Hw11.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "4764ce34",
   "metadata": {},
   "outputs": [],
   "source": [
    "# all includes\n",
    "from tensorflow.keras.datasets import imdb\n",
    "from tensorflow.keras.preprocessing import sequence\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Embedding, SimpleRNN, Dense, Dropout, LSTM, Bidirectional\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "import matplotlib.pyplot as plt\n",
    "import time\n",
    "from tensorflow import keras\n",
    "from keras.utils import pad_sequences"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3b55028f",
   "metadata": {},
   "source": [
    "### Prepare data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "fa4719fd",
   "metadata": {},
   "outputs": [],
   "source": [
    "# define some hyperparameters\n",
    "max_features = 10000  # amount of worlds in text considered like features\n",
    "maxlen = 500  # max amount of worlds in text\n",
    "batch_size = 32"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "9bb2dd7a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/imdb.npz\n",
      "17464789/17464789 [==============================] - 4s 0us/step\n"
     ]
    }
   ],
   "source": [
    "# load data\n",
    "(x_train, y_train), (x_test, y_test) = imdb.load_data(num_words=max_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e3f5885c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ensure they all have the same length\n",
    "x_train = sequence.pad_sequences(x_train, maxlen=maxlen)\n",
    "x_test = sequence.pad_sequences(x_test, maxlen=maxlen)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "772db9ad",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(25000, 500)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train.shape\n",
    "x_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "550301bf",
   "metadata": {},
   "source": [
    "### Build RNN model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "262a3c47",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "model.add(Embedding(max_features, 32))\n",
    "model.add(SimpleRNN(32, activation='relu'))\n",
    "model.add(Dense(256, activation='relu'))\n",
    "model.add(Dropout(0.5))  # Reducing the risk of overfitting\n",
    "model.add(Dense(1, activation='sigmoid'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "2014decf",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    }
   ],
   "source": [
    "# Compile the model\n",
    "model.compile(optimizer=Adam(learning_rate=1e-4), loss='binary_crossentropy', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "3b61d948",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-04-27 20:44:30.397596: W tensorflow/tsl/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "157/157 [==============================] - 15s 93ms/step - loss: 0.6929 - accuracy: 0.5074 - val_loss: 0.6922 - val_accuracy: 0.5312\n",
      "Epoch 2/10\n",
      "157/157 [==============================] - 15s 93ms/step - loss: 0.6908 - accuracy: 0.5500 - val_loss: 0.6895 - val_accuracy: 0.5584\n",
      "Epoch 3/10\n",
      "157/157 [==============================] - 15s 95ms/step - loss: 0.6831 - accuracy: 0.6023 - val_loss: 0.6773 - val_accuracy: 0.5880\n",
      "Epoch 4/10\n",
      "157/157 [==============================] - 15s 93ms/step - loss: 0.5925 - accuracy: 0.7014 - val_loss: 0.4697 - val_accuracy: 0.7932\n",
      "Epoch 5/10\n",
      "157/157 [==============================] - 15s 93ms/step - loss: 0.4921 - accuracy: 0.7708 - val_loss: 0.5047 - val_accuracy: 0.7638\n",
      "Epoch 6/10\n",
      "157/157 [==============================] - 15s 94ms/step - loss: 0.3830 - accuracy: 0.8450 - val_loss: 0.3845 - val_accuracy: 0.8334\n",
      "Epoch 7/10\n",
      "157/157 [==============================] - 15s 93ms/step - loss: 0.3150 - accuracy: 0.8693 - val_loss: 0.3783 - val_accuracy: 0.8378\n",
      "Epoch 8/10\n",
      "157/157 [==============================] - 15s 93ms/step - loss: 0.2789 - accuracy: 0.8882 - val_loss: 0.3789 - val_accuracy: 0.8352\n",
      "Epoch 9/10\n",
      "157/157 [==============================] - 15s 94ms/step - loss: 0.2392 - accuracy: 0.9082 - val_loss: 0.3418 - val_accuracy: 0.8600\n",
      "Epoch 10/10\n",
      "157/157 [==============================] - 15s 93ms/step - loss: 0.2199 - accuracy: 0.9175 - val_loss: 0.3398 - val_accuracy: 0.8608\n"
     ]
    }
   ],
   "source": [
    "# Training the model\n",
    "history = model.fit(x_train, y_train, epochs=10, batch_size=128, validation_split=0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "69fd070e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "782/782 [==============================] - 9s 11ms/step - loss: 0.3497 - accuracy: 0.8546\n",
      "Test accuracy: 0.8546000123023987\n",
      "Test loss: 0.3496982157230377\n"
     ]
    }
   ],
   "source": [
    "# find loss and accuracy of the model on test data\n",
    "test_loss, test_acc = model.evaluate(x_test, y_test)\n",
    "print(f'Test accuracy: {test_acc}')\n",
    "print(f'Test loss: {test_loss}' )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "5141b420",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " embedding (Embedding)       (None, None, 32)          320000    \n",
      "                                                                 \n",
      " simple_rnn (SimpleRNN)      (None, 32)                2080      \n",
      "                                                                 \n",
      " dense (Dense)               (None, 256)               8448      \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 256)               0         \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 1)                 257       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 330,785\n",
      "Trainable params: 330,785\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# displaying the summary of the RNN model\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "72d4b9f7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# line graphic for loss\n",
    "plt.title('Loss : Training and Validation')\n",
    "plt.plot(history.history['loss'], label='Train Loss')\n",
    "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "84ac5e49",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# line graphic for accuracy\n",
    "plt.title('Training and Validation Accuracy')\n",
    "plt.plot(history.history['accuracy'], label='Train Accuracy')\n",
    "plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "efbda50c",
   "metadata": {},
   "source": [
    "### Deep LSTM Mode"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "7c49e85c",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-04-27 21:45:27.086124: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_2_grad/concat/split_2/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_2_grad/concat/split_2/split_dim}}]]\n",
      "2024-04-27 21:45:27.087263: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_grad/concat/split/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_grad/concat/split/split_dim}}]]\n",
      "2024-04-27 21:45:27.087879: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_1_grad/concat/split_1/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_1_grad/concat/split_1/split_dim}}]]\n",
      "2024-04-27 21:45:27.196039: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_2_grad/concat/split_2/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_2_grad/concat/split_2/split_dim}}]]\n",
      "2024-04-27 21:45:27.196669: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_grad/concat/split/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_grad/concat/split/split_dim}}]]\n",
      "2024-04-27 21:45:27.197241: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_1_grad/concat/split_1/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_1_grad/concat/split_1/split_dim}}]]\n",
      "2024-04-27 21:45:27.305734: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_2_grad/concat/split_2/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_2_grad/concat/split_2/split_dim}}]]\n",
      "2024-04-27 21:45:27.306581: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_grad/concat/split/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_grad/concat/split/split_dim}}]]\n",
      "2024-04-27 21:45:27.307179: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_1_grad/concat/split_1/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_1_grad/concat/split_1/split_dim}}]]\n"
     ]
    }
   ],
   "source": [
    "# creating a deep LSTM model\n",
    "model_deep_lstm = Sequential()\n",
    "model_deep_lstm.add(Embedding(max_features, 32))\n",
    "model_deep_lstm.add(Dropout(0.2))\n",
    "model_deep_lstm.add(LSTM(64, return_sequences=True))\n",
    "model_deep_lstm.add(Dropout(0.2))\n",
    "model_deep_lstm.add(LSTM(64, return_sequences=True))\n",
    "model_deep_lstm.add(Dropout(0.2))\n",
    "model_deep_lstm.add(LSTM(64))\n",
    "model_deep_lstm.add(Dense(1, activation='sigmoid'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "cce82313",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    }
   ],
   "source": [
    "# Compiling the deep LSTM model\n",
    "model_deep_lstm.compile(optimizer=Adam(1e-4), loss='binary_crossentropy', metrics=['acc'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "97bd64e5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-04-27 21:47:09.957378: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_2_grad/concat/split_2/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_2_grad/concat/split_2/split_dim}}]]\n",
      "2024-04-27 21:47:09.958377: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_grad/concat/split/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_grad/concat/split/split_dim}}]]\n",
      "2024-04-27 21:47:09.959281: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_1_grad/concat/split_1/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_1_grad/concat/split_1/split_dim}}]]\n",
      "2024-04-27 21:47:10.082720: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_2_grad/concat/split_2/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_2_grad/concat/split_2/split_dim}}]]\n",
