hyperparameter_optimization-Copy1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "bca3be57",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import pickle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "522acb44",
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_dataset(dsname):\n",
    "    metadata = pd.read_csv(f\"/datasets/nicolas_facchinetti/processed_data/{dsname}/metadata_final.csv\")\n",
    "    x = pickle.load(open(f\"/datasets/nicolas_facchinetti/processed_data/{dsname}/processed_data0.p\", \"rb\" ))\n",
    "    y = pickle.load(open(f\"/datasets/nicolas_facchinetti/processed_data/{dsname}/processed_labels.p\", \"rb\" ))\n",
    "    return metadata, x, y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "8c4a903c",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "def standardize(data):\n",
    "    scaler = StandardScaler()\n",
    "    n = data.shape\n",
    "    return scaler.fit_transform(data.reshape(n[0],-1)).reshape(n)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "8c97d4e1",
   "metadata": {},
   "outputs": [],
   "source": [
    "datasets = [\"emodb\", \"emovo\", \"ravdess\"]\n",
    "data = {}\n",
    "for d in datasets:\n",
    "    md, x, y = load_dataset(d)\n",
    "    data[d] = {}\n",
    "    data[d][\"x\"] = standardize(x)\n",
    "    data[d][\"y\"] = y\n",
    "    data[d][\"metadata\"] = md"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8b18f3a7",
   "metadata": {},
   "source": [
    "Get train/test data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "2d647052",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "train = {}\n",
    "test = {}\n",
    "for d in datasets:\n",
    "    train[d] = {}\n",
    "    test[d] = {}\n",
    "    train_df, test_df = train_test_split(data[d]['metadata'], \n",
    "                                       test_size = 0.2, \n",
    "                                       random_state = 1938,\n",
    "                                       stratify = data[d]['metadata']['label'])\n",
    "    train_index = train_df.index\n",
    "    test_index = test_df.index\n",
    "    train[d]['x'] = data[d]['x'][train_index]\n",
    "    train[d]['y'] = data[d]['y'][train_index]\n",
    "    test[d]['x']= data[d]['x'][test_index]\n",
    "    test[d]['y']= data[d]['y'][test_index]\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "a4d0ca86",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((4181, 5), (3344, 5), (837, 5))"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['emodb']['y'].shape, train['emodb']['y'].shape, test['emodb']['y'].shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bd2bc608",
   "metadata": {},
   "source": [
    "# Model hyperparameter"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "89a3d08d",
   "metadata": {},
   "source": [
    "Find the optimal number of LSTM unit for each dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "3f35aef4",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-02-28 11:36:01.188418: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.10.1\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "import tensorflow.keras\n",
    "import keras_tuner as kt\n",
    "from tensorflow.keras.models import Sequential \n",
    "from tensorflow.keras.callbacks import EarlyStopping, ReduceLROnPlateau\n",
    "from tensorflow.keras.layers import Conv2D, Flatten, Dense, Dropout, MaxPooling2D, LSTM, TimeDistributed, InputLayer, Reshape, BatchNormalization, Bidirectional"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "3a88a8c6",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_m1(hp):\n",
    "    # Define hyper model architecture\n",
    "    m = Sequential([\n",
    "        InputLayer(input_shape=(261,128,1)),\n",
    "        Reshape((9,29,128,1)),\n",
    "        TimeDistributed(Conv2D(16, kernel_size=(5,5), activation='relu')),\n",
    "        TimeDistributed(BatchNormalization()),\n",
    "        TimeDistributed(MaxPooling2D(pool_size=(4,4), strides=2)),\n",
    "\n",
    "        TimeDistributed(Conv2D(32, kernel_size=(3,3), activation='relu')),\n",
    "        TimeDistributed(MaxPooling2D(pool_size=(2,2), strides=2)),\n",
    "\n",
    "        TimeDistributed(Conv2D(64, kernel_size=(3,3), activation='relu')),\n",
    "        TimeDistributed(MaxPooling2D(pool_size=(2,2), strides=1)),\n",
    "        TimeDistributed(Flatten()),\n",
    "        \n",
    "        # Tune number of units (4, 8, 16, 32, 64, 128, 256, 512, 1024)\n",
    "        Bidirectional(LSTM(hp.Choice(\"lstm_units\",  values=[4, 8, 16, 32, 64, 128, 256, 512, 1024]),\n",
    "                           dropout=0.2,\n",
    "                           return_sequences=False)),\n",
    "        Dense(5, activation='softmax')\n",
    "    ])\n",
    "    \n",
    "    \n",
    "    # Define optimizer, loss, and metrics\n",
    "    m.compile(optimizer='adam',\n",
    "              loss='categorical_crossentropy',\n",
    "              metrics=[\"accuracy\"])\n",
    "    return m"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "08580683",
   "metadata": {},
   "outputs": [],
   "source": [
    "def search_lstm(ds):\n",
    "    # instantiate the tuner\n",
    "    tuner = kt.tuners.Hyperband(get_m1,\n",
    "                                objective=\"val_loss\",\n",
    "                                max_epochs=120,\n",
    "                                # number of models that should be built and fit for each trial with the same hyperparameter values\n",
    "                                executions_per_trial=3,\n",
    "                                # the reduction factor for the number of epochs and number of models for each bracket\n",
    "                                factor=15,\n",
    "                                # the number of times to iterate over the full Hyperband algorithm\n",
    "                                hyperband_iterations=3,\n",
    "                                overwrite=True,\n",
    "                                directory=\"/datasets/nicolas_facchinetti/kt_dir1\",\n",
    "                                project_name=\"cnnlstm1\")\n",
    "\n",
    "    earlyStopping = EarlyStopping(monitor='val_loss', patience=10, verbose=1, mode='min', restore_best_weights=True)\n",
    "    reduce_lr_loss = ReduceLROnPlateau(monitor='val_loss', factor=0.1, patience=6, verbose=1, min_delta=1e-4, mode='min')\n",
    "\n",
    "    tuner.search(train[ds]['x'], train[ds]['y'],\n",
    "                 epochs=100,\n",
    "                 validation_split=0.2,\n",
    "                 callbacks=[earlyStopping, reduce_lr_loss],\n",
    "                 verbose=2)\n",
    "\n",
    "    tuner.results_summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "5a256432",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trial 9 Complete [00h 01m 41s]\n",
      "val_loss: 0.3975821038087209\n",
      "\n",
      "Best val_loss So Far: 0.2849500874678294\n",
      "Total elapsed time: 00h 14m 02s\n",
      "INFO:tensorflow:Oracle triggered exit\n",
      "Results summary\n",
      "Results in /datasets/nicolas_facchinetti/kt_dir1/cnnlstm1\n",
      "Showing 10 best trials\n",
      "<keras_tuner.engine.objective.Objective object at 0x7f7912190f70>\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 512\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.2849500874678294\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 256\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.2930381695429484\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 128\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.30164288481076557\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 1024\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.3155254324277242\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 64\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.3293354610602061\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 32\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.3975821038087209\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 16\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.4528856774171193\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 8\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.626187781492869\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 4\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.8504553039868673\n"
     ]
    }
   ],
   "source": [
    "res1 = search_lstm('emodb')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "07835e06",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trial 9 Complete [00h 02m 06s]\n",
      "val_loss: 0.4782815972963969\n",
      "\n",
      "Best val_loss So Far: 0.47662363449732464\n",
      "Total elapsed time: 00h 18m 23s\n",
      "INFO:tensorflow:Oracle triggered exit\n",
      "Results summary\n",
      "Results in /datasets/nicolas_facchinetti/kt_dir1/cnnlstm1\n",
      "Showing 10 best trials\n",
      "<keras_tuner.engine.objective.Objective object at 0x7f7701d71040>\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 512\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.47662363449732464\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 256\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.4782815972963969\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 1024\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.4845413863658905\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 128\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.5194460848967234\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 64\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.5630144874254862\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 32\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.7065879305203756\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 16\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.7795344193776449\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 8\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 1.0190823276837666\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 4\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 1.293078859647115\n"
     ]
    }
   ],
   "source": [
    "res2 = search_lstm('emovo')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "743c6820",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trial 9 Complete [00h 02m 40s]\n",
      "val_loss: 0.4444417556126912\n",
      "\n",
      "Best val_loss So Far: 0.30376137296358746\n",
      "Total elapsed time: 00h 25m 11s\n",
      "INFO:tensorflow:Oracle triggered exit\n",
      "Results summary\n",
      "Results in /datasets/nicolas_facchinetti/kt_dir1/cnnlstm1\n",
      "Showing 10 best trials\n",
      "<keras_tuner.engine.objective.Objective object at 0x7f77308c0700>\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 1024\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.30376137296358746\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 256\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.3299727439880371\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 512\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.33017777403195697\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 128\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.3831322391827901\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 64\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.4444417556126912\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 32\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.604045569896698\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 16\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.7923407753308614\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 8\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.9179666042327881\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "lstm_units: 4\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 1.2668561538060505\n"
     ]
    }
   ],
   "source": [
    "res3 = search_lstm('ravdess')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7535191f",
   "metadata": {},
   "source": [
    "Going with 256 LSTM units enure great perfomance boost with a lower risk of overfitting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "46b8ec80",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "reshape_1 (Reshape)          (None, 9, 29, 128, 1)     0         \n",
      "_________________________________________________________________\n",
      "time_distributed_8 (TimeDist (None, 9, 25, 124, 16)    416       \n",
      "_________________________________________________________________\n",
      "time_distributed_9 (TimeDist (None, 9, 25, 124, 16)    64        \n",
      "_________________________________________________________________\n",
      "time_distributed_10 (TimeDis (None, 9, 11, 61, 16)     0         \n",
      "_________________________________________________________________\n",
      "time_distributed_11 (TimeDis (None, 9, 9, 59, 32)      4640      \n",
      "_________________________________________________________________\n",
      "time_distributed_12 (TimeDis (None, 9, 4, 29, 32)      0         \n",
      "_________________________________________________________________\n",
      "time_distributed_13 (TimeDis (None, 9, 2, 27, 64)      18496     \n",
      "_________________________________________________________________\n",
      "time_distributed_14 (TimeDis (None, 9, 1, 26, 64)      0         \n",
      "_________________________________________________________________\n",
      "time_distributed_15 (TimeDis (None, 9, 1664)           0         \n",
      "_________________________________________________________________\n",
      "bidirectional_1 (Bidirection (None, 512)               3934208   \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 5)                 2565      \n",
      "=================================================================\n",
      "Total params: 3,960,389\n",
      "Trainable params: 3,960,357\n",
      "Non-trainable params: 32\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "def get_m():\n",
    "    # Define hyper model architecture\n",
    "    m = Sequential([\n",
    "        InputLayer(input_shape=(261,128,1)),\n",
    "        Reshape((9,29,128,1)),\n",
    "        TimeDistributed(Conv2D(16, kernel_size=(5,5), activation='relu')),\n",
    "        TimeDistributed(BatchNormalization()),\n",
    "        TimeDistributed(MaxPooling2D(pool_size=(4,4), strides=2)),\n",
    "\n",
    "        TimeDistributed(Conv2D(32, kernel_size=(3,3), activation='relu')),\n",
    "        TimeDistributed(MaxPooling2D(pool_size=(2,2), strides=2)),\n",
    "\n",
    "        TimeDistributed(Conv2D(64, kernel_size=(3,3), activation='relu')),\n",
    "        TimeDistributed(MaxPooling2D(pool_size=(2,2), strides=1)),\n",
    "        TimeDistributed(Flatten()),\n",
    "        \n",
    "        Bidirectional(LSTM(256,\n",
    "                           dropout=0.2,\n",
    "                           return_sequences=False)),\n",
    "        Dense(5, activation='softmax')\n",
    "    ])\n",
    "    \n",
    "    \n",
    "    return m\n",
    "\n",
    "get_m().summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d1873f74",
   "metadata": {},
   "source": [
    "### Test LSTM architecture"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "ab1505c7",
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_evaluate(ds):\n",
    "    m = get_m()\n",
    "    m.compile(optimizer='adam',\n",
    "              loss='categorical_crossentropy',\n",
    "              metrics=[\"accuracy\"])\n",
    "    earlyStopping = EarlyStopping(monitor='val_loss', patience=10, verbose=1, mode='min', restore_best_weights=True)\n",
    "    reduce_lr_loss = ReduceLROnPlateau(monitor='val_loss', factor=0.1, patience=6, verbose=1, min_delta=1e-4, mode='min')\n",
    "    \n",
    "    # fit & predict\n",
    "    history = m.fit(train[ds]['x'], train[ds]['y'],\n",
    "                    epochs=50,\n",
    "                    batch_size=32,\n",
    "                    callbacks=[earlyStopping, reduce_lr_loss],\n",
    "                    validation_split=0.2,\n",
    "                    verbose=2)\n",
    "    # test\n",
    "    results = m.evaluate(test[ds]['x'], test[ds]['y'], verbose=0)\n",
    "    print(\"\\t\\tTest loss, test acc:\", results)\n",
    "    return history, results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9d829c56",
   "metadata": {
    "scrolled": true
   },
   "source": [
    "res = {}\n",
    "for ds in data:\n",
    "    print(f'\\nTraining on {ds}')\n",
    "    res[ds] = train_evaluate(ds)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "0e84c0cb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "For emodb test loss: 0.37915724515914917 accuracy: 0.8888888955116272\n",
      "For emovo test loss: 0.3849480450153351 accuracy: 0.8733536005020142\n",
      "For ravdess test loss: 0.35736197233200073 accuracy: 0.87325519323349\n"
     ]
    }
   ],
   "source": [
    "start_df = []\n",
    "for d in res:\n",
    "    print(f'For {d} test loss: {res[d][1][0]} accuracy: {res[d][1][1]}')\n",
    "    start_df.append([res[d][1][0], res[d][1][1]])\n",
    "start_df = pd.DataFrame(start_df, columns=['S loss', 'S acc'], index=res.keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "8811acd2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>S loss</th>\n",
       "      <th>S acc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>emodb</th>\n",
       "      <td>0.379157</td>\n",
       "      <td>0.888889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>emovo</th>\n",
       "      <td>0.384948</td>\n",
       "      <td>0.873354</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ravdess</th>\n",
       "      <td>0.357362</td>\n",
       "      <td>0.873255</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           S loss     S acc\n",
       "emodb    0.379157  0.888889\n",
       "emovo    0.384948  0.873354\n",
