IMDB Movie Review Classification using Simple Neural Network.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "7b102648",
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.datasets import imdb\n",
    "from keras import models\n",
    "from keras import layers\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "2173320e",
   "metadata": {},
   "outputs": [],
   "source": [
    "(train_data, train_labels), (test_data, test_labels) = imdb.load_data(num_words=10000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "86ea11c2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[1,\n",
       " 14,\n",
       " 22,\n",
       " 16,\n",
       " 43,\n",
       " 530,\n",
       " 973,\n",
       " 1622,\n",
       " 1385,\n",
       " 65,\n",
       " 458,\n",
       " 4468,\n",
       " 66,\n",
       " 3941,\n",
       " 4,\n",
       " 173,\n",
       " 36,\n",
       " 256,\n",
       " 5,\n",
       " 25,\n",
       " 100,\n",
       " 43,\n",
       " 838,\n",
       " 112,\n",
       " 50,\n",
       " 670,\n",
       " 2,\n",
       " 9,\n",
       " 35,\n",
       " 480,\n",
       " 284,\n",
       " 5,\n",
       " 150,\n",
       " 4,\n",
       " 172,\n",
       " 112,\n",
       " 167,\n",
       " 2,\n",
       " 336,\n",
       " 385,\n",
       " 39,\n",
       " 4,\n",
       " 172,\n",
       " 4536,\n",
       " 1111,\n",
       " 17,\n",
       " 546,\n",
       " 38,\n",
       " 13,\n",
       " 447,\n",
       " 4,\n",
       " 192,\n",
       " 50,\n",
       " 16,\n",
       " 6,\n",
       " 147,\n",
       " 2025,\n",
       " 19,\n",
       " 14,\n",
       " 22,\n",
       " 4,\n",
       " 1920,\n",
       " 4613,\n",
       " 469,\n",
       " 4,\n",
       " 22,\n",
       " 71,\n",
       " 87,\n",
       " 12,\n",
       " 16,\n",
       " 43,\n",
       " 530,\n",
       " 38,\n",
       " 76,\n",
       " 15,\n",
       " 13,\n",
       " 1247,\n",
       " 4,\n",
       " 22,\n",
       " 17,\n",
       " 515,\n",
       " 17,\n",
       " 12,\n",
       " 16,\n",
       " 626,\n",
       " 18,\n",
       " 2,\n",
       " 5,\n",
       " 62,\n",
       " 386,\n",
       " 12,\n",
       " 8,\n",
       " 316,\n",
       " 8,\n",
       " 106,\n",
       " 5,\n",
       " 4,\n",
       " 2223,\n",
       " 5244,\n",
       " 16,\n",
       " 480,\n",
       " 66,\n",
       " 3785,\n",
       " 33,\n",
       " 4,\n",
       " 130,\n",
       " 12,\n",
       " 16,\n",
       " 38,\n",
       " 619,\n",
       " 5,\n",
       " 25,\n",
       " 124,\n",
       " 51,\n",
       " 36,\n",
       " 135,\n",
       " 48,\n",
       " 25,\n",
       " 1415,\n",
       " 33,\n",
       " 6,\n",
       " 22,\n",
       " 12,\n",
       " 215,\n",
       " 28,\n",
       " 77,\n",
       " 52,\n",
       " 5,\n",
       " 14,\n",
       " 407,\n",
       " 16,\n",
       " 82,\n",
       " 2,\n",
       " 8,\n",
       " 4,\n",
       " 107,\n",
       " 117,\n",
       " 5952,\n",
       " 15,\n",
       " 256,\n",
       " 4,\n",
       " 2,\n",
       " 7,\n",
       " 3766,\n",
       " 5,\n",
       " 723,\n",
       " 36,\n",
       " 71,\n",
       " 43,\n",
       " 530,\n",
       " 476,\n",
       " 26,\n",
       " 400,\n",
       " 317,\n",
       " 46,\n",
       " 7,\n",
       " 4,\n",
       " 2,\n",
       " 1029,\n",
       " 13,\n",
       " 104,\n",
       " 88,\n",
       " 4,\n",
       " 381,\n",
       " 15,\n",
       " 297,\n",
       " 98,\n",
       " 32,\n",
       " 2071,\n",
       " 56,\n",
       " 26,\n",
       " 141,\n",
       " 6,\n",
       " 194,\n",
       " 7486,\n",
       " 18,\n",
       " 4,\n",
       " 226,\n",
       " 22,\n",
       " 21,\n",
       " 134,\n",
       " 476,\n",
       " 26,\n",
       " 480,\n",
       " 5,\n",
       " 144,\n",
       " 30,\n",
       " 5535,\n",
       " 18,\n",
       " 51,\n",
       " 36,\n",
       " 28,\n",
       " 224,\n",
       " 92,\n",
       " 25,\n",
       " 104,\n",
       " 4,\n",
       " 226,\n",
       " 65,\n",
       " 16,\n",
       " 38,\n",
       " 1334,\n",
       " 88,\n",
       " 12,\n",
       " 16,\n",
       " 283,\n",
       " 5,\n",
       " 16,\n",
       " 4472,\n",
       " 113,\n",
       " 103,\n",
       " 32,\n",
       " 15,\n",
       " 16,\n",
       " 5345,\n",
       " 19,\n",
       " 178,\n",
       " 32]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e29c1cbb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_labels[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "97c956d9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((25000,), (25000,), (25000,), (25000,))"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data.shape, train_labels.shape, test_data.shape, test_labels.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "5c5613c4",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "9999"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "max([max(sequence) for sequence in train_data])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3e007084",
   "metadata": {},
   "source": [
    "### Decoding one of the Reviews to English"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "31c153f3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "? this film was just brilliant casting location scenery story direction everyone's really suited the part they played and you could just imagine being there robert ? is an amazing actor and now the same being director ? father came from the same scottish island as myself so i loved the fact there was a real connection with this film the witty remarks throughout the film were great it was just brilliant so much that i bought the film as soon as it was released for ? and would recommend it to everyone to watch and the fly fishing was amazing really cried at the end it was so sad and you know what they say if you cry at a film it must have been good and this definitely was also ? to the two little boy's that played the ? of norman and paul they were just brilliant children are often left out of the ? list i think because the stars that play them all grown up are such a big profile for the whole film but these children are amazing and should be praised for what they have done don't you think the whole story was so lovely because it was true and was someone's life after all that was shared with us all\n"
     ]
    }
   ],
   "source": [
    "# Reverse word index mapping\n",
    "word_index = imdb.get_word_index()\n",
    "reverse_word_index = {index: word for word, index in word_index.items()}\n",
    "#reverse_word_index = {index + 3: word for word, index in word_index.items()}\n",
    "\n",
    "# Choose a review to decode\n",
    "review_index = 0\n",
    "review = train_data[review_index]\n",
    "\n",
    "# Decode the review\n",
    "decoded_review = ' '.join(reverse_word_index.get(index - 3, '?') for index in review)\n",
    "\n",
    "# Print the decoded review\n",
    "print(decoded_review)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "d5ef16dd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "? a total waste of time just throw in a few explosions non stop fighting exotic cars a deranged millionaire slow motion computer generated car crashes and last but not least a hugh ? like character with wall to wall hot babes and mix in a ? and you will have this sorry excuse for a movie i really got a laugh out of the dr evil like heavily ? compound the plot was somewhere between preposterous and non existent how many ? are willing to make a 25 million dollar bet on a car race answer 4 but didn't they become ? through ? responsibility this was written for ? males it plays like a video game i did enjoy the ? ii landing in the desert though\n"