      "2024-04-27 21:47:10.083558: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_grad/concat/split/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_grad/concat/split/split_dim}}]]\n",
      "2024-04-27 21:47:10.084283: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_1_grad/concat/split_1/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_1_grad/concat/split_1/split_dim}}]]\n",
      "2024-04-27 21:47:10.212232: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_2_grad/concat/split_2/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_2_grad/concat/split_2/split_dim}}]]\n",
      "2024-04-27 21:47:10.213169: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_grad/concat/split/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_grad/concat/split/split_dim}}]]\n",
      "2024-04-27 21:47:10.214095: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_1_grad/concat/split_1/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_1_grad/concat/split_1/split_dim}}]]\n",
      "2024-04-27 21:47:10.715504: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_2_grad/concat/split_2/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_2_grad/concat/split_2/split_dim}}]]\n",
      "2024-04-27 21:47:10.716339: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_grad/concat/split/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_grad/concat/split/split_dim}}]]\n",
      "2024-04-27 21:47:10.717246: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_1_grad/concat/split_1/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_1_grad/concat/split_1/split_dim}}]]\n",
      "2024-04-27 21:47:10.838278: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_2_grad/concat/split_2/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_2_grad/concat/split_2/split_dim}}]]\n",
      "2024-04-27 21:47:10.839308: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_grad/concat/split/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_grad/concat/split/split_dim}}]]\n",
      "2024-04-27 21:47:10.840169: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_1_grad/concat/split_1/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_1_grad/concat/split_1/split_dim}}]]\n",
      "2024-04-27 21:47:10.964681: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_2_grad/concat/split_2/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_2_grad/concat/split_2/split_dim}}]]\n",
      "2024-04-27 21:47:10.965657: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_grad/concat/split/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_grad/concat/split/split_dim}}]]\n",
      "2024-04-27 21:47:10.966517: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_1_grad/concat/split_1/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_1_grad/concat/split_1/split_dim}}]]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "625/625 [==============================] - ETA: 0s - loss: 0.5417 - acc: 0.6931"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-04-27 21:52:23.022395: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_2_grad/concat/split_2/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_2_grad/concat/split_2/split_dim}}]]\n",
      "2024-04-27 21:52:23.023207: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_grad/concat/split/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_grad/concat/split/split_dim}}]]\n",
      "2024-04-27 21:52:23.024334: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_1_grad/concat/split_1/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_1_grad/concat/split_1/split_dim}}]]\n",
      "2024-04-27 21:52:23.158749: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_2_grad/concat/split_2/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_2_grad/concat/split_2/split_dim}}]]\n",
      "2024-04-27 21:52:23.159545: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_grad/concat/split/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_grad/concat/split/split_dim}}]]\n",
      "2024-04-27 21:52:23.160212: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_1_grad/concat/split_1/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_1_grad/concat/split_1/split_dim}}]]\n",
      "2024-04-27 21:52:23.327677: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_2_grad/concat/split_2/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_2_grad/concat/split_2/split_dim}}]]\n",
      "2024-04-27 21:52:23.328693: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_grad/concat/split/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_grad/concat/split/split_dim}}]]\n",
      "2024-04-27 21:52:23.329514: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_1_grad/concat/split_1/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_1_grad/concat/split_1/split_dim}}]]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "625/625 [==============================] - 334s 532ms/step - loss: 0.5417 - acc: 0.6931 - val_loss: 0.3810 - val_acc: 0.8350\n",
      "Epoch 2/10\n",
      "625/625 [==============================] - 460s 737ms/step - loss: 0.3112 - acc: 0.8712 - val_loss: 0.3202 - val_acc: 0.8664\n",
      "Epoch 3/10\n",
      "625/625 [==============================] - 334s 535ms/step - loss: 0.2494 - acc: 0.9031 - val_loss: 0.3017 - val_acc: 0.8776\n",
      "Epoch 4/10\n",
      "625/625 [==============================] - 1166s 2s/step - loss: 0.2043 - acc: 0.9237 - val_loss: 0.2943 - val_acc: 0.8792\n",
      "Epoch 5/10\n",
      "625/625 [==============================] - 333s 533ms/step - loss: 0.1784 - acc: 0.9348 - val_loss: 0.3226 - val_acc: 0.8830\n",
      "Epoch 6/10\n",
      "625/625 [==============================] - 334s 535ms/step - loss: 0.1571 - acc: 0.9427 - val_loss: 0.3308 - val_acc: 0.8802\n",
      "Epoch 7/10\n",
      "625/625 [==============================] - 1311s 2s/step - loss: 0.1412 - acc: 0.9501 - val_loss: 0.3753 - val_acc: 0.8776\n",
      "Epoch 8/10\n",
      "625/625 [==============================] - 342s 548ms/step - loss: 0.1263 - acc: 0.9545 - val_loss: 0.3822 - val_acc: 0.8678\n",
      "Epoch 9/10\n",
      "625/625 [==============================] - 350s 559ms/step - loss: 0.1136 - acc: 0.9604 - val_loss: 0.3728 - val_acc: 0.8756\n",
      "Epoch 10/10\n",
      "625/625 [==============================] - 377s 604ms/step - loss: 0.1064 - acc: 0.9628 - val_loss: 0.3743 - val_acc: 0.8774\n"
     ]
    }
   ],
   "source": [
    "# training the deep LSTM model\n",
    "history_deep_lstm = model_deep_lstm.fit(x_train, y_train, epochs=10, batch_size=32, validation_split=0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "d827fda0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "782/782 [==============================] - 106s 136ms/step - loss: 0.4065 - acc: 0.8552\n",
      "Test Accuracy of deep LSTM: 0.8551999926567078\n",
      "Test Loss of deep LSTM: 0.4065380394458771\n"
     ]
    }
   ],
   "source": [
    "# find accuracy and loss of the deep LSTM model on test data\n",
    "test_loss_deep_lstm, test_acc_deep_lstm = model_deep_lstm.evaluate(x_test, y_test)\n",
    "print(f'Test Accuracy of deep LSTM: {test_acc_deep_lstm}')\n",
    "print(f'Test Loss of deep LSTM: {test_loss_deep_lstm}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "3a9e8284",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " embedding_1 (Embedding)     (None, None, 32)          320000    \n",
      "                                                                 \n",
      " dropout_1 (Dropout)         (None, None, 32)          0         \n",
      "                                                                 \n",
      " lstm (LSTM)                 (None, None, 64)          24832     \n",
      "                                                                 \n",
      " dropout_2 (Dropout)         (None, None, 64)          0         \n",
      "                                                                 \n",
      " lstm_1 (LSTM)               (None, None, 64)          33024     \n",
      "                                                                 \n",
      " dropout_3 (Dropout)         (None, None, 64)          0         \n",
      "                                                                 \n",
      " lstm_2 (LSTM)               (None, 64)                33024     \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 1)                 65        \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 410,945\n",
      "Trainable params: 410,945\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# displaying the summary of the deep LSTM model\n",
    "model_deep_lstm.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "ba0c87a4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Line graph of accuracy (deep LSTM)\n",
    "plt.title('Training and Validation Accuracy (deep LSTM)')\n",
    "plt.plot(history_deep_lstm.history['acc'], label='Training Accuracy (deep LSTM)')\n",
    "plt.plot(history_deep_lstm.history['val_acc'], label='Validation Accuracy (deep LSTM)')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "d2d308ef",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Line graph of loss (deep LSTM)\n",
    "plt.title('Training and Validation Loss (deep LSTM)')\n",
    "plt.plot(history_deep_lstm.history['loss'], label='Training Loss (deep LSTM)')\n",
    "plt.plot(history_deep_lstm.history['val_loss'], label='Validation Loss (deep LSTM)')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fb74e854",
   "metadata": {},
   "source": [
    "### Bidirectional LSTM Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "39bf8be8",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-04-28 00:34:24.986452: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_2_grad/concat/split_2/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_2_grad/concat/split_2/split_dim}}]]\n",
      "2024-04-28 00:34:24.987442: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_grad/concat/split/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_grad/concat/split/split_dim}}]]\n",
      "2024-04-28 00:34:24.988549: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_1_grad/concat/split_1/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_1_grad/concat/split_1/split_dim}}]]\n",
      "2024-04-28 00:34:25.079593: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/ReverseV2_grad/ReverseV2/ReverseV2/axis' with dtype int32 and shape [1]\n",