       "ravdess  0.357362  0.873255"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "start_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "26a3a420",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "def plot_res_acc(h, x, y):\n",
    "    plt.subplot(x, 2, y)\n",
    "    plt.plot(h.history['accuracy'])\n",
    "    plt.plot(h.history['val_accuracy'])\n",
    "    plt.title('model accuracy')\n",
    "    plt.ylabel('accuracy')\n",
    "    plt.xlabel('epoch')\n",
    "    plt.legend(['train', 'val'], loc='upper left')\n",
    "    \n",
    "def plot_res_loss(h, x, y):\n",
    "    plt.subplot(x, 2, y)\n",
    "    plt.plot(h.history['loss'])\n",
    "    plt.plot(h.history['val_loss'])\n",
    "    plt.title('model loss')\n",
    "    plt.ylabel('loss')\n",
    "    plt.xlabel('epoch')\n",
    "    plt.legend(['train', 'val'], loc='upper left')\n",
    "    \n",
    "def plot_stats(h):\n",
    "    fig = plt.gcf()\n",
    "    fig.set_size_inches(15, 6)\n",
    "    for num, c in enumerate(h):\n",
    "        n = (num+1)*2\n",
    "        plot_res_acc(c, len(h), n-1)\n",
    "        plot_res_loss(c, len(h), n)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "77423b62",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "emodb\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABNEAAAIhCAYAAABg7ixdAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAA9hAAAPYQGoP6dpAAC+YUlEQVR4nOzdd3hUZfr/8c/MpHdCSAFCTSGAAoLSREUEBEFxLayrKK7ourqryOoq66or64oVsZefYtbvuoouFlQsqDQFC1WlBEILJSHU9Dpzfn+czEAMJeUkk0zer+s61zlzcs4z9wR2PdxzP89tMwzDEAAAAAAAAIATsns7AAAAAAAAAKC5I4kGAAAAAAAAnAJJNAAAAAAAAOAUSKIBAAAAAAAAp0ASDQAAAAAAADgFkmgAAAAAAADAKZBEAwAAAAAAAE6BJBoAAAAAAABwCiTRAAAAAAAAgFMgiQag2dixY4dsNpvS09PrfO/ixYtls9m0ePFiy+MCAABA89QUz488ZwJwI4kGAAAAAAAAnAJJNABoxkpKSmQYhrfDAAAAAIBWjyQaAI9//OMfstls+umnn3TFFVcoMjJS0dHRmjZtmiorK5WRkaELL7xQ4eHh6tKlix577LEaY2RlZemaa65RbGysAgMDlZaWpieffFIul6vadXv37tWVV16p8PBwRUZGauLEicrJyTluXCtXrtTFF1+s6OhoBQUFqV+/fnrnnXfq9Rn379+vW265RT179lRYWJhiY2N1/vnna9myZTWuLSsr04wZM5SWlqagoCC1bdtWw4cP1/Llyz3XuFwuPfvss+rbt6+Cg4MVFRWlQYMGaf78+Z5rbDab/vGPf9QYv0uXLpo8ebLndXp6umw2m7744gv9/ve/V7t27RQSEqKysjJlZmbq+uuvV3JyskJCQtShQweNHz9eP//8c41xjxw5or/85S/q1q2bAgMDFRsbq7Fjx2rTpk0yDEPJyckaPXp0jfsKCwsVGRmpW2+9tY6/VQAA0Fq1hufHE5k/f74GDx6skJAQhYeHa+TIkVqxYkW1a/bv36+bbrpJiYmJCgwMVLt27TR06FB9+eWXnmvWrFmjcePGeT5/+/btddFFF2n37t2Wxgug4fy8HQCA5ufKK6/UNddcoz/84Q9auHChHnvsMVVUVOjLL7/ULbfcojvvvFP//e9/dffddyspKUm/+c1vJJkPCUOGDFF5ebn++c9/qkuXLvr444915513auvWrXrhhRckmdVVF1xwgfbu3auZM2cqJSVFn3zyiSZOnFgjlkWLFunCCy/UwIED9dJLLykyMlJvv/22Jk6cqOLi4mpJqNo4dOiQJOmBBx5QfHy8CgsL9f777+u8887TV199pfPOO0+SVFlZqTFjxmjZsmWaOnWqzj//fFVWVuq7775TVlaWhgwZIkmaPHmy/vOf/+iGG27QjBkzFBAQoNWrV2vHjh31++VL+v3vf6+LLrpI//d//6eioiL5+/tr7969atu2rR555BG1a9dOhw4d0r///W8NHDhQa9asUWpqqiSpoKBAZ599tnbs2KG7775bAwcOVGFhoZYuXars7Gz16NFDf/7znzV16lRt2bJFycnJnvd94403lJ+fTxINAADUmS8/Px7Pf//7X1199dUaNWqU3nrrLZWVlemxxx7zPFOeffbZkqRJkyZp9erV+te//qWUlBQdOXJEq1ev1sGDByVJRUVFGjlypLp27arnn39ecXFxysnJ0aJFi1RQUNDgOAFYzACAKg888IAhyXjyySerne/bt68hyXjvvfc85yoqKox27doZv/nNbzzn7rnnHkOS8f3331e7/49//KNhs9mMjIwMwzAM48UXXzQkGR9++GG162688UZDkvH66697zvXo0cPo16+fUVFRUe3acePGGQkJCYbT6TQMwzAWLVpkSDIWLVpUp89cWVlpVFRUGCNGjDAuvfRSz/k33njDkGT8v//3/05479KlSw1Jxr333nvS95BkPPDAAzXOd+7c2bjuuus8r19//XVDknHttdfWKu7y8nIjOTnZuOOOOzznZ8yYYUgyFi5ceMJ78/PzjfDwcOP222+vdr5nz57G8OHDT/neAAAAbq3h+fHX1zmdTqN9+/bGaaed5hnLMAyjoKDAiI2NNYYMGeI5FxYWZkydOvWEY69cudKQZHzwwQcnjQFA88B0TgA1jBs3rtrrtLQ02Ww2jRkzxnPOz89PSUlJ2rlzp+fc119/rZ49e+qss86qdv/kyZNlGIa+/vprSea3g+Hh4br44ourXfe73/2u2uvMzExt2rRJV199tSSzOsy9jR07VtnZ2crIyKjz53vppZd0xhlnKCgoSH5+fvL399dXX32ljRs3eq759NNPFRQUpN///vcnHOfTTz+VJMsrty677LIa5yorK/Xwww+rZ8+eCggIkJ+fnwICArRly5YacaekpOiCCy444fjh4eG6/vrrlZ6erqKiIknmn92GDRv0pz/9ydLPAgAAWgdff348VkZGhvbu3atJkybJbj/6T+qwsDBddtll+u6771RcXCxJOuuss5Senq6HHnpI3333nSoqKqqNlZSUpDZt2ujuu+/WSy+9pA0bNjQoNgCNiyQagBqio6OrvQ4ICFBISIiCgoJqnC8tLfW8PnjwoBISEmqM1759e8/P3fu4uLga18XHx1d7vW/fPknSnXfeKX9//2rbLbfcIkk6cOBAnT7brFmz9Mc//lEDBw7UvHnz9N133+nHH3/UhRdeqJKSEs91+/fvV/v27as9GP3a/v375XA4asTdUMf7HU6bNk333XefJkyYoI8++kjff/+9fvzxR/Xp06dG3B07djzle/z5z39WQUGB3nzzTUnSc889p44dO+qSSy6x7oMAAIBWw5efH3/NHdOJ4na5XDp8+LAkae7cubruuuv06quvavDgwYqOjta1117rWcstMjJSS5YsUd++ffW3v/1NvXr1Uvv27fXAAw/USLgB8D7WRANgmbZt2yo7O7vG+b1790qSYmJiPNf98MMPNa779cKw7uunT5/uWTfj19xrgdXWf/7zH5133nl68cUXq53/9ZoT7dq10zfffCOXy3XCRFq7du3kdDqVk5Nz3Icot8DAQJWVldU4734A+zWbzXbcuK+99lo9/PDD1c4fOHBAUVFR1WKqzSK0SUlJGjNmjJ5//nmNGTNG8+fP14MPPiiHw3HKewEAAKzSEp4fjxezpBPGbbfb1aZNG088s2fP1uzZs5WVlaX58+frnnvuUW5urj777DNJ0mmnnaa3335bhmHop59+Unp6umbMmKHg4GDdc889DYoVgLWoRANgmREjRmjDhg1avXp1tfNvvPGGbDabhg8fLkkaPny4CgoKqnWwlMwFWo+Vmpqq5ORkrVu3TgMGDDjuFh4eXqcYbTabAgMDq5376aefanRSGjNmjEpLS5Wenn7CsdzTE36dkPu1Ll266Keffqp27uuvv1ZhYWGD4v7kk0+0Z8+eGjFt3rzZM/XhZG6//Xb99NNPuu666+RwOHTjjTfWOh4AAAArtITnx19LTU1Vhw4d9N///leGYXjOFxUVad68eZ6Onb/WqVMn/elPf9LIkSNrfF7JfN7r06ePnnrqKUVFRR33GgDeRSUaAMvccccdeuONN3TRRRdpxowZ6ty5sz755BO98MIL+uMf/6iUlBRJ0rXXXqunnnpK1157rf71r38pOTlZCxYs0Oeff15jzJdfflljxozR6NGjNXnyZHXo0EGHDh3Sxo0btXr1ar377rt1inHcuHH65z//qQceeEDnnnuuMjIyNGPGDHXt2lWVlZWe66666iq9/vrruvnmm5WRkaHhw4fL5XLp+++/V1pamn77299q2LBhmjRpkh566CHt27dP48aNU2BgoNasWaOQkBD9+c9/lmR2Zbrvvvt0//3369xzz9WGDRv03HPPKTIysk5xp6enq0ePHjr99NO1atUqPf744zWmbk6dOlVz587VJZdconvuuUdnnXWWSkpKtGTJEo0bN87zICpJI0eOVM+ePbVo0SJPW3kAAICm1BKeH3/Nbrfrscce09VXX61x48bpD3/4g8rKyvT444/ryJEjeuSRRyRJeXl5Gj58uH73u9+pR48eCg8P148//qjPPvvMUyX38ccf64UXXtCECRPUrVs3GYah9957T0eOHNHIkSMbFCeARuDVtgYAmhV3d6X9+/dXO3/dddcZoaGhNa4/99xzjV69elU7t3PnTuN3v/ud0bZtW8Pf399ITU01Hn/88WqdiwzDMHbv3m1cdtllRlhYmBEeHm5cdtllxvLly2t0VzIMw1i3bp1x5ZVXGrGxsYa/v78RHx9vnH/++cZLL73kuaa23ZXKysqMO++80+jQoYMRFBRknHHGGcYHH3xgXHfddUbnzp2rXVtSUmLcf//9RnJyshEQEGC0bdvWOP/8843ly5d7rnE6ncZTTz1l9O7d2wgICDAiIyONwYMHGx999FG19/zrX/9qJCYmGsHBwca5555rrF279oTdOX/88ccacR8+fNi44YYbjNjYWCMkJMQ4++yzjWXLlhnnnnuuce6559a49vbbbzc6depk+Pv7G7GxscZFF11kbNq0qca4//jHPwxJxnfffXfS3xsAAMDxtIbnxxNd98EHHxgDBw40goKCjNDQUGPEiBHGt99+6/l5aWmpcfPNNxunn366ERERYQQHBxupqanGAw88YBQVFRmGYRibNm0yrrrqKqN79+5GcHCwERkZaZx11llGenr6SWMC4B02wzim/hQA0KoMGDBANptNP/74o7dDAQAAAIBmjemcANDK5Ofn65dfftHHH3+sVatW6f333/d2SAAAAADQ7JFEA4BWZvXq1Ro+fLjatm2rBx54QBMmTPB2SAAAAADQ7DGdEwAAAAAAADgFu7cDAAAAAAAAAJo7kmgAAAAAAADAKZBEAwAAAAAAAE6h1TUWcLlc2rt3r8LDw2Wz2bwdDgAAaCEMw1BBQYHat28vu711fw+5dOlSPf7441q1apWys7P1/vvvn7RJyeTJk/Xvf/+7xvmePXtq/fr1kqT09HRdf/31Na4pKSlRUFBQreLiOQ8AANRHbZ/zWl0Sbe/evUpMTPR2GAAAoIXatWuXOnbs6O0wvKqoqEh9+vTR9ddfr8suu+yU1z/99NN65JFHPK8rKyvVp08fXXHFFdWui4iIUEZGRrVztU2gSTznAQCAhjnVc16rS6KFh4dLMn8xERERXo4GAAC0FPn5+UpMTPQ8S7RmY8aM0ZgxY2p9fWRkpCIjIz2vP/jgAx0+fLhG5ZnNZlN8fHy94+I5DwAA1Edtn/NaXRLNXdofERHBwxUAAKgzpgk23GuvvaYLLrhAnTt3rna+sLBQnTt3ltPpVN++ffXPf/5T/fr1O+E4ZWVlKisr87wuKCiQxHMeAACon1M957XuBT0AAADQpLKzs/Xpp59qypQp1c736NFD6enpmj9/vt566y0FBQVp6NCh2rJlywnHmjlzpqfKLTIykqmcAACgUZFEAwAAQJNJT09XVFRUjUYEgwYN0jXXXKM+ffpo2LBheuedd5SSkqJnn332hGNNnz5deXl5nm3Xrl2NHD0AAGjNWt10TgAAAHiHYRiaM2eOJk2apICAgJNea7fbdeaZZ560Ei0wMFCBgYFWhwkAAHBcJNGOwzAMVVZWyul0ejuUFsnhcMjPz481YwAAQDVLlixRZmambrjhhlNeaxiG1q5dq9NOO83SGHjOaxie8wAArRlJtF8pLy9Xdna2iouLvR1KixYSEqKEhIRTfssMAABansLCQmVmZnpeb9++XWvXrlV0dLQ6deqk6dOna8+ePXrjjTeq3ffaa69p4MCB6t27d40xH3zwQQ0aNEjJycnKz8/XM888o7Vr1+r555+3LG6e86zBcx4AoLUiiXYMl8ul7du3y+FwqH379goICOBbtjoyDEPl5eXav3+/tm/fruTkZNntLL0HAIAvWblypYYPH+55PW3aNEnSddddp/T0dGVnZysrK6vaPXl5eZo3b56efvrp44555MgR3XTTTcrJyVFkZKT69eunpUuX6qyzzrIkZp7zGo7nPABAa2czDMPwdhBNKT8/X5GRkcrLy6vR+ry0tFTbt29X586dFRIS4qUIfUNxcbF27typrl27KigoyNvhAADQYCd7hkDzwHNe0+A5DwDga2r7nMdXR8fBN2oNx+8QAAA0RzyjNBy/QwBAa8V/AQEAAAAAAIBTIIkGAAAAAAAAnAJJNNTQpUsXzZ4929thAAAAwGI85wEAUH905/QR5513nvr27WvJQ9GPP/6o0NDQhgcFAACABuM5DwCA5oEkWithGIacTqf8/E79R96uXbsmiAgAAABW4DkPAICmwXTOUzAMQ8XllV7ZDMOoVYyTJ0/WkiVL9PTTT8tms8lmsyk9PV02m02ff/65BgwYoMDAQC1btkxbt27VJZdcori4OIWFhenMM8/Ul19+WW28X5f522w2vfrqq7r00ksVEhKi5ORkzZ8/38pfMwAAQJPjOY/nPAAA6sKrlWhLly7V448/rlWrVik7O1vvv/++JkyYcNJ7lixZomnTpmn9+vVq3769/vrXv+rmm29utBhLKpzqef/njTb+yWyYMVohAaf+I3r66ae1efNm9e7dWzNmzJAkrV+/XpL017/+VU888YS6deumqKgo7d69W2PHjtVDDz2koKAg/fvf/9b48eOVkZGhTp06nfA9HnzwQT322GN6/PHH9eyzz+rqq6/Wzp07FR0dbc2HBQAAaGI855l4zgMAoHa8WolWVFSkPn366LnnnqvV9du3b9fYsWM1bNgwrVmzRn/729902223ad68eY0cafMWGRmpgIAAhYSEKD4+XvHx8XI4HJKkGTNmaOTIkerevbvatm2rPn366A9/+INOO+00JScn66GHHlK3bt1O+Y3j5MmTddVVVykpKUkPP/ywioqK9MMPPzTFxwMAAGi1eM4DAKD58Gol2pgxYzRmzJhaX//SSy+pU6dOnhL0tLQ0rVy5Uk888YQuu+yyRokx2N+hDTNGN8rYtXnvhhowYEC110VFRXrwwQf18ccfa+/evaqsrFRJSYmysrJOOs7pp5/uOQ4NDVV4eLhyc3MbHB8AoH5cLkNOw5DLMORySU7DkNNlyOUyVLtJYt5lGIZchuSqitvpMjzH5l6/en30/LHnnIb5md3nQgL8dE4Kaz6hdur8nFdRJlWWSI5AKSC4we/dUDznAQDQtFpUY4EVK1Zo1KhR1c6NHj1ar732mioqKuTv71/jnrKyMpWVlXle5+fn1+k9bTZbrUrtm6tfd1+666679Pnnn+uJJ55QUlKSgoODdfnll6u8vPyk4/z6d2uz2eRyuSyPF0DzUFbp1OGiCh0qKtehonIdLCrT4arjw8UVqnRZm6ax26QObYKV1C5MSbFh6hQdIj+Hd5ftLK1wKjO3UJv3FShjX4E25xRo1+GSWq9jVFuGcTQBZhiqnhg6JjHm/FWyDMeXHBumhdPO9XYYaCHq/JxXkiMV7ZfCYqWA8MYLrJZ4zgMAoGm1qOxQTk6O4uLiqp2Li4tTZWWlDhw4oISEhBr3zJw5Uw8++GBTheg1AQEBcjqdp7xu2bJlmjx5si699FJJUmFhoXbs2NHI0QE4ltNlqMLpqtrM4/JK87UhyWGzyWG3yWaTHHabHDab7MfuPcfHXms74fsZhqGCskodKizXoeJyc19UdVx0/K2wrLLpfiHHEeCwq2tMqJJiw9Q9NkzJsWZyrWtMqIIsqN44VqXTpZ2HipWRU6CMnAJP0mzHgSKRq2p8v/77fOzfcZvNJsdxzlf/34OqnUuMDvH2R4Ivc1QlmyormvRtec4DAKB5aFFJNEk1/qHorgg40T8gp0+frmnTpnle5+fnKzExsfEC9JIuXbro+++/144dOxQWFnbCbw+TkpL03nvvafz48bLZbLrvvvv4phGWq3S6VFhWqYLSSuWXVqiw1DwuKKsw957NfG1ee/RnpRXOE/7j2Z1Yslclj6odu/8hXuOcTXabZNOJE0314TSOSYZVGir3JMZcqnQe+9pQRaXL87oxEjM2m6p91mN/b0Vllapw1v1NHXab2oQEqG1ogNqE+qttaKDahPorOiRA/hZXiVW4DO08WKTM3EJt3V+o0gqXMqqSWcey26RO0SGe5FpSuzAlx4Wre7tQhQfVrEY+lmEY2ptXqs055rjupFnm/kKVVx7//wcjg/2VGh+u1LhwpcSHq2vbUPk5rP17ZFPV39mqv6/V/36b/33znD9OAvV499mtDbHRnCz5CzRLjgBz7zp5ZZfVeM4DAKB5aFFJtPj4eOXk5FQ7l5ubKz8/P7Vt2/a49wQGBiowMLApwvOqO++8U9ddd5169uypkpISvf7668e97qmnntLvf/97DRkyRDExMbr77rvrPMUVMAxDm3IK9MX6fVqddVh5JRXVEmLF5af+thxH+dlt8nfY5ecw03yu40zpO9UMQsOQKg1DJ8vQhQQ4zKRYWICiQwMUHVK1DzvmuGprGxqo8CA/2b2QjXG5DO05UqLM3ELPtiW3QJm5hcovrdSOg8XacbBYX26svl5PfESQkuPC1L1qSmj7qCBlHSxWxr5CZeTka8u+QhWcoMIu2N+hlLgwpcSFKzU+XClx4eoRH6524YEkegAc5U6iOZu2Eo3nPAAAmgebYfXiLvVks9n0/vvva8KECSe85u6779ZHH32kDRs2eM798Y9/1Nq1a7VixYpavU9+fr4iIyOVl5eniIiIaj8rLS3V9u3b1bVrVwUFBdXrc8DE79L3VDpdWrnzsL5Yv09fbMjR7sMlp7wnyN+u8CB/hQf6KTzIzzwO8lNY4NHjo9vRnwUHODxrQ9VlQXHXr+759QLlVrPbbArws8vfYVOAw+5JhAU47PL3M18f+zP/Y671q/qZv91eq0SVcbzPd8znP3a9rOrnDIUE+qltaIDl0yCbmmEY2l9YVi255t5yC8pOPYDMhGX3dmFKiQ9XalXSrEd8hDq2CfZKwhAty8meIdA8NPpzXmW5lLtekk1K6GOWAbdCPOcBAHxNbZ/zvFqJVlhYqMzMTM/r7du3a+3atYqOjlanTp00ffp07dmzR2+88YYk6eabb9Zzzz2nadOm6cYbb9SKFSv02muv6a233vLWRwB8WnF5pZZuPqCFG/bpq037dKT46DfvgX52DUuO0XmpsYqLCKpKjPkpwp0MC/KzfMpfa2ZO6TOn/bVWNptNseFBig0P0pDuMdV+lldSYU4FzS1U5n4zsbb3SIkSo0M8UzFT48LVNSZUAX78vQRQT+410WRIroqjlWkAAKBV8GoSbeXKlRo+fLjntXvtsuuuu07p6enKzs6u1pK7a9euWrBgge644w49//zzat++vZ555hlddtllTR474KsOFJbp6425+mJDjpZtOaCyY9aKigrx14gecRrZM07npMS06M618C2Rwf7q37mN+ndu4+1QAPgym02y+5sJNCdJNAAAWhuv/gv4vPPO08lmk6anp9c4d+6552r16tWNGBXQ+uw4UKQvNuRo4YZ9WrnzcLX1tzq2CdaonvEa1StOAzq3kR/VZQCA1swRUJVEK5cU6u1oAABAE6KMBGiFXC5DP+3J08INOfpi/T5tyS2s9vPeHSI0Ms1MnPWID2dhdQAA3Bz+UoWavLkAAADwPpJogI9yuQyVVjpVVOZUcXmlisqc2pdfqq827dPCDfu0L//oQuwOu02DukVrVM94XdAzTh2igr0YOQAAzZinQ2e5d+MAAABNjiQa0AwdqOpAmF9SoZKK6omw4vJKFZc7VVReqeKyqn25s2qrfs3JhAY4dG5qO43qGa/hqbGKDPE/6fUAAEBHmwuQRAMAoNUhiQZ4UUFphTbvK9TmfQXKyCnw7A8WWftgHhLgUEiAnyKC/DSwW1uN6hmnwd3bKsjfYen7AADg8zyVaEznBACgtSGJBjSB0gqntu53J8uOJs32HCk57vU2m5TYJkTRoQEKDTQTYKEBDgVX7UMCj+5D/B1Hrwl0KNjfr9rrID+H7HbWNAMAwBKeSjSSaAAAtDYk0QALVTpd2nmoWJtzCpSxz6ws25RToB0HiuQ6QSPa+IggpcSHKzUuTKnxEUqNC1dSbJiCA6gSAwCg2XFXorkqJMMl2ehaDQBAa0ESDZKkLl26aOrUqZo6daq3Q2lxdhwo0stLt+qn3Xnakluo8krXca+LDPZXany4UuPClRIfrh7x4UqJDWctMgAAWhK7nySbJMOsRvML9HZEp8RzHgAA1iCJBtRTQWmFnluUqTnfbFeF82iZWbC/QylxYUqJC1dqfLhS4syEWbvwQNlsTKsEAKBFs9nMKZ3O8haTRAMAANYgiQbUkdNl6H+rdunxzzfrQGGZJOmclHa6emAnpcVHqGObYNYgAwDAlzkCqpJodOgEAKA1YRGHUzEMqbzIO5txgkW0fuXll19Whw4d5HJVn0Z48cUX67rrrtPWrVt1ySWXKC4uTmFhYTrzzDP15ZdfNsZvy+f9sP2QLnn+G90972cdKCxTt5hQzZk8QP++/kyN7hWvTm1DSKABANBS1Pc5z1UpVZRIpXk85wEA0IpQiXYqFcXSw+29895/2ysFhJ7ysiuuuEK33XabFi1apBEjRkiSDh8+rM8//1wfffSRCgsLNXbsWD300EMKCgrSv//9b40fP14ZGRnq1KlTY38Kn7D7cLFmfrpJn/yULUkKD/TT7Rck69rBXRTgRy4aAIAWiec8AABQByTRfEB0dLQuvPBC/fe///U8XL377ruKjo7WiBEj5HA41KdPH8/1Dz30kN5//33Nnz9ff/rTn7wVdotQXF6plxZv1ctLt6ms0iWbTfrtmZ30l1EpigljDRQAANC4eM4DAKD5IIl2Kv4h5jeF3nrvWrr66qt100036YUXXlBgYKDefPNN/fa3v5XD4VBRUZEefPBBffzxx9q7d68qKytVUlKirKysRgy+ZXO5DH24bo8e/TRDOfmlkqRB3aJ1/7he6tk+wsvRAQAAS9T3Oa80Xzq8XfILltql1P+9a4nnPAAAmgeSaKdis9Wq1N7bxo8fL5fLpU8++URnnnmmli1bplmzZkmS7rrrLn3++ed64oknlJSUpODgYF1++eUqL2cx3ONZu+uIHvxovdZkHZEkdWwTrHvHpunC3vF01wQAwJfU9znPZpcKgyWbo0meE3nOAwCgeSCJ5iOCg4P1m9/8Rm+++aYyMzOVkpKi/v37S5KWLVumyZMn69JLL5UkFRYWaseOHV6Mtnnal1+qRz/bpPdW75EkhQQ4dOvwJN1wdlcF+Tu8HB0AAGg2HP7m3nBKLqdkb9znBJ7zAABoHkii+ZCrr75a48eP1/r163XNNdd4ziclJem9997T+PHjZbPZdN9999Xo8NSalVY49eqybXph8VYVlzslSZed0VF/vTBVcRFBXo4OAAA0OzaHWY1muCRnRaMn0SSe8wAAaA5IovmQ888/X9HR0crIyNDvfvc7z/mnnnpKv//97zVkyBDFxMTo7rvvVn5+vhcjbR4Mw9Cnv+ToX59s1J4jJZKkMzpF6f7xvdQ3Mcq7wQEAgObLZjOr0SrLJFeFpMb/0o3nPAAAvI8kmg9xOBzau7fm4rhdunTR119/Xe3crbfeWu11ayv7X783Tw9+tEE/bD8kSUqIDNI9Y3ro4j7tWfcMAACcmj1AUpnkbJq1x3jOAwDA+0iioVXZX1CmWQsz9PaPu2QYUqCfXX84t7tuPrebQgL4nwMAAKglvwCpXOZ0TgAA0CqQNUCrUFrh1GvfbNcLizJVVLXu2bjTE3TPmB7q2Kb2LeYBAAAkHW0u0ESVaAAAwPtIosGnGYahj37K1qOfbvKse3Z6x0jdN66nzuwS7eXoAABAi2UPMPck0QAAaDVIosFnrc46rH9+vEFrso5IMtc9++uFqbqkTwfZ7ax7BgAAGsBTicZ0TgAAWguSaMdhGIa3Q2jxvPk73H24WI9+lqGP1pmL7wb7O/TH87rrxmHdFBzQ+C3oAQBA82XZM4qj9Vai8awMAGitSKIdw9/f/EaxuLhYwcHBXo6mZSsuLpZ09HfaFApKK/Ti4q169ZvtKq90yWaTLj+jo+4cnaq4iMZvPQ8AAJovy5/z3JVohktyVUr21vNY7Y3nPAAAmoPW81/7WnA4HIqKilJubq4kKSQkRDYb0/7qwjAMFRcXKzc3V1FRUXI4Gr/yy+kyNPfHXZq1MEMHCs1vgwd1i9bfL+qp3h0iG/39AQBA89coz3mVdklOqahQ8vf9L+y88ZwHAEBzQhLtV+Lj4yXJ84CF+omKivL8LhvTsi379a9PNmpTToEkqWtMqKaP6aGRPeNIgAIAgGosf84rOGRO5zwiyb/1zGJoquc8AACaG5Jov2Kz2ZSQkKDY2FhVVLBQbH34+/s3+jeTmbkF+tcnG7UoY78kKTLYX7ePSNY1gzorwM/eqO8NAABaJsuf8z56Wtr5jXTe36SU3zR8vBagKZ7zAABorkiinYDD4eABoRk6VFSu2V9u1pvfZ8npMuRnt2nS4M66fUSyokICvB0eAABoASx7zgsOlgp3SfnbpCDfn84JAEBrRxINLUJZpVNvLN+pZ77eooLSSknSBWlx+tvYHurWLszL0QEAgFYpooO5z9vj3TgAAECTIImGZs0wDH2+PkczP92knQfNTlBpCRG676I0DUmK8XJ0AACgVYvsaO7zd3s3DgAA0CRIoqHZKi6v1B/+b5WWbTkgSWoXHqi7RqXqsv4d5bDTNAAAAHiZO4lGJRoAAK0CSTQ0S5VOl259c7WWbTmgQD+7/nBON/3h3O4KDeSvLAAAaCbc0znz90iGIdEZHAAAn0ZGAs2OYRi678NftChjvwL97PrvjQPVv3O0t8MCAACoLqK9ua8slYoPSaFtvRsPAABoVHZvBwD82rNfZ+qtH3bJbpOeuaofCTQAANA8+QVKobHmcd4u78YCAAAaHUk0NCvvrNylWQs3S5IevLiXRveK93JEAADg15YuXarx48erffv2stls+uCDD056/eLFi2Wz2WpsmzZtqnbdvHnz1LNnTwUGBqpnz556//33G/FTWCTymCmdAADAp5FEQ7OxOCNX09/7WZL0x/O6a9LgLt4NCAAAHFdRUZH69Omj5557rk73ZWRkKDs727MlJyd7frZixQpNnDhRkyZN0rp16zRp0iRdeeWV+v77760O31ruddFoLgAAgM9jTTQ0Cz/vztMtb66W02Xo0n4d9NfRqd4OCQAAnMCYMWM0ZsyYOt8XGxurqKio4/5s9uzZGjlypKZPny5Jmj59upYsWaLZs2frrbfeaki4jcvdoTN/t3fjAAAAjY5KNHjdrkPFuj79RxWXO3V2Uowevex02ehuBQCAz+nXr58SEhI0YsQILVq0qNrPVqxYoVGjRlU7N3r0aC1fvvyE45WVlSk/P7/a1uSoRAMAoNUgiQavOlRUruvm/KADhWVKS4jQi9ecoQA//loCAOBLEhIS9Morr2jevHl67733lJqaqhEjRmjp0qWea3JychQXF1ftvri4OOXk5Jxw3JkzZyoyMtKzJSYmNtpnOCHWRAMAoNVgOie8pqTcqSn//lHbDhSpQ1Sw0q8/U+FB/t4OCwAAWCw1NVWpqUeXahg8eLB27dqlJ554Quecc47n/K8r0Q3DOGl1+vTp0zVt2jTP6/z8/KZPpEVUTeekEg0AAJ9HyQ+8wukydPvba7Q664gigvyUfv2ZiosI8nZYAACgiQwaNEhbtmzxvI6Pj69RdZabm1ujOu1YgYGBioiIqLY1OXclWsFeyeVs+vcHAABNhiQampxhGPrH/PX6YsM+BfjZ9ep1Zyo5LtzbYQEAgCa0Zs0aJSQkeF4PHjxYCxcurHbNF198oSFDhjR1aHUTFi/Z7JKrUirM9XY0AACgETGdE03uxSVb9X/f7ZTNJs2e2FdndY32dkgAAKAOCgsLlZmZ6Xm9fft2rV27VtHR0erUqZOmT5+uPXv26I033pBkdt7s0qWLevXqpfLycv3nP//RvHnzNG/ePM8Yt99+u8455xw9+uijuuSSS/Thhx/qyy+/1DfffNPkn69OHH5SeHuzO2f+Hiki4dT3AACAFokkGprUe6t367HPMiRJf7+op8aexoMmAAAtzcqVKzV8+HDPa/e6ZNddd53S09OVnZ2trKwsz8/Ly8t15513as+ePQoODlavXr30ySefaOzYsZ5rhgwZorffflt///vfdd9996l79+6aO3euBg4c2HQfrL4iO5hJtLzdUscB3o4GAAA0EpthGIa3g2hK+fn5ioyMVF5ennfWzWjFvtlyQJNf/0GVLkNTzu6qv4/r6e2QAACoNZ4hmj+v/Rm9e720/j1p1L+kIX9quvcFAACWqO0zBGuioUls2Juvm/+zSpUuQ+NOT9DfxqZ5OyQAAABruJsL5NOhEwAAX0YSDY1uz5ESXZ/+gwrLKjWwa7SevLKP7PYTt6sHAABoUSI6mvu83d6NAwAANCqSaGhUecUVum7OD9qXX6aUuDC9cu0ABfo5vB0WAACAdahEAwCgVSCJhkZTWuHUjW+sVGZuoeIjgpR+/VmKDPb3dlgAAADWiqhKouWRRAMAwJeRREOjcLkM/eWddfphxyGFB/op/fdnqn1UsLfDAgAAsF5k1XTOwn1SZbl3YwEAAI2GJBoaxUOfbNQnP2fL32HTy9f2V494upgBAAAfFRIjOQIkGVJBtrejAQAAjYQkGiz36rJtmvPtdknSE1f00ZDuMV6OCAAAoBHZ7VJEe/OYddEAAPBZft4OAL7lo3V79dAnGyVJ08f00CV9O3g5IqAFKDkibf5M2vKFFNpOGjpVikjwdlQAgLqI6Cgd3sG6aAAA+DCvV6K98MIL6tq1q4KCgtS/f38tW7bspNc///zzSktLU3BwsFJTU/XGG280UaQ4lRVbD+ov76yTJE0e0kU3ndPNyxEBzVjhfmnl69L//UZ6vLv0/h+kX+ZJ378kPXuGtOhhqazQ21ECAGrLvS5a/m7vxgEAABqNVyvR5s6dq6lTp+qFF17Q0KFD9fLLL2vMmDHasGGDOnXqVOP6F198UdOnT9f/+3//T2eeeaZ++OEH3XjjjWrTpo3Gjx/vhU8At7ySCt3639Uqd7p0Ya943Teup2w2m7fDApqXvN3Sxo+ljfOlrBWS4Tr6s3ZpUuoYacc30u4fpCWPmkm24X+T+k2SHBQOw4e5nFJ50TFboVRRfPS4vEgqLz7m+NfXuK8rltp0lq5+19ufCK1RJB06AQDwdV79V9msWbN0ww03aMqUKZKk2bNn6/PPP9eLL76omTNn1rj+//7v//SHP/xBEydOlCR169ZN3333nR599FGSaF720pKtOlRUru7tQjX7t33lsJNAAyRJB7eaSbONH0l7VlX/WUJfqefFUtrFUkyyec4wzOsXPiAd3i59PFX67kVp5AwpZbREcto3GYZUWVozUVTxq8TS8ZJJ1a6p2pwtoDugYUjOMjPeylILx3VaNxZQFxFVSTTWRAMAwGd5LYlWXl6uVatW6Z577ql2ftSoUVq+fPlx7ykrK1NQUFC1c8HBwfrhhx9UUVEhf3//495TVlbmeZ2fn29B9DjW3iMlmvON2Uhg+pg0Bfk7vBwR4EWGIeVukDZUJc5y1x/zQ5vUaZCZNEsbJ0XVrLiVzSb1vERKGSOtnGNWpB3IkN6aKHUZJo36p9S+X5N9nGqyf5LWvS1t/tTsRNdpoJQ4SEocKIW1805MLZHLKWWvlTK/lrZ+Le37xUyKHVuZ2FrZHFJAmBQQKgWEVO2rXvuHHPMz98+Pee1ftQ+O8vanQGvlns6Zt8u7cQAAgEbjtSTagQMH5HQ6FRcXV+18XFyccnJyjnvP6NGj9eqrr2rChAk644wztGrVKs2ZM0cVFRU6cOCAEhJqLsQ9c+ZMPfjgg43yGWCatXCzyipdOqtrtEakxXo7HOD4DKPxqrgMQ9qzuqribL50aNvRn9kcUtdzpLTxUo9xUnjcicc5ll+ANOhmqc9vpW9mSd+9JO1YJr1ynnTaldKI+46fhLNaQY7087tm8mzfL0fPH9pmTjvVs+br6G5mQs2dWItJMbvVwZS/10yYZX4lbVsslRw68bV+wTUTSO5j/5Dqr4+XTAoIkxz+klpA1aJfYPXP5hdItSVargimcwIA4Ou8vsjOr9fNMgzjhGtp3XfffcrJydGgQYNkGIbi4uI0efJkPfbYY3I4jl/9NH36dE2bNs3zOj8/X4mJidZ9gFZuY3a+5q02F9D929g01kFD06oolYpypUL3ts/cF7mP95v7ov3mlLGAsGOSEyercAmt3TW5G49O1Tx2+o4jUOp+vjlVM+VCKSS6/p8xOMqcynnmFOnrh6Sf5ko/vyNt+NBMsp09zfrKm4oSadMn0rq3zMSPu0LKEWCu23baFWbTg13fSVnfS/s3mkm1Q9ukdf81rw2KMivU3Em1DmdI/sHWxtmcVZRIO5cfTZzt31j954ERZnI1aYTUabD5+3L/vbJTzQu0SO410UoOmdOuA0K8Gw8AALCc15JoMTExcjgcNarOcnNza1SnuQUHB2vOnDl6+eWXtW/fPiUkJOiVV15ReHi4YmJijntPYGCgAgMDLY8fpkc+3STDkC46PUF9E6O8HQ58gWGYVTvu5Jc7MeZOknnO7ZfK8uo2dnmBuTUG/1ApZZRZcZY8SgoMt3b8qE7Sb16RBv1R+uI+syrt26el1f8nnXu3NOD3ZvVafblcZlJs7X/NBF3ZMVPfEweaFXG9LpWC2xw93/cqc19yWNq9Usr6Ttr1vXlcekTa8rm5SZLdT0roU71arbZVeS2BYZhJ1a1fS1u/MhNo1db5spmJxO4jzMRZh/5V1WIAfEZQlPnfgooi879jMUnejggAAFjMa0m0gIAA9e/fXwsXLtSll17qOb9w4UJdcsklJ73X399fHTua6068/fbbGjdunOxMG2py32w5oCWb98vfYdNfR6d6Oxy0ZIX7pW2LzIqdrV+blWS15QiQwuKk0HbmPqxqHxorhcVWnYs1K8mO7eR3vMXYf72Ae8WvuwEe89pZJgVFSqljzTXOug9vmkqr9v2k6z6StnxhJtMOZEif3S398LI04gFzPbW6VIQe3GpWt617Wzqy8+j5yE5m4qzPb6W23U8+RnAbKXmkuUmSs0LK+cmsUnNXqxXmmI0V9qySvnvevK5Nl2OSagOltskNSwQ2teJDVUmzRea+YG/1n4e3l5LONxNn3c5rWEUigObPZjOr0Q5slvJ3k0QDAMAHeXU657Rp0zRp0iQNGDBAgwcP1iuvvKKsrCzdfPPNksypmHv27NEbb7whSdq8ebN++OEHDRw4UIcPH9asWbP0yy+/6N///rc3P0ar5HIZmvmpOT3p6oGd1bltqJcjQotSWW4mV9xT3XJ+qv5zu19VEuzXCbFjEmPuc0GR3llDyVkp2ezeWffLZjM7dXYfIa15Q1o005xK+e51UsezpNH/khLPOvH9JUek9e+bibNd3x09HxAu9bpE6vM7c4phfT+bw9+stOrQXxp8i1mldWRn9aRa7gbp8A5z++ntqs9lNxfmbtPVXGMtumrfpqt5HODF/59xuaTig9LBLUf/3u5dI8k4eo1fsNRlqDmVt/sIqV0q63sBrU1EVRKNddEAAPBJXk2iTZw4UQcPHtSMGTOUnZ2t3r17a8GCBercubMkKTs7W1lZWZ7rnU6nnnzySWVkZMjf31/Dhw/X8uXL1aVLFy99gtZr/rq9Wr83X2GBfvrz+XzTilMwDLPiaetXZvJhxzdmJdix4k83kw9JI8yqJL9mPg3b4fUlJc0YBvzeXKNs+bPmtvsH6bWRZkXaiAeOVpE5K8zkz7q3pE0LzEo6yUxcdT9f6nOVWVXXGGv42Gxm1VmbLlKfiea5kiPmtM9dVYm13avMvxNHssxt+5Ka44TFHU2oeZJrVcm24DZ1T1gZhjkVtcZU4WOnD1cdFx2QDGfNMWJ7VVWbnS91GiL5B9W8BkDr4V4XLZ8kGgAAvshmGIZx6st8R35+viIjI5WXl6eIiAhvh9MilVU6df4TS7TnSInuGp2qW4eTRMNxlByRti81E2dbvzYTI8cKbXe0Yqf7cLOqDA2Tny0t+pe09k2zGYDd30yy2f3MZgRF+49eG9vTTJyddoUUUbOzcZMzDDNpdWi7WVV3eHv145LDJ78/MLIqudb1aHItor1Umnf8ZhOFuebvw1VRhyBtZiKvy9lmsrfb8Obxu0OT4Rmi+fP6n9GimdKSR6QzrpMufqbp3x8AANRLbZ8hmkEpBVqa/1uxU3uOlCguIlC/H9rV2+GguXA5pT2rjy6svntl9codR4DUadDRxFlcb+9MhfRlEQnSJc+ZzQcW3i9lfmmuleYW2s5MmvW5Soo/rXlNNbTZpPB4c+s8uObPSw6bSTVPcu2Y44K9ZpOJ7LXmVlfBbX41ZbhqHb3Q2Orr7IXENI8KRADNF5VoAAD4NP41gDrJK67Qs19nSpL+MjJVwQEOL0eEWisvNqdRuqfxWaX4oLmw+rbFZkfGY7VNNit2up9vVu94c02r1iSul3TNPPPP5bsXpIAws0FA9/NbbkfI4DZShzZmh8tfKy8211z7dRVbQbbZLe9E6+mFxZqJxeY+dRhAyxFpNr5iTTQAAHwTSTTUyQuLM5VXUqGUuDBd1r+jt8NBbRQdlH54xdxKDjXuewVGSt3OPZo4i+rUuO+Hk+s+3Nx8XUCIFJtmbgDgTRFVz0ZUogEA4JNIoqHWdh8u1uvLd0iSpo9Jk8PejKaCoaZD26UVz0tr/iNVlpjnIjoenWpiFb9Ac0H17ueb3RiZ7gYAaK3c/40tyzfXZAyK9G48AADAUvxrF7U264vNKq90aXC3tjovtZ23w8GJ7F0jffu0tOFDc3F5SUroKw29XUq7mCQXAACNJSDUnEZeesSc0kkSDQAAn8K/plErv+zJ0/trzakJ08f2kK05LUgOs7Ph1q/M5Nn2pUfPJ11gJs+6DGtei8gDAOCrIjuaSbT8PVJcT29HAwAALEQSDbXy6GebZBjSxX3a6/SOUd4OB27OCumX96Tlz0j7fjHP2f2k3pdJQ/5sdmAEAABNJ6KD+d/kvN3ejgQAAFiMJBpOaenm/Vq25YD8HTbdNTrV2+FAksoKpdVvmGue5Vc9pPuHSv0nS4P+KEUlejU8AABaLfe6aDQXAADA55BEw0k5XYZmfrpJknTt4C5KjA7xckStXGGu9P3L0o+vmlNFJCm0nTTwZunMG6TgNl4NDwCAVi+iKomWRxINAABfQxINJ/XBmj3amJ2v8CA//Wl4krfDab0OZEornpXWviU5y8xz0d2lobdJp/9W8g/ybnwAAMAU2dHc5zOdEwAAX0MSDSdUWuHUk19kSJJuOS9JbUIDvBxRK7R7pfTtbGnjx5IM81yHAdLZU6XUsZLd4cXgAABADVSiAQDgs0ii4YT+vXyH9uaVKiEySNcP7eLtcFqP8iJp40fSqn9LWcuPnk8ZY1aedRpMp00AAJqrY9dEMwz+mw0AgA8hiYbjOlxUrucWZUqS/jIqVUH+VDw1KpdL2rFMWve2tOFDqaLIPG/3l06faHbajO3h3RgBAMCpuSvRKkul4kNSaFvvxgMAACxDEg3H9fyiTBWUVqpHfLgu7dfB2+FYJ+t76YeXJZtD6n6+uYXHeS+e/Zuln96W1s2tvnZKdDepz1VSv2ukiPbeiw8AANSNX6AUGisV5Zr/bSeJBgCAzyCJhhp2HSrWGyt2SpKmj02Tw+4D0xB2r5QWPSxt/erouZ/fMfdxvc1kWtIIc6qkX2DjxlJ8SPplnrTuLWnPqqPngyKlXr8xk2eJZzH9AwCAliqyg5lEy9sjJfTxdjQAAMAiJNFQwxNfZKjc6dLZSTE6JznG2+E0zJ7V0uKZ0pYvzNd2P6nv76TQdlLmV1L2OmnfL+a2/BnJL1jqcraZUOt+vhSTYk0yq7LcjGHdW9LmzyVXhXne5pCSR0p9fmuueUaXTQAAWr6IDtLeNVIeHToBAPAlJNFQzc+78/Th2r2SpHvG9JCtpVZD7V0rLX5E2vyp+drmkPpeJZ1zl9Smi3luxP1S0QFp22Izobb1a6kwR8pcaG6SFNFRSqqa9tntPCm4Te1jMAxp72pp7Vtm5VnJoaM/iz/dTOb1vlwKa9fwzwsAAJqPyI7mPp8kGgAAvoQkGjwMw9DDCzZKki7t10G9O0R6OaJ6yPnZTJ5t+th8bbObC/Ofc5fUtnvN60NjpNMuNzfDkHI3VCXUvpJ2rjAffle/YW42u9Shf9VaaiPMY8dx/ieUt1v6aa7ZJODA5qPnw+Kl0680q87iejXO5wcAAN7nbi6Qt8e7cQAAAEuRRIPH4s37tWLbQQU47PrLqBRvh1M3+zaY0zY3zq86YZNOu0I6924pJql2Y9hsZnIrrpc09DapvFjaudxMqG39Wtq/Sdr9o7kteVQKjJS6nWMm1Lqcba67tu4taftSSYY5pl+wlDbOTJx1Gy7Z6XIKAIDPi6xKouWTRAMAwJeQRIMkyeky9MiCTZKkyUO7qGObEC9HVEu5m6Qlj0jrP5CZuLJJvX9jJs/apTZs7IAQKfkCc5PMCrOtX1dti6TSI9LGj8zt1zqfbU4fTbtYCopoWBwAAKBliaiazkklGgAAPoUkGiRJ81bvVsa+AkUG++vW82pZueVNB7aY1WA//0+eqq+eE6Tz7pFi0xrnPSM7Smdca24up7lgsHsttd0/mmut9bnKnLLZpnPjxAAAAJo/dyVawV7zmYFKdAAAfAJJNKik3KlZX5hrd/1peJIiQ/y9HNFJHNwqLXlM+vkdyXCZ59LGS+feI8X3bro47A6p4wBzO+9uyVlpnmupjRgAAIB1wuLNtVRdlVJhrhSR4O2IAACABUiiQXO+3a6c/FJ1iArWpMHNtILq0HZp6ePmYv2G0zyXepFZeZZwundjk47fYAAAALRODj8pPMFcEy1/D0k0AAB8BP/yb+UOFZXrpcVbJUl3jk5RkH8zm25waJu0bJa09r9Hk2cpF5rJs/b9vBsbAADAiUR0MBNoebvNynUAANDikURr5Z79eosKyirVMyFCl/Tp4O1wTAcypY0fmgv2711z9HzSBdJ5f5M69vdebAAAALUR2VHa/QMdOgEA8CEk0VqxnQeL9J/vdkqS/jY2TXa7l9bzMgxp3y9m0mzDfGn/xmN+aJO6n29WniWe5Z34AAAA6srdXIAOnQAA+AySaK3Y459nqMJp6JyUdjo7OaZp39zlkvaskjbON5Nnh7cf/ZndT+p6rtkwoMdFUlhs08YGAADQUBEdzX3+bu/GAQAALEMSrZVau+uIPv4pWzabdM+FPZrmTZ2VUtaKqsTZx2bbdze/IKn7CKnnxVLKaCm4TdPEBAAA0Bg8lWgk0QAA8BUk0VqpJ7/IkCT9pl9H9Wwf0XhvVFkmbV8qbfhQylggFR88+rOAMDNhlnaxud5ZYFjjxQEAANCUIpjOCQCAryGJ1grlFpTqm8wDkqSpFyRb/wblxVLml+Y0zc2fSWX5R38W3EZKvcicqtntPMk/yPr3BwAA8LbIqumchfukynLJL8C78QAAgAYjidYKLdywT4Yh9UmMUmJ0iHUDb18q/fCKtOVLqbLk6PmweCltnJk46zxUcvhb954AAADNUUiM5AiQnOVSQbbUprO3IwIAAA1EEq0V+uyXHEnSmN7x1g1aXiz997dSRZH5OqqTOU0z7WKp45mS3W7dewEAADR3drsU0V46vEPK30MSDQAAH0Bmo5U5UlyuFVvNdcku7GVhEm3rV2YCLaKjdNMS6fafpNH/kjoNJIEGAICPWbp0qcaPH6/27dvLZrPpgw8+OOn17733nkaOHKl27dopIiJCgwcP1ueff17tmvT0dNlsthpbaWlpI36SRubu0Mm6aAAA+ASyG63MlxtzVeky1CM+XF1iQq0beONH5r7XBKl9X8lms25sAADQrBQVFalPnz567rnnanX90qVLNXLkSC1YsECrVq3S8OHDNX78eK1Zs6badREREcrOzq62BQW14PVT3R068+nQCQCAL2A6Zyvz2S/ZkqQxvROsG7SyXMr4zDxOG2/duAAAoFkaM2aMxowZU+vrZ8+eXe31ww8/rA8//FAfffSR+vXr5zlvs9kUH29hpby30aETAACfQiVaK1JYVqmlW8yunBdauR7ajqVSWZ4UFid1PMu6cQEAgE9yuVwqKChQdHR0tfOFhYXq3LmzOnbsqHHjxtWoVPu1srIy5efnV9uaFXeHznySaAAA+AKSaK3Iok25Kq90qVtMqFLiwqwb2D2Vs8dFrH8GAABO6cknn1RRUZGuvPJKz7kePXooPT1d8+fP11tvvaWgoCANHTpUW7ZsOeE4M2fOVGRkpGdLTExsivBrz51Ey2M6JwAAvoCMRyvi7sp5Ye942axas8zllDZ9Yh4zlRMAAJzCW2+9pX/84x+aO3euYmNjPecHDRqka665Rn369NGwYcP0zjvvKCUlRc8+++wJx5o+fbry8vI8265du5riI9SeezonlWgAAPgE1kRrJUornFqUkSvJ4qmcu36QivZLQZFSl2HWjQsAAHzO3LlzdcMNN+jdd9/VBRdccNJr7Xa7zjzzzJNWogUGBiowMNDqMK3jbixQfFAqL5YCQrwbDwAAaBAq0VqJpZv3q7jcqQ5RwTqtQ6R1A7uncqaMkRz+1o0LAAB8yltvvaXJkyfrv//9ry666KJTXm8YhtauXauEBAubITW1oCjJv6obev5er4YCAAAajkq0VqJRpnIaxtEkGlM5AQBoNQoLC5WZmel5vX37dq1du1bR0dHq1KmTpk+frj179uiNN96QZCbQrr32Wj399NMaNGiQcnLM55Lg4GBFRppf7j344IMaNGiQkpOTlZ+fr2eeeUZr167V888/3/Qf0Co2m1mNdmCzlL9biknydkQAAKABqERrBcorXVq4cZ8ki6dyZq+T8rIk/xCp+/nWjQsAAJq1lStXql+/furXr58kadq0aerXr5/uv/9+SVJ2draysrI817/88suqrKzUrbfeqoSEBM92++23e645cuSIbrrpJqWlpWnUqFHas2ePli5dqrPOauGdv93rouWxLhoAAC0dlWitwPKtB1RQWql24YHq36mNdQO7q9CSLmCNDwAAWpHzzjtPhmGc8Ofp6enVXi9evPiUYz711FN66qmnGhhZMxRJcwEAAHwFlWitwOfrzSkTo3vFyW63aCqnJG362NynXWzdmAAAAL4koqO5z9vt3TgAAECDkUTzcU6XoS/WV03l7GXhwrz7N0v7N0l2fylllHXjAgAA+BIq0QAA8Bkk0XzcD9sP6WBRuaJC/DWwW7R1A2+qmsrZ7VwpyMJunwAAAL6ENdEAAPAZJNF8nHsq58i0OPk7LPzjpisnAADAqUVWTeekEg0AgBaPJJoPc7kMffaLmUSztCvnkV3S3jWSbFLqRdaNCwAA4GvclWhl+VJpvndjAQAADUISzYet3X1EOfmlCgv009nJMdYNvOkTc995iBTWzrpxAQAAfE1gmBQUZR5TjQYAQItGEs2HfV5VhXZ+j1gF+jmsG9g9lbPHOOvGBAAA8FXuKZ2siwYAQItGEs1HGYahTxtjKmfhfilruXmcRhINAADglDzNBXZ5Nw4AANAgJNF81IbsfGUdKlaQv13npVo45TJjgWS4pIS+UlQn68YFAADwVZFVSTSmcwIA0KKRRPNR7oYC56a0U0iAn3UD05UTAACgbjyVaCTRAABoybyeRHvhhRfUtWtXBQUFqX///lq2bNlJr3/zzTfVp08fhYSEKCEhQddff70OHjzYRNG2HO4k2pjeCdYNWponbV9iHqddbN24AAAAvsy9Jlr+bu/GAQAAGsSrSbS5c+dq6tSpuvfee7VmzRoNGzZMY8aMUVZW1nGv/+abb3Tttdfqhhtu0Pr16/Xuu+/qxx9/1JQpU5o48uYtM7dQW3IL5e+waXiPWOsG3rJQcpZLMSlSuxTrxgUAAPBlVKIBAOATvJpEmzVrlm644QZNmTJFaWlpmj17thITE/Xiiy8e9/rvvvtOXbp00W233aauXbvq7LPP1h/+8AetXLmyiSNv3j77JVuSNDQpRpHB/tYNvHG+uWcqJwAAQO0duyaaYXg3FgAAUG9eS6KVl5dr1apVGjVqVLXzo0aN0vLly497z5AhQ7R7924tWLBAhmFo3759+t///qeLLrrohO9TVlam/Pz8apuv+2y9eyqnhV05K0rMSjSJJBoAAEBduCvRKkul4kPejQUAANSb15JoBw4ckNPpVFxcXLXzcXFxysnJOe49Q4YM0ZtvvqmJEycqICBA8fHxioqK0rPPPnvC95k5c6YiIyM9W2JioqWfo7nZdahYv+zJl90mXZAWd+obamvr11JFsRSZaHbmBAAAQO34BUqhVd3SWRcNAIAWy+uNBWw2W7XXhmHUOOe2YcMG3Xbbbbr//vu1atUqffbZZ9q+fbtuvvnmE44/ffp05eXlebZdu3ZZGn9z424oMLBrW7UNC7Ru4I0fm/u08dIJ/nwAAAB82fq9eXpjxQ6t2nm47jezLhoAAC2en7feOCYmRg6Ho0bVWW5ubo3qNLeZM2dq6NChuuuuuyRJp59+ukJDQzVs2DA99NBDSkio2YkyMDBQgYEWJpOaOc9UztMsnMrprJAyFpjHPcZZNy4AAEAL8s6Pu/TvFTt10znd1L9zm7rdHNlRyl5rrosGAABaJK9VogUEBKh///5auHBhtfMLFy7UkCFDjntPcXGx7PbqITscDklmBVtrty+/1PPN6KieFibRdnwjlR6RQmKkToOsGxcAAKAFSYkPlyRl5BTU/ebIjuY+j+mcAAC0VF6dzjlt2jS9+uqrmjNnjjZu3Kg77rhDWVlZnumZ06dP17XXXuu5fvz48Xrvvff04osvatu2bfr2229122236ayzzlL79u299TGajc+rqtDO6BSl+Mgg6wbe+JG573GRZHdYNy4AAEALkhpnJtG27KtHEs0znZMkGgAALZXXpnNK0sSJE3Xw4EHNmDFD2dnZ6t27txYsWKDOnTtLkrKzs5WVleW5fvLkySooKNBzzz2nv/zlL4qKitL555+vRx991FsfoVlxr4c2pnfNaa315nJJmz4xj9Mutm5cAACAFia5Kom2N69U+aUVigjyr/3NkVVJNKZzAgDQYnk1iSZJt9xyi2655Zbj/iw9Pb3GuT//+c/685//3MhRtTyHisr1/XazZfqFvS2cyrlnpVSYIwVGSF3PsW5cAACAFiYy2F/xEUHKyS/Vln2FdVsXLcI9nZMkGgAALZXXu3PCGgs35MjpMtSrfYQSo0OsG3jjfHOfMlryC7BuXAAAgBYoOS5MkrS5rlM63ZVoBXsll9PiqAAAQFMgieYjPvVM5bSwCs0wjq6HljbeunEBAABaKPe6aHVuLhAWL9nskqtSKsxthMgAAEBjI4nmA/JLK/Rt5gFJFk/l3PeLdHiH5BckJV1g3bgAAAAtlLtD55bcOibRHH5SeNW6tayLBgBAi0QSzQd8vTFXFU5DSbFhSooNt25gdxVa0gVSQKh14wIAALRQKZ5KtMK630yHTgAAWjSSaD7g01+yJVk8lVOSNn5s7pnKCQAAIElKjjXXRDtQWKZDReV1u5kOnQAAtGgk0Vq44vJKLdm8X5I0upeFSbSDW6Xc9ZLdz2wqAAAAAIUG+qljm2BJ9Wgu4KlEI4kGAEBLRBKthVuSsV+lFS4lRgerV/sI6wZ2T+XsMkwKrkP7dgAAAB/nbi5Q9w6dieY+n+mcAAC0RCTRWrijXTkTZLPZrBuYrpwAAADH5W4uUPckGpVoAAC0ZCTRWrCySqe+3mS2SLd0Kmf+XmnPSkk2qcdF1o0LAADgA1LizHXRNte1uUAEa6IBANCSkURrwb7NPKDCskrFRQSqX2KUdQNv+sTcJw6Uwi1uVgAAANDCuTt0bs4tkGEYtb8xsqO5L8iRKuvYlAAAAHgdSbQW7NOfzamcF/aKl91u5VTO+eaeqZwAAAA1dG8XJrtNOlJcof0FZbW/MSRGcgRIMqSC7EaLDwAANA6SaC1UpdOlhRv3SZJG97awWqzooLTjW/M4bZx14wIAAPiIIH+HurQNlSRl1GVdNLtdimhvHjOlEwCAFockWgv1/fZDOlJcoejQAJ3VJdq6gTd/KhlOKf40qU0X68YFAADwIZ4pnfvqui5a1ZROmgsAANDikERroT79xZwCMKpnnPwcFv4xerpyXmzdmAAAAD7maHOBenbozN9tcUQAAKCxkURrgVwuQ5+vN6dyXmjlVM6yAmnrIvOY9dAAAABOKCX+aHOBOnF36KQSDQCAFockWgu0Ouuw9heUKTzIT0O6x1g38JaFkrNMapskteth3bgAAAA+xjOdM6euHTrdlWgk0QAAaGlIorVAn/5iduW8IC1OAX6NMJWzxzjJZmG3TwAAAB/TpW2o/B02FZU7tedISe1v9KyJxnROAABaGpJoLYxhGPqsKolm6VTOilJpyxfmMeuhAQAAnFSAn13dYsx10bbUpbkAlWgAALRYJNFamF/25GvPkRIF+zt0TnI76wbetlgqLzTX6Wjfz7pxAQAAfFRyVXOBjH11WBctsqoSrfigVFGHCjYAAOB1JNFaGHdXzuE92ik4wGHdwMdO5bTz1wIAAOBUUt3rotUliRYUJfmHmsf5e60PCgAANBqyJS1I9amcCdYN7KyUMhaYx3TlBAAAqJXk+iTRbLajUzrzdjVCVAAAoLGQRGtBtuQWatuBIgU47BqeauFUzqzlUskhKTha6jTYunEBAAB8WGq8mUTLzC2U01WHDp0R7iQa66IBANCSkERrQT792axCG5Yco/Agf+sG9kzlHCs5/KwbFwAAwId1ig5RoJ9dpRUu7TpUXPsbaS4AAECLRBKtBXGvh2ZpV06XS9r4sXlMV04AAIBac9htSoqtR3OBiKrmAnm7GyEqAADQWEiitRA7DhRpU06BHHabLkiLs27gvaulgr1SQLjU9VzrxgUAAGgF3M0FttSpQyeVaAAAtEQk0VqIz9abUzkHd2urNqEB1g3snsqZMkryD7JuXAAAgFbA3VwgY19h7W9iTTQAAFokkmgtxKeerpwWTuU0DGnjfPO4xzjrxgUAAGglUuPN6Zx1q0Srms5JJRoAAC0KSbQWYO+REq3bdUQ2mzSql4VTOXM3Soe2SY5AKXmkdeMCAAC0EilVlWhb9xeqwumq3U3uSrSyfKk0v5EiAwAAViOJ1gJ8XjWVc0DnNooNt3DKpXsqZ/fzpcBw68YFAABoJTpEBSs0wKEKp6EdB4pqd1NgmBQUaR5TjQYAQItBEq0F+HpTriRpdC8Lp3JKR5NoaeOtHRcAAKCVsNlsnnXRNtdlXbTIRHPPumgAALQYJNFagC1VD2T9OrWxbtBD26V9P0s2h5Q6xrpxAQAAWpmUOHNdtIy6rIvmntKZv7sRIgIAAI2BJFozV1hWqZz8UklSUrsw6wbe9LG57zJUCom2blwAAODzli5dqvHjx6t9+/ay2Wz64IMPTnnPkiVL1L9/fwUFBalbt2566aWXalwzb9489ezZU4GBgerZs6fef//9Rojeeu510erWXIAOnQAAtDQk0Zq5rblmFVpMWKAiQ/ytG9gzlfNi68YEAACtQlFRkfr06aPnnnuuVtdv375dY8eO1bBhw7RmzRr97W9/02233aZ58+Z5rlmxYoUmTpyoSZMmad26dZo0aZKuvPJKff/99431MSyTGm8m0epViZZHJRoAAC2Fn7cDwMlt3W8m0bq3C7Vu0KID0q4fzOPUsdaNCwAAWoUxY8ZozJjaLwfx0ksvqVOnTpo9e7YkKS0tTStXrtQTTzyhyy67TJI0e/ZsjRw5UtOnT5ckTZ8+XUuWLNHs2bP11ltvWf4ZrOSuRNtxoEilFU4F+TtOfVNkR3PPdE4AAFoMKtGaucyqSrSkWAuncm5ZKMmQ4k8/OpUAAACgkaxYsUKjRo2qdm706NFauXKlKioqTnrN8uXLTzhuWVmZ8vPzq23eEBseqMhgf7kMadv+WnbojGA6JwAALQ1JtGbuaCWahUm0zZ+Z+5QLrRsTAADgBHJychQXF1ftXFxcnCorK3XgwIGTXpOTk3PCcWfOnKnIyEjPlpiYaH3wtWCz2TzNBTbXdkqn+4vM/D2SYTRSZAAAwEok0Zq5rVXfZlpWiVZZLm392jwmiQYAAJqIzWar9tqoShwde/541/z63LGmT5+uvLw8z7Zr1y4LI64b95TOWifR3JVolaVS8aFGigoAAFiJNdGasQqnSzsOmEm07lYl0bKWS2X5Umis1L6fNWMCAACcRHx8fI2KstzcXPn5+alt27YnvebX1WnHCgwMVGBgoPUB14O7uUCtk2h+gVJoO6lov7kuWmjbRowOAABYgUq0ZizrULEqXYaC/R1KiAiyZtDNn5v7lFGSnT9+AADQ+AYPHqyFCxdWO/fFF19owIAB8vf3P+k1Q4YMabI4GyI5tiEdOlkXDQCAloAsSjO2taqpQPfYUNntJ57KUGuGIWV8ah4zlRMAANRTYWGh1q5dq7Vr10qStm/frrVr1yorK0uSOc3y2muv9Vx/8803a+fOnZo2bZo2btyoOXPm6LXXXtOdd97pueb222/XF198oUcffVSbNm3So48+qi+//FJTp05tyo9Wb+410XYdKlFxeWXtbvJ06CSJBgBAS0ASrRnLtLqpwMFM6fB2yREgdTvPmjEBAECrs3LlSvXr10/9+plLQ0ybNk39+vXT/fffL0nKzs72JNQkqWvXrlqwYIEWL16svn376p///KeeeeYZXXbZZZ5rhgwZorfffluvv/66Tj/9dKWnp2vu3LkaOHBg0364emobFqiYsABJ0pZ9hbW7yVOJtruRogIAAFZiTbRmbGtu1XpoViXR3F05u5wtBYZbMyYAAGh1zjvvPE9jgONJT0+vce7cc8/V6tWrTzru5Zdfrssvv7yh4XlNSly4DhQe1OZ9BeqTGHXqG6hEAwCgRaESrRlzV6JZ1pkzoyqJxlROAAAAy9W5Q2cka6IBANCSkERrpgzD0LZcC6dzlhyWslaYxymjGz4eAAAAqnEn0TJqPZ2zqhKN6ZwAALQIJNGaqf0FZSooq5TdJnWJCWn4gJlfSYZTapcmtenS8PEAAABQTWq8+cXnlrpWohXslVzORooKAABYhSRaM5VZVYXWKTpEgX6Ohg+4+XNzTxUaAABAo0iKNSvRsvNKlVdSceobwuIlm11yVUqFuY0cHQAAaKh6JdEWL15scRj4ta1WduZ0VkpbvjCPWQ8NAACgUUQG+yshMkiSlJlbi2o0h58UnmAe01wAAIBmr15JtAsvvFDdu3fXQw89pF27dlkdEyRt3W925rSkqcDuH6TSI1JwG6njmQ0fDwAAAMflWRctp7brormbC7AuGgAAzV29kmh79+7V7bffrvfee09du3bV6NGj9c4776i8vNzq+FqtTCubCmyu6sqZPMr8xhMAAACNIiXOfHardYfOqERzvz+jkSICAABWqVcSLTo6WrfddptWr16tlStXKjU1VbfeeqsSEhJ02223ad26dVbH2ep4pnPGhjZ8MNZDAwAAaBLuSrRaJ9G6nWfuN33UOAEBAADLNLixQN++fXXPPffo1ltvVVFRkebMmaP+/ftr2LBhWr9+vRUxtjqFZZXKziuVZEEl2qHt0v5Nks0hdR9hQXQAAAA4kTon0VIvMp/Tcn6WDm5txMgAAEBD1TuJVlFRof/9738aO3asOnfurM8//1zPPfec9u3bp+3btysxMVFXXHGFlbG2GtuqqtBiwgIUFRLQsMHcVWidh0jBUQ0bCwAAACeVXDWd80BhuQ4Wlp36htC2UtdzzOMNHzZiZAAAoKHqlUT785//rISEBN18881KSUnRmjVrtGLFCk2ZMkWhoaFKTEzUI488ok2bNlkdb6vgnsrZzcr10JjKCQAA0OhCAvzUKTpEkrR5Xy2bC/S8xNyTRAMAoFmrVxJtw4YNevbZZ7V3717Nnj1bvXv3rnFN+/bttWjRogYH2Bq5mwo0uDNnWYG04xvzOGVMA6MCAABAbdS5uUDaeMlml7LXmktxAACAZqleSbSvvvpKV111lQICTjzV0M/PT+eee+4px3rhhRfUtWtXBQUFqX///lq2bNkJr508ebJsNluNrVevXvX5GM3W1twiSRash7Z1keSqkKK7SzFJFkQGAACAU6nzumihMVKXs83jjfMbKSoAANBQ9UqizZw5U3PmzKlxfs6cOXr00UdrPc7cuXM1depU3XvvvVqzZo2GDRumMWPGKCsr67jXP/3008rOzvZsu3btUnR0tM+tveaeztngSjRPV84LGxgRAAAAaqvOSTRJ6jnB3K//wPJ4AACANeqVRHv55ZfVo0ePGud79eqll156qdbjzJo1SzfccIOmTJmitLQ0zZ49W4mJiXrxxRePe31kZKTi4+M928qVK3X48GFdf/319fkYzVKl06UdB92VaKH1H8jlkra4k2ishwYAANBUjibRCmUYRu1uck/p3LtaOnL8L5QBAIB31SuJlpOTo4SEhBrn27Vrp+zs7FqNUV5erlWrVmnUqFHVzo8aNUrLly+v1RivvfaaLrjgAnXu3PmE15SVlSk/P7/a1pxlHSpWhdNQsL9D7SOD6z/Q3tVS0X4pMELqNNi6AAEAAHBS3dqFymG3Ka+kQrkFtejQKUlhsVLnoeYxDQYAAGiW6pVES0xM1Lffflvj/Lfffqv27dvXaowDBw7I6XQqLi6u2vm4uDjl5OSc8v7s7Gx9+umnmjJlykmvmzlzpiIjIz1bYmJireLzlq37zSq0bu1CZbfb6j+Quytn0gjJ78Rr1wEAAMBaQf4OdW7r7tBZlymddOkEAKA5q1cSbcqUKZo6dapef/117dy5Uzt37tScOXN0xx136MYbb6zTWDZb9USRYRg1zh1Penq6oqKiNGHChJNeN336dOXl5Xm2Xbt21Sm+pubuzNngpgLuJBrroQEAADS51KopnRk5dUiipY2XZJN2/yjl7W6cwAAAQL351eemv/71rzp06JBuueUWlZeXS5KCgoJ09913a/r06bUaIyYmRg6Ho0bVWW5ubo3qtF8zDENz5szRpEmTTtohVJICAwMVGBhYq5iaA3dTgQYl0fJ2Szk/S7JJSSOtCQwAAAC1lhwXrk9/yalbJVp4vLkMR9ZyacN8afAtjRcgAACos3pVotlsNj366KPav3+/vvvuO61bt06HDh3S/fffX+sxAgIC1L9/fy1cuLDa+YULF2rIkCEnvXfJkiXKzMzUDTfcUJ/wmzV3JVqDOnO6u3ImniWFtrUgKgAAANRF6jHNBeqk1wRzv+EDS+MBAAANV68kmltYWJjOPPNM9e7du17VXtOmTdOrr76qOXPmaOPGjbrjjjuUlZWlm2++WZI5FfPaa6+tcd9rr72mgQMHqnfv3g0Jv9kxDONoJVpsAzpzbqYrJwAAgDelxptfiG7ZV1D7Dp1S1ZROSbu+l/L3NkJkAACgvuo1nVOSfvzxR7377rvKysryTOl0e++992o1xsSJE3Xw4EHNmDFD2dnZ6t27txYsWODptpmdna2srOotvvPy8jRv3jw9/fTT9Q292dpfWKaC0krZbVKXtvVMopUXS9uXmMcpY6wLDgAAALXWuW2o/B02FZU7tedIiTq2CandjRHtpcRB0q7vzCmdg25u3EABAECt1asS7e2339bQoUO1YcMGvf/++6qoqNCGDRv09ddfKzIysk5j3XLLLdqxY4fKysq0atUqnXPOOZ6fpaena/HixdWuj4yMVHFxcZ0bGLQE7qmcidEhCvJ31G+Q7UulylIpspMUm2ZhdAAAAKgtf4fds8ZtndZFk+jSCQBAM1WvJNrDDz+sp556Sh9//LECAgL09NNPa+PGjbryyivVqVMnq2NsNbbuL5LUwKYCmz819ymjpVp0OQUAAEDjSPZ06Kzjumg9Lzb3WSukgpyTXwsAAJpMvZJoW7du1UUXXSTJ7H5ZVFQkm82mO+64Q6+88oqlAbYmWxvaVMAwjlkP7UKLogIAAL7i3//+tz755BPP67/+9a+KiorSkCFDtHPnTi9G5ptS446ui1YnkR2ljmdKMqSNH1kfGAAAqJd6JdGio6NVUGA+DHTo0EG//PKLJOnIkSMqLi62LrpWxtNUoF0910PL+UkqyJb8Q6UuZ1sYGQAA8AUPP/ywgoODJUkrVqzQc889p8cee0wxMTG64447vByd70lxV6LVNYkmST0nmPv1H1gWDwAAaJh6JdGGDRumhQsXSpKuvPJK3X777brxxht11VVXacSIEZYG2Jq4K9HqPZ3TXYXWfbjkH2RRVAAAwFfs2rVLSUlJkqQPPvhAl19+uW666SbNnDlTy5Yt83J0vsedRMvMLZTTVYcOndLRKZ07v5UKcy2ODAAA1Ee9kmjPPfecfvvb30qSpk+frjvvvFP79u3Tb37zG7322muWBthaFJVVam9eqaSGJNE+M/cpoy2KCgAA+JKwsDAdPHhQkvTFF1/oggsukCQFBQWppKTEm6H5JLNZlF1llS5lHarjbI2oTlKH/jKndM5vlPgAAEDd1DmJVllZqY8++kh2u3mr3W7XX//6V82fP1+zZs1SmzZtLA+yNdhW1VSgbWiA2oQG1H2Agn3SnlXmcfIoCyMDAAC+YuTIkZoyZYqmTJmizZs3e9a4Xb9+vbp06eLd4HyQw27zrHWbkVOfKZ106QQAoDmpcxLNz89Pf/zjH1VWVtYY8bRanvXQ6ttUYMsX5r59Pyk83qKoAACAL3n++ec1ePBg7d+/X/PmzVPbtm0lSatWrdJVV13l5eh8k3tKZ52bC0hHk2g7vpEK91sYFQAAqA+/+tw0cOBArVmzRp07d7Y6nlYrs8HrobmnctKVEwAAHF9UVJSee+65GucffPBBL0TTOqQ2pLlAmy5SQl8pe6206WNpwPVWhgYAAOqoXkm0W265RX/5y1+0e/du9e/fX6Gh1btJnn766ZYE15o0qDNnZZm0dZF5TBINAACcwGeffaawsDCdfbbZxfv555/X//t//089e/bU888/z7IcjeBoJVph/QboeYmZRNvwAUk0AAC8rF6NBSZOnKjt27frtttu09ChQ9W3b1/169fPs0fduZNoSfWZzrljmVRRJIUnSAl9LI4MAAD4irvuukv5+fmSpJ9//ll/+ctfNHbsWG3btk3Tpk3zcnS+KSXeTKJt3V+o8kpX3QdwT+ncvkwqOmhhZAAAoK7qVYm2fft2q+No1SqdLm0/YDYWqNd0zs2fm/vkUZLNZmFkAADAl2zfvl09e/aUJM2bN0/jxo3Tww8/rNWrV2vs2LFejs43tY8MUlignwrLKrXjYJGnMq3W2naX4k+Tcn42p3T2v65xAgUAAKdUryQaa6FZa9fhElU4DQX529UhKrhuNxsG66EBAIBaCQgIUHFxsSTpyy+/1LXXXitJio6O9lSowVo2m03JcWFak3VEm/cV1D2JJkk9J5hJtA0fkkQDAMCL6pVEe+ONN076c/cDGWpna1VTgW4xYbLb61hJtn+TdCRLcgRK3c5thOgAAICvOPvsszVt2jQNHTpUP/zwg+bOnStJ2rx5szp27Ojl6HxXaly4mUTLKZDqs3RwzwnS1/+Uti+Rig9JIdFWhwgAAGqhXkm022+/vdrriooKFRcXKyAgQCEhISTR6ijT3VSgPuuhuavQup0rBdSjKQEAAGg1nnvuOd1yyy363//+pxdffFEdOnSQJH366ae68EIq2htLclX12eb6NheISZLiekv7fpEyFkj9rrEwOgAAUFv1SqIdPny4xrktW7boj3/8o+66664GB9XauCvR6tWZM8M9lXO0hREBAABf1KlTJ3388cc1zj/11FNeiKb1SPUk0QrqP0jPS8wk2voPSKIBAOAl9UqiHU9ycrIeeeQRXXPNNdq0aZNVw7YKmfXtzFl0UNr9g3mcTBINAACcmtPp1AcffKCNGzfKZrMpLS1Nl1xyiRwOh7dD81kpceYz3o6DRSqtcCrIvx6/654TpEX/krYtlkoOS8FtLI0RAACcmmVJNElyOBzau3evlUP6PMMwjqlEq2MSLfNLyXCZ5f1RiY0QHQAA8CWZmZkaO3as9uzZo9TUVBmGoc2bNysxMVGffPKJunfv7u0QfVK78EBFhfjrSHGFtu4vVK/2kfUYJEVqlybt3yhlfCr1/Z31gQIAgJOqVxJt/vz51V4bhqHs7Gw999xzGjp0qCWBtRYHCsuVX1opm03qGlPH6ZybmcoJAABq77bbblP37t313XffKTraXJz+4MGDuuaaa3Tbbbfpk08+8XKEvslmsyklLlw/bD+kzfsK6pdEk6ReE6TFG80unSTRAABocvVKok2YMKHaa5vNpnbt2un888/Xk08+aUVcrUZmVRVaYpuQupX2OyukzK/M45QxjRAZAADwNUuWLKmWQJOktm3b6pFHHuGL0EaWEhdWlUSrZ3MByVwXbfFMaevXUmmeFFTPZBwAAKiXeiXRXC6X1XG0Wlv317OpQNYKqSxPComROpzRCJEBAABfExgYqIKCmovbFxYWKiAgwAsRtR6e5gI5DWguEJsmxaRKBzLM5lJ9JloUHQAAqA27twNo7bbWt6nA5s/NffIoyc5CwAAA4NTGjRunm266Sd9//70Mw5BhGPruu+9088036+KLL/Z2eD4tuSqJltGQDp2SWY0mSRs+aNg4AACgzuqVRLv88sv1yCOP1Dj/+OOP64orrmhwUK1JZn2bCrAeGgAAqKNnnnlG3bt31+DBgxUUFKSgoCANGTJESUlJmj17trfD82kpVUm03YdLVFRWWf+Bek0w95lfSaX5DQ8MAADUWr2SaEuWLNFFF11U4/yFF16opUuXNjio1mTb/iJJUve6VKIdyJQOZkp2f6n7+Y0UGQAA8DVRUVH68MMPtXnzZv3vf//Tu+++q82bN+v9999XVFSUt8PzadGhAWoXHihJ2pLbgHXRYntKbZMkZ9nRmQkAAKBJ1GtNtBOtm+Hv76/8fL4Rq62iskrtOVIiSUqqSyXalqoHpi5DpaCIRogMAAD4imnTpp3054sXL/Ycz5o1q5Gjad1S4sK0v6BMm/