     ]
    }
   ],
   "source": [
    "#same things in another way\n",
    "word_index = imdb.get_word_index()\n",
    "reverse_word_index = dict(\n",
    "                           [(value, key) for (key, value) in word_index.items()] \n",
    "                            )\n",
    "# Decode the review\n",
    "decoded_review = ' '.join(reverse_word_index.get(index - 3, '?') for index in train_data[15])\n",
    "\n",
    "# Print the decoded review\n",
    "print(decoded_review)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f6740ffa",
   "metadata": {},
   "source": [
    "### Encoding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "18a2a027",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "07be2c56",
   "metadata": {},
   "outputs": [],
   "source": [
    "def vectorize_sequences (sequences, dimension = 10000):\n",
    "    results = np.zeros((len(sequences), dimension))\n",
    "    for i, sequence in enumerate(sequences):\n",
    "        results[i, sequence] =1\n",
    "    return results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "301fc49b",
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train = vectorize_sequences(train_data)\n",
    "x_test = vectorize_sequences(test_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "5e070df9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(25000, 10000)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "6e506980",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 1., 1., ..., 0., 0., 0.],\n",
       "       [0., 1., 1., ..., 0., 0., 0.],\n",
       "       [0., 1., 1., ..., 0., 0., 0.],\n",
       "       ...,\n",
       "       [0., 1., 1., ..., 0., 0., 0.],\n",
       "       [0., 1., 1., ..., 0., 0., 0.],\n",
       "       [0., 1., 1., ..., 0., 0., 0.]])"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "ec8dd645",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(25000, 10000)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "084869f4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 0, 0, ..., 0, 1, 0], dtype=int64)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "32c486a2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 1, 1, ..., 0, 0, 0], dtype=int64)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "003fb9ca",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Vectorizing the labels also\n",
    "y_train = np.asarray(train_labels).astype('float32')\n",
    "y_test = np.asarray(test_labels).astype('float32')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "d0701382",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1., 0., 0., ..., 0., 1., 0.], dtype=float32)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f2735fd3",
   "metadata": {},
   "source": [
    "## Building a NN Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "7da7b1c3",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = models.Sequential()\n",
    "model.add(layers.Dense(16, activation='relu', input_shape=(10000,)))\n",
    "model.add(layers.Dense(16, activation='relu'))\n",
    "model.add(layers.Dense(1, activation= 'sigmoid'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "36847628",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(optimizer ='rmsprop', loss='binary_crossentropy', metrics = ['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "d03d0cc5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# from keras import optimizers\n",
    "# from keras import metrics\n",
    "# from keras import losses\n",
    "\n",
    "# model.compile(optimizer = optimizers.RMSprop(lr=0.001), loss='binary_crossentropy', metrics=[mertrics.binary_accuracy])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "9c8403c2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000, 10000)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_val = x_train[:10000]\n",
    "partial_x_train = x_train[10000:]\n",
    "\n",
    "x_val.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "9e8fd43a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000,)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_val = y_train[:10000]\n",
    "partial_y_train= y_train[10000:]\n",
    "\n",
    "y_val.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "d9e4b1cd",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "30/30 [==============================] - 3s 47ms/step - loss: 0.5376 - accuracy: 0.7701 - val_loss: 0.4132 - val_accuracy: 0.8552\n",
      "Epoch 2/20\n",
      "30/30 [==============================] - 1s 21ms/step - loss: 0.3320 - accuracy: 0.8939 - val_loss: 0.3526 - val_accuracy: 0.8572\n",
      "Epoch 3/20\n",
      "30/30 [==============================] - 1s 23ms/step - loss: 0.2511 - accuracy: 0.9188 - val_loss: 0.2886 - val_accuracy: 0.8883\n",
      "Epoch 4/20\n",
      "30/30 [==============================] - 1s 23ms/step - loss: 0.2029 - accuracy: 0.9325 - val_loss: 0.2787 - val_accuracy: 0.8891\n",
      "Epoch 5/20\n",
      "30/30 [==============================] - 1s 22ms/step - loss: 0.1704 - accuracy: 0.9434 - val_loss: 0.2775 - val_accuracy: 0.8872\n",
      "Epoch 6/20\n",
      "30/30 [==============================] - 1s 23ms/step - loss: 0.1444 - accuracy: 0.9537 - val_loss: 0.2807 - val_accuracy: 0.8873\n",
      "Epoch 7/20\n",
      "30/30 [==============================] - 1s 21ms/step - loss: 0.1245 - accuracy: 0.9613 - val_loss: 0.3082 - val_accuracy: 0.8818\n",
      "Epoch 8/20\n",
      "30/30 [==============================] - 1s 22ms/step - loss: 0.1101 - accuracy: 0.9660 - val_loss: 0.3293 - val_accuracy: 0.8781\n",
      "Epoch 9/20\n",
      "30/30 [==============================] - 1s 22ms/step - loss: 0.0930 - accuracy: 0.9730 - val_loss: 0.3524 - val_accuracy: 0.8745\n",
      "Epoch 10/20\n",
      "30/30 [==============================] - 1s 26ms/step - loss: 0.0820 - accuracy: 0.9784 - val_loss: 0.3899 - val_accuracy: 0.8628\n",
      "Epoch 11/20\n",
      "30/30 [==============================] - 1s 26ms/step - loss: 0.0706 - accuracy: 0.9819 - val_loss: 0.4181 - val_accuracy: 0.8647\n",
      "Epoch 12/20\n",
      "30/30 [==============================] - 1s 22ms/step - loss: 0.0606 - accuracy: 0.9858 - val_loss: 0.3667 - val_accuracy: 0.8796\n",
      "Epoch 13/20\n",
      "30/30 [==============================] - 1s 25ms/step - loss: 0.0540 - accuracy: 0.9865 - val_loss: 0.3858 - val_accuracy: 0.8791\n",
      "Epoch 14/20\n",
      "30/30 [==============================] - 1s 22ms/step - loss: 0.0456 - accuracy: 0.9911 - val_loss: 0.4127 - val_accuracy: 0.8732\n",
      "Epoch 15/20\n",
      "30/30 [==============================] - 1s 25ms/step - loss: 0.0397 - accuracy: 0.9916 - val_loss: 0.4247 - val_accuracy: 0.8744\n",
      "Epoch 16/20\n",
      "30/30 [==============================] - 1s 22ms/step - loss: 0.0328 - accuracy: 0.9943 - val_loss: 0.4497 - val_accuracy: 0.8735\n",
      "Epoch 17/20\n",
      "30/30 [==============================] - 1s 24ms/step - loss: 0.0296 - accuracy: 0.9945 - val_loss: 0.4683 - val_accuracy: 0.8729\n",
      "Epoch 18/20\n",
      "30/30 [==============================] - 1s 23ms/step - loss: 0.0257 - accuracy: 0.9952 - val_loss: 0.4927 - val_accuracy: 0.8709\n",
      "Epoch 19/20\n",
      "30/30 [==============================] - 1s 22ms/step - loss: 0.0207 - accuracy: 0.9975 - val_loss: 0.5119 - val_accuracy: 0.8723\n",
      "Epoch 20/20\n",
      "30/30 [==============================] - 1s 25ms/step - loss: 0.0176 - accuracy: 0.9983 - val_loss: 0.5711 - val_accuracy: 0.8666\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(partial_x_train, partial_y_train, epochs=20, batch_size=512, validation_data=(x_val, y_val))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8f5e911c",
   "metadata": {},
   "source": [
    "#### The fit() function returns a History object (history in this case), which contains information about the training process, such as the loss and accuracy values at each epoch. This object can be used to analyze and visualize the training history."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "59fd14b1",
   "metadata": {},
   "outputs": [],
   "source": [
    "history_dict = history.history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "50d7cde3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['loss', 'accuracy', 'val_loss', 'val_accuracy'])"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "history_dict.keys()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3db9485e",
   "metadata": {},
   "source": [
    "### Plotting the training loss and validation loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "0147b206",
   "metadata": {},
   "outputs": [],
   "source": [
    "loss_values = history_dict['loss']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "84492531",
   "metadata": {},
   "outputs": [],
   "source": [
    "val_loss_values = history_dict['val_loss']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "11030573",
   "metadata": {},
   "outputs": [],
   "source": [
    "acc = history_dict['val_accuracy']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "176f8f39",
   "metadata": {},
   "outputs": [],
   "source": [
    "epochs = range(1, len(acc)+1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "537d6000",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "range(1, 21)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "epochs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "bf55edc9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1e2b958a890>"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjcAAAHHCAYAAABDUnkqAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABlJ0lEQVR4nO3deXhMZ/sH8O8ksohIYs0iIZbYSQjS8MbSpo2llsZeJVSpnYa+qjQJ2sZOqdpqrbUIqgihKBrltS+hqF0i1oSIRGbO74/nl2FkT2bmzPL9XNdc5jzzzJn7ZDLmzrMqJEmSQERERGQiLOQOgIiIiEibmNwQERGRSWFyQ0RERCaFyQ0RERGZFCY3REREZFKY3BAREZFJYXJDREREJoXJDREREZkUJjdERERkUpjcEMmgb9++8PT0LNRzIyIioFAotBuQgblx4wYUCgVWrFih99dWKBSIiIhQH69YsQIKhQI3btzI87menp7o27evVuMpyu9KUcj5HhAVFZMbojcoFIp83Q4cOCB3qGZvxIgRUCgUuHr1ao51xo8fD4VCgbNnz+oxsoK7d+8eIiIicPr0ablDITIJxeQOgMiQ/PLLLxrHq1atQkxMTJbyWrVqFel1lixZApVKVajnTpgwAV999VWRXt8U9OrVC/PmzcPatWsRFhaWbZ1169ahXr16qF+/fqFfp3fv3ujRowdsbGwKfY683Lt3DxMnToSnpyd8fHw0HivK7wqRuWJyQ/SGTz75ROP46NGjiImJyVL+thcvXsDOzi7fr2NlZVWo+ACgWLFiKFaMH10/Pz9Uq1YN69atyza5iY2NxfXr1zFlypQivY6lpSUsLS2LdI6iKMrvCpG5YrcUUQG1bNkSdevWxYkTJ9C8eXPY2dnh66+/BgBs27YN7dq1g5ubG2xsbFC1alVMnjwZSqVS4xxvj6PIHN8wY8YMLF68GFWrVoWNjQ0aN26M48ePazw3uzE3CoUCw4YNw9atW1G3bl3Y2NigTp06iI6OzhL/gQMH0KhRI9ja2qJq1apYtGhRvsfxHDp0CF27dkXFihVhY2MDDw8PfPHFF0hNTc1yffb29rh79y46deoEe3t7lCtXDmPGjMnys3j69Cn69u0LR0dHODk5ISQkBE+fPs0zFkC03ly6dAknT57M8tjatWuhUCjQs2dPpKenIywsDL6+vnB0dESJEiUQEBCA/fv35/ka2Y25kSQJ3377Ldzd3WFnZ4dWrVrhwoULWZ77+PFjjBkzBvXq1YO9vT0cHBzQpk0bnDlzRl3nwIEDaNy4MQCgX79+6q7PzLEu2Y25SUlJwejRo+Hh4QEbGxvUqFEDM2bMgCRJGvUK8nuRX3/88QcCAgJQokQJODk5oWPHjoiLi9Oo8+zZM4waNQqenp6wsbFB+fLl8f7772u8T1euXEHnzp3h4uICW1tbuLu7o0ePHkhKSip0bESZ+OcfUSE8evQIbdq0QY8ePfDJJ5/A2dkZgPgitLe3R2hoKOzt7fHHH38gLCwMycnJmD59ep7nXbt2LZ49e4bPP/8cCoUC06ZNQ3BwMP799988/4I/fPgwoqKiMGTIEJQsWRJz585F586dcevWLZQpUwYAcOrUKbRu3Rqurq6YOHEilEolJk2ahHLlyuXrujdu3IgXL15g8ODBKFOmDI4dO4Z58+bhzp072Lhxo0ZdpVKJoKAg+Pn5YcaMGdi7dy9mzpyJqlWrYvDgwQBEktCxY0ccPnwYgwYNQq1atbBlyxaEhITkK55evXph4sSJWLt2LRo2bKjx2r/++isCAgJQsWJFPHz4ED///DN69uyJAQMG4NmzZ1i6dCmCgoJw7NixLF1BeQkLC8O3336Ltm3bom3btjh58iQ++OADpKena9T7999/sXXrVnTt2hWVK1fG/fv3sWjRIrRo0QIXL16Em5sbatWqhUmTJiEsLAwDBw5EQEAAAKBp06bZvrYkSejQoQP279+P/v37w8fHB7t378aXX36Ju3fvYvbs2Rr18/N7kV979+5FmzZtUKVKFURERCA1NRXz5s1Ds2bNcPLkSXUSNmjQIGzatAnDhg1D7dq18ejRIxw+fBhxcXFo2LAh0tPTERQUhLS0NAwfPhwuLi64e/cufv/9dzx9+hSOjo4FiosoC4mIcjR06FDp7Y9JixYtJADSwoULs9R/8eJFlrLPP/9csrOzk16+fKkuCwkJkSpVqqQ+vn79ugRAKlOmjPT48WN1+bZt2yQA0vbt29Vl4eHhWWICIFlbW0tXr15Vl505c0YCIM2bN09d1r59e8nOzk66e/euuuzKlStSsWLFspwzO9ldX2RkpKRQKKSbN29qXB8AadKkSRp1GzRoIPn6+qqPt27dKgGQpk2bpi7LyMiQAgICJADS8uXL84ypcePGkru7u6RUKtVl0dHREgBp0aJF6nOmpaVpPO/JkyeSs7Oz9Omnn2qUA5DCw8PVx8uXL5cASNevX5ckSZISExMla2trqV27dpJKpVLX+/rrryUAUkhIiLrs5cuXGnFJknivbWxsNH42x48fz/F63/5dyfyZffvttxr1unTpIikUCo3fgfz+XmQn83fyzZh8fHyk8uXLS48ePdI4n4WFhdSnTx91maOjozR06NAcz33q1CkJgLRx48ZcYyAqLHZLERWCjY0N+vXrl6W8ePHi6vvPnj3Dw4cPERAQgBcvXuDSpUt5nrd79+4oVaqU+jjzr/h///03z+cGBgaiatWq6uP69evDwcFB/VylUom9e/eiU6dOcHNzU9erVq0a2rRpk+f5Ac3rS0lJwcOHD9G0aVNIkoRTp05lqT9o0CCN44CAAI1r2blzJ4oVK6ZuyQHEGJfhw4fnKx5AjJO6c+cO/vzzT3XZ2rVrYW1tja5du6rPaW1tDQBQqVR4/PgxMjIy0KhRo2y7tHKzd+9epKenY/jw4RpdeaNGjcpS18bGBhYW4r9ZpVKJR48ewd7eHjVq1Cjw62bauXMnLC0tMWLECI3y0aNHQ5Ik7Nq1S6M8r9+L/IqPj8fp06fRt29flC5dWuN877//Pnbu3Kkuc3Jywt9//4179+5le67Mlpndu3fjxYsXBYqDKD+Y3BAVQoUKFdRflm+6cOECPvroIzg6OsLBwQHlypVTD0bOz1iCihUrahxnJjpPnjwp8HMzn5/53MTERKSmpqJatWpZ6mVXlp1bt26pv9wyx9G0aNECQNbrs7W1zdLd9WY8AHDz5k24urrC3t5eo16NGjXyFQ8A9OjRA5aWlli7di0A4OXLl9iyZQvatGmjkSiuXLkS9evXh62tLcqUKYNy5cphx44dBR7jcfPmTQCAl5eXRnm5cuU0Xg8QidTs2bPh5eUFGxsblC1bFuXKlcPZs2cLPbbk5s2bcHNzQ8mSJTXKM2fwZcaXKa/fi4K8LpD9e1OrVi08fPgQKSkpAIBp06bh/Pnz8PDwQJMmTRAREaGRTFWuXBmhoaH4+eefUbZsWQQFBWH+/Pkcb0Naw+SGqBDebMHI9PTpU7Ro0QJnzpzBpEmTsH37dsTExGDq1KkAkK/pvDnNypHeGiiq7efmh1KpxPvvv48dO3Zg7Nix2Lp1K2JiYtQDX9++Pn3NMMocrLp582a8evUK27dvx7Nnz9CrVy91ndWrV6Nv376oWrUqli5diujoaMTExODdd9/V6TTr77//HqGhoWjevDlWr16N3bt3IyYmBnXq1NHb9G5d/15kp1u3bvj3338xb948uLm5Yfr06ahTp45Gq9LMmTNx9uxZfP3110hNTcWIESNQp04d3LlzR2dxkfnggGIiLTlw4AAePXqEqKgoNG/eXF1+/fp1GaN6rXz58rC1tc120bvcFsLLdO7cOfzzzz9YuXIl+vTpoy6PiYkpdEyVKlXCvn378Pz5c43Wm8uXLxfoPL169UJ0dDR27dqFtWvXwsHBAe3bt1c/vmnTJlSpUgVRUVEaXUnh4eGFihkQs32qVKmiLn/w4EGW1pBNmzahVatWWLp0qUb506dPUbZsWfVxQVacrlSpEvbu3Ytnz55ptN5kdntmxqdtmefN7r25dOkSypYtixIlSqjLXF1dMWTIEAwZMgSJiYlo2LAhvvvuO40u0Hr16qFevXqYMGEC/vrrLzRr1gwLFy7Et99+q5NrIPPBlhsiLcn8C/nNv4jT09Px008/yRWSBktLSwQGBmLr1q0aYyGuXr2aZZxGTs8HNK9PkiT88MMPhY6pbdu2yMjIwIIFC9RlSqUS8+bNK9B5OnXqBDs7O/z000/YtWsXgoODYWtrm2vsf//9N2JjYwscc2BgIKysrDBv3jyN882ZMydLXUtLyywtJBs3bsTdu3c1yjKTgvxMgW/bti2USiV+/PFHjfLZs2dDoVDke/xUQbm6usLHxwcrV67UiPP8+fPYs2cP2rZtC0C8f293L5UvXx5ubm5IS0sDACQnJyMjI0OjTr169WBhYaGuQ1QUbLkh0pKmTZuiVKlSCAkJUW8N8Msvv+i0+b+gIiIisGfPHjRr1gyDBw9Wf0nWrVs3z6X/a9asiapVq2LMmDG4e/cuHBwcsHnz5gKP3XhT+/bt0axZM3z11Ve4ceMGateujaioqAKPvbC3t0enTp3U427e7JICgA8//BBRUVH46KOP0K5dO1y/fh0LFy5E7dq18fz58wK9VuZ6PZGRkfjwww/Rtm1bnDp1Crt27dJojcl83UmTJqFfv35o2rQpzp07hzVr1mi0+ABA1apV4eTkhIULF6JkyZIoUaIE/Pz8ULly5Syv3759e7Rq1Qrjx4/HjRs34O3tjT179mDbtm0YNWqUxuBhbZs+fTratGkDf39/9O/fXz0V3NHRUb0f17Nnz+Du7o4uXbrA29sb9vb22Lt3L44fP46ZM2cCEGvlDBs2DF27dkX16tWRkZGBX375BZaWlujcubPO4ifzwZYbIi0pU6YMfv/9d7i6umLChAmYMWMG3n//fUybNk3u0NR8fX2xa9culCpVCt988w2WLl2KSZMm4b333tNo6ciOlZUVtm/fDh8fH0RGRmLixInw8vLCqlWrCh2PhYUFfvvtN/Tq1QurV6/G+PHjUaFCBaxcubLA58pMaFxdXfHuu+9qPNa3b198//33OHPmDEaMGIHdu3dj9erVaNSoUaHi/vbbbzFx4kScOnUKX375Ja5du4Y9e/ZodMsAwNdff43Ro0dj9+7dGDlyJE6ePIkdO3bAw8NDo56VlRVWrlwJS0tLDBo0CD179sTBgwezfe3Mn9moUaPw+++/Y9SoUbh48SKmT5+OWbNmFep68iswMBDR0dEoU6YMwsLCMGPGDLzzzjs4cuSIOhGzs7PDkCFDcPr0aYSHh+OLL77A5cuX8dNPPyE0NBQA4O3tjaCgIGzfvh2hoaGIiIiAvb09du3ahXfeeUen10DmQSEZ0p+VRCSLTp064cKFC7hy5YrcoRARFRlbbojMzNtbJVy5cgU7d+5Ey5Yt5QmIiEjL2HJDZGZcXV3Rt29fVKlSBTdv3sSCBQuQlpaGU6dOZVm7hYjIGHFAMZGZad26NdatW4eEhATY2NjA398f33//PRMbIjIZbLkhIiIik8IxN0RERGRSmNwQERGRSTG7MTcqlQr37t1DyZIlC7TkOREREclHkiQ8e/YMbm5usLDIvW3G7JKbe/fuZVlAi4iIiIzD7du34e7unmsds0tuMjeau337NhwcHGSOhoiIiPIjOTkZHh4eGhvG5sTskpvMrigHBwcmN0REREYmP0NKOKCYiIiITAqTGyIiIjIpTG6IiIjIpDC5ISIiIpPC5IaIiIhMCpMbIiIiMilMboiIiMikMLkhIiIik8LkhoiIiEwKkxsiIiIyKUxuiIiIyKQwuSEiIiKTwuSGiIiItGb3buDVK3ljYHJDREREWnHwINC6NfDOO0BqqnxxMLkhIiKiIktJAfr3F/d9fYHixeWLhckNERERFdn48cC1a4C7OzB9uryxMLkhIiKiIjlyBJg7V9xfsgRwdJQ3HiY3REREVGipqcCnnwKSBPTrJ8bcyI3JDRERERVaWBjwzz+Amxswa5bc0QhMboiIiKhQ/v77dUKzaBHg5CRrOGpMboiIiKjAXr4U3VAqFfDJJ8CHH8od0WtMboiIiKjAJk0C4uIAZ2fghx/kjkYTkxsiIiIqkBMngGnTxP0FC4DSpeWN521MboiIiCjf0tNFd5RSCXTvDnz0kdwRZcXkhoiIiPLtu++Ac+eAcuWAefPkjiZ7TG6IiIgoX06fBr7/Xtz/8UeR4BgiJjdERESUp1evRHdURgYQHAx07Sp3RDljckNERER5mjJFtNyULg389BOgUMgdUc6Y3BAREVGuzp0DJk8W9+fOFdO/DVkxuQMwFUolcOgQEB8PuLoCAQGApaXcURERERVNRobojnr1CmjfHvj4Y7kjyhuTGy2IigJGjgTu3Hld5u4uFjUKDpYvLiIioqKaMUOsa+PkBCxcaNjdUZnYLVVEUVFAly6aiQ0A3L0ryqOi5ImLiIioqOLigPBwcX/OHLE5pjFgclMESqVosZGkrI9llo0aJeoREREZE6US+PRTsWhfmzZAnz5yR5R/TG6K4NChrC02b5Ik4PZtUY+IiMiYzJkDHD0KODgAixcbR3dUJiY3RRAfr916REREhuCff4AJE8T9mTPFOFJjwuSmCFxdtVuPiIhIbioV0L8/8PIl8P774r6xYXJTBAEBIpvNqalOoQA8PEQ9IiIiY/Djj8Dhw4C9PbBkiXF1R2ViclMElpZiujeQ9c3PPJ4zh+vdEBGRcbh2DRg3TtyfNg2oVEneeAqLyU0RBQcDmzYBFSpolru7i3Kuc0NERMYgszvqxQugZUvg88/ljqjwuIifFgQHAx07coViIiIyXgsXAgcPAnZ2wNKlgIURN38wudESS0uR6RIRERmbGzeA//5X3I+MBKpUkTWcIjPivIyIiIiKSpKAAQOAlBTgP/8Bhg2TO6KiY3JDRERkxn7+Gdi7F7C1BZYtM+7uqEwmcAlERERUGLdvA6NHi/vffgt4eckbj7YwuSEiIjJDkgQMHAg8ewa8847YC9FUMLkhIiIyQytXAtHRgI2N6I4ypRm+TG6IiIjMzL17wBdfiPsTJwK1askbj7YxuSEiIjIjkgQMGgQ8fQo0bvx6zI0pYXJDRERkRtauBbZvB6ysRHdUMRNc8Y7JDRERkRm4cwcYMeL1Lt9hYUDduvLGpCsmmK8RERFRplu3gClTxJYK6emirE0bYOxYeePSJbbcEBERmaDr18VU72rVgAULRGLTogWwbx+wY4foljJVbLkhIiIyIVevAt9/D6xaBSiVouy994BvvhHJjTkwiJab+fPnw9PTE7a2tvDz88OxY8dyrLtixQooFAqNm62trR6jJSIiMjyXLwN9+gA1agDLl4vEJigIOHxYbK9gLokNYADJzYYNGxAaGorw8HCcPHkS3t7eCAoKQmJiYo7PcXBwQHx8vPp28+ZNPUZMRERkOC5cAD7+WKxV88svgEoFtGsHHD0qFulr1kzuCPVP9uRm1qxZGDBgAPr164fatWtj4cKFsLOzw7Jly3J8jkKhgIuLi/rm7Oysx4iJiIjkd/Ys0K0bUK8esG6dWL+mY0fgf/8Dfv8d8POTO0L5yJrcpKen48SJEwgMDFSXWVhYIDAwELGxsTk+7/nz56hUqRI8PDzQsWNHXLhwIce6aWlpSE5O1rgREREZq5MngY8+Ary9gY0bRVLTuTNw6hSwdSvg6yt3hPKTNbl5+PAhlEpllpYXZ2dnJCQkZPucGjVqYNmyZdi2bRtWr14NlUqFpk2b4s6dO9nWj4yMhKOjo/rm4eGh9esgIiLStWPHgPbtRfKydSugUAA9egDnzgGbNgE+PnJHaDhk75YqKH9/f/Tp0wc+Pj5o0aIFoqKiUK5cOSxatCjb+uPGjUNSUpL6dvv2bT1HTEREVHixsWJdGj8/0d1kYQF88okYa7NunekuxFcUsk4FL1u2LCwtLXH//n2N8vv378PFxSVf57CyskKDBg1w9erVbB+3sbGBjY1NkWMlIiLSp0OHgEmTxEwnQOza3bs38PXXgJeXvLEZOllbbqytreHr64t9+/apy1QqFfbt2wd/f/98nUOpVOLcuXNwdXXVVZhERER6ExsLtGoFNG8uEptixYDPPgP++UdM8WZikzfZF/ELDQ1FSEgIGjVqhCZNmmDOnDlISUlBv379AAB9+vRBhQoVEBkZCQCYNGkS3nnnHVSrVg1Pnz7F9OnTcfPmTXz22WdyXgYREVGRPHkCfPUVsHixOLayEvtAffUVUKmSvLEZG9mTm+7du+PBgwcICwtDQkICfHx8EB0drR5kfOvWLVhYvG5gevLkCQYMGICEhASUKlUKvr6++Ouvv1C7dm25LoGIiKjQJEmMnfniCyBzibd+/YCJEwHOgSkchSRJktxB6FNycjIcHR2RlJQEBwcHucMhIiIzdvUqMGQIEBMjjmvVAhYtAgIC5I3LEBXk+9voZksREREZu7Q04NtvxUynmBjA1hb47jvg9GkmNtoge7cUERGROfnzT+Dzz4FLl8TxBx8AP/0EVK0qb1ymhC03REREevDoEfDpp2IDy0uXAGdnYO1asf8TExvtYnJDRESkQ5IErFwJ1KwppnIDouUmLg7o2VOsNEzaxW4pIiIiHbl8GRg0CDhwQBzXrSsGDDdtKmtYJo8tN0RERFr28iUQEQHUry8Sm+LFgalTxaaXTGx0jy03REREWvTHH6K15soVcdy2LfDjj0DlyvLGZU7YckNERKQFiYli76f33hOJjasrsHGj2OySiY1+MbkhIiIqApUK+PlnMWB49WoxQHjYMDFguEsXDhiWA7uliIiICunCBdEFdfiwOPbxEQOGmzSRNSyzx5YbIiKiAkpNBcaPF8nM4cNAiRLAzJnA8eNMbAwBW26IiIjyKSMD2LABCAsD/v1XlHXoAMybB1SsKG9s9BqTGyIiojy8fAmsWAFMmwZcvy7KKlQQs6A6dZIzMsoOkxsiIqIcPHsmxtDMnAkkJIiysmWBUaOAESOAkiVlDY9ywOSGiIjoLY8eAXPniu6mJ09EmYcHMGYM8NlngJ2dvPFR7pjcEBER/b+7d0UrzeLFQEqKKKteHfjqK6BXL8DaWt74KH+Y3BARkdm7elVsj7ByJfDqlShr0AD4+mvgo48AS0t546OCYXJDRERm68wZYMoU4NdfxWJ8ANC8uUhqPviAC/AZKyY3RERkdo4cASIjgR07Xpe1aweMGwc0ayZfXKQdTG6IiMgsSBKwZw/w/ffAn3+KMgsLoGtXMabGx0fW8EiLmNwQEZFJUyqBLVtEUnPqlCizsgJCQoD//hfw8pI3PtI+JjdERGSS0tOBNWvEQOHLl0WZnR3w+edAaCjg7i5vfKQ7TG6IiMikZGSIhfemTgVu3xZlTk5i0b3hw8UifGTamNwQEZHJOHlSLLKX2f3k7AyMHi1aaxwc5I2N9IfJDRERGb0XL4CICGDWLDHGplQpYPJkoH9/wNZW7uhI35jcEBGRUfvjD2DgQODaNXHcvTvwww+i1YbMk4XcARARERXG48eiZea990Ri4+4O/PYbsH49Extzx+SGiIiMiiSJFYVr1QKWLROrCA8dCly4ALRvL3d0ZAjYLUVEREbjzh1gyBBg+3ZxXKsWsGQJVxUmTWy5ISIig6dSAT/9BNSuLRIbKysgPFzMimJiQ29jyw0RERm0uDhgwACxHxQA+PuL1po6deSNiwwXW26IiMggpacDkyaJPZ+OHAHs7YEffwQOH2ZiQ7ljyw0RERmc2FjRWnPhgjhu1w5YsADw8JA3LjIObLkhIiKD8eyZ2CahWTOR2JQrJ6Z2b9/OxIbyjy03RERkEHbsAAYPfr0fVN++wIwZQJkysoZFRojJDRERySoxERg1Cli3ThxXriw2vnz/fVnDIiPGbikiIpKFJAErV4q1atatAywsgDFjgHPnmNhQ0bDlhoiI9O7SJTG2JiZGHHt7A0uXAr6+8sZFpoHJDRER6ZxKBZw4AWzbJm7nz4tyW1uxm3doqFiYj0gbmNwQEZFOpKWJHbu3bROzne7de/2YpSXQujUwezbg5SVfjGSamNwQEZHWPH4sZj1t2wbs3g08f/76MXt7kdB07Ai0bQuULi1fnGTamNwQkdl6/lysq+LqKnckxu369dfdTYcOAUrl68fc3IAOHURC06oVYGMjX5xkPpjcEJFZ2r4d+PRTkdzExAABAXJHZDxyGj+TqV49kcx06CAGCFtwXi7pGZMbIjIrqanAf/8r9ijK1KOH2F26fHn54jJ0eY2fCQh4ndBUqSJfnEQAkxut+vNP0eTq5yd3JESUnfPngZ49X7c0jBwJ7Nkjdp3++GMxRsTSUt4YDUnm+JnffgOiozl+howHkxstWbMG+OQToGZN8Regra3cERFRJkkSmy6OHg28fClaaFauFF/OFy8CjRsD+/YBkyeLacnm7tkz4MsvgZ9/zn78TIcOwLvvcvwMGS72hGpJ27aAi4tYmOrbb+WOhogyPXwIdOoEDB0qEpvWrYGzZ8W/AFC7tljqHwAmTRItOebs8GGxoN6iRSKxqVsXGD8eOHZM7Pm0YAHQpg0TGzJsTG60pFSp1334U6cCZ87IGw8RiTEi3t6iW8XaWqypsmMH4OysWe+TT4CBA0ULT69ewJ078sQrp7Q0YOxYoHlzMfupYkXx8zt3TvzB1rgxBwaT8eCvqhZ17gwEBwMZGUD//uJfItK/V6+Ar74CAgPFwNcaNYCjR8XmjDl9Qf/wA+DjI1p6uncX5zAXZ86I5GXaNJHg9e0rWrdatZI7MqLCMYjkZv78+fD09IStrS38/Pxw7NixfD1v/fr1UCgU6NSpk24DLIAffwScnMQ0yTlz5I6GyPxcvQo0ayZaUCUJGDBAfB4bNMj9eba2wKZNgIMD8NdfwNdf6ydeOSmVwJQpIrE5dw4oVw7YsgVYvhxwdJQ7OqLCkz252bBhA0JDQxEeHo6TJ0/C29sbQUFBSExMzPV5N27cwJgxYxBgYItTuLoCM2aI+2Fh4j9aItI9SQJWrRJJzPHj4o+MjRuBxYuBEiXyd46qVcUXOyA+x9u26Sxc2V27Jrqgxo0TrVQdO4pZZAb0tyJRocme3MyaNQsDBgxAv379ULt2bSxcuBB2dnZYtmxZjs9RKpXo1asXJk6ciCoGuKDCp5+KmQSpqa/78YlId5KSxLiZkBAxXbl5c9Gt0qVLwc8VHAx88YW4HxIC/PuvdmOVmySJwcLe3qKFqmRJkdBt2cJ1fsh0yJrcpKen48SJEwgMDFSXWVhYIDAwELGxsTk+b9KkSShfvjz69++vjzALTKEAliwBihcH9u8Hli6VOyIi03X0qGitWbtWrFEzaZIYCOvhUfhzTp0K+PuLpKlrVzHLyhTExwPt2gGDBgEpKUDLlqI7qm9f8f8WkamQNbl5+PAhlEolnN+auuDs7IyEhIRsn3P48GEsXboUS5YsyddrpKWlITk5WeOmD1WqiDUzAGDMGM3VPImo6JRK4LvvgP/8R8zu8fQUC2l+803RF+KzsgI2bADKlAFOnnzdkmPMfv1VTOvetUtM4549W6ztU6mS3JERaZ/s3VIF8ezZM/Tu3RtLlixB2bJl8/WcyMhIODo6qm8eRflzroBGjhQD9ZKSxBob7J4i0o7bt4H33gMmTBBJTo8ewOnTQNOm2nsNDw9g9WrRorFwoWgZMkaPH4vVl7t3F/cbNhQJW24zx4iMnay/2mXLloWlpSXu37+vUX7//n24uLhkqX/t2jXcuHED7du3R7FixVCsWDGsWrUKv/32G4oVK4Zr165lec64ceOQlJSkvt2+fVtn1/O2YsXECp/FigFbtwKbN+vtpYlMVlSUGC9y8KAYKLxihUg8dDG7p3VrsYAdIMbPxcVp/zV0ac8esYnlunWiNeubb0Q3Xu3ackdGpFuyJjfW1tbw9fXFvn371GUqlQr79u2Dv79/lvo1a9bEuXPncPr0afWtQ4cOaNWqFU6fPp1tq4yNjQ0cHBw0bvpUv75YbwMAhg0TfzkRUcG9eAF8/rlYT+rJE6BRI7HVSUiIbseLRESICQIpKWKAckqK7l5LW1JSRGtxUJDoEq9eHThyRIxHsrKSOzoi3ZO9UTI0NBRLlizBypUrERcXh8GDByMlJQX9+vUDAPTp0wfjxo0DANja2qJu3boaNycnJ5QsWRJ169aFtbW1nJeSowkTxJ5T9++LvW2IqGDOnAF8fcW0bkDs6n3kCODlpfvXtrQULUOurmIfqsGDDbuLOXOA9U8/iePhw0USyA19yZzIntx0794dM2bMQFhYGHx8fHD69GlER0erBxnfunUL8fHxMkdZNDY2YsaUQiGa0GNi5I6IyDhIklg5uEkTsW+bq6v4/EydKrZT0BdnZ2D9ejFG5ZdfDHMGZHq6+EOqWTPgyhXA3V38rObOBezs5I6OSL8UkmTIf4NoX3JyMhwdHZGUlKT3LqoRI4B588SsjnPnAHt7vb48kVHJyBBTlNesEcft24ukolw5+WKaMkUsemdjI1pIfHzki+VNFy4AvXuLFhpArPkzb55YyJDIVBTk+1v2lhtz8v33YjO6GzfEwD4iyl56OtCtm0hsihUTX9Tbtsmb2ACiO+zDD8Umk126iJmQclIqgZkzRZfdqVNi6vrGjaJ1iYkNmTMmN3pkby9WBgVEU/vRo/LGQ2SIUlPFFgBbtoiup6goMRjfEBaZs7AAVq4Ua8NcuyZWI5er7fvUKTHQecwYkWy1aydahAuzKjORqWFyo2etW4vmY0kCPvtM/IVKRMLz5+JLetcuscL377+L7ihDUrq0WBDPykokXj/8oN/X//tv0XrUsKFYtNDeXqyIvn27GJNERExuZDF7tmhev3ABiIyUOxoiw5CUJKYu798vvrCjo4H335c7quw1aSK6gwDgyy/10wp76BDwwQfAO+8AO3aIVqRevcRMss8+M4yWLSJDweRGBmXKiBkMgFg+/sIFeeMhktujR2LF4b/+EmNF9u0Tm18asmHDxL5TGRlifNCjR9p/DUkC9u4FWrQQP4+YGDEG6dNPxeyx1avFVi9EpInJjUy6dxfN7a9eAf37i4GBROYoIUFs4HjiBFC2rGi5adJE7qjyplCIFci9vMR2EL17AyqVds4tSaJ1pmlT0Xr1559i/NGgQWKa99Kl+lnjh8hYMbmRiUIhFtkqWVL0of/4o9wREenfnTuiVeL8eTFe5M8/DWd6dX44OACbNgG2tmKcUFG7mVUqMZC6USMxruboUXHukSPFAOYFC8RSEkSUOyY3MnJ3B6ZPF/e//lrsbExkLv79FwgIAP75RyyR8OefQK1ackdVcPXrv14NOCxMtDwVlFIpFgn09gaCg8XGliVKiPE8N24Ac+aI/y+IKH+Y3MhswADRl565b455LalI5urSJfF7f+MGUK2aGCxbrZrcURVev35iwUGVCujZE8jvouoZGcCqVUCdOuJ558+L1qDx48XPZto0sToyERUMkxuZWViIaZw2NmKw4KpVckdEpFvnzomuqLt3xe7Uf/4pWm6M3fz5Ygfu+/dFopKRkXPd9HTxua9RQ2z8efkyUKqU2Njy5k3g22/F+CMiKhwmNwagenVg4kRx/4svxH+ORKbof/8Tg4cTE8XYmgMHTGdtFjs7sTqwvT1w8KDoonrby5ciCapWDRg4UHTNlSsntnW4eVOsXM6VhYmKjsmNgRg9Wuzk++SJ2MWXyNQcOSKmez9+LHao/uMP+bdT0LYaNcQMKkAMLt6xQ9xPSQFmzQIqVxZTyG/fFknd7NlirN3YsWJyARFpB5MbA1GsmJjeaWkp/vrbulXuiIi0548/xAJ0ycmiSyomRnTDmKLu3YGhQ8X93r2BiAiR1IweLaa9e3iI1pt//wVGjRIDh4lIu7gruIEZN040Ubu6AhcvsomajN/OnWIGUFqaSHC2bBFdOKYsLU3MBDt+/HVZlSpiVmTv3mLNGiIqGO4KbsTCwsTiXPHxYgdiImO2ebPYBDMtDejQAfjtN9NPbAAxQeDXX8UGm7Vri4kCly+LBTuZ2BDpHltuDNCff4qme0A057dqJW88RIWxZo2YCaRUiq6aX34Rm02aE0nink9E2sKWGyPXvLlYZh0Q6+C8eCFvPEQFtWSJ6H5RKsX6L2vWmF9iAzCxIZILkxsDNXUqUKGCWHI9IkLuaIjy74cfxDRnSQKGDHk9UJ6ISF+Y3BgoBwexjwwAzJwp1gchMnRTpogZQAAwZozYM82C/8sQkZ7xvx0D1r490KOHWNK9f3+xgziRIZIkMRh+3DhxHB4utg5gtwwRyYHJjYH74QegdGng7NnXm2wSGRJJEq00kyeL46lTRVcqExsikguTGwNXvrxIcACxRcOlS/LGQ/SmjAyxYN2sWeJ43jwuYUBE8mNyYwR69QLatBGb7Q0axJ3DyTAcPAg0bCjGhikUYuDwsGFyR0VExOTGKCgUwE8/AcWLiy+UNWvkjojM2Z07Ytfrli3FDt+lSwMbNgCffip3ZEREApMbI+HpKXYMBsQeNU+fyhkNmaO0NLEZZM2awPr1YhbU4MHAP/8AXbvKHR0R0WtMbozI6NHiiyUxERg/Xu5oyJzs3AnUrSv2RkpJAZo1E8sT/PQTUKaM3NEREWlicmNErK3FbsKAGOfAtW9I165eFUsStGsn7ru6im0UDh0CGjSQOzoiouwxuTEy774rBhhLkugSUCrljohMUUqKaB2sUwf4/XegWDHgyy/F5o+ffMJp3kRk2JjcGKEZM8QKxv/7H7B4sdzRkCmRJDE4uGZN4PvvxQy9Dz4QA4enTQNKlpQ7QiKivDG5MUIuLsB334n748YB9+/LGw+ZhvPnRctgjx5iRpSnJ7BlCxAdLZIdIiJjweTGSCiVwIEDwLp14t+BA8WYh6QkLppGRfP0KTByJODjI363bG3FgpEXLwKdOrELioiMD5MbIxAVJf6KbtUK+Phj8W/VqkC3buKLZ9Uqsf4NUUGoVMCyZUD16sDcuSKBDg4G4uLEPlHFi8sdIRFR4TC5MXBRUUCXLqKb4E1374ppue+/L46HDBHjI4jy49gxwN9fbMj64IHodoqJATZvFok0EZExY3JjwJRK0V2Q3XYLmWUXLgDlyokuhDlz9BoeGaHERJHQ+PmJBKdkSWDmTLExa2Cg3NEREWkHkxsDduhQ1habN0mSaMHJXPZ+4kTg1i39xEbGJSNDbMBavbroigKAPn3E6sKhoYCVlbzxERFpE5MbAxYfn7969esDAQHAixfAqFE6DYmM0P79YvD5qFFiAHrDhsCRI8DKlWLmHRGRqWFyY8BcXfNXz81NLINvaSmm7u7Yodu4yDicOwd8+KGY3n3+vNgmYdEi0R3VtKnc0RER6Q6TGwMWEAC4u+c8FVehADw8RL26dYEvvhDlw4cDqan6i5MMy61bQN++gLe3SHSLFQOGDhVdUAMHiiSYiMiUMbkxYJaWYpwEkDXByTyeM+f1l1V4uEiGrl8Xq8uSeXn8WGyRUL266HKSJLFb98WLwI8/AqVLyx0hEZF+MLkxcMHBwKZNQIUKmuXu7qI8OPh1mb3962Ro2jSxDxCZvtRU8X5XrSq25khLA1q0AP7+G/j1V8DLS+4IiYj0SyFJ2U00Nl3JyclwdHREUlISHBwc5A4n35RKMXsqPl6MxQkIyL57QZLEDs67dompvXv2cIVZU6VUigUcw8Jez6qrVw+YMgVo04bvOxGZloJ8fzO5MUHXrondnNPSgPXrge7d5Y6ItEmSgJ07ga++EgOFATH26ttvxY7xHFNDRKaoIN/f7JYyQVWritWLATHIODlZ3nhIe/7+G2jZUsyCOn8eKFUKmD5dDBbu04eJDRERwOTGZP33v0C1aqIbKzxc7mioqP75R2zD8c47wJ9/AjY24j2+dg0YM0ZsdklERAKTGxNlawvMny/uz50LnD4tazhUSAkJwODBQO3aYt8nCwugXz/gyhVg6lTRckNERJqY3JiwDz4QO4erVOILUqWSOyLKr+RkMVC4alVg4UIxePjDD4EzZ8T2CR4eckdIRGS4CpXc3L59G3fe2PTo2LFjGDVqFBYvXqy1wEg7Zs0SU8SPHn29pxAZrvR0YN480aU4ebLYUuOdd4CDB4Ht28VijURElLtCJTcff/wx9u/fDwBISEjA+++/j2PHjmH8+PGYNGmSVgOkoqlQAch8S8aOBR4+lDceyp5KJWa21aoFjBgBPHggFuPbvBn46y+geXO5IyQiMh6FSm7Onz+PJk2aAAB+/fVX1K1bF3/99RfWrFmDFStWaDM+0oLhw8Xmmo8fi+nDZFgOHwaaNAF69gT+/VdsZrlwoZgNFRzM9WqIiAqqUMnNq1evYGNjAwDYu3cvOnToAACoWbMm4vO7lTXpTbFiwIIF4v7SpaIlgOSXni6SzebNgRMngJIlRVfU1avA558DVlZyR0hEZJwKldzUqVMHCxcuxKFDhxATE4PWrVsDAO7du4cyZcoU+Hzz58+Hp6cnbG1t4efnh2PHjuVYNyoqCo0aNYKTkxNKlCgBHx8f/PLLL4W5DLPStCnQv7+4P3gwkJEhbzzmLi5OjKWZOlUsytevn5jWPWECUKKE3NERERm3QiU3U6dOxaJFi9CyZUv07NkT3t7eAIDffvtN3V2VXxs2bEBoaCjCw8Nx8uRJeHt7IygoCImJidnWL126NMaPH4/Y2FicPXsW/fr1Q79+/bB79+7CXIpZmTJFbJ549qwYtEr6J0liin7DhsCpU0CZMkBUlBjsXa6c3NEREZmGQm+/oFQqkZycjFJvLLRx48YN2NnZoXz58vk+j5+fHxo3bowff/wRAKBSqeDh4YHhw4fjq3wOEGnYsCHatWuHyZMn51nXHLZfyM3PPwMDBogZVJcuZd2Qk3QnIQH49FOx7xcABAUBy5eLvcKIiCh3Ot9+ITU1FWlpaerE5ubNm5gzZw4uX75coMQmPT0dJ06cQGBg4OuALCwQGBiI2NjYPJ8vSRL27duHy5cvo3kO00nS0tKQnJyscTNnn34qukOePwdCQ+WOxnxs2yY2tdy1S6wuPHeu2B+KiQ0RkfYVKrnp2LEjVq1aBQB4+vQp/Pz8MHPmTHTq1AkLMkeu5sPDhw+hVCrh7OysUe7s7IyEhIQcn5eUlAR7e3tYW1ujXbt2mDdvHt5///1s60ZGRsLR0VF98zDz1c8sLMTgYgsL4Ndfxa7hpDvPnwMDBwKdOolp+N7eYvDw8OHiPSAiIu0r1H+vJ0+eREBAAABg06ZNcHZ2xs2bN7Fq1SrMnTtXqwFmp2TJkjh9+jSOHz+O7777DqGhoThw4EC2dceNG4ekpCT17fbt2zqPz9D5+Ii1VABg6FDg5UtZwzFZf/8NNGgALFkipnP/97+irE4duSMjIjJtxQrzpBcvXqBkyZIAgD179iA4OBgWFhZ45513cPPmzXyfp2zZsrC0tMT9+/c1yu/fvw8XF5ccn2dhYYFq1aoBAHx8fBAXF4fIyEi0bNkyS10bGxv1tHV6beJEYMMGMe14+nTgm2/kjsh0ZGQA338vFk9UKgF3d2DVKqBVK7kjIyIyD4VqualWrRq2bt2K27dvY/fu3fjggw8AAImJiQUapGttbQ1fX1/s27dPXaZSqbBv3z74+/vn+zwqlQppaWn5vwCCgwMwe7a4/913YhoyFd21a2LdmvBwkdj06CFmpzGxISLSn0IlN2FhYRgzZgw8PT3RpEkTdSKyZ88eNGjQoEDnCg0NxZIlS7By5UrExcVh8ODBSElJQb9+/QAAffr0wbhx49T1IyMjERMTg3///RdxcXGYOXMmfvnlF3zyySeFuRSz1q0bEBgIpKUBw4aJacpUOJIkZj75+ACxsSJ5XLMGWLeOO3cTEelbobqlunTpgv/85z+Ij49Xr3EDAO+99x4++uijAp2re/fuePDgAcLCwpCQkAAfHx9ER0erBxnfunULFm+MvExJScGQIUNw584dFC9eHDVr1sTq1avRvXv3wlyKWVMoxJor9eoB0dHAli1iuX8qmEePxKDhqChx3Ly56IaqVEneuIiIzFWh17nJlLk7uLu7u1YC0jVzX+cmO2FhYtn/ChWAOXOAFi24oFx+7dkD9O0LxMeL7RImTwbGjAEsLeWOjIjItOh8nRuVSoVJkybB0dERlSpVQqVKleDk5ITJkydDpVIVKmiSz7hxQOXKwN27QNeuQPnyQN26Yrry5s1ih2rSlJoKjBolFuKLjwdq1gSOHhU7rzOxISKSV6G6pcaPH4+lS5diypQpaNasGQDg8OHDiIiIwMuXL/Hdd99pNUjSreLFgT/+AGbNAg4cAM6dAy5cELf/XzgadeuKQbEtW4pul7Jl5YxYXmfOAL16iZ8PIKbTT5sG2NnJGxcREQmF6pZyc3PDwoUL1buBZ9q2bRuGDBmCu3fvai1AbWO3VN4ePgT+/BPYv18kO+fPZ61Tr55mslOI/VKNjkolZph9/bXY0dvZWewJ1bat3JEREZm+gnx/Fyq5sbW1xdmzZ1G9enWN8suXL8PHxwepqakFPaXeMLkpuAcPNJOdzBaLTAoFUL++SHQyk53SpWUIVIdu3wZCQsTPAAA6dhSL83FsEhGRfug8ufHz84Ofn1+W1YiHDx+OY8eO4e+//y7oKfWGyU3RJSZqJjsXL2o+rlCIbQbeTHYMdTp0RoaY7fTwobg9eJD9/b//BpKSRNfTnDnAZ5+J6yQiIv3QeXJz8OBBtGvXDhUrVlSvcRMbG4vbt29j586d6q0ZDBGTG+27fx84eFAkOgcOAHFxmo8rFGL9l7p1xfgeW1txs7F5ff/tW34fK/bGqDFJApKT805U3jx+8iT/19mkCbB6NeDlpYUfGhERFYjOkxsAuHfvHubPn49Lly4BAGrVqoWBAwfi22+/xeLFiwtzSr1gcqN7CQmayc7//4rohKXl64Tn2TPg1auCn0OhEN1o5cqJgdJly2a97+YmWqCsrLR/DURElDe9JDfZOXPmDBo2bAilUqmtU2odkxv9y0x2bt0SqyG/fJn1VpDyvBKYEiVyTlLevl+unOgy4/RtIiLDVpDv70JNBScqCBcXQJsLSCuVIul5O/GxtxcJS/Hi2nstIiIyPkxuyOhYWoqBvVxXhoiIslOoFYqJiIiIDFWBWm6C89hV8enTp0WJhYiIiKjICpTcODo65vl4nz59ihQQERERUVEUKLlZvny5ruIgIiIi0gqOuSEiIiKTwuSGiIiITAqTGyIiIjIpTG6IiIjIpHARP4JSCRw6BMTHA66uQEAAtyMgIiLjxeTGzEVFASNHAnfuvC5zdwd++AHIY1kjIiIig8RuKTMWFQV06aKZ2ADA3buiPCpKnriIiIiKgsmNmVIqRYtNdnvCZ5aNGiXqERERGRMmN2bq0KGsLTZvkiTg9m1Rj4iIyJgwuTFT8fHarUdERGQomNyYKVdX7dYjIiIyFExuzFRAgJgVpVBk/7hCAXh4iHpERETGhMmNmbK0FNO9gawJTubxnDlc74aIiIwPkxszFhwMbNoEVKigWe7uLsq5zg0RERkjLuJn5oKDgY4duUIxERGZDiY3BEtLoGVLuaMgIiLSDnZLERERkUlhckNEREQmhckNERERmRQmN0RERGRSmNwQERGRSWFyQ0RERCaFyQ0RERGZFCY3REREZFKY3BAREZFJYXJDREREJoXJDREREZkUJjdERERkUpjcEBERkUlhckNEREQmhckNERERmRQmN0RERGRSmNwQERGRSSkmdwBk+pRK4NAhID4ecHUFAgIAS0u5oyIiIlPF5IZ0KioKGDkSuHPndZm7O/DDD0BwsHxxERGR6WK3FOlMVBTQpYtmYgMAd++K8qgoeeIiIiLTZhDJzfz58+Hp6QlbW1v4+fnh2LFjOdZdsmQJAgICUKpUKZQqVQqBgYG51id5KJWixUaSsj6WWTZqlKhHRESkTbInNxs2bEBoaCjCw8Nx8uRJeHt7IygoCImJidnWP3DgAHr27In9+/cjNjYWHh4e+OCDD3D37l09R065OXQoa4vNmyQJuH1b1CMiItIm2ZObWbNmYcCAAejXrx9q166NhQsXws7ODsuWLcu2/po1azBkyBD4+PigZs2a+Pnnn6FSqbBv3z49R065iY/Xbj0iIqL8kjW5SU9Px4kTJxAYGKgus7CwQGBgIGJjY/N1jhcvXuDVq1coXbp0to+npaUhOTlZ40a65+qq3XpERET5JWty8/DhQyiVSjg7O2uUOzs7IyEhIV/nGDt2LNzc3DQSpDdFRkbC0dFRffPw8Chy3JS3gAAxK0qhyP5xhQLw8BD1iIiItEn2bqmimDJlCtavX48tW7bA1tY22zrjxo1DUlKS+nb79m09R2meLC3FdG8ga4KTeTxnDte7ISIi7ZM1uSlbtiwsLS1x//59jfL79+/DxcUl1+fOmDEDU6ZMwZ49e1C/fv0c69nY2MDBwUHjRvoRHAxs2gRUqKBZ7u4uyrnODRER6YKsyY21tTV8fX01BgNnDg729/fP8XnTpk3D5MmTER0djUaNGukjVCqk4GDgxg1g/35g7Vrx7/XrTGyIiEh3ZF+hODQ0FCEhIWjUqBGaNGmCOXPmICUlBf369QMA9OnTBxUqVEBkZCQAYOrUqQgLC8PatWvh6empHptjb28Pe3t72a6DcmZpCbRsKXcURERkLmRPbrp3744HDx4gLCwMCQkJ8PHxQXR0tHqQ8a1bt2Bh8bqBacGCBUhPT0eXLl00zhMeHo6IiAh9hk5EREQGSCFJ2a0ha7qSk5Ph6OiIpKQkjr8hIiIyEgX5/jbq2VJEREREb2NyQ0RERCaFyQ0RERGZFCY3REREZFKY3BAREZFJYXJDREREJkX2dW6IikqpBA4dAuLjxS7jAQHcs4qIyJwxuSGjFhUFjBwJ3LnzuszdXWzayS0eiIjME7ulyGhFRQFdumgmNgBw964oj4qSJy4iIpIXkxsySkqlaLHJbn3tzLJRo0Q9IiIyL0xuyCgdOpS1xeZNkgTcvi3qERGReWFyQ0YpPl679YiIyHQwuSGj5Oqq3XpERGQ6mNyQUQoIELOiFIrsH1coAA8PUY+IiMwLkxsySpaWYro3kDXByTyeM4fr3RARmSMmN2S0goOBTZuAChU0y93dRTnXuSEiMk9cxI+MWnAw0LEjVygmIqLXmNyQ0bO0BFq2lDsKIiIyFOyWIiIiIpPC5IaIiIhMCpMbIiIiMilMboiIiMikcEAxUS6USs7EIiIyNkxuiHIQFSV2Hn9zg053d7F4INfQISIyXOyWIspGVBTQpUvWncfv3hXlUVHyxEVERHljckP0FqVStNhIUtbHMstGjRL1iIjI8DC5IXrLoUNZW2zeJEnA7duiHhERGR4mN0RviY/Xbj0iItIvJjdEb3F11W49IiLSLyY3RG8JCBCzohSK7B9XKAAPD1GPiIgMD5MbordYWorp3kDWBCfzeM4crndDRGSomNwQZSM4GNi0CahQQbPc3V2Uc50bIiLDxUX8iHIQHAx07MgViomIjA2TG6JcWFoCLVvKHQURERUEkxsiGXHvKiIi7WNyQyQT7l1FRKQbHFBMJAPuXUVEpDtMboj0jHtXERHpFpMbIj3j3lVERLrF5IZIz7h3FRGRbjG5IdIz7l1FRKRbTG6I9Ix7VxER6RaTGyI9495VRES6xeSGSAb62rtKqQQOHADWrRP/cgYWEZkDLuJHJBNd713FRQKJyFwpJCm71TZMV3JyMhwdHZGUlAQHBwe5wyHSicxFAt/+dGd2e3FncyIyNgX5/ma3FJGJ4SKBRGTumNwQmRguEkhE5o7JDZGJ4SKBRGTumNwQmRguEkhE5k725Gb+/Pnw9PSEra0t/Pz8cOzYsRzrXrhwAZ07d4anpycUCgXmzJmjv0CJjAQXCSQicydrcrNhwwaEhoYiPDwcJ0+ehLe3N4KCgpCYmJht/RcvXqBKlSqYMmUKXFxc9BwtkXHgIoFEZO5kTW5mzZqFAQMGoF+/fqhduzYWLlwIOzs7LFu2LNv6jRs3xvTp09GjRw/Y2NjoOVoi46GvRQKJiAyRbIv4paen48SJExg3bpy6zMLCAoGBgYiNjdXa66SlpSEtLU19nJycrLVzExkyXS8SCIjp5Lo8PxFRYciW3Dx8+BBKpRLOzs4a5c7Ozrh06ZLWXicyMhITJ07U2vmIjImlJdCypW7OzRWQichQyT6gWNfGjRuHpKQk9e327dtyh0Rk9DJXQH57PZ27d0V5VJQ8cRERATImN2XLloWlpSXu37+vUX7//n2tDha2sbGBg4ODxo2ICo8rIBORoZMtubG2toavry/27dunLlOpVNi3bx/8/f3lCouI8sAVkInI0Mm6K3hoaChCQkLQqFEjNGnSBHPmzEFKSgr69esHAOjTpw8qVKiAyMhIAGIQ8sWLF9X37969i9OnT8Pe3h7VqlWT7TqIzAlXQCYiQydrctO9e3c8ePAAYWFhSEhIgI+PD6Kjo9WDjG/dugULi9eNS/fu3UODBg3UxzNmzMCMGTPQokULHDhwQN/hE5klroBMRIZOIUnZ9ZybroJsmU5EWSmVgKenGDyc3f8eCoWYNXX9etGmhXOaORG9qSDf3yY/W4qItEsfKyBHRYkEqlUr4OOPxb+enpyFRUT5w+SGiApMlysgc5o5ERUVu6WIqNC03XWU2eWV02wsbXV5EZHxKcj3t6wDionIuGl7BeSCTDPX1crLRGT82C1FRAaD08yJSBuY3BCRweA0cyLSBnZL5UCpVOLVq1dyh0FGzsrKCpYcHJJvAQFiTE1e08wDAor+WpxqTmS6mNy8RZIkJCQk4OnTp3KHQibCyckJLi4uULw9b5qyyJxm3qWLSGTeTHC0Nc0c4I7mRKaOyc1bMhOb8uXLw87Ojl9IVGiSJOHFixdITEwEALiyLyVfMqeZZ5d8zJlT9OQjc6r52y1DmVPNizqVnYjkx6ngb1Aqlfjnn39Qvnx5lClTRqYIydQ8evQIiYmJqF69OruoCkAX3Uacak5kvDgVvJAyx9jY2dnJHAmZkszfp1evXjG5KQBtTzMHONWcyFxwtlQ22BVF2sTfJ8PBqeZE5oHJDeXI09MTc+bMyXf9AwcOQKFQ6Hww9ooVK+Dk5KTT1yDTpM+p5kolcOAAsG6d+FepLPo5iSh/mNzoiD7/Y1MoFLneIiIiCnXe48ePY+DAgfmu37RpU8THx8PR0bFQr0eka5lTzXNqTFMoAA+Pok8158afRPLimBsd0Pc00/g32tA3bNiAsLAwXL58WV1mb2+vvi9JEpRKJYoVy/utL1euXIHisLa2houLS4GeQ6RP+phqztlYRPJjy42WybGjsYuLi/rm6OgIhUKhPr506RJKliyJXbt2wdfXFzY2Njh8+DCuXbuGjh07wtnZGfb29mjcuDH27t2rcd63u6UUCgV+/vlnfPTRR7Czs4OXlxd+++039eNvd0tldh/t3r0btWrVgr29PVq3bq2RjGVkZGDEiBFwcnJCmTJlMHbsWISEhKBTp04F+hksWLAAVatWhbW1NWrUqIFffvlF/ZgkSYiIiEDFihVhY2MDNzc3jBgxQv34Tz/9BC8vL9ja2sLZ2RldunQp0GuTcdHljuZKpfjDJrs5qJllo0axi4pI15jcaJEh/8f21VdfYcqUKYiLi0P9+vXx/PlztG3bFvv27cOpU6fQunVrtG/fHrdu3cr1PBMnTkS3bt1w9uxZtG3bFr169cLjx49zrP/ixQvMmDEDv/zyC/7880/cunULY8aMUT8+depUrFmzBsuXL8eRI0eQnJyMrVu3FujatmzZgpEjR2L06NE4f/48Pv/8c/Tr1w/79+8HAGzevBmzZ8/GokWLcOXKFWzduhX16tUDAPzvf//DiBEjMGnSJFy+fBnR0dFo3rx5gV6fjE9wMHDjBrB/P7B2rfj3+vWit6gUZDZWUXFMD1EuJDOTlJQkAZCSkpKyPJaamipdvHhRSk1NLdS59++XJPHfV+63/fuLdg25Wb58ueTo6PhGTPslANLWrVvzfG6dOnWkefPmqY8rVaokzZ49W30MQJowYYL6+Pnz5xIAadeuXRqv9eTJE3UsAKSrV6+qnzN//nzJ2dlZfezs7CxNnz5dfZyRkSFVrFhR6tixY76vsWnTptKAAQM06nTt2lVq27atJEmSNHPmTKl69epSenp6lnNt3rxZcnBwkJKTk3N8vaIq6u8VGY+1a/P3f8DatUV7nc2bJcndXfOc7u6inMhU5fb9/Ta23GiRIU8zbdSokcbx8+fPMWbMGNSqVQtOTk6wt7dHXFxcni039evXV98vUaIEHBwc1CvwZsfOzg5Vq1ZVH7u6uqrrJyUl4f79+2jSpIn6cUtLS/j6+hbo2uLi4tCsWTONsmbNmiEuLg4A0LVrV6SmpqJKlSoYMGAAtmzZgoyMDADA+++/j0qVKqFKlSro3bs31qxZgxcvXhTo9Yky6WM2lhxd30TGhsmNFhnyjsYlSpTQOB4zZgy2bNmC77//HocOHcLp06dRr149pKen53oeKysrjWOFQgGVSlWg+pKeF8X28PDA5cuX8dNPP6F48eIYMmQImjdvjlevXqFkyZI4efIk1q1bB1dXV4SFhcHb25t7i1Gh6Ho2liF3fRMZEiY3WqSvaabacOTIEfTt2xcfffQR6tWrBxcXF9y4cUOvMTg6OsLZ2RnHjx9XlymVSpw8ebJA56lVqxaOHDmiUXbkyBHUrl1bfVy8eHG0b98ec+fOxYEDBxAbG4tz584BAIoVK4bAwEBMmzYNZ8+exY0bN/DHH38U4crIXGXOxgKy/j+gjdlY+hzTQ2TMOBVci/S1o7E2eHl5ISoqCu3bt4dCocA333yTawuMrgwfPhyRkZGoVq0aatasiXnz5uHJkycFWtX3yy+/RLdu3dCgQQMEBgZi+/btiIqKUs/+WrFiBZRKJfz8/GBnZ4fVq1ejePHiqFSpEn7//Xf8+++/aN68OUqVKoWdO3dCpVKhRo0aurpkMnG63PhTX13futjXi0ifmNxoma53NNaWWbNm4dNPP0XTpk1RtmxZjB07FsnJyXqPY+zYsUhISECfPn1gaWmJgQMHIigoqEB7MHXq1Ak//PADZsyYgZEjR6Jy5cpYvnw5Wv7/5kBOTk6YMmUKQkNDoVQqUa9ePWzfvh1lypSBk5MToqKiEBERgZcvX8LLywvr1q1DnTp1dHTFZA6Cg4GOHbWfIOhrTI8+1+ki0gXuCv6Gly9f4vr166hcuTJsbW2L9Dr8y6dwVCoVatWqhW7dumHy5Mlyh6MV2vy9IvOWuav53bvZj7sp6q7mOS1AmNmQygUISU7cFdwA6GJHY1N08+ZN7NmzBy1atEBaWhp+/PFHXL9+HR9//LHcoREZHF12fec1WFmhEIOVO3bkH2pk+DigmGRlYWGBFStWoHHjxmjWrBnOnTuHvXv3olatWnKHRmSQdLXCMhcgJFPClhuSlYeHR5aZTkSUO12M6dHXYGWO6SF9YHJDRGSEtN31rc8FCLmpKOkau6WIiIgLEJJJYXJDREQmtQAhx/QQkxsiIgKgu8HKgH7H9Hh6Aq1aAR9/LP719OSeW+aGY26IiEjN2Bcg1MeYHq5jZviY3BARkQZdrNOVOaYnrwUIdTWmR1vr9HC2l3FgtxSptWzZEqNGjVIfe3p6Ys6cObk+R6FQYOvWrUV+bW2dJzcRERHw8fHR6WsQUfZMYUxPZsvQ26+T2TKkja4vjhfSDiY3JqB9+/Zo3bp1to8dOnQICoUCZ8+eLfB5jx8/joEDBxY1PA05JRjx8fFo06aNVl+LiAyLMY/p0cdsL44X0h4mNyagf//+iImJwZ1s/mxZvnw5GjVqhPr16xf4vOXKlYOdnZ02QsyTi4sLbGxs9PJaRCSf4GDgxg1g/35g7Vrx7/XrRe/S0fWYHl23DOmjVcicMLkxAR9++CHKlSuHFStWaJQ/f/4cGzduRP/+/fHo0SP07NkTFSpUgJ2dHerVq4d169blet63u6WuXLmC5s2bw9bWFrVr10ZMTEyW54wdOxbVq1eHnZ0dqlSpgm+++QavXr0CAKxYsQITJ07EmTNnoFAooFAo1DG/3S117tw5vPvuuyhevDjKlCmDgQMH4vnz5+rH+/bti06dOmHGjBlwdXVFmTJlMHToUPVr5YdKpcKkSZPg7u4OGxsb+Pj4IDo6Wv14eno6hg0bBldXV9ja2qJSpUqIjIwEAEiShIiICFSsWBE2NjZwc3PDiBEj8v3aROYsc0xPz57iX20MxtX1Oj26bBnS5xpA5tLtxQHFeZAk4MULeV7bzi7nD+qbihUrhj59+mDFihUYP348FP//pI0bN0KpVKJnz554/vw5fH19MXbsWDg4OGDHjh3o3bs3qlatiiZNmuT5GiqVCsHBwXB2dsbff/+NpKQkjfE5mUqWLIkVK1bAzc0N586dw4ABA1CyZEn897//Rffu3XH+/HlER0dj7969AABHR8cs50hJSUFQUBD8/f1x/PhxJCYm4rPPPsOwYcM0Erj9+/fD1dUV+/fvx9WrV9G9e3f4+PhgwIABef/QAPzwww+YOXMmFi1ahAYNGmDZsmXo0KEDLly4AC8vL8ydOxe//fYbfv31V1SsWBG3b9/G7du3AQCbN2/G7NmzsX79etSpUwcJCQk4c+ZMvl6XiLRPl5uKArptGSpIq1BRBnqb1WBoycwkJSVJAKSkpKQsj6WmpkoXL16UUlNT1WXPn0uS+NXS/+358/xfV1xcnARA2r9/v7osICBA+uSTT3J8Trt27aTRo0erj1u0aCGNHDlSfVypUiVp9uzZkiRJ0u7du6VixYpJd+/eVT++a9cuCYC0ZcuWHF9j+vTpkq+vr/o4PDxc8vb2zlLvzfMsXrxYKlWqlPT8jR/Ajh07JAsLCykhIUGSJEkKCQmRKlWqJGVkZKjrdO3aVerevXuOsbz92m5ubtJ3332nUadx48bSkCFDJEmSpOHDh0vvvvuupFKpspxr5syZUvXq1aX09PQcXy9Tdr9XRKQbmzdLkru75v+lHh6ivCgyMsR5FYrs/79WKMTrvPFfUr6tXZu/74S1awsf/+bN2ceuUIhbUX8+mTIyJGn/fhHr/v2F+3nkJLfv77exW8pE1KxZE02bNsWyZcsAAFevXsWhQ4fQv39/AIBSqcTkyZNRr149lC5dGvb29ti9ezdu3bqVr/PHxcXBw8MDbm5u6jJ/f/8s9TZs2IBmzZrBxcUF9vb2mDBhQr5f483X8vb2RokSJdRlzZo1g0qlwuXLl9VlderUgeUbf4a5uroiMTExX6+RnJyMe/fuoVmzZhrlzZo1Q1xcHADR9XX69GnUqFEDI0aMwJ49e9T1unbtitTUVFSpUgUDBgzAli1bkJGRUaDrJCLt09WYHl3O9tL1eCF9dXsZ0oBoJjd5sLMDnj+X51bQsbz9+/fH5s2b8ezZMyxfvhxVq1ZFixYtAADTp0/HDz/8gLFjx2L//v04ffo0goKCkJ6errWfVWxsLHr16oW2bdvi999/x6lTpzB+/HitvsabrKysNI4VCgVUKpXWzt+wYUNcv34dkydPRmpqKrp164YuXboAELuZX758GT/99BOKFy+OIUOGoHnz5gUa80NEuqGLMT2A7mZ76Xq8kKlMky8IJjd5UCiAEiXkueVnvM2bunXrBgsLC6xduxarVq3Cp59+qh5/c+TIEXTs2BGffPIJvL29UaVKFfzzzz/5PnetWrVw+/ZtxL8xWu7o0aMadf766y9UqlQJ48ePR6NGjeDl5YWbN29q1LG2toYyjz8PatWqhTNnziAlJUVdduTIEVhYWKBGjRr5jjk3Dg4OcHNzw5EjRzTKjxw5gtq1a2vU6969O5YsWYINGzZg8+bNePz4MQCgePHiaN++PebOnYsDBw4gNjYW586d00p8RGSYdNEypOs1gExhmnxBcUCxCbG3t0f37t0xbtw4JCcno2/fvurHvLy8sGnTJvz1118oVaoUZs2ahfv372t8kecmMDAQ1atXR0hICKZPn47k5GSMHz9eo46Xlxdu3bqF9evXo3HjxtixYwe2bNmiUcfT0xPXr1/H6dOn4e7ujpIlS2aZAt6rVy+Eh4cjJCQEERERePDgAYYPH47evXvD2dm5cD+cbHz55ZcIDw9H1apV4ePjg+XLl+P06dNYs2YNAGDWrFlwdXVFgwYNYGFhgY0bN8LFxQVOTk5YsWIFlEol/Pz8YGdnh9WrV6N48eKoVKmS1uIjIsOkixWcM1uFshvwO2dO0ZInQ5omr+2fW07YcmNi+vfvjydPniAoKEhjfMyECRPQsGFDBAUFoWXLlnBxcUGnTp3yfV4LCwts2bIFqampaNKkCT777DN89913GnU6dOiAL774AsOGDYOPjw/++usvfPPNNxp1OnfujNatW6NVq1YoV65cttPR7ezssHv3bjx+/BiNGzdGly5d8N577+HHH38s2A8jDyNGjEBoaChGjx6NevXqITo6Gr/99hu8vLwAiJlf06ZNQ6NGjdC4cWPcuHEDO3fuhIWFBZycnLBkyRI0a9YM9evXx969e7F9+3aUKVNGqzESkfnQ1XghY54mX1gKScquIcl0JScnw9HREUlJSXBwcNB47OXLl7h+/ToqV64MW1tbmSIkU8PfKyKSW+aYGCD7afJFGTN04IAYPJyX/fuL1nKT2/f329hyQ0REZOJ0ufWFrluGCoNjboiIiMxAcLDYFf3QIdFF5OoqEo6izibT9QKKhcHkhoiIyEzoYjA0oNsB0YXB5IaIiIiKTFctQ4VhEGNu5s+fD09PT9ja2sLPzw/Hjh3Ltf7GjRtRs2ZN2Nraol69eti5c6eeIiUiIqKc6GoBxYKSPbnZsGEDQkNDER4ejpMnT8Lb2xtBQUE5LqP/119/oWfPnujfvz9OnTqFTp06oVOnTjh//rzWYjKzCWSkY/x9IiLSL9mngvv5+aFx48bqNUxUKhU8PDwwfPhwfPXVV1nqd+/eHSkpKfj999/VZe+88w58fHywcOHCPF8vt6lkSqUS//zzD8qXL8/1SkhrHj16hMTERFSvXl1jLywiIsq/gkwFl3XMTXp6Ok6cOIFx48apyywsLBAYGIjY2NhsnxMbG4vQ0FCNsqCgIGzdurXI8VhaWsLJyUndamRnZ6fevoCooCRJwosXL5CYmAgnJycmNkREeiJrcvPw4UMolcosS+o7Ozvj0qVL2T4nISEh2/oJCQnZ1k9LS0NaWpr6ODk5OdeYXFxcACDfu0sT5cXJyUn9e0VERLpn8rOlIiMjMXHixHzXVygUcHV1Rfny5bnDMxWZlZUVW2yIiPRM1uSmbNmysLS0xP379zXK79+/n+Nfui4uLgWqP27cOI1urOTkZHh4eOQZm6WlJb+UiIiIjJCss6Wsra3h6+uLffv2qctUKhX27dsHf3//bJ/j7++vUR8AYmJicqxvY2MDBwcHjRsRERGZLtm7pUJDQxESEoJGjRqhSZMmmDNnDlJSUtCvXz8AQJ8+fVChQgVERkYCAEaOHIkWLVpg5syZaNeuHdavX4///e9/WLx4sZyXQURERAZC9uSme/fuePDgAcLCwpCQkAAfHx9ER0erBw3funULFhavG5iaNm2KtWvXYsKECfj666/h5eWFrVu3om7dunJdAhERERkQ2de50bekpCQ4OTnh9u3b7KIiIiIyEpljZp8+fQpHR8dc68recqNvz549A4B8DSomIiIiw/Ls2bM8kxuza7lRqVS4d+8eSpYsadIL9GVmuObSQmVO18trNV3mdL28VtOlq+uVJAnPnj2Dm5ubxnCV7Jhdy42FhQXc3d3lDkNvzG2GmDldL6/VdJnT9fJaTZcurjevFptMsm+cSURERKRNTG6IiIjIpDC5MVE2NjYIDw+HjY2N3KHohTldL6/VdJnT9fJaTZchXK/ZDSgmIiIi08aWGyIiIjIpTG6IiIjIpDC5ISIiIpPC5IaIiIhMCpMbIxQZGYnGjRujZMmSKF++PDp16oTLly/n+pwVK1ZAoVBo3GxtbfUUcdFERERkib1mzZq5Pmfjxo2oWbMmbG1tUa9ePezcuVNP0RaNp6dnlmtVKBQYOnRotvWN7X39888/0b59e7i5uUGhUGDr1q0aj0uShLCwMLi6uqJ48eIIDAzElStX8jzv/Pnz4enpCVtbW/j5+eHYsWM6uoL8y+1aX716hbFjx6JevXooUaIE3Nzc0KdPH9y7dy/Xcxbms6APeb2vffv2zRJ369at8zyvIb6vQN7Xm91nWKFQYPr06Tme0xDf2/x817x8+RJDhw5FmTJlYG9vj86dO+P+/fu5nrewn/OCYHJjhA4ePIihQ4fi6NGjiImJwatXr/DBBx8gJSUl1+c5ODggPj5efbt586aeIi66OnXqaMR++PDhHOv+9ddf6NmzJ/r3749Tp06hU6dO6NSpE86fP6/HiAvn+PHjGtcZExMDAOjatWuOzzGm9zUlJQXe3t6YP39+to9PmzYNc+fOxcKFC/H333+jRIkSCAoKwsuXL3M854YNGxAaGorw8HCcPHkS3t7eCAoKQmJioq4uI19yu9YXL17g5MmT+Oabb3Dy5ElERUXh8uXL6NChQ57nLchnQV/yel8BoHXr1hpxr1u3LtdzGur7CuR9vW9eZ3x8PJYtWwaFQoHOnTvnel5De2/z813zxRdfYPv27di4cSMOHjyIe/fuITg4ONfzFuZzXmASGb3ExEQJgHTw4MEc6yxfvlxydHTUX1BaFB4eLnl7e+e7frdu3aR27dpplPn5+Umff/65liPTvZEjR0pVq1aVVCpVto8b8/sKQNqyZYv6WKVSSS4uLtL06dPVZU+fPpVsbGykdevW5XieJk2aSEOHDlUfK5VKyc3NTYqMjNRJ3IXx9rVm59ixYxIA6ebNmznWKehnQQ7ZXWtISIjUsWPHAp3HGN5XScrfe9uxY0fp3XffzbWOMby3b3/XPH36VLKyspI2btyorhMXFycBkGJjY7M9R2E/5wXFlhsTkJSUBAAoXbp0rvWeP3+OSpUqwcPDAx07dsSFCxf0EZ5WXLlyBW5ubqhSpQp69eqFW7du5Vg3NjYWgYGBGmVBQUGIjY3VdZhalZ6ejtWrV+PTTz/NdZNXY35f33T9+nUkJCRovHeOjo7w8/PL8b1LT0/HiRMnNJ5jYWGBwMBAo3u/k5KSoFAo4OTklGu9gnwWDMmBAwdQvnx51KhRA4MHD8ajR49yrGtK7+v9+/exY8cO9O/fP8+6hv7evv1dc+LECbx69UrjfapZsyYqVqyY4/tUmM95YTC5MXIqlQqjRo1Cs2bNULdu3Rzr1ahRA8uWLcO2bduwevVqqFQqNG3aFHfu3NFjtIXj5+eHFStWIDo6GgsWLMD169cREBCAZ8+eZVs/ISEBzs7OGmXOzs5ISEjQR7has3XrVjx9+hR9+/bNsY4xv69vy3x/CvLePXz4EEql0ujf75cvX2Ls2LHo2bNnrhsNFvSzYChat26NVatWYd++fZg6dSoOHjyINm3aQKlUZlvfVN5XAFi5ciVKliyZZ1eNob+32X3XJCQkwNraOktCntv7VJjPeWGY3a7gpmbo0KE4f/58nn2z/v7+8Pf3Vx83bdoUtWrVwqJFizB58mRdh1kkbdq0Ud+vX78+/Pz8UKlSJfz666/5+mvIWC1duhRt2rSBm5tbjnWM+X0l4dWrV+jWrRskScKCBQtyrWusn4UePXqo79erVw/169dH1apVceDAAbz33nsyRqZ7y5YtQ69evfIc6G/o721+v2sMBVtujNiwYcPw+++/Y//+/XB3dy/Qc62srNCgQQNcvXpVR9HpjpOTE6pXr55j7C4uLllG69+/fx8uLi76CE8rbt68ib179+Kzzz4r0POM+X3NfH8K8t6VLVsWlpaWRvt+ZyY2N2/eRExMTK6tNtnJ67NgqKpUqYKyZcvmGLexv6+ZDh06hMuXLxf4cwwY1nub03eNi4sL0tPT8fTpU436ub1PhfmcFwaTGyMkSRKGDRuGLVu24I8//kDlypULfA6lUolz587B1dVVBxHq1vPnz3Ht2rUcY/f398e+ffs0ymJiYjRaOAzd8uXLUb58ebRr165AzzPm97Vy5cpwcXHReO+Sk5Px999/5/jeWVtbw9fXV+M5KpUK+/btM/j3OzOxuXLlCvbu3YsyZcoU+Bx5fRYM1Z07d/Do0aMc4zbm9/VNS5cuha+vL7y9vQv8XEN4b/P6rvH19YWVlZXG+3T58mXcunUrx/epMJ/zwgZPRmbw4MGSo6OjdODAASk+Pl59e/HihbpO7969pa+++kp9PHHiRGn37t3StWvXpBMnTkg9evSQbG1tpQsXLshxCQUyevRo6cCBA9L169elI0eOSIGBgVLZsmWlxMRESZKyXuuRI0ekYsWKSTNmzJDi4uKk8PBwycrKSjp37pxcl1AgSqVSqlixojR27Ngsjxn7+/rs2TPp1KlT0qlTpyQA0qxZs6RTp06pZwhNmTJFcnJykrZt2yadPXtW6tixo1S5cmUpNTVVfY53331Xmjdvnvp4/fr1ko2NjbRixQrp4sWL0sCBAyUnJycpISFB79f3ptyuNT09XerQoYPk7u4unT59WuNznJaWpj7H29ea12dBLrld67Nnz6QxY8ZIsbGx0vXr16W9e/dKDRs2lLy8vKSXL1+qz2Es76sk5f17LEmSlJSUJNnZ2UkLFizI9hzG8N7m57tm0KBBUsWKFaU//vhD+t///if5+/tL/v7+GuepUaOGFBUVpT7Oz+e8qJjcGCEA2d6WL1+urtOiRQspJCREfTxq1CipYsWKkrW1teTs7Cy1bdtWOnnypP6DL4Tu3btLrq6ukrW1tVShQgWpe/fu0tWrV9WPv32tkiRJv/76q1S9enXJ2tpaqlOnjrRjxw49R114u3fvlgBIly9fzvKYsb+v+/fvz/Z3N/OaVCqV9M0330jOzs6SjY2N9N5772X5OVSqVEkKDw/XKJs3b57659CkSRPp6NGjerqinOV2rdevX8/xc7x//371Od6+1rw+C3LJ7VpfvHghffDBB1K5cuUkKysrqVKlStKAAQOyJCnG8r5KUt6/x5IkSYsWLZKKFy8uPX36NNtzGMN7m5/vmtTUVGnIkCFSqVKlJDs7O+mjjz6S4uPjs5znzefk53NeVIr/f2EiIiIik8AxN0RERGRSmNwQERGRSWFyQ0RERCaFyQ0RERGZFCY3REREZFKY3BAREZFJYXJDREREJoXJDRGZJYVCga1bt8odBhHpAJMbItK7vn37QqFQZLm1bt1a7tCIyAQUkzsAIjJPrVu3xvLlyzXKbGxsZIqGiEwJW26ISBY2NjZwcXHRuJUqVQqA6DJasGAB2rRpg+LFi6NKlSrYtGmTxvPPnTuHd999F8WLF0eZMmUwcOBAPH/+XKPOsmXLUKdOHdjY2MDV1RXDhg3TePzhw4f46KOPYGdnBy8vL/z222/qx548eYJevXqhXLlyKF68OLy8vLIkY0RkmJjcEJFB+uabb9C5c2ecOXMGvXr1Qo8ePRAXFwcASElJQVBQEEqVKoXjx49j48aN2Lt3r0bysmDBAgwdOhQDBw7EuXPn8Ntvv6FatWoarzFx4kR069YNZ8+eRdu2bdGrVy88fvxY/foXL17Erl27EBcXhwULFqBs2bL6+wEQUeFpdRtOIqJ8CAkJkSwtLaUSJUpo3L777jtJksQuwoMGDdJ4jp+fnzR48GBJkiRp8eLFUqlSpaTnz5+rH9+xY4dkYWGh3m3azc1NGj9+fI4xAJAmTJigPn7+/LkEQNq1a5ckSZLUvn17qV+/ftq5YCLSK465ISJZtGrVCgsWLNAoK126tPq+v7+/xmP+/v44ffo0ACAuLg7e3t4oUaKE+vFmzZpBpVLh8uXLUCgUuHfvHt57771cY6hfv776fokSJeDg4IDExEQAwODBg9G5c2ecPHkSH3zwATp16oSmTZsW6lqJSL+Y3BCRLEqUKJGlm0hbihcvnq96VlZWGscKhQIqlQoA0KZNG9y8eRM7d+5ETEwM3nvvPQwdOhQzZszQerxEpF0cc0NEBuno0aNZjmvVqgUAqFWrFs6cOYOUlBT140eOHIGFhQVq1KiBkiVLwtPTE/v27StSDOXKlUNISAhWr16NOXPmYPHixUU6HxHpB1tuiEgWaWlpSEhI0CgrVqyYetDuxo0b0ahRI/znP//BmjVrcOzYMSxduhQA0KtXL4SHhyMkJAQRERF48OABhg8fjt69e8PZ2RkAEBERgUGDBqF8+fJo06YNnj17hiNHjmD48OH5ii8sLAy+vr6oU6cO0tLS8Pvvv6uTKyIybExuiEgW0dHRcHV11SirUaMGLl26BEDMZFq/fj2GDBkCV1dXrFu3DrVr1wYA2NnZYffu3Rg5ciQaN24MOzs7dO7cGbNmzVKfKyQkBC9fvsTs2bMxZswYlC1bFl26dMl3fNbW1hg3bhxu3LiB4sWLIyAgAOvXr9fClRORrikkSZLkDoKI6E0KhQJbtmxBp06d5A6FiIwQx9wQERGRSWFyQ0RERCaFY26IyOCwt5yIioItN0RERGRSmNwQERGRSWFyQ0RERCaFyQ0RERGZFCY3REREZFKY3BAREZFJYXJDREREJoXJDREREZkUJjdERERkUv4PFmUeh6DiEyMAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(epochs, loss_values, 'bo', label='Training loss')\n",
    "plt.plot(epochs, val_loss_values, 'b', label= 'Validation loss')\n",
    "plt.title('Training and Validation loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "99173299",
   "metadata": {},
   "source": [
    "### Training Accuracy vs Validation Accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "4d9f4540",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.clf()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "21d62503",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_acc_values=history_dict['accuracy']\n",
    "val_acc_values = history_dict['val_accuracy']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "f825abbf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1e2b9507a30>"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(epochs, train_acc_values, 'bo', label='Training Accuracy')\n",
    "plt.plot(epochs, val_acc_values, 'b', label= 'Validation Accuracy')\n",
    "plt.title('Training and Validation Accuracy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c351b0b9",
   "metadata": {},
   "source": [
    "### Retraining a Model from Scratch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "acfab7c7",
   "metadata": {},
   "outputs": [],
   "source": [
    "model2 = models.Sequential()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "782d5a04",
   "metadata": {},
   "outputs": [],
   "source": [
    "model2.add(layers.Dense(16, activation = 'relu', input_shape=(10000,)))\n",
    "model2.add(layers.Dense(16, activation ='relu'))\n",
    "model2.add(layers.Dense(1, activation = 'sigmoid'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "3214bb57",
   "metadata": {},
   "outputs": [],
   "source": [
    "model2.compile(optimizer='rmsprop', loss='binary_crossentropy', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "db4bd03e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/4\n",
      "49/49 [==============================] - 2s 14ms/step - loss: 0.4633 - accuracy: 0.8157\n",
      "Epoch 2/4\n",
      "49/49 [==============================] - 1s 13ms/step - loss: 0.2765 - accuracy: 0.9019\n",
      "Epoch 3/4\n",
      "49/49 [==============================] - 1s 16ms/step - loss: 0.2141 - accuracy: 0.9234\n",
      "Epoch 4/4\n",
      "49/49 [==============================] - 1s 16ms/step - loss: 0.1808 - accuracy: 0.9363\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1e2b72904f0>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model2.fit(x_train, y_train, epochs=4, batch_size = 512)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "54349c16",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "782/782 [==============================] - 2s 2ms/step - loss: 0.3010 - accuracy: 0.8796\n"
     ]
    }
   ],
   "source": [
    "results= model2.evaluate(x_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "769ee95d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.3009743392467499, 0.8795999884605408]"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results #Loos & accuracy of Model2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "754625f9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "782/782 [==============================] - 2s 2ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[0.16490719],\n",
       "       [0.99977785],\n",
       "       [0.6789138 ],\n",
       "       ...,\n",
       "       [0.09131444],\n",
       "       [0.06476244],\n",
       "       [0.52512085]], dtype=float32)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model2.predict(x_test) #Likelihood of reviews being Positive or Negative"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c6121956",
   "metadata": {},
   "source": [
    "### Build the model using 3 hidden layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "3db7892d",
   "metadata": {},
   "outputs": [],
   "source": [
    "model3 = models.Sequential()\n",
    "model3.add(layers.Dense(16, activation = 'relu', input_shape=(10000,)))\n",
    "model3.add(layers.Dense(16, activation ='relu'))\n",
    "model3.add(layers.Dense(16, activation ='relu'))\n",
    "model3.add(layers.Dense(1, activation = 'sigmoid'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "af2fa748",
   "metadata": {},
   "outputs": [],
   "source": [
    "model3.compile(optimizer='rmsprop', loss='binary_crossentropy', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "f3cf1bd9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/4\n",
      "49/49 [==============================] - 2s 14ms/step - loss: 0.4526 - accuracy: 0.8117\n",
      "Epoch 2/4\n",
      "49/49 [==============================] - 1s 14ms/step - loss: 0.2614 - accuracy: 0.9050\n",
      "Epoch 3/4\n",
      "49/49 [==============================] - 1s 15ms/step - loss: 0.2046 - accuracy: 0.9246\n",
      "Epoch 4/4\n",
      "49/49 [==============================] - 1s 13ms/step - loss: 0.1735 - accuracy: 0.9372\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1e2b5dc18d0>"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model3.fit(x_train, y_train, epochs=4, batch_size = 512)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "a39d0bd4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "782/782 [==============================] - 2s 3ms/step - loss: 0.2943 - accuracy: 0.8836\n"
     ]
    }
   ],
   "source": [
    "results= model3.evaluate(x_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "405777fb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.2942780554294586, 0.8835999965667725]"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2dc5c9d3",
   "metadata": {},
   "source": [
    "### Let's use different activation functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "dd2e6815",
   "metadata": {},
   "outputs": [],
   "source": [
    "model4 = models.Sequential()\n",
    "model4.add(layers.Dense(16, activation = 'tanh', input_shape=(10000,)))\n",
    "model4.add(layers.Dense(16, activation ='tanh'))\n",
    "model4.add(layers.Dense(1, activation = 'sigmoid'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "1caa319e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/4\n",
      "49/49 [==============================] - 2s 15ms/step - loss: 0.4380 - accuracy: 0.8161\n",
      "Epoch 2/4\n",
      "49/49 [==============================] - 1s 17ms/step - loss: 0.2488 - accuracy: 0.9073\n",
      "Epoch 3/4\n",
      "49/49 [==============================] - 1s 18ms/step - loss: 0.1923 - accuracy: 0.9279\n",
      "Epoch 4/4\n",
      "49/49 [==============================] - 1s 16ms/step - loss: 0.1624 - accuracy: 0.9404\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1e2b43cd840>"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model4.compile(optimizer='rmsprop', loss='binary_crossentropy', metrics=['accuracy'])\n",
    "model4.fit(x_train, y_train, epochs=4, batch_size = 512)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "23e0aa66",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "782/782 [==============================] - 6s 7ms/step - loss: 0.3067 - accuracy: 0.8794\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.3067285418510437, 0.879360020160675]"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results= model4.evaluate(x_test, y_test)\n",
    "results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff426612",
   "metadata": {},
   "source": [
    "### Using mse as loss function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "27ef9bc2",
   "metadata": {},
   "outputs": [],
   "source": [
    "model5 = models.Sequential()\n",
    "model5.add(layers.Dense(32, activation = 'tanh', input_shape=(10000,)))\n",
    "model5.add(layers.Dense(32, activation ='tanh'))\n",
    "model5.add(layers.Dense(1, activation = 'sigmoid'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "0f43705a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/4\n",
      "30/30 [==============================] - 6s 104ms/step - loss: 0.6847 - accuracy: 0.5549 - val_loss: 0.6723 - val_accuracy: 0.6081\n",
      "Epoch 2/4\n",
      "30/30 [==============================] - 2s 62ms/step - loss: 0.6612 - accuracy: 0.6441 - val_loss: 0.6507 - val_accuracy: 0.6827\n",
      "Epoch 3/4\n",
      "30/30 [==============================] - 2s 57ms/step - loss: 0.6393 - accuracy: 0.7049 - val_loss: 0.6305 - val_accuracy: 0.7204\n",
      "Epoch 4/4\n",
      "30/30 [==============================] - 1s 46ms/step - loss: 0.6182 - accuracy: 0.7401 - val_loss: 0.6106 - val_accuracy: 0.7456\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1e2b84572b0>"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from keras import optimizers\n",
    "\n",
    "model5.compile(optimizer=optimizers.SGD(), loss='binary_crossentropy', metrics=['accuracy'])\n",
    "model5.fit(partial_x_train, partial_y_train, epochs=4, batch_size=512, validation_data=(x_val, y_val))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "bd6641c7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "782/782 [==============================] - 3s 3ms/step - loss: 0.6119 - accuracy: 0.7397\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.6118904948234558, 0.7397199869155884]"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results= model5.evaluate(x_test, y_test)\n",
    "results"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}