      "\t [[{{node gradients/ReverseV2_grad/ReverseV2/ReverseV2/axis}}]]\n",
      "2024-04-28 00:34:25.104054: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_2_grad/concat/split_2/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_2_grad/concat/split_2/split_dim}}]]\n",
      "2024-04-28 00:34:25.104738: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_grad/concat/split/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_grad/concat/split/split_dim}}]]\n",
      "2024-04-28 00:34:25.105444: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_1_grad/concat/split_1/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_1_grad/concat/split_1/split_dim}}]]\n"
     ]
    }
   ],
   "source": [
    "# Creating a Bidirectional LSTM model\n",
    "model_bidirectional = Sequential()\n",
    "model_bidirectional.add(Embedding(max_features, 32))\n",
    "model_bidirectional.add(Dropout(0.2))\n",
    "model_bidirectional.add(Bidirectional(LSTM(64)))\n",
    "model_bidirectional.add(Dense(1, activation='sigmoid'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "ebff8c42",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    }
   ],
   "source": [
    "# compiling the Bidirectional LSTM model\n",
    "model_bidirectional.compile(optimizer=Adam(1e-4), loss='binary_crossentropy', metrics=['acc'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "45b0c344",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "625/625 [==============================] - 99s 159ms/step - loss: 0.1358 - acc: 0.9532 - val_loss: 0.3417 - val_acc: 0.8844\n",
      "Epoch 2/10\n",
      "625/625 [==============================] - 100s 159ms/step - loss: 0.1089 - acc: 0.9657 - val_loss: 0.3385 - val_acc: 0.8862\n",
      "Epoch 3/10\n",
      "625/625 [==============================] - 99s 158ms/step - loss: 0.1031 - acc: 0.9672 - val_loss: 0.3227 - val_acc: 0.8852\n",
      "Epoch 4/10\n",
      "625/625 [==============================] - 5940s 10s/step - loss: 0.0954 - acc: 0.9692 - val_loss: 0.3427 - val_acc: 0.8850\n",
      "Epoch 5/10\n",
      "625/625 [==============================] - 13908s 22s/step - loss: 0.0897 - acc: 0.9718 - val_loss: 0.3692 - val_acc: 0.8826\n",
      "Epoch 6/10\n",
      "625/625 [==============================] - 8717s 14s/step - loss: 0.0827 - acc: 0.9737 - val_loss: 0.3590 - val_acc: 0.8706\n",
      "Epoch 7/10\n",
      "625/625 [==============================] - 2583s 4s/step - loss: 0.0833 - acc: 0.9728 - val_loss: 0.3402 - val_acc: 0.8814\n",
      "Epoch 8/10\n",
      "625/625 [==============================] - 102s 164ms/step - loss: 0.0684 - acc: 0.9790 - val_loss: 0.4432 - val_acc: 0.8786\n",
      "Epoch 9/10\n",
      "625/625 [==============================] - 109s 175ms/step - loss: 0.0675 - acc: 0.9798 - val_loss: 0.3754 - val_acc: 0.8750\n",
      "Epoch 10/10\n",
      "625/625 [==============================] - 100s 160ms/step - loss: 0.0610 - acc: 0.9823 - val_loss: 0.4386 - val_acc: 0.8802\n"
     ]
    }
   ],
   "source": [
    "# training the Bidirectional LSTM model\n",
    "history_bidirectional = model_bidirectional.fit(x_train, y_train, epochs=10, batch_size=32, validation_split=0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "d26a4852",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "782/782 [==============================] - 32s 41ms/step - loss: 0.5005 - acc: 0.8599\n",
      "Bidirectional LSTM Test Accuracy: 0.8599200248718262\n",
      "Bidirectional LSTM Test Loss: 0.5005342364311218\n"
     ]
    }
   ],
   "source": [
    "# Find loss and accuracy of the Bidirectional LSTM model on test data\n",
    "test_loss_bidirectional, test_acc_bidirectional = model_bidirectional.evaluate(x_test, y_test)\n",
    "print(f'Bidirectional LSTM Test Accuracy: {test_acc_bidirectional}')\n",
    "print(f'Bidirectional LSTM Test Loss: {test_loss_bidirectional}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "a10d575b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_2\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " embedding_2 (Embedding)     (None, None, 32)          320000    \n",
      "                                                                 \n",
      " dropout_4 (Dropout)         (None, None, 32)          0         \n",
      "                                                                 \n",
      " bidirectional (Bidirectiona  (None, 128)              49664     \n",
      " l)                                                              \n",
      "                                                                 \n",
      " dense_3 (Dense)             (None, 1)                 129       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 369,793\n",
      "Trainable params: 369,793\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# displaying the summary of the Bidirectional LSTM model\n",
    "model_bidirectional.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "cebd2415",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Line graph of accuracy for Bidirectional LSTM\n",
    "plt.title('Training and Validation Accuracy (Bidirectional LSTM)')\n",
    "plt.plot(history_bidirectional.history['acc'], label='Training Accuracy (Bidirectional LSTM)')\n",
    "plt.plot(history_bidirectional.history['val_acc'], label='Validation Accuracy (Bidirectional LSTM)')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "0cd6ad81",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkAAAAHFCAYAAAAaD0bAAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAACIyElEQVR4nO3dd3xN9//A8dfN3hEhA0lEjNgjMSJmEbNfoy211aqWquqiqkarqotq0WpLWpRQ1foVJapqRFHEKLWJkogEWWTe8/vjyOW6SWTfxH0/H4/zyL2fc87nfM69N7nvfKZGURQFIYQQQggTYmbsAgghhBBClDYJgIQQQghhciQAEkIIIYTJkQBICCGEECZHAiAhhBBCmBwJgIQQQghhciQAEkIIIYTJkQBICCGEECZHAiAhhBBCmBwJgB4zGo0mX9vOnTuLdJ2ZM2ei0WgKde7OnTuLpQxl3YgRI6hevXqu+2/cuIGVlRXPPvtsrsckJiZiZ2fH//73v3xfNzQ0FI1Gw6VLl/JdlgdpNBpmzpyZ7+tlu3btGjNnziQyMtJgX1E+L0VVvXp1evXqZZRrF9T58+extrZm3759urQRI0bo/e6am5tTrVo1+vfvz4kTJ/TOL8jvVk6ficK+90UVERHBzJkzuX37tsG+Dh060KFDh1IvU04uXbqERqMhNDQ0X8d9/PHHeR6XkpLCvHnzaNy4MU5OTjg6OuLn50f//v35888/AfXzm5+/6dllyn4+YsSIHK85e/Zs3TEP/o0YOnQoffr0yecr8XiwMHYBRPF68A8nwLvvvssff/zBjh079NLr1atXpOuMHj2abt26FercZs2asW/fviKXobyrXLky//vf//j555+5desWLi4uBsesWbOGu3fvMmrUqCJda/r06bz88stFyuNRrl27xqxZs6hevTpNmjTR21eUz4spee211+jSpQtBQUF66ba2trrf4czMTM6dO8d7771H69atOXXqFFWrVgWK/ru1b98+qlWrVrSbKISIiAhmzZrFiBEjqFChgt6+xYsXl3p5SkNWVhYhISEcP36c119/nRYtWgBw9uxZ/u///o/du3fTvn17NmzYQFpamu68b775hm+//ZbffvsNZ2dnXbqfn5/usaOjI+vWrePzzz/H0dFRl64oCqGhoTg5OZGYmKhXnpkzZ+Lv78+OHTt44oknSuq2yxQJgB4zrVq10nteuXJlzMzMDNIfdufOHezs7PJ9nWrVqhX6D6WTk9Mjy2MqRo0axfr161m1ahUTJkww2L9s2TLc3d3p2bNnka7z4B9HYyjK58VUnDp1ip9//pnffvvNYN/Dv8Nt2rTB29ubTp06sWnTJsaOHQsU/XcrP+fevXsXGxubUqvRe1z/Udq1axcREREsW7aM5557TpfetWtXJkyYgFarBaBp06Z652V/PgICAqhUqVKOeffu3Zv169ezZs0axowZo0vfsWMHFy9eZMyYMXz99dd65/j5+dGtWzc++OADkwmApAnMBHXo0IEGDRqwa9cuWrdujZ2dHSNHjgQgLCyMkJAQPD09sbW1pW7dukyZMoWUlBS9PHJq0shuavjtt99o1qwZtra2+Pv7s2zZMr3jcqqmHzFiBA4ODpw7d44ePXrg4OCAl5cXr776qt5/PwD//fcfTz/9NI6OjlSoUIHBgwdz8ODBfFVN37hxgxdffJF69erh4OCAm5sbTzzxBLt379Y77sEq7E8//RRfX18cHBwICgrir7/+Msg3NDSUOnXqYG1tTd26dfn+++/zLEe2rl27Uq1aNZYvX26w79SpU+zfv59hw4ZhYWFBeHg4vXv3plq1atjY2FCzZk2ef/554uLiHnmdnJo7EhMTGTNmDK6urjg4ONCtWzfOnDljcO65c+d47rnnqFWrFnZ2dlStWpUnn3yS48eP647ZuXMnzZs3B+C5557TVbFnN6fk9HnRarV8+OGH+Pv7Y21tjZubG8OGDeO///7TOy7783rw4EHatm2LnZ0dNWrU4IMPPtB9SRRVamoqU6dOxdfXFysrK6pWrcr48eMNmmR27NhBhw4dcHV1xdbWFm9vb5566inu3LmjO2bJkiU0btwYBwcHHB0d8ff356233npkGZYsWYKHhwddunTJV5mz//u3tLTUpeXWBJbfz+fDTWDZzanbtm1j5MiRVK5cGTs7O93vZFhYGEFBQdjb2+Pg4EDXrl05cuSIQb779+/nySefxNXVFRsbG/z8/Jg0aRKgfjZef/11AHx9fQ2a6XNqArt58yYvvvgiVatWxcrKiho1ajBt2jSDvxUajYYJEyawYsUK6tati52dHY0bN+bXX3/VOy4/n/HiFh8fD4Cnp2eO+83MCv/17OzsTN++fQ3+9i5btozg4GBq166d43lDhw5l+/btnD9/vtDXLk8kADJR0dHRDBkyhEGDBrF582ZefPFFQK1+7dGjh66KddKkSaxdu5Ynn3wyX/kePXqUV199lVdeeYVffvmFRo0aMWrUKHbt2vXIczMyMvjf//5Hp06d+OWXXxg5ciTz589n3rx5umNSUlLo2LEjf/zxB/PmzWPt2rW4u7szYMCAfJXv5s2bAMyYMYNNmzaxfPlyatSoQYcOHXLsN7Fo0SLCw8NZsGABq1atIiUlhR49epCQkKA7JjQ0lOeee466deuyfv163n77bd59912DZsecmJmZMWLECA4fPszRo0f19mUHRdnB6fnz5wkKCmLJkiVs27aNd955h/3799OmTRsyMjLydf/ZFEWhT58+rFixgldffZUNGzbQqlUrunfvbnDstWvXcHV15YMPPuC3335j0aJFWFhY0LJlS06fPg2oTS/Z5X377bfZt28f+/btY/To0bmW4YUXXuDNN9+kS5cubNy4kXfffZfffvuN1q1bGwR1MTExDB48mCFDhrBx40a6d+/O1KlTWblyZYHuO6/X4uOPP2bo0KFs2rSJyZMn89133/HEE0/ovlQvXbpEz549sbKyYtmyZfz222988MEH2Nvbk56eDqhNli+++KKu6eLnn3/mlVdeMfgHIiebNm2iXbt2uX7xZWZmkpmZSWpqKidOnOD111/HxcXlkbWDRfl8Zhs5ciSWlpasWLGCH3/8EUtLS95//30GDhxIvXr1WLt2LStWrCApKYm2bdty8uRJ3blbt26lbdu2REVF8emnn7Jlyxbefvttrl+/DqjNoy+99BIAP/30k+6z06xZsxzLkpqaSseOHfn++++ZPHkymzZtYsiQIXz44Yf069fP4PhNmzbxxRdfMHv2bNavX0/FihXp27cvFy5c0B2Tn894cQsMDMTS0pKXX36ZVatWER0dXaz5jxo1ir/++otTp04BcPv2bX766ac8m9M7dOiAoihs3ry5WMtSZinisTZ8+HDF3t5eL619+/YKoPz+++95nqvVapWMjAzlzz//VADl6NGjun0zZsxQHv74+Pj4KDY2Nsrly5d1aXfv3lUqVqyoPP/887q0P/74QwGUP/74Q6+cgLJ27Vq9PHv06KHUqVNH93zRokUKoGzZskXvuOeff14BlOXLl+d5Tw/LzMxUMjIylE6dOil9+/bVpV+8eFEBlIYNGyqZmZm69AMHDiiAsnr1akVRFCUrK0upUqWK0qxZM0Wr1eqOu3TpkmJpaan4+Pg8sgwXLlxQNBqNMnHiRF1aRkaG4uHhoQQHB+d4TvZ7c/nyZQVQfvnlF92+5cuXK4By8eJFXdrw4cP1yrJlyxYFUD777DO9fOfMmaMAyowZM3Itb2ZmppKenq7UqlVLeeWVV3TpBw8ezPU9ePjzcurUKQVQXnzxRb3j9u/frwDKW2+9pUvL/rzu379f79h69eopXbt2zbWc2Xx8fJSePXvmuv+3335TAOXDDz/USw8LC1MAZenSpYqiKMqPP/6oAEpkZGSueU2YMEGpUKHCI8v0sOvXryuA8sEHHxjsy/7deHjz9PRU9uzZo3fsw79bBf18PvzeZ3+Whg0bpndcVFSUYmFhobz00kt66UlJSYqHh4fSv39/XZqfn5/i5+en3L17N9f7/+ijjww+s9nat2+vtG/fXvf8yy+/zPFvxbx58xRA2bZtm979uLu7K4mJibq0mJgYxczMTJk7d26u5cntM579d+FRf2eyj/voo4/yPO7bb79VHBwc9N7TYcOGKbt27cr1nOzfpRs3buS4H1DGjx+vaLVaxdfXV3nttdcURVH/djo4OChJSUl5vt5Vq1ZVBgwYkGe5HxdSA2SiXFxccmznvXDhAoMGDcLDwwNzc3MsLS1p3749gO4/ibw0adIEb29v3XMbGxtq167N5cuXH3muRqMxqGlq1KiR3rl//vknjo6OBh1qBw4c+Mj8s3355Zc0a9YMGxsbLCwssLS05Pfff8/x/nr27Im5ubleeQBdmU6fPs21a9cYNGiQXhOPj48PrVu3zld5fH196dixI6tWrdLVJGzZsoWYmBhd7Q9AbGws48aNw8vLS1duHx8fIH/vzYP++OMPAAYPHqyXPmjQIINjMzMzef/996lXrx5WVlZYWFhgZWXF2bNnC3zdh6//8EiVFi1aULduXX7//Xe9dA8PD10n0WwPfzYKK7sm5OGyPPPMM9jb2+vK0qRJE6ysrBg7dizfffedXg3Cg+W/ffs2AwcO5JdffslX8ySoNRAAbm5uOe63tbXl4MGDHDx4kP379/PTTz9Ru3ZtevToYTDw4UHF8fkEeOqpp/Seb926lczMTIYNG6armcrMzMTGxob27dvralPPnDnD+fPnGTVqFDY2Nvm+Xl527NiBvb09Tz/9tF569vv38GenY8eOeh2B3d3dcXNz0/vslMRnPD9GjhzJf//9xw8//MDEiRPx8vJi5cqVtG/fno8++qhIeWePBFuxYgWZmZl8++239O/fHwcHhzzPc3Nz4+rVq0W6dnkhAZCJyqndOTk5mbZt27J//37ee+89du7cycGDB/npp58AtfPjo7i6uhqkWVtb5+tcOzs7gz+S1tbWpKam6p7Hx8fj7u5ucG5OaTn59NNPeeGFF2jZsiXr16/nr7/+4uDBg3Tr1i3HMj58P9bW1sD91yK7Hd/Dw8Pg3JzScjNq1Cji4+PZuHEjoDZ/OTg40L9/f0DtLxMSEsJPP/3EG2+8we+//86BAwd0/ZHy8/o+KD4+HgsLC4P7y6nMkydPZvr06fTp04f/+7//Y//+/Rw8eJDGjRsX+LoPXh9y/hxWqVJFtz9bUT5X+SmLhYUFlStX1kvXaDR4eHjoyuLn58f27dtxc3Nj/Pjx+Pn54efnx2effaY7Z+jQoSxbtozLly/z1FNP4ebmRsuWLQkPD8+zDNn3kVuQYGZmRmBgIIGBgbRo0YK+ffuyefNmLCwsmDx5cp73BkX/fD78PmU3XzVv3hxLS0u9LSwsTBf43bhxA6BYO8DHx8fj4eFh0KfMzc0NCwuLQn12SuIznl/Ozs4MHDiQzz77jP3793Ps2DHc3d2ZNm1ajtMCFMRzzz3HjRs3eP/99zl8+HC+RpPa2NiU+D2XFTIKzETlNIJjx44dXLt2jZ07d+pqfYAi/xIWJ1dXVw4cOGCQHhMTk6/zV65cSYcOHViyZIleelJSUqHLk9v181smgH79+uHi4sKyZcto3749v/76K8OGDdP9t3bixAmOHj1KaGgow4cP15137ty5Qpc7MzOT+Ph4vS+InMq8cuVKhg0bxvvvv6+XHhcXZzBkuSDXB7Uv2sNfjteuXct1dEtJyH4tbty4oRcEKYpCTEyMrnM3QNu2bWnbti1ZWVn8/ffffP7550yaNAl3d3fdfE7PPfcczz33HCkpKezatYsZM2bQq1cvzpw5o6uxe1j2/Wb3UcsPOzs7/Pz8DPqOPXxvUPTP58N/L7LL++OPP+Z6T4Du9Xy4Y3tRuLq6sn//fhRF0StXbGwsmZmZhfrslMRnvLDq16/Ps88+y4IFCzhz5oxBzWdBeHl50blzZ2bNmkWdOnXyVet38+bNfM8ZVt5JDZDQyf5jkl3Lke2rr74yRnFy1L59e5KSktiyZYte+po1a/J1vkajMbi/Y8eO5dmMkJc6derg6enJ6tWrURRFl3758mUiIiLynY+NjQ2DBg1i27ZtzJs3j4yMDL3mr+J+bzp27AjAqlWr9NJ/+OEHg2Nzes02bdpkUE3+cO1YXrKbXx/uxHzw4EFOnTpFp06dHplHccm+1sNlWb9+PSkpKTmWxdzcnJYtW7Jo0SIADh8+bHCMvb093bt3Z9q0aaSnp/PPP//kWgYfHx9sbW0LNPomOTmZc+fO5dpsBsX3+XxY165dsbCw4Pz587qaqYc3gNq1a+Pn58eyZcsMRmg9qCCfnU6dOpGcnMzPP/+sl549sq0wn538fsaLU3x8vK7J+2H//vsvoNaGFtWrr77Kk08+yfTp0x95bGZmJleuXHlspx54mNQACZ3WrVvj4uLCuHHjmDFjBpaWlqxatSrP/zBL2/Dhw5k/fz5Dhgzhvffeo2bNmmzZsoWtW7cCjx462qtXL959911mzJhB+/btOX36NLNnz8bX15fMzMwCl8fMzIx3332X0aNH07dvX8aMGcPt27eZOXNmgZoYQG0GW7RoEZ9++in+/v56/635+/vj5+fHlClTUBSFihUr8n//93+PbFrJTUhICO3ateONN94gJSWFwMBA9u7dy4oVKwyO7dWrF6Ghofj7+9OoUSMOHTrERx99ZFBz4+fnh62tLatWraJu3bo4ODhQpUqVHP+I16lTh7Fjx/L5559jZmZG9+7duXTpEtOnT8fLy4tXXnmlUPeVm5iYGH788UeD9OrVq9OlSxe6du3Km2++SWJiIsHBwRw7dowZM2bQtGlThg4dCqh9x3bs2EHPnj3x9vYmNTVVN8y4c+fOAIwZMwZbW1uCg4Px9PQkJiaGuXPn4uzsrFeT9DArK6tcp1gAtQk0e59Wq+Xq1assXLiQW7du5Tlzc3F+Ph9UvXp1Zs+ezbRp07hw4QLdunXDxcWF69evc+DAAezt7Zk1axagjqR88sknadWqFa+88gre3t5ERUWxdetWXQDesGFDAD777DOGDx+OpaUlderU0eu7k23YsGEsWrSI4cOHc+nSJRo2bMiePXt4//336dGjh+69KIj8fsYL6vjx4zl+7po3b87Bgwd5+eWXGTx4MK1bt8bV1ZXY2FhWr17Nb7/9xrBhw4ql6TAkJISQkJB8HXvs2DHu3Lmj+wfpsWfULtiixOU2Cqx+/fo5Hh8REaEEBQUpdnZ2SuXKlZXRo0crhw8fNhj5kNsosJxG2zw8iiO3UWAPlzO360RFRSn9+vVTHBwcFEdHR+Wpp55SNm/ebDAaKidpaWnKa6+9plStWlWxsbFRmjVrpvz8888Go6TyGsVBDqOkvvnmG6VWrVqKlZWVUrt2bWXZsmUGeeZH06ZNcxyRpCiKcvLkSaVLly6Ko6Oj4uLiojzzzDNKVFRUriN38hoFpiiKcvv2bWXkyJFKhQoVFDs7O6VLly7Kv//+a5DfrVu3lFGjRilubm6KnZ2d0qZNG2X37t0G76uiKMrq1asVf39/xdLSUi+fnN7HrKwsZd68eUrt2rUVS0tLpVKlSsqQIUOUK1eu6B2X2+c1v6+vj49PjqOoAGX48OGKoqijFd98803Fx8dHsbS0VDw9PZUXXnhBuXXrli6fffv2KX379lV8fHwUa2trxdXVVWnfvr2yceNG3THfffed0rFjR8Xd3V2xsrJSqlSpovTv3185duzYI8v57bffKubm5sq1a9cM7vPhcru5uSnt27dXNmzYoHdsTr9bipL/z2dun6WDBw/mWOaff/5Z6dixo+Lk5KRYW1srPj4+ytNPP61s375d77h9+/Yp3bt3V5ydnRVra2vFz89Pb3SVoijK1KlTlSpVqihmZmZ695DT5yw+Pl4ZN26c4unpqVhYWCg+Pj7K1KlTldTUVIP7GT9+vEG5fXx8dO+9ouT/M17QUWC5bcuXL1euXLmivP3220pwcLDi4eGhWFhYKI6OjkrLli2Vzz//XG/06YPyOwosL7mNAps+fbpSqVIlg9fxcaVRlAfqRYUop95//33efvttoqKiZMZhUS6lpqbi7e3Nq6++yptvvmns4ggTk5WVRc2aNRk0aBBz5swxdnFKhTSBiXLniy++ANRmoYyMDHbs2MHChQsZMmSIBD+i3LKxsWHWrFnMnDmTCRMmYG9vb+wiCROycuVKkpOTdbNymwIJgES5Y2dnx/z587l06RJpaWl4e3vz5ptv8vbbbxu7aEIUydixY7l9+zYXLlzQ9YsRojRotVpWrVpV6qPejEmawIQQQghhcmQYvBBCCCFMjgRAQgghhDA5EgAJIYQQwuRIJ+gcaLVarl27hqOjY45LRgghhBCi7FEUhaSkJKpUqfLIiXElAMrBtWvX8PLyMnYxhBBCCFEIV65ceeS0KBIA5SB7+vUrV67g5ORk5NIIIYQQIj8SExPx8vLKcRmVh0kAlIPsZi8nJycJgIQQQohyJj/dV6QTtBBCCCFMjgRAQgghhDA5EgAJIYQQwuRIH6AiyMrKIiMjw9jFEEIIHSsrq0cO/xVClIEAaPHixXz00UdER0dTv359FixYQNu2bR953t69e2nfvj0NGjQgMjJSlx4aGspzzz1ncPzdu3exsbEpljIrikJMTAy3b98ulvyEEKK4mJmZ4evri5WVlbGLIkSZZtQAKCwsjEmTJrF48WKCg4P56quv6N69OydPnsTb2zvX8xISEhg2bBidOnXi+vXrBvudnJw4ffq0XlpxBT+ALvhxc3PDzs5OJksUQpQJ2ZO4RkdH4+3tLX+bhMiDUQOgTz/9lFGjRjF69GgAFixYwNatW1myZAlz587N9bznn3+eQYMGYW5uzs8//2ywX6PR4OHhUSJlzsrK0gU/rq6uJXINIYQorMqVK3Pt2jUyMzOxtLQ0dnGEKLOM1lCcnp7OoUOHCAkJ0UsPCQkhIiIi1/OWL1/O+fPnmTFjRq7HJCcn4+PjQ7Vq1ejVqxdHjhzJsyxpaWkkJibqbbnJ7vNjZ2eXZ55CCGEM2U1fWVlZRi6JEGWb0QKguLg4srKycHd310t3d3cnJiYmx3POnj3LlClTWLVqFRYWOVde+fv7ExoaysaNG1m9ejU2NjYEBwdz9uzZXMsyd+5cnJ2ddVt+lsGQqmUhRFkkf5uEyB+jDxV4+JdVUZQcf4GzsrIYNGgQs2bNonbt2rnm16pVK4YMGULjxo1p27Yta9eupXbt2nz++ee5njN16lQSEhJ025UrVwp/Q0IIIYQo84wWAFWqVAlzc3OD2p7Y2FiDWiGApKQk/v77byZMmICFhQUWFhbMnj2bo0ePYmFhwY4dO3K8jpmZGc2bN8+zBsja2lq37IUsf1EwHTp0YNKkSfk+/tKlS2g0Gr2Re6YgPT2dmjVrsnfv3gKdN3PmTJo0aZLnMSNGjKBPnz665wV9T4qTsa6t0Why7A9Ylr322mtMnDjR2MUQwmQZLQCysrIiICCA8PBwvfTw8HBat25tcLyTkxPHjx8nMjJSt40bN446deoQGRlJy5Ytc7yOoihERkbi6elZIvdRXmg0mjy3ESNGFCrfn376iXfffTffx3t5eREdHU2DBg0Kdb38KmuB1tKlS/Hx8SE4OFiX9uDrb2Fhgbe3N5MnTyYtLU13zGuvvcbvv/9eoGsV9D0pjJ07d6LRaAymgiiNaxfGw0Hiw44cOUKvXr1wc3PDxsaG6tWrM2DAAOLi4pg5c+Yjf38uXbqkO65bt24G+X/44YdoNBo6dOigS3vjjTdYvnw5Fy9eLIE7FkI8ilFHgU2ePJmhQ4cSGBhIUFAQS5cuJSoqinHjxgFq09TVq1f5/vvvMTMzM/jSzP5j9WD6rFmzaNWqFbVq1SIxMZGFCxcSGRnJokWLSvXeypro6Gjd47CwMN555x29qQJsbW31js/IyMjXCJKKFSsWqBzm5uYlNkKvLPv888+ZOXOmQfry5cvp1q0bGRkZHD16lOeeew57e3tdEOHg4ICDg0OBrvWo9yQ9Pb3E5ogp6OehLIiNjaVz5848+eSTbN26lQoVKnDx4kU2btzInTt3eO2113R/kwCaN2/O2LFjGTNmjC6tcuXKAHh6evLHH3/w33//Ua1aNd3+5cuXG0zt4ebmRkhICF9++SXz5s0r4bsUpCWB9aNXCBemw6h9gAYMGMCCBQuYPXs2TZo0YdeuXWzevBkfHx9A/dKOiooqUJ63b99m7Nix1K1bl5CQEK5evcquXbto0aJFSdxCueHh4aHbnJ2ddVMFeHh4kJqaSoUKFVi7di0dOnTAxsaGlStXEh8fz8CBA6lWrRp2dnY0bNiQ1atX6+X7cJNH9erVef/99xk5ciSOjo54e3uzdOlS3f6Ha2ayaxJ+//13AgMDsbOzo3Xr1gbzOL333nu4ubnh6OjI6NGjmTJlyiObhvKSlpbGxIkTdUF0mzZtOHjwoG7/rVu3GDx4MJUrV8bW1pZatWqxfPlyQA0gJkyYgKenp662IK9pGw4fPsy5c+fo2bOnwb4KFSrg4eGBl5cXvXr14n//+x+HDx/W7X+4CSwrK4vJkydToUIFXF1deeONN1AURS/PnN6T9957jxEjRuDs7Kz74o6IiKBdu3bY2tri5eXFxIkTSUlJ0XuN3njjDby8vLC2tqZWrVp8++23XLp0iY4dOwLg4uKiV4P48LVv3brFsGHDcHFxwc7Oju7du+s1R4eGhlKhQgW2bt1K3bp1cXBwoFu3bnoB+8GDB+nSpQuVKlXC2dmZ9u3b671GRRUREUFiYiLffPMNTZs2xdfXlyeeeIIFCxbg7e2Ng4OD3u+Pubk5jo6OBmlwP6j57rvv9PKPi4vL8f3/3//+Z/A7JUrA4e9hbjX4vezVTgrjMXon6BdffJFLly6RlpbGoUOHaNeunW5faGgoO3fuzPXcmTNnGjRxzJ8/n8uXL5OWlkZsbCxbt24lKCiohEqvUhSFO+mZRtke/vIrijfffJOJEydy6tQpunbtSmpqKgEBAfz666+cOHGCsWPHMnToUPbv359nPp988gmBgYEcOXKEF198kRdeeIF///03z3OmTZvGJ598wt9//42FhQUjR47U7Vu1ahVz5sxh3rx5HDp0CG9vb5YsWVKke33jjTdYv3493333HYcPH6ZmzZp07dqVmzdvAjB9+nROnjzJli1bOHXqFEuWLKFSpUoALFy4kI0bN7J27VpOnz7NypUrqV69eq7X2rVrF7Vr135k37IzZ87wxx9/5NqcC+pru2zZMr799lv27NnDzZs32bBhwyPv96OPPqJBgwYcOnSI6dOnc/z4cbp27Uq/fv04duwYYWFh7NmzhwkTJujOGTZsGGvWrGHhwoWcOnWKL7/8EgcHB7y8vFi/fj0Ap0+fJjo6ms8++yzH644YMYK///6bjRs3sm/fPhRFoUePHnpLyNy5c4ePP/6YFStWsGvXLqKionjttdd0+5OSkhg+fDi7d+/mr7/+olatWvTo0YOkpKRH3nd+eHh4kJmZyYYNG4rl92nkyJGEhobqni9btozBgwfnWOvWokULrly5wuXLl4t8XZGLu7ch/B318e6P4chKoxZHlB1GXwrjcXA3I4t672w1yrVPzu6KnVXxvI2TJk2iX79+emkPfhG99NJL/Pbbb6xbty7PL+kePXrw4osvAmpQNX/+fHbu3Im/v3+u58yZM4f27dsDMGXKFHr27Elqaio2NjZ8/vnnjBo1SrfEyTvvvMO2bdtITk4u1H2mpKSwZMkSQkND6d69OwBff/014eHhfPvtt7z++utERUXRtGlTAgMDAfQCnKioKGrVqkWbNm3QaDS6GsvcXLp0iSpVquS4b+DAgZibm5OZmUlaWhq9evVi6tSpuea1YMECpk6dylNPPQXAl19+ydatj/7sPfHEE3rv5bBhwxg0aJCutqZWrVosXLiQ9u3bs2TJEqKioli7di3h4eF07twZgBo1aujOz27qcnNzo0KFCjle8+zZs2zcuJG9e/fq+vWtWrUKLy8vfv75Z5555hlAbW798ssv8fPzA2DChAnMnj1br+wP+uqrr3BxceHPP/+kV69ej7z3R2nVqhVvvfUWgwYNYty4cbRo0YInnniCYcOG5Tgg41F69erFuHHj2LVrFwEBAaxdu5Y9e/awbNkyg2OrVq0KqJ+RR32ORCHt/gTu3gJLO8i4A/83CVyqQ/U2xi6ZMDKj1wCJsiP7yz5bVlYWc+bMoVGjRri6uuLg4MC2bdse2SzZqFEj3ePsprbY2Nh8n5PdYT37nNOnTxs0YRalSfP8+fNkZGTodUi2tLSkRYsWnDp1CoAXXniBNWvW0KRJE9544w29yTlHjBhBZGQkderUYeLEiWzbti3P6+W1Dt38+fOJjIzk6NGj/Prrr5w5c4ahQ4fmeGxCQgLR0dF6NZoWFhYG71tOHj7m0KFDhIaG6voYOTg40LVrV7RaLRcvXiQyMhJzc3NdUFoYp06dwsLCQi9YdnV1pU6dOrrXGdRJRbODH1Df/wc/L7GxsYwbN47atWvr5upKTk4ucPN4XubMmUNMTAxffvkl9erV48svv8Tf35/jx48XOC9LS0uGDBnC8uXLWbduHbVr19b7fD8ou+/dnTt3ilR+kYtbl2H/V+rjp5dD/b6gzYCwIRB/3rhlM3VntkG6cT/3UgNUDGwtzTk5u6vRrl1c7O3t9Z5/8sknzJ8/nwULFtCwYUPs7e2ZNGkS6enpeebzcOdpjUaDVqvN9znZ80A9eE5O80UVVva5ec1B1b17dy5fvsymTZvYvn07nTp1Yvz48Xz88cc0a9aMixcvsmXLFrZv307//v3p3LkzP/74Y47Xq1SpUq5fpB4eHtSsWROAOnXqkJSUxMCBA3nvvfd06cXh4fdWq9Xy/PPP5zgM29vbm3PnzhX5mrm9Rw/P9ZXT5+XBc0eMGMGNGzdYsGABPj4+WFtbExQU9MjPYUG5urryzDPP8MwzzzB37lyaNm3Kxx9/rNefJ79GjhxJy5YtOXHihF5z7sOym1yzO1GLYrbjXchKA992ULsr1GivBkXXDsPqZ2FUONhWMHYpTc/JjbB2GHi1gKEbwMr+0eeUAKkBKgYajQY7KwujbCU56+vu3bvp3bu3bmLJGjVq5DmfUkmpU6cOBw4c0Ev7+++/C51fzZo1sbKyYs+ePbq0jIwM/v77b+rWratLq1y5MiNGjGDlypUsWLBArzO3k5MTAwYM4OuvvyYsLIz169frvswe1rRpU/799998BW3ZnWnv3r1rsM/Z2RlPT0/++usvXVpmZiaHDh169E0/pFmzZvzzzz/UrFnTYLOysqJhw4ZotVr+/PPPHM/Pz3IL9erVIzMzU6/PWHx8PGfOnNF7nR9l9+7dTJw4kR49elC/fn2sra2Ji4vL9/mFYWVlhZ+fn16n8IKoX78+9evX58SJEwwaNCjX406cOIGlpSX169cvbFFFbq4eguPrAA2EvAcaDVjawsDV4FQV4s7AuhGQlWnskpqWS3tg/WhAAbe6atOkkUgNkMhVzZo1Wb9+PREREbi4uPDpp58SExNToC+v4vDSSy8xZswYAgMDad26NWFhYRw7dkyvT0puHh5NBuoX8wsvvMDrr79OxYoV8fb25sMPP+TOnTuMGjUKUPsZBQQEUL9+fdLS0vj111919z1//nw8PT1p0qQJZmZmrFu3Dg8Pj1z7wnTs2JGUlBT++ecfg6kcbt++TUxMDFqtlrNnzzJ79mxq166d62v88ssv88EHH1CrVi3q1q3Lp59+ajAXT368+eabtGrVivHjxzNmzBjs7e05deoU4eHhfP7551SvXp3hw4czcuRIFi5cSOPGjbl8+TKxsbH0798fHx8fNBoNv/76Kz169MDW1tZguH6tWrXo3bs3Y8aM4auvvsLR0ZEpU6ZQtWpVevfune+y1qxZkxUrVhAYGEhiYiKvv/66wbQN+ZGQkGAwaKJixYocO3aMNWvW8Oyzz1K7dm0UReH//u//2Lx5s27kX2Hs2LGDjIyMXD8XoAZ3bdu2LdT9iDwoCmybrj5u/Cx4Nr6/z9FDDYKWdYMLf8CWN6DnJ2qAJErW9X9g9SC1Vs6/F/T81KivuwRAIlfTp0/n4sWLdO3aFTs7O8aOHUufPn1ISEgo1XIMHjyYCxcu8Nprr5Gamkr//v0ZMWKEQa1QTp599lmDtIsXL/LBBx+g1WoZOnQoSUlJBAYGsnXrVlxcXAC1BmDq1KlcunQJW1tb2rZty5o1awB1bp558+Zx9uxZzM3Nad68OZs3b8bMLOcKVVdXV/r168eqVasMhstnd+zO7ivVrl073n///VzXunv11VeJjo5mxIgRmJmZMXLkSPr27Vvg96RRo0b8+eefTJs2jbZt26IoCn5+fgwYMEB3zJIlS3jrrbd48cUXiY+Px9vbm7feegtQO+/OmjWLKVOm8NxzzzFs2DC9kU/Zli9fzssvv0yvXr1IT0+nXbt2bN68uUCrlC9btoyxY8fStGlTvL29ef/99/U6dOfXzp07adq0qV7a8OHDeeedd7Czs+PVV1/lypUruiH/33zzTa79sfLj4WbHnKxevZpZs2YV+hoiF6e3wOW9YGEDT7xtuN+zMfT7Wu0L9Pe3ULkOtHy+9MtpSm5HwcqnIC0BvIPgqW/ArPi6cBSGRinOcdSPicTERJydnUlISDAYupyamsrFixfx9fXNtWOrKHldunTBw8ODFStWGLso+XL8+HE6d+7MuXPncHSUydgEbNq0iddff51jx47lGvAWhsn/jcrKgMVBEH8W2kyGzjNyP3bPAtg+AzRmMGgd1OpcasU0KXduwrch6ntSuS6M3AK2LiVyqby+vx8mfYBEmXfnzh0+/fRT/vnnH/79919mzJjB9u3bGT58uLGLlm8NGzbkww8/5NKlS8YuiigjUlJSWL58ebEGPwI4FKp+0dpVgjav5H1s8MvQZAgoWvjxOYjNe74yUQjpd+CH/up74lQNhqwvseCnoOQ3T5R5Go2GzZs3895775GWlkadOnVYv369bn6a8qI8BWyi5PXv39/YRXj8pCbCzg/Uxx2mgM0jFrbWaKDXfLh1UW0y+6E/jNkB9pVKvqymICtTDSz/Owg2FdTgx7mqsUulIwGQKPNsbW3Zvn27sYshhCjr9i6AO3HgWgsCRuTvHAsr6L8CvnkCbl2CNYNh+EawsC7BgpoARYFfX4Yzv6l9sQatBbfcJ8M1BmkCE0IIUf4l/Af77i163WUWmOe/oz32ruoXtLUzXPkL/u9l9QtcFN6O99RlRzRm6iSU3rmvHmAsEgAJIYQo/3a8B5mp4BMMdXoU/PzKdeCZ5aAxh6OrYc/84i+jqdi/VF13DaDXAvAvxPtRCiQAEkIIUb5FH4Wj6jQVhLxb+LllanaC7vPUx7/PUmcsFgXzzwZ1biWAjtMgoOz2fZQASAghRPmlKLDtbUCBBk9D1YCi5ddiDLQYqz7e8DxciyxqCU3Hxd3w01hAgeajod3rxi5RniQAEkIIUX6dDYeLu8DcCjq9Uzx5dp0Lfp3U1eNXPwuJ0cWT7+Ms5jisGQRZ6VD3Sej+YZmfXVsCICGEEOVTViaE31vyouU4cPEpnnzNLdT+QJX9ISlaDYKMvHJ5mXbr8r1ZnhPVPlj9jD/Lc35IACQKpEOHDkyaNEn3vHr16ixYsCDPczQaDT///HORr11c+ZQnp0+fxsPDg6SkpAKd9/D7lJOH3ztjvr7GuPbOnTvRaDSFWkvNmJo3b85PP/1k7GKUDZEr4ca/6sR6bV8t3rxtnGHgGrBzhehI+HkcaLXFe43HQUo8rOwHydfBrT48+wNYlo8ZyCUAMhFPPvlkrhMH7tu3D41Gw+HDhwuc78GDBxk7dmxRi6dn5syZNGnSxCA9Ojqa7t27F+u1HhYaGprn4pWlbdq0aYwfP163fEb2l3b2ZmtrS/369fVWqgf46aefePfddwt0rdJ4fY353hbGowL89evX07JlS5ydnXF0dKR+/fq8+qr6RdyhQwe99+rhrXr16nrHffDBBwb59+jRA41Gw8yZM3Vp06dPZ8qUKWhN/cs4LRl2zFEft38TbCsU/zUq+sKAlWBmCSd/gT/mFP81yrP0FPjhGYg/B85eMOTHknkfSogEQCZi1KhR7Nixg8uXLxvsW7ZsGU2aNKFZs2YFzrdy5crY2dkVRxEfycPDA2tr05mc7L///mPjxo26BVMfdPr0aaKjozl58iTPP/88L7zwAr///rtuf8WKFQu85tijXt+MjIwC5Vec1y6Ltm/fzrPPPsvTTz/NgQMHOHToEHPmzCE9PR1Qg9Do6Giio6N1C/du375dl3bw4EFdXl5eXgYrz1+7do0dO3bg6empl96zZ08SEhLYunVrCd9hGRexEFJiwcUXAkeV3HV8WsP/FqqPd38MR8NK7lrlSVYGrB0OVw+pNXBDfgKnKsYuVYFIAGQievXqhZubm8GK3Xfu3CEsLIxRo0YRHx/PwIEDqVatGnZ2djRs2JDVq1fnme/D/yGfPXuWdu3aYWNjQ7169QgPDzc4580336R27drY2dlRo0YNpk+frvtyDQ0NZdasWRw9elT3n3J2mR9uJjl+/DhPPPEEtra2uLq6MnbsWJKTk3X7R4wYQZ8+ffj444/x9PTE1dWV8ePHF+mLPCoqit69e+Pg4ICTkxP9+/fn+vXruv1Hjx6lY8eOODo64uTkREBAAH///TcAly9f5sknn8TFxQV7e3vq16/P5s2bc73W2rVrady4MdWqVTPY5+bmhoeHB76+vkycOJHq1avr1eA93AQWGxvLk08+ia2tLb6+vqxatcogzwdf30uXLqHRaFi7di0dOnTAxsaGlStXAuoK73Xr1sXGxgZ/f38WL16sl89///3Hs88+S8WKFbG3tycwMJD9+/eX+nu7cuVKAgMDcXR0xMPDg0GDBhEbG5vr611Qv/76K23atOH111+nTp061K5dmz59+vD5558DahDq4eGBh4cHlStXBsDV1dUgDdTfz/j4ePbu3atLCw0NJSQkBDc3N73rmpub06NHj0f+bj7WEqMhQn2d6TJLnc25JDUZdH9dsY0TIGp/yV6vrFMU2DgRzoWDha26kGzl2sYuVYHJUhjFQVHU0QLGYGmXr572FhYWDBs2jNDQUN555x00985Zt24d6enpDB48mDt37hAQEMCbb76Jk5MTmzZtYujQodSoUYOWLR89i6dWq6Vfv35UqlSJv/76i8TExBz7oTg6OhIaGkqVKlU4fvw4Y8aMwdHRkTfeeIMBAwZw4sQJfvvtN93yF87OzgZ53Llzh27dutGqVSsOHjxIbGwso0ePZsKECXpB3h9//IGnpyd//PEH586dY8CAATRp0oQxY8Y88n4epigKffr0wd7enj///JPMzExefPFFBgwYwM6dOwEYPHgwTZs2ZcmSJZibmxMZGYmlpToj7fjx40lPT2fXrl3Y29tz8uRJHBwccr3erl27CAwMfGSZtm7dypUrV/J8j0aMGMGVK1fYsWMHVlZWTJw4MV/BwJtvvsknn3zC8uXLsba25uuvv2bGjBl88cUXNG3alCNHjjBmzBjs7e0ZPnw4ycnJtG/fnqpVq7Jx40Y8PDw4fPgwWq221N/b9PR03n33XerUqUNsbCyvvPIKI0aMyDPoLAgPDw9++OEHTpw4QYMGDYqUl5WVFYMHD2b58uUEBwcDagD04Ycf6jV/ZWvRogUffvhhka5Zrv0xR/2b69US6v6vdK75xDsQdxb+/VUd7TRmR/F1ui5vfp8FR39QJ418JhS8mhu7RIUiAVBxyLgD7xup6u+ta2Bln69DR44cyUcffcTOnTvp2LEjoDZ/9evXDxcXF1xcXHjttdd0x7/00kv89ttvrFu3Ll8B0Pbt2zl16hSXLl3S1Vq8//77Bn073n77bd3j6tWr8+qrrxIWFsYbb7yBra0tDg4OWFhY4OHhkeu1Vq1axd27d/n++++xt1fv/4svvuDJJ59k3rx5uLu7A+Di4sIXX3yBubk5/v7+9OzZk99//71QAdD27ds5duwYFy9exMvLC4AVK1ZQv359Dh48SPPmzYmKiuL111/H319d86ZWrVq686Oionjqqado2LAhADVq1MjzepcuXSIgIOc5TbJf37S0NLRaLbNnz6Zdu3Y5HnvmzBm2bNnCX3/9pXsfv/32W+rWrfvIe540aRL9+vXTPX/33Xf55JNPdGm+vr6cPHmSr776iuHDh/PDDz9w48YNDh48SMWKFQGoWbOm7vzSfG9Hjhypy7NGjRosXLiQFi1akJycnGfgmV8vvfQSu3fvpmHDhvj4+NCqVStCQkIYPHhwoZrzRo0aRZs2bfjss884dOgQCQkJ9OzZM8cAqGrVqkRFRaHVajEzM7GK/JgT6hILACHvld5QazMz6LcUlnWDmGPwwwAYte3RC64+bv768v4s2f9bCHW6Gbc8RWBivzmmzd/fn9atW7Ns2TIAzp8/z+7du3VfFFlZWcyZM4dGjRrh6uqKg4MD27ZtIyoqKl/5nzp1Cm9vb70mm6CgIIPjfvzxR9q0aYOHhwcODg5Mnz4939d48FqNGzfWfUECBAcHo9VqOX36tC6tfv36mJvfH47p6elZ6GaQU6dO4eXlpQt+AOrVq0eFChU4deoUAJMnT2b06NF07tyZDz74gPPnz+uOnThxIu+99x7BwcHMmDGDY8eO5Xm9u3fvYmOT82iK3bt3ExkZSWRkJN988w3vv/8+S5YsybXcFhYWerVJ/v7++ers/eA5N27c4MqVK4waNQoHBwfd9t577+nuMzIykqZNm+qCn8Iorvf2yJEj9O7dGx8fHxwdHenQoQNAgT9rubG3t2fTpk2cO3eOt99+GwcHB1599VVatGjBnTsFrxFu1KgRtWrV4scff2TZsmUMHTpUV3v4MFtbW7RaLWlpaUW9jfIn/B1AgXp9wKtF6V7byl4dGebgATdOwfpRoM0q3TIY04n18NsU9fET06HpEOOWp4ikBqg4WNqpNTHGunYBjBo1igkTJrBo0SKWL1+Oj48PnTp1AuCTTz5h/vz5LFiwgIYNG2Jvb8+kSZN0nTofRclh8UDNQ/+d/fXXXzz77LPMmjWLrl274uzszJo1a/jkk08KdB+KohjkndM1H/4C0Wg0hR49k9s1H0yfOXMmgwYNYtOmTWzZsoUZM2awZs0a+vbty+jRo+natSubNm1i27ZtzJ07l08++YSXXnopx+tVqlSJW7du5bjP19dXF8DUr1+f/fv3M2fOHF544YUcyweG70V+PBiEZL9uX3/9tUGNYHYgYmtrW+BrPKw43tuUlBRCQkIICQlh5cqVVK5cmaioKLp27Zrvz3N++fn54efnx+jRo5k2bRq1a9cmLCwsx87rjzJy5EgWLVrEyZMndR2nc3Lz5k3s7OyK5fUuV879Dud/V0dldZ5hnDI4V4WBq2F5Dzi7TZ2Futtc45SlNF34EzaMAxR1puzinnbACKQGqDhoNOp/BsbYCvil1r9/f8zNzfnhhx/47rvveO6553RfKrt376Z3794MGTKExo0bU6NGDc6ePZvvvOvVq0dUVBTXrt0PBvft26d3zN69e/Hx8WHatGkEBgZSq1Ytg5FpVlZWZGXl/V9VvXr1iIyMJCUlRS9vMzMzatcumc542fd35coVXdrJkydJSEjQa06qXbs2r7zyCtu2baNfv356o3u8vLwYN24cP/30E6+++ipff/11rtdr2rQpJ0+ezFfZzM3NuXv3bo776tatS2Zmpq4zNqijyAo6/427uztVq1blwoUL1KxZU2/z9fUF1FqMyMhIbt68mWMepfXe/vvvv8TFxfHBBx/Qtm1b/P39i7UDdG6qV6+OnZ2dXtkLYtCgQRw/fpwGDRpQr169XI87ceJEoUZtlmvarHu1P6hfwBXzbkIuUVWbQd8v1cd/LYa/lxmvLKUh+iisGazO8lyvD3T7oMzP8pwfEgCZGAcHBwYMGMBbb73FtWvXGDFihG5fzZo1CQ8PJyIiglOnTvH8888TExOT77w7d+5MnTp1GDZsGEePHmX37t1MmzZN75iaNWsSFRXFmjVrOH/+PAsXLmTDhg16x1SvXp2LFy8SGRlJXFxcjtX8gwcPxsbGhuHDh3PixAn++OMPXnrpJYYOHarrI1JYWVlZuual7O3kyZN07tyZRo0aMXjwYA4fPsyBAwcYNmwY7du3JzAwkLt37zJhwgR27tzJ5cuX2bt3LwcPHtQFR5MmTWLr1q1cvHiRw4cPs2PHjjz74XTt2pV9+/blGDDExsYSExPD5cuXWbduHStWrKB379455lOnTh26devGmDFj2L9/P4cOHWL06NGFqj2YOXMmc+fO5bPPPuPMmTMcP36c5cuX8+mnnwIwcOBAPDw86NOnD3v37uXChQusX79eFwiX1nvr7e2NlZUVn3/+ORcuXGDjxo0Fnhcp29WrVw0+Dzdv3mTmzJm88cYb7Ny5k4sXL3LkyBFGjhxJRkYGXbp0KdS1XFxciI6O1pvSICe7d+8mJCSkUNcot46uhusn1AkK27326ONLWv0+0PFef8ZNr8GFncYsTcm5dQlWPg3pSVC9LfT9qlzM8pwfEgCZoFGjRnHr1i06d+6Mt7e3Ln369Ok0a9aMrl270qFDB90XWX6ZmZmxYcMG0tLSaNGiBaNHj2bOHP2Jw3r37s0rr7zChAkTaNKkCREREUyfPl3vmKeeeopu3brRsWNHKleunONwXzs7O7Zu3crNmzdp3rw5Tz/9NJ06deKLL74o2IuRg+TkZJo2baq3ZU9I9/PPP+Pi4kK7du3o3LkzNWrUICxMnRfE3Nyc+Ph4hg0bRu3atenfvz/du3dn1qxZgBpYjR8/nrp169KtWzfq1KljMIT8QT169MDS0lI3YupBderUwdPTk5o1a/Lmm2/y/PPP64Zf52T58uV4eXnRvn17+vXrx9ixYw2GV+fH6NGj+eabbwgNDaVhw4a0b9+e0NBQXQ2QlZUV27Ztw83NjR49etCwYUM++OADXRNZab23lStXJjQ0lHXr1lGvXj0++OADPv744wLfL8DHH39s8HnYuHEj7du358KFCwwbNgx/f3+6d+9OTEwM27Zto06dOoW6FkCFChX0mh4fdvXqVSIiIgrVxFZupafAjvfUx+1eB7vC9zErVu1eg4b9QcmCtcPUUWKPk5Q4WNFPnW/JvSE8u6rczPKcHxolp44bJi4xMRFnZ2cSEhJwctLv4Z+amsrFixfx9fXNtYOqEMVl8eLF/PLLLzLpndB5/fXXSUhIMJj9O9tj+Tfqz4/gj/eggjdM+BssytCkmRmp8P3/4Mp+tVlu9O9lJ0ArirRk+O5JuHYYnL1hdDg45j56s6zI6/v7YVIDJEQZNnbsWNq1a1fgtcDE48vNza3QzXnlUtL1+8OuO80oW8EPqDUiA1apQcLNC2pNUGbxdrQvdVkZ6n1cOwy2FWHoT+Ui+CkoCYCEKMMsLCyYNm1agZe1EI+v119/vcj93MqVnXMhIwWqBkCDp4xdmpw5VIZBa8DKAS7ths2vqhPklkdaLfwyQR1tZ2kHg9dBpVqPPq8ckgBICCFE2RT7Lxz+Tn1cmpMeFoZ7fXh6GWjM4PD3sG+RsUtUONtnwLE16izP/b+HannPRl+eSQAkhBCibNo+AxQt+PdSFyUt62p3hZB7Az+2vQ2ntxi3PAW1b5G6yCxA7y+gVuFGM5YXEgAVkvQdF0KURY/N36YLf8KZ38DMAjrPMnZp8q/VCxAwAlBg/Wh16Y7y4PiPsPUt9XHnmeoCsI85owdAixcv1o1WCAgIYPfu3fk6b+/evVhYWNCkSRODfevXr6devXpYW1tTr149g3lmiiJ79tnCTHUvhBAlLXum6weXCSl3tFq1BgUgcCRUqpn38WWJRgM9PgbfdpCeDKufheSSn4SzSM7vuDfLM9DyBQieZNTilBajLoURFhbGpEmTWLx4McHBwXz11Vd0796dkydP6s1P87CEhASGDRtGp06duH79ut6+ffv2MWDAAN5991369u3Lhg0b6N+/P3v27MnXgp6PYm5uToUKFXSzytrZ2RVqiQEhhChuWq2WGzduYGdnh4VFOV7p6PhadcFRaydo/6axS1Nw5pZq/5lvOkP8OXX1+OG/ls05dK4dgbChoM2A+v2g6/tlu69VMTLqPEAtW7akWbNmeos41q1blz59+jB3bu5rqzz77LPUqlULc3Nzfv75ZyIjI3X7BgwYQGJiIlu23G977datGy4uLjlOupaTR80joCgKMTExBV5KQAghSpqZmRm+vr5YWVkZuyiFk3EXPg+ExP/Uppg2rxi7RIUXfx6+fgJSb0ODp+Gpb8pWcHHzAnwbAik3wLe9OuKrrE0zUEAFmQfIaP8ipKenc+jQIaZMmaKXHhISQkRERK7nLV++nPPnz7Ny5Uree+89g/379u3jlVf0f2G6du3KggULcs0zLS1Nb0r+xMTEPMuu0Wjw9PTEzc2NjIyMPI8VQojSZGVlhZmZ0Xs3FN5fS9Tgx6katBxn7NIUjasfDFgBK/rCiR+hUm3oUEZqtJJj783yfAM8GsKAleU++CkoowVAcXFxZGVlGcxn4e7unuv6U2fPnmXKlCns3r071+rdmJiYAuUJMHfuXN1yBQVhbm5evtvZhRCiLEmJg93qunJ0egcsH4PV7n3bQc9P4f8mws731Tl1GvQzbpnSkmDV03DrIlTwgcHrwSbv2pLHkdH/TXi4/4yiKDn2qcnKymLQoEHMmjXrkStC5zfPbFOnTiUhIUG3PbjatxBCiFLy5zx10U3PxtDwGWOXpvgEDIegCerjn1+A/w4ZryyZ6Wqfn+ijYFcJhm4ARxOaWPMBRqsBqlSpEubm5gY1M7GxsTnOcpqUlMTff//NkSNHmDBB/SBptVoURcHCwoJt27bxxBNP4OHhke88s1lbW2NtbVpVf0IIUabEnYW/l6mPQ96D8tyMl5Mus9UO0Wd+gzUDYcwOcK5WumXQauGXF+HCH2Bpr/b5cfUr3TKUIUb7hFlZWREQEEB4eLheenh4OK1bG0545eTkxPHjx4mMjNRt48aNo06dOkRGRupGeAUFBRnkuW3bthzzFEIIUUZsnwnaTKjdTW02etyYmaudoN3qQ/J1+OFZdcHR0hQ+HY6vU+dWGvA9VG1WutcvY4w6TnLy5MkMHTqUwMBAgoKCWLp0KVFRUYwbp3Z8mzp1KlevXuX777/HzMyMBg0a6J3v5uaGjY2NXvrLL79Mu3btmDdvHr179+aXX35h+/bt7Nmzp1TvTQghRD5djoB/f1WXX+gy29ilKTnWjuqaYV8/AdePw09j1c7HpVHbFfE57PtCfdx7MdTsXPLXLOOMWsc4YMAAFixYwOzZs2nSpAm7du1i8+bN+Pj4ABAdHU1UVFSB8mzdujVr1qxh+fLlNGrUiNDQUMLCwoplDiAhhBDFTKuFrdPUxwHDoXId45anpFXwhmdXg7k1nN4Ev88s+WseDbs/sWSXd6HxgJK/Zjlg1HmAyqqCzCMghBCiCI7/COtHqSupTzwCDm7GLlHpOLYOfhqtPu69CJoOKZnrnNsOPwxQmxeDJkDXOSVznTKiIN/fj1kvMyGEEOVGRipsvzcFSfAk0wl+ABo9c3+W6/+bBJdKoJvG1cMQNkwNfho+o9b+CB0JgIQQQhjHgaWQEAWOnhA03tilKX3tp0D9vuoyFGFD1JmZi0v8eVj1DGSkQI2Oar+fx21kXRHJqyGEEKL03bkJuz9WHz8xHazsjFseYzAzgz5LoEozuHtLbaq6e7vo+SZdV2efvhOnzqk0YAVYlNOlUUqQBEBCCCFK366PIDUB3BtA42eNXRrjsbSFgavBqSrEnYF1IyArs/D5pSaqszzfvgwuvjD4R3X0mTAgAZAQQojSdfMCHPhafRzyrjpHjilz9ICBa9TJCS/8Ab8Vcr2wzDS1KS3mGNhXhqE/mVa/qgKSAEgIIUTp2j5L7fdSszP4PWHs0pQNno3gqa8BDRz8BvYvLdj5Wi1sGAcX/1RH1A1eBxVrlEhRHxcSAAkhhCg9Vw7AyZ9BYyajkh7m3xM6z1Qf//YmnN2ev/MUBba+Bf/8BGaWap+fKk1LrJiPCwmAhBBClA5FuT/pYZPB4F7PuOUpi4JfhiZDQNHCj89B7L+PPmfvZ7B/ifq475dSq5ZPEgAJIYQoHSd/gf8OgKUddJxm7NKUTRoN9JoPPsGQlgg/9IeUuNyPj1wN22eoj7u+Dw2fLp1yPgYkABJCCFHyMtPVBU8BWk8EJ0+jFqdMs7CC/ivApbo6mitsiNrB+WFnw+GXe/MntX7JNOdSKgIJgIQQQpS8v7+FWxfBwV39shZ5s3eFQWvB2hmi9qmzRT+4ctV/h2DtMFCyoNEA6PwYLyJbQiQAEkIIUbLu3oI/56mPO74F1g7GLU95UbkOPLMcNOZw9AfYM19NjzsHPzwDGXfAr5O6lpjM8lxg8ooJIYQoWbs/UYOgynXVDr4i/2p2gu73gsffZ8HBb2FlX7gTr4706v89mFsat4zllARAQgghSs6ty7D/K/VxyLtgbmHc8pRHLcZAi7Hq402T4XaUOsfPoHVSm1YEEgAJIYQoOb/Phqx08G2vTnwoCqfrXLW5C8DeDYb8BA6VjVumck5CcSGEECXj6iE48SOggZD31CHeonDMLaD/d+qw91qdoaKvsUtU7kkAJIQQovgpCmybrj5uPFBd6kEUjbUjtBxr7FI8NqQJTAghRPE7vRku7wULG3jibWOXRggDEgAJIcoORYH0O8YuhSiqrAwIf0d9HDQenKsatzxC5ECawIQQxpeRCkdXQ8RCuHkBnKqCe31wqwfuDdQ1o1xrqTPkirLvUCjEnwO7ShA8ydilESJHEgAJIYwnNUGd1+SvJZASez898aq6nd12P83MUp0Yzq2eGhxlb46e0rm2LElNgJ1z1ccdp4KNk3HLI0QuJAASQpS+xGj4azH8vRzSk9Q0p2rQegLU7wu3LsH1E3D9H7h+Uv2ZnnQv7QQcfyAvWxdwyw6I7tUYVfaX+VGMZc8CdZK+SrWh2XBjl0aIXEkAJIQoPXHnIOIzOLpGnRsG1NmB20yCBk/dn9HW0QO8W90/T1Eg4cq9gOjE/aAo/pw6w/DlPer2IBdf/Zoit/rq0GEz81K5VZN0+4oa2AJ0niUzFIsyTQIgIUTJ++8Q7J0Pp34F7i3o6B2k9g+pFfLodYw0GqjgrW51ut9Pz0iFuNP3AqJ7NUaxJyH5urrw5q2L8O+v94+3sAW3uvdrirL7GNm7Fvcdm6Yd70FmKvi00X+fhCiDJAASQpQMRYHzv6tNIpd230+v00MNfLxbFv0aljbg2VjdHpQSd6+26N4W+w/EnoLMu3DtsLo9yMFdv6bIvb7a38jCuuhlNBXXIuFYmPo45F3plyXKPAmAhBDFKysTTv4MexdAzL3OOmYW0LA/BL8Mbv4lXwb7SlCjvbpl02bBzYtqTVHsyfvNabcuqTVGydfh/I77x2vMoVItw9Fozl7y5f4wRYFtbwMKNHwGqjYzdomEeCQJgIQQxSP9DkSugojP4fZlNc3SHgJGQNCL4FzNqMXDzBwq1VS3+n3up6clQey/ai3Rg7VGqbfhxr/qxvr7x1s7q4HQg6PR3OqZ9mins9vUWj5za3hiurFLI0S+SAAkhCiaOzfVoez7v4Q7cWqanSu0HAfNR4NdReOW71GsHcGrubplUxRIvHavpuiB0WhxpyEtAaL2qduDnL0f6HRdDzwag6vf419blJV5f8mLVuPAxce45REinyQAEkIUTsJ/sG+xOuldRoqaVsEbWk+EJoPBys6oxSsSjUadvdi5KtTqcj89Mx3izxqORku6BglR6nZmy/3jXWtBvd5qjZN7g8czGDqyQg0MbStCm8nGLo0Q+aZRFEUxdiHKmsTERJydnUlISMDJyYSrtYXISey/sPczOL4WtJlqmntDdSh7vT7qqtWm5s7Ne7VFD9QYxRy7P9QfoGIN9fWp11vttP04BENpSbCwmTqJZbd5ag2QEEZUkO9vCYByIAGQEDmI2q92bD69+X5a9bbqiK6anR6PL/TilJoIZ7aqHcLPhkNW2v19LtXVQKheH6jStPy+djvmwK4P1eDuxf2yVIkwOgmAikgCICHuURS1g+ue+Q/0edFA3V4Q/ApUCzBq8cqNtCT9YCgz9f6+Ct73gqG+6uip8hIMJV5Ta38y70L/FVDvf8YukRASABWVBEDC5GVlwIn1alNX7Ek1zcwSGj+rDmWvVMu45SvP0pLVoPLkL+rPjDv39zl73QuGekPVwEdPEGlMv4yHIyvBqxWM/K38BG7isVaQ72+j/3YtXrwYX19fbGxsCAgIYPfu3bkeu2fPHoKDg3F1dcXW1hZ/f3/mz5+vd0xoaCgajcZgS01NzSVXIYROeoq6MOnCprDheTX4sXJUOzZPOg69v5Dgp6isHaBBP+j/Hbx+Dvp/D/X7qVMGJFyBfV/At11gQQPYMgWi/gKt1til1hdzAo6sUh+HvCfBjyiXjNpbMSwsjEmTJrF48WKCg4P56quv6N69OydPnsTb29vgeHt7eyZMmECjRo2wt7dnz549PP/889jb2zN27FjdcU5OTpw+fVrvXBsbmxK/HyHKrZR4OLAUDnylrq0FYO8GrV6AwJFgW8GoxXtsWdnfr/HJuAvntqs1Q6e3QOJV2L9E3Rw9oe7/1OO8Wxl/PbPw6YCiLlz74PQBQpQjRm0Ca9myJc2aNWPJkiW6tLp169KnTx/mzp2brzz69euHvb09K1asANQaoEmTJnH79u1Cl0uawITJuB0FEV/A4e/VvhygLiIaPBEaD1KXmhClLyNVnZX65M9qMJSWeH+fg/v9YMindekHQ+e2w8qn1CbRCQfVBWaFKCMK8v1ttBqg9PR0Dh06xJQpU/TSQ0JCiIiIyFceR44cISIigvfee08vPTk5GR8fH7KysmjSpAnvvvsuTZs2zTWftLQ00tLuj9BITEzM9VghHgsxJ9T+PSfWg5Klpnk2hjavqF+uxq5hMHWWNuDfQ90y0+D8H2ow9O9mdcmOg1+rm31lqPukOprMJ7jkpyDQZsG2d9THLZ+X4EeUa0YLgOLi4sjKysLd3V0v3d3dnZiYmDzPrVatGjdu3CAzM5OZM2cyevRo3T5/f39CQ0Np2LAhiYmJfPbZZwQHB3P06FFq1cq578LcuXOZNWtW0W9KiLJMUeByhDqU/ey2++k1OqhD2Wt0kL4cZZGFNdTppm6Z6XBhp9pM9u+vkHID/l6mbnaV1NF59XpD9XYlEwxF/qAuGWLjDG1fLf78hShFRmsCu3btGlWrViUiIoKgoCBd+pw5c1ixYgX//vtvrudevHiR5ORk/vrrL6ZMmcIXX3zBwIEDczxWq9XSrFkz2rVrx8KFC3M8JqcaIC8vL2kCE48HrVadu2fvAvjvoJqmMVO/KINfVuehEeVPZjpc2gX//KwGQ9l9t0Cdldm/pzoDtW97MLcs+vXSU9Rh78kxEDIHWk8oep5CFLNy0QRWqVIlzM3NDWp7YmNjDWqFHubrq1a7NmzYkOvXrzNz5sxcAyAzMzOaN2/O2bNnc83P2toaa2vrAt6BEGVcZro6W/PezyDujJpmbg1NBkHrl9R1qkT5ZWEFNTurW6/56mKkJ3+BU/8Hd+LVJSqOrACbCuDf634wVNjJCvctUoOfCj7QYkxx3okQRmG0AMjKyoqAgADCw8Pp27evLj08PJzevXvnOx9FUfRqb3LaHxkZScOGDYtUXiHKjbQkdX2ufYvVNapAXcG8+Sh1gVLHvP/BEOWQuSX4PaFuPT6By3vuB0MpNyBypbrZOEOdnmrtn19HtXktP5Kuw54F6uPOM/J/nhBlmFGHwU+ePJmhQ4cSGBhIUFAQS5cuJSoqinHj1PVkpk6dytWrV/n+++8BWLRoEd7e3vj7+wPqvEAff/wxL730ki7PWbNm0apVK2rVqkViYiILFy4kMjKSRYsWlf4NClGakmPVFdkPfgOpCWqaoye0ehECRoCNNOeaBHMLtT9XjQ7Q42O139fJn9VgKPk6HP1B3aydoE53tQO13xN5j/jb+b664G3VQHXOIiEeA0YNgAYMGEB8fDyzZ88mOjqaBg0asHnzZnx8fACIjo4mKipKd7xWq2Xq1KlcvHgRCwsL/Pz8+OCDD3j++ed1x9y+fZuxY8cSExODs7MzTZs2ZdeuXbRo0aLU70+IEqfNUps+jq2DEz/eX2LBtZY6lL3RAPlv3ZSZmYNvW3Xr/qE6qeLJX+DURkiKhmNh6mblqHayrtdbbVKztL2fR+y/6jQJIJMeiseKLIWRgxKbByjmBHzXC6wd1f++rB0f2pzySHdU/4O3dgRLO/kjZMoUBa4ehuPr4J+f1P/qs1UNUIey1+lZtpdREMal1cJ/B9QO1Cd/ud9UCuqM1LW7qn2GanaBdSPg7FZ1uP2AlUYqsBD5I2uBFVGJBUCX9kJoj6LnozHLJYhyyuXxA2k2D6Rb2suXZHly47Qa9Bz/EW5dvJ9u66I2YzQaoM4SLMGxKAitFq7+rQZCJ39Rl+PIZmGrTpBpZgHjD0jHeVHmSQBURCUWAGXcVWfeTU1UZ3ZNS3poS8zl8QPPleJcE0iTe21TfoIoZy91XSNRcm5fUScrPPEjxBy/n25ppw5zbvC02n+jsCN7hHhQdu3iyQ1qMHT7XheEFs9Djw+NWzYh8kECoCIqs0thKIq6cnR2UFTgQCrx/nnZs/8WhZmlWuPg9wTU7ATuDaVGqTikxKudVo//CFEPzIpuZqH2z2j4jNp51creaEUUJkBR4NoRiD0FDZ+WvmSiXJAAqIjKbABUXBRF7SyrC6QSHhFEPfj83s+7tyH1tn6+9pXvDcXtpP50qGyMuyuf0pLVyQqPr1PXgNJm3tuhUZc4aPi02kHVrqJRiymEEGVZuZgIURiRRqOO8rC0BQe3wucTf179sj63HS7uVucbyR5VAuraUn6d1Nohr5bFMxvt4yQzDc79rgY9p7fcX4wU1Neu4TPqkGPnqsYroxBCPKakBigHj30NUEnITIMr+9Uv9PO/6/dXAXWYrW87qHmvhshUF1HUZsHlvWrQc/KX+/P1AFT0U4Oehk9DpZzXrRNCCJE7aQIrIgmAikHSdbV26Pzv6s878fr7K9a4XztUve3j3Zk6uy/F8R/VYetJ0ff3OXpCg6fUrUpTGcElhBBFIAFQEUkAVMy0Wog5eq92aIdaU6Tr48L9ztQ1O6lBkUfDxyMQuHFGHb11fB3cvHA/3cZZHbbe8Gm1f4+ZudGKKIQQjxMJgIpIAqASlpoIF3eptUPnfofbl/X3O7g/0Jm6I9hXMk45CyPh6v1h69FH76db2Kojtxo+owZ6MqJGCCGKnQRARSQBUClSFLV2JLvv0MVd6lB/HY3aITi7dsirRdnrTH3n5r1h6+vV/j3c+5Uys1ADuYbPQJ0ej3cznxBClAESABWRBEBGlJmmrld0/nc4twOu59CZukb7+3MPuVQ3SjFJT1FHbh1fp46Ce7BJz7v1vWHrfcDe1TjlE0IIEyQBUBFJAFSGJMXcG2r/O1z4w7AztWvNBzpTtynZyQEz09WyHF+nztnzYE2VR8P7w9YreJVcGYQQQuRKAqAikgCojNJqITryfu3Qlf36M1qbW92bmfpeQOTeoOidqbVadTbm7GHrd2/d3+fie3/YeuU6RbuOEEKIIpMAqIgkAConUhPUPkPZ/Yey1y3K5uBxv6msRsf8N0cpitqB+fg6OPGT/krZDu73hq0/DVWbPR6j1YQQ4jEhAVARSQBUDinKvZmpf1f75FzaY9iZukrT+52pqzUH84cmQo87d2/Y+o8Qf/Z+urUz1HtSre2p3laGrQshRBklAVARSQD0GMhMg6h99+ceun5Cf7+1072ZqTtB+h018Ll25P5+Cxuo3U0Nemp1kWHrQghRDkgAVEQSAD2GEqMfmJn6D7h70/AYjbk671D2sHUbee+FEKI8kcVQhXiYkyc0Haxu2iy1M/W5HerIMo2ZutJ6/b7la9JFIYQQhSYBkDA9ZuZQNUDd2r9u7NIIIYQwAjNjF0AIIYQQorRJACSEEEIIkyMBkBBCCCFMjgRAQgghhDA5EgAJIYQQwuRIACSEEEIIkyMBkBBCCCFMjgRAQgghhDA5EgAJIYQQwuRIACSEEEIIkyMBkBBCCCFMjgRAQgghhDA5EgAJIYQQwuQYPQBavHgxvr6+2NjYEBAQwO7du3M9ds+ePQQHB+Pq6oqtrS3+/v7Mnz/f4Lj169dTr149rK2tqVevHhs2bCjJWxBCCCFEOWPUACgsLIxJkyYxbdo0jhw5Qtu2benevTtRUVE5Hm9vb8+ECRPYtWsXp06d4u233+btt99m6dKlumP27dvHgAEDGDp0KEePHmXo0KH079+f/fv3l9ZtCSGEEKKM0yiKohjr4i1btqRZs2YsWbJEl1a3bl369OnD3Llz85VHv379sLe3Z8WKFQAMGDCAxMREtmzZojumW7duuLi4sHr16nzlmZiYiLOzMwkJCTg5ORXgjoQQQghhLAX5/jZaDVB6ejqHDh0iJCRELz0kJISIiIh85XHkyBEiIiJo3769Lm3fvn0GeXbt2jXfeQohhBDi8WdhrAvHxcWRlZWFu7u7Xrq7uzsxMTF5nlutWjVu3LhBZmYmM2fOZPTo0bp9MTExBc4zLS2NtLQ03fPExMSC3IoQQgghyhmjd4LWaDR6zxVFMUh72O7du/n777/58ssvWbBggUHTVkHznDt3Ls7OzrrNy8urgHchhBBCiPLEaDVAlSpVwtzc3KBmJjY21qAG52G+vr4ANGzYkOvXrzNz5kwGDhwIgIeHR4HznDp1KpMnT9Y9T0xMlCBICCGEeIwZrQbIysqKgIAAwsPD9dLDw8Np3bp1vvNRFEWv+SooKMggz23btuWZp7W1NU5OTnqbEEIIIR5fRqsBApg8eTJDhw4lMDCQoKAgli5dSlRUFOPGjQPUmpmrV6/y/fffA7Bo0SK8vb3x9/cH1HmBPv74Y1566SVdni+//DLt2rVj3rx59O7dm19++YXt27ezZ8+e0r9BIYQQQpRJRg2ABgwYQHx8PLNnzyY6OpoGDRqwefNmfHx8AIiOjtabE0ir1TJ16lQuXryIhYUFfn5+fPDBBzz//PO6Y1q3bs2aNWt4++23mT59On5+foSFhdGyZctSvz8hhBBClE1GnQeorJJ5gIQQQojyp1zMAySEEEIIYSwSAAkhhBDC5EgAJIQQQgiTIwGQEEIIIUyOBEBCCCGEMDkSAAkhhBDC5EgAJIQQQgiTIwGQEEIIIUyOBEBCCCGEMDkSAAkhhBDC5EgAJIQQQgiTIwGQEEIIIUyOBEBCCCGEMDkSAAkhhBDC5EgAJIQQQgiTIwGQEEIIIUyOBEBCCCGEMDkSAAkhhBDC5EgAJIQQQgiTIwGQEEIIIUyOBEBCCCGEMDkSAAkhhBDC5EgAJIQQQgiTIwGQEEIIIUxOoQKgK1eu8N9//+meHzhwgEmTJrF06dJiK5gQQgghREkpVAA0aNAg/vjjDwBiYmLo0qULBw4c4K233mL27NnFWkAhhBBCiOJWqADoxIkTtGjRAoC1a9fSoEEDIiIi+OGHHwgNDS3O8gkhhBBCFLtCBUAZGRlYW1sDsH37dv73v/8B4O/vT3R0dPGVTgghhBCiBBQqAKpfvz5ffvklu3fvJjw8nG7dugFw7do1XF1di7WAQgghhBDFrVAB0Lx58/jqq6/o0KEDAwcOpHHjxgBs3LhR1zQmhBBCCFFWaRRFUQpzYlZWFomJibi4uOjSLl26hJ2dHW5ubsVWQGNITEzE2dmZhIQEnJycjF0cIYQQQuRDQb6/C1UDdPfuXdLS0nTBz+XLl1mwYAGnT58u98GPEEIIIR5/hQqAevfuzffffw/A7du3admyJZ988gl9+vRhyZIlxVpAIYQQQojiVqgA6PDhw7Rt2xaAH3/8EXd3dy5fvsz333/PwoULC5TX4sWL8fX1xcbGhoCAAHbv3p3rsT/99BNdunShcuXKODk5ERQUxNatW/WOCQ0NRaPRGGypqakFv1EhhBBCPJYKFQDduXMHR0dHALZt20a/fv0wMzOjVatWXL58Od/5hIWFMWnSJKZNm8aRI0do27Yt3bt3JyoqKsfjd+3aRZcuXdi8eTOHDh2iY8eOPPnkkxw5ckTvOCcnJ6Kjo/U2GxubwtyqEEIIIR5DhQqAatasyc8//8yVK1fYunUrISEhAMTGxhao0/Cnn37KqFGjGD16NHXr1mXBggV4eXnl2oy2YMEC3njjDZo3b06tWrV4//33qVWrFv/3f/+nd5xGo8HDw0NvE0IIIYTIVqgA6J133uG1116jevXqtGjRgqCgIECtDWratGm+8khPT+fQoUO64ClbSEgIERER+cpDq9WSlJRExYoV9dKTk5Px8fGhWrVq9OrVy6CGSAghhBCmzaIwJz399NO0adOG6Oho3RxAAJ06daJv3775yiMuLo6srCzc3d310t3d3YmJiclXHp988gkpKSn0799fl+bv709oaCgNGzYkMTGRzz77jODgYI4ePUqtWrVyzCctLY20tDTd88TExHxdXwghhBDlU6ECIEDXtPTff/+h0WioWrVqoSZB1Gg0es8VRTFIy8nq1auZOXMmv/zyi97Q+1atWtGqVSvd8+DgYJo1a8bnn3+eawftuXPnMmvWrAKXXQghhBDlU6GawLRaLbNnz8bZ2RkfHx+8vb2pUKEC7777LlqtNl95VKpUCXNzc4PantjYWINaoYeFhYUxatQo1q5dS+fOnfM81szMjObNm3P27Nlcj5k6dSoJCQm67cqVK/m6ByGEEEKUT4WqAZo2bRrffvstH3zwAcHBwSiKwt69e5k5cyapqanMmTPnkXlYWVkREBBAeHi4XrNZeHg4vXv3zvW81atXM3LkSFavXk3Pnj0feR1FUYiMjKRhw4a5HmNtba1b3FUIIYQQj79CBUDfffcd33zzjW4VeIDGjRtTtWpVXnzxxXwFQACTJ09m6NChBAYGEhQUxNKlS4mKimLcuHGAWjNz9epV3aSLq1evZtiwYXz22We0atVKV3tka2uLs7MzALNmzaJVq1bUqlWLxMREFi5cSGRkJIsWLSrMrQohhBDiMVSoAOjmzZv4+/sbpPv7+3Pz5s185zNgwADi4+OZPXs20dHRNGjQgM2bN+Pj4wNAdHS03pxAX331FZmZmYwfP57x48fr0ocPH05oaCigzkw9duxYYmJicHZ2pmnTpuzatUsWaRVCCCGETqEWQ23ZsiUtW7Y06FT80ksvceDAAfbv319sBTQGWQxVCCGEKH8K8v1dqBqgDz/8kJ49e7J9+3aCgoLQaDRERERw5coVNm/eXKhCCyGEEEKUlkKNAmvfvj1nzpyhb9++3L59m5s3b9KvXz/++ecfli9fXtxlFEIIIYQoVoVqAsvN0aNHadasGVlZWcWVpVFIE5gQQghR/hTk+7tQNUBCCCGEEOWZBEBCCCGEMDkSAAkhhBDC5BRoFFi/fv3y3H/79u2ilEUIIYQQolQUKADKnm05r/3Dhg0rUoGEEEIIIUpagQIgGeIuhBBCiMeB9AESQgghhMmRAEgIIYQQJkcCICGEEEKYHAmAhBBCCGFyJAASQgghhMmRAEgIIYQQJkcCICGEEEKYHAmAhBBCCGFyJAASQgghhMmRAEgIIYQQJkcCICGEEEKYHAmAhBBCCGFyJAASQgghhMmRAEgIIYQQJkcCICGEEEKYHAmAhBBCCGFyJAASQgghhMmRAEgIIYQQJkcCICGEEEKYHAmAhBBCCGFyJAASQgghhMmRAEgIIYQQJkcCICGEEEKYHAmAhBBCCGFyjB4ALV68GF9fX2xsbAgICGD37t25HvvTTz/RpUsXKleujJOTE0FBQWzdutXguPXr11OvXj2sra2pV68eGzZsKMlbEEIIIUQ5Y9QAKCwsjEmTJjFt2jSOHDlC27Zt6d69O1FRUTkev2vXLrp06cLmzZs5dOgQHTt25Mknn+TIkSO6Y/bt28eAAQMYOnQoR48eZejQofTv35/9+/eX1m0JIYQQoozTKIqiGOviLVu2pFmzZixZskSXVrduXfr06cPcuXPzlUf9+vUZMGAA77zzDgADBgwgMTGRLVu26I7p1q0bLi4urF69Ol95JiYm4uzsTEJCAk5OTgW4IyGEEEIYS0G+v41WA5Sens6hQ4cICQnRSw8JCSEiIiJfeWi1WpKSkqhYsaIubd++fQZ5du3aNd95CiGEEOLxZ2GsC8fFxZGVlYW7u7teuru7OzExMfnK45NPPiElJYX+/fvr0mJiYgqcZ1paGmlpabrniYmJ+bq+EEIIIcono3eC1mg0es8VRTFIy8nq1auZOXMmYWFhuLm5FSnPuXPn4uzsrNu8vLwKcAdCCCGEKG+MFgBVqlQJc3Nzg5qZ2NhYgxqch4WFhTFq1CjWrl1L586d9fZ5eHgUOM+pU6eSkJCg265cuVLAuxFCCCFEeWK0AMjKyoqAgADCw8P10sPDw2ndunWu561evZoRI0bwww8/0LNnT4P9QUFBBnlu27Ytzzytra1xcnLS24QQQgjx+DJaHyCAyZMnM3ToUAIDAwkKCmLp0qVERUUxbtw4QK2ZuXr1Kt9//z2gBj/Dhg3js88+o1WrVrqaHltbW5ydnQF4+eWXadeuHfPmzaN379788ssvbN++nT179hjnJoUQQghR5hi1D9CAAQNYsGABs2fPpkmTJuzatYvNmzfj4+MDQHR0tN6cQF999RWZmZmMHz8eT09P3fbyyy/rjmndujVr1qxh+fLlNGrUiNDQUMLCwmjZsmWp358QQgghyiajzgNUVsk8QEIIIUT5Uy7mARJCCCGEMBYJgIQQQghhciQAEkIIIYTJkQBICCGEECZHAiAhhBBCmBwJgIQQQghhciQAEkIIIYTJkQBICCGEECZHAiAhhBBCmBwJgIQQQghhciQAEkIIIYTJkQBICCGEECZHAiAhhBBCmBwJgIQQQghhciQAEkIIIYTJkQBICCGEECZHAiAhhBBCmBwJgIQQQghhciQAEkIIIYTJkQBICCGEECZHAiAhhBBCmBwJgIQQQghhciQAEkIIIYTJkQBICCGEECZHAiAhhBBCmBwJgIQQQghhciQAEkIIIYTJkQBICCGEECZHAiAhhBBCmBwJgIQQQghhciQAEkIIIYTJkQBICCGEECZHAiAhhBBCmByjB0CLFy/G19cXGxsbAgIC2L17d67HRkdHM2jQIOrUqYOZmRmTJk0yOCY0NBSNRmOwpaamluBdCCGEEKI8MWoAFBYWxqRJk5g2bRpHjhyhbdu2dO/enaioqByPT0tLo3LlykybNo3GjRvnmq+TkxPR0dF6m42NTUndhhBCCCHKGaMGQJ9++imjRo1i9OjR1K1blwULFuDl5cWSJUtyPL569ep89tlnDBs2DGdn51zz1Wg0eHh46G1CCCGEENmMFgClp6dz6NAhQkJC9NJDQkKIiIgoUt7Jycn4+PhQrVo1evXqxZEjR4qUnxBCCCEeL0YLgOLi4sjKysLd3V0v3d3dnZiYmELn6+/vT2hoKBs3bmT16tXY2NgQHBzM2bNncz0nLS2NxMREvU0IIYQQjy+jd4LWaDR6zxVFMUgriFatWjFkyBAaN25M27ZtWbt2LbVr1+bzzz/P9Zy5c+fi