gc-ds1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-08-21 00:53:57.200909: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
      "To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
      "2023-08-21 00:53:57.789912: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n",
      "WARNING:root:No module named 'tempeh': LawSchoolGPADataset will be unavailable. To install, run:\n",
      "pip install 'aif360[LawSchoolGPA]'\n",
      "WARNING:root:No module named 'fairlearn': ExponentiatedGradientReduction will be unavailable. To install, run:\n",
      "pip install 'aif360[Reductions]'\n",
      "WARNING:root:No module named 'fairlearn': GridSearchReduction will be unavailable. To install, run:\n",
      "pip install 'aif360[Reductions]'\n",
      "WARNING:root:No module named 'fairlearn': GridSearchReduction will be unavailable. To install, run:\n",
      "pip install 'aif360[Reductions]'\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import cv2\n",
    "import os\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import accuracy_score, confusion_matrix, precision_score, recall_score\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.models import Sequential # type: ignore\n",
    "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, Dense, Dropout # type: ignore\n",
    "from tensorflow.keras.callbacks import EarlyStopping # type: ignore\n",
    "from aif360.datasets import BinaryLabelDataset\n",
    "#from aif360.metrics import BinaryLabelDatasetMetric, ClassificationMetric\n",
    "from aif360.algorithms.preprocessing import Reweighing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-08-21 00:53:58.335859: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-08-21 00:53:58.355226: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-08-21 00:53:58.355388: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n"
     ]
    }
   ],
   "source": [
    "#limit VRAM usage\n",
    "gpus = tf.config.experimental.list_physical_devices('GPU')\n",
    "for gpu in gpus:\n",
    "    tf.config.experimental.set_memory_growth(gpu, True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Load the training set and create a dataframe with image paths and labels\n",
    "dataset_path = \"datasets/dataset1/data/\"\n",
    "images = []\n",
    "labels = []\n",
    "for folder in os.listdir(dataset_path):\n",
    "    if os.path.isdir(os.path.join(dataset_path, folder)):\n",
    "        for file in os.listdir(os.path.join(dataset_path, folder)):\n",
    "            if file.endswith(\".jpg\"):\n",
    "                images.append(os.path.join(dataset_path, folder, file))\n",
    "                labels.append(folder)\n",
    "df = pd.DataFrame({\"image\": images, \"label\": labels})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Load the validation set and create a dataframe with image paths and labels\n",
    "validation_path = \"datasets/dataset1/validation/\"\n",
    "images = []\n",
    "labels = []\n",
    "for folder in os.listdir(validation_path):\n",
    "    if os.path.isdir(os.path.join(validation_path, folder)):\n",
    "        for file in os.listdir(os.path.join(validation_path, folder)):\n",
    "            if file.endswith(\".jpg\"):\n",
    "                images.append(os.path.join(validation_path, folder, file))\n",
    "                labels.append(folder)\n",
    "vf = pd.DataFrame({\"image\": images, \"label\": labels})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Define a function to preprocess the images: resize, grayscale, histogram equalization\n",
    "def preprocess_image(image_path):\n",
    "  image = cv2.imread(image_path)\n",
    "  image = cv2.resize(image, (64, 64))\n",
    "  image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)\n",
    "  image = cv2.equalizeHist(image)\n",
    "  image = image / 255.0\n",
    "  image = np.expand_dims(image, axis=2)\n",
    "  return image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Apply the preprocessing function to the images and convert the labels to numeric values\n",
    "X_train = np.array([preprocess_image(image) for image in df[\"image\"]])\n",
    "y_train = np.array([0 if label == \"male\" else 1 for label in df[\"label\"]])\n",
    "\n",
    "X_test = np.array([preprocess_image(image) for image in vf[\"image\"]])\n",
    "y_test = np.array([0 if label == \"male\" else 1 for label in vf[\"label\"]])\n",
    "#label for male is 0 & female is 1\n",
    "#0 is privileged group & 1 is unprivileged group"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv2d (Conv2D)             (None, 62, 62, 32)        320       \n",
      "                                                                 \n",
      " max_pooling2d (MaxPooling2D  (None, 31, 31, 32)       0         \n",
      " )                                                               \n",
      "                                                                 \n",
      " conv2d_1 (Conv2D)           (None, 29, 29, 64)        18496     \n",
      "                                                                 \n",
      " max_pooling2d_1 (MaxPooling  (None, 14, 14, 64)       0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " conv2d_2 (Conv2D)           (None, 12, 12, 128)       73856     \n",
      "                                                                 \n",
      " max_pooling2d_2 (MaxPooling  (None, 6, 6, 128)        0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " flatten (Flatten)           (None, 4608)              0         \n",
      "                                                                 \n",
      " dense (Dense)               (None, 256)               1179904   \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 256)               0         \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 1)                 257       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 1,272,833\n",
      "Trainable params: 1,272,833\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-08-21 00:54:12.100025: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-08-21 00:54:12.100196: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-08-21 00:54:12.100317: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-08-21 00:54:12.399188: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-08-21 00:54:12.399409: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-08-21 00:54:12.399535: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-08-21 00:54:12.399628: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1635] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 7400 MB memory:  -> device: 0, name: NVIDIA GeForce RTX 3080, pci bus id: 0000:07:00.0, compute capability: 8.6\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "\n",
    "model.add(Conv2D(32, (3, 3), activation=\"relu\", input_shape=(64, 64, 1)))\n",
    "model.add(MaxPooling2D((2, 2)))\n",
    "\n",
    "model.add(Conv2D(64, (3, 3), activation=\"relu\"))\n",
    "model.add(MaxPooling2D((2, 2)))\n",
    "\n",
    "model.add(Conv2D(128, (3, 3), activation=\"relu\"))\n",
    "model.add(MaxPooling2D((2, 2)))\n",
    "\n",
    "model.add(Flatten())\n",
    "\n",
    "model.add(Dense(256, activation=\"relu\"))\n",
    "model.add(Dropout(0.5))\n",
    "model.add(Dense(1, activation=\"sigmoid\"))\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-08-21 00:54:14.329877: I tensorflow/compiler/xla/stream_executor/cuda/cuda_dnn.cc:424] Loaded cuDNN version 8600\n",
      "2023-08-21 00:54:14.861306: I tensorflow/compiler/xla/stream_executor/cuda/cuda_blas.cc:637] TensorFloat-32 will be used for the matrix multiplication. This will only be logged once.\n",
      "2023-08-21 00:54:14.863511: I tensorflow/compiler/xla/service/service.cc:169] XLA service 0x3b565580 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
      "2023-08-21 00:54:14.863524: I tensorflow/compiler/xla/service/service.cc:177]   StreamExecutor device (0): NVIDIA GeForce RTX 3080, Compute Capability 8.6\n",
      "2023-08-21 00:54:14.866882: I tensorflow/compiler/mlir/tensorflow/utils/dump_mlir_util.cc:269] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable.\n",
      "2023-08-21 00:54:14.976897: I ./tensorflow/compiler/jit/device_compiler.h:180] Compiled cluster using XLA!  This line is logged at most once for the lifetime of the process.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1470/1470 [==============================] - 7s 3ms/step - loss: 0.2414 - accuracy: 0.9010 - val_loss: 0.1493 - val_accuracy: 0.9414\n",
      "Epoch 2/50\n",
      "1470/1470 [==============================] - 4s 3ms/step - loss: 0.1549 - accuracy: 0.9441 - val_loss: 0.1177 - val_accuracy: 0.9579\n",
      "Epoch 3/50\n",
      "1470/1470 [==============================] - 4s 3ms/step - loss: 0.1327 - accuracy: 0.9527 - val_loss: 0.1136 - val_accuracy: 0.9595\n",
      "Epoch 4/50\n",
      "1470/1470 [==============================] - 4s 3ms/step - loss: 0.1174 - accuracy: 0.9581 - val_loss: 0.1109 - val_accuracy: 0.9599\n",
      "Epoch 5/50\n",
      "1470/1470 [==============================] - 4s 2ms/step - loss: 0.1049 - accuracy: 0.9630 - val_loss: 0.1234 - val_accuracy: 0.9574\n",
      "Epoch 6/50\n",
      "1470/1470 [==============================] - 4s 3ms/step - loss: 0.0946 - accuracy: 0.9668 - val_loss: 0.1144 - val_accuracy: 0.9567\n",
      "Epoch 7/50\n",
      "1470/1470 [==============================] - 4s 2ms/step - loss: 0.0845 - accuracy: 0.9701 - val_loss: 0.1101 - val_accuracy: 0.9608\n",
      "Epoch 8/50\n",
      "1470/1470 [==============================] - 4s 2ms/step - loss: 0.0760 - accuracy: 0.9729 - val_loss: 0.1221 - val_accuracy: 0.9585\n",
      "Epoch 9/50\n",
      "1470/1470 [==============================] - 4s 2ms/step - loss: 0.0670 - accuracy: 0.9753 - val_loss: 0.1176 - val_accuracy: 0.9586\n",
      "Epoch 10/50\n",
      "1470/1470 [==============================] - 4s 2ms/step - loss: 0.0613 - accuracy: 0.9776 - val_loss: 0.1201 - val_accuracy: 0.9597\n",
      "Epoch 11/50\n",
      "1470/1470 [==============================] - 4s 2ms/step - loss: 0.0524 - accuracy: 0.9806 - val_loss: 0.1366 - val_accuracy: 0.9559\n",
      "Epoch 12/50\n",
      "1470/1470 [==============================] - 4s 2ms/step - loss: 0.0465 - accuracy: 0.9824 - val_loss: 0.1522 - val_accuracy: 0.9577\n"
     ]
    }
   ],
   "source": [
    "#Compile and fit the model on the train set\n",
    "model.compile(optimizer=\"adam\", loss=\"binary_crossentropy\", metrics=[\"accuracy\"])\n",
    "early_stopping = EarlyStopping(monitor=\"val_loss\", patience=5)\n",
    "history = model.fit(X_train, y_train, batch_size=32, epochs=50, validation_data=(X_test, y_test), callbacks=[early_stopping])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA/IAAAIQCAYAAAAimVUFAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAADdL0lEQVR4nOzdd3xT1f/H8Ve6dym0dLBayixC2RUHKFQKiAIiS5Sh4sTVrwtFQBwgIqKo8BMFEUFw4FZUqiACAgIFpOxVoIsyOulK8vsjEKzMQtvblvfz8biPJjcn934SSnM/Oed8jslqtVoRERERERERkUrBwegAREREREREROTiKZEXERERERERqUSUyIuIiIiIiIhUIkrkRURERERERCoRJfIiIiIiIiIilYgSeREREREREZFKRIm8iIiIiIiISCWiRF5ERERERESkElEiLyIiIiIiIlKJKJEXERERERERqUSUyIuIiMhleffddwkNDcXNzY2oqCjWrFlzzrYzZ87k+uuvx8/PDz8/P6Kjo89oP2zYMEwmU7GtW7duxdocPXqUwYMH4+PjQ7Vq1bjnnnvIzs4uk9cnIiJS0TgZHUBFZLFYSEpKwtvbG5PJZHQ4IiIiWK1WsrKyCAkJwcGh4nwPv3DhQmJjY5kxYwZRUVFMnTqVmJgYtm/fTs2aNc9ov3TpUgYNGsQ111yDm5sbr732Gl27dmXLli3UqlXL3q5bt27Mnj3bft/V1bXYcQYPHkxycjK//vorhYWFDB8+nPvuu4/58+dfVNz6rBcRkYqmRJ/1VjnDgQMHrIA2bdq0adNW4bYDBw4Y/TFZTPv27a0PP/yw/b7ZbLaGhIRYJ0yYcFHPLyoqsnp7e1vnzJlj3zd06FBrr169zvmchIQEK2Bdu3atfd9PP/1kNZlM1kOHDl3UefVZr02bNm3aKup2MZ/16pE/C29vbwAOHDiAj4+PwdGIiIhAZmYmderUsX9GVQQFBQWsW7eOUaNG2fc5ODgQHR3NqlWrLuoYubm5FBYWUr169WL7ly5dSs2aNfHz86Nz5868/PLL1KhRA4BVq1ZRrVo12rZta28fHR2Ng4MDq1evpk+fPmecJz8/n/z8fPt9q9UK6LNeREQqjpJ81iuRP4tTQ+x8fHz04S4iIhVKRRoGnp6ejtlsJjAwsNj+wMBAtm3bdlHHeOaZZwgJCSE6Otq+r1u3btx2222EhYWxe/dunnvuObp3786qVatwdHQkJSXljGH7Tk5OVK9enZSUlLOeZ8KECbz44otn7NdnvYiIVDQX81mvRF5EREQMMXHiRBYsWMDSpUtxc3Oz7x84cKD9dvPmzWnRogXh4eEsXbqULl26XNK5Ro0aRWxsrP3+qV4PERGRyqjiVMsRERGRSsXf3x9HR0dSU1OL7U9NTSUoKOi8z508eTITJ07kl19+oUWLFudtW79+ffz9/dm1axcAQUFBpKWlFWtTVFTE0aNHz3leV1dXe++7euFFRKSyUyIvIiIil8TFxYU2bdoQFxdn32exWIiLi6NDhw7nfN6kSZN46aWXWLx4cbF57udy8OBBjhw5QnBwMAAdOnTg+PHjrFu3zt7mt99+w2KxEBUVdRmvSEREpHLQ0PpLZLVaKSoqwmw2Gx2KSKlydHTEycmpQs3DFZGKKzY2lqFDh9K2bVvat2/P1KlTycnJYfjw4QAMGTKEWrVqMWHCBABee+01xowZw/z58wkNDbXPaffy8sLLy4vs7GxefPFF+vbtS1BQELt37+bpp5+mQYMGxMTEANC0aVO6devGiBEjmDFjBoWFhYwcOZKBAwcSEhJizBshIlWWrvultJTmdbYS+UtQUFBAcnIyubm5RociUiY8PDwIDg7GxcXF6FBEpIIbMGAAhw8fZsyYMaSkpNCyZUsWL15sL4CXmJhYbC3c6dOnU1BQwO23317sOGPHjmXcuHE4OjqyadMm5syZw/HjxwkJCaFr16689NJLxdaSnzdvHiNHjqRLly44ODjQt29f3n777fJ50SJyxdB1v5S20rrONllPrb8idpmZmfj6+pKRkXHGHDqLxcLOnTtxdHQkICAAFxcX9VxKlWG1WikoKODw4cOYzWYaNmxY7AJcRIxzvs8mKTm9nyJyIbrul9J0MdfZJflsUo98CRUUFGCxWKhTpw4eHh5GhyNS6tzd3XF2dmb//v0UFBQUqyQtIiIicqXQdb+UttK8zlZX2yVSL6VUZfr9FhEREbHRdZGUptL6fdJvpYiIiIiIiEglokReREREREREpBJRIi+XLDQ0lKlTpxodhoiIiIiIlKHSuO5X7lC6VOzuCnLDDTfQsmXLUvsPtHbtWjw9PUvlWCIiIiIiUjp03V/1KZGXYqxWK2azGSenC/9qBAQElENE5askr19EREREpLK60q/7KzsNrS8FVquV3IKict+sVutFxzhs2DCWLVvGW2+9hclkwmQysW/fPpYuXYrJZOKnn36iTZs2uLq68ueff7J792569epFYGAgXl5etGvXjiVLlhQ75n+Hx5hMJj744AP69OmDh4cHDRs25Ntvvz1vXHPnzqVt27Z4e3sTFBTEHXfcQVpaWrE2W7ZsoWfPnvj4+ODt7c3111/P7t277Y/PmjWLZs2a4erqSnBwMCNHjgRg3759mEwm4uPj7W2PHz+OyWRi6dKlAJf1+vPz83nmmWeoU6cOrq6uNGjQgA8//BCr1UqDBg2YPHlysfbx8fGYTCZ27dp13vdERERERComo677S3LtX1Gv+/8rMTGRXr164eXlhY+PD/379yc1NdX++MaNG7nxxhvx9vbGx8eHNm3a8PfffwOwf/9+brnlFvz8/PD09KRZs2b8+OOPJTp/Zadux1JwotBMxJify/28CeNj8HC5uH/Ct956ix07dnDVVVcxfvx4wPbN2r59+wB49tlnmTx5MvXr18fPz48DBw7Qo0cPXnnlFVxdXfn444+55ZZb2L59O3Xr1j3neV588UUmTZrE66+/zrRp0xg8eDD79++nevXqZ21fWFjISy+9ROPGjUlLSyM2NpZhw4bZ/yMeOnSIjh07csMNN/Dbb7/h4+PDihUrKCoqAmD69OnExsYyceJEunfvTkZGBitWrLjYt9DuUl7/kCFDWLVqFW+//TaRkZHs3buX9PR0TCYTd999N7Nnz+bJJ5+0n2P27Nl07NiRBg0alDg+ERERETGeUdf9cPHX/hX1uv/fLBaLPYlftmwZRUVFPPzwwwwYMMDe4TZ48GBatWrF9OnTcXR0JD4+HmdnZwAefvhhCgoK+OOPP/D09CQhIQEvL68LnrcqUSJ/hfD19cXFxQUPDw+CgoLOeHz8+PHcdNNN9vvVq1cnMjLSfv+ll17iq6++4ttvv7X3eJ/NsGHDGDRoEACvvvoqb7/9NmvWrKFbt25nbX/33Xfbb9evX5+3336bdu3akZ2djZeXF++++y6+vr4sWLDA/h+3UaNG9ue8/PLL/O9//+Oxxx6z72vXrt2F3o4zlPT179ixg88++4xff/2V6Ohoe/z/fh/GjBnDmjVraN++PYWFhcyfP/+MXnoRERERkdJUUa/7/y0uLo7Nmzezd+9e6tSpA8DHH39Ms2bNWLt2Le3atSMxMZGnnnqKJk2aANCwYUP78xMTE+nbty/NmzcHil+HXymUyJcCd2dHEsbHGHLe0tK2bdti97Ozsxk3bhw//PADycnJFBUVceLECRITE897nBYtWthve3p64uPjc8ZQ+X9bt24d48aNY+PGjRw7dgyLxQLY/nNGREQQHx/P9ddfb0/i/y0tLY2kpCS6dOlSkpd6ViV9/fHx8Tg6OtKpU6ezHi8kJISbb76ZWbNm0b59e7777jvy8/Pp16/fZccqIpVDxolC/tpzhGruzkTVr2F0OFKG0rLy2JB4HD8PF9qHXbgnSkQqL6Ou+0+duzQYdd3/b1u3bqVOnTr2JB4gIiKCatWqsXXrVtq1a0dsbCz33nsvc+fOJTo6mn79+hEeHg7Ao48+yoMPPsgvv/xCdHQ0ffv2LRbPlUCJfCkwmUwXPcS9ovpvFconn3ySX3/9lcmTJ9OgQQPc3d25/fbbKSgoOO9x/ptwm0wme3L+Xzk5OcTExBATE8O8efMICAggMTGRmJgY+3nc3d3Pea7zPQbg4GArAfHv+USFhYVnbVvS13+hcwPce++93HXXXbz55pvMnj2bAQMG4OHhccHniUjllFdoZv3+Y/y5K50Vu4+w+eBxLFboGhGoRL6KW7T+EBN/2kbPFsFK5EWqOF33n1aS6/5LMW7cOO644w5++OEHfvrpJ8aOHcuCBQvo06cP9957LzExMfzwww/88ssvTJgwgTfeeINHHnmk1M5f0VXu30IpERcXF8xm80W1XbFiBcOGDaNPnz6A7Zu6U/NqSsu2bds4cuQIEydOtH8bd6qAxSktWrRgzpw5FBYWnvHHwtvbm9DQUOLi4rjxxhvPOP6p6prJycm0atUKoFjhu/O50Otv3rw5FouFZcuW2YfW/1ePHj3w9PRk+vTpLF68mD/++OOizi0ilYPZYmVLUgYrdh1hxa501u47Sn5R8QuY+gGeNKh5Zc3ZuxJFBPsAkJCcaXAkIiI2Fe26/7+aNm3KgQMHOHDggD0PSEhI4Pjx40RERNjbNWrUiEaNGvHEE08waNAgZs+ebY+zTp06PPDAAzzwwAOMGjWKmTNnKpGXqik0NJTVq1ezb98+vLy8zluIomHDhixatIhbbrkFk8nECy+8UKrfsAHUrVsXFxcXpk2bxgMPPMA///zDSy+9VKzNyJEjmTZtGgMHDmTUqFH4+vry119/0b59exo3bsy4ceN44IEHqFmzJt27dycrK4sVK1bwyCOP4O7uztVXX83EiRMJCwsjLS2N0aNHX1RsF3r9oaGhDB06lLvvvtte7G7//v2kpaXRv39/ABwdHRk2bBijRo2iYcOGdOjQofTePBEpd1arlX1HcvlzVzord6WzcvcRMk4UH+VT09uVaxv4n9xqEOx74dE7Uvk1PZnI703PIbegqNL31olI5VfRrvv/Kzo6mubNmzN48GCmTp1KUVERDz30EJ06daJt27acOHGCp556ittvv52wsDAOHjzI2rVr6du3LwCPP/443bt3p1GjRhw7dozff/+dpk2blmnMFY2Wn7uCPPnkkzg6OhIREWEfxn4uU6ZMwc/Pj2uuuYZbbrmFmJgYWrduXarxBAQE8NFHH/H5558TERHBxIkTzygGV6NGDX777Teys7Pp1KkTbdq0YebMmfbe+aFDhzJ16lTee+89mjVrRs+ePdm5c6f9+bNmzaKoqIg2bdrw+OOP8/LLL19UbBfz+qdPn87tt9/OQw89RJMmTRgxYgQ5OTnF2txzzz0UFBQwfPjwS3mLRMRgaVl5fBN/iKc+38i1E3/jxslLeeHrf/jpnxQyThTi7epEdNNAxt0Swa9PdGT1c114c0BLbm9TW0n8FSTA25Wa3q5YrbAtJcvocEREKtx1/3+ZTCa++eYb/Pz86NixI9HR0dSvX5+FCxcCtg6xI0eOMGTIEBo1akT//v3p3r07L774IgBms5mHH36Ypk2b0q1bNxo1asR7771XpjFXNCZrSRYjv0JkZmbi6+tLRkYGPj4+xR7Ly8tj7969hIWF4ebmZlCEUlksX76cLl26cODAAQIDA40O56Lp91yuVNn5Razec+Rkr/sRtqcWT8pcHB1oXa8a14b7c21Df1rU8sXJsXy+Ez/fZ5OUXGm/n8Nmr2Hp9sO83Psq7ry6XilEKCJG0/WQlIXz/V6V5LNJY79EykB+fj6HDx9m3Lhx9OvXr1Il8SJXkoIiC/EHjtsK1O1KZ+OB4xRZTn+/bTLZ5j9fd3K4fLvQ6ri7lN6KIVJ1RAT7sHT7Yc2TFxGRcqFEXqQMfPrpp9xzzz20bNmSjz/+2OhwROQki8XKtpQsVuxKZ8XudNbsPUpuQfFiQPVqeNjmuIf70yG8BtU9XQyKViqTiJCTBe+SlMiLiEjZUyIvUgaGDRvGsGHDjA5DRIADR3NZsSvdNlx+9xGO5hRfTqeGpwvXNPDnugY1uCbcnzrVtUyklNypyvXbUjIxW6w4OpgMjkhERKoyJfIiIlKlHM0pYOXudPuycIlHc4s97uHiSFRYdXt1+caB3jgo6ZLLVK+GJ+7OjpwoNLM3PUfLDoqISJkyvGr9u+++S2hoKG5ubkRFRbFmzZpzti0sLGT8+PGEh4fj5uZGZGQkixcvLtbGbDbzwgsvEBYWhru7O+Hh4bz00kuopp+ISNW2PvEYI+evp90rSxg5fwOfrkkk8WguTg4m2oX68ViXhnz+QAfix3Rl9vD23Ht9fZoG+yiJl1Lh6GCiSbA3oPXkRUSk7BnaI79w4UJiY2OZMWMGUVFRTJ06lZiYGLZv307NmjXPaD969Gg++eQTZs6cSZMmTfj555/p06cPK1eupFWrVgC89tprTJ8+nTlz5tCsWTP+/vtvhg8fjq+vL48++mh5v0QRESlDhWYLP/2Twqw/9xJ/4Lh9f+NAb65raFvLvX1YDbxcNQBNyl5EsA8bEo+TkJTJrZEhRocjIiJVmKFXNlOmTGHEiBH2NbZnzJjBDz/8wKxZs3j22WfPaD937lyef/55evToAcCDDz7IkiVLeOONN/jkk08AWLlyJb169eLmm28GIDQ0lE8//fS8Pf0iIlK5HMsp4NO1iXy8cj8pmXmAbWm4W1uGMPzaUJqF+BocoVyJ7AXv1CMvIiJlzLBEvqCggHXr1jFq1Cj7PgcHB6Kjo1m1atVZn5Ofn3/GWnvu7u78+eef9vvXXHMN77//Pjt27KBRo0Zs3LiRP//8kylTppwzlvz8fPLz8+33MzP1ASwiUhHtTM1i1op9fLXhIHmFFgD8vVy56+p63BFVlwBvV4MjlCvZqYJ3qlwvIiJlzbA58unp6ZjN5jPW1w4MDCQlJeWsz4mJiWHKlCns3LkTi8XCr7/+yqJFi0hOTra3efbZZxk4cCBNmjTB2dmZVq1a8fjjjzN48OBzxjJhwgR8fX3tW506dUrnRVZBoaGhTJ061X7fZDLx9ddfn7P9vn37MJlMxMfHX9Z5S+s4IlL5WCxWft+exl0fruamN//g0zWJ5BVaaBbiwxv9Ilnx7I08Ft1QSbwYrkmQDw4mSM/OJy0rz+hwREQuS1W/7h82bBi9e/cu03OUpUo1afCtt95ixIgRNGnSBJPJRHh4OMOHD2fWrFn2Np999hnz5s1j/vz5NGvWjPj4eB5//HFCQkIYOnToWY87atQoYmNj7fczMzOVzF+k5ORk/Pz8SvWYw4YN4/jx48X+UNSpU4fk5GT8/f1L9VwiUnHlFhTx5fpDzF6xlz2HcwBwMEHXiCCGXxtK+7DqmEwqVCcVh7uLI2H+nuw+nENCUiY1G7td+EkiIpWErvsrFsMSeX9/fxwdHUlNTS22PzU1laCgoLM+JyAggK+//pq8vDyOHDlCSEgIzz77LPXr17e3eeqpp+y98gDNmzdn//79TJgw4ZyJvKurK66u6sm5FOf6typtjo6O5XauiqawsBBnZ2ejwxApN4eOn+DjVfv4dHUimXlFAHi7OjGgXR2GXhOqdd6lQosI8bUl8smZ3ND4zMK9IiKVla77KxbDhta7uLjQpk0b4uLi7PssFgtxcXF06NDhvM91c3OjVq1aFBUV8eWXX9KrVy/7Y7m5uTg4FH9Zjo6OWCyW0n0Blcz7779PSEjIGe9Dr169uPvuuwHYvXs3vXr1IjAwEC8vL9q1a8eSJUvOe9z/DrFZs2YNrVq1ws3NjbZt27Jhw4Zi7c1mM/fcc499ecDGjRvz1ltv2R8fN24cc+bM4ZtvvsFkMmEymVi6dOlZh9gsW7aM9u3b4+rqSnBwMM8++yxFRUX2x2+44QYeffRRnn76aapXr05QUBDjxo077+tZu3YtN910E/7+/vj6+tKpUyfWr19frM3x48e5//77CQwMxM3Njauuuorvv//e/viKFSu44YYb8PDwwM/Pj5iYGI4dOwacOUQJoGXLlsXiMplMTJ8+nVtvvRVPT09eeeWVC75vp8yaNYtmzZrZ35ORI0cCcPfdd9OzZ89ibQsLC6lZsyYffvjhed8TkfJgtVpZt/8oD89bT8dJv/N/y/aQmVdEvRoejLslglXPdWF0zwgl8VLhaZ68iBhN1/0Xd93/X/n5+Tz66KPUrFkTNzc3rrvuOtauXWt//NixYwwePJiAgADc3d1p2LAhs2fPBmz130aOHElwcDBubm7Uq1ePCRMmlOj8JWXo0PrY2FiGDh1K27Ztad++PVOnTiUnJ8dexX7IkCHUqlXL/iasXr2aQ4cO0bJlSw4dOsS4ceOwWCw8/fTT9mPecsstvPLKK9StW5dmzZqxYcMGpkyZYv+lLRNWKxTmlt3xz8XZAy5yWGm/fv145JFH+P333+nSpQsAR48eZfHixfz4448AZGdn06NHD1555RVcXV35+OOPueWWW9i+fTt169a94Dmys7Pp2bMnN910E5988gl79+7lscceK9bGYrFQu3ZtPv/8c2rUqMHKlSu57777CA4Opn///jz55JNs3bqVzMxM+3+M6tWrk5SUVOw4hw4dokePHgwbNoyPP/6Ybdu2MWLECNzc3Ir9p50zZw6xsbGsXr2aVatWMWzYMK699lpuuumms76GrKwshg4dyrRp07Barbzxxhv06NGDnTt34u3tjcVioXv37mRlZfHJJ58QHh5OQkICjo6OAMTHx9OlSxfuvvtu3nrrLZycnPj9998xm80X9e90yrhx45g4cSJTp07Fycnpgu8bwPTp04mNjWXixIl0796djIwMVqxYAcC9995Lx44dSU5OJjg4GIDvv/+e3NxcBgwYUKLYREpTQZGFn/5JZtafe9l4MMO+/9oGNbj72jBubFxT67xLpaLK9SJVnFHX/XDR1/667r+46/7/evrpp/nyyy+ZM2cO9erVY9KkScTExLBr1y6qV6/OCy+8QEJCAj/99BP+/v7s2rWLEydOAPD222/z7bff8tlnn1G3bl0OHDjAgQMHLuq8l8rQRH7AgAEcPnyYMWPGkJKSQsuWLVm8eLG9AF5iYmKx3vW8vDxGjx7Nnj178PLyokePHsydO5dq1arZ20ybNo0XXniBhx56iLS0NEJCQrj//vsZM2ZM2b2Qwlx41YD1Yp9LAhfPi2rq5+dH9+7dmT9/vv0/9BdffIG/vz833ngjAJGRkURGRtqf89JLL/HVV1/x7bff2nt2z2f+/PlYLBY+/PBD3NzcaNasGQcPHuTBBx+0t3F2dubFF1+03w8LC2PVqlV89tln9O/fHy8vL9zd3cnPzz/vkJr33nuPOnXq8M4772AymWjSpAlJSUk888wzjBkzxv5706JFC8aOHQtAw4YNeeedd4iLizvnf+jOnTsXu//+++9TrVo1li1bRs+ePVmyZAlr1qxh69atNGrUCKDY1I5JkybRtm1b3nvvPfu+Zs2aXfC9+6877rjD/oXWKed73wBefvll/ve//xX7I9quXTvAtppD48aNmTt3rv2Lr9mzZ9OvXz+8vLxKHJ/I5TqaU8CnaxL5eNU+UjNtq4a4ODnQp2Uthl8XSpMgH4MjFLk0p3rk96bnkFtQhIdLpSpHJCIXYtR1P1z0tb+u+y/uuv/fcnJymD59Oh999BHdu3cHYObMmfz66698+OGHPPXUUyQmJtKqVSvatm0L2EbanpKYmEjDhg257rrrMJlM1KtX74LnvFyGf7qMHDnynL8sS5cuLXa/U6dOJCQknPd43t7eTJ069YzhywKDBw9mxIgRvPfee7i6ujJv3jwGDhxo/+XPzs5m3Lhx/PDDDyQnJ1NUVMSJEydITEy8qONv3bqVFi1aFFsi8GzTJN59911mzZpFYmIiJ06coKCggJYtW5botWzdupUOHToUK3R17bXXkp2dzcGDB+3fJLZo0aLY84KDg0lLSzvncVNTUxk9ejRLly4lLS0Ns9lMbm6u/T2Ij4+ndu3a9iT+v+Lj4+nXr1+JXsvZnPoD8W/ne9/S0tJISkqy/7E+m3vvvZf333+fp59+mtTUVH766Sd+++23y45VpCR2pGYxe8VeFq0/RH6RbchfTe/Ty8fV8FK9EqncArxdCfB25XBWPttSsmhdt3QLQ4mIXAxd91/4uv/fdu/eTWFhIddee619n7OzM+3bt2fr1q0APPjgg/Tt25f169fTtWtXevfuzTXXXAPYivbddNNNNG7cmG7dutGzZ0+6du1aotdZUoYn8lWCs4ftGzIjzlsCt9xyC1arlR9++IF27dqxfPly3nzzTfvjTz75JL/++iuTJ0+mQYMGuLu7c/vtt1NQUFBqIS9YsIAnn3ySN954gw4dOuDt7c3rr7/O6tWrS+0c//bfInEmk+m89RKGDh3KkSNHeOutt6hXrx6urq506NDB/h64u7uf93wXetzBwQGr1VpsX2Fh4RntPD2Lf9t6offtQucF21SVZ599llWrVrFy5UrCwsK4/vrrL/g8kctlsVhZtuMws1bsZfnOdPv+5rV8uee6MHo0D8bFybCSLSKlLiLYh2VZh0lIylQiL1LVGHXdf+rcF0nX/Re+7i+p7t27s3//fn788Ud+/fVXunTpwsMPP8zkyZNp3bo1e/fu5aeffmLJkiX079+f6Ohovvjii1I7/38pkS8NJtNFD3E3kpubG7fddhvz5s1j165dNG7cmNatW9sfX7FiBcOGDaNPnz6A7Zu6ffv2XfTxmzZtyty5c8nLy7N/O/fXX38Va7NixQquueYaHnroIfu+3bt3F2vj4uJywTnlTZs25csvv8Rqtdq/nVuxYgXe3t7Url37omP+rxUrVvDee+/Ro0cPAA4cOEB6+unEo0WLFhw8eJAdO3actVe+RYsWxMXFFRtG9G8BAQEkJyfb72dmZrJ3796Liut875u3tzehoaHExcXZh0z9V40aNejduzezZ89m1apVZwzdFyltOflFfLn+IB+t2Mee9NPLx3W7Koi7rw2jTT0/LR8nVVJEiA/LdhzWPHmRqkjX/UDVuO7/t/DwcFxcXFixYoV9WHxhYSFr167l8ccft7cLCAhg6NChDB06lOuvv56nnnqKyZMnA+Dj48OAAQMYMGAAt99+O926dePo0aNUr169VGL8L3WBXGEGDx7MDz/8wKxZsxg8eHCxxxo2bMiiRYuIj49n48aN3HHHHSX6FuuOO+7AZDIxYsQIEhIS+PHHH+2/2P8+x99//83PP//Mjh07eOGFF4pVgwTbfJNNmzaxfft20tPTz9pj/dBDD3HgwAEeeeQRtm3bxjfffMPYsWOJjY09Y9WCkmjYsCFz585l69atrF69msGDBxfr7e7UqRMdO3akb9++/Prrr/Zv3hYvXgzAqFGjWLt2LQ899BCbNm1i27ZtTJ8+3f5lQOfOnZk7dy7Lly9n8+bNDB061F4o70JxXeh9GzduHG+88QZvv/02O3fuZP369UybNq1Ym3vvvZc5c+awdevWcy7HKHK5Dh7L5dUft3L1hDjGfLOFPek5eLs5cV/H+vzx9I28N7gNbUO1BrxUXapcLyIVga77L56npycPPvggTz31FIsXLyYhIYERI0aQm5vLPffcA8CYMWP45ptv2LVrF1u2bOH777+nadOmAEyZMoVPP/2Ubdu2sWPHDj7//HOCgoKK1XIrbUrkrzCdO3emevXqbN++nTvuuKPYY1OmTMHPz49rrrmGW265hZiYmGLf3F2Il5cX3333HZs3b6ZVq1Y8//zzvPbaa8Xa3H///dx2220MGDCAqKgojhw5UuxbOoARI0bQuHFj2rZtS0BAgL3y+r/VqlWLH3/8kTVr1hAZGckDDzzAPffcw+jRo0vwbpzpww8/5NixY7Ru3Zq77rrLvgTFv3355Ze0a9eOQYMGERERwdNPP23/JrFRo0b88ssvbNy4kfbt29OhQwe++eYbnJxsg19GjRpFp06d6NmzJzfffDO9e/cmPDz8gnFdzPs2dOhQpk6dynvvvUezZs3o2bMnO3fuLNYmOjqa4OBgYmJiCAkxqFCLVElWq5W/9x3loXnr6Djpd97/Yw9ZeUXU9/dkfK9m/DWqC8/1aEptPy0fJ1Xfqcr121IyMVusF2gtIlI2dN1fMhMnTqRv377cddddtG7dml27dvHzzz/j52ebIuXi4sKoUaNo0aIFHTt2xNHRkQULFgC20bGnil63a9eOffv28eOPP5baFw1nY7L+d8KukJmZia+vLxkZGfj4FK+cnJeXx969ewkLCytW3EGkMsjOzqZWrVrMnj2b22677Zzt9HsuFys1M4/vNyXz1YaD/HPodO/j9Q39ufvaMDo1CtDycaXkfJ9NUnJl+X6aLVauGvszJwrNLIntRIOaWh1EpDLS9ZCUhfP9XpXks0lz5EWuABaLhfT0dN544w2qVavGrbfeanRIUokdzy3gp39S+DY+ib/2HuHU18GuTg7c1roWw64Jo3GQt7FBihjI0cFEk2BvNiQeJyE5U4m8iIiUOiXyIleAxMREwsLCqF27Nh999JF9qL/IxcotKGLJ1jS+jT/Esh2HKTSfHszVtp4ft7YMoWeLEKp7uhgYpUjFERHsY0vkkzK5NVJTmUREpHTpal7kChAaGnrGsnciF1JQZOGPHYf5dmMSvyakcqLwdFXZpsE+3BoZwi2RwZr3LnIWp+bJq3K9iIiUBSXyIiJiZ7ZYWb33CN9tTOLHzSlknDhdPbZeDQ9ujQzh1sgQGgZq6LzI+ahyvYiIlCUl8iIiVzir1cqmgxl8uzGJ7zclkZqZb3+sprcrPVuEcGvLECJr+2rJOJGL1CTIBwcTpGfnk5aVR01vFcoSEZHSo0T+EmmYslRl+v2+MuxKy+Lb+CS+3ZjEviO59v0+bk70aB7MrZEhRNWvgaOqzouUmLuLI2H+nuw+nMPW5Cwl8iKVmK6LpDSV1u+TEvkScnZ2BiA3Nxd3d3eDoxEpG7m5tqTu1O+7VB2Hjp/gu41JfBOfxNZ/zd11d3YkOiKQWyND6NjIH1cnRwOjFKkaIkJ82X04h4SkTDo1CjA6HBEpIV33S1koretsJfIl5OjoSLVq1UhLSwPAw8NDQ02lyrBareTm5pKWlka1atVwdFQyVxUcyc7nx83JfBOfxN/7j9n3OzmY6NQogFtbhhDdNBBPV30kyKV59913ef3110lJSSEyMpJp06bRvn37s7adOXMmH3/8Mf/88w8Abdq04dVXX7W3LywsZPTo0fz444/s2bMHX19foqOjmThxIiEhp6u/h4aGsn///mLHnjBhAs8++2wZvcqSiwj24buNSSp4J1JJ6bpfSlNpX2frqu0SBAUFAdj/U4tUNdWqVbP/nkvllJVXyC9bUvlmYxIrdqVjttiGcZlMEBVWnVsja9H9qiD8tFycXKaFCxcSGxvLjBkziIqKYurUqcTExLB9+3Zq1qx5RvulS5cyaNAgrrnmGtzc3Hjttdfo2rUrW7ZsoVatWuTm5rJ+/XpeeOEFIiMjOXbsGI899hi33norf//9d7FjjR8/nhEjRtjve3tXrCKM9sr1SRkGRyIil0rX/VLaSus622TVpI8zZGZm4uvrS0ZGBj4+PudsZzabKSwsPOfjIpWRs7OzeuIrqbxCM0u3p/FNfBK/bUsjv8hif6xFbV9ujbSt9R7kq7m6ldHFfjaVt6ioKNq1a8c777wDgMVioU6dOjzyyCMX1TtuNpvx8/PjnXfeYciQIWdts3btWtq3b8/+/fupW7cuYOuRf/zxx3n88ccvKe7yeD8PZ+XT7pUlmEyw5cUYPFzUfyJSWem6X0rDha6zS/LZpE+Uy+Do6KiER0QMVWS2sGL3Eb6NT+KXLSlk5RfZH6sf4EmvyFrcEhlM/QAvA6OUqqqgoIB169YxatQo+z4HBweio6NZtWrVRR0jNzeXwsJCqlevfs42GRkZmEwmqlWrVmz/xIkTeemll6hbty533HEHTzzxBE5OZ7+0yc/PJz//9IoMmZllP9w9wNuVAG9XDmflsz0li1Z1/cr8nCJSNnTdLxWNEnkRkUrGarWyPvE438Yf4ofNyaRnF9gfC/Z149bIEG6JDKFZiI/m8kmZSk9Px2w2ExgYWGx/YGAg27Ztu6hjPPPMM4SEhBAdHX3Wx/Py8njmmWcYNGhQsd6JRx99lNatW1O9enVWrlzJqFGjSE5OZsqUKWc9zoQJE3jxxRcv8pWVnohgH5ZlHSYhOVOJvIiIlBol8iIilcSutGy+iT/EN/FJJB49vVycn4czN7cI5tbIWrSt54eDlouTSmLixIksWLCApUuX4uZ25pSPwsJC+vfvj9VqZfr06cUei42Ntd9u0aIFLi4u3H///UyYMAFXV9czjjVq1Khiz8nMzKROnTql+GrOLiLEh2U7DpOQpIJ3IiJSepTIi4hUYKmZeXy3MYmv4w/xz6HTiYCHiyNdIwLp1bIW1zX0x9nRwcAo5Url7++Po6MjqampxfanpqZesJDP5MmTmThxIkuWLKFFixZnPH4qid+/fz+//fbbBecKRkVFUVRUxL59+2jcuPEZj7u6up41wS9rEcEnC96pcr2IiJQiJfIiIhVMZl4hizen8M3GQ6zcfYRTJUmdHEx0bBRAr5Yh3BQRqMJZYjgXFxfatGlDXFwcvXv3BmzF7uLi4hg5cuQ5nzdp0iReeeUVfv75Z9q2bXvG46eS+J07d/L7779To0aNC8YSHx+Pg4PDWSvlG+lU5fptyVmYLVYcNWJGRERKga4CRUQqgPwiM79vO8w38YeI25ZGwb8qzrep50fvliH0aB5MDa/y71EUOZ/Y2FiGDh1K27Ztad++PVOnTiUnJ4fhw4cDMGTIEGrVqsWECRMAeO211xgzZgzz588nNDSUlJQUALy8vPDy8qKwsJDbb7+d9evX8/3332M2m+1tqlevjouLC6tWrWL16tXceOONeHt7s2rVKp544gnuvPNO/Pwq1jz00BqeuDs7cqLQzL4jOYSr8KSIiJQCJfIiIgaxWKys3nuUb+IP8ePmZDLzTlecb1DTi94tQ+jVshZ1qnsYGKXI+Q0YMIDDhw8zZswYUlJSaNmyJYsXL7YXwEtMTMTB4fTUj+nTp1NQUMDtt99e7Dhjx45l3LhxHDp0iG+//RaAli1bFmvz+++/c8MNN+Dq6sqCBQsYN24c+fn5hIWF8cQTTxSbA19RODqYaBLszYbE4yQkZSqRFxGRUqF15M+ioq7VKyKVn9VqZWtyFt/EH+LbjUkkZ+TZHwv0ceXWSFvyrorz8l/6bCpd5fl+Pv/VZuatTuTBG8J5pluTMj2XiIhUXlpHXkSkgjl4LJdv4pP4Jv4QO1Kz7fu93ZzoflUQvVvWIqp+Dc2fFamCTs2TV+V6EREpLUrkRUTKyLGcAn7YnMw38YdYu++Yfb+LowOdm9Skd6sQbmhcEzdnRwOjFJGypsr1IiJS2pTIi4iUohMFZpZsTeWb+EMs23GYQrNt9pLJBFeH1aB3qxC6XRWMr7uzwZGKSHlpHOSNyQSHs/JJy8qjpreb0SGJiEglp0ReROQyFZktrNx9hK/jD/HzPynkFJjtj0UE+9C7VQi3RIYQ7OtuYJQiYhQPFyfC/D3ZcziHrclZSuRFROSyKZEXEbkEVquVTQcz+Dr+EN9tTCY9O9/+WG0/d3q1DKF3y1o0DPQ2MEoRqSgign3YcziHhKRMOjUKMDocERGp5JTIi4iUwP4jOXy14RDfxCexNz3Hvt/Pw5meLULo1TKENvX8VHFeRIqJCPHh+03JmicvIiKlQom8iMgFFJotxG1N5ZO/EvlzV7p9v5uzA10jgujdKoTrGwbg7OhwnqOIyJXMXvAuKcPgSEREpCpQIi8icg7JGSf4dM0BFqxJJC3LNnTeZILrGwbQp1UIXSOC8HTVn1ERubBTS9DtSc8ht6AIDxf97RARkUunTxERkX+xWKws35XOJ3/tJ25rKhZb0Xn8vVwY0K4Og9rXpbafh7FBikilU9PbDX8vV9Kz89mekkWrun5GhyQiIpWYEnkREeBIdj6frzvI/NWJJB7Nte+/un51BkfVI6ZZEC5OGjovIpcuIsSHP3YcJiE5U4m8iIhcFiXyInLFslqt/L3/GJ/8tZ+fNqdQYLYA4O3mxO1tajM4qi4NaqrqvIiUjojgk4l8kgreiYjI5VEiLyJXnKy8Qr7acIh5fyWyPTXLvj+yti+Do+pxS2QI7i6OBkYoIlXRqXnyqlwvIiKXS4m8iFwx/jmUwbzV+/kmPoncAjNgqzzfK7IWd15dj+a1fQ2OUESqslOV67clZ2G2WHF00DKVIiJyaZTIi0iVlldo5ruNScxbnUj8geP2/Q1rejE4qi59WtfG193ZuABF5IoR5u+Jm7MDJwrN7DuSQ3iAl9EhiYhIJaVEXkSqpN2Hs5n3VyJfrDtAZl4RAM6OJrpdFcydUXVpH1Ydk0m9YSJSfhwdTDQJ8iH+wHESkjKVyIuIyCVTIi8iVUah2cIvW1KZt3o/K3cfse+v7efOHVF16d+2Dv5ergZGKCJXuoiQk4l8cia3RIYYHY6IiFRSSuRFpNI7dPwEC9YksmDtAQ5n5QPgYILOTWoy+Op6dGwYoLmoIlIhnJonr8r1IiJyOSrEosjvvvsuoaGhuLm5ERUVxZo1a87ZtrCwkPHjxxMeHo6bmxuRkZEsXry4WJvQ0FBMJtMZ28MPP1zWL0VEyonZYuX3bWncO2ct17/2G9N+28XhrHwCvF15pHMDlj/TmQ+GtuPGxjWVxItIhaHK9SIiUhoM75FfuHAhsbGxzJgxg6ioKKZOnUpMTAzbt2+nZs2aZ7QfPXo0n3zyCTNnzqRJkyb8/PPP9OnTh5UrV9KqVSsA1q5di9lstj/nn3/+4aabbqJfv37l9rpEpGykZ+fz2d8HmL86kYPHTtj3XxNegzuvrsdNEYE4O1aI7yhFRM7QJMgbkwkOZ+WTlpVHTW83o0MSEZFKyGS1Wq1GBhAVFUW7du145513ALBYLNSpU4dHHnmEZ5999oz2ISEhPP/888V61/v27Yu7uzuffPLJWc/x+OOP8/3337Nz586LKm6VmZmJr68vGRkZ+Pj4XOIrE5HSYrVaWb33KPNWJ7L4n2QKzbY/Wz5uTtzepg6Dr66rolFS5emzqXQZ+X52fmMpew7nMOfu9nRqFFCu5xYRkYqrJJ9NhvbIFxQUsG7dOkaNGmXf5+DgQHR0NKtWrTrrc/Lz83FzK/7ttbu7O3/++ec5z/HJJ58QGxurCtUilYzZYuXHzcm8+/sutqVk2fe3rFONwVF1uSUyBDdnRwMjFBEpuYhgH/YcziEhKVOJvIiIXBJDE/n09HTMZjOBgYHF9gcGBrJt27azPicmJoYpU6bQsWNHwsPDiYuLY9GiRcWG0v/b119/zfHjxxk2bNg548jPzyc/P99+PzNT89ZEjFRktvDdpiTe+W0Xuw/nAODu7EjvViEMjqrHVbV8DY5QROTSRYT48P2mZM2TFxGRS2b4HPmSeuuttxgxYgRNmjTBZDIRHh7O8OHDmTVr1lnbf/jhh3Tv3p2QkHMv8TJhwgRefPHFsgpZRC5SodnCVxsO8d7vu9h3JBcAX3dn7r42jGHXhOLr4WxwhCIil+905foMgyMREZHKytBE3t/fH0dHR1JTU4vtT01NJSgo6KzPCQgI4OuvvyYvL48jR44QEhLCs88+S/369c9ou3//fpYsWcKiRYvOG8eoUaOIjY2138/MzKROnTqX8IpE5FIUFFn4Yt1B3lu6y17Azs/DmXuvr8+QDvXwdlMCLyJVx6nK9XvSc8gtKMLDpdL1q4iIiMEM/eRwcXGhTZs2xMXF0bt3b8BW7C4uLo6RI0ee97lubm7UqlWLwsJCvvzyS/r3739Gm9mzZ1OzZk1uvvnm8x7L1dUVV1fXS34dInJp8grNfPb3AWYs3U1SRh4A/l4u3NexPoOj6uHpqotbEal6anq74e/lSnp2PttTsmhV18/okEREpJIx/Co5NjaWoUOH0rZtW9q3b8/UqVPJyclh+PDhAAwZMoRatWoxYcIEAFavXs2hQ4do2bIlhw4dYty4cVgsFp5++ulix7VYLMyePZuhQ4fi5GT4yxSRfzlRYGb+mkT+b9lu0rJs9SlqervyQKdwBrWvi7uLCtiJSNUWEeLDHzsOk5CcqUReRERKzPAMd8CAARw+fJgxY8aQkpJCy5YtWbx4sb0AXmJiIg4Op9eEzsvLY/To0ezZswcvLy969OjB3LlzqVatWrHjLlmyhMTERO6+++7yfDkich45+UXMW72f9//YQ3p2AQAhvm48eEM4/drWUQV6EbliRASfTOSTVPBORERKzvB15CsirdUrUrqy8gr5eNV+Pli+h2O5hQDU9nPn4Rsb0Ld1bVycHC5wBBHRZ1PpMvr9/HZjEo9+uoFWdavx1UPXlvv5RUSk4qk068iLSNWWcaKQj1bsY9aKvWScsCXwoTU8ePjGBvRuVQtnRyXwInJlOlW5fltyFmaLFUcHk8ERiYhIZaJEXkRK3bGcAmat2MtHK/aRlV8EQHiAJ490bkjPFsE4KYEXkStcmL8nbs4OnCg0s+9IDuEBXkaHJCIilYgSeREpNenZ+XywfC9zV+0jp8AMQONAb0Z2bkCP5sHqcRIROcnRwUSTIB/iDxwnISlTibyIiJSIEnkRuWxpmXm8/8cePlm9n7xCC2AbNvpolwZ0jQjCQQm8iMgZIkJOJvLJmdwSGWJ0OCIiUokokReRS5accYL/W7aHT9ckkl9kS+Aja/vySOeGdGlaE5NJCbyIyLmcmievyvUiIlJSSuRFpMQOHstl+tLdfP73QQrMtgS+dd1qPNqlIZ0aBSiBFxG5CBEhJxP5ZCXyIiJSMkrkReSi7T+Sw3u/7+bL9QcpsthWrowKq86jXRpyTXgNJfAiIiXQJMgbkwkOZ+VzOCufAG9Xo0MSEZFKQom8iFzQnsPZvPP7Lr6JT8J8MoG/tkENHunckKvr1zA4OhGRysnDxYkwf0/2HM5ha3ImAd4BRockIiKVhBJ5ETmnnalZTPttF99vSuJk/k6nRgE82qUBbepVNzY4EZEqICLYhz2Hc0hIzqRjIyXyIiJycZTIi8gZEo/kMunnbfywORnryQQ+umlNHunckMg61QyNTUSkKokI8eH7TckqeCciIiWiRF5E7I7nFjDtt118vGofhWZbBt+tWRAjOzfgqlq+BkcnIlL1NA1WwTsRESk5JfIiQn6RmY9X7mfabzvJzCsC4PqG/ozq3tReVVlEREpfs5OJ/J7D2ZwoMOPu4mhwRCIiUhkokRe5glmtVr7blMzrP2/jwNETgK2K8qgeTemkuZoiImUuwNsVfy8X0rML2J6aRUtNXxIRkYugRF7kCrVm71Fe+XErGw8cByDQx5X/3dSYvm1q4+igZeRERMqDyWSiabAPy3emk5CUqUReREQuihJ5kSvMnsPZTPxpG78kpALg4eLIA53Cuff6MDxc9CdBRKS8RYScTOSTM4wORUREKgkHowMQkfJxJDufsd/8Q9c3/+CXhFQcTHBHVF2WPnUDj3ZpqCReRC7Zu+++S2hoKG5ubkRFRbFmzZpztp05cybXX389fn5++Pn5ER0dfUZ7q9XKmDFjCA4Oxt3dnejoaHbu3FmszdGjRxk8eDA+Pj5Uq1aNe+65h+zs7DJ5fWUt4lTBO1WuFxGRi6REXqSKyys0897SXdzw+lLmrNpPkcVKlyY1+fnxjrzapzk1vd2MDlFEKrGFCxcSGxvL2LFjWb9+PZGRkcTExJCWlnbW9kuXLmXQoEH8/vvvrFq1ijp16tC1a1cOHTpkbzNp0iTefvttZsyYwerVq/H09CQmJoa8vDx7m8GDB7NlyxZ+/fVXvv/+e/744w/uu+++Mn+9ZaHZyaKi21KyMFusBkcjIiKVgclqteoT4z8yMzPx9fUlIyMDHx9V7JbKyWKx8nX8ISb/vJ2kDNvFb7MQH57v0ZRrGvgbHJ2IlFRF/WyKioqiXbt2vPPOOwBYLBbq1KnDI488wrPPPnvB55vNZvz8/HjnnXcYMmQIVquVkJAQ/ve///Hkk08CkJGRQWBgIB999BEDBw5k69atREREsHbtWtq2bQvA4sWL6dGjBwcPHiQkJOSC561I76fZYqXZ2MXkFVr47X+dqB/gZWg8IiJijJJ8NqlHXqQKWrkrnVvf/ZPYzzaSlJFHiK8bbw6I5LuR1ymJF5FSU1BQwLp164iOjrbvc3BwIDo6mlWrVl3UMXJzcyksLKR69eoA7N27l5SUlGLH9PX1JSoqyn7MVatWUa1aNXsSDxAdHY2DgwOrV68+63ny8/PJzMwstlUUjg4mGgdpPXkREbl4SuRFqpCdqVnc/dFa7vhgNf8cysTb1YmnuzXmtydvoE+r2jioGr2IlKL09HTMZjOBgYHF9gcGBpKSknJRx3jmmWcICQmxJ+6nnne+Y6akpFCzZs1ijzs5OVG9evVznnfChAn4+vratzp16lxUfOVF8+RFRKQkVN1KpApIy8pj6pKdLFiTiMUKTg4mBkfV5dEuDanh5Wp0eCIiZzVx4kQWLFjA0qVLcXMr23odo0aNIjY21n4/MzOzQiXzESHqkRcRkYunRF6kEsstKOKD5XuZsWw3uQVmAGKaBfJMtyaaYykiZc7f3x9HR0dSU1OL7U9NTSUoKOi8z508eTITJ05kyZIltGjRwr7/1PNSU1MJDg4udsyWLVva2/y3mF5RURFHjx4953ldXV1xda24X2yqR15EREpCQ+tFKiGzxcpnaw9w4+SlTPl1B7kFZiLrVOOz+zvwf3e1VRIvIuXCxcWFNm3aEBcXZ99nsViIi4ujQ4cO53zepEmTeOmll1i8eHGxee4AYWFhBAUFFTtmZmYmq1evth+zQ4cOHD9+nHXr1tnb/Pbbb1gsFqKiokrr5ZWrJkHemEyQlpXP4ax8o8MREZEKTj3yIpXMHzsO8+qPW9mWkgVAbT93nunWhJ4tgjGZNAdeRMpXbGwsQ4cOpW3btrRv356pU6eSk5PD8OHDARgyZAi1atViwoQJALz22muMGTOG+fPnExoaap/T7uXlhZeXFyaTiccff5yXX36Zhg0bEhYWxgsvvEBISAi9e/cGoGnTpnTr1o0RI0YwY8YMCgsLGTlyJAMHDryoivUVkaerE2E1PNmTnsPW5EwCvAOMDklERCowJfIilcTW5Exe/XEry3emA+Dj5sQjnRsy5Jp6uDo5GhydiFypBgwYwOHDhxkzZgwpKSm0bNmSxYsX24vVJSYm4uBwegDg9OnTKSgo4Pbbby92nLFjxzJu3DgAnn76aXJycrjvvvs4fvw41113HYsXLy42j37evHmMHDmSLl264ODgQN++fXn77bfL/gWXoaYhPuxJzyEhOZOOjZTIi4jIuWkd+bOoSGvLiqRk5DHl1+18vu4gVis4O5oY0iGURzo3oJqHi9HhiUg50WdT6aqI7+e7v+/i9Z+3c2tkCG8PamV0OCIiUs5K8tmkHnmRCio7v4j3l+3m/eV7yCu0AHBzi2CejmlMvRqeBkcnIiKlTZXrRUTkYimRF6lgiswWFv59gDd/3Ul6tq3gUZt6fjzXoylt6vkZHJ2IiJSVZicr1+85nM2JAjPuLpo2JSIiZ6dEXqQC+WPHYcZ/n8CutGwAQmt48Gz3JsQ0C1IhOxGRKi7A2xV/LxfSswvYnppFyzrVjA5JREQqKCXyIhVAfpGZ137azqwVewHw83DmsS4NuSOqHi5OWiVSRORKYDKZaBrsw/Kd6SQkZSqRFxGRc1IiL2Kw3YezeWT+BvucyKEd6hHbtTG+7s4GRyYiIuUtIuRkIp+cYXQoIiJSgSmRFzGI1Wrl83UHGfvNFk4Umqnu6cLrt7egS9NAo0MTERGDRJycJ5+QpIJ3IiJybkrkRQyQmVfI81/9w3cbkwC4JrwGbw5oSaCP2wWeKSIiVVmzk5Xrt6VkYbZYcXRQfRQRETmTEnmRcrY+8RiPfrqBg8dO4Ohg4n9dG3F/x3BdrImICGH+Xrg5O5BbYGb/kRzqB3gZHZKIiFRASuRFyonZYmXGst1M+XUHZouVOtXdeXtgK1rV1ZJyIiJi4+hgonGQDxsPHCchOVOJvIiInJXKYYuUg5SMPO78YDWv/7wds8XKrZEh/PDo9UriRUTkDJonLyIiF6IeeZEytiQhlae+2Mix3EI8XBwZ3+sq+raupXXhRUTkrCJOzpM/tZqJiIjIfymRFykjeYVmJv60jY9W7gNsBYymDWqlYZIiInJe6pEXEZELUSIvUgZ2pWUxcv4GtqVkAXDvdWE81a0xrk6OBkcmIiIVXZMgb0wmSMvK53BWPgHerkaHJCIiFYwSeZFSZLVaWbj2AOO+20JeoYUani5M7h/JjY1rGh2aiIhUEp6uToTV8GRPeg5bkzMJ8A4wOiQREalglMiLlJKME4U8t2gzP2xOBuD6hv680T+Smt5aG/6yHNkNf7wO+VnQ9FZo3B3cfIyOSkSkTDUN8WFPeg4JyZl0bKREXkREijO8av27775LaGgobm5uREVFsWbNmnO2LSwsZPz48YSHh+Pm5kZkZCSLFy8+o92hQ4e48847qVGjBu7u7jRv3py///67LF+GXOHW7T9Kj7eW88PmZJwcTIzq3oQ5w9srib8cBbkQ9xK8dzVs/BS2fQ9f3QeTG8LCO2HLV7Y2IiJVkObJi4jI+RjaI79w4UJiY2OZMWMGUVFRTJ06lZiYGLZv307NmmcORR49ejSffPIJM2fOpEmTJvz888/06dOHlStX0qpVKwCOHTvGtddey4033shPP/1EQEAAO3fuxM9Py3xJ6TNbrLz7+y7eituJ2WKlXg0P3h7Yisg61YwOrfKyWmHrd/Dzc5BxwLav/o1Qux1sWQRHdtke3/odOHvaeuiv6gsNuoCT5pGKSNWgyvUiInI+JqvVajXq5FFRUbRr14533nkHAIvFQp06dXjkkUd49tlnz2gfEhLC888/z8MPP2zf17dvX9zd3fnkk08AePbZZ1mxYgXLly+/5LgyMzPx9fUlIyMDHx8N4ZWzS844weML4lm99ygAfVrV4qXeV+Hlqhkrlyx9J/z0NOz+zXbftw7EvApNbwGTyZbkp2yGf760JfXHE08/19UXmvaEq26DsE7g6GzMaxApI/psKl0V/f1My8yj/atxOJhgy4vdcHdRsVQRkaquJJ9NhmUcBQUFrFu3jlGjRtn3OTg4EB0dzapVq876nPz8fNzcig9Vdnd3588//7Tf//bbb4mJiaFfv34sW7aMWrVq8dBDDzFixIhzxpKfn09+fr79fmamvv2usgrzIC/jP9txcPaA8M7gfHFD4X/eksIzX27ieG4hni6OvNT7Km5rXbtsY6/K8rNt8+BXvQuWQnB0gWsfg+tiwcXjdDuTCYJb2LbocXBo3cmk/ivISob4ebbNo4ZtPv1VfaHeNeCgC2ARqVwCvF3x93IhPbuA7alZtNRILxER+RfDEvn09HTMZjOBgYHF9gcGBrJt27azPicmJoYpU6bQsWNHwsPDiYuLY9GiRZjNZnubPXv2MH36dGJjY3nuuedYu3Ytjz76KC4uLgwdOvSsx50wYQIvvvhi6b04KTvmQsjLtCXfeceLJ+Qnjp89Sf/3/aK8cx/b1Reu6gORg6BOlC1p/I+8QjOv/LCVuX/tB6BFbV/eHtiKUH/Psni1VZ/VautZ/3k0ZCXZ9jXsCt0mQo3w8z/XZILabW1b11cgcZUtqU/4BnLTYd1s2+YVBM1625L62u3O+u8qpcBi1hcmIqXIZDLRNNiH5TvTSUjKVCIvIiLFVKoxwG+99RYjRoygSZMmmEwmwsPDGT58OLNmzbK3sVgstG3blldffRWAVq1a8c8//zBjxoxzJvKjRo0iNjbWfj8zM5M6deqU7Yu50hXk2OY6nzMJ/+/9k20Kc0rh5CZb1XM3X3CrZvt5dC9kHoR1H9k2vzCIHAgtBkD1MAB2pGbxyPwNbE+1rQ1/f8f6/K9rY1ycDK8ZWTmlbYUfn4J9J6fBVKsH3V+DRt1Knmw7OEDotbat+yTY94ctqd/6HWSnwOoZts23ru3Lmma3QXCkkvqSsloh85BtCsSRXSd/7oT0XbZ6BsGR0PUlCOtodKQiVULEqUQ+OcPoUEREpIIxLJH39/fH0dGR1NTUYvtTU1MJCgo663MCAgL4+uuvycvL48iRI4SEhPDss89Sv359e5vg4GAiIiKKPa9p06Z8+eWX54zF1dUVV1cVySpTJ45B4l+wfwXsXwXJ8WApuvTjuXifTMT/tblX+8++amd/3MXblvj9m8UC+/+EjQtsPbrH9sLSCbB0Ata6HVjt3ZWHN9bjSJEb/l6uTOkfqeWALlVeJix7zZZYW4rAyc02hP7aR8HZ/fKP7+hkmyYR3hluftM23/6fL2H7j5CRCCvesm3Vw2299FfdBjWbXv55q5L8rJOJ+q6TifrJhP3Ibig8z0oByfEw5xZo1B1uGg8BjcotZJGqyF7wTpXrRUTkPwxL5F1cXGjTpg1xcXH07t0bsPWmx8XFMXLkyPM+183NjVq1alFYWMiXX35J//797Y9de+21bN++vVj7HTt2UK9evVJ/DXIeWSmwf+XpLS0B+E9dRQ9/21zmMxLw8yThbtXA1ceWrJUmBwdbL2JYR+jxOmz9HjZ+inXPUkyJq7iaVaxwdCbe51qaxtyHb7hWQSgxqxU2fQa/vgDZJ7/Aa3wzdHsV/ELL5pxOLtC4m20rPAE7f7El9Tt+hqO74Y9Jtq1mhC2hb3bbhYf0VxUWs61YYLGe9ZM97VnJ536eg5Pt36tGQ/BvcPJnQ/AKtH05s/ZD2PGT7b1uOxxuGAWe/uX2skSqklNL0G1LycJsseLooFFEIiJiY2jV+oULFzJ06FD+7//+j/bt2zN16lQ+++wztm3bRmBgIEOGDKFWrVpMmDABgNWrV3Po0CFatmzJoUOHGDduHHv37mX9+vVUq1YNgLVr13LNNdfw4osv0r9/f9asWcOIESN4//33GTx48EXFVdEr2VY4Visc22ebo7x/hS1xP7rnzHY1GtoKj9W7Fup1gGp1yz3Ukli95wivLIijQ85v3O70Bw1Nh04/6FkTWvS3Db8Pam5ckJVFyj+2YfSJK233q9e3DYFveJMx8eRnwfbFtqR+1xJbgb1Tglvaeuqb9YFqVWCKzYljZ/asp++y/R8155/7eZ4BxZP1Gg1sCbtf6PlXBEjfCb+OsY2AANsXb9fHQtSDF11MUs5On02lqzK8n0VmC83G/kx+kYXf/teJ+gFeRockIiJlqFJUrQcYMGAAhw8fZsyYMaSkpNCyZUsWL15sL4CXmJiIw7+GQOfl5TF69Gj27NmDl5cXPXr0YO7cufYkHqBdu3Z89dVXjBo1ivHjxxMWFsbUqVMvOomXi2CxQPr208Pk9688XajMzgRBV51M2q+Buh3Aq6Yh4ZZUkdnCtN92Me23nVisnmT5D6TngInguA82fgqbP4ecNFj1jm0LvMqW0DfvD96BFzz+FeXEcdsUhTUzwWoGJ3fo+CRc84ixa767ekOLfrbtxDHY9oMtqd+zzDY8PDneNnKgTpQtqY/oXbH/bc2FtjoP/x0Gn77TVvjvXBxdbSMQTiXpp3rXa4SD+yWOOvFvCIM+hb1/wC+jIXkjLBkHa2dB9Fjb+6naBCIXxcnRgSZB3mw8mEFCcqYSeRERsTO0R76iqgzf0pcrcxGkbLIl7IknE/cTR4u3cXCCkNane9zrtLcNh69kDh0/weMLNrB23zEA+rauzfhezfD899rw5kJbL+7GT2H7T2AusO03OUB4F1tS3+Tm0pnvXVlZLLb3Z8lYyDls2xfRy1ZdviL3cuek22ok/LPI9kWVfTqICUKvsyWhTW8FzxpnPtdqtc35L8q3/U6c2opO3c63/e6U5uNF+ZBx0DYixmo+M6ZTvEOKD4M/1dPuW6dsK81bLLBpIcSNP/1lX602EPMq1L267M5bRemzqXRVlvdz1KJNfLrmAA/dEM7T3ZoYHY6IiJShknw2KZE/i8ry4V5mCvMgaf3pYfIH1kBBdvE2Tu5Qp93pHvdabYuv910J/bQ5mWe+3ERmXhFerk680ucqerWsdf4nnThmW8N84wI4sPr0flcfW+IaOcg2GuG/xfWqsqR42zD6g2ts9/0b2arRh3c2NKwSy0yGhK9tPfUH157eb3IE31q2pPq/ibaRnD1tPenFetYb2DZXg3vxCnLhr3dh+ZunV55oeivc9KJtmoVclCv+s6mUVZb3c+6qfbzwzRZuaBzAR8PbGx2OiIiUISXyl6myfLiXmvwsW7J+qjDdoXVnJiWuvrYetFM97sGRtkJiVcCJAjMv/ZDA/NWJAETWqca0ga2oW6OEX0wc2W1L6DctsBURO6VavdNL2VXlQmq5R+G3l+HvWYDVllje8IxtbnRl/105tt/2hc0/X9pGp1wUk236gKPL6c3p1G1X2zxzp5M/HV3/8/ip9hd43NHFNuS/RkPwCan4Q9azUmHpq7D+Y7BawMEZ2t9nm27hUd3o6Cq8K+6zqYxVlvdz3f6j9J2+iprerqx5PtrocEREpAwpkb9MleXD/ZLlHj09RH7/CkjedOawXM+atoJ0p3rca0aU7RBcg+w/ksP9c9exLSULkwke6BRO7E2NcHa8jB50i8X2/m6cD1u+gYKs04/VibL10jfrfelzkCsaiwU2fAxLXjw95eKq223rifuEGBtbWTi61zYE356InyNRd3Cs+Im1UVITbDUIdi2x3XerBp2egXb3Vv4vfcpQlf9sKmeV5f3Mzi+i+bifsVrh79HR+HtpuVwRkapKifxlqiwf7iW29Xtbj+nhrWc+Vq2uLWmvezJ5rxFe5ZOQ37en8dinG8jMK8Lfy5WpA1pyXcNSXiarINdWvXvjp7b1zK0W235HV2jc3ZbUN+hy/irgFdnBdfDjk7apGAABTW3L94Vdb2xcUjnsioNfXoC0Lbb7fmG24fZNb61af3/MhaXyf7zKfjYZpDK9nzdOXsre9Bzm3tOe6xsGGB2OiIiUkUpTtV7KkbnQlnCdWh/av/HJYfInK8pX5AJkpcxqtfLu77t449cdWK3Qum41pt/ZhkCfMlgay8UDmt9u27JSbBXv4z+1JS4JX9s2D39o3g9aDoKgFpUjgck5AnHjYP1cwGqrCXDDKGg/ovJ+KSHlr0EXqH8DbPgEfn8Fju2Fz4bY/iZ1fQVqtzE6wktz4pitav+epbatYQx0n2h0VFKJRQT7sDc9h4SkTCXyIiICKJG/cmz/yZbEe/jDQ6sqzVJwpS07v4j/fRbPz1tSAbgjqi5jb4nA1akcpg14B9mWXeswElI22+bTb/7MVtV99XTbVjMCGvcA39rgHWx7jncwePpXjKkNFjOsmw1xL0Hecdu+yEEQ/WLFXp5NKi4HR2gz1LYiwIq3YOU029SUDzrbpmhEj7WNGKrICvNsxS5PJe5JGzi94gGwb7lBgUlVERHiww+bk0lIzjQ6FBERqSCUyF8p/v7Q9rP1XVdsEr/7cDb3z13HrrRsXBwdGN+rGQPbG5AgmEwQ3MK23TTeNuR+46e2tczTEmzbGc9xBK/A04n9uX56VC+7Hv0Da+CH/50u9hbYHG6erGXEpHS4ekHn56HNMFvvfPx8+OcL2PodXP0gXB8Lbr5GR2ljsUDq5tOJ+/5VUHSieJuAJrbRBvVvsE1XErkMEcG24ZUJSUrkRUTERon8lSB9l+1iExO0GW50NIb4NSGV2IXxZOUXEeTjxvQ7W9OqbgUoNufoBI262rYTx21rmCett1X3zkq2DcfPSbMVI8xKOr0W9zmP5wJeQScT+/Mk/W6+F5/wZ6fBr2NtxfvA9tzOL9h+lxz1J0RKmW8t6P0eRN0Pv4y2DVFfMRU2zLVN32gzzJjpG8f2n0zcf4c9y04XdjzFK+h04l7/BvAJLv8YpcqKCLEl8rsPZ5NXaMbNuQKM0BIREUPpKvxK8Pcs28+GXcGvnrGxlDOLxcrUuJ28HbcTgPah1Xl3cGsCvCtg1V/3arYhxm2GFt9vLrINvz+V2J/x8+Tt3HTbeuYZibbtfJzcz5Ho/+sLAM8A2/D/31+F/Azb81rdBV3GgpfmaEoZC46EId/Cjp9tFe7Td9jqfKx5H256CRrFlG09idyjxee5H9tb/HEXLwi9DurfaEvcAxpXjvoWUinV9HalhqcLR3IK2J6SRWSdakaHJCIiBlMiX9UVnoD4ebbb7e4xNpZylnGikNiF8cRtSwNg2DWhPH9z08tbWs4Ijk623r0L9fAVFUB26jmS/X/9zDtuGwZ8bO+Zycm5BLeEm9+A2m0v99WIXDyTCRp3sxXFW/cRLJ1gS+g/HQBhHaHry7aEvzQU5sGBv/41zz2eYvPcTY5Qux2En0zca7VRYUcpNyaTiYgQH5bvTCchOVOJvIiIKJGv8v5ZZEvcfOtCg2ijoyk3O1KzuH/uOvam5+Dq5MCrfZrTt01to8MqW04uttUHLrQCQeGJ4j355+rlL8gC9+rQ5QVoPbRiFNuTK5Ojs21FhBb9YfkU+Gu6rbf8/zrZii12eQF8Qkp2TIvFVu/hVOKeuAqK8oq3CWj6r3nu14BbxV6iTKq2iOCTibzmyYuICErkq75TRe7aDrtiErEfNyfz5OcbyS0wU6uaOzPubEPz2hWkSFZF4OwO1cNs2/nkZ9mG4GsevFQUbr62debb3g1x423F8DbOhy1f2VaEuPYxW9G8czm273TifrZ57t7BpxP3sE6a5y4Vyql58qpcLyIioES+akuKh0PrwMEZWg0xOpoyZ7ZYmfzLdqYv3Q3ANeE1mDaoFTW8KuB8+MrA1dvoCETOzq8e3P4hXP0Q/PK8rTf9j0mwfg7c+Dy0utP2xWWxee6/2xL5f3PxPjnP/QbNc5cK71Tl+q3JmVgsVhwc9LsqInIlUyJflZ3qjY+4tcoXJzueW8Ajn25g+c50AEZcH8Yz3ZrgVNnmw4vIxavdBob/ZFui7tcxtpoP3z1qG3rv5ArJGyk2z93ByTbPvf4NtiJ1tVprnrtUGmH+nrg6OZBbYGb/0VzC/D2NDklERAykRL6qysuAzV/Ybret2kXuEpIyuf+Tvzlw9ARuzg5Muj2SWyNLOF9WRConk8n2ZWWjbrD2A1j2GhzeevrxgKanC9TVu0YjTaTScnJ0oEmQNxsPZpCQlKlEXkTkCqdEvqrauAAKc20XsfWuMTqaMvNN/CGe+XITeYUW6lb34P/uakPTYBWkErniOLlAh4cgciBs/hzcqkH9TrblFEWqiIgQH1sin5zBzS1Uw0FE5EqmRL4qslph7akid3dXyTmfRWYLE3/axgd/2pZP69gogLcHtqSah4vBkYmIoTyqQ9T9RkchUiZOzZNX5XoREdEE4qpo/wpI3w7OHhA5wOhoSt2R7HyGzFpjT+IfuiGc2cPaKYkXETHIu+++S2hoKG5ubkRFRbFmzZpztt2yZQt9+/YlNDQUk8nE1KlTz2hz6rH/bg8//LC9zQ033HDG4w888EBZvLwKQ5XrRUTkFCXyVdHaD2w/W/S3LddUhWw+mMGt76xg5e4jeLg4Mn1wa57u1gRHVe8VETHEwoULiY2NZezYsaxfv57IyEhiYmJIS0s7a/vc3Fzq16/PxIkTCQo6+9SHtWvXkpycbN9+/fVXAPr161es3YgRI4q1mzRpUum+uAqmcZAPJhOkZuaTnp1vdDgiImIgJfJVTVaqrYIzVLkid1+sO0jfGSs5dPwEYf6efP3wtXRvrjmCIiJGmjJlCiNGjGD48OFEREQwY8YMPDw8mDVr1lnbt2vXjtdff52BAwfi6nr25UEDAgIICgqyb99//z3h4eF06tSpWDsPD49i7Xx8qnaNFC9XJ0Jr2IrcbVWvvIjIFU2JfFWz4WOwFNmWWApuYXQ0paLQbGHsN//w5OcbKSiy0KVJTb5++FoaBar6tIiIkQoKCli3bh3R0dH2fQ4ODkRHR7Nq1apSO8cnn3zC3Xffjek/NV/mzZuHv78/V111FaNGjSI3N/ecx8nPzyczM7PYVhlpnryIiICK3VUtFjOsm2O7XUV649Oy8hg5bwNr9h0F4LEuDXmsS0McNJReRMRw6enpmM1mAgMDi+0PDAxk27ZtpXKOr7/+muPHjzNs2LBi+++44w7q1atHSEgImzZt4plnnmH79u0sWrTorMeZMGECL774YqnEZKSIEB9+2JysefIiIlc4JfJVyc5fIOMAuPtBsz5GR3PZ1ice48FP1pGamY+3qxNTBrTkpojACz9RRESqjA8//JDu3bsTEhJSbP99991nv928eXOCg4Pp0qULu3fvJjw8/IzjjBo1itjYWPv9zMxM6tSpU3aBlxH1yIuICCiRr1pOLTnXcjA4uxkby2X6dE0iY7/ZQoHZQniAJ+8PaUt4gJfRYYmIyL/4+/vj6OhIampqsf2pqannLGRXEvv372fJkiXn7GX/t6ioKAB27dp11kTe1dX1nHPyK5NTlet3H84mr9CMm7OjwRGJiIgRNEe+qji2D3Ytsd1ue7ehoVyO/CIzoxZtZtSizRSYLcQ0C+Trh69VEi8iUgG5uLjQpk0b4uLi7PssFgtxcXF06NDhso8/e/Zsatasyc0333zBtvHx8QAEB1ftIqg1vV2p4emCxQrbU7KMDkdERAyiHvmq4u/ZgBXq3wg1zuyJqAxSM/N44JN1bEg8jskET3ZtzIOdwjUfXkSkAouNjWXo0KG0bduW9u3bM3XqVHJychg+fDgAQ4YMoVatWkyYMAGwFa9LSEiw3z506BDx8fF4eXnRoEED+3EtFguzZ89m6NChODkVv1zZvXs38+fPp0ePHtSoUYNNmzbxxBNP0LFjR1q0qBqFXs/FZDIREeLD8p3pJCRnElmnmtEhiYiIAZTIVwVF+bBhru12u8pZ5G7tvqM8+Ml60rPz8XFz4q1BrbixcU2jwxIRkQsYMGAAhw8fZsyYMaSkpNCyZUsWL15sL4CXmJiIg8PpAYBJSUm0atXKfn/y5MlMnjyZTp06sXTpUvv+JUuWkJiYyN13nznKzMXFhSVLlti/NKhTpw59+/Zl9OjRZfdCK5CI4JOJvObJi4hcsZTIVwUJ30LuEfAOgUbdjY6mRKxWK5/8tZ8Xv0ugyGKlSZA3/3dXG+qdXCdXREQqvpEjRzJy5MizPvbv5BwgNDQUq9V6wWN27dr1nO3q1KnDsmXLShxnVXFqnrwq14uIXLmUyFcFaz+w/WwzFBwrzz9pXqGZ0V//wxfrDgLQs0Uwk25vgYdL5XkNIiIi5e1U5fqtyZlYLFZNQRMRuQIpY6rsUrfAgb/A5AithxodzUVLOn6CBz5Zx6aDGTiY4NnuTRhxfX1MJl2MiIiInE+YvyeuTg7kFpjZfzSXMH+NYhMRudKoan1ld2rJuSY3g0/lqNSbcaKQO2b+xaaDGfh5OPPx3VHc1zFcSbyIiMhFcHJ0oHGQN6D15EVErlRK5Cuz/CzYtNB2u5IUubNYrPzvs3j2HcmlVjV3vh15Hdc19Dc6LBERkUrl1PD6hOQMgyMREREjKJGvzDZ9BgXZUKMBhHUyOpqLMn3ZbpZsTcPFyYHpd7amTnUPo0MSERGpdOwF79QjLyJyRVIiX1lZrfD3LNvttndDJRiW/seOw0z+ZTsAL/VqRova1YwNSEREpJI63SOvRF5E5EqkRL6yOrAGUv8BJzeIHGR0NBd08Fgujy3YgNUKA9vVYUC7ukaHJCIiUmk1OZnIp2bmk56db3A0IiJS3pTIV1Z/nyxyd1Vf8KhubCwXkFdo5qF56zmWW0jzWr6Mu7WZ0SGJiIhUal6uToTWsE1P26peeRGRK44S+coo5whs+cp2u23FL3L34ndb2HQwg2oezky/szVuzo5GhyQiIlLpaZ68iMiVS4l8ZRT/CZgLIDgSarU2OprzWrg2kU/XHMBkgrcHtqK2n4rbiYiIlAbNkxcRuXIpka9sLJZ/Fbm7p0IXudt8MIMXvtkCwJNdG9OxUYDBEYmIiFQd6pEXEblyKZGvbHb/Bsf2gasvNL/d6GjO6VhOAQ98so6CIgvRTQN5sFO40SGJiIhUKRHBvgDsPpxNXqHZ4GhERKQ8KZGvbE4VuWs5CFw8jY3lHMwWK48tjOfQ8ROE1vDgjf6RODhU3JEDIiIilVGgjyvVPV2wWGF7SpbR4YiISDmqEIn8u+++S2hoKG5ubkRFRbFmzZpzti0sLGT8+PGEh4fj5uZGZGQkixcvLtZm3LhxmEymYluTJk3K+mWUveMHYMfJ19r2bmNjOY+pS3bwx47DuDk7MOOuNvi6OxsdkoiISJVjMpk0T15E5ApleCK/cOFCYmNjGTt2LOvXrycyMpKYmBjS0tLO2n706NH83//9H9OmTSMhIYEHHniAPn36sGHDhmLtmjVrRnJysn37888/y+PllK31c8BqgdDrIaCx0dGc1ZKEVKb9tguAibe1oEmQj8ERiYiIVF2aJy8icmUyPJGfMmUKI0aMYPjw4URERDBjxgw8PDyYNWvWWdvPnTuX5557jh49elC/fn0efPBBevTowRtvvFGsnZOTE0FBQfbN39+/PF5O2TEXwvqPbbcraG/8vvQcnvgsHoBh14TSu1UtYwMSERGp4tQjLyJyZTI0kS8oKGDdunVER0fb9zk4OBAdHc2qVavO+pz8/Hzc3NyK7XN3dz+jx33nzp2EhIRQv359Bg8eTGJi4jnjyM/PJzMzs9hW4Wz7HrJTwbMmNOlpdDRnOFFg5oFP1pGVV0Sben4816Op0SGJiIhUead65LcmZ2KxWA2ORkREyouhiXx6ejpms5nAwMBi+wMDA0lJSTnrc2JiYpgyZQo7d+7EYrHw66+/smjRIpKTk+1toqKi+Oijj1i8eDHTp09n7969XH/99WRlnb0QzIQJE/D19bVvderUKb0XWVrWnixy13oIOLkYG8t/WK1WnvtqM9tSsvD3cuW9wa1xcTJ8sIeIiEiVV9/fExcnB3ILzOw/mmt0OCIiUk4qXbb11ltv0bBhQ5o0aYKLiwsjR45k+PDhODicfindu3enX79+tGjRgpiYGH788UeOHz/OZ599dtZjjho1ioyMDPt24MCB8no5F+fwDti3HEwO0GaY0dGcYe5f+/lqwyEcHUy8c0crAn3cLvwkERERuWxOjg40CfIGNE9eRORKYmgi7+/vj6OjI6mpqcX2p6amEhQUdNbnBAQE8PXXX5OTk8P+/fvZtm0bXl5e1K9f/5znqVatGo0aNWLXrl1nfdzV1RUfH59iW4Xy98l6AQ1joFrFGi2wbv9Rxn+XAMCo7k24un4NgyMSERG5spyeJ59hcCQiIlJeDE3kXVxcaNOmDXFxcfZ9FouFuLg4OnTocN7nurm5UatWLYqKivjyyy/p1avXOdtmZ2eze/dugoODSy32clOQA/Hzbbfb3WNsLP9xOCufh+atp8hi5ebmwdxzXZjRIYmIiFxxVLleROTKY/jQ+tjYWGbOnMmcOXPYunUrDz74IDk5OQwfPhyAIUOGMGrUKHv71atXs2jRIvbs2cPy5cvp1q0bFouFp59+2t7mySefZNmyZezbt4+VK1fSp08fHB0dGTRoULm/vsv2z5eQnwHV6kF4F6OjsSsyWxg5fz2pmfk0qOnFa7e3wGQyGR2WiIjIFedUj/zW5LPXAhIRkarHyegABgwYwOHDhxkzZgwpKSm0bNmSxYsX2wvgJSYmFpv/npeXx+jRo9mzZw9eXl706NGDuXPnUq1aNXubgwcPMmjQII4cOUJAQADXXXcdf/31FwEBAeX98i7fqSJ3be8GB8O/d7Gb9PN2Vu89iqeLIzPubIOXq+G/SiIiIlekJicT+ZTMPI5k51PDy9XgiEREpKyZrFar1ir5j8zMTHx9fcnIyDB2vvyhdTCzMzi6QOxW8PQ3LpZ/+XFzMg/NWw/A9MGt6d68Ek5ZEBGpZCrMZ1MVUdXezxte/519R3L55J4ormtYMa4XRESkZEry2VRxunjlTGtPFrmL6F1hkvhdaVk89flGAO7vVF9JvIiISAVgnyevgnciIlcEJfIV1YljtvnxUGGK3GXnF3H/3HXkFJjpUL8GT3VtbHRIIiIiwr8q16vgnYjIFUGJfEUV/ykUnYCazaBOlNHRYLVaefqLjew+nEOQjxvT7miFk6N+fURERCqC0z3ySuRFRK4EysQqIqv19Nrx7e6GClAN/oPle/lxcwrOjibeu7M1/iqkIyIiUmFEBPsCsPtwDnmFZoOjERGRsqZEviLa+wcc2QkuXtBigNHRsGr3ESYu3gbAmFua0bqun8ERiYiIyL8F+rhS3dMFs8XKjlQtQyciUtUpka+I/j655FyL/uDqbWgoKRl5PPLpeswWK7e1rsWdUXUNjUdERETOZDKZNE9eROQKokS+oslMhm0/2G63NbbIXUGRhYfmrSM9u4CmwT680rs5pgowzF9ERETOpHnyIiJXjhIn8qGhoYwfP57ExMSyiEfWfwyWIluBu6CrDA3llR8SWJ94HB83J2bc2Rp3F0dD4xEREZFzU4+8iMiVo8SJ/OOPP86iRYuoX78+N910EwsWLCA/P78sYrvymItg3Ue22+3uNTSUrzYcZM6q/QBMHdiSejU8DY1HREREzu9Uj/zW5EwsFqvB0YiISFm6pEQ+Pj6eNWvW0LRpUx555BGCg4MZOXIk69evL4sYrxw7FkNWEnjUgIhehoWRkJTJqEWbAXi0S0M6Nwk0LBYRERG5OPX9PXFxciCnwEzi0VyjwxERkTJ0yXPkW7duzdtvv01SUhJjx47lgw8+oF27drRs2ZJZs2Zhteqb4BI7VeSu1Z3gZMzybhm5hTzwyTryCi10ahTAY10aGhKHiIiIlIyTowNNgmxFcjVPXkSkarvkRL6wsJDPPvuMW2+9lf/973+0bduWDz74gL59+/Lcc88xePDg0oyz6juyG3b/BpigzXBDQrBYrMR+Fk/i0Vxq+7nz1sCWODqouJ2IiEhlcWqefPyB48YGIiIiZcqppE9Yv349s2fP5tNPP8XBwYEhQ4bw5ptv0qRJE3ubPn360K5du1INtMpbN9v2s0EXqB5mSAjv/r6LuG1puDg5MOPONlTzcDEkDhEREbk0V9evwYK1B5izch99W9emcZCxy9iKiEjZKHGPfLt27di5cyfTp0/n0KFDTJ48uVgSDxAWFsbAgQNLLcgqrzAPNsyz3TZoybllOw4zZckOAF7ufRVX1fI1JA4RERG5dLdGhtCxUQD5RRYe+XQ9eYVmo0MSEZEyUOJEfs+ePSxevJh+/frh7Ox81jaenp7Mnj37soO7YiR8DSeOgk9taBRT7qc/cDSXxxZswGqFO6Lq0r9tnXKPQURERC6fg4OJN/pF4u/lyo7UbF76PsHokEREpAyUOJFPS0tj9erVZ+xfvXo1f//9d6kEdcVZe7LIXZth4FC+a7XnFZp5cN46jucWElnbl7G3RJTr+UVERKR0BXi7MqV/JADzVify0+ZkgyMSEZHSVuJE/uGHH+bAgQNn7D906BAPP/xwqQR1RUneBAfXgIMTtB5S7qcf+80W/jmUSXVPF967sw2uTuX7RYKIiIiUvo6NAri/U30AnvlyEwePaTk6EZEykbYVju0v99OWOJFPSEigdevWZ+xv1aoVCQkavlVip5aca9ITvMt3vfYFaxJZ+PcBHEwwbVAralVzL9fzi4iISNl5smtjIutUIzOviMcXxFNkthgdkohI1ZIUD7N7wMe3QmZSuZ66xIm8q6srqampZ+xPTk7GyanERfCvbHmZsOlz2+125VvkbuOB44z5ZgsAT8Y05toG/uV6fhERESlbzo4OTBvYCi9XJ/7ef4y343YaHZKISNVxYA3MudVW68yjBjiXb6doiRP5rl27MmrUKDIyMuz7jh8/znPPPcdNN91UqsFVeZsWQmEO+DeC0OvL7bRHcwp4aN56CswWukYE8mCn8HI7t4iIVD3vvvsuoaGhuLm5ERUVxZo1a87ZdsuWLfTt25fQ0FBMJhNTp049o824ceMwmUzFtv+ukJOXl8fDDz9MjRo18PLyom/fvmftaLjS1a3hwSt9rgJg2u+7WLX7iMERiYhUAXuXw8e9IT8D6l4Dd30N7n7lGkKJE/nJkydz4MAB6tWrx4033siNN95IWFgYKSkpvPHGG2URY9VktZ4uctf2HjCZyuW0ZouVxxZs4NDxE4T5ezK5fySmcjq3iIhUPQsXLiQ2NpaxY8eyfv16IiMjiYmJIS0t7aztc3NzqV+/PhMnTiQoKOicx23WrBnJycn27c8//yz2+BNPPMF3333H559/zrJly0hKSuK2224r1ddWVfRqWYt+bWpjtcLjCzdwNKfA6JBERCqvXUtg3u22Dtn6N8KdX4KbT7mHUeJEvlatWmzatIlJkyYRERFBmzZteOutt9i8eTN16mjZsouWuAoObwVnD4gcWG6nffPXHSzfmY67syMz7myDj9vZlxAUERG5GFOmTGHEiBEMHz6ciIgIZsyYgYeHB7NmzTpr+3bt2vH6668zcOBAXF1dz3lcJycngoKC7Ju//+kpYBkZGXz44YdMmTKFzp0706ZNG2bPns3KlSv566+/Sv01VgUv9mpG/QBPUjPzefqLjVitVqNDEhGpfLb9AJ8OgqI8aNQNBi0AFw9DQrmkSe2enp7cd999pR3LleVUb/xVfcG9WrmcMm5rKu/8vguAiX2b0zjIu1zOKyIiVVNBQQHr1q1j1KhR9n0ODg5ER0ezatWqyzr2zp07CQkJwc3NjQ4dOjBhwgTq1q0LwLp16ygsLCQ6OtrevkmTJtStW5dVq1Zx9dVXn3G8/Px88vPz7fczMzMvK74zWMxgtYBjxfyC3MPFiWmDWtHn3ZUs2ZrGRyv3MfzaMKPDEhGpPP75EhbdB5YiiOgNt80EJxfDwilxj/wpCQkJLF68mG+//bbYJhch+zAkfGO7XY5F7v5v2R4Ahl0TSq+WtcrtvCIiUjWlp6djNpsJDCy+6kpgYCApKSmXfNyoqCg++ugjFi9ezPTp09m7dy/XX389WVlZAKSkpODi4kK1atUu+rwTJkzA19fXvpXqKMIdP8N7HeDv2aV3zDLQLMSX53rYag1M+HEbW5IyLvAMEREBIH4+fHmvLYlvMRD6fmhoEg+X0CO/Z88e+vTpw+bNmzGZTPahWafmWZvN5tKNsCraMBcshRDSGkJalcsprVYrO9JsF0D92tYul3OKiEjFdeDAAUwmE7Vr2z4T1qxZw/z584mIiDB81F337t3tt1u0aEFUVBT16tXjs88+4557Lu0L8FGjRhEbG2u/n5mZWXrJfMZBSN8Oy9+A1neVe+Xikhh6TSh/7jrCkq2pPPLpBr5/5Do8XLTqkIjIOa39EH44+fnRZhjc/CY4XHJ/eKkpcQSPPfYYYWFhpKWl4eHhwZYtW/jjjz9o27YtS5cuLYMQqxiLGdad/Ma+HHvjj+YUcDy3EJMJ6vt7ldt5RUSkYrrjjjv4/fffAVsP90033cSaNWt4/vnnGT9+/EUdw9/fH0dHxzOqxaempp63kF1JVatWjUaNGrFrl216WFBQEAUFBRw/fvyiz+vq6oqPj0+xrdS0uguq1YXslNNT5yook8nE67e3IMjHjT2Hcxh7cilaERE5i1Xvnk7iox6AnlMrRBIPl5DIr1q1ivHjx+Pv74+DgwMODg5cd911TJgwgUcffbQsYqxadsXB8URw84Vm5Vddd/fhHABqVXPH3cWx3M4rIiIV0z///EP79u0B+Oyzz7jqqqtYuXIl8+bN46OPPrqoY7i4uNCmTRvi4uLs+ywWC3FxcXTo0KHUYs3Ozmb37t0EBwcD0KZNG5ydnYudd/v27SQmJpbqeS+akwt0esZ2+88pkJ9d/jGUgJ+nC28OaInJBJ+vO8g38YeMDklEpOL543X4+Tnb7etiodvEcltp7GKUOJE3m814e9uKpPn7+5OUlARAvXr12L59e+lGVxWt/cD2s+Xgcq1wuPuw7aIiPEC98SIiAoWFhfaq8UuWLOHWW28FbEXjkpOTL/o4sbGxzJw5kzlz5rB161YefPBBcnJyGD58OABDhgwpVgyvoKCA+Ph44uPjKSgo4NChQ8THx9t72wGefPJJli1bxr59+1i5ciV9+vTB0dGRQYMGAeDr68s999xDbGwsv//+O+vWrWP48OF06NDhrIXuykWLgVA9HHKPwJr/MyaGEugQXoNHbmwAwPNf/cP+IzkGRyQiUkFYrRA3Hn572Xb/xtEQPbZCJfFwCYn8VVddxcaNGwFbMZpJkyaxYsUKxo8fT/369Us9wCrl2H7Y+Yvtdtu7y/XUu9KUyIuIyGnNmjVjxowZLF++nF9//ZVu3boBkJSURI0aNS76OAMGDGDy5MmMGTOGli1bEh8fz+LFi+0F8BITE4t9MZCUlESrVq1o1aoVycnJTJ48mVatWnHvvffa2xw8eJBBgwbRuHFj+vfvT40aNfjrr78ICAiwt3nzzTfp2bMnffv2pWPHjgQFBbFo0aLLfVsunaMT3PCs7faKtyGv4heSe7RLQ9qF+pGdX8Sjn26goMhidEgiIsayWm298MvfsN3v+jJ0esrYmM7BZC3hQqI///wzOTk53HbbbezatYuePXuyY8cOatSowcKFC+ncuXNZxVpuMjMz8fX1JSMjo3Tn0C150TbkLqwTDC3fCv/DZq9h6fbDvNqnOXdE1S3Xc4uIyOUr7c+mpUuX0qdPHzIzMxk6dKh93ffnnnuObdu2GZsUl4My+ay3mGH6NXB4G3R6Fm4cdeHnGOzQ8RP0eGs5GScKub9TfUZ1b2p0SCIixrBYbPPhT9Uz6zEZ2o8o1xBK8tlU4jKlMTEx9tsNGjRg27ZtHD16FD8/P3vlejmLogJbtXoo1yJ3p5weWu9Z7ucWEZGK54YbbiA9PZ3MzEz8/Pzs+++77z48PMpv6leV4uAIN4yCz4faCiRF3Q8e1Y2O6rxqVXPntb4teOCTdfzfsj1cG+5Px0YBF36iiEhVYi6Cb0fCxk/B5AC3vgOtBhsd1XmVaGh9YWEhTk5O/PPPP8X2V69eXUn8hWz9FnIOg1cQNO5RrqfOKzRz8NgJAMJrami9iIjAiRMnyM/Ptyfx+/fvZ+rUqWzfvp2aNWsaHF0l1vRWCGoOBVmw8m2jo7ko3a4K4s6rbaP1Yj/byOGsfIMjEhEpR+ZCWHTvySTeEW6bWeGTeChhIu/s7EzdunW1Vvyl+Ns2ZJE2Q8HRuVxPvTc9B6sVfN2dqeHpUq7nFhGRiqlXr158/PHHABw/fpyoqCjeeOMNevfuzfTp0w2OrhJzcIAbn7fdXv1/kJ1mbDwXafTNETQO9CY9O5/Yz+KxWEo081JEpHIqzIOFd8GWr8DRBfp/DM1vNzqqi1LiYnfPP/88zz33HEePHi2LeKqmtK2wf4XtG57WQ8v99KcK3TWo6aWREyIiAsD69eu5/vrrAfjiiy8IDAxk//79fPzxx7z9duXoSa6wGnWDWm2gMBf+nGp0NBfFzdmRaXe0ws3ZgeU705m5fI/RIYmIlK2CXFgwCHb8BE5uMPBTaNrT6KguWokT+XfeeYc//viDkJAQGjduTOvWrYttchaneuMbdwffWuV+es2PFxGR/8rNzbUvJ/vLL79w22234eDgwNVXX83+/fsNjq6SM5lO98qv/QAyk4yN5yI1CvRmTM9mALz+83Y2HjhubEAiImUlPwvm3Q67fwNnTxj8OTSMNjqqEilxsbvevXuXQRhVWH42bFxgu13OS86dsvuwbW1YLT0nIiKnNGjQgK+//po+ffrw888/88QTTwCQlpZWuiu2XKnCO0PdayBxpW0Zo5vfMDqiizKofR1W7Ernh83JPPLpBn549Dq83cp3SqCISJk6cRw+6QuH/gZXHxj8BdSNMjqqEitxIj927NiyiKPq+ucLyM8EvzCof6MhIezWGvIiIvIfY8aM4Y477uCJJ56gc+fOdOjQAbD1zrdq1crg6KoAkwk6Pw8f3Qzr5sC1j0G1ir/8q8lk4tXbmhN/4DiJR3MZ/fU/TB3QUlPzRKRqyDkCc3tDyiZw94O7voKQyvmZV+Kh9VJCaz+0/Wx7t60ATjmzWKzsST+ZyKtivYiInHT77beTmJjI33//zc8//2zf36VLF958800DI6tCQq+DsE5gKYRlk4yO5qL5ujvz9qCWODqY+CY+iS/WHTQ6JBGRy5eVYvtyNWUTeAbAsB8qbRIPl5DIOzg44OjoeM5N/uO296H9/dDqTkNOf+j4CfIKLbg4OlDHz92QGEREpGIKCgqiVatWJCUlcfCgLVlr3749TZo0MTiyKqTzaNvP+PlwZLexsZRAm3rVeSK6IQBjvtlir7cjIlIpZRyE2T3g8FbwDoHhP0FgM6OjuiwlTuS/+uorFi1aZN8WLlzIs88+S3BwMO+//35ZxFi51WwKPSaBR3VDTn/qgzfU3wMnRw3AEBERG4vFwvjx4/H19aVevXrUq1ePatWq8dJLL2GxWIwOr+qo0x4adgWrGZa9ZnQ0JfLgDQ3oUL8GJwrNPDJ/A/lFWn5YRCqho3thdnc4uts2xWn4j+Df0OioLluJM7tevXoV226//XZeeeUVJk2axLfffntJQbz77ruEhobi5uZGVFQUa9asOWfbwsJCxo8fT3h4OG5ubkRGRrJ48eJztp84cSImk4nHH3/8kmKr7FToTkREzub555/nnXfeYeLEiWzYsIENGzbw6quvMm3aNF544QWjw6tabnzO9nPTZ5C2zdhYSsDRwcTUgS2p7ulCQnImE3+qPLGLiABweIetJ/54IlQPt/XEVw8zOqpSUWpdtFdffTVxcXElft7ChQuJjY1l7NixrF+/nsjISGJiYkhLSztr+9GjR/N///d/TJs2jYSEBB544AH69OnDhg0bzmi7du1a/u///o8WLVqUOK6qYpcK3YmIyFnMmTOHDz74gAcffJAWLVrQokULHnroIWbOnMlHH31kdHhVS0graNITsMLSCUZHUyKBPm5M7me7jpq9Yh9xW1MNjkhE5CKlboGPekBWEgQ0tSXxvrWNjqrUlEoif+LECd5++21q1Sr5GulTpkxhxIgRDB8+nIiICGbMmIGHhwezZs06a/u5c+fy3HPP0aNHD+rXr8+DDz5Ijx49eOON4su6ZGdnM3jwYGbOnImfn98lva6q4NTQ+gYqdCciIv9y9OjRs86Fb9KkCUePHjUgoiruxucAEyR8DSmbjY6mRDo3CeTua209WE9+vpGUjDyDIxIRuYCkDbbCdjmHIaiFrbCdd6DRUZWqEifyfn5+VK9e3b75+fnh7e3NrFmzeP3110t0rIKCAtatW0d0dPTpgBwciI6OZtWqVWd9Tn5+Pm5ubsX2ubu78+effxbb9/DDD3PzzTcXO/a55Ofnk5mZWWyrKvYcVo+8iIicKTIyknfeeeeM/e+8884VPZKtzAQ2g6tus93+/VVjY7kEz3RvTLMQH47lFvL4wg2YLVajQxIRObvE1TDnVjhxDGq3g6HfgWcNo6MqdSVeR/7NN98stpaog4MDAQEBREVFlbjnOz09HbPZTGBg8W9HAgMD2bbt7POwYmJimDJlCh07diQ8PJy4uDgWLVqE2Xy6AMuCBQtYv349a9euvag4JkyYwIsvvlii2CuD47kFpGcXAFA/wNPgaEREpCKZNGkSN998M0uWLLGvIb9q1SoOHDjAjz/+aHB0VdQNo2DLV7D9Rzi0Dmq1MTqii+bq5Mi0Qa3oOe1P/tpzlOlLdzGyc+UvFiUiVczeP2D+QCjMgXrXwR0LwNXb6KjKRIl75IcNG8bQoUPt21133UW3bt3Kbfj6W2+9RcOGDWnSpAkuLi6MHDmS4cOH43ByjfYDBw7w2GOPMW/evDN67s9l1KhRZGRk2LcDBw6U5UsoN6cK3QX7uuHpWuLvbEREpArr1KkTO3bsoE+fPhw/fpzjx49z2223sWXLFubOnWt0eFWTf0NoMdB2+7dXjI3lEtQP8GJ8r6sAeHPJTtbt1xQMEalAdi6Bef1sSXx4Zxj8eZVN4uESEvnZs2fz+eefn7H/888/Z86cOSU6lr+/P46OjqSmFi+ckpqaSlBQ0FmfExAQwNdff01OTg779+9n27ZteHl5Ub9+fQDWrVtHWloarVu3xsnJCScnJ5YtW8bbb7+Nk5NTsZ77U1xdXfHx8Sm2VQW70zQ/XkREzi0kJIRXXnmFL7/8ki+//JKXX36ZY8eO8eGHHxodWtXV6WlwcILdcbD/7NMIK7K+rWvRu2UIZouVRz+NJyO30OiQRERg6/fw6UAoyoPGPWDQAnDxMDqqMlXiRH7ChAn4+/ufsb9mzZq8+mrJ5ny5uLjQpk2bYtXuLRYLcXFx9mF+5+Lm5katWrUoKiriyy+/pFevXgB06dKFzZs3Ex8fb9/atm3L4MGDiY+Px9HRsUQxVma7NT9eRESkYqkeBq3utN3+vfL1yptMJl7u05x6NTw4dPwEzy7ahNWq+fIiYqDNX8BnQ8BSCM36QP+PwcnV6KjKXIkT+cTERMLCzlx7r169eiQmJpY4gNjYWGbOnMmcOXPYunUrDz74IDk5OQwfPhyAIUOGMGrUKHv71atXs2jRIvbs2cPy5cvp1q0bFouFp59+GgBvb2+uuuqqYpunpyc1atTgqquuKnF8ldnpRF7z40VERCqMjk+BowvsWw57lhkdTYl5uToxbVArnB1N/PRPCvPXlPz6T0SkVGz4BL68F6xmiLwD+n4Ijs5GR1UuSpzI16xZk02bNp2xf+PGjdSoUfJqgAMGDGDy5MmMGTOGli1bEh8fz+LFi+0F8BITE0lOTra3z8vLY/To0URERNCnTx9q1arFn3/+SbVq1Up87qru1Bx59ciLiIhUIL61oY2tw4LfXoZK2KPdonY1no6xLV84/rsEdqRmGRyRiFxx1syEbx4GrND2buj1LjhcOaOvS1wBbdCgQTz66KN4e3vTsWNHAJYtW8Zjjz3GwIEDLymIkSNHMnLkyLM+tnTp0mL3O3XqREJCQomO/99jXAnyi8zsP3IykdcceREROem222477+PHjx8vn0CudNfHwvo5cHAN7FoCDW8yOqISu+e6MJbvSuePHYcZOX893468DjfnK+ciWkQMtPId+OV52+2rH4aYV+BfK6tdCUrcI//SSy8RFRVFly5dcHd3x93dna5du9K5c+cSz5GXsrP/SC4WK3i7OlHTu+rPERERkYvj6+t73q1evXoMGTLE6DCrPu8gaD/CdruS9so7OJh4o18k/l6u7EjN5qXvS9bRIiJySVa9ezqJv/7JKzKJh0vokXdxcWHhwoW8/PLLxMfH4+7uTvPmzalXr15ZxCeX6FTF+vo1vTBdgb/YIiJydrNnzzY6BDnl2sdh7SxIjodtP0DTnkZHVGIB3q68OSCSuz5cw7zViVzXwJ/uzYONDktEqqrV78PPz9lu3zAKbnjW2HgMVOIe+VMaNmxIv3796Nmzp5L4CkiF7kRERCo4T3+4+kHb7d9fAYvF2Hgu0fUNA3igUzgAz3y5iYPHcg2OSESqpL9nwU9P2W5f/yR0esbYeAxW4kS+b9++vPbaa2fsnzRpEv369SuVoOTy7UrT0nMiIiIV3jUjwdUX0hIg4Sujo7lk/+vaiJZ1qpGZV8TjC+IpMlfOLyVEpIJaPxe+f8J2+5pHofPoK3I4/b+VOJH/448/6NGjxxn7u3fvzh9//FEqQcnlO1WxvoEK3YmIiFRc7n62ZB7g9wlgLjI2nkvk7OjAtEGt8HZ14u/9x3g7bqfRIYlIVbFxAXz7iO321Q/BTeOv+CQeLiGRz87OxsXF5Yz9zs7OZGZmlkpQcnmsVuu/htYrkRcREanQoh6wJfRHdsLmz42O5pLVqe7BK7c1B2Da77tYtfuIwRGJSKW3+Qv4+kHACu3uhZhXlcSfVOJEvnnz5ixcuPCM/QsWLCAiIqJUgpLLk5KZR26BGScHE/VqeBgdjoiIiJyPm4+t8B3AsolgLjQ0nMtxa2QI/dvWxmqFxxdu4GhOgdEhiUhllfANLLoPrBZoPRS6v64k/l9KXLX+hRde4LbbbmP37t107twZgLi4OObPn88XX3xR6gFKye1Osw2rr1vDA2fHS65nKCIiIuWl/QhY9Q4c2wfx86DNMKMjumTjbm3Guv3H2H04h6e/2MjMIW21go6IlMy2H+GLu8FqhpaDoedUcFBe828lfjduueUWvv76a3bt2sVDDz3E//73Pw4dOsRvv/1GgwYNyiJGKaFdaVkANNCwehERkcrBxROui7XdXvY6FOUbG89l8HBxYtqg1rg4ObBkaxpvLtmJ1Wo1OiwRqSx2/AKfDQFLETTvD7dOUxJ/Fpf0jtx8882sWLGCnJwc9uzZQ//+/XnyySeJjIws7fjkEpwqdBeuQnciIiKVR9u7wTsEMg/CujlGR3NZIkJ8GNPTNuXy7bidvPrjViXzInJhu+Jg4Z1gKYRmfaD3dHBwNDqqCumSv9r4448/GDp0KCEhIbzxxht07tyZv/76qzRjk0ukQnciIiKVkLMbdPyf7fbyyVBQuddjv/PqevZkfubyvTz31WbMFiXzInIOe5bBgjvAnA9NesJtM8GxxDPBrxglSuRTUlKYOHEiDRs2pF+/fvj4+JCfn8/XX3/NxIkTadeuXVnFKSVwOpH3NDgSERERKZFWQ8C3LmSnwt8fGh3NZbv7ujAm9W2Bgwk+XXOAxxZsoFBrzIvIf+1bAZ8OhKI8aNQdbp8Njs5GR1WhXXQif8stt9C4cWM2bdrE1KlTSUpKYtq0aWUZm1yCzLxCUjNt8+o0tF5ERKSScXKBTk/bbv/5JuRnGxtPKejfrg7TBrXG2dHE95uSuX/uOvIKzUaHJSIVReJqmNcPCnOhQTT0n2P7WyjnddGJ/E8//cQ999zDiy++yM0334yjo+YqVER7Ts6Pr+ntio+bvsUSERGpdCIHQfX6kHsEVs8wOppScXOLYGYOaYubswO/bUtj6Kw1ZOVV3mX2RKSUHFwH826Hwpz/b+/O46Kq9z+Ov2bYQcAFAUEUt9wFcyG01JJCbbOsrGtpZoulltGtq2Xacm90s8xMU9u9mWmWmm2W4l64oZjmvivKpgmKss78/jiK8RPLBTgz8H4+Huchc+bMmc85IF8+8/1+P19o2A36TgdXD7OjcgoXncivXLmSEydO0K5dO6Kiopg4cSKZmZnlGZtcht3pmh8vIiLi1FxcodtI4+tfJ8Dp46aGU1a6NQ3kfw9FUc3DldV7j3H/h6v5Q+vMi1Rdh5Nh+h2Qlw3h18G9X4Cbl9lROY2LTuSvueYaPvjgA44cOcJjjz3GzJkzCQkJwWazsXDhQk6cOFGeccpFKp4fH6j58SIiIk6rVR+o3Qxys2DVe2ZHU2Y6NqjJF49cQw1vNzYeyqLv+4mkZ+eaHZaIVLTUTfBZb+N3XNg1cN9McPc2OyqncslV6318fHjooYdYuXIlmzZt4plnnuH1118nMDCQ2267rTxilEuwSz3yIiIizs/qcq5XPvE9OHXM3HjKUOu6/nz5WDRBfh7sSDvJXVMSOXjMuSv0i8glSN8K/7sdTv8BdTtAv9ngodzlUl328nMATZs25Y033uDQoUN88cUXZRWTXIGzPfKNVehORETEuTW/DYJaQ/4J+OUds6MpU02CfPlqcCfq1fTmwLFT3D0lkV3pGt0pUull7IBptxk1QELaQr+vwNPP7Kic0hUl8me5uLjQu3dv5s+fXxank8tUUGRj/1HjE231yIuIiDg5qxVueMH4es37cDLd3HjKWFhNb2YPjqZJYDVSs3O5Z+oqNqdkmR2WiJSXo7th2q2Qkw7BreH+OeBV3eyonFaZJPLiGA4cO0WhzY63uwvBfp5mhyMiIiJX6qoeENrOWJZp5dtmR1Pmgvw8mfVYNG3q+nMsJ5/73l/F2n2VZxqBiJxxbK+RxJ9MhcAW8MA34F3T7KicmhL5SuRsxfqGtX2wWi0mRyMiIiJXzGKB68/0yq/9CLIPmxtPOajp487nD0fRsUFNTuQV8sBHq1m2I8PssESkrBw/YAynz06BgKbQfz741DI7KqenRL4S2XV2fryG1YuISAWaNGkS4eHheHp6EhUVxZo1ay547O+//06fPn0IDw/HYrEwfvz4846Jj4+nQ4cO+Pr6EhgYSO/evdm+fXuJY7p164bFYimxDR48uKwvzTE0ugHqRUNRHix/0+xoyoWvpxv/e6gj1zetTW6BjYenreXHTUfMDktErlRWitETn3UAajWGAfOhWm2zo6oUlMhXIrvTcwDNjxcRkYoza9Ys4uLiGDNmDOvXryciIoLY2FjS00ufz33q1CkaNmzI66+/TnBwcKnHLFu2jCFDhrBq1SoWLlxIQUEBN910Ezk5OSWOe+SRRzhy5Ejx9sYbb5T59TkEiwVuGGV8vf5/8Md+c+MpJ55uLkx9oD03t6lDQZGdITPWM3vdQbPDEpHLlX3ESOL/2Ac1GsCAb8G39N/7cumUyFci59aQVyIvIiIVY9y4cTzyyCMMHDiQFi1aMGXKFLy9vfn4449LPb5Dhw6MHTuWe++9Fw8Pj1KPWbBgAQ8++CAtW7YkIiKCTz/9lAMHDpCUlFTiOG9vb4KDg4s3P79KXPk4/Fpo0BVsBbC8kn5gAbi7Wplwb1vu7RCGzQ7PfvUbn/yy1+ywRORSnUyH/90Gx3ZD9XpGEu8XYnZUlYoS+UrCbrefS+TVIy8iIhUgPz+fpKQkYmJiivdZrVZiYmJITEwss/fJyjIqmdesWbIw0ueff05AQACtWrVi5MiRnDpVydciP9srn/yFUf25knKxWoi/szUPX9sAgJe/3cKEhJ3Y7XaTIxORi5KTacyJz9wBfnWNJL56mNlRVTquZgcgZSPjRB4ncguxWiA8wNvscEREpArIzMykqKiIoKCgEvuDgoLYtm1bmbyHzWZj+PDhdO7cmVatWhXv/8c//kH9+vUJCQnht99+41//+hfbt29nzpw5pZ4nLy+PvLy84sfZ2dllEl+FCusITW6CnT/D0tehzwdmR1RuLBYLL9zcHF9PN95etINxC3dwIreA53s1x2JRQV8Rh3XqGPyvN2RsBd86xpz4GuFmR1UpKZGvJM4WuqtX0xsPVxeToxERESkbQ4YMYfPmzaxcubLE/kcffbT469atW1OnTh26d+/O7t27adSo0XnniY+P5+WXXy73eMvd9c8bifym2XDdMxDYzOyIyo3FYuGpmCb4erryyndb+GDFXk7mFfLv3q1x0eo8Io7n9HH4rDekbQKfQKMnvtb5v4+lbGhofSWxO0OF7kREpGIFBATg4uJCWlpaif1paWkXLGR3KYYOHcp3333HkiVLqFu37l8eGxUVBcCuXbtKfX7kyJFkZWUVbwcPOmkRtZC20OwWwA5L482OpkI8dG0D3ujTBqsFvlhzkKdmbqCgyGZ2WCLyZ7nZMP1OOLIRvAOMJD6gidlRVWpK5CuJs2vIq9CdiIhUFHd3d9q1a0dCQkLxPpvNRkJCAtHR0Zd9XrvdztChQ5k7dy6LFy+mQYMGf/ua5ORkAOrUqVPq8x4eHvj5+ZXYnNb1zwMW2DIPjvxmdjQV4p4OYbx739W4uVj47rcjPPZZErkFRWaHdekK8+GnF+DNprDkNcg7aXZEIlcu7yR8fhekJIFXDej/TaUeLeQolMhXEucK3fmYHImIiFQlcXFxfPDBB0ybNo2tW7fy+OOPk5OTw8CBAwHo378/I0eOLD4+Pz+f5ORkkpOTyc/PJyUlheTk5BI96UOGDGH69OnMmDEDX19fUlNTSU1N5fTp0wDs3r2bV199laSkJPbt28f8+fPp378/Xbp0oU2bNhV7A8wQ1BJa3Wl8XUV65QFublOHD/q3x9PNyuJt6Qz4eA0ncgvMDuviZR2CT3tB4kQ4mQrL/gsT2kLSp1BUaHZ0IpcnPwdm3AMHV4Onv5HEB7f6+9fJFVMiX0mc7ZFvrB55ERGpQH379uXNN99k9OjRREZGkpyczIIFC4oL4B04cIAjR44UH3/48GHatm1L27ZtOXLkCG+++SZt27bl4YcfLj5m8uTJZGVl0a1bN+rUqVO8zZo1CzBGAixatIibbrqJZs2a8cwzz9CnTx++/fbbir14M3UbCRYrbP8BDiX9/fGVRLemgfzvoSiqebiyeu8x7v9wNX/k5Jsd1t/blQBTu8ChtUayc8OLxrraOenw7VMwpTPs+BlUmd8xFOQaK0PYNIXjLxWchi/uhf2/gIcfPDAX6kSYHVWVYbFrLY/zZGdn4+/vT1ZWllMMvcvJK6TlmJ8A2PDijdTwcTc5IhERKWvO1jY5ukpxP+c+DhtnQKPu8EDp1forq02Hsuj/8Wr+OFXAVUHVmD4oikA/T7PDOp/NBsvfMFYZwG4kOff8z6jiXZgP6z4yeuZP/2Ec36Ar3PSqkiEz7V0BXz9sjJrwCYTGMdC4OzS6Abxr/v3rq4qCXJh5H+xeDO7VjCQ+rKPZUTm9S2mb1CNfCezNNArd1fJxVxIvIiJSVXR9DqyusDsB9ieaHU2Fal3Xny8fiybIz4MdaSe5a0oiB4+dMjusknKOGvOGl8YDdmj3IDz087mluFzd4ZrH4clk6PQkuLjD3mUwtSvMHWwMxZeKYysy6hZMu9VI4sEYMbFxBnw9CN5oCB90hyXxcHCtcXxVVZgHX/Y3kng3b+g3W0m8CZTIVwLn5sdrWL2IiEiVUbMBtL3f+Hrxv6vcsOwmQb58NbgT9Wp6c+DYKe6eksiu9BNmh2U4tM4YSr87AVy9oPcUuPUdcCtl1IBXdaMXfug6aHUXYIeNX8C77WDRy0Y1cClf2Ydh2m3G6Ajs0PYB+Nd+o/J656cgqJWxP2UdLHsdPoqBsY3gq4cgeQacSPu7d6g8igpg9kDY+ZPxs/2PWVC/k9lRVUkaWl8KZxtu99bP23l38S7u61iP+Dtbmx2OiIiUA2drmxxdpbmfWYeMgmlF+UaRqYbdzI6owqVl53L/h6vZmX6Smj7u/O+hjrQK9TcnGLsd1rxvVKa3FUDNRtD3M6NA4cVKSYKfXzTmHYOxlFe3EUaPvotbuYRdpe34GeYNhlNHjSHit4yHNneff1z2YaPWwa5FsGcJ5GaVfD649Zlh+DEQFlU5v1dFhfD1Q7DlG3DxgH/MNKYcSJm5lLZJiXwpnK1xf+LzJH7YlMqom5vz8HUNzQ5HRETKgbO1TY6uUt3PH56DNVOhbkcY9DNYLGZHVOGO5eTz4Cdr+O1QFr4ernw8sAMdwit4PnPeSfj2Sdj8tfG4xe1w20TwvIyfL7vdKGS4cAwc3Wnsq9UYbnwFmvaqkt/jMleYD4tfgV/fNR4Ht4G7P4Vajf7+tUWFxgcuuxbBroVweEPJ5919oWHXc4l99bAyD7/c2e1wMg0yd0DmTmM7tMa4bhd3uHcGNLnR7CgrHSXyV8jZGvfYt5ezPe0EnwzswPVNA80OR0REyoGztU2OrlLdzxOp8E4EFObCP2bDVTeZHZEpTuQWMGjaOtbsPYanm5WpD7Sn61W1K+bN07fBlw8YSY/VFW581Zj/fqUJd1GBsTzd0nijxxigfmdjKH5ouysOu8r6Y58xLD7lzIoPHR8z7qmrx+Wd72SG0Uu/a5HRa38qs+TzAU3PFc2r37n0KRZmKcwzKvQf3Xkmad9l/Ht0F+SVMq3D6gr3fAbNelV8rFWAEvkr5EyNe5HNTvMXF5BfZGPFc9cTVtPb7JBERKQcOFPb5Awq3f38eZTRs1gnAh5dVmV7bHMLinh8ehJLtmfg5mJhwr1t6dm6Tvm+6W+zjZ74glPgG2L06taLKtv3yM2CleNh1XvGBzZgzKfvPhpq1C/b96rsfp8H85+EvCxjKcDb34Pmt5Td+W02SN0IOxcZif2hNWD/0zJ2rl7Q4LpzvfU1G5b//1e7HXIyzvSsn0nSz/a0H99fMr4/s1ihen0IaAIBVxn/1r8WAhqXb7xVmBL5K