Brain Tumor Malignancy Classification

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "63bdfd88-dd37-470d-a2a4-19b242dbd11d",
   "metadata": {},
   "source": [
    "# Klasifikasi Tingkat Keganasan Tumor Otak"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "af46491a-c944-402e-9103-686cadcd696f",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import time\n",
    "\n",
    "import pathlib\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import tensorflow as tf\n",
    "import matplotlib.pyplot as plt\n",
    "from tensorflow.keras.applications import ResNet50V2\n",
    "from tensorflow.keras.applications import MobileNetV2\n",
    "from tensorflow.keras.applications.inception_resnet_v2 import InceptionResNetV2\n",
    "from tensorflow.keras import Sequential\n",
    "from tensorflow.keras.layers import Flatten, Dense, Input\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "from tensorflow.keras.models import Model\n",
    "from tensorflow.keras.applications.resnet_v2 import preprocess_input\n",
    "from tensorflow.keras.applications.mobilenet_v2 import preprocess_input\n",
    "from tensorflow.keras.preprocessing.image import img_to_array,  load_img\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import classification_report\n",
    "\n",
    "from imblearn.over_sampling import SMOTE\n",
    "\n",
    "from tensorflow.keras.utils import to_categorical\n",
    "from sklearn.preprocessing import LabelBinarizer\n",
    "\n",
    "from IPython.display import display, Markdown"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "78448e7a-3d0d-4d0c-8391-d4a946d5060c",
   "metadata": {},
   "source": [
    "## Pengaturan Hyperparameter dan Konfigurasi Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "60c96950-1067-47ba-bbfa-a08e1c20a332",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Menginisialisasi learning rate, jumlah epoch, dan jumlah batch size\n",
    "INIT_LR = 1e-4\n",
    "EPOCHS = 30\n",
    "BS = 32\n",
    "SIZE_X = 112\n",
    "SIZE_Y = 112\n",
    "\n",
    "NETWORK = \"ResNet50V2_Tesis_SMOTE\"\n",
    "\n",
    "dataset = r\"C:\\Users\\Lab129\\Downloads\\Indi\\Dataset Kla 128\";\n",
    "model_path = NETWORK + \"_\" + str(EPOCHS) + \".model\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "21aa5ea8-63a6-4187-b031-faf12456d778",
   "metadata": {},
   "source": [
    "## Pre-processing Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "4e44e9b2-45d4-41e0-b42a-a1c22265f4fe",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[INFO] loading images...\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 47232 entries, 0 to 47231\n",
      "Data columns (total 2 columns):\n",
      " #   Column  Non-Null Count  Dtype   \n",
      "---  ------  --------------  -----   \n",
      " 0   image   47232 non-null  string  \n",
      " 1   label   47232 non-null  category\n",
      "dtypes: category(1), string(1)\n",
      "memory usage: 415.4 KB\n"
     ]
    }
   ],
   "source": [
    "data_dir = pathlib.Path(dataset)\n",
    "\n",
    "print(\"[INFO] loading images...\")\n",
    "df = pd.DataFrame(\n",
    "    [(str(dir), str(dir).split(os.path.sep)[-3]) for dir in data_dir.glob(\"**/*.jpg\")],\n",
    "    columns=[\"image\", \"label\"],\n",
    ")\n",
    "\n",
    "# Mengatur tipe data\n",
    "df[\"image\"] = df[\"image\"].astype('string')\n",
    "df[\"label\"] = df[\"label\"].astype('category')\n",
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b4d68bd4-53b0-4b95-b8e9-9598ec1ece02",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(Index(['HGG', 'LGG'], dtype='object'), array([0, 1], dtype=int8))"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"label\"].cat.categories, df[\"label\"].cat.codes.unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "539d578b-9197-4d1c-a49a-19653b55a9a5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "HGG    37504\n",
       "LGG     9728\n",
       "Name: label, dtype: int64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Memeriksa ketidakseimmbangan kelas\n",
    "class_count = df[\"label\"].value_counts()\n",
    "display(class_count)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "49119cac-69fa-4e84-8497-9edf68d09603",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Lakukan pengkodean one-hot encoding pada label\n",
    "label_encoder = LabelBinarizer()\n",
    "label_encoded = label_encoder.fit_transform(df[\"label\"])\n",
    "label_one_hot_encoded = to_categorical(label_encoded)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "e8782262-9f39-454c-865f-0e77e0d5fa6a",
   "metadata": {},
   "outputs": [],
   "source": [
    "def image_reader(path: str):\n",
    "    image = tf.io.read_file(path)\n",
    "    image = tf.image.decode_jpeg(image, channels=3)\n",
    "    image = tf.image.resize(image, [SIZE_X, SIZE_Y])\n",
    "    image = preprocess_input(image)\n",
    "    return image\n",
    "\n",
    "def image_generator(path: str, label):\n",
    "    image = image_reader(path)\n",
    "    return image, label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "551eaefa-a7a8-42af-ada9-8addf0dd3fe3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/markdown": [
       "**Train images count**: 37785\n",
       "\n",
       "**Test images count**: 9447"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Membagi data menjadi set pelatihan dan set pengujian\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    df[\"image\"], label_one_hot_encoded, test_size=0.20, random_state=42, stratify=df[\"label\"]\n",
    ")\n",
    "\n",
    "display(Markdown((\n",
    "    f\"**Train images count**: {X_train.count()}\\n\\n\"\n",
    "    f\"**Test images count**: {X_test.count()}\"\n",
    ")))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "eb5bb98a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/markdown": [
       "**Train images count**: 60006\n",
       "\n",
       "**Test images count**: 9447"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Menginisialisasi objek SMOTE\n",
    "sm = SMOTE(random_state=42)\n",
    "\n",
    "# Melakukan resample dataset\n",
    "X_train_idx, y_train = sm.fit_resample(X_train.index.values.reshape(-1, 1), y_train)\n",
    "\n",
    "X_train = df.iloc[X_train_idx.flatten()][\"image\"]\n",
    "y_train = to_categorical(y_train)\n",
    "\n",
    "\n",
    "display(Markdown((\n",
    "    f\"**Train images count**: {X_train.count()}\\n\\n\"\n",
    "    f\"**Test images count**: {X_test.count()}\"\n",
    ")))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "f772bfca-e5ff-4e1b-9e81-143bc4f7afac",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_total_items = len(X_train)\n",
    "train_num_batches = train_total_items // BS\n",
    "train_dataset = tf.data.Dataset.from_tensor_slices((X_train, y_train)).map(\n",
    "    image_generator, num_parallel_calls=tf.data.AUTOTUNE\n",
    ").shuffle(train_total_items).repeat(EPOCHS).batch(BS).prefetch(tf.data.AUTOTUNE)\n",
    "\n",
    "val_total_items = len(X_test)\n",
    "val_num_batches = val_total_items // BS\n",
    "val_dataset =  tf.data.Dataset.from_tensor_slices((X_test, y_test)).map(\n",
    "    image_generator, num_parallel_calls=tf.data.AUTOTUNE\n",
    ").batch(BS).prefetch(tf.data.AUTOTUNE)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ccecc2d4-551f-4c89-8567-19738d504448",
   "metadata": {},
   "source": [
    "## Pengembangan Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "2aa92969-c858-4894-9269-83e073a09395",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:`input_shape` is undefined or non-square, or `rows` is not in [96, 128, 160, 192, 224]. Weights for input shape (224, 224) will be loaded as the default.\n",
      "(None, 4, 4, 1280)\n"
     ]
    }
