pooled_data.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tensor Flow Version: 2.3.2\n",
      "Keras Version: 2.4.0\n",
      "\n",
      "Python 3.8.8 | packaged by conda-forge | (default, Feb 20 2021, 16:22:27) \n",
      "[GCC 9.3.0]\n",
      "GPU is NOT AVAILABLE\n"
     ]
    }
   ],
   "source": [
    "import sys\n",
    "# Before using tensorflow:\n",
    "import os\n",
    "os.environ['TF_FORCE_GPU_ALLOW_GROWTH'] = 'true'\n",
    "import tensorflow as tf\n",
    "import tensorflow.keras.backend as KB\n",
    "from matplotlib import pylab as plt\n",
    "import tensorflow.keras\n",
    "#from tensorflow.python.framework.ops import disable_eager_execution\n",
    "#disable_eager_execution()\n",
    "\n",
    "import pickle\n",
    "import numpy as np\n",
    "print(f\"Tensor Flow Version: {tf.__version__}\")\n",
    "print(f\"Keras Version: {tensorflow.keras.__version__}\")\n",
    "print()\n",
    "print(f\"Python {sys.version}\")\n",
    "\n",
    "gpu = len(tf.config.list_physical_devices('GPU'))>0\n",
    "print(\"GPU is\", \"available\" if gpu else \"NOT AVAILABLE\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create the model / CNN architecture"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow\n",
    "from client.models.AMLmodel import construct_model\n",
    "model = construct_model()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d (Conv2D)              (None, 100, 100, 32)      1184      \n",
      "_________________________________________________________________\n",
      "activation (Activation)      (None, 100, 100, 32)      0         \n",
      "_________________________________________________________________\n",
      "conv2d_1 (Conv2D)            (None, 98, 98, 32)        9248      \n",
      "_________________________________________________________________\n",
      "activation_1 (Activation)    (None, 98, 98, 32)        0         \n",
      "_________________________________________________________________\n",
      "max_pooling2d (MaxPooling2D) (None, 49, 49, 32)        0         \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 49, 49, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 49, 49, 64)        18496     \n",
      "_________________________________________________________________\n",
      "activation_2 (Activation)    (None, 49, 49, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 47, 47, 64)        36928     \n",
      "_________________________________________________________________\n",
      "activation_3 (Activation)    (None, 47, 47, 64)        0         \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 23, 23, 64)        0         \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 23, 23, 64)        0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 33856)             0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 512)               17334784  \n",
      "_________________________________________________________________\n",
      "batch_normalization (BatchNo (None, 512)               2048      \n",
      "_________________________________________________________________\n",
      "activation_4 (Activation)    (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "dropout_2 (Dropout)          (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 15)                7695      \n",
      "_________________________________________________________________\n",
      "activation_5 (Activation)    (None, 15)                0         \n",
      "=================================================================\n",
      "Total params: 17,410,383\n",
      "Trainable params: 17,409,359\n",
      "Non-trainable params: 1,024\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train a model usign data from one clinic"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from client.data.datagenerator import DataGenerator\n",
    "\n",
    "labels0 = np.load('dataset/processed/data_partitions/partition0/labels.npy', allow_pickle=True).item()\n",
    "data_path0 = 'dataset/processed/data_partitions/partition0/data_singlets'\n",
    "\n",
    "ids0 = [l for l in labels0]\n",
    "np.random.shuffle(ids0)\n",
    "train_split_index = int(len(ids0)*0.9)\n",
    "train_ids = ids0[:train_split_index]\n",
    "val_ids = ids0[train_split_index:]\n",
    "\n",
    "train_gen = DataGenerator(train_ids,labels0, data_path0, dim=(100,100), batch_size=32)\n",
    "val_gen = DataGenerator(val_ids,labels0, data_path0, dim=(100,100), batch_size=32)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "17/17 [==============================] - 12s 715ms/step - loss: 1.9728 - accuracy: 0.4393\n",
      "Epoch 2/20\n",
      "17/17 [==============================] - 12s 707ms/step - loss: 1.2845 - accuracy: 0.6746\n",
      "Epoch 3/20\n",
      "17/17 [==============================] - 12s 697ms/step - loss: 0.9901 - accuracy: 0.7096\n",
      "Epoch 4/20\n",
      "17/17 [==============================] - 12s 697ms/step - loss: 0.8417 - accuracy: 0.7224\n",
      "Epoch 5/20\n",
      "17/17 [==============================] - 12s 697ms/step - loss: 0.5939 - accuracy: 0.8254\n",
      "Epoch 6/20\n",
      "17/17 [==============================] - 12s 701ms/step - loss: 0.4592 - accuracy: 0.8401\n",
      "Epoch 7/20\n",
      "17/17 [==============================] - 12s 702ms/step - loss: 0.3362 - accuracy: 0.8695\n",
      "Epoch 8/20\n",
      "17/17 [==============================] - 12s 692ms/step - loss: 0.2665 - accuracy: 0.9099\n",
      "Epoch 9/20\n",
      "17/17 [==============================] - 12s 685ms/step - loss: 0.2073 - accuracy: 0.9393\n",
      "Epoch 10/20\n",
      "17/17 [==============================] - 12s 686ms/step - loss: 0.1534 - accuracy: 0.9540\n",
      "Epoch 11/20\n",
      "17/17 [==============================] - 12s 694ms/step - loss: 0.1080 - accuracy: 0.9706\n",
      "Epoch 12/20\n",
      "17/17 [==============================] - 12s 678ms/step - loss: 0.1300 - accuracy: 0.9540\n",
      "Epoch 13/20\n",
      "17/17 [==============================] - 12s 708ms/step - loss: 0.0829 - accuracy: 0.9798\n",
      "Epoch 14/20\n",
      "17/17 [==============================] - 12s 695ms/step - loss: 0.0576 - accuracy: 0.9908\n",
      "Epoch 15/20\n",
      "17/17 [==============================] - 12s 715ms/step - loss: 0.0440 - accuracy: 0.9890\n",
      "Epoch 16/20\n",
      "17/17 [==============================] - 12s 688ms/step - loss: 0.0365 - accuracy: 0.9890\n",
      "Epoch 17/20\n",
      "17/17 [==============================] - 12s 685ms/step - loss: 0.0245 - accuracy: 0.9982\n",
      "Epoch 18/20\n",
      "17/17 [==============================] - 12s 697ms/step - loss: 0.0425 - accuracy: 0.9853\n",
      "Epoch 19/20\n",
      "17/17 [==============================] - 12s 691ms/step - loss: 0.0131 - accuracy: 1.0000\n",
      "Epoch 20/20\n",
      "17/17 [==============================] - 12s 689ms/step - loss: 0.0252 - accuracy: 0.9963\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x7fd04455c430>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(train_gen,epochs=20)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model validation - plot a confusion matrix"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We reserve partition1 as a validation set "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.sort([k for k in labels0])\n",
    "\n",
    "labels1 = np.load('dataset/processed/data_partitions/partition1/labels.npy', allow_pickle=True).item()\n",
    "data_path1 = 'dataset/processed/data_partitions/partition1/data_singlets'\n",
    "\n",
    "ids1 = [l for l in labels1]\n",
    "np.random.shuffle(ids1)\n",
    "\n",
    "test_gen = DataGenerator(ids1,labels1, data_path1, dim=(100,100), batch_size=32)\n",
    "labels, preds = test_gen.predict_all(model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "def confusion_matrix(model, data_gen):\n",
    "    \n",
    "    labels, preds = data_gen.predict_all(model)\n",
    "\n",
    "    y_pred = np.argmax(preds,1)\n",
    "    y_data = np.argmax(labels,1)\n",
    "    M = np.zeros((16,16))\n",
    "\n",
    "    for pred_, true_ in zip(y_pred,y_data):\n",
    "        M[true_,pred_] +=1\n",
    "    \n",
    "    M[15,:15] = np.sum(M[:15,:15],0)\n",
    "    M[:15,15] = np.sum(M[:15,:15],1)\n",
    "    M[15,15] = np.sum(M[:15,:15])\n",
    "        \n",