cVqG9iVP0GsdmknhOkZU+YUzpPZxYIAABNpV5JtN69e2vu3Lm6//77q51/++231bNnT0sCaw22HzCr0KJDA9QmtA6L+WZ8au5TLmyEqAAAgC9Zs2ZNra6z2WyNHAlS4sL1bebBhjUXkMx10ZY9IWV+KZUVSoF1XBYEAADUS72SaPfdd58uu+wybd26Veefb04n/Oqrr/TWW2/p3XfftTRAX+ZpKlCXKrSSI2ZnTon10AAAwCktWrTI2yGgSopVzQXiT5Oiu0mHtpkzFHpfZkF0AADgVOq1JtrFF1+sDz74QJmZmbrlllv0l7/8Rbt379aXX36pCRMmWByi7/I0FYgNrf1NW7+WXJVSTIr58AQAAIAWwZ1E27KvAWuiSVVTOqu6dK7/oGFjAQCAWqtXJZokXXTRRcdtLoDac1ei1akzp3sBWaZyAgAAtCgpceYzX05+qfJKKhQZ7F//wXpOkL55StqyUCovkgLq8KUsAACol3pVov3444/6/vvva5z//vvvtXLlygYH1Vpsza1jZ06XU9ryhXlMEg0AAHjZCy+8oK5duyooKEj9+/fXsmXLTnjt5MmTZbPZamy9evXyXJOenn7ca0pLS5vi4zS68CB/tY8MkiRtaeiUzoQ+UlRnqbLk6PMhAABoVPVKot16663atWtXjfN79uzRrbfe2uCgWoNKp8vTWKDWa6Lt/lEqOSQFRUqJAxsxOgAAgJObO3eupk6dqnvvvVdr1qzRsGHDNGbMGGVlZR33+qefflrZ2dmebdeuXYqOjtYVV1TvLhkREVHtuuzsbAUFBTXFR2oSKfEWrYtms0m9JpjHGz5s2FgAAKBW6pVE27Bhg84444wa5/v166cNGzY0OKjWYPfhEpU7XQr0s6t9VHDtbtr8mblPGik56j0TFwAAoMFmzZqlG264QVOmTFFaWppmz56txMREvfjii8e9PjIyUvHx8Z5t5cqVOnz4sK6//vpq19lstmrXxcfHN8XHaTKWrYsmHV0XbfPnUnlxw8cDAAAnVa8kWmBgoPbt21fjfHZ2tvz8SO7Uhns9tG7twuSw17KlPOuhAQCAZqC8vFyrVq3SqFGjqp0fNWqUli9fXqsxXnvtNV1wwQXq3LlztfOFhYXq3LmzOnbsqHHjxmnNmjUnHKOsrEz5+fnVtubO06Ezp4GVaJLU/gwpspNUUSxlftnw8QAAwEnVK4k2cuRITZ8+XXl5eZ5zR44c0d/+9jeNHDnSsuB8maczZ7taLgJ7eKeUu0GyOaSkEY0YGQAAwMkdOHBATqdTcXFx1c7HxcUpJyfnlPdnZ2fr008/1ZQpU6qd79Gjh9LT0zV//ny99dZbCgoK0tChQ7Vly5bjjjNz5kxFRkZ6tsTExPp/qCaSWpVE29zQ6ZxSVZfOi83jDR80fDwAAHBS9UqiPfnkk9q1a5c6d+6s4cOHa/jw4eratatycnL05JNPWh2jT3JXoiXVtqmAuwqt0yApJLqRogIAAKg9m616Nb1hGDXOHU96erqioqI0YcKEaucHDRqka665Rn369NGwYcP0zjvvKCUlRc8+++xxx3F/qevejrdmb3OTFBsmm006WFSuA4VlDR+w16XmfvPnUkVJw8cDAAAnVK+5lx06dNBPP/2kN998U+vWrVNwcLCuv/56XXXVVfL3b0Cr7lbkaCVabZNoVeuhpYxupIgAAABqJyYmRg6Ho0bVWW5ubo3qtF8zDENz5szRpEmTFBAQcNJr7Xa7zjzzzBNWogUGBiowMLBuwXtZcIBDnaJDtPNgsTbvK1BMWAPj79Bfiugo5e+WMr+S0sZZEygAAKihXpVokhQaGqqzzz5b48eP1znnnKOoqCh9+umnmj9/vpXx+STDMLR1v9mZs1ZJtLJCaUdVy3jWQwMAAF4WEBCg/v37a+HChdXOL1y4UEOGDDnpvUuWLFFmZqZuuOGGU76PYRhau3atEhISGhRvc5Mca2FzAZvtaIMBunQCANCo6lWJtm3bNl166aX6+eefZbPZapTuO51OywL0RQeLypVXUiGbTepWmzXRti2WnOVSmy5STEpjhwcAAHBK06ZN06RJkzRgwAANHjxYr7zyirKysnTzzTdLMqda7tmzR2+88Ua1+1577TUNHDhQvXv3rjHmgw8+qEGDBik5OVn5+fl65plntHbtWj3//PNN8pmaSmp8mL7cuE8ZVqyLJplJtO+elzI+lSrLJL+WVZ0HAEBLUa9KtNtvv11du3bVvn37FBISol9++UVLlizRgAEDtHjxYotD9D3uqZwd2wQryN9x6hs8UznHmN82AgAAeNnEiRM1e/ZszZgxQ3379tXSpUu1YMECT7fN7OxsZWVlVbsnLy9P8+bNO2EV2pEjR3TTTTcpLS1No0aN0p49e7R06VKdddZZjf55mpK7Q+dmKzp0SlLHM6Xw9lJ5gbT1a2vGBAAANdSrEm3FihX6+uuv1a5dO9ntdjkcDp199tmaOXOmbrvttpO2IsfRpgK1Xg9t3y/mvsvQRooIAACg7m655Rbdcsstx/1Zenp6jXORkZEqLi4+4XhPPfWUnnrqKavCa7ZSjunQWdtmDCdlt5tdOr9/yZzSmTrGgigBAMCv1asSzel0KizMTADFxMRo7969kqTOnTsrIyPDuuh81NZccz20pNom0fL2mPuoTo0UEQAAAJpKt3ahcthtyi+t1L58Czp0SlLPCeZ+0wJzSicAALBcvZJovXv31k8//SRJGjhwoB577DF9++23mjFjhrp162ZpgL4o012JFluLJFplmVSUax5HdGzEqAAAANAUAv0c6tI2RJJZjWaJxIFSWLxUlidtW2LNmAAAoJp6JdH+/ve/y+VySZIeeugh7dy5U8OGDdOCBQv0zDPPWBqgL9qaW4fpnPlVVWh+wVJIdCNGBQAAgKaSGn90Sqcl3FM6JWnDB9aMCQAAqqnXmmijR4/2HHfr1k0bNmzQoUOH1KZNm4av6eDjissrtedIiSQpqTaVaHm7zX1kB5oKAAAA+IiUuHAt+DlHGVY1F5DMLp0/vCJt+liqnC35BVg3NgAAqF8l2vFER0eTQKuFbfvN9dDahPgrOrQWDzbu9dAimcoJAADgKzzNBapmKFii02ApNFYqzZO2L7VuXAAAIMnCJBpqx92Zs1ZVaJKUX1WJxnpoAAAAPsOdRNuyr0Aul2HNoHaHlDbePGZKJwAAliOJ1sTqtB6adMx0TpJoAAAAvqJL2xAFOOwqLnd6lvqwRK8J5n7Tx5KzwrpxAQAASbSmtrVqOmftk2ju6ZwdGikiAAAANDU/h13d2oVKsrC5gCR1GiKFxEglh6Udy6wbFwAAkERranWezumuRIsgiQYAAOBL3B06M6xMojn8pNQLzeMtX1o3LgAAIInWlJwuQ9sO1LESLd9diZbYSFEBAADAG46ui2ZhcwFJShpp7jMXWjsuAACtHEm0JrT7cLHKK10K8LOrQ5vgU99QmieV5ZvHTOcEAADwKe4kWkaOhZVoktTtPMnmkA5slg7vtHZsAABaMZJoTcg9lbNbTKgcdtupb3CvhxbcRgoIbcTIAAAA0NRS4syZCZn7C+W0qkOnJAVHSYlnmcdbv7JuXAAAWjmSaE0o092Zs7brobmnckbQmRMAAMDXJLYJUZC/XeWVLu08WGTt4EkjzD3rogEAYBmSaE1oa675cJRU686cu8x9JEk0AAAAX2O32zxTOi3t0ClJSReY++1LpMpya8cGAKCV8noS7YUXXlDXrl0VFBSk/v37a9myk7fiLisr07333qvOnTsrMDBQ3bt315w5c5oo2obJ3F/HSjT3dE7WQwMAAPBJvdpHSJKWbjlg7cDxfaTQdlJ5obTrO2vHBgCglfJqEm3u3LmaOnWq7r33Xq1Zs0bDhg3TmDFjlJWVdcJ7rrzySn311Vd67bXXlJGRobfeeks9evRowqjrxzCMo9M529VyfbO83eaeSjQAAACfNL5Pe0nSR2v3qqTcad3AdrvUvWpKZyZTOgEAsIJXk2izZs3SDTfcoClTpigtLU2zZ89WYmKiXnzxxeNe/9lnn2nJkiVasGCBLrjgAnXp0kVnnXWWhgwZ0sSR192honLllVTIZpO6xbAmGgAAAKRBXduqY5tgFZRV6vP1OdYOnjzS3LMuGgAAlvBaEq28vFyrVq3SqFGjqp0fNWqUli9fftx75s+frwEDBuixxx5Thw4dlJKSojvvvFMlJSUnfJ+ysjLl5+dX27zBXYXWISpYwQGO2t3kWRON6ZwAAAC+yG636Yr+iZKkd1ftsnbwbsMl2aTc9VL+XmvHBgCgFfJaEu3AgQNyOp2Ki4urdj4uLk45Ocf/Fm7btm365ptv9Msvv+j999/X7Nmz9b///U+33nrrCd9n5syZioyM9GyJiYmWfo7a2rrfbCrQvbZNBVyuow87TOcEAADwWZf17yCbTfo286B2HSq2buDQtlKH/uZx5lfWjQsAQCvl9cYCNput2mvDMGqcc3O5XLLZbHrzzTd11llnaezYsZo1a5bS09NPWI02ffp05eXlebZduyz+hq+WtlY1FUiqbVOBov2Ss1ySTQpPaLzAAAAA4FUd24RoaPcYSdK81butHdzdpTNzobXjAgDQCnktiRYTEyOHw1Gj6iw3N7dGdZpbQkKCOnTooMjISM+5tLQ0GYah3buP/8ARGBioiIiIaps3HG0qUNv10Ko+T3iC5PBvpKgAAADQHFwxwJx58O7K3XK5DOsGdq+LtnWx5Ky0blwAAFohryXRAgIC1L9/fy1cWP1bsYULF56wUcDQoUO1d+9eFRYWes5t3rxZdrtdHTs27ymP7kq02nfmrGoqwHpoAAAAPm90r3iFB/lpz5ESfbftoHUDt+8nBbeRyvKk3T9aNy4AAK2QV6dzTps2Ta+++qrmzJmjjRs36o477lBWVpZuvvlmSeZUzGuvvdZz/e9+9zu1bdtW119/vTZs2KClS5fqrrvu0u9//3sFBwd762OcUkm5U3uOmNNNaz2dM6+qEo310AAAAHxekL9DF/dpL0l6Z6WFy4/YHVL3883jTLp0AgDQEF5Nok2cOFGzZ8/WjBkz1LdvXy1dulQLFixQ586dJUnZ2dnKysryXB8WFqaFCxfqyJEjGjBggK6++mqNHz9ezzzzjLc+Qq1sO1Aow5CiQvwVHRpQu5vyqyrRIqhEAwAAaA2uHGA2wPr0lxzll1ZYN3BS1ZRO1kUDAKBB/LwdwC233KJbbrnluD9LT0+vca5Hjx41poA2d+7OnEntwk7YNKGGvKpvIKlEAwAAaBVO7xiplLgwbd5XqI/W7dXVAztbM3DSCHOfvU4qzJXCYq0ZFwCAVsbr3Tlbgzo3FZCOWRONJBoAAEBrYLPZPNVo7660sEtnWKyU0Mc83vq1deMCANDKkERrAp6mArG1bCogHV0TjemcAAAArcaEfh3kZ7dp7a4j2rKvwLqBky4w91ta1owOAACaE5JoTWBrVSVarZsKVJZLhfvM48jERooKAAAAzU1MWKDO72FOt3x3lYXVaO510bZ+Lbmc1o0LAEArQhKtkTldhrYdMNdEq/V0zoK9kgzJESiFxjRecAAAAGh2rqia0vne6t2qcLqsGbTjmVJgpFRySNq7xpoxAQBoZUiiNbI9h0tUXulSgJ9dHduE1O4mz3poHaTaNiIAAACATzgvtZ1iwgJ1oLBcizP2WzOow0/qdq55nPmlNWMCANDKkERrZO710LrFhMphr2VCLL8qicZ6aAAAAK2Ov8Ou35xhPge+s3KXdQMnV03pZF00AADqhSRaI6tfZ86qhyXWQwMAAGiVruhvdmhftClX+wvKrBm0+whzv2eVVHzImjEBAGhFSKI1sqOdOeuSRDtmOicAAABaneS4cPVNjFKly9AHa/ZYM2hkBym2lyTDbDAAAADqhCRaIztaiRZa+5vyqjoxMZ0TAACg1bqyqsHAOyt3yTAMawZNqqpGY100AADqjCRaI/NUotVlOqd7TTSmcwIAALRa4/okKMjfri25hVq3O8+aQd3romV+Jbks6vwJAEArQRKtER0qKtfh4gpJ9V0TjUo0AACA1ioiyF9jeidIkt61qsFA4iDJP1QqypVyfrJmTAAAWgmSaI3IPZWzQ1SwggMctbuprFAqrfqmkemcAAAArZq7wcD8dXtVWuFs+IB+AVK3c81jpnQCAFAnJNEaUb2aCrincgZGSkERjRAVAAAAWopB3dqqY5tgFZRW6vP1OdYMmnSBuSeJBgBAnZBEa0RbqyrRkuo1lbNjI0QEAACAlsRut+nyqmq0d6ya0ulOou36QSo5Ys2YAAC0AiTRGlGmpxKtLp053U0FmMoJAAAA6fL+HWWzScu3HtSuQ8UNH7BNZykmRTKc0vYlDR8PAIBWgiRaI3JP56xbJdpuc896aAAAAJDUsU2IhnRvK8OQ5q3ebc2g7mq0LQutGQ8AgFaAJFojKa1wavfhEkn1XBON6ZwAAACocuWAREnS/1btlstlNHxAz7poX0mGBeMBANAKkERrJNv2F8kwpMhgf7UNDaj9jayJBgAAgF8Z3Ste4UF+2n24RN9tO9jwATsPlfyCpYK9Uu6Gho8HAEArQBKtkXimcsaGyWaz1f7GPCrRAAAAUF2Qv0MX92kvSXp3lQVTOv2DpC5nm8d06QQAoFZIojWSzKrOnN3b1aGpgGEcnc7JmmgAAAA4xhVVUzoX/Jyt/NKKhg+YPNLcsy4aAAC1QhKtkbgr0brXpalA8UGpslSSTYpo3ziBAQAAoEXq0zFSKXFhKqt06eN12Q0f0L0uWtZ3UllBw8cDAMDHkURrJFv3F0kyp3PWmrszZ1is5BfYCFEBAACgpbLZbLqiv1mN9s7KXQ0fsG13qU1XyVUhbV/W8PEAAPBxJNEagdNlaFt9KtHcSTTWQwMAAMBxTOjXQX52m9buOqIt+yyoHvN06WRKJwAAp0ISrRHsPVKiskqXAhx2dWwTXPsbWQ8NAAAAJ9EuPFDDe8RKsqjBgHtdtMwvzfV5AQDACZFEawSZVVVoXWNC5eeow684r6osn0o0AAAAnMCVVQ0G3lu9RxVOV8MG63K25AiQjmRJB7ZYEB0AAL6LJFoj2OruzBlbh86ckpRXVYlGEg0AAAAncF5qO8WEBehAYZkWZ+xv2GABoVLnIeZx5pcNDw4AAB9GEq0RuDtzJtVlPTTp6JpoTOcEAADACfg77PrNGeaXru9a0WAgyT2lk3XRAAA4GZJojSDTU4lWxySae020yESLIwIAALDeCy+8oK5duyooKEj9+/fXsmUn7vC4ePFi2Wy2GtumTZuqXTdv3jz17NlTgYGB6tmzp95///3G/hgt0hX9zSTa15tydaCwrGGDuZsL7PhWKi9uYGQAAPgukmiNYOv+Ikl17MzprJQKss3jSCrRAABA8zZ37lxNnTpV9957r9asWaNhw4ZpzJgxysrKOul9GRkZys7O9mzJycmen61YsUITJ07UpEmTtG7dOk2aNElXXnmlvv/++8b+OC1Ocly4+iZGqdJl6IM1exo2WLtU80tcZ5m081trAgQAwAeRRLPYoaJyHSoqlyR1a1eHNdEKsiXDJdn9pdDYRooOAADAGrNmzdINN9ygKVOmKC0tTbNnz1ZiYqJefPHFk94XGxur+Ph4z+ZwODw/mz17tkaOHKnp06erR48emj59ukaMGKHZs2c38qdpma4YYFajzf1xl4yGdNa02aSkEebxFqZ0AgBwIiTRLOZeD61DVLBCAvxqf6N7KmdEe8nOHwsAAGi+ysvLtWrVKo0aNara+VGjRmn58uUnvbdfv35KSEjQiBEjtGjRomo/W7FiRY0xR48efcIxy8rKlJ+fX21rTcb3aa9AP7u25BZq3e68hg3mWReN5gIAAJwI2RqLuTtz1qkKTTraVIDOnAAAoJk7cOCAnE6n4uLiqp2Pi4tTTk7Oce9JSEjQK6+8onnz5um9995TamqqRowYoaVLl3quycnJqdOYM2fOVGRkpGdLTGxd68pGBPlrTO94SRY0GOh6jmT3kw5tlQ5tsyA6AAB8D0k0i3k6c9a1qQBJNAAA0MLYbLZqrw3DqHHOLTU1VTfeeKPOOOMMDR48WC+88IIuuugiPfHEE/Uec/r06crLy/Nsu3ZZ0KmyhbligJk4nL9ur0ornPUfKChC6jTYPM78yoLIAADwPSTRLLbzoNnRqE5NBaSjSbQImgoAAIDmLSYmRg6Ho0aFWG5ubo1KspMZNGiQtmzZ4nkdHx9fpzEDAwMVERFRbWttBndrqw5RwSoordTn649fsVdrrIsGAMBJkUSz2EvX9Nc3dw/XuNMT6naje000KtEAAEAzFxAQoP79+2vhwurJloULF2rIkCG1HmfNmjVKSDj6zDR48OAaY37xxRd1GrO1sdttury/+fz4TkOndCZdYO53LJMqShsYGQAAvqcOK9+jNux2mzq2Can7jXlVDz0k0QAAQAswbdo0TZo0SQMGDNDgwYP1yiuvKCsrSzfffLMkc6rlnj179MYbb0gyO2926dJFvXr1Unl5uf7zn/9o3rx5mjdvnmfM22+/Xeecc44effRRXXLJJfrwww/15Zdf6ptvvvHKZ2wpLu/fUU9/tUXLtx7UrkPFSoyux7OoJMX1lsLipcIcKWuF1H24tYECANDCUYnWXORRiQYAAFqOiRMnavbs2ZoxY4b69u2rpUuXasGCBercubMkKTs7W1lZWZ7ry8vLdeedd+r000/XsGHD9M033+iTTz7Rb37zG881Q4YM0dtvv63XX39dp59+utLT0zV37lwNHDiwyT9fS5IYHaIh3dvKMKR5q3fXfyCb7Wg1Gl06AQCowWYYhuHtIJpSfn6+IiMjlZeX13zWzSgvlh6umspw904pOMqr4QAAgJqa5TMEqmnNf0YfrNmjqXPXqmObYC29a7js9uM3Yzil9e9L706W2vWQbv3e0hgBAGiuavsMQSVac+BeDy0gXAqK9G4sAAAAaHFG94pXeKCfdh8u0XfbDtZ/oG7nSTa7tH+TdKT1dTsFAOBkSKI1B+7OnJEdzDJ6AAAAoA6CAxwa37e9JOndVQ2Y0hncRup4lnnMlE4AAKohidYcuJNoER28GwcAAABarCuqunQu+Dlb+aUV9R+IddEAADgukmjNQT5NBQAAANAwfROjlBwbprJKlz5el13/gZJGmPttS6TKcmuCAwDAB5BEaw7yqtabIIkGAACAerLZbLpigPk8+c7KBqxnltBXComRyguk3T9YExwAAD6AJFpzkEclGgAAABru0n4d5bDbtHbXEWXkFNRvELv9aDXaloXWBQcAQAtHEq05YE00AAAAWKBdeKBGpsVJkp5blFn/gZJGmvvMryyICgAA30ASzdsMgzXRAAAAYJnbRiRLkj5at1cbs/PrN0j38yXZpH0/S/kNWF8NAAAfQhLN20oOSxXF5jGVaAAAAGignu0jdNHpCZKkpxZurt8goW2lDmeYx1upRgMAQCKJ5n3uqZwhMZJ/kHdjAQAAgE+444IU2W3SFxv2ad2uI/UbJOkCc8+6aAAASCKJ5n1M5QQAAIDFkmLDdGk/8/nyyfpWo7mTaNsWSc5KiyIDAKDlIonmbe5KNJJoAAAAsNDtI5LlZ7dp6eb9+mH7oboP0KG/FBQlleZJe1ZZHh8AAC0NSTRvI4kGAACARtCpbYiuPDNRkvTEFxkyDKNuA9gdVQ0GJGUypRMAAJJo3uZOotFUAAAAABb78/lJCvCz64fth/RN5oG6D5A80txnfmltYAAAtEAk0byNNdEAAADQSBIig3XNwM6SpCe+2Fz3arTuI8z93jVS4X6LowMAoGXxehLthRdeUNeuXRUUFKT+/ftr2bJlJ7x28eLFstlsNbZNmzY1YcQWyyOJBgAAgMbzx/O6K9jfoXW7juirjbl1uzk8Too/3Tze+rX1wQEA0IJ4NYk2d+5cTZ06Vffee6/WrFmjYcOGacyYMcrKyjrpfRkZGcrOzvZsycnJTRSxxVxOKtEAAADQqNqFB2ry0C6SzLXRXK46VqO5u3SyLhoAoJXzahJt1qxZuuGGGzRlyhSlpaVp9uzZSkxM1IsvvnjS+2JjYxUfH+/ZHA5HE0VsscJ9kuGU7H5SWJy3owEAAICP+sM53RQe6KdNOQVa8Et23W72JNG+knaukPL3Si6X9UECANDM+XnrjcvLy7Vq1Srdc8891c6PGjVKy5cvP+m9/fr1U2lpqXr27Km///3vGj58+AmvLSsrU1lZmed1fn5+wwK3krupQHiC2f0IAAAAaARRIQGaMqybnvpys2Yt3KwLe8XLz1HL79MTz5ICI6SSQ9LrF5rnHAFSZKLUprMU1VmK6lR13MXch7SVbLZG+zwAAHiD15JoBw4ckNPpVFzc/2/vzsOjKu/3j98zk5nJQlZCFiAJYRUBsQaEsBaFWLTWhf7E2irW7UurVkitVdGqaMVatdRaqNatakWsWpdC1VhlE3ChoIgoCEgiJA0EspA9M+f3x8kMCUlIkMycCXm/rutcc+bMmZnPcDLwcOdZmvfASk5OVlFRUavPSU1N1WOPPaasrCzV1tbq2Wef1ZlnnqkVK1Zo0qRJrT5nwYIFuuuuuzq9/k7hC9EYygkAAIAAu2JCPz21dpd27qvUq5v26odZHWyDOpzS9/8gbXhaKt1tzunrqZMO7DC31jijmgRr6WbQ1jRwi4jrrI8FAEDQWBai+diO+A2VYRgtjvkMGTJEQ4YM8d/Pzs5WQUGBHnjggTZDtFtuuUW5ubn+++Xl5UpLS+uEyjuBL0SL6WNtHQAAADjhRYc7NXvyAN337y/0x/9s0w9G9pYrrIO90Ub80NwkydNgzutbulsqzZcO7m6+X1Eo1VdK+7aaW2vCY81ArU+WNGGOFN+vMz4iAAABZVmIlpiYKIfD0aLXWXFxcYveaUczduxYPffcc20+7na75Xa7v3WdAcWiAgAAAAiiWdn99PjqXSo4UK1/bCjQj8dkHPuLOMLMXmXxbTy3odb8ZfHBr1sP2ir3STVlUtGn5rbxOWnUT6WJN5qrgQIAEKIsC9FcLpeysrKUl5enCy64wH88Ly9P5513XodfZ+PGjUpNTQ1EiYHHcE4AAAAEUYTLoeumDNCdb3yuP/3nK804ra/CnZ08N2+YW+o5wNxaU1dphmkHdkof/lXa+Z704WNmmDb2Z9K4XzDcEwAQkixdnTM3N1ePP/64nnzySW3dulVz585Vfn6+Zs+eLckcinnZZZf5z1+4cKFeffVVbd++XVu2bNEtt9yil19+Wdddd51VH+H4EKIBAAAgyH40Jl29Y8NVVF6jv3+QH/wCXFFS0lDppHOky16VLnvdHNZZXyWtflD640hpzR+kuqrg1wYAwFFYOifazJkzVVJSovnz56uwsFDDhw/X8uXLlZFhdg0vLCxUfv7hf9jr6up04403as+ePYqIiNCwYcO0bNkynX322VZ9hOPjG87JnGgAAAAIEneYQ9efOUi3vLJZi1d8pYtHpynKbeF/C/pPljL/I32xTHr3bmnfF9I7d0rr/yJN/pV02ixzcQMAACxmMwzDsLqIYCovL1dsbKzKysoUExNjXSH1NdJvG+d8uGmXFJlgXS0AAKBdIdOGQJu4Rh1X7/Fq6kMrtbukSr86a4iunTLQ6pJMXo/06YvSinvNIZ+SuejAlHnS8B9K9iAPpPF6pP3bpZjeUjg/UwBwoupoG8LS4Zzdmq8XmjNSioi3thYAAAB0K06HXXOmDpIkPbpyh8qq6y2uqJHdIZ36I+m6j6Xpv5eikswFCl65WvrLBOnLf0uB7ANgGFLJDumjx6WlP5Huz5QWjZEWjpA2PR/Y9wYAhDxCNKv45kOL6SPZbNbWAgAAgG7nByP7aFBSD5XXNOiJNbusLqe5MLc05hrphk3SGbdL7lipeIu05GLpybOkr9d03nsdKpY2vyS9dq0Zlv3pNGnZL6Wtb5iriNrDpJpS6dWfSc9daK40CgDolgjRrOLricaiAgAAALCAw25T7rTBkqQnVu/Ugco6iytqhStKmnSjGaaNnyOFRUgFH0hPnyM9e6G0d9Oxv2btIWl7nvTWPGnxeOmBQdLLV5qrg5YVSHan1G+idMZt0pXvSLfskabeKYWFSzvelRaNldYtMod6AgC6FUsXFujW/CtzsqgAAAAArHHWsBQN6x2jLXvL9ejKHbrl7KFWl9S6yARp2l3SmNnSqt9L//2btOM/5jbsAmnKbVJiG/O6eRqkvf+Vdq4wt4IPJe8Rw1dTRkj9v2tu6dlmeNfUhLnS0B9Ir/9C2r1GeusW6bOXpR/8SUo+ufM/LwAgJBGiWcUfoqVZWwcAAAC6Lbvdpl/mDNYVT3+sv637WldOyFRSTLjVZbUtJlX6/kPSuOuk9xZIm/8hbfmn9Pnr0nd+LE2+2VwEYP+2w6HZ12uk2vLmrxObLg34rhmaZU6WohLbf++eA6RZb5gBXt5vpD0fS49OkibmShN/aQ5BBQCc0AjRrNJ0TjQAAADAIlOGJOk76XHamF+qRSt26M4fDLO6pPYl9Jdm/FUaf4P07j3Stn9L/31G+mSp2WutorD5+RHxUuakw73N4jO/3bzEdrs06qfS4LPMedO+XC6t/J30+Wtmr7S00zvj0wEAQhRzolmFOdEAAAAQAmw2m36VM0SS9PwH+dpTWm1xRccgZbh0yQvSFW9LGeMlT60ZoIWFm2HZ1Dula1ZIv9ohXfSMNOoKM4A73oW9YnpLFz8v/fApKaqXtO8L6Ykc6d+/NudcAwCckOiJZgXDaDKckxANAAAA1ho3MFHZ/Xtq3c4S/ek/23XfjFOsLunYpI+RLl8mffOx1FAt9T1dcgZ4WKrNJg2/0Azr3ponffK89MFfpC+WS+f+QRo4NbDvDzTUSoZXckZYXUlw1FdLe/4r5a+T/veZGZZHJEiR8Y23CS1vu8ufDYKGEM0KNWVSXeNvqBjOCQAAgBBw41mDNWPxOv1jwzeaPXmA+iVGtf+kUGKzSWmjg/++kQnSBYulET+U/jVHKs2XnpshnXKx9L0F5uPoOgxDKvnKXH016SSrq2ldbYW0ZqG07hFJNmnI96ThM6SB0wIfHgfToWIpf725Im/+eqnwk5aLgrQnLMIczh2Z0OS2jcAtIl5yx0juHpIz8vh7rOKERIhmBd9QzogEyRVpbS0AAACApKyMBE0Z0kvvfblPC9/ZpoUXf8fqkrqWgWdKP1snvfdbaf1i6dMXpK/ekc6+Xxp2If8hD2V1VdLXq6Xtb0vb86TS3ebx/lOkM26T+o6ytj4fT4O08RnpvXulyn2Hj2/5p7m5Y6STvm8Gav0nSw6ndbUeK6/XXBCkYL2U/4F5e2Bny/OiU6W0MVKfLLMXXvUBqeqAVH2w8fbA4Vtvg9kztaJaqth7bPXY7JIrWnJHm6GaO1pyNd423Y481uKcmBMr2AQhmiX8QznphQYAAIDQkTttiN77cp9e+2Svfj5loAYnR1tdUtfi7mH2Phs+Q3rtOmnfVumlK6RP/yGd8yDt/1BSssMMzL7Kk3atNufT83G4zIBm53vmNni6NOVWKdWiYc6GYQZ8b98u7f/SPJYwQJo23/yZ+uxl6bNXzM4anzxvbpE9pZPPM38W08eZi2KEkvpqae9Gc2hm/gdmb7Oa0iNOsklJJ5vDtdPGmrdxGR0LpA3D7LHXNFSrLm0ZtDW7LW1cydcwr39tmbkdL2eUFNVTikw0r0tU09vEw/d9x9wxhO4hzGYYhmF1EcFUXl6u2NhYlZWVKSYmxpoiPnpCWpYrDTlb+tESa2oAAADHJCTaEDgqrlHnmP3sBr25pUjTh6do8U+yrC6n62qok9b8QVr1e3MImitamnaXlPXT0As0uoP6Gmn3GjM4254nHdjR/PHYdGnQNHPLnGT29Fr5ezOQMrzmOSefJ3331uAO8yz8VHp7nrRrlXk/IkH67s3mIhlNe5p5vWYQ9dnLZq+0qv2HH4tONXtDDp8h9TnNmoDm0L7GXmaNwzP3bmo5NDMswuz1lzZGSs829yPiglunYUj1VWYAV3vIDNVqK8zpmGorDm/++62dc+jwfX2LuMXhagzVEhvDt57Nw7bwWMkeJtkdks3R5NZ+xP2jHbc3vy+Zc+x56s1AuaFO8tSZ+576xscajzXd99S1fa7ve6PGnzebrcl+W8dth5/T2r4rygyPA6CjbQhCNCu8c5e05iFp9NXSOQ9YUwMAADgmIdGGwFFxjTrHtv9V6KyFq2QY0r+un6DhfWKtLqlrK94qvX699M1H5v2M8dK5D0uJA62tqzs4uNvswfXVO9LOlebQPh+7U8rIlgblmHOJ9RrSeri0/ytp5X3S5pdkBiI26ZSLpMm/lnoOCFztZXukd++RPllivq/DJY39mTQht/1gydMgfb3KDNQ+f6N5b6r4fmaYNnyG2curMwO1hjpzKOyBXeZQTN9Wsl06+HXL83skS+ljD/cySzmlaw1BbY/XK9VVSJX7zZ5uVfsb9323JYdvq/ZLlSVSfaXVVYe2iHjp118H5KUJ0doQEo2rV66RPl0qTb1LmjDHmhoAAMAxCYk2BI6Ka9R55rywUa9u2qspQ3rpqZ+ebnU5XZ/XI334V+k/883/JDvcUvIwmcPGjCNudfi+4W3jHKPlOTa7lDzc7EXVb6KUNLT7DQlrqDWHB/p6m/mGPvpE927S22yyFH4Mf0/873Npxb3S1jfM+zaH9J0fS5N+JcWld95n8C8a8OfDod/wH0pn/kaKzzj212uolb76jxmofbnc7GXl0+ukw4FaRwPBuiozEDuwUzp4RFhW9k2T3ketSDq5sZfZWPM2vl/3+xltT311k6CtpEnA1uRYbbn5d4rhOeLW28pxbyvntXJcMv9eCnOZga3DbQaaYY23Drd53P9449bscd/5jY/ZHY1/V0mH/95qbV+H96XDf7+1tu8Mlyb+MiB/9IRobQiJxtVT55hdiWc8Ya7iAwAAQl5ItCFwVFyjzrNrf6WmPrRSHq+hl3+WrawMVpjsFAd3S/+aK+34T+DfKzJR6jdBypwo9ZskJQ468QKLhjpzJc2CD8zQbNfKxiF0jWwOM7AZONXscZY87Pj/DPZuNCf23/62ed/ulLIuN/9jH5P67V+3tUUD0rOlnN9KfTtpWHVdpbTtTXP+tO1vm0PufFJPNf9vOuwCc06uZgHZrsO9y9qboN8ZJSX0lxIyG7f+UnymOZ9cRHznfA4gAAjR2hASjauFp5jdXH/6ptmFGAAAhLyQaEPgqLhGnevXL32qpR8XKLt/Ty25ZqzV5Zw4DEMq+NCczNw/14/N7EnmmyfIf6zx+JHHWnusvlr65kNzkvz89c2HLkrm0Ll+ExtDtYlmuNFVQjVfWLZvq7TvS3OI7L4vzMUBfL1ofKKSDvc26z8lcHNq5X8gvXfP4bnKwsKl0VdJE+aac1d1lGGYAWDe7eZnksxrM22+udJmoK5Rdan0xTKzh9rOFS3/HI/GHSv17N8YljWGZL79Hkld5+cKaIIQrQ2WN668XumeJHMSxTmfSXFpwa8BAAAcM8vbEGgX16hz7Smt1pTfr1Cdx6vnrxqjcQOPIRiAtRrqpD0bpK9XmyFPwYfNV5+UpJg+Zk81X7AW38+SUptpqDMn/C9uDMv2bZWKvzCPeRtaf447xuxhNuBMMzhLOSW4CzfsWmXOXVbwgXnfGWXOXTbuuvZ7XhV+Kr19m9mDTjq8aEDWT81hc8FyaJ+09TWzh9ru981jkYmHgzH/1hiWRcQTlOGEQ4jWBssbVxVF0oNDzN8a3bZPcoQFvwYAAHDMLG9DoF1co853x2uf6W/rduu09Di9/LNxsvEf566pvsZc2ODr1WZPtW8+arkyYmz64V5qmROl2L6Bq8dTb/Yi84Vk+xq3kq/aDstc0eaqmL0at6STpF5DpZje1gc6hmHOO/bu3VLhJvOYO9YM0sbMbjn3WmuLBoyZbQ4JDfZqlEeqOmCu/Hgs88UBJwBCtDZY3rj6ZoP0+Bnmb35yPw/++wMAgG/F8jYE2sU16nzF5TWaeP97qm3w6qnLR2vKSUlWl4TOUFd1eOjn16vNXmtHhlfxmeacXM6IJhOQG80nMTeaTmbubT6xeVvHaw8dvWeZK9pcKdMXkvkCs5g+1odl7TEMc4jke7+Vihv/rxeRYC4mN/pq8/O//0dp7SOds2gAgE7T0TYE3aCCrazAvI3pY20dAAAAQDuSYsI1a1w/PbZqpx54+0tNHtxLdnuIBxlonytS6v9dc5PMYKtg/eFQbe9Gc2L5g7sCWEOPlr3KukpY1habTRr6fWnI2dKWV6QVC8zedXm/MYMzSaosNm/Ts6Wce6S+o6yrF8AxI0QLtvI95m0gu0cDAAAAnWT25AH6+/rd2rK3XPcs26rbvz+UYZ0nGncPcwXLgVPN+zXlUv46ae8msxeZ3dG4kIGjcd9+xL6948fD3FLPQeb/h07UnyO73Vzp8uTzpc0vmmFaab75WDAWDQAQMIRowVb2jXkbS080AAAAhL6EKJd+e8EIzVm6SU++v0vxkU5df+Ygq8tCIIXHSIPPMjd8e44w6dRLzCGbm/9hDmEd+aPgLhoAoFMRogWbP0RjVU4AAAB0Ded/p48OVNZp/r8+14N52xQX6dSl2f2sLgvoGsJc0nd+bHUVADpBENf+haTDwzmZEw0AAABdyBUTMvWLxh5ov3l9i17btMfiigAACC5CtGBjOCcAAAC6qLlTB2lWdoYMQ/rli5/ovS+KrS4JAICgIUQLpoZa6dD/zH2GcwIAAKCLsdlsuuPcYTrv1N5q8Bqa/dwGffT1AavLAgAgKAjRgql8r3kbFi5F9rS2FgAAgOO0aNEiZWZmKjw8XFlZWVq9enWb577yyiuaNm2aevXqpZiYGGVnZ+utt95qds7TTz8tm83WYqupqQn0R8ExsNtteuD/jdQZJyWptsGrK57+SJ/vLbe6LAAAAo4QLZiazofGcsYAAKALW7p0qebMmaN58+Zp48aNmjhxoqZPn678/PxWz1+1apWmTZum5cuXa8OGDZoyZYrOPfdcbdy4sdl5MTExKiwsbLaFh4cH4yPhGDgddv35ktM0ul+8KmoadNmTH+rr/ZVWlwUAQEARogUT86EBAIATxEMPPaQrr7xSV111lYYOHaqFCxcqLS1NixcvbvX8hQsX6qabbtLo0aM1aNAg3XvvvRo0aJDeeOONZufZbDalpKQ02xCaIlwOPT5rtE5OjdH+Q7X6yRMfqKiMXoMAgBMXIVow+UM05kMDAABdV11dnTZs2KCcnJxmx3NycrR27doOvYbX61VFRYUSEhKaHT906JAyMjLUt29fff/732/RU62p2tpalZeXN9sQXLERTv3titPVr2ekvjlYrUuf+EAHK+usLgsAgIAgRAsmX4gWQ080AADQde3fv18ej0fJycnNjicnJ6uoqKhDr/Hggw+qsrJSF110kf/YSSedpKefflqvv/66lixZovDwcI0fP17bt29v9TUWLFig2NhY/5aWxi8qrdAr2q1nrxyjlJhwbS8+pJ8+/ZEqaxusLgsAgE5HiBZMvjnRGM4JAABOALYj5ng1DKPFsdYsWbJEd955p5YuXaqkpCT/8bFjx+onP/mJRo4cqYkTJ+rFF1/U4MGD9ac//anV17nllltUVlbm3woKCo7vA+FbS0uI1LNXnq64SKc2FZTq/57doNoGj9VlAQDQqQjRgqnMF6L1tbYOAACA45CYmCiHw9Gi11lxcXGL3mlHWrp0qa688kq9+OKLmjp16lHPtdvtGj16dJs90dxut2JiYpptsM6g5Gg9/dPTFelyaM1X+zXnhU3yeA2rywIAoNMQogWTfzgnIRoAAOi6XC6XsrKylJeX1+x4Xl6exo0b1+bzlixZossvv1zPP/+8zjnnnHbfxzAMbdq0SampqcddM4Lj1LQ4/fWyUXI57Pr3Z0W69ZXNMgyCNADAiYEQLVhqyqXaMnOf4ZwAAKCLy83N1eOPP64nn3xSW7du1dy5c5Wfn6/Zs2dLModaXnbZZf7zlyxZossuu0wPPvigxo4dq6KiIhUVFamsrMx/zl133aW33npLO3fu1KZNm3TllVdq06ZN/tdE1zB+YKIe/tGpstukpR8X6L43v7C6JAAAOgUhWrD45kMLj5Xc0dbWAgAAcJxmzpyphQsXav78+Tr11FO1atUqLV++XBkZGZKkwsJC5efn+89/9NFH1dDQoGuvvVapqan+7YYbbvCfU1paqmuuuUZDhw5VTk6O9uzZo1WrVun0008P+ufD8fne8FTdd+EpkqRHV+7U4hU7LK4IAIDjZzO6Wf/q8vJyxcbGqqysLLjzZmx/R/r7DCl5uPSz94P3vgAAoFNY1oZAh3GNQs9jq3bo3uVmT7QFF47Qj05Pt7giAABa6mgbgp5owVLWuFpUDEM5AQAA0D1cM2mAfvbdAZKkef/crOWbCy2uCACAb48QLVjKWZkTAAAA3c9NZw3Rj05Pl9eQbnhho1Zv32d1SQAAfCuEaMFS5gvR6IkGAACA7sNms+me84frnFNSVe8xdM0zG/Tf/INWlwUAwDEjRAsW/3BOeqIBAACge3HYbfrDRadq4qBEVdd79NOnPtKXRRVWlwUAwDEhRAsWhnMCAACgG3OF2fXopVk6LT1OZdX1uvSJD1RwoMrqsgAA6DBCtGAwDIZzAgAAoNuLdIXpyctHa0hytIoranXxY+u1fmeJ1WUBANAhhGjBULlf8tRKsknRva2uBgAAALBMXKRLz155ujJ6RmpPabUufmy95v1zsypq6q0uDQCAoyJECwbffGg9kqUwl7W1AAAAABZLignXG9dP0CVj0iVJf/8gXzl/WKX3vii2uDIAANpGiBYMzIcGAAAANBMT7tS9F4zQ81ePUUbPSBWW1einT3+kOS9s1IHKOqvLAwCgBUK0YCj7xrxlPjQAAACgmXEDEvXmDZN09cRM2W3Sq5v2atpDK/XGJ3tlGIbV5QEA4EeIFgy+EC2GnmgAAADAkSJcDs0752S98vPxGpIcrZLKOl2/ZKOufmaD/ldeY3V5AABIIkQLDoZzAgAAAO06NS1Ob1w/QXOmDpLTYdM7W/+nqQ+t1Asf5tMrDQBgOUK0YGA4JwAAANAhrjC75kwdrH9dP1Ej0+JUUdOgm1/ZrB8//oHyS6qsLg8A0I0RogVDGT3RAAAAgGMxJCVar/xsnG47Z6jCnXat3VGinIUr9fjqnfJ46ZUGAAg+QrRA89RLFYXmPnOiAQAAAB3msNt01cT+evOGSRrbP0E19V7ds2yrZixeq23/q7C6PABAN0OIFmgVhZIMyeGSonpZXQ0AAADQ5fRLjNLzV43VvReMULQ7TJsKSnXOw6v18H+2q67Ba3V5AIBughAt0Pwrc/aW7PxxAwAAAN+G3W7TJWPS9XbuJJ15UpLqPYYeytumHzyyRp8UlFpdHgCgG7A81Vm0aJEyMzMVHh6urKwsrV69ukPPe//99xUWFqZTTz01sAUeL/98aGnW1gEAAACcAFJjI/T4rFH648WnKiHKpS+KKnTBove1YPlWVdd5rC4PAHACszREW7p0qebMmaN58+Zp48aNmjhxoqZPn678/PyjPq+srEyXXXaZzjzzzCBVehzKfT3RWJkTAAAA6Aw2m03nndpHeXMn6bxTe8trSI+u2qnpf1yl9TtLrC4PAHCCsjREe+ihh3TllVfqqquu0tChQ7Vw4UKlpaVp8eLFR33e//3f/+mSSy5RdnZ2kCo9Dr7hnLGEaAAAAEBn6tnDrT9e/B09ftkopcSE6+uSKl382Hrd8spmlVXXW10eAOAEY1mIVldXpw0bNignJ6fZ8ZycHK1du7bN5z311FPasWOH7rjjjg69T21trcrLy5ttQeUfzsnKnAAAAEAgTD05WW/nTtKPTk+XJC35MF/THlqpNz8rsrgyAMCJxLIQbf/+/fJ4PEpOTm52PDk5WUVFrf9jt337dt188836+9//rrCwsA69z4IFCxQbG+vf0tKCPDeZf2EBQjQAAAAgUGLCnVpw4Qi9cM1YZSZGqbiiVrOf26DZz25QcXmN1eUBAE4Ali8sYLPZmt03DKPFMUnyeDy65JJLdNddd2nw4MEdfv1bbrlFZWVl/q2goOC4az4mvjnR6IkGAAAABNzY/j317xsm6topAxRmt+nNLUU686GVeuHDfBmGYXV5AIAurGPduQIgMTFRDoejRa+z4uLiFr3TJKmiokIff/yxNm7cqOuuu06S5PV6ZRiGwsLC9Pbbb+uMM85o8Ty32y232x2YD9Geukqp+qC5z5xoAAAAQFCEOx361Vkn6ZwRvXXzK5/q02/KdPMrm/Xqpj1acOEpykyMsrpEAEAXZFlPNJfLpaysLOXl5TU7npeXp3HjxrU4PyYmRps3b9amTZv82+zZszVkyBBt2rRJY8aMCVbpHeebD80dI4XHWlsLAAAA0M2c3DtGr/xsnG47Z6jCnXat33lAZy1cpUUrvlK9x2t1eQCALsaynmiSlJubq0svvVSjRo1Sdna2HnvsMeXn52v27NmSzKGYe/bs0TPPPCO73a7hw4c3e35SUpLCw8NbHA8ZZY1DR2PohQYAAABYIcxh11UT+yvn5BTNe3WzVm/fr/vf/FL/+qRQv5txikb05ZfdAICOsTREmzlzpkpKSjR//nwVFhZq+PDhWr58uTIyMiRJhYWFys/Pt7LE41PuW5mTEA0AAACwUnrPSD1zxel65b97dPeyz/V5YbnO+/MaXTWxv+ZOHawIl8PqEgEAIc5mdLPZNcvLyxUbG6uysjLFxMQE9s3eWyCtvE/Kulw694+BfS8AABBQQW1D4FvhGqGj9h+q1fw3Ptfrn+yVJKUnROreC0ZowqBEiysDAFiho20Iy1fnPKGVNa7MGcPKnAAAAECoSOzh1sM/+o6emDVKqbHhyj9QpZ888YF+9Y9PVFpVZ3V5AIAQRYgWSOWNIVosIRoAAAAQas4cmqy83MmalZ0hm036x4ZvNPWhlfrXp3vVzQbsAAA6gBAtkHw90ZgTDQAAAAhJPdxhuuu84XppdrYGJvXQ/kN1uu75jbr6mY9VWFZtdXkAgBBCiBYohiGV+RYWoCcaAAAAEMqyMhK07BcTdMOZg+R02PTO1mJNe2iVnl33tbxeeqUBAAjRAqfqgNTQ+JurGHqiAQAAAKHOHebQ3GmDtewXE/Wd9Dgdqm3Q7a9t0eQH3tPtr36mdz7/nyprG6wuEwBgkTCrCzhh+eZDi0qSwtzW1gIAAACgwwYnR+ul2eP03Prduv/NL1RwoFrPrt+tZ9fvltNh06iMBE0e0kuTB/fSSSnRstlsVpcMAAgCQrRAYT40AAAAoMty2G2aNa6fZmT11bodJVq5rVgrt+1TwYFqrdtZonU7S3Tfv79QUrRbkwabgdqEgYmKj3JZXToAIEAI0QLFNx8aQzkBAACALquHO0zTTk7WtJOTZRiGvi6p0sovi7Vq+36t21Gi4opavbThG7204RvZbdIpfeM0eXAvTR7SSyP7xslhp5caAJwoCNECxTecMzbN2joAAAAAdAqbzabMxChlJmbq8vGZqm3w6OOvD2rltn1a+eU+ffm/Cm0qKNWmglL98T/bFRvh1IRBiWaoNriXkmPCrf4IAIDjQIgWKAznBAAAAE5o7jCHxg9M1PiBibr17KEqLKvW6m37tXLbPq3evk9l1fVa9mmhln1aKEk6KSXaP/RzYFIPxUU65Q5zWPwpAAAdRYgWKL7hnLF9ra0DAAAAQFCkxkbootFpumh0mho8Xn3yTZlWbdunldv26ZNvSvVFUYW+KKrQY6t2+p/Twx2mhCiX4qNcSoh0Kj7KpZ7++y4lRLn8j/eMcikm3Ck7Q0QBwBKEaIHi64kWQ4gGAAAAdDdhDruyMuKVlRGvudMG62BlndZ8ZfZSW7ejREXlNfJ4DR2qbdCh2gblH6jq0OvabVJ8ZGPI1hi0xUe51KuHS2P699SYzASFOewB/nQA0D0RogWCp0GqMLts0xMNAAAAQHyUS+eO7K1zR/aWJHm9hipqGnSgqk4HKs3tYGVdi/sllXU62HisoqZBXkMqaTzewrtfqWeUS2cNT9E5I1IJ1ACgkxGiBcKhIsnwSPYwqUeS1dUAAAAACDF2u02xkU7FRjqVmRjVoefUNXhVWtUYtB0yb31BW8GBar37xf9UUlmn5z/I1/Mf5KtnlEs5w1L0/VMI1ACgMxCiBYJvPrTo3pKdiUIBAAAAHD9XmF1JMeFKamOVz3qPV+t3lmj55kK9+VmRSirrtOTDfC35MF8JUS6dNczsoTa2P4EaAHwbhGiBUO5bmZOhnAAAAACCw+mwa+KgXpo4qJfuPm+41u88oGWb9+rNz4p0gEANAI4bIVog+BYViO1jbR0AAAAAuqUwh10TBiVqwqDEDgRqyTpnRG8CNQBoByFaIPiGc9ITDQAAAIDFWg/UCvXWFl+gVqAlHxb4A7WzR6Qqu39PAjUAOAIhWiD4eqLF0BMNAAAAQOhoHqgNazNQi490atrJyRqaGqO0+EilJUQqLSFCkS7+Cwmg++JvwEDwz4mWZm0dAAAAANCGIwO1D3Yd0L8+PRyovfjxNy2e0zPKpb4JkUqLjzCDtXgzXEuLj1TvuAi5wui9BuDERYgWCMyJBgAAAKALCXPYNX5gosYPPByordq2T7tLqlRwsEoFB6pUXtOgkso6lVTW6ZOC0havYbdJKTHhjSHb4XDN14stOTpcdrst+B8OADoJIVpnq6+WqkrMfeZEAwAAANDFNA3UmiqrrlfBgSp9c7BKBQeq/eFawcFqfXOwSjX1Xu0tq9Heshp9uOtAi9d1OexKjnUrPMwhV5jd3BxH3IbZ5WxyzN3KsSPPd9htMgxDHq/kMQx5vYa8hiGP/9Y8bjQe8x33GjL3vYb5PEP+fafdprhIl+KjnOZtpEsJkS7FRTkV7Q6TzUYYGGxl1fXaXVKp3SVVKiqrUUKUS33iI9Q3PkIpMeHM4YegIETrbL5FBZxRUnicpaUAAAAAQGeJjXAqtk+shveJbfGYYRjad6hWBQeqG0O2JkHbwSrtLa1RncerggPVFlTeucLsNsVFOhXfGK7596Ncim/cj4t0KiHK1RjAmUHc4bDPUIPXUJ3HqwaPoQaP9/C+16t6j6F6j3nb0Hhb7zUfN48f3g93OhQX6Wz2PjHhXTPk8/0M7S6patzMwGz3AXO/tKq+zec67DalxIT7Q7W+cRHqGx/pv58ay1BjdA5CtM5W3mQoZxf8iwsAAKCjFi1apN///vcqLCzUsGHDtHDhQk2cOLHN81euXKnc3Fxt2bJFvXv31k033aTZs2c3O+fll1/W7bffrh07dmjAgAH67W9/qwsuuCDQHwXAcbLZbEqKDldSdLiyMuJbPN7g8aqwrEbFFTWqrTdDo7qGw7f1jbe1jcfqGwzVeTzmOQ1e1XmMJud7VO+73+BVg9crh90mm80mh83WuG8GKw6bTXa7TfbG+/bGx+02375afW69x6uDVfUqrarTwUrz9kBVnWrqvWrwGtp/qE77D9Ud05+Ry2FXncfbWX/kbXLYbYqLcPrDvSNDtvjG/dgjgsBwpyPgtXm8hvaWVjeGY5XNwrL8A1WqqvMc9fm9ot3KSIhUalyEDlbWaU9ptfYcrFadx2vul1brw10tn2ezScnRTUK2+Aj1iYs0b+Mj1CcuIiifH10fIVpn8/VEYygnAAA4gS1dulRz5szRokWLNH78eD366KOaPn26Pv/8c6Wnp7c4f9euXTr77LN19dVX67nnntP777+vn//85+rVq5dmzJghSVq3bp1mzpypu+++WxdccIH++c9/6qKLLtKaNWs0ZsyYYH9EAJ0ozGFvnBst0upSjktNvUcHG4O1g1V1jVu9SivNkK20qt5/7GCl+XhFTYMktRmg2W3mn4/LYVeYw6Ywu10uh01hjfeddrucYeZxp8Mmp8OuMIddTrtNNQ0ef8h3sKpe1fUeebyGf+46qbLDny3C6VB8pFNup6NZABlmN28dNinMbpe9MXh02O1y+MJKu61ZSOl7vi+ULCyrUX7j/Hr1HqPNGuw2qXdchDJ6RiqjZ5QyEiL9++kJkYpyt4wwvF5D+w/V+ocV7ymt1jcHzXDNd7+m3qui8hoVlddow+6Drb53zyiXIt0OhYc5FOEyb91OuyKcDoU7HQpvsu92Ohr37c0eczubPN9p9w9dPnI4soO5Abssm2EYbf8En4DKy8sVGxursrIyxcTEdP4brPidtOJe6TuXSuc90vmvDwAALBHwNkQXM2bMGJ122mlavHix/9jQoUN1/vnna8GCBS3O//Wvf63XX39dW7du9R+bPXu2PvnkE61bt06SNHPmTJWXl+vf//63/5zvfe97io+P15IlS9qtiWsEIBQ1eLwqra5XXYNXTocZhIU13jrt9k5dbKGm3uMP8kqrDodr5v3GwM9/vPGc6np5vMGLBVwOu9ISIsyQrGdkY1Bm7veNj+z0YZeGYYaKZqjWetBW2U4PuM5mt6nFHH9O322T+f58jzsdNrnCHP6fGZvNN/DN/Nmx2cw932A4m2xNjh3++TKPtXyOzWaeb/bQNG9tTfbtjec0f7zJ+fbm59uavtcRdRx+z+a1qrVzjqjVFWbXmUOTA3JNOtqGoCdaZ/MP50yztg4AAIAAqaur04YNG3TzzTc3O56Tk6O1a9e2+px169YpJyen2bGzzjpLTzzxhOrr6+V0OrVu3TrNnTu3xTkLFy5s9TVra2tVW1vrv19eXv4tPg0ABFaYw67EHu6gvFe406GUWIdSYsM7/BzDMFRR26DSxt51vvnZvE0WYvA0LrjQdGGGBo/hX8ihockxb5N538xFHqSkGHMYZkZilFJiwoPaE8tmsymxh1uJPdwamRbX6ucvq65XYVmNqus9qqnzqKbBo5p6r6ob96vrPKpt8Kqm3uM/VlPvVXW9R7X1h/dr6j2Nx8xza+o95vDkI3rfeQ2ppt6rmvrAD+89kcRFOrXpNzntnxhAhGidLS5D6jta6jXY6koAAAACYv/+/fJ4PEpObv7b4OTkZBUVFbX6nKKiolbPb2ho0P79+5WamtrmOW295oIFC3TXXXcdxycBANhsNsWEOxUT7lR6z6493PbbsNnMlVjjIl0Bew/DMPxhWtM5AI+cE/Dw/bbPa/CaK80ahmT4X18y1PyYDEPGEY9JanbMd7K38fW8hm/f8O97DTXeN5o8riPuG/J6mx/zfe7mNZrHdETd/n1/jUd+FvNYj1aG8wab9RWcaCbdaG4AAAAnuCNXfzMM46grwrV2/pHHj+U1b7nlFuXm5vrvl5eXKy2N0QAAgNBis9nkDnPIHSYpOJ0SESCEaAAAADgmiYmJcjgcLXqIFRcXt+hJ5pOSktLq+WFhYerZs+dRz2nrNd1ut9xu/jcCAACCo3Nn7AMAAMAJz+VyKSsrS3l5ec2O5+Xlady4ca0+Jzs7u8X5b7/9tkaNGiWn03nUc9p6TQAAgGCiJxoAAACOWW5uri699FKNGjVK2dnZeuyxx5Sfn6/Zs2dLModa7tmzR88884wkcyXORx55RLm5ubr66qu1bt06PfHEE81W3bzhhhs0adIk/e53v9N5552n1157Te+8847WrFljyWcEAABoihANAAAAx2zmzJkqKSnR/PnzVVhYqOHDh2v58uXKyMiQJBUWFio/P99/fmZmppYvX665c+fqz3/+s3r37q2HH35YM2bM8J8zbtw4vfDCC7rtttt0++23a8CAAVq6dKnGjBkT9M8HAABwJJvRdGmEbqC8vFyxsbEqKytTTEyM1eUAAIAugjZE6OMaAQCAb6OjbQjmRAMAAAAAAADaQYgGAAAAAAAAtIMQDQAAAAAAAGgHIRoAAAAAAADQDkI0AAAAAAAAoB2EaAAAAAAAAEA7CNEAAAAAAACAdhCiAQAAAAAAAO0gRAMAAAAAAADaQYgGAAAAAAAAtIMQDQAAAAAAAGgHIRoAAAAAAADQDkI0AAAAAAAAoB1hVhcQbIZhSJLKy8strgQAAHQlvraDry2B0EM7DwAAfBsdbed1uxCtoqJCkpSWlmZxJQAAoCuqqKhQbGys1WWgFbTzAADA8WivnWczutmvU71er/bu3avo6GjZbLZOf/3y8nKlpaWpoKBAMTExnf76OH5co9DG9Ql9XKPQxvUJHMMwVFFRod69e8tuZ0aMUBTodp7EdyzUcX1CG9cn9HGNQhvXJ3A62s7rdj3R7Ha7+vbtG/D3iYmJ4Yc6xHGNQhvXJ/RxjUIb1ycw6IEW2oLVzpP4joU6rk9o4/qEPq5RaOP6BEZH2nn8GhUAAAAAAABoByEaAAAAAAAA0A5CtE7mdrt1xx13yO12W10K2sA1Cm1cn9DHNQptXB8gsPiOhTauT2jj+oQ+rlFo4/pYr9stLAAAAAAAAAAcK3qiAQAAAAAAAO0gRAMAAAAAAADaQYgGAAAAAAAAtIMQDQAAAAAAAGgHIVonW7RokTIzMxUeHq6srCytXr3a6pIg6c4775TNZmu2paSkWF1Wt7Zq1Sqde+656t27t2w2m1599dVmjxuGoTvvvFO9e/dWRESEvvvd72rLli3WFNsNtXd9Lr/88hbfqbFjx1pTbDe0YMECjR49WtHR0UpKStL555+vL7/8stk5fIeAzkc7L3TR1gsttPNCG+280EY7L7QRonWipUuXas6cOZo3b542btyoiRMnavr06crPz7e6NEgaNmyYCgsL/dvmzZutLqlbq6ys1MiRI/XII4+0+vj999+vhx56SI888og++ugjpaSkaNq0aaqoqAhypd1Te9dHkr73ve81+04tX748iBV2bytXrtS1116r9evXKy8vTw0NDcrJyVFlZaX/HL5DQOeinRf6aOuFDtp5oY12XmijnRfiDHSa008/3Zg9e3azYyeddJJx8803W1QRfO644w5j5MiRVpeBNkgy/vnPf/rve71eIyUlxbjvvvv8x2pqaozY2FjjL3/5iwUVdm9HXh/DMIxZs2YZ5513niX1oKXi4mJDkrFy5UrDMPgOAYFAOy+00dYLXbTzQhvtvNBHOy+00BOtk9TV1WnDhg3KyclpdjwnJ0dr1661qCo0tX37dvXu3VuZmZm6+OKLtXPnTqtLQht27dqloqKiZt8nt9utyZMn830KIStWrFBSUpIGDx6sq6++WsXFxVaX1G2VlZVJkhISEiTxHQI6G+28roG2XtfAv1FdA+280EE7L7QQonWS/fv3y+PxKDk5udnx5ORkFRUVWVQVfMaMGaNnnnlGb731lv7617+qqKhI48aNU0lJidWloRW+7wzfp9A1ffp0/f3vf9e7776rBx98UB999JHOOOMM1dbWWl1at2MYhnJzczVhwgQNHz5cEt8hoLPRzgt9tPW6Dv6NCn2080IH7bzQE2Z1AScam83W7L5hGC2OIfimT5/u3x8xYoSys7M1YMAA/e1vf1Nubq6FleFo+D6FrpkzZ/r3hw8frlGjRikjI0PLli3ThRdeaGFl3c91112nTz/9VGvWrGnxGN8hoHPxnQpdtPW6Hr5PoYt2XuignRd66InWSRITE+VwOFokv8XFxS0SYlgvKipKI0aM0Pbt260uBa3wrabF96nrSE1NVUZGBt+pILv++uv1+uuv67333lPfvn39x/kOAZ2Ldl7XQ1svdPFvVNdDO88atPNCEyFaJ3G5XMrKylJeXl6z43l5eRo3bpxFVaEttbW12rp1q1JTU60uBa3IzMxUSkpKs+9TXV2dVq5cyfcpRJWUlKigoIDvVJAYhqHrrrtOr7zyit59911lZmY2e5zvENC5aOd1PbT1Qhf/RnU9tPOCi3ZeaGM4ZyfKzc3VpZdeqlGjRik7O1uPPfaY8vPzNXv2bKtL6/ZuvPFGnXvuuUpPT1dxcbHuuecelZeXa9asWVaX1m0dOnRIX331lf/+rl27tGnTJiUkJCg9PV1z5szRvffeq0GDBmnQoEG69957FRkZqUsuucTCqruPo12fhIQE3XnnnZoxY4ZSU1P19ddf69Zbb1ViYqIuuOACC6vuPq699lo9//zzeu211xQdHe3/TWRsbKwiIiJks9n4DgGdjHZeaKOtF1po54U22nmhjXZeiLNqWdAT1Z///GcjIyPDcLlcxmmnneZfhhbWmjlzppGammo4nU6jd+/exoUXXmhs2bLF6rK6tffee8+Q1GKbNWuWYRjm0s133HGHkZKSYrjdbmPSpEnG5s2brS26Gzna9amqqjJycnKMXr16GU6n00hPTzdmzZpl5OfnW112t9HatZFkPPXUU/5z+A4BnY92XuiirRdaaOeFNtp5oY12XmizGYZhBD6qAwAAAAAAALou5kQDAAAAAAAA2kGIBgAAAAAAALSDEA0AAAAAAABoByEaAAAAAAAA0A5CNAAAAAAAAKAdhGgAAAAAAABAOwjRAAAAAAAAgHYQogEAAAAAAADtIEQDgOO0YsUK2Ww2lZaWWl0KAAAAOhHtPABNEaIBAAAAAAAA7SBEAwAAAAAAANpBiAagyzMMQ/fff7/69++viIgIjRw5Ui+99JKkw13wly1bppEjRyo8PFxjxozR5s2bm73Gyy+/rGHDhsntdqtfv3568MEHmz1eW1urm266SWlpaXK73Ro0aJCeeOKJZuds2LBBo0aNUmRkpMaNG6cvv/wysB8cAADgBEc7D0AoIUQD0OXddttteuqpp7R48WJt2bJFc+fO1U9+8hOtXLnSf86vfvUrPfDAA/roo4+UlJSkH/zgB6qvr5dkNoouuugiXXzxxdq8ebPuvPNO3X777Xr66af9z7/sssv0wgsv6OGHH9bWrVv1l7/8RT169GhWx7x58/Tggw/q448/VlhYmK644oqgfH4AAIATFe08AKHEZhiGYXURAPBtVVZWKjExUe+++66ys7P9x6+66ipVVVXpmmuu0ZQpU/TCCy9o5syZkqQDBw6ob9++evrpp3XRRRfpxz/+sfbt26e3337b//ybbrpJy5Yt05YtW7Rt2zYNGTJEeXl5mjp1aosaVqxYoSlTpuidd97RmWeeKUlavny5zjnnHFVXVys8PDzAfwoAAAAnHtp5AEINPdEAdGmff/65ampqNG3aNPXo0cO/PfPMM9qxY4f/vKYNr4SEBA0ZMkRbt26VJG3dulXjx49v9rrjx4/X9u3b5fF4tGnTJjkcDk2ePPmotZxyyin+/dTUVElScXHxcX9GAACA7oh2HoBQE2Z1AQBwPLxeryRp2bJl6tOnT7PH3G53swbWkWw2myRzrg3fvk/TTroREREdqsXpdLZ4bV99AAAAODa08wCEGnqiAejSTj75ZLndbuXn52vgwIHNtrS0NP9569ev9+8fPHhQ27Zt00knneR/jTVr1jR73bVr12rw4MFyOBwaMWKEvF5vs7k3AAAAEFi08wCEGnqiAejSoqOjdeONN2ru3Lnyer2aMGGCysvLtXbtWvXo0UMZGRmSpPnz56tnz55KTk7WvHnzlJiYqPPPP1+S9Mtf/lKjR4/W3XffrZkzZ2rdunV65JFHtGjRIklSv379NGvWLF1xxRV6+OGHNXLkSO3evVvFxcW66KKLrProAAAAJzTaeQBCDSEagC7v7rvvVlJSkhYsWKCdO3cqLi5Op512mm699VZ/N/v77rtPN9xwg7Zv366RI0fq9ddfl8vlkiSddtppevHFF/Wb3/xGd999t1JTUzV//nxdfvnl/vdYvHixbr31Vv385z9XSUmJ0tPTdeutt1rxcQEAALoN2nkAQgmrcwI4oflWVDp48KDi4uKsLgcAAACdhHYegGBjTjQAAAAAAACgHYRoAAAAAAAAQDsYzgkAAAAAAAC0g55oAAAAAAAAQDsI0QAAAAAAAIB2EKIBAAAAAAAA7SBEAwAAAAAAANpBiAYAAAAAAAC0gxANAAAAAAAAaAchGgAAAAAAANAOQjQAAAAAAACgHf8fgoUy3Nj7ADgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1500x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "emovo\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ravdess\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for s in res:\n",
    "    print(s)\n",
    "    plot_stats([res[s][0]])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6bb1c78c",
   "metadata": {},
   "source": [
    "# Algorithm hyperparameter"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7740df92",
   "metadata": {},
   "source": [
    "Adjust algorithm training hyperparameter plus minor parameter like dropout"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "id": "2ec1275e",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_m2(hp):\n",
    "    # Define hyper model architecture\n",
    "    m = Sequential([\n",
    "        InputLayer(input_shape=(261,128,1)),\n",
    "        Reshape((9,29,128,1)),\n",
    "        TimeDistributed(Conv2D(16, kernel_size=(5,5), activation='relu')),\n",
    "        TimeDistributed(BatchNormalization()),\n",
    "        TimeDistributed(MaxPooling2D(pool_size=(4,4), strides=2)),\n",
    "\n",
    "        TimeDistributed(Conv2D(32, kernel_size=(3,3), activation='relu')),\n",
    "        TimeDistributed(MaxPooling2D(pool_size=(2,2), strides=2)),\n",
    "\n",
    "        TimeDistributed(Conv2D(64, kernel_size=(3,3), activation='relu')),\n",
    "        TimeDistributed(MaxPooling2D(pool_size=(2,2), strides=1)),\n",
    "        TimeDistributed(Flatten()),\n",
    "        \n",
    "        # Tune dropout layer with values in 0 - 0.6 with stepsize of 0.3\n",
    "        Dropout(hp.Float(\"dropout\", min_value=0, max_value=0.6, step=0.3)),\n",
    "        \n",
    "        # Tune dropout probability (values in 0 - 0.4 with stepsize of 0.2)\n",
    "        Bidirectional(LSTM(256,\n",
    "                           dropout=hp.Float(\"lstm_dropout\", min_value=0, max_value=0.4, step=0.2),\n",
    "                           return_sequences=False)),\n",
    "        Dense(5, activation='softmax')\n",
    "    ])\n",
    "    \n",
    "    # Tune learning rate for Adam optimizer with values from 0.01, 0.001, or 0.0001\n",
    "    hp_learning_rate = hp.Choice(\"learning_rate\", values=[1e-2, 1e-3, 1e-4])\n",
    "    \n",
    "    # Define optimizer, loss, and metrics\n",
    "    m.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=hp_learning_rate),\n",
    "              loss='categorical_crossentropy',\n",
    "              metrics=[\"accuracy\"])\n",
    "    return m"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "id": "cf23690f",
   "metadata": {},
   "outputs": [],
   "source": [
    "class MyTuner(kt.tuners.Hyperband):\n",
    "      def run_trial(self, trial, *args, **kwargs):\n",
    "        # You can add additional HyperParameters for preprocessing and custom training loops via overriding `run_trial`\n",
    "        kwargs['batch_size'] = trial.hyperparameters.Choice(\"batch_size\", values=[8, 16, 32, 64, 128])\n",
    "        return super(MyTuner, self).run_trial(trial, *args, **kwargs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "id": "57687c08",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trial 84 Complete [00h 06m 33s]\n",
      "val_loss: 0.2943439881006877\n",
      "\n",
      "Best val_loss So Far: 0.24255329370498657\n",
      "Total elapsed time: 01h 56m 43s\n",
      "INFO:tensorflow:Oracle triggered exit\n"
     ]
    }
   ],
   "source": [
    "# instantiate the tuner\n",
    "tuner1 = MyTuner(get_m2,\n",
    "                 objective=\"val_loss\",\n",
    "                 max_epochs=120,\n",
    "                 # number of models that should be built and fit for each trial with the same hyperparameter values\n",
    "                 executions_per_trial=3,\n",
    "                 # the reduction factor for the number of epochs and number of models for each bracket\n",
    "                 factor=15,\n",
    "                 # the number of times to iterate over the full Hyperband algorithm\n",
    "                 hyperband_iterations=3,\n",
    "                 overwrite=True,\n",
    "                 directory=\"/datasets/nicolas_facchinetti/kt_dir2\",\n",
    "                 project_name=\"cnnlstm2\")\n",
    "tuner1.search_space_summary()\n",
    "\n",
    "earlyStopping = EarlyStopping(monitor='val_loss', patience=10, verbose=1, mode='min', restore_best_weights=True)\n",
    "reduce_lr_loss = ReduceLROnPlateau(monitor='val_loss', factor=0.1, patience=6, verbose=1, min_delta=1e-4, mode='min')\n",
    "\n",
    "tuner1.search(train['emodb']['x'], train['emodb']['y'],\n",
    "             epochs=100,\n",
    "             validation_split=0.2,\n",
    "             callbacks=[earlyStopping, reduce_lr_loss],\n",
    "             verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 233,
   "id": "3ef8d689",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Results summary\n",
      "Results in /datasets/nicolas_facchinetti/kt_dir2/cnnlstm2\n",
      "Showing 10 best trials\n",
      "<keras_tuner.engine.objective.Objective object at 0x7f6e440fa5e0>\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.3\n",
      "lstm_dropout: 0.2\n",
      "learning_rate: 0.001\n",
      "batch_size: 32\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 8\n",
      "tuner/bracket: 1\n",
      "tuner/round: 1\n",
      "tuner/trial_id: 0050\n",
      "lstm_units: 2\n",
      "Score: 0.24255329370498657\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.3\n",
      "lstm_dropout: 0.2\n",
      "learning_rate: 0.001\n",
      "batch_size: 16\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 8\n",
      "tuner/bracket: 1\n",
      "tuner/round: 1\n",
      "tuner/trial_id: 0058\n",
      "Score: 0.24965617060661316\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.3\n",
      "lstm_dropout: 0.4\n",
      "learning_rate: 0.001\n",
      "batch_size: 16\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 0\n",
      "tuner/round: 0\n",
      "Score: 0.2517334371805191\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.3\n",
      "lstm_dropout: 0.4\n",
      "learning_rate: 0.001\n",
      "batch_size: 32\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 8\n",
      "tuner/bracket: 1\n",
      "tuner/round: 1\n",
      "tuner/trial_id: 0073\n",