   ],
   "source": [
    "starttime = time.time()\n",
    "# Memuat jaringan MobileNetV2, memastikan adanya kumpulan lapisan\n",
    "baseModel = ResNet50V2(weights=\"imagenet\", include_top=False,\n",
    "                       input_tensor=Input(shape=(SIZE_X, SIZE_Y, 3)))\n",
    "\n",
    "# baseModel = MobileNetV2(weights=\"imagenet\", include_top=False,\n",
    "#                         input_tensor=Input(shape=(SIZE_X, SIZE_Y, 3)))\n",
    "\n",
    "# baseModel = InceptionResNetV2(weights=\"imagenet\", include_top=False,\n",
    "#                               input_tensor=Input(shape=(SIZE_X, SIZE_Y, 3)))\n",
    "\n",
    "baseModel.trainable = False\n",
    "\n",
    "# Membangun head model\n",
    "headModel = baseModel.output  # outshape = 4 x 4 x 1280 channel\n",
    "print(headModel.shape)\n",
    "\n",
    "headModel = Flatten(name=\"flatten\")(headModel)\n",
    "headModel = Dense(512, activation=\"relu\")(headModel)\n",
    "headModel = Dense(2, activation=\"softmax\")(headModel)\n",
    "\n",
    "# Tempatkan head model di atas base model\n",
    "model = Model(inputs=baseModel.input, outputs=headModel)\n",
    "# model.summary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "9b343efc-9d7f-440f-ab43-2fa8cc730bfa",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[INFO] compiling model...\n"
     ]
    }
   ],
   "source": [
    "# Lakukan kompilasi model (mengukur accuracy)\n",
    "print(\"[INFO] compiling model...\")\n",
    "opt = Adam(learning_rate=INIT_LR, decay=INIT_LR / EPOCHS)\n",
    "model.compile(loss=\"categorical_crossentropy\", optimizer=opt,\n",
    "              metrics=[\"accuracy\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a72c2f03-fa66-4878-80ea-c51d6a9bec28",
   "metadata": {},
   "source": [
    "## Proses Pelatihan Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "3adf2316-45aa-4e59-9ccc-695f0fcb1d08",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[INFO] training head...\n",
      "Epoch 1/30\n",
      "1875/1875 [==============================] - 212s 74ms/step - loss: 0.3876 - accuracy: 0.8220 - val_loss: 0.4250 - val_accuracy: 0.8097\n",
      "Epoch 2/30\n",
      "1875/1875 [==============================] - 128s 68ms/step - loss: 0.1848 - accuracy: 0.9207 - val_loss: 0.1870 - val_accuracy: 0.9221\n",
      "Epoch 3/30\n",
      "1875/1875 [==============================] - 115s 62ms/step - loss: 0.1131 - accuracy: 0.9518 - val_loss: 0.1431 - val_accuracy: 0.9338\n",
      "Epoch 4/30\n",
      "1875/1875 [==============================] - 101s 54ms/step - loss: 0.0802 - accuracy: 0.9650 - val_loss: 0.1952 - val_accuracy: 0.9280\n",
      "Epoch 5/30\n",
      "1875/1875 [==============================] - 90s 48ms/step - loss: 0.0637 - accuracy: 0.9715 - val_loss: 0.1239 - val_accuracy: 0.9538\n",
      "Epoch 6/30\n",
      "1875/1875 [==============================] - 79s 42ms/step - loss: 0.0521 - accuracy: 0.9763 - val_loss: 0.1295 - val_accuracy: 0.9530\n",
      "Epoch 7/30\n",
      "1875/1875 [==============================] - 74s 39ms/step - loss: 0.0420 - accuracy: 0.9805 - val_loss: 0.1560 - val_accuracy: 0.9478\n",
      "Epoch 8/30\n",
      "1875/1875 [==============================] - 68s 36ms/step - loss: 0.0422 - accuracy: 0.9804 - val_loss: 0.1099 - val_accuracy: 0.9588\n",
      "Epoch 9/30\n",
      "1875/1875 [==============================] - 64s 34ms/step - loss: 0.0338 - accuracy: 0.9840 - val_loss: 0.1402 - val_accuracy: 0.9441\n",
      "Epoch 10/30\n",
      "1875/1875 [==============================] - 61s 33ms/step - loss: 0.0338 - accuracy: 0.9834 - val_loss: 0.1221 - val_accuracy: 0.9526\n",
      "Epoch 11/30\n",
      "1875/1875 [==============================] - 58s 31ms/step - loss: 0.0356 - accuracy: 0.9833 - val_loss: 0.1121 - val_accuracy: 0.9635\n",
      "Epoch 12/30\n",
      "1875/1875 [==============================] - 57s 31ms/step - loss: 0.0261 - accuracy: 0.9862 - val_loss: 0.1211 - val_accuracy: 0.9576\n",
      "Epoch 13/30\n",
      "1875/1875 [==============================] - 55s 30ms/step - loss: 0.0308 - accuracy: 0.9854 - val_loss: 0.1229 - val_accuracy: 0.9554\n",
      "Epoch 14/30\n",
      "1875/1875 [==============================] - 54s 29ms/step - loss: 0.0230 - accuracy: 0.9879 - val_loss: 0.1259 - val_accuracy: 0.9518\n",
      "Epoch 15/30\n",
      "1875/1875 [==============================] - 54s 29ms/step - loss: 0.0283 - accuracy: 0.9864 - val_loss: 0.1218 - val_accuracy: 0.9675\n",
      "Epoch 16/30\n",
      "1875/1875 [==============================] - 53s 28ms/step - loss: 0.0232 - accuracy: 0.9876 - val_loss: 0.1008 - val_accuracy: 0.9640\n",
      "Epoch 17/30\n",
      "1875/1875 [==============================] - 53s 28ms/step - loss: 0.0256 - accuracy: 0.9877 - val_loss: 0.1009 - val_accuracy: 0.9603\n",
      "Epoch 18/30\n",
      "1875/1875 [==============================] - 53s 28ms/step - loss: 0.0205 - accuracy: 0.9886 - val_loss: 0.1070 - val_accuracy: 0.9655\n",
      "Epoch 19/30\n",
      "1875/1875 [==============================] - 52s 28ms/step - loss: 0.0216 - accuracy: 0.9879 - val_loss: 0.1488 - val_accuracy: 0.9619\n",
      "Epoch 20/30\n",
      "1875/1875 [==============================] - 52s 28ms/step - loss: 0.0207 - accuracy: 0.9884 - val_loss: 0.1257 - val_accuracy: 0.9597\n",
      "Epoch 21/30\n",
      "1875/1875 [==============================] - 51s 27ms/step - loss: 0.0211 - accuracy: 0.9890 - val_loss: 0.0947 - val_accuracy: 0.9761\n",
      "Epoch 22/30\n",
      "1875/1875 [==============================] - 51s 27ms/step - loss: 0.0220 - accuracy: 0.9888 - val_loss: 0.0906 - val_accuracy: 0.9781\n",
      "Epoch 23/30\n",
      "1875/1875 [==============================] - 51s 27ms/step - loss: 0.0200 - accuracy: 0.9889 - val_loss: 0.0944 - val_accuracy: 0.9772\n",
      "Epoch 24/30\n",
      "1875/1875 [==============================] - 51s 27ms/step - loss: 0.0223 - accuracy: 0.9882 - val_loss: 0.1139 - val_accuracy: 0.9631\n",
      "Epoch 25/30\n",
      "1875/1875 [==============================] - 50s 27ms/step - loss: 0.0188 - accuracy: 0.9903 - val_loss: 0.1295 - val_accuracy: 0.9694\n",
      "Epoch 26/30\n",
      "1875/1875 [==============================] - 50s 27ms/step - loss: 0.0188 - accuracy: 0.9891 - val_loss: 0.1247 - val_accuracy: 0.9610\n",
      "Epoch 27/30\n",
      "1875/1875 [==============================] - 50s 27ms/step - loss: 0.0134 - accuracy: 0.9914 - val_loss: 0.1080 - val_accuracy: 0.9659\n",
      "Epoch 28/30\n",
      "1875/1875 [==============================] - 50s 27ms/step - loss: 0.0205 - accuracy: 0.9889 - val_loss: 0.0976 - val_accuracy: 0.9677\n",
      "Epoch 29/30\n",
      "1875/1875 [==============================] - 50s 27ms/step - loss: 0.0186 - accuracy: 0.9904 - val_loss: 0.1026 - val_accuracy: 0.9768\n",
      "Epoch 30/30\n",
      "1875/1875 [==============================] - 49s 26ms/step - loss: 0.0199 - accuracy: 0.9893 - val_loss: 0.1397 - val_accuracy: 0.9578\n"
     ]
    }
   ],
   "source": [
    "# Lakukan pelatihan untuk head network\n",
    "print(\"[INFO] training head...\")\n",
    "H = model.fit(train_dataset,\n",
    "              batch_size=BS, \n",
    "              verbose=1, \n",
    "              steps_per_epoch=train_num_batches,\n",
    "              epochs=EPOCHS, \n",
    "              validation_data=val_dataset,\n",
    "              validation_steps=val_num_batches,\n",
    "              validation_batch_size=BS,\n",
    "              shuffle=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4a84722c-005d-4ecd-b9e8-b86bf974f20c",
   "metadata": {},
   "source": [
    "## Uji Coba dan Evaluasi Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "15ad82a6-6317-43ad-8bfa-f5514aa8e148",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[INFO] evaluating network...\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         HGG       0.99      0.96      0.97      7501\n",
      "         LGG       0.85      0.96      0.90      1946\n",
      "\n",
      "    accuracy                           0.96      9447\n",
      "   macro avg       0.92      0.96      0.94      9447\n",
      "weighted avg       0.96      0.96      0.96      9447\n",
      "\n",
      "Running time: 2043.5096752643585 s\n"
     ]
    }
   ],
   "source": [
    "# Buat prediksi pada set pengujian\n",
    "print(\"[INFO] evaluating network...\")\n",
    "predIdxs = model.predict(val_dataset, batch_size=BS)\n",
    "\n",
    "# Untuk setiap gambar dalam set pengujian kita perlu menemukan indeks file label dengan probabilitas prediksi terbesar yang sesuai\n",
    "predIdxs = np.argmax(predIdxs, axis=1)\n",
    "\n",
    "endtime = time.time()\n",
    "\n",
    "# Tampilkan classification report\n",
    "print(classification_report(y_test.argmax(axis=1), predIdxs,\n",
    "                            target_names=df[\"label\"].cat.categories))\n",
    "\n",
    "# Tampilkan running time\n",
    "print(\"Running time: {} s\".format(endtime - starttime))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "948d8802-783f-425b-9251-c908e64c4bf9",
   "metadata": {},
   "source": [
    "## Penyimpanan Model dan Visualisasi"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "721f4a1d-ccb9-4bb7-a8f3-1a19a50f321f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[INFO] saving brain tumor malignancy detector model...\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAj0AAAHMCAYAAAA+txkgAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAA9hAAAPYQGoP6dpAAB/OElEQVR4nO3dd3zTdeLH8dc3TdOdDtpSaKGFlr1VUEGZDlDcuNBzo556d96dW1HcP/T0vDu9pZ53eA7QExQEHIigoICy994U2tJ0rzTf3x9pI7UFOtIkbd/Px4MHzTfffPPJp2nz7mcapmmaiIiIiLRyFn8XQERERMQXFHpERESkTVDoERERkTZBoUdERETaBIUeERERaRMUekRERKRNUOgRERGRNkGhR0RERNoEhR4RERFpExR6RBrBMAxGjhzZ5OuMHDkSwzCaXqBWxlv16y033XQThmGwe/duz7Hdu3djGAY33XRTva/z73//G8Mw+Pe//+31Mh6rrvKKiEKPtFCGYTToX3N/yIh/XHfddRiGwV//+teTnnveeedhGAYzZ870Qcma15QpUzAMg6+//trfRam36iCmn0XxJ6u/CyDSGE888UStY6+88gp5eXn85je/ISYmpsZ9AwcO9Orzb9q0ifDw8CZfZ9q0aRQXF3uhRG3TpEmTePfdd3njjTe46667jnve7t27+fLLL+nQoQMXXXSRV547OTmZTZs2ER0d7ZXredPzzz/PQw89RHJysr+LIhJQFHqkRZoyZUqtY//+97/Jy8vj3nvvJS0trVmfv2fPnl65TufOnb1ynbZq5MiRdO/enVWrVrFy5UpOOeWUOs978803MU2Tm2++GavVO7/2goODvfY+8LYOHTrQoUMHfxdDJOCoe0tavepxM+Xl5Tz11FP06NGDkJAQz1iMvLw8XnzxRUaPHk1KSgo2m42EhAQuvvhivvvuuzqvWdeYk2O7HD788EOGDBlCeHg4cXFxXHPNNRw4cOC4ZTvW119/jWEYTJkyhdWrV3PhhRcSExNDeHg4I0aMYOnSpXWW6dChQ9x8880kJiYSFhbGwIED+c9//lPjevXRlPrIzs7m9ttvp0OHDoSEhNCnTx/eeuutOh9TXl7O008/TXp6OiEhIXTp0oXHHnuMsrKyepWz2qRJkwB4/fXX67y/srKSt956C8MwuO222wCYNWsW119/Pd27dyciIoKIiAhOPfVU/vznP+Nyuer1vCca07N9+3auvPJKYmNjiYiIYOjQoXz66afHvdbChQu5/fbb6d27N3a7nbCwMPr27cuTTz5JaWlpjXPT0tJ48sknARg1alSNbtxqJxrTM2PGDIYPH050dDRhYWH069eP559/vs56T0tLIy0tjaKiIu6//346d+5MSEgIGRkZTJ06FdM061VXjbFgwQLGjh1LXFwcISEhdO/enYceeoi8vLxa5+7cuZPbb7+djIwMwsLCiIuLo1+/ftx5553k5OR4zisvL+fPf/4zp5xyCrGxsYSHh5OWlsYll1zCl19+2WyvRQKHWnqkzbjiiitYsWIF48aN49JLLyUxMRFwd1U9+uijDB8+nAsvvJDY2Fj27t3LJ598wrx585g9ezZjx46t9/P89a9/5ZNPPuHiiy9mxIgRLFu2jOnTp7NmzRpWr15NSEhIva7zww8/8MILL3DmmWdy2223sXfvXv73v/8xZswYVq9eTY8ePTznHjlyhDPPPJM9e/YwfPhwhg4dSmZmJnfddRfnnXdeg+qpsfXhcDgYNmwYNpuNCRMmUFZWxgcffMAtt9yCxWLhxhtv9JxrmiZXXXUVH3/8Menp6dxzzz2Ul5fzr3/9i3Xr1jWovDfeeCOPPvoo7733Hi+99FKtbsd58+Zx4MABzj33XLp06QLAQw89hMVi4fTTTyc5OZm8vDy++uorfvOb37BixQrefvvtBpXhWNu2bePMM88kJyeHcePGMXDgQLZv386ll17KuHHj6nzM1KlT2bx5M0OHDuXCCy+ktLSUJUuWMGXKFL7++mu+/PJLgoKCALj33nuZNWsWixYt4sYbb2xQq+YjjzzC888/T3x8PBMnTiQyMpJ58+bxyCOP8Nlnn/H5559js9lqPKaiooLzzz+fgwcPMm7cOKxWK7NmzeKhhx6itLS0zq7mpvrHP/7BL3/5SyIiIrjyyitJTEzk66+/ZurUqcyePZslS5Z4urAPHTrE4MGDyc/P54ILLuCKK66gtLSUXbt28fbbb3PPPffQrl07wB0G33vvPfr27csNN9xAWFgYBw8e5Ntvv2X+/Pmcc845Xn8tEmBMkVYiNTXVBMxdu3bVOD5ixAgTMPv162dmZWXVepzD4ajz+L59+8wOHTqYPXv2rHUfYI4YMaLGsSeeeMIEzKioKHPt2rU17rv22mtNwJw+fXqdZTvWwoULTcAEzLfeeqvGfX//+99NwPzlL39Z4/gtt9xiAuYDDzxQ4/jq1atNm81mAuYTTzxR63XUpbH1AZi33nqr6XQ6Pcc3bNhgBgUFmb169apx/jvvvGMC5hlnnGGWlJR4jufk5Jhdu3ats35P5KqrrqqzvkzTNC+++GITMD/44APPse3bt9c6r7Ky0rzhhhtMwPz+++9r3HfjjTfWem/t2rXLBMwbb7yxxrnnnnuuCZivvPJKjeOzZs067vd1x44dpsvlqlWmxx57zATM999/v8bx6vfawoULaz3meOVdunSpCZidOnUyDx065DleUVFhjh8/3gTMZ599tsZ1qn+mxo0bZxYXF3uOHz582IyOjjajo6PN8vLyOstwvDLV9T061u7du02bzWZGRUWZmzZtqnHfL3/5SxMwJ02a5Dn25z//uc76Nk3TLCws9JTb4XCYhmGYp556ao33aLXs7Ox6vQ5p2dS9JW3G008/TXx8fK3j0dHRdR5PSUlhwoQJbN68mb1799b7eX7961/Tr1+/Gsequ2CWL19e7+sMGzasVtfJLbfcgtVqrXGd8vJy3nvvPaKjo3nsscdqnD9gwABuuOGGej8nNL4+wsPDefnllz0tEgC9e/dm2LBhbNq0icLCQs/x6i6v5557jtDQUM/xuLg4Jk+e3KDyAtx+++0AvPHGGzWOHzp0iLlz55KYmMgll1ziOZ6enl7rGhaLhd/85jcAfPbZZw0uA8D+/fv54osv6NKlC/fcc0+N+y655BJGjBhR5+O6du1a59IFv/3tb5tUnmP961//AuCxxx4jKSnJc9xqtfLSSy9hsVhq1V+1P//5z4SFhXluV9dnXl4eW7ZsaXLZjvXf//6X8vJy7rnnnlpjpp599lmioqJ4++23a3XHHVu+ahEREZ7jhmFgmiYhISFYLLU/+qpbg6R1U+iRNmPIkCHHvW/JkiVcddVVdOrUiZCQEM8Yib/85S8AdY7HOZ7TTjut1rFOnToBkJub26TrBAcH0759+xrX2bJlCyUlJfTv35+oqKhajznrrLPq/ZzVGlMf3bp1w2631zpe12tfuXIlFoulzrI1Zn2e0aNHk56ezpIlS9i0aZPn+FtvvYXT6eSmm24iODjYczwnJ4eHHnqI/v37ExkZ6Xl9p5566nFfX32sWrUKcNf5seGv2vFeW1FREc899xyDBw8mOjoai8WCYRieD+LGludYK1euBNx19XPdu3cnJSWFXbt21RozEx0dTUZGRq3HNOY93dRyxsbGMmjQIEpLS9m8eTMAF198MZGRkdx9991cccUV/POf/2TDhg21xhvZ7XYuuugili5dysCBA3nqqadYuHChZk+2MRrTI23GsX/dHmvmzJlMmDCB0NBQzj33XNLT04mIiMBisfD111+zaNGiBg2u/fl0ecAzY6iysrJJ16m+1rHXqf6Qat++fZ3nH+/48TS2Pk5UXqBWmePi4moEkWrH+z6dSPUg5Ycffpg33niDl156CdM0efPNNzEMw9PSBu6xR4MHD2bXrl0MGTKEG264gbi4OKxWKw6Hgz/96U8NHkx97OuC49d5Xa+toqKC0aNHs3z5cvr27cvVV19NQkKCp26efPLJRpenrrIdb1ZXhw4d2Lt3Lw6Ho8Y0/IZ8X72hPuUE9/cRIDU1leXLlzNlyhTmz5/PRx99BLhD2X333cevf/1rz2OnT5/O1KlTeffddz1jkUJDQ5kwYQJ/+MMfGvyzIi2PQo+0Gcdb+Xjy5MnYbDZ++OEHevXqVeO+O+64g0WLFvmieI1W3bpy+PDhOu8/3vHj8UV9REdHc/ToUSoqKmoFn8zMzEZd8+abb+bxxx9n2rRpPP/883zzzTfs3LmT0aNH12ipeOONN9i1axdPPPFErRlt3333HX/6058a9fyAJywcr87rem0ff/wxy5cv56abbqo10+3QoUOemVpNVV22zMzMOrv3Dh06VOM8fzm2nH369Kl1f13l7NWrF9OnT8fpdLJmzRq+/PJL/vKXv/Cb3/yGiIgIbr31VsDdBTZlyhSmTJnCvn37WLx4Mf/+97/573//y+7du/nmm2988ArFn9S9JW3e9u3b6d27d60PeJfLxbfffuunUtVfz549CQsLY+3atRQUFNS6v6GvwRf1ccoppxz3eo1dZbh9+/ZcfPHFZGdnM2vWLM/4lOrxPtW2b98OuGfz/VxTA92gQYMAd53X1QJS12urLs/ll19e7/JUd501pJWlumzHK8P+/fvp0qXLcVt2fOVE5XQ4HKxevZrQ0NBa709wtz6deuqpPPjgg7z33nuAe3mCunTq1InrrruOzz77jIyMDL799tsa09uldVLokTYvLS2Nbdu2cfDgQc8x0zSZMmUKGzdu9GPJ6sdms3H11VeTl5fHM888U+O+NWvWMG3atAZdzxf1cfPNNwPw6KOP1liH5ujRo7VeQ0NUd2O99NJLzJw5k/j4eC677LIa51RP8f75h+qqVat4/vnnG/3c4B7sfe6557Jr1y5effXVGvd9/PHHdYaY45Vn586dPPjgg3U+T/VYn4YMsL/lllsAeOaZZ8jKyvIcr6ys5L777sPlcnlaRPzp+uuvJzg4mL/85S+eQFht8uTJ5Ofnc/3113uWfvjxxx/rXLunurWtegmDrKysOpdDKCoqorCwEKvVWmu6vrQ+6t6SNu+3v/0td955J4MGDeKKK64gODiYJUuWsHHjRi666CJmz57t7yKe1P/93//x1Vdf8cILL7Bs2TKGDh3KoUOHmDFjBhdccAGzZs2qc8ZKXXxRH9deey3Tp0/nk08+oW/fvlxyySVUVFTw4YcfMnjwYHbs2NGo65533nmkpaV5Zrfdc889tT7IbrjhBl588UXuvfdeFi5cSLdu3di2bRtz5szh8ssvZ/r06U16ba+99hpnnnkm9957L59//jkDBgxg+/btzJw5s876u+iii8jIyODll19m3bp1DBo0iL179zJnzhwuvPDCOoPNqFGjsFgsPPzww6xfv57Y2FiAWrP3jjV06FAeeOABXnjhBfr27cuECROIiIhg3rx5rF+/nrPOOov777+/Sa+9Pt54443jtuZNnDiR8847j1deeYW7776bU045hauuuoqEhAQWLVrEd999R8+ePZk6darnMW+//Tb/+Mc/OOuss0hPTyc2NpYdO3Ywe/ZsQkJCuPfeewH3YPBBgwbRr18/+vfvT6dOncjPz2fOnDlkZmby61//us6JANLK+HXCvIgXnWydnhN56623zAEDBpjh4eFmu3btzEsvvdRcu3btcddD4QTr9NS1dsrx1nQ50To9x1tXJzU11UxNTa11fP/+/eYNN9xgxsfHm6GhoeaAAQPMf//73+YHH3xgAuYf//jHE9bBsbxRH9XqWjPGNE2zrKzMfPLJJ80uXbqYNpvNTE1NNR955BGztLS0wev0HOuZZ57xrIezefPmOs/ZsGGDedFFF5kJCQlmeHi4ecopp5ivv/76cb9PDVmnxzRNc9u2beYVV1xhRkdHm+Hh4eYZZ5xhzpkzx3zrrbfqXKtm79695sSJE82OHTuaoaGhZu/evc2pU6eaFRUVx62Lt99+2xwwYIAZGhrqeb0nKm+19957zxw2bJgZGRlphoSEmL179zafeeaZGuslVTvee800T75