+RMjfu+zBy6vbkUD1crW17pgYu1ajbcIiKVnTO1Tc6g0t3PnEwY3wYKcqDvdGh+q9kRmSa/0MbTXybz/W9HsFrgv33acHf7chjaXJgHPz0Paz80HjfoCn0+gmrlOAog65BR2HDjTMBuDHGOGgzXPWMUzZMLK8g1vl/rPjIeh0VBnw+her3yfd/Tf8CeZed6608cLvl8jfAzSf2NEH4teFxB8erCfPhjb8nh8Gd72v//nP4/8/AzkvRaTc4k7WcS95oNL3+UglwWJfJXyJka94StaQyato7mdfz48anrzA5HRETKiTO1Tc6gUt7PhFdgxVsQ2AIG/wLWqrs4UZHNzgtzNzFz7UEAxtzagoGdG5TdGxw/AF8OgMPrjcddnjMK0lldyu49/sqRjcYojL3LjcdeNaDrv6D9IGNZOykpYwd8NRDSNhuPr42D65+v+IJ0djukbzmT1C8ylo20FZx73sUd6kWf660PbF56b33O0TM96/9vOPwf+8B+oWXxLMaHFmeT9FqNz/SyXwXVAqvsKB5Ho0T+CjlT4/7+8t289sM2bmlTh4n/uNrscEREpJw4U9vkDCrl/Tz9B4yPMIYMR95vDBeu38kYPlwF2e12/vP9Vj5cuReAuBuvYtgNjbFcacKy42eY8wjkHjcS6Ds/MKfol90OOxfCwhchY5uxr0YDiHnJKLSnxMyQ/AV8/4wxWsWnNtwx1Zir7gjyTsK+FUZSv3OhMcz9z3xDjFhrNoRju88NjT/9x4XP6V7tT0n6/+tdd/Mq3+uRK3YpbZNrBcX0lyZNmsTYsWNJTU0lIiKCd999l44dO5Z6bEFBAfHx8UybNo2UlBSaNm3Kf//7X3r06FF8zOTJk5k8eTL79u0DoGXLlowePZqePXtWxOVUqN3pOYDWkBcREanyvGrAtcMh4WVInm5sFqsxbz78OmjQBepdAx6+ZkdaISwWCy/c3BxfTzfeXrSDcQt3kPLHaV6+vSWebpfRc24rMorOLR9rPA65Gu6ZVv5Dsy/EYjEKGza6wfheL/6PMax69gBjBYPY/0BY6X9PVwl5J+GHf8LGL4zHDboYH7r4Bpsb1595VIOmPY3NbjeKzp3trd+3whiGv+Gz0l/rH1ZK73oT8K2jD3GqCNN75GfNmkX//v2ZMmUKUVFRjB8/ntmzZ7N9+3YCA8+vwP6vf/2L6dOn88EHH9CsWTN++ukn4uLi+PXXX2nbti0A3377LS4uLjRp0gS73c60adMYO3YsGzZsoGXLv1/H05k+pe8z+VeS9v/BhPvacltEiNnhiIhIOXGmtskZVNr7abfDtu+MRGDvCqMX788sLka18wbXGcl9WBS4V/5CuZ/8spdXv9uCzQ6tQv2Y3K/dpRUIPpkBXw+CvcuMxx0eMRJlR5o/nHcSfp1gFD0sOGXsa9EbYsYYvbFVSepmmP2gMfTcYoVuz8N1cRU39aEsFJyG/b8a8+pzMoxl8c7OY6/VuEr8v62KnGpofVRUFB06dGDixIkA2Gw2wsLCGDZsGCNGjDjv+JCQEF544QWGDBlSvK9Pnz54eXkxffr0C75PzZo1GTt2LIMGDfrbmJylcbfb7US+spCs0wX88OR1tAhx3FhFROTKOEvb5CyqzP3MPmwk9PuWG//+/6G7Vjeo2+FcYl+3g2MtjVWGVu7M5MmZGziWk4+fpyvj743khmZBf//CA6uNpPDEYXDzhlsnQJu7yz3ey5Z9BJb8BzZMB+zG97jjI9DlWfCuaXZ05ctuN4rZLXgeivKMoel3fWRMMRFxAk4ztD4/P5+kpCRGjhxZvM9qtRITE0NiYmKpr8nLy8PTs2QD4+XlxcqVK0s9vqioiNmzZ5OTk0N0dPQFz5mXl1f8ODu7lDUTHdCxnHyyThdgsUCDAB+zwxERERFH4xcCEX2NDYwibXtXGMN2966A7ENw4FdjW/ZfcPEwhmM36GIk9qHtKk3xtGubBPDdsGt54vP1JB88zkOfruPJGxrzVMxVpa/6Y7fDqsnGHHRboTF0+Z7PILBZxQd/KfzqwO0TjWr2C0fD7gRj2brkz+G6f0LHRyvnhzWnj8P8YbB1vvH4qh7G0nI+tUwNS6S8mJrIZ2ZmUlRURFBQyU9Dg4KC2LZtW6mviY2NZdy4cXTp0oVGjRqRkJDAnDlzKCoqWaFx06ZNREdHk5ubS7Vq1Zg7dy4tWrQo9Zzx8fG8/PLLZXNRFWh3hjE/PrS6F17uTjRUSERERMxRvR607Wdsdjsc23Muqd+3Ak6mGf/uW2Ec7+ZtDL9vcB2Ed4GQtuDiECWWLktIdS++fCya/3y/hWmJ+5mweBcbDh7nnXvbUtPnTx9Y5GbD/KGw5Rvjcas+Rk/8lSwNVtGCW8EDc4yh2QtHGxXbF74Iaz+A7mOMa6osc6kPJcFXDxofVFnd4MaX4ZonKs/1iZTC6X4Tv/POOzzyyCM0a9YMi8VCo0aNGDhwIB9//HGJ45o2bUpycjJZWVl89dVXDBgwgGXLlpWazI8cOZK4uLjix9nZ2YSFlcN6o2Vsd8ZJQIXuRERE5DJYLMa821qNoN2DRmKfufPcMPx9K+FUJuxZYmxgVMSu3+lM8bzrILiNc807Btxdrbx8eyva1qvByDmbWLEzk1smrGBSv6tpW68GpP0Osx4w6gtY3SD2NWNourMmhY27Q8NuRtG3xf82kt2vB0HiJLjp3xDe2ewIL5/NBqsmwaKXjFET1evD3Z8YI0lEKjlTE/mAgABcXFxIS0srsT8tLY3g4NIrStauXZt58+aRm5vL0aNHCQkJYcSIETRsWLKIh7u7O40bNwagXbt2rF27lnfeeYepU6eed04PDw88PByoWMlF2pVuJPKNA5XIi4iIyBWyWKD2VcbW4eEza15vPdNjv9xI7HOPw86fjQ2Mpe3qdz6X2Ae2dJr163u3DaV5HT8en57Enswc7pmayKdtd9Np23+wFJ4Gv7pGVfq67c0O9cpZXaDt/dDyDkh8D34ZD4fXw6e9jCHoLe+Ehl0dq6L738k5CvMGn/tZbHkH3PpOlV1uUaoeUxN5d3d32rVrR0JCAr179waMYncJCQkMHTr0L1/r6elJaGgoBQUFfP3119xzzz1/ebzNZisxD74yUI+8iIiIlBuLBYJaGFvUY0bvZ9rmc0Px9/8CuVmw/QdjA/CqafTwhncxEvvazRy6J7tpsC/fDO3MyFnr6LTzDTpvXgxAUcPuuNz1YeUrDufuA12fhXYDjKX0kj6FHQuMDYzvV8Nuxla/M3g6aCHIfSvh64fhxBFw9YQerxujShz4Z02krJk+tD4uLo4BAwbQvn17OnbsyPjx48nJyWHgwIEA9O/fn9DQUOLj4wFYvXo1KSkpREZGkpKSwksvvYTNZuO5554rPufIkSPp2bMn9erV48SJE8yYMYOlS5fy008/mXKN5eVcIq9CdyIiIlLOrFao08bYoodAUSGkbjw3v35/Ipw+Blu/NTYAn9oQ1MpIEGs3PfevAyXIvqdTeDf3X1hcN2KzWxhf2IefMx/gvRx3GlbWFb6qBcItbxsF8ZI/hz3L4MhGyNhmbKunnFuqsGE3o7e+bgfzl9uzFcHyN2HZ62C3GQUI7/4Ugv5+eWmRysb0RL5v375kZGQwevRoUlNTiYyMZMGCBcUF8A4cOID1T0O0cnNzGTVqFHv27KFatWr06tWLzz77jOrVqxcfk56eTv/+/Tly5Aj+/v60adOGn376iRtvvLGiL6/c5BYUceiP0wA00tB6ERERqWgurkaiF9oOrh0ORQVweMOZYfgrjGXbcjJKzrE/yyewZGJfu5mx+QRUbK/q9h9h7mNYcrPAuxbbO43ji2XVyEjP4baJv/Dm3W3o0apOxcVT0Wo3hRtfMb4+dcz4vu1ZaiT2x3bDoTXGtvwNo/Bh/U7QoKuR3Ae1qthpFNlHYM4j5woxRt4Pvd4wRhmIVEGmryPviJxhbdkth7PpNWEF/l5uJI++EYuGEomIVGrO0DY5E93PClCYB0d+O9fLm7Hd2LIOXPg1XjXP772v3cyYu12Wf+sUFcKSf8PKt43HdTsYPbv+dUnPzmXojA2s2XcMgMe6NOTZ2Ka4ujjH3P8yc/yAkdDvXWYk9zkZJZ/3rmUsU3h2KH6N8PKLZecimPuYUXzRzccYTXB2SUWRSuRS2iYl8qVwhsb9242HGfbFBtrVr8HXj3cyOxwRESlnztA2ORPdTxPlnYTMHWcS+23n/v1jH3CBP0s9/M8k9X/qva/dFPzrXnqCfyLNqNp+tmc36nGjV9r13PJzBUU2xv60nfeX7wGgY4OaTPxHWwJ9K+H66xfDbof0LUZiv2epUR8h/2TJY6rXP5fUN+haNuu3FxXA4lfhl3eMx8Gt4a5PIaDxlZ9bxAEpkb9CztC4j1+0g/GLdnJP+7q8cVeE2eGIiEg5c4a2yZnofjqggtPG8nfFCf6ZJP/YHrAXlf4a92rGPOn/34tfvX7pw773/QJfPQQnU43X3vYutLrzgiH9uOkIz371GyfzCgn09WBSv6vpEO448/tNU1QAKUlnhuEvhUNrjeXf/iy49ZmkvhvUj770IfB/7Dc+cDm01njc8VG48VVwq6IfpkiVcCltk+lz5OXy7M7IAVSxXkRERCoJN69zxfT+rDAPju4u2XufsR2O7jJ6hQ+vN7Y/c/WCgCYlE/zMHcY66vYiqN0c+n5mHPMXerauw1XBvjw+PYkdaSe59/1VjOzZjEHXNqja0xpd3KDeNcbWbYQxymL/r0ZSv3eZsbpB6iZj+/VdsLpBWJRRNK9hNwi52qixcCFb5sP8ocaqCJ7+cNtEaHFbRV2diFNQj3wpnOFT+p7vrGDrkWw+7N+emBZBZocjIiLlzBnaJmei+1kJFBUYvfX/P8HP3AlFf7HkcJu+xhzrS+ghPpVfyMg5m/gm+TAAvVoH88ZdEVTzUJ9YqU6mG0UPz/bYZx0s+byHn7G83dmh+LWbGlMkCnLh51Gw9gPjuLodoM9HUKN+xcYvYhL1yFdyNpudPWeWnmusivUiIiJSFbm4nZs3/2dFhXB8//8rsrcNcrOh0zBo/9Alz6v3dndlfN9I2tWvwavfbeGHTalsSz3B1Pvb0STItwwvqpKoFgit7zI2u934wOVs0by9y+H0H7DjR2MDqBZs9NanbzF68QE6PwU3vGh8n0XkPErknVDK8dPkFdpwd7FSt4aX2eGIiIiIOA4XV6jVyNia3Vxmp7VYLPSPDqdVqD9DPl/Pnowcbp/0C/F3tub2yNAye59Kx2I59/1o/xDYbJD627ne+gOJRs2C32YZx3sHwB1ToUmMmVGLODwl8k5o95ne+PAA76q3FIqIiIiIia6uV4Pvhl3LkzM38Muuozw1M5kNB47zfK/muLvq77K/ZbVCSKSxXTvcGE5/aI2R1BfmQfRQ8KtjbowiTkCJvBPalW4k8ip0JyIiIlLxalXz4H8PRfH2wh1MXLKLT3/dx2+HjjOp39XU8ddoyUvi5mmsR9+gi9mRiDgVfWzohM5WrNf8eBERERFzuFgt/DO2KR/2b4+vpyvrDxznlgkr+XVXptmhiUgVoETeCZ0dWq8eeRERERFzxbQI4vth19Gijh9Hc/K5/6PVvLd0FzabFoYSkfKjRN4J7VEiLyIiIuIw6tXyZs4Tnbi7XV1sdnhjwXYe/SyJrNMFZocmIpWUEnknc/xUPpkn8wFoWPvi1z8VERERkfLj6ebC2LsjeP3O1ri7Wlm0NY3bJq7k98NZZocmIpWQEnknc3ZYfR1/T3w8VKtQRERExJHc27EeXw/uRN0aXuw/eoo73/uVr5IOmR2WiFQySuSdzO50FboTERERcWSt6/rz3bBrub5pbfIKbfxz9kZGztlEbkGR2aGJSCWhRN7JqNCdiIiIiOOr7u3ORwM6EHfjVVgs8MWaA9w9JZGDx06ZHZqIVAJK5J3MuURe8+NFREREHJnVauHJ7k2YNrAjNbzd2JSSxa0TV7J0e7rZoYmIk1Mi72R2patHXkRERMSZdLmqNt8Ou5aIuv4cP1XAwE/XMuabzZzMKzQ7NBFxUkrknUheYREHzgzH0hx5EREREedRt4Y3Xw6O5v5r6mG3w7TE/dw4bhkLt6SZHZqIOCEl8k5k/9FT2Ozg6+FKbV8Ps8MRERERkUvg4erCv3u3ZvqgKOrV9OZIVi6P/G8dT3yeRHp2rtnhiYgTUSLvRHafGVbfMLAaFovF5GhERERE5HJc2ySAn4Z3YXDXRrhYLfywKZXu45YxY/UBbDa72eGJiBNQIu9Ezs2PV6E7EREREWfm5e7CiJ7N+HaoMXf+RG4hz8/dxL3vryr+m09E5EKUyDuRsxXrNT9eREREpHJoEeLHnCc68+ItLfB2d2HNvmP0emcF7yzaSX6hzezwRMRBKZF3IrszcgBVrBcRERGpTFysFgZd24Cfn+7C9U1rk19k4+1FO7h5wgrW7Ttmdngi4oCUyDsJu93+pzXklciLiIjjmDRpEuHh4Xh6ehIVFcWaNWsueOzvv/9Onz59CA8Px2KxMH78+Ms6Z25uLkOGDKFWrVpUq1aNPn36kJam6t/i3OrW8ObjBzvw7n1tCajmzs70k9w1JZFR8zaRnVtgdngi4kCUyDuJ1OxcTuUX4Wq1UL+Wt9nhiIiIADBr1izi4uIYM2YM69evJyIigtjYWNLT00s9/tSpUzRs2JDXX3+d4ODgyz7n008/zbfffsvs2bNZtmwZhw8f5s477yyXaxSpSBaLhVsjQlgU15V72tcFYPqqA9w4bhkLNqeaHJ2IOAqL3W5Xacz/Jzs7G39/f7KysvDz8zM7HABW7MzggY/W0LC2D4uf6WZ2OCIiUsEcsW0CiIqKokOHDkycOBEAm81GWFgYw4YNY8SIEX/52vDwcIYPH87w4cMv6ZxZWVnUrl2bGTNmcNdddwGwbds2mjdvTmJiItdcc83fxu2o91Pk//t1dyYvzN3M3kxjimVsyyBevq0Vwf6eJkcmImXtUtom9cg7ibNLzzXWsHoREXEQ+fn5JCUlERMTU7zParUSExNDYmJiuZ0zKSmJgoKCEsc0a9aMevXqXfB98/LyyM7OLrGJOINOjQL48anrGHJ9I1ytFn76PY0bxy3js1X7tVSdSBWmRN5JFBe6U8V6ERFxEJmZmRQVFREUFFRif1BQEKmplzcE+GLOmZqairu7O9WrV7/o942Pj8ff3794CwsLu6z4RMzg6ebCs7HN+O7Ja4kMq86JvEJenLeZu6cmsjPthNnhiYgJlMg7CRW6ExERuXwjR44kKyureDt48KDZIYlcsmbBfnz9eCdevq0lPu4uJO3/g14TVjBu4Q7yCovMDk9EKpASeSexK/1sIu9jciQiIiKGgIAAXFxczqsWn5aWdsFCdmVxzuDgYPLz8zl+/PhFv6+Hhwd+fn4lNhFn5GK1MKBTOAvjuhLTPJCCIjsTEnbS850VrNmrpepEqgol8k4gO7eA9BN5gIbWi4iI43B3d6ddu3YkJCQU77PZbCQkJBAdHV1u52zXrh1ubm4ljtm+fTsHDhy47PcVcTYh1b34oH973ut3NbV9PdiTkcM9UxMZOWcTWae1VJ1IZedqdgDy9/acmR8f6OuBn6ebydGIiIicExcXx4ABA2jfvj0dO3Zk/Pjx5OTkMHDgQAD69+9PaGgo8fHxgFHMbsuWLcVfp6SkkJycTLVq1WjcuPFFndPf359BgwYRFxdHzZo18fPzY9iwYURHR19UxXqRysJisdCrdR06Nwrg9QVb+WLNQb5Yc4BFW9N46daW9GodjMViMTtMESkHSuSdwO50zY8XERHH1LdvXzIyMhg9ejSpqalERkayYMGC4mJ1Bw4cwGo9NwDw8OHDtG3btvjxm2++yZtvvknXrl1ZunTpRZ0T4O2338ZqtdKnTx/y8vKIjY3lvffeq5iLFnEw/t5uxN/Zht6RoYycu4k9GTkMmbGemOaBvHJ7K0Kqe5kdooiUMa0jXwpHW1v2vwu2MXnpbu6/ph7/7t3a7HBERMQEjtY2OTvdT6mscguKeG/pbiYv3UVBkR0fdxeejW3KA9HhuFjVOy/iyLSOfCWjNeRFRERE5GJ4urkQd+NVfP/kdbSrX4Oc/CJe+nYLfSb/yrbUbLPDE5EyokTeCRQvPadCdyIiIiJyEa4K8mX2Y9G82rsV1TxcST54nFsmrGTsT9vILdBSdSLOTom8gysosrH/6ClAc+RFRERE5OJZrRYeuKY+i+K6clOLIAptdiYt2U3Pd1bw6+5Ms8MTkSugRN7BHTh2ikKbHW93F4L9PM0OR0REREScTLC/J+/3b8+U+9sR6OvB3swc/vHBap77aiPHT+WbHZ6IXAYl8g5u15n58Q1r+2BVgRIRERERuUw9WgWz6Jmu3H9NPQC+XHeImHHL+CY5BdW/FnEuSuQd3Nn58Sp0JyIiIiJXys/TjX/3bs1Xg6NpHFiNzJP5PDUzmQc+WsPezByzwxORi6RE3sHtTjd+oWp+vIiIiIiUlfbhNfn+yWt55sarcHe1snJXJrHjlzN+0Q7yClUMT8TROUQiP2nSJMLDw/H09CQqKoo1a9Zc8NiCggJeeeUVGjVqhKenJxERESxYsKDEMfHx8XTo0AFfX18CAwPp3bs327dvL+/LKBeqWC8iIiIi5cHD1YVh3Zvw8/AuXNckgPxCG+MX7aTn+BX8ukvF8EQcmemJ/KxZs4iLi2PMmDGsX7+eiIgIYmNjSU9PL/X4UaNGMXXqVN599122bNnC4MGDueOOO9iwYUPxMcuWLWPIkCGsWrWKhQsXUlBQwE033UROjnMNF7Lb7cVryKtHXkRERETKQ3iAD/97qCPv3teW2r4e7MnM4R8frubpWclknswzOzwRKYXFbnJli6ioKDp06MDEiRMBsNlshIWFMWzYMEaMGHHe8SEhIbzwwgsMGTKkeF+fPn3w8vJi+vTppb5HRkYGgYGBLFu2jC5duvxtTNnZ2fj7+5OVlYWfn99lXtmVS8/OpeNrCVgtsPXVHni4upgWi4iImMtR2qbKQvdTpHRZpwt486ftTF+9H7sd/L3cGNGzGX3bh6nwskg5u5S2ydQe+fz8fJKSkoiJiSneZ7VaiYmJITExsdTX5OXl4elZchk2Ly8vVq5cecH3ycrKAqBmzZoXPGd2dnaJzRHsOjOsvl5NbyXxIiIiIlLu/L3ceLV3K+Y+0ZkWdfzIOl3AyDmbuHtqIttSHeNvZBExOZHPzMykqKiIoKCgEvuDgoJITU0t9TWxsbGMGzeOnTt3YrPZWLhwIXPmzOHIkSOlHm+z2Rg+fDidO3emVatWpR4THx+Pv79/8RYWFnZlF1ZGdmeo0J2IiIiIVLzIsOrMH9qZUTc3x9vdhaT9f3DLhJXE/7iVU/mFZocnUuWZPkf+Ur3zzjs0adKEZs2a4e7uztChQxk4cCBWa+mXMmTIEDZv3szMmTMveM6RI0eSlZVVvB08eLC8wr8kxfPjVehORERERCqYq4uVh69ryKK4rsS2DKLQZmfqsj3cOG45CVvTzA5PpEozNZEPCAjAxcWFtLSSvwjS0tIIDg4u9TW1a9dm3rx55OTksH//frZt20a1atVo2LDheccOHTqU7777jiVLllC3bt0LxuHh4YGfn1+JzRFoDXkRERERMVtIdS+mPtCeD/u3J7S6FynHTzNo2joGf5bEkazTZocnUiWZmsi7u7vTrl07EhISivfZbDYSEhKIjo7+y9d6enoSGhpKYWEhX3/9Nbfffnvxc3a7naFDhzJ37lwWL15MgwYNyu0aytO5HnkfkyMRERERkaoupkUQC+O68FiXhrhYLSz4PZWYt5bx0cq9FBbZzA5PpEoxfWh9XFwcH3zwAdOmTWPr1q08/vjj5OTkMHDgQAD69+/PyJEji49fvXo1c+bMYc+ePaxYsYIePXpgs9l47rnnio8ZMmQI06dPZ8aMGfj6+pKamkpqaiqnTzvPJ4Y5eYUczsoFoGGAeuRFRERExHze7q6M7NWc74Zdy9X1qpOTX8Sr323h9km/sPHgcbPDE6kyXM0OoG/fvmRkZDB69GhSU1OJjIxkwYIFxQXwDhw4UGL+e25uLqNGjWLPnj1Uq1aNXr168dlnn1G9evXiYyZPngxAt27dSrzXJ598woMPPljel1Qm9mYahe5q+bhTw8fd5GhERERERM5pXsePrwZ3Yubag7z+41Z+P5xN7/d+4YFr6vPP2Kb4ebqZHaJIpWb6OvKOyBHWlp23IYXhs5LpGF6TLwf/9TQDERGp/ByhbapMdD9Fyk7GiTxe+2ErczekABDo68HoW1twc+s6WCxae17kYjnNOvJyYWcL3alivYiIiIg4stq+HrzdN5LPH46iQYAP6SfyGDpjAw9+spYDR0+ZHZ5IpaRE3kEVJ/K1VehORERERBxf58YB/PjUdQyPaYK7i5VlOzK48e1lTFqyi/xCFcMTKUtK5B3U7nRjjrx65EVERETEWXi6uTA85ioWDL+Ozo1rkVdoY+xP2+k1YQWr9xw1OzyRSkOJvAMqLLIVF7vTGvIiIiIi4mwa1q7G9EFRjO8bSUA1d3aln6Tv+6t4dvZGjuXkmx2eiNNTIu+ADv1xmvwiGx6uVkKre5kdjoiIiIjIJbNYLPRuG0pCXDf+EVUPgNlJh+j+1lK+XHcQ1dwWuXxK5B3Q2fnxDWtXw2pVpU8RERERcV7+3m68dkdrvn68E82CffnjVAHPffUbfaeuYmfaCbPDE3FKSuQdkArdiYiIiEhl065+Db4ddi3P92qGl5sLa/Ydo9eEFYz9aRun84vMDk/EqSiRd0C70s8m8pofLyIiIiKVh5uLlUe7NGJhXBdimgdSUGRn0pLd3DR+GUu2p5sdnojTUCLvgHZnnCl0p4r1IiIiIlIJ1a3hzYcDOjD1gXbU8ffk4LHTDPxkLfd/uJqNB4+bHZ6Iw1Mi72Dsdrt65EVERESkSohtGcyiuK48fG0D3FwsrNyVye2TfmHwZ0nsStf8eZELUSLvYI7l5JN1ugCLBRoEaI68iIiIiFRuPh6ujLqlBYuf6cadV4discCC31O56e3l/HP2Rg4eO2V2iCIOR4m8gzk7rD60uhde7i4mRyMiIiIiUjHCanoz7p5IfhrehdiWQdjs8FXSIW54aykvzf+djBN5Zoco4jCUyDsYDasXERERkarsqiBfpj7QnnlDOtO5cS0Kiux8+us+uo5dwps/bSfrdIHZIYqYTom8gzm79JwK3YmIiIhIVRYZVp3PH76Gzx+OIiKsOqfyi5i4ZBdd3ljClGW7tWSdVGlK5B3MuTXklciLiIiIiHRuHMC8Jzox5f52NAmsRtbpAl7/cRtdxy7hs1X7KSiymR2iSIVTIu9gziXyKnQnIiIiIgJgsVjo0SqYBcO78NbdEdSt4UX6iTxenLeZ7m8tY96GFGw2u9lhilQYJfIOJLegiEN/nAagkYbWi4iIiIiU4GK10KddXRKe6crLt7UkoJoHB46dYvisZHpNWMGiLWnY7UropfJTIu9A9mTkYLdDdW83avm4mx2OiIiIiIhD8nB1YUCncJY/141nY5vi6+nKttQTPPy/dfSZ/CuJu4+aHaJIuVIi70D+PD/eYrGYHI2IiIiIiGPzdndlyPWNWfncDTzerRGeblbWHzjOfR+s4oGPVrPpUJbZIYqUCyXyDkTz40VERERELp2/txv/6tGM5c9ezwPX1MfVamHFzkxunbiSJz5PKl7iWaSyUCLvQLSGvIiIiIjI5Qv08+TV3q1Y/Ew37mgbisUCP2xK5aa3l/HcVxtJOX7a7BBFyoQSeQeyOyMH0BryIiIiIiJXol4tb97uG8mPT13HjS2CsNnhy3WHuH7sUl7+9ncyT+aZHaLIFVEi7yBsNjt7tIa8iIiIiEiZaRbsxwf92zPniU5EN6xFfpGNT37ZR9c3ljDu5+1k5xaYHaLIZVEi7yBSjp8mr9CGu4uVujW8zA5HRERERKTSuLpeDWY8EsVngzrSpq4/OflFTFi8iy5vLOH95bvJLSgyO0SRS6JE3kGcLXQXHuCNq4u+LSIiIiIiZclisXBdk9p8M6Qzk/tdTaPaPhw/VcBrP2yj69glfL56PwVFNrPDFLkoyhgdhArdiYiIs5o0aRLh4eF4enoSFRXFmjVr/vL42bNn06xZMzw9