj8cwOzuLpqamDSkHSKKPUi5GupbU95AqwSSHmHqKWVhYwK1btxAKhWAymdDZ2ZlzdVf6dzg/P4/Z2VlYrVa0tLRsmUy9x+NBV1cXyQutrKxAr9ffU95nI+yGv8FcyepMzGYzSf6FQiFFLHVlZQUvvvgiPv/5zyuee+LECZw4cQLA3Wl2u4mNxGp3K7vhGpNJMZRyL3Zu9PpaWlpQX1+PZDIJg8GAUCiUdQKiZFtfXx9mZmaQSCRQUFCAtrY2MAwDnucxPj4uK21dXFzE+Pg4CgsLyXNTK838fj+uXr1K/j07O4sTJ06AZVlSmsvzPGpqalBTU5PT9SWTSQQCViSTSXR3d2N4eBj79++Hw+HI6XMZHx9XJPUXFhbIe/t8PnR1dZHNppRrMhgM4HkeV69eJfdFIhFcu3YNZ8+ezUlOJfU7HBkZkfUATU1N4fjx41ty0unp6ZF91/F4HCMjIxkLDLaK3fA3mE6mjUjWb7OlpQW9vb0AgNu3b8tmHoTDYXz729/Gxz/+ccWcAwplt6LVatdtilxdXcXrr7+OYDCIYDCInp4erKyswOv1YnJyEufPnwfP8+B5XpGk5nleNtVxvSS21IAXCoVw+fJlTE5OYnp6GlevXlWdg6LG5OSk7H0ikciGcpxqi35qOHB0dFQWtfD5fBgdHQUgRjfSx/0GAgHVHFU2BEHAwsKC7Do8Hk/G4WIbJV2wkrJxsjqT2tpa2Gw2PPHEE5ibm8Px48fx/e9/HwDwk5/8BMvLy3j++efx1FNPYWBAXTmVQnmY4HkePT098Hg8GR1GIBDAyMgINBqN4jRvMplkuQ/phJKNaDSKsbEx2c45Go0qpixmQu10FY/Hsba2hv7+fkxPT6tqji0vL+PSpUuYmJhQ7P5jsRgpUFCT35fuMxgMilChwWDYsECnlHNJvy0X1ehcqKmpkdnJcRytAtsg65Y5fPjDH5b9+xOf+AQA4P3vfz/e//733x+rKJQ7fOQjwfUf9ADx+/1kV/3II5k3UNLAqM7OTllneFNTk6y6qKmpCTzPw+12k/kkqRInUlmzFCFIJdfO8PLycrzpTcMyp8cwDK5cuUJeY3FxEceOHSNOI5FIoL+/X3GqSH1vScbfZrPJJFAYhiFOUqfToaqqikxh1Ol0qKys3PB8d5ZlYbPZZM7DZDJl1DTbKPn5+ejs7MTU1BQEQUBZWRmqqqq25LXfKNCaOcqO5rHHMucstgODwUCaAzPNMQFAFkuDwZBVzoNhGLS2tpJKImleSDKZhE6nQ2NjI7RaLcrLy7G8vEwWf4ZhFDtnQRDg9XqRSCRk6sNFRUV4/HHvne5wE2w2GwKBgMwZrayskHkmgFgIkMmRSEjhr5aWFsRiMXg8HjAMA4fDgZqaGvK4lpYWVFZWwuVykbLde0FyzH6/X/bZbBV0psrmoM6EQtkAer0eZWVlZCqgGnl5eTINrPXw+/24ceMG2XVzHIe9e/fKRteWlpZiz549mJ+fB8/zKC4uRnt7OzkBJZNJXL9+HR6POErYarXi8OHDJJzU0NBAKpOSyST+9V//VWaDIAgy52E2m6HRaDKefqSSY0B0Kh0dHSTvoJYQT22svFe0Wm3G6ZeU7Yc6E8qO5rHHxGqa1Fnw283+/fvBMAy++MVWCALwh394nmhj8TyPysrKDTXKTU5OysI3yWQS8/PzijnoUgWXROqiPTo6Kgs1eb1eDA4OEr2s1M9xbGxMkedhGAZutxtOpxNmsxnt7e0oKSnB4uIieJ6HRqNBYWEhWJaFVqtVhOvS7aG88aDOhLKjmZ7emj6CrWbfvn0IhYqQTCZx9uzZDecAUtmKSiK1kFSq/Erq55g+tx2460wAMd8Ti8Vw9OhRlJeXw+PxoLy8nFZtUrJCnQmFco8wDAONRrMpRwKIVZPLy8tk8b+XSiK73Y7FxUXZbZk6+NXsTa/m8vl8SCQSKCkpQUlJSU42+P1+TE5OgmEY1NfXbzqsRdldUGdCoWwzNpsNBw4cwNTUFHieR0lJiSyclQt1dXXwer1k9LDdbs848ra1tRU+n480IppMJkQiEZlD2eiM9pWVFXR1dRGHuLy8jMOHD9PTzBsI6kwolB1AQUEBYrEY4vF4zlIjqUgjjWOxGHiez9p4aTAYcPr0aXKSKS0txfXr10mYi2GYnGeISExOTsrCalK3Pk2Yv3GgzoRC2Wbi8TiuXr1KKrMmJydx8ODBe1IZzjXkxrKszGkdOXIEk5OTmJmZQSQSwdLSEi5evIh9+/bl1Fip1vSodhvl4YU6E8qO5sknNz8c636xVbaNjo4SRwLc7aDfynGz2WwVBAGjo6NEtiXdDkljLxsOh0M2oph2kL/xoM6EsqN59NHNj+29X2yFbV6vV7VnZb1ZKhslm61jY2MYHx9XrSrLVa6ktrYWyWQSTqcTDMOgvLz8nsJ1lN0LdSYUyjaSKk+fisViwfT0NGZmZhCPx2GxWNDR0bHpyjE1VlZWMpYnb+T9UhsjKW88ci/XoFC2gXPninHu3M6UuNgK29R2/mazGZWVlRgaGoLX60UoFMLS0hJ6enru+X2y2ZptJojX693U+1LeOFBnQqFsI+lVVxzHob29HYuLiwo13lxk2xOJBAYGBnD9+nUMDQ3lNKe9oaEhY/UXz/NYWFhQzDOhUNKhzoRC2Ub27NkDq9UKhmGg1+tRWVmJoqIi1dLc9UJOgiDgxo0bmJiYwNLSEsbGxtDV1bWuDQUFBTh58iQcDofqKSWRSFBnQlkXmjOhULYRu92OM2fOwOfzQa/XkxNCRUUF5ufnSe+HXq8nwoqZ8Pv98HjkGmYejyenJLrJZEIymVQ9yRgMhpynMlLeuFBnQqFsMwzDKDrFWZbFsWPHsLCwgEAggKqqqozyKBJqSXRBEHLW/lIbcqXVatHS0nJfEv+UhwvqTChveARBwNzcHFZWVmAymdDQ0JA1Kf2gYBhmQ+W1VqsVNptNNlM8Pz8/56mGVqtVFs7SaDQ4fPhwTk2L6yEIAmKxGLRa7YZkWii7B+pMKDua55+//9LzAwMDZMIeIJbKHj9+fF1J9Qdh20ZgGAZHjhzB4OAgIpEITCYT9uzZAyA3W/ft2wee5+H1esFxHKqqqrbEkXi9XvT29iIYDEKv16O6uhr19fWbfl3KzoI6E8qOpq5u/WqkzZBMJrG0tCQLBXk8HrhcrnXzBPfbNgmv1wuXy4XS0lLFTPl0dDodOjo6FLfnYivHcWT+yVbS399POvzj8TjGxsZQUlJCVYUfMqgzoexoJifFcNP9WrgFQVAknaWQzHbbBgB9fX2Ym5tDIpHA+Pg46urqNjTFUeJB2Lq6uoqRkRHE43Hk5eWhvb0dLMvKJFoAsbt/cXERjY2N980WyoOHOhPKjuZjHxMnBL7yiuu+vL5Go4HdbsfS0hK5LS8vj8xC307bwuEwFhYWSId8PB7H3Nwc6uvrNzTJEbj/tsbjcdy+fZsM6VpbW0MikcChQ4eg1+tlFWUcx6GgoOC+2EHZPmgmjPKG58CBA6isrITdbkdJSQkOHDiwIxLwwWBQcUKKRqM562U9SObn5xXTHtfW1iAIAhoaGkgRgEajQXl5OXUmDyH0ZEJ5w6PRaNDZ2bndZijIz8+H2WyWLdJms3lH5hr0ej0YhpHlnjiOI6KPxcXFWF5ehsViuSdpfcrOh55MKJQdCsdxaGtrg81mg8FgQH5+Pvbt27duldl2UFpaivz8fPJvlmVRVlZGbNVqtaioqKCO5CGGnkwolB1MSUkJHA4HBEHY0f0ZDMPg2LFjmJiYQDAYRElJSU55J8rDA3UmFMoOh2GYHXkaSYfjODQ1NW23GZRtgjoTyo7mflUfbQU72bZ0dpOtlN0JdSYUykNIJBLByMgIYrEY8vPzUV9fvytON5TdC3UmlB3Nq6/qAezM8b071bZkMolr164RnS2n04kLF4yoq6vbcbZSHh6oM6HsaJ5+Wqz+efTRnRem2am2zc/PK+aP/I//0Qiz2bzjbKU8POzc8hAKhXJP5Co5T6FsJdSZUCgPGZWVlYpJjet19AeDQSwtLeU05pdCUYOGuSiUhwyO43D48GEMDw8jHo+Tpkc1BEFAT08PnE4n4vE4zGYz9u7dSycrUjYMdSYUykOI2WzOSU5+ZWUFCwsL5EQSDAYxMjJCnQllw1BnQtnR1NTs3LDLTrYtnUy2rqysKEJb0WgUgiDQUmLKhqDOhLKjeeGFnTXNMJWdbFs6mWwtKyvD5OQkkbkHAJPJRB0JZcPQBDyFch8QBAEul0sWQtqJWK1W1NfXw2Qykdkue/fu3W6zKLsQejKh7GheeMEEAHjssdA6j3zwZLJNahr0eDwQBAEWiwUHDhzYVsXcbJ9jc3Mz6uvrEY/HYTAY6KmEck/QkwllR/Pii2a8+OLOm98BZLZtfHwcKysrpN/D7/djZGTkQZsnY73PUaPRwGg0UkdCuWfoyYRC2WLSJw4CYlJ7t5BMJjExMYFQKITS0lKUlJRst0mUXQB1JhTKFlNYWIiFhQVZJ/pOnI6oBs/zuHr1KjweMWG/sLCAhoYGNDc3b7NllJ0ODXNRKFtMVVUVampqYDKZoNfr4XA40N7evt1m5cTi4iJxJIB4Skl3jBSKGvRkQqFsMQzDoL29HclkEjzPQ6vVbrdJOROJRBS37eRqNMrOgToTyo7m7Nmdm2tYzzaO49bVxHpQ5Po5VlVVYXp6GqHQ3aovm81GE/OUdWGEB3B+jcfjcLl2l/R1fn4+VldXt9uM+8rDfo30+u6NlZUVjI2NIRaLwWKxoL29HRrN9uw76Xe48ygvL1e9nZ5MKBSKjMLCQhQWFm63GZRdBk3AU3Y0Tz5pxZNPbl+zXzZ2sm3p7CZbKbsTejKh7GguXNBvtwkZ2cm2pbObbKXsTtZ1Ji+99BJGRkZQXFyMT37ykyR2yvM8nnvuOTidTtTX1+Oxxx6737ZSKJQ78DyPqakphMNhlJSU0AQ5ZdvJGuaampqCx+PBM888g4qKCly5coXcd/PmTRQUFOCZZ55BJBLZdrkICuWNQigUwmuvvYaLFy/ixo0buHTpEmKx2HabRXmDk9WZjIyMoKOjAwDQ2dmJ4eFhct/w8DD2799P7hsaGrqPZlIoFImhoSH4fD7y77W1NbqZo2w7WcNcgUAAdrsdgDjjIBAIkPuCwSBMJpPqfQBw+fJlXL58GQDw+OOPIz8/fyvtvu9otdpdZ/NG2Q3XKPVp3Iud9/v6NmPbZuB5XnFbMpnMasd22bpZdsNvdDM8TNeX1ZmYzWaEw2EA4tE6Ly9Pdp/U2JR+HwCcOHECJ06cAED7THYqu+EaP/QhccOyurpxCfr7fX2bsW0zGI1GxW15eXlZr3W7bN0su+E3uhl24/Wp/f6AdZxJS0sLfvzjH+ORRx7B7du30dLSQu5rbm5Gb28v2tra0N3djXPnzm2txRQKduYcE4ntsq21tRXRaBRra2sAgKKiItTX12d9zk7+HCkPB1lzJrW1tbDZbHjiiScwNzeH48eP4/vf/z4A4NChQ3C73XjiiSeg1WqpqiiF8oDgOA4HDx7Ee97zHjzyyCPYv38/reaibDtUTiUDu/H4uVF2wzU+9lgBgHubt36/r28ztm0FG7m+7bb1XtkNv9HNsBuvj8qpUHYl09M7QyhRjZ1sWzq7yVbK7oTKqVAoFApl01BnQqFQKJRNQ50JhUKhUDYNdSYUCoVC2TQ0AU/Z0Tz5pG/9B20TO9m2dHaTrZTdCXUmlB3No4/u3LG9O9m2dHaTrZTdCQ1zUSgUCmXTUGdC2dGcO1eMc+eKt9sMVXaybensJlspuxPqTCgUCoWyaagzoVAoFMqmoc6EQqFQKJuGOhMKhUKhbBrqTCgUCoWyaWifCWVH8/zzO1cyfSfbls5uspWyO6HOhLKjqatLbrcJGdnJtqWzm2yl7E5omIuyo5mc5DA5uTNncexk29LZTbZSdif0ZELZ0XzsY+KEwFde2XmTOneybensJlspuxN6MqFQKBTKpqHOhEKhUCibhjoTCoVCoWwa6kwoFAqFsmmoM6FQKBTKpqHVXJQdzU6uPtrJtqWzm2yl7E7oyYRCoVAom4Y6EwqFQqFsGupMKBQKhbJpqDOhUCgUyqahzoRCoVAom4Y6EwqFQqFsGupMKBQKhbJpqDOhUCgUyqahzoRCoVAom4Y6EwqFQqFsGupMKBQKhbJpqDOhUCgUyqahzoRCoVAom4Y6EwqFQqFsGupMKBQKhbJpqDOhUCgUyqZhBEEQttuInUgoFILJZNpuM+4rD/s10uvb/Tzs1/gwXR89mWTgueee224T7jsP+zXS69v9POzX+DBdH3UmFAqFQtk01Jlk4MSJE9ttwn3nYb9Gen27n4f9Gh+m66M5EwqFQqFsGnoyoVAoFMqm0Wy3ATuBl156CSMjIyguLsYnP/lJaDTix8LzPJ577jk4nU7U19fjscce215D75FM13fjxg28/PLL4DgO9fX1+OhHP7rNlt47ma5R4kc/+hGuXLmCr33ta9tk4ebIdn2XLl3CL3/5SwiCgN/+7d9Gc3PzNlp672S6xlgshm9+85sIh8PgOA6f/exnYbfbt9fYDRIKhfDss89ibm4OX/nKV1BdXU3ue1jWmTf8yWRqagoejwfPPPMMKioqcOXKFXLfzZs3UVBQgGeeeQaRSAQjIyPbaOm9ke36amtr8eyzz+LZZ5+F1+vF+Pj4Nlp672S7RgAIh8OYmZnZJus2T7br83g8uH79Op544gk89dRTu9aRZLvGW7duobq6Gk8//TQeffRR/OpXv9pGS+8NnU6HL3/5yzh+/LjivodhnQGoM8HIyAg6OjoAAJ2dnRgeHib3DQ8PY//+/eS+oaGhbbFxM2S7vqKiInAcBwDQaDRgGGZbbNws2a4RAH7605/i7W9/+3aYtiVku77bt29Dq9XiT//0T/Gd73wHkUhku8zcFNmusbS0lFxXIBCA1WrdFhs3g0ajyWj3w7DOANSZIBAIwGg0AgBMJhMCgQC5LxgMkoai9Pt2C9muT2JsbAw+nw/19fUP2rwtIds1hkIhzMzM7NodO5D9+rxeL/x+P/74j/8Yzc3N+NnPfrZdZm6KbNdYVlaG+fl5fO5zn8MvfvELnD59ervMvC88DOsMQJ0JzGYzwuEwAHHhycvLk90XCoVU79stZLs+AFhZWcGLL76IT3/609th3paQ7Rp/8pOf4B3veMd2mbYlrPcb3bt3LxiGwb59+zA7O7tdZm6KbNf46quvoqWlBd/85jfxwQ9+ED/4wQ+2y8z7wsOwzgDUmaClpQW9vb0AxJBBS0sLua+5uZnc193djdbW1m2xcTNku75wOIxvf/vb+PjHPw6bzbZdJm6abNfodDrxwx/+EF/5ylewuLiIl19+ebvMvGeyXV9LSwumpqYAiHmHkpKS7TBx02S7RgAkRGSxWMjC+7DwMKwzAO0zAQD89V//NUZHR1FUVIRPfepTeP755/GJT3wCyWQSzz33HJaWllBbW4uPfexj223qPZHp+n7wgx/gF7/4BcrKygAAH/jAB9DW1rbN1t4bma4xlS996Uu7tpor2/X97d/+LUZGRqDVavGZz3xmV+YUgMzXGAqF8K1vfQvxeBw8z+P3f//3UV5evt3mbpivfvWrmJqaQlFREd761rdiZGTkoVpnqDOhUCgUyqZ5w4e5KBQKhbJ5qDOhUCgUyqahzoRCoVAom4Y6EwqFQqFsGupMKBQKhbJpqDOhUCgUyqahzoRCoVAom+b/B+0+0iMScexWAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 480x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot data\n",
    "plt.figure(figsize=(8, 8), dpi=60)\n",
    "plt.scatter(moons_X[:, 0], moons_X[:, 1], cmap=plt.cm.Set1, c=moons_y)\n",
    "\n",
    "# Plot explained instance\n",