W0M9Vl+lE/459j3722Wfmueeea8bExJg2m81MT08377//fjM3N7fGdb///nvzzjvvNPv372/GxsaaoaGhZnp6unnTTTeZ69at85yXm5trPvnkk+aoUaPMjh07mjabzUxKSjJHjBhhvvvuu3WukyStj2GazbiOuIj43aOPPspzzz3H/PnzOf/88/1dHBERv1HoEWklDh48SMeOHWscW7duHUOHDsVms3HgwIEaCwGKiLQ1GtMj0kqcdtppZGRk0LdvXyIiIti2bRuffvopLpeLf/zjHwo8ItLmqaVHpJV48sknmTVrFrt376agoICYmBjOOOMM7rvvvkatciwi0too9IiIiEiboHV6REREpE1Q6BEREZE2QaFHRERE2gSFHhEREWkTNGW9Sm5uLk6n0+vXTUhIqLHPjdSP6q3hVGeNo3prHNVb46jeGu54dWa1Wj1bsNSXQk8Vp9NJRUWFV69pGIbn2pokV3+qt4ZTnTWO6q1xVG+No3prOG/Xmbq3REREpE1Q6BEREZE2QaFHRERE2gSFHhEREWkTFHpERESkTVDoERERkTZBoUdERETaBIUeERERaRMUekRERKRNUOgRERGRNkGhR0RERNoEhR4RERFpE7ThaDOpdJnklzmpdJQQ5O/CiIiIiFp6msuGI8Xc9NF2fvfRWn8XRURERFDoaTb2EHf7jqOk3M8lEREREVDoaTb2UHfPYV5JBS7T9HNpRERERKGnmVS39LhMKCyv9HNpRERERKGnmVgtBhHB7urNL1XoERER8TeFnmZkD3W39uSXKfSIiIj4m0JPM7KHuMf1KPSIiIj4n0JPM4quGteTV+r0c0lEREREoacZRal7S0REJGAo9DSjaHVviYiIBAyFnmZUPW1ds7dERET8T6GnGUV7urc0pkdERMTfFHqakd0zkFktPSIiIv6m0NOMojSmR0REJGAo9DQjdW+JiIgEDoWeZlTdvVXqNClzuvxcGhERkbZNoacZhQdbsFoMQF1cIiIi/haQoWf+/PncfffdXHfddTzyyCNs3769Xo9bsmQJV111FS+88EIzl7B+DMMgJiwYgAKFHhEREb8KuNCzdOlSpk2bxoQJE5g6dSqpqak8++yz5OXlnfBxR44c4e2336ZXr14+Kmn9xIbbAMhT6BEREfGrgAs9c+bMYcyYMYwaNYqUlBQmTZqEzWZj4cKFx32My+XiL3/5C1dddRWJiYk+LO3JRVe19ORr/y0RERG/svq7AMdyOp3s3LmTSy+91HPMYrHQr18/tm7detzHffjhh9jtdkaPHs2mTZtO+BwVFRVUVFR4bhuGQVhYmOdrbzIMg9jwqtBT7vL69Vur6npSfdWf6qxxVG+No3prHNVbw3m7zgIq9OTn5+NyuYiJialxPCYmhoMHD9b5mM2bN/PVV1/VexzPzJkz+fDDDz23u3TpwtSpU0lISGh0uU8kJszdLecKDqNDhw7N8hytVVJSkr+L0OKozhpH9dY4qrfGUb01nLfqLKBCT0OVlJTwl7/8hTvuuAO73V6vx1x22WWMHz/ec7s6PWZlZeF0ercLyjAMYsPcY3oOZjs4dOiQV6/fWhmGQVJSEpmZmZim6e/itAiqs8ZRvTWO6q1xVG8Nd6I6s1qtDW6wCKjQY7fbsVgsOByOGscdDket1h+Aw4cPk5WVxdSpUz3Hqivlmmuu4ZVXXqmVDoODgwkODq7z+ZvjTVg9pievtFJv8gYyTVN11kCqs8ZRvTWO6q1xVG8N5606C6jQY7Va6dq1K+vXr2fIkCGAe5Dy+vXrGTt2bK3zO3bsyB/+8Icax95//31KS0u56aabiI+P90m5T8QzpkerMouIiPhVQIUegPHjx/Paa6/RtWtXMjIymDt3LmVlZYwcORKAV199lbi4OCZOnIjNZqNz5841Hh8REQFQ67i/VK/To8UJRURE/CvgQs/QoUPJz89nxowZOBwO0tLSeOSRRzzdW9nZ2S1q5Hv1Oj352mldRETErwIu9ACMHTu2zu4sgClTppzwsXfffXczlKjxPCsyl1fiMk0sLSiwiYiItCYBtzhha1MdelwmFJZr01ERERF/UehpZtYgCxHB7mrWYGYRERH/CcjurdbAPHIQ1+LPyItPwB7ai6IKl3tcT/2WExIREREvU0tPcynIx/xsJkVffoo9xJ0tNYNLRETEfxR6mos9BgCXIxt7SBCg0CMiIuJPCj3NxR4LgFlWht3qXkVS09ZFRET8R6GnmRghIRAWDkA07l3d8zSQWURExG8UeppTVWtPVGUJoO4tERERf1LoaU7RMQDYK4oBdW+JiIj4k0JPMzKqWnrsZQWAWnpERET8SaGnOUVXhZ6SXECLE4qIiPiTQk8zMqpCT1RBDqCWHhEREX9S6GlOVd1b0QVHACh1mpQ5tf+WiIiIPyj0NKeqgcyhjmysVTWt1h4RERH/UOhpRtXdW0Z+rraiEBER8TOFnuZU1b1Ffh72kOqd1hV6RERE/EGhpzlFRYNhgOnCXrWffV6pZnCJiIj4g0JPMzKCgrBUT1s33GGnQC09IiIifqHQ08yCYtsBYHeVAZCnVZlFRET8QqGnmQXFxgNg1/5bIiIifqXQ08ws1S095YWAVmUWERHxF4WeZubp3irNB9TSIyIi4i8KPc3ME3qKjgIa0yMiIuIvCj3NrDr0RBW699/S7C0RERH/UOhpZp4xPY5MAArKK3GZpj+LJCIi0iYp9DSz6tlbUbnu0OMyobBcm46KiIj4mkJPM6vu3rIWFxARXLUVhVZlFhER8TmFnmZmRESCNRgAu/s/zeASERHxA4WeZmYYBlRvRWFxd2vlKfSIiIj4nEKPL1SHHioAzeASERHxB4UeHzDsMQBEuUoB7bQuIiLiDwo9vlDV0hPtLAI0pkdERMQfFHp8we4OPVGlVftvaVVmERERn1Po8QGjekxPqQNQS4+IiIg/KPT4QnQMAPaqrSg0e0tERMT3FHp8wKjq3rLnZwFQUKaBzCIiIr6m0OML1d1bjsOAdloXERHxB4UeX6gOPSUOAMoqTcqc2n9LRETElxR6fMAItkF4BGGVZVgN9zENZhYREfEthR5fscdiAHarCSj0iIiI+JpCj69Ud3EZ7kHMCj0iIiK+pdDjI9VbUdgpB7QVhYiIiK8p9PhKdUuPswTQpqMiIiK+ptDjK9Vr9VS4t6LQtHURERHfUujxlapVmaNKCgCN6REREfE1hR4f8azKXHwUgHytyiwiIuJTCj2+Uj2mp2r/LbX0iIiI+JZCj69Ubzpa4N5/S2N6REREfEuhx1ci7WBYsJcXAZq9JSIi4msKPT5iWILAHu2ZvVVQXonLNP1cKhERkbZDoceX7DFEVRQD4DKhsFybjoqIiPiKQo8vRcdiNV1EWNxhJ1+rMouIiPiMQo8PeaatUwFoBpeIiIgvKfT4UvUMLlcZAHkKPSIiIj6j0ONL1S09Tve4Hs3gEhER8R2FHl+KjgPAXubeikI7rYuIiPiOQo8PGZ79t/IAjekRERHxJYUeX6ru3iqq2n9LqzKLiIj4jEKPL1Xvv1WSC6ilR0RExJcUenwpNAxsNs9WFJq9JSIi4jsKPT5kGAbYY7FXVO+/pYHMIiIivqLQ42vRP4Ue7bQuIiLiOwo9vmaP8YSeskqTMqf23xIREfEFhR4fM6JjCassw0rV/lsa1yMiIuITCj2+Zo/FAOxmOaDQIyIi4isKPb5Wvf9WZSmgVZlFRER8RaHHxzw7rVeN61FLj4iIiG8o9Pha9QKFZfmAQo+IiIivKPT4WlVLT1SxA9BWFCIiIr6i0ONr9hgAoqt2WldLj4iIiG8o9PiYERwM4ZFEecb0aCCziIiIL1j9XYC6zJ8/n9mzZ+NwOEhNTeWWW24hIyOjznOXLVvGzJkzyczMpLKykqSkJC666CKGDx/u41I3QHQs0RWFgFZlFhER8ZWACz1Lly5l2rRpTJo0iW7duvHpp5/y7LPP8sorrxAdHV3r/MjISC6//HI6duyI1Wpl5cqV/PWvf8VutzNw4EDfv4D6sMcQdbgYUPeWiIiIrwRc99acOXMYM2YMo0aNIiUlhUmTJmGz2Vi4cGGd5/fp04chQ4aQkpJCUlISF1xwAampqWzevNnHJa8/IzoWe1VLj0KPiIiIbwRUS4/T6WTnzp1ceumlnmMWi4V+/fqxdevWkz7eNE3Wr1/PwYMHue666+o8p6KigoqKCs9twzAICwvzfO1N1derdd3oOM86PYXllbhMCLJ497lbsuPWmxyX6qxxVG+No3prHNVbw3m7zgIq9OTn5+NyuYiJialxPCYmhoMHDx73ccXFxdxxxx04nU4sFgu33nor/fv3r/PcmTNn8uGHH3pud+nShalTp5KQkOCV11CXpKSkGrfzO3WmosLdveUyITI2nphwW7M9f0v183qTk1OdNY7qrXFUb42jems4b9VZQIWexgoNDeXFF1+ktLSUdevWMW3aNNq3b0+fPn1qnXvZZZcxfvx4z+3q9JiVlYXT6d2ZVIZhkJSURGZmJqZpeo67jCCsposIVzlFFhvb9h4kJTrEq8/dkh2v3uT4VGeNo3prHNVb46jeGu5EdWa1WhvcYBFQocdut2OxWHA4HDWOOxyOWq0/x7JYLJ4UmJaWxoEDB5g1a1adoSc4OJjg4OA6r9Ncb0LTNGteu2qtHruzmCKbDUepk2S7Wnp+rla9yUmpzhpH9dY4qrfGUb01nLfqLKAGMlutVrp27cr69es9x1wuF+vXr6d79+71vo7L5aoxbifgVO+/Va7BzCIiIr4SUC09AOPHj+e1116ja9euZGRkMHfuXMrKyhg5ciQAr776KnFxcUycOBFwj9FJT0+nffv2VFRUsGrVKr755htuu+02P76Kk6jafyuqNB8itRWFiIiILwRc6Bk6dCj5+fnMmDEDh8NBWloajzzyiKd7Kzs7u8Yo7rKyMt544w1ycnKw2WwkJyfzq1/9iqFDh/rpFdRDRBRYLJ4FCrUqs4iISPMLuNADMHbsWMaOHVvnfVOmTKlx+5prruGaa67xQam8x7BY3AsUVk1bz1P3loiISLMLqDE9bYo91rNWT4G6t0RERJqdQo+/RP8UetTSIyIi0vwUevzEsMdgL6/eaV2hR0REpLkp9PjLMS09+aUayCwiItLcFHr85ZgxPWrpERERaX4KPX5iRMd4Qk9ZpUmZ0+XnEomIiLRuCj3+Yo8lrLIMq8vdtaXWHhERkeal0OMv0bEYuPffAsjTtHUREZFmpdDjL9ExwLH7b2kws4iISHNS6PGXkDCwhWgws4iIiI8o9PiJYRjuaetaq0dERMQnFHr8KTr2p/23NKZHRESkWSn0+JM9lujq/bfU0iMiItKsFHr8yIg+dqd1DWQWERFpTgo9/mSPJbqiavaWurdERESalUKPP0XHElXhXqdHA5lFRESal0KPHxn2WOzVLT0KPSIiIs1Kocefjtl/q7C8kkqX6ecCiYiItF4KPf5k/6l7y2VCUblae0RERJqLQo8/2aOxmi4iNK5HRESk2Sn0+JFhDYbIKE8XV55Cj4iISLNR6PE3e6z23xIREfEBhR5/iz4m9GitHhERkWaj0ONnhj3mmJYercosIiLSXBR6/O2Ylh6N6REREWk+Cj3+Zv9pp/UCdW+JiIg0G4Uef4uOIbpcLT0iIiLNTaHHz4xjWno0e0tERKT5KPT4W3Qc0Z7ZWxrILCIi0lwUevwtOkYtPSIiIj6g0ONv4ZFEu0oBKKs0KXO6/FwgERGR1kmhx88Mi4XQiHCsLnfXllp7REREmodCTwBwL1BYCECepq2LiIg0C4WeQBAdi92z07oGM4uIiDQHhZ4AYETHYi93t/Soe0tERKR5KPQEAnvMMS09Cj0iIiLNQaEnEETHakyPiIhIM1PoCQCG/acxPQVq6REREWkWCj2B4JgFCvM0kFlERKRZKPQEAnss0VXdW/nq3hIREWkWCj2BwB5DVPVA5tIKPxdGRESkdVLoCQBGaBh2qlZkVkuPiIhIs1DoCRD2MCsABRUuKl2mn0sjIiLS+ij0BIioiFAATAyKytXaIyIi4m3Wpjw4Ozub7Oxsevbs6Tm2e/du5syZQ0VFBcOGDWPIkCFNLmRbYI2OJqKimKLgcPLKKrGHNulbIyIiIj/TpJaef/3rX3zwwQee2w6HgyeffJJly5axadMmXnrpJZYtW9bkQrYF7rV63NPWtSqziIiI9zUp9OzYsYN+/fp5bi9evJjy8nJefPFF/v73v9OvXz9mz57d5EK2CdHHhB4NZhYREfG6JoWewsJCoqOjPbd//PFHevfuTVJSEhaLhSFDhnDgwIEmF7JNiFZLj4iISHNqUuix2+1kZWUBUFRUxLZt2xgwYIDnfpfLhcvlaloJ2wjjmNCjVZlFRES8r0mjZfv168e8efMIDw9nw4YNmKZZY+Dy/v37adeuXZML2SZoTI+IiEizalLomThxIocOHeLtt9/GarXyi1/8gsTERAAqKir47rvvGDZsmFcK2uods/9WfolaekRERLytSaEnJiaGp59+muLiYmw2G1brT5czTZPJkycTHx/f5EK2CZHRRFeHnuIyPxdGRESk9fHKYjDh4eG1jtlsNtLS0rxx+TbBsFqJCnJ/nVei/bdERES8rUmhZ926dezatYuLL77Yc+yrr77igw8+wOl0MmzYMG644QYsFi38XB/RIe56KtCYHhEREa9rUhr54IMP2L17t+f23r17ef3117Hb7fTu3Zt58+bxySefNLWMbUZUuA2APA3pERER8bomhZ4DBw6Qnp7uub148WLCwsJ46qmn+O1vf8uYMWNYvHhxkwvZVsREhgFQblooc2qqv4iIiDc1KfSUlpYSFhbmub169WoGDhxISEgIABkZGZ51fOTkQu1RWF3uZp48rcosIiLiVU0KPfHx8ezYsQOAzMxM9u3bR//+/T33FxYWEhwc3LQStiHuBQoLAa3VIyIi4m1NGsh81lln8eGHH3L06FH2799PREQEgwcP9ty/c+dOOnTo0ORCthnRsdh3FXM0JIZ8rcosIiLiVU0KPZdffjlOp5NVq1YRHx/PXXfdRUREBOBu5dmwYQMXXHCBVwraFhj2GOzlewC19IiIiHhbk0JPUFAQ1157Lddee22t+yIjI3n99debcvm2JzoWe8UmQGN6REREvM0rixOCe1BzdnY24B7rExoa6q1Ltx3HjukpKfdzYURERFqXJoee7du3884777B582bPjuoWi4WePXty/fXX15jSLicRHom9shSAvIISPxdGRESkdWlS6Nm2bRtTpkzBarUyevRokpOTAff6PUuWLOGJJ55gypQpZGRkeKWwrZ1hGNiD3MGxQPtviYiIeFWTQs/7779PXFwcTz/9NDExMTXuu/LKK5k8eTLvvfcekydPbsrTtCl2m3sVgbxSzd4SERHxpiat07Nt2zbOPffcWoEH3Duwn3POOWzbtq0pT9Hm2MPc6xrll2tFZhEREW9qUugxDIPKyuPPMnK5XBiG0ZSnaHPsEe7VrPMrtUmriIiINzXpk7VHjx589tlndW41kZ2dzeeff07Pnj2b8hRtTnRkOACFZhCVLtPPpREREWk9mjSm59prr+WJJ57g3nvvZciQIZ7Vlw8ePMgPP/yAxWKpcw0fOb7ImEgoBhODwvJKokO9tqqAiIhIm9akT9QuXbrw3HPP8d577/HDDz9QXu5eW8ZmszFw4ECuvPJKoqKivFLQtiI4OpbIPcUUBoeTX6bQIyIi4i1N/kRNSUnh/vvvx+VykZ+fD4DdbsdisfDRRx8xffp0pk+f3qBrzp8/n9mzZ+NwOEhNTeWWW2457rT3L7/8ksWLF7Nv3z4AunbtyrXXXttyp8lHxxJVkeUOPaWVEO3vAomIiLQOXhsta7FYiImJISYmBoul8ZddunQp06ZNY8KECUydOpXU1FSeffZZ8vLy6jx/48aNDBs2jCeeeIJnnnmGdu3a8cwzz3D06NFGl8Gv7DHYK4oAyNOmoyIiIl4TcFOE5syZw5gxYxg1ahQpKSlMmjQJm83GwoUL6zz/17/+Neeffz5paWkkJydz5513Ypom69at83HJvcQe6wk9+VqVWURExGsCKvQ4nU527txJv379PMcsFgv9+vVj69at9bpGWVkZTqeTyMjI5ipmszJCQrC73Ksx5+cX+bk0IiIirUdAjZLNz8/H5XLVWuwwJiaGgwcP1usa77zzDnFxcTWC07EqKiqoqKjw3DYMg7CwMM/X3lR9vYZe125xL0yYX1TaJtc5amy9tWWqs8ZRvTWO6q1xVG8N5+06a3Do2blzZ73P9fW4mlmzZrFkyRKmTJmCzWar85yZM2fy4Ycfem536dKFqVOnkpCQ0GzlSkpKatD5cWFBAJQ68SwD0BY1tN5EddZYqrfGUb01juqt4bxVZw0OPQ8//LBXnrgu1bO+HA5HjeMOh6POrS6O9cknnzBr1iwmT55Mamrqcc+77LLLGD9+vOd2dXrMysrC6fTuwGHDMEhKSiIzMxPTrP9Cg+HuzEN2QQmHDh3yaplagsbWW1umOmsc1VvjqN4aR/XWcCeqM6vV2uAGiwaHnl/+8pcNfUi9Wa1Wunbtyvr16xkyZAjg3spi/fr1jB079riP+/jjj/noo4949NFHSU9PP+FzBAcHExwcXOd9zfUmNE2zQdeODg8GF+RVNF+ZTsRR6uSL7Q7GdY8l0hbk8+ev1tB6E9VZY6neGkf11jiqt4bzVp01OPSMHDmyyU96IuPHj+e1116ja9euZGRkMHfuXMrKyjzP++qrrxIXF8fEiRMBd5fWjBkz+PWvf01iYqKnlSg0NJTQ0NBmLWtziYoIhQIocPlnnPmbPxxh8Z58Cstd3HxKol/KICIi4m0BNZAZYOjQoeTn5zNjxgwcDgdpaWk88sgjnu6t7OzsGgOavvjiC5xOJy+//HKN60yYMIGrrrrKl0X3mmh7BBRAHnW3SDWnovJKvt9fAMCqg0XcfIrPiyAiItIsAi70AIwdO/a43VlTpkypcfu1117zQYl8KzrGDgeg3LBS5nQRYvVdi893+woor3Q3Ie7JKyO3xElsWEC+TURERBokoNbpEbfQ2GiCXe5p9XmllT597oU7a658vSZTawWJiEjroNATgIzoOKKqt6IoLffZ8x4uLGf9kRIMYHiqHVDoERGR1kOhJxBFRWOvKAYg31Hgs6ddtNu9YWy/pHDGpLt3Ol1zqFizDEREpFVQ6AlARlCQZyuKPEehT57TNE0W7nSHnlFdoumVEEawxSCnxMmBfN+1NomIiDQXhZ4A1cF0dyutPFzqk+fbllPKwYJyQoIMzugUSYjVQq9E9/YcazKLfVIGERGR5qTQE6DOq9wHwFJHEDnFFSc5u+kW7nIPYD6jUxThwe4FCQckRQCwWuN6RESkFVDoCVBdEyLp7dhJJQZztzqa9bkqKk2+2eMeOzSqa7Tn+ICkcADWZRZT6dK4HhERadkUegKU0bM/4/d/C8Bn2x2UOV3N9lwrDxZSUFZJbJiV/u3DPce7xoYSZbNQ4nSxLcc33WwiIiLNRaEnUHXvy+Cjm0goPUpBWSWLq2ZWNYeFu9zXHpFmJ8jy02rXQRaDfuriEhGRVkKhJ0AZ4REEdenGuANLAZi9JbdZpo4XlFWy4oB7htioLvZa91d3ca05pNAjIiItm0JPADN6DeCcQysIMZ3scZSx7rD3Z1F9uycfp8ukS2wIabG1N2gdWNXSsyW7hJKK5utiExERaW4KPQHM6DWASGcJI7PWADBnS67Xn+Prqq6tkXW08gAkRdloHxlMpQkbjmjquoiItFwKPYGsaw+whXDh7q8AWL6/kMwC7y0UeKignM3ZJVgMGJ4Wfdzzqru4NK5HRERaMoWeAGZYg6F7X1KKsxgUXIgJfLrVe609X1etzTMgKYK4E+ykXt3FtfaQWnpERKTlUugJcEavAQBcmP0jAF/uyKO4ouk7r5um6enaqmsA87H6tQ/HAPbklXG0xNnk5xYREfEHhZ4AZ/R2h56BGxbQMSqY4gqXZ4+spticVUJmYQWhVgtndIo64bn2UCtd40IAWKsuLhERaaEUegJdx1SIisZSXsr4WPcCgXO2HMXVxOnr1WvzDO0cRYj15G+D6i0p1ij0iIhIC6XQE+AMi8XTxTUyew0RwRYOFlSw8mDjw0d5pYtv99Sva6uaZx+uQ8XNsl6QiIhIc1PoaQmqQk/o5lWck+6eZTW7CdPXVxwopKjCRXy4lb7HbDtxwiIkhBFsMTha4mR/vvdmkImIiPiKQk8LYPQa6P5i1zYu6ByKxYDVh4rYl1fWqOtVjwka2SUai2Gc5Gy3EKuF3olhgLq4RESkZVLoaQGMdgmQ2BFMF+0PbGJwciTQuMUK80qdrDzo3nbieAsSHs+xXVwiIiItjUJPC1E9i8vcuIaLesYCsHBnHoVlDZu+/s2efCpNyIgLpVN0SIMeWx161h8uxunSuB4REWlZFHpaiOouLnPTGvomhpMWE0JZpcnnOxwNus7Jtp04ka5xIUTZLJQ4XWzLKWnw40VERPxJoael6NEPDAtk7ofcHE9rz9wtuVTWs9Vlf14Z23JKCTLg7LSGhx6LYdC/euq6urhERKSFUehpIYyISEjLAMDcvIbhaXbsIUFkFTtZtr+gXteoXpvnlI4RxIQef9uJE9F6PSIi0lIp9LQg1ev1sGkNtiAL52fEADB788kHNLtM07PX1qgux99c9GQGdnBPcd+SXeKV7TBERER8RaGnBakOPeamNZimybjuMQQZsDGrhJ1HS0/42PWHi8kudhIRbGFwSmSjy9A+0kZSZDCVJmw4rHE9IiLScij0tCTpPcFmg7xcOLiPduHBDOvsHpsze8vREz60egDzsNQobEFN+7ari0tERFoihZ4WxAi2QUYfAMxNqwEYXzWgefHuAhzH2QG9zOliyV73uJ+mdG1VG1DVxRWIoWfD4WLPFhsiIiLHUuhpYTzr9WxaA0CP+DC6twvF6TKZv91R52OW7S+k1OmifWQwvRLCmlyGfu0jMIC9eeXkFFc0+Xresiu3lMe/2seL3x5k0xHNLhMRkZoUeloYz2DmLesxne6WnYt6xgEwf2suFZW1p68v3OkewDyyix2jnttOnIg9JIiucaEArM0MjHBR5nTxh28PehZNnLvN4d8CiYhIwFHoaWlSukBkFJSVwO6tAAztHEVcmJXc0kqW7K3ZtZNb4mR1VTfUyLSmd21VG5gUWF1c/1p5hP355UTY3G/ppXvzcZTW3d0nIiJtk0JPC2NYLBg9f9qSAsBqMRjXPQZwT183zZ9aexbvzsdlurvBOtptXivHgA5V+3BlFtd4Pn/4fl8B86tadh44K5lu7UJxuuDL7Xl+LZeIiAQWhZ6WqFfNcT0A52fEEGwx2H60lM3ZP00lX+hZm6fhKzCfsAgJYdiCDHJLnOzLL/fqtRsip7iCV78/BMBlveIY2CGCC7q7B3fP31b/1apFRKT1U+hpgTzjenZtwSx1j6mJDrUyoirYVC9WuDu3lF25ZVgtcFaqd0OPLchC76pB0WsO+aeLq9Jl8selhygod5EeF8J1AxIAGNY5iiibhaxiJz9W7Sgv0poUlVfy2rJDrNT7W6RBFHpaICMhCRKSoLIStm7wHL+oh7uF47t9BWQVVXi2nTgtOZKokCCvl8Pf6/XM2nSUdYeLCbUa/H5YMsFB7kHaIVYLY9JjAJi31eGXsok0p482HuXz7Xk8t+gAW7K1SKhIfSn0tFBGHV1cabGh9G0fjsuET7fksmi3O/R4Y22eulSP61l3uMQza8pXtuWU8M6aLAAmndae5J+NVxrbLQaAlYeKOFTgv+43EW8rqXAxf5u7NbfCZfL84gMBtXSESCBT6Gmh6go98FNrz8ebj5Jb4iTKZuHUjo3fduJEusSGEBUSRKnTxTYf/rVZXFHJS0sOUmm6u7LGdK0d6jpE2TilKpTN1/R1aUW+3OGgsNxFh6hgOkfbyC1x8tyiA5Q5Xf4umkjAU+hpqXr0B8OAA3sw837acHRwciTtI4Opbng5K9Xu6fbxNoth0L+9e+r6ah92cb3+wxEOFVSQEG7lriFJx117qHpG24IdDn0gSKtQ6TL5pGrM3qW94nh0RApRNgvbj5by6veZfp9JKRLoFHpaKCPKDp26Aj9tSQEQZDG4sGr2EsCoOlpBvGlgh+pxPb5ZpHDx7ny+2pmHxYDfDutI5AnGKp3aMZKEcCsF5T9twyHSki3dW8CRogqiQ4IY1SWapCgbD5ydTJABi/fk87+NJ96DT6StU+hpwTyzuDbW7OI6Jz2ajlE2+ieF071daLOWYUDVIoVbsksorqhs1uc6XFjO35dnAnBl33b0SQw/4flBFoOx3dwBcN7W3BOeKxLoTNNk5iZ3qLmgRywhVvev7/5JEUw6rT0A/12dxfL9Cvgix6PQ04Iduw/Xsc3aEbYg/nZxV54e09kr206cSPtIG0lV3WnrDzdfa0/19PSiChc94sO4um98vR53TkY0VgtszSlle05ps5UPYMX+Qt5aeYTC8uYNf9I2rT9SzI6jpdiCDC6oGqhfbVz3WMZ1i8EEXlpyiD2OMr+UUeTnTNPk8+2OZv+juL4UelqyjN5gDQZHDmQe8Fsxfpq63nyh54P1OWzKKiE82MLvh3UgyFK/MBcTamVoJ/caRfO2NV9rz4H8cqZ+c4BZm47y+IK95GsLDPGymVVdV2O6RmMPtda6/7bT2tO3fTilThfPLtpPfllgfMhI27byYBGvLcvkV3N21bk3pK8p9LRghi0EMnoBNcf1+NrADs27D9emI8VMX58NwC+HJNE+smHbaVQPaF68O5/CZvggME2Tvy3PpKJq9PiOo2U88uVejpYo+Ih37HWU8ePBIgzgkl5xdZ5jtRg8eFZH2kcGc7iwgqnfHPD5UhIixzJNk3fWun93N+ekmoZQ6GnhjN4DgdpT132pX/sIDGBfXrnX1wspLK/k5aUHcZnurTSGpzV8ZeleCWGkxoRQXmny1S7v78e1cFc+6w4XYwsyeHxkCnFhVvbllfPw53s4Uqj1U6TpZlWN5TmjUyQdoo4f+u2hVh4bkUKo1cL6w8W88cNhXxVRpJZl+wvZcbSUUKuFy3vXHdZ9TaGnhfMMZt6yDrPSP83ZUSFBpMe5B0x7s4urugXlSJGTpMhgbh/cvlHXMQyDcVVjIOZtzcXlxWm9+aVO/rXyCADX9ovn1ORInj+3M+0jg8ksrODhL/Zw0I97k0nLd7TEyaLd7rB+We92Jz2/c0wIvx/WAQOYt82hQfziFy7T5N2qVp6LesQSXUeXrD8o9LR0nbtCeCSUFMPubX4rxk9T173XxfXVzjy+3VNAkAG/H9aR8ODGb6UxooudMKuFgwUVrPViMPvXyiMUlFWSFhPCxVXdDklRNp47tzPJdhvZxU4e/mKPBpZKo83ZfBSny91i2SM+rF6PGZISxfUD3XvRvf7DYdb6aasYabuW7Clgj6OMiGALlx6nS9YfFHpaOMMSBD37A/7t4qqeur4ms9grC6Ttyy3mHyvc09Mn9k+gez1/2R9PeHAQo7p6d0DzmswiFu7KxwDuOj0J6zGDq+PDg3nu3M6kxYTgKK3k0S/2NPvsMWl9iisqmb/dAcBlDfzguKJ3HMPT7FSa8MI3B8jUdiziI5Uuk/fWuVt5LukVd8L11HxNoacVON6WFL7UMyEMW5BBbomTfXlN++VaUWny6JwNlDpN+rYP5zIv9QWPq1qzZ/n+QrKbOPaovNLF36rWDBrXPabOv8BjQq08c05nurULpaDcxeQFe9l4xDeLOErrsGBHHkXlLjpG2Ric0rDtZAzD4J7Tkzzvv2cX7Q+YacPSui3anc+B/HKibBYu6hl78gf4kEJPK1C9Xg87NmOW+ac1wRZkoXeid2Zxvbs2i02ZBUTaLPx2aP2np59M55gQ+iaG4TLh86q/nhvrg/U5HCqoIC7Myi+quhHqEhUSxFNjOtEnMYziChdTvtrH6kPqapCTc2854R7AfGmvOCyNWHMrxGrh4eHJxIZZ2ZtXzstLDnl1TJvIzzldJtOrWnku792uScMSmoNCT2uQ0AHaJUKlE7Zt8FsxfuriaviHenFFJUv35vOn7w7x0YYcAO45owPx4cFeLeO4qi06Pt/maPR03r15ZXy00V3G209rf9If6vDgIJ4Y1YlTOkRQVmny9Nf7tWqunNSSvQUcKXISHRLEyC4Nn7VYrV14MI8MTybYYrDiQCHvrMn2YinbjqMlTv5v8QF+P283b/54mGX7C5plCYyW7qudeWQWVhAdGsQFPQKrlQcgMIZTS5MYhoHRawDmt19gblqD0fdUv5RjYFIE/yGLdYdLcLrMGmNc6nIgv5wfDhTyw4FCNmYVc+yeoBMGJjO0s93rGyienhJFTGgQuaWVLNtXwLDUhn2YuEyTvy7LxOlyb+56Rqf6dTmEWC08MiKZPyw5yPf7Cvm/xQf47dCOnN2IKfjS+pmmyaxN7mB97JYTjdU9Pox7zkjij0sP8eGGHDpH2xjRpXn35WtNtuWU8NyiA561t7YfLeWTzbkYQJfYEPq2D6dv+3D6JIQH1PgVX6uodHlaea7s047QJr5vm4NCT2vRawB8+wXmRv+N60mLDcEeEkR+WSVbs0s83V3VKipdbDhS4g46Bws5VFBzXE3HKBunJUcwOCWK8wemk5mZ6fUyBgcZnJcRw4z1Oczd5mhw6Pliex6bskoItRrcMbh9g7b5CA6y8MBZyfz5u0N8vTufl5cepKzSxTnpMQ18FdLarTtczI6jZXVuOdFYI7tEs8dRxkcbj/Lqskw62m10a9e0CQJtwde78nj1e/fioyl2G5f1jmNrdinrjxRzIL+cnbll7MwtUwgCPt+eR3axk3ZhVs730vvW2xR6WgmjZ39MgP27MPMdGPYYn5fBYhj0Twrn2z0FrMksondiOEdLnPxYFXJWHyqm9JjmHKsF+iSGc1pyJKd1jKSj3b3ommEYzbpn2HkZMXy4IYf1h4vZm1dG5+iQej0ut8TJf1a71+S5bkACCREN73oLshj8ZmgHQqwWPtvu4C/fZ1LmNLkwAJuBvc3pMsksKGdffjkH8srZn1/G/vxyDuSXYwsy6B4fRo92YXSPDyWjXWjAjQXwperFCI+35URjXT8ggX15Zaw4UMRziw7wxwvSiAmQ9VMCTaXLZNrqLM/3YnByJL8b1oHw4CDOSXefc7TEyfrDxe5/9QhBfRMj/PeCmlmZ08UHVSvnX9m3HbagwGvlAYWeVsOwx0BKF3fo2bwWY8hwv5RjQFIE3+4pYMGOPFYcKGLH0ZoDq2NDgzg1OZLTkiMZkBTulw+2hIhgBidHsmx/IfO35nL74KR6Pe7NHw9TVO4iPS6UC7s3PqRYDINfDmlPqNXg4825/POHw5Q4XUzoc/KF51qCovJKDuSXsz+/nP157mCzP7+czIJyjrf1TnGFe1bd8v2FAFgM6BQdQo/4ULq3c69PkxJta9Rg3pZmTz22nGisIIvB74Z15IHP9rAvr5y3Vh7ht0M7evU5WoPC8kpe+vYgK6smHVzZpx0TB8TXev/FhVkZnvbTSvEnC0EWA245o5yLuob6/DU1t3nbcsktrSQxwhrQrdcKPa2I0XsA5v5dsGkN+Cn0DKzafDSr2ElWsRMDyGgX6mnN6RoXEhAfXBd0j2XZ/kK+2pnPLwYmEhZ84r9KVh4s5Js9BVgMuPv0pCbPKDMMg5tPSSQ02ML0dTm8vTqL0goX1w2Ib9ZWrqaqdJkUlFVytMSJo9RJbomT3JJKsosrPEHnRHuOhVotpNhtpNhtJEfb6GQPIdluo7jCxdacErZkl7A1u4QjRU72OMrY4yjj8+3u1YjDgy1ktAulR1UI6h4fGjCrvHrTT1tORJ1wy4nGCg8O4jdnduD++Xv4elc+56RH0699622BaKj9eWU8u+gABwvcLZC/ObMDZ9WzG7w+IeiN73az43A0dw9JCoi9qLyhpMLF/za437dX94sP6NfV+n5jtGFGrwGYn8/C3Lga0zT98uGZGBnMdf3j2ZdXzqCOEZzSMSIgm8/7J4XTMSqYgwUVLNqdx9hux2+5KXW6+Nty9x5G43vEerbcaCrDMJjYP4FQq4X/rMrigw05HC6s4BcDE0iM9O6stZOpqHSRXezEUeLkaKkTR8nPg42T3NJK8kqd1GfSW2yYlU52G8l2GynRNlLsIaRE22gXZj3u+7Jnwk/jS46WONlaFYC25JSyPaeE4goXazOLa6yonRQZTM+EMK7tF09SMwQEX8sprmCxZ8uJ5lvFtlu7MMZ2i2HeNgd/X36YVy7oEtAfVL7yw4FCXlpykOIKF/HhVh4dkULXJvy8/zwEfbHdwd+WH2bhzjyOFlfw4NnJRNhafjfup1tyyS+rpENUMKMCfIB84H0aSeN16wNBVjiaBVmHINE/zdZX9Yv3y/M2hMUwGNstln+tPMK8rQ7Oz4g57ofx9HXZHCmqICHcysT+x1+Tp7Eu7+2e5fDPFYdZvCefpfvyOT8jhiv7xhMb1rw/ojnFFXyyOZf52xw1xludiAFEhwYRG2YlNtRKTJiVuDCrO+BUBZ2m/iKPC7NyRqcozugUBbhbmPbllbElu5Qt2e4Wof355WQWVpBZWMGGw8VMPT+Vdl5e4sDX5mzJxemC3g3YcqKxrh+YwNJ9BezPL+fjTUeZ0Ld1dK82hmmafLTxKG+vzsLEXf8PDk/2+h9s53WLJSOlPQ99vJY1mcU88sVeHh+V0qLft0Xllcysmml4Tb94r62r1lwUeloRIyQU0nvC1vWYG9dg+Cn0tBRjukbz3zVZ7HaUsTmrhF4/m20GsCu31NPdcMfgpJN2gzXWBd1jyYgL5b9rsliTWcynWx18sSOPi3rEclnvdkR5eQbIoYJyPtqYw1c78z3rFYUEGe4gE2YlJtRKXFiQJ9C4b7sDTnRIkM9/sQVZDNJiQ0mLDfXMCiksr2RbTin/XHGYgwXlPPnVfp47t3OLnS1TXFHJZ9scAD7ZqyjSFsQtpyTyx6WHmL4+m7PTomgf2fJbyxqqzOni1e8zWbwnH4DzM2KYdFr7Zmv5Gta1Hc+dm8pTC/ex21HG/Z/t4fGRKaTFtsxxPp9sPkphuYtO0TbObuBsWH9Q6GlljF4DMLeud29JMXKcv4sT0CJDghieZufLHXnM3eaoFXoqXSavLcvEZcKZnaIavA1AQ3WPD+OpMZ1Zm1nEf9dksSW7lP9tPMq8bQ4u6xXH+J6xTR74vfNoKR9uyOG7fQWebqreCWFc0acdp3aMCOjxRD8XaQtiUIcIpoxO4aHP97Inr4xnFu3nydGdmryujT98sT2PoorGbTnRWCPS7CzYkcfaw8X8c8VhHhuZ0qLeA02VVVTB84v3s+NoGUEGTDqtvWcB0+aU0S6MF85P5amF+9mfX87DX+zl4eHJ9E9qWWOr8ssq+XiTey/Da/sHfisPaEXmVqd6Hy42r8V0abXQk6nej2vp3gIcpTUH4M7f5mBbTinhwRYmnZboszL1T4pg6nmpPDYihbSYEIorXLyzNps7P97Jx5uOUl5Zv26oaqZpsv5wMVO+2sdv5+1myV534DmtYwTPn9uZ589L5bTkyBb7Ydc+0sYTo1KIsFnYlFXCi98eoLKRq237i9NlMruJW040hmG415uyWuCHg0V8XzV7ri3YdKSY++bvZsfRMuwhQTw1prNPAk+19pE2/u+8VHonuLeoeXLhPr7eleez5/eGWRtzKHG66BIbwplVXdGBTqGntUnrBmHhUFwIe3f6uzQBL6NdKN3aheJ0mXy546dfODnFFby9OguAXwxM8Hmfu2EYDE6J5I8XpPH7YR3pGBVMXlkl/1p5hDs/2cnn20++jYbLNFm2v4AHP9/Do1/uZdWhIiyG+6/7P12QxuRRnWotINlSpcWG8tiIFGxBBisOFPHaskyvr+bdnJbuLSCr2El0aBCjuvq2iyAlOoTLernH87z+w2FKKhoWqluiL7Y7eGzBXhyllaTFhPCHsan0be/7n4WokCCeHNOJYZ2jcLpwr5i9PqdFvHcdJU7mbHG38kzsX3s6f6BS6GlljKAg6NEP8O+u6y3JuKoxIp9ty/W0ELxetXZO93ahjPXjyqIWw2B4mp1Xx3flntOTiA+3klPs5LVlmdw9eyeLduXV2kDS6TJZuDOP33y6i+cWHWBLdinBFoNx3WL420Vd+d2wji12/MCJ9E4M576zOmIxYMHOPE9oDXSmaTKzai+3C7vH+mVRtyv7tiMpMpicYifvr2u9e3M5XSb//OEwr1ZtJXNmpyj+77xUv45lsgVZuO+sjp5xXG+vyeJvyw8HfGvl/zbmUFZp0q1dKIOTfdMd6w0KPa1QdReXQk/9nJVqJ9Jm4UiRk5UHi1i2v4Dv9hUSVLUmTyD8BRNkMTg3I4a/XdyV205NJDo0iMzCCl5eeoh75+5m2b4CSisqmbPlKL/8ZAevfHeIvXnlhAdbmNCnHW9cms6dQ5JaxbTuEzk9JYq7T3cvNvm/jUc9u5QHsnWHi9mZ695ywpfdK8cKsVq4/bT2gHtg6u7c0pM8ouVxukxe/PYAnx7TOvHA2R2bbXJCQ1iq1u2adFoiBvDZdgfPL95f7xmVvpZTXMG8rQ7AvTp9S+oa9/93W7zO6D3Q/cXW9ZiH9vu1LC1BiNXiWUF01qYc/rHCvSbPJb3iAq5FxBZk4aKecfzj4nSuHxBPRLCFPY4ynl20nzF/+YZ/rjjMkSInMaFB3DAwgTcuTecXAxOIaeap74HknPQYfjHQvbTAmz8eCfhxEjM3uoPZOenR2P048+zU5EjO7BSFy4S/LT9cqwWxJat0mbxctdlvsMXgoeHJXN0v8LpkxveI48HhyZ5u2ke/2IvjBIt9+ssH63OocJn0TghjYFLL6iJX6GmFjKQUGDAEKitxzXijRfQP+1t1F9b6IyXkFDtpHxnMNQG83lBYsIUr+8bzz0vSmdCnHSFBBuWVLtpHBnPn4Pb885J0rujTrlUsfNYYV/SO46Kq/cz+/N0hVh4MzAG6exxlrKwaa3Vxz+afpn4yt52WSKjVwubskhpj3FqySpfJn747xJK9BVgt8NDw5IAedHtmpyieHtOZqJAgth8t5YHP97A/v8zfxfI4UljBFzscQMtr5QGFnlbLcuUt7oUK16+EdT/4uzgBr0OUjUEdfpoueufg9i1i2nNkSBC/GJjAG5dl8O/rT+PvF6czrntsiyh7czIMg1tOTWR4mp1KE/5v8QG2ZJf4u1i1zKpa1K25tpxoqPjwYCb2d4f9aauOkF8aeK0MDeEyTV5dlsmi3fkEGfDAWcmc1gLGn/RMCGPqeakkRQZzuLCChz7bw6YjxSd/oA9MX5+N0+Ve1d4fg7+bqm3/ZmzFjPYdMc65CADX9DcxnRV+LlHgu7x3HBbD3c1wSsfA/8V4rOhQK3062FvEOhm+YjEMfn1GBwZ1iKCs0uTpr/ezPy9w/mJ2bznhXhDPF4sR1tf4HrF0iQ2hoNzFv1e1jMHgdXGZJn9bnslXO/OwGHDfWR05PYBbeH4u2W5j6vmpdGsXSkG5i8kL9rGkagFFfzlUUM5XO90tgNc1w+r0vqDQ04oZF14N9hg4chDzqzn+Lk7A658UwdtXdOOe0+u367oEvuAggwfPTnZ/cJRV8sRX+8gu9v8fAIVllby7NttnW040RJDF4M7B7p+BBTvz2BAgLQwNYZomr/9wmM+3uwPPb4d2ZGjnwF8t+OdiQq08c05nBidHUuEyeeHbg+6VnP000Pz9tdm4TDi1Y0SNvfJaEoWeVswIC8e4/AYAzDnTMfNz/VyiwBcZEtTi+qjlxMKCLTw+MoVku43sYidTvtpHQZnvF+7MLXEyb2suTyzYyw3/2+YZM3NpM24s2lg9E8I4L8O9ceTfl2eedE2oQGKaJm+uPMLcrQ4M4FdndPBs+NkShVotPDw8mct7xxFkwI8Hi7h37m5eWXqQrCLfBfi9eWUsqmqZbI49CH1FoaeVM84cDakZUFKMOfO//i6OiF/YQ61MGdWJdmFW9uWV88zX+ynzwXTgw4XuzTwf+nwPN3+0nb+vOMzqzGIqTUiNDmHSaYkMCdAxJjcMTMQeEsTevHI+2RT4U//BHXimrc5i9mb3H3h3n57E6K6Bvet3fQRZDG4clMhrF3VlWOcoTGDhrnx++clO3lp5xCch/v212ZjAGZ0iyWgXWLNaGyLg5rHOnz+f2bNn43A4SE1N5ZZbbiEjI6POc/ft28f06dPZtWsXWVlZ3HjjjVx44YU+LnFgMywWLNfejuv/HsBc8iXmyHEYqXXXp0hrlhgZzBOjO/HwF3vYnF3CC98c4JGRnbz6HKZpsi+/nO/3FvDdvgJ25tYcQ9S9XShnVu0e39Hu/4HLJxIVEsTNpyTyp+8O8f66bM5KtZMYGdi7gb+7NpuPqpYAuHNwe87NiPFvgbysQ5SNB85OZltOCf9ZlcW6w8XM2nSUL7Y7uKJPO8b3aJ5JDLtyS1mytwADuDaAZ7XWR0C19CxdupRp06YxYcIEpk6dSmpqKs8++yx5eXVPnSwrK6N9+/ZMnDiRmJgY3xa2BTHSe2KcPgJME9f7r2sKu7RZqTEhTK7aruKHg0W8+v2hJv88mKbJtpwSpq06wl2zd/GrObt4Z202O3PLsBjQr304t5/WnjcvS+fFsWlc3qddwAeeaqO62OmTGEZZpcnrPx72d3FOaPq6bGasd8+Gu+3URL8t9OgL3dqF8fSYTjwxyr0/X1GFi2mrs/jlJzv5cofDK6s5F1dUsmJ/IW/+eJjnFh0A4KzUqIBbu6yhAqqlZ86cOYwZM4ZRo0YBMGnSJFauXMnChQu59NJLa52fkZHhaQV69913fVnUFse4/EbMVd/D9k2Yyxe7Q5BIG9QrMZwHzkrmucX7+WpnHg9/sp6YYFed4efnh35+RkmFix8PFpJd/NPUbqvFYFCHcM7oFMWQ5EjsoQH1a7ZBDMM9qPneubtYvr+QZfsLOD0l8GZA/W9DDu+udW+fcfMpCVwUAGseNTfDMDilYyQDkiJYvDufd9ZkkVXs5C/fZ/LxpqP8YmACgxuwkXB5pYvNWSWsySxm3eEituWUcmx2irJZuLYFj+WpFjA/jU6nk507d9YINxaLhX79+rF161avPU9FRQUVFT8N/jIMg7CwMM/X3lR9vUAYGGu0S4ALrsQ167+Y//sPDDoDIyQwE3sg1VtLoTprmCGdovjVGR3403eHWLC16dOyQ60GpyVHcmYnO6d2jCC8FS0KmRobyqW92/G/DTm8/sNhBnaI9GzdEAjvt4835TDtmM2BL+sduN0vzfFzag0yGJ0ew1lpduZuyeWD9TnszSvn2UUH6J0Yxk2DEumZUHs9nUqXybacUtZmFrH2cBGbjpRQ8bMWoqTIYAYkRdA/KYIBHcKxh/g+Mni7zgIm9OTn5+NyuWp1U8XExHDw4EGvPc/MmTP58MMPPbe7dOnC1KlTSUhovgSblBQYU6BdN9xJ5ndfUXn4IBHffEb0L+70d5FOKFDqrSVRndXf9R060KVjAst358Ixv09//qv1579sj71lsRj07xjN6WmxhFhbT9D5uXvjE1m6bxmH8kuZs7OEX41wt7D7+/02Y+V+3vzxCAC3D+3CpGFd/Fqe+mquevtlSjLXD6vgP8v38v6P+9h4pIQHPtvDqG4J/PLsrjgrTVbsPcqKvbms2uegqLzmAOj4CBuDU2M5rXMsgzvH0iE6cKale6vOAib0+Mpll13G+PHjPberf6FlZWXhdHp39VHDMEhKSiIzMzNgxtGYl98Af/s/8v83jaKBZ2DEt/d3kWoJxHoLdKqzxkkPNxg2ulsT662Co1lHvFquQHTrKfE88/V+3lmxlyGJVk7vlebX99v8bbn8dVkm4N4l/sIuIRw6dMgvZakvX/2cXtEtnBEdu/Le2iwW7Mxj4bYsFm6r3aIZabPQr30E/ZPCGZAUQbLd9lPIL3ZwqNjRbGWsrxPVmdVqbXCDRcCEHrvdjsViweFw1DjucDi8Okg5ODiY4OC6ZyA015vQNM3A+SAadCb06Adb1uH64C0sdz7o7xIdV0DVWwuhOmsc1dvJDU6O5PSUSJbtL+Svyw8xpGeq3+rtyx0OT+C5tFcc11VtndFSvoe+qLd24VbuOaMDF/eM4+01WSzfX0hIkEGfxHD6VYWctJiQWqu4B2odeqvOAib0WK1Wunbtyvr16xkyZAgALpeL9evXM3bsWD+XrvUwDAPLNbfheuq3mD8uwdyyHqNHX38XS0RagEmntWf1oSI2Hinh70t2cWq8hY5RwT7brbykwsXi3fn8bbk78IzvEctNg1reppe+1DkmhEdHpJBX6iQ8OIjgoLZdVwETegDGjx/Pa6+9RteuXcnIyGDu3LmUlZUxcuRIAF599VXi4uKYOHEi4B78vH//fs/XR48eZffu3YSGhvq9rzmQGSldMEacj/n1PFzvv45l8ssYltY7HkFEvCMhIphr+sfzn1VZ/Ou73fwLiAi20K1dKN3jw+jeLozu8aFEe2HGWpnTxc7cUrbnlLLjaCnbj5ayP6/cM4NuXLcYbjs1UYGnnrzxPWkNAqoWhg4dSn5+PjNmzMDhcJCWlsYjjzzi6d7Kzs6u8QY/evQoDzzwgOf27NmzmT17Nr1792bKlCk+Ln3LYlx8HebyxbB/F+Y3X2CMUGuaiJzcJT3jAIO1WWVsOJRPUYWL1ZnFrM78aY+u9pHBdK8KQj3iw+gSG4It6PjLwpVXutiVW8b2HHe42XG0lH15ZdS13Ey7MCujukZz3YB4BR5pMMMM1A48H8vKyqoxld0bDMOgQ4cOHDrU9AXQmoNrwWzM91+HyCgsz/wDIyIwlsMP9HoLRKqzxlG9NU51ve0/cJDduaVsyS5ha04pW7NL2J9fXut8qwW6xIZ6glD7iGB2O8o8AWevo4zKOqo/NjSIjHahZMSFkdEulPS4UGLDAupv9QbR+63hTlRnwcHBLXcgs/ieMWIc5qL5cGgf5uz3MK6Z5O8iiUgLEmQx6BoXSte4UMZVHSsqr2RbTilbc0rYml3C1uxS8srcx7bllPLpVked14oOCfIEG3fQCaVdeGBveyEtj0JPG2ZYre5BzX98AnPhp5jDz8fo2NnfxRKRFizCFsTADhEM7BABuGfdHCmqYEt2dRAqJbuogs4xITUCTny4Vd1V0uwUeto4o/cgGDAE1izHNf1NLPdO0S8eEfEawzBoH2mjfaSN4Wl2fxdH2riA2nBU/MNy1S0QZIWNq2DtCn8XR0REpFko9AhGYkeMcy4GwDXjTUwvD+gWEREJBAo9AoAx/iqIjoUjhzC/mu3v4oiIiHidQo8AYISGY1x2AwDmnOmYebl+LpGIiIh3KfSIh3HmKEjrBqUlmDOn+bs4IiIiXqXQIx6GxYKlaq0ec8kCzK3r/VwiERER71HokRqM9J4YZ4wCwPWnJzFXL/NziURERLxDoUdqMa67A/oMgvIyXH99HtdXc/xdJBERkSZT6JFajNBwLPdMxjj7PDBdmO/9E9f0NzFdLn8XTUREpNEUeqROhtWK8Yu7MS77BQDmlx/j+sdUzLIyP5dMRESkcRR65LgMw8BywZUYt/0erFZY+R2ulx7FzHf4u2giIiINptAjJ2U5fQSW3z4F4ZGwayuu/3sAM3O/v4slIiLSIAo9Ui9G975YHn4B4ttDViau/3sQc+sGfxdLRESk3hR6pN6MpBQsD78IXbpDUQGuP07GteIbfxdLRESkXhR6pEEMewyW3z8Lg84ApxPzny/imvc/TNP0d9FEREROSKFHGswICcFy54MY51wCgPnRfzD/+1fMyko/l0xEROT4FHqkUQxLEJarb8W45nYwDMzFn+F69WnM0mJ/F01ERKROCj3SJJYx47Hc9TDYbLB+Ja4XHsbMzfF3sURERGpR6JEmMwaegeW+58EeA/t24Xr+fsz9u/xdLBERkRoUesQrjC7d3DO7OnSC3Gxcz9+P69MZmBUV/i6aiIgIoNAjXmTEt8fy4FToPQjKyzFn/RfXU7/G3LTG30UTERFR6BHvMiIisdw7xb11hT0GMg/genkyrn++iOnQWB8REfEfhR7xOsMw3FtXPP1XjNHjwbBgrvgG1+S7cH35iaa2i4iIXyj0SLMxwiOxXHs7lkdfcq/iXFqCOf0NXM/8DnPHZn8XT0RE2hirvwvQEjidToqLG7f+TElJCeXl5V4uUQsTm4B592Rw5MCRQ1BZCVlHoMIJiR0xrLXfht6uN9M0sVqtREREeO2aIiLSsij0nITT6aSoqIioqCgsloY3jAUHB1OhGUxu0dGYKZ0hNwcK893HCh0Q2w4i7RiG4Tm1OeqtqKiIsrIyQkJCvHpdERFpGdS9dRLFxcWNDjxSmxFkxYhvD0kpYAsBVyXkHIHM/ZhlZc363OHh4ZQ183OIiEjg0id5PSjweJ8RGuZe0ycuASwWKCuFQ/swj2a5g1BzPOcxLUkiItL2qHtL/MYwDLDHYIZHQm42FBVAvgOzqJDK+EQIDQcFFRER8RKFHvE7w2qFhCTMSDsczYKKcioPH4RgG8S0g/AItdKIiEiTqd9GTur000/n9ddf98q1li5dSnJyMnl5ebXuM8LC3V1esfEQFAQV5ZB1yD3ep0S7t4uISNOopaeVmjBhAr179+app55q8rXmzp1LeHi4F0p1cobFAtGxWOPaUZGTBfkO93ifwwcwQ8Mhth1GSKhPyiIiIq2LQk8bZZomlZWVWOtYI+fn2rVr54MS1WRYgjBi2mFGRUNeLhTkQWkxHCp2jwGKaYdhs/m8XCIi0nKpe6sVuvfee/nuu+948803SU5OJjk5menTp5OcnMxXX33F2LFj6dKlC8uXL2f37t3cfPPNDBgwgG7dunHBBRewePHiGtf7efdWcnIy7777Lrfeeivp6ekMGzaMzz//vNHl/fTTTxk1ahRdunTh9NNP5+9//7vnPiPIyn8++ZSzJt5I+qjzGDj+Mm7/7e/g4F7M7MPM/vhjxowZQ3p6On369OHqq69u9EKSIiLSuqmlp4FM04Ty+q/1YroqMb21yJ4tpF4Dep966il27txJz549ue+++wDYsmULAM899xyPP/44nTt3Jjo6moMHDzJ69GgefPBBbDYbH374ITfffDOLFy8mOTn5uM/x8ssv89hjj/HYY4/x1ltvcc8997Bs2TJiY2Mb9JLWrl3LnXfeye9+9zsuvvhifvjhBx555BESEhK44oorWLNmDY8//jh//vOfOe2008jNOsKyRYsAk8O7d3HPr37Fo7+7l7GXXk5RSQnLli1zf49ERER+RqGnocrLcN1zVb1P9+ZSeJZXZ0A9xrPY7XZsNhuhoaEkJiYCsH37dgDuv/9+hg8f7jk3NjaWPn36eG4/8MADzJ8/n88//5ybb775uM9x1VVXcemllwLw0EMP8eabb7J69WpGjRrVoNf0z3/+k7POOovf/va3AKSnp7Nt2zZee+01rrjiCg4cOEB4eDjnnHMOkZGRpKSk0G/QKZilJRzZtQdnZSXjzhhMiqUSOrSn5w2/wLAENagMIiLSNij0tDH9+/evcbuoqIiXXnqJBQsWcOTIEZxOJ6WlpRw4cOCE1+nVq5fn6/DwcKKiosjOzm5webZt28b5559f49jgwYN54403qKysZPjw4aSkpHDmmWcycuRIRo0axbhx4wgLC6P32SM5a+iZnHPDLYwYMpjhQwZz4ZjRxHRIPmZ9HxM8DT8mZnY2rjnvgWmC6XL/73K5v3aZ7sf16Icx+GyM4OAGvx4REQlcCj0NZQtxt7jUk1f3kLI1fc+on8/Ceuqpp/jmm2+YPHkyaWlphIaGcvvtt590s8/gnwUCwzBwuVxNLt/PRUZGMn/+fJYuXcrixYv5wx/+wEsvvcTcuXOJjo7m/RkfsGLFChZ9+QVv/W8mL/zjDWa//jc6d+xQ9wWPZmEu/PTET7p0Aeb//o0x8gKMEWMx7DHeflkiIuIHCj0NZBhGvbqYPOcHB/uluyU4OLheIeSHH37gyiuvZNy4cYC75Wf//v3NXTyPbt26sWLFihrHVqxYQXp6OkFB7nqzWq0MHz6c4cOH87vf/Y5evXqxZMkSLrjgAgzDYMiQIQwZMoTfPvAgpw8ZwvxlK7j9hhvAM/zJ+Ok/pwvjwqvcW18YFnfLTvXXFgOKizCXfgWOHMxP3sWc+wHG6cMxxlyM0amLz+pFRES8T6GnlerUqROrVq1i3759REREHDcAdenShXnz5nHuuediGAYvvvhis7TYHM8dd9zBBRdcwB//+EcuvvhifvzxR9566y1eeOEFAL744gv27t3L6aefTkxMDAsWLMDlcpGens7KlSv59ttvGTFiBPHx8axcuZKjubl0GzAIo11Cnc9nBNmwXHr9CctkXnQt5sqlmF9+Aru2Yi5ZgLlkAfToh+Wci6H/aRo3JCLSAin0tFJ33HEH9957LyNHjqS0tJSXX365zvOeeOIJfve733HJJZcQFxfH3XffTWFhoc/K2a9fP/7+97/zhz/8gT/96U8kJiZy//33c80111BRUUF0dDTz5s3j5ZdfprS0lC5duvDaa6/Ro0cPtm3bxrJly3jjjTcoLCwkOTmZxx9/nNGjRzepTIbVijFkOAwZjrljM+aC2Zg/LoEt63BtWQcJSRhjLsIYNgYj1DeLNoqISNMZpub3ApCVlVXn2Jv8/Hzsdnujr+vVMT1tSHPVW2O/n+bRLMyFczEXfwbFVaEwLBxj2LkYoy/ESEjycknrWS6XC4oKMArySEhMJMuq1aobwjAMOnTowKFDh7TUQQOo3hpH9dZwJ6qz4OBgEhLqbtU/HrX0iNSDEZeAccWNmOOvxvxuIeaCTyDzAOaXH2MumA0Dh7i7vrr1afLmqKarEgrz3Vtw5Dswq/6vfTsPChzu2WdAJkBGLywXXQO9BmqTVhGRn1HoEa968MEH+eijj+q87/LLL2fq1Kk+LpF3GSGhGCPHYQ4/HzauwvXlJ7BhFaz6Hteq7yEuoUED3WswTSgqgMIC9xT6hoiIcu9Rtn0Trj8+AV17uMNPn1MUfkREqqh7q4q6t7wjOzubgoKCOu+LiooiPj6+XtcJtO6tEzEP7sVcMAfz+6/gJFP9680wINIO9hiwx2BExXi+xh7jnkZf/S/KjiXYRqLNyqFpf8NcNN+9Qz1Al+5Yxl8N/U5T+KmDuhsaR/XWOKq3hvN295ZCTxWFnsDSkkJPNbOoAA7sOWYxxEYIj3AHmUg7RlD9Z4gd+4vB5TiK+dlHmIvm/RTCUjPc4WfAEK+GH9M0ISsTco5Aek8ML6wl5Uv6EGoc1VvjqN4aTmN6RAKUEREF3fv6uxgY0bEYV92KOfYKzM9nYi6cC3u243rtWejUBcv4a2Dg6RiWhu83bJomHNqHuXUDbNvg/t+R474zPNI9o234WIyk4+/bJiLiLwo9Iq2UYY/BmHAz5vmXY34xC/OrubBvF66/PQ8pae7wM+iME4Yf01UJ+3djbt2AuW0DbN3gHmR9rCArRES6B1l/8THmFx9DrwFYRoxztyxZm+/XjFlUCAd2Q3KqO3SKiJyAQo9IK2dERWNcfiPmeZe5Q8lXc2D/blx//z93WLjwaoxTh2JYLJhOJ+zZjlndirN9E5QU1bygzQZde2J064PRoy906Q5WK2xYhevrebDuB9i0BtemNRAdh3H2ee5/cfUbz3UipqsSdm/HXL8Sc+Mq2LnVPejbMKBzOkbvgRi9B0J6L+2dJiK1aExPFY3pCSwtcUyPPzVkrIBZVID55SfuqfYlxe6DHTpBTBzs2AzlZTUfEBoGGb0xuvfB6NYH0jIwrMcPFGbOEczFn2F+8zkU5LkPWizQfwiWEWOh98AGda2ZjhzMDatg/UrMTWvcM9yOFR0Lebk1j9ls7uUDqkNQclqdY5maMsbCdDqhMA/ssY3qKmzJNDalcVRvDaeBzM1EoSewKPQ0TGN+mZrFhZhfznavOVR8TGtORBR0643RvS9G9z6Q0qVBg6o913dWYK76HvPrebB1/U93JCS5N3Ideg5GVO3vhVlR4R4vtGEV5oaV7sHhxwqLgN4DMPqcgtFnEEZcgjsYbVwDm1a7g9HPQ5A9BqPXAPf6Rb0HYsS2A05cb6ZputdDyj6MmX3YPWC7+uvsw5Cb7V4jKTIKo+cAd5jrffwtUFoTfXg3juqt4RR6molCT02nn346t912G5MmTTrpucnJybz55puMHTvWa8+v0NMwTWqxKC7CXPa1+zrd+kDHzl5vuTAP7sVcNB/zu69+al2yBmOcNgxjxDiItGNuWOlu0dmyrmZrk2FAWjdPyKFL9xOGMNM04cAezE1rMDeudgeun7dedejkDj+9BpLQoxdZm9djZrnDjFkVbsg5Uvtx9ZGUjNFroLusPfq2yq1K9OHdOKq3htPsLRHxKiM8AmPUhc37HB07Y1x7O+blN2AuX+xeS2jPdszvv8b8/uvaD4iOc4eGPoPc4SSy/kHVMAxIScNISYNzL3G3HO3cjLmxqhVo93b3DLRD+zAXzObwiS8Gse0gvj1GfBLEt3d/neD+n0g77N6GuWE15qbV7jFGmQcwMw9gLvwUgoKga4+qbrZB7q7BNrpZrVlZ6V5As6wUKp0QHYsRbGv+5zVN94KfOYchJwuzMB+jfUfo1EWD39sghR4R8RkjJBTj7PPg7PMwd23DXDQXc/k37sHI3fq4u6v6DDruGJxGPWdwMPToh9GjH1z2C/eMr81rPSHIUlaMKy4Bo117SEiqCjhVoSYu4eQDojN6Y2T0hksmYhYXwuZ1mBtXuVuZsjJh20bMbRsxP37XvQ5TzwGesUZGQpJ7/7SyUveA8ZLiqv9LMGvcruN4aYk7VIWGgS0EIyQMQkKgjv+NWsdDITjYHT4qK8Hp/OnrY/93ur82f3afUekkPzSEyqwjUFoKZSVQVoZZVvJTsCktcbeUlZW4z3HW0XIbEeUeSxbbDiPa/T8x7TBi2lUdj4PI6BPPMDy2G/JoFmQfgaNHMLOPuFvr6mix87QXxMVDp64YnbpgpHSBTl3c3/82NkarLVHoaYX++9//8vLLL/PDDz9gOeaH9+abbyY2NpZf//rXPPnkk6xcuZLi4mK6devGQw89xPDhw73y/Js2beLxxx9n5cqVhIaGcuGFF/LEE08QEREBwNKlS3n22WfZsmULwcHBdO/enddee42UlBQ2bNjAE088wdq1azEMgy5dujB16lQGDBjglbJJ4DC6dMPo8hvM634JpumzhQ2NiEg4dah7xpqXuxuM8Eg45UyMU84EwMzKdIerjatg81r32KmVSzFXLnV/8IaEQXmpewuSJjrRFbzdkWICeY19sMXi/ud0ugelVy3qeWwZa5Q3yOoerB4T5w5Ese2gvAwz5wjkZLlDTV2B6uei4yA+EcIj4dA+dxfm0Ww4mo25ZvlPzxka5m4p7NTFHYhSukBy5xa38CZUBcJ9OzGXfoW5dgVYg90tbNGx7vqIiau6HVdVx7FN6o41KyrcGzIXF1Z9b4vci7YWF0JUNJbTR3jx1TWOQk8DmaZJWWX9f4VU4qLC2cB9lI4jJMio11+/48ePZ/LkySxZsoSzzz4bgNzcXL7++mumTZtGUVERo0eP5sEHH8Rms/Hhhx9y8803s3jxYpKTm7aoXHFxMddddx2nnnoqn376KdnZ2dx///08+uijvPLKKzidTm699VYmTpzIa6+9RkVFBatWrfK8rl/96lf06dOHP/zhD7hcLjZs2IC1Gdd5Ef/zRReHvxhVg7YZMdbdvbNn+0+tQDu3uFtBqgVZISz8mH8REBbu/hCqvh1e9XVoOEZYOFQ6McvKPC0tnlaV8lIoLcUsL/2p5aWs9KdzysrcW5VYre7nDQr66f+6jgXVPGZYgwiLjqXEZbpbjX72z6hqfSK0qlXp2Fam6p/n4kJwHIXcHExHjnuRS8dRzKpjOHLcs/8qnXA0y/2P4wQ4w+JuFWqXiNEuEar+eb6Oi6/1PjOLi9xrUO3bBft3uf8/sMfdQrV9E+b2TT89n2Fxj9Xq1MXdGuW5SI0rHvOlWfc5QUEUn3o6ZlKq+/vYTMz8XMzvF2EuXVB7IsChfbXqsMbtkFB3AKoRhuLcx4uL3GGmuNDdYlpcCEVV/4oLTzwGLr0nKPS0PGWVJldP3+qX555+dXdCrScPPTExMYwaNYpZs2Z5Qs+nn35KXFwcw4YNw2Kx0KdPH8/5DzzwAPPnz+fzzz/n5ptvblIZZ86cSVlZGX/6058ID3f/UD/zzDPcdNNNPProo1itVvLz8znnnHNIS0sDoFu3bp7HHzhwgDvvvJNu3bpRUVFB165dm1QekUBhVI/v6doDxl+DWVoMjtyfQk6wrVFdev7YUc0wDNo1tYUsIsr9Lzn1uK/BdDohP7cqBFUFIkeOu2uuXXv3TLl2iRAb3+BFMI3wCOjexz1Dsfr5KivdY7L27fwpCO3b5Q5fVePAmirns4/c4TG9J0bfUzH6nupuWWpid65ZUQFrl+Na+hWs/9E9sxDcEwYGno5xxkiwhWDm5ULeUfd7L+9o1e2qY6VV3ZNHDsGRQ41rITQMd0CPiHS3qkVEusdOdejUpNfnLQo9rdRll13GAw88wHPPPUdISAgzZ87k4osvxmKxUFRUxEsvvcSCBQs4cuQITqeT0tJSDhw40OTn3bZtG7169fIEHoDBgwfjcrnYsWMHZ5xxBldddRXXXXcdZ599NmeffTYXXXQR7du3B+D222/n/vvv56OPPuKss85i/PjxnnAk0poYoeGQ1PpmdnmTYbVCXIL7H80f8IygIHdXVnJnYCRQ1UWUlwv7drnDUFHhzx5U6yo/u3nM7dJigrZtwHlgL2x1LwBqfjQNYuLcsxP7nQa9BrgDWT2YpukeSL/0K8zli92tLdW69sA4czTG4LPdXbrHK+6x1yst8QSgmuEoF7O81N19WxVk3P9HYUREeL4mPNLdOhnAY6IUehooJMhg+tXd631+sDWYivr0N9fzuevr3HPPxTRNFixYwIABA1i2bBlTpkwB4KmnnuKbb75h8uTJpKWlERoayu233065t3YIP4k//vGP3HrrrSxcuJBPPvmEF154gffee49TTz2V3//+91x66aV8/fXXfPnll7z00kv89a9/Zdy4cT4pm4jIsQzDqBpPFIfR79QmX6tDhw4cXLsK17ofMNf9CFvWuluxlnyJueTLerUCmbk57pmP333lHp9ULaYdxpmjMIaOxkhKaXj5QsPcXZLtO/qlBdEXFHoayDCMenUxVQsOthCE71NvaGgo48aNY+bMmezevZv09HT69esHwA8//MCVV17pCRJFRUXs37/fK8/brVs3PvjgA4qLiz2tPStWrMBisZCenu45r2/fvvTt25df/epXXHTRRcyaNYtTT3X/QklPT6dnz57ceuut3HXXXUyfPl2hR0RaDSMhCcuoC2HUhZgV5e5Wn/U/Yq7/ETIP1GwFio7D6DsIo++pmJWV7qCzcY17xiOAzYYx8EyMYaOhZ/82uyRCfSn0tGKXXXYZN910E1u2bOHyyy/3HO/SpQvz5s3j3HPPxTAMXnzxRVwu7wy2vvzyy3nppZf4zW9+w+9//3tycnKYPHkyV1xxBQkJCezdu5d33nmHc889l6SkJHbs2MGuXbuYMGECJSUlPPPMM1x44YV07dqVffv2sWbNGi644AKvlE1EJNAYwTb3elR9BsHVt7ln/K1f6Q5Am9e6u5qWLMBcsqDmAzN6u1t0TjvLPahd6kWhpxU766yziImJYceOHVx22WWe40888QS/+93vuOSSS4iLi+Puu++msLDwBFeqv7CwMN555x0ef/xxLrzwwhpT1qvv3759Ox988AG5ubkkJiZy00038Ytf/AKn00lubi6/+c1vyM7OJi4ujnHjxvH73//eK2UTEQl0RkISxqgLYNQF7lagbRsw1610b8nicmEMPsvdhZXY0d9FbZG0DUUVbUMRWLQNRcNoefvGUb01juqtcVRvDeftbSgCd4i1iIiIiBepe0tO6KOPPuLBBx+s876UlBQWLlzo4xKJiIg0jkKPnNB5553HoEGD6rwv+GR7EomIiAQQhR45ocjISCIjI09+ooiISIDTmB4RERFpEwKypWf+/PnMnj0bh8NBamoqt9xyCxkZGcc9/7vvvmP69OlkZWWRlJTEddddxymnnOLDEouIiEigC7iWnqVLlzJt2jQmTJjA1KlTSU1N5dlnnyUvL6/O87ds2cKf/vQnRo8ezdSpUxk8eDAvvvgie/fu9VqZvLVwn/iXpoiKiLRtARd65syZw5gxYxg1ahQpKSlMmjQJm8123FlCc+fOZeDAgVx88cWkpKRwzTXX0LVrV+bPn++V8oSHh1NQUKDg0woUFxcTEhLi72KIiIifBFT3ltPpZOfOnVx66aWeYxaLhX79+rF169Y6H7N161bGjx9f49iAAQNYsWJFnedXVFTUWPTOMAzCwsI8X/9ccHAwERERjV6x2Gaz+Wwjz9bE2/VmmiZWq5XQ0FCvXTOQVL9363oPy/Gp3hpH9dY4qreG83adBVToyc/Px+VyERMTU+N4TEwMBw8erPMxDoeD6OjoGseio6NxOBx1nj9z5kw+/PBDz+0uXbowderUBq/qKBKIkpKS/F2EFkn11jiqt8ZRvTWct+osoEKPL1x22WU1Woaq02NWVhZOp9Orz2UYBklJSWRmZmo8SQOo3hpOddY4qrfGUb01juqt4U5UZ1artcENFgEVeux2OxaLpVYrjcPhqNX6Uy0mJqbWIOe8vLzjnh8cHHzcRfWa601omqbe4I2gems41VnjqN4aR/XWOKq3hvNWnQXUQGar1UrXrl1Zv36955jL5WL9+vV07969zsd0796ddevW1Ti2du1aunXr1qxlFRERkZYloEIPwPjx41mwYAFff/01+/fv54033qCsrIyRI0cC8Oqrr/Luu+96zr/gggtYs2YNs2fP5sCBA8yYMYMdO3YwduxYP70CERERCUQB1b0FMHToUPLz85kxYwYOh4O0tDQeeeQRT3dVdnZ2jVHcPXr04Ne//jXvv/8+7733Hh06dOD++++nc+fODXpeq7X5qqI5r92aqd4aTnXWOKq3xlG9NY7qreHqqrPG1KNhqmNRRERE2oCA695qTUpKSnjwwQcpKSnxd1FaFNVbw6nOGkf11jiqt8ZRvTWct+tMoacZmabJrl27NEq/gVRvDac6axzVW+Oo3hpH9dZw3q4zhR4RERFpExR6REREpE1Q6GlGwcHBTJgw4biLIUrdVG8NpzprHNVb46jeGkf11nDerjPN3hIREZE2QS09IiIi0iYo9IiIiEiboNAjIiIibYJCj4iIiLQJ2gCkmcyfP5/Zs2fjcDhITU3llltuISMjw9/FClgzZszgww8/rHGsY8eOvPLKK/4pUIDauHEjn3zyCbt27SI3N5f77ruPIUOGeO43TZMZM2awYMECioqK6NmzJ7fddhsdOnTwY6n972T19tprr7Fo0aIajxkwYACPPvqor4saMGbOnMny5cs5cOAANpuN7t27c/3119OxY0fPOeXl5UybNo2lS5dSUVHBgAEDuO222zx7JbZF9am3KVOmsHHjxhqPO+ecc7j99tt9XdyA8fnnn/P555+TlZUFQEpKChMmTGDQoEGA995rCj3NYOnSpUybNo1JkybRrVs3Pv30U5599lleeeUVoqOj/V28gNWpUycmT57suW2xqCHy58rKykhLS2P06NH84Q9/qHX/xx9/zLx587j77rtJTExk+vTpPPvss7z88svYbDY/lDgwnKzeAAYOHMhdd93lud3WN4XcuHEj559/Punp6VRWVvLee+/xzDPP8PLLLxMaGgrAf/7zH1auXMnvfvc7wsPDefPNN3nppZd4+umn/Vx6/6lPvQGMGTOGq6++2nO7Lf98AsTFxTFx4kQ6dOiAaZosWrSIF154gRdeeIFOnTp57b2mT5VmMGfOHMaMGcOoUaNISUlh0qRJ2Gw2Fi5c6O+iBTSLxUJMTIznn91u93eRAs6gQYO45pprarRSVDNNk7lz53L55ZczePBgUlNTueeee8jNzWXFihV+KG3gOFG9VbNarTXef5GRkT4sYeB59NFHGTlyJJ06dSItLY27776b7Oxsdu7cCUBxcTFfffUVN954I3379qVr167cddddbNmyha1bt/q59P5zsnqrFhISUuP9Fh4e7qcSB4bTTjuNU045hQ4dOtCxY0euvfZaQkND2bZtm1ffa237T5lm4HQ62blzJ5deeqnnmMVioV+/fm36F0F9ZGZmcscddxAcHEz37t2ZOHEi8fHx/i5Wi3HkyBEcDgf9+/f3HAsPDycjI4OtW7cybNgwP5Yu8G3cuJHbbruNiIgI+vbtyzXXXENUVJS/ixUwiouLATxhcOfOnVRWVtKvXz/POcnJycTHx7N161a6d+/ul3IGmp/XW7VvvvmGb775hpiYGE499VSuuOIKQkJC/FHEgONyufjuu+8oKyuje/fuXn2vKfR4WX5+Pi6Xq1Y/Y0xMDAcPHvRPoVqAbt26cdddd9GxY0dyc3P58MMPefzxx3nppZcICwvzd/FaBIfDAVCrCzU6Otpzn9Rt4MCBnH766SQmJpKZmcl7773Hc889x7PPPqtuVtwfQv/+97/p0aMHnTt3BtzvN6vVSkRERI1z9X77SV31BnDWWWcRHx9PXFwce/bs4Z133uHgwYPcd999fiyt/+3du5dHH32UiooKQkNDue+++0hJSWH37t1ee68p9EhAqB6sBpCamuoJQd999x2jR4/2Y8mkLTi2Faxz586kpqbyq1/9ig0bNtT467KtevPNN9m3bx9PPfWUv4vSohyv3s455xzP1507dyY2NpannnqKzMxMkpKSfF3MgNGxY0defPFFiouL+f7773nttdd48sknvfoc+hPGy+x2OxaLpVb6dDgcbXpGQ0NFRETQsWNHMjMz/V2UFqP6/ZWXl1fjeF5ent57DdS+fXuioqL0/sP9wb1y5UqeeOIJ2rVr5zkeExOD0+mkqKioxvl6v7kdr97qUj2zt62/36xWK0lJSXTt2pWJEyeSlpbG3LlzvfpeU+jxMqvVSteuXVm/fr3nmMvlYv369erjboDS0lIyMzP1y7MBEhMTiYmJYd26dZ5jxcXFbN++Xe+9BsrJyaGwsJDY2Fh/F8VvTNPkzTffZPny5Tz++OMkJibWuL9r164EBQXVeL8dPHiQ7OzsNv1+O1m91WX37t0Abfr9VheXy0VFRYVX32vq3moG48eP57XXXqNr165kZGQwd+5cysrKGDlypL+LFrCmTZvGaaedRnx8PLm5ucyYMQOLxcJZZ53l76IFlOowWO3IkSPs3r2byMhI4uPjueCCC/joo4/o0KEDiYmJvP/++8TGxjJ48GA/ltr/TlRvkZGRfPDBB5x++unExMRw+PBh/vvf/5KUlMSAAQP8WGr/evPNN/n222954IEHCAsL87Reh4eHY7PZCA8PZ/To0UybNo3IyEjCw8P517/+Rffu3dt06DlZvWVmZvLtt99yyimnEBkZyd69e/nPf/5Dr169SE1N9W/h/ejdd99l4MCBxMfHU1payrfffsvGjRt59NFHvfpe0y7rzWT+/Pl88sknOBwO0tLSuPnmm+nWrZu/ixWwXnnlFTZt2kRBQQF2u52ePXtyzTXXtOn+7bps2LChzj7uESNGcPfdd3sWJ/zyyy8pLi6mZ8+e3HrrrTUWRmuLTlRvkyZN4sUXX2TXrl0UFRURFxdH//79ufrqq9t0S+NVV11V5/G77rrL8wdc9YJxS5Yswel0anFCTl5v2dnZ/OUvf2Hfvn2UlZXRrl07hgwZwuWXX96mp63/7W9/Y/369eTm5hIeHk5qaiqXXHKJZzaqt95rCj0iIiLSJmhMj4iIiLQJCj0iIiLSJij0iIiISJug0CMiIiJtgkKPiIiItAkKPSIiItImKPSIiIhIm6DQIyJyAl9//TVXXXUVO3bs8HdRRKSJtA2FiPjd119/zV//+tfj3v/MM8+06a0NRMQ7FHpEJGBcddVVdW7QqO1IRMQbFHpEJGAMGjSI9PR0fxdDRFophR4RaRGOHDnCPffcw/XXX4/FYmHu3Lnk5eWRkZHBrbfeSufOnWucv379embMmMGuXbsICgqid+/eTJw4kZSUlBrnHT16lOnTp7N69WoKCgqIjY1l4MCB3HzzzVitP/2KrKio4D//+Q+LFy+mvLyc/v37c8cdd2C3233y+kWk6TSQWUQCRnFxMfn5+TX+FRQU1Dhn8eLFzJs3j/PPP5/LLruMffv28dRTT+FwODznrF27lmeffZa8vDyuvPJKxo8fz5YtW5g8eTJHjhzxnHf06FEefvhhli5dyplnnsnNN9/M8OHD2bhxI2VlZTWe96233mLPnj1ceeWVnHvuufz444+8+eabzVofIuJdaukRkYDx9NNP1zoWHBzMO++847mdmZnJn//8Z+Li4gAYOHAgjzzyCB9//DE33ngjAP/973+JjIzk2WefJTIyEoDBgwfzwAMPMGPGDO655x4A3n33XRwOB88991yNbrWrr74a0zRrlCMyMpLHHnsMwzAAME2TefPmUVxcTHh4uBdrQUSai0KPiASMW2+9lQ4dOtQ4ZrHUbJAePHiwJ/AAZGRk0K1bN1atWsWNN95Ibm4uu3fv5uKLL/YEHoDU1FT69+/PqlWrAHC5XKxYsYJTTz21znFE1eGm2jnnnFPjWK9evfj000/JysoiNTW18S9aRHxGoUdEAkZGRsZJBzL/PBRVH/vuu+8AyMrKAqBjx461zktOTmbNmjWUlpZSWlpKSUlJrbFAxxMfH1/jdkREBABFRUX1eryI+J/G9IiI1MPPW5yq/bwbTEQCl1p6RKRFOXToUJ3HEhISADz/Hzx4sNZ5Bw8eJCoqitDQUGw2G2FhYezdu7d5CywiAUMtPSLSoqxYsYKjR496bm/fvp1t27YxcOBAAGJjY0lLS2PRokU1up727t3LmjVrGDRoEOBuuRk8eDA//vhjnVtMqAVHpPVRS4+IBIxVq1Zx4MCBWsd79OjhGUSclJTE5MmTOe+886ioqGDu3LlERUVxySWXeM6//vrref7553nssccYNWoU5eXlzJ8/n/DwcK666irPeRMnTmTt2rVMmTKFMWPGkJKSQm5uLt9//z1PPfWUZ9yOiLQOCj0iEjBmzJhR5/G77rqL3r17AzB8+HAsFguffvop+fn5ZGRkcMsttxAbG+s5v3///jzyyCPMmDGDGTNmeBYnvO6662pscxEXF8dzzz3H+++/z7fffktJSQlxcXEMHDiQkJCQ5n2xIuJzhqk2XBFpAY5dkfniiy/2d3FEpAXSmB4RERFpExR6REREpE1Q6BEREZE2QWN6REREpE1QS4+IiIi0CQo9IiIi0iYo9IiIiEiboNAjIiIibYJCj4iIiLQJCj0iIiLSJij0iIiISJug0CMiIiJtgkKPiIiItAn/D6gutzQjE1vUAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Lakukan penyimpanan model\n",
    "print(\"[INFO] saving brain tumor malignancy detector model...\")\n",
    "model.save(model_path, save_format=\"h5\")  # Save dalam format .h5\n",
    "\n",
    "# Munculkan hasil klasifikasi dan kompilasi dari pelatihan data\n",
    "epochs_range = range(EPOCHS)\n",
    "\n",
    "plt.style.use(\"ggplot\")\n",
    "plt.figure()\n",
    "plt.plot(epochs_range, H.history[\"accuracy\"], label=\"train_acc\")\n",
    "plt.plot(epochs_range, H.history[\"val_accuracy\"], label=\"val_acc\")\n",
    "plt.legend(loc='lower left')\n",
    "plt.title('Training and Validation Accuracy')\n",
    "plt.xlabel(\"Epoch\")\n",
    "plt.ylabel(\"Accuracy\")\n",
    "plt.show()\n",
    "\n",
    "plt.style.use(\"ggplot\")\n",
    "plt.figure()\n",
    "plt.plot(epochs_range, H.history[\"loss\"], label=\"train_loss\")\n",
    "plt.plot(epochs_range, H.history[\"val_loss\"], label=\"val_loss\")\n",
    "plt.legend(loc=\"lower left\")\n",
    "plt.title(\"Training and Validation Loss\")\n",
    "plt.xlabel(\"Epoch\")\n",
    "plt.ylabel(\"Loss\")\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}