{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e4fc6dc9-8a30-4e4d-b419-2f2c10f5e5c7",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from torch.utils.data import DataLoader, TensorDataset\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "import matplotlib.pyplot as plt\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "80e073d1-1180-421b-ace5-ba7a5cdf7859",
   "metadata": {},
   "source": [
    "# Load Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "4134a135-76fa-424b-b887-cd8786b22301",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load Data\n",
    "data = pd.read_csv('data/task2/diabetes_binary.csv', header=None)\n",
    "X = data.drop(7, axis=1).values  \n",
    "y = data[7].values \n",
    "\n",
    "# Normalize the features\n",
    "scaler = StandardScaler()\n",
    "X_scaled = scaler.fit_transform(X)\n",
    "\n",
    "# Split Data\n",
    "X_train, X_test, y_train, y_test = train_test_split(X_scaled, y, test_size=0.3, random_state=42)\n",
    "\n",
    "# Convert to PyTorch tensors\n",
    "train_dataset = TensorDataset(torch.tensor(X_train, dtype=torch.float32), torch.tensor(y_train, dtype=torch.long))\n",
    "test_dataset = TensorDataset(torch.tensor(X_test, dtype=torch.float32), torch.tensor(y_test, dtype=torch.long))\n",
    "\n",
    "# Specify a batch size\n",
    "batch_size = 16\n",
    "\n",
    "train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=True)\n",
    "test_loader = DataLoader(test_dataset, batch_size=batch_size, shuffle=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b4866035-59d1-4513-8c56-be703191b600",
   "metadata": {},
   "source": [
    "# 1-1. CNN Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c44bcdc5-4f51-4818-ad12-dc36752a3c0b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define the CNN Model\n",
    "class CNN(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(CNN, self).__init__()\n",
    "        self.conv1 = nn.Conv1d(1, 32, 3)  # Assuming each data point is treated as a 1D \"image\" with one channel\n",
    "        self.fc1 = nn.Linear(32 * 5, 128)  # Adjust the dimension accordingly\n",
    "        self.fc2 = nn.Linear(128, 1)  # Binary classification\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = torch.relu(self.conv1(x))\n",
    "        x = x.view(x.size(0), -1)  # Flatten the output for dense layer\n",
    "        x = torch.relu(self.fc1(x))\n",
    "        x = torch.sigmoid(self.fc2(x))  # Sigmoid for binary classification\n",
    "        return x\n",
    "\n",
    "\n",
    "# Initialize the CNN\n",
    "model = CNN()\n",
    "criterion = nn.BCEWithLogitsLoss()  # Binary Cross Entropy loss\n",
    "optimizer = optim.Adam(model.parameters(), lr=0.001)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "725582b4-44f7-456a-8158-5f3097b1702f",
   "metadata": {},
   "source": [
    "# 1-2. CNN Train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "01d6edaa-9b9d-49f2-a49e-20e06486d455",
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_accuracy(model, loader):\n",
    "    model.eval()\n",
    "    correct = 0\n",
    "    with torch.no_grad():\n",
    "        for data, target in loader:\n",
    "            data = data.unsqueeze(1)\n",
    "            output = model(data)\n",
    "            pred = output.argmax(dim=1, keepdim=True)\n",
    "            correct += pred.eq(target.view_as(pred)).sum().item()\n",
    "    accuracy = 100. * correct / len(loader.dataset)\n",
    "    return accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "30d68047-971a-4a77-b6d5-2b37f57e170b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0/30, Loss: 0.6387, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 1/30, Loss: 0.6382, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 2/30, Loss: 0.6380, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 3/30, Loss: 0.6378, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 4/30, Loss: 0.6376, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 5/30, Loss: 0.6375, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 6/30, Loss: 0.6374, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 7/30, Loss: 0.6374, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 8/30, Loss: 0.6373, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 9/30, Loss: 0.6375, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 10/30, Loss: 0.6372, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 11/30, Loss: 0.6371, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 12/30, Loss: 0.6371, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 13/30, Loss: 0.6370, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 14/30, Loss: 0.6370, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 15/30, Loss: 0.6368, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 16/30, Loss: 0.6369, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 17/30, Loss: 0.6368, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 18/30, Loss: 0.6368, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 19/30, Loss: 0.6368, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 20/30, Loss: 0.6368, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 21/30, Loss: 0.6368, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 22/30, Loss: 0.6367, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 23/30, Loss: 0.6366, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 24/30, Loss: 0.6367, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 25/30, Loss: 0.6366, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 26/30, Loss: 0.6365, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 27/30, Loss: 0.6365, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 28/30, Loss: 0.6365, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n",
      "Epoch 29/30, Loss: 0.6365, Training Accuracy: 50.01%, Test Accuracy: 49.99%\n"
     ]
    }
   ],
   "source": [
    "epochs = 30\n",
    "train_losses = []\n",
    "test_accuracies = []\n",
    "\n",
    "for epoch in range(epochs):\n",
    "    model.train()\n",
    "    running_loss = 0.0\n",
    "    for data, target in train_loader:\n",
    "        data = data.unsqueeze(1)  # Add channel dimension\n",
    "        target = target.float()  # Convert target to Float\n",
    "        optimizer.zero_grad()\n",
    "        output = model(data)\n",
    "        output = torch.squeeze(output)  # Remove extra dimension\n",
    "        loss = criterion(output, target)\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        running_loss += loss.item()\n",
    "    train_loss = running_loss / len(train_loader)\n",
    "    train_losses.append(train_loss)\n",
    "    \n",
    "    train_accuracy = compute_accuracy(model, train_loader)\n",
    "    test_accuracy = compute_accuracy(model, test_loader)\n",
    "    test_accuracies.append(test_accuracy)\n",
    "    \n",
    "    print(f'Epoch {epoch}/{epochs}, Loss: {train_loss:.4f}, Training Accuracy: {train_accuracy:.2f}%, Test Accuracy: {test_accuracy:.2f}%')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "123e138c-0026-4f90-bf7f-315de7a3f2c5",
   "metadata": {},
   "source": [
    "# 1-3. CNN Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "71e16790-cc64-4c3a-96f7-40b280b49c97",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnYAAAHVCAYAAAB8NLYkAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAABzGUlEQVR4nO3dd3hU1b7G8e/09N6AJHRpCthQRFCsiKKooAh6sKEo2NAjYKMognotiA1FrIgKgtLsiHoULCigdJGSQkjvZZKZff8IjkZqQsJOeT/PM8/J7Fmz9m/2neu8rL3X2hbDMAxEREREpMGzml2AiIiIiNQOBTsRERGRRkLBTkRERKSRULATERERaSQU7EREREQaCQU7ERERkUZCwU5ERESkkbCbXUBjVlFRwa+//kpsbCxWqzK0iIiIWbxeL3v27OH444/Hbm+88afxfrJ64Ndff6VHjx5mlyEiIiJ7/fjjj5x88slml1FnFOzqUGxsLFD5JWrWrJnJ1YiIiDRdu3fvpkePHr7f5sZKwa4O/XX6tVmzZsTHx5tcjYiIiDT2S6Ma96cTERERaUIU7EREREQaCQU7ERERkUZCwU5ERESkkVCwExEREWkkFOxEREREGgkFOxEREZFGQsFOREREpJFQsBMRERFpJBTsRERERBoJBTsRERGRRkLBTkRERKSRULATERERaSQU7EREREQaCbvZBUj1lG7aRP6SJdhj44i45mqzyxEREZF6RCN2DYz7zz/JmvUqeUsWm12KiIiI1DMKdg2Mq2NHAMq2bMXweEyuRkREROoTBbsGxtmyJRY/P4ySEsqTkswuR0REROoRBbsGxmKz4WrfHoDSTZtNrkZERETqEwW7BsivYwcASjdvMrkSERERqU8U7BogV4e919lpxE5ERET+QcGuAfLrcAygETsRERGpSsGuAXJ1qDwVW5G6G09ensnViIiISH2hYNcA2UJCcDRvDkDpZp2OFRERkUoKdg2Ubz07XWcnIiIieynYNVCaGSsiIiL/pmDXQGlmrIiIiPybgl0D9deIXdnWrRgVFSZXIyIiIvWBgl0D5UhIwBoQgOF2496xw+xyREREpB5QsGugLFYrrmP2rmen07EiIiKCgl2D5vrrdKwmUIiIiAgKdg2a394lTzRiJyIiIqBg16D9dQeKsk0asRMREREFuwbN75hjwGKhIiODiuxss8sRERERkynYNWDWwEAciQmARu1EREREwa7B8+ug6+xERESkkoJdA6eZsSIiIvIXBbsGzq/DX/eM3WJyJSIiImI2BbsGznfP2G3bMNxuk6sRERERMynYNXCOFs2xBgdDeTll27ebXY6IiIiYSMGugbNYLLg6VN5aTDNjRUREmjYFu0ZAM2NFREQEFOwaBc2MFREREQC72QVkz5lD9quzqcjMxNWxI3EP3I9/164HbO/JzyfjmWfI//xzvLl5OJo3J/a+8QSdcQYAOXPnkjP3XcpTUgBwtWtH1KhbCerTx9dHRUYGe554gqLvV+ItKsLZuhVRN48k5PzzfG3+OOtsylNTq+w7eswYom4aUZsfv1b8856xhmFgsVhMrkhERETMYGqwy1+2jPRpjxE3cSL+3bqS/cab7LpxBG0/XoY9MnKf9obbza7rb8AWGUH89OnYY2IpT03BFhLia2OPjSPm7jE4W7bEMAzyPvyIpFGjabPgA1zt2wOQOnYcnoICEl54Hlt4OPlLlpBy110458/Dr3NnX19Rt99G+ODBvufWwMA6PBo152rfHqxWPNnZVGRk4IiJMbskERERMYGpp2KzXn+DsMGDCbv8Mlzt2hE3aSJWPz9yP1iw3/a5Cxbgycsj4bnnCDjhBJzxLQjs0cM3YgUQfFZfgs44A2erVrhatybmrjuxBgRQsnatr03xmjVEXD0M/65dcSYkEHXLLdiCgylZv77K/myBgdijo30Pa0DAQT9PWVkZ+fn5vkdBQcERHJ3DZ/Xzw9mqVWUNm3WdnYiISFNlWrAz3G5K168n8LSevm0Wq5XAnj0pWbNmv+8pWL4c/+7dSZv8MFt6nc6fAwaQ+dJMDI9n//vweMhbuhSjuBj/7t192wO6dyd/2cd4cnMxvF7yli7F63YT2KNHlfdnvjKLLaecyp+XXkbWq69iVFQc9DNNnTqV0NBQ36PzP0b/6prf3uvsSjUzVkREpMky7VRsRU4ueDzY/nXK1RYVecD12MqTkile9QMhAy4iYeZMynftJG3SZIyKCqJHj/K1K928hR1XXYVRVoY1IID452bgatfO93qLZ54m5a4xbDm1J9jtWP38iJ8xA2fLlr424ddcg1/nztjCQin59VfSn3qaivQMYsePO+BnGj9+PGPGjPE9T0lJOWrhztWhIyz7mDLNjBUREWmyTJ88US1eL7bISJpNnozFZsP/2C6U70kna/arVYKdq3Ur2ixcgKegkIJPPyV13HhavvWmL9xlTH8WT0EBia/NxhYeTsEXX5Jy1120fPtt/PauCRd53bW+/vw6dMDicLB7wkSi7x6D1encb3kulwuXy+V7np+fXwcHYf98I3aaGSsiInLEnv58C9O/3FplW5voQJbffSa5xW6e/nwL327NJCW3hMhAJ+d1iWPMeccQ4ucwqeJKpgU7e3gY2Gx4srKqbPdkZmGPitr/e6KjwWHHYrP5trnatsGTkYnhdmPZG7gsTqdv9M3/2C6U/P4b2W++RbPJk3Dv2kXOnDm0WbzIN5nCr2NHilf/TM4779Bs0sT97tu/a1eoqKA8OQVXm9ZH+Olrn2vvdYbu7TvwlpVh/UfAFBERkeo7JjaIt288xffcbq28gm1Pfhl78su4r38n2scGkZJTwv0f/s6e/FJevPpEs8oFTLzGzuJ04telC0UrV/m2GV4vRatWVbke7p/8TziB8p27MLxe3zb3jh3Yo6N9oW6/vIbvPqrektLKbdaqH91itcE/+v230k2bwGrFHhlxiE9mDntMDLawMPB4KNv6h9nliIiI1EsFBQVVJjqWlZUdsK3NaiUm2M/3iAiszBod4oJ56ZoTOadzLC0jAzmtXRT3nNeBLzemU+E5cJY4GkydFRt57XBy580jd+GHlG3bRtrESXhLSgi77FIAUseOJf3Jp3ztw68agicvjz1THqVs+3YKVqwgc+bLhA8b6muT/uRTFP/0E+7kFEo3b6l8/uOPhAy4CABXm9Y4Wiaye8IEStatw71rF1mzX6Po++8JPudsAIp//ZXsN96gdNMm3ElJ5C1ezJ6p0wgdMABbaOhRPEKHz2Kx+EbttFCxiIjI/nXu3LnKRMepU6cesO2OzCJ6TPmC3o8v5453fyUlt+SAbQtKywnys2O3mXvvB1OvsQvp35+K7BwyZjyLJyMTV6dOJL7ysu9UbHnqbrD8fYAczZqRMOsV9kybRu4lA7HHxhJxzTVEjrjR16YiO4vUseOoyMjAGhyMq8MxJMx6haBevQCwOBwkzpxJ+pNPkXTLrXiLi3EmJtJ82lTfIscWp5O8ZcvIeO55DLcbR3w8EcOHE/GP6+7qI78OHShetUq3FhMRETmADRs20KJFC99z1wEuXeqeGMb/De5Gm+hA0gvKmP7FFq54aSWf3tWHIFfV+JRd5GbG8j+4qkdCndZ+OCyGYRhmF9FYJScnk5CQQFJSEvHx8XW+v9wFC9l9330EnHwyLd96s873JyIi0lAc6W9yXkk5p09bzgMXdeLKkxN92wtKy7n61R8J83cwa/hJOEwesdO9YhuRv2fGVt5aTERERGpHqL+D1tGB7Mgq9m0rLKtg+OwfCXLZmHnNiaaHOlCwa1Sc7dqB3Y43P5+K3bvNLkdERKTRKCqrYGdWMTHBladuC0rLuebVH3DYrMz6z8n4OWyH6OHoULBrRKxOJ67WlUux6Do7ERGRmpuydAOr/swiKbuY1Tuzufmt1disFi7u1nxvqPuREreHxwd1paCsnPSCUtILSvF4zT1j1rAWKJZDcnXsSNnWrZRt2UzwWX3NLkdERKRB2p1Xyu1zfyW3uJyIQCcntQpn4a2nERnkYuW2LNYk5QJwxhMrqrzv23v7khBx8HvL1yUFu0bGr2MH8hcv1oidiIjIEXhu6AkHfK1n20h2TLvwKFZz+HQqtpFxddi7lt0mrWUnIiLS1CjYNTJ/zYx179yJt7j4EK1FRESkMVGwa2TsUVHYoqLAMCjbuvXQbxAREZFGQ8GuEfLrsHc9O11nJyIi0qQo2DVCrr2nY3XPWBERkaZFwa4R8utYOYFCI3YiIiJNi4JdI+Tq8NeI3WYMr9fkakRERORoUbBrhFytW2NxOPAWFVGekmJ2OSIiInKUKNg1QhaHA2f7dgCUaj07ERGRJkPBrpHy8y1UrOvsREREmgoFu0bK1eEYAEo1M1ZERKTJULBrpP6aGasROxERkaZDwa6R+mtmbHlyMp7CQpOrERERkaNBwa6RsoeHY4+NBSqXPREREZHGT8GuEfvrDhSaGSsiItI0KNg1YpoZKyIi0rQo2DVifn+N2OlUrIiISJOgYNeIuf6aGbtlC4bHY3I1IiIiUtcU7BoxZ8uWWPz8MEpLce/aZXY5IiIiUscU7Boxi82Gq317QDNjRUREmgIFu0bOTzNjRUREmgwFu0bOpZmxIiIiTYaCXSOnmbEiIiJNh4JdI/fXrcUqdu/Gk5trbjEiIiJSpxTsGjlbcDCOFi0AKN28xeRqREREpC4p2DUBvvXsNmsChYiISGOmYNcE+HU4BoBSTaAQERFp1BTsmoC/Z8ZqxE5ERKQxU7BrAv6aGVv2xx8YFRUmVyMiIiJ1RcGuCXAkJGANCMBwu3Fv3252OSIiIlJHFOyaAIvViusYXWcnIiLS2CnYNRGuv07HamasiIhIo6Vg10T47V3yRCN2IiIijZeCXRPx1x0oSjViJyIi0mgp2DURfsccAxYLnoxMKrKyzC5HRERE6oCCXRNhDQzEkZgAQKnWsxMREWmUFOyaED/fQsW6zk5ERKQxUrBrQnwzY7co2ImIiDRGCnZNiGbGioiING4Kdk2IX6dOAJRt3Yp7xw5zixEREZFap2DXhDiaNSPojDPA6yXj+RfMLkdERERqmYJdExN1+20A5C9ZQtnWrSZXIyIiIrVJwa6J8e/SheDzzgPDIGPGc2aXIyIiIrVIwa4Jir5tNFgsFHz2GSXr15tdjoiIiNQSBbsmyNW+PSEXXQRA5rMzTK5GREREaouCXRMVPepWsNko/Pprin/91exyREREpBYo2DVRzlatCL10IAAZzz5rbjEiIiJSKxTsmrDoW24Bh4PilasoWvWD2eWIiIjIEVKwa8IcLVoQPngwUDlqZxiGyRWJiIjIkVCwa+Iib74Zi8tFyS+/UPS//5ldjoiIiBwBBbsmzhEbQ/jQoQBkPDNdo3YiIiINmIKdEDniRqwBAZSuX0/BF1+YXY6IiIjUkIKdYI+IIHz4fwDIfPZZDI/H5IpERESkJhTsBIDI667DGhJC2dY/yP/4E7PLERERkRpQsBMAbCEhRF5/HQCZM2ZgVFSYXJGIiIhUl4Kd+IRffQ228HDcO3eS99Eis8sRERGRalKwEx9bUCCRI0YAkPn88xhut8kViYiISHUo2EkV4VcNwR4dTXlqKrkffGB2OSIiIlINCnZShdXfn8iRNwOQ+eJLeEtLTa5IREREDpeCnewjbPBg7M2bUZGeTs6775pdjoiIiBwmBTvZh9XpJPrWWwHIevkVvEVFJlckIiIih0PBTvYr9JJLcLRMxJOdTfZbb5tdjoiIiBwGBTvZL4vDQfTo0QBkzZ6NJz/f5IpERETkUOxmFyD1V0j//mS9/DJlW/8g+/XXib79drNLEhEROSqe/nwL07/cWmVbm+hAlt99JgDv/LCLj9aksD41n8KyCtZOOI9Qf4cJlValYCcHZLHZiLrtNlJuv4Ps198g/JprsIeHm12WiIjIUXFMbBBv33iK77nd+veJzpJyD2d0iOaMDtE8/slmM8rbLwU7Oajgc8/F1bkTZRs2kjVrFrH//a/ZJYmIiNRYQUEB+f+4vMjlcuFyufbb1ma1EhPst9/Xbji9NQArt2XVfpFHQNfYyUFZLBZi7rgDgJw571Cenm5yRSIiIjXXuXNnQkNDfY+pU6cesO2OzCJ6TPmC3o8v5453fyUlt+QoVlozpo/YZc+ZQ/ars6nIzMTVsSNxD9yPf9euB2zvyc8n45lnyP/8c7y5eTiaNyf2vvEEnXEGADlz55Iz913KU1IAcLVrR9SoWwnq08fXR0VGBnueeIKi71fiLSrC2boVUTePJOT88/7eT24uaY9MofCrr8BqJfi8c4m77z6sgYF1dCTqr8A+ffDv3p2SNWvIevkV4h643+ySREREamTDhg20aNHC9/xAo3XdE8P4v8HdaBMdSHpBGdO/2MIVL63k07v6EOQyPT4dkKkjdvnLlpE+7TGiRo2i9YIP8OvQgV03jqAia//Dmobbza7rb8CdkkL89Om0+fhj4h6ejD021tfGHhtHzN1jaP3BfFrNn0fAqaeSNGo0ZVv/vgAydew43Nt3kPDC87RZ9BEh555Lyl13Ubphg69Nyn/vpeyPP0ic/SoJL71I8c8/s/uhCXV3MOoxi8VC9J2Vo3a5771H+R6N2omISMMUHBxMSEiI73GgYNe3QwwXdm1Gp2YhnHFMNK9d14P8knKWrks9yhVXj6nBLuv1NwgbPJiwyy/D1a4dcZMmYvXzI/eDBfttn7tgAZ68PBKee46AE07AGd+CwB498OvY0dcm+Ky+BJ1xBs5WrXC1bk3MXXdiDQigZO1aX5viNWuIuHoY/l274kxIIOqWW7AFB1Oyfj0AZdu2UfTttzR7+GH8u3Uj4MQTiXvgAfKXLWuyoSbw1FPx794do7yc/KVLzS5HRETkqAr1d9A6OpAdWcVml3JQpgU7w+2mdP16Ak/r6dtmsVoJ7NmTkjVr9vueguXL8e/enbTJD7Ol1+n8OWAAmS/NxPB49r8Pj4e8pUsxiovx797dtz2ge3fyl32MJzcXw+slb+lSvG43gT16AFCyZg3WkBD8jzvW957Anj3BaqVk3dp/78anrKyM/Px836OgoKAaR6T+C73kYgDyliw2uRIREZGjq6isgp1ZxcQE73+Er74w7SRxRU4ueDzYIiOrbLdFRVK2fft+31OelEzxqh8IGXARCTNnUr5rJ2mTJmNUVBA9epSvXenmLey46iqMsjKsAQHEPzcDV7t2vtdbPPM0KXeNYcupPcFux+rnR/yMGThbtqysLSMTe0RElX1b7HZsoaF4MjMP+JmmTp3KpEmTqnsoGozgfv1Im/IoZRs2UrZtG662bc0uSUREpE5MWbqBszvF0iLMn/SCUp7+fCs2q4WLuzUHIL2glIyCMnZmVd52c3NaAYEuGy3C/AkLcJpWd/29+m9/vF5skZE0mzwZi82G/7FdKN+TTtbsV6sEO1frVrRZuABPQSEFn35K6rjxtHzrTV+4y5j+LJ6CAhJfm40tPJyCL74k5a67aPn22/h1OKbG5Y0fP54xY8b4nqekpNC5c+eaf956xh4eTtDpp1O4YgV5ixcTc+edZpckIiJSJ3bnlXL73F/JLS4nItDJSa3CWXjraUQGVY7YzVm1q8oCxlfMXAnAE4O6MvikBFNqBhODnT08DGw2PP+aKOHJzMIeFbX/90RHg8OOxWbzbXO1bYMnIxPD7cbirEzIFqfTN/rmf2wXSn7/jew336LZ5Em4d+0iZ84c2ixehKt9ewD8OnakePXP5LzzDs0mTcQeHUVFdnaVfRsVFXjy8rAdoDbYdy2c/EZ4G66QARdRuGIF+UuWEn3HHVgsFrNLEhERqXXPDT3hoK/fde4x3HVuzQeD6opp19hZnE78unShaOUq3zbD66Vo1aoq18P9k/8JJ1C+cxeG1+vb5t6xA3t0tC/U7ZfXwHC7K/8sKa3cZq360S1WG+zt1797d7z5+ZT8vt73etGqH8Drxb9rt+p8zEYnuG9fLAEBlCcnH/BaSBERETGHqbNiI68dTu68eeQu/JCybdtImzgJb0kJYZddCkDq2LGkP/mUr334VUPw5OWxZ8qjlG3fTsGKFWTOfJnwYUN9bdKffIrin37CnZxC6eYtlc9//JGQARcB4GrTGkfLRHZPmEDJunW4d+0ia/ZrFH3/PcHnnF3Zpm1bAnv3ZvdDD1Kybh3Fv/zCnocfJqR/fxyxMUfxCNU/1oAA33HKX7zE5GpERETkn0y9xi6kf38qsnPImPEsnoxMXJ06kfjKy75TseWpu8Hyd/Z0NGtGwqxX2DNtGrmXDMQeG0vENdcQOeJGX5uK7CxSx46jIiMDa3Awrg7HkDDrFYJ69QLA4nCQOHMm6U8+RdItt+ItLsaZmEjzaVN9ixwDtHjicdIefoRd1163d4Hi84i7/76jdGTqt9ABA8hftJj8jz8mdvw4LA7zb3osIiIiYDEMwzC7iMYqOTmZhIQEkpKSiI+PN7ucWmNUVLC1zxl4srNJeHlmlbt6iIiI1EeN9Tf533SvWKk2i91OyAUXAJCn07EiIiL1hoKd1Ejo3msWC778Em9x/V6FW0REpKlQsJMa8evWDUdiIkZxMQVfLje7HBEREUHBTmrIYrEQetGFAOQv0elYERGR+kDBTmos5KLK07GF331HRU6OydWIiIiIgp3UmKtNG/w6d4aKCvI//tjsckRERJo8BTs5IiEDBgBarFhERKQ+ULCTIxLSvz9YLJT8+ivu5GSzyxEREWnSFOzkiDhiYwg49RQA8pcsNbkaERGRpk3BTo5Y6N5JFHlLFqMbmYiIiJhHwU6OWPB552FxOnH/sY2yTZvMLkdERKTJUrCTI2YLDibozDMB3WJMRETETAp2UitC9t5iLH/pUgyv1+RqREREmiYFO6kVQX36YA0OpmLPHop/+tnsckRERJokBTupFVaXi+DzzwMgf8lik6sRERFpmhTspNaEXrR3seJPP8PrdptcjYiISNOjYCe1JuDkk7DHxuLNz6fom2/MLkdERKTJUbCTWmOx2SrvRIFmx4qIiJhBwU5qVeje2bGFX32Fp6DA5GpERETqp6c+30JyTnGt96tgJ7XK1akTzrZtMdxuCj773OxyRERE6qXPN+zhjCdWMPSVVXy0JoWyCk+t9KtgJ7XKYrH4Ru3yl+p0rIiIyP58fEdvPhrVi2Nig5m0eAMnP/IF9y/8jbVJuUfUb7WDXe7CDylYscL3fM8TT7D55B7sGHIV5SkpR1SMNA4he+8dW7TqB8rT002uRkREpH46tkUoEy/uwg/3nc3jg7qSllfKoJe+p98z3zD7f9vJLy2vdp/VDnZZM2di9fMDoPjXX8l5Zy4x99yDLTycPdOmVbsAaXyc8fH4d+8OXi/5y5aZXY6IiEi9ZhhQ7jFwe7wYBoT4O3hz5Q5Om7qcxWtTq9WXvbo7L09Lw5mYCEDhl18Sct65hF95BQEnHM/O/wyvbnfSSIUMuIiSNWvIX7yEyGuvNbscERGReue35DzmrU5i0dpUnDYrl50Qz8OXHEurqEAAXv9uO5MWr2dAt+aH3We1R+ysAQF4cnMBKPzuewJPOw0Ai8uFt6ysut1JIxVywQVgs1G6fj1lf243uxwREZF65fynv+HSF74jKbuYxy7vysrxZzPugo6+UAdwcfcWZBVVb8H/ao/YBZ52GrsfeBBX5064d+wgsE8fAMr++ANni8NPlNK42SMiCDy9F0Vff0P+kiVE336b2SWJiIjUGxd2bcYVJyUQF+p3wDYRgU62T72wWv1We8Qu7qEH8e/eHU92DvHPTsceHg5A6e/rCbmwejuXxi107ySKvCVLMAzD5GpERETqj9vPbn/QUFdT1R6xs4WEEPfQg/ts14iM/FvwWWdh8fenfNcuStetw79bN7NLEhERqRdGvrWabglh3HJm2yrbX/p6G+uSc3lh2Ik16rfaI3aF335L8erVvufZc+bw58BLSbn7Hjx5eTUqQhona2AgwWefDUDekqUmVyMiIlJ//Lgjm74do/fZfmaHaH7cnl3jfqsd7NIffwJvYSEApZu3kP7Y4wT16UN5cjJ7pj1W40KkcfItVrxsGUZFhcnViIiI1A9FZRU4bPvGMLvVSkFpzX8vqx3s3CkpONu2A6Dgs88IOvNMYsbcRexDD1L47bc1LkQap8DTTsMWHo4nK4uilavMLkdERKRe6BgXzJK1u/fZvnhtKu1jg2rcb7WvsbM4HBilJQAUrVxJ6CWXAGALDfON5In8xeJwEHJBP3LemUv+ksUE9T7d7JJERERMd9tZ7Rn59mp2ZhdxWtsoAL7/I5NFa1N5ftgJNe632iN2ASecwJ5pj5HxwguU/PYbQWeeAYB7xw4csbE1LkQar5CLBgBQ8PkXeEtKTK5GRETEfOd0juXl/5zIzqxiHvzwd6Ys3cDuvFLevvEUzu8SV+N+qz1iF/fgA6RNmkzBp5/RbMJDvjBX9O03BPbuXeNCpPHyP747jvh4ypOTKfj8c0IvvtjskkREREx3VsdYzupYu4Ni1Q52jubNSZj50j7bY8ePr5WCpPGxWCyEXnYpmc/OIOvV2YQMGIDFYjG7LBERkUan2sEOwPB4KPjiS9x/bgPA2a5d5ZplNlutFieNR8TQoWTPepWyzZsp/Pprgs880+ySRERETOPxGrz6vz9Zum43KbmllHu8VV5fO+G8GvVb/VmxO3fyZ/8LSR03jvzPPyf/889JvXcsf140APeuXTUqQho/W1gYYUOGAJD18ismVyMiImKu6V9sYda327moa3MKSsu58fTW9OsSh9UCd57Tvsb9VjvYpU2ZgiMxkfZfLafNggW0WbCAdsu/xBEfT9qUKTUuRBq/iOHDsTgclPzyC8U//2x2OSIiIqb5cE0q0y4/jhF92mC3Wri4e3MeG9SV289uz6+7cmvcb7WDXfFPPxNzzz3YwsJ82+zh4cTcPYbin/RjLQfmiI0h9LLLAMic+bLJ1YiIiJgno6CMDnEhAAS47L5Fic/uGMtXm9Jr3G+1g53F6cRbVLTPdm9xMRaHo8aFSNMQecP1YLVS9O23lG7YYHY5IiIipmgW6kd6fikALSMC+GZrBgBrk3Nx2qsdz3yq/c7gM88gbcJDlKxdi2EYGIZByZo1pE2YSHDfvjUuRJoGZ2IiIf37A5Cpa+1ERKSJOq9LHN9vywJg+GmteOqzLZz5xFfc/f5aBp+UUON+LYZhGNV5gyc/n9Rx4yn86iss9spJtYbHQ9BZfWn+6KPYQkJqXExjk5ycTEJCAklJScTHx5tdTr1RunkL2y+5BCwW2ixbiqt1a7NLEhGRRq6+/yb/siuHX3bm0CoykHM613xtu2ovd2ILCSHhhedx79xJ2bY/AXC1bYOzZcsaFyFNi1+HYwjq25fCr74ia9YsmmvSjYiINCHlHi/3LfiN289uT0JEAAAnJIZzQmL4Efdd45O4zpYtCT6rL8Fn9cXZsiWlmzez8biuR1yQNA2RN40AIG/RYsrT0kyuRkRE5Ohx2Kx88nvd/PbV/Oq8fzMM8HhqrTtp3AKOP56Ak0+G8nKyX3vN7HJERESOqnO7xPLp+toPdzW684RIbYi8+WaKf/qJnPfnETlyJPbwIx+CFhERaQhaRwby7JdbWb0zh2NbhBLgrHr3rut61ez6cwU7MU1gr9Pw69yZ0g0byHnrLaJvv93skkRERI6K935OIsTfwW8pefyWklflNYvlKAQ7T2HhQV/f39p2IgdjsViIvPlmUu64g+y35xBx/fXYgoLMLktERKTO/W/sWXXS72EHuy0n96iMkAdiGAd/XWQ/gs89B2fr1ri3byf3vfeIvOEGs0sSERFpsA472CW+8XodliFNlcVqJXLECHbfdx9Zr79O+NVXY3W5zC5LRESkTv133tqDvv7E4G416vewg11gjx412oHIoYRedCEZM2ZQsXs3eQsXEj5kiNkliYiI1Km8kvIqzyu8BpvTCsgvLee0tpE17leTJ8R0FqeTyOuvZ8+UKWTNepWwQYN8dzURERFpjF7+z0n7bPN6De7/8HdaRgbUuN/aW8dO5AiEDbocW0QE5cnJ5H/8sdnliIiIHHVWq4Ube7fm1f9tr3kftViPSI1Z/f2J+M81AGS9/AqG12tyRSIiIkffrqxiPF6jxu/X+S6pN8KHDiXrlVmUbd1K4YoVBJ9VN1PBRUREzPbwkg1VnhsGpBeU8tWmdC4/Mb7G/SrYSb1hCwkhfOhVZL0yi8yZMwnq2xeLltAREZFGaH1q1UWJrRYLEYFO7r+wM1ecdBSDXdLo0Qf4sbVgcblwJiYSctFFuNrUbMVkadoihg8n+823KF27juIffiTw1FPMLklERKTWvXtTzzrpt9rX2NmCgila9QOl6zcAFsBC6YaNFP3wA4angvyPP2b7wIEU//JL7VcrjZ49Koqwyy8DIOvll02uRkREpG4kZRezPXPfu3ZtzywiKbu4xv1WO9jZo6MIuehC2n7xOfEzniV+xrO0/fwzQgcMwJmQSJtlSwkdOJD0/3uyxkVJ0xZx/Q1gs1H0/feU/Pa72eWIiIjUurvnrWX1zpx9tq9JyuGeQyxefDDVDna58z8g4j//wWL9+60Wq5Xwq4eR+/77WCwWwocNo2zr1hoXJU2bM74FoRddCGjUTkREzPH051toNW5plcdZT67wvV5a7uHBD3+n++TP6PzQJ4x8azUZBWWH3f+G1HxOahm+z/bjE8LZsDu/xnVX+xo7w+PBvX07rtZVr6Fzb9/uW6LC6nLqvrFyRCJvvJG8jxZR8MUXlG3bhqttW7NLEhGRJuaY2CDevvHva73t/xjUenjJBr7alM4LQ08g2M/BQ4t+Z+Tbq/ngltMOq28LUFhWsc/2gtIKvEdzuZPQiy9m9/0P4L75JvyPOw6Akt9+I2vmy4RecgkART/9hKtduxoXJQdmGAYl5R6zy6h7LVtjO+c8ir5aQfIrs2k2eZLZFYmIyFHk77DVycoIBQUF5Of/PSLmcrlwHeAe5TarlZhgv32255eW8/7PSUwfcjyntYsC4IlB3Tjnqa/5ZVcOJyTuOxL3bz1aR/Diim08e9Xx2KyVn9PjNXhhxR+c1CqiJh8NqEGwix0/DntUJFmvvoonMwsAW1QkEcOHEzniRgCCevUiqHfvGhclB1ZS7qHzQ5+aXcbREXQeDDiv8u+m8plFRASADZPPJ8BZ+6uyde7cucrzCRMmMHHixP223ZFZRI8pX+ByWDkhMZx7+3WkRZg/vyfnUe4x6LU31AG0iwmiRZg/v+w8vGA37oKOXDFzJWc9uYKT9wa5n3ZkU1hawTsjTq3x56v2EbPYbESNHEnUyJF4CgsBsAUFVWnjaN68xgWJiIiI1JUNGzbQokUL3/MDjdZ1Twzj/wZ3o010IOkFZUz/YgtXvLSST+/qQ0ZhGU6blVB/R5X3RAU5ySg8vOvs2scG88mdfXjj+x1s3J2Pn8PGZcfHM/y0loQFOGv8+Y4oCv870End83fY2DD5fLPLOGqKfviR5JG3YHE5abN0KfbImg9Pi4hIw+HvsNVJv8HBwYSEhByyXd8OMb6/OzWD7glhnD5tOUvXpeJXS7XFhvhxb7+OtdLXX6od7CoyM9nz+OMUr1xFRXZ25T0w/qHThvW1Vpzsy2Kx1MnQdH3lf3pPCrt0pHTdOvIffZjmjz+O1d/f7LJERKSJCfV30Do6kB1ZxfRuF4Xb4yWvpLzKqF1moZvooP2PAP7b+z8nEei0c2HXZlW2L123m5JyD4NqeFuxaieE1PH3Ub47lahbb8EeHa3Zr1KnLBYLMXfewa4RN1Hw+RfsGHIV8c/NwJmQYHZpIiLShBSVVbAzq5hLj3dxbHwoDpuF7//I5ILjKoPZtoxCUnJLOGE/S5jsz4srtjHl0mP32R4Z5OS+Bb8dvWBXsno1Lee8jV+nTjXaoUh1BZ52GomvzSblrjGUbd7M9ssH0eL/niCoTx+zSxMRkUZqytINnN0plhZh/qQXlPL051uxWS1c3K05IX4OrjgpgUeWbiQ0wEGwy8GERb9zQmLYYU2cAEjJLSEhPGCf7S3C/EnJLalx3dUOdvZmzfY5/SpS1wJ79KD1B/NJvuMOSteuI+nmkUTffhuRN99cZbFsERGR2rA7r5Tb5/5KbnE5EYFOTmoVzsJbTyNy76nWBy/qjNWykVve/gV3hZc+x0Tx8MB9R+AOJCrQyaa0AhIiqoa7jbvzCT+CyRMWw6heSiv833dkv/YacZMm4Yxvceg3NGHJyckkJCSQlJREfHzNhlSlKq/bzZ4pj5L73nsABJ19Ns2nTcUWHGxyZSIiUp/Vt9/kqR9vZMna3TwxuCuntI4E4Ic/s/jv/HX0Py6O+y/sfIge9q/awW5zj1MwSkowPB6sfn7gqDrVt8MPq2pUSGNU375EjUnu/PmkTZqMUV6Os1Ur4p+boUWxRUTkgOrbb7K7wstd769h2W+7se9doNhrwGXHt+CRS4/FZa/ZzNsaLFA8vkY7EqlNYYMG4erQgeTbbse9Ywfbr7iS5o8+Ski/prMUjIiINFxOu5Xnh57A9swiNqTm4+ew0iEumPj9XHdXHdUesasL2XPmkP3qbCoyM3F17EjcA/fj37XrAdt78vPJeOYZ8j//HG9uHo7mzYm9bzxBZ5wBQM7cueTMfZfylBQAXO3aETXqVt/F9u7kFLadc85++27xzNOE9OsHwMaO+04Qaf7k/xF64YWH9bnq278OGqOKrCxSxtxN8Q8/ABB54w1E33knFnvTWRJGREQOrSH8JheUlvPhmlTe/ymJxbedXqM+DuvXz1NY6FuM+K+7TRxIdRctzl+2jPRpjxE3cSL+3bqS/cab7LpxBG0/XoY9MnKf9obbza7rb8AWGUH89OnYY2IpT03B9o/FBu2xccTcPQZny5YYhkHehx+RNGo0bRZ8gKt9exzN4mj/7TdV+s15/32yX529z63Qmj36KEG9/z641sNY1FCOHntkJImvziL9qafJnj2brFmvUrphA82ffBJ7+OHNTBIRETHT99symfdzMp/8nkawn53zu8TVuK/DCnZbepxC+2+/wR4ZyZaTe+x/7TrDAIul2gsUZ73+BmGDBxN2+WUAxE2aSOHXX5P7wQKibhqxT/vcBQvw5OXRau47WPZe3/fvSRzBZ/Wt8jzmrjvJefddStauxdW+PRabrXINvn8o+OJLgi/ohzUwsMp2W0jwPm2lfrHY7cTe+1/8j+1C6gMPUvT9SrZffjnxz87A/9guZpcnIiKyj7S8UuavTmLe6mTyS8rJKyln+pDjuahrMyxHsEbwYQW7xNdfwxYaWvn3G6/XeGf/ZrjdlK5fXyXAWaxWAnv2pGTNmv2+p2D5cvy7dydt8sMULF+OPSKckAsvInLEjVhs+15oaHg85H/yCUZxMf7du++3z5Lf11O2cSNxDz64z2tpkx9m9wMP4khIIHzIlYRedtkBD3hZWRllZX/fI66goOAgn15qW0j//jjbtSP5ttso37mLnUOHEjdpEmGXDjS7NBEREQA+/m037/2cxI/bszmzQzT39+/EmR1i6PzQJ3SMCz6iUAeHGewCe/TY799HqiInFzwebP865WqLiqRs+/b9vqc8KZniVT8QMuAiEmbOpHzXzsrZkRUVRI8e5WtXunkLO666CqOsDGtAwEFnTeZ+MB9n27YEnHB8le1Rt99G4KmnYvXzo/C770ibNBlvUTER/7lmv/1MnTqVSZMmVeMISG3zO+YYWs+bR+q9YylcsYLd48dT+ts6YseNw+Ks+bpAIiIitWH03F8ZeUYbnht6AkGu2r8evEY9evLzKVn3G57sLAyvt8prYQMH1kZdB+b1YouMpNnkyVhsNvyP7UL5nnSyZr9aJdi5WreizcIFeAoKKfj0U1LHjaflW2/uE+68paXkL1lK1C237LOr6Ftv9f3t17kzRkkJWbNnHzDYjR8/njFjxviep6Sk0LlzzdahkZqzhYQQ/8LzZL74IpnPPU/OO3Nx79xFwsyXNKlCRERMdcVJCby5cier/szm0uNbMKBrc0IDHId+42Gq9q9cwfKvSP3vf/EWF2MNCqp6vZ3FUq1gZw8PA5sNT1ZWle2ezCzsUVH7f090NDjsVU67utq2wZORieF2+0ZlLE4nzpYtAfA/tgslv/9G9ptv0Wxy1RG1gk8/xVtaSujASw5Zr1/XrlS88CJetxvrfkZ/XC4XLtffN//Nz88/ZJ9SNyxWK9GjRuHXpQspY+6m6LvvSH/iCS3XIyIippp62XFMGNCZJet28/7PSUxesoE+7aMxqFzH7khVO9ilP/YYoZdfRsxdd2H19z+inVucTvy6dKFo5SqC9y4/Yni9FK1aRfiwYft9j/8JJ5C/ZAmG1+u7lZR7xw7s0dEHP9XmNTDc7n02587/gOC+fbFHRByy3rJNm7CGhu431En9FHzmmTR/bBopt91O9htv4urUqe5HlUVERA7Cz2Fj0InxDDoxnu2ZRcz7OYnfUnIZ9OL39O0YQ//j4uh3bLMa9V3tm2yWp6cTcc01Rxzq/hJ57XBy580jd+GHlG3bRtrESXhLSgi77FIAUseOJf3Jp3ztw68agicvjz1THqVs+3YKVqwgc+bLhA8b6muT/uRTFP/0E+7kFEo3b6l8/uOPhAy4qMq+3Tt3Uvzzz4QNHrRPXQXLvyJn3jxKt2zBvXMnOXPnkjnzZSIOEDil/go591yi9p5WT3toAiXr1plckYiISKXWUYHc268jK8edzdNXdqek3MPtc9fUuL9qj9gFnd6L0t9/x5mQUOOd/lNI//5UZOeQMeNZPBmZuDp1IvGVl32nYstTd4Pl7/zpaNaMhFmvsGfaNHIvGYg9NpaIa64hcsSNvjYV2Vmkjh1HRUYG1uBgXB2OIWHWKwT16lVl37kfLMAeF0fgv7YDWBx2ct6ZS/rUaRiAMzGR2LFjCbticK18bjm6okaPonTzZgq//JLk226n9fx5WsZGRETqDavVwjmdYzmncyyZhWWHfsMBVPvOE7nz55PxwguEXXoZrmOOweKomg2DzzqrxsU0Ng1hleumxFNYyI4rh+Detg3/448n8Y3XG+VpdU9uLulPP0NI//4EnlJ7s9hFRBqypvKbXO1gt7HTQWZ51mCB4sasqXyJGpK/7ivrzc8nbPAg4iZPPuI1g+qbtMmTyXlnLo7ERNp+8rHvWlQRkaasqfwmV/tUbKeNG+qiDpGjwtmqFS2e/D+Sbh5J7rz5uDp1ImLo0EO/sYEoT0khZ978yr937aL4xx8JPPVUk6sSEZGjRf+UlyYnqHdvYsbcBcCeR6dS/NNPJldUezJfegnKy33LEOW+P8/kikRE5Gg6rGCX/eZbePfeKiv7zbcO+hBpCCJuuIGQCy+EigqS77iT8tRUs0s6Yu6dO8ldsBCA2AfuB6Dg88+pyMkxsywREdmP3o8vJ6do32XY8krK6f348hr3e1inYrPfeIOQARdhdbnIfuONAze0WA54VwaR+sRisdDskYcp2/4nZRs2kjR6NK3mzKm1ZXzMkPnCC+DxENinNxHDhpG3YCGl69eT9+FHRF53rdnliYjIPyTnlODZzzQHd4WXPXk1nxV7WMGu3Zdf7PdvkYbM6u9PwowZbB80mLING9n9wIM0/78nGuRkirJt28hbvASA6NtuByBs8GDS1q8nd948Iq4d3iA/l4hIY/P5hj2+v7/ZkkGw39+3E/N4Db7flkl8eM0HGXTjTGnSHC1a0GL6M+y6/gbyly7Fr3MnIm+4weyyqi3juefA6yXo7LPxP+5YAEIuupA9jz2G+88/KVm9moCTTjK5ShERuemtnwGwAHfPW1vlNYfVSny4P/df2KnG/dco2JWnpVGwfDkVu3djuMurvBY7flyNixExQ2CPHsSOH8eehx8h/cmncB3TgaDep9e4P8MwKFmzhvzFi6nIziFuwkPYw8NrseKqSjdvpuDjTwCIvv0233ZbUBAhF/Ynb/4H5M6bp2AnIlIPbJ96IQCnP7acRaNPJyKwdtdTrXawK1q5kqRbR+GMj6ds+3Zc7dtTnpIChoFf54OscSdSj4UPHUrpxo3kzf+AlLvvpvW893G2bFmtPtxJSeQtWkTeokWU79zl225UlBM/Y0adnQrNmDEDgOAL+uHXoUOV18IHDyZv/gfkf/Ipsffdhy00tE5qEBGR6vnf2H1v6JBXUk6ov2M/rQ9ftZc7SX/qaSKvu442ixdhdTqJf3Y67b9aTsDJJxPS7/wjKkbELBaLhbiHHsK/e3e8+fkkjRqFp7DokO/z5OWR89777Bh2NdvOPY/MGc9RvnMXFn9/gi/oBw4HhV98Sd6CBXVSd8nv6yn84kuwWokePXqf1/26dsV1zDEYZWXkLVpcJzWIiEj1vbhiG4vX/r0iw61zVtN98mec8ugXbEjNr3G/1Q527m3bCB14SeUTux2jtBRrYCDRt99G1iuzalyIiNmsTictnp2OPSYG9x/bSB07FsPr3aed4XZTsHw5yXfcydbTe5M2YQIlq1eDxULgaT1pNm0qx/zvW+Kfftp3anTPlEdxJyXVes0Zz04HIHTARbjatt3ndYvFQtgVVwCQO28e1bzRjIiI1JE5P+ykeZgfAN9uzeB/WzN547oenHlMDFM/3ljjfqsd7CwBARjlldfV2aOjq/xYVeTm1rgQkfrAERND/HMzsDidFH75JZnPvwDsvW5u3TrSHn6ErX3OIPnWURR8+ilGeTmu9u2Juedu2n21nMTZswkbOBBrYCAAkddfj/9JJ+ItLib13rEYFRW1Vmvxr79S9M23YLMRdeutB2wXOuAiLC4XZVu2ULpuXa3tX0REai6joIxmoZWzX7/cmM6FXZvT55hobj6jDWuTcmvcb7WvsfPv1o3i1atxtW1LUJ8+7HnsMcq2bKHgs8/x79a1xoWI1Bf+XbsSN2kSu8ePJ/P556nIyqR41Q+4d+zwtbFFRxF64UWEXnIxro4dD3j9nMVmo/m0x9h+ySWU/PorWbNmETVyZK3UmfHsswCEXjrwoNcD2kJDCel3PnkfLSLn/ffx79atVvYvIiI1F+rvYHdeCc3D/PlmSwZ3n1d5jbQBeI/g5Eq1R+xix431/TBE3zaawFN7kr/sYxwtWtD8kUdqXolIPRJ26UDC9y62nfvue7h37MDi50fIRReR8MrLtP/qK2LHjcWvU6dDTopwxrcg9sEHAMh47nlKfvv9iOsr+uFHileuAoeD6FtuOfTn2Xs6Nn/Zx3gKC494/yIicmT6HRvH7XPXcPWsH8gpdnNmh2gA1qfm0zIyoMb9VmvEzvB4qEhLw7V35p01IIBmkybWeOci9VnsvffiLS6mYk86If37E3zuudiCAmvUV+gll1D41QoKPv2U1HvvpfWCD2p8lwvDMHyjdeGDB+Fo0eKQ7/E/4QScbdvi3raN/CVLCR9yZY32LSIitePBizoTH+5Pam4p4y7oSKCrMpKl55dyzanVW5XhnyxGNa+m3tS1G22WLcUZH1/jnTYVycnJJCQkkJSURLyOV5NXkZPD9osvoSIjg/ChQ4l76MEa9VP4v+9IuvFGLE4nbT//DEds7GG9L+u110l/7DH8Onem9YIParRvEZGGqqn8Jlf7VKyrfXvK62B2n0hjZw8Pp9nUqQDkvPMOhd9+W+0+qozWXTXksEMdQOjAS7A4HJRu2EDJ+vXV3reIiNSuBb8kM+jF7+kx5QuSc4oBePV/2/lsfVqN+6x2sIu+8w72PP4EBV99RXl6Op7CwioPETmwoNN7EX711QCk3ncfFTk51Xp/4VcrKF23Dou/P5EjRlTrvfbwcILPPReoXPpERETM89aqnTyydCNndogmv7Scv1bXCvGzM/u77TXu97CDXcbzz+MtLibpppsp27SJ5FtH8ceZfdnS45TKx8k92NLjlBoXItJUxNxzN862bfFkZJL20ITDXlvO8Hp9d5mIuHoY9qioau877IrBAOQvXoK36NALMIuISN144/sdTL3sOEaf1R7bPybhdY0PY3NaQY37PezJE5nPv0D4kCEkvvF6jXcmImD186P544+x48ohFHz+OXkLPyTssksP+b6Cz7+gbONGrIGBRFx/fY32HdCjB46WiZTv3EX+J58QdvnlNepHRESOTFJ2MV2ah+yz3Wm3Uuz21Ljfwz8Vu3dUIbBHj4M+ROTQ/Lt0Ifq2v+5KMQV3cvJB2xseD5nP7R2tG/4f7OHhNdqvxWolbNAgAHLf1+lYERGzJEQE7PfWYV9vTqddTFCN+63eNXZ1dBNzkaYo8sYb8D/xRLxFRZV3pfAc+F9o+cs+pmzrH1hDQoi49toj2m/YpZeC3U7J2rWUbt5yRH2JiEj1TP9iKyVuDzee3pqHPlrP4rWpGMCa5FyeW76Vxz/dzM1n7HuLyMNVrXXstvW74JDhrsMPq2pcjEhTYrHZaP7YNLZfMpCSX34ha9arRN180z7tjIoKMp97DoDI66/DFrLv0H112KOiCD7rLAo++4zcefOIe+D+I+pPREQO3/QvtzDs1ESG9EjEz2Hjyc82U1Lu4Y53fyU22I8JAzpzcbfmNe6/WsEuevRorMHBNd6ZiFTljI8n9v772X3ffWTMmEHg6b3w79KlSpu8RYtx79yJLSyM8KuvqZX9hg0eTMFnn5G3aBEx99yN1c+vVvoVEZGD++d0uYHHt2Dg8S0ocXsoclcQFeQ64v6rFexCLuyPPTLyiHcqIn8LvXQghStWUPDZZ6TeO5bWH8z3BS3D7SbzhRcAiBxxY43vfPFvgb1Ow9G8OeWpqRR8+imhl1xSK/2KiMih/fvcp7/Thr/TVit9H36w0/V1InXCYrEQN2kiJb/+invbNtL/70nf6dHcBQspT07GFhVF+NChtbdPq5WwwYPImP4sOfPmKdiJiBxFff9vxSHvM752wnk16vvwg1317jwmItVgDw+n2aOPkjRiBDlvv03QmWcScPJJZL70EgBRN91U43vLHkjoZZeRMeM5Sn5eTdmff+Jq06ZW+xcRkf2769xjCPZz1Enfhx3sOm3cUCcFiEiloN6nEz5sGDlz5rB7/HjCrriCirQ07HFxhF15Ra3vzxEbS9AZZ1D41Vfkvj+P2HFja30fIiKyrwHdmtfK9XT7U+1biolI3Ym5526cbdpQkZFB5vPPAxA18masrrr5D8Bfd6LI+/BDvG53nexDRET+VtcXtinYidQjVn9/mj/xONgrB9MdLVoQdtlldba/oN69scfG4snNpfCLL+psPyIiUqmuL2xTsBOpZ/y7dCH23nvB4SBm7L1YnM4625fFbifs8srgmKM7UYiI1LntUy+ss9OwoGAnUi9F/OcaOq5bS8h5NZsVVR1hl18OFgvFq1bh3rWrzvcnIiJ1R8FOpJ461FT42uJo0YLA008HIHfe/KOyTxERqRsKdiJC2OBBAOQuXIhRXm5yNSIiUlMKdiJCcN++2KKi8GRmUvDVV2aXIyIiNaRgJyJYHA7CLr0UgFxNohARabAU7EQE+Pt0bNF33+FOTjG5GhERqQkFOxEBwJmYSEDPU8EwyFvwgdnliIhIDSjYiYhP+ODKO1HkzHmHsm3bTK5GRESqS8FORHyCzzkHvy5d8OTlseva63Dv3Gl2SSIiUg0KdiLiY3E6SZj1Cq727anIyGDndddRnppqdlkiInKYFOxEpAp7eDiJs1/F2aoVFam72XntdZTvSTe7LBEROQwKdiKyD3t0NImvv4YjPp7yXbvYdf31VGRnm12WiIgcgoKdiOyXIy6OxNdfwx4bi3vbNnZdfwOevDyzyxIRkYNQsBORA3LGx5P4+mvYoqIo27SJXSNuwlNYaHZZIiJyAAp2InJQrtatSZz9KrawMErXrSPp5pF4i4vNLktERPZDwU5EDsnvmGNIeHUW1uBgSlavJmnUKLxlZWaXJSIi/6JgJyKHxb9LFxJfeRlrQADFK1eRcvsdGG632WWJiMg/KNiJyGHz796d+JdexOJyUfj116Tc81+MigqzyxIRkb0U7ESkWgJ79CD++eexOBwUfPYZqePvw/B4zC5LRERQsBORGgg6vRctpk8Hu538xYtJmzgRwzDMLktEpM68sOIPWo1byqTF633bdmYVcdObP3PCw59z7IRPGTXnFzIKzL3+WMFORGok+Ky+tHjicbBayZ03nz2PTq1RuPMUFFC6cSP5n31GwYoVOrUrIvXO2qRc3vlhFx3jgn3bit0VXPPqj1gs8M6IU5h/S0/cHi83vvETXq95/9C1m7ZnEWnwQi64AG9ZGbvHjSfnrbew+rmIHjMGi8Xia+N1uylPSaE8OYXylGTcSUmVfycn405OxvuvRY9DLryQ5o8/hsVmO9ofR0SagIKCAvLz833PXS4XLpfrgO2Lyiq48701TLusKzOWb/Vt/3lHDsk5xSy9/XSC/RwAPHlFN7pN+ozvt2VxevuouvsQB6FgJyJHJGzgQIzSMtImTiTrlVmUp+3BYrPhTq4McBV79sAhRvJsERE4WrSoHLlbuhSLy0WzRx7GYtVJBRGpXZ07d67yfMKECUycOPGA7R/86Hf6dojh9PZRVYKdu8KLxWLBaf/7v1MuuxWrxcJPO7IV7ESk4QofciVGWSl7pk4jf/HifV63+PvjjI/HsffhTNj7d4t4nPEtsAYGApD/yaekjBlD3oIFWFxO4h56qMron4jIkdqwYQMtWrTwPT/YaN2itamsT8nno9G99nnt+MQwAhw2pn28iXvP74iBwWMfb8LjNUg38To7BTsRqRURw4djDQqmZM0aHC1aVAlwtoiIwwpoIf3OxyifRuq9Y8md+y5Wlx8xY+9VuBORWhMcHExISMgh26XmljB58XreuuEU/Bz7XhoSGeTi+WEn8MCHv/P69zuwWixc3K05x7YIwWrif7Ishqay1Znk5GQSEhJISkoiPj7e7HJEGozc+fPZ/cCDAESOvJmYO+80tyARafCq+5v86fo0bn5rNbZ/pDSP18BiAavFwpZHLvC9ll3kxma1EOrv4KRHvmBE79bcfEbbOvssB6MROxGpd8IGDcJbVsaehx8h66WZWP38iBo50uyyRKQJ6dUuik/v7FNl23/nr6VtdBAjz2hbJfBFBDoB+P6PTLKKyjinc+xRrfWfFOxEpF6KGDYMo7SM9CeeIOOZ6VhcfkRed63ZZYlIExHkstPhH8ubAPg7bIQFOHzb3/85iXYxQUQGOvllVw6TFm/ghl6taRsdZEbJgIKdiNRjkTdcj7eslMxnZ5D+2GNYXE4ihg41uywREQD+zCji8U82k1fiJj48gNF923HD6a1NrUnX2NUhXWMncuQMwyDjqafJeuUVAJpNmULY5ZeZXJWINDRN5TdZi0SJSL1msViIHnMX4f+5BoDdDzxA3pKlJlclIlI/KdiJSL1nsViIHT+esCuvBMMgdexY8j/7zOyyRETqHQU7EWkQLBYLcRMeInTgQPB4SLn7Hgq//trsskRE6hUFOxFpMCxWK82mPEJI/wugvJzk226n6PvvzS5LRKTeULATkQbFYrPR/LHHCDrnbAy3m6RbR1H8889mlyUiUi9ouRMRaXAsDgctnnqK5FGjKfr2W5JuupnE12bj363bftsbFRV4i4rwFhbiKSzCW1T4j+eFGCUlBPbujau1ucsUiIgcKQU7EWmQrE4n8TOeJWnkLRSvWsWuG0cQcPLJeAv/Edr2/q9RWnrI/pxt3qPN0iW6L62INGgKdiLSYFn9/Eh44Xl23TiCkl9+oXD58oO2t7hcWIOCsAYGYg0KxBYYhDUoiKJVq3D/+Sclq1cTcNJJR6l6EZHap2AnIg2aNSCAxFmvkP/pZxjlbqyBgdiCKgNbZYgLwhoYgC0wEIvTud8+Uu+/n7wPFpA7/wMFOxFp0BTsRKTBswYEEHbpwBq/P+zyQeR9sID8Tz4h9v77sAUHH/pNIiL1kGbFikiT5398d5xt22KUlpK/dJnZ5YiI1Fi9CHbZc+bwx1lns6lrN7ZfcSUl69YdtL0nP5+0yZPZ0rs3m47ryrbz+1VZqDRn7lz+vPgSNp94EptPPIkdVw6h8JtvfK+7k1PY2LHTfh/5n3zia1eemsqum29mU/fj2XJaL/Y8/gRGRUXtHwARMZXFYiHs8ssByP3gA5OrERGpOdNPxeYvW0b6tMeImzgR/25dyX7jTXbdOIK2Hy/DHhm5T3vD7WbX9Tdgi4wgfvp07DGxlKemYAsJ8bWxx8YRc/cYnC1bYhgGeR9+RNKo0bRZ8AGu9u1xNIuj/bffVOk35/33yX51NkG9e1fux+Mh6eaR2KOjaDX3HSoyMkgdOw6L3U7MmLvq9qCIyFEXesnFpD/9NKW//Ubp5s34dehgdkkiItVm+ohd1utvEDZ4MGGXX4arXTviJk3E6udH7gcL9ts+d8ECPHl5JDz3HAEnnIAzvgWBPXrg17Gjr03wWX0JOuMMnK1a4Wrdmpi77sQaEEDJ2rVA5QKn9ujoKo+CL74k+IJ+WAMDASj67jvKtm2j+eOP49epE0F9+hB9x+3kvPMOhttd9wdGRI4qe2QkwX37ApA7X6N2ItIwmRrsDLeb0vXrCTytp2+bxWolsGdPStas2e97CpYvx797d9ImP8yWXqfz54ABZL40E8Pj2f8+PB7yli7FKC7Gv3v3/bYp+X09ZRs3Enb5oL+3rVmD65hjsEdF+bYFnn463sJCyv74Y7/9lJWVkZ+f73sUFBQc4giISH0SNqjydGzeokV4y8pMrkZEpPpMDXYVObng8WD71ylXW1QkFZmZ+31PeVIyBZ9+iuH1kDBzJlG33EL2a6+R+eJLVdqVbt7CphNOZFPXbqRNnET8czNwtWu33z5zP5iPs21bAk44/u/aMjL3ORX81/MD1TZ16lRCQ0N9j86dOx/084tI/RLYqxf2uDi8eXkUfPGF2eWIiFSb6adiq83rxRYZSbPJk/E/tgsh/fsTOXIkOe+9W6WZq3Ur2ixcQKv33iN8yBBSx43f70ibt7SU/CVLfRdOH4nx48eTl5fne2zYsOGI+xSRo8disxF22aUA5GkShYg0QKYGO3t4GNhseLKyqmz3ZGZVOQVa5T3R0ThbtcRis/m2udq2wZORWeXaN4vTibNlS/yP7ULM3WNwdexA9ptv7dNfwaef4i0tJXTgJf/aTxQV/6rrr+cHqs3lchESEuJ7BGstLJEGJ/Syy8Fioej7lbiTk80uR0SkWkwNdhanE78uXShaucq3zfB6KVq16oDXw/mfcALlO3dheL2+be4dO7BHRx9wVXkAvMZ+Jz3kzv+A4L59sUdEVN1P9+6UbdlSJdwVffc91qAgnAc4pSsiDZ8zvgWBPU8FIG/B/idxiYjUV6afio28dji58+aRu/BDyrZtI23iJLwlJb7TIaljx5L+5FO+9uFXDcGTl8eeKY9Stn07BStWkDnzZcKHDfW1SX/yKYp/+gl3cgqlm7dUPv/xR0IGXFRl3+6dOyn++WfCBg/i3wJ79cLVti2p946ldNMmCr/9HxnTpxM+dCjWgwVIEWnwQv9a027BwgNOzBIRqY9MX8cupH9/KrJzyJjxLJ6MTFydOpH4ysu+053lqbvB8nf+dDRrRsKsV9gzbRq5lwzEHhtLxDXXEDniRl+biuwsUseOoyIjA2twMK4Ox5Aw6xWCevWqsu/cDxZgj4sj8F/bofJam4SXXmT3pEnsGHIVVn9/QgcOJPr22+roSIhIfRF8zjnYQkOpSEuj6LvvCOrTx+ySREQOi8UwDMPsIhqr5ORkEhISSEpKIj4+3uxyRKQa0qY8Ss5bbxF83nnEPzvd7HJE5Ag1ld9k00/FiojUR3+taVewfPk+E6lEROorBTsRkf3w69ABv+OOg4oK8j5aZHY5IiKHRcFOROQA/lrfMveDD9BVKyLSECjYiYgcQMhFF2Lx98e9bRslv64xuxwRkUNSsBMROQBbUBAh558PVN56sLYZFRV4cnNrvV8RaboU7EREDuKvdS7zP/4ET2FRrfXrKSxixxVXsqV3H/IWL661fkWkaVOwExE5CP8TTsDZujVGcTH5Hy+rlT6NigpS7h5D6YYNUF5O6r1jyXn//VrpW0SaNgU7EZGDsFgshF1+GQB58z+olT73PPY4RV9/g8XlIrhfPzAM0h6aQPabb9ZK/yLSdCnYiYgcQugll4DdTsnatZRt3XpEfWXPmUPOW28B0Pyxx2jx9FNE3HA9AHsenUrmzJePuF4RaboU7EREDsEeHU3QmWcAkHsEo3aF337LnimPAhB9112E9Dsfi8VCzD33EDV6NAAZTz9N+jPPaHkVEakRBTsRkcMQNqhyEkXeokV43e5qv790yxZS7rwLvF5CBw4k8qYRvtcsFgvRo0cR8997AMh6aSbp06Yp3IlItSnYiYgchqDTT8ceE4MnJ4fC5cur9d6KzEySR96Ct6iIgJNPptnkSVgsln3aRd5wA7EPPgBA9htvkjZxEobXWyv1i0jToGAnInIYLHY7oZdeClTvdKy3tJSkUaMoT03F2bIlLZ6djsXpPGD7iGHDaDZlClit5L73HrvHj8eoqDji+kWkaVCwExE5TH/Nji367jvKU1IO2d7wetl9332Url2HNTSU+JdexB4eflj7af7E42CzkffRIlLG3I1Rg9O/ItL0KNiJiBwmZ2IiAaecAoZB7sIPD9k+87nnyF/2MdjtxD/7LK7WrQ97X6EXXkj8s9OxOBwUfPYZybfdjres7AiqF5GmQMFORKQa/ppEkbvgAwyP54Dt8j76iMwXXgSg2aRJBJ7So9r7Cj77bOJfeAGLy0Xh11+TNHIk3uLimhUuIk2Cgp2ISDUEn3sO1pAQKlJ3U7Ry1X7bFK9eze4HHgQgcsSNvlO4NRHU+3QSXnkZa0AAxStXsevGEXgKCmrcn4g0bgp2IiLVYPXzI3TAAAByP5i/z+vuXbtIHjUao7yc4HPPJfquu454n4E9epD42mysISGU/PILu669joqcnCPuV0QaHwU7EZFqCht0OQCFX3xZJWB58vJIGnkLntxc/I49luaPP4bFWjv/mfXv1o2Wb7yOLTyc0vXr2fWf4VRkZtZK3yLSeCjYiYhUk1+nTvh17oxRXk7+okUAGOXlJN9xJ+4//8QeF0f8C89j9fev9f22fOtN7NHRlG3dys5hV5P95psU/fAjntzcWt2XiDRMdrMLEBFpiMIGDyJt0mRy539A+H/+Q9rkhyletQprQAAJL72IIyamTvbrateOlm+/xc7rrsO9cyd7Hp3qe80eF4dfhw64OnbEr2MHXB064myZiMVmq5NaaoNhGBhuN0ZZGd7S0qp/l7kx3FX/toWFE9jrtFobCRVpbCyG7llTZ5KTk0lISCApKYn4+HizyxGRWuTJz2dr7z4YZWWEXnIxeR8tAquV+OefI7hv3zrff3l6Ornvz6N040bKNm064Lp6Fn9/XO3b7w18HfDr2BFXhw7YgoLqvMZ/K9+zhz2PTqX4l9WVQa2sDKMGS7iEDhxIs0cexmLX2IQcvqbym6z/rxARqQFbSAgh/c4n76NFlaEOiB039qiEOgBHTAzRo0f5nnsKCijbvJnSTZsp27yp8n+3bsUoKaF03TpK162r+v6EBMIGDSLyumsPeieM2mAYBvmLF5P2yBS8+fkHbmixYPHzw+p0YnG5Kv92ObE4K/+2OBwU//QTeR9+iKewgBZPPonV5arT2kUaGo3Y1aGm8q8Dkaaq6Mcf2fWf4QCEXTWEuIce2u89YM1ieDy4d+6kbFNl0CvdvImyTZup2LPH18bZpg1xDz1I4Kmn1kkNFVlZpE2cSMHnXwDgd9xxxN77X2yRkVhdrr8DnNMJDschj1/B8uWk3HkXhttNwKmnEv/cc9iCAuukdmlcmspvsoJdHWoqXyKRpsowDPY8/DCG10vcAw80mFODFTk5FK74mvT/+z88WVkAhFx4ITFj763VawPzP/2MtIkT8eTkgMNB9KhbibzxxiM+TkWrfiD51lvxFhfj17UriS/PxBYWVjtFS6PVVH6TFezqUFP5EolIw+TJzydj+rPkzJ0LXi/WwECi77iD8KFXHVH48uTmkvbIFPKXLAHA1aEDzadNxa9Tp9oqnZLffiPpxhF48vJwtW9HwqxXccTWzYQVaRyaym+yphWJiDRRtpAQ4h58gFbvv4/fccfhLSpiz6OPsn3wFZSsWVOjPgu//po/B1xcGeqsViJvvplW896v1VAH4H/ccbR8+y3sMTGUbf2DnVdfjTspqVb3IdIQKdiJiDRx/sd2odW7c4mbOBFraChlGzeyY8hV7H7wwcO+w4WnsJDUBx4g6eaRVGRk4GzdmlZz3yHmrjsrr5+rA6727Wn5zhwciYmUJyWxc+gwSrdsqZN9iTQUCnYiIoLFZiN8yJW0/XgZoZdV3ts2d958/rygP7nz52N4vQd8b9GqVfx58cXkzf8ALBYihg+n9cIF+HfrVud1O+Pjafn2W7iOOYaKjAx2XvMfStaurfP9itRXCnYiIuJjj4ig+aNTaDnnbVzt2+PJzWX3Aw9WjoZt2lSlrbe4mLSHH6m8d23qbhzx8bR88w1ix4/D6ud31Gp2xMTQ8s038O/WDW9eHjuvu56ilSuP2v5F6hMFOxER2UfAiSfSesEHxIwdizUggJI1a9h+2eWkPfoonsJCin/5lT8vvZScOXMACBtyJW0++pCAk082pV5bWBiJs18l8LSeGMXFJN10MwVffGFKLSJm0qzYOtRUZuCISONWnpbGnmmPUfDJJwDYwsMr701rGNhjY2n2yCME9T7d3CL38rrdpN59DwWffw5WK82mTCHs0oFmlyX1QFP5TdaInYiIHJQjLo74Z54mYdYsHC0TK9elMwxCL7mENosX1ZtQB2B1Omnx9FOEXnopeL3sHj+e7DffMrsskaOmYaymKSIipgs6vRdtFi0ib+FCHC3i61Wg+yeL3U6zKY9gCwkh+4032PPoo3jy84kadWu9ujOISF1QsBMRkcNmdbkIHzLE7DIOyWK1EjNuLNbQEDKfnUHmc8/hyc8jdtw4LFadrJLGS99uERFplCwWC9G33krs/fcDkPPmW2S9+qrJVYnULQU7ERFp1CKuuZrYBx8AIGvWq3gKC02uSKTuKNiJiEijFz5kCM42bfDm5ZHz9hyzyxGpMwp2IiLS6FlsNqJuGQlA9muv4SksMrkikbqhYCciIk1CSP/+OFu1wpOXR8477xyVfRpeL+nTp5P54osYbvdR2ac0bQp2IiLSJFQZtZs9G29R3Y/a5X34EVkvvkTG9GfZdf0NVGRl1fk+pWlTsBMRkSYj5MILKxdZzs0lZ+7cOt2Xp7CQ9KeeqnxitVL8889sHzSYkvXr63S/UjdeWPEHrcYtZdLiv//vl15Qyl3vreGkR76g04OfcOGz3/Lxb7tNrFLBTkREmhCL3U7UyFsAyJr9Gt7i4jrbV9bMmXgyM3G2bEnrhQtwtmpFxe7d7Bx2NXlLltbZfqX2rU3K5Z0fdtExLrjK9rvfX8ufGYXMGn4Sn97Zh35d4hj1zi/8npJnUqUKdiIi0sSEDrgIR2Iinuxscua+Wyf7cO/cSfbrbwAQM24sfh060Or99wjs0xujtJTUe+4h/f/+D8PjqZP9y4EVFBSQn5/ve5SVlR20fVFZBXe+t4Zpl3Ul1N9R5bXVO3MYfloruieEkRgZwG1ntyfE36FgJyIicrRY7Haibr4ZgKzZs/GWlNT6PvY89jhGeTmBvXsTdOaZANhCQkh48UUiR4yo3PesV0kaeQuePPNCQFPUuXNnQkNDfY+pU6cetP2DH/1O3w4xnN4+ap/XTmwZzpJ1u8ktduP1Gixam0pZuZdT20TWVfmHpGAnIiJNTujFA3DEx+PJyiLnvfdqte/C/31H4fLlYLcTO25slfvTWmw2Yu4eQ/Mn/w+Lnx9F337LjiuupGzbtlqtQQ5sw4YN5OXl+R7jx48/YNtFa1NZn5LPvf067Pf154aeQLnHS/fJn3PMAx9z/4LfmHnNibSKCqyr8g9JwU5ERJoci8NB5M03AZD16qt4S0trpV+jvJw9e0eAIoYNxdW27X7bhV54Ia3emYO9eTPcO3ey44orKVj+Va3UIAcXHBxMSEiI7+FyufbbLjW3hMmL1/PMkO74OWz7bfPUZ5vJL61gzo2nsGj06dzQuzWj3vmFTWn5dfkRDkrBTkREmqSwSy7B0bw5noxMct9/v1b6zJn7Lu5t27CFhxM1atRB2/p17kzr+fMJOOkkvEVFJI8aVbnenWHUSi1yZH5LySOz0M1FM/5H2/uW0fa+ZfywPZvXv99B2/uWsTOriDdW7uSJQV3p1S6Kzs1DuPOcY+gaH8qbK3eaVreCnYiINEkWp5PIv661e2UW3kNcRH8oFTk5ZDz3HADRd96JLSTkkO+xR0SQ+NpswocOBcMgY/qzpNxx51FZY08Orle7KD69sw/Lbu/te3SND2Vg9xYsu703JeWVE1+slqrvs1ospoZzBTsREWmywi4diL15MyoyMsh9f94R9ZUxfTre/HxcHTsSNujyw36fxeEg7qEHiXt4MjgcFHz2GTuuGoo7OfmI6pEjE+Sy0yEuuMrD32EjLMBBh7hg2kYH0SoygPsW/M6apFx2ZhXxyjd/8r8/Mjmvc5xpdSvYiYhIk2VxOom6ae+1dq+8UuNRu9JNm3zBMO7++7DY9n9N1sGEDx5MyzfewBYVRdmWLey4fBBFK1fWqB6pew6bldeu60FEoJMb3/iJfs98ywe/JPPk4G707RhjWl0WQyfz60xycjIJCQkkJSURHx9vdjkiIrIfXrebbeedT0VaGrEPPkDEsGHVer9hGOwafi3FP/5I8AX9iH/66SOqpzwtjeTbbqf0t9/AZiN27Fgi/nPNEfUpTec3WSN2IiLSpFmdTiJv2ru23Cuz8Lrd1Xp/waefUfzjj1hcLmLvueeI63HExdHy7bcIveQS8HjY8+ijFHylGbNyeBTsRESkyQsbNAh7bCwVaWnkLVhw2O/zlpaS/vjjAETecAOOFi1qpR6ry0WzaVMJv6ZypG7PlEdrbUkWadwU7EREpMmzOp2+O0JkvvwyxmGO2mW/9hrlqanY4+KIHHFjrdZksViIufMO7HFxlCcnk/Xyy7XavzROCnYiIiJA2OBB2KOjqUjdTe7CDw/ZvjwtjcyXXwEg5r/3YPX3r/WarIGBxI4bB1SeJnbvNG99NGkYFOxERESoPP3516hb1syZhxy1S/+/JzFKSvA/8URC+vevs7qCzz+PwF69MMrLSXtkihYwloNSsBMREdkr7IorsEVHUZ6aSu5HHx2wXfEvv5C/ZAlYLMTeN77K/WBrm8ViIe7BB7A4HBR9+y0Fn39eZ/uShk/BTkREZC+rnx+RN9wAQNZLMzHKy/dpY3i97JnyKABhgy7Hv0uXOq/L2aoVETdW1rVn6jS8xcV1vk9pmBTsRERE/iH8yiuxRUVRnpJC3qJF+7yet3AhpevXYw0KIvrOO49aXVE33YSjRQsqdu8m88UXj9p+pWFRsBMREfkHq78/kddfD0Dmv0btPIWFpD9VuQBx1KhR2CMjj2pdsfffD0DWa69Ttm3bUdu3NBwKdiIiIv8SPuRKbBERlCclkbd4iW975gsv4snKqjw1OmzoUa8r+Ky+BPXtCxUVpE1+WBMpZB8KdiIiIv9iDQgg8oa/Ru1ewqiooGz7drLfeguA2PHjsDidptQWe/99WFwuin/4gfyly0ypQeovBTsREZH9CB8yBFt4OOW7dpG/dCnpjz0O5eUEntGHoDPOMK0uZ3w8kTffBED6Y4/hKSw0rRapfxTsRERE9sMaGEjEddcBsOfRqRSuWAF2O7Fjx5lbGHtvX9YykYqMDDJnPGd2OVKPKNiJiIgcQPjQodhCQ/Hk5QEQcfXVuNq0NrmqysWU4x54EIDst9+mdPMWkyuS+kLBTkRE5ABsQX+P2tkiIogadavJFf0tqPfpBJ93Hng8pE2erIkUAijYiYiIHFTEddcSNXo08c8/hy042OxyqogdPw5LQAAlq1eT9+GB75RRXRU5OeS89z7le/bUWp9ydCjYiYiIHITV5SJ69CgCjj/e7FL24WjWjOhbbwEg/YknfKeMa8rwesmZN48/L+hP2oQJbB80iNJNm2qjVDlKFOxEREQasIj//Adn27Z4srPJmD69xv2UbtzIzquGkvbgQ3hyc7E4HHgyMtl59TUUrfqhFiuWuqRgJyIi0oBZnE7iHqycSJEz911Kfl9frfd7CgpIe2QK2y8fRMnatVgDAogZO5Z233xNwMkn4y0sJGnECPI//rguypdapmAnIiLSwAWeegohF14IhkHaw5MxvN5DvscwDPIWL2bbBf3Jeftt8HoJ6X8BbT5eRuR112IPDydh1isEn38+Rnk5KWPuJvvNt47Cp5EjYXqwy54zhz/OOptNXbux/YorKVm37qDtPfn5pE2ezJbevdl0XFe2nd+Pwq+/9r2eM3cuf158CZtPPInNJ57EjiuHUPjNN/v0U/zrr+wcfi2bjj+hst3VV+MtLfW9/sdZZ7OxY6cqj8yXX6m9Dy4iIlKLYsbeizUwkNK168idP/+gbcv++INdw68l9b/34snMxNmqFYmzX6XFU0/hiI31tbO6XLR46knChw0Dw2DPo4+S/uSTdT4D11tcXKf9N2Z2M3eev2wZ6dMeI27iRPy7dSX7jTfZdeMI2n68bL83VjbcbnZdfwO2yAjip0/HHhNLeWoKtpAQXxt7bBwxd4/B2bJl5b9GPvyIpFGjabPgA1zt2wOVoS5pxE1E3nQTcQ/cDzY7ZZs3gbVqzo26/TbCBw/2PbcGBtbRkRARETkyjpgYom+/jT1Tp5Hx5FMEn3su9vDwKm28RUVkvPAC2W+8CRUVWPz8iBo5kojrr8N6gFukWWw2Yh+4H3tsLBlPPUXWK7OoSM+g2SMPY3E4avUzlG3bRsYz03Hv2kXrhQuwWE0ff2pwTA12Wa+/QdjgwYRdfhkAcZMmUvj11+R+sICom0bs0z53wQI8eXm0mvuO78vkjG9RpU3wWX2rPI+5605y3n2XkrVrfcFuz7RphF9zdZV97G/BSVtgIPbo6CP7kCIiIkdJ+LBh5C5YSNnmzWQ89RTNHn4YqDztWvDZ5+yZOpWKtDQAgs46i9j77tvnd3R/LBYLUTeNwB4Vxe4HHyTvo4+oyMoifvoztTLoUZ6SQsZzz5P30Ufg9YLFQsmatQScUP9mItd3pkVhw+2mdP16Ak/r6dtmsVoJ7NmTkjVr9vueguXL8e/enbTJD7Ol1+n8OWAAmS/NxPB49r8Pj4e8pUsxiovx794dgIqsLErXrsMeEcmOIVexpdfp7Lz6GopXr97n/ZmvzGLLKafy56WXkfXqqxgVFQf9TGVlZeTn5/seBQUFh3cwREREaoHFbifuocqJFLnz5lOyZg3uHTtIGnETKXfcQUVaGo4WLYh/8QUSXnj+sELdP4VddikJL76Axd+fov/9j53Dr6UiK6vG9VZkZpL2yBT+6HcBeQsXgtdL0Dln02bRRwp1NWTaiF1FTi54PNj+dcrVFhVJ2fbt+31PeVIyxat+IGTARSTMnEn5rp2kTZqMUVFB9OhRvnalm7ew46qrMMrKsAYEEP/cDFzt2u3tIwmAzOeeI+bee/Hr1JG8jz5i17XX0WbxIpytWgEQfs01+HXujC0slJJffyX9qaepSM8gdvyB7xE4depUJk2adARHRURE5MgEnHgioQMHkvfhhyTfeReerCyM8nIsDgeRI24k8qabsPr51bj/oD59aPn6ayTdPJLS339nx9ChJM6ahTMh4bD78OTnkzV7NtlvvoWx93q6gJ6nEnPXXfh37Vrj2qQeTJ6oFq8XW2QkzSZPxv/YLoT070/kyJHkvPdulWau1q1os3ABrd57j/AhQ0gdN56yP/4AwPBWXvAZduWVhF1+GX6dOxM7fjzO1q3J/WCBr4/I664l8JQe+HXoQPiQIcSOvZfsOXPwut0HLG/8+PHk5eX5Hhs2bKiDgyAiInJwMf+9B2tICBVpaRjl5QT26kWbxYuIvv32Iwp1f/Hv1o2Wc9/B0aIF5Tt3sWPIVZSsP/QyK96SEjJfeYU/zj2PrJdmYhQX49e1K4mvzabla68p1NUC00bs7OFhYLPh+dcQriczC3tU1P7fEx0NDjsWm823zdW2DZ6MTAy3G8veCz8tTifOli0B8D+2CyW//0b2m2/RbPIk7DGV18y52rWt0rezbRvKd+8+YL3+XbtCRQXlySkHvAG0y+XC5XL5nufn5x+wPxERkbpij4yk+bRpZL/xBuFXXUXw+edhsVhqdR+u1q1pOfcdkm4eSdnGjey65j/EPzeDwNNO26et4XaT+8EHZL7wIhUZGQA427Ul5s47CTr77FqvrSkzbcTO4nTi16ULRStX+bYZXi9Fq1b5rof7N/8TTqB8564q6/O4d+zAHh3tC3X75TUw9o60OVq0wB4Ts8/pXveOnTiaNz9gF6WbKmfN2iMjDuPTiYiImCv4rL60fON1QvqdX2fByRETQ8u33iTg1FPxFhez6+aR5C1e4nvd8HjIW7SIbRdeRNqkyVRkZOBo0YJm06bS5qOPCD7nHIW6WmbqrNjIa4eTOm48fscei3/X48h+4028JSWEXXYpAKljx2KPiSXm7jEAhF81hJw5c9gz5VHCrx6Ge+dOMme+TMQ1V/v6TH/yKYL69MberDneoiLylyyh+McfSZhVuQadxWIh8obryZjxHH4dOuLXqSO5H36I+88/CZv+DFC5HErpunUEnHIK1sBAStasYc/UaYQOGIAtNPToHiQREZF6zBYURMLLM9k9bjz5y5aR+t//UpGRgbNVSzKefoayrVsr20VFEXXLSMIHDz74YIwcEVODXUj//lRk55Ax41k8GZm4OnUi8ZWXfadiy1N3g+XvQUVHs2YkzHqFPdOmkXvJQOyxsURccw2RI270tanIziJ17DgqMjKwBgfj6nAMCbNeIahXL1+biOHD8Za52TNtGp68PPw6dCBx9qs4ExOBytHEvGXLyHjueQy3G0d8PBHDhxNx3bVH58CIiIg0IFank+b/9wT26Ciy33iT9Mcf//u1kBAib7iBiGuuxhoQYGKVTYPFqOvlo5uw5ORkEhISSEpKIj4+3uxyRERE6pRhGGTPfo30J57A4u9fOfhyw/X14mxXU/lNNnXETkRERBqPvy53Cup7JrawMOwRui79aFOwExERkVrlatPG7BKarIa1jp2IiIiIHJCCnYiIiEgjoWAnIiIi0kgo2ImIiIg0Egp2IiIiIo2Egp2IiIhII6FgJyIiItJIKNiJiIiINBIKdiIiIiKNhIKdiIiISCOhYCciIiLSSCjYiYiIiDQSCnYiIiIijYSCnYiIiEgjoWAnIiIi0kjYzS6gMfN6vQDs3r3b5EpERESatr9+i//6bW6sFOzq0J49ewDo0aOHyZWIiIgIVP42JyYmml1GnbEYhmGYXURjVVFRwa+//kpsbCxWa9Wz3gUFBXTu3JkNGzYQHBxsUoUNh45X9eh4VY+OV/XoeFWPjlf11NXx8nq97Nmzh+OPPx67vfGOaynYmSQ/P5/Q0FDy8vIICQkxu5x6T8erenS8qkfHq3p0vKpHx6t6dLyOjCZPiIiIiDQSCnYiIiIijYSCnUlcLhcTJkzA5XKZXUqDoONVPTpe1aPjVT06XtWj41U9Ol5HRtfYiYiIiDQSGrETERERaSQU7EREREQaCQU7ERERkUZCwU5ERESkkVCwM8Hzzz9Pq1at8PPz45RTTuHHH380u6R6aeLEiVgsliqPjh07ml1WvfHNN98wYMAAmjdvjsVi4cMPP6zyumEYPPTQQzRr1gx/f3/OOecctm7dak6x9cShjtm11167z3euX79+5hRrsqlTp3LyyScTHBxMTEwMAwcOZPPmzVXalJaWMmrUKCIjIwkKCuLyyy/33UqxqTmc43XmmWfu8/0aOXKkSRWb68UXX6Rr166EhIQQEhJCz549+fjjj32v67tVcwp2R9l7773HmDFjmDBhAr/88gvdunXj/PPPJz093ezS6qUuXbqwe/du3+N///uf2SXVG0VFRXTr1o3nn39+v68//vjjPPvss7z00kv88MMPBAYGcv7551NaWnqUK60/DnXMAPr161flOzd37tyjWGH98fXXXzNq1ChWrVrF559/Tnl5Oeeddx5FRUW+NnfddReLFy9m3rx5fP3116SmpnLZZZeZWLV5Dud4AYwYMaLK9+vxxx83qWJzxcfHM23aNFavXs3PP//MWWedxSWXXML69esBfbeOiCFHVY8ePYxRo0b5nns8HqN58+bG1KlTTayqfpowYYLRrVs3s8toEABj4cKFvuder9eIi4sznnjiCd+23Nxcw+VyGXPnzjWhwvrn38fMMAxj+PDhxiWXXGJKPfVdenq6ARhff/21YRiV3yeHw2HMmzfP12bjxo0GYKxcudKsMuuNfx8vwzCMM844w7jjjjvMK6qeCw8PN2bNmqXv1hHSiN1R5Ha7Wb16Neecc45vm9Vq5ZxzzmHlypUmVlZ/bd26lebNm9OmTRuGDRvGrl27zC6pQdi+fTtpaWlVvmuhoaGccsop+q4dwooVK4iJiaFDhw7ccsstZGVlmV1SvZCXlwdAREQEAKtXr6a8vLzKd6xjx44kJibqO8a+x+svc+bMISoqimOPPZbx48dTXFxsRnn1isfj4d1336WoqIiePXvqu3WE7GYX0JRkZmbi8XiIjY2tsj02NpZNmzaZVFX9dcopp/D666/ToUMHdu/ezaRJk+jduze///47wcHBZpdXr6WlpQHs97v212uyr379+nHZZZfRunVrtm3bxn333ccFF1zAypUrsdlsZpdnGq/Xy5133kmvXr049thjgcrvmNPpJCwsrEpbfcf2f7wAhg4dSsuWLWnevDnr1q1j7NixbN68mQULFphYrXl+++03evbsSWlpKUFBQSxcuJDOnTuzZs0afbeOgIKd1FsXXHCB7++uXbtyyimn0LJlS95//31uuOEGEyuTxmrIkCG+v4877ji6du1K27ZtWbFiBWeffbaJlZlr1KhR/P7777rG9TAd6HjddNNNvr+PO+44mjVrxtlnn822bdto27bt0S7TdB06dGDNmjXk5eUxf/58hg8fztdff212WQ2eTsUeRVFRUdhstn1m9uzZs4e4uDiTqmo4wsLCOOaYY/jjjz/MLqXe++v7pO/akWnTpg1RUVFN+js3evRolixZwldffUV8fLxve1xcHG63m9zc3Crtm/p37EDHa39OOeUUgCb7/XI6nbRr144TTzyRqVOn0q1bN6ZPn67v1hFSsDuKnE4nJ554Il9++aVvm9fr5csvv6Rnz54mVtYwFBYWsm3bNpo1a2Z2KfVe69atiYuLq/Jdy8/P54cfftB3rRqSk5PJyspqkt85wzAYPXo0CxcuZPny5bRu3brK6yeeeCIOh6PKd2zz5s3s2rWrSX7HDnW89mfNmjUATfL7tT9er5eysjJ9t46QTsUeZWPGjGH48OGcdNJJ9OjRg2eeeYaioiKuu+46s0urd+655x4GDBhAy5YtSU1NZcKECdhsNq666iqzS6sXCgsLq/xLf/v27axZs4aIiAgSExO58847eeSRR2jfvj2tW7fmwQcfpHnz5gwcONC8ok12sGMWERHBpEmTuPzyy4mLi2Pbtm3ce++9tGvXjvPPP9/Eqs0xatQo3nnnHT766COCg4N91zaFhobi7+9PaGgoN9xwA2PGjCEiIoKQkBBuu+02evbsyamnnmpy9UffoY7Xtm3beOedd+jfvz+RkZGsW7eOu+66iz59+tC1a1eTqz/6xo8fzwUXXEBiYiIFBQW88847rFixgk8//VTfrSNl9rTcpmjGjBlGYmKi4XQ6jR49ehirVq0yu6R66corrzSaNWtmOJ1Oo0WLFsaVV15p/PHHH2aXVW989dVXBrDPY/jw4YZhVC558uCDDxqxsbGGy+Uyzj77bGPz5s3mFm2ygx2z4uJi47zzzjOio6MNh8NhtGzZ0hgxYoSRlpZmdtmm2N9xAozXXnvN16akpMS49dZbjfDwcCMgIMC49NJLjd27d5tXtIkOdbx27dpl9OnTx4iIiDBcLpfRrl0747///a+Rl5dnbuEmuf76642WLVsaTqfTiI6ONs4++2zjs88+872u71bNWQzDMI5mkBQRERGRuqFr7EREREQaCQU7ERERkUZCwU5ERESkkVCwExEREWkkFOxEREREGgkFOxEREZFGQsFOREREpJFQsBMRERFpJBTsRETqiMVi4cMPPzS7DBFpQhTsRKRRuvbaa7FYLPs8+vXrZ3ZpIiJ1xm52ASIidaVfv3689tprVba5XC6TqhERqXsasRORRsvlchEXF1flER4eDlSeJn3xxRe54IIL8Pf3p02bNsyfP7/K+3/77TfOOuss/P39iYyM5KabbqKwsLBKm9mzZ9OlSxdcLhfNmjVj9OjRVV7PzMzk0ksvJSAggPbt27No0aK6/dAi0qQp2IlIk/Xggw9y+eWXs3btWoYNG8aQIUPYuHEjAEVFRZx//vmEh4fz008/MW/ePL744osqwe3FF19k1KhR3HTTTfz2228sWrSIdu3aVdnHpEmTuOKKK1i3bh39+/dn2LBhZGdnH9XPKSJNiCEi0ggNHz7csNlsRmBgYJXHlClTDMMwDMAYOXJklfeccsopxi233GIYhmG8/PLLRnh4uFFYWOh7fenSpYbVajXS0tIMwzCM5s2bG/fff/8BawCMBx54wPe8sLDQAIyPP/641j6niMg/6Ro7EWm0+vbty4svvlhlW0REhO/vnj17VnmtZ8+erFmzBoCNGzfSrVs3AgMDfa/36tULr9fL5s2bsVgspKamcvbZZx+0hq5du/r+DgwMJCQkhPT09Jp+JBGRg1KwE5FGKzAwcJ9To7XF39//sNo5HI4qzy0WC16vty5KEhHRNXYi0nStWrVqn+edOnUCoFOnTqxdu5aioiLf69999x1Wq5UOHToQHBxMq1at+PLLL49qzSIiB6MROxFptMrKykhLS6uyzW63ExUVBcC8efM46aSTOP3005kzZw4//vgjr776KgDDhg1jwoQJDB8+nIkTJ5KRkcFtt93GNddcQ2xsLAATJ05k5MiRxMTEcMEFF1BQUMB3333HbbfddnQ/qIjIXgp2ItJoffLJJzRr1qzKtg4dOrBp0yagcsbqu+++y6233kqzZs2YO3cunTt3BiAgIIBPP/2UO+64g5NPPpmAgAAuv/xynnrqKV9fw4cPp7S0lKeffpp77rmHqKgoBg0adPQ+oIjIv1gMwzDMLkJE5GizWCwsXLiQgQMHml2KiEit0TV2IiIiIo2Egp2IiIhII6Fr7ESkSdJVKCLSGGnETkRERKSRULATERERaSQU7EREREQaCQU7ERERkUZCwU5ERESkkVCwExEREWkkFOxEREREGgkFOxEREZFG4v8BY8+bziWMTYQAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting\n",
    "fig, ax1 = plt.subplots()\n",
    "\n",
    "color = 'tab:red'\n",
    "ax1.set_xlabel('Epoch')\n",
    "ax1.set_ylabel('Training Loss', color=color)\n",
    "ax1.plot(range(1, epochs+1), train_losses, color=color)\n",
    "ax1.tick_params(axis='y', labelcolor=color)\n",
    "\n",
    "ax2 = ax1.twinx()\n",
    "color = 'tab:blue'\n",
    "ax2.set_ylabel('Test Accuracy', color=color)\n",
    "ax2.plot(range(1, epochs+1), test_accuracies, color=color)\n",
    "ax2.tick_params(axis='y', labelcolor=color)\n",
    "\n",
    "fig.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cbc2102a-9fb5-4fec-ae67-78d3054d2be7",
   "metadata": {},
   "source": [
    "# 2-1. MLP Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "c6ee5ef8-56ff-4019-99db-7e90f103d246",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define MLP Model\n",
    "class MLP(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(MLP, self).__init__()\n",
    "        self.fc1 = nn.Linear(7, 128)  # 7 input features\n",
    "        self.fc2 = nn.Linear(128, 64)\n",
    "        self.fc3 = nn.Linear(64, 1)  # Binary classification\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = torch.relu(self.fc1(x))\n",
    "        x = torch.relu(self.fc2(x))\n",
    "        x = torch.sigmoid(self.fc3(x))\n",
    "        return x\n",
    "\n",
    "        \n",
    "# Train MLP\n",
    "model = MLP()\n",
    "criterion = nn.BCEWithLogitsLoss()\n",
    "optimizer = optim.Adam(model.parameters(), lr=0.001)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d11caa6a-9652-4972-9f7a-c42ccdcc3019",
   "metadata": {},
   "source": [
    "# 2-2. MLP Train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "927653d5-3a20-4396-b4c2-9eb12bbc94fa",
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_accuracy(model, loader):\n",
    "    model.eval()\n",
    "    correct = 0\n",
    "    with torch.no_grad():\n",
    "        for data, target in loader:\n",
    "            output = model(data)\n",
    "            pred = (output > 0.5).float()\n",
    "            correct += pred.eq(target.float().view_as(pred)).sum().item()\n",
    "    accuracy = 100. * correct / len(loader.dataset)\n",
    "    return accuracy"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bb8957ca-959c-4fa6-9095-9128a0586a5a",
   "metadata": {},
   "source": [
    "# Training loop\n",
    "epochs = 30\n",
    "train_losses = []\n",
    "test_accuracies = []\n",
    "\n",
    "for epoch in range(1, epochs+1):\n",
    "    model.train()\n",
    "    running_loss = 0\n",
    "    for batch_idx, (data, target) in enumerate(train_loader):\n",
    "        optimizer.zero_grad()\n",
    "        output = model(data)\n",
    "        loss = criterion(output, target.unsqueeze(1).float())\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        running_loss += loss.item()\n",
    "    train_loss = running_loss / len(train_loader)\n",
    "    train_losses.append(train_loss)\n",
    "    \n",
    "    train_accuracy = compute_accuracy(model, train_loader)\n",
    "    test_accuracy = compute_accuracy(model, test_loader)\n",
    "    test_accuracies.append(test_accuracy)\n",
    "    \n",
    "    print(f'Epoch {epoch}/{epochs}, Loss: {train_loss:.4f}, Training Accuracy: {train_accuracy:.2f}%, Test Accuracy: {test_accuracy:.2f}%')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "cb427445-a6b3-4bba-9248-db2e06658aca",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting\n",
    "fig, ax1 = plt.subplots()\n",
    "\n",
    "color = 'tab:red'\n",
    "ax1.set_xlabel('Epoch')\n",
    "ax1.set_ylabel('Training Loss', color=color)\n",
    "ax1.plot(range(1, epochs+1), train_losses, color=color)\n",
    "ax1.tick_params(axis='y', labelcolor=color)\n",
    "\n",
    "ax2 = ax1.twinx()\n",
    "color = 'tab:blue'\n",
    "ax2.set_ylabel('Test Accuracy', color=color)\n",
    "ax2.plot(range(1, epochs+1), test_accuracies, color=color)\n",
    "ax2.tick_params(axis='y', labelcolor=color)\n",
    "\n",
    "fig.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1e760fb7-8350-468b-b690-ababac7868d0",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7ed7e618-48e4-401e-9fdd-386a934ff70f",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "prac",
   "language": "python",
   "name": "prac"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}