    "plt.scatter(instance_m[0], instance_m[1], c='green', marker='*', s=500)\n",
    "\n",
    "# Plot x-axis quartiles\n",
    "x_quartiles_m = q_discretiser_moons.feature_bin_boundaries.get(0)\n",
    "for i in x_quartiles_m:\n",
    "    plt.plot([i, i], [-0.1, 1.1], linewidth=2, linestyle='--', c='blue')\n",
    "\n",
    "# Plot y-axis quartiles\n",
    "y_quartiles_m = q_discretiser_moons.feature_bin_boundaries.get(1)\n",
    "for i in y_quartiles_m:\n",
    "    plt.plot([-0.1, 1.1], [i, i], linewidth=2, linestyle='--', c='blue')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "VezOVcpKcUbi"
   },
   "source": [
    "We can also get a discretised version of the explained instance *m*."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1898,
     "status": "ok",
     "timestamp": 1598703181396,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "9VS2cIdrSq5P",
    "outputId": "610c0ab3-5d01-464f-af7d-9646fff6094f"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 3], dtype=int8)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "instance_m_discrete = q_discretiser_moons.discretise(instance_m)\n",
    "\n",
    "display(instance_m_discrete)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "amOci5KpvIFN"
   },
   "source": [
    "The faithfulness of this discretisation (with respect to the labels of the\n",
    "training data) seems to be really good.\n",
    "In this particular case, it is possible to assess this property by means of\n",
    "visual inspection since the data set is two-dimensional.\n",
    "However, when dealing with a higher number of dimensions, we need a proxy to\n",
    "measure the quality of our discretisation.\n",
    "To this end, we propose two metrics that can be applied to each individual\n",
    "hyper-rectangle:\n",
    "\n",
    "* **Gini Index** for *crisp* classification tasks; and\n",
    "* **Mean Squared Error** for *probabilistic* classification and regression.\n",
    "\n",
    "Let's count the labels and measure *Gini Index* of the hyper-rectangle\n",
    "enclosing our explained instance *m*."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "GHmgp_Bry83M"
   },
   "outputs": [],
   "source": [
    "moons_X_hr_m_indices = fatf_ecml.get_hyperrectangle_indices(\n",
    "    moons_X_discrete, instance_m_discrete)\n",
    "\n",
    "moons_X_hr_m = moons_X_discrete[moons_X_hr_m_indices, :]\n",
    "moons_y_hr_m = moons_y[moons_X_hr_m_indices]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1592,
     "status": "ok",
     "timestamp": 1598703181818,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "jeYeTpBXWFdy",
    "outputId": "610391b8-ea4d-4827-ed7a-92ff053869fa"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([0, 1]), array([134,   1]))"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.unique(moons_y_hr_m, return_counts=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 985,
     "status": "ok",
     "timestamp": 1598703181821,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "w-0GdlzBXWxJ",
    "outputId": "126b33af-7daa-40ea-b97a-dbf99b2d2ae0"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.014705075445816198"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.gini_index(moons_y_hr_m)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "h7S_7ov-XYrN"
   },
   "source": [
    "It seems to be fairly pure, which agrees with the label count and our\n",
    "intuition based on the visualisations.\n",
    "\n",
    "Now, we need to do the same for all of the hyper-rectangles.\n",
    "Since some partitions may hold a relatively small number of data points, e.g.,\n",
    "the top-right and bottom-left hyper-rectangles, this observation needs to be\n",
    "taken into account to avoid biasing our metric.\n",
    "To this end, we employ an average **weighted** by the number of instances in\n",
    "each partition."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1561,
     "status": "ok",
     "timestamp": 1598703184977,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "zwqDFBYdpaP1",
    "outputId": "20998b3b-f49d-48fe-a883-cc029a1347e8"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.13028591315141477"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.weighted_purity(moons_X_discrete, moons_y, 'gini')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "eZVSu5OZqqlN"
   },
   "source": [
    "This number is consistent with what we have seen so far -- it is a pretty good\n",
    "discretisation.\n",
    "\n",
    "Next, we move on to **binarisation** -- creating a binary interpretable\n",
    "representation with respect to the explained instance."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "TMIHZfjtdUJn"
   },
   "source": [
    "#### Step 2: Binarisation ####\n",
    "\n",
    "So far so good.\n",
    "However, discretisation is not the representation used for fitting a local surrogate model.\n",
    "While Step 1 is universally applicable to any instance we may wish to explain,\n",
    "Step 2 is specific to the explained data point.\n",
    "To be more precise, it is specific to each unique *hyper-rectangle*, i.e.,\n",
    "a partition of the feature space resulting from the discretisation step.\n",
    "\n",
    "---\n",
    "\n",
    "The binarisation procedure is *XNOR* -- 1 if the same, 0 if different --\n",
    "applied between the discrete representation of a data sample and the explained\n",
    "data point.\n",
    "Therefore, for each dimension -- *x* and *y* axes independently in our example --\n",
    "it encodes whether a given partition of this feature is the one holding the\n",
    "explained data point.\n",
    "\n",
    "Before processing the data, let's see a visualisation of this step for our\n",
    "explained instance *m*."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 423
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1868,
     "status": "ok",
     "timestamp": 1598703185954,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "Pmvh7PJ9cd4N",
    "outputId": "47deb6c9-1e48-43be-9715-ddbffbd91209"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.patches.Polygon at 0x7f568de82910>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 480x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot data\n",
    "plt.figure(figsize=(8, 8), dpi=60)\n",
    "plt.scatter(moons_X[:, 0], moons_X[:, 1], cmap=plt.cm.Set1, c=moons_y)\n",
    "\n",
    "# Plot explained instance\n",
    "plt.scatter(instance_m[0], instance_m[1], c='green', marker='*', s=500)\n",
    "\n",
    "# Plot x-axis quartiles\n",
    "x_quartiles_m = q_discretiser_moons.feature_bin_boundaries.get(0)\n",
    "for i in x_quartiles_m:\n",
    "    plt.plot([i, i], [-0.1, 1.1], linewidth=2, linestyle='--', c='blue')\n",
    "\n",
    "# Plot y-axis quartiles\n",
    "y_quartiles_m = q_discretiser_moons.feature_bin_boundaries.get(1)\n",
    "for i in y_quartiles_m:\n",
    "    plt.plot([-0.1, 1.1], [i, i], linewidth=2, linestyle='--', c='blue')\n",
    "\n",
    "# Plot binarisation of the x-axis\n",
    "hyper_rectangle_m_x = instance_m_discrete[0]\n",
    "if 0 < hyper_rectangle_m_x < 3:\n",
    "    hyper_rectangle_m_x_min = x_quartiles_m[hyper_rectangle_m_x - 1]\n",
    "    hyper_rectangle_m_x_max = x_quartiles_m[hyper_rectangle_m_x]\n",
    "elif hyper_rectangle_m_x == 0:\n",
    "    hyper_rectangle_m_x_min = -0.1\n",
    "    hyper_rectangle_m_x_max = x_quartiles_m[hyper_rectangle_m_x]\n",
    "elif hyper_rectangle_m_x == 3:\n",
    "    hyper_rectangle_m_x_min = x_quartiles_m[hyper_rectangle_m_x - 1]\n",
    "    hyper_rectangle_m_x_max = 1.1\n",
    "plt.axvspan(hyper_rectangle_m_x_min, hyper_rectangle_m_x_max,\n",
    "            facecolor='blue', alpha=0.2)\n",
    "\n",
    "# Plot binarisation of the y-axis\n",
    "hyper_rectangle_m_y = instance_m_discrete[1]\n",
    "if 0 < hyper_rectangle_m_y < 3:\n",
    "    hyper_rectangle_m_y_min = y_quartiles_m[hyper_rectangle_m_y - 1]\n",
    "    hyper_rectangle_m_y_max = y_quartiles_m[hyper_rectangle_m_y]\n",
    "elif hyper_rectangle_m_y == 0:\n",