      "Score: 0.25494201978047687\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.6\n",
      "lstm_dropout: 0.0\n",
      "learning_rate: 0.001\n",
      "batch_size: 64\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 0\n",
      "tuner/round: 0\n",
      "Score: 0.27260248859723407\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.0\n",
      "lstm_dropout: 0.0\n",
      "learning_rate: 0.001\n",
      "batch_size: 16\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 8\n",
      "tuner/bracket: 1\n",
      "tuner/round: 1\n",
      "tuner/trial_id: 0018\n",
      "Score: 0.28204230467478436\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.0\n",
      "lstm_dropout: 0.2\n",
      "learning_rate: 0.001\n",
      "batch_size: 16\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 0\n",
      "tuner/round: 0\n",
      "Score: 0.2833857238292694\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.3\n",
      "lstm_dropout: 0.4\n",
      "learning_rate: 0.001\n",
      "batch_size: 32\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.2875823477904002\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.3\n",
      "lstm_dropout: 0.2\n",
      "learning_rate: 0.001\n",
      "batch_size: 32\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.28841229279836017\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.0\n",
      "lstm_dropout: 0.4\n",
      "learning_rate: 0.001\n",
      "batch_size: 8\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 8\n",
      "tuner/bracket: 1\n",
      "tuner/round: 1\n",
      "tuner/trial_id: 0047\n",
      "Score: 0.2936298648516337\n"
     ]
    }
   ],
   "source": [
    "tuner1.results_summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "id": "eb4b33a0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trial 84 Complete [00h 02m 01s]\n",
      "val_loss: 1.5282885630925496\n",
      "\n",
      "Best val_loss So Far: 0.36417991916338605\n",
      "Total elapsed time: 02h 22m 06s\n",
      "INFO:tensorflow:Oracle triggered exit\n"
     ]
    }
   ],
   "source": [
    "tuner2 = MyTuner(get_m2,\n",
    "                 objective=\"val_loss\",\n",
    "                 max_epochs=120,\n",
    "                 # number of models that should be built and fit for each trial with the same hyperparameter values\n",
    "                 executions_per_trial=3,\n",
    "                 # the reduction factor for the number of epochs and number of models for each bracket\n",
    "                 factor=15,\n",
    "                 # the number of times to iterate over the full Hyperband algorithm\n",
    "                 hyperband_iterations=3,\n",
    "                 overwrite=True,\n",
    "                 directory=\"/datasets/nicolas_facchinetti/kt_dir3\",\n",
    "                 project_name=\"cnnlstm3\")\n",
    "tuner2.search_space_summary()\n",
    "\n",
    "tuner2.search(train['emovo']['x'], train['emovo']['y'],\n",
    "             epochs=100,\n",
    "             validation_split=0.2,\n",
    "             callbacks=[earlyStopping, reduce_lr_loss],\n",
    "             verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "id": "3ea26673",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Results summary\n",
      "Results in /datasets/nicolas_facchinetti/kt_dir3/cnnlstm3\n",
      "Showing 10 best trials\n",
      "<keras_tuner.engine.objective.Objective object at 0x7f6f10eb15b0>\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.3\n",
      "lstm_dropout: 0.0\n",
      "learning_rate: 0.0001\n",
      "batch_size: 8\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 8\n",
      "tuner/bracket: 1\n",
      "tuner/round: 1\n",
      "tuner/trial_id: 0001\n",
      "Score: 0.36417991916338605\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.0\n",
      "lstm_dropout: 0.2\n",
      "learning_rate: 0.0001\n",
      "batch_size: 8\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 8\n",
      "tuner/bracket: 1\n",
      "tuner/round: 1\n",
      "tuner/trial_id: 0057\n",
      "Score: 0.3771323362986247\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.3\n",
      "lstm_dropout: 0.4\n",
      "learning_rate: 0.0001\n",
      "batch_size: 16\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 0\n",
      "tuner/round: 0\n",
      "Score: 0.3842690885066986\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.3\n",
      "lstm_dropout: 0.2\n",
      "learning_rate: 0.0001\n",
      "batch_size: 32\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 0\n",
      "tuner/round: 0\n",
      "Score: 0.3907069265842438\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.0\n",
      "lstm_dropout: 0.0\n",
      "learning_rate: 0.0001\n",
      "batch_size: 8\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 8\n",
      "tuner/bracket: 1\n",
      "tuner/round: 1\n",
      "tuner/trial_id: 0003\n",
      "Score: 0.3994199534257253\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.6\n",
      "lstm_dropout: 0.2\n",
      "learning_rate: 0.001\n",
      "batch_size: 128\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 0\n",
      "tuner/round: 0\n",
      "Score: 0.4018554389476776\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.6\n",
      "lstm_dropout: 0.2\n",
      "learning_rate: 0.0001\n",
      "batch_size: 64\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 0\n",
      "tuner/round: 0\n",
      "Score: 0.4111494521299998\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.0\n",
      "lstm_dropout: 0.2\n",
      "learning_rate: 0.0001\n",
      "batch_size: 8\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.431589017311732\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.3\n",
      "lstm_dropout: 0.0\n",
      "learning_rate: 0.001\n",
      "batch_size: 32\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 8\n",
      "tuner/bracket: 1\n",
      "tuner/round: 1\n",
      "tuner/trial_id: 0036\n",
      "Score: 0.4319891035556793\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.6\n",
      "lstm_dropout: 0.4\n",
      "learning_rate: 0.0001\n",
      "batch_size: 128\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 0\n",
      "tuner/round: 0\n",
      "Score: 0.43790172537167865\n"
     ]
    }
   ],
   "source": [
    "tuner2.results_summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "id": "cf314325",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trial 84 Complete [00h 03m 28s]\n",
      "val_loss: 0.3047107756137848\n",
      "\n",
      "Best val_loss So Far: 0.24022861818472543\n",
      "Total elapsed time: 03h 52m 46s\n",
      "INFO:tensorflow:Oracle triggered exit\n"
     ]
    }
   ],
   "source": [
    "tuner3 = MyTuner(get_m2,\n",
    "                 objective=\"val_loss\",\n",
    "                 max_epochs=120,\n",
    "                 # number of models that should be built and fit for each trial with the same hyperparameter values\n",
    "                 executions_per_trial=3,\n",
    "                 # the reduction factor for the number of epochs and number of models for each bracket\n",
    "                 factor=15,\n",
    "                 # the number of times to iterate over the full Hyperband algorithm\n",
    "                 hyperband_iterations=3,\n",
    "                 overwrite=True,\n",
    "                 directory=\"/datasets/nicolas_facchinetti/kt_dir4\",\n",
    "                 project_name=\"cnnlstm4\")\n",
    "tuner3.search_space_summary()\n",
    "\n",
    "tuner3.search(train['ravdess']['x'], train['ravdess']['y'],\n",
    "             epochs=100,\n",
    "             validation_split=0.2,\n",
    "             callbacks=[earlyStopping, reduce_lr_loss],\n",
    "             verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "id": "1ab49211",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Results summary\n",
      "Results in /datasets/nicolas_facchinetti/kt_dir4/cnnlstm4\n",
      "Showing 10 best trials\n",
      "<keras_tuner.engine.objective.Objective object at 0x7f6e4542bb50>\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.6\n",
      "lstm_dropout: 0.2\n",
      "learning_rate: 0.001\n",
      "batch_size: 128\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 0\n",
      "tuner/round: 0\n",
      "Score: 0.24022861818472543\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.3\n",
      "lstm_dropout: 0.0\n",
      "learning_rate: 0.001\n",
      "batch_size: 64\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 8\n",
      "tuner/bracket: 1\n",
      "tuner/round: 1\n",
      "tuner/trial_id: 0039\n",
      "Score: 0.27419041593869525\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.0\n",
      "lstm_dropout: 0.4\n",
      "learning_rate: 0.001\n",
      "batch_size: 64\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 8\n",
      "tuner/bracket: 1\n",
      "tuner/round: 1\n",
      "tuner/trial_id: 0066\n",
      "Score: 0.27864599227905273\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.3\n",
      "lstm_dropout: 0.2\n",
      "learning_rate: 0.001\n",
      "batch_size: 64\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 8\n",
      "tuner/bracket: 1\n",
      "tuner/round: 1\n",
      "tuner/trial_id: 0011\n",
      "Score: 0.2835317353407542\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.0\n",
      "lstm_dropout: 0.4\n",
      "learning_rate: 0.0001\n",
      "batch_size: 64\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 0\n",
      "tuner/round: 0\n",
      "Score: 0.2998693684736888\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.6\n",
      "lstm_dropout: 0.2\n",
      "learning_rate: 0.0001\n",
      "batch_size: 64\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 0\n",
      "tuner/round: 0\n",
      "Score: 0.30245624979337055\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.0\n",
      "lstm_dropout: 0.2\n",
      "learning_rate: 0.001\n",
      "batch_size: 128\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 0\n",
      "tuner/round: 0\n",
      "Score: 0.3047107756137848\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.0\n",
      "lstm_dropout: 0.2\n",
      "learning_rate: 0.001\n",
      "batch_size: 32\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 8\n",
      "tuner/bracket: 1\n",
      "tuner/round: 1\n",
      "tuner/trial_id: 0075\n",
      "Score: 0.30553631981213886\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.3\n",
      "lstm_dropout: 0.2\n",
      "learning_rate: 0.001\n",
      "batch_size: 32\n",
      "tuner/epochs: 120\n",
      "tuner/initial_epoch: 8\n",
      "tuner/bracket: 1\n",
      "tuner/round: 1\n",
      "tuner/trial_id: 0040\n",
      "Score: 0.30712713797887164\n",
      "Trial summary\n",
      "Hyperparameters:\n",
      "dropout: 0.0\n",
      "lstm_dropout: 0.2\n",
      "learning_rate: 0.001\n",
      "batch_size: 32\n",
      "tuner/epochs: 8\n",
      "tuner/initial_epoch: 0\n",
      "tuner/bracket: 1\n",
      "tuner/round: 0\n",
      "Score: 0.3135736385981242\n"
     ]
    }
   ],
   "source": [
    "tuner3.results_summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 184,
   "id": "7a902da6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get the optimal hyperparameters from the results\n",
    "param_found = {'emodb': tuner1,\n",
    "               'emovo': tuner2,\n",
    "               'ravdess': tuner3}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "id": "3913c074",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "For emodb the best found paramters are:\n",
      "{'dropout': 0.3, 'lstm_dropout': 0.2, 'learning_rate': 0.001, 'batch_size': 32, 'tuner/epochs': 120, 'tuner/initial_epoch': 8, 'tuner/bracket': 1, 'tuner/round': 1, 'tuner/trial_id': '0050', 'lstm_units': 2}\n",
      "For emovo the best found paramters are:\n",
      "{'dropout': 0.3, 'lstm_dropout': 0.0, 'learning_rate': 0.0001, 'batch_size': 8, 'tuner/epochs': 120, 'tuner/initial_epoch': 8, 'tuner/bracket': 1, 'tuner/round': 1, 'tuner/trial_id': '0001', 'lstm_units': 2}\n",
      "For ravdess the best found paramters are:\n",
      "{'dropout': 0.6, 'lstm_dropout': 0.2, 'learning_rate': 0.001, 'batch_size': 128, 'tuner/epochs': 120, 'tuner/initial_epoch': 0, 'tuner/bracket': 0, 'tuner/round': 0, 'lstm_units': 2}\n"
     ]
    }
   ],
   "source": [
    "for d in param_found:\n",
    "    print(f'For {d} the best found paramters are:')\n",
    "    print(param_found[d].get_best_hyperparameters()[0].get_config()[\"values\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ae4be72b",
   "metadata": {},
   "source": [
    "### Test hyperparameter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "9013c007",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "emodb\n",
      "Model: \"sequential_8\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "reshape_8 (Reshape)          (None, 9, 29, 128, 1)     0         \n",
      "_________________________________________________________________\n",
      "time_distributed_64 (TimeDis (None, 9, 25, 124, 16)    416       \n",
      "_________________________________________________________________\n",
      "time_distributed_65 (TimeDis (None, 9, 25, 124, 16)    64        \n",
      "_________________________________________________________________\n",
      "time_distributed_66 (TimeDis (None, 9, 11, 61, 16)     0         \n",
      "_________________________________________________________________\n",
      "time_distributed_67 (TimeDis (None, 9, 9, 59, 32)      4640      \n",
      "_________________________________________________________________\n",
      "time_distributed_68 (TimeDis (None, 9, 4, 29, 32)      0         \n",
      "_________________________________________________________________\n",
      "time_distributed_69 (TimeDis (None, 9, 2, 27, 64)      18496     \n",
      "_________________________________________________________________\n",
      "time_distributed_70 (TimeDis (None, 9, 1, 26, 64)      0         \n",
      "_________________________________________________________________\n",
      "time_distributed_71 (TimeDis (None, 9, 1664)           0         \n",
      "_________________________________________________________________\n",
      "dropout_3 (Dropout)          (None, 9, 1664)           0         \n",
      "_________________________________________________________________\n",
      "bidirectional_8 (Bidirection (None, 512)               3934208   \n",
      "_________________________________________________________________\n",
      "dense_8 (Dense)              (None, 5)                 2565      \n",
      "=================================================================\n",
      "Total params: 3,960,389\n",
      "Trainable params: 3,960,357\n",
      "Non-trainable params: 32\n",
      "_________________________________________________________________\n",
      "Epoch 1/100\n",
      "84/84 - 5s - loss: 1.1908 - accuracy: 0.5196 - val_loss: 1.6160 - val_accuracy: 0.3109\n",
      "Epoch 2/100\n",
      "84/84 - 2s - loss: 0.7037 - accuracy: 0.7211 - val_loss: 1.0701 - val_accuracy: 0.5874\n",
      "Epoch 3/100\n",
      "84/84 - 1s - loss: 0.5101 - accuracy: 0.8030 - val_loss: 0.6846 - val_accuracy: 0.7414\n",
      "Epoch 4/100\n",
      "84/84 - 2s - loss: 0.3469 - accuracy: 0.8699 - val_loss: 0.5254 - val_accuracy: 0.7937\n",
      "Epoch 5/100\n",
      "84/84 - 2s - loss: 0.2378 - accuracy: 0.9140 - val_loss: 0.3462 - val_accuracy: 0.8655\n",
      "Epoch 6/100\n",
      "84/84 - 2s - loss: 0.1789 - accuracy: 0.9413 - val_loss: 0.2812 - val_accuracy: 0.8939\n",
      "Epoch 7/100\n",
      "84/84 - 2s - loss: 0.1221 - accuracy: 0.9600 - val_loss: 0.2745 - val_accuracy: 0.8984\n",
      "Epoch 8/100\n",
      "84/84 - 2s - loss: 0.0926 - accuracy: 0.9742 - val_loss: 0.2629 - val_accuracy: 0.8984\n",
      "Epoch 9/100\n",
      "84/84 - 2s - loss: 0.0624 - accuracy: 0.9839 - val_loss: 0.2612 - val_accuracy: 0.9193\n",
      "Epoch 10/100\n",
      "84/84 - 2s - loss: 0.0523 - accuracy: 0.9836 - val_loss: 0.3393 - val_accuracy: 0.8894\n",
      "Epoch 11/100\n",
      "84/84 - 2s - loss: 0.0759 - accuracy: 0.9761 - val_loss: 0.2602 - val_accuracy: 0.9133\n",
      "Epoch 12/100\n",
      "84/84 - 2s - loss: 0.0425 - accuracy: 0.9899 - val_loss: 0.2992 - val_accuracy: 0.8999\n",
      "Epoch 13/100\n",
      "84/84 - 2s - loss: 0.0298 - accuracy: 0.9933 - val_loss: 0.2652 - val_accuracy: 0.9088\n",
      "Epoch 14/100\n",
      "84/84 - 2s - loss: 0.0222 - accuracy: 0.9951 - val_loss: 0.2598 - val_accuracy: 0.9013\n",
      "Epoch 15/100\n",
      "84/84 - 2s - loss: 0.0195 - accuracy: 0.9955 - val_loss: 0.2554 - val_accuracy: 0.9193\n",
      "Epoch 16/100\n",
      "84/84 - 2s - loss: 0.0179 - accuracy: 0.9959 - val_loss: 0.2678 - val_accuracy: 0.9193\n",
      "Epoch 17/100\n",
      "84/84 - 2s - loss: 0.0180 - accuracy: 0.9959 - val_loss: 0.2864 - val_accuracy: 0.8984\n",
      "Epoch 18/100\n",
      "84/84 - 2s - loss: 0.0146 - accuracy: 0.9966 - val_loss: 0.2602 - val_accuracy: 0.9148\n",
      "Epoch 19/100\n",
      "84/84 - 2s - loss: 0.0072 - accuracy: 0.9996 - val_loss: 0.2690 - val_accuracy: 0.9148\n",
      "Epoch 20/100\n",
      "84/84 - 2s - loss: 0.0106 - accuracy: 0.9989 - val_loss: 0.2760 - val_accuracy: 0.9103\n",
      "Epoch 21/100\n",
      "84/84 - 2s - loss: 0.0161 - accuracy: 0.9959 - val_loss: 0.2620 - val_accuracy: 0.9133\n",
      "\n",
      "Epoch 00021: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.\n",
      "Epoch 22/100\n",
      "84/84 - 2s - loss: 0.0079 - accuracy: 0.9985 - val_loss: 0.2466 - val_accuracy: 0.9208\n",
      "Epoch 23/100\n",
      "84/84 - 2s - loss: 0.0042 - accuracy: 0.9993 - val_loss: 0.2434 - val_accuracy: 0.9238\n",
      "Epoch 24/100\n",
      "84/84 - 2s - loss: 0.0040 - accuracy: 0.9996 - val_loss: 0.2451 - val_accuracy: 0.9253\n",
      "Epoch 25/100\n",
      "84/84 - 2s - loss: 0.0038 - accuracy: 0.9993 - val_loss: 0.2495 - val_accuracy: 0.9253\n",
      "Epoch 26/100\n",
      "84/84 - 2s - loss: 0.0030 - accuracy: 0.9996 - val_loss: 0.2461 - val_accuracy: 0.9223\n",
      "Epoch 27/100\n",