7Oys27y8vAp9fSGEEEKUfUYLgCpVqoS5ublBbU9sbKxBrVBRmJmZ0bx58zxrgKZOnUpCQoJuu3LlSrFdXwghhBBlj9ECICsrKwICAggPD9dLDw8Pp3Xr1sV2HUVRiIyMxNPTM9djrK2tcXJy0tuEEEII8fiyMObFJ0+ezNChQwkMDCQoKIilS5cSFRXFuHHjALVm5urVq3z//fe6cyIjIwG1o/ONGzeIjIzEysqKevXqATBr1ixatWpFrVq1SExMZOHChURGRrJo0aJSvz8hhBBClE1GDYAGDBhAfHw8s2fPJjo6mgYNGrB582Z8fHwAdeLDh+cEatq0qe7xoUOH+OGHH/Dx8eHSpUsA3L59m7FjxxITE4OzszNNmzZl165dtGjRotTuSwghhBBlm0ZRFMXYhShrEhMTcXZ2JiEhQZrDhBBCiHKiIN/fRh8FJoQQQghR2iQAEkIIIYTJkQBICCGEECZHAiAhhBBCmBwJgEpZfHKasYsghBBCmDwJgErRrZR02n34B6O/O8ihy7eMXRwhhBDCZBl1HiBTs+dcHHcysth+Kpbtp2Jp6VuRFzvWpF2tSkVa/0wIIYQQBSPzAOWgJOcBOhebzFd/nmfDkatkatWXvkFVJ15oX5NuDTwwN5NASAghhCiMgnx/SwCUg9KYCPHa7bt8vfsCaw5c4W5GFgA1KtnzfPsa9G1aDSsLaZ0UQgghCkICoCIqzZmgb6akE7r3IqERl0hMzQTAw8mG0W19GdjCG3traaUUQggh8kMCoCIyxlIYyWmZ/LD/Mt/svkhskjpSrIKdJSNaV2dE6+pUsLMqlXIIIYQQ5ZUEQEVkzLXA0jKzWH/oKl/tOs/l+DsA2FmZM6iFN6Pb1sDD2aZUyyOEEEKUFxIAFVFZWAw1M0vL5hMxLNl5nlPRiQBYmZvRr1lVnm/vh28le6OUSwghhCirJAAqorIQAGVTFIWdp2+weOc5Dl5S5w4y00D3hp682MGP+lWcjVo+IYQQoqyQAKiIylIA9KCDl26y+I9z/HH6hi6tfe3KjO9Ykxa+FY1YMiGEEML4JAAqorIaAGU7eS2RJX+eZ9Oxa9ybSohAHxde6ODHE/5uMqmiEEIIkyQBUBGV9QAo26W4FL7adYH1h/4jPUsLgL+HIy908KNnQ08szGUuISGEEKZDAqAiKi8BULbrial8u+ciq/66TEq6Oqmid0U7xrarwdMB1bCxNDdyCYUQQoiSJwFQEZW3AChbwp0Mvtt3ieV7L3LrTgYAlR2tGdXGl8EtvXG0sTRyCYUQQoiSIwFQEZXXACjbnfRM1hy4wte7LxCdkAqAk40Fw4Kq81xwdVwdrI1cQiGEEKL4SQBUROU9AMqWnqnl58irfPnneS7cSAHAxtKMZ5t7M6ZdDapWsDVyCYUQQojiIwFQET0uAVC2LK3Ctn9iWLzzPMevJgBgYaahT9OqjGvvR003ByOXUAghhCg6CYCK6HELgLIpisKec3Es/uM8+y7EA6DRQNd6HrzY0Y9G1SoYt4BCCCFEEUgAVESPawD0oCNRt1i88zzhJ6/r0trUrMSLHfwI8nOVuYSEEEKUOxIAFZEpBEDZzlxP4ss/z/NL5DWy7s2q2NirAi928KNLXXfMzCQQEkIIUT5IAFREphQAZbty8w5f775A2MErpGWqkyrWcnNgXHs//tekCpYyqaIQQogyTgKgIjLFACjbjaQ0lu+9yIp9l0lKywSgagVbxrarQf9AL2ytZFJFIYQQZZMEQEVkygFQtsTUDFb9FcW3ey4Sl5wGgKu9FSPb+DKklQ/OtjKpohBCiLJFAqAikgDovtSMLNYd+o+v/jzPf7fuAuBgbcGQVj70aVoFTydbnGwtpNO0EEIIo5MAqIgkADKUmaXl/45dY8nO85y5nqy3z9rCDDcna9wdbXB3slEfO9ng5qj+dHeyxs3JBkdrCZSEEEKUHAmAikgCoNxptQo7/o3lmz0X+Dcmidv31hzLDxtLMzUgcjQMkrKfuzvZ4GBtUYJ3IIQQ4nElAVARSQCUf6kZWdxISuN6Yiqx935eT0wjNjGV60mpxCaqaYmpmfnO087K3LAGyVE/SHJztMZeAiUhhBAPKMj3t3yDiCKxsTTHq6IdXhXt8jzubnoWsUn3gqOkB4KkxPtpsYlpJKVlcic9i4txKVyMS8kzTwdriwea3tRmtvtB0/3ASUauCSGEeJjRA6DFixfz0UcfER0dTf369VmwYAFt27bN8djo6GheffVVDh06xNmzZ5k4cSILFiwwOG79+vVMnz6d8+fP4+fnx5w5c+jbt28J34nIi62VOT6u9vi42ud5XEpaJrFJ2TVIOQdJMYmp3EnPIjktk+QbmbqFXnPjaGOhC4jcHW2o/EB/JXcnazwr2OLhZIO5TPoohBAmw6gBUFhYGJMmTWLx4sUEBwfz1Vdf0b17d06ePIm3t7fB8WlpaVSuXJlp06Yxf/78HPPct28fAwYM4N1336Vv375s2LCB/v37s2fPHlq2bFnStySKyN7aAl9rC3wr5R0oJadl3guMUnVNcNfvNbfF3guWYhJTSc3QkpSaSVJqMudik3PNz8rcjGoutnhVtMPH1Q7viurm42qPd0U7qUUSQojHjFH7ALVs2ZJmzZqxZMkSXVrdunXp06cPc+fOzfPcDh060KRJE4MaoAEDBpCYmMiWLVt0ad26dcPFxYXVq1fnq1zSB+jxoCgKSWmZxN4Liq4n5RwkxSSkkpGV969BZUdrfO4FRd6uDwZJ9lRysJLRbUIIUQaUiz5A6enpHDp0iClTpuilh4SEEBERUeh89+3bxyuvvKKX1rVr1xybysTjTaPR4GRjiZONJTXdHHM9LkurEJ1wl6j4O0TdvMPlm3fuP45PITE1kxtJadxISuPvy7cMzrezMtfVGHln1yDdqzmqWsEWKwtZRkQIIcoaowVAcXFxZGVl4e7urpfu7u5OTExMofONiYkpcJ5paWmkpaXpnicmJhb6+qL8MTfTUM3FjmoudrTOYf/tO+n3giE1KIqKv8PlmylcuXmXawl3uZOexb8xSfwbk2RwrpkGqlSwvR8YVbTXPfaqaCczagshhJEYvRP0w00HiqIUuTmhoHnOnTuXWbNmFema4vFVwc6KCnZWNKpWwWBfWmYW/926ez8wyg6SbqYQdfMOqRla/rt1l/9u3SXifHwOeVvic28UnY+rHT4V7XWPPZxsMJOO2UIIUSKMFgBVqlQJc3Nzg5qZ2NhYgxqcgvDw8ChwnlOnTmXy5Mm654mJiXh5eRW6DMJ0WFuY41fZAb/KDgb7FEXhRlKarklN/ZlyL0C6Q1xyOrfvZHD7TgJH/0swON/K3IxqFW0f6Htkrz6+1//IxlI6ZgshRGEZLQCysrIiICCA8PBwvSHq4eHh9O7du9D5BgUFER4ertcPaNu2bbRunVPjhsra2hpra+tCX1OInGg0GnVuIicbmlevaLA/JS1T17R25abarJb9+L9bd0nP0nLhRkquw/zdnax1HbG9K9rhVdEWN8f78x/JGm1CCJE7ozaBTZ48maFDhxIYGEhQUBBLly4lKiqKcePGAWrNzNWrV/n+++9150RGRgKQnJzMjRs3iIyMxMrKinr16gHw8ssv065dO+bNm0fv3r355Zdf2L59O3v27Cn1+xMiL/bWFtT1dKKup+FIhcwsLdEJqfp9j+4FSFHxd0hKy7w3oi2Ng5cMO2bD/TXaHgyKHn7u7mSNs62lBEpCCJNj9KUwFi9ezIcffkh0dDQNGjRg/vz5tGvXDoARI0Zw6dIldu7cqTs+pz/UPj4+XLp0Sff8xx9/5O233+bChQu6iRD79euX7zLJMHhRlimKwu07GWqT2gPNaldv3703vD+NhLv5X6PNysIMN0fre5v+rNpuD8yo7WIngZIQomyTtcCKSAIgUd7ltEZb9s8H0wuymK2VuRmVHa2p7GitV4P0cM2Si52VdN4WQhhFuZgHSAhRcvK7Rlt2oJS9/EhuwdKtOxmkZ2m5evsuV2/fzTNPCzMNlR0fXJst52DJ1V4CJSGE8UgAJIQJy2+glJZpGCjFJhoGS/Ep6WRqFaITUolOSM0zTwszDZUcrHUBUfbCth7O1lStYEc1F1s8K9hgbSGj3YQQxU8CICHEI1lbmOsmi8xLeqaWuOT7gZFhrVIaN5JSdYFSTKK6HAkYTgMAoNGAu6MNVV1sqabb1Bm2q7nYUqWCrUwHIIQoFAmAhBDFxsrCjCoV1MAkLxlZaqD0YC1SdrB0LSGVq7fUTt2pGVpdkHQoh2VIANwcre8FSHY5BkkSIAkhciIBkBCi1Fmam+HpbIunc+6BkqIoxKek35tJW50b6eoDj/+7dZe7GVlq8JSUxpGo2znmU8nBmmoutg/UIqmBkpeLLVUr2GFrJQGSEKZIAiAhRJmk0ah9hCo5WNPEq4LBfkVRuHUn44GA6M69AOmu7nlKehZxyWnEJacReeV2jtdxtbe6X2uUQ1ObvbX8mRTicSS/2UKIckmj0VDR3oqK9jmv06YoCgl3M/RqkB4Mjv67dZfktEziU9KJT0nPcTkSABc7S73mNbVpzY5qFdXHjjayoK0Q5ZEEQEKIx5JGo9EtZNugqrPBfkVRSLybyX+37+gFRlcfeJyYmsmtOxncupPA8as5B0jOtpY59j1yc7IBIEuroFUU9adWISv7saKQpX1o/72fevsV9TztQ+cZHqvojtXbr0t74LGCQZpWyTndwlxDEy8X2tSsRGB1F+lTJR4bMhFiDmQiRCEEQMLdDK7eUuc+0mtqu60GSQWZSPJxYGVhRoC3C8E1XWldsxKNqjpjYW5m7GIJoSMzQReRBEBCiPxISs1QJ4d8qGnt6u273EhKw0yjwcwMzDUazMw0mGs0mJtpMMv+aabBXINemt5+jQZzs4f2P5CXmZm630yT0/nkcOyD5/NAGR4qn+4xJKZm8teFeCLOxd+bsuA+R2sLWtZwJbimK8E1K1HLzUGWSxFGJQFQEUkAJIQQ+hRF4UJcChHn4th7Lp6I83EkpmbqHVPZ0ZpgP7V2KLhmJao+YjoEIYqbBEBFJAGQEELkLUur8M+1BF0wdODiTdIytXrH+Fayp7WfWjsUVMMVF3srI5VWmAoJgIpIAiAhhCiY1IwsDkfdIuJcPHvPx3H0ym20D3y7aDRQv4oTwX6VaF2zEs2ru2BnJeNwRPGSAKiIJAASQoiiSUzNYP+Fm+w9F0fE+TjOXE/W229prqGptzq6LLimK42qVcBSOlSLIpIAqIgkABJCiOIVm5hKxPn4ewFRPFdv39Xbb29lfq9DtRoQ1XF3lA7VosAkACoiCYCEEKLkKIrC5fg77LlXOxRxPt5gSoFKDlYE+VWiTU1XWvtVwqti3gvxCgESABWZBEBCCFF6tFqFk9GJRJyPY8+5eA5evMndjCy9Y7wr2umG2wfVcMXVwdpIpRVlmQRARSQBkBBCGE96ppYjUbfYe6/JLPLKbbK0+l9VdT2dCL43wqyFb0VZs00AEgAVmQRAQghRdiSnZXLgYjx7z6kB0b8xSXr7Lcw0NPWuQGs/df6hJl4VsLKQDtWmSAKgIpIASAghyq645DQizscTcS6OPefi+O+WfodqOytzWvhWvDfk3pW6Hk6YmUmHalMgAVARSQAkhBDlR1T8Hfaej2PvuTj2nY8nPiVdb39FeytaVK9IBTtLzM00WJhpMDczu7fMiNm95/c3C4PH6jFmD+wzfG6GmRlYmJnlmEdBjpXRb4UnAVARSQAkhBDlk1ar8G9MEhH3AqL9F29yJz3r0SeWIWb31mnLDpYeDM4szTRUcrSmirMtnhVsqFrBFk9nW6pUsKFKBVsqO1ibdG2XBEBFJAGQEEI8HjKytBy9cpsjUbdJy8wiU6uQpVXI1Cpo7/3MeiAtS6s12Gf4XItWC5la7QPnPbApCplZD5yrKGRmafWeZ2SVzFevhZkGD2cbqtwLijwr2FKlgi1VnG3u/bTFydbisa1lKsj3t3SbF0II8diyNDcjsHpFAqtXNHZRDGgfDJC0CllZanCVpdwLnrIe2HfveUaWltikNK7dvsu1hLtcu51K9O27XLt9l+tJaWRqFf67ddegX9SD7K3M8axgi6ezYQ1SlXvpNpbmpfhKGIcEQEIIIYQRmJlpsCrG5qrMe8FR9L3A6Nrtu0QnpHL19l1d2s2UdFLSszgXm8y52ORc83K1t8KzQnZNkhoUqQGS+tPN0Qbzct7UJgGQEEII8RiwMDfT1eIE+OR8TGpGli4wunb7Xg1Swt17QZKadic9i/iUdOJT0jlxNTHHfMzNNHg42egCo4eDpaoVbKlgZ1mmm9okABJCCCFMhI2lOTUqO1CjskOO+xVFIfFu5r3mtbtcuxcURd8Llq4l3CUmIZVMrcLV22rgxOVbuVzLTNfvqEoFGzydbdUmt3uPq1Swwc7KeGGIBEBCCCGEAECj0eBsZ4mznSV1PXPuRJylVYhLTjOsQboXIF27nUpcchqpGVou3Ejhwo2UHPOp5eZA+OT2JXk7eZIASAghhBD5Zm6mwd3JBncnG5p653xMWmYWMQmpD/RFusvVe8HStXvBUpUKtqVb8IdIACSEEEKIYmVtYY6Pqz0+rva5HpOWadz5mWSxFCGEEEKUOmsL4w61lwBICCGEECZHAiAhhBBCmByjB0CLFy/G19cXGxsbAgIC2L17d57H//nnnwQEBGBjY0ONGjX48ssv9faHhoai0WgMttTU1JK8DSGEEEKUI0YNgMLCwpg0aRLTpk3jyJEjtG3blu7duxMVFZXj8RcvXqRHjx60bduWI0eO8NZbbzFx4kTWr1+vd5yTkxPR0dF6m42NTWnckhBCCCHKAaMuhtqyZUuaNWvGkiVLdGl169alT58+zJ071+D4N998k40bN3Lq1Cld2rhx4zh69Cj79u0D1BqgSZMmcfv27UKXSxZDFUIIIcqfgnx/G60GKD09nUOHDhESEqKXHhISQkRERI7n7Nu3z+D4rl278vfff5ORkaFLS05OxsfHh2rVqtGrVy+OHDlS/DcghBBCiHLLaAFQXFwcWVlZuLu766W7u7sTExOT4zkxMTE5Hp+ZmUlcXBwA/v7+hIaGsnHjRlavXo2NjQ3BwcGcPXs217KkpaWRmJiotwkhhBDi8WX0TtAPL5SmKEqei6fldPyD6a1atWLIkCE0btyYtm3bsnbtWmrXrs3nn3+ea55z587F2dlZt3l5eRX2doQQQghRDhgtAKpUqRLm5uYGtT2xsbEGtTzZPDw8cjzewsICV1fXHM8xMzOjefPmedYATZ06lYSEBN125cqVAt6NEEIIIcoTowVAVlZWBAQEEB4erpceHh5O69atczwnKCjI4Pht27YRGBiIpaVljucoikJkZCSenp65lsXa2honJye9TQghhBCPL6M2gU2ePJlvvvmGZcuWcerUKV555RWioqIYN24coNbMDBs2THf8uHHjuHz5MpMnT+bUqVMsW7aMb7/9ltdee013zKxZs9i6dSsXLlwgMjKSUaNGERkZqctTCCGEEMKoi6EOGDCA+Ph4Zs+eTXR0NA0aNGDz5s34+PgAEB0drTcnkK+vL5s3b+aVV15h0aJFVKlShYULF/LUU0/pjrl9+zZjx44lJiYGZ2dnmjZtyq5du2jRokWp358QQgghyiajzgNUVsk8QEIIIUT5U5Dvb6PWAJVV2TGhDIcXQgghyo/s7+381O1IAJSDpKQkABkOL4QQQpRDSUlJODs753mMNIHlQKvVcu3aNRwdHfOck6gwEhMT8fLy4sqVK9K8VgbI+1G2yPtRtsj7UfbIe5I3RVFISkqiSpUqmJnlPc5LaoByYGZmRrVq1Ur0GjLcvmyR96NskfejbJH3o+yR9yR3j6r5yWb0maCFEEIIIUqbBEBCCCGEMDkSAJUya2trZsyYgbW1tbGLIpD3o6yR96Nskfej7JH3pPhIJ2ghhBBCmBypARJCCCGEyZEASAghhBAmRwIgIYQQQpgcCYCEEEIIYXIkACpFixcvxtfXFxsbGwICAti9e7exi2Sy5s6dS/PmzXF0dMTNzY0+ffpw+vRpYxdLoL43Go2GSZMmGbsoJu3q1asMGTIEV1dX7OzsaNKkCYcOHTJ2sUxSZmYmb7/9Nr6+vtja2lKjRg1mz56NVqs1dtHKNQmASklYWBiTJk1i2rRpHDlyhLZt29K9e3eioqKMXTST9OeffzJ+/Hj++usvwsPDyczMJCQkhJSUFGMXzaQdPHiQpUuX0qhRI2MXxaTdunWL4OBgLC0t2bJlCydPnuSTTz6hQoUKxi6aSZo3bx5ffvklX3zxBadOneLDDz/ko48+4vPPPzd20co1GQZfSlq2bEmzZs1YsmSJLq1u3br06dOHuXPnGrFkAuDGjRu4ubnx559/0q5dO2MXxyQlJyfTrFkzFi9ezHvvvUeTJk1YsGCBsYtlkqZMmcLevXullrqM6NWrF+7u7nz77be6tKeeego7OztWrFhhxJKVb1IDVArS09M5dOgQISEheukhISFEREQYqVTiQQkJCQBUrFjRyCUxXePHj6dnz5507tzZ2EUxeRs3biQwMJBnnnkGNzc3mjZtytdff23sYpmsNm3a8Pvvv3PmzBkAjh49yp49e+jRo4eRS1a+yWKopSAuLo6srCzc3d310t3d3YmJiTFSqUQ2RVGYPHkybdq0oUGDBsYujklas2YNhw8f5uDBg8YuigAuXLjAkiVLmDx5Mm+99RYHDhxg4sSJWFtbM2zYMGMXz+S8+eabJCQk4O/vj7m5OVlZWcyZM4eBAwcau2jlmgRApUij0eg9VxTFIE2UvgkTJnDs2DH27Nlj7KKYpCtXrvDyyy+zbds2bGxsjF0cAWi1WgIDA3n//fcBaNq0Kf/88w9LliyRAMgIwsLCWLlyJT/88AP169cnMjKSSZMmUaVKFYYPH27s4pVbEgCVgkqVKmFubm5Q2xMbG2tQKyRK10svvcTGjRvZtWsX1apVM3ZxTNKhQ4eIjY0lICBAl5aVlcWuXbv44osvSEtLw9zc3IglND2enp7Uq1dPL61u3bqsX7/eSCUyba+//jpTpkzh2WefBaBhw4ZcvnyZuXPnSgBUBNIHqBRYWVkREBBAeHi4Xnp4eDitW7c2UqlMm6IoTJgwgZ9++okdO3bg6+tr7CKZrE6dOnH8+HEiIyN1W2BgIIMHDyYyMlKCHyMIDg42mBbizJkz+Pj4GKlEpu3OnTuYmel/XZubm8sw+CKSGqBSMnnyZIYOHUpgYCBBQUEsXbqUqKgoxo0bZ+yimaTx48fzww8/8Msvv+Do6KirnXN2dsbW1tbIpTMtjo6OBn2v7O3tcXV1lT5ZRvLKK6/QunVr3n//ffr378+BAwdYunQpS5cuNXbRTNKTTz7JnDlz8Pb2pn79+hw5coRPP/2UkSNHGrto5ZoMgy9Fixcv5sMPPyQ6OpoGDRowf/58GXJtJLn1vVq+fDkjRowo3cIIAx06dJBh8Eb266+/MnXqVM6ePYuvry+TJ09mzJgxxi6WSUpKSmL69Ols2LCB2NhYqlSpwsCBA3nnnXewsrIydvHKLQmAhBBCCGFypA+QEEIIIUyOBEBCCCGEMDkSAAkhhBDC5EgAJIQQQgiTIwGQEEIIIUyOBEBCCCGEMDkSAAkhhBDC5EgAJIQQudBoNPz888/GLoYQogRIACSEKJNGjBiBRqMx2Lp162bsogkhHgOyFpgQoszq1q0by5cv10uztrY2UmmEEI8TqQESQpRZ1tbWeHh46G0uLi6A2jy1ZMkSunfvjq2tLb6+vqxbt07v/OPHj/PEE09ga2uLq6srY8eOJTk5We+YZcuWUb9+faytrfH09GTChAl6++Pi4ujbty92dnbUqlWLjRs36vbdunWLwYMHU7lyZWxtbalVq5ZBwCaEKJskABJClFvTp0/nqaee4ujRowwZMoSBAwdy6tQpAO7cuUO3bt1wcXHh4MGDrFu3ju3bt+sFOEuWLGH8+PGMHTuW48ePs3HjRmrWrKl3jVmzZtG/f3+OHTtGjx49GDx4MDdv3tRd/+TJk2zZsoVTp06xZMkSKlWqVHovgBCi8BQhhCiDhg8frpibmyv29vZ62+zZsxVFURRAGTdunN45LVu2VF544QVFURRl6dKliouLi5KcnKzbv2nTJsXMzEyJiYlRFEVRqlSpokybNi3XMgDK22+/rXuenJysaDQaZcuWLYqiKMqTTz6pPPfcc8Vzw0KIUiV9gIQQZVbHjh1ZsmSJXlrFihV1j4OCgvT2BQUFERkZCcCpU6do3Lgx9vb2uv3BwcFotVpOnz6NRqPh2rVrdOrUKc8yNGrUSPfY3t4eR0dHYmNjAXjhhRd46qmnOHz4MCEhIfTp04fWrVsX6l6FEKVLAiAhRJllb29v0CT1KBqNBgBFUXSPczrG1tY2X/lZWloanKvVagHo3r07ly9fZtOmTWzfvp1OnToxfvx4Pv744wKVWQhR+qQPkBCi3Prrr78Mnvv7+wNQr149IiMjSUlJ0e3fu3cvZmZm1K5dG0dHR6pXr87vv/9epDJUrlyZESNGsHLlShYsWMDSpUuLlJ8QonRIDZAQosxKS0sjJiZGL83CwkLX0XjdunUEBgbSpk0bVq1axYEDB/j2228BGDx4MDNmzGD48OHMnDmTGzdu8NJLLzF06FDc3d0BmDlzJuPGjcPNzY3u3buTlJTE3r17eemll/JVvnfeeYeAgADq169PWloav/76K3Xr1i3GV0AIUVIkABJClFm//fYbnp6eeml16tTh33//BdQRWmvWrOHFF1/Ew8ODVatWUa9ePQDs7OzYunUrL7/8Ms2bN8fOzo6nnnqKTz/9VJfX8OHDSU1NZf78+bz22mtUqlSJp59+Ot/ls7KyYurUqVy6dAlbW1vatm3LmjVriuHOhRAlTaMoimLsQgghREFpNBo2bNhAnz59jF0UIUQ5JH2AhBBCCGFyJAASQgghhMmRPkBCiHJJWu+FEEUhNUBCCCGEMDkSAAkhhBDC5EgAJIQQQgiTIwGQEEIIIUyOBEBCCCGEMDkSAAkhhBDC5EgAJIQQQgiTIwGQEEIIIUyOBEBCCCGEMDn/D+CEVkGlVt5HAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Line graph of loss for Bidirectional LSTM\n",