PWndujU//PBDiectFkup29ixY4uPCQ8PP+/5119/vVyuT0QqF4vFQs/WdfhpeBfG3tWG0OpepGXn8cLczdw4bhnfJKdgs9nNDlPkLymRdxAqdCciIs5o1qxZxMXFMWbMGNavX09ERASxsbGkp6eXevyvv/7Kfffdx6BBg9iwYQO9e/emd+/ebN68ufiYI0eOlNg+/vhjLBYLffr0KXGuV155pcRxw4YNK9drFZHKxdXFyt3tw1j8z66MubUFtXzc2Xf0FE/NTObmd1eyYmeG2SGKXJDFbrfr46b/Jzs7G39/f7KysvDz86uQ97xnaiJr9h5jfN9IercNrZD3FBER52FG23QxoqKi6NChAxMnTgTAZrMRFhbGsGHDGDFixHnH9+3bl5ycHL777rvifddccw2RkZFMmTKl1Pfo3bs3J06cICEhoXhfeHg4w4cPZ/jw4ZcVt6PeTxExT05eIR+v3Mv7y/dwIq8QgG5NazOyZ3OaBvuaHJ1UBZfSNqlH3kGoYr2IiDib/Px8kpKSiImJKd5ntVqJiYkhMTGx1NckJiaWOB4gNjb2gsenpaXx/fffM2jQoPOee/3116lVqxZt27Zl7NixFBYWXjDWvLw8srOzS2wiIn/m4+HKsO5NWP7c9TzUuQFuLhaWbs+g5zvLGTnnN9JP5JodokgxJfIO4PipfDJP5gPQsLaPydGIiIhcnMzMTIqKiggKCiqxPygoiNTU1FJfk5qaeknHT5s2DV9fX+68884S+5988klmzpzJkiVLeOyxx3jttdd47rnnLhhrfHw8/v7+xVtYWNjFXKKIVEE1fNwZfWsLFj7dlZ6tgrHZ4Ys1B+k2dikTEnZyOl8V7sV8rmYHIOcq1of4e+LjoW+JiIjIWR9//DH9+vXD09OzxP64uLjir9u0aYO7uzuPPfYY8fHxeHh4nHeekSNHlnhNdna2knkR+UvhAT5Mvr8d6/Yd49/fbyX54HHGLdzB56v388+bmnLn1XVxsVrMDlOqKPXIO4Dd6Uahu0YqdCciIk4kICAAFxcX0tLSSuxPS0sjODi41NcEBwdf9PErVqxg+/btPPzww38bS1RUFIWFhezbt6/U5z08PPDz8yuxiYhcjPbhNZn7RCfeva8tdWsYFe6f/eo3bn13Jb/syjQ7PKmilMg7gN2aHy8iIk7I3d2ddu3alShCZ7PZSEhIIDo6utTXREdHlzgeYOHChaUe/9FHH9GuXTsiIiL+Npbk5GSsViuBgYGXeBUiIn/PYrFwa0QICc905flezfD1dGXLkWz6fbiagZ+sYUfaCbNDlCpG47gdwLk15DU/XkREnEtcXBwDBgygffv2dOzYkfHjx5OTk8PAgQMB6N+/P6GhocTHxwPw1FNP0bVrV9566y1uvvlmZs6cybp163j//fdLnDc7O5vZs2fz1ltvnfeeiYmJrF69muuvvx5fX18SExN5+umnuf/++6lRo0b5X7SIVFkeri482qURd7cLY8LinXyWuJ8l2zNYtiODezvW4+mYq6jte/70HpGypkTeART3yGtovYiIOJm+ffuSkZHB6NGjSU1NJTIykgULFhQXtDtw4ABW67kBgJ06dWLGjBmMGjWK559/niZNmjBv3jxatWpV4rwzZ87Ebrdz3333nfeeHh4ezJw5k5deeom8vDwaNGjA008/XWIOvIhIearh486YW1vSPzqc//64jQW/pzJj9QG+2ZDC490aMejahni5u5gdplRiWke+FBW5tmxeYRHNX1yAzQ5rnu9OoJ/n379IRESqHK17XrZ0P0WkLK3Ze4z/fL+FjYeyAAj28+SfsU25s20oVhXEk4ukdeSdyP6jp7DZwdfDVcNwREREREScUMcGNZn7RGcm3NeW0OpepGbn8s/ZG7nl3ZX8qoJ4Ug6UyJts95n58Q0Dq2Gx6NM6ERERERFnZLVauO1MQbyRPc8VxPvHh6sZ9OladqWrIJ6UHSXyJlOhOxERERGRysPTzYXHujZi2bPX82CncFytFhK2pRM7fgUvzN1E5sk8s0OUSkCJvMnOFrprrEJ3IiIiIiKVRk0fd166rSU/P92Fm1oEUWSz8/nqA3Qbu5RJS3aRW1BkdojixExP5CdNmkR4eDienp5ERUWxZs2aCx5bUFDAK6+8QqNGjfD09CQiIoIFCxaUOGb58uXceuuthISEYLFYmDdvXjlfwZXZnZEDaA15EREREZHKqGHtarzfvz2zHr2GNnX9OZlXyNiftnP9m0uZs/4QNptqj8ulMzWRnzVrFnFxcYwZM4b169cTERFBbGws6enppR4/atQopk6dyrvvvsuWLVsYPHgwd9xxBxs2bCg+Jicnh4iICCZNmlRRl3HZ7Hb7uaXnlMiLiIiIiFRaUQ1rMe+JzrxzbySh1b04kpVL3JcbuW3SSn7drYJ4cmlMXX4uKiqKDh06MHHiRABsNhthYWEMGzaMESNGnHd8SEgIL7zwAkOGDCne16dPH7y8vJg+ffp5x1ssFubOnUvv3r0vKa6KWpLm8PHTdHp9Ma5WC1tf7YGbi+kDJERExEFpubSypfspImbKLSjik1/28d6SXZzIKwQgpnkgI3o215TbKswplp/Lz88nKSmJmJiYc8FYrcTExJCYmFjqa/Ly8vD0LLnOupeXFytXrryiWPLy8sjOzi6xVYSzvfH1a3kriRcRERERqSI83Vx4vFsjlj7bjQHR9XGxWli0NZ3Y8ct5cd5mFcSTv2Va9piZmUlRURFBQUEl9gcFBZGamlrqa2JjYxk3bhw7d+7EZrOxcOFC5syZw5EjR64olvj4ePz9/Yu3sLCwKzrfxdqdrmH1IiIiIiJVVa1qHrx8eyt+froLN54piPfZqv10G7uU95aqIJ5cmFN1A7/zzjs0adKEZs2a4e7uztChQxk4cCBW65VdxsiRI8nKyireDh48WEYR/7XiQncaPiMiIiIiUmU1ql2ND/q354tHrqF1qFEQ740F27nhzaV8ue4gp/ILzQ5RHIxpiXxAQAAuLi6kpaWV2J+WlkZwcHCpr6lduzbz5s0jJyeH/fv3s23bNqpVq0bDhg2vKBYPDw/8/PxKbBVhl3rkRURERETkjOhGtfhmSGfe7htBiL8nh7Nyee6r32j36iKe/GIDCVvTyC+0mR2mOADTEnl3d3fatWtHQkJC8T6bzUZCQgLR0dF/+VpPT09CQ0MpLCzk66+/5vbbby/vcMuF1pAXEREREZE/s1ot3NG2Lov/2Y0RPZtRv5Y3pwuKmL/xMIOmraPja4sYOWcTq/Yc1dJ1VZirmW8eFxfHgAEDaN++PR07dmT8+PHk5OQwcOBAAPr3709oaCjx8fEArF69mpSUFCIjI0lJSeGll17CZrPx3HPPFZ/z5MmT7Nq1q/jx3r17SU5OpmbNmtSrV69iL/AvZOcWkH7CKGLRsLaPydGIiIiIiIgj8XRzYXDXRjzWpSEbD2UxP/kw3/52mIwTeXyx5gBfrDlAsJ8nt0bU4fbIUFqG+GGxWMwOWyqIqYl83759ycjIYPTo0aSmphIZGcmCBQuKC+AdOHCgxPz33NxcRo0axZ49e6hWrRq9evXis88+o3r16sXHrFu3juuvv774cVxcHAADBgzg008/rZDruhh7zsyPD/T1wM/TzeRoRERERETEEVksFiLDqhMZVp0Xbm7O6j1H+Sb5MD9sPkJqdi4frNjLByv20jDAh9siQ7gtIoSGmrpb6Zm6jryjqoi1Zb9KOsQ/Z28kumEtvnj0mnJ5DxERqTy07nnZ0v0UEWeXV1jEsu0ZfLPxMAlb08gtODd3vnWoP7dFhHBLRB3q+HuZGKVciktpm0ztka/Kzs6PbxSoYfUiIiIiInJpPFxduKllMDe1DOZkXiELt6QyP/kwy3dmsikli00pWbz241Y6htfk9shQerUOprq3u9lhSxlRIm+Ss2vIN9awFxERERERuQLVPFy5o21d7mhbl6Mn8/hhcyrfJh9mzb5jrN5rbGPmb6ZLk9rcFhnCjS2C8HZXKujM9N0zybkeeSXyIiIiIiJSNmpV8+CBa+rzwDX1STl+mm83HmZ+8mG2HMkmYVs6CdvS8XJz4cYWQdweGcJ1TWrj7mraYmZymZTIm6CgyMb+o6cArSEvIiIiIiLlI7S6F4O7NmJw10bsSj/B/OTDfLPxMPuPnmL+xsPM33iY6t5u9GxVh9siQohqUBOrVZXvnYESeRPsP3qKQpsdb3cX6vh7mh2OiIiIiIhUco0DfYm7qSlP33jVXy5nd0sbYzm7VqFazs6RKZE3QfGw+trV9J9DREREREQqzN8tZ/fhyr18uNJYzu7WiBBuiwzRKGIHpETeBOcSeVWsFxERERERc7hYLXRqHECnxgG80rtlieXs9mTm8E7CTt5J2EmrUD9ujwjltsgQgvw0otgRKJE3we70HEDz40VERERExDH81XJ2m1Oy2ZySzdiftjOgU32GXt8Ef283s0Ou0pTIm2CXKtaLiIiIiIiDKm05u7nrD7H+wHE+WLGXL9cdYtgNjXkguj4eri5mh1slaZ2BCma329lzdg15JfIiIiIiIuLAzi5n9/XjnfhkYAeaBvmSdbqAf3+/lZhxy/jut8PY7Xazw6xylMhXsIwTeZzIK8Rqgfq1vM0OR0RERERE5G9ZLBaubxrID09dx3/7tCbQ14ODx04zdMYG7njvV9buO2Z2iFWKEvkKdnZYfb2a3hqGIiIiIiIiTsXFaqFvh3osfbYbT8dchbe7C8kHj3P3lEQe+2wde87kO1K+lMhXsN3p55aeExERERERcUbe7q48FdOEpc92476O9bBa4Kff07jp7eWM+WYzR0/mmR1ipaZEvoLtzjhTsV7z40VERERExMkF+noSf2drFgzvwg3NAim02ZmWuJ9uY5cyackucguKzA6xUlIiX8HOriHfWD3yIiIiIiJSSVwV5MvHD3ZgxsNRtAzx40ReIWN/2s4Nby7l66RD2GwqiFeWlMhXsOKh9YE+JkciIiIiIiJStjo1DuDbodfydt8IQqt7cTgrl2dmb+TWiSv5ZVem2eFVGkrkK1BOXiGHs3IBaBigHnkREREREal8rFYLd7StS8IzXflXj2b4erjy++Fs+n24mgc/WcP21BNmh+j0lMhXoD1n5sfX8nGnho+7ydGIiIiIiIiUH083Fx7v1ohlz13Pg53CcbVaWLo9g57vLGfE17+Rnp1rdohOS4l8BTo7P16F7kREREREpKqo6ePOS7e1ZGFcV3q2CsZmh5lrD9J17FLeXriDnLxCs0N0OkrkK1BxIq9CdyIiIiIiUsU0CPBh8v3t+GpwNG3rVed0QRHvJOyk25tL+WLNAQqLbGaH6DSUyFegc4m8Ct2JiIiIiEjV1D68JnMe78R7/a6mXk1vMk7kMXLOJnpNWMGSbenY7apw/3eUyFegXekaWi8iIiIiImKxWOjVug6L4roy+pYWVPd2Y0faSQZ+upZ+H65mc0qW2SE6NCXyFaSwyMa+zFOA1pAXEREREREBcHe18tC1DVj27PU81qUh7i5Wft19lFsnriRuVjIpx0+bHaJDUiJfQQ79cZr8IhserlZCq3uZHY6IiIiIiIjD8PdyY2Sv5iQ805XbI0Ow22HOhhSuf3Mp/12wjezcArNDdChK5CvI2fnxDWtXw2q1mByNiIiIiIiI4wmr6c0797Zl/tDORDWoSX6hjclLd9Nt7FKm/bqPAhXEA5TIV5ji+fEqdCciIiIiIvKX2tStzsxHr+HD/u1pVNuHYzn5jJn/Oze9vZwFm1OrfEE8V7MDqCq09JyIiIiIiMjFs1gsxLQIolvT2sxce5Dxi3awNzOHwdOTaB3qz93t63Jz6zrUquZhdqgVTol8BdmdkQNAY1WsFxERERERuWiuLlbuv6Y+vduGMnXZbj5YsYdNKVlsSsnilW+30PWq2vRuG8qNLYLwdHMxO9wKoaH1FcBut/9paL0SeRERqVwmTZpEeHg4np6eREVFsWbNmr88fvbs2TRr1gxPT09at27NDz/8UOL5Bx98EIvFUmLr0aNHiWOOHTtGv3798PPzo3r16gwaNIiTJ0+W+bWJiIjjqObhyjM3NWXFczfw4i0taB3qT6HNTsK2dIZ9sYH2/17EP2dv5NddmRTZKvfQeyXyFeBYTj5ZpwuwWKBBgObIi4hI5TFr1izi4uIYM2YM69evJyIigtjYWNLT00s9/tdff+W+++5j0KBBbNiwgd69e9O7d282b95c4rgePXpw5MiR4u2LL74o8Xy/fv34/fffWbhwId999x3Lly/n0UcfLbfrFBERx1Hb14NB1zbg22HXsiiuC0Ovb0xodS9O5hXyVdIh/vHhajq/vpj4H7ay9Ui22eGWC4u9qlcJKEV2djb+/v5kZWXh5+d3xedbvecofd9fRd0aXqz81w1lEKGIiFQ1Zd02lZWoqCg6dOjAxIkTAbDZbISFhTFs2DBGjBhx3vF9+/YlJyeH7777rnjfNddcQ2RkJFOmTAGMHvnjx48zb968Ut9z69attGjRgrVr19K+fXsAFixYQK9evTh06BAhISF/G7ej3k8REbk8Npuddfv/YO6GFL7/7TDZuYXFzzUL9qV321Bujwyhjr/jLgV+KW2TeuQrgObHi4hIZZSfn09SUhIxMTHF+6xWKzExMSQmJpb6msTExBLHA8TGxp53/NKlSwkMDKRp06Y8/vjjHD16tMQ5qlevXpzEA8TExGC1Wlm9enWp75uXl0d2dnaJTUREKg+r1ULHBjWJv7M1a0fFMOX+dvRoGYy7i5VtqSd4/cdtdHp9Mfe9v4ov1x50+nXpVeyuAqhivYiIVEaZmZkUFRURFBRUYn9QUBDbtm0r9TWpqamlHp+amlr8uEePHtx55500aNCA3bt38/zzz9OzZ08SExNxcXEhNTWVwMDAEudwdXWlZs2aJc7zZ/Hx8bz88suXc5kiIuJkPFxd6NEqmB6tgsk6VcAPm48wd0MKa/YeI3HPURL3HOXFbzYT0yKIOyJD6XJVbdxdnauPW4l8BVAiLyIicvHuvffe4q9bt25NmzZtaNSoEUuXLqV79+6Xdc6RI0cSFxdX/Dg7O5uwsLArjlVERBybv7cb93Wsx30d63Hoj1N8k3yYOesPsTsjh+9/O8L3vx2hhrcbt7QJoXfbUK6uVx2LxWJ22H9LiXwFOFexXoXuRESk8ggICMDFxYW0tLQS+9PS0ggODi71NcHBwZd0PEDDhg0JCAhg165ddO/eneDg4POK6RUWFnLs2LELnsfDwwMPj6q3zrCIiJxTt4Y3Q65vzBPdGvH74Wzmbkjhm+TDZJ7M47NV+/ls1X7q1/Lm9shQ7mgb6tCFyp1r/IATOp1fRMrx04DmyIuISOXi7u5Ou3btSEhIKN5ns9lISEggOjq61NdER0eXOB5g4cKFFzwe4NChQxw9epQ6deoUn+P48eMkJSUVH7N48WJsNhtRUVFXckkiIlIFWCwWWoX68+ItLVg18gb+91BH7mwbire7C/uPnmJCwk6uf3MpvSf9wrRf93H0ZJ7ZIZ9HPfLlbG9mDnY7VPd2o6aPu9nhiIiIlKm4uDgGDBhA+/bt6dixI+PHjycnJ4eBAwcC0L9/f0JDQ4mPjwfgqaeeomvXrrz11lvcfPPNzJw5k3Xr1vH+++8DcPLkSV5++WX69OlDcHAwu3fv5rnnnqNx48bExsYC0Lx5c3r06MEjjzzClClTKCgoYOjQodx7770XVbFeRETkLFcXK12uqk2Xq2rz7/xCFm5JY+6GFFbszCT54HGSDx7nle+20PWq2vRuG8qNzYPwcncxO2wl8uXtz/PjnWGuhYiIyKXo27cvGRkZjB49mtTUVCIjI1mwYEFxQbsDBw5gtZ4bANipUydmzJjBqFGjeP7552nSpAnz5s2jVatWALi4uPDbb78xbdo0jh8/TkhICDfddBOvvvpqiaHxn3/+OUOHDqV79+5YrVb69OnDhAkTKvbiRUSkUvF2d+X2yFBujwwl40Qe3248zLzkFH47lMXibeks3paOj7sLPVrV4Y62oUQ3qoWL1ZwcT+vIl6Is15Z9e+EO3knYyT3t6/LGXRFlFKGIiFQ1Wve8bOl+iojIxdqVfpJvklOYuyGFQ3+cLt4f5OfB7ZGh9I4MpXkd3yvuuNU68g5EFetFREREREScV+PAajxzU1NWPHc9Xw2Opl9UPfy93EjLzuP95XvoNWEFR7JyKzQmDa0vZ+G1fGgd6k+zOvq0X0RERERExFlZLBbah9ekfXhNRt/agmXbM5iXnELW6QJCqntVaCwO0SM/adIkwsPD8fT0JCoqijVr1lzw2IKCAl555RUaNWqEp6cnERERLFiw4IrOWZ7+GduUb4ddS9erapvy/iIiIiIiIlK2PFxduKllMO/1a8dnD1X8iimmJ/KzZs0iLi6OMWPGsH79eiIiIoiNjT1vfdizRo0axdSpU3n33XfZsmULgwcP5o477mDDhg2XfU4RERERERGRy2E1oeCd6cXuoqKi6NChAxMnTgSM9WfDwsIYNmwYI0aMOO/4kJAQXnjhBYYMGVK8r0+fPnh5eTF9+vTLOuf/pwI4IiLiaNQ2lS3dTxERcTROU+wuPz+fpKQkYmJiivdZrVZiYmJITEws9TV5eXl4enqW2Ofl5cXKlSuv6JzZ2dklNhERERERERFHZGoin5mZSVFRUfFas2cFBQWRmppa6mtiY2MZN24cO3fuxGazsXDhQubMmcORI0cu+5zx8fH4+/sXb2FhYWVwdSIiIiIiIiJlz/Q58pfqnXfeoUmTJjRr1gx3d3eGDh3KwIEDsVov/1JGjhxJVlZW8Xbw4MEyjFhERERERESk7JiayAcEBODi4kJaWlqJ/WlpaQQHB5f6mtq1azNv3jxycnLYv38/27Zto1q1ajRs2PCyz+nh4YGfn1+JTURERERERMQRmZrIu7u7065dOxISEor32Ww2EhISiI6O/svXenp6EhoaSmFhIV9//TW33377FZ9TRERERERExNG5mh1AXFwcAwYMoH379nTs2JHx48eTk5PDwIEDAejfvz+hoaHEx8cDsHr1alJSUoiMjCQlJYWXXnoJm83Gc889d9HnFBEREREREXFWpifyffv2JSMjg9GjR5OamkpkZCQLFiwoLlZ34MCBEvPfc3NzGTVqFHv27KFatWr06tWLzz77jOrVq1/0OUVERERERESclenryDsirS0rIiKORm1T2dL9FBERR+M068iLiIiIiIiIyKVRIi8iIiIiIiLiRJTIi4iIiIiIiDgRJfIiIiIiIiIiTkSJvIiIiIiIiIgTUSIvIiIiIiIi4kSUyIuIiIiIiIg4ESXyIiIiIiIiIk7E1ewAHJHdbgcgOzvb5EhEREQMZ9uks22UXBm19SIi4mgupa1XIl+KEydOABAWFmZyJCIiIiWdOHECf39/s8NwemrrRUTEUV1MW2+x66P989hsNg4fPoyvry8Wi+WKzpWdnU1YWBgHDx7Ez8+vjCJ0frovpdN9KZ3uS+l0X0pXWe+L3W7nxIkThISEYLVqZtyVKsu2Hirvz92V0n05n+5J6XRfSqf7UrrKel8upa1Xj3wprFYrdevWLdNz+vn5VaofsrKi+1I63ZfS6b6UTveldJXxvqgnvuyUR1sPlfPnrizovpxP96R0ui+l030pXWW8Lxfb1usjfREREREREREnokReRERERERExIkokS9nHh4ejBkzBg8PD7NDcSi6L6XTfSmd7kvpdF9Kp/siZtDPXel0X86ne1I63ZfS6b6UTvdFxe5EREREREREnIp65EVERERERESciBJ5ERERERERESeiRF5ERERERETEiSiRFxEREREREXEiSuTL2aRJkwgPD8fT05OoqCjWrFljdkimio+Pp0OHDvj6+hIYGEjv3r3Zvn272WE5lNdffx2LxcLw4cPNDsUhpKSkcP/991OrVi28vLxo3bo169atMzss0xQVFfHiiy/SoEEDvLy8aNSoEa+++ipVrW7p8uXLufXWWwkJCcFisTBv3rwSz9vtdkaPHk2dOnXw8vIiJiaGnTt3mhOsVHpq60tSW39x1N6fo7b+fGrvDWrvL0yJfDmaNWsWcXFxjBkzhvXr1xMREUFsbCzp6elmh2aaZcuWMWTIEFatWsXChQspKCjgpptuIicnx+zQHMLatWuZOnUqbdq0MTsUh/DHH3/QuXNn3Nzc+PHHH9myZQtvvfUWNWrUMDs00/z3v/9l8uTJTJw4ka1bt/Lf//6XN954g3fffdfs0CpUTk4OERERTJo0qdTn33jjDSZMmMCUKVNYvXo1Pj4+xMbGkpubW8GRSmWntv58auv/ntr7c9TWl07tvUHt/V+wS7np2LGjfciQIcWPi4qK7CEhIfb4+HgTo3Is6enpdsC+bNkys0Mx3YkTJ+xNmjSxL1y40N61a1f7U089ZXZIpvvXv/5lv/baa80Ow6HcfPPN9oceeqjEvjvvvNPer18/kyIyH2CfO3du8WObzWYPDg62jx07tnjf8ePH7R4eHvYvvvjChAilMlNb//fU1pek9r4ktfWlU3t/PrX3JalHvpzk5+eTlJRETExM8T6r1UpMTAyJiYkmRuZYsrKyAKhZs6bJkZhvyJAh3HzzzSV+Zqq6+fPn0759e+6++24CAwNp27YtH3zwgdlhmapTp04kJCSwY8cOADZu3MjKlSvp2bOnyZE5jr1795Kamlri/5K/vz9RUVH6/StlSm39xVFbX5La+5LU1pdO7f3fq+rtvavZAVRWmZmZFBUVERQUVGJ/UFAQ27ZtMykqx2Kz2Rg+fDidO3emVatWZodjqpkzZ7J+/XrWrl1rdigOZc+ePUyePJm4uDief/551q5dy5NPPom7uzsDBgwwOzxTjBgxguzsbJo1a4aLiwtFRUX85z//oV+/fmaH5jBSU1MBSv39e/Y5kbKgtv7vqa0vSe39+dTWl07t/d+r6u29EnkxzZAhQ9i8eTMrV640OxRTHTx4kKeeeoqFCxfi6elpdjgOxWaz0b59e1577TUA2rZty+bNm5kyZUqVbdy//PJLPv/8c2bMmEHLli1JTk5m+PDhhISEVNl7IiKOS239OWrvS6e2vnRq7+XvaGh9OQkICMDFxYW0tLQS+9PS0ggODjYpKscxdOhQvvvuO5YsWULdunXNDsdUSUlJpKenc/XVV+Pq6oqrqyvLli1jwoQJuLq6UlRUZHaIpqlTpw4tWrQosa958+YcOHDApIjM9+yzzzJixAjuvfdeWrduzQMPPMDTTz9NfHy82aE5jLO/Y/X7V8qb2vq/pra+JLX3pVNbXzq193+vqrf3SuTLibu7O+3atSMhIaF4n81mIyEhgejoaBMjM5fdbmfo0KHMnTuXxYsX06BBA7NDMl337t3ZtGkTycnJxVv79u3p168fycnJuLi4mB2iaTp37nzekkU7duygfv36JkVkvlOnTmG1lvzV7eLigs1mMykix9OgQQOCg4NL/P7Nzs5m9erVVfr3r5Q9tfWlU1tfOrX3pVNbXzq193+vqrf3GlpfjuLi4hgwYADt27enY8eOjB8/npycHAYOHGh2aKYZMmQIM2bM4JtvvsHX17d4/oq/vz9eXl4mR2cOX1/f8+YN+vj4UKtWrSo/n/Dpp5+mU6dOvPbaa9xzzz2sWbOG999/n/fff9/s0Exz66238p///Id69erRsmVLNmzYwLhx43jooYfMDq1CnTx5kl27dhU/3rt3L8nJydSsWZN69eoxfPhw/v3vf9OkSRMaNGjAiy++SEhICL179zYvaKmU1NafT2196dTel05tfenU3hvU3v8Fs8vmV3bvvvuuvV69enZ3d3d7x44d7atWrTI7JFMBpW6ffPKJ2aE5FC1Hc863335rb9Wqld3Dw8PerFkz+/vvv292SKbKzs62P/XUU/Z69erZPT097Q0bNrS/8MIL9ry8PLNDq1BLliwp9XfJgAED7Ha7sSTNiy++aA8KCrJ7eHjYu3fvbt++fbu5QUulpba+JLX1F0/tvUFt/fnU3hvU3l+YxW632yvygwMRERERERERuXyaIy8iIiIiIiLiRJTIi4iIiIiIiDgRJfIiIiIiIiIiTkSJvIiIiIiIiIgTUSIvIiIiIiIi4kSUyIuIiIiIiIg4ESXyIiIiIiIiIk5EibyIOCSLxcK8efPMDkNERETKidp6kcunRF5EzvPggw9isVjO23r06GF2aCIiIlIG1NaLODdXswMQEcfUo0cPPvnkkxL7PDw8TIpGREREypraehHnpR55ESmVh4cHwcHBJbYaNWoAxlC4yZMn07NnT7y8vGjYsCFfffVViddv2rSJG264AS8vL2rVqsWjjz7KyZMnSxzz8ccf07JlSzw8PKhTpw5Dhw4t8XxmZiZ33HEH3t7eNGnShPnz55fvRYuIiFQhautFnJcSeRG5LC+++CJ9+vRh48aN9OvXj3vvvZetW7cCkJOTQ2xsLDVq1GDt2rXMnj2bRYsWlWi8J0+ezJAhQ3j00UfZtGkT8+fPp3HjxiXe4+WXX+aee+7ht99+o1evXvTr149jx45V6HWKiIhUVWrrRRyYXUTk/xkwYIDdxcXF7uPjU2L7z3/+Y7fb7XbAPnjw4BKviYqKsj/++ON2u91uf//99+01atSwnzx5svj577//3m61Wu2pqal2u91uDwkJsb/wwgsXjAGwjxo1qvjxyZMn7YD9xx9/LLPrFBERqarU1os4N82RF5FSXX/99UyePLnEvpo1axZ/HR0dXeK56OhokpOTAdi6dSsRERH4+PgUP9+5c2dsNhvbt2/HYrFw+PBhunfv/pcxtGnTpvhrHx8f/Pz8SE9Pv9xLEhERkT9RWy/ivJTIi0ipfHx8zhv+Vla8vLwu6jg3N7cSjy0WCzabrTxCEhERqXLU1os4L82RF5HLsmrVqvMeN2/eHIDmzZuzceNGcnJyip//5ZdfsFqtNG3aFF9fX8LDw0lISKjQmEVEROTiqa0XcVzqkReRUuXl5ZGamlpin6urKwEBAQDMnj2b9u3bc+211/L555+zZs0aPvroIwD69evHmDFjGDBgAC+99BIZGRkMGzaMBx54gKCgIABeeuklBg8eTGBgID179uTEiRP88ssvDBs2rGIvVEREpIpSWy/ivJTIi0ipFixYQJ06dUrsa9q0Kdu2bQOMKrMzZ87kiSeeoE6dOnzxxRe0aNECAG9vb3766SeeeuopOnTogLe3N3369GHcuHHF5xowYAC5ubm8/fbb/POf/yQgIIC77rqr4i5QRESkilNbL+K8LHa73W52ECLiXCwWC3PnzqV3795mhyIiIiLlQG29iGPTHHkRERERERERJ6JEXkRERERERMSJaGi9iIiIiIiIiBNRj7yIiIiIiIiIE1EiLyIiIiIiIuJElMiLiIiIiIiIOBEl8iIiIiIiIiJORIm8iIiIiIiIiBNRIi8iIiIiIiLiRJTIi4iIiIiIiDgRJfIiIiIiIiIiTkSJvIiIiIiIiIgT+T8MfPF6sBlQXQAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 1200x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Plot the training and validation accuracy and loss curves\n",
    "plt.figure(figsize=(12,6))\n",
    "plt.subplot(1,2,1)\n",
    "plt.plot(history.history[\"accuracy\"], label=\"train accuracy\")\n",
    "plt.plot(history.history[\"val_accuracy\"], label=\"validation accuracy\")\n",
    "plt.xlabel(\"Epoch\")\n",
    "plt.ylabel(\"Accuracy\")\n",
    "plt.legend()\n",
    "plt.subplot(1,2,2)\n",
    "plt.plot(history.history[\"loss\"], label=\"train loss\")\n",
    "plt.plot(history.history[\"val_loss\"], label=\"validation loss\")\n",
    "plt.xlabel(\"Epoch\")\n",
    "plt.ylabel(\"Loss\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "365/365 [==============================] - 0s 773us/step\n",
      "Test accuracy: 0.9576787707099322\n",
      "Test Precision: 0.9435965494359655\n",
      "Test Recall: 0.9738058551617874\n",
      "Test confusion matrix:\n",
      "[[5468  340]\n",
      " [ 153 5688]]\n"
     ]
    }
   ],
   "source": [
    "#Evaluate the model on the test set\n",
    "y_pred = model.predict(X_test).round().ravel()\n",
    "test_acc = accuracy_score(y_test, y_pred)\n",
    "precision = precision_score(y_test, y_pred)\n",
    "recall = recall_score(y_test, y_pred)\n",
    "test_cm = confusion_matrix(y_test, y_pred)\n",
    "print(\"Test accuracy:\", test_acc)\n",
    "print(\"Test Precision:\", precision)\n",
    "print(\"Test Recall:\", recall)\n",
    "print(\"Test confusion matrix:\")\n",
    "print(test_cm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Create a binary label dataset from the train set for aif360\n",
    "train_df = pd.DataFrame(X_train.reshape(-1, 64*64))\n",
    "train_df[\"label\"] = y_train\n",
    "train_df[\"gender\"] = 0\n",
    "train_dataset = BinaryLabelDataset(favorable_label=0, unfavorable_label=1, df=train_df, label_names=[\"label\"], protected_attribute_names=[\"gender\"], unprivileged_protected_attributes=[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(47009, 4097)\n",
      "192595873\n"
     ]
    }
   ],
   "source": [
    "print(train_dataset.features.shape)\n",
    "print(train_dataset.features.size)\n",
    "\n",
    "#Extract the features and labels from the train set\n",
    "X_train = train_dataset.features[:, :-1].reshape(-1, 64, 64, 1)\n",
    "y_train = train_dataset.labels.ravel()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#from aif360.metrics import BinaryLabelDatasetMetric\n",
    "#Compute the disparate impact and statistical parity difference metrics for the train set\n",
    "#metric_dataset = BinaryLabelDatasetMetric(train_dataset, unprivileged_groups=[{\"gender\": 1}], privileged_groups=[{\"gender\": 0}])\n",
    "#print(\"Disparate impact:\", metric_dataset.disparate_impact())\n",
    "#print(\"Statistical parity difference:\", metric_dataset.statistical_parity_difference())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Apply the reweighing algorithm to mitigate bias in the training set\n",
    "RW = Reweighing(unprivileged_groups=[{\"gender\": 1}], privileged_groups=[{\"gender\": 0}])\n",
    "train_dataset_rw = RW.fit_transform(train_dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Compute the metrics for the reweighted train set\n",
    "#metric_dataset_rw = BinaryLabelDatasetMetric(train_dataset_rw, unprivileged_groups=[{\"gender\": 1}], privileged_groups=[{\"gender\": 0}])\n",
    "#print(\"Disparate impact after reweighing:\", metric_dataset_rw.disparate_impact())\n",
    "#print(\"Statistical parity difference after reweighing:\", metric_dataset_rw.statistical_parity_difference())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(47009, 4097)\n",
      "192595873\n"
     ]
    }
   ],
   "source": [
    "print(train_dataset_rw.features.shape)\n",
    "print(train_dataset_rw.features.size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Extract the features and labels from the reweighted train set\n",
    "X_train_rw = train_dataset_rw.features[:, :-1].reshape(-1, 64, 64, 1)\n",
    "y_train_rw = train_dataset_rw.labels.ravel()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv2d_3 (Conv2D)           (None, 62, 62, 32)        320       \n",
      "                                                                 \n",
      " max_pooling2d_3 (MaxPooling  (None, 31, 31, 32)       0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " conv2d_4 (Conv2D)           (None, 29, 29, 64)        18496     \n",
      "                                                                 \n",
      " max_pooling2d_4 (MaxPooling  (None, 14, 14, 64)       0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " conv2d_5 (Conv2D)           (None, 12, 12, 128)       73856     \n",
      "                                                                 \n",
      " max_pooling2d_5 (MaxPooling  (None, 6, 6, 128)        0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " flatten_1 (Flatten)         (None, 4608)              0         \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 256)               1179904   \n",
      "                                                                 \n",
      " dropout_1 (Dropout)         (None, 256)               0         \n",
      "                                                                 \n",
      " dense_3 (Dense)             (None, 1)                 257       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 1,272,833\n",
      "Trainable params: 1,272,833\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "\n",
    "model.add(Conv2D(32, (3, 3), activation=\"relu\", input_shape=(64, 64, 1)))\n",
    "model.add(MaxPooling2D((2, 2)))\n",
    "\n",
    "model.add(Conv2D(64, (3, 3), activation=\"relu\"))\n",
    "model.add(MaxPooling2D((2, 2)))\n",
    "\n",
    "model.add(Conv2D(128, (3, 3), activation=\"relu\"))\n",
    "model.add(MaxPooling2D((2, 2)))\n",
    "\n",
    "model.add(Flatten())\n",
    "\n",
    "model.add(Dense(256, activation=\"relu\"))\n",
    "model.add(Dropout(0.5))\n",
    "model.add(Dense(1, activation=\"sigmoid\"))\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "1470/1470 [==============================] - 5s 3ms/step - loss: 0.2476 - accuracy: 0.8947 - val_loss: 0.1369 - val_accuracy: 0.9488\n",
      "Epoch 2/50\n",
      "1470/1470 [==============================] - 4s 3ms/step - loss: 0.1525 - accuracy: 0.9455 - val_loss: 0.1375 - val_accuracy: 0.9496\n",
      "Epoch 3/50\n",
      "1470/1470 [==============================] - 4s 3ms/step - loss: 0.1308 - accuracy: 0.9538 - val_loss: 0.1196 - val_accuracy: 0.9573\n",
      "Epoch 4/50\n",
      "1470/1470 [==============================] - 4s 2ms/step - loss: 0.1190 - accuracy: 0.9583 - val_loss: 0.1092 - val_accuracy: 0.9601\n",
      "Epoch 5/50\n",
      "1470/1470 [==============================] - 4s 2ms/step - loss: 0.1059 - accuracy: 0.9623 - val_loss: 0.1195 - val_accuracy: 0.9562\n",
      "Epoch 6/50\n",
      "1470/1470 [==============================] - 4s 3ms/step - loss: 0.0938 - accuracy: 0.9661 - val_loss: 0.1323 - val_accuracy: 0.9498\n",
      "Epoch 7/50\n",
      "1470/1470 [==============================] - 4s 3ms/step - loss: 0.0841 - accuracy: 0.9695 - val_loss: 0.1136 - val_accuracy: 0.9590\n",
      "Epoch 8/50\n",
      "1470/1470 [==============================] - 4s 3ms/step - loss: 0.0749 - accuracy: 0.9722 - val_loss: 0.1259 - val_accuracy: 0.9591\n",
      "Epoch 9/50\n",
      "1470/1470 [==============================] - 4s 3ms/step - loss: 0.0659 - accuracy: 0.9759 - val_loss: 0.1230 - val_accuracy: 0.9591\n"
     ]
    }
   ],
   "source": [
    "#Compile and fit the model on the reweighted train set\n",
    "model.compile(optimizer=\"adam\", loss=\"binary_crossentropy\", metrics=[\"accuracy\"])\n",
    "early_stopping = EarlyStopping(monitor=\"val_loss\", patience=5)\n",
    "history = model.fit(X_train_rw, y_train_rw, batch_size=32, epochs=50, validation_data=(X_test, y_test), callbacks=[early_stopping])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Plot the training and validation accuracy and loss curves\n",
    "plt.figure(figsize=(12,6))\n",
    "plt.subplot(1,2,1)\n",
    "plt.plot(history.history[\"accuracy\"], label=\"train accuracy\")\n",
    "plt.plot(history.history[\"val_accuracy\"], label=\"validation accuracy\")\n",
    "plt.xlabel(\"Epoch\")\n",
    "plt.ylabel(\"Accuracy\")\n",
    "plt.legend()\n",
    "plt.subplot(1,2,2)\n",
    "plt.plot(history.history[\"loss\"], label=\"train loss\")\n",
    "plt.plot(history.history[\"val_loss\"], label=\"validation loss\")\n",
    "plt.xlabel(\"Epoch\")\n",
    "plt.ylabel(\"Loss\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "365/365 [==============================] - 0s 758us/step\n",
      "Test accuracy: 0.9590522791655937\n",
      "Test Precision: 0.9597188892697978\n",
      "Test Recall: 0.9585687382297552\n",
      "Test confusion matrix:\n",
      "[[5573  235]\n",
      " [ 242 5599]]\n"
     ]
    }
   ],
   "source": [
    "#Evaluate the model on the test set\n",
    "y_pred = model.predict(X_test).round().ravel()\n",
    "test_acc = accuracy_score(y_test, y_pred)\n",
    "precision = precision_score(y_test, y_pred)\n",
    "recall = recall_score(y_test, y_pred)\n",
    "test_cm = confusion_matrix(y_test, y_pred)\n",
    "print(\"Test accuracy:\", test_acc)\n",
    "print(\"Test Precision:\", precision)\n",
    "print(\"Test Recall:\", recall)\n",
    "print(\"Test confusion matrix:\")\n",
    "print(test_cm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Create a binary label dataset from the test set for aif360\n",
    "test_df = pd.DataFrame(X_test.reshape(-1, 64*64))\n",
    "test_df[\"label\"] = y_test\n",
    "test_df[\"gender\"] = 0\n",
    "test_dataset = BinaryLabelDataset(favorable_label=1, unfavorable_label=0, df=test_df, label_names=[\"label\"], protected_attribute_names=[\"gender\"], unprivileged_protected_attributes=[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Compute the classification metrics for the test set\n",
    "#metric_classifier = ClassificationMetric(test_dataset, test_dataset.copy(), unprivileged_groups=[{\"gender\": 0}], privileged_groups=[{\"gender\": 1}])\n",
    "#print(\"Accuracy:\", metric_classifier.accuracy())\n",
    "#print(\"Balanced accuracy:\", metric_classifier.balanced_accuracy())\n",
    "#print(\"Equal opportunity difference:\", metric_classifier.equal_opportunity_difference())\n",
    "#print(\"Average odds difference:\", metric_classifier.average_odds_difference())"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "tf",
   "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": "3.9.16"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}