    "    hyper_rectangle_m_y_min = -0.1\n",
    "    hyper_rectangle_m_y_max = y_quartiles_m[hyper_rectangle_m_y]\n",
    "elif hyper_rectangle_m_y == 3:\n",
    "    hyper_rectangle_m_y_min = y_quartiles_m[hyper_rectangle_m_y - 1]\n",
    "    hyper_rectangle_m_y_max = 1.1\n",
    "plt.axhspan(hyper_rectangle_m_y_min, hyper_rectangle_m_y_max,\n",
    "            facecolor='yellow', alpha=0.2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "XrxETgJY6Frb"
   },
   "source": [
    "Since we have two dimensions:\n",
    "\n",
    "* **blue** encodes the x-axis partition that holds the explained instance; and\n",
    "* **yellow** encodes the y-axis partition that holds the explained instance.\n",
    "\n",
    "This *binarisation* results in four distinct encodings in the\n",
    "*binary interpretable representation*:\n",
    "\n",
    "* **blue & yellow** -- (1, 1);\n",
    "* **blue & not yellow** -- (1, 0);\n",
    "* **not blue & yellow** -- (0, 1); and\n",
    "* **not blue & not yellow** -- (0, 0).\n",
    "\n",
    "---\n",
    "\n",
    "Now, let's compute this representation and visualise it with different markers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1834,
     "status": "ok",
     "timestamp": 1598703186545,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "ZwHPf9Hb7zOz",
    "outputId": "ba30ca4b-1bea-4e0e-db3b-2a986cb2084a"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 1], dtype=int8)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "moons_X_m_binary = fatf_transformation.dataset_row_masking(\n",
    "    moons_X_discrete, instance_m_discrete)\n",
    "\n",
    "instance_m_binary = np.ones_like(instance_m_discrete, dtype=np.int8)\n",
    "\n",
    "display(instance_m_binary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 550
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1637,
     "status": "ok",
     "timestamp": 1598703186546,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "yh2Q1i6j8E8C",
    "outputId": "ac0729da-80ad-46c4-99c7-5fb521890131"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.patches.Polygon at 0x7f568ddc4a90>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x640 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8, 8), dpi=80)\n",
    "\n",
    "# Plot data using a unique marker for each binary encoding\n",
    "moons_unique_binary = np.unique(moons_X_m_binary, axis=0)\n",
    "for rep in moons_unique_binary:\n",
    "    indices = fatf_ecml.get_hyperrectangle_indices(\n",
    "        moons_X_m_binary, rep)\n",
    "\n",
    "    if np.array_equal(rep, [1, 1]):\n",
    "        marker = '*'\n",
    "    elif np.array_equal(rep, [1, 0]):\n",
    "        marker = 'x'\n",
    "    elif np.array_equal(rep, [0, 1]):\n",
    "        marker = 'v'\n",
    "    else:\n",
    "        marker = 'o'\n",
    "\n",
    "    plt.scatter(moons_X[indices, 0], moons_X[indices, 1],\n",
    "                cmap=plt.cm.Set1, c=moons_y[indices], marker=marker)\n",
    "\n",
    "# Plot explained instance\n",
    "plt.scatter(instance_m[0], instance_m[1], c='green', marker='*', s=500)\n",
    "\n",
    "# Plot x-axis quartiles\n",
    "x_quartiles_m = q_discretiser_moons.feature_bin_boundaries.get(0)\n",
    "for i in x_quartiles_m:\n",
    "    plt.plot([i, i], [-0.1, 1.1], linewidth=2, linestyle='--', c='blue')\n",
    "\n",
    "# Plot y-axis quartiles\n",
    "y_quartiles_m = q_discretiser_moons.feature_bin_boundaries.get(1)\n",
    "for i in y_quartiles_m:\n",
    "    plt.plot([-0.1, 1.1], [i, i], linewidth=2, linestyle='--', c='blue')\n",
    "\n",
    "# Plot binarisation of the x-axis\n",
    "hyper_rectangle_m_x = instance_m_discrete[0]\n",
    "if 0 < hyper_rectangle_m_x < 3:\n",
    "    hyper_rectangle_m_x_min = x_quartiles_m[hyper_rectangle_m_x - 1]\n",
    "    hyper_rectangle_m_x_max = x_quartiles_m[hyper_rectangle_m_x]\n",
    "elif hyper_rectangle_m_x == 0:\n",
    "    hyper_rectangle_m_x_min = -0.1\n",
    "    hyper_rectangle_m_x_max = x_quartiles_m[hyper_rectangle_m_x]\n",
    "elif hyper_rectangle_m_x == 3:\n",
    "    hyper_rectangle_m_x_min = x_quartiles_m[hyper_rectangle_m_x - 1]\n",
    "    hyper_rectangle_m_x_max = 1.1\n",
    "plt.axvspan(hyper_rectangle_m_x_min, hyper_rectangle_m_x_max,\n",
    "            facecolor='blue', alpha=0.2)\n",
    "\n",
    "# Plot binarisation of the y-axis\n",
    "hyper_rectangle_m_y = instance_m_discrete[1]\n",
    "if 0 < hyper_rectangle_m_y < 3:\n",
    "    hyper_rectangle_m_y_min = y_quartiles_m[hyper_rectangle_m_y - 1]\n",
    "    hyper_rectangle_m_y_max = y_quartiles_m[hyper_rectangle_m_y]\n",
    "elif hyper_rectangle_m_y == 0:\n",
    "    hyper_rectangle_m_y_min = -0.1\n",
    "    hyper_rectangle_m_y_max = y_quartiles_m[hyper_rectangle_m_y]\n",
    "elif hyper_rectangle_m_y == 3:\n",
    "    hyper_rectangle_m_y_min = y_quartiles_m[hyper_rectangle_m_y - 1]\n",
    "    hyper_rectangle_m_y_max = 1.1\n",
    "plt.axhspan(hyper_rectangle_m_y_min, hyper_rectangle_m_y_max,\n",
    "            facecolor='yellow', alpha=0.2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "23phWtUV_M8D"
   },
   "source": [
    "Everything is as expected.\n",
    "\n",
    "---\n",
    "\n",
    "Having seen this visualisation, you may have noticed an undesired consequence\n",
    "of applying the bianrisation step -- it partially breaks the purity of\n",
    "our discretisation.\n",
    "Since this step combines multiple hyper-rectangles that may hold different\n",
    "classes, the purity of the binary representation may suffer despite having\n",
    "nearly perfect discretisation.\n",
    "This is especially visible for the (0, 0) encoding in the binary interpretable\n",
    "representation -- the regions without any shading, i.e., the\n",
    "bottom-right partitions, which mix red and grey instances.\n",
    "\n",
    "A complementary observation is the *unnecessity* of splitting each numerical\n",
    "feature into more than *three* partitions.\n",
    "Two sections are enough for bordering partitions -- see the x-axis and y-axis\n",
    "splits -- and three are required when the explained instance is \"sandwiched\"\n",
    "between two partitions.\n",
    "\n",
    "Finally, let's see how these visual observations translate into our proxy\n",
    "metrics for split purity."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "rNHIxc9tcd-c"
   },
   "outputs": [],
   "source": [
    "moons_X_binary_m_indices = fatf_ecml.get_hyperrectangle_indices(\n",
    "    moons_X_m_binary, instance_m_binary)\n",
    "\n",
    "moons_X_binary_m = moons_X_discrete[moons_X_binary_m_indices, :]\n",
    "moons_y_binary_m = moons_y[moons_X_binary_m_indices]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 927,
     "status": "ok",
     "timestamp": 1598703189987,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "hZYLfdDpcd7k",
    "outputId": "9ea798ad-ef2c-4907-b7fd-8e9670754ff8"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([0, 1]), array([134,   1]))"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.unique(moons_y_binary_m, return_counts=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1497,
     "status": "ok",
     "timestamp": 1598703190910,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "HcYMHx-3kIZJ",
    "outputId": "141211c5-c95b-463a-cf3d-256a0fe96a30"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.014705075445816198"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.gini_index(moons_y_binary_m)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "g4ZkH8TBMsFg"
   },
   "source": [
    "Unsurprisingly, the label count and purity of the explained hyper-rectangle\n",
    "have not changed -- it was not (and never is) merged with any other partition\n",
    "unlike the remaining hyper-rectangles.\n",
    "\n",
    "Let's compute the overall purity then."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1776,
     "status": "ok",
     "timestamp": 1598703192531,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "xj93WlxOMdyf",