      "84/84 - 2s - loss: 0.0032 - accuracy: 0.9993 - val_loss: 0.2502 - val_accuracy: 0.9238\n",
      "Epoch 28/100\n",
      "84/84 - 2s - loss: 0.0030 - accuracy: 0.9996 - val_loss: 0.2452 - val_accuracy: 0.9238\n",
      "Epoch 29/100\n",
      "84/84 - 2s - loss: 0.0026 - accuracy: 0.9996 - val_loss: 0.2455 - val_accuracy: 0.9238\n",
      "\n",
      "Epoch 00029: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.\n",
      "Epoch 30/100\n",
      "84/84 - 2s - loss: 0.0031 - accuracy: 0.9993 - val_loss: 0.2459 - val_accuracy: 0.9238\n",
      "Epoch 31/100\n",
      "84/84 - 2s - loss: 0.0021 - accuracy: 1.0000 - val_loss: 0.2455 - val_accuracy: 0.9238\n",
      "Epoch 32/100\n",
      "84/84 - 2s - loss: 0.0025 - accuracy: 0.9996 - val_loss: 0.2448 - val_accuracy: 0.9238\n",
      "Epoch 33/100\n",
      "84/84 - 2s - loss: 0.0025 - accuracy: 0.9993 - val_loss: 0.2445 - val_accuracy: 0.9238\n",
      "Restoring model weights from the end of the best epoch.\n",
      "Epoch 00033: early stopping\n",
      "27/27 [==============================] - 0s 7ms/step - loss: 0.3126 - accuracy: 0.9068\n",
      "Test loss, test acc: [0.3126097321510315, 0.9068100452423096]\n",
      "emovo\n",
      "Model: \"sequential_9\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "reshape_9 (Reshape)          (None, 9, 29, 128, 1)     0         \n",
      "_________________________________________________________________\n",
      "time_distributed_72 (TimeDis (None, 9, 25, 124, 16)    416       \n",
      "_________________________________________________________________\n",
      "time_distributed_73 (TimeDis (None, 9, 25, 124, 16)    64        \n",
      "_________________________________________________________________\n",
      "time_distributed_74 (TimeDis (None, 9, 11, 61, 16)     0         \n",
      "_________________________________________________________________\n",
      "time_distributed_75 (TimeDis (None, 9, 9, 59, 32)      4640      \n",
      "_________________________________________________________________\n",
      "time_distributed_76 (TimeDis (None, 9, 4, 29, 32)      0         \n",
      "_________________________________________________________________\n",
      "time_distributed_77 (TimeDis (None, 9, 2, 27, 64)      18496     \n",
      "_________________________________________________________________\n",
      "time_distributed_78 (TimeDis (None, 9, 1, 26, 64)      0         \n",
      "_________________________________________________________________\n",
      "time_distributed_79 (TimeDis (None, 9, 1664)           0         \n",
      "_________________________________________________________________\n",
      "dropout_4 (Dropout)          (None, 9, 1664)           0         \n",
      "_________________________________________________________________\n",
      "bidirectional_9 (Bidirection (None, 512)               3934208   \n",
      "_________________________________________________________________\n",
      "dense_9 (Dense)              (None, 5)                 2565      \n",
      "=================================================================\n",
      "Total params: 3,960,389\n",
      "Trainable params: 3,960,357\n",
      "Non-trainable params: 32\n",
      "_________________________________________________________________\n",
      "Epoch 1/100\n",
      "395/395 - 9s - loss: 1.4177 - accuracy: 0.3886 - val_loss: 1.2394 - val_accuracy: 0.5120\n",
      "Epoch 2/100\n",
      "395/395 - 4s - loss: 1.1144 - accuracy: 0.5563 - val_loss: 1.0720 - val_accuracy: 0.5716\n",
      "Epoch 3/100\n",
      "395/395 - 4s - loss: 0.8778 - accuracy: 0.6662 - val_loss: 0.8790 - val_accuracy: 0.6679\n",
      "Epoch 4/100\n",
      "395/395 - 4s - loss: 0.6804 - accuracy: 0.7471 - val_loss: 0.6753 - val_accuracy: 0.7529\n",
      "Epoch 5/100\n",
      "395/395 - 4s - loss: 0.5301 - accuracy: 0.8165 - val_loss: 0.6457 - val_accuracy: 0.7630\n",
      "Epoch 6/100\n",
      "395/395 - 4s - loss: 0.4072 - accuracy: 0.8653 - val_loss: 0.6515 - val_accuracy: 0.7364\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 7/100\n",
      "395/395 - 4s - loss: 0.2982 - accuracy: 0.9071 - val_loss: 0.4948 - val_accuracy: 0.8251\n",
      "Epoch 8/100\n",
      "395/395 - 4s - loss: 0.2230 - accuracy: 0.9407 - val_loss: 0.4449 - val_accuracy: 0.8555\n",
      "Epoch 9/100\n",
      "395/395 - 4s - loss: 0.1716 - accuracy: 0.9525 - val_loss: 0.4937 - val_accuracy: 0.8251\n",
      "Epoch 10/100\n",
      "395/395 - 4s - loss: 0.1390 - accuracy: 0.9677 - val_loss: 0.4403 - val_accuracy: 0.8454\n",
      "Epoch 11/100\n",
      "395/395 - 4s - loss: 0.0982 - accuracy: 0.9762 - val_loss: 0.4638 - val_accuracy: 0.8466\n",
      "Epoch 12/100\n",
      "395/395 - 4s - loss: 0.0763 - accuracy: 0.9848 - val_loss: 0.4343 - val_accuracy: 0.8492\n",
      "Epoch 13/100\n",
      "395/395 - 4s - loss: 0.0616 - accuracy: 0.9880 - val_loss: 0.4145 - val_accuracy: 0.8707\n",
      "Epoch 14/100\n",
      "395/395 - 4s - loss: 0.0492 - accuracy: 0.9889 - val_loss: 0.3887 - val_accuracy: 0.8669\n",
      "Epoch 15/100\n",
      "395/395 - 4s - loss: 0.0530 - accuracy: 0.9880 - val_loss: 0.3884 - val_accuracy: 0.8745\n",
      "Epoch 16/100\n",
      "395/395 - 4s - loss: 0.0393 - accuracy: 0.9911 - val_loss: 0.4481 - val_accuracy: 0.8568\n",
      "Epoch 17/100\n",
      "395/395 - 4s - loss: 0.0447 - accuracy: 0.9905 - val_loss: 0.4119 - val_accuracy: 0.8783\n",
      "Epoch 18/100\n",
      "395/395 - 4s - loss: 0.0335 - accuracy: 0.9946 - val_loss: 0.4048 - val_accuracy: 0.8872\n",
      "Epoch 19/100\n",
      "395/395 - 4s - loss: 0.0274 - accuracy: 0.9930 - val_loss: 0.4222 - val_accuracy: 0.8771\n",
      "Epoch 20/100\n",
      "395/395 - 4s - loss: 0.0291 - accuracy: 0.9924 - val_loss: 0.4262 - val_accuracy: 0.8745\n",
      "Epoch 21/100\n",
      "395/395 - 4s - loss: 0.0186 - accuracy: 0.9962 - val_loss: 0.4051 - val_accuracy: 0.8809\n",
      "\n",
      "Epoch 00021: ReduceLROnPlateau reducing learning rate to 9.999999747378752e-06.\n",
      "Epoch 22/100\n",
      "395/395 - 4s - loss: 0.0138 - accuracy: 0.9952 - val_loss: 0.3883 - val_accuracy: 0.8923\n",
      "Epoch 23/100\n",
      "395/395 - 4s - loss: 0.0105 - accuracy: 0.9975 - val_loss: 0.3913 - val_accuracy: 0.8885\n",
      "Epoch 24/100\n",
      "395/395 - 4s - loss: 0.0108 - accuracy: 0.9968 - val_loss: 0.3891 - val_accuracy: 0.8885\n",
      "Epoch 25/100\n",
      "395/395 - 4s - loss: 0.0096 - accuracy: 0.9978 - val_loss: 0.3854 - val_accuracy: 0.8923\n",
      "Epoch 26/100\n",
      "395/395 - 4s - loss: 0.0087 - accuracy: 0.9978 - val_loss: 0.3874 - val_accuracy: 0.8897\n",
      "Epoch 27/100\n",
      "395/395 - 4s - loss: 0.0090 - accuracy: 0.9978 - val_loss: 0.3867 - val_accuracy: 0.8885\n",
      "Epoch 28/100\n",
      "395/395 - 4s - loss: 0.0094 - accuracy: 0.9981 - val_loss: 0.3885 - val_accuracy: 0.8897\n",
      "Epoch 29/100\n",
      "395/395 - 4s - loss: 0.0080 - accuracy: 0.9984 - val_loss: 0.3942 - val_accuracy: 0.8948\n",
      "Epoch 30/100\n",
      "395/395 - 4s - loss: 0.0081 - accuracy: 0.9971 - val_loss: 0.3890 - val_accuracy: 0.8986\n",
      "Epoch 31/100\n",
      "395/395 - 4s - loss: 0.0082 - accuracy: 0.9975 - val_loss: 0.3910 - val_accuracy: 0.8986\n",
      "\n",
      "Epoch 00031: ReduceLROnPlateau reducing learning rate to 9.999999747378752e-07.\n",
      "Epoch 32/100\n",
      "395/395 - 4s - loss: 0.0071 - accuracy: 0.9978 - val_loss: 0.3907 - val_accuracy: 0.8986\n",
      "Epoch 33/100\n",
      "395/395 - 4s - loss: 0.0073 - accuracy: 0.9978 - val_loss: 0.3923 - val_accuracy: 0.8961\n",
      "Epoch 34/100\n",
      "395/395 - 4s - loss: 0.0068 - accuracy: 0.9981 - val_loss: 0.3919 - val_accuracy: 0.8973\n",
      "Epoch 35/100\n",
      "395/395 - 4s - loss: 0.0071 - accuracy: 0.9981 - val_loss: 0.3933 - val_accuracy: 0.8935\n",
      "Restoring model weights from the end of the best epoch.\n",
      "Epoch 00035: early stopping\n",
      "124/124 [==============================] - 1s 6ms/step - loss: 0.3450 - accuracy: 0.8956\n",
      "Test loss, test acc: [0.34495365619659424, 0.8956433534622192]\n",
      "ravdess\n",
      "Model: \"sequential_10\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "reshape_10 (Reshape)         (None, 9, 29, 128, 1)     0         \n",
      "_________________________________________________________________\n",
      "time_distributed_80 (TimeDis (None, 9, 25, 124, 16)    416       \n",
      "_________________________________________________________________\n",
      "time_distributed_81 (TimeDis (None, 9, 25, 124, 16)    64        \n",
      "_________________________________________________________________\n",
      "time_distributed_82 (TimeDis (None, 9, 11, 61, 16)     0         \n",
      "_________________________________________________________________\n",
      "time_distributed_83 (TimeDis (None, 9, 9, 59, 32)      4640      \n",
      "_________________________________________________________________\n",
      "time_distributed_84 (TimeDis (None, 9, 4, 29, 32)      0         \n",
      "_________________________________________________________________\n",
      "time_distributed_85 (TimeDis (None, 9, 2, 27, 64)      18496     \n",
      "_________________________________________________________________\n",
      "time_distributed_86 (TimeDis (None, 9, 1, 26, 64)      0         \n",
      "_________________________________________________________________\n",
      "time_distributed_87 (TimeDis (None, 9, 1664)           0         \n",
      "_________________________________________________________________\n",
      "dropout_5 (Dropout)          (None, 9, 1664)           0         \n",
      "_________________________________________________________________\n",
      "bidirectional_10 (Bidirectio (None, 512)               3934208   \n",
      "_________________________________________________________________\n",
      "dense_10 (Dense)             (None, 5)                 2565      \n",
      "=================================================================\n",
      "Total params: 3,960,389\n",
      "Trainable params: 3,960,357\n",
      "Non-trainable params: 32\n",
      "_________________________________________________________________\n",
      "Epoch 1/100\n",
      "45/45 - 6s - loss: 1.5216 - accuracy: 0.3471 - val_loss: 1.5997 - val_accuracy: 0.2512\n",
      "Epoch 2/100\n",
      "45/45 - 2s - loss: 1.3066 - accuracy: 0.4591 - val_loss: 1.5804 - val_accuracy: 0.3475\n",
      "Epoch 3/100\n",
      "45/45 - 2s - loss: 1.1348 - accuracy: 0.5414 - val_loss: 1.8445 - val_accuracy: 0.2882\n",
      "Epoch 4/100\n",
      "45/45 - 2s - loss: 0.9871 - accuracy: 0.6091 - val_loss: 1.7294 - val_accuracy: 0.3692\n",
      "Epoch 5/100\n",
      "45/45 - 2s - loss: 0.8624 - accuracy: 0.6699 - val_loss: 1.4686 - val_accuracy: 0.4327\n",
      "Epoch 6/100\n",
      "45/45 - 2s - loss: 0.6917 - accuracy: 0.7399 - val_loss: 1.4297 - val_accuracy: 0.4731\n",
      "Epoch 7/100\n",
      "45/45 - 2s - loss: 0.5901 - accuracy: 0.7830 - val_loss: 1.1915 - val_accuracy: 0.5541\n",
      "Epoch 8/100\n",
      "45/45 - 2s - loss: 0.4712 - accuracy: 0.8307 - val_loss: 0.8938 - val_accuracy: 0.6755\n",
      "Epoch 9/100\n",
      "45/45 - 2s - loss: 0.3865 - accuracy: 0.8680 - val_loss: 0.6965 - val_accuracy: 0.7369\n",
      "Epoch 10/100\n",
      "45/45 - 2s - loss: 0.3142 - accuracy: 0.8973 - val_loss: 0.5682 - val_accuracy: 0.7969\n",
      "Epoch 11/100\n",
      "45/45 - 2s - loss: 0.2632 - accuracy: 0.9097 - val_loss: 0.5744 - val_accuracy: 0.7858\n",
      "Epoch 12/100\n",
      "45/45 - 2s - loss: 0.2127 - accuracy: 0.9317 - val_loss: 0.4148 - val_accuracy: 0.8507\n",
      "Epoch 13/100\n",
      "45/45 - 2s - loss: 0.1861 - accuracy: 0.9384 - val_loss: 0.3958 - val_accuracy: 0.8562\n",
      "Epoch 14/100\n",
      "45/45 - 2s - loss: 0.1545 - accuracy: 0.9564 - val_loss: 0.3569 - val_accuracy: 0.8793\n",
      "Epoch 15/100\n",
      "45/45 - 2s - loss: 0.1385 - accuracy: 0.9564 - val_loss: 0.3291 - val_accuracy: 0.8849\n",
      "Epoch 16/100\n",
      "45/45 - 2s - loss: 0.1131 - accuracy: 0.9661 - val_loss: 0.3584 - val_accuracy: 0.8918\n",
      "Epoch 17/100\n",
      "45/45 - 2s - loss: 0.0988 - accuracy: 0.9724 - val_loss: 0.3079 - val_accuracy: 0.8897\n",
      "Epoch 18/100\n",
      "45/45 - 2s - loss: 0.0929 - accuracy: 0.9731 - val_loss: 0.3342 - val_accuracy: 0.8897\n",
      "Epoch 19/100\n",
      "45/45 - 2s - loss: 0.0838 - accuracy: 0.9771 - val_loss: 0.3110 - val_accuracy: 0.8967\n",
      "Epoch 20/100\n",
      "45/45 - 2s - loss: 0.0691 - accuracy: 0.9804 - val_loss: 0.2803 - val_accuracy: 0.9135\n",
      "Epoch 21/100\n",
      "45/45 - 2s - loss: 0.0693 - accuracy: 0.9808 - val_loss: 0.2845 - val_accuracy: 0.9072\n",
      "Epoch 22/100\n",
      "45/45 - 2s - loss: 0.0608 - accuracy: 0.9817 - val_loss: 0.3073 - val_accuracy: 0.9100\n",
      "Epoch 23/100\n",
      "45/45 - 2s - loss: 0.0581 - accuracy: 0.9841 - val_loss: 0.3054 - val_accuracy: 0.9058\n",
      "Epoch 24/100\n",
      "45/45 - 2s - loss: 0.0575 - accuracy: 0.9834 - val_loss: 0.2939 - val_accuracy: 0.9093\n",
      "Epoch 25/100\n",
      "45/45 - 2s - loss: 0.0573 - accuracy: 0.9839 - val_loss: 0.2861 - val_accuracy: 0.9086\n",
      "Epoch 26/100\n",
      "45/45 - 2s - loss: 0.0529 - accuracy: 0.9845 - val_loss: 0.2747 - val_accuracy: 0.9121\n",
      "Epoch 27/100\n",
      "45/45 - 2s - loss: 0.0525 - accuracy: 0.9848 - val_loss: 0.3053 - val_accuracy: 0.9044\n",
      "Epoch 28/100\n",
      "45/45 - 2s - loss: 0.0447 - accuracy: 0.9880 - val_loss: 0.3120 - val_accuracy: 0.9086\n",
      "Epoch 29/100\n",
      "45/45 - 2s - loss: 0.0447 - accuracy: 0.9871 - val_loss: 0.2591 - val_accuracy: 0.9191\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 30/100\n",
      "45/45 - 2s - loss: 0.0371 - accuracy: 0.9907 - val_loss: 0.2590 - val_accuracy: 0.9246\n",
      "Epoch 31/100\n",
      "45/45 - 2s - loss: 0.0387 - accuracy: 0.9900 - val_loss: 0.2674 - val_accuracy: 0.9100\n",
      "Epoch 32/100\n",
      "45/45 - 2s - loss: 0.0385 - accuracy: 0.9897 - val_loss: 0.2891 - val_accuracy: 0.9149\n",
      "Epoch 33/100\n",
      "45/45 - 2s - loss: 0.0418 - accuracy: 0.9890 - val_loss: 0.2871 - val_accuracy: 0.9184\n",
      "Epoch 34/100\n",
      "45/45 - 2s - loss: 0.0388 - accuracy: 0.9869 - val_loss: 0.2653 - val_accuracy: 0.9239\n",
      "Epoch 35/100\n",
      "45/45 - 2s - loss: 0.0361 - accuracy: 0.9871 - val_loss: 0.3089 - val_accuracy: 0.9191\n",
      "Epoch 36/100\n",
      "45/45 - 2s - loss: 0.0385 - accuracy: 0.9895 - val_loss: 0.3160 - val_accuracy: 0.9093\n",
      "\n",
      "Epoch 00036: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.\n",
      "Epoch 37/100\n",
      "45/45 - 2s - loss: 0.0329 - accuracy: 0.9902 - val_loss: 0.2778 - val_accuracy: 0.9191\n",
      "Epoch 38/100\n",
      "45/45 - 2s - loss: 0.0234 - accuracy: 0.9930 - val_loss: 0.2703 - val_accuracy: 0.9281\n",
      "Epoch 39/100\n",
      "45/45 - 2s - loss: 0.0234 - accuracy: 0.9937 - val_loss: 0.2669 - val_accuracy: 0.9295\n",
      "Epoch 40/100\n",
      "45/45 - 2s - loss: 0.0212 - accuracy: 0.9935 - val_loss: 0.2651 - val_accuracy: 0.9295\n",
      "Restoring model weights from the end of the best epoch.\n",
      "Epoch 00040: early stopping\n",
      "14/14 [==============================] - 0s 18ms/step - loss: 0.2817 - accuracy: 0.9129\n",
      "Test loss, test acc: [0.2817017436027527, 0.9128978252410889]\n"
     ]
    }
   ],
   "source": [
    "final_res = {}\n",
    "\n",
    "for d in param_found:\n",
    "    print(d)\n",
    "    # Build model\n",
    "    params = param_found[d].get_best_hyperparameters()[0]\n",
    "    h_model = param_found[d].hypermodel.build(params)\n",
    "    h_model.summary()\n",
    "    # Train the hypertuned model\n",
    "    earlyStopping = EarlyStopping(monitor='val_loss', patience=10, verbose=1, mode='min', restore_best_weights=True)\n",
    "    reduce_lr_loss = ReduceLROnPlateau(monitor='val_loss', factor=0.1, patience=6, verbose=1, min_delta=1e-4, mode='min')\n",
    "    history = h_model.fit(train[d]['x'], train[d]['y'],\n",
    "                          epochs=100,\n",
    "                          batch_size=params.get_config()[\"values\"][\"batch_size\"],\n",
    "                          validation_split=0.2,\n",
    "                          callbacks=[earlyStopping, reduce_lr_loss], verbose=2)\n",
    "    \n",
    "    # Evaluate model on test set\n",
    "    results = h_model.evaluate(test[d]['x'], test[d]['y'],\n",
    "                               batch_size=params.get_config()[\"values\"][\"batch_size\"]\n",
    "                               )\n",
    "    print(\"Test loss, test acc:\", results)\n",
    "    final_res[d] = [history, results]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "fd7a88cb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "For emodb test loss: 0.3126097321510315 accuracy: 0.9068100452423096\n",
      "For emovo test loss: 0.34495365619659424 accuracy: 0.8956433534622192\n",
      "For ravdess test loss: 0.2817017436027527 accuracy: 0.9128978252410889\n"
     ]
    }
   ],
   "source": [
    "final_df = []\n",
    "for d in final_res:\n",
    "    print(f'For {d} test loss: {final_res[d][1][0]} accuracy: {final_res[d][1][1]}')\n",
    "    final_df.append([final_res[d][1][0], final_res[d][1][1]])\n",
    "final_df = pd.DataFrame(final_df, columns=['F loss', 'F acc'], index=final_res.keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "6311430b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>F loss</th>\n",
       "      <th>F acc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>emodb</th>\n",
       "      <td>0.312610</td>\n",
       "      <td>0.906810</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>emovo</th>\n",
       "      <td>0.344954</td>\n",
       "      <td>0.895643</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ravdess</th>\n",
       "      <td>0.281702</td>\n",
       "      <td>0.912898</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           F loss     F acc\n",