    "plt.title('Training and Validation Loss (Bidirectional LSTM)')\n",
    "plt.plot(history_bidirectional.history['loss'], label='Training Loss (Bidirectional LSTM)')\n",
    "plt.plot(history_bidirectional.history['val_loss'], label='Validation Loss (Bidirectional LSTM)')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "898a8d5d",
   "metadata": {},
   "source": [
    "### LSTM Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "32bc445e",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-04-28 10:50:46.703295: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_2_grad/concat/split_2/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_2_grad/concat/split_2/split_dim}}]]\n",
      "2024-04-28 10:50:46.704474: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_grad/concat/split/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_grad/concat/split/split_dim}}]]\n",
      "2024-04-28 10:50:46.705088: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_1_grad/concat/split_1/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_1_grad/concat/split_1/split_dim}}]]\n"
     ]
    }
   ],
   "source": [
    "# creating an LSTM model\n",
    "lstm_model = Sequential()\n",
    "lstm_model.add(Embedding(max_features, 32))\n",
    "lstm_model.add(LSTM(32))\n",
    "lstm_model.add(Dense(1, activation='sigmoid'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "c05c4964",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    }
   ],
   "source": [
    "# compiling the LSTM model\n",
    "lstm_model.compile(optimizer=Adam(1e-4), loss='binary_crossentropy', metrics=['acc'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "25c0b67e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-04-28 10:50:52.946837: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_2_grad/concat/split_2/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_2_grad/concat/split_2/split_dim}}]]\n",
      "2024-04-28 10:50:52.948058: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_grad/concat/split/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_grad/concat/split/split_dim}}]]\n",
      "2024-04-28 10:50:52.948861: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_1_grad/concat/split_1/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_1_grad/concat/split_1/split_dim}}]]\n",
      "2024-04-28 10:50:53.361131: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_2_grad/concat/split_2/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_2_grad/concat/split_2/split_dim}}]]\n",
      "2024-04-28 10:50:53.361938: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_grad/concat/split/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_grad/concat/split/split_dim}}]]\n",
      "2024-04-28 10:50:53.362687: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_1_grad/concat/split_1/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_1_grad/concat/split_1/split_dim}}]]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "157/157 [==============================] - ETA: 0s - loss: 0.6919 - acc: 0.5666"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-04-28 10:51:29.905426: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_2_grad/concat/split_2/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_2_grad/concat/split_2/split_dim}}]]\n",
      "2024-04-28 10:51:29.906273: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_grad/concat/split/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_grad/concat/split/split_dim}}]]\n",
      "2024-04-28 10:51:29.907095: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'gradients/split_1_grad/concat/split_1/split_dim' with dtype int32\n",
      "\t [[{{node gradients/split_1_grad/concat/split_1/split_dim}}]]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "157/157 [==============================] - 40s 245ms/step - loss: 0.6919 - acc: 0.5666 - val_loss: 0.6898 - val_acc: 0.6206\n",
      "Epoch 2/10\n",
      "157/157 [==============================] - 37s 235ms/step - loss: 0.6312 - acc: 0.7136 - val_loss: 0.5735 - val_acc: 0.7860\n",
      "Epoch 3/10\n",
      "157/157 [==============================] - 37s 233ms/step - loss: 0.5219 - acc: 0.8133 - val_loss: 0.4913 - val_acc: 0.8242\n",
      "Epoch 4/10\n",
      "157/157 [==============================] - 36s 232ms/step - loss: 0.4346 - acc: 0.8480 - val_loss: 0.4098 - val_acc: 0.8580\n",
      "Epoch 5/10\n",
      "157/157 [==============================] - 36s 232ms/step - loss: 0.3602 - acc: 0.8770 - val_loss: 0.3708 - val_acc: 0.8710\n",
      "Epoch 6/10\n",
      "157/157 [==============================] - 36s 232ms/step - loss: 0.3211 - acc: 0.8950 - val_loss: 0.3572 - val_acc: 0.8694\n",
      "Epoch 7/10\n",
      "157/157 [==============================] - 37s 237ms/step - loss: 0.2895 - acc: 0.9046 - val_loss: 0.3361 - val_acc: 0.8790\n",
      "Epoch 8/10\n",
      "157/157 [==============================] - 40s 253ms/step - loss: 0.2754 - acc: 0.9115 - val_loss: 0.3306 - val_acc: 0.8784\n",
      "Epoch 9/10\n",
      "157/157 [==============================] - 38s 244ms/step - loss: 0.2479 - acc: 0.9212 - val_loss: 0.3210 - val_acc: 0.8798\n",
      "Epoch 10/10\n",
      "157/157 [==============================] - 38s 240ms/step - loss: 0.2333 - acc: 0.9247 - val_loss: 0.3369 - val_acc: 0.8782\n"
     ]
    }
   ],
   "source": [
    "# Training the LSTM model \n",
    "lstm_history = lstm_model.fit(x_train, y_train, epochs=10, batch_size=128, validation_split=0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "860f0134",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "782/782 [==============================] - 19s 24ms/step - loss: 0.3528 - acc: 0.8677\n",
      "Test Accuracy (LSTM): 0.867680013179779\n",
      "Test Loss (LSTM): 0.35277149081230164\n"
     ]
    }
   ],
   "source": [
    "# Find loss, accuracy of the LSTM model on test data\n",
    "test_loss, test_acc = lstm_model.evaluate(x_test, y_test)\n",
    "print(f'Test Accuracy (LSTM): {test_acc}')\n",
    "print(f'Test Loss (LSTM): {test_loss}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "fc17e94f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_3\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " embedding_3 (Embedding)     (None, None, 32)          320000    \n",
      "                                                                 \n",
      " lstm_4 (LSTM)               (None, 32)                8320      \n",
      "                                                                 \n",
      " dense_4 (Dense)             (None, 1)                 33        \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 328,353\n",
      "Trainable params: 328,353\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# summary of the LSTM model\n",
    "lstm_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "e0e7dd35",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Line graph of loss for LSTM\n",
    "plt.title('Training and Validation Loss (LSTM)')\n",
    "plt.plot(lstm_history.history['loss'], label='Train Loss')\n",
    "plt.plot(lstm_history.history['val_loss'], label='Validation Loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "613eca87",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Line graph of accuracy for LSTM\n",
    "plt.title('Training and Validation Accuracy (LSTM)')\n",
    "plt.plot(lstm_history.history['acc'], label='Training Accuracy')\n",
    "plt.plot(lstm_history.history['val_acc'], label='Validation Accuracy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a8673ea8",
   "metadata": {},
   "source": [
    "### Conclusion"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "61877651",
   "metadata": {},
   "source": [
    "In this task was created 4 model (RNN, LSTM, Bidirectional LSTM, Deep LSTM)—and evaluating their performance. Here are the results and recommendations for each model:\n",
    "\n",
    "**Simple RNN Model:**\n",
    "* Test Accuracy: 0.8546000123023987\n",
    "* Test loss: 0.3496982157230377\n",
    "* Architecture: Embedding -> SimpleRNN -> Dense\n",
    "* Recommendations: Find better hyperparameters for improved accuracy.\n",
    "\n",
    "**Deep LSTM Model:**\n",
    "* Test Accuracy of deep LSTM: 0.8551999926567078\n",
    "* Test Loss of deep LSTM: 0.4065380394458771\n",
    "* Architecture: Embedding -> LSTM (x3) -> Dense\n",
    "* Recommendations: Improve time compilation, managing network depth.\n",
    "\n",
    "**Bidirectional LSTM Model:**\n",
    "* Test Accuracy: 0.8599200248718262\n",
    "* Test Loss: 0.5005342364311218\n",
    "* Architecture: Embedding -> Bidirectional LSTM -> Dense\n",
    "* Recommendations: Improve time compilation and find better hyperparameters.\n",
    "\n",
    "**LSTM Model:**\n",
    "* Test Accuracy (LSTM): 0.867680013179779\n",
    "* Test Loss (LSTM): 0.35277149081230164\n",
    "* Architecture: Embedding -> LSTM -> Dense\n",
    "* Recommendations: Find better hyperparameters for better performance.\n",
    "\n",
    "Each model exhibits its own strengths and weaknesses, and there is room for improvement in all of them. Better hyperparameter tuning, optimization of compilation time, and refining of network architecture can lead to improved performance across the board."
   ]
  }
 ],
 "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": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}