    "outputId": "bc3b17dc-b3f1-43a5-cf57-4d823179a532"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.26466978171740074"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.weighted_purity(moons_X_m_binary, moons_y, 'gini')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "baTqjcu4NwXP"
   },
   "source": [
    "As expected, the impurity has seen a considerable jump."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "tT7pe36bw3pk"
   },
   "source": [
    "### Tree-based Interpretable Representation ###\n",
    "\n",
    "Let's try something different -- generating an interpretable representation\n",
    "based on a feature space partition created by a decision tree.\n",
    "Conceptually, such an approach has the advantage over quartile-based\n",
    "discretisation since it has access to data labels in the selected region\n",
    "(either ground truth or black-box predictions).\n",
    "To check our intuition, we can calculate purity of this discretisation and\n",
    "compare it visually to our previous results."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "0pZCdA2f9sGg"
   },
   "source": [
    "#### Tree-based Discretisation ####\n",
    "Since each feature was split into 4 bins with quartile-based discretisation,\n",
    "this discrete space had the cardinality of $4^d$, where $d$ is the number of\n",
    "features in the original representation.\n",
    "Here, we limit the number of tree leaves to $4^2 = 16$ to give the tree\n",
    "comparable expressiveness."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 196
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 820,
     "status": "error",
     "timestamp": 1598798843738,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "RLavmy9y_lSZ",
    "outputId": "b3e5aee3-b3b3-4c9d-caee-bfb67a206f7d"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier(max_leaf_nodes=16, random_state=42)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tree_discretiser_moons = sklearn.tree.DecisionTreeClassifier(\n",
    "    random_state=42, max_leaf_nodes=16)\n",
    "tree_discretiser_moons.fit(moons_X, moons_y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "ZUFwMMY5BIe4"
   },
   "source": [
    "Let's plot the decision tree splits."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 423
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2359,
     "status": "ok",
     "timestamp": 1598703196154,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "I-y_vaFu9ptj",
    "outputId": "6e6aed8b-075d-4d45-a7e6-dd0f6e3afac0"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f568dcfe990>"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 480x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8, 8), dpi=60)\n",
    "\n",
    "# Compute the decision boundary of the tree\n",
    "x_y_step = 0.005\n",
    "x_min, x_max = -0.1, 1.1\n",
    "y_min, y_max = -0.1, 1.1\n",
    "xx, yy = np.meshgrid(np.arange(x_min, x_max, x_y_step),\n",
    "                     np.arange(y_min, y_max, x_y_step))\n",
    "\n",
    "Z = tree_discretiser_moons.apply(np.c_[xx.ravel(), yy.ravel()])\n",
    "\n",
    "Z = Z.reshape(xx.shape)\n",
    "Z_norm = np.zeros_like(Z)\n",
    "for i, v in enumerate(np.unique(Z)):\n",
    "    mask = (Z == v)\n",
    "    Z_norm[mask] = i\n",
    "\n",
    "# Plot the boundary\n",
    "plt.contourf(xx, yy, Z_norm, cmap=plt.cm.tab20, alpha=0.5)\n",
    "\n",
    "# Plot data\n",
    "plt.scatter(moons_X[:, 0], moons_X[:, 1], cmap=plt.cm.Set1, c=moons_y)\n",
    "\n",
    "# Plot explained instance\n",
    "plt.scatter(instance_m[0], instance_m[1], c='green', marker='*', s=500)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "-Nyv3LQiOti-"
   },
   "source": [
    "Also, let's compute the purity of this discretisation.\n",
    "\n",
    "> **Note that we are using the `apply` method of our tree classifier -- it\n",
    "  returns the leaf ID of each instance, i.e., its hyper-rectangle ID.**\n",
    "\n",
    "We start with purity of the partition assigned to the explained instance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "oc6p0Tl89pwH"
   },
   "outputs": [],
   "source": [
    "moons_X_discrete_tree = tree_discretiser_moons.apply(moons_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1440,
     "status": "ok",
     "timestamp": 1598703196155,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "5oUSQ7ZDRNHm",
    "outputId": "7c2984b2-2213-42ed-d756-e9472e669efe"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.04032871972318343"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "instance_m_discrete_tree = tree_discretiser_moons.apply([instance_m])\n",
    "\n",
    "moons_X_tree_m_indices = fatf_ecml.get_hyperrectangle_indices(\n",
    "    moons_X_discrete_tree, instance_m_discrete_tree)\n",
    "\n",
    "moons_y_tree_m = moons_y[moons_X_tree_m_indices]\n",
    "\n",
    "fatf_ecml.gini_index(moons_y_tree_m)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "FeU2rauTRNel"
   },
   "source": [
    "It is slightly worse than the one achieved with quartiles (0.015).\n",
    "\n",
    "---\n",
    "\n",
    "Now, let's compute purity of the whole discretisation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 985,
     "status": "ok",
     "timestamp": 1598703198116,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "r6--P70q9prV",
    "outputId": "9b4ca2d7-dbb3-42ce-8e22-52089db1e339"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.06319989994928496"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.weighted_purity(moons_X_discrete_tree, moons_y, 'gini')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "5MKKFuc0Q4cv"
   },
   "source": [
    "Which gives a better result than quartiles (0.13)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "43Znin7-9xdo"
   },
   "source": [
    "#### Binarisation ####\n",
    "Binarisation of the tree-based discretisation is not a well-defined concept.\n",
    "The best we can do is one-hot encode each tree partitioni, i.e., hyper-rectangle.\n",
    "Such a representation, however, would have higher expressiveness than\n",
    "binarised quartile-based discretisation, hence the two should not be compared\n",
    "directly.\n",
    "**Nonetheless, the purity calculations should not change in this case.**\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "KhvoFlR59pml"
   },
   "outputs": [],
   "source": [
    "moons_X_binary_tree = fatf_ecml.one_hot_encode(moons_X_discrete_tree)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 523,
     "status": "ok",
     "timestamp": 1598703200729,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "fcScGP-a9_Fm",
    "outputId": "b410278d-9e54-4fd5-e8fa-35794a15ea72"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.06319989994928496"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.weighted_purity(moons_X_binary_tree, moons_y, 'gini')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "l0-FPyvLVZG-"
   },
   "source": [
    "Alternatively, a tree-based representation that is comparable to a\n",
    "quartile-based binarisation is generated by a tree with no more than\n",
    "$2^d$ leaves, where $d$ is the number of features in the original\n",
    "data representation.\n",
    "Let's compute purity of such a partition."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 120
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 624,
     "status": "ok",
     "timestamp": 1598703202639,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "EGgBy410VY6E",
    "outputId": "b27a4f8d-d958-496d-90e3-da09eabe0f35"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier(max_leaf_nodes=4, random_state=42)"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tree_binariser_moons = sklearn.tree.DecisionTreeClassifier(\n",
    "    random_state=42, max_leaf_nodes=4)\n",
    "tree_binariser_moons.fit(moons_X, moons_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 423
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1297,
     "status": "ok",
     "timestamp": 1598703204656,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "bG8l8EMQ9AtL",