       "emodb    0.312610  0.906810\n",
       "emovo    0.344954  0.895643\n",
       "ravdess  0.281702  0.912898"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "34a06b86",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "emodb\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "emovo\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ravdess\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABNEAAAIhCAYAAABg7ixdAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAA9hAAAPYQGoP6dpAADccklEQVR4nOzdd3hUZf7+8ffMpHdCOgRCr9I7IiJFUbGXtSGudXVVZNVd1q+6sv7E1VVRse2KRnddK4rYQaUpiHSQXgIJkJAQICE9mTm/P04SCKGkTHJmkvt1XXOdMydnztwT3OXwmefzPDbDMAxERERERERERETklOxWBxAREREREREREfF0KqKJiIiIiIiIiIicgYpoIiIiIiIiIiIiZ6AimoiIiIiIiIiIyBmoiCYiIiIiIiIiInIGKqKJiIiIiIiIiIicgYpoIiIiIiIiIiIiZ6AimoiIiIiIiIiIyBmoiCYiIiIiIiIiInIGKqKJiMfYvXs3NpuN5OTkWr924cKF2Gw2Fi5c6PZcIiIiIuKZGuP+UfeZIlJBRTQREREREREREZEzUBFNRMSDFRYWYhiG1TFERERERESaPRXRRKTS3/72N2w2G+vXr+fqq68mPDycyMhIpkyZQllZGVu3buWCCy4gNDSUpKQknnnmmWrXSE1N5cYbbyQmJgZ/f3+6devGc889h8vlqnLe/v37ueaaawgNDSU8PJxrr72WjIyMk+ZauXIll1xyCZGRkQQEBNC3b18++uijOn3GrKws7r77brp3705ISAgxMTGcd955LFmypNq5xcXFTJs2jW7duhEQEEDLli0ZNWoUS5curTzH5XLx8ssv06dPHwIDA4mIiGDIkCHMnTu38hybzcbf/va3atdPSkpi0qRJlc+Tk5Ox2WzMmzeP3//+90RHRxMUFERxcTE7duzglltuoVOnTgQFBdGqVSsmTJjAhg0bql33yJEj/OlPf6J9+/b4+/sTExPDhRdeyJYtWzAMg06dOnH++edXe11eXh7h4eHcc889tfytioiISHPVHO4fT2Xu3LkMHTqUoKAgQkNDGTt2LMuWLatyTlZWFnfccQeJiYn4+/sTHR3N8OHD+f777yvPWbNmDRdffHHl509ISOCiiy5i7969bs0rIvXnY3UAEfE811xzDTfeeCN33nkn8+fP55lnnqG0tJTvv/+eu+++mwcffJD//e9//PnPf6Zjx45cccUVgHmTMGzYMEpKSvj73/9OUlISX375JQ8++CA7d+7k1VdfBczRVWPGjGH//v1Mnz6dzp0789VXX3HttddWy7JgwQIuuOACBg8ezOuvv054eDgffPAB1157LQUFBVWKUDVx6NAhAB5//HHi4uLIy8vjs88+49xzz+WHH37g3HPPBaCsrIzx48ezZMkSJk+ezHnnnUdZWRm//PILqampDBs2DIBJkybx3//+l1tvvZVp06bh5+fH6tWr2b17d91++cDvf/97LrroIv7zn/+Qn5+Pr68v+/fvp2XLljz99NNER0dz6NAh3nnnHQYPHsyaNWvo0qULAEePHuXss89m9+7d/PnPf2bw4MHk5eWxePFi0tPT6dq1K/feey+TJ09m+/btdOrUqfJ93333XXJzc1VEExERkVpryvePJ/O///2PG264gXHjxvH+++9TXFzMM888U3lPefbZZwNw0003sXr1av7f//t/dO7cmSNHjrB69Wqys7MByM/PZ+zYsbRr145XXnmF2NhYMjIyWLBgAUePHq13ThFxM0NEpNzjjz9uAMZzzz1X5XifPn0MwPj0008rj5WWlhrR0dHGFVdcUXnsL3/5iwEYy5cvr/L6P/zhD4bNZjO2bt1qGIZhvPbaawZgfP7551XOu/322w3AePvttyuPde3a1ejbt69RWlpa5dyLL77YiI+PN5xOp2EYhrFgwQIDMBYsWFCrz1xWVmaUlpYao0ePNi6//PLK4++++64BGP/+979P+drFixcbgPHII4+c9j0A4/HHH692vG3btsbNN99c+fztt982AGPixIk1yl1SUmJ06tTJeOCBByqPT5s2zQCM+fPnn/K1ubm5RmhoqHH//fdXOd69e3dj1KhRZ3xvERERkQrN4f7xxPOcTqeRkJBgnHXWWZXXMgzDOHr0qBETE2MMGzas8lhISIgxefLkU1575cqVBmDMmTPntBlExDOonVNEqrn44ourPO/WrRs2m43x48dXHvPx8aFjx47s2bOn8tiPP/5I9+7dGTRoUJXXT5o0CcMw+PHHHwHz28HQ0FAuueSSKuddf/31VZ7v2LGDLVu2cMMNNwDm6LCKx4UXXkh6ejpbt26t9ed7/fXX6devHwEBAfj4+ODr68sPP/zA5s2bK8/55ptvCAgI4Pe///0pr/PNN98AuH3k1pVXXlntWFlZGU899RTdu3fHz88PHx8f/Pz82L59e7XcnTt3ZsyYMae8fmhoKLfccgvJycnk5+cD5p/dpk2b+OMf/+jWzyIiIiLNQ1O/fzze1q1b2b9/PzfddBN2+7F/UoeEhHDllVfyyy+/UFBQAMCgQYNITk7mySef5JdffqG0tLTKtTp27EiLFi3485//zOuvv86mTZvqlU1EGpaKaCJSTWRkZJXnfn5+BAUFERAQUO14UVFR5fPs7Gzi4+OrXS8hIaHy5xXb2NjYaufFxcVVeX7gwAEAHnzwQXx9fas87r77bgAOHjxYq8/2/PPP84c//IHBgwcze/ZsfvnlF1asWMEFF1xAYWFh5XlZWVkkJCRUuTE6UVZWFg6Ho1ru+jrZ73DKlCk8+uijXHbZZXzxxRcsX76cFStW0Lt372q5W7dufcb3uPfeezl69CjvvfceADNnzqR169Zceuml7vsgIiIi0mw05fvHE1VkOlVul8vF4cOHAfjwww+5+eabefPNNxk6dCiRkZFMnDixci638PBwFi1aRJ8+ffjrX/9Kjx49SEhI4PHHH69WcBMR62lONBFxm5YtW5Kenl7t+P79+wGIioqqPO/XX3+tdt6JE8NWnD916tTKeTNOVDEXWE3997//5dxzz+W1116rcvzEOSeio6P56aefcLlcpyykRUdH43Q6ycjIOOlNVAV/f3+Ki4urHa+4ATuRzWY7ae6JEyfy1FNPVTl+8OBBIiIiqmSqySS0HTt2ZPz48bzyyiuMHz+euXPn8sQTT+BwOM74WhERERF38Yb7x5NlBk6Z226306JFi8o8M2bMYMaMGaSmpjJ37lz+8pe/kJmZybfffgvAWWedxQcffIBhGKxfv57k5GSmTZtGYGAgf/nLX+qVVUTcSyPRRMRtRo8ezaZNm1i9enWV4++++y42m41Ro0YBMGrUKI4ePVplBUswJ2g9XpcuXejUqRPr1q1jwIABJ32EhobWKqPNZsPf37/KsfXr11dbSWn8+PEUFRWRnJx8ymtVtCecWJA7UVJSEuvXr69y7McffyQvL69eub/66iv27dtXLdO2bdsqWx9O5/7772f9+vXcfPPNOBwObr/99hrnEREREXEHb7h/PFGXLl1o1aoV//vf/zAMo/J4fn4+s2fPrlyx80Rt2rThj3/8I2PHjq32ecG83+vduzcvvPACERERJz1HRKylkWgi4jYPPPAA7777LhdddBHTpk2jbdu2fPXVV7z66qv84Q9/oHPnzgBMnDiRF154gYkTJ/L//t//o1OnTnz99dd899131a75xhtvMH78eM4//3wmTZpEq1atOHToEJs3b2b16tV8/PHHtcp48cUX8/e//53HH3+ckSNHsnXrVqZNm0a7du0oKyurPO+6667j7bff5q677mLr1q2MGjUKl8vF8uXL6datG7/73e8YMWIEN910E08++SQHDhzg4osvxt/fnzVr1hAUFMS9994LmKsyPfroozz22GOMHDmSTZs2MXPmTMLDw2uVOzk5ma5du9KrVy9WrVrFs88+W611c/LkyXz44Ydceuml/OUvf2HQoEEUFhayaNEiLr744sobUYCxY8fSvXt3FixYULmsvIiIiEhj8ob7xxPZ7XaeeeYZbrjhBi6++GLuvPNOiouLefbZZzly5AhPP/00ADk5OYwaNYrrr7+erl27EhoayooVK/j2228rR8l9+eWXvPrqq1x22WW0b98ewzD49NNPOXLkCGPHjq1XThFpAJYuayAiHqVidaWsrKwqx2+++WYjODi42vkjR440evToUeXYnj17jOuvv95o2bKl4evra3Tp0sV49tlnq6xcZBiGsXfvXuPKK680QkJCjNDQUOPKK680li5dWm11JcMwjHXr1hnXXHONERMTY/j6+hpxcXHGeeedZ7z++uuV59R0daXi4mLjwQcfNFq1amUEBAQY/fr1M+bMmWPcfPPNRtu2baucW1hYaDz22GNGp06dDD8/P6Nly5bGeeedZyxdurTyHKfTabzwwgtGz549DT8/PyM8PNwYOnSo8cUXX1R5z4cffthITEw0AgMDjZEjRxpr16495eqcK1asqJb78OHDxq233mrExMQYQUFBxtlnn20sWbLEGDlypDFy5Mhq595///1GmzZtDF9fXyMmJsa46KKLjC1btlS77t/+9jcDMH755ZfT/t5ERERETqY53D+e6rw5c+YYgwcPNgICAozg4GBj9OjRxs8//1z586KiIuOuu+4yevXqZYSFhRmBgYFGly5djMcff9zIz883DMMwtmzZYlx33XVGhw4djMDAQCM8PNwYNGiQkZycfNpMImINm2EcN/5URESalQEDBmCz2VixYoXVUURERERERDya2jlFRJqZ3NxcfvvtN7788ktWrVrFZ599ZnUkERERERERj6cimohIM7N69WpGjRpFy5Ytefzxx7nsssusjiQiIiIiIuLx1M4pIiIiIiIiIiJyBnarA4iIiIiIiIiIiHg6FdFERERERERERETOQEU0ERERERERERGRM2h2Cwu4XC72799PaGgoNpvN6jgiIiLiJQzD4OjRoyQkJGC363tIT6T7PBEREamLmt7nNbsi2v79+0lMTLQ6hoiIiHiptLQ0WrdubXUMOQnd54mIiEh9nOk+r9kV0UJDQwHzFxMWFmZxGhEREfEWubm5JCYmVt5LiOfRfZ6IiIjURU3v85pdEa1iaH9YWJhurkRERKTW1CbouXSfJyIiIvVxpvs8TeghIiIiIiIiIiJyBiqiiYiIiIiIiIiInIGKaCIiIiIiIiIiImfQ7OZEqwnDMCgrK8PpdFodxSs5HA58fHw0Z4yIiIh4HN3n1Y/u80REpDlTEe0EJSUlpKenU1BQYHUUrxYUFER8fDx+fn5WRxEREREBdJ/nLrrPExGR5kpFtOO4XC5SUlJwOBwkJCTg5+enb9lqyTAMSkpKyMrKIiUlhU6dOmG3q2tYRERErKX7vPrTfZ6IiDR3KqIdp6SkBJfLRWJiIkFBQVbH8VqBgYH4+vqyZ88eSkpKCAgIsDqSiIiINHO6z3MP3eeJiEhzpq+OTkLfqNWffociIiLiiXSPUn/6HYqISHOlvwFFRERERERERETOQEU0ERERERERERGRM7C0iLZ48WImTJhAQkICNpuNOXPmnPE1ixYton///gQEBNC+fXtef/31hg/azCQlJTFjxgyrY4iIiIiIm+k+T0REpO4sLaLl5+fTu3dvZs6cWaPzU1JSuPDCCxkxYgRr1qzhr3/9K/fddx+zZ89u4KSe79xzz2Xy5MluudaKFSu444473HItEREREakf3eeJiIh4BktX5xw/fjzjx4+v8fmvv/46bdq0qfz2rFu3bqxcuZJ//vOfXHnllQ2UsmkwDAOn04mPz5n/yKOjoxshkYiIiIi4g+7zREREGodXzYm2bNkyxo0bV+XY+eefz8qVKyktLT3pa4qLi8nNza3yqA3DMCgoKbPkYRhGjTJOmjSJRYsW8eKLL2Kz2bDZbCQnJ2Oz2fjuu+8YMGAA/v7+LFmyhJ07d3LppZcSGxtLSEgIAwcO5Pvvv69yvROH+dtsNt58800uv/xygoKC6NSpE3Pnzq3V71FERETE0+g+T/d5IiIitWHpSLTaysjIIDY2tsqx2NhYysrKOHjwIPHx8dVeM336dJ544ok6v2dhqZPuj31X59fXx6Zp5xPkd+Y/ohdffJFt27bRs2dPpk2bBsDGjRsBePjhh/nnP/9J+/btiYiIYO/evVx44YU8+eSTBAQE8M477zBhwgS2bt1KmzZtTvkeTzzxBM888wzPPvssL7/8MjfccAN79uwhMjLSPR9WREREpJHpPs+k+zwREZGa8aqRaGB+W3a8im/xTjxeYerUqeTk5FQ+0tLSGjxjYwsPD8fPz4+goCDi4uKIi4vD4XAAMG3aNMaOHUuHDh1o2bIlvXv35s477+Sss86iU6dOPPnkk7Rv3/6M3zhOmjSJ6667jo4dO/LUU0+Rn5/Pr7/+2hgfT0RERKTZ0n2eiIiI5/CqkWhxcXFkZGRUOZaZmYmPjw8tW7Y86Wv8/f3x9/ev83sG+jrYNO38Or++PgJ9HfW+xoABA6o8z8/P54knnuDLL79k//79lJWVUVhYSGpq6mmv06tXr8r94OBgQkNDyczMrHc+ERFpflwugxKni+JSF0VlTopLXRSXOSkuM7dF5c9LylzYbDbsNht2G9htNmzl24pjtvKtw262uQX6OuieEGb1RxQv4ZH3ec5ScDnBN+CM19B9noiISOPyqiLa0KFD+eKLL6ocmzdvHgMGDMDX17dB3tNms9VoqL2nCg4OrvL8oYce4rvvvuOf//wnHTt2JDAwkKuuuoqSkpLTXufE36/NZsPlcrk9r4iIeC/DMDiUX0La4ULSDhWw93AhaYcLKvcPHi2muMxFibPh/v7oGBPC91NGNtj1pWnxyPu87J1QVgRRncEv6LSn6j5PRESkcVl615CXl8eOHTsqn6ekpLB27VoiIyNp06YNU6dOZd++fbz77rsA3HXXXcycOZMpU6Zw++23s2zZMmbNmsX7779v1UfwGH5+fjidzjOet2TJEiZNmsTll18OmH8Gu3fvbuB0IiKNx+Uy2Hu4kG0HjrIt8ygpWfkABPg68Pex4+9rx9/H3D/VMV+HHYfdho/dZm4dFft2fMqfmz83z/N12PD3ceCwn3xqAW9S5nRRVOaiuLRiZNhxo8OOO1ZU6uRAbpFZKDtUQNphs1BWUHLmv4uOZ7cd92fj4yj/szD3/XzsGIaBy6By66ry/Ngxo3wbH37m0TsiHstZBmWF5n7eAYhsB+g+T0RExFNYWkRbuXIlo0aNqnw+ZcoUAG6++WaSk5NJT0+vMvy8Xbt2fP311zzwwAO88sorJCQk8NJLL3HllVc2enZPk5SUxPLly9m9ezchISGn/PawY8eOfPrpp0yYMAGbzcajjz6qbxpFxDKlThelThcBPg7stSxAuVwG+44Usj3zKNsO5LHtwFG2H8hjR2YehaW1K+S4S0UxzSwC2fGvLNIdd8zHQYCvnZbBfsSEBRAd6k90qD8xof7EhAYQGezXIMW43KJS0o8UsT+nkPQjRWTkFLI/p4j08ueZR4spLHXidNVsxcBTsdkgNjSA1i0CSYwMIrFFIK0jg0hsEURMmD+BJ/xOfMrbMEUEKC04tl90BMqKwcdf93kiIiIewtIi2rnnnnva5b2Tk5OrHRs5ciSrV69uwFTe6cEHH+Tmm2+me/fuFBYW8vbbb5/0vBdeeIHf//73DBs2jKioKP785z+Tm5vbyGlFpDkpc7rYe7iQlOx8dh80HynZBew+mM++I4WVRZsAXzuBvg4CfR0E+DkI8ivfLz8W5Ocg0M9BqdNge2YeOw4cJf8Uo578HHY6xITQOTaEDtEh+DhsJ8y/ddwcXBXzcZUeG3FV6nLhdBmUOQ1z63JR5jJwOg1zW37sxHpTqdOg1FlGXnHdf18Ou42WwX5VCmvRof4E+jlwuY4feXX6kVkFxU6zYJZTREZOEXnFZbXO4uewVx2pd9yIPX8fO1Eh/rSODCSxRVBlwaxVi0D8feo/p6dIs3R8EQ0gLxMiEnWfJyIi4iFsxumqWE1Qbm4u4eHh5OTkEBZWdeLhoqIiUlJSaNeuHQEBagepD/0uRbyTYZhFolKni9Iyc/L3ioKR02VQelxRydwaVYpNJU6nWTArL5btzjbnwyqr5+imU/F12GgfFUKn2BA6x4bSOTaETrGhtI0MwsfR8AtQu1zHfl8lx7U5nlikO/FYQYmT7LxiMo8Wk3W0YltEdn4JDfm3cnigL/HhASREBBIXHkBCeADx4YHERwQQGxZAiL9PlVbKptCe6k6nu4cQz+D193mHdkFRDviHQ3EOYIPYHuBomLl/68orfpciIiK1UNP7PA+bSVVEpOkqdbrIyCkivaKFLqeI9CNmS11GThGH8ksq2wEDfKvPD1VlFFD5vmEYlJRP1F5R+Cp1uih2uigtKz/mNCp/XnL8scr9iuNGg0347u9jJ6llMElRQSRFBZv7LYNpFxVMaIAPhaVOCkucFJU6K/cLSp0UlZQ/Lz9WWD7yrH20OcosKSoY30Yolp2K3W7Dz27Dz8dOcN0Xgq5U5nSRnV9SXlgrIjP3WJGtpMyF3X5sNcqKFSorVqs0V6ekctVKfx9HeaGsvGAWEeB5E6iLSFUl5SPRQmLAVWqOTMvPgrAEa3OJiIgIoCKaiEiNOF0GxWVOCkqOFXsKjivwFJUce15UXvA5VFBC+pFjBbOsvOIGHWXUUHxPmES/crJ9uw2Hw4Zv+fGKCfjjwgJpV14sa9cymKSoYOLCAk4751mwv/46AvBx2IkNM0eFQbjVcUSkMTlLzcIZgG8ghMTC4RTIP2ju29UmLSIiYjX9q0VEmrzD+SXsOpjHzqx8dmXlsysrj72HCyl1Vm1JPPH58S2L7ip++TnsxIUHVLbUxZfvx4cHEhXqj9PlMldBPGHurhOPVbQHVqwM6edjx9dhPvwcdnwdNnx9zP3jf+bjsOHvsFf+zNdhx8/Hduy1PlWv4dCk7yIijaNiPjSfALNgFhAODn9wFkNBtjk6TURERCylIpqINAklZS5SD+VXKZTtOmhuDxeUuvW9KibAD/LzMffLJ8AP9PMhsGJyfD8fwgJ9jrXSlc871TLYT0UpERGprqKI5htkbm02s3CWk2YuMBAcBTbr2tdFRERERTQR8TKGYXAgt5jN6blsSs9lc/ljd3ZB5SqPJ5MQHkD76BDaRwfTPiqYti2D8fc1R11VbVG0V47uOrGFMcDXnI/sdG2JIiIidVIxH5pv4LFjQZFwNMNs8yw8DEEtrckmIiIigIpoIuLBSspc7MjMqyyUbc7IZdP+3FOOLAv2cxxXKCvfRpuT12tCdRER8Wilhea2YiQamCPPgqPh6H5zNFpgpDlCTURERCyhf1WKSIM6WlTKhr055Jc4Tzu3V3GZs/J5UamTXVn57MzKo9RZfXSZw26jfVQw3eLDyh+hdI0LIzbMX62SIiLifZwlVRcVOF5wS8g7AGVFUJxrzpUmIiIillARTUTcyjAMdmTmsWBrJgu2ZLFyz6GTFsJqKjTAh27xYXQvL5Z1jw+nU2wIAb5apUxERJqIkvJRaBWLChzP7lNeSMuEowdURBMREbGQimgiUm8FJWUs25ldWTjbd6Swys9btwgkKsQffx975bxi/uXbAF87/j7lx3wc+Pva8fex0yoikO4JYbSKCNToMhERadpOXFTgRMExkJcFpflQnAf+IY2XTURERCqpiCYAJCUlMXnyZCZPnmx1FPESKQfzWbAlk4XbsvhlVzYlZa7Kn/n52BnSviWjukRzbpcY2kUFW5hURETEw1UU0fxOUURz+JqLDBRkm62dtSyi6T5PRETEPVREE5EayzxaxH+W7eGLdfvZnV1Q5WetIgI5r2sMo7pGM6R9S03kLyIiUhOGceaRaGCORivINudFKy2sPneaiIiINDj9K1dEzmhrxlHeXLKLz9fup8RpjjjzddgYmBTJqC5m4axDdIjaLkVERGrLWQquMsAGPqcpjPkGmPOhFeWY86O1aNtoEUVERMSkItqZHP/tYGPzDarRMuZvvPEG06ZNIy0tDbvdXnn8kksuoUWLFjz22GNMmTKFX375hfz8fLp168b06dMZM2ZMQ6YXL2cYBj/tOMi/l6SweFtW5fH+bVswaVgSo7rGEOKv/wsREREv5gn3eRXv7xMAx93HVahynxcSaxbRCg9zycR7aBHZUvd5IiIijUj/Aj6T0gJ4KsGa9/7rfvA781xSV199Nffddx8LFixg9OjRABw+fJjvvvuOL774gry8PC688EKefPJJAgICeOedd5gwYQJbt26lTZs2Df0pxMsUlzmZu3Y/s35KYUvGUQDsNrigZxy3jWhPvzYtLE4oIiLiJp5wn3eG+dCq3ef5hXA4cz/fzZuv+zwREZFGVv3rLvE6kZGRXHDBBfzvf/+rPPbxxx8TGRnJ6NGj6d27N3feeSdnnXUWnTp14sknn6R9+/bMnTvXwtTiaY4UlPDKgh2M+McCHvpkPVsyjhLk52DSsCQWPjiKV2/orwKaiIgAsHjxYiZMmEBCQgI2m405c+ac9vxJkyZhs9mqPXr06FF5TnJy8knPKSoqauBPY7HK+dBO3spZ7T4vJJaPv5xPZEQYo88dqfs8ERGRRqSRaGfiG2R+U2jVe9fQDTfcwB133MGrr76Kv78/7733Hr/73e9wOBzk5+fzxBNP8OWXX7J//37KysooLCwkNTW1AcOLVfZk55NXXFbj84vLXMxZs4+PV+6lsNQJQGyYP5OGteP6QW0ID/JtqKgiIuKl8vPz6d27N7fccgtXXnnlGc9/8cUXefrppyufl5WV0bt3b66++uoq54WFhbF169YqxwICAtwT+mSsvs8zDCg586ICVe/zQnlvznf87pJxOIoPk18aqvs8ERGRRqIi2pnYbDVqqbTahAkTcLlcfPXVVwwcOJAlS5bw/PPPA/DQQw/x3Xff8c9//pOOHTsSGBjIVVddRUlJicWpxZ2KSp389dMNfLpmX52v0S0+jNtHtOPiXgn4+WigqoiInNz48eMZP358jc8PDw8nPDy88vmcOXM4fPgwt9xyS5XzbDYbcXFxNb5ucXExxcXFlc9zc3Nr/NryN7T2Pq+sGAwnYDvtapvV7vN+WcXzj02G/CweeuIJvps3T/d5IiIijUBFtCYiMDCQK664gvfee48dO3bQuXNn+vfvD8CSJUuYNGkSl19+OQB5eXns3r3bwrTibpm5Rdzxn1WsTTuC3QZRIf61en3PVuHcdnY7hnZoqRU2RUSkwc2aNYsxY8bQtm3VFSbz8vJo27YtTqeTPn368Pe//52+ffue8jrTp0/niSeeaOi4DaeylTMAbKf+8uqk93l9+4CzhCWLF+k+T0REpJGoiNaE3HDDDUyYMIGNGzdy4403Vh7v2LEjn376KRMmTMBms/Hoo4/icrksTCrutGFvDre/u5KM3CLCA3159YZ+DO8YZXUsERGRk0pPT+ebb76pMpcrQNeuXUlOTuass84iNzeXF198keHDh7Nu3To6dep00mtNnTqVKVOmVD7Pzc0lMTGxQfO7VemZWzkrVLvPC46B3L10bBuv+zwREZFGoiJaE3LeeecRGRnJ1q1buf766yuPv/DCC/z+979n2LBhREVF8ec//7n27Q7ikb5Yt5+HPllHUamLDtHBzLp5IElRnt9+LCIizVdycjIRERFcdtllVY4PGTKEIUOGVD4fPnw4/fr14+WXX+all1466bX8/f3x96/d6GuPUoP50CpUu88LioS8DF54fAq/f/hp3eeJiIg0AhXRmhCHw8H+/dUnx01KSuLHH3+scuyee+6p8lzD/r2Ly2Uw4/ttvPTjDgDO7RLNS9f1JSxAiwCIiIjnMgyDt956i5tuugk/P7/Tnmu32xk4cCDbt29vpHSNzDCgtNDcr0ER7aT3ecFRJCWW8ePH/4boLuYcb+g+T0REpKFo5nARL5NfXMYf3ltVWUC745z2zLp5oApoIiLi8RYtWsSOHTu49dZbz3iuYRisXbuW+Pj4RkhmAefxiwrUcQXSoGhzLrWyQig+6tZ4IiIiUp1Gool4kb2HC7jtnZVsyTiKn8POU1ecxVX9W1sdS0REmpm8vDx27NhR+TwlJYW1a9cSGRlJmzZtmDp1Kvv27ePdd9+t8rpZs2YxePBgevbsWe2aTzzxBEOGDKFTp07k5uby0ksvsXbtWl555ZUG/zyWqGzlDDztogKn5fCBwEgoOAhFORAQ5r58IiIiUo2KaCJeYuXuQ9z5n1Vk55cQFeLHGzf1p3/bSKtjiYhIM7Ry5UpGjRpV+bxicv+bb76Z5ORk0tPTSU1NrfKanJwcZs+ezYsvvnjSax45coQ77riDjIwMwsPD6du3L4sXL2bQoEEN90GsVItWztPyDzGLaBWLFIiIiEiDURFNxAt8tCKNR+ZsoNRp0D0+jH/fPIBWEYFWxxIRkWbq3HPPxTCMU/48OTm52rHw8HAKCk5d6HnhhRd44YUX3BHPO1QUvfzqWUSrKMKVFoLhqvuoNhERETkjFdFO4nQ3hVIz+h26R5nTxfRvtjDrpxQAxveM47lrehPkp//pioiI1IVH3KMYxrEiWn1Hojn8wOYw51crLap/Ua4GPOJ3KCIiYgH9S/w4vr7mxOwFBQUEBmqUT31UfNNc8TuVmnO6DFanHub7TQeYt+kAKQfzAbh/dCfuH90Ju91mcUIRERHv41H3eWXF5qgxbOBTx0UFKthsZiGu5KhZmGuEIpru80REpLlSEe04DoeDiIgIMjMzAQgKCsJmU8GiNgzDoKCggMzMTCIiInA4HFZH8gr5xWUs2Z7F/E2ZLNiayaH8ksqfBfk5ePaq3lzUq4muTiYiItIIPOo+r/AIlBlmAa242A0X9DOvl58LjhA3XO/kdJ8nIiLNnYpoJ4iLiwOovMGSuomIiKj8XcrJZeQU8f3mA3y/+QBLd2ZTUuaq/FlYgA/ndY1hTPdYzukcTViAvukVERGpL4+5zys8DMVHwT8UctxwvdICyD8IjhwIdUdR7vR0nyciIs2VimgnsNlsxMfHExMTQ2lpqdVxvJKvr6++mTyFXVl5fLk+ne83H2D93qp3zW1bBjGmWyxjusUyIKkFvg5NDCwiIuJOHnOfN/spSF8Do/8G7frV/3pHD8A7NwMOuHMR+NazRfQ0dJ8nIiLNmYpop+BwOHSDIG41f9MB7n5vFaVOczJemw36JkYwpnssY7vF0jEmRO3DIiLHc7nOfM7J2PUlhJyepfd5Lifs/gFK86FVTwhwQ8HLvw0YxZCfCUe2QeKg+l9TREREqlERTaQR/LD5WAFtcLtIruzXmlFdY4gO9bc6mohI4yg8DAc2QdERKMox54QqyjnJ8+OOleTV/n2iusAff3VrdBG3OrjdLKD5BkNUJ/dc02aDVv1g27ewf42KaCIiIg1ERTSRBrZgSyZ/+O9qSp0GF/eKZ8a1ffBRq6aINBf718Cvb8Jvn0BZkdVpRKy3f425je8FdjeOhkvoaxbR9q123zVFRESkChXRRBrQwq2Z3PmfVZQ4XVx4VpwKaCKNrbQIjuyBFu3Ax8/qNM1HaRFs/AxW/Bv2rTp2PDwRQmIgIAICwiGwfFvx/MRj/qFgq2WRQa2c4ukqimgJfd173YR+Va8vIiIibqcimkgDWbQtizvKC2gX9Ijjxd/1VQFNpKE5S81/QKYsgpQlkLbcHP0U1Rkuex1a97c6Yc2V5EPqL7B7CexZCj7+0HqQ2abVeiAERVqdsLrDu2Hl27DmP1CQbR6z+0KPy2Dg7WZ2zf0ozV2DFdHKr3dw27GVP0VERMStVEQTaQBLtmdxx7srKSlzMa57LC9f31erbYo0BJcTMjZAymLzkbqs+jxaNrv5j8pZY2D4ZDj3L2ZBytOUFcPeFeWfZYm57zph9cCUxcf2W3aCxMGQONDcRnWxZhSWywU7f4AVb8K27wBz8RTCWsOAW6DfRHP0mYiAs8z8/yxwfxEtJNoc7ZmTBunrIOls915fREREVEQTcbefdxzktndWUlzmYky3WGZe308FNBF3MQzI2nKsaLb7J3MS+uMFtjD/8dhuJLQ7B4Ki4JuHzTm5fnrenDPostcgoY8Vn+AYZ9lxo+YWHxs1d7yw1uZnaDfiWJEtbTlk74Ds7eZj7X/Nc/3DofUAc7RX4iBoNQACwhouf8EhWPserJgFh1OOHW8/CgbeBp0vAIduM0SqOLgVygrBLxQiO7j/+gl9zCLavtUqoomIiDQA3d2KuNHSnQe59Z0VFJe5GN01hldv6IefjwpoIvXmcsKmz2HJ83BgQ9Wf+YVC0nCz2JQ0AmJ7Vh+RddUs6H4JfDkFMjfBv8+Dcx6EEQ82zlxpRbnmaLisreY/og9sNFs1Txw1FxxzrGjW7hxzLrfj2x8H3GJu87PNgtreXyHtV3PeseIcc0TYzh/Mcxz+0PtaGPpHiO7ivs+SvROWvQJr/2cWA8As4PW9AQb83n2rDYo0RZWLCvRumJGjCf1g8xeaF01ERKSBqIgm4ia/7Mrm98krKCp1MapLNK/eqAKaSL05S2H9R/DTC+aoKwCfAGgztLzYNNL8x2hNRjx1vxTaDoevppgFuUX/gK1fm6PS4s6qf1bDgLxMs0iWtfW4otk2OJp+8tecOGouqnPN5gwLbgldLjAfYI5qy9xoFtTSfjVHqx3ZA6vfNR8dx8KwP5rvU5c5yQzDnJdt2UzY+g2VLZuxZ8Gg2+Csq8EvuPbXFWluKudD69Mw169oEd2vFTpFREQagopoIm7wa8ohbnnbLKCN7BzNazf2x9/HjcvWizQ3pUXm5PQ/vwQ5qeaxgAgYfBcMvrPuk+oHR8HV78DGT+GrP5lzE/1rFIz8M5w9GRy+NbuOYZhFqrTy9sr0dWbxrCjn1K8JiYPozubcZdFdzHnMTjZqri4cPmYxMb43DLrdzJe2HJa+DFu+gh3zzUfcWebItB5X1GwEnrPULDgum1l1ZEvnC8zrJJ2thQJEaqOhFhWoUFGcO7zbbLn2xAVIREREvJiKaCL1tGL3ISa9/SuFpU5GdIrijZv6E+CrAprHcrng5xnmCoKt+kKncdBxDITGWZ3M85QWme2GwVGN957FebDyLbNok3fAPBYcbRZsBt7qntXmbDboeSW0PRu+fAC2fgULnoQtX8Llr0NMt+qvKS2C9LVmYSrtV7OVsiJflWvbIaKtWSSL6ly+7WK2OAZG1D97Tdls0GaI+cjeCctfhzX/NYuGn90J3/8NBt1htocGtqj++qIccwTb8jfM+ZXAHAHY+3cw5B6zGCgitVNWAhm/mfsNVUQLbAGR7eHQLrNg13F0w7yPiIhIM2UzDMOwOkRjys3NJTw8nJycHMLCGnDCZWkWVu05xMRZv5Jf4uTsjlG8efMAFdA8WeFh+Owuc2L5E8X1MgtqncaZk7Pbm9GfY1EOZG2r3oZ4ZA8YLojvY7br9bwCwhIaJkPhYVj+L1j+mrkP5qT6w++HfjeBb2DDvK9hmO2i3zxk/h4cfjDqr9DzKti38lh7ZPq66itl2n0hvhe0HmT+NxPdFVp2BN+AhslaXwWHYNXb5u85L8M85hsMfW+EIX+AyHZwJM0suK16B0qOmucERZkFt4G3Nm5B1QPpHsLzefSfUfo6eOMccw7Bv+xpuFGcn9xqLqRy3qPm3I8iIiJyRjW9h1ARTaQOXC6Dz9ft49E5G8krLmNYh5bMunkggX7NqPDibfavgY8mwpFUc8L10Y+ao562z6s+d0xgC+gwGjqNNUepnapw4CyDo/vh8B6z4FSxPZJq7ttsZrtefC+zzS6uF0S0qd8/nAqPHCtyHdppFrl8AsDH/4Tt8Y/yY74BZiHl+Anus7YdK6ickc2c8P6sq6HbJe4ZWZWzD379l7nCY0XRJrIDnP0A9Lq2cSb9B8hNhy/uM/97OJXgmGMrX7YeZLZNNVRxryGVFcNvs2HpTHMeNTBH0LUaYC5QYDjNY1FdYOg95p+DpxYGG5nuITyfR/8ZrXrH/P+ZdiPh5rkN9z5LZ8K8R6DrxfC79xrufURERJqQmt5DqJ1TpJZW7TnMtC83sS7tCABD2keqgObJDANWJcM3D4OzxGy1u+bdY/PGjJpqTga/4wezgLLzB3Mk1G+fmA9s0KqfWUxz+FYtmOXuA1fZ6d8/dx9s/+7Y84AIc16qiqJafG+z1e/4kW+GAUczjhW5jh8hdrIWQnc4cb6uilZEuw9s/Aw2fAJpv0DKYvPx1Z/MUXtnXQ2dzz9zMckwzDl6MtZD+npzREbG+qqfJ6YHjJgCPS5v/JGAYfFw/UfmipPfTjULerE9y4tmg6H1QGiR1DTm//Lxhz7XQ+/rYNcC8x/cO38wV/oEc4GDofea/803xOqBIs1VQ8+HVqFVv6rvJyIiIm6jkWgiNbTvSCH/+GYLc9ftByDYz8Hdozpy69nt1MLpqUoKzJUY171vPu9yIVz26snngKrgLDPb+LbPMx8ZG07/HnZfiEg0i3Mt2prbiDZmwcVVZhaMMtaZRaPMLdVbAgF8AiG2h/maI3vMwlnxaSaoD403i1xRnc2CSFlR+aP42La0sOrzinN8g+o+X9fh3eYIpvUfQ9bmY8f9QqHbBOh1NSSdYx7L3m5+5vT1ZrEsY/0pJt23mYWq4ZPNyeo9oWjjLANncfNabfLAJkhZZK5eGt/L6jQeS/cQns+j/4zeGGnOrXh1svllQUMpzoOnE82Ryn/aBqGxDfdeIiIiTYTXtHO++uqrPPvss6Snp9OjRw9mzJjBiBEjTnn+K6+8wsyZM9m9ezdt2rThkUceYeLEiTV+P4++uRKPlF9cxuuLdvKvxbsoLnNhs8HV/Vvz4LguxISpxcljHdxhtm9mbjRb1UY/BsPur32RJjcddnxvFhgcfscVycoLZqHxNb9mWYlZfDp+JFbGb1CaX/1cm90sqkV1OWGEWCcICK/dZ2gIBzbCho/NEWoVE88DBEZCaYFZsDuRw8+ctD/uuPbW2B7gH9J4uUXqQfcQns9j/4zKiuGpVuYXKfevM///vSG9MsT8++a6D6HLBQ37XiIiIk2AV7Rzfvjhh0yePJlXX32V4cOH88YbbzB+/Hg2bdpEmzZtqp3/2muvMXXqVP79738zcOBAfv31V26//XZatGjBhAkTLPgE0pS5XAazV+/l2e+2knm0GIDB7SJ59OLu9GzlAUUMObVNn8Oce8yWvOAYuOotcy6vugiLNye273dT/XP5+JnFo/jeQPn1XE5zFbX0dWYxqmJlx8gOnj0PVWwP83HeY+aKlRs+Nts+Cw+ZP/cLMdtW43qZI5viepkT7zfWHGciIp7kwEazgBbYwvz/+YaW0Ncsou1foyKaiIiIG1k6Em3w4MH069eP1157rfJYt27duOyyy5g+fXq184cNG8bw4cN59tlnK49NnjyZlStX8tNPP9XoPT32G0rxKL+mHOLvX25iwz6z/axNZBB/vbAr5/eIw9YU5kRqqpylMP9x+OUV83mbYXD12xAaZ22u5sJZCvtWQ1BLiGzvGa2ZIm6kewjP57F/RitmmdMLdDgPbvqs4d/v13/D1w+ac1fe8HHDv5+IiIiX8/iRaCUlJaxatYq//OUvVY6PGzeOpUuXnvQ1xcXFBARUHZkRGBjIr7/+SmlpKb6+vid9TXFxceXz3NxcN6SXpirtUAFPf7OFrzakAxDi78O953Vk0vAk/H0075lHy90PH99iTn4PMOw+GP04OLR+SqNx+EKbwVanEBHxPOlrzW1DLypQoeJ99q8xF3bRF4AiIiJuYdkwgYMHD+J0OomNrTrZaWxsLBkZGSd9zfