{ "cells": [ { "cell_type": "code", "execution_count": 32, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "100%|██████████| 50/50 [00:00<00:00, 929.73it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 988.24it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 993.78it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 790.39it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 953.19it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 992.44it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 996.24it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 999.91it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1476.15it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1005.97it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1011.11it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1471.92it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1001.95it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 997.82it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1014.84it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1070.94it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1403.33it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1684.13it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1378.91it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1039.67it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1429.53it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1010.23it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1036.58it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1453.24it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1017.11it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1500.61it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1495.53it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1674.80it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1006.44it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1399.07it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1490.16it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1561.58it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1613.39it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1410.76it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1486.50it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1323.15it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1826.63it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 1496.93it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 2904.28it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 21874.96it/s]\n", "100%|██████████| 50/50 [00:00<00:00, 4161.43it/s]\n", "100%|██████████| 50/50 [00:00 128x128)\n", " data_resized = cv2.resize(data, (128, 128)).astype(np.uint8) # cv2로 128x128로 리사이즈하고 데이터 타입을 uint8로 변경\n", "\n", " # 각 레이블에 따라 리사이즈된 이미지 추가\n", " if np_path.split(\"_\")[0] in [\"낙상\", \"도움요청\", \"폭력범죄\", \"강도범죄\"]:\n", " label1_batch.append(data_resized)\n", " elif np_path.split(\"_\")[0] == \"실내\":\n", " label0_batch.append(data_resized)\n", "\n", " return label1_batch, label0_batch\n", "\n", "# 전체 파일을 배치 단위로 나누어 처리\n", "for i in range(0, len(np_paths), batch_size):\n", " np_paths_batch = np_paths[i:i + batch_size]\n", " \n", " # 배치 처리\n", " label1_batch, label0_batch = process_batch(np_paths_batch)\n", " \n", " # 결과를 전체 리스트에 추가\n", " label1.extend(label1_batch)\n", " label0.extend(label0_batch)\n", "\n", "# 레이블 배열 결합\n", "label1_resized = np.array(label1, dtype=np.uint8) # 데이터 타입을 uint8로 지정하여 메모리 사용량 감소\n", "label0_resized = np.array(label0, dtype=np.uint8)\n", "\n", "# 배열 및 레이블 결합\n", "X = np.concatenate([label1_resized, label0_resized], axis=0)\n", "y = np.concatenate([np.ones(len(label1_resized)), np.zeros(len(label0_resized))], axis=0)\n", "\n", "# train-test split 수행\n", "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n", "\n", "# 결과 확인\n", "print(\"X_train shape:\", X_train.shape) # 예상: (n_train, 128, 128, 3)\n", "print(\"X_test shape:\", X_test.shape) # 예상: (n_test, 128, 128, 3)\n", "print(\"y_train shape:\", y_train.shape)\n", "print(\"y_test shape:\", y_test.shape)" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "100%|██████████| 128554/128554 [18:10<00:00, 117.87it/s]\n" ] }, { "ename": "MemoryError", "evalue": "Unable to allocate 23.0 GiB for an array with shape (125848, 128, 128, 3) and data type float32", "output_type": "error", "traceback": [ "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[1;31mMemoryError\u001b[0m Traceback (most recent call last)", "Cell \u001b[1;32mIn[70], line 32\u001b[0m\n\u001b[0;32m 29\u001b[0m \u001b[38;5;66;03m# label0=label1\u001b[39;00m\n\u001b[0;32m 30\u001b[0m \u001b[38;5;66;03m# 레이블 배열 결합\u001b[39;00m\n\u001b[0;32m 31\u001b[0m label1_resized \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39marray(label1)\n\u001b[1;32m---> 32\u001b[0m label0_resized \u001b[38;5;241m=\u001b[39m \u001b[43mnp\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43marray\u001b[49m\u001b[43m(\u001b[49m\u001b[43mlabel0\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m 34\u001b[0m \u001b[38;5;66;03m# 배열 및 레이블 결합\u001b[39;00m\n\u001b[0;32m 35\u001b[0m X \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39mconcatenate([label1_resized, label0_resized], axis\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m0\u001b[39m)\n", "\u001b[1;31mMemoryError\u001b[0m: Unable to allocate 23.0 GiB for an array with shape (125848, 128, 128, 3) and data type float32" ] } ], "source": [ "\n", "'''import os\n", "import numpy as np\n", "from tqdm import tqdm\n", "import cv2\n", "from sklearn.model_selection import train_test_split\n", "\n", "# base_path 경로 설정\n", "base_path = r\"C:\\Users\\KOSA\\Desktop\\AUDIO_FEATURES\"\n", "\n", "# 파일 목록 가져오기\n", "np_paths = os.listdir(base_path)\n", "\n", "label1 = []\n", "label0 = []\n", "\n", "# 파일을 순차적으로 탐색하면서 \"낙상\"으로 시작하는 파일 로드\n", "for np_path in tqdm(np_paths[:]):\n", " full_path = os.path.join(base_path, np_path)\n", " data = np.load(full_path) # data는 (128, 157, 3) 크기의 배열\n", "\n", " # 리사이즈 과정 (128x157 -> 128x128)\n", " data_resized = cv2.resize(data, (128, 128)) # cv2로 128x128로 리사이즈\n", "\n", " # 각 레이블에 따라 리사이즈된 이미지 추가\n", " if np_path.split(\"_\")[0] in [\"낙상\", \"도움요청\", \"폭력범죄\", \"강도범죄\"]:\n", " label1.append(data_resized)\n", " elif np_path.split(\"_\")[0] == \"실내\":\n", " label0.append(data_resized)\n", "# label0=label1\n", "# 레이블 배열 결합\n", "label1_resized = np.array(label1)\n", "label0_resized = np.array(label0)\n", "\n", "# 배열 및 레이블 결합\n", "X = np.concatenate([label1_resized, label0_resized], axis=0)\n", "y = np.concatenate([np.ones(len(label1_resized)), np.zeros(len(label0_resized))], axis=0)\n", "\n", "# train-test split 수행\n", "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n", "\n", "# 결과 확인\n", "print(\"X_train shape:\", X_train.shape) # 예상: (n_train, 128, 128, 3)\n", "print(\"X_test shape:\", X_test.shape) # 예상: (n_test, 128, 128, 3)\n", "print(\"y_train shape:\", y_train.shape)\n", "print(\"y_test shape:\", y_test.shape)\n", "'''" ] }, { "cell_type": "code", "execution_count": 33, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1/10\n", "\u001b[1m1047/1047\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m58s\u001b[0m 55ms/step - accuracy: 0.9249 - loss: 17.5027 - val_accuracy: 0.9830 - val_loss: 0.0661\n", "Epoch 2/10\n", "\u001b[1m1047/1047\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m59s\u001b[0m 56ms/step - accuracy: 0.9825 - loss: 0.0669 - val_accuracy: 0.9907 - val_loss: 0.0359\n", "Epoch 3/10\n", "\u001b[1m1047/1047\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m59s\u001b[0m 56ms/step - accuracy: 0.9855 - loss: 0.0526 - val_accuracy: 0.9894 - val_loss: 0.0378\n", "Epoch 4/10\n", "\u001b[1m1047/1047\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m81s\u001b[0m 78ms/step - accuracy: 0.9904 - loss: 0.0343 - val_accuracy: 0.9973 - val_loss: 0.0105\n", "Epoch 5/10\n", "\u001b[1m1047/1047\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m106s\u001b[0m 101ms/step - accuracy: 0.9915 - loss: 0.0296 - val_accuracy: 0.9952 - val_loss: 0.0190\n", "Epoch 6/10\n", "\u001b[1m1047/1047\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m78s\u001b[0m 75ms/step - accuracy: 0.9907 - loss: 0.0353 - val_accuracy: 0.9990 - val_loss: 0.0049\n", "Epoch 7/10\n", "\u001b[1m1047/1047\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m58s\u001b[0m 55ms/step - accuracy: 0.9924 - loss: 0.0270 - val_accuracy: 0.9987 - val_loss: 0.0064\n", "Epoch 8/10\n", "\u001b[1m1047/1047\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m58s\u001b[0m 56ms/step - accuracy: 0.9934 - loss: 0.0262 - val_accuracy: 0.9985 - val_loss: 0.0052\n", "Epoch 9/10\n", "\u001b[1m1047/1047\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m58s\u001b[0m 56ms/step - accuracy: 0.9940 - loss: 0.0279 - val_accuracy: 0.9988 - val_loss: 0.0034\n", "Epoch 10/10\n", "\u001b[1m1047/1047\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m58s\u001b[0m 55ms/step - accuracy: 0.9937 - loss: 0.0225 - val_accuracy: 0.9984 - val_loss: 0.0052\n", "\u001b[1m210/210\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 17ms/step - accuracy: 0.9984 - loss: 0.0047\n", "Test accuracy: 0.9983579516410828\n" ] } ], "source": [ "from tensorflow.keras.models import Sequential\n", "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, Dense, Dropout\n", "from tensorflow.keras.optimizers import Adam\n", "from tensorflow.keras.utils import to_categorical\n", "\n", "# X_train, Y_train 준비 (label1은 1, label0은 0으로 설정)\n", "X_train = np.concatenate([label1, label0], axis=0)\n", "y_train = np.concatenate([np.ones(len(label1)), np.zeros(len(label0))], axis=0)\n", "\n", "# 레이블을 원-핫 인코딩\n", "y_train = to_categorical(y_train, num_classes=2)\n", "y_test = to_categorical(y_test, num_classes=2) # y_test도 원-핫 인코딩\n", "# CNN 모델 정의\n", "model = Sequential()\n", "model.add(Conv2D(32, (3, 3), activation='relu', input_shape=(128, 128, 3)))\n", "model.add(MaxPooling2D(pool_size=(2, 2)))\n", "model.add(Conv2D(64, (3, 3), activation='relu'))\n", "model.add(MaxPooling2D(pool_size=(2, 2)))\n", "model.add(Flatten())\n", "model.add(Dense(128, activation='relu'))\n", "model.add(Dropout(0.5))\n", "model.add(Dense(2, activation='softmax')) # 2개의 클래스로 분류\n", "\n", "# 모델 컴파일\n", "model.compile(optimizer=Adam(), loss='categorical_crossentropy', metrics=['accuracy'])\n", "\n", "# 모델 학습\n", "history = model.fit(X_train, y_train, epochs=10, batch_size=32, validation_data=(X_test, y_test))\n", "\n", "# 모델 평가\n", "test_loss, test_acc = model.evaluate(X_test, y_test)\n", "print(f\"Test accuracy: {test_acc}\")" ] }, { "cell_type": "code", "execution_count": 34, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "WARNING:absl:You are saving your model as an HDF5 file via `model.save()` or `keras.saving.save_model(model)`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')` or `keras.saving.save_model(model, 'my_model.keras')`. \n" ] } ], "source": [ "model.save(\"jhyaudio2.h5\")" ] }, { "cell_type": "code", "execution_count": 38, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "#jhyaudio2\n", "import matplotlib.pyplot as plt\n", "\n", "# 학습 정확도와 검증 정확도 시각화\n", "plt.figure(figsize=(12, 4))\n", "\n", "# 정확도 그래프\n", "plt.subplot(1, 2, 1)\n", "plt.plot(history.history['accuracy'], label='Train Accuracy')\n", "plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n", "plt.title('Model Accuracy')\n", "plt.xlabel('Epoch')\n", "plt.ylabel('Accuracy')\n", "plt.legend()\n", "\n", "# 손실 그래프\n", "plt.subplot(1, 2, 2)\n", "plt.plot(history.history['loss'], label='Train Loss')\n", "plt.plot(history.history['val_loss'], label='Validation Loss')\n", "plt.title('Model Loss')\n", "plt.xlabel('Epoch')\n", "plt.ylabel('Loss')\n", "plt.legend()\n", "\n", "plt.tight_layout()\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 36, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\u001b[1m210/210\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 16ms/step\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "#jhyaudio2\n", "from sklearn.metrics import confusion_matrix\n", "import seaborn as sns\n", "import matplotlib.pyplot as plt\n", "\n", "# 모델 예측 수행\n", "y_pred = model.predict(X_test)\n", "y_pred_classes = np.argmax(y_pred, axis=1) # 예측된 클래스 (0 또는 1)\n", "y_true = np.argmax(y_test, axis=1) # 실제 클래스 (0 또는 1)\n", "\n", "# 혼동 행렬 계산\n", "conf_matrix = confusion_matrix(y_true, y_pred_classes)\n", "\n", "# 혼동 행렬 시각화\n", "plt.figure(figsize=(8, 6))\n", "sns.heatmap(conf_matrix, annot=True, fmt='d', cmap='Blues', cbar=False)\n", "plt.xlabel('Predicted Label')\n", "plt.ylabel('True Label')\n", "plt.title('Confusion Matrix')\n", "plt.show()\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAABKUAAAGGCAYAAACqvTJ0AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAA9hAAAPYQGoP6dpAADJDElEQVR4nOzdd3gU1dfA8e9uek8gIR2S0EJJoQVBQdBgKCK9qnR4VVARRUUREAuKiAgiKNJBUAT5oXRCR3rvHUJ6CKRC2u68fyxZCKEksMmmnM/z7LOzM3fvnF0CzJzce65KURQFIYQQQgghhBBCCCGKkdrYAQghhBBCCCGEEEKI8keSUkIIIYQQQgghhBCi2ElSSgghhBBCCCGEEEIUO0lKCSGEEEIIIYQQQohiJ0kpIYQQQgghhBBCCFHsJCklhBBCCCGEEEIIIYqdJKWEEEIIIYQQQgghRLGTpJQQQgghhBBCCCGEKHaSlBJCCCGEEEIIIYQQxU6SUkKIUkelUjFu3LhCv+/KlSuoVCrmzZtn8JiEEEIIIcoyuf4SQhQFSUoJIZ7IvHnzUKlUqFQqdu7cme+4oih4e3ujUql4+eWXjRChYaxZswaVSoWHhwdardbY4QghhBCiHCvL119bt25FpVLx119/GTsUIUQxkqSUEOKpWFpa8vvvv+fbv23bNiIjI7GwsDBCVIazePFifHx8iImJYfPmzcYORwghhBCizF9/CSHKD0lKCSGeStu2bVm2bBk5OTl59v/+++80aNAANzc3I0X29NLT0/nf//7HiBEjqFevHosXLzZ2SA+Vnp5u7BCEEEIIUUzK8vWXEKJ8kaSUEOKp9OrVi8TERDZu3Kjfl5WVxV9//UXv3r0f+J709HTef/99vL29sbCwoGbNmkyaNAlFUfK0y8zM5L333sPFxQU7OzteeeUVIiMjH9hnVFQUAwYMwNXVFQsLC+rUqcOcOXOe6rP9/fff3L59m27dutGzZ09WrFhBRkZGvnYZGRmMGzeOGjVqYGlpibu7O507d+bixYv6Nlqtlh9//JGAgAAsLS1xcXGhdevWHDhwAHh0vYX7aziMGzcOlUrFqVOn6N27N05OTjz33HMAHDt2jH79+uHn54elpSVubm4MGDCAxMTEB35nAwcOxMPDAwsLC3x9fXnzzTfJysri0qVLqFQqfvjhh3zv+++//1CpVCxZsqSwX6kQQgghDKAsX389zqVLl+jWrRsVKlTA2tqaZ555htWrV+drN23aNOrUqYO1tTVOTk40bNgwz+iy1NRUhg8fjo+PDxYWFlSqVIlWrVpx6NChIo1fCJGXqbEDEEKUbj4+PjRp0oQlS5bQpk0bANauXUtycjI9e/Zk6tSpedorisIrr7zCli1bGDhwIMHBwaxfv56RI0cSFRWVJwkyaNAgFi1aRO/evWnatCmbN2+mXbt2+WKIi4vjmWeeQaVSMWzYMFxcXFi7di0DBw4kJSWF4cOHP9FnW7x4MS1btsTNzY2ePXvy8ccf888//9CtWzd9G41Gw8svv0x4eDg9e/bk3XffJTU1lY0bN3LixAmqVq0KwMCBA5k3bx5t2rRh0KBB5OTksGPHDvbs2UPDhg2fKL5u3bpRvXp1vv76a/0F5caNG7l06RL9+/fHzc2NkydP8uuvv3Ly5En27NmDSqUCIDo6mpCQEJKSkhgyZAj+/v5ERUXx119/cevWLfz8/Hj22WdZvHgx7733Xr7vxc7Ojg4dOjxR3EIIIYR4OmX5+utR4uLiaNq0Kbdu3eKdd96hYsWKzJ8/n1deeYW//vqLTp06ATBr1izeeecdunbtyrvvvktGRgbHjh1j7969+qTdG2+8wV9//cWwYcOoXbs2iYmJ7Ny5k9OnT1O/fn2Dxy6EeAhFCCGewNy5cxVA2b9/v/LTTz8pdnZ2yq1btxRFUZRu3bopLVu2VBRFUapUqaK0a9dO/76VK1cqgPLll1/m6a9r166KSqVSLly4oCiKohw5ckQBlLfeeitPu969eyuAMnbsWP2+gQMHKu7u7sr169fztO3Zs6fi4OCgj+vy5csKoMydO/exny8uLk4xNTVVZs2apd/XtGlTpUOHDnnazZkzRwGUyZMn5+tDq9UqiqIomzdvVgDlnXfeeWibR8V2/+cdO3asAii9evXK1zb3s95ryZIlCqBs375dv69Pnz6KWq1W9u/f/9CYfvnlFwVQTp8+rT+WlZWlODs7K3379s33PiGEEEIUrbJ8/bVlyxYFUJYtW/bQNsOHD1cAZceOHfp9qampiq+vr+Lj46NoNBpFURSlQ4cOSp06dR55PgcHB2Xo0KGPbCOEKHoyfU8I8dS6d+/O7du3+ffff0lNTeXff/996NDxNWvWYGJiwjvvvJNn//vvv4+iKKxdu1bfDsjX7v7fuimKwvLly2nfvj2KonD9+nX9IywsjOTk5Ccahr106VLUajVdunTR7+vVqxdr167l5s2b+n3Lly/H2dmZt99+O18fuaOSli9fjkqlYuzYsQ9t8yTeeOONfPusrKz02xkZGVy/fp1nnnkGQP89aLVaVq5cSfv27R84Sis3pu7du2NpaZmnltb69eu5fv06r7322hPHLYQQQoinVxavvx5nzZo1hISE6MsWANja2jJkyBCuXLnCqVOnAHB0dCQyMpL9+/c/tC9HR0f27t1LdHS0weMUQhScJKWEEE/NxcWF0NBQfv/9d1asWIFGo6Fr164PbHv16lU8PDyws7PLs79WrVr647nParVaP/0tV82aNfO8TkhIICkpiV9//RUXF5c8j/79+wMQHx9f6M+0aNEiQkJCSExM5MKFC1y4cIF69eqRlZXFsmXL9O0uXrxIzZo1MTV9+Gzoixcv4uHhQYUKFQodx6P4+vrm23fjxg3effddXF1dsbKywsXFRd8uOTkZ0H1nKSkp1K1b95H9Ozo60r59+zz1FxYvXoynpycvvPCCAT+JEEIIIQqrLF5/Pc7Vq1fzxfKgz/HRRx9ha2tLSEgI1atXZ+jQoezatSvPeyZOnMiJEyfw9vYmJCSEcePGcenSJYPHLIR4NKkpJYQwiN69ezN48GBiY2Np06YNjo6OxXJerVYLwGuvvUbfvn0f2CYwMLBQfZ4/f17/m7Xq1avnO7548WKGDBlSyEgf7WEjpjQazUPfc++oqFzdu3fnv//+Y+TIkQQHB2Nra4tWq6V169b676ow+vTpw7Jly/jvv/8ICAhg1apVvPXWW6jV8jsNIYQQwtjK0vWXIdWqVYuzZ8/y77//sm7dOpYvX87PP//MmDFj+PzzzwHdNVOzZs34+++/2bBhA9999x3ffvstK1as0NfpEkIUPUlKCSEMolOnTvzf//0fe/bs4Y8//nhouypVqrBp0yZSU1Pz/LbuzJkz+uO5z1qtVj8SKdfZs2fz9Je7MoxGoyE0NNQgn2Xx4sWYmZmxcOFCTExM8hzbuXMnU6dOJSIigsqVK1O1alX27t1LdnY2ZmZmD+yvatWqrF+/nhs3bjx0tJSTkxMASUlJefbn/savIG7evEl4eDiff/45Y8aM0e8/f/58nnYuLi7Y29tz4sSJx/bZunVrXFxcWLx4MY0bN+bWrVu8/vrrBY5JCCGEEEWnLF1/FUSVKlXyxQL5PweAjY0NPXr0oEePHmRlZdG5c2e++uorRo0ahaWlJQDu7u689dZbvPXWW8THx1O/fn2++uorSUoJUYzkV91CCIOwtbVlxowZjBs3jvbt2z+0Xdu2bdFoNPz000959v/www+oVCr9RUDu8/2rx0yZMiXPaxMTE7p06cLy5csfmGRJSEgo9GdZvHgxzZo1o0ePHnTt2jXPY+TIkQAsWbIEgC5dunD9+vV8nwfQr4jXpUsXFEXR/2buQW3s7e1xdnZm+/bteY7//PPPBY47N4Gm3Le08/3fmVqtpmPHjvzzzz8cOHDgoTEBmJqa0qtXL/7880/mzZtHQECAUX/zKYQQQoi7ytL1V0G0bduWffv2sXv3bv2+9PR0fv31V3x8fKhduzYAiYmJed5nbm5O7dq1URSF7OxsNBqNvqxBrkqVKuHh4UFmZmaRxC6EeDAZKSWEMJiHDd++V/v27WnZsiWffvopV65cISgoiA0bNvC///2P4cOH62sYBAcH06tXL37++WeSk5Np2rQp4eHhXLhwIV+f33zzDVu2bKFx48YMHjyY2rVrc+PGDQ4dOsSmTZu4ceNGgT/D3r17uXDhAsOGDXvgcU9PT+rXr8/ixYv56KOP6NOnDwsWLGDEiBHs27ePZs2akZ6ezqZNm3jrrbfo0KEDLVu25PXXX2fq1KmcP39eP5Vux44dtGzZUn+uQYMG8c033zBo0CAaNmzI9u3bOXfuXIFjt7e3p3nz5kycOJHs7Gw8PT3ZsGEDly9fztf266+/ZsOGDTz//PMMGTKEWrVqERMTw7Jly9i5c2ee4f99+vRh6tSpbNmyhW+//bbA8QghhBCi6JWF6697LV++XD/y6f7P+fHHH7NkyRLatGnDO++8Q4UKFZg/fz6XL19m+fLl+vICL730Em5ubjz77LO4urpy+vRpfvrpJ9q1a4ednR1JSUl4eXnRtWtXgoKCsLW1ZdOmTezfv5/vv//+ieIWQjwh4yz6J4Qo7e5dkvhR7l+SWFF0S/e+9957ioeHh2JmZqZUr15d+e677xStVpun3e3bt5V33nlHqVixomJjY6O0b99euXbtWr4liRVFUeLi4pShQ4cq3t7eipmZmeLm5qa8+OKLyq+//qpvU5Alid9++20FUC5evPjQNuPGjVMA5ejRo4qiKMqtW7eUTz/9VPH19dWfu2vXrnn6yMnJUb777jvF399fMTc3V1xcXJQ2bdooBw8e1Le5deuWMnDgQMXBwUGxs7NTunfvrsTHx+f7vGPHjlUAJSEhIV9skZGRSqdOnRRHR0fFwcFB6datmxIdHf3A7+zq1atKnz59FBcXF8XCwkLx8/NThg4dqmRmZubrt06dOoparVYiIyMf+r0IIYQQomiV1esvRVGULVu2KMBDHzt27FAURVEuXryodO3aVXF0dFQsLS2VkJAQ5d9//83T1y+//KI0b95cqVixomJhYaFUrVpVGTlypJKcnKwoiqJkZmYqI0eOVIKCghQ7OzvFxsZGCQoKUn7++edHxiiEMDyVotw3z0MIIYS4T7169ahQoQLh4eHGDkUIIYQQQghRRkhNKSGEEI904MABjhw5Qp8+fYwdihBCCCGEEKIMkZFSQgghHujEiRMcPHiQ77//nuvXr3Pp0iX9ajVCCCGEEEII8bRkpJQQQogH+uuvv+jfvz/Z2dksWbJEElJCCCGEEEIIg5KRUkIIIYQQQgghhBCi2MlIKSGEEEIIIYQQQghR7CQpJYQQQgghhBBCCCGKnamxAyittFot0dHR2NnZoVKpjB2OEEIIIYqJoiikpqbi4eGBWi2/33sUuV4SQgghyqeCXi9JUuoJRUdH4+3tbewwhBBCCGEk165dw8vLy9hhlGhyvSSEEEKUb4+7XpKk1BOys7MDdF+wvb29kaMRQgghRHFJSUnB29tbfy0gHk6ul4QQQojyqaDXS5KUekK5Q9Dt7e3lIksIIYQoh2Q62uPJ9ZIQQghRvj3uekkKIQghhBBCCCGEEEKIYidJKSGEEEIIIYQQQghR7CQpJYQQQgghhBBCCCGKndSUKmIajYbs7GxjhyGEwZmZmWFiYmLsMIQQQgghhBCPoNVqycrKMnYYoowx1P2gJKWKiKIoxMbGkpSUZOxQhCgyjo6OuLm5SbFfIYQQQgghSqCsrCwuX76MVqs1diiiDDLE/aAkpYpIbkKqUqVKWFtby027KFMUReHWrVvEx8cD4O7ubuSIhBBCCCGEEPdSFIWYmBhMTEzw9vZGrZbqPcIwDHk/KEmpIqDRaPQJqYoVKxo7HCGKhJWVFQDx8fFUqlRJpvIJIYQQQghRguTk5HDr1i08PDywtrY2djiijDHU/aBRU6Xbt2+nffv2eHh4oFKpWLly5WPfs3XrVurXr4+FhQXVqlVj3rx5+dpMnz4dHx8fLC0tady4Mfv27ctzPCMjg6FDh1KxYkVsbW3p0qULcXFxBvpU6GtIyV98Udbl/oxL3TQhhBBCCCFKFo1GA4C5ubmRIxFllSHuB42alEpPTycoKIjp06cXqP3ly5dp164dLVu25MiRIwwfPpxBgwaxfv16fZs//viDESNGMHbsWA4dOkRQUBBhYWH6YWUA7733Hv/88w/Lli1j27ZtREdH07lzZ4N/PpmyJ8o6+RkXQgghhBCiZJNrdlFUDPGzZdSkVJs2bfjyyy/p1KlTgdrPnDkTX19fvv/+e2rVqsWwYcPo2rUrP/zwg77N5MmTGTx4MP3796d27drMnDkTa2tr5syZA0BycjKzZ89m8uTJvPDCCzRo0IC5c+fy33//sWfPniL5nEIIIYQQRe1xI8UfZunSpahUKjp27Jhnv6IojBkzBnd3d6ysrAgNDeX8+fNFEPmT02oVtFrF2GEIIYQQ4gmVqkpnu3fvJjQ0NM++sLAwdu/eDehWFjh48GCeNmq1mtDQUH2bgwcPkp2dnaeNv78/lStX1rd5kMzMTFJSUvI8xOP5+PgwZcoUY4chhBCiNNPkQHoiJF6EyINwYRMc/wv2z4Yd38OGz2DV2/DH6zC/PcxsBlMCYM2Hxo682BRkpPiDXLlyhQ8++IBmzZrlOzZx4kSmTp3KzJkz2bt3LzY2NoSFhZGRkVFUH6NQhv5+iKDxGzgWlWzsUIQQQpRwcl9acpWqQuexsbG4urrm2efq6kpKSgq3b9/m5s2baDSaB7Y5c+aMvg9zc3McHR3ztYmNjX3ouSdMmMDnn39umA9SAj1u2N3YsWMZN25cofvdv38/NjY2TxhVXkuWLOG1117jjTfeKPCUTyGEECVEThZkJMHtpLzPGcn5991/LCv1yc6ZEvX0cZcS944UB93o8tWrVzNnzhw+/vjjB75Ho9Hw6quv8vnnn7Njxw6SkpL0xxRFYcqUKYwePZoOHToAsGDBAlxdXVm5ciU9e/Ys8s/0OOmZOaRm5HAsMolgb0djhyOEEMIASvJ9aYsWLQgODpbkloGVqqSUMY0aNYoRI0boX6ekpODt7W3EiAwrJiZGv/3HH38wZswYzp49q99na2ur31YUBY1Gg6np4398XFxcDBbj7Nmz+fDDD/nll1/4/vvvsbS0NFjfhZWVlSUFA4UQ5U/27fsSR8kPTiY9qE32rac/v7kdWDmCpQNYOt7ZfsCzpYNu287t6c9ZCuSOFB81apR+3/0jxR9k/PjxVKpUiYEDB7Jjx448xy5fvkxsbGyekeUODg40btyY3bt3PzQplZmZSWZmpv51UY4sD/RyZOvZBI5eS4YmRXYaIYQQxag03JcKwypV0/fc3NzyrZIXFxeHvb09VlZWODs7Y2Ji8sA2bm5u+j6ysrLy/Dbw/jYPYmFhgb29fZ5HWeLm5qZ/ODg4oFKp9K/PnDmDnZ0da9eupUGDBlhYWLBz504uXrxIhw4dcHV1xdbWlkaNGrFp06Y8/d4/TFKlUvHbb7/RqVMnrK2tqV69OqtWrXpsfJcvX+a///7j448/pkaNGqxYsSJfmzlz5lCnTh0sLCxwd3dn2LBh+mNJSUn83//9H66urlhaWlK3bl3+/fdfAMaNG0dwcHCevqZMmYKPj4/+db9+/ejYsSNfffUVHh4e1KxZE4CFCxfSsGFD7OzscHNzo3fv3vmmSpw8eZKXX34Ze3t77OzsaNasGRcvXmT79u2YmZnlG6E3fPjwB06jEEKIIpMaCwfnw+avYM1IWD4YFneD2S/BTyEwqQZ8UQm+coPJ/vDzMzC3NSzpAX//H6z7CLZOgL0z4OgSOLcWInZDwmlIjb4nIaXSJYwcK4NbIPg2h1qvQP0+0PQdeHEMtJsMXWbDa8th0GZ4+xCMvASfJcInkfDeCXhzF/RfDT0XQ8fpEPYVPD8SQgZDYDeo8RJ4h+jOUw5cv379oSPFHzYKfOfOncyePZtZs2Y98Hju+wrTJ+hGljs4OOgfRfkLvCAvBwCORSYV2TmEEEIUr5J+X/ooy5cv19+P+vj48P333+c5/vPPP1O9enUsLS1xdXWla9eu+mN//fUXAQEBWFlZUbFiRUJDQ0lPT3+qeEqLUjVSqkmTJqxZsybPvo0bN9Kkie7XY+bm5jRo0IDw8HB9sU6tVkt4eLg+QdGgQQPMzMwIDw+nS5cuAJw9e5aIiAh9P4amKAq3szVF0vfjWJmZGGy1hY8//phJkybh5+eHk5MT165do23btnz11VdYWFiwYMEC2rdvz9mzZ6lc+eE3Ap9//jkTJ07ku+++Y9q0abz66qtcvXqVChUqPPQ9c+fOpV27djg4OPDaa68xe/ZsevfurT8+Y8YMRowYwTfffEObNm1ITk5m165dgO5noE2bNqSmprJo0SKqVq3KqVOnMDExKdTnDw8Px97eno0bN+r3ZWdn88UXX1CzZk3i4+MZMWIE/fr10/+cRkVF0bx5c1q0aMHmzZuxt7dn165d5OTk0Lx5c/z8/Fi4cCEjR47U97d48WImTpxYqNiEEKLQUqLh9D9wcqUugUQBi0Wr1AUfqXT/MQt7UBfu315heKmpqbz++uvMmjULZ2dng/ZdnCPLA70cAbiQkEZaZg62FqXqslYIIYqd3Jfm9ST3pQ9z8OBBunfvzrhx4+jRowf//fcfb731FhUrVqRfv34cOHCAd955h4ULF9K0aVNu3LihH6UcExNDr169mDhxIp06dSI1NZUdO3agKOVjIQ+j/u+dlpbGhQsX9K8vX77MkSNHqFChApUrV2bUqFFERUWxYMECAN544w1++uknPvzwQwYMGMDmzZv5888/Wb16tb6PESNG0LdvXxo2bEhISAhTpkwhPT1dX2PBwcGBgQMHMmLECCpUqIC9vT1vv/02TZo04ZlnnimSz3k7W0PtMeuLpO/HOTU+DGtzw/wxjx8/nlatWulfV6hQgaCgIP3rL774gr///ptVq1blGaV0v379+tGrVy8Avv76a6ZOncq+ffto3br1A9trtVrmzZvHtGnTAOjZsyfvv/8+ly9fxtfXF4Avv/yS999/n3fffVf/vkaNGgGwadMm9u3bx+nTp6lRowYAfn5+hf78NjY2/Pbbb3mm7Q0YMEC/7efnx9SpU2nUqBFpaWnY2toyffp0HBwcWLp0KWZmZgD6GAAGDhzI3Llz9Umpf/75h4yMDLp3717o+IQQ4rGSI+HUKji1Eq7tzXvMsyF4BD8+4WRuC+pSNdC6zCvISPF7Xbx4kStXrtC+fXv9Pq1WC4CpqSlnz57Vvy8uLg53d/c8fd4/uvheFhYWWFhYPM3HKTAXOws8HCyJTs7gRFQyz/hVLJbzCiFEaSX3pXkV9r70USZPnsyLL77IZ599Buju+U6dOsV3331Hv379iIiIwMbGhpdffhk7OzuqVKlCvXr1AF1SKicnh86dO1OlShUAAgICCh1DaWXUpNSBAwdo2bKl/nXub9b69u3LvHnziImJISIiQn/c19eX1atX89577/Hjjz/i5eXFb7/9RlhYmL5Njx49SEhIYMyYMcTGxhIcHMy6devyDD//4YcfUKvVdOnShczMTMLCwvj555+L4ROXbg0bNszzOi0tjXHjxrF69Wr9X6Tbt2/n+TN7kMDAQP22jY0N9vb2j1wdaOPGjaSnp9O2bVtAd/HdqlUr5syZwxdffEF8fDzR0dG8+OKLD3z/kSNH8PLyypMMehIBAQH56kgdPHiQcePGcfToUW7evKm/qI+IiKB27docOXKEZs2a6RNS9+vXrx+jR49mz549PPPMM8ybN4/u3bsbrDi8EEJw8yqcXqUbERV1IO8x72egdgeo1R4cy06dxPKmICPF7+Xv78/x48fz7Bs9ejSpqan8+OOPeHt7Y2ZmhpubG+Hh4fokVEpKCnv37uXNN98s6o9UYIFejkQnx3IsMkmSUkIIUU4Y6770UU6fPq1fGCTXs88+y5QpU9BoNLRq1YoqVarg5+dH69atad26tX7qYFBQEC+++CIBAQGEhYXx0ksv0bVrV5ycnJ4oltLGqEmpFi1aPHJI2rx58x74nsOHDz+y32HDhj0yI2ppacn06dOLbQU3KzMTTo0Pe3zDIjq3odyfKPnggw/YuHEjkyZNolq1alhZWdG1a1eysrIe2c/9CRqVSqVP5jzI7NmzuXHjBlZWVvp9Wq2WY8eO8fnnn+fZ/yCPO65Wq/P9HGZnZ+drd//nT09PJywsjLCwMBYvXoyLiwsRERGEhYXpv4PHnbtSpUq0b9+euXPn4uvry9q1a9m6desj3yOEEI914zKc+p/uEX3ongMqqNwE6nTUJaLsPYwVoTCwx40U79OnD56enkyYMEFfW/FeuasS37t/+PDhfPnll1SvXh1fX18+++wzPDw89ImvkiDQ24F1J2M5Gpls7FCEEKLEk/vSvAp7X/o07OzsOHToEFu3bmXDhg2MGTOGcePGsX//fhwdHdm4cSP//fcfGzZsYNq0aXz66afs3btXPzOoLJPJ98VApVIZbKhiSbJr1y769etHp06dAF2G+sqVKwY9R2JiIv/73/9YunQpderU0e/XaDQ899xzbNiwgdatW+Pj40N4eHiekXe5AgMDiYyM5Ny5cw8cLeXi4kJsbCyKoujnOR85cuSxsZ05c4bExES++eYbfb2MAwfyjkIIDAxk/vz5ZGdnP3S01KBBg+jVqxdeXl5UrVqVZ5999rHnFkKIfBIv6qblnfofxBy9u1+lhirP3h0RVU5WpCtvHjdSPCIiAnUhp11++OGHpKenM2TIEJKSknjuuedYt26dUVe/vV/QnbpSR68lGTUOIYQoDeS+tOjUqlVLX9P43rhq1Kihr2VsampKaGgooaGhjB07FkdHRzZv3kznzp1RqVQ8++yzPPvss4wZM4YqVarw999/56nTWFaVvZ9IUWyqV6/OihUraN++PSqVis8++8zgmeWFCxdSsWJFunfvnq8wXtu2bZk9ezatW7dm3LhxvPHGG1SqVElf1HzXrl28/fbbPP/88zRv3pwuXbowefJkqlWrxpkzZ1CpVLRu3ZoWLVqQkJDAxIkT6dq1K+vWrWPt2rWPXWGxcuXKmJubM23aNN544w1OnDjBF198kafNsGHDmDZtGj179mTUqFE4ODiwZ88eQkJC9Cv4hYWFYW9vz5dffsn48eMN+v0JIcq4hHN3R0TF3TMdS6UGn2a6EVH+L4NtJaOFKIrPo0aKP24U7oNGp6tUKsaPH1+i/2+q66lbgS/y5m0S0zKpaFs89ayEEEKUHMVxX5orISEh3wAGd3d33n//fRo1asQXX3xBjx492L17Nz/99JO+TNC///7LpUuXaN68OU5OTqxZswatVkvNmjXZu3cv4eHhvPTSS1SqVIm9e/eSkJBArVq1iuQzlDRSqVQ8scmTJ+Pk5ETTpk1p3749YWFh1K9f36DnmDNnDp06dXrgSg1dunRh1apVXL9+nb59+zJlyhR+/vln6tSpw8svv8z58+f1bZcvX06jRo3o1asXtWvX5sMPP0Sj0a08UatWLX7++WemT59OUFAQ+/bt44MPPnhsbC4uLsybN49ly5ZRu3ZtvvnmGyZNmpSnTcWKFdm8eTNpaWk8//zzNGjQgFmzZuUZNaVWq+nXrx8ajYY+ffo86VclhCgv4s/A1m/g5yYwvRFs+VKXkFKZQNUXoP2P8MF56LsKGg6QhJQo0xyszPBz1k3jOBYlU/iEEKI8Ko770ly///479erVy/OYNWsW9evX588//2Tp0qXUrVuXMWPGMH78ePr16wfopsmvWLGCF154gVq1ajFz5kyWLFlCnTp1sLe3Z/v27bRt25YaNWowevRovv/+e9q0aVMkn6GkUSnlZZ1BA0tJScHBwYHk5OR8I2oyMjL0K8OVpCHuouQaOHAgCQkJrFq1ytihFIr8rAtRDBQF4k/pRkOdXAnXz949pjYFv5a6qXn+7cC68EsYi8J71DWAyKs4vqvhSw+z8kg074XW4N3Q6kVyDiGEKI3kWl0UtUf9jBX0GkCm7wlhRMnJyRw/fpzff/+91CWkhBBFSFEg9vidqXkrIfHC3WMm5roRUbU7QM02YFU+VmYprKwcLTlabZmsnSHyCvJ2ZOWRaI5FJhk7FCGEEEIUklypCWFEHTp0YN++fbzxxhu0atXK2OEIIYxJUSDmyN0aUTcu3T1mYgHVXoTaHaFma7B0MFaUJd61G7dYuj+CP/ZH0v9ZH4a2rGbskEQRC8wtdh6ZnGfREiGEEEKUfJKUEsKIHld4VghRxikKRB26u2pe0tW7x0wtoVoo1OkE1V8CS5km9jAarcK2c/Es2hPBlrPx5BYm2HQ6TpJS5UAdD3tM1Squp2USk5yBh6OVsUMSQgghRAFJUkoIIYQoTlotRB24OyIq+drdY6ZWUOMl3Yio6i+Bha3RwiwN4lMzWHYgkt/3RhCVdFu//9lqFXmtcRVCa7saMTpRXCzNTKjhasepmBSORSZJUkoIIYQoRSQpJYQQQhQ1rRau7dUloU6vgpSou8fMbKBGmK5GVPVWYG5jvDhLAUVR2HPpBov2XmX9iVhytLphUQ5WZnRr4EXvxpXxc5FkXnkT5O3AqZgUjkYm07quu7HDEUIIIUQBSVJKCCFE0UmN09VJijkK0Ufg1nVdPSRLB7B0BCvHRzw7gLktlNb6MFoNROy+MyJqFaTF3j1mbqerDVW7g26KnpmM7Hic5FvZLD8UyeK9V7mYkK7fX6+yI681rkK7QHcszUyMGKEwpkAvR5bsuybFzoUQQohSRpJSQgghnp6iQGqMLvEUc1SXiIo+kjcR8yTUpg9PYFk6PDypZekAFvagVj/d+QtLkwMR/8HJlXD6H0iPv3vMwh5qttUloqq+AGayNHNBHL2WxKI9V/nnWDQZ2VoArM1N6FjPk1cbV6aOhxR9FxDopfs5OBaZjFaroFaX0mS2EEIIUc5IUkoIIUThKAokR+ZNPsUczZuAyaVSg3MNcA8C92Bw8ISMFMhIhowkuJ308GdtNmhz4Fai7lFYKrUuEfSwUViPGqll6QDqAo660eTAlR26YuWn/9WNBstl6QD+L+sSUX4twNSi8J+jHLqVlcOqI9Es3hvB8ahk/X5/NztefaYKHYM9sLM0M2KEoqSp4WqHhama1IwcriSmyxROIYQQopSQpJQQQoiHUxTdinC50+9yp+I9KEmkUoOLvy755BGsS0S5BTxZjSRFgexbuuTVoxJXGUkPbpOTAYr2zvGkwp8fdAktS0ewelAC686+2GO6RNTtG3ffZ+UE/u2gdifwbQ6m5k92/nLoXFwqi/dcZcWhKFIzcwAwN1HTLtCdVxtXpkEVJ1SldTqnKFJmJmrqeNhzKCKJo5FJkpQSQgghSglJSgmDatGiBcHBwUyZMgUAHx8fhg8fzvDhwx/6HpVKxd9//03Hjh2f6tyG6keIcktR4OblvMmnmKNw+2b+tmpTcKkFHndGQLkHg2sdMLc2TCwqlS6ZZW4D9h6Ff392xuMTVw9LbmXfqVeUmaJ7JOfvPh/riroRUXU6gk8zMJFRPAWVmaNh3YlYFu+NYN/lu8m9KhWtebVxZbo28KaCjST2xOMFejnqklLXkulUz8vY4QghhDAiuS8tPSQpJQBo37492dnZrFu3Lt+xHTt20Lx5c44ePUpgYGCh+t2/fz82NoZdSWrcuHGsXLmSI0eO5NkfExODk5OTQc/1MLdv38bT0xO1Wk1UVBQWFjIlR5QyWi3cuHRn+t3hOwmoY5D5gAyM2gxca99JPgXpRkFVqlOyayKZWYKZG9i5Ff69OVkPn154/z47N6j1ClR5Fkzkv9TCuHbjFr/vi+DP/ddITM8CwEStIrRWJV5tXIXnqjlLXSBRKEHeuXWlkowbiBBCiCcm96UFM2/ePIYPH05SUlKRnqc4yBW0AGDgwIF06dKFyMhIvLzy/nZx7ty5NGzYsNB/8QFcXFwMFeJjubk9wc3nE1q+fDl16tRBURRWrlxJjx49iu3c91MUBY1Gg6mp/HUWD6HVQOKFvEXIY45BVmr+tiYWuhFPudPv3IOhUu3yNQXN1BxsXXQPYVAarcLmM/Es3nuVbecSUBTdfld7C3qFVKZno8q4OZTgZKco0QK9HAE4GZ1CtkaLmUkxL3QghBDiqcl9afkj/1sLAF5++WVcXFyYN29env1paWksW7aMgQMHkpiYSK9evfD09MTa2pqAgACWLFnyyH59fHz0QyYBzp8/T/PmzbG0tKR27dps3Lgx33s++ugjatSogbW1NX5+fnz22WdkZ2cDuozw559/ztGjR1GpVKhUKn3MKpWKlStX6vs5fvw4L7zwAlZWVlSsWJEhQ4aQlpamP96vXz86duzIpEmTcHd3p2LFigwdOlR/rkeZPXs2r732Gq+99hqzZ8/Od/zkyZO8/PLL2NvbY2dnR7Nmzbh48aL++Jw5c6hTpw4WFha4u7szbNgwAK5cuYJKpcqTbU9KSkKlUrF161YAtm7dikqlYu3atTRo0AALCwt27tzJxYsX6dChA66urtja2tKoUSM2bdqUJ67MzEw++ugjvL29sbCwoFq1asyePRtFUahWrRqTJk3K0/7IkSOoVCouXLjw2O9ElBCaHIg7BUeWwNqPYHYYTPCG6SHw9xDYMx2u7tIlpEwtwasRNBoMr/wEb+yET6JgyBZ4+Qdo0E+XnCpPCSlRJOJTMpgWfp5m325m8IIDbD2rS0g1q+7MzNcasOujFxgeWkMSUuKp+Fa0wc7SlMwcLefiHpB0F0IIUeLJfWnh7ksfJiIigg4dOmBra4u9vT3du3cnLi5Of/zo0aO0bNkSOzs77O3tadCgAQcOHADg6tWrtG/fHicnJ2xsbKhTpw5r1qx54lgeR4ZWFIfcgr3GYGatq83yGKampvTp04d58+bx6aef6gvJLlu2DI1GQ69evUhLS6NBgwZ89NFH2Nvbs3r1al5//XWqVq1KSEjIY8+h1Wrp3Lkzrq6u7N27l+Tk5AfO6bWzs2PevHl4eHhw/PhxBg8ejJ2dHR9++CE9evTgxIkTrFu3Tp9wcXDIvxx4eno6YWFhNGnShP379xMfH8+gQYMYNmxYnn/gtmzZgru7O1u2bOHChQv06NGD4OBgBg8e/NDPcfHiRXbv3s2KFStQFIX33nuPq1evUqVKFQCioqJo3rw5LVq0YPPmzdjb27Nr1y5ycnRFe2fMmMGIESP45ptvaNOmDcnJyezateux39/9Pv74YyZNmoSfnx9OTk5cu3aNtm3b8tVXX2FhYcGCBQto3749Z8+epXLlygD06dOH3bt3M3XqVIKCgrh8+TLXr19HpVIxYMAA5s6dywcffKA/x9y5c2nevDnVqlUrdHyiGGiyIeHMPUXIj0Lscci5nb+tmTW4Bd6dfucerFsVT6aciSKiKAq7LyayaO9VNpyMI0erGxblZG1Gt4be9A6pjI+zYYfRi/JNrVYR6OXArguJHItMpo5H/usDIYQo1+S+FCg796WP+ny5Calt27aRk5PD0KFD6dGjh36gw6uvvkq9evWYMWMGJiYmHDlyBDMzXU3UoUOHkpWVxfbt27GxseHUqVPY2hbdAiJyN1Icsm/B109QqNcQPoku8MpXAwYM4LvvvmPbtm20aNEC0CUlunTpgoODAw4ODnkSFm+//Tbr16/nzz//LNBf/k2bNnHmzBnWr1+Ph4fu+/j6669p06ZNnnajR4/Wb/v4+PDBBx+wdOlSPvzwQ6ysrLC1tcXU1PSRwyJ///13MjIyWLBggX7u8E8//UT79u359ttvcXV1BcDJyYmffvoJExMT/P39adeuHeHh4Y/8yz9nzhzatGmjnyccFhbG3LlzGTduHADTp0/HwcGBpUuX6v9i16hRQ//+L7/8kvfff593331Xv69Ro0aP/f7uN378eFq1aqV/XaFCBYKCgvSvv/jiC/7++29WrVrFsGHDOHfuHH/++ScbN24kNDQUAD8/P337fv36MWbMGPbt20dISAjZ2dn8/vvv+UZPCSPJyYKE03mLkMeeAE1m/rbmtroEVG7yyT0InKuD2qR4YxblUtKtLP46GMnveyO4dD1dv79BFSdee6Yybeq6Y2kmP4uiaAR6Od5JSiXRK6SyscMRQoiSRe5LgbJzX/ow4eHhHD9+nMuXL+Pt7Q3AggULqFOnDvv376dRo0ZEREQwcuRI/P39Aahevbr+/REREXTp0oWAgAAg7z1jUZCklNDz9/enadOmzJkzhxYtWnDhwgV27NjB+PHjAdBoNHz99df8+eefREVFkZWVRWZmJtbWBVtt6/Tp03h7e+v/4gM0adIkX7s//viDqVOncvHiRdLS0sjJycHe3r5Qn+X06dMEBQXlKWb37LPPotVqOXv2rP4vf506dTAxuXtz5O7uzvHjxx/ar0ajYf78+fz444/6fa+99hoffPABY8aMQa1Wc+TIEZo1a6ZPSN0rPj6e6OhoXnzxxUJ9ngdp2LBhntdpaWmMGzeO1atXExMTQ05ODrdv3yYiIgLQTcUzMTHh+eeff2B/Hh4etGvXjjlz5hASEsI///xDZmYm3bp1e+pYxSPcv0rc7aS8RbZTonQJqPhToMnK/34L+zu1n4LAo57uuUJVUMvsbFF8FEXhyLUkFu2J4N9j0WTmaAGwMTehU31PXm1chVruhft3XIgnEeSl+y310WsFWTZTCCFESST3pY+/L33cOb29vfUJKYDatWvj6OjI6dOnadSoESNGjGDQoEEsXLiQ0NBQunXrRtWqVQF45513ePPNN9mwYQOhoaF06dLliep4FZQkpYqDmbUuM2yscxfCwIEDefvtt5k+fTpz586latWq+iTGd999x48//siUKVMICAjAxsaG4cOHk5X1gBvlJ7R7925effVVPv/8c8LCwvQjjr7//nuDneNe9yeOVCoVWq32oe3Xr19PVFRUvsLmGo2G8PBwWrVqhZWV1UPf/6hjAOo7iQQlt/ovPHQu8f2rR3zwwQds3LiRSZMmUa1aNaysrOjatav+z+dx5wYYNGgQr7/+Oj/88ANz586lR48eBf7HvdzKHQadb5W25Pz7HpR4ysko+LksHfNOv3MPAidfSUAJo0nPzOF/R6JZvPcqJ6NT9Ptrudvz2jOV6RDsia2FXGqI4pNb7PxsXCoZ2RoZlSeEEPeS+9ICK+n3pU9r3Lhx9O7dm9WrV7N27VrGjh3L0qVL6dSpE4MGDSIsLIzVq1ezYcMGJkyYwPfff8/bb79dJLHIlWJxUKkKPFTR2Lp37867777L77//zoIFC3jzzTf183h37dpFhw4deO211wDdXNVz585Ru3btAvVdq1Ytrl27RkxMDO7u7gDs2bMnT5v//vuPKlWq8Omnn+r3Xb16NU8bc3NzNBrNY881b9480tPT9cmbXbt2oVarqVmzZoHifZDZs2fTs2fPPPEBfPXVV8yePZtWrVoRGBjI/Pnzyc7OzvePi52dHT4+PoSHh9OyZct8/eeuChETE0O9evUA8i0x+jC7du2iX79+dOrUCdCNnLpy5Yr+eEBAAFqtlm3btumn792vbdu22NjYMGPGDNatW8f27dsLdO5ST1EgM/XRI5Ye9JybeNI+eRFCAFRqsHS483AEK8e7zzYu4FpXl4hyrFKgufhCFLWzsaks2nOVvw9HkZapq5dnbqrm5UB3XnumCvW8HfX/dwhRnNwdLHG2teB6WiYno1NoUKVol+QWQohSRe5LgbJxX/q4c167do1r167pR0udOnWKpKSkPN9RjRo1qFGjBu+99x69evVi7ty5+ntJb29v3njjDd544w1GjRrFrFmzJCklioetrS09evRg1KhRpKSk0K9fP/2x6tWr89dff/Hff//h5OTE5MmTiYuLK/Bf/tDQUGrUqEHfvn357rvvSElJyZfcqV69OhERESxdupRGjRqxevVq/v777zxtfHx8uHz5MkeOHMHLyws7OzssLCzytHn11VcZO3Ysffv2Zdy4cSQkJPD222/z+uuv64dIFlZCQgL//PMPq1atom7dunmO9enTh06dOnHjxg2GDRvGtGnT6NmzJ6NGjcLBwYE9e/YQEhJCzZo1GTduHG+88QaVKlWiTZs2pKamsmvXLt5++22srKx45pln+Oabb/D19SU+Pj7PXOZHqV69OitWrKB9+/aoVCo+++yzPNl1Hx8f+vbty4ABA/SFzq9evUp8fDzdu3cHwMTEhH79+jFq1CiqV6/+wGGsJV5KNKTFFXLEUjIoj/4P5bFUJnmTSfmeH5Bwyn02t5PRTqLEy8zRsPZ4LIv2XOXA1Zv6/b7ONrzauDJdG3jhaC0rNQrjUqlUBHk5EH4mnmORSZKUEkKIUkruSx9Po9HkG8BgYWFBaGgoAQEBvPrqq0yZMoWcnBzeeustnn/+eRo2bMjt27cZOXIkXbt2xdfXl8jISPbv30+XLl0AGD58OG3atKFGjRrcvHmTLVu2UKtWraeK9VEkKSXyGThwILNnz6Zt27Z55tmOHj2aS5cuERYWhrW1NUOGDKFjx44kJxesboNarebvv/9m4MCBhISE4OPjw9SpU2ndurW+zSuvvMJ7773HsGHDyMzMpF27dnz22Wf6IuIAXbp0YcWKFbRs2ZKkpCTmzp2b5x8pAGtra9avX8+7775Lo0aNsLa2pkuXLkyePPmJv5fc4nQPqgf14osvYmVlxaJFi3jnnXfYvHkzI0eO5Pnnn8fExITg4GCeffZZAPr27UtGRgY//PADH3zwAc7OznTt2lXf15w5cxg4cCANGjSgZs2aTJw4kZdeeumx8U2ePJkBAwbQtGlTnJ2d+eijj0hJScnTZsaMGXzyySe89dZbJCYmUrlyZT755JM8bQYOHMjXX39N//79n+RrMq6D8+Cfdx/b7KFMzB+cOLJ0eEzCyVH3WycZGSLKoKuJ6fy+L4JlByK5ka4bFm+iVvFSbVdee6YKTfwqolbLz74oOQK9HO8kpaSulBBClGZyX/poaWlp+tk1uapWrcqFCxf43//+x9tvv03z5s1Rq9W0bt2aadOmAbqBCImJifTp04e4uDicnZ3p3Lkzn3/+OaBLdg0dOpTIyEjs7e1p3bo1P/zww1PH+zAq5d7iNaLAUlJScHBwIDk5OV+xs4yMDC5fvoyvry+WlpZGilCIJ7Njxw5efPFFrl279tjsfYn6WddqYVo9uHlFN+XN2vnxI5TuH8VkZiWJJSGAHI2W8DPxLN4bwfZzCfr97g6W9AqpTI9G3rjal9//3x51DSDyMsZ3teVsPP3n7sfP2YbNH7QolnMKIURJVKKu1UWZ9KifsYJeA8hIKSEEAJmZmSQkJDBu3Di6dev21MNJi92FTbqElKUDvHsMzKVAuxCFFZeSwdJ911i6P4KYZF0RfpUKmld34bVnqtCypgumJjLVtKSaPn063333HbGxsQQFBTFt2rSHLo29YsUKvv76ay5cuEB2djbVq1fn/fff5/XXX9e36devH/Pnz8/zvrCwMNatW1ekn+NpBd0pdn7pejrJt7NxsMq/Gq4QQgghSgZJSgkhAFiyZAkDBw4kODiYBQsWGDucwtv3q+653uuSkBKiENIzc9h8Jp5/jkYTfiYejVY3gLqCjTndG3rTO6QylSvK36mS7o8//mDEiBHMnDmTxo0bM2XKFMLCwjh79iyVKlXK175ChQp8+umn+Pv7Y25uzr///kv//v2pVKkSYWFh+natW7dm7ty5+tf318ooiSrYmOPlZEXkzduciErm2WrOxg5JCCGEEA8hSSkhBKD7jfj9c6BLjcSLcGEjoIJGA40djRAlXmpGNpvPxLPmeAxbzyaQmXN3UYQQnwq8+kxlWtd1w8LUxIhRisKYPHkygwcP1tcDnDlzJqtXr2bOnDl8/PHH+dq3aNEiz+t3332X+fPns3PnzjxJKQsLC9zc3Io09qIQ5OVI5M3bHI1MkqSUEEIIUYJJUkoIUfrtn617rt4KKvgZNxYDURSFf4/FcPDqTRr5VKClvwvW5vJPtnhyKRnZbDoVx5rjsWw/n0DWPYkon4rWtA1wp0OwJzXd7IwYpXgSWVlZHDx4kFGjRun3qdVqQkND2b1792PfrygKmzdv5uzZs3z77bd5jm3dupVKlSrh5OTECy+8wJdffknFihUN/hkMLcjbgdXHYzh2TYqdCyGEECWZ3OEIIUq3rHQ4vEi3HTLEuLEYyKGIm3zx7ykORyQBMO+/K1iaqWlRoxJtA915wb8Sthbyz7d4vORb2Ww4FcvaE7HsOJ9Atubu2iZ+Lja0C3CnTV13arnboZIi/6XW9evX0Wg0+WoBurq6cubMmYe+Lzk5GU9PTzIzMzExMeHnn3+mVatW+uOtW7emc+fO+Pr6cvHiRT755BPatGnD7t27MTF58Ci6zMxMMjMz9a/vXwW2uATeqSt1LDLJKOcXQgghRMHIXU0R0mq1j28kRClWIn7Gjy+DzGRw8oWqLxo7mqcSlXSbb9eeYdXRaACszU1oU9ed/VduEHHjFutOxrLuZCwWpmqer+FC2wB3XqxVCTtLKeIr7rqZnsWGU7GsOR7LrgvXydHeTURVr2RL2wB32ga4U8PVVhJR5ZydnR1HjhwhLS2N8PBwRowYgZ+fn35qX8+ePfVtAwICCAwMpGrVqmzdupUXX3zwv7cTJkzQLyltTHU9HVCpIDo5g4TUTFzsSn4tLCGEKCqKojy+kRBPwBD3g5KUKgLm5uao1Wqio6NxcXHB3NxcLvxFmaIoCllZWSQkJKBWqzE3NzdWILBvlm47ZDCoS+eqYOmZOczYepFZOy6RmaNFpYJuDbz44KWaVLK3RFEUTkansOZ4DGuOx3Al8RYbTsWx4VQc5iZqmtdwpk1dd0Jru8oqU+VUYlomG07FseZ4DP9dTNQXKwfwd7OjbYA7beq6Ud1VpuaVRc7OzpiYmBAXF5dnf1xc3CPrQanVaqpVqwZAcHAwp0+fZsKECfnqTeXy8/PD2dmZCxcuPDQpNWrUKEaMGKF/nZKSgre3dyE/0dOztTClmost5+PTOBaZxIu1StmKskIIYQBmZmaoVCoSEhJwcXGRe1JhMIa8H5SkVBFQq9X4+voSExNDdHS0scMRoshYW1tTuXJl1MZKBkXshrgTYGYNwb2NE8NT0GoV/joUyXfrz5KQqpvu0ti3Ap+9XJu6ng76diqVirqeDtT1dGBkWE1Ox6Sy9kQMq4/HcCkhnU2n49l0Oh4zExXPVXOmTYA7L9V2xdHaSMlCUSwSUjNZfzKWtSdi2HPpRp5EVG13e9oGuNEmwJ2qLrZGjFIUB3Nzcxo0aEB4eDgdO3YEdL+5DA8PZ9iwYQXuR6vV5pl6d7/IyEgSExNxd3d/aBsLC4sSs0JfoJcj5+PTOBqZLEkpIUS5ZGJigpeXF5GRkVy5csXY4YgyyBD3g5KUKiLm5uZUrlyZnJwcNBqNscMRwuBMTEwwNTU17m9c9v2qew7sDlZOxovjCey5lMgX/57iZLSu3kqVitaMalOLsDquj/xOVSoVtT3sqe1hz4hWNTgXl6YfQXU+Po0tZxPYcjaBT9QqmlZzpm1dN16q40YFG0lQlQXxKRmsOxnLmuMx7Lt8g3vyUAR4OtAmwI22dd3xcbYxXpDCKEaMGEHfvn1p2LAhISEhTJkyhfT0dP1qfH369MHT05MJEyYAuml2DRs2pGrVqmRmZrJmzRoWLlzIjBkzAEhLS+Pzzz+nS5cuuLm5cfHiRT788EOqVauWZ3W+kizI24HlhyKlrpQQolyztbWlevXqZGdnGzsUUcYY6n5QklJFSKVSYWZmhpmZTKcRwuBSouH0P7rtRoONG0shXE1M5+s1p1l/UjfNxs7SlHdeqE6fplWwMH1w4eCHUalU1HSzo6abHe+1qsH5uFTWntAlLM7EprL9XALbzyXw6coTNPGrSNsAd16q44qzbckYxSAKJjY5g7UnYlh7PJb9V29wb1mIIG9H2tZ1o01ddypXtDZekMLoevToQUJCAmPGjCE2Npbg4GDWrVunL34eERGR57eY6enpvPXWW0RGRmJlZYW/vz+LFi2iR48egO5C89ixY8yfP5+kpCQ8PDx46aWX+OKLL0rMSKjHuVvsPBlFUWTaihCi3DIxMXnoAhVCGJtKkapnTyQlJQUHBweSk5Oxt7c3djhClD9bvoZt30LlpjBgrbGjeazk29n8tPk88/67QrZGQa2C3o0r815oDSoWQZLoYkIaa4/HsOZ4LKdi7q5+pVZBY9+KtA10J6yOK5XsLA1+bvH0opNus+Z4DGtPxHLw6s08x+pXdqRtgDut67rh5SSJKGOQa4CCM+Z3lZmjoe7Y9WRrFHZ82BLvCvL3RQghhCguBb0GkJFSQojSJycLDszVbYeU7FFSORotS/Zf44eN57iRngVA8xoujG5XixpFWHS6qostw16ozrAXqnPlejpr7oy0OR6VzO5Liey+lMiY/50gxKeCPsHhai8JKmO6duMWa0/oEolHriXlOdawipP+z8nD0co4AQpRyliYmuDvZs/xqGSORiZJUkoIIYQogSQpJYQofU6vgvR4sHWDWu2NHc1DbTuXwFerT3EuLg2AapVs+bRdLVrWrFSscfg42/BWi2q81aIa127c0tWgOhHL0WtJ7L18g72XbzDun5M0rOJEm7rutAlww91BEh/FISLxFmtO6GqCHYtM1u9XqaCRTwXaBbgTVscNNwdJGArxJAK9HDgelcyxyGReDvQwdjhCCCGEuI8kpYQQpc++WbrnhgPApOTVbLsQn8qXq0+z9WwCAE7WZrzXqga9QipjZmKklQrv8K5gzf89X5X/e74qkTdvse5ODapDEUnsv3KT/VduMv7fU/opYm0C3PGUkTkGdfl6ur44fW6he7hnamWAG2F13WRqpRAGEOTlyOK9ERy9b/ShEEIIIUoGSUoJUViKohvGIIwj5ihc2wNqM2jQz9jR5HEjPYspm86xeG8EGq2CmYmKPk18eOeF6jhYl7zkmZeTNYOa+TGomR/RSbdZdyKWtSdiOHD1JocikjgUkcSXq0/ri2m3DXCX6S9P6EK8rsbX6jtF6HOZqFU08atImwA3wuq4SRF6IQws0NsBgBNRyWi0CiZq+f9bCCGEKEkkKSVEYdy4DPNeBrcA6LVEklPGkDtKqnYHsHM1bix3ZOVoWbD7ClPDz5OSkQNAq9qufNK2Fr7ONkaOrmA8HK0Y8JwvA57zJS4lg3UnYll9PIb9V25w9FoSR68lMWHtGQI8HWgb4E7bADeqVCwdn81YzsWl6oqVH4/lbNzdRJSpWkXTas60revGS3XcqGBjbsQohSjbqleyw9rchPQsDZcS0qhehLX8hBBCCFF4kpQSoqCyM2BZX0iJ1D0ubYGqLxg7qvLl1g04vky3HTLEuLEAiqKw8VQcX685zZXEWwDUcrfns3a1aFrN2cjRPTlXe0v6NvWhb1Mf4lMzWH8yjjXHYth7OZHjUckcj0rm23VnqO1uT7tAd9rUdcPPxdbYYRudoiicjUtlzTFdza4L8Wn6Y2YmKp6t5kzbAHdequ2Ko7UkooQoDiZqFXU9HNh35QZHI5MlKSWEEEKUMJKUEqKg1n+imzqWa9tE8Gspo6WK0+FFkJOhG6nmHWLUUE5GJ/Plv6fZfSkRAGdbC0aG1aBrA+8yNT2kkp0lrz9ThdefqcL1tEw2nIxjzfEYdl9K5FRMCqdiUvhu/Vn83ez0I6iqVSo/N32KonAqJkU/IurS9XT9MXMTNc2q6xJRobVcS+QUTiHKg0AvXVLqWGQSXRt4GTscIYQQQtxDklJCFMTxv+DAbEAFr0yD1SMgYjdc2Qm+zYwdXfmg1cD+33TbIUOMlgyMT83g+/Xn+PPgNRQFzE3VDG7my5stqmFrUbb/SXW2taB348r0blyZG+lZbDgZy5oTsfx34TpnYlM5E5vK5I3nqF7J9k6Cyp0arraoSnDiNluj5Xa2howsDbez7zyy7jzueZ2h39beec4hPUvDgSs39KPkQPfz8HwNF9oFuPNCrUrYW0oiSghjC/R2BODoPStcCiGEEKJkKNt3UEIYQsI5WPWObrv5B1D/dYg+rEtSbZ8oSanicn4jJF0FS0eo27XYT5+RrWH2zsv8vOUC6VkaAF4OdOej1v7lsvh3BRtzeoZUpmdIZZJuZbHhVBxrj8ew88J1zsen8WP4eX4MP09VFxt9gsrfza7ACSpFUcjM0d6TDHrwc4Z+W6t/fSsrh9tZ+d+b+/pW1t0kVI5WeervwsJUTcualWgT4MaLtVzLfHJSiNImyEtX7Px0dApZOVrMTY27CqoQQggh7pIrZyEeJeuWro5Udjr4NIMWo3T7n3sPDi2Ay9shYg9Ufsa4cZYH++8UOK//OpgXXxJIURT+PRbDN2vPEJV0G4Agb0fGvFyLBlUqFFscJZmjtTndG3rTvaE3ybey2XQ6jrUnYth+7joXE9KZtvkC0zZfwNfZhkY+TmRrlDwjkTIelHTK1qA8fb6owNQqsDY3xdLMBCtzNVZmJliZmdx5bYK1+Z3t3Med1z4VbWhR0wUbSUQJUWJVrmCNo7UZSbeyORubSsCdJJUQQgghjE+uooV4lDUjIf4U2FSCLrNBbaLb7+gNwb10ialtE+H1FcaNs6xLvAgXNgEqaDiw2E575FoSX/x7ioNXbwLg7mDJR639eSXIA3UZqhtlSA7WZnRp4EWXBl6kZGSz+XQ8a47HsPVcApevp3P5nppLBWVmosLSTJcYujdRpE8c3dkuyHHLe5JKd4+rMTdRl+hphkKIJ6dSqQjwdGDH+escjUySpJQQQghRgkhSSoiHObwIjiwClRq6zgE717zHnxsBhxfDxXCIPAheDYwTZ3mQW0uqRhhU8C3y00Un3WbiujOsPBINgJWZCW+2qMrgZn5YmZsU+fnLCntLMzrW86RjPU/SMnPYfCaeK9fT8ySKrO6MTMpNFlmbm+oTRbkJJjMTmWojhHg6QV6OuqTUtSRee6aKscMRQgghxB2SlBLiQWJPwOr3ddstP31w3agKvhDYA47+Dtu/g95LizfG8iIzTZf8AwgZXKSnSs/M4ZdtF/l1xyUysrUAdKnvxciwmrg5WBbpucs6WwtTXgnyMHYYQohyKvDO6KhjUuxcCCGEKFEkKSXE/TJTdXWkcjKgWqhuRNTDNHsfji2Fc2sh5ii4BxVfnOXF8T8hMxkq+IHfC0VyCq1WYfmhSL5bf5b41EwAQnwq8NnLtWWahxBClAFBd1bgOx+fyq2sHKzN5RJYCCGEKAlkToQQ91IU+OddSLwA9p7Q6VdQP+KviXM1qNNZt739u+KJsTxRFNh3p8B5o8GP/rN4Qvsu36DD9F2M/OsY8amZeFewYsar9fnj/56RhJQQQpQRrvaWuNpboFXgRFSKscMRQgghxB2SlBLiXgdmw4nloDaFrnPBpuLj39P8A93z6X8g7lTRxlfeXN2lKzRvZg3BvQ3adUTiLd5cdJDuv+zmeFQythamjGrjz6YRz9MmwF2KXgshRBkT6OUIwLHIJKPGIYQQQoi7JCklRK7ow7BulG479HOo3Lhg76tUC2p30G3vmFQ0sZVX+37VPQf2ACtHg3SZkpHNhDWnCZ28jbUnYlGroHfjymwd2YL/e74qFqZSyFwIIcqi4DtT+I5KXSkhhBCixJAJ9UIA3L4Jf/YFTRbUbAdNhhbu/c1Hwqn/wYkV8PzH4FKjaOIsT1Ki4fS/um0DFDjP0WhZuv8aP2w8R2J6FgDNqjvzabta+LvZP3X/QgghSra7xc6TjBuIEEIIIfSMPlJq+vTp+Pj4YGlpSePGjdm3b99D22ZnZzN+/HiqVq2KpaUlQUFBrFu3Lk+b1NRUhg8fTpUqVbCysqJp06bs378/T5t+/fqhUqnyPFq3bl0kn0+UAooCK4dC0lVwrAIdf4bCTt1yC4CabQEFdnxfJGGWOwfmgqKBKs+Ba52n6mrH+QTaTd3J6JUnSEzPws/Fhjn9GrJgQIgkpIQQopwI9HQE4GriLZJuZRk3GCGEEEIARk5K/fHHH4wYMYKxY8dy6NAhgoKCCAsLIz4+/oHtR48ezS+//MK0adM4deoUb7zxBp06deLw4cP6NoMGDWLjxo0sXLiQ48eP89JLLxEaGkpUVFSevlq3bk1MTIz+sWTJkiL9rKIE2z0dzq4GE3PoPv/Jp4k1H6l7Pr4MblwyWHjlUk4mHJyr236KUVIX4tMYMG8/r8/ex9m4VByszBjXvjbrhzfnBX9XqRslhBDliIO1GT4VrQE4JlP4hBBCiBLBqEmpyZMnM3jwYPr370/t2rWZOXMm1tbWzJkz54HtFy5cyCeffELbtm3x8/PjzTffpG3btnz/vW5kyu3bt1m+fDkTJ06kefPmVKtWjXHjxlGtWjVmzJiRpy8LCwvc3Nz0DycnpyL/vKIEitgLm8bqtltPAI96T96XZ32o1ko3umfHZMPEV16dWgXpCWDnAf7tCv32m+lZjFt1ktZTtrP5TDymahX9n/Vh28gW9HvWFzMTow8SFUIIYQRS7FwIIYQoWYxWUyorK4uDBw8yatQo/T61Wk1oaCi7d+9+4HsyMzOxtLTMs8/KyoqdO3cCkJOTg0ajeWSbXFu3bqVSpUo4OTnxwgsv8OWXX1Kx4sNXWsvMzCQzM1P/OiVFlhMu9dIT4a/+oM2Bul2g4cCn7/P5D+HCRji6RDdyyqnK0/dZHuUWOG/YH0zMUBSFHK1CVo5W99Dc95yjJfvO9qmYFKZtvkDy7WwAQmtVYlTbWlR1sTXiBxJCCFESBHo5sOpotBQ7F0IIIUoIoyWlrl+/jkajwdXVNc9+V1dXzpw588D3hIWFMXnyZJo3b07VqlUJDw9nxYoVaDQaAOzs7GjSpAlffPEFtWrVwtXVlSVLlrB7926qVaum76d169Z07twZX19fLl68yCeffEKbNm3YvXs3JiYPXnlrwoQJfP755wb69MLotFr4ewikREHFatD+x8LXkXoQ7xDwfR4ub4NdU+DlH56+z2J0b/InN8mTeSf5k31PAuhBiaF87XMUsjSae9or97TXkK25m2TK1GjJvnPMJ+scv2XsIxtTWm2uQvTGtWRrtChK4T6Lv5sdo9vV5rnqzkXzZQkhhCh1gu6swCcjpYQQQoiSoVStvvfjjz8yePBg/P39UalUVK1alf79++eZ7rdw4UIGDBiAp6cnJiYm1K9fn169enHw4EF9m549e+q3AwICCAwMpGrVqmzdupUXX3zxgeceNWoUI0aM0L9OSUnB29u7CD6lKBY7v4cLm8DUErovAAs7w/X9/Ie6pNThRbrRUvYehuvbQDRahb2XE1l7PJYtZ+NJvpWtSxY9QfLH0IaY/gOmsFoTwpVsW0Cbr41KBeYmasxN1ViYqjE3UWN259nW0pRuDbzp0cgbE7XUjBJClB/Tp0/nu+++IzY2lqCgIKZNm0ZISMgD265YsYKvv/6aCxcukJ2dTfXq1Xn//fd5/fXX9W0URWHs2LHMmjWLpKQknn32WWbMmEH16tWL6yMZXB0Pe9QqiEvJJDY5AzcHy8e/SQghhBBFxmhJKWdnZ0xMTIiLi8uzPy4uDjc3twe+x8XFhZUrV5KRkUFiYiIeHh58/PHH+Pn56dtUrVqVbdu2kZ6eTkpKCu7u7vTo0SNPm/v5+fnh7OzMhQsXHpqUsrCwwMLC4gk+qShxLm+HLV/rttt9/9Qru+Xj8xxUeRau7oJdP0Kbbw3b/xPK0WjZc+kGa07EsOFkLNfTHr/ykFoF5qZqzEzyJ3/MTdX5juW+zj2mb3dfe3NTNRYmasxMVZibmOiPWeUk0XD5HtBAvS4fsdmzof6YhYnJnfZqTKUmlBBC5JG7eMzMmTNp3LgxU6ZMISwsjLNnz1KpUqV87StUqMCnn36Kv78/5ubm/Pvvv/Tv359KlSoRFhYGwMSJE5k6dSrz58/H19eXzz77jLCwME6dOpWvVEJpYW1uSg1XO87EpnI0Mgk3hwdfcwohhBCieBgtKWVubk6DBg0IDw+nY8eOAGi1WsLDwxk2bNgj32tpaYmnpyfZ2dksX76c7t2752tjY2ODjY0NN2/eZP369UycOPGh/UVGRpKYmIi7u/tTfSZRCqTGwV8DQdFC8KtQ77WiOU/zkbBwFxycB8+NADvXx76lKGRrtOy+mMia4zGsPxnLzVvZ+mOO1ma8VNuVNnXdqVLRumQkf3b9DppMcA+iStDzhplSKYQQ5cC9i8cAzJw5k9WrVzNnzhw+/vjjfO1btGiR5/W7777L/Pnz2blzJ2FhYSiKwpQpUxg9ejQdOnQAYMGCBbi6urJy5co8o85Lm0AvB87EpnIsMomwOpKUEkIIIYzJqNP3RowYQd++fWnYsCEhISFMmTKF9PR0/QVVnz598PT0ZMKECQDs3buXqKgogoODiYqKYty4cWi1Wj788EN9n+vXr0dRFGrWrMmFCxcYOXIk/v7++j7T0tL4/PPP6dKlC25ubly8eJEPP/yQatWq6X8zKMoorQaWD4T0eKhUG9pOKrpz+bUAr0YQuR/+mwphXxXdue6TlaNl18XrrDkWw4ZTcfqC3wAVbMwJq6NLRDWpWrFkrUKn1cD+33TbIUMkISWEEAX0JIvH3EtRFDZv3szZs2f59lvd6N7Lly8TGxtLaGiovp2DgwONGzdm9+7dpTwp5cifByI5JsXOhRBCCKMzalKqR48eJCQkMGbMGGJjYwkODmbdunX64ucRERGo1XdvmjMyMhg9ejSXLl3C1taWtm3bsnDhQhwdHfVtkpOTGTVqFJGRkVSoUIEuXbrw1VdfYWZmBoCJiQnHjh1j/vz5JCUl4eHhwUsvvcQXX3wh0/PKuq0T4MoOMLfV1ZEyty66c6lU0PxD+L0bHJgDz70HNkVXcDszR8PO89dZczyWjadiScnI0R9ztjUnrI4bbQPcaexboeROfTu/AZIiwMpJtxqiEEKIAnmSxWNAd83k6elJZmYmJiYm/Pzzz7Rq1QqA2NhYfR/395l77EFKw2rFQV6OAByLTEZRFFTySxAhhBDCaIxe6HzYsGEPna63devWPK+ff/55Tp069cj+unfv/sDpfLmsrKxYv359oeMUpdz5TbD9O912+x/BuRiKtFZvBe7BEHMEdk+H0LEG7T4jW8P2cwmsPRHLplNxpGbeTUS52FnQ+k4iKsS3Quko+L3vV91zvdfBzMq4sQghRDlgZ2fHkSNHSEtLIzw8nBEjRuDn55dval9hlIbVimu62WFuqib5djZXE2/h42xj7JCEEEKIcsvoSSkhilxyJKwYrNtuOBACuhbPeVUq3Up8S3vrEi5N3wbrCk/VZUa2hq1n41lzPJbw03GkZ2n0x1ztLWhT1522Ae40qOJUOhJRua6fh4ubARU0GmjsaIQQolR5ksVjQDfFr1q1agAEBwdz+vRpJkyYQIsWLfTvi4uLy1NzMy4ujuDg4If2WRpWKzY3VVPb3Z4j15I4GpkkSSkhhBDCiCQpJco2TTYs6w+3b4B7EIR9Xbznr9kWXOtC3AnYOxNaflLoLm5l5bDlTAJrTsSw5Uw8t+5JRLk7WNKmrjvtAt2o5+2EujQlou6VW0uqRmtw8jFqKEIIUdo8zeIx99Jqtfqpd76+vri5uREeHq5PQqWkpLB3717efPPNh/ZRWlYrDvJy4Mi1JI5FJtMh2NPY4QghhBDlliSlRNm2aRxE7gMLB+g2H8yKeQlrlUq3Et+yvrBnJjQZCpYOj31bemYOm8/Es+Z4DFvOxpORrdUf83S0om2AG20C3An2ciy9iahcmalw5Hfddshg48YihBClVGEXj5kwYQINGzakatWqZGZmsmbNGhYuXMiMGTMAUKlUDB8+nC+//JLq1avj6+vLZ599hoeHhz7xVZoFejkCVzkWmWTkSIQQQojyTZJSouw6sxp2/6Tb7jgdKvgaJ45ar4CLPyScgb2/wvMjH9gsNSObzWfiWX0shm3nEsjMuZuI8q5gRdsAd9rWdSfQy6FsFWU99gdkpkDFauDX0tjRCCFEqVTYxWPS09N56623iIyMxMrKCn9/fxYtWkSPHj30bT788EPS09MZMmQISUlJPPfcc6xbtw5Ly2L+BU8RCPLW/YLoRFQKORptyV0ERAghhCjjVIqiKMYOojRKSUnBwcGB5ORk7O3tjR2OuN+Ny/DL85CZDE2GQdhXxo3n2DJYMUi3stzw42BhB0Dy7WzCT8ex5ngM289dJ0tzNxHlU9Fal4gKcKeOh33ZSkTlUhT4uQkknIbW38Izbxg7IiGEeCy5Bii4kvpdabUKgZ9vIC0zh3XDm+HvVnJiE0IIIcqCgl4DyEgpUfbkZMKyfrqElFcIhI4zdkRQtzNsnQA3LnL7v1/51647a47HsPPCdbI1d/PCfi42tAtwp01dd2q525XNRNS9ruzUJaTMbCC4l7GjEUIIUU6o1Srqetqz59INjl1LlqSUEEIIYSSSlBJlz/pPIOYIWFWAbnPBxMzYEXHztoZz3v1pfGM0t7ZO4bNMXzLQFYKtXsmWNgHutAtwp4arbdlPRN1r36+656CeBaq1JYQQQhhKkJcjey7d4EhkEt0blawVAoUQQojyQpJSomw5/tfdldw6/woOXkYLJTEtk/Un41h7Iob/Liai0lZms7kLldUJvOe4k4yGb9I2wI3qrnZGi9GokiN1db9ACpwLIYQodrpi50ixcyGEEMKIJCklyo7r5+Gfd3Xbzd6H6q2KPYSE1EzWnYxl7fEY9lxKRHtPxbZa7hW44DyEyue/4v9MV8PzXxf/aoAlyYG5oGjApxlUqmXsaIQQQpQzgV66EbpnYlLJyNZgaWZi5IiEEEKI8keSUqJsyLoFf/aFrDSo8hy0+KTYTh2fksG6k7GsPhbDvis3uHfpgLqe9rS9UyPK19kGchrD1PmQEgmHF5bfEUI5mXBwnm67vH4HQgghjMrLyYoKNubcSM/idEwK9So7GTskIYQQotyRpJQoG9aOhPiTYFMJus4Gk6L90Y5Jvs26E7GsOR7Dgas38ySigrwcaBPgTtu67lSuaJ33jabm8NxwWPMB7PwB6vcBU4sijbVEOrkSbl0He0+o2c7Y0QghhCiHVCoVgV4ObD2bwLHIZElKCSGEEEYgSSlR+h1eDIcXgUqtS0jZuRXZqW6mZ/HO0sPsOH89z/56lR1pW9ed1nXd8K5g/ZB35zZ+HbZPgpQoOPI7NOxfZPGWWLkFzhv2L/IEohBCCPEwQV6ObD2bwFGpKyWEEEIYhdwNitIt7iSsfl+33eIT8G1eZKdKzcimz5x9HI9KBqBhFSfaBLjTpq4bHo5WBe/IzBKefRfWj4Kdk6HeayVihcBiE3UIog6AiTnU72fsaIQQQpRjQd66ulLHIpONHIkQQghRPklSSpRemam6OlI5t6Hqi7ri5kXkdpaGgfMOcDwqmQo25vw+uDH+bvZP3mGDfrqEVFIEHPsT6r1qsFhLvNzVEet0AlsX48YihBCiXMtdge9iQhppmTnYWsilsRBCCFGc1MYOQIgnoii6lfYSz4OdB3T+FdRF8+OcmaNhyMID7LtyAztLUxYMCHm6hBSAuTU0fVu3vWMSaHKePtDSID0Rjv+l224kBc6FEEIYl7OtBZ6OVigKHJfRUkIIIUSxk6SUKJ0OzIETy0FlAt3mgo1zkZwmR6PlnSW6GlJWZibM69+Iup4Ohum84UCwqgA3LsHJFYbps6Q7vAA0meAeDF4NjR2NEEIIQaBX7hS+JOMGIoQQQpRDkpQSpU/0EVj3sW47dBxUfqZITqPVKoz86xjrT8ZhbqJmVp+GNKhSwXAnsLCFJm/ptrdPAq3WcH2XRFoN7J+t2w4ZAiqVceMRQgghuDuFT+pKCSGEEMVPklKidLmdBMv6giYLara7OwXOwBRF4bP/neDvw1GYqFVMf7U+z1UvgtFYIUPA0gGun4XT/zN8/yXJuXWQfE03OqxuZ2NHI4QQQgAQdGek1JFrScYNRAghhCiHJCklSg9Fgf8NhZtXwLEydJxeJKNtFEXhm7VnWLw3ApUKJncPolVtV4OfB9AlpBq/qdve9l3ZHi2171fdc/0+YFaI1QqFEEKIIlT3TlIqKuk2iWmZRo5GCCGEKF8kKSVKjz0/w5l/wcQcus0HK6ciOc1Pmy/wy/ZLAHzdKYAOwZ5Fch69Z94AczuIPwln1xTtuYwl4Rxc2goqNTQcYOxohBBCCD17SzP8XGwAmcInhBBCFDdJSonS4do+2DhGtx32NXjWL5LTzN55me83ngNgdLta9AqpXCTnycPKCULurES3faJuRFhZs/833XONNuBUxbixCCGEEPcJulNX6qgUOxdCCCGKlSSlRMmXngjL+oE2B+p0gkaDiuQ0f+yP4It/TwHwXmgNBjXzK5LzPFCTYWBmAzFH4fzG4jtvcchMhSO/67Zzk29CCCFECXJ3BT4ZKSWEEEIUJ0lKiZJNq4W//w9SoqBCVWg/tUjqSP1zNJqPVxwHYHAzX955sZrBz/FINhWh0Z1pbdu+LVujpY4uhaxUqFgd/FoYOxohhBAin7sr8CWhlKX/g4UQQogSTpJSomTb9QNc2AimltB9AVjaG/wUm07F8d4fR1AU6N24Mp+0rYWqCBJfj9Xkbd3njDoAl7YU//mLgqLAvlm67ZDBRZJQFEIIIZ5WHQ97TNUqrqdlEZ2cYexwhBBCiHJDklKi5Lq8AzZ/qdtuOwnc6hr8FLsuXOet3w+Ro1XoGOzBlx3qGichBWDnCg366ba3fWecGAzt8na4fhbMbSGol7GjEUIIIR7I0syEmm52ABy7lmTcYIQQQohyRJJSomRKjYPlA0HRQlBvqPeawU9x8OpNBi84QFaOlpdquzKpWxBqtZFH8jz7rm51wYj/4MpO48ZiCPt+1T0H9SySUW5CCCGEoQTqi51LXSkhhBCiuEhSSpQ8Wo0uIZUWBy61oN0kg0/7OhmdTL+5+7iVpaFZdWem9a6HqUkJ+Otg7wH1Xtdtb/vWuLE8raRrcHaNbruRFDgXQghRsgXpi50nGTcQIYQQohwpAXfhQtxn6zdwZYduNbru88HcxqDdX4hPo8/sfaRm5NDIx4lfXm+AhamJQc/xVJ4bDmpT3dS3iL3GjubJHZyrG+nm2xwq+Rs7GiGEEOKRckdKHY9MRquVYudCCCFEcZCklChZLmyC7XfqKbX/EVxqGrT7azdu8dpve0lMz6Kupz2z+zXC2tzUoOd4ao6V79Zf2j7RuLE8qewMODhPtx0yxKihCCGEEAVRw9UWSzM1qZk5XE5MN3Y4QgghRLkgSSlRciRHwYohgAIN+kNgN4N2H5ucQe/f9hCbkkH1SrYsGNAYe0szg57DYJqNAJWJLkkXedDY0RTeqZVwKxHsvaBGG2NHI4QQ5cL06dPx8fHB0tKSxo0bs2/fvoe2nTVrFs2aNcPJyQknJydCQ0Pzte/Xrx8qlSrPo3Xr1kX9MYzG1ERNHQ/dFL6jUuxcCCGEKBaSlBIlgyYb/uqvS2S4BULrbwzafWJaJq/N3su1G7epUtGaRYMaU8HG3KDnMKgKfhDYXbe9vRSuxJdb4LxhfzApYSPRhBCiDPrjjz8YMWIEY8eO5dChQwQFBREWFkZ8fPwD22/dupVevXqxZcsWdu/ejbe3Ny+99BJRUVF52rVu3ZqYmBj9Y8mSJcXxcYwmUF9XSoqdCyGEEMVBklKiZAj/HK7tBQt7XR0pM0uDdZ18O5s+c/ZxIT4NdwdLFg1sjKu94fovMs3eB1Rwbi3EHDN2NAUXeRCiDupWEazf19jRCCFEuTB58mQGDx5M//79qV27NjNnzsTa2po5c+Y8sP3ixYt56623CA4Oxt/fn99++w2tVkt4eHiedhYWFri5uekfTk5OxfFxjCZIvwJfklHjEEIIIcoLSUoJ4zuzGv6bptvuMF03SshAbmXlMGDefk5Gp1DRxpxFgxrjXcHaYP0XKefqULezbrs0jZbaP0v3XKcz2LoYNxYhhCgHsrKyOHjwIKGhofp9arWa0NBQdu/eXaA+bt26RXZ2NhUqVMizf+vWrVSqVImaNWvy5ptvkpiY+Mh+MjMzSUlJyfMoTXJHSp2KTiFbozVyNEIIIUTZJ0kpYVw3r8DKN3Xbz7wFtV8xWNcZ2RqGLDjIwas3sbc0ZeHAxlR1sTVY/8Wi+Ujd8+lVEHfKuLEURPp1OLFcty0FzoUQolhcv34djUaDq6trnv2urq7ExsYWqI+PPvoIDw+PPImt1q1bs2DBAsLDw/n222/Ztm0bbdq0QaPRPLSfCRMm4ODgoH94e3s/2YcyEp+KNthZmpKZo+VsbKqxwxFCCCHKPElKCePJyYRl/SAjGTwbQujnBus6W6Nl2O+H2XnhOtbmJswbEEJtD3uD9V9sKtWCWncSdTsmGTeWgjg0HzRZ4FEfvBoYOxohhBAF8M0337B06VL+/vtvLC3vTm/v2bMnr7zyCgEBAXTs2JF///2X/fv3s3Xr1of2NWrUKJKTk/WPa9euFcMnMBy1WiV1pYQQQohiJEkpYTwbRkP0YbBygm7zwNQwhcc1WoX3/zzKptNxmJuq+a1vQ+pXLsU1MHJHS51YAdfPGzeWR9HkwIG5um0ZJSWEEMXG2dkZExMT4uLi8uyPi4vDzc3tke+dNGkS33zzDRs2bCAwMPCRbf38/HB2dubChQsPbWNhYYG9vX2eR2mTW1fqmNSVEkIIIYqcJKWEcZxYfneFtk6/gKNhhvcrisLolcdZdTQaU7WKma/Vp2lVZ4P0bTTugVCzLaDAju+NHc3DnVsHydfAuiLU6WTsaIQQotwwNzenQYMGeYqU5xYtb9KkyUPfN3HiRL744gvWrVtHw4YNH3ueyMhIEhMTcXd3N0jcJVWgvti5jJQSQgghilqhk1I+Pj6MHz+eiIiIoohHlAfXL8Cqd3Tbz70HNcIM0q2iKHy1+jRL9l1DrYIpPYN5wd/18W8sDXJHSx37E25cMm4sD5ObZKzfx6CrJwohhHi8ESNGMGvWLObPn8/p06d58803SU9Pp3///gD06dOHUaNG6dt/++23fPbZZ8yZMwcfHx9iY2OJjY0lLS0NgLS0NEaOHMmePXu4cuUK4eHhdOjQgWrVqhEWZpj/t0uqIG/d9L1zcancznp4/SwhhBBCPL1CJ6WGDx/OihUr8PPzo1WrVixdupTMzMyiiE2URdm34c8+kJUGVZ6FlqMN1vWUTef5bedlAL7pHMjLgR4G69voPOtDtVBQNLBjsrGjyS/hLFzeBio1NBxg7GiEEKLc6dGjB5MmTWLMmDEEBwdz5MgR1q1bpy9+HhERQUxMjL79jBkzyMrKomvXrri7u+sfkybp6heamJhw7NgxXnnlFWrUqMHAgQNp0KABO3bswMLCwiifsbi42VviYmeBRqtwKkZGSwkhhBBFSaUoivIkbzx06BDz5s1jyZIlaDQaevfuzYABA6hfv76hYyyRUlJScHBwIDk5uVTWSzCa/w2DwwvBxgX+bwfYG2YKwKztl/hqzWkAxravTf9nfQ3Sb4kSsRfmvARqU3jnMDhWNnZEd63+APbPAv+XoediY0cjhBBFSq4BCq60fleD5u9n0+l4xrxcmwHPlcFrCiGEEKKIFfQa4IlrStWvX5+pU6cSHR3N2LFj+e2332jUqBHBwcHMmTOHJ8x1ibLsyO+6hBQq6PKbwRJSv++N0CekPnipRtlMSAFUbgy+z4M2B3b+YOxo7spIgaNLdNshg40bixBCCGEAgVLsXAghhCgWT5yUys7O5s8//+SVV17h/fffp2HDhvz222906dKFTz75hFdffdWQcYrSLu4U/DtCt91iFPi1MEi3Kw9H8enK4wC88XxVhrasZpB+S6znP9Q9H14EKdHGjSXX0aW66ZjONXRJMyGEEKKUC/TS1ZWSYudCCCFE0TIt7BsOHTrE3LlzWbJkCWq1mj59+vDDDz/g7++vb9OpUycaNWpk0EBFKZaZBsv6Qs5t8GsJzT8wSLcbTsby/rKjKAq8/kwVPmpdE5VKZZC+Syyf56ByU4j4D3b9CG2+NW48iqKbtgcQMgTK+vcvhBCiXMgdKXX5ejrJt7NxsDIzbkBCCCFEGVXokVKNGjXi/PnzzJgxg6ioKCZNmpQnIQXg6+tLz549DRakKMW0Wlj1Nlw/B3bu0HkWqE2eutsd5xMY9vthNFqFzvU9+fyVOmU/IZXr+Tsr8R2cB6lxRg2Fy9t0f7bmdhAkf+eFEEKUDRVszPGuYAXAcRktJYQQQhSZQielLl26xLp16+jWrRtmZg/+rZGNjQ1z58596uBEKZebkDq5AlQm0HUO2Lo8dbf7r9xgyIKDZGm0tKnrxsQugajV5SQhBbrRZl6NICcD/ptq3Fj23RklFdwLLOyMG4sQQghhQLmjpY5KXSkhhBCiyBQ6KRUfH8/evXvz7d+7dy8HDhwwSFCiDNBq4H9D4cgiUKmh0y9QpelTd3s8MpkBc/dzO1vD8zVcmNIzGFOTJy6NVjqpVND8Tm2pA3Mg/bpx4kiKgLNrdNuNBhknBiGEEKKIBN2pKyXFzoUQQoiiU+i7+aFDh3Lt2rV8+6Oiohg6dKhBghKlnFYDK9+Co7/rRkh1+Q0Cuz11t+fjUukzZy+pmTmE+FZg5msNsDB9+qmApVL1VuAeDNm3YPd048RwYA4oWl1xc5eaxolBCCGEKCJ3V+CT6XtCCCFEUSl0UurUqVPUr18/3/569epx6tQpgwQlSjFNDvz9f3Bs6Z0pe7Ohbpen7vZqYjqv/raXm7eyCfJyYHbfhliZl9OEFNwZLXWnttS+WXDrRvGePzsDDs7XbYcMKd5zCyGEEMUgwNMBtQpikjOIT80wdjhCCCFEmVTopJSFhQVxcfmLK8fExGBqWujF/ERZosmBv4fA8WWgNoVu86BOp6fuNib5Nr1n7SU+NZOarnbM6x+CnaWsgkPNtuBaF7JSYe/M4j33yRVw+wY4eEON1sV7biGEEKIY2FiYUq2SLQDHrsloKSGEEKIoFDop9dJLLzFq1CiSk+/+55yUlMQnn3xCq1atDBqcKEU02bB8IJxYDmoz6DYfar/y1N1eT8vk1d/2EpV0G5+K1iwcFIKTjbkBAi4D1Gpo/oFue89MyCimC2ZFgb2/6LYbDgATSUYLIYQom+5O4UsyahxCCCFEWVXopNSkSZO4du0aVapUoWXLlrRs2RJfX19iY2P5/vvviyJGUdJpsuGvAXBqpS4h1WMh1Hr5qbtNvpXN67P3cSkhHQ8HSxYNakwlO8unj7csqdUBnGtCZjLs+7V4zhl1EGKOgIkF1O9TPOcUQgghjCC32PlRqSslhBBCFIlCJ6U8PT05duwYEydOpHbt2jRo0IAff/yR48eP4+3tXRQxipIsJwuW9YPTq8DEHHouhpptnrrbtMwc+s3bx+mYFJxtLVg8+Bm8nKyfPt6yRq2+W1tq93TITC36c+6bpXuu2wVsnIv+fEIIIYSR3DtSSlEU4wYjhBBClEFPNO/GxsaGIUOkuHG5l5MFy/rC2TW6UTM9F+tWhXtKGdkaBs8/wOGIJByszFg0KARfZxsDBFxG1e0MWyfAjYuwfzY8N7zozpWWoKsnBRAyqOjOI4QQQpQA/u52mJmouHkrm8ibt/GuIL8gE0IIIQzpiYvBnDp1ioiICLKysvLsf+WVp68jJEqBnEz4sw+cW6dLSPX6HaqFPnW32RotQxcfYvelRGzMTZg/IAR/N3sDBFyGqU2g2fvwv7dg90+61fDMi+ii+dB80GSBZwPdQwghhCjDLExNqOVuz7HIZI5cS5KklBBCCGFghZ6+d+nSJYKCgqhbty7t2rWjY8eOdOzYkU6dOtGpU+FXWps+fTo+Pj5YWlrSuHFj9u3b99C22dnZjB8/nqpVq2JpaUlQUBDr1q3L0yY1NZXhw4dTpUoVrKysaNq0Kfv378/TRlEUxowZg7u7O1ZWVoSGhnL+/PlCx15uZWfAH6/pElKmltB7qUESUhqtwnt/HCH8TDwWpmpm92tEsLfj08dbHgR2B8fKkJ4AB+cVzTk0OXBgjm47REZKCiGEKB8C79SVkmLnQgghhOEVOin17rvv4uvrS3x8PNbW1pw8eZLt27fTsGFDtm7dWqi+/vjjD0aMGMHYsWM5dOgQQUFBhIWFER8f/8D2o0eP5pdffmHatGmcOnWKN954g06dOnH48GF9m0GDBrFx40YWLlzI8ePHeemllwgNDSUqKkrfZuLEiUydOpWZM2eyd+9ebGxsCAsLIyMjo7BfR/mTnQF/vArnN4CpFfT+A6q+8NTdarUKo1Yc499jMZiZqPjl9QY841fRAAGXEyZmutFSALt+1P05GdrZNZASBdbOULuj4fsXQohy6Nq1a0RGRupf79u3j+HDh/Prr8W0eIV4rNy6UlLsXAghhDC8Qieldu/ezfjx43F2dkatVqNWq3nuueeYMGEC77zzTqH6mjx5MoMHD6Z///7Url2bmTNnYm1tzZw5cx7YfuHChXzyySe0bdsWPz8/3nzzTdq2batf9e/27dssX76ciRMn0rx5c6pVq8a4ceOoVq0aM2bMAHSjpKZMmcLo0aPp0KEDgYGBLFiwgOjoaFauXFnYr6N8yb4NS3rChU1gZg2v/gl+LZ66W0VRGP/vKf48EIlaBVN71qNFzUpPH295E9Qb7L0gLRYOLzR8/7mr+zXoC2ayCqIQQhhC79692bJlCwCxsbG0atWKffv28emnnzJ+/HgjRycAgu4kpU5EJaPRSrFzIYQQwpAKnZTSaDTY2dkB4OzsTHR0NABVqlTh7NmzBe4nKyuLgwcPEhp6d9qXWq0mNDSU3bt3P/A9mZmZWFrmvRm2srJi586dAOTk5KDRaB7Z5vLly8TGxuY5r4ODA40bN37oeXPPnZKSkudRrmTdgt97wKUtYGYDry4D3+YG6XryxnPM++8KABO7BtEmwN0g/ZY7puZ3i5zvnKIrRG8o8afhyg5QqaHhAMP1K4QQ5dyJEycICQkB4M8//6Ru3br8999/LF68mHnz5hk3OAFAtUq2WJubcCtLw8WENGOHI4QQQpQphU5K1a1bl6NHjwLQuHFjJk6cyK5duxg/fjx+fn4F7uf69etoNBpcXV3z7Hd1dSU2NvaB7wkLC2Py5MmcP38erVbLxo0bWbFiBTExMQDY2dnRpEkTvvjiC6Kjo9FoNCxatIjdu3fr2+T2XZjzAkyYMAEHBwf9w9vbu8CftdTLSoffu8PlbWBuC6/9BT7PGaTrmdsuMm3zBQC+6FCHrg28DNJvuVXvdbB1g5RIOPq74frd/5vu2b8dOMifkRBCGEp2djYWFhYAbNq0Sb9gjL+/v/7aRRiXiVpFXQ9dXamj15KMG4wQQghRxhQ6KTV69Gi0Wi0A48eP5/LlyzRr1ow1a9YwdepUgwd4rx9//JHq1avj7++Pubk5w4YNo3///qjVdz/GwoULURQFT09PLCwsmDp1Kr169crT5kmMGjWK5ORk/ePatWtP+3FKh6x0WNxdN0rG3A5eWw5Vmhqk64W7r/DN2jMAfNTan9eb+Bik33LNzBKefVe3veN70GQ/fZ8ZyXBkiW5bCpwLIYRB1alTh5kzZ7Jjxw42btxI69atAYiOjqZiRamtWFIEeecWO5e6UkIIIYQhFTpTExYWRufOnQGoVq0aZ86c4fr168THx/PCCwUveO3s7IyJiQlxcXF59sfFxeHm5vbA97i4uLBy5UrS09O5evUqZ86cwdbWNs8IrapVq7Jt2zbS0tK4du0a+/btIzs7W98mt+/CnBfAwsICe3v7PI8yLzMNFnWFqzt1CanXV0DlZwzS9YpDkXz2v5MADG1ZlTdbVDVIvwJo0A9sXCApAo79+fT9HV0K2eng4g8+zZ6+PyGEEHrffvstv/zyCy1atKBXr14EBQUBsGrVKv20PmF8ucXOZQU+IYQQwrAKlZTKzs7G1NSUEydO5NlfoUIFVCpVoU5sbm5OgwYNCA8P1+/TarWEh4fTpEmTR77X0tIST09PcnJyWL58OR06dMjXxsbGBnd3d27evMn69ev1bXx9fXFzc8tz3pSUFPbu3fvY85YrmamwqAtE/AcW9vD63+BtmIvjdSdi+GCZbgpov6Y+fPBSTYP0K+4wt4Ymw3TbO74HrebJ+9Jq7xY4bzQICvn3XAghxKO1aNGC69evc/369TwLvQwZMoSZM2caMTJxr9xi56djUsnK0Ro3GCGEEKIMKVRSyszMjMqVK6PRPMVN7j1GjBjBrFmzmD9/PqdPn+bNN98kPT2d/v37A9CnTx9GjRqlb793715WrFjBpUuX2LFjB61bt0ar1fLhhx/q26xfv55169Zx+fJlNm7cSMuWLfH399f3qVKpGD58OF9++SWrVq3i+PHj9OnTBw8PDzp27GiQz1XqZaTAws5wbQ9YOMDrK8G7kUG63nYugbeXHEarQLcGXox5uXahE5qiABoNBCsnuHERTqx48n4ub4XEC7qRckE9DRaeEEIIndu3b5OZmYmTkxMAV69eZcqUKZw9e5ZKlWQl2pLCu4IVTtZmZGm0nIktZ4vdCCGEEEWo0NP3Pv30Uz755BNu3Ljx1Cfv0aMHkyZNYsyYMQQHB3PkyBHWrVunL0IeERGRp8hnRkYGo0ePpnbt2nTq1AlPT0927tyJo6Ojvk1ycjJDhw7F39+fPn368Nxzz7F+/XrMzMz0bT788EPefvtthgwZQqNGjUhLS2PdunX5Vu0rlzKSYVFniNwHlg7QZyV4NTBI13svJfJ/Cw+QrVFoF+DON10CUaslIVUkLOygyVDd9vbvdCOensS+Wbrn4N66PoUQQhhUhw4dWLBgAQBJSUk0btyY77//no4dOzJjxoxC9TV9+nR8fHywtLSkcePG7Nu376FtZ82aRbNmzXBycsLJyYnQ0NB87RVFYcyYMbi7u2NlZUVoaCjnz58v/IcsA1QqFQF3RksdlbpSQgghhMGoFEVRCvOGevXqceHCBbKzs6lSpQo2NjZ5jh86dMigAZZUKSkpODg4kJycXHbqS91O0iWkog6CpSP0+R94BBuk66PXknj1t72kZebwgn8lZr7WAHPTpys+Lx4jIxmmBOieu82DOp0K9/6bV+HHIECBofvBpUZRRCmEEKWOIa8BnJ2d2bZtG3Xq1OG3335j2rRpHD58mOXLlzNmzBhOnz5doH7++OMP+vTpw8yZM2ncuDFTpkxh2bJlDx1x9eqrr/Lss8/StGlTLC0t+fbbb/n77785efIknp6egK7e1YQJE5g/fz6+vr589tlnHD9+nFOnThX4F3ll6Xrp+w1nmbb5Al0beDGpW5CxwxFCCCFKtIJeA5gWtmOZ4lZG3b4JCztB9GGwqqBLSLkHGqTrs7Gp9J27j7TMHJr4VeTnV+tLQqo4WDpA4zdg27ewfRLU6gCFWYXywGxAAb+WkpASQogicuvWLezsdCNRN2zYQOfOnVGr1TzzzDNcvXq1wP1MnjyZwYMH68sVzJw5k9WrVzNnzhw+/vjjfO0XL16c5/Vvv/3G8uXLCQ8Pp0+fPiiKwpQpUxg9erS+LueCBQtwdXVl5cqV9OxZ/qZ0S7FzIYQQwvAKnZQaO3ZsUcQhjOnWDVjYEWKOgnVF6LMK3OoapOtsjZYB8/aTdCubYG9HZvVtiKWZiUH6FgXQ+A3Y/TPEnYCza6DWywV7X/ZtOLRQtx0ypOjiE0KIcq5atWqsXLmSTp06sX79et577z0A4uPjCzyyKCsri4MHD+apw6lWqwkNDWX37t0F6uPWrVtkZ2dToUIFAC5fvkxsbCyhoaH6Ng4ODjRu3Jjdu3eXy6RUkJcDABfi00jPzMHGotCX0UIIIYS4jwxXKe9u3YAFr9xJSDlD338NlpAC2Ho2gaik2zjbmjOvfyNs5QKueFlXgJDBuu3tE6Ggs3VPrIDbN8ChMtQIK7r4hBCinBszZgwffPABPj4+hISE6FcC3rBhA/Xq1StQH9evX0ej0ehrcuZydXUlNja2QH189NFHeHh46JNQue8rbJ+ZmZmkpKTkeZQVlewtcbO3RKvAiSipKyWEEEIYQqGTUmq1GhMTk4c+RCmSngjzX4HY42DjAv3+BdfaBj3FXwevAdC5vheO1uYG7VsUUJOhYGatSzye3/j49ooC+37RbTcaAGr5ey2EEEWla9euREREcODAAdavX6/f/+KLL/LDDz8USwzffPMNS5cu5e+//37qRV8mTJiAg4OD/uHt7W2gKEuGwDujpY5JsXMhhBDCIAo9bOXvv//O8zo7O5vDhw8zf/58Pv/8c4MFJopY+nVdQir+JNhU0iWkXGoa9BSJaZmEn44HoEt9L4P2LQrBxhkaDoDdP+nqS1VvBapHrHoYeUCXwDKxgHp9ii9OIYQop9zc3HBzcyMyMhIALy8vQkJCCvx+Z2dnTExMiIuLy7M/Li4ONze3R7530qRJfPPNN2zatInAwLu1JHPfFxcXh7u7e54+g4ODH9rfqFGjGDFihP51SkpKmUpMBXk7suFUHEelrpQQQghhEIUeKdWhQ4c8j65du/LVV18xceJEVq1aVRQxCkNLi4d5L+sSUrZu0G+1wRNSAP87Ek2OViHQy4GabnYG718UQtN3wNQSog7ApS2PbrvvV91zQFewqVj0sQkhRDmm1WoZP348Dg4OVKlShSpVquDo6MgXX3yBVqstUB/m5uY0aNCA8PDwPP2Gh4frpwM+yMSJE/niiy9Yt24dDRs2zHPM19cXNze3PH2mpKSwd+/eR/ZpYWGBvb19nkdZIiOlhBBCCMMyWE2pZ555Js+FiyihUuN0CamE02DnfichVTQrq/11UPcb364NZJSU0dm5QoN+uu1t3z28XVo8nLwzGjK3FpUQQogi8+mnn/LTTz/xzTffcPjwYQ4fPszXX3/NtGnT+Oyzzwrcz4gRI5g1axbz58/n9OnTvPnmm6Snp+tX4+vTp0+eQujffvstn332GXPmzMHHx4fY2FhiY2NJS0sDQKVSMXz4cL788ktWrVrF8ePH6dOnDx4eHuV6JeZAT0cAIm7c4mZ6lnGDEUIIIcoAg1Sdvn37NlOnTsXT09MQ3YmikhoL89vD9XNg56GbslexapGc6mR0MqdiUjA3UfNKkEeRnEMU0rPvwoE5EPEfXNkJPs/lb3NwPmizwasReBSswK4QQognN3/+fH777TdeeeUV/b7AwEA8PT156623+OqrrwrUT48ePUhISGDMmDHExsYSHBzMunXr9IXKIyIiUKvv/i5yxowZZGVl0bVr1zz9jB07lnHjxgHw4Ycfkp6ezpAhQ0hKSuK5555j3bp1T113qjRzsDbD19mGy9fTORaVzPM1XIwdkhBCCFGqFTop5eTkhOqeejSKopCamoq1tTWLFi0yaHDCgFKidQmpxAtg7wX9/oEKfkV2uuUHowBoVdtVCpyXFPYeUO81XWJq28T8SSlNju4YQMiQ4o9PCCHKoRs3buDv759vv7+/Pzdu3ChUX8OGDWPYsGEPPLZ169Y8r69cufLY/lQqFePHj2f8+PGFiqOsC/Ry0CWlriVJUkoIIYR4SoVOSv3www95klJqtRoXFxcaN26Mk5OTQYMTBpIcBfNfhhuXwMEb+v4DFXyL7HRZOVpWHtElpWTqXgnz3HtwaAFc3gYRe6Fy47vHzq6G1GjdSoy1OxgvRiGEKEeCgoL46aefmDp1ap79P/30U57C46LkCPRy5H9HojkqdaWEEEKIp1bopFS/fv2KIAxRZJIjdTWkbl4Gx8rQ919wqlKkp9x6Np4b6Vm42FnQrLpzkZ5LFJJjZQjqBYcXwvaJ8Nryu8f2zdI91+8LphbGiU8IIcqZiRMn0q5dOzZt2qQvIL57926uXbvGmjVrjBydeJCgO8XOj0YmoShKnl/WCiGEEKJwCl3ofO7cuSxbtizf/mXLljF//nyDBCUMJOkazGt3JyFVRVfUvIgTUnC3wHnnep6Ymhislr4wlGYjQGUCFzZB1EHdvrhTcGWHbn/D/saNTwghypHnn3+ec+fO0alTJ5KSkkhKSqJz586cPHmShQsXGjs88QB1PBwwUatISM0kNiXD2OEIIYQQpVqhMwYTJkzA2Tn/6JdKlSrx9ddfGyQoYQA3r8K8tnDzCjj5Qv81ulEyRex6Wiabz8QD0EWm7pVMFfwgoJtuO3clvv13Rkn5twMH+XMTQoji5OHhwVdffcXy5ctZvnw5X375JTdv3mT27NnGDk08gJW5CdUr2QJw9JpM4RNCCCGeRqGTUhEREfj65q9HVKVKFSIiIgwSlHhKN6/oRkglRegSEP1WF1ui4X9HosnRKgR5OVDD1a5YzimeQPMPABWcW6tbie/oUt1+KXAuhBBCPFaQlyMAxyKTjBqHEEIIUdoVOilVqVIljh07lm//0aNHqVixokGCEk/hxiWY2w6Sr0HFancSUp7FdvrcqXtS4LyEc64OdTvrtpf2huxb4FIr/4p8QgghhMgn0FtXV+qYFDsXQgghnkqhk1K9evXinXfeYcuWLWg0GjQaDZs3b+bdd9+lZ8+eRRGjKKjEi7qi5in/396dx0dV3/sff81MkpnsLCEbBBMCgiAJewRttUoNoFQUF/xxC9JetChWGpcLlkWrFlcuVSxcuS5URXEDra14bSoqAoIg4MKiILJmA5LJvsyc3x+TTDKQQAIhZ5K8n4/HeczMme858zmZtH755PP9nIPQuZcnIRUR32If/+3hAnYccRJkszI2teU+V87Qz+7xPJZVT6iHTQU1axURETmtupVShmGYG4yIiEgr1uS77z300EPs27ePK664goAAz+Fut5tJkyapp5SZ8n6AZVdD4RGI6g2T/w7hMS0aQk2V1C/7xdAhJKhFP1vOQExfuGAs7Pg72CMg5SazIxIRaTeuu+66U76fn5/fMoHIGekdG05QgBVnWRX7jpaQFBVqdkgiIiKtUpOTUkFBQaxYsYKHH36YrVu3EhwcTP/+/TnvvHN/VzdpQN73ngqpoizPEqzJ70FYdIuGUFHl5t2thwEt3WtVrpjnqbAbfAvYw8yORkSk3YiMjDzt+5MmTWqhaKSpAm1W+sZFsPVAPtsP5ispJSIicoaanJSq0atXL3r16tWcsciZyN0Fy8ZCUTZE94NJ70JYlxYP4+NdORwrriA63M7Pep58d0bxU1G94Pb1ZkchItLuvPjii2aHIGdpQEIHth7IZ9uBAq4Z0HL9O0VERNqSJveUGj9+PI899thJ+x9//HFuuOGGZglKGilnZ3WFVDbEXOhZsmdCQgpql+5dO6grAbYm/1qJiIiItCop3WqaneebG4iIiEgr1uTswaeffsqYMWNO2j969Gg+/fTTZglKGiH7O3jpKijOgdj+noRUqDl3P8wrKufjnTkAXD9IS/dERESk7Uupbnb+zeECqlxuc4MRERFppZqclCoqKiIo6OQm1oGBgTidzmYJSk4j6xtPU/OSPIhLhUnvQUgn08JZ9dUhqtwGqQkd6BUTblocIiIiIi2lR1Qo4fYAyirdfJ9TZHY4IiIirVKTk1L9+/dnxYoVJ+1//fXX6du3b7MEJadwZLunh1TJUYgb4OkhZWJCyjAM79I9NTgXERGR9sJqtXBhVy3hExERORtNbnQ+Z84crrvuOvbs2cPll18OQGZmJsuXL+ett95q9gCljsNb4eVxUHoc4gfBr1dCcAdTQ/r2sJOdWYUE2az8KiXe1FhEREREWlJKQiTr9x5l64ECbhpqdjQiIiKtT5OTUmPHjmXVqlX8+c9/5q233iI4OJjU1FT+/e9/06mTeRU7bd7hr+Bv10BZAXQdAr9+Bxynvp10S6ipkvplvxgiQwJNjkZERESk5aRW95VSpZSIiMiZOaPbpF111VV8/vnnFBcXs3fvXm688UbuueceUlNTmzs+ATi0uTYh1W2Yp0LKDxJSFVVu3t16CIAbtHRPRERE2pmaO/DtyiqkrNJlcjQiIiKtzxklpcBzF77JkycTHx/PU089xeWXX86GDRuaMzYBOPgl/O1aT0Iq4aLqCqkIs6MC4N87czheUklMhJ2f9epidjgiIiIiLaprh2A6hwZR5Tb47ohu+CMiItJUTVq+l5WVxUsvvcTzzz+P0+nkxhtvpLy8nFWrVqnJ+blwYCO8Mh7KndB9BEx8A+z+c3e7mqV71w7shs1qMTkaERERkZZlsVhI6RbJx7ty2X4gn0HdO5odkoiISKvS6EqpsWPH0rt3b7Zv387ChQs5fPgwzzzzzLmMrX3b/wW8fJ0nIXXeJTDxTb9KSOUWlvPxrhwArh/c1eRoRERERMyR4u0rVWBuICIiIq1QoyulPvjgA37/+98zbdo0evXqdS5jkp/Ww6vXQ0URJP4M/t8KCAo1Oyof7249hMttMCChAz2j/SdZJiIiItKSUhM8faW2qdm5iIhIkzW6Umrt2rUUFhYyePBg0tLSWLRoEXl5eecytvap9Dgsv8mTkOpxGfy/N/wuIWUYBm9+6Vm6d70anIuIiEg7VlMptTevmMKySnODERERaWUanZS66KKLWLp0KUeOHOG2227j9ddfJz4+HrfbzUcffURhYeG5jLP9CO4IYxdCz1/Cza9DUIjZEZ3k28NOdmUXEhRgZWxKvNnhiIiIiJgmKsxO1w7BGAZ8fUhL+ERERJqiyXffCw0N5Te/+Q1r167l66+/5u677+bRRx8lOjqaX/3qV+cixvbnwus8PaQCg82OpF41Dc6v7BtDZEigydGIiIiImKtmCZ/6SomIiDRNk5NSdfXu3ZvHH3+cgwcP8tprrzVXTAJg8c+72ZVXuVi19RAANwxJMDkaERERqfHss8+SmJiIw+EgLS2NjRs3Njj222+/Zfz48SQmJmKxWFi4cOFJYx544AEsFovP1qdPn3N4Ba1XbbPzfFPjEBERaW3OKilVw2azMW7cON57773mOJ34sY935pBfUklMhJ1LekaZHY6IiIgAK1asICMjg3nz5rFlyxZSU1NJT08nJyen3vElJSX06NGDRx99lNjY2AbP269fP44cOeLd1q5de64uoVVL6Vbd7PyAKqVERESaolmSUtJ+1Czdu25QN2xW/6zmEhERaW8WLFjA1KlTmTJlCn379mXJkiWEhITwwgsv1Dt+6NChPPHEE0yYMAG73d7geQMCAoiNjfVuUVH6g1R9+neNxGKBQ/ml5BWVmx2OiIhIq6GklDRabmE5H+/KBWD8IN11T0RExB9UVFSwefNmRo4c6d1ntVoZOXIk69evP6tzf//998THx9OjRw8mTpzI/v37Tzm+vLwcp9Pps7UH4Y5AekR57pasJXwiIiKNp6SUNNqqrw7hchsM7N6BntFhZocjIiIiQF5eHi6Xi5iYGJ/9MTExZGVlnfF509LSeOmll1i9ejWLFy/mxx9/5Gc/+9kp77g8f/58IiMjvVtCQvvpP5la3VdKS/hEREQaT0kpaRTDMLxL964frCopERGRtm706NHccMMNpKSkkJ6ezj//+U/y8/N54403Gjxm1qxZFBQUeLcDBw60YMTmqukrpUopERGRxgswOwBpHb455GRXdiFBAVauTok3OxwRERGpFhUVhc1mIzs722d/dnb2KZuYN1WHDh04//zz+eGHHxocY7fbT9mjqi1LSegAwPaDBRiGgcVP76QsIiLiT1QpJY3y1mbPXzrT+8USGRxocjQiIiJSIygoiMGDB5OZmend53a7yczMZPjw4c32OUVFRezZs4e4uLhmO2db0jcuggCrhaPFFRzKLzU7HBERkVZBSSk5rfIqF+9uOwzADVq6JyIi4ncyMjJYunQpy5YtY8eOHUybNo3i4mKmTJkCwKRJk5g1a5Z3fEVFBVu3bmXr1q1UVFRw6NAhtm7d6lMFdc899/DJJ5+wb98+1q1bx7XXXovNZuPmm29u8etrDRyBNnrHhgOeaikRERE5PS3fk9P6944c8ksqiY1wcHFP3QpaRETE39x0003k5uYyd+5csrKyGDBgAKtXr/Y2P9+/fz9Wa+3fIg8fPszAgQO9r5988kmefPJJLr30UtasWQPAwYMHufnmmzl69ChdunThkksuYcOGDXTp0qVFr601SU3owLeHnWw7mM+Y/qooExEROR0lpeS0ahqcXzeoKzar+iOIiIj4o+nTpzN9+vR636tJNNVITEzEMIxTnu/1119vrtDajdRukSz/ArbrDnwiIiKNouV7cko5hWWs2Z0LwHgt3RMRERFpUEq3DgB8c6gAt/vUST8RERFRUkpO492vDuNyGwzq3oHkLmFmhyMiIiLit3pFh+EItFJYXsXevGKzwxEREfF7SkpJgwzD4M3qu+5dPzjB5GhERERE/FuAzcqF8ZEAbDuQb24wIiIirYCSUtKgrw8VsDu7CHuAlatS1KxTRERE5HRqlvBtP5hvahwiIiKtgZJS0qCaBufp/WKJDA40ORoRERER/5eaUF0pdVDNzkVERE5HSSmpV3mVi3e3HgbghiFqcC4iIiLSGDWVUt8dcVJR5TY3GBERET+npJTUK3NHDgWllcRFOhiRHGV2OCIiIiKtQmLnECIcAVRUudmdXWh2OCIiIn5NSSmpV83SvesGdcVmtZgcjYiIiEjrYLFYvNVS29RXSkRE5JSUlJKT5DjL+GR3LgDjB2npnoiIiEhTpHTz9JXafkB9pURERE5FSSk5yaqth3C5DQaf15EeXcLMDkdERESkVVGllIiISOOYnpR69tlnSUxMxOFwkJaWxsaNGxscW1lZyZ/+9CeSk5NxOBykpqayevVqnzEul4s5c+aQlJREcHAwycnJPPTQQxiG4R1zyy23YLFYfLZRo0ads2tsTQzD4M0vPUv3rh+sKikRERGRphqQ0AGA73OKKK1wmRuMiIiIHwsw88NXrFhBRkYGS5YsIS0tjYULF5Kens6uXbuIjo4+afzs2bN55ZVXWLp0KX369OHDDz/k2muvZd26dQwcOBCAxx57jMWLF7Ns2TL69evHl19+yZQpU4iMjOT3v/+991yjRo3ixRdf9L622+3n/oJbge0HC/g+pwh7gJWrUuLMDkdERESk1YmNdBAdbiensJxvDxcwJLGT2SGJiIj4JVMrpRYsWMDUqVOZMmUKffv2ZcmSJYSEhPDCCy/UO/7ll1/m/vvvZ8yYMfTo0YNp06YxZswYnnrqKe+YdevWcc0113DVVVeRmJjI9ddfz5VXXnlSBZbdbic2Nta7dezY8Zxea2tR0+B81IWxRDgCTY5GREREpHWqXcKnvlIiIiINMS0pVVFRwebNmxk5cmRtMFYrI0eOZP369fUeU15ejsPh8NkXHBzM2rVrva9HjBhBZmYmu3fvBmDbtm2sXbuW0aNH+xy3Zs0aoqOj6d27N9OmTePo0aPNdWmtVlmli/e2HQbghsEJJkcjIiIi0nql1jQ7V18pERGRBpm2fC8vLw+Xy0VMTIzP/piYGHbu3FnvMenp6SxYsICf//znJCcnk5mZyTvvvIPLVbtWf+bMmTidTvr06YPNZsPlcvHII48wceJE75hRo0Zx3XXXkZSUxJ49e7j//vsZPXo069evx2az1fvZ5eXllJeXe187nc6zuXy/lLkjh4LSSuIjHQxP7mx2OCIiIiKtVkp1X6ltB/JNjUNERMSfmdpTqqn+8pe/MHXqVPr06YPFYiE5OZkpU6b4LPd74403ePXVV1m+fDn9+vVj69atzJgxg/j4eCZPngzAhAkTvOP79+9PSkoKycnJrFmzhiuuuKLez54/fz4PPvjgub1Ak721+QAA1w3qhs1qMTkaERERkdYrpaunUmrf0RIKSiqJDFFbBBERkROZtnwvKioKm81Gdna2z/7s7GxiY2PrPaZLly6sWrWK4uJifvrpJ3bu3ElYWBg9evTwjrn33nuZOXMmEyZMoH///vz617/mD3/4A/Pnz28wlh49ehAVFcUPP/zQ4JhZs2ZRUFDg3Q4cONDEK/ZvOc4yPtmdC8B43XVPRERE5Kx0DA2ie6cQALYfyjc3GBERET9lWlIqKCiIwYMHk5mZ6d3ndrvJzMxk+PDhpzzW4XDQtWtXqqqqePvtt7nmmmu875WUlGC1+l6WzWbD7XY3eL6DBw9y9OhR4uIavtuc3W4nIiLCZ2tLVn51CLcBQ87rSFJUqNnhiIiIiLR6Kd6+Ump2LiIiUh9T776XkZHB0qVLWbZsGTt27GDatGkUFxczZcoUACZNmsSsWbO847/44gveeecd9u7dy2effcaoUaNwu93cd9993jFjx47lkUce4R//+Af79u1j5cqVLFiwgGuvvRaAoqIi7r33XjZs2MC+ffvIzMzkmmuuoWfPnqSnp7fsD8BPGIbBm9V33bteVVIiIiIizSK15g586islIiJSL1N7St10003k5uYyd+5csrKyGDBgAKtXr/Y2P9+/f79P1VNZWRmzZ89m7969hIWFMWbMGF5++WU6dOjgHfPMM88wZ84cbr/9dnJycoiPj+e2225j7ty5gKdqavv27Sxbtoz8/Hzi4+O58soreeihh7Db7S16/f5i28ECfsgpwhFoZUxKw9ViIiIiItJ4qpQSERE5NYthGIbZQbRGTqeTyMhICgoKWv1SvtmrvuaVDfsZNyCehRMGmh2OiIiIX2tLc4Bzrb3/rIrLq+j/wIe4Ddh4/xVERzjMDklERKRFNHYOYOryPTFfWaWL97YeBuD6wQkmRyMiIiLSdoTaA+gZHQZ4KtNFRETEl5JS7dy/dmTjLKsiPtLBiOTOZocjIiIi0qbU9JXafjDf1DhERET8kZJS7dxb1Q3Oxw/uhtVqMTkaEREROVPPPvssiYmJOBwO0tLS2LhxY4Njv/32W8aPH09iYiIWi4WFCxee9TmlfikJHQBVSomIiNRHSal2LNtZxqe7cwEYP0h33RMREWmtVqxYQUZGBvPmzWPLli2kpqaSnp5OTk5OveNLSkro0aMHjz76KLGxsc1yTqlfqrfZeT5q5SoiIuJLSal2bOVXh3AbMDSxI4lRoWaHIyIiImdowYIFTJ06lSlTptC3b1+WLFlCSEgIL7zwQr3jhw4dyhNPPMGECRMavPtwU88p9esTG0GQzUp+SSUHjpWaHY6IiIhfUVKqnTIMw7t07/rBqpISERFprSoqKti8eTMjR4707rNarYwcOZL169e36DnLy8txOp0+W3sXFGDlgrhwALaqr5SIiIgPJaXaqa0H8vkhpwhHoJUx/ePMDkdERETOUF5eHi6Xi5iYGJ/9MTExZGVlteg558+fT2RkpHdLSNCdfQFSapqdH8g3NQ4RERF/o6RUO1VTJTX6wjjCHYEmRyMiIiJtwaxZsygoKPBuBw4cMDskv5Di7SulZuciIiJ1BZgdgLS8skoX7207DGjpnoiISGsXFRWFzWYjOzvbZ392dnaDTczP1TntdnuDParas9TqO/B9c7gAl9vApjsei4iIAKqUapc++i6bwrIqunYIZniPzmaHIyIiImchKCiIwYMHk5mZ6d3ndrvJzMxk+PDhfnPO9iy5SxghQTZKKlz8kFNkdjgiIiJ+Q5VS7VDN0r3xg7pi1V/qREREWr2MjAwmT57MkCFDGDZsGAsXLqS4uJgpU6YAMGnSJLp27cr8+fMBTyPz7777zvv80KFDbN26lbCwMHr27Nmoc0rj2awWLuwaycYfj7HtYD69Y8PNDklERMQvKCnVzmQVlPHZ97kAjNfSPRERkTbhpptuIjc3l7lz55KVlcWAAQNYvXq1t1H5/v37sVprC+QPHz7MwIEDva+ffPJJnnzySS699FLWrFnTqHNK06R28ySlth/M58YhagAvIiICSkq1Oyu/OoTbgGGJnTivc6jZ4YiIiEgzmT59OtOnT6/3vZpEU43ExEQMwzirc0rTeO/Ap2bnIiIiXuop1Y4YhsFbmz13wVGDcxEREZGWM6C62fmOI07Kq1zmBiMiIuInlJRqR746kM+e3GKCA22MSYkzOxwRERGRdqNbx2A6hgRS6TLYeaTQ7HBERET8gpJS7UhNg/PRF8YSZtfKTREREZGWYrFY6izhyzc1FhEREX+hpFQ7UVbp4u/bDgNauiciIiJihtRukQBsPaC+UiIiIqCkVLvxf99lU1hWRdcOwVzUo7PZ4YiIiIi0O6qUEhER8aWkVDtRs3Rv/OBuWK0Wk6MRERERaX9SEjyVUj/kFlFUXmVyNCIiIuZTUqodyCooY+33uQCMH9TV5GhERERE2qfocAdxkQ4MA745pCV8IiIiSkq1A+98dRC3AcOSOnFe51CzwxERERFpt1Kq+0ppCZ+IiIiSUm2eYRjepXtqcC4iIiJirpq+UtsOqlJKRERESak2bsv+fPbmFhMcaGNM/zizwxERERFp11LV7FxERMRLSak2rqZKanT/WMLsASZHIyIiItK+9a9evnfgWCnHiitMjkZERMRcSkq1YWWVLt7fdhjQ0j0RERERfxAZHEhSlKfHp6qlRESkvVNSqg378NssCsur6NYxmIuSOpsdjoiIiIgAqd5m5+orJSIi7ZuSUm1YzdK98YO6YbVaTI5GRERERKC22bkqpUREpL1TUqqNOlJQytof8gBPUkpERERE/ENqgqdSauuBAgzDMDkaERER8ygp1Ua9s+UQhgFpSZ3o3jnE7HBEREREpFrfuEhsVgt5ReUcKSgzOxwRERHTKCnVBhmGwdvVS/fU4FxERETEvwQH2Tg/JhzQEj4REWnflJRqg7bsP87evGJCgmyM6R9ndjgiIiIicoKaZufb1OxcRETaMSWl2qCaBuejL4wj1B5gcjQiIiIiciI1OxcREVFSqs0prXDx/rYjgJbuiYiIiPirlOpKqe0HC3C71excRETaJyWl2pj/+y6LwvIqunUMJi2pk9nhiIiIiEg9eseGYw+wUlhWxb6jxWaHI6ey59/wfDp8+EeoLDU7GhGRNkVJqTbmrToNzq1Wi8nRiIiIiEh9Am1W+sZHAJ5qKfFDxXnwzq3w8rVwYAOsXwTPXQZHtpsdmYhIm6GkVBtyOL+UtT/kATB+kJbuiYiIiPiz1Oq+UtvUV8q/GAZ89QosGgLbVwAWGDARwmIgdycsvRw+fxrcbrMjFRFp9ZSUakNWfnUIw4CLenQioVOI2eGIiIhIC3r22WdJTEzE4XCQlpbGxo0bTzn+zTffpE+fPjgcDvr3788///lPn/dvueUWLBaLzzZq1KhzeQntTt2+UuIn8n6AZWPh3Tug9DjE9If/zIRxf4Vp66D3VeCuhI/mwN9+BQUHzY5YRKRVU1KqjTAMo87SvQSToxEREZGWtGLFCjIyMpg3bx5btmwhNTWV9PR0cnJy6h2/bt06br75Zn7729/y1VdfMW7cOMaNG8c333zjM27UqFEcOXLEu7322mstcTntRmpCBwC+PVxAlUtVN6aqqoBPnoDFI2DfZxAQDL/8E9z6MXQb7BkTGgUTXoWxT0NgiGfc4hHwzdvmxi4i0oopKdVGbNl/nB/zigkJsjH6wlizwxEREZEWtGDBAqZOncqUKVPo27cvS5YsISQkhBdeeKHe8X/5y18YNWoU9957LxdccAEPPfQQgwYNYtGiRT7j7HY7sbGx3q1jx44tcTntRlLnUMLtAZRVutmdXWR2OO3X/g3wPz+Djx8GVzkkXwF3bICL7wJboO9YiwUGT4bfrYWug6GsAN76jaf3VJkq3kREmkpJqTbizS89VVJj+scRag8wORoRERFpKRUVFWzevJmRI0d691mtVkaOHMn69evrPWb9+vU+4wHS09NPGr9mzRqio6Pp3bs306ZN4+jRo81/Ae2Y1Wqhv3cJX765wbRHpfnw9xnwQrqnV1RIFIx/Hv7jbeiYeOpjOyfDbz6ES/8LLFZP76nFl8BP61ogcBGRtkNJqTagtMLF+9uPAJ677omIiEj7kZeXh8vlIiYmxmd/TEwMWVlZ9R6TlZV12vGjRo3ib3/7G5mZmTz22GN88sknjB49GpfL1WAs5eXlOJ1On01OLUXNzlueYcC3K+HZYbD5Rc++gb+G6Zug//WeaqjGsAXCL+73JKc6JkLBfnjpKsj8k2c5oIiInJZKatqAD7/Noqi8ioROwQxL7GR2OCIiItIGTJgwwfu8f//+pKSkkJyczJo1a7jiiivqPWb+/Pk8+OCDLRVim5BaXSm17YCWfrWI/P3wj3vg+w89rzv3hLF/gcRLzvycCcM8y/k+mAlbX4HPnoIfMuG6pdDl/OaJW0SkjVKlVBvgbXA+KAGrtZF/2REREZE2ISoqCpvNRnZ2ts/+7OxsYmPr7zMZGxvbpPEAPXr0ICoqih9++KHBMbNmzaKgoMC7HThwoAlX0j6lVDc735VdSFllw1VocpbcLlj/V3j2Ik9CyhroWXr3u8/PLiFVwx4O456FG5aBowMc2Qr/83PY9LynMktEROqlpFQrdyi/lM/35AFw3aCuJkcjIiIiLS0oKIjBgweTmZnp3ed2u8nMzGT48OH1HjN8+HCf8QAfffRRg+MBDh48yNGjR4mLi2twjN1uJyIiwmeTU4uPdBAVFoTLbfDtYS13PCeObIOll8OHs6CyGLoP91Q2/eJ+CHQ072f1Gwe3r4cel0FVKfwjA16bAEW5zfs5IiJthJJSrdzKLQcxDBjeozMJnULMDkdERERMkJGRwdKlS1m2bBk7duxg2rRpFBcXM2XKFAAmTZrErFmzvOPvuusuVq9ezVNPPcXOnTt54IEH+PLLL5k+fToARUVF3HvvvWzYsIF9+/aRmZnJNddcQ8+ePUlPTzflGtsqi8Xi7SulZufNrKIYPvwjPHeZp3LJHulZqnfLPyG6z7n73Ih4+I+VkD4fbHbYvRoWD4ddq8/dZ4qItFLqKdWKGYZRu3RPDc5FRETarZtuuonc3Fzmzp1LVlYWAwYMYPXq1d5m5vv378dqrf1b5IgRI1i+fDmzZ8/m/vvvp1evXqxatYoLL7wQAJvNxvbt21m2bBn5+fnEx8dz5ZVX8tBDD2G32025xrYspVsk/96Zw/aD6ivVbHb/H/zjbk/zcYB+18GoRyE85tTHNRerFYbfDj0uhbenQs638NpNMOS3cOXDEKQ/JouIAFgMQ4ucz4TT6SQyMpKCggLTStO/3HeM65esJzTIxqbZIwkJUo5RRETkXPOHOUBroZ9V43y8M4cpL22iR5dQ/n33ZWaH07oVZsPqmfDtO57Xkd3hqqfg/CvNi6myzHNHvg3Pel537gXjl0L8QPNiEhE5xxo7B1AWoxV780tPldSY/nFKSImIiIi0UinVd+Dbm1vMxP/dwAWxEfSNj+CCuAiSu4QRFKCOG6fldsNXf4OP5kJZAViscNHtnr5RQaHmxhbogFF/hl6/hFXT4Oj38L8jPbFdPAOsNnPjExExkTIZrVRJRRX/+PoIoKV7IiIiIq1Z5zA7QxM7smnfcT7/4Sif/3DU+16gzULP6HAuiAunb5wnUXVBXASdQoNMjNjP5O6Cv98F+9d7XscN8PSOih9gZlQnS/4FTFsH78+A7971VE99/y+4dgl0PM/s6ERETKGkVCv14bdZFJVX0b1TCMOSOpkdjoiIiIicheVTL2LHESc7jjj57rCTHUcK2ZHlpLCsyrv/HQ55x8dE2L0JqgviIugbF0FSVCg2q8XEq2hhlWWwdgF8tgDclRAYCpf/EYbdBjY//WdOSCe4YRlsew3+eS/sXwdLLoExT0LKjWBpR9+fiAhKSrVadRucW/QfLxEREZFWLdBmJaVbB++d+MBzU5uDx0urk1KFnscsJz8dLSHbWU62M5c1u3K94x2BVnrHhPskq/rEhRPhCDThis6xfWvh7zM8S+EAeqXDVU9Ch+6mhtUoFgsM+H/QfTisvA0OfAErb/Xcpe/qBRDc0ewIRURajJJSrdDB4yWs2+Mp675uUFeToxERERExQWk+uF21rxv6I53Pfsvp95/JMc2132rzec9isZDQKYSETiFc2S/Wu7+ovIpdWU6+q0lUHXGy80ghpZUuth0sYNsJd/Hr1jG4TkWVJ2mV0DEEa2usqio5Bh/Nga9e8bwOi4HRj0Hfca2vyqhTEtzyT1j737Bmvqc5+4EvPMv5kn5udnQiIi1CSalWaOWWQxgGjEjuTLeOup2siIiItEOv3exZ+tSWOCIh8Wee3kM9fgGdetSbaAmzBzD4vE4MPq+2hYPbbfDTsZLqpX+12+GCMg4eL+Xg8VI++i7b5xy9Y2t6VUVyQVw4vWPD/ffmOYYBX78FH86C4urqsMFTYOQDENzBzMjOji0ALr0Xki+Hd6bCsT2w7Fcw4k64fDYE2M2OUETknPLT/+pIQwzD4K0ttUv3RERERKSNKCuAne97NoDIBOhxmWdLuhTCujR4qNVqISkqlKSoUK5KifPuzy+pqF36V738b3d2EUXlVWz+6TibfzruHWuxQFLn0OqqqtplgHGRDnPbRRzfB+9nwJ5Mz+uo3p5G5ucNNy+m5tZtMNz2KXx4P2xZBuuehj0fw/ilEH2B2dGJiJwzFsMwDDMDePbZZ3niiSfIysoiNTWVZ555hmHDhtU7trKykvnz57Ns2TIOHTpE7969eeyxxxg1apR3jMvl4oEHHuCVV14hKyuL+Ph4brnlFmbPnu39j6lhGMybN4+lS5eSn5/PxRdfzOLFi+nVq1ej43Y6nURGRlJQUEBERMTZ/RCaYNO+Y9ywZD2hQTY2zR7pv3/NEhERaaPMmgO0Ruf0Z+V213lRZzrrM7U9i/3Nea5G7QeO7YW9H8OeNZ5lXO5K3/dj+kPyZZ4kVfcREHRmFfNVLjd784o9TdWr+1V9d9hJXlF5veM7hATSJzbcp6l6r5gw7AG2M/r8RnNVwoa/wsfzoaoUbEHw83vh4rvadgXRzn/Ae3dCyVEIcMAv/wTDbm19yxNFpF1r7BzA1IzGihUryMjIYMmSJaSlpbFw4ULS09PZtWsX0dHRJ42fPXs2r7zyCkuXLqVPnz58+OGHXHvttaxbt46BAwcC8Nhjj7F48WKWLVtGv379+PLLL5kyZQqRkZH8/ve/B+Dxxx/n6aefZtmyZSQlJTFnzhzS09P57rvvcDgcLfozaKo3vzwAwFUpcUpIiYiISPtltZodQfML6QTdhngSLxXF8NN6T5Jq7xrI/gayv/Zs657xJGgS0qorqX4B8QM8PakaIcBm5fyYcM6PCeeaAbX9SXMLy32W/u04Usie3CLySyrZsPcYG/Yeqz2H1UJylzCfiqoL4iLoEt5MyaJDm+G9uzzXC55ljVcvhKiezXN+f9bnKug6BN69A374CD64D3Z/COP+CuGxpz9eRKQVMbVSKi0tjaFDh7Jo0SIA3G43CQkJ3HnnncycOfOk8fHx8fzxj3/kjjvu8O4bP348wcHBvPKKp9nh1VdfTUxMDM8//3y9YwzDID4+nrvvvpt77rkHgIKCAmJiYnjppZeYMGFCo2I346+kJRVVDH34XxRXuHjjtuEMS+p0+oNERESkWalSqvH0s2pGRTmw9xNPgmrvx+A85Pu+I9LTHLsmSdVAP6qmKq9y8X120Ul3AMwvqax3fFSYvbpPVQTREQ5sFrBZLVitFmyW2se6+2xWsFbvC3QVk7x9IfG7/4YFg8qgDhwcej/Hz78Bm9XqOa56bN3javdZsFio5zMsdcbSOu5ebRiw6X/h/2ZDVRkEd4JfPQMXXG12ZCIip+X3lVIVFRVs3ryZWbNmefdZrVZGjhzJ+vXr6z2mvLz8pEqm4OBg1q5d6309YsQInnvuOXbv3s3555/Ptm3bWLt2LQsWLADgxx9/JCsri5EjR3qPiYyMJC0tjfXr1zc6KWWG1d9kUVzh4rzOIQxN1K1iRURERNqNsGhIucGzGQYc3VNbRfXjp55+VDv+7tkAIrtDj0s9TdOTLoXQqDP6WHuAjQu7RnJh10jvPsMwyHKWeRNVNc3VfzxaTF5ROZ99X85n3+c1+bNGWjfzp8AXibd4KrJWui7mYed/cDQzEjLr//fBmbLWJMtqklk1CbPqpNWJya4Aq4WI4ECiw+10qd6iwx11ntuJCrMTFNCMFXwWCwyb6qkSe+c/IetrWDERBk2C9PlgD2u+zxIRMYlpSam8vDxcLhcxMTE++2NiYti5c2e9x6Snp7NgwQJ+/vOfk5ycTGZmJu+88w4uV+3tgGfOnInT6aRPnz7YbDZcLhePPPIIEydOBCArK8v7OSd+bs179SkvL6e8vHadvdPpbNoFN4O3Nlc3OB/UrXX8dUdEREREmp/F4lnGFtXTk7RwVcHhr6qrqNZ4+lEV7IevXvZsALH9PRVUPS6D7sPPuB+V5+MtxEUGExcZzOV9aufUJRVV7MoqZMeRQnZmOTleUonbbeByG7gNz+ZyG7gMvPtdhkGHylymFC5meLnnbopHrDE87bidTQEDiXQbhFUf63bjPcZd/ehy1z53u/HuOx23AW6XwUl9xM5Sh5A6iaswO9ERDrqE1SauapJYkcGBjZ/PR/eB//w3fPwwfP40bPkb/PgZjP9fz3JPEZFWrFU1JfrLX/7C1KlT6dOnDxaLheTkZKZMmcILL7zgHfPGG2/w6quvsnz5cvr168fWrVuZMWMG8fHxTJ48+Yw/e/78+Tz44IPNcRln5ODxEtbtOYrFAtfprnsiIiIiUsMWAAlDPdul90J5Eexf70lQ7fkYcr71VNlkfe25q1tNP6rk6iRV3IBG96M6lZCgAAZ278jA7o2s6He74cvn4V8PQkUhWGww4k7iLv0v5p9F0sxz6tqklWFQTwKrblKrzvs1ibM6z92Gpzl8fmklOYXl5Hq3strnReVUugzySyrJL6lkd3bRKeMLslnpEm4nqm6yKsxOdIS9NokV4SAqLMjTUD4gyNPwvOcvYeXv4PiP8PyVcOl/wc/u9vwOiLSgSpebskoXZZWex/Kq2ufexyoX5ZVurFaIcAQSGRxIRHCg97kj0KpiCzEvKRUVFYXNZiM7O9tnf3Z2NrGx9Tfw69KlC6tWraKsrIyjR48SHx/PzJkz6dGjh3fMvffey8yZM73L8Pr3789PP/3E/PnzmTx5svfc2dnZxMXV3i43OzubAQMGNBjvrFmzyMjI8L52Op0kJCQ0+brP1DtbPH0DRiR3pmuH4Bb7XBERERFpZexh0OuXng2gMNuzxK9muZ/zEOz7zLNl/gkcHer0o7qs2fpRnVL2t/D3u+DgJs/rroNh7F88FV3NwGq1YMVC4Dm+QWANt9ugoLSS3KJycpzl5BZ5Elae557EVU1Cq6C0kgqXm0P5pRzKLz3tuSN9lg2GkZD8AuMOL6Bn9mpY82dKd3xIxdjFRHQ9X//Ab6fcboMyn6RQ9fMqz/PyOkmisko35ZUuyqpOlVSqe3z1+EoX5TXHVLkbVZF4OoE2iydR5ahOVgUHEuEI8CavIusksCKCA05IbAUQYGuDN7xoh0xLSgUFBTF48GAyMzMZN24c4Gl0npmZyfTp0095rMPhoGvXrlRWVvL2229z4403et8rKSnBesLdWGw2G+7q2wYnJSURGxtLZmamNwnldDr54osvmDZtWoOfabfbsdvNufWsYRi1S/dUJSUiIiIiTREec0I/qh9qq6j2fQZl+bDjPc8G0KF7bYIq6TII7dx8sVSWwiePeyq23FUQFAZXzIOhv22Wai2zWK0WOoYG0TE0iPNjwk85trzKVafaqjZZVZvQKievel+Fy01BaSUFpZV8n1NbfbWISfzKmszDgS8Qkb0Z19KfM8s1mU9DrqRLhIMuJ/S7OrH3laOlsnVySm63wdHiCrKdZWQVlJHlLCPbWcax4gpvUqj8hMqjutVI5dWVSBUut6nXYQ+w4gi04QisfgzwPLcH2LAHWnEbBs7SKpxlnt9lZ2klbgMqXQZ5RRXkFVWc0eeGBtl8qq8iqpNXdRNdkfUluoIDCQ2yKYnrJ0yt88zIyGDy5MkMGTKEYcOGsXDhQoqLi5kyZQoAkyZNomvXrsyfPx+AL774gkOHDjFgwAAOHTrEAw88gNvt5r777vOec+zYsTzyyCN0796dfv368dVXX7FgwQJ+85vfAJ418DNmzODhhx+mV69eJCUlMWfOHOLj473JMX+zad9x9h8rIcweQHo/3QZWRERERM6QxQJRvTybTz+q6iqqAxshf7+nb9GWv3mOiU2pTVKdNwICz7Bqf+8a+PsMz9IzgD5Xw+jHIbLrWV9Wa2IPsNGtYwjdOp56iaJhVFdf1U1cFZaTU7NssOhXTMsfyB+KnmKIZQePBvwPH5RuYZbzP9nGqRNjEY4Ab6KqS7hnmWDNEsIu1Ymr6HA7nUKDVI1yhsoqXT6JphOfZzs932Wlq3n7mgXZrNirE0LeJFGgtTpRVJ0sqpM4cgTa6k0q2b3H2nAE1D4/caw9oOlL8AzDoKi8CmdZFc7S2kRVQWklzrIq72tnWfVjafW+6tfFFZ6e0sUVLoorXBwuKGvyz8lmtRDhCGiwIisi2DepVXdcp9AgbFYltJqLqUmpm266idzcXObOnUtWVhYDBgxg9erV3ibk+/fv96l6KisrY/bs2ezdu5ewsDDGjBnDyy+/TIcOHbxjnnnmGebMmcPtt99OTk4O8fHx3HbbbcydO9c75r777qO4uJhbb72V/Px8LrnkElavXn3Snf38xVubDwBwVf84QoK0XlxEREREmolPP6r7PP2oflpX2zQ951vI2u7Z1j0NNjt0T6tOUv0C4lJPX+FUfBT+74+w7TXP6/A4GPMEXDD2HF9c62axWOgQEkSHkCB6NVh9dRG4r6Nq7V+wrfkzo9nE5aE/sbb/Q3xtH3RSQqum+spZ5kkI7MktPk0M0Dk0iKgw+0lJqy51KrCiwux0CA7E2g7+oe52GxwrqahOLFUnmqoTTlnOcu/zgtLKRp3PYoGoMDuxEQ5iIhzERtrpHGonOKj+ZJD9hGokb+KoOhHVGpIlFouFcEcg4Y7AM2pNU+lyU1gneVWbsKpNXtUmtjz7CqvHFZRWUlXdN+54SSXHSxr3PdUVFGClR1QoyV3C6NHF85jcJYykLqGE2fXv9aayGIbRvKnZdsLpdBIZGUlBQQERERHn7HNKKqoY+vC/KK5w8ebvhjM0sdM5+ywRERE5vZaaA7QF+lm1AYXZ8OMntcv9Cg/7vl/Tj6qmaXrHpNp+VIYB216HD++H0mOABYb+J1wxFxz6fWh2h7fCO1Mhb7fn9UW3e5ZGBtb+4d2oXkaVU1jm7XWVW1hOXlGFdwmh53U5R4vKaUrboACrxZus8lZe1U1g1bwXbifcHuCXS6fKKl31VDWVe5NPWQVlTapuCg60ERvpICaiOukU6SA2wuHzvEu4nUBVo7UYwzAoq3SflLw6MYnVULKrsKzqlOePjXDUSVSF0qNLGMnRYcRFONpF0rauxs4BlJQ6Qy01yXp780HufnMbiZ1D+Piey/zy/7xFRETaEyVaGk8/qzbGMCDv+9oqqn2fQbnTd0yH7p4KqvNGwNblnoQWQHRfGPu0pyJLzp2KEvhoDmz6X8/r6L5w3VKIvbDJp3K5DY4VV3iTVCcmrereeTC/idUm9gBrw0mrMN/+V8FBZ9//yjA811I30VS3wqkm6dTY6/BUkNmJjaxT4VQ36RTp2Rfh8M/km5y5quqbFOzNLWZPbhF7qh/35hadsjdWcKCNpKhQkqPDPFVW0dVJq6iwZvkd90dKSp1jLTXJuvm5Dazfe5R7rjyf6Zf3OmefIyIiIo2jREvj6WfVxrmq4PCW2iqqgxs9zcvrCnDApf8FI+4EW6ApYbZLu/8P3r0dinPBFuSpmLrodrCem4qciio3R4vrVl75Jq3yCiu8Ca2i8lNXmpwozB5wQtIq6KSlg2H2APKKKk5YSlf7PMdZ3uhm4I5Aa52ldA6f5zWP0apuknoUlFSyJ6+IPTlF7M0r9j7uyyum6hRlh107BJ9cXdUljJgIe6tOaiopdY61xCTrwLESfvb4x1gssPa/Lj+j9bYiIiLSvJRoaTz9rNqZuv2oflrrqZr65Z+gUw+zI2ufinLhvTth9wee10mXwrjFpjeWL61wkVfke9fBvDpVWDVJrZzCMoyqSoIpx0EFwZZyQij3vLZUEIzntcNSQSAunEYIxwkj3wijgDDyjVCKCAZq/1EfFRZ0clXTCc8jglXdJM2r0uXmwLESb3VVbZVV0Sl7WoUG2aoTVLWJquToUBI7h7aKO2g2dg6gLlx+7J0thwC4ODlKCSkRERER8W/2MDj/Ss8m5gvrAje/Bptf8vT1+vETWDwCxi6Eftee3bkNA1yVUFkClaXVj3Wfl0JFcfXruu+XEFxZSkJlKQne9+s5h7sUI7AYS4DrrMJ0WwJw2SMhuCO20M5YQzpBcEcI6QTBHTzPgzuBrSO4O0JF9fOgsNreaCJnKdBmpUeXMHp0CWMkMT7vHSuuYG91gqpu0uqnYyUUV7j4+lABXx8q8DnGYoFuHYM9jdajPImqmscuYa2vukpJKT/ldhu8tcVz173rB3czORoREREREWl1LBYYMgUSfwbv/Ccc/grevAV2rYbES05KGHmTSKdKGNW8b5xdwui0ofu8sEJgKAQGQ1AIBIZ4ntc8WgOhrABKj3ua6pccA1c5VqMKa9lRKDsKx39o/IdbA2oTVsEd6ySyOlYns+rbr2SWNF2n0CA6hXZiyAk3NKuocrP/WLG3Z9WenGL2Vi8NdJZVceBYKQeOlbJmV67PceGOAG91VXKdx+6dQ7AH+Gd1lZJSfmrTvmMcOFZKmD2A9H6xZocjIiIiIiKtVVRP+O1HsOZRWLsAtr/u2ZqDxQZB1Qkjb6LohKRRYEh1MumEfSeO854npM75Qj39yJqa7Kks9SSnSo/XJqu8z4/XeS+/9r3qZBbuKk8/ruLc036MD2tgbYLqxITViZuSWXIKQQFWekaH0zM63Ge/YRgcLa5gT46nyXpNldWe3GIOHi+hsKyKbQfy2XYg3+c4qwW6dwrxVFfVJKyqm653Cg0ytbpKSSk/9dbmgwBcnRLXZrvxi4iISPN59tlneeKJJ8jKyiI1NZVnnnmGYcOGNTj+zTffZM6cOezbt49evXrx2GOPMWbMGO/7hmEwb948li5dSn5+PhdffDGLFy+mVy/deEWkVbIFwhVzoOdIWL/Is/zOp+oouLYaqe6+kxJFJySTziRh1BICgz39s5raQ6uixDd5VTeZVXJCYqvuflc5uCuhOMezNUXdZFZIJ0+SCsOzTBLqPK9vH40c18A+7/En7mvsuEZ8ni3Q83sUFFb9GOr5XQuqu4U14nkoWNv3v40tFgtRYZ4G/2k9Ovu8V1bp4qejJT6Jqr3Vj0XlVew7WsK+oyVk7vQ957CkTrxx2/AWvApfSkr5oeLyKv7x9RFAS/dERETk9FasWEFGRgZLliwhLS2NhQsXkp6ezq5du4iOjj5p/Lp167j55puZP38+V199NcuXL2fcuHFs2bKFCy/03Dr+8ccf5+mnn2bZsmUkJSUxZ84c0tPT+e6773A4HC19iSLSXM4b7tmkfkHVVV1NSWYZhqcyqymJrOZIZrVHAcEnJKxC6k9eNSrJVf3cFuSfidUmcgTa6B0bTu/Yk6urcgvL+eGERNXe3CIO5ZcSHW43KWIP3X3vDJ3Lu8m8tfkg97y5jcTOIXx8z2WtrlGZiIhIW+aPd5RLS0tj6NChLFq0CAC3201CQgJ33nknM2fOPGn8TTfdRHFxMe+//75330UXXcSAAQNYsmQJhmEQHx/P3XffzT333ANAQUEBMTExvPTSS0yYMKFRcfnjz0pExG/4JLPqJK/Ki6qTJJbaRzhhHw3ss5zlPhr4vLPY56rwVKBVFHn6kVUUn+J5A6/PZQ8za8BpKrdOeG0L9PQ5q7k+n+d1X1vrf+3z3omvm2ts42Iqq3JTgoNO8c1/l1Tdfa8Ve2tzbYNzJaRERETkVCoqKti8eTOzZs3y7rNarYwcOZL169fXe8z69evJyMjw2Zeens6qVasA+PHHH8nKymLkyJHe9yMjI0lLS2P9+vUNJqXKy8spLy/3vnY6nWd6WSIibZ/FcmaVWe2NYUBV+amTWZWnSWrV97yqzHN+dxWUF3i2dsYBOJIvh1+vNC0GJaX8TEWVm0CblQCrhesGaemeiIiInFpeXh4ul4uYGN/bTMfExLBz5856j8nKyqp3fFZWlvf9mn0NjanP/PnzefDBB5t8DSIiIg2yWCDQ4dlCO59+fGO5qk6RzDpFJZerEk/PLHed/lnVzw2359w+r40Tnp84tqH36vsMTvHeicfRiM83wO673K+lKSnlZ4ICrLz82zSOFVfQKTTI7HBEREREGm3WrFk+FVhOp5OEhAQTIxIREWmALQBskeCINDuSds1qdgBSPyWkREREpDGioqKw2WxkZ2f77M/OziY2NrbeY2JjY085vuaxKecEsNvtRERE+GwiIiIiDVFSSkRERKQVCwoKYvDgwWRmZnr3ud1uMjMzGT68/jtsDR8+3Gc8wEcffeQdn5SURGxsrM8Yp9PJF1980eA5RURERJpKy/dEREREWrmMjAwmT57MkCFDGDZsGAsXLqS4uJgpU6YAMGnSJLp27cr8+fMBuOuuu7j00kt56qmnuOqqq3j99df58ssvee655wCwWCzMmDGDhx9+mF69epGUlMScOXOIj49n3LhxZl2miIiItDFKSomIiIi0cjfddBO5ubnMnTuXrKwsBgwYwOrVq72Nyvfv34/VWlsgP2LECJYvX87s2bO5//776dWrF6tWreLCCy/0jrnvvvsoLi7m1ltvJT8/n0suuYTVq1fjcDha/PpERESkbbIYhmGYHURr5HQ6iYyMpKCgQP0SRERE2hHNARpPPysREZH2qbFzAPWUEhERERERERGRFqeklIiIiIiIiIiItDglpUREREREREREpMUpKSUiIiIiIiIiIi1OSSkREREREREREWlxSkqJiIiIiIiIiEiLCzA7gNbKMAzAc5tDERERaT9q/ttfMxeQhmm+JCIi0j41dr6kpNQZKiwsBCAhIcHkSERERMQMhYWFREZGmh2GX9N8SUREpH073XzJYujPfGfE7XZz+PBhwsPDsVgszXpup9NJQkICBw4cICIiolnPLWdP349/0/fjv/Td+Dd9P41nGAaFhYXEx8djtaoTwqlovtR+6fvxb/p+/Je+G/+m76fxGjtfUqXUGbJarXTr1u2cfkZERIR+0f2Yvh//pu/Hf+m78W/6fhpHFVKNo/mS6Pvxb/p+/Je+G/+m76dxGjNf0p/3RERERERERESkxSkpJSIiIiIiIiIiLU5JKT9kt9uZN28edrvd7FCkHvp+/Ju+H/+l78a/6fuR1ka/s/5N349/0/fjv/Td+Dd9P81Pjc5FRERERERERKTFqVJKRERERERERERanJJSIiIiIiIiIiLS4pSUEhERERERERGRFqeklB969tlnSUxMxOFwkJaWxsaNG80Oqd2bP38+Q4cOJTw8nOjoaMaNG8euXbvMDksa8Oijj2KxWJgxY4bZoUi1Q4cO8R//8R907tyZ4OBg+vfvz5dffml2WO2ey+Vizpw5JCUlERwcTHJyMg899BBqNymtgeZL/klzptZD8yX/o/mSf9J86dxSUsrPrFixgoyMDObNm8eWLVtITU0lPT2dnJwcs0Nr1z755BPuuOMONmzYwEcffURlZSVXXnklxcXFZocmJ9i0aRP/8z//Q0pKitmhSLXjx49z8cUXExgYyAcffMB3333HU089RceOHc0Ord177LHHWLx4MYsWLWLHjh089thjPP744zzzzDNmhyZySpov+S/NmVoHzZf8j+ZL/kvzpXNLd9/zM2lpaQwdOpRFixYB4Ha7SUhI4M4772TmzJkmRyc1cnNziY6O5pNPPuHnP/+52eFItaKiIgYNGsRf//pXHn74YQYMGMDChQvNDqvdmzlzJp9//jmfffaZ2aHICa6++mpiYmJ4/vnnvfvGjx9PcHAwr7zyiomRiZya5kuth+ZM/kfzJf+k+ZL/0nzp3FKllB+pqKhg8+bNjBw50rvParUycuRI1q9fb2JkcqKCggIAOnXqZHIkUtcdd9zBVVdd5fO/ITHfe++9x5AhQ7jhhhuIjo5m4MCBLF261OywBBgxYgSZmZns3r0bgG3btrF27VpGjx5tcmQiDdN8qXXRnMn/aL7knzRf8l+aL51bAWYHILXy8vJwuVzExMT47I+JiWHnzp0mRSUncrvdzJgxg4svvpgLL7zQ7HCk2uuvv86WLVvYtGmT2aHICfbu3cvixYvJyMjg/vvvZ9OmTfz+978nKCiIyZMnmx1euzZz5kycTid9+vTBZrPhcrl45JFHmDhxotmhiTRI86XWQ3Mm/6P5kv/SfMl/ab50bikpJdJEd9xxB9988w1r1641OxSpduDAAe666y4++ugjHA6H2eHICdxuN0OGDOHPf/4zAAMHDuSbb75hyZIlmmSZ7I033uDVV19l+fLl9OvXj61btzJjxgzi4+P13YjIWdOcyb9ovuTfNF/yX5ovnVtKSvmRqKgobDYb2dnZPvuzs7OJjY01KSqpa/r06bz//vt8+umndOvWzexwpNrmzZvJyclh0KBB3n0ul4tPP/2URYsWUV5ejs1mMzHC9i0uLo6+ffv67Lvgggt4++23TYpIatx7773MnDmTCRMmANC/f39++ukn5s+fr0mW+C3Nl1oHzZn8j+ZL/k3zJf+l+dK5pZ5SfiQoKIjBgweTmZnp3ed2u8nMzGT48OEmRiaGYTB9+nRWrlzJv//9b5KSkswOSeq44oor+Prrr9m6dat3GzJkCBMnTmTr1q2aYJns4osvPul24Lt37+a8884zKSKpUVJSgtXqOxWw2Wy43W6TIhI5Pc2X/JvmTP5L8yX/pvmS/9J86dxSpZSfycjIYPLkyQwZMoRhw4axcOFCiouLmTJlitmhtWt33HEHy5cv59133yU8PJysrCwAIiMjCQ4ONjk6CQ8PP6lXRWhoKJ07d1YPCz/whz/8gREjRvDnP/+ZG2+8kY0bN/Lcc8/x3HPPmR1auzd27FgeeeQRunfvTr9+/fjqq69YsGABv/nNb8wOTeSUNF/yX5oz+S/Nl/yb5kv+S/Olc8tiGIZhdhDia9GiRTzxxBNkZWUxYMAAnn76adLS0swOq12zWCz17n/xxRe55ZZbWjYYaZTLLrtMtzj2I++//z6zZs3i+++/JykpiYyMDKZOnWp2WO1eYWEhc+bMYeXKleTk5BAfH8/NN9/M3LlzCQoKMjs8kVPSfMk/ac7Uumi+5F80X/JPmi+dW0pKiYiIiIiIiIhIi1NPKRERERERERERaXFKSomIiIiIiIiISItTUkpERERERERERFqcklIiIiIiIiIiItLilJQSEREREREREZEWp6SUiIiIiIiIiIi0OCWlRERERERERESkxSkpJSIiIiIiIiIiLU5JKRGRFmSxWFi1apXZYYiIiIj4Lc2XRNoPJaVEpN245ZZbsFgsJ22jRo0yOzQRERERv6D5koi0pACzAxARaUmjRo3ixRdf9Nlnt9tNikZERETE/2i+JCItRZVSItKu2O12YmNjfbaOHTsCnlLxxYsXM3r0aIKDg+nRowdvvfWWz/Fff/01l19+OcHBwXTu3Jlbb72VoqIinzEvvPAC/fr1w263ExcXx/Tp033ez8vL49prryUkJIRevXrx3nvvnduLFhEREWkCzZdEpKUoKSUiUsecOXMYP34827ZtY+LEiUyYMIEdO3YAUFxcTHp6Oh07dmTTpk28+eab/Otf//KZRC1evJg77riDW2+9la+//pr33nuPnj17+nzGgw8+yI033sj27dsZM2YMEydO5NixYy16nSIiIiJnSvMlEWk2hohIOzF58mTDZrMZoaGhPtsjjzxiGIZhAMbvfvc7n2PS0tKMadOmGYZhGM8995zRsWNHo6ioyPv+P/7xD8NqtRpZWVmGYRhGfHy88cc//rHBGABj9uzZ3tdFRUUGYHzwwQfNdp0iIiIiZ0rzJRFpSeopJSLtyi9+8QsWL17ss69Tp07e58OHD/d5b/jw4WzduhWAHTt2kJqaSmhoqPf9iy++GLfbza5du7BYLBw+fJgrrrjilDGkpKR4n4eGhhIREUFOTs6ZXpKIiIhIs9J8SURaipJSItKuhIaGnlQe3lyCg4MbNS4wMNDntcViwe12n4uQRERERJpM8yURaSnqKSUiUseGDRtOen3BBRcAcMEFF7Bt2zaKi4u973/++edYrVZ69+5NeHg4iYmJZGZmtmjMIiIiIi1J8yURaS6qlBKRdqW8vJysrCyffQEBAURFRQHw5ptvMmTIEC655BJeffVVNm7cyPPPPw/AxIkTmTdvHpMnT+aBBx4gNzeXO++8k1//+tfExMQA8MADD/C73/2O6OhoRo8eTWFhIZ9//jl33nlny16oiIiIyBnSfElEWoqSUiLSrqxevZq4uDiffb1792bnzp2A504vr7/+OrfffjtxcXG89tpr9O3bF4CQkBA+/PBD7rrrLoYOHUpISAjjx49nwYIF3nNNnjyZsrIy/vu//5t77rmHqKgorr/++pa7QBEREZGzpPmSiLQUi2EYhtlBiIj4A4vFwsqVKxk3bpzZoYiIiIj4Jc2XRKQ5qaeUiIiIiIiIiIi0OCWlRERERERERESkxWn5noiIiIiIiIiItDhVSomIiIiIiIiISItTUkpERERERERERFqcklIiIiIiIiIiItLilJQSEREREREREZEWp6SUiIiIiIiIiIi0OCWlRERERERERESkxSkpJSIiIiIiIiIiLU5JKRERERERERERaXFKSomIiIiIiIiISIv7/4TDL1gWQsqGAAAAAElFTkSuQmCC", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "'''\n", "#jhyaudio1\n", "import matplotlib.pyplot as plt\n", "\n", "# 학습 정확도와 검증 정확도 시각화\n", "plt.figure(figsize=(12, 4))\n", "\n", "# 정확도 그래프\n", "plt.subplot(1, 2, 1)\n", "plt.plot(history.history['accuracy'], label='Train Accuracy')\n", "plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n", "plt.title('Model Accuracy')\n", "plt.xlabel('Epoch')\n", "plt.ylabel('Accuracy')\n", "plt.legend()\n", "\n", "# 손실 그래프\n", "plt.subplot(1, 2, 2)\n", "plt.plot(history.history['loss'], label='Train Loss')\n", "plt.plot(history.history['val_loss'], label='Validation Loss')\n", "plt.title('Model Loss')\n", "plt.xlabel('Epoch')\n", "plt.ylabel('Loss')\n", "plt.legend()\n", "\n", "plt.tight_layout()\n", "plt.show()\n", "'''" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\u001b[1m120/120\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 13ms/step\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "'''\n", "#jhyaudio1\n", "from sklearn.metrics import confusion_matrix\n", "import seaborn as sns\n", "import matplotlib.pyplot as plt\n", "\n", "# 모델 예측 수행\n", "y_pred = model.predict(X_test)\n", "y_pred_classes = np.argmax(y_pred, axis=1) # 예측된 클래스 (0 또는 1)\n", "y_true = np.argmax(y_test, axis=1) # 실제 클래스 (0 또는 1)\n", "\n", "# 혼동 행렬 계산\n", "conf_matrix = confusion_matrix(y_true, y_pred_classes)\n", "\n", "# 혼동 행렬 시각화\n", "plt.figure(figsize=(8, 6))\n", "sns.heatmap(conf_matrix, annot=True, fmt='d', cmap='Blues', cbar=False)\n", "plt.xlabel('Predicted Label')\n", "plt.ylabel('True Label')\n", "plt.title('Confusion Matrix')\n", "plt.show()\n", "'''\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.12.5" } }, "nbformat": 4, "nbformat_minor": 2 }