    "outputId": "213ba2ea-6cc2-47a7-a311-92cf6688c5d1"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f568da71b10>"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 480x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8, 8), dpi=60)\n",
    "\n",
    "# Compute the decision boundary of the tree\n",
    "x_y_step = 0.005\n",
    "x_min, x_max = -0.1, 1.1\n",
    "y_min, y_max = -0.1, 1.1\n",
    "xx, yy = np.meshgrid(np.arange(x_min, x_max, x_y_step),\n",
    "                     np.arange(y_min, y_max, x_y_step))\n",
    "\n",
    "Z = tree_binariser_moons.apply(np.c_[xx.ravel(), yy.ravel()])\n",
    "\n",
    "Z = Z.reshape(xx.shape)\n",
    "Z_norm = np.zeros_like(Z)\n",
    "for i, v in enumerate(np.unique(Z)):\n",
    "    mask = (Z == v)\n",
    "    Z_norm[mask] = i\n",
    "\n",
    "# Plot the boundary\n",
    "plt.contourf(xx, yy, Z_norm, cmap=plt.cm.tab20, alpha=0.5)\n",
    "\n",
    "# Plot data\n",
    "plt.scatter(moons_X[:, 0], moons_X[:, 1], cmap=plt.cm.Set1, c=moons_y)\n",
    "\n",
    "# Plot explained instance\n",
    "plt.scatter(instance_m[0], instance_m[1], c='green', marker='*', s=500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "J3SZ62u19Aqh"
   },
   "outputs": [],
   "source": [
    "moons_X_binary_tree = tree_binariser_moons.apply(moons_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 685,
     "status": "ok",
     "timestamp": 1598703205779,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "2aC2dTkn9pkP",
    "outputId": "0e27e44a-afca-4483-cecb-5dae81bd11a2"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.13809164792390438"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.weighted_purity(moons_X_binary_tree, moons_y, 'gini')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "WKA3G0QR92vb"
   },
   "source": [
    "This result is on a par with the binary representation achieved with\n",
    "quartile-based discretisation (0.13).\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "nj8kuXmKSkNC"
   },
   "source": [
    "## Real Data -- Bike Sharing ##\n",
    "Let's repeat the same procedure with a real-life data set.\n",
    "In the example above, we generated the representations based on the **entire**\n",
    "data set and used **ground truth** labels, neither of which is done when\n",
    "building a *local* surrogate.\n",
    "Therefore, with the Bike Sharing data set we will:\n",
    "\n",
    "* generate a data sample around a data point chosen to be explained and\n",
    "  use these data to evaluate purity of the interpretable representation; and\n",
    "* evaluate purity based on predictions of a black box.\n",
    "\n",
    "Since this data set has more than two feature, we cannot visualise the\n",
    "behaviour of the interpretable representation.\n",
    "Instead, we will rely on the numerical purity: Gini Index for *crisp* and\n",
    "Mean Squared Error for *probabilistic* classifiers.\n",
    "\n",
    "First, let's get the data set and build one probabilistic and one crisp\n",
    "black-box model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 50
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 5920,
     "status": "ok",
     "timestamp": 1598798303078,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "fzZuATLrelgZ",
    "outputId": "fc4caa71-a460-4784-f435-4c779c4017e5"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "22-Sep-08 12:51:30 fatf         INFO     Seeding RNGs using the input parameter.\n",
      "22-Sep-08 12:51:30 fatf         INFO     Seeding RNGs with 42.\n"
     ]
    }
   ],
   "source": [
    "(bikes_train_X,\n",
    " bikes_test_X,\n",
    " bikes_train_y,\n",
    " bikes_test_y,\n",
    " bikes_feature_names,\n",
    " bikes_target_name) = fatf_ecml.generate_bikes(random_seed=42)\n",
    "\n",
    "bikes_target_classes = ['low', 'medium', 'high']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 50
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1213,
     "status": "ok",
     "timestamp": 1598798308435,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "YQUkEH3YAgFX",
    "outputId": "3bb7ac41-0cbd-4a21-ed97-d870b0a7671a"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "22-Sep-08 12:51:32 fatf         INFO     Seeding RNGs using the input parameter.\n",
      "22-Sep-08 12:51:32 fatf         INFO     Seeding RNGs with 42.\n"
     ]
    }
   ],
   "source": [
    "clf_bikes_probabilistic = fatf_ecml.get_random_forest(\n",
    "    bikes_train_X, bikes_train_y, random_seed=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 50
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1090,
     "status": "ok",
     "timestamp": 1598798332871,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "0SUu6FoyAgMn",
    "outputId": "a8694359-a653-47e2-b091-48f8d1d97e69"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "22-Sep-08 12:51:32 fatf         INFO     Seeding RNGs using the input parameter.\n",
      "22-Sep-08 12:51:32 fatf         INFO     Seeding RNGs with 42.\n"
     ]
    }
   ],
   "source": [
    "clf_bikes_crisp = fatf_ecml.get_svc(\n",
    "    bikes_train_X, bikes_train_y, random_seed=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "PqEZ6PSvw6pA"
   },
   "source": [
    "Now, let's select a data point to be explained and generate a sample of 10,000\n",
    "instances around it (see the *Data Sampling* notebook for more details on\n",
    "sampling).\n",
    "\n",
    "We also predict the sample with both of our black boxes to get crisp\n",
    "predictions and probabilities, which we use to investigate purity\n",
    "of the interpretable representations.\n",
    "This can serve as a proxy for how well interpretable representations capture\n",
    "a black-box decision boundary, which we try to approximate with our surrogate."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 67
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1091,
     "status": "ok",
     "timestamp": 1598798336338,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "fsqGw5SwxC2c",
    "outputId": "9f059878-285a-49bd-9141-f91065a00c1e"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2.      , 0.      , 6.      , 0.      , 6.      , 0.      ,\n",
       "       1.      , 0.696667, 0.643954, 0.670417, 0.119408], dtype=float32)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "instance_b_index = 42\n",
    "instance_b = bikes_test_X[instance_b_index, :]\n",
    "instance_b_label = bikes_test_y[instance_b_index]\n",
    "\n",
    "display(instance_b)\n",
    "display(instance_b_label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 50
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1155,
     "status": "ok",
     "timestamp": 1598798396479,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "afUleQWbBZ9X",
    "outputId": "8d9a7589-d66f-4e13-cb7b-98359694218a"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "22-Sep-08 12:51:34 fatf         INFO     Seeding RNGs using the input parameter.\n",
      "22-Sep-08 12:51:34 fatf         INFO     Seeding RNGs with 42.\n"
     ]
    }
   ],
   "source": [
    "fatf.setup_random_seed(42)\n",
    "\n",
    "augmenter_bikes = fatf_augmentation.NormalSampling(bikes_train_X)\n",
    "\n",
    "sample_bikes_b = augmenter_bikes.sample(\n",
    "    instance_b, samples_number=10000)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "_eUmOy9SC2aF"
   },
   "source": [
    "Note that for the probabilistic classifier, we only choose the probabilities\n",
    "of a single class -- the one predicted by the black box for the explained\n",
    "instance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1058,
     "status": "ok",
     "timestamp": 1598798403224,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "YwNhjha8Cs-K",
    "outputId": "7c080e84-eea6-4a07-ecc8-840708ec193d"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "instance_b_prediction_proba = clf_bikes_probabilistic.predict_proba(\n",
    "    [instance_b])\n",
    "instance_b_prediction = int(np.argmax(instance_b_prediction_proba))\n",
    "display(instance_b_prediction)\n",
    "\n",
    "sample_bikes_b_predictions_proba = clf_bikes_probabilistic.predict_proba(\n",
    "    sample_bikes_b)\n",
    "\n",
    "sample_bikes_b_predictions_proba = (\n",
    "    sample_bikes_b_predictions_proba[:, instance_b_prediction]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1068,
     "status": "ok",
     "timestamp": 1598798405827,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "VpH67EnvGEN9",
    "outputId": "2557e6b0-bc71-4ea0-bea5-91d5e5fad56e"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "instance_b_prediction_crisp = clf_bikes_crisp.predict([instance_b])\n",
    "display(instance_b_prediction_crisp[0])\n",
    "\n",
    "sample_bikes_b_predictions_crisp = clf_bikes_crisp.predict(\n",
    "    sample_bikes_b)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "IroQBJdSVbly"
   },
   "source": [
    "As a point of reference, let's measure the purity of the sample without either\n",
    "discretisation or binarisation.\n",
    "We start with the probabilistic sample."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1015,
     "status": "ok",
     "timestamp": 1598798411019,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "QT6hfgTaVd5Q",
    "outputId": "7b69b163-aeae-4fc7-c1da-5554dd7b1772"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.05288705751874819"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.mse(sample_bikes_b_predictions_proba)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "_QM5EqMyJM9t"
   },
   "source": [
    "This number may be difficult to read.\n",
    "Let's visualise the probability distribution of this class."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 269
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 916,
     "status": "ok",
     "timestamp": 1598798413056,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "dskhhPuyJNTQ",
    "outputId": "26442317-be22-4d02-dd30-b2f0a2be8f1c"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "_ = plt.hist(sample_bikes_b_predictions_proba, bins=100)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "8CqgW1K6JiyM"
   },
   "source": [
    "It is quite a diverse sample, which is desirable (see the *Data Sampling*\n",
    "notebook for more details).\n",
    "\n",
    "---\n",
    "\n",
    "Let's move on to the crisp sample.\n",
    "(Recall that *Gini Impurity* for *three* classes can have the maximum value of\n",
    "$\\frac{2}{3}$.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1254,
     "status": "ok",
     "timestamp": 1598798506444,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "Osbz3a_aVl67",
    "outputId": "a85a96cf-6c49-4a15-d5d2-8b2d70001a90"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5654105"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.gini_index(sample_bikes_b_predictions_crisp)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "03H1rP79I2zR"
   },
   "source": [
    "The latter score is quite high.\n",
    "Let's inspect the class counts."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 923,
     "status": "ok",
     "timestamp": 1598798512017,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "XONEa7Mv3ldm",
    "outputId": "35872826-5874-440d-835b-e2665858107c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([0, 1, 2]), array([4385, 4865,  750]))"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.unique(sample_bikes_b_predictions_crisp, return_counts=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "sRE0s5eBKoAn"
   },
   "source": [
    "Similar to probabilities, this sample is fairly diverse as well."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "MjECm1wVuxy4"
   },
   "source": [
    "### Interpretable Representation Based on Quartile Discretisation ###\n",
    "For quartile-based discretisation, we are facing with a choice of data to be\n",
    "used for computing the quartiles.\n",
    "Shall we use data from the same distribution as our entire Bikes data set,\n",
    "or instead use the data sampled in the neighbourhood of the explained instance?\n",
    "\n",
    "While you may think that the latter approach is desired, the choice is not obvious\n",
    "because of the curse of dimensionality.\n",
    "With 11 features and 4 bins per feature, we will be creating\n",
    "$4^{11} = 4,194,304$ hyper-rectangles.\n",
    "This granularity is overwhelming when compared to the number of instances\n",
    "in the data set or the number of sampled data points.\n",
    "With this many partitions, it may actually be better to apply discretisation\n",
    "step to the entire feature space.\n",
    "\n",
    "The next step is binarisation.\n",
    "Similarly, it will give $2^{11} = 2,048$ partitions --\n",
    "it is more manageable but still a lot.\n",
    "\n",
    "Let's do both approaches -- the entire data distribution and the local sample\n",
    "-- and compare the purity of each step: discretisation and binarisation."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "sSzrgrNeGxd7"
   },
   "source": [
    "#### Whole Data Discretisation ####\n",
    "In this case the quartiles are computed based on the distribution of the\n",
    "entire data set (this is how LIME operates).\n",
    "To this end, we will use the *Bikes Sharing* **test** set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "5-dgRF9Bv9z3"
   },
   "outputs": [],
   "source": [
    "bikes_test_q_discretiser = fatf_discretisation.QuartileDiscretiser(bikes_test_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "bcqBEIzmKISm"
   },
   "outputs": [],
   "source": [
    "instance_b_discrete_t = bikes_test_q_discretiser.discretise(instance_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "11k1iE5EG8Kg"
   },
   "outputs": [],
   "source": [
    "sample_bikes_b_discrete_t = bikes_test_q_discretiser.discretise(sample_bikes_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3879,
     "status": "ok",
     "timestamp": 1598798535935,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "H9hx8n_7G8Hw",
    "outputId": "c683d704-f3ec-4750-cf42-f29f8516bfa5"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.05524230158730158"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.weighted_purity(\n",
    "    sample_bikes_b_discrete_t,\n",
    "    sample_bikes_b_predictions_crisp,\n",
    "    'gini')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "luhC0fV0O1CX"
   },
   "source": [
    "Discretisation based on quartiles that are determined by the **entire data\n",
    "distribution** give a considerable reduction in impurity.\n",
    "Recall that the entire sample had Gini Index of 0.565.\n",
    "However, this is not the representation used for computing surrogate\n",
    "explanations.\n",
    "We still need to binarise this space, which, as we have seen above, is the\n",
    "step that causes troubles."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "2CASfWnLG8FJ"
   },
   "outputs": [],
   "source": [
    "sample_bikes_b_binary_t = fatf_transformation.dataset_row_masking(\n",
    "    sample_bikes_b_discrete_t, instance_b_discrete_t)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1705,
     "status": "ok",
     "timestamp": 1598798562562,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "VOhqClxxKVcp",
    "outputId": "b495fb7d-6746-4167-cd9b-cb2a22518eec"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.41021482545580185"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.weighted_purity(\n",
    "    sample_bikes_b_binary_t,\n",
    "    sample_bikes_b_predictions_crisp,\n",
    "    'gini')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "FimCQ7oVP_zf"
   },
   "source": [
    "It seems that by combining hyper-rectangles we are not too far off from\n",
    "where we started.\n",
    "This does not look promising for the quality of our future explanations.\n",
    "\n",
    "---\n",
    "\n",
    "Let's have a look at the probabilistic black box now, before we move on\n",
    "to quartiles based on the sampled data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3317,