    "    return M"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA5wAAAMvCAYAAAC+7N8KAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAACmOUlEQVR4nOzdfbxVZZ3//9dHkbTUjAQzwBBRRBBRDprdkKZmN2Z5U0JWGk5OTjWVX5ufM9NM2TTGlJUm1UyZN2lJk5WUJWYqamYRKN6RRoYF6CiUmprKjZ/fH2sd3eIBzoGzzubC1/Px2A/2utbde1+ss8/57GuttSMzkSRJkiSpt23W7gCSJEmSpE2TBackSZIkqREWnJIkSZKkRlhwSpIkSZIaYcEpSZIkSWqEBackSZIkqREWnCpGRKyKiHkRcUtE3BQRr1pt/kcj4omIeHFL2wsj4tsRcVtE3B4Rv4iIrduYvfNxat0+KyLuqtt+GxEntqwzJCJmRMSCiLg7Is6KiP59nX0d+ftHxJkR8fs654yIGNKy3r9GxB0RcWu93n5tyv9oy/M3R8TvIuIVETGy/j/o7P+vtyy3b0RcV///3BwR50TEC9uRv87T+hq+HREntUzvV/fxFhHRLyKWRsTU9iTtWjfzz63/L/5Uv4bO421YG/JmRFzUMt3Zr5fV08fXyxzcsszb67ajI2Loaj8z8yLirxHxXxtD/pa8t9bH/m0R8faWeedHxMI6950R8cm+yN2L2ZdExAvq6e0j4p6+yr+u1xARo+v3oK1a5v8kIibXx1XnsX9HRFzS1+87G5D9xRHxrfr3wd318xd3vZdGs3+hZfqUiPhUy/S76+Pmjqj+ljgnIrar563x93Eh+Q+L6nfVLRExPyL+vg35O/9WuD0ivtd57Hbz/bTdx/16Za/b1vh+1Ae5XxrP/I75v/q9r3O6/2rLfrQ7/Vr/LHT0atDM9OGjiAfwaMvzQ4FrV5v/a+B64H0tbf8MfLFleiTwgnZmX619FtBRPx8APAj0BwKY3flagM2BbwKfb3ffr9Z+Rp1r83r6fXXuAPYHbuzsb2B74OXtzA8cBPwe2KWevgJ4W8tye9b/7gD8Edi/Zd7RwA7tyL/6/0Gd7w/AQKoPDn8DvKae9ybgBuBuINqVd33z1/OPB6a1Oy8wD9iqpV/nAZe1ZLwVOKdlne/Wyxzdxfb2BBYBL9tI8u9V/yzsXE/vXE+PrafP73wdwJb1/9fOBWX/E3BSPb09cM9GdvxMBT5TP3878LOW42pay3a+Q8vvtI08+yXAp1q2cxrwvT7O/gSwENi+nj6lMxPwRmAuMLie3hyYAoysp2fRxe/jEvIDWwD3AkPqeS/ofF19fey0PP82cHI3j6mN4rhfz+xrfT/q49fwKeCUtcy/p/PYWsd2nv5Z6K2HI5wq1bZUvwwAiIhdgK2BTwCTW5bbEVjSOZGZd2Xmk30Vsoe2Bh4DVgGvB57IzPMAMnMV8DFgSl9/6rcmdY73AR+r81HnfZIq/47Ass7+zsxlmXlvG/NOBL4BHJaZd9fNOwKLO5fJzNvqpx8ELsjMG1vmXZKZ9/dV3rWpc5wBfA74AHBrZv6inj0ZOIvqD+7925Nw7daRf2PyU+At9fPJwMWrzb8e2DeqkeWtgRFUf4g8S0RsSfUH1Acz8/+ai/sca8t/CnB6Zi4EqP/9LPDxLrazZf3vYw3l7MqGZj8T+FhE9Gs+6hqt7TV8GnhHRIyjKuA+uPrKdfYX0fK7rg/1KHtEjADGA/+x2nId9e/nvrIS+DrV78vV/SvVH+NLoPq9mpnnZuZdXSzb+vu4L61v/m2AfsCf63lPruF19aXrqd4TO63r/RRo+3HfqSfZe/Je2ici4qB6tPu2iDg3Il4QEf8IvBy4JiKuqZf7WkTMqUeVT2sykwWnSrJVfYrAncA5PPsX2yRgOtWbxMiI2KFuPxf4/yLixoj4TETs2reRn9aZvfNxTMu8b0fErcBdwH/Uxdtoqk8yn5aZf6UqIlrfBPtKV/lHAH+qc7WaQ5X/Z8DQ+vSrr0bE6/o6dIsXAJcCb8/MO1vavwRcHRGXR8THOk9NAsawWv9vhP4b2IPql9o/wdOFzcHAj6l+IU5e49rt95z8G6HpwKS6X8dSnUXRKoGfU51x8TbgR2vYzueAX2TmmuY3ZW35n/MewzM/u50+HxHzqD6UmZ6ZDzSYdXUbmv1PwC+A9zQZch3W+Boy829Uf6heR9W3C1rWO6bu9yVUI20/7rPEz+hp9j2AeZ0fPtbLraL6AKb1/6UvfAU4Np57Ou9o4KZ1rNvV7+O+1uP8mfkXqvefP0bExRFxbES07W/8umh8E3BbS/O63k83huN+fbJ35/2oL21JdZbHMZm5J9UHESdl5pepRsEPzMwD62X/NTM7qF7T6yJibFOhLDhVksczc1xm7k51asm3IiLqeZOpfvE9BXwfeAdAZs4DhgOfp3oD+01EjOrz5M9k73x8t2XesZk5FtgJOCUiXtGGfOuytvxdysxHqT7xPhFYCnw3Io5vOOearAB+CZzQ2liPyI4CvgccAPwq6uu+Nnb1sf4/wOWZ+ee6+TDgmsx8nOrn4O0RsXm7Mq7NGvJvVDLzVmAY1fvLT9ew2HSqD7wm0cUn9hHxJqoPAU5uJuWadTP/2nw8M8cBLwMOitWum29SL2SHZ0YZ2vK3zrpeQ2b+GHgI+Opqs77b0u+30YaRkg3I3nb1h6DfAv5xTctExJ71h6d3r/YBcNt/H69v/sz8O6rLRmZTfSBwbl/kXc1WddE4h+pDn292zujGz3S7j/sNyb4x2RxYmJm/q6cvACauYdl3RsRNwM1UBfIeTYWy4FSR6lMdtwcGRsSewK7AlVHdGGISLSM7mfloZv4gM/8BuAh4cxsir1NmLqX69HI/YD5Vsfa0iNiW6pfg7/s+XZfuBnaKiG1Wax8P3AFPn/IzKzM/CXwIOKqPM3Z6Cngn1emP/9I6IzPvrU9LehvV6UxjqPKPf+5mNjpP1Y9Ok4GD65+DucBLqU5v3litnn9j9COq03+7PP0rM2dTXZ+5fcsveAAiYhBVUX1s/SFAO6wp/3PeY2j52W1Vf3g0C3hNA/nWZoOy1yNv86h+9ttlrccPa/kZyMykGuVZ0x+LTetJ9vnAuNZRtfr5uHpeXzuT6gPGF7W03QHsA9XlE3Vxczmw1eorr/b7uB3OZD3y1+1fAg6hPb9vWz+c/nBmLl9t/rqOqXYe9+ubvdvvpRuTiNiZ6oOJg+oPWX7CM5dP9DoLThUpInan+hTnz1R/ZH8qM4fVj5cDL4/qLqSvjoiX1Ov0p/r05o9tC74W9TWRe1MVclcBL4yI99bzNge+AJxfn87Udpn5GNUnZ1/sHEWr876Q6jTVkaudwjyONvZ93W9voTpV6QSAiHhjRGxRP38ZVYG2BJgGHBctd9WNiCNbTtXe6NQfSLwW2KnzZ4HqurCN+bTaEpwLnNZyfW9XTgX+pYv2c4GzM/PmRpJ1z5rynwH8c9R3AK7//Req95lnqU8x24/qvakvbXB24D+p/qhql+4cP2vzGvq+3zt1O3tm/p5qlOQTLc2fAG6q5/Wp+hTT/+XZZ7V8FjgjWu6kThfFJjzn93Gf62n+iNg6Ig5oaR/Hxvm3TnePqXYe92vSG+9HfWEVMKy+rhqqywqurZ8/QnW9L1T3QnkMeLj+2+ZNTYZq58X0Uk91nu4A1V1Qj8vMVRExieeOWv6QaqTzPuBr9am3m1F9gvP9PsrbqjU7wMzMPLV+/u2IeJzqOsPzM3MuQEQcAXw1Iv6NKvtP6fqP2r6wpvz/TPVm+7uIeAq4EzgiMzOqm6icXV8XuZJqZLbPbzPfKjP/EhFvBK6LiKVUp9GeFRFP1It8POubutTH1Rn1KNVTVNcrzWxD7E4vjIjFLdNfBP7SMn0EcHU++6ZYM4DPRcQLsv03y1pX/o1SZi4GvryOZS5fvS0i9qf6gGNoRBzbMuvKzOyzU8XWlD8z50XE/wf8uP7QZQXwT/VlCJ0+HxGfoLpz9lXAD/ogcmvGDcneuewd9Slj+zSdtyvdOX66cExEvIbqfX8x1R08+9x6ZD+B6j2/s1C4kdUuY+hjX6A6swaAzPxpRAwELq8/JH0IuJ3qbuWduvx93CY9yR/AP0XE/wCPUxUSx/d14HVZxzG1URz3a9Ib70d95AmqGzp+r/6w8DdU90yA6oZUMyPi3sw8MCJupvq7bRHV3e0bE9XItSRJkiRJvctTaiVJkiRJjbDglCRJkiQ1woJTkiRJktQIC05JkiRJUiMsOCVJkiRJjbDg1CYlItr6tRsbquT8JWeHsvOXnB3M304lZ4ey85ecHcrOX3J2MH87lZwd2pffglObmqLfCCg7f8nZoez8JWcH87dTydmh7PwlZ4ey85ecHczfTiVnhzblt+CUJEmSJDUiMrPdGfQ8sv322+ewYcMa2/7SpUsZOHBgY9tvWsn5S84ODef/w9xmtltb+iQMfEGDOxg+vsGNe+y0U8nZoez8JWeHPsh/W3Pvm0tXwcDNG9s87Ol75tqUnL+p7HPnPvd4Hz++94+jJvt+7ty5yzKzy41bcKpPdXR05Jw5c9odQ+pbk6PdCTbMxf6ekNTHdi74fXOh75nqmYjnHu+l1WgRMTczO7qa5ym1kiRJkqRGWHBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRGWHBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRGWHBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRGWHBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRGWHBqkzFz5kxGjhzJiBEjmDp1arvj9EjJ2aHs/CVnn/JrGPRDGHN5u5Osv5L7v+TsUHb+krND2flLzj5lGQz6E4xZ0u4k66/k/i85O5Sbf9GiRRx44IHssccejB49mrPOOqvPM0Rm9vlOtf4iYhVwGxDAKuBDmfnLPtjv+cBlmXnJau0vB76cmUdHxAHAKZl52Jq209HRkXPmzOn1fKtWrWK33XbjyiuvZMiQIUyYMIGLL76YPfbYo9f31dtKzg5l5++z7JOjd7dXu+4B2LofvPfXcPubGtlF5eJmfk947LRPyflLzg5l5+/T7Dv3/vvmdU/A1gHvXQa3D+71zT9joe+Zqys5OzSfP+K5x3tv1Wj33Xcf9913H/vssw+PPPII48eP59JLL+31vo+IuZnZ0dU8RzjL83hmjsvMvYB/Bj7bzjCZeW9mHt3ODACzZ89mxIgRDB8+nP79+zNp0iRmzJjR7ljdUnJ2KDt/ydkBJg6CAf3bnWL9ldz/JWeHsvOXnB3Kzl9ydoCJW8KAgv/yLbn/S84OZeffcccd2WeffQDYZpttGDVqFEuW9O0wf8E/dgK2BR4EiIitI+KqiLgpIm6LiLfV7S+KiJ9ExC0RcXtEHFO3HxQRN9fLnhsRL6jb74mIz9XtsyNiRMv+JkbELyPiDxFxdL38sIi4vW9f9nMtWbKEoUOHPj09ZMiQPv9hWl8lZ4ey85ecfVNQcv+XnB3Kzl9ydig7f8nZNwUl93/J2aH8/J3uuecebr75Zvbbb78+3W+/Pt2besNWETEP2BLYEXh93f4EcERm/jUitgd+FRE/At4I3JuZbwGIiBdHxJbA+cBBmfm7iPgWcBJwZr2thzNzz4h4b93WeYrsjsBrgN2BHwHPOr1WkiRJ0sbn0Ucf5aijjuLMM89k22237dN9O8JZns5TanenKia/FdWJ3wGcHhG3Aj8HBgM7UF3veUhE/FdEvDYzHwZGAgsz83f1Ni8AJrbs4+KWf/dvab80M5/KzPn1trslIk6MiDkRMWfp0qU9f8XdMHjwYBYtWvT09OLFixk8uMkLNHpPydmh7PwlZ98UlNz/JWeHsvOXnB3Kzl9y9k1Byf1fcnYoP/+KFSs46qijOPbYYznyyCP7fP8WnAXLzBuB7YGBwLH1v+MzcxxwP7BlXVTuQ1V4fiYi/r07m17D8ydbnnf7av7M/HpmdmRmx8CBA7u7Wo9MmDCBBQsWsHDhQpYvX8706dM5/PDDG9lXbys5O5Sdv+Tsm4KS+7/k7FB2/pKzQ9n5S86+KSi5/0vODmXnz0xOOOEERo0axcknn9yWDJ5SW7CI2B3YHPgz8GLggcxcEREHAq+ol3k58JfMvCgiHgL+DvgcMCwiRmTm74H3ANe2bPoYYGr974199Xo2RL9+/Zg2bRqHHnooq1atYsqUKYwePbrdsbql5OxQdv6SswNM/iXMegCWPQlDZsBpY+CEXdqdqvtK7v+Ss0PZ+UvODmXnLzk7wOSlMOsJWLYKhiyC07aDE7Zpd6ruK7n/S84OZee/4YYbuPDCC9lzzz0ZN24cAKeffjpvfvOb+yyDX4tSmJavRYFqlPFfMvMn9XWbPwa2BuYArwTeRHX67OeBp4AVwEmZOSciDgLOoPrQ4Td1+5MRcQ/w3XrdJ4HJmfn71b8WJSIezcytI2JY3T6mnV+LIm3UGvpalD7T0NeiSNIaNfC1KH2moa9F0aarya9F6Str+1oUC049S11wdmTmsia2b8Gp5yULTknqGQtOPY9s6gWn13BKkiRJkhrhNZx6lswc1u4MkiRJkjYNjnBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRGWHBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRGWHBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRGWHBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRG9Gt3AKkc0e4AGyjbHeD5643tDiDpeef9hf/Oel27A0jqLY5wSpIkSZIaYcEpSZIkSWqEBackSZIkqREWnJIkSZKkRlhwSpIkSZIaYcEpSZIkSWqEBackSZIkqREWnJIkSZKkRlhwSpIkSZIaYcEpSZIkSWqEBackSZIkqREWnJIkSZKkRlhwSpIkSZIaYcEpSZIkSWqEBackSZIkqREWnJIkSZKkRlhwSpIkSZIaYcEpSZIkSWqEBackSZIkqREWnNpkzJw5k5EjRzJixAimTp3a7jjPsWgRHHgg7LEHjB4NZ51Vtd9yC+y/P+y5J7z1rfDXvz6zzq23VvNGj67mP/FEe7Kvy8be92tTcnaAYf8P9vwEjPs36PhUu9P0XMn9X3J2KDt/ydmh3PyLHoUDfwJ7XAKjL4Gzbm93ovWz6inY+8dw2FXtTtJzpR47UHZ22PjzP/HEE+y7777stddejB49mk9+8pMALFy4kP32248RI0ZwzDHHsHz5cgA+9rGPMW7cOMaNG8duu+3Gdttt11i2yMzGNq5nRMQq4LaWpumZ2e2jNSI+Cnw9M/9WTz+amVtvQJ5ZwCmZOWe19uOBjsz8UA+3Nwx4VWZ+Z23LdXR05Jw5c9a2yHpZtWoVu+22G1deeSVDhgxhwoQJXHzxxeyxxx69uJfYoLXvu6967LMPPPIIjB8Pl14Kxx0HZ5wBr3sdnHsuLFwI//EfsHJlteyFF8Jee8Gf/wzbbQebb76+CZr5We+bvm9Gn2W/YMOOnbUZ9v9gzqdg+20a2wUc57GzupKzQ9n5S84OfZj//b3/vnPf36rHPtvDI8th/KVw6SGwx0t6fVewooFt1r54B8z5M/x1BVx2UAM7ON/3zNWVnB2azx/x3J/XntZomcljjz3G1ltvzYoVK3jNa17DWWedxRe/+EWOPPJIJk2axAc+8AH22msvTjrppGete/bZZ3PzzTdz7rnnbshrmJuZHV3Nc4Sz7zyemeNaHs8pNiNibaXER4EX9mSHEdGvhxk3xDDgXX24v2eZPXs2I0aMYPjw4fTv359JkyYxY8aMdsXp0o47VgUkwDbbwKhRsGQJ/O53MHFi1X7IIfD971fPf/YzGDu2KjYBXvrSDSk2m1NC369Jydk3BSX3f8nZoez8JWeHsvPv+MKq2ATYpj+M2g6WPNbWSD22+DH4yWL4u13bnaTnSj52Ss4OZeSPCLbeuhqLWrFiBStWrCAiuPrqqzn66KMBOO6447j00kufs+7FF1/M5MmTG8tmwdlmEXFPRPxXRNwEnFr/2zlv14i4KSL+EXg5cE1EXNMy/z8j4paI+FVE7FC3nR8R/x0RvwY+FxHj6vm3RsQPI6L1c8j3RMS8iLg9IvbtIttbI+LXEXFzRPy8ZR+vq9ebV8/bBpgKvLZu+1gjnbUWS5YsYejQoU9PDxkyhCVLlvR1jG675x64+WbYb7/qdNnO96zvfa869RaqQjQCDj20KlQ/97m2xV2r0vq+VcnZO0XAG86A8Z+Er89qd5qeKbn/S84OZecvOTuUn7/TPY/AzX+G/Qa1O0nPfPQ38LkO2Ky5E08aU/KxU3J2KCf/qlWrGDduHIMGDeKQQw5hl112YbvttqNfv2oMqqvcf/zjH1m4cCGvf/3rG8tlwdl3tmop0uZFxDEt8/6cmftk5n8CD0fEuLr9fcB5mfll4F7gwMw8sJ73IuBXmbkXcB3w/pbtDaE6vfVk4FvA/5eZY6lO6f1ky3IvzMxxwD8AXY2h/wJ4ZWbuDUwH/qluPwX4YL3ua4HHgVOB6+vR2y+1biQiToyIORExZ+nSpd3pq03ao4/CUUfBmWfCtttWp9F+9avVKbaPPAL9+1fLrVwJv/gFfPvb1b8//CFcVeD1JmrWL/4VbjoNLv9/8JWr4Lq72p1I0qbu0RVw1M/hzFfCtv3bnab7LlsEg7aE8S9tdxKpGZtvvjnz5s1j8eLFzJ49mzvvvHOd60yfPp2jjz6azRs8jc6Cs++sfkrtd1vmtT4/B3hffXrtMcCarolcDlxWP59LdUprp+9l5qqIeDGwXWZeW7dfAExsWe5igMy8Dtg2IrZbbR9DgCsi4jbg48Douv0G4Iv1yOt2mblybS88M7+emR2Z2TFw4MC1LbreBg8ezKLOoUFg8eLFDB48uJF9bYgVK6pi89hj4cgjq7bdd69On507FyZPhl12qdqHDKlOtd1+e3jhC+HNb4abblrzttullL7vSsnZOw2uz1kYtC0csQ/M/kN78/REyf1fcnYoO3/J2aH8/CueqorNY0fAkTu3O03P3PAA/GgRDLsEJl0LV98H776+3am6r+Rjp+TsUF7+7bbbjgMPPJAbb7yRhx56iJUrqz/Vu8o9ffr0Rk+nBQvOjUXrFRDfB94EHAbMzcw/r2GdFfnM1cSrgNbrNbt7RcXqVyOvPn02MC0z9wT+HtgSoL7+9O+ArYAbImL3bu6vMRMmTGDBggUsXLiQ5cuXM336dA4//PB2x3qWTDjhhOrazZNPfqb9gQeqf596Cj7zGfjAB6rpQw+F226Dv/2tGu289trqDrcbmxL6fk1Kzg7w2JPwyOPPPP/ZHTBm4/399xwl93/J2aHs/CVnh7LzZ8IJ11XXbp68Z7vT9Nxnx8Pid8A9R8P018Hrd4SLXtvuVN1X8rFTcnYoI//SpUt56KGHAHj88ce58sorGTVqFAceeCCXXHIJABdccAFve9vbnl7nzjvv5MEHH2T//fdvNFtf3lRG3ZCZT0TEFcDXgBNaZj0CbAMs68G2Ho6IByPitZl5PfAe4NqWRY6hui70NcDD9fKtm3gx0Hmi93GdjRGxS2beBtwWEROA3YFFdb626NevH9OmTePQQw9l1apVTJkyhdGjR697xT50ww3VHWf33BPGjavaTj8dFiyAr3ylmj7ySHjf+6rnL3lJVZhOmFBdp/fmN8Nb3tKW6GtVQt+vScnZAe5/GI44u3q+chW865XwxrHtzdQTJfd/ydmh7PwlZ4ey899wP1z4e9jzJTDuB1Xb6RPgzUPXvp56R8nHTsnZoYz89913H8cddxyrVq3iqaee4p3vfCeHHXYYe+yxB5MmTeITn/gEe++9Nyec8Ex5MX36dCZNmtTlXXJ7k1+L0ke6+FqUmZl5akTcQ/U1JMtaln0lcAnwisxcVbd9GPgQcG9mHtj6tSgRcTRwWGYeHxHnA5dl5iX1vHHAf1Pd4fYPwPsy88H6a1HmAa8DtgCmZObs1q9FiYi3AV8CHgSuBiZk5gERcTZwIPAUcAdwfP38CuClwPmrX8fZqamvRekbBd5h4Fn8WW+bBr8WpU809LUokhrUwNei9KkGvxalcQ19LYo2Xb3xtSjttravRbHg3AhFxCnAizPz39qdpbdZcLaTP+ttY8Epqa9ZcLaPBad6aFMvOD2ldiMTET8EdgGauzexJEmSJPUBC86NTGYe0e4MkiRJktQbvEutJEmSJKkRFpySJEmSpEZYcEqSJEmSGmHBKUmSJElqhAWnJEmSJKkRFpySJEmSpEZYcEqSJEmSGmHBKUmSJElqhAWnJEmSJKkRFpySJEmSpEZYcEqSJEmSGmHBKUmSJElqhAWnJEmSJKkRFpySJEmSpEZYcEqSJEmSGtGv3QGkcmS7A6hUx3nsSOpj3yj8fWfnaHcCSb3EEU5JkiRJUiMsOCVJkiRJjbDglCRJkiQ1woJTkiRJktQIC05JkiRJUiMsOCVJkiRJjbDglCRJkiQ1woJTkiRJktQIC05JkiRJUiMsOCVJkiRJjbDglCRJkiQ1woJTkiRJktQIC05JkiRJUiMsOCVJkiRJjbDglCRJkiQ1woJTkiRJktQIC05JkiRJUiMsOCVJkiRJjbDglCRJkiQ1woJTm4yZM2cycuRIRowYwdSpU9sdp0dKzg5l5y85O5i/nUrODmXnLzk7lJ2/5OxTlsGgP8GYJe1Osv5K7v+Ss0PZ+adMmcKgQYMYM2ZMewJkpo+N5AGsAua1PE5dj218AHhvL+caBty+hnmfBg6un88COta2rfHjx2cTVq5cmcOHD8+77747n3zyyRw7dmzecccdjeyrt5WcPbPs/CVnzzR/O5WcPbPs/CVnzyw7f59mH0avP659GTl3R3L0Fr2/7Wc9GuKx0z5N5wee8+hN1157bc6dOzdHjx7dq9ttBczJNfz97wjnxuXxzBzX8ujxxyeZ+d+Z+a0mwq1hf/+emT/vq/2tyezZsxkxYgTDhw+nf//+TJo0iRkzZrQ7VreUnB3Kzl9ydjB/O5WcHcrOX3J2KDt/ydkBJm4JAwr+y7fk/i85O5Sff+LEiQwYMKBt+y/4x+75IyIOioibI+K2iDg3Il5Qt0+NiPkRcWtEnFG3fSoiTqmfz4qI/4qI2RHxu4h4bd2+ZUScV2/v5og4sG4/PiJm1OstiIhPtsTYPCK+ERF3RMTPImKrep3zI+LoPu2QLixZsoShQ4c+PT1kyBCWLCnjnJmSs0PZ+UvODuZvp5KzQ9n5S84OZecvOfumoOT+Lzk7lJ+/3Sw4Ny5bRcS8lscxEbElcD5wTGbuCfQDToqIlwJHAKMzcyzwmTVss19m7gt8FOgsID9INVS/JzAZuKDeD8C+wFHAWOAdEdFRt+8KfCUzRwMP1ctIkiRJ0hpZcG5cVj+l9rvASGBhZv6uXuYCYCLwMPAE8M2IOBL42xq2+YP637lU12ICvAa4CCAz7wT+COxWz7syM/+cmY/X676mbl+YmfO62NY6RcSJETEnIuYsXbq0u6v1yODBg1m0aNHT04sXL2bw4MGN7Ku3lZwdys5fcnYwfzuVnB3Kzl9ydig7f8nZNwUl93/J2aH8/O1mwVmozFxJNRp5CXAYMHMNiz5Z/7uKanR0nZtew/STLW3d3Va1gcyvZ2ZHZnYMHDiwu6v1yIQJE1iwYAELFy5k+fLlTJ8+ncMPP7yRffW2krND2flLzg7mb6eSs0PZ+UvODmXnLzn7pqDk/i85O5Sfv926XTSobe4ChkXEiMz8PfAe4NqI2Bp4YWb+NCJuAP7Qg21eDxwLXB0RuwE71fvZBzgkIgYAjwNvB6b03ktpTr9+/Zg2bRqHHnooq1atYsqUKYwePbrdsbql5OxQdv6Ss4P526nk7FB2/pKzQ9n5S84OMHkpzHoClq2CIYvgtO3ghG3anar7Su7/krND+fknT57MrFmzWLZsGUOGDOG0007jhBNO6LP9R3UXW20MImIVcFtL08zMPDUiDgLOoPqA4DfAScAAYAawJRDAGZl5QUR8Cng0M8+IiFnAKZk5JyK2p7pd8bD6es2vAR3ASuDkzLwmIo6nKjJfDAwBLsrM0yJiGHBZZo6pc54CbJ2Zn4qI8+t5l7Tub02vsaOjI+fMWeNsSZIk2DnanWD9LfRva/VMxHOP99JqtIiYm5kdXc4r7cWoOXXB2ZGZH2pqHxackiRpnSw49TyyqRecXsMpSZIkSWqE13DqaZl5PtVXsEiSJEnSBnOEU5IkSZLUCAtOSZIkSVIjLDglSZIkSY2w4JQkSZIkNcKCU5IkSZLUCAtOSZIkSVIjLDglSZIkSY2w4JQkSZIkNcKCU5IkSZLUCAtOSZIkSVIjLDglSZIkSY2w4JQkSZIkNcKCU5IkSZLUCAtOSZIkSVIjLDglSZIkSY2w4JQkSZIkNaJfuwNIkiRJzzKs3QEk9RZHOCVJkiRJjbDglCRJkiQ1woJTkiRJktQIC05JkiRJUiMsOCVJkiRJjbDglCRJkiQ1woJTkiRJktQIC05JkiRJUiMsOCVJkiRJjbDglCRJkiQ1woJTkiRJktQIC05JkiRJUiMsOCVJkiRJjbDglCRJkiQ1woJTkiRJktQIC05JkiRJUiMsOCVJkiRJjbDglCRJkiQ1woJTkiRJktQIC05tMmbOnMnIkSMZMWIEU6dObXecHik5O5Sdv+TsYP52Kjk7lJ2/5OxQdv6Ssy96Ag6cB3vMhtGz4azF7U7UcyX3f8nZodz8ixYt4sADD2SPPfZg9OjRnHXWWX0fIjN9NPwA/hW4A7gVmAfs18A+DgAua8NrGwbc3t0M48ePzyasXLkyhw8fnnfffXc++eSTOXbs2Lzjjjsa2VdvKzl7Ztn5S86eaf52Kjl7Ztn5S86eWXb+Ps1+AL3+uHd/cu746vlfX0PuuhV5x4Te309TPHbap+n8wHMeveXee+/NuXPnZmbmX//619x1110b6XtgTq7h739HOBsWEfsDhwH7ZOZY4GBgUXtTbXpmz57NiBEjGD58OP3792fSpEnMmDGj3bG6peTsUHb+krOD+dup5OxQdv6Ss0PZ+UvODrDjC2Cfbarn2/SDUS+EJU+2N1NPlNz/JWeHsvPvuOOO7LPPPgBss802jBo1iiVLlvRpBgvO5u0ILMvMJwEyc1lm3gsQEfdExGcjYl5EzImIfSLiioi4OyI+UC9zQERc1rmxiJgWEcfXz98YEXdGxE3AkS3LDIyIKyPijog4JyL+GBHb1/PeHRGz633+T0RsvnrgiJgQEb+MiFvqZbeJiM0j4vMR8ZuIuDUi/r7BPuuxJUuWMHTo0KenhwwZ0uc/TOur5OxQdv6Ss4P526nk7FB2/pKzQ9n5S86+unseh5sfhf22bXeS7iu5/0vODuXn73TPPfdw8803s99++/Xpfi04m/czYGhE/C4ivhoRr1tt/p8ycxxwPXA+cDTwSuC0tW00IrYEvgG8FRgPvKxl9ieBqzNzNHAJsFO9zijgGODV9T5XAceutt3+wHeBj2TmXlQjso8DJwAPZ+YEYALw/ojYuTsdEBEn1gX1nKVLl3ZnFUmSpEY8uhKOugPOHAHb9mt3GqlvPProoxx11FGceeaZbLtt337SYsHZsMx8lKogPBFYCny3c4Sy9qP639uAX2fmI5m5FHgyIrZby6Z3BxZm5oL6vOmLWua9Bphe738m8GDdflCd5TcRMa+eHr7adkcC92Xmb+r1/5qZK4E3AO+t1/s18FJg1272wdczsyMzOwYOHNidVXps8ODBLFr0zJnKixcvZvDgwY3sq7eVnB3Kzl9ydjB/O5WcHcrOX3J2KDt/ydk7rXiqKjaP3QGObOZPksaU3P8lZ4fy869YsYKjjjqKY489liOPPHLdK/QyC84+kJmrMnNWZn4S+BBwVMvszqsHnmp53jndD1jJs/+fttyAKAFckJnj6sfIzPxUD9b9cMu6O2fmzzYgS6+aMGECCxYsYOHChSxfvpzp06dz+OGHtztWt5ScHcrOX3J2MH87lZwdys5fcnYoO3/J2QEy4YS7qms3Tx667uU3NiX3f8nZoez8mckJJ5zAqFGjOPnkk9uSwRMJGhYRI4GnMnNB3TQO+GMPNvFHYI+IeAGwFdWo5C+AO4FhEbFLZt4NTG5Z5wbgncB/RcQbgJfU7VcBMyLiS5n5QEQMALbJzNY8dwE7RsSEzPxNRGxDdUrtFcBJEXF1Zq6IiN2Ajebk9X79+jFt2jQOPfRQVq1axZQpUxg9enS7Y3VLydmh7PwlZwfzt1PJ2aHs/CVnh7Lzl5wd4IaH4cL7Yc8XwbjfVG2nD4c3v7S9ubqr5P4vOTuUnf+GG27gwgsvZM8992TcuHEAnH766bz5zW/uswxRnY2ppkTEeOBsYDuq0crfAydm5rKIuAfoqJ8fXz//UL1e67zPAUcAC4FHgR9l5vkR8UbgTOBvVNeA7pKZh0XEIOBiYAfgRqq75A7LzCcj4hjgn6lGTVcAH8zMX62WeUKdeSuqYvPgeh+fobpmNKhOD347VTF7WWaOiYgDgFMy87A19UdHR0fOmTOn5x0pSZKePw6MdidYf9f4t7V6JuK5x3tpNVpEzM3Mji7nlfZitG71aOiqzFxZfy3L1+qbBLWdBackSVonC049j2zqBaen1G6adgL+NyI2A5YD729zHkmSJEnPQxacm6D6etG9251DkiRJ0vObd6mVJEmSJDXCglOSJEmS1AgLTkmSJElSIyw4JUmSJEmNsOCUJEmSJDXCglOSJEmS1AgLTkmSJElSIyw4JUmSJEmNsOCUJEmSJDXCglOSJEmS1AgLTkmSJElSIyw4JUmSJEmNsOCUJEmSJDXCglOSJEmS1AgLTkmSJElSI/q1O4BUjA9FuxNsmGnZ7gTPX3cXfuzs4rEjqY/9Q7sDSOotjnBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRGWHBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRGWHBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRGWHBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRGWHBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERac2mTMnDmTkSNHMmLECKZOndruOOs05SoYdC6MufiZtu/9HkZ/Bzb7Csx5oH3Zeqq0vm9VWvYpp8KgfWHMm55pu+W3sP/RsOeb4a3vh78+0r58PVVa/7cqOTuUnb/k7FB2/tKyL1oGB54Ge3wMRp8MZ/20av/ejdX0ZsfAnLvbm7EnSuv/ViVnh/LyDxs2jD333JNx48bR0dEBwPe+9z1Gjx7NZpttxpw5c/osy0ZfcEbEo320n1kR0dHg9v9lPde7JCKGr2OZj0bEC9cvGUTEORGxx3qsd0BEXFY/PywiPr2+GTbUqlWr+OAHP8jll1/O/Pnzufjii5k/f3674nTL8aNg5luf3TZmAPzgTTDx5e3JtD5K7PtOJWY//kiYee6z2/7uX2Dqx+G2n8IRb4DPn9OebD1VYv93Kjk7lJ2/5OxQdv4Ss/fbHL7wHpj/JfjVf8JXroD5i2HMUPjBKTBxVLsTdl+J/d+p5OxQbv5rrrmGefPmPV1cjhkzhh/84AdMnDixT3Ns9AXnJqTHBWdEjAY2z8w/rGPRjwLrXXBm5t9l5ob+1PwEeOuGFL4bYvbs2YwYMYLhw4fTv39/Jk2axIwZM9oRpdsmvhwGvODZbaMGwMiXtCfP+iqx7zuVmH3ivjBgu2e3/W5h1Q5wyKvh+zP7PNZ6KbH/O5WcHcrOX3J2KDt/idl3fAnsU39sv81WMGowLPkLjBoCIwv6cBfK7P9OJWeH8vN3GjVqFCNHjuzz/RZVcEbEZhGxICIGtkz/PiIGRsT5EfG1iPhVRPyhHn07NyJ+GxHnt2zj0Yj4UkTcERFXdW6r9o6ImB0Rv4uI19bLbxkR50XEbRFxc0QcWLdvHhFnRMTtEXFrRHw4Il4fEZe27OuQiPhhREwFtoqIeRHx7Xreu+t9zYuI/4mIzbt4yccCM1q297WImFNnP61u+0fg5cA1EXFNREyJiDNb1nl//XqHRcSdEfHtuk8u6SwOW0d3I+KNEXFTRNwSEVfVbftGxI316/9lRDznSM3MBGYBh3X3/7M3LVmyhKFDhz49PWTIEJYsWdKOKM87Jfd9ydlbjd4VZvy8ev69y2HR/7U3T3eV3P8lZ4ey85ecHcrOX3J2gHsegJsXwn4j2p1k/ZTc/yVnhzLzRwRveMMbGD9+PF//+tfbmqWogjMznwIuoirEAA4GbsnMpfX0S4D9gY8BPwK+BIwG9oyIcfUyLwLmZOZo4Frgky276JeZ+1KNGHa2f7Dade4JTAYuiIgtgROBYcC4zBwLfBu4Bti9pYh9H3BuZp4KPJ6Z4zLz2IgYBRwDvDozxwGrWl5Tq1cDc1um/zUzO4CxwOsiYmxmfhm4FzgwMw8E/pdqpHGL1gz185HAVzNzFPBX4B9ad1bn/gZwVGbuBbyjnnUn8NrM3Bv4d+D0LrICzAFeu4Z5khpy7lT46kUw/m3wyGPQf4t1ryNJzyePPgFHfQHOPB62bcu5WFLf+sUvfsFNN93E5Zdfzle+8hWuu+66tmUpquCsnQu8t34+BTivZd6P65G224D7M/O2uki9g6o4BHgK+G79/CLgNS3r/6D+d27L8q+plyMz7wT+COxGVez+T2aurOf9pd73hcC7I2I7quL38i5ew0HAeOA3ETGvnu7qOs0dgaUt0++MiJuAm6kK6edcd5mZjwJXA4dFxO7AFpl5Wz17UWbesIbXDvBK4LrMXNj5mur2FwPfi4jbeaaI78oDVKOtzxIRJ9Yjs3OWLl3axWobbvDgwSxatOjp6cWLFzN48OBG9qVnK7nvS87eavdd4GcXwNwZMPmtsMtO7U7UPSX3f8nZoez8JWeHsvOXmn3FyqrYPPa1cOR+7U6z/krtfyg7O5SZvzPfoEGDOOKII5g9e3bbshRXcGbmIuD+iHg9sC/PLuierP99quV553S/NW2yi/VXrWX5dTkPeDfVaOj3OgvS1QRwQT3iOS4zR2bmp7pY7nFgS4CI2Bk4BTioHlH9See8LpwDHE81utlakOdqy60+vSb/AVyTmWOAt65lv1vWmZ+9k8yvZ2ZHZnYMHDiwi9U23IQJE1iwYAELFy5k+fLlTJ8+ncMPP7yRfenZSu77krO3euDP1b9PPQWf+Qp8YHJ783RXyf1fcnYoO3/J2aHs/CVmz4QT/ru6dvPktlz003tK7P9OJWeH8vI/9thjPPLII08//9nPfsaYMWPalmd9i6p2O4dqhO7CzFzVw3U3A44GpgPvAn6xjuWvpzrd9eqI2A3YCbgLuBL4+4i4JjNXRsSAepTz3oi4F/gE1ShopxURsUVmrgCuAmZExJcy84GIGABsk5l/XG3fvwVGAPcA2wKPAQ9HxA7Am6iumQR4BNgGWAaQmb+OiKHAPlSn33baKSL2z8wb1/DafwV8NSJ2zsyFna+JaoSz80T149fSV7sBt69lfmP69evHtGnTOPTQQ1m1ahVTpkxh9Og1DcRuHCb/DGYtgWVPwJDz4bR9YcCW8OHrYOnj8JbLYNz2cMXG+34GlNn3nUrMPvmjMOvXsOxBGPJqOO0j8Ojf4CsXVfOPfAO87+i2Ruy2Evu/U8nZoez8JWeHsvOXmP2Gu+DC62DPnWDcx6u20yfDkyvhw+fC0r/CW6bCuGFwxb+2Neo6ldj/nUrODuXlv//++zniiCMAWLlyJe9617t44xvfyA9/+EM+/OEPs3TpUt7ylrcwbtw4rrjiisbzRHUW6MYrIp6iukax0xeBs4E/A/vWp7lS3xjossy8JCKG1c/HdDHvUeDrwBuoTgE9JjOXRsQs4JTMnBMR21Nd5zmsvl7za0AHsBI4OTOviYh+wOeANwIrgG9k5rR6f5OAj2bmK1tex38BhwM31ddxHgP8M1UBvAL4YGb+arXX/h5gZGZ+ouV1vApYBDwM/Cgzz4+IDwMfAu6tr+MkIk6lur50Uj09DJhJdZ3leGA+8J7M/Ntqr/1NVNdobgY8kJmHRMT+wAVUBe9PgHfXfXNAvd5h9T4uA/655RTe5+jo6Mi+/N6fXvWhaHeCDTNt4/5Z36TdXfixs4vHjqQ+9r2C3zff4Xumeibiucf7xl6jrS4i5tb3mnnuvNJeDEB9R9UvZWaPb1ATEY9m5tYNxGrdxzTg5sz85gZuZyuqGxG9uqcjuXXx96XM7LzT7DBaivDeVo+6ficzD1rbchacbWTB2T4WnJLUMxaceh7Z1AvO4q7hrEfuvk81OrjRiYi5VKexXrSh28rMx6nultvtq5IjYruI+B3VXXGv2tAMPbAT8P/6cH+SJEmSNnLFXcOZmVOBqRuwfqOjm5k5vpe316MTqzPzIaprKVdvvwdo7GrhzPxNU9uWJEmSVKbiRjglSZIkSWWw4JQkSZIkNcKCU5IkSZLUCAtOSZIkSVIjLDglSZIkSY2w4JQkSZIkNcKCU5IkSZLUCAtOSZIkSVIjLDglSZIkSY2w4JQkSZIkNcKCU5IkSZLUCAtOSZIkSVIjLDglSZIkSY2w4JQkSZIkNcKCU5IkSZLUCAtOSZIkSVIj+rU7gFSMr7Q7wAaa1u4Az2O7ZLsTSHq+OSTanWDDDG53gA3wjnYHkDYujnBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRGWHBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRGWHBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRGWHBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRGWHBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERac2mTMnDmTkSNHMmLECKZOndruOD0yBRgEjGl3kPVUct+XnB3M304lZ4ey85ecHcrOP+UOGHQtjLmx3UnW36qnYO8fw2FXtTtJz5V87JScHcrO/9BDD3H00Uez++67M2rUKG68sW9/gIstOCPi0T7az6yI6Ghw+/+ynutdEhHDN2QbLdv6aURstx7rHR8R0+rnH4qIKRuSY0OsWrWKD37wg1x++eXMnz+fiy++mPnz57crTo8dD8xsd4j1VHLfl5wdzN9OJWeHsvOXnB3Kz3/8y2Hm3u1OsWHO+i2MenG7U/RcycdOydmh/Pwf+chHeOMb38idd97JLbfcwqhRo/p0/8UWnJuQHheLETEa2Dwz/7C+22iVmW/OzIc2ZBvAucCHN3Ab62327NmMGDGC4cOH079/fyZNmsSMGTPaFafHJgID2h1iPZXc9yVnB/O3U8nZoez8JWeH8vNPfAkM2KLdKdbf4sfgJ4vh73Ztd5KeK/nYKTk7lJ3/4Ycf5rrrruOEE04AoH///my33XZ9mmGTKDgjYrOIWBARA1umfx8RAyPi/Ij4WkT8KiL+EBEHRMS5EfHbiDi/ZRuPRsSXIuKOiLiqc1u1d0TE7Ij4XUS8tl5+y4g4LyJui4ibI+LAun3ziDgjIm6PiFsj4sMR8fqIuLRlX4dExA8jYiqwVUTMi4hv1/PeXe9rXkT8T0Rs3sVLPhaYUS//rG1ExKcj4qMt+/rPiPhI/bqvi4ifRMRdEfHfEbFZvcw9EbF9/fy9de5bIuLCuu2tEfHr+nX+PCJ2WD1QZv4NuCci9u3xf2AvWLJkCUOHDn16esiQISxZsqQdUZ53Su77krOD+dup5OxQdv6Ss0P5+Uv30d/A5zpgs2h3kp4r+dgpOTuUnX/hwoUMHDiQ973vfey999783d/9HY899lifZtgkCs7MfAq4iKoQAzgYuCUzl9bTLwH2Bz4G/Aj4EjAa2DMixtXLvAiYk5mjgWuBT7bsol9m7gt8tKX9g9Wuc09gMnBBRGwJnAgMA8Zl5ljg28A1wO4tRez7gHMz81Tg8cwcl5nHRsQo4Bjg1Zk5DljV8ppavRqYW7/2Z22DaqTxvVAV3sCkum8A9qUahdwD2AU4snWj9cjpJ4DXZ+ZewEfqWb8AXpmZewPTgX/qIhPAHOC1qzdGxIkRMSci5ixdurSL1SRJkpp12SIYtCWMf2m7k0h9Z+XKldx0002cdNJJ3HzzzbzoRS/q82tQN4mCs/Z0oUV1D5bzWub9ODMTuA24PzNvq4vUO6iKQ4CngO/Wzy8CXtOy/g/qf+e2LP+aejky807gj8BuVMXu/2TmynreX+p9Xwi8u75Wcn/g8i5ew0HAeOA3ETGvnh7exXI7Al1Wbpl5D/DniNgbeANwc2b+uZ49OzP/kJmrgItXe40Arwe+l5nLOrPX7UOAKyLiNuDjVMV6Vx4AXt5Fpq9nZkdmdgwcOLCL1Tbc4MGDWbRo0dPTixcvZvDgwY3sS89Wct+XnB3M304lZ4ey85ecHcrPX7IbHoAfLYJhl8Cka+Hq++Dd17c7VfeVfOyUnB3Kzj9kyBCGDBnCfvvtB8DRRx/NTTfd1KcZNpmCMzMXAfdHxOupRvJaC7on63+fanneOd1vTZvsYv1Va1l+Xc4D3k01Gvq9zoJ0NQFcUI9WjsvMkZn5qS6WexzYci37OofqPjTvoyrEO+Vqy60+vSZnA9Pq0dy/X8u+t6yz9bkJEyawYMECFi5cyPLly5k+fTqHH354O6I875Tc9yVnB/O3U8nZoez8JWeH8vOX7LPjYfE74J6jYfrr4PU7wkXPOS9r41XysVNydig7/8te9jKGDh3KXXfdBcBVV13FHnvs0acZ1rd42lidQzXqeGE9itcTmwFHU50y+i6q00jX5nqq012vjojdgJ2Au4Argb+PiGsyc2VEDKhHOe+NiHupTlk9uGU7KyJii8xcAVwFzIiIL2XmAxExANgmM/+42r5/C4wA7uliGwA/BD4NbFG/lk77RsTOVKOxxwBfX227VwM/jIgvZuafO7MDLwY6T1Q/bi19shtww1rmN6Zfv35MmzaNQw89lFWrVjFlyhRGj17TQOzGZzIwC1hGNZx8GnBCOwP1QMl9X3J2MH87lZwdys5fcnYoP//k22DWg7BsBQy5Hk4bDieUMdBTvJKPnZKzQ/n5zz77bI499liWL1/O8OHDOe+889a9Ui+K6mzP8kTEU8C9LU1fpBqJ+zOwb32aK/WNgS7LzEsiYlj9fEwX8x6lKsDeQHVq6DGZuTQiZgGnZOac+sY6czJzWH295teADmAlcHJmXhMR/YDPAW8EVgDfyMzOrw6ZBHw0M1/Z8jr+CzgcuKm+jvMY4J+pCuAVwAcz81ervfb3ACMz8xNdbaNu+2/gofoaTyLiAKoi9BGqYvUa4B8y86mIuAfoyMxlEXEc1Wmzq6hOxz0+It5Gdd3rg1RF6YTMPCAijq/X+1C9j5uAQ1pO4X2Ojo6OnDNnzppmb9yiwDsMtCr0Z12StB4OKfx3VslF7Pn+vlXPRBd/Y5ZWo0XE3Mzs8qskiy04uxLV92V+KTN7fIJERDyamVs3EKt1H9OoirhvbuB2tqIqGF/d1UhufbOgm4B3ZOaCuu0AqsL5sA3Z91oy7U1VdL9nbctZcLbRJvSzLklaBwvO9rHgVA9t6gXnJnMNZ0ScCnyfanRwoxMRc4GxPHPH2PWWmY9T3S33OW/HEbEH8Hvgqs5is49sD/xbH+5PkiRJ0kZukxrh1MbPEc428mddkp4/HOFsH0c41UOOcEqSJEmStB4sOCVJkiRJjbDglCRJkiQ1woJTkiRJktQIC05JkiRJUiMsOCVJkiRJjbDglCRJkiQ1woJTkiRJktQIC05JkiRJUiMsOCVJkiRJjbDglCRJkiQ1woJTkiRJktQIC05JkiRJUiMsOCVJkiRJjbDglCRJkiQ1woJTkiRJktSIfu0OIBUjs90JJEl6fjjf37nSpsIRTkmSJElSIyw4JUmSJEmNsOCUJEmSJDXCglOSJEmS1AgLTkmSJElSIyw4JUmSJEmNsOCUJEmSJDXCglOSJEmS1AgLTkmSJElSIyw4JUmSJEmNsOCUJEmSJDXCglOSJEmS1AgLTkmSJElSIyw4JUmSJEmNsOCUJEmSJDXCglOSJEmS1AgLTkmSJElSIyw4JUmSJEmNsODUJmPmzJmMHDmSESNGMHXq1HbH6ZGSs0PZ+UvODuZvp5KzQ9n5S84OZeefcgcMuhbG3NjuJOun5L6HsvOXnB3Kzj9lyhQGDRrEmDFj2hMgM3104wEkcFHLdD9gKXDZOtYbB7y5jbnPB47uxe0dALxqfdcfP358NmHlypU5fPjwvPvuu/PJJ5/MsWPH5h133NHIvnpbydkzy85fcvZM87dTydkzy85fcvbMPsx/MI08rh1Pzt2XHP2iZrb/9KMBHjvtU3L2zObz13XGsx696dprr825c+fm6NGje3W7rYA5uYa//x3h7L7HgDERsVU9fQiwpBvrjQPe3FSoNjgAeFW7Q6xu9uzZjBgxguHDh9O/f38mTZrEjBkz2h2rW0rODmXnLzk7mL+dSs4OZecvOTuUn3/iS2DAFu1OsX5K7/uS85ecHcrPP3HiRAYMGNC2/Vtw9sxPgbfUzycDF3fOiIgXRcS5ETE7Im6OiLdFRH/g08AxETEvIo6JiH0j4sZ6mV9GxMh6/eMj4tKIuDIi7omID0XEyfVyv4qIARGxS0Tc1LLPXTunI2J8RFwbEXMj4oqI2HH18BFxUL292+qsL6jbJ9RZbqnzbxMR10XEuJZ1fxERewEfAD5Wv57XRsTAiPh+RPymfry6tzu9O5YsWcLQoUOfnh4yZAhLlnTn84D2Kzk7lJ2/5Oxg/nYqOTuUnb/k7FB+/pKV3vcl5y85O5Sfv90sOHtmOjApIrYExgK/bpn3r8DVmbkvcCDweWAL4N+B72bmuMz8LnAn8NrM3Lued3rLNsYARwITgP8E/lYvdyPw3sy8G3i4pRB8H3BeRGwBnE116ux44Nx6/afVmc8HjsnMPalOCT6pLoq/C3wkM/cCDgYeB74JHF+vuxuwZWbeAvw38KX69VwPnFVPTwCOAs7pebdKkiRJ2hT1a3eAkmTmrRExjGp086erzX4DcHhEnFJPbwns1MVmXgxcEBG7Up2j3XpiyjWZ+QjwSEQ8DPy4br+NqsCFqqB7X0ScDBwD7AuMpCpWr4wIgM2B+1bb70hgYWb+rp6+APggcBVwX2b+pn6NfwWIiO8B/xYRHwemUBWrXTkY2KPeL8C2EbF1Zj7a2RARJwInAuy0U1ddsuEGDx7MokWLnp5evHgxgwcPbmRfva3k7FB2/pKzg/nbqeTsUHb+krND+flLVnrfl5y/5OxQfv52c4Sz534EnEHL6bS1AI6qR/7GZeZOmfnbLtb/D6rCcgzwVqrCtNOTLc+fapl+imc+HPg+8CbgMGBuZv653vcdLfveMzPfsAGvkcz8G3Al8DbgncC317DoZsArW/Y9uLXYrLf19czsyMyOgQMHbkisNZowYQILFixg4cKFLF++nOnTp3P44Yc3sq/eVnJ2KDt/ydnB/O1UcnYoO3/J2aH8/CUrve9Lzl9ydig/f7s5wtlz5wIPZeZtEXFAS/sVwIcj4sOZmRGxd2beDDwCbNOy3It55mZDx/d055n5RERcAXwNOKFuvgsYGBH7Z+aN9Sm2u2XmHS2r3gUMi4gRmfl74D3AtXX7jhExITN/ExHbAI9n5kqq0dQfA9dn5oP1dh4Btm3Z7s+AD1OdQkxEjMvMeT19XRuqX79+TJs2jUMPPZRVq1YxZcoURo8e3dcx1kvJ2aHs/CVnB/O3U8nZoez8JWeH8vNPvg1mPQjLVsCQ6+G04XBCIQM9pfd9yflLzg7l5588eTKzZs1i2bJlDBkyhNNOO40TTjhh3Sv2kqjuYqt1iYhHM3Pr1doOAE7JzMPqu9eeSXUH182oTl89LCIGUBWjWwCfBf5EdTrrY8BPgHdn5rCIOB7oyMwP1du+p55e1sW8VwKXAK/IzFV12zjgy1QFbT/gzMz8RkScT/XVLZdExEFUo7P9gN8AJ2XmkxExgeoa0K2ort88uHOUMiLuBD6amTPr6d3qfT9FVWj+FvgKMKre7nWZ+YE19WNHR0fOmTOnu90uSZLWxyGx7mU2Zlf696meP1ouTXtaaTVaRMzNzI4u55X2YgT1daIvzsx/a3g/LwdmAbtn5lO9sU0LTkmS+oAFp1SMTb3g9JTawkTED4FdgNc3vJ/3Ut3p9uTeKjYlSZIkPb9YcBYmM4/oo/18C/hWX+xLkiRJ0qbJu9RKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRGWHBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRGWHBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRGWHBKkiRJkhphwSlJkiRJaoQFpyRJkiSpERackiRJkqRGWHBKkiRJkhrR7YIzIoZFxJsj4kUtbf0i4rSIuCUifhkRRzQTU5IkSZJUmn49WPaTwOHADi1tnwD+rWX6fyPitZn5q94IJ0mSJEkqV09Oqd0fuCozVwJExGbAPwB3AjsB+wKPAR/r7ZCSJEmSpPL0pODcAfhjy/Q4YHvgK5m5ODPnADOACb0XT5IkSZJUqp4UnFsA2TL96nr66pa2xcCOvZBLkiRJklS4nhSci4GxLdNvBpZl5m9b2gYBf+2NYJIkSZKksvXkpkGXAR+LiDOAJ4BDgPNWW2Y3nn3arSRJkiTpeaonBefngLcDJ9fTS6juXAtARAyiurHQl3srnCRJkiSpXN0uODPzgYjYEziobro2Mx9pWWR74OPAFb2YT5IkSZJUqJ6McJKZj1OdWtvVvPnA/N4IJUmSJEkqX48Kzq5ExO7Am4C/AdMz8+ENTiVJkiRJKl6371IbEf8eEfdFxICWtoOBm4EzgK8CN0XES3s/piRJkiSpND35WpQ3AXdm5l9a2j5L9V2cnwS+BuwMfKT34kmSJEmSStWTgnMY8PR3bkbEYGA88NXM/Exmfgi4mupOtpIkSZKk57meFJwvAVpHN19NNbrZehOhucBOvZBLkiRJklS4nhScS4HBLdMHAiuAX7e09e/hNiVJkiRJm6ie3KV2HnB4RIwBngCOAX5Rf1VKp2HAfb2WTpIkSZJUrJ6MRn4OeDFwC3BX/fwLnTMjYnOq02zn9GZASZIkSVKZuj3CmZnXR8RhwPuprt38dmZe3rLIq4AlwA97N6IkSZIkqUQ9ut4yM2dm5lGZeXRm/nC1eddn5t6ZeUnvRpS6Z+bMmYwcOZIRI0YwderUdsfpkZKzQ9n5S84O5m+nkrND2flLzg5l559yBwy6Fsbc2O4k66fkvoey85ecHcrOP2XKFAYNGsSYMWPaEyAzfWwkD6qR44tapvtR3azpsl7ezzDg9jXMmwV0rMc23w7ssa7lxo8fn01YuXJlDh8+PO++++588sknc+zYsXnHHXc0sq/eVnL2zLLzl5w90/ztVHL2zLLzl5w9sw/zH0wjj2vHk3P3JUe/qJntP/1ogMdO+5ScPbP5/HUN8KxHb7r22mtz7ty5OXr06F7dbitgTq7h7//1uqNsRAyJiP0iYmJXj/WsfQWPAWMiYqt6+hCq05RL8HZgj3btfPbs2YwYMYLhw4fTv39/Jk2axIwZM9oVp0dKzg5l5y85O5i/nUrODmXnLzk7lJ9/4ktgwBbtTrF+Su/7kvOXnB3Kzz9x4kQGDBjQtv33qOCMiDdExB3AH4FfAtes4aH191PgLfXzycDFnTMiYkBEXBoRt0bEryJibN3+qYg4NyJmRcQfIuIfW9Y5OSJurx8fbdlPv4j4dkT8NiIuiYgXrh4kIr4WEXMi4o6IOK2lfWpEzK9znBERrwIOBz4fEfMiYpde7ZFuWLJkCUOHDn16esiQISxZUkatXnJ2KDt/ydnB/O1UcnYoO3/J2aH8/CUrve9Lzl9ydig/f7t1u+CMiFcClwHbAdOAAK4DvgHcWU//GPh0r6d8fpkOTIqILYGxPPt7Tk8Dbs7MscC/AN9qmbc7cCiwL/DJiNgiIsYD7wP2A14JvD8i9q6XHwl8NTNHAX8F/qGLLP+amR11jtdFxNiIeClwBDC6zvGZzPwl8CPg45k5LjPvbt1IRJxYF65zli5dut4dI0mSJKksPRnh/Geq79+ckJkfqduuycwPAGOAzwAHA940aANk5q1U11hOphrtbPUa4MJ6uauBl0bEtvW8n2Tmk5m5DHgA2KFe/oeZ+VhmPgr8AHhtvfyizLyhfn5Rvezq3hkRNwE3A6OpTpl9mOo4+GZEHAn8rRuv6euZ2ZGZHQMHDlxnH6yPwYMHs2jRoqenFy9ezODBgxvZV28rOTuUnb/k7GD+dio5O5Sdv+TsUH7+kpXe9yXnLzk7lJ+/3XpScO4P/Cgz7119/fpa0X8Hfks1CqcN8yPgDFpOp+2GJ1uer2LdX3mTa5uOiJ2BU4CD6pHMnwBbZuZKqlHUS4DDgJk9yNiYCRMmsGDBAhYuXMjy5cuZPn06hx9+eLtjdUvJ2aHs/CVnB/O3U8nZoez8JWeH8vOXrPS+Lzl/ydmh/Pzt1u3v4QReDPypZXo58KLVlrkBeNeGhhLnAg9l5m0RcUBL+/XAscB/1O3LMvOvEbGm7VwPnB8RU6lOeT4CeE89b6eI2D8zb6T6P/vFautuS3UTo4cjYgfgTcCsiNgaeGFm/jQibgD+UC//CLDNer7eDdavXz+mTZvGoYceyqpVq5gyZQqjR49uV5weKTk7lJ2/5Oxg/nYqOTuUnb/k7FB+/sm3wawHYdkKGHI9nDYcTihkoKf0vi85f8nZofz8kydPZtasWSxbtowhQ4Zw2mmnccIJJ/TZ/qO6i203FoxYRPX1HCfV03+iuv3tkS3LfAV4b2a2rfAoWUQ8mplbr9Z2AHBKZh4WEQOoitHhVKeynpiZt0bEp4BHM/OMep3bgcMy856IOBmYUm/unMw8MyKGUY1MzgHGA/OB92Tm3yJiVr2/ORFxPvAqYBHVqbQ/Aq4AZgBbUhWxZ2TmBRHxaqrreZ8Ejl79Os5OHR0dOWfOnA3sKUmStFaHrPHD6DJc2b2/T6VNQVeDR92t0TYWETG3vvfLc+f1oOC8CliVmW+op/+XatRrfGb+LiJeBswD7s3MfXoluTY5FpySJPUBC06pGJt6wdmTazhnUt2ptPNLXM4CtgJujojfUN2pdiBw5gZklSRJkiRtInpScP4PMBFYAVDf4fQdwEKqu9TeB5yUmd9a4xYkSZIkSc8b3b5pUGb+lWd/JySZ+UPgh70dSpIkSZJUvp6McEqSJEmS1G0WnJIkSZKkRqzxlNqI+MOa5q1DZuYu67muJEmSJGkTsbZrODcD1ud+vIXfh1uSJEmS1BvWWHBm5rA+zCFJkiRJ2sR4DackSZIkqREWnJIkSZKkRnS74IyIT0TEioh4+RrmD46I5RHx//VePEmSJElSqXoywvlWYFZm3tvVzMxcAlwDvL0XckmSJEmSCteTgnMEMH8dy8yvl5MkSZIkPc/1pODcCvjbOpZ5Athm/eNIkiRJkjYVPSk4FwOvXMcyrwSWrH8cSZIkSdKmoicF50xgYkQc09XMiJgEvA64vDeCSZIkSZLK1q8Hy/4XcCzwnbronEk1mjkYeBNwOPAXYGpvh5QkSZIklafbBWdmLomIQ4HvUd2J9m0tswO4B3hHZi7uzYCSJEmSpDL1ZISTzJwTEbtRfUXKK4HtgIeAXwE/zswVvR1QkiRJPXRltjvBhtk52p1g/S0svO+lXtajghOgLip/UD8kSZIkSepST24aJEmSJElSt1lwSpIkSZIaYcEpSZIkSWqEBackSZIkqREWnJIkSZKkRlhwSpIkSZIaYcEpSZIkSWpEj7+HMyLGAu8CRgEvysyD6/ZhwL7AlZn5YG+GlCRJkiSVp0cFZ0R8GvgXnhkZzZbZmwEXAx8Fzu6NcJIkSZKkcnX7lNqImAR8ArgSGAd8tnV+Zv4BmAMc3ov5JEmSJEmF6sk1nP8I/B54W2beCizvYpnfArv2RjBJkiRJUtl6UnDuCVyRmV0Vmp3uBXbYsEiSJEmSpE1BTwrOAJ5axzI7AE+sfxxJkiRJ0qaiJwXnAuBVa5oZEZsBrwHu2NBQkiRJkqTy9aTg/F9gn4j4f2uY/y/ACOA7G5xKkiRJklS8nnwtypnAO4DPRcQ7qb8SJSLOAF4LdAC/Ar7eyxklSZIkSQXqdsGZmY9HxIHAWcCxwOb1rJOpru28CPhQZq7s9ZSSJEmSpOL0ZISTzHwYOD4iTgYmAC8FHgZmZ+bSBvJJkiRJkgrVk2s4n5aZf8nMKzLzO5n5E4tNbQxmzpzJyJEjGTFiBFOnTm13nB4pOTuUnb/k7GD+dio5O5Sdv+TsUHb+krNPWQaD/gRjlrQ7yforuf9Lzg5l558yZQqDBg1izJgx7QmQmT42wgfVNbIXtUz3A5YCl7U51/HAy9d3/fHjx2cTVq5cmcOHD8+77747n3zyyRw7dmzecccdjeyrt5WcPbPs/CVnzzR/O5WcPbPs/CVnzyw7f59mH0avP659GTl3R3L0Fr2/7Wc9GuKx0z5N56//7n/Wozdde+21OXfu3Bw9enSvbrcVMCfX8Pd/t0c4I+Lcbj6+2Yv18PPZY8CYiNiqnj4E2Bg+kzseeHm7Q6xu9uzZjBgxguHDh9O/f38mTZrEjBkz2h2rW0rODmXnLzk7mL+dSs4OZecvOTuUnb/k7AATt4QB63Vu38ah5P4vOTuUn3/ixIkMGDCgbfvvyY/d8et4HNfyXL3jp8Bb6ueTgYs7Z0TEgIi4NCJujYhfRcTYuv1TdeE/KyL+EBH/2LLOyRFxe/34aEv7e+vt3BIRF0bENhGxMCK2qOdvW0+/g+puxN+OiHkRsVVEjI+IayNibkRcERE7Nt8tz7VkyRKGDh369PSQIUNYsmRjqM/XreTsUHb+krOD+dup5OxQdv6Ss0PZ+UvOvikouf9Lzg7l52+3nhScO6/hsTdwIrAY+C4wvJczPp9NByZFxJbAWODXLfNOA27OzLFU34H6rZZ5uwOHAvsCn4yILSJiPPA+YD/glcD7I2LviBgNfAJ4fWbuBXwkMx8BZvFMsTsJ+EFmfg+YAxybmeOAlcDZwNGZOR44F/jP1V9ERJwYEXMiYs7SpV7uK0mSJD1f9ORrUf64hll/BG6JiCuAW4GfA55W2wsy89aIGEY1uvnT1Wa/BjiqXu7qiHhpRGxbz/tJZj4JPBkRDwA71Mv/MDMfA4iIH1B9f2oC38vMZfW2/lJv4xzgn4BLqQrV93cRcSQwBrgyIqD6qpz7ungdX6f+ftaOjo7sWS90z+DBg1m0aNHT04sXL2bw4MFN7KrXlZwdys5fcnYwfzuVnB3Kzl9ydig7f8nZNwUl93/J2aH8/O3Wa2eyZ+Yi4MfAR3prmwLgR8AZtJxO2w1PtjxfRQ+//gYgM28AhkXEAcDmmXl7F4sFcEdmjqsfe2bmG3q6r94wYcIEFixYwMKFC1m+fDnTp0/n8MMPb0eUHis5O5Sdv+TsYP52Kjk7lJ2/5OxQdv6Ss28KSu7/krND+fnbrceFyDrcD+zay9t8vjsXeCgzb6uLv07XA8cC/1G3L8vMv9YjjV25Hjg/IqZSFYpHAO8BlgM/jIgvZuafI2JAyyjnt4DvAP/Rsp1HgG3q53cBAyNi/8y8sb7mc7fMvGODXvF66NevH9OmTePQQw9l1apVTJkyhdGjR/d1jPVScnYoO3/J2cH87VRydig7f8nZoez8JWcHmLwUZj0By1bBkEVw2nZwwjbrXG2jUXL/l5wdys8/efJkZs2axbJlyxgyZAinnXYaJ5xwQp/tP6q72PbChiI2B2YDgzPzZb2y0eexiHg0M7dere0A4JTMPCwiBlAVo8OBvwEn1qfgfgp4NDPPqNe5HTgsM++JiJOBKfXmzsnMM+tljgM+TjUaenNmHl+3vwxYCOyYmQ/VbUcBpwOPA/tTnVb7ZeDFVB9gnJmZ31jT6+ro6Mg5c+asf8dIkqRN385r/AB947ewkauHtAnrasCot2q0vhIRczOzo8t53X0xETFxDbP6AUOprvN7LVUh8/frE1Qbl4g4GnhbZr6nt7ZpwSlJktbJglPPI5t6wdmTU2pnUd1gZo37Aa6jGilT4SLibOBNwJvbnUWSJElSmXpScH6argvOp4AHgdmZObtXUqntMvPD7c4gSZIkqWw9+VqUTzWYQ5IkSZK0ien216JExLkR8bEmw0iSJEmSNh09+R7OdwGDmgoiSZIkSdq09KTgvAcLTkmSJElSN/Wk4PwO8KaIeElTYSRJkiRJm46eFJyfBeYA10TEYRGxQ0OZJEmSJEmbgLXepTYi3gvMy8xbgSc6m4EZ9fyuVsvM7MnXrUiSJEmSNkHrKgzPBz4J3ApcT9ffwylJkiRJ0nN0ZyQyADLzgGajSJIkSZI2JT25hlOSJEmSpG6z4JQkSZIkNaI7p9RuFxE79WSjmfmn9cwjSZIkSdpEdKfg/Ej96K7s5nYlSZIkSZuw7hSGfwUeajiHJEmSJGkT052C80uZ+enGk0iSJEmSNineNEiSJEmS1AivtZQkSdLGZWG2O4GkXuIIpyRJkiSpERackiRJkqRGrPWU2sy0IJUkSZIkrRcLSkmSJElSIyw4JUmSJEmNsOCUJEmSJDXCglOSJEmS1AgLTkmSJElSIyw4JUmSJEmNsOCUJEmSJDXCglOSJEmS1AgLTkmSJElSIyw4JUmSJEmNsOCUJEmSJDXCglOSJEmS1AgLTkmSJElSIyw4JUmSJEmNsOCUJEmSJDXCglObjJkzZzJy5EhGjBjB1KlT2x2nR0rODmXnLzk7mL+dSs4OZecvOTuUnb/k7GD+dio5O5Sdv+3ZM9PHRvQAErioZbofsBS4bD23dzwwrRfzDQPetb7rjx8/PpuwcuXKHD58eN5999355JNP5tixY/OOO+5oZF+9reTsmWXnLzl7pvnbqeTsmWXnLzl7Ztn5S86eaf52Kjl7ZvP567//n/XoLX3V98CcXMPf/45wbnweA8ZExFb19CHAkjbmWd0w4F3tDrG62bNnM2LECIYPH07//v2ZNGkSM2bMaHesbik5O5Sdv+TsYP52Kjk7lJ2/5OxQdv6Ss4P526nk7FB2/o0huwXnxumnwFvq55OBiwEiYrOIWBARA1umfx8RA+vH9yPiN/Xj1atvNCKGRcTVEXFrRFwVETvV7TtExA8j4pb68aqI+HREfLRl3f+MiI8AU4HXRsS8iPhYRGweEZ+v93lrRPx9s13TtSVLljB06NCnp4cMGcKSJRtTnb5mJWeHsvOXnB3M304lZ4ey85ecHcrOX3J2MH87lZwdys6/MWS34Nw4TQcmRcSWwFjg1wCZ+RRwEXBsvdzBwC2ZuRQ4C/hSZk4AjgLO6WK7ZwMXZOZY4NvAl+v2LwPXZuZewD7AHcC5wHuhKmyBSfW+TwWuz8xxmfkl4ATg4Xq/E4D3R8TOvdYTkiRJkorVr90B9FyZeWtEDKMa3fzparPPBWYAZwJTgPPq9oOBPSKic7ltI2Lr1dbdHziyfn4h8Ln6+eupi8vMXAU8DDwcEX+OiL2BHYCbM/PPLdvv9AZgbEQcXU+/GNgVWNi5QEScCJwIsNNOO627A9bD4MGDWbRo0dPTixcvZvDgwY3sq7eVnB3Kzl9ydjB/O5WcHcrOX3J2KDt/ydnB/O1UcnYoO/9GkX1NF3f6aM8DeLT+99+BPwN7AgfQctMg4HKqIvEPwOZ12zJgyy62dzz1TYPqZbaon28BLKufLwVe0MW6x1CNnH4XeHPdtnqW7wOHdvf1NXXToBUrVuTOO++cf/jDH56+IPr2229vZF+9reTsmWXnLzl7pvnbqeTsmWXnLzl7Ztn5S86eaf52Kjl7ZvP5afCmQX3V96zlpkGOcG68zgUeyszbIuKA1eadQ3V664VZjUgC/Az4MPB5gIgYl5nzVlvvl1Snxl5IdVru9XX7VcBJwJkRsTmwdWY+DPwQ+DRVcdp5o6BHgG1atnkFcFJEXJ2ZKyJiN2BJZj62vi98ffTr149p06Zx6KGHsmrVKqZMmcLo0aP7MsJ6Kzk7lJ2/5Oxg/nYqOTuUnb/k7FB2/pKzg/nbqeTsUHb+jSF7VAWpNhYR8Whmbr1a2wHAKZl5WD29BdXo576ZeWfdtj3wFWAU1anS12XmByLieKAjMz8UEa+gOgV3e6pRzfdl5p8iYgfg68BwYBVwUmbeWG/3v6kK31Nb9n0F8FLgfKoR0M8AbwWi3u7b64L1OTo6OnLOnDkb2k2SJEnSJqGLS9YorUaLiLmZ2dHlvNJejCAiOqhuEPTahvezGXAT8I7MXNAb27TglCRJkp6xqRec3qW2MBFxKtV1k//c8H72AH4PXNVbxaYkSZKk5xev4SxMZk6l+i7Mpvczn+oUW0mSJElaL45wSpIkSZIaYcEpSZIkSWqEBackSZIkqREWnJIkSZKkRlhwSpIkSZIaYcEpSZIkSWqEBackSZIkqREWnJIkSZKkRlhwSpIkSZIaYcEpSZIkSWqEBackSZIkqREWnJIkSZKkRlhwSpIkSZIaYcEpSZIkSWqEBackSZIkqREWnJIkSZKkRvRrdwCpGKdEuxNsmDOy3Qmev75b+LFzjMeOVJ7C33cua3eADXCY75lSK0c4JUmSJEmNsOCUJEmSJDXCglOSJEmS1AgLTkmSJElSIyw4JUmSJEmNsOCUJEmSJDXCglOSJEmS1AgLTkmSJElSIyw4JUmSJEmNsOCUJEmSJDXCglOSJEmS1AgLTkmSJElSIyw4JUmSJEmNsOCUJEmSJDXCglOSJEmS1AgLTkmSJElSIyw4JUmSJEmNsOCUJEmSJDXCglOSJEmS1AgLTm0yZs6cyciRIxkxYgRTp05td5x1mjITBn0Vxpz/TNu/3QBjL4Bx34I3XAL3Ptq2eD1SWt+3Kjk7wEOPwdFfhN1PhlEnw42/a3einim5/0vODmXnLzk7lJf/oYfg6KNh991h1Ci48cZn5n3hCxABy5a1Ld5aLVoKB/4L7PEPMPof4KwfPTPv7B/D7h+o2v/pvPZl7InSjp1WJWeHjT//lClTGDRoEGPGjHm67S9/+QuHHHIIu+66K4cccggPPvggAA8++CBHHHEEY8eOZd999+X2229vNJsF50YgIjIiLmqZ7hcRSyPisvXc3vERMW0dy5wfEUd30X7A+uw3IraLiH/o6Xq9ZdWqVXzwgx/k8ssvZ/78+Vx88cXMnz+/XXG65fgxMPOoZ7d9vANuPQ7mvRcOGw6fvrHrdTcmJfZ9p5Kzd/rIBfDGcXDnF+GWz8Gowe1O1H0l93/J2aHs/CVnhzLzf+Qj8MY3wp13wi23VEUnwKJF8LOfwU47tTff2vTbHL4wBeZ/FX51BnzlJzD/T3DNrTDj13DL2XDHV+GUI9qddN1KPHY6lZwdysh//PHHM3PmzGe1TZ06lYMOOogFCxZw0EEHPV0on3766YwbN45bb72Vb33rW3zkIx9pNJsF58bhMWBMRGxVTx8CLGljnvWxHdC2gnP27NmMGDGC4cOH079/fyZNmsSMGTPaFadbJg6BAVs+u23bFzzz/LEV1afGG7sS+75TydkBHv4bXPdbOOHAarp/P9juRe3N1BMl93/J2aHs/CVnh/LyP/wwXHcdnHBCNd2/P2y3XfX8Yx+Dz31u4/5dteMA2GdE9XybF8KoobDkz/C1n8KpR8MLtqjmDdqubRG7rbRjp1XJ2aGM/BMnTmTAgAHPapsxYwbHHXccAMcddxyXXnopAPPnz+f1r389ALvvvjv33HMP999/f2PZLDg3Hj8F3lI/nwxcDBARm0XEgogY2DL9+4gYWD++HxG/qR+vXn2jETEsIq6OiFsj4qqIaP0c8uCImBMRv4uIw7pYd9+IuDEibo6IX0bEyLp9dETMjoh59XZ3BaYCu9Rtn+/drlm3JUuWMHTo0KenhwwZwpIlpdXslX/9BQz9H/j2b+HTr2p3mnUrue9Lzg6w8AEYuC2872uw96nwd/8Djz3R7lTdV3L/l5wdys5fcnYoL//ChTBwILzvfbD33vB3fwePPQYzZsDgwbDXXu1O2H333A833w37jYTf3QvX3wH7/T943anwmwIuRyjt2GlVcnYoN//999/PjjvuCMDLXvayp4vKvfbaix/84AdAVUz/8Y9/ZPHixY3lsODceEwHJkXElsBY4NcAmfkUcBFwbL3cwcAtmbkUOAv4UmZOAI4Czuliu2cDF2TmWODbwJdb5g0D9qUqdP+73nerO4HXZubewL8Dp9ftHwDOysxxQAewGDgVuDszx2Xmx1s3EhEn1oXtnKVLl/agS56f/vM1sOjv4dhRMO3mdqfRxmzlKrhpIZx0CNw8FV70Api6cX3gKqlwK1fCTTfBSSfBzTfDi14En/oUnH46fPrT7U7XfY8+Dkd9Fs58P2z7wur98y+PVqfZfn4KvPO/ILPdKaXmRARRn45w6qmn8tBDDzFu3DjOPvts9t57bzbffPPG9m3BuZHIzFupCsDJVKOdrc4F3ls/nwJ0Xtp+MDAtIuYBPwK2jYitV1t3f+A79fMLgde0zPvfzHwqMxcAfwB2X23dFwPfi4jbgS8Bo+v2G4F/iYj/D3hFZj6+jtf29czsyMyOgQMHrm3R9TZ48GAWLVr09PTixYsZPLigi9m6cOwo+P6CdqdYt5L7vuTsAENeCkMGwH67VtNH7wc33dPWSD1Scv+XnB3Kzl9ydigv/5Ah1WO//arpo4+uCtCFC6vRzWHDYPFi2Gcf+L//a2vUNVqxsio2jz0AjqzPHBqyPRy5f3U68L67wWabwbK/tjXmOpV27LQqOTuUm3+HHXbgvvvuA+C+++5j0KBBAGy77bacd955zJs3j29961ssXbqU4cOHN5bDgnPj8iPgDOrTaTtl5iLg/oh4PdWI5OX1rM2AV9ajiuMyc3Bm9uS+pqt/lrf69H8A12TmGOCtwJZ1nu8AhwOPAz+tc7XVhAkTWLBgAQsXLmT58uVMnz6dww8/vN2xemzBg888n/F72H3AmpfdWJTc9yVnB3jZdjD0pXDXvdX0VbfDHhv/77+nldz/JWeHsvOXnB3Ky/+yl8HQoXDXXdX0VVdVxeUDD8A991SPIUOqIvRlL2tn0q5lwglfrq7dPPntz7S//ZXVjYMAfrcElq+E7bdtS8RuK+3YaVVydig3/+GHH84FF1wAwAUXXMDb3vY2AB566CGWL18OwDnnnMPEiRPZdtvmfgD6NbZlrY9zgYcy87aIOGC1eedQnVp7YWauqtt+BnwY+DxARIzLzHmrrfdLYBLV6OaxwPUt894RERcAOwPDgbuAV7bMfzHP3Lzo+M7GiBgO/CEzv1xfEzoWuAXYpmcvt/f069ePadOmceihh7Jq1SqmTJnC6NGj171iG02+DGYthmWPw5D/gdNeBT9dCHf9BTYLeMW28N8HtzvlupXY951Kzt7p7PfBsdOqP5aGD4LzPtDuRN1Xcv+XnB3Kzl9ydigz/9lnw7HHwvLlMHw4nFfIV4gA3DAfLrwG9hwG4/6xajv9vTDlYJjyZRjzweqGaxd8dOO++RGUeex0Kjk7lJF/8uTJzJo1i2XLljFkyBBOO+00Tj31VN75znfyzW9+k1e84hX87//+LwC//e1vOe6444gIRo8ezTe/+c1Gs0V6wnrbRcSjmbn1am0HAKdk5mH19BbAn4F9M/POum174CvAKKoPD67LzA9ExPFAR2Z+KCJeQXUK7vbAUuB9mfmniDgfeILqGsxtgZMz87LW/UbE/sAFVHfR/Qnw7swcFhGnAu8BVgD/B7wrM/8SEd+hKj4vX/06zk4dHR05Z86cDe6ztjhlI/9NtC5n+LPeNt8t/Ng5xmNHKk/h7zvr9cVwG4nDfM9Uz0QXn3aUVqNFxNzM7OhyXmkv5vkqIjqobhD02nZn2RAWnG1kwdk+FpyS+lzh7zsWnHoe2dQLTk+pLUA9ongSz9ypVpIkSZI2et40qACZOTUzX5GZv2h3FkmSJEnqLgtOSZIkSVIjLDglSZIkSY2w4JQkSZIkNcKCU5IkSZLUCAtOSZIkSVIjLDglSZIkSY2w4JQkSZIkNcKCU5IkSZLUCAtOSZIkSVIjLDglSZIkSY2w4JQkSZIkNcKCU5IkSZLUCAtOSZIkSVIjLDglSZIkSY2w4JQkSZIkNcKCU5IkSZLUiH7tDiAV46p2B3ieuy/anWD97dbuAJKed97f7gAb6LPtDiCptzjCKUmSJElqhAWnJEmSJKkRFpySJEmSpEZYcEqSJEmSGmHBKUmSJElqhAWnJEmSJKkRFpySJEmSpEZYcEqSJEmSGmHBKUmSJElqhAWnJEmSJKkRFpySJEmSpEZYcEqSJEmSGmHBKUmSJElqhAWnJEmSJKkRFpySJEmSpEZYcEqSJEmSGmHBKUmSJElqhAWnJEmSJKkRFpzaZMycOZORI0cyYsQIpk6d2u443fbEU7Dv72CvO2H0nfDJ+9qdqOdK7XuAKR+DQXvCmAPbnaR7pnwKBh0EY97x3HlfuBBiH1j2YF+nWn8lHzslZ4ey85ecHcrNv+hROPAnsMclMPoSOOv2difqni/9N4x+DYx5LUw+EZ54Aq6+HvZ5fdV23Adh5cp2p+yeUo8dKDs7bPz5p0yZwqBBgxgzZszTbX/5y1845JBD2HXXXTnkkEN48MHqD4QHH3yQI444grFjx7Lvvvty++3N/jA/bwvOiMiI+ELL9CkR8an13NZ2EfEPvZhtWER0+T8fEZ+OiIPr57MiomMNy10SEcPr54/2VrbV9nF+RBxdP58eEbs2sZ/uWLVqFR/84Ae5/PLLmT9/PhdffDHz589vV5weeUHA1bvALbvDvJEw8xH41WPtTtV9Jfc9wPHHwMxvtztF9x3/Vpg57bnti/4PfnYj7PSyvs+0vko+dkrODmXnLzk7lJ2/32bwhf1g/tHwq8PhK/Nh/kb+AdeS++DL34A5V8Lt18OqVfCd78NxH4Lp36jaXjEULpje7qTrVvKxU3J2KCP/8ccfz8yZM5/VNnXqVA466CAWLFjAQQcd9HShfPrppzNu3DhuvfVWvvWtb/GRj3yk0WzP24ITeBI4MiK274VtbQd0WXBGRL9e2P7TMvPfM/Pna1smIkYDm2fmH3pz3+vwNeCf+nB/zzJ79mxGjBjB8OHD6d+/P5MmTWLGjBntitMjEbD15tXzFVk9or2ReqTkvgeY+EoY8JJ2p+i+ieNhwIuf2/6xL8DnPlodT6Uo+dgpOTuUnb/k7FB2/h1fCPvUfzVt0x9GbQdLCviAdOVKePyJ6t+/PQ4veiH07w+77VLNP+R18P3L2puxO0o+dkrODmXknzhxIgMGDHhW24wZMzjuuOMAOO6447j00ksBmD9/Pq9//esB2H333bnnnnu4//77G8v2fC44VwJfBz62+oyIGBgR34+I39SPV9ftn4qIU1qWuz0ihgFTgV0iYl5EfD4iDoiI6yPiR8D8iNgyIs6LiNsi4uaIOLBe//iImFGPVC6IiE+2xNg8Ir4REXdExM8iYqt6nadHFdfiWOBZPwUR8aV6W1dFxMC67f3167ulfr0vbNnHlyPilxHxh5ZRzIiIaRFxV0T8HBjUsovrgYN7u8DuriVLljB06NCnp4cMGcKSJUvaEWW9rEoYdycMuh0O2Qb2e1G7E3Vf6X2/KZgxCwYPgr12a3eSnin52Ck5O5Sdv+TsUH7+Tvc8Ajf/GfYbtO5l22nwjnDKP8BO42DHMfDibeGdb6+KzznzqmUu+TEsureNIbup5GOn5OxQbv7777+fHXfcEYCXvexlTxeVe+21Fz/4wQ+Aqpj+4x//yOLFixvL8XwuOAG+AhwbEauPF5wFfCkzJwBHAeesYzunAndn5rjM/Hjdtg/wkczcDfggkJm5JzAZuCAitqyX27fex1jgHS2nyO4KfCUzRwMP1ct016uBuS3TLwLm1Nu6FugsbH+QmRMycy/gt8AJLevsCLwGOIyqoAY4AhgJ7AG8F3hV58KZ+RTwe2CvHuRUbfOAebvD4j1g9t/g9sfbnUil+NvjcPq58OkPtDuJpOeLR1fAUT+HM18J2/Zvd5q1e/AhmDETFs6Fe2+Dx/4G374Epn8dPvYJ2PcNsM3WsPnz/S9ibfIigqhPgzr11FN56KGHGDduHGeffTZ77703m2++eWP7bsto1MYiM/8aEd8C/hFo/RP/YGCPeObctG0jYusebn52Zi6sn78GOLve550R8Uegcyziysz8M0BE/KBe9lJgYWbOq5eZCwzrwb53BJa2TD8FfLd+fhHwg/r5mIj4DNUpwVsDV7Ssc2ldRM6PiB3qtonAxZm5Crg3Iq5ebb8PAC/n2cUuEXEicCLATjvt1IOX0X2DBw9m0aJFT08vXryYwYMHN7KvJm3XDw7curqOc8xW7U7TPZtK35fq7sWwcAnsNamaXvwA7HMszP4WvKw3LhhoUMnHTsnZoez8JWeH8vOveKoqNo8dAUfu3O406/bza2HnnWBg/X545Fvgl7+Bd78Drq9Po/3ZNfC7u9uXsbtKPnZKzg7l5t9hhx2477772HHHHbnvvvsYNKg6JWHbbbflvPPOAyAz2XnnnRk+fHhjOfw8B86kGtlrPYlxM+CV9YjluMwcnJmPUp2G29pnW7Jm3b2qIdcw/WRL2yp69uHA46w9W+c+zgc+VI+8nrbaOq377+5VYVvy7MK92lnm1zOzIzM7Bg4c2M1N9cyECRNYsGABCxcuZPny5UyfPp3DDz+8kX31tqUr4aH67niPPwVXPgK7v6C9mXqi5L7fFOy5KzxwFdzzk+oxZBDc9O2Nv9iEso+dkrND2flLzg5l58+EE66rrt08ec92p+menYbAr+bC3/5W5b/qOhi1KzxQfyz/5JPwX2fDB45va8xuKfnYKTk7lJv/8MMP54ILLgDgggsu4G1vexsADz30EMuXLwfgnHPOYeLEiWy77baN5XjeF5yZ+Rfgf3n26aQ/Az7cORER4+qn91CdKktE7AN0frb3CLDNWnZzPdV1lUTEbsBOwF31vEMiYkB9jebbgRvW+8U847fAiJbpzYDO6z7fBfyifr4NcF9EbNGZbx2uA46JiM0jYkdg9S+S2A1oy03S+/Xrx7Rp0zj00EMZNWoU73znOxk9enQ7ovTYfSvgwLth7J0w4XfVNZyHdXFTmI1VyX0PMPkk2P+tcNfdMGQ8fPM77U60dpP/GfY/Hu76Iwx5I3zz0nYnWn8lHzslZ4ey85ecHcrOf8P9cOHv4ep7YdwPqsdPF617vXbabzwc/VbY5yDYcyI8lXDie+HzX4FRr4Kxr4O3vgFe/9p2J123ko+dkrNDGfknT57M/vvvz1133cWQIUP45je/yamnnsqVV17Jrrvuys9//nNOPfVUAH77298yZswYRo4cyeWXX85ZZ53VaLbIXH2A7fkhIh7NzK3r5zsAC4HPZean6jvXfgUYRTWyeF1mfqAuCmcAg4FfA/sDb8rMeyLiO1TXYV4O/AQ4JTMPq7e/JdVdXDuoRklPzsxrIuJ4qiLzxcAQ4KLMPK2+EdFlmTmmXv8UYOs62/n1vEsiYla9nzmrvbb3ACMz8xOdr5XqBklvoDrt9ZjMXBoRJ1HdWXZp/Xq2yczjW/fR2ldRnWN8NnAI8CdgBXBunWUH4MeZue/a+r2joyPnzJmztkU2XnsXdPvPrtxc+M/6fQX3//+1O8AG2rvwY0d6Pnp/we+ZAJ9td4ANsL3vmeqZ6OIW86XVaBExNzO7/rrG0l7MpqQuODsy80O9vN2tgGuAV9fXWzYuIj4G/DUzv7m25Sw428iCs30sOCX1NQvO9rHgVA9t6gXn8/6U2k1RZj5OdSfavrya+SHggj7cnyRJkqSN3PP6LrXtlpnnU924p4ltX7HupXp1f+f15f4kSZIkbfwc4ZQkSZIkNcKCU5IkSZLUCAtOSZIkSVIjLDglSdL/3969x1tV0Pn/f30UkbwPJYoHb4AiHm5yyVuhZGp9U8tLCdlF0ZqcsotZzXem+aoz9pOxeqSNXaa8dZVGmYIyUSevOToIoqEmGUFxaUpLRRBB4PP7Yy1we+DAOXAWm4Wv5+OxH2fvdX2vxT6H8z7rsiVJqoSFU5IkSZJUCQunJEmSJKkSFk5JkiRJUiUsnJIkSZKkSlg4JUmSJEmVsHBKkiRJkiph4ZQkSZIkVcLCKUmSJEmqhIVTkiRJklQJC6ckSZIkqRIWTkmSJElSJSyckiRJkqRKdGt2AKk2nm12gNe43tnsBJuud7MDSHrN+XaNf2ZK2qZ4hFOSJEmSVAkLpyRJkiSpEhZOSZIkSVIlLJySJEmSpEpYOCVJkiRJlbBwSpIkSZIqYeGUJEmSJFXCwilJkiRJqoSFU5IkSZJUCQunJEmSJKkSFk5JkiRJUiUsnJIkSZKkSlg4JUmSJEmVsHBKkiRJkiph4ZQkSZIkVcLCKUmSJEmqhIVTkiRJklQJC6ckSZIkqRIWTkmSJElSJSyc2mZMnTqVAQMG0L9/fyZMmNDsOB02fyWM+V84dBG0LoKrFjc7UefVdd9DvbOD+Zupztmh3vnrnB3qnb/O2cH8zVTn7FDf/PPnz2fMmDEceuihtLa2ctVVV235EJnZtAeQwJcbXl8EXLKJy9oD+LsuzHYA8Fg74/4ZeGv5/G5gZDvT3Qz0beY+rnLfAZcAF5XPvwS8ZWPzjBgxIquwcuXK7Nu3b86ZMyeXL1+eQ4YMyccff7xrV7I/lTwWtZAz9i6eL96XPKgb+XjvCtZVkS2y7ytS5+yZ5m+mOmfPrHf+OmfPrHf+OmfPNH8z1Tl7ZvX5y070qkdXWbRoUc6YMSMzMxcvXpwHHXRQJfsemJ7t/P7f7COcy4HTIuINXbCsPYC/W9+IiOjWBctfKzP/X2b+14amiYhWYPvM/F1Xrrsie9DOvuuEfwP+fvOjbJpp06bRv39/+vbtS/fu3Rk7diyTJ09uVpxO6d0Nhu9YPN91Oxi4Ayxc1dxMnVHnfV/n7GD+Zqpzdqh3/jpnh3rnr3N2MH8z1Tk71Dt/7969GT58OAC77rorAwcOZOHChVs0Q7ML50rgW8Cn2o6IiD0jYlJEPFQ+ji6HXxIRFzVM91hEHABMAPpFxCMR8cWIODYi7ouIKcATEdEjIq6PiFkRMTMixpTznx0RkyPi7oh4KiIuboixfUR8OyIej4jbI+J15Tw3RMQZG9m2s4DJ5fTbl/M8Vq7/U+XwfhExNSJmlFkPaRj+YDntZRGxpBx+bETcU+b9XURMiIizImJaOW2/Duy768pt/V1EfLzM+qp9V077mXLeX0XEpQ37+x8j4jcR8UtgwJrhmfl74PURsfdG9kslFi5cyL777rv2dZ8+fbb4N1NXmLcSZq6Aw3dsdpKOq/O+r3N2MH8z1Tk71Dt/nbNDvfPXOTuYv5nqnB3qn3+NefPmMXPmTA4//PAtut4uPfK3ib4G/Coirmgz/CrgK5n5y4jYD7gNGLiB5fw9MCgzh0FRzoDh5bC5EfFpisPTg8tid3tEHFzO+0ZgEPAi8FBE3AI8AxwEjMvMD0XEfwCnA9/v4HYdDdxYPh8GtGTmoDLbHuXwbwEfycynIuJw4OvAW8ptvyozb4yIj7RZ7tByP/wV+B1wTWa+MSI+AVwAfJIN77tDgDHArsDsiPjGevbdCeW2vxEIYEpEjAaWAmPL7ekGPAzMaMj2cLndkxoDR8SHgQ8D7Lfffh3aea9FS1bD6U/DlT1ht2b/KUiSJEnbjCVLlnD66adz5ZVXsttuu23RdTe9cGbm4oj4LvBxYFnDqLcCh0bEmte7RcQunVz8tMycWz5/E8Vpn2TmkxHxe2BN4bwjM/8CEBH/WU77E2BuZj5STjOD4rrOjuoNPF0+/x3QNyL+DbiFouzuAhwF3NSwjWuOax0JvKt8/kOK6yPXeCgz/1hmnQPcXg6fRVEkYcP77pbMXA4sj4g/A3utJ/sJ5WNm+XoXigK6K/DjzHyxXP+UNvP9Gdin7cIy81sU5ZqRI0fmeta32VpaWpg/f/7a1wsWLKClpaWKVVXi5SzK5lk7w2k7NTtN59R539c5O5i/meqcHeqdv87Zod7565wdzN9Mdc4O9c//8ssvc/rpp3PWWWdx2mmnbfH1by3HUa4EzgV2bhi2HXBEZg4rHy2ZuYTiNNzG3D02sNylHVx/2xK05vXyhmGr6FxBX0aZLTOfpTgyeTfwEeAaim14rmH7hmXmho7grtGYaXXD69UN+drbdx3dpgAub5i/f2Ze24FsPXj1Hw22mFGjRvHUU08xd+5cVqxYwcSJEznllFOaEaXTMuHcvxTXbl64Zf/g1CXqvO/rnB3M30x1zg71zl/n7FDv/HXODuZvpjpnh3rnz0zOPfdcBg4cyIUXXtiUDFtF4czMvwL/QVE617id4hRRACJiWPl0HsWpskTEcODAcvgLFEfg2nMfxXWVlKfS7gfMLscdHxE9y2s03wXcv8kb84pfA/3L9b0B2C4zJwGfB4Zn5mJgbkS8u5wmImJoOe+DFKfvQnEKa2e1t+/a03bf3QaMX3NUNCJaIqIXcC/wroh4XUTsCpzcZjkHA49tQt7N1q1bN66++mpOPPFEBg4cyHve8x5aW1ubEaXT7l8O31sKd74EwxYVj583pbZvmjrv+zpnB/M3U52zQ73z1zk71Dt/nbOD+Zupztmh3vnvv/9+vve973HnnXcybNgwhg0bxs9//vMtmiGKu9g2R0Qsycw1pWYvYC5wRWZeUpa0r1Fce9gNuDczP1KWwslAC/A/FKefvj0z50XED4EhwK0Up65elJknlcvvAXwDGElxlPTCzLwrIs6mKJm7A32A72fmpeWNiH7WcN3lRcAuZbYbynE3R8Td5Xqmt9m29wMDMvPzZZG8nlcK/v/NzFsj4sAyU29gB2BiZv5zRBxEca3o64CpwFmZ2VJel9q4TWvX3ThuA/vuEmBJZn6pnP8x4KS2+y4zP1NeE3pemXcJ8L7MnBMR/wh8kOL02T8AD2fmlyJiB+BXwODMXNnev/nIkSNz+vTp7Y3euh0QG59mazaved/rkiRJWr+Gy+DWamZH2xQRMSMzR653XN02pquVhXNkZn6si5f7OuAu4OjM7NSHXETETsCyzMyIGEtx46J3dmW+rhYRp1Icuf2nDU1n4WwiC6ckSdJWZ1svnE2/adC2KjOXRfERKy0URwI7YwRwdRTvvueA8V0crwrdgC83O4QkSZKkrcdrvnBm5g3ADRUt+7ZNnO8+ipsM1UZm3tTsDJIkSZK2LlvFTYMkSZIkSdseC6ckSZIkqRIWTkmSJElSJSyckiRJkqRKWDglSZIkSZWwcEqSJEmSKmHhlCRJkiRVwsIpSZIkSaqEhVOSJEmSVAkLpyRJkiSpEhZOSZIkSVIlLJySJEmSpEpYOCVJkiRJlbBwSpIkSZIqYeGUJEmSJFXCwilJkiRJqkS3ZgeQauOIZgd4jftjNDvBpruq2QE204RsdgJJnXV8jX9mAhzd7ACb4RJ/ZkqNPMIpSZIkSaqEhVOSJEmSVAkLpyRJkiSpEhZOSZIkSVIlLJySJEmSpEpYOCVJkiRJlbBwSpIkSZIqYeGUJEmSJFXCwilJkiRJqoSFU5IkSZJUCQunJEmSJKkSFk5JkiRJUiUsnJIkSZKkSlg4JUmSJEmVsHBKkiRJkiph4ZQkSZIkVcLCKUmSJEmqhIVTkiRJklQJC6e2GVOnTmXAgAH079+fCRMmNDtOh81eDMNue+Wx2yS4cnazU3VOXfc9wPhPQa/BMGhMs5N03vzFMOZGOPQaaL0Grpre7ESdV+f3Tp2zQ73z1zk71Dv/+Meh1z0w6IFmJ9k0X3kAWr8Og74O4ybBSyubnahz6vzeqXN2qFf+8ePH06tXLwYNGrR22KOPPsqRRx7J4MGDOfnkk1m8ePEWy9P0whkRqyLikYh4LCJuioidtoJMN0TEGV24vGMj4qhNmO+wiLi2fH5JRFy0nmn2iYibG9bzs40ss73lHBARj3Uy39r9FBETI+KgzszflVatWsVHP/pRbr31Vp544gluvPFGnnjiiWbF6ZQBu8EjJxaPGcfDTt3g1D7NTtVxdd73AGefCVN/0OwUm6bbdvDlMfDEefDg++FrD8MTzzQ7VcfV+b1T5+xQ7/x1zg71z3/2PjD1sGan2DQLF8NXp8H0D8FjfwerVsPETv3m01x1fu/UOTvUL//ZZ5/N1KlTXzXsvPPOY8KECcyaNYtTTz2VL37xi1ssT9MLJ7AsM4dl5iBgBfCRxpER0a05sbrUsUCnCyfwD8BXNzRBZi7KzC4rx5vhG8Bnm7XyadOm0b9/f/r27Uv37t0ZO3YskydPblacTfaLP0O/nWH/nZudpOPqvu9HHwE9/6bZKTZN711g+N7F8113hIGvh4UvNDdTZ9T5vVPn7FDv/HXODvXPP/pvoOcOzU6x6VauhmUri68vvgz77NrsRB1X5/dOnbND/fKPHj2anj17vmrYb37zG0aPHg3A8ccfz6RJk7ZYnq2hcDa6D+hfHqm7LyKmAE9ERI+IuD4iZkXEzIgYAxARZ0fETyLijoiYFxEfi4gLy2kejIieEdEvIh5es4KIOGjN64gYERH3RMSMiLgtInq3DRQRx5XLmxUR10XEjuXwURHx3xHxaERMi4hdI+LeiBjWMO8vI2IoRYn+VHkk980RsWdETIqIh8rH0etZ767AkMx8tGHw0Ih4ICKeiogPldOt98hkue0/iYhflftiyIaW02be7SPii2W2X0XE35bDIyKujojZEfFfQK82/3ZvbdYfCBYuXMi+++679nWfPn1YuHBhM6Jslol/gHH7NztF52wr+77u5j0PM/8Eh+/T7CQdV+f3Tp2zQ73z1zk71D9/nbXsBhcdCft9BXp/GXbvASf0a3aqjqvze6fO2aH++QFaW1vXluSbbrqJ+fPnb7F1bzWFsywqbwdmlYOGA5/IzIOBjwKZmYOBccB3IqJHOd0g4DRgFPAF4MXMPAx4APhAZs4Bnm8ogucA10fEDsC/AWdk5gjgunL+xkw9gBuAM8t1dwPOj4juwI/KfEOBtwLLgGuBs8t5DwZ6lIXxm8BXyiO59wFXla9HAacD16xnl4wE2hbJIcBbgCOB/xcRG/rV8lJgZmYOoThS+t1OLOdc4Pky3yjgQxFxIHAqMAA4FPgADUdtM3M18Ftg6AYyaQNWrIIpC+Hd+258WqnRkhVw+o/hyuNgtx2bnUaStk7PLoPJs2HuJ2DRhbB0BXz/V81OJW0Z1113HV//+tcZMWIEL7zwAt27d99i694aCufrIuIRYDrwB4rSBjAtM+eWz98EfB8gM58Efg8cXI67KzNfyMyngeeBn5bDZwEHlM+vAc6JiO2BM4EfUhSnQcAd5fo/D7S9cm4AMDczf1O+/g4wuhz+x8x8qMy0ODNXAjcBJ5VldjxFWV2ftwJXl+udAuwWEbu0maY38HSbYZMzc1lmPgPcBbyxneVDsc++V+a7E3h9ROzWweWcAHygzPc/wOuBg8ptvzEzV2XmIuDONvP9GVinBEfEhyNiekRMf/rptpvUNVpaWl71l5oFCxbQ0tJSybqqcuv/wvC/gb16bHzarcm2sO/r7OVVRdk861A4bUCz03ROnd87dc4O9c5f5+xQ//x19l+/gwP3gD13hh22h9MGwn9vuYM8m63O7506Z4f65wc45JBDuP3225kxYwbjxo2jX78td3h/ayica67hHJaZF2TminL40g7Ov7zh+eqG16spjkgCTKI4enoSMCMz/wIE8HjDugdn5gmbsyGZ+SJwB/BO4D1Ae7ci2Q44omHdLZm5pM00y4C21SM38rrDUTfyOoALGvIdmJm3d2C5PShyv3rhmd/KzJGZOXLPPffcxMgbNmrUKJ566inmzp3LihUrmDhxIqecckol66rKjb+Hcfs1O0XnbQv7vq4y4dxbi2s3L9zQn5+2UnV+79Q5O9Q7f52zQ/3z19l+u8ODC4trNzPhF3Nh4Buanarj6vzeqXN2qH9+gD//+c8ArF69mssuu4yPfOQjG5mj62wNhbMj7gPOgrWnqu4HdPiDIzLzJeA2ihvbXF8Ong3sGRFHlsvdISJa28w6GzggIvqXr98P3FMO7x0Ro8p5d224dvEaihv9PJSZz5bDXgAaL0u/HbhgzYvG6z4b/Bro32bYO8vrWV9PcSOihzaw2Y377Fjgmcxcc//jjS3nNopTh3co5z84InYG7gXOLK/x7A20/SCJg1n3NOAtolu3blx99dWceOKJDBw4kPe85z20trb959x6LV0Jd/wJTqvR3WnXqPu+H3c+HHkyzJ4DfUbAtT9sdqKOu38hfO9xuPMPMOz64vHzOc1O1XF1fu/UOTvUO3+ds0P984+bBUc+BLNfhD73wbU1uozt8D5wxkAY/u8w+BuwOuHDI5qdquPq/N6pc3aoX/5x48Zx5JFHMnv2bPr06cO1117LjTfeyMEHH8whhxzCPvvswznnnLPF8kTmph4k66IAEUsyc5c2w44FLsrMk8rXPSjK4khgJXBhZt4VEWcDIzPzY+V088rXz6xn3BHAzcD+mbmqHDaMohzuTnE09MrM/HZE3AD8LDNvjojjgC+V4x8Czs/M5WXZ/DfgdRRH9d665ihlRDwJfDIzp5avDy7XvZqiaP4a+BowsFzuvZm5zp8ZImIWcFRmvhARlwB9KU5tfQNwRZn1gDLroMb9FhE9Ka5L7Qu8CHw4M3/VweVsB1wGnExxtPNp4F3A4nKbj6c4/fll4LpyP+0F/DQzN3icZeTIkTl9eg0/LBBgbDQ7weaZ2Nzv9c32xxrv/6uaHWAzTaj5e0d6LTq+xj8zAda5nWKNXOLPTHVOxLrfr83uaJ0VETMyc+R6x9VtYzZVFJ89uXtm/lPF69kHuBs4pLyRzuYs61PAC5m5vpsKbVXKrIsz89oNTWfhbCILZ/NYOCVtaRbO5rFwqpO29cJZl1NqN0tE/JjirqqV/toXER+guMnOP25u2Sx9g1dfo7o1e47ipkqSJEmSBLxyU51tWmaeuoXW811e/fEjm7u8lyjvNLu1y8zrNz6VJEmSpNeS18QRTkmSJEnSlmfhlCRJkiRVwsIpSZIkSaqEhVOSJEmSVAkLpyRJkiSpEhZOSZIkSVIlLJySJEmSpEpYOCVJkiRJlbBwSpIkSZIqYeGUJEmSJFXCwilJkiRJqoSFU5IkSZJUCQunJEmSJKkSFk5JkiRJUiUsnJIkSZKkSlg4JUmSJEmVsHBKkiRJkirRrdkBpNqYmM1O8NrWu8b7f0KzA0h6zbmjxj8zAQ6MZifYdJc0O4C0dfEIpyRJkiSpEhZOSZIkSVIlLJySJEmSpEpYOCVJkiRJlbBwSpIkSZIqYeGUJEmSJFXCwilJkiRJqoSFU5IkSZJUCQunJEmSJKkSFk5JkiRJUiUsnJIkSZKkSlg4JUmSJEmVsHBKkiRJkiph4ZQkSZIkVcLCKUmSJEmqhIVTkiRJklQJC6ckSZIkqRIWTkmSJElSJSyc2mZMnTqVAQMG0L9/fyZMmNDsOJ1S5+xQ7/x1zg7mb6Y6Z4d6569zdqh3/jpnH/8M9PoDDFrY7CSbrs77v87Zod75x48fT69evRg0aFBzAmSmjy56AKuAR4DHgJuAnbaCTDcAZ3Th8o4FjtrU+UeMGJFVWLlyZfbt2zfnzJmTy5cvzyFDhuTjjz9eybq6Wp2zZ9Y7f52zZ5q/meqcPbPe+eucPbPe+bdo9gPo8sc9e5MzepOtO3T9sl/1qIjvneapOj+wzqMr3XPPPTljxoxsbW3t0uU2AqZnO7//e4Szay3LzGGZOQhYAXykcWREdGtOrC51LHBUs0O0NW3aNPr370/fvn3p3r07Y8eOZfLkyc2O1SF1zg71zl/n7GD+Zqpzdqh3/jpnh3rnr3N2gNE9oGeNf/Ot8/6vc3aof/7Ro0fTs2fPpq2/xt92W737gP4RcWxE3BcRU4AnIqJHRFwfEbMiYmZEjAGIiLMj4icRcUdEzIuIj0XEheU0D0ZEz4joFxEPr1lBRBy05nVEjIiIeyJiRkTcFhG92waKiOPK5c2KiOsiYsdy+KiI+O+IeDQipkXErhFxb0QMa5j3lxExlKJEfyoiHomIN0fEnhExKSIeKh9HV7pX27Fw4UL23Xffta/79OnDwoX1OGemztmh3vnrnB3M30x1zg71zl/n7FDv/HXOvi2o8/6vc3aof/5ms3BWoDyS+XZgVjloOPCJzDwY+CjFYfLBwDjgOxHRo5xuEHAaMAr4AvBiZh4GPAB8IDPnAM83FMFzgOsjYgfg3yhOnR0BXFfO35ipB8XptWeW6+4GnB8R3YEflfmGAm8FlgHXAmeX8x4M9MjMR4FvAl8pj+TeB1xVvh4FnA5cs5798eGImB4R059++unO71BJkiRJtWTh7Fqvi4hHgOnAHyhKG8C0zJxbPn8T8H2AzHwS+D1wcDnursx8ITOfBp4HfloOnwUcUD6/BjgnIrYHzgR+CAygKKt3lOv/PNCnTbYBwNzM/E35+jvA6HL4HzPzoTLT4sxcSXEN6kllmR1PUVbX563A1eV6pwC7RcQujRNk5rcyc2Rmjtxzzz3bWczmaWlpYf78+WtfL1iwgJaWlkrW1dXqnB3qnb/O2cH8zVTn7FDv/HXODvXOX+fs24I67/86Z4f65282C2fXWnMN57DMvCAzV5TDl3Zw/uUNz1c3vF5NcUQSYBLF0dOTgBmZ+RcggMcb1j04M0/YnA3JzBeBO4B3Au8BftDOpNsBRzSsuyUzl2zOujfFqFGjeOqpp5g7dy4rVqxg4sSJnHLKKVs6xiapc3aod/46ZwfzN1Ods0O989c5O9Q7f52zbwvqvP/rnB3qn7/ZtoWb2NTNfcBZwJ3lqar7AbMpTrvdqMx8KSJuA74BnFsOng3sGRFHZuYD5VHJgzPz8YZZZwMHRET/zPwt8H7gnnJ474gYlZkPRcSuFMV5JcXR1J8C92Xms+VyXgB2a1ju7cAFwBcBImJYZj7SmR3SFbp168bVV1/NiSeeyKpVqxg/fjytra1bOsYmqXN2qHf+OmcH8zdTnbNDvfPXOTvUO3+dswOMexrufgmeWQV95sOle8C5uzY7VcfVef/XOTvUP/+4ceO4++67eeaZZ+jTpw+XXnop55577sZn7CJR3MVWXSEilmTmLm2GHQtclJknla97UJTFkcBK4MLMvCsizgZGZubHyunmla+fWc+4I4Cbgf0zc1U5bBjwVWB3ij8kXJmZ346IG4CfZebNEXEc8KVy/EPA+Zm5PCJGUVwD+jqK6zffuuYoZUQ8CXwyM6eWrw8u172aomj+GvgaMLBc7r2Z+aq78zYaOXJkTp8+vZN7VpIkvaYcGM1OsOnm+ru1Oidi3fd73TpaRMzIzJHrHVe3jRFExEXA7pn5TxWvZx/gbuCQzFzdFcu0cEqSpI2ycOo1ZFsvnJ5SWzMR8WOgH/CWitfzAYo73V7YVWVTkiRJ0muLhbNmMvPULbSe7wLf3RLrkiRJkrRt8i61kiRJkqRKWDglSZIkSZWwcEqSJEmSKmHhlCRJkiRVwsIpSZIkSaqEhVOSJEmSVAkLpyRJkiSpEhZOSZIkSVIlLJySJEmSpEpYOCVJkiRJlbBwSpIkSZIqYeGUJEmSJFXCwilJkiRJqoSFU5IkSZJUicjMZmfQa0hE+IaTJEmSti0zMnPk+kZ4hFOSJEmSVAkLpyRJkiSpEhZOSZIkSVIlLJySJEmSpEp0a3YAvbaMGDGC6dOnNzuGJEmStFWIiHWG1e3GruvbhjU8wilJkiRJqoSFU5IkSZJUCQunJEmSJKkSFk5JkiRJUiUsnJIkSZKkSlg4JUmSJEmVsHBKkiRJkiph4ZQkSZIkVcLCKUmSJEmqhIVTkiRJklQJC6ckSZIkqRIWTkmSJElSJSyckiRJkqRKWDglSZIkSZWwcEqSJEmSKmHh1DZj6tSpDBgwgP79+zNhwoRmx1nHSy+9xBvf+EaGDh1Ka2srF198MQDnnnsuQ4cOZciQIZxxxhksWbIEgG9+85sMHjyYYcOG8aY3vYknnniimfE3aGvf9+2ZP38+Y8aM4dBDD6W1tZWrrrqq2ZE6bNWqVRx22GGcdNJJALz5zW9m2LBhDBs2jH322Yd3vetdzQ3YQXV970C9s0O989c5O9Qr/+zZs9f+bBk2bBi77bYbV155JZ/5zGc45JBDGDJkCKeeeirPPfdcs6OuNX78eHr16sWgQYPWDrvkkktoaWlZux0///nPAXj55Zf54Ac/yODBgxk4cCCXX355s2J3SJ3eO23VOTvUI/9zzz3HGWecwSGHHMLAgQN54IEHuOmmm2htbWW77bZj+vTpa6e94447GDFiBIMHD2bEiBHceeed1QXLTB+vwQfweuCR8vG/wMKG193bTPtJYKcOLPNuYOSGphkxYkRWYeXKldm3b9+cM2dOLl++PIcMGZKPP/54JevaVKtXr84XXnghMzNXrFiRb3zjG/OBBx7I559/fu00n/rUp/Lyyy/PzHzV8MmTJ+eJJ564ZQN3UB32fXsWLVqUM2bMyMzMxYsX50EHHVSb7F/+8pdz3Lhx+Y53vGOdcaeddlp+5zvfaUKqzqnze6fO2TPrnb/O2TPrnX/lypW511575bx58/K2227Ll19+OTMzP/vZz+ZnP/vZJqd7xT333JMzZszI1tbWtcMuvvji/OIXv7jOtD/4wQ/yzDPPzMzMpUuX5v77759z587dUlE7pe7vnbpmz6w+P7DOY1N84AMfyG9/+9uZmbl8+fJ89tln84knnsgnn3wyjznmmHzooYfWTvvwww/nwoULMzNz1qxZuc8++2zuNkzPdn7/9wjna1Rm/iUzh2XmMOCbwFfWvM7MFW0m/ySw05bO2BnTpk2jf//+9O3bl+7duzN27FgmT57c7FivEhHssssuQPEX1ZdffpmIYLfddgOKP/4sW7aMiABYOxxg6dKla4dvbeqw79vTu3dvhg8fDsCuu+7KwIEDWbhwYZNTbdyCBQu45ZZbOO+889YZt3jxYu68885aHOGs83unztmh3vnrnB3qnf8Xv/gF/fr1Y//99+eEE06gW7duABxxxBEsWLCgyeleMXr0aHr27NmhaSOCpUuXsnLlSpYtW0b37t1f9f/v1qTO7506Z4d65H/++ee59957OffccwHo3r07e+yxBwMHDmTAgAHrTH/YYYexzz77ANDa2sqyZctYvnx5JdksnForIo6LiJkRMSsirouIHSPi48A+wF0RcVc53TciYnpEPB4RlzY3dWHhwoXsu+++a1/36dNnqywOq1atYtiwYfTq1Yvjjz+eww8/HIBzzjmHvffemyeffJILLrhg7fRf+9rX6NevH5/97Gf56le/2qzYG1SXfb8x8+bNY+bMmWv/TbZmn/zkJ7niiivYbrt1f4T/5Cc/4bjjjttqf2FqVOf3Tp2zQ73z1zk71Dv/xIkTGTdu3DrDr7vuOt7+9rc3IVHnXH311QwZMoTx48fz7LPPAnDGGWew884707t3b/bbbz8uuuiiDpfVLa3O7506Z4d65J87dy577rkn55xzDocddhjnnXceS5cu7dC8kyZNYvjw4ey4446VZLNwao0ewA3AmZk5GOgGnJ+ZXwUWAWMyc0w57T9m5khgCHBMRAxpRuA62n777XnkkUdYsGAB06ZN47HHHgPg+uuvZ9GiRQwcOJAf/ehHa6f/6Ec/ypw5c/jXf/1XLrvssmbF3uYtWbKE008/nSuvvHKrL2o/+9nP6NWrFyNGjFjv+BtvvHG9vxBK0uZYsWIFU6ZM4d3vfverhn/hC1+gW7dunHXWWU1K1jHnn38+c+bM4ZFHHqF37958+tOfBoojV9tvvz2LFi1i7ty5fPnLX+Z3v/tdk9NKnbdy5Uoefvhhzj//fGbOnMnOO+/coWtNH3/8cT73uc/x7//+75Vls3Bqje2BuZn5m/L1d4DR7Uz7noh4GJgJtAKHbmjBEfHh8ojo9KeffrrLAjdqaWlh/vz5a18vWLCAlpaWStbVFfbYYw/GjBnD1KlT1w7bfvvtGTt2LJMmTVpn+rFjx/KTn/xkCybsuLrt+7ZefvllTj/9dM466yxOO+20ZsfZqPvvv58pU6ZwwAEHMHbsWO68807e9773AfDMM88wbdo03vGOdzQ5ZcfU+b1T5+xQ7/x1zg71zX/rrbcyfPhw9tprr7XDbrjhBn72s5/xgx/8YKu97GONvfbai+23357tttuOD33oQ0ybNg2AH/7wh7ztbW9jhx12oFevXhx99NGvurHK1qSu7x2od3aoR/4+ffrQp0+ftWdqnXHGGTz88MMbnGfBggWceuqpfPe736Vfv36VZbNwqlMi4kDgIuC4zBwC3EJxdLRdmfmtzByZmSP33HPPSnKNGjWKp556irlz57JixQomTpzIKaecUsm6NtXTTz+99i5+y5Yt44477mDAgAH89re/BYprOKdMmcIhhxwCwFNPPbV23ltuuYWDDjpoi2fuiDrs+/ZkJueeey4DBw7kwgsvbHacDrn88stZsGAB8+bNY+LEibzlLW/h+9//PgA333wzJ510Ej16bPBbcqtR5/dOnbNDvfPXOTvUN3/bsyemTp3KFVdcwZQpU9hpp636Ng8A/PGPf1z7/Mc//vHaO9jut99+a+/OuXTpUh588MG1/w9vber63oF6Z4d65N97773Zd999mT17NlBcc33ooe0fE3ruued4xzvewYQJEzj66KOrDdfe3YR8vHYewCXA54E/AP3LYTcAnyifzwIOLJ8PBR6l+GPFXsCfgLPLcXfTpLvUZmbecsstedBBB2Xfvn3zsssuq2w9m+rRRx/NYcOG5eDBg7O1tTUvvfTSXLVqVR511FE5aNCgbG1tzfe+971r70778Y9/PA899NAcOnRoHnvssfnYY481eQvat7Xv+/bcd999CeTgwYNz6NChOXTo0LzllluaHavD7rrrrlfdpfaYY47JW2+9tYmJOq+u753MemfPrHf+OmfPrF/+JUuWZM+ePfO5555bO6xfv37Zp0+ftT87//Zv/7aJCV9t7Nixuffee2e3bt2ypaUlr7nmmnzf+96XgwYNysGDB+fJJ5+cixYtyszMF154Ic8444w89NBDc+DAgXnFFVc0Of2G1e2906jO2TOrzU8X3aV25syZOWLEiBw8eHC+853vzL/+9a/5n//5n9nS0pLdu3fPXr165QknnJCZmf/yL/+SO+2009rv4aFDh+af/vSnzdmGdu9SG8V4vZZFxCXAEopTZL9Ecf3mQxTXcC6PiAuAjwGLMnNMRNwAHAXMB54HpmTmDRFxN3BRZrZ7LsrIkSNzaz1VRZIkSdrS1ndKet06WkTMyOIeL+uOq9vGqN4snJIkSdIrtvXC6TWckiRJkqRKWDglSZIkSZWwcEqSJEmSKmHhlCRJkiRVwsIpSZIkSaqEhVOSJEmSVAkLpyRJkiSpEhZOSZIkSVIlLJySJEmSpEpYOCVJkiRJlbBwSpIkSZIqYeGUJEmSJFXCwilJkiRJqoSFU5IkSZJUCQunJEmSJKkSFk5JkiRJUiUsnJIkSZKkSlg4JUmSJEmVsHBKkiRJkioRmdnsDHoNiYingd9XuIo3AM9UuPyq1Tl/nbNDvfPXOTuYv5nqnB3qnb/O2aHe+eucHczfTHXODtXm3z8z91zfCAuntikRMT0zRzY7x6aqc/46Z4d6569zdjB/M9U5O9Q7f52zQ73z1zk7mL+Z6pwdmpffU2olSZIkSZWwcEqSJEmSKmHh1LbmW80OsJnqnL/O2aHe+eucHczf5SIiI+LuNsMuKYcf2zB4q8u+Rjt529qk/BFxQ7nsAzZl/i6y1e77Dqpz/jpnB/M3U52zQ5Pyew2nJEmbICLa/ge6GngW+BVwTWb+cMunKpTZ7snMYxuGXQJcDIzJzLu7arlV6Yq8G1j2DcAHgQMzc15XLluS9Grdmh1AkqSau7T8ugNwCPBOYExEjMzMC5sXax1XAxOBPzQ7iCTptcPCKUnSZsjMSxpfR8RxwB3AJyPiq1vLEbTMfIZ6385fklRDXsMpSVIXysxfAE8CAYyCV1+PGBHvjYj/iYglETFvzXwRsVNE/N+IeCQilpbjH4iIcetbT0R0j4h/iog5EbE8IuZGxGURsWM707d7TWREHBIR10XEvHJZf46I+yLi/HL82Q2nEB9TLmfN45I2yzo8Im6OiP+NiBURMT8i/j0i9mkn14iImBoRL0TE4oj4r4g4csN7ef3Kffi5iJheLm9JRPw6Ir4aEXt1YP6zI2JSRPwuIpaVee6PiPe1M33fiPhWRPy2nP6vETErIr4ZEa9vmK57RHw8Ih6OiGcj4sVyX0+OiLduyrZKUl14hFOSpK4X5de213l+Gjge+ClwF7A7QETsAdwJHAY8DFxH8UfhE4EfRkRrZn5+7cIjAvgPitN351CcLtsdGA8M7lTQiHcANwE7AlOBG4E9gKHAZ4FvAI9QnDp8MfB74IaGRdzdsKzxFDelWA5MAeYDBwHnASdHxBGZ+YeG6Y8C/qvM/p/Ab4Fh5TLv7OR2/A3FPh0KzKbYhyuAfsA55fL/tJHFfAN4HLgX+CPweuD/AN+LiAGZ+U8N6+sNPATsBvwcmAT0AA4E3k/xb/KXcvIbgHHAY8B3gWXAPsCbgLeV+0CStkkWTkmSulB5xGoARdl8qM3otwBHZubMNsOvpCibn8vMKxqW1QP4CfAPEXFzZj5SjhpHUTYfpLipzkvl9BevZ50byvoG4IcUvw+8JTPvaTO+D0C53kfK5c9rexpxOe3BwDeBecAxmbmwYdxxwO3AVcCp5bCgKIWvA96VmZMbpv9EuU8642sUZfObwEczc3XD8nYBtu/AMgZl5pw229UduBX4+4j4ZsN2nQH0BD6ZmVe1mWdniptIERG7A2OBGcDhmbmqzbSvR5K2YZ5SK0nSZihPVb0kIr4QETdTHCUM4MrM/H2byb/VtmyWheN9wPTGsglQFsnPlct7b8Ooc8qv/7CmbJbT/xX4l07E/yDFEbpvtC2b5fIWdGJZ51PcOOkTjWWzXM4vKI54nhwRu5aDj6Io5vc2ls3S1RRHbjskInoBZ1IclbyosWyW61+Smc9vbDlty2Y5bAVFme0GHLee2ZatZ56lmblmeFL8+y2nLKFtpv1L22GStC3xCKckSZvn4vJrAs8B9wHXZub31zPttPUMG0Vx9G2d6yFLO5RfBzYMG05RXn65nunv3mjiVxxRfr21E/O0Z811l8dExKj1jO9FsZ0HUxztG14OX1/RXRURv6Q4HbYjRlH8Ef3ezFzaqdQNImI/ioJ/HLAfxdHXRi0Nz6cA/x/wtYg4EbgNuB94Ihs+cy4zF0fET4GTKY4ST6J4j/xPZr64qVklqS4snJIkbYbMjI1Ptdb/rmfYmlMqR5WP9uzS8Hx34K+Z+XIH19GePcqvCzc0UQet2Y7PbGS6Nduxe/m1vesqt+h2RERfij8I/A1FIbwdeB5YBRxAcTR47Q2ZMvP3EfFG4BKK6zBPK0fNj4gvZeZXGxZ/JkWRfS+vfIzOS+UR8Ysyc2PXlkpSbVk4JUnactreRAiKUgPwlU58bufzQM+I2GE9pXPvTuR5rvzaAszqxHztZQLYPTMXd2L69u4eu6nbsakupCjN52TmDY0jyjsFf7DtDJn5a+DMiOhGcf3oW4ELgKsiYmlmXltOt4yimF4SEfsCo4GzKU6lPgB482bklqStmtdwSpLUXNMoTo/tTOl4mOL/8DetZ9yxnVjOg+XXt3dw+tW0f/OdNcvq6HY8XH49pu2IiNie9W9be9bsw9HlDXs2Rf/y66T1jFsnY6PMXJmZMzLzXylu6ATwrnamnZ+ZP6C4A/FvgTd54yBJ2zILpyRJTZSZfwZ+AIwsP1dznUIXEf0i4sCGQdeXX79Q3sl2zXQ9gc/Tcd8BFgPnR8To9ay3T5tBfwH2bWdZVwMvA18p71jbdlndI6KxjP43xceXjI6Id7aZ/GN0/PpNMvNpYCLQG/hSRLzq95uI2KW8W+yGzCu/Httm3hMpPtaFNsNHtLPMNUdsXyyn2zMi1vdRNTtTnF68kuLjWyRpm+QptZIkNd/HKD6v8p+B95c3zPkTxWc1DqS4tnMcMLec/kaK6wJPAR6LiMkUNxc6g+JjUTpU1jLzmYh4L3AzcFdE3Ar8iuLOtUMoymVj0f0FMLa8Cc7DFAXz3sy8NzOfLD+H8zrg8YiYCvymzLUfxZHPp4FDynVnRJwL3AFMiojGz+E8juJuv2/r0N4rfAwYBHwEODYibqMocgdSHE08hQ3fUOnrFHf/vam8tnJRuby3UXzm6Zltpn8/8Lflv9Uc4FmK/X4yxR1pryynawFmRsQsin07n2L/nkRx2vBXM/OFTmynJNWKhVOSpCYr72R6DPBhihvLnA70oCidTwGfoihma6bPiHg38PcU1wJ+jOIjQa6nKK0v0UGZeUtEjOSVu7OeQFGengQubzP5JyiuQz0O+D8UZ0pdCtxbLuv7EfEo8GlgTLmspRTl7WbgR23WfX951PMLvHJa7/9QHGU8kU4Uzsx8NiKOAj5JUQ4/THHDn/kUJfiJjcz/q4gYA1wGvIPid6RHKW4G9BzrFs4bKW4idBQwguKOtgspjrR+OTMfK6ebR3En42Mp9skbgL9SHN39+3J6SdpmRcOduyVJkiRJ6jJewylJkiRJqoSFU5IkSZJUCQunJEmSJKkSFk5JkiRJUiUsnJIkSZKkSlg4JUmSJEmVsHBKkiRJkiph4ZQkSZIkVcLCKUmSJEmqhIVTkiRJklSJ/x/ySEBfLMj4gAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1440x1008 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib\n",
    "from client.models.AMLmodel import classes as aml_classes\n",
    "\n",
    "class_names_y = [aml_classes[k] for k in aml_classes] + ['Total']\n",
    "class_names_x = [k for k in aml_classes] + ['Total']\n",
    "\n",
    "def plot_confusion_matrix(matrix, title='Confusion matrix', cmap=plt.cm.autumn):\n",
    "    \n",
    "    f, ax = plt.subplots(1,1,figsize=(20,14))\n",
    "    color_mat = np.zeros(matrix.shape)\n",
    "    color_mat[:15,:15] = matrix[:15,:15]/np.maximum(1,np.sum(matrix[:15,:15],0))\n",
    "    ax.matshow(color_mat, cmap=cmap, norm=matplotlib.colors.LogNorm()) \n",
    "\n",
    "    tick_mark = np.arange(16)\n",
    "    ax.set_xticks(tick_mark)\n",
    "    ax.set_yticks(tick_mark)\n",
    "    ax.set_xticklabels(class_names_x)\n",
    "    ax.set_yticklabels(class_names_y)\n",
    "    \n",
    "    for (i, j), z in np.ndenumerate(matrix):\n",
    "        ax.text(j, i, '{:0.0f}'.format(z), ha='center', va='center')\n",
    "\n",
    "    ax.axvline(x=14.5,color='black',lw=5)\n",
    "    ax.axhline(y=14.5,color='black', lw=5)\n",
    "    ax.set_xlabel('Predicted class', fontsize=20)\n",
    "    ax.set_ylabel('True class', fontsize=20)\n",
    "\n",
    "M = confusion_matrix(model, test_gen)\n",
    "plot_confusion_matrix(M)\n",
    "plt.savefig('confusion_sample.png')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create a model object in STACKn "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: models/1/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "10/13/2021 11:29:14 AM [builder_impl.py:774] Assets written to: models/1/assets\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "models/\n",
      "models/1/\n",
      "models/1/variables/\n",
      "models/1/variables/variables.data-00000-of-00001\n",
      "models/1/variables/variables.index\n",
      "models/1/saved_model.pb\n",
      "models/1/assets/\n",
      "Released model: clinic1, release_type: major\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We here create a model object in STACKn for use with Tensorflow Serving\n",
    "from stackn import stackn\n",
    "tf.saved_model.save(model, 'models/1/')\n",
    "!tar -czvf clinic1.tar.gz models\n",
    "stackn.create_object('clinic1', model_file='clinic1.tar.gz', release_type=\"major\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "import stackn.auth as sa\n",
    "import requests\n",
    "import numpy as np\n",
    "\n",
    "# If you are running locally with self signed certificate, then CHANGE the secure and verify variables to False\n",
    "secure = {\"STACKN_SECURE\": True}\n",
    "conf, status = sa.get_token(secure)\n",
    "\n",
    "# Load image to predict on / form input\n",
    "imgtest = np.load('dataset/processed/data_partitions/partition0/data_singlets/MON_1683.npy')\n",
    "imgtest=np.expand_dims(imgtest,0)\n",
    "inp = {\"inputs\": imgtest.tolist()}\n",
    "\n",
    "endpoint = 'https://r5c50cce0.studio.scaleoutsystems.com/v1/models/models:predict'\n",
    "res = requests.post(endpoint, json=inp, headers={\"Authorization\": \"Bearer \"+conf['STACKN_ACCESS_TOKEN']})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction: Lymphocyte (atypical)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(imgtest[0])\n",
    "print(\"Prediction:\", class_names_y[np.argmax(res.json()['outputs'])])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Save weights for use as an initial model in FEDn "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'weights_single_clinic.npz'"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from fedn.utils.kerashelper import KerasHelper\n",
    "helper = KerasHelper()\n",
    "weights = model.get_weights()\n",
    "helper.save_model(weights, \"weights_single_clinic.npz\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}