     "status": "ok",
     "timestamp": 1598798629631,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "9JnXadWLG8Ck",
    "outputId": "e16dc151-dd82-4f8a-865b-0a04be61b204"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.007986073566028934"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.weighted_purity(\n",
    "    sample_bikes_b_discrete_t,\n",
    "    sample_bikes_b_predictions_proba,\n",
    "    'mse')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1352,
     "status": "ok",
     "timestamp": 1598798632090,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "KkRJ9YMEK2RC",
    "outputId": "1d8d37b3-52a9-405b-afe0-d2e72cbcee33"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.03765456999193213"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.weighted_purity(\n",
    "    sample_bikes_b_binary_t,\n",
    "    sample_bikes_b_predictions_proba,\n",
    "    'mse')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "827W6-8XQofT"
   },
   "source": [
    "Again, we notice a similar trend.\n",
    "Discretisation gives us a considerable reduction (0.008 from 0.053), followed\n",
    "by a binarisation-induced increase to 0.038."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Y_bcxOcIwcB7"
   },
   "source": [
    "#### Sampled Data Discretisation ####\n",
    "In this case, the quartiles are computed based on the distribution of the\n",
    "**data sampled in the neighbourhood** of the explained instance.\n",
    "Let's compute purity of this discretisation and binarisation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "UsANLun0wFMN"
   },
   "outputs": [],
   "source": [
    "bikes_sample_q_discretiser = fatf_discretisation.QuartileDiscretiser(\n",
    "    sample_bikes_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "q4GtoiXUwk1b"
   },
   "outputs": [],
   "source": [
    "instance_b_discrete_s = bikes_sample_q_discretiser.discretise(instance_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Ub_jKAtQLp7d"
   },
   "outputs": [],
   "source": [
    "sample_bikes_b_discrete_s = bikes_sample_q_discretiser.discretise(\n",
    "    sample_bikes_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3980,
     "status": "ok",
     "timestamp": 1598798673882,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "C0q86v0ALx7m",
    "outputId": "f0bf6573-0c03-4f3a-9861-35fb10648c88"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0005"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.weighted_purity(\n",
    "    sample_bikes_b_discrete_s,\n",
    "    sample_bikes_b_predictions_crisp,\n",
    "    'gini')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "OuF9vj_RL5bT"
   },
   "outputs": [],
   "source": [
    "sample_bikes_b_binary_s = fatf_transformation.dataset_row_masking(\n",
    "    sample_bikes_b_discrete_s, instance_b_discrete_s)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1029,
     "status": "ok",
     "timestamp": 1598798677256,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "B2Bi33XQMI0b",
    "outputId": "d05efd09-4988-4757-8803-97f01d6f638f"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.4294403699716251"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.weighted_purity(\n",
    "    sample_bikes_b_binary_s,\n",
    "    sample_bikes_b_predictions_crisp,\n",
    "    'gini')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "hj1LcwReMvn9"
   },
   "source": [
    "In this case the discretisation purity is nearly perfect.\n",
    "However, it should not come as a surprise.\n",
    "With 4,194,304 hyper-rectangles and just 10,000 data points points, each one\n",
    "can be assigned to a separate partition.\n",
    "When compared to the quartiles computed with the entire data set, local\n",
    "partition is simply **overfitted**.\n",
    "We can observe this lack of generalisation when moving from the discrete into\n",
    "the binary representation, which has considerable impurity of 0.429.\n",
    "\n",
    "We can validate our theory by inspecting counts."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 265
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1409,
     "status": "ok",
     "timestamp": 1598798711655,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "7EjvuBUUMIya",
    "outputId": "ab9a7f49-a64a-4f02-a4b3-72f53455d25f"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "_, counts = np.unique(sample_bikes_b_discrete_s, return_counts=True, axis=0)\n",
    "_ = plt.hist(counts, bins=3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "aiSvsCbo1I0L"
   },
   "source": [
    "The vast majority of hyper-rectangles enclose just a *single* data point.\n",
    "\n",
    "---\n",
    "\n",
    "Let's check purity for the probabilistic model.\n",
    "We should observe a similar behaviour."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3879,
     "status": "ok",
     "timestamp": 1598798785384,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "kf28DupLMwXy",
    "outputId": "f585ef05-521c-49ed-b8a3-e330a0cc8c97"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3.746483187175399e-05"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.weighted_purity(\n",
    "    sample_bikes_b_discrete_s,\n",
    "    sample_bikes_b_predictions_proba,\n",
    "    'mse')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1215,
     "status": "ok",
     "timestamp": 1598798788961,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "XgjTH6yrMwSG",
    "outputId": "ce676cb4-9298-4e84-c02a-128ef4899a81"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.04412780561123212"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.weighted_purity(\n",
    "    sample_bikes_b_binary_s,\n",
    "    sample_bikes_b_predictions_proba,\n",
    "    'mse')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "IBtI5EYHMJAg"
   },
   "source": [
    "As expected, we observe a reduction of the Mean Squared Error to 0.000037\n",
    "for discretisation and increase to 0.044 for binarisation (the overall purity\n",
    "is 0.053)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "R4tdJspuNxAS"
   },
   "source": [
    "### Interpretable Representation Based on a Decision Tree ###\n",
    "For tree-based interpretable representations it is only reasonable to\n",
    "fit the tree to the local sample.\n",
    "Otherwise, we would be competing with the black box, which is not\n",
    "any different from actually using a decision tree to model the entire\n",
    "data in a more transparent way.\n",
    "(And the black box was chosen for a reason...)\n",
    "\n",
    "There are two possible approaches to ensure comparability with the\n",
    "quartile-based interpretable representation.\n",
    "One is to limit the number of tree leaves to $2^{11} = 2,048$, which\n",
    "corresponds to the number of unique representations in the binarised\n",
    "quartile-based representation.\n",
    "Alternatively, if we choose to one-hot-encode the tree partitions, we\n",
    "should limit the number of leaves to 11.\n",
    "\n",
    "Let's see how both of these scenarios play out."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 118
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1058,
     "status": "ok",
     "timestamp": 1598798865395,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "nXINo7gBMIvY",
    "outputId": "0b91710e-fe7a-4c3b-95c7-2c7caa5dd834"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier(max_leaf_nodes=2048, random_state=42)"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tree_bikes_crisp_2048 = sklearn.tree.DecisionTreeClassifier(\n",
    "    random_state=42, max_leaf_nodes=2048)\n",
    "tree_bikes_crisp_2048.fit(sample_bikes_b, sample_bikes_b_predictions_crisp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "83TbmnQNOpHm"
   },
   "outputs": [],
   "source": [
    "sample_bikes_b_tree_crisp_2048 = tree_bikes_crisp_2048.apply(sample_bikes_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1153,
     "status": "ok",
     "timestamp": 1598798873855,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "oKIAFLZ1O5yl",
    "outputId": "51b29e69-9548-41d5-e19f-f0b6b414110e"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.weighted_purity(\n",
    "    sample_bikes_b_tree_crisp_2048,\n",
    "    sample_bikes_b_predictions_crisp,\n",
    "    'gini')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1351,
     "status": "ok",
     "timestamp": 1598798876750,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "qryMrHgigkid",
    "outputId": "1756664c-5367-4b42-ce19-2b35180c5342"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Leaves #: 577.\n"
     ]
    }
   ],
   "source": [
    "unique = np.unique(sample_bikes_b_tree_crisp_2048)\n",
    "print('Leaves #: {}.'.format(unique.shape[0]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Zv4yxiOMhBHe"
   },
   "source": [
    "Therefore, with \"just\" 577 partitions we can reduce the impurity to 0,\n",
    "compared to the original impurity of 0.565 and best quartile-based binarised\n",
    "score of 0.410.\n",
    "\n",
    "Let's see how we will do with just 11 leaves."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 118
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1064,
     "status": "ok",
     "timestamp": 1598798963754,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "qRG5_AxmP75J",
    "outputId": "817b5b6d-108e-435a-d97e-03b8a8518b05"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier(max_leaf_nodes=11, random_state=42)"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tree_bikes_crisp_11 = sklearn.tree.DecisionTreeClassifier(\n",
    "    random_state=42, max_leaf_nodes=11)\n",
    "tree_bikes_crisp_11.fit(sample_bikes_b, sample_bikes_b_predictions_crisp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "AkarmZ9qhje0"
   },
   "outputs": [],
   "source": [
    "sample_bikes_b_tree_crisp_11 = tree_bikes_crisp_11.apply(sample_bikes_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 926,
     "status": "ok",
     "timestamp": 1598798967248,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "Ksw7ZvBahv5R",
    "outputId": "c2660ba0-b880-499f-e6a4-d7103d2b1fc4"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.2340876374955482"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.weighted_purity(\n",
    "    sample_bikes_b_tree_crisp_11,\n",
    "    sample_bikes_b_predictions_crisp,\n",
    "    'gini')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "THbSFusShjpt"
   },
   "source": [
    "Not too bad.\n",
    "We managed to reduce the impurity 2.5 times with just 11 leaves.\n",
    "\n",
    "---\n",
    "\n",
    "Let's execute the same experiments for the probabilistic model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 118
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1233,
     "status": "ok",
     "timestamp": 1598799025921,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "Wu-BBFuHOZHh",
    "outputId": "90b80eef-1402-4230-b16e-c781e7508137"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeRegressor(max_leaf_nodes=2048, random_state=42)"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tree_bikes_proba_2048 = sklearn.tree.DecisionTreeRegressor(\n",
    "    random_state=42, max_leaf_nodes=2048)\n",
    "tree_bikes_proba_2048.fit(sample_bikes_b, sample_bikes_b_predictions_proba)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "-rg1kAKaOi7Y"
   },
   "outputs": [],
   "source": [
    "sample_bikes_b_tree_proba_2048 = tree_bikes_proba_2048.apply(sample_bikes_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1270,
     "status": "ok",
     "timestamp": 1598799028377,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "NSKoNd1yPYao",
    "outputId": "1f14dc4d-1719-4842-f37e-74edeaf4b5fb"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2.8717372707635692e-05"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.weighted_purity(\n",
    "    sample_bikes_b_tree_proba_2048,\n",
    "    sample_bikes_b_predictions_proba,\n",
    "    'mse')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1162,
     "status": "ok",
     "timestamp": 1598799032060,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "jwju-crnhjbF",
    "outputId": "04598e6d-78d5-423b-eb5f-749af0fee357"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Leaves #: 2048.\n"
     ]
    }
   ],
   "source": [
    "unique = np.unique(sample_bikes_b_tree_proba_2048)\n",
    "print('Leaves #: {}.'.format(unique.shape[0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 118
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1683,
     "status": "ok",
     "timestamp": 1598799034909,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "TofF7J0RPdCx",
    "outputId": "cca53e4a-5be4-4c47-895f-2a554e245e68"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeRegressor(max_leaf_nodes=11, random_state=42)"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tree_bikes_proba_11 = sklearn.tree.DecisionTreeRegressor(\n",
    "    random_state=42, max_leaf_nodes=11)\n",
    "tree_bikes_proba_11.fit(sample_bikes_b, sample_bikes_b_predictions_proba)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "jF8LDiXJi60v"
   },
   "outputs": [],
   "source": [
    "sample_bikes_b_tree_proba_11 = tree_bikes_proba_11.apply(sample_bikes_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 917,
     "status": "ok",
     "timestamp": 1598799037354,
     "user": {
      "displayName": "Kacper Sokol",
      "photoUrl": "",
      "userId": "08424478775456813676"
     },
     "user_tz": -60
    },
    "id": "zd6O41Bmi6wd",
    "outputId": "5101c841-3777-47e0-ccd6-659e60d92a5e"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.026033229911113745"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fatf_ecml.weighted_purity(\n",
    "    sample_bikes_b_tree_proba_11,\n",
    "    sample_bikes_b_predictions_proba,\n",
    "    'mse')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "78zzvNjSjUjt"
   },
   "source": [
    "In case of the probabilistic model, we can reduce the MSE to 0.000029 with\n",
    "the full expressiveness of 2,048 partitions, which increases to 0.026\n",
    "with just 11 partitions -- recall that the whole sample has the Mean Squared\n",
    "Error of 0.053."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "LPx3bNyAeASz"
   },
   "source": [
    "## Summary ##\n",
    "This notebook showed two distinct approaches to generating interpretable\n",
    "representations of tabular data and variants thereof.\n",
    "It is worth noting that these insights should **not** be generalised to other\n",
    "data sets and use cases.\n",
    "Instead, a similar **exploratory** procedure should be employed, with various approaches\n",
    "analysed to choose the most suitable interpretable representation for the\n",
    "problem at hand.\n",
    "\n",
    "> In summary, the core premise of analysing purity (or uniformity) of **black-box\n",
    "  predictions** when distributed across **partitions** of a tabular interpretable\n",
    "  representation is to ensure that our hyper-rectangles **align** with black-box\n",
    "  decision boundaries in the modelled neighbourhood.\n",
    "  If such a partition is pure, the decision boundary coincides with our feature space\n",
    "  separation, hence our approximation is faithful.\n",
    "  If a black-box decision boundary runs across a hyper-rectangle, on the other hand,\n",
    "  its purity score should reflect it.\n",
    "\n",
    "---\n",
    "\n",
    "Now, let's move on to *sampling* data in the neighbourhood of the explained\n",
    "instance, which is the next step of building local surrogate explainers.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "cajmYUjtQp2a"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "authorship_tag": "ABX9TyNRUh2LJNJsjE47wzX1FvSF",
   "collapsed_sections": [],
   "name": "2-interpretable_representations.ipynb",
   "provenance": []
  },
  "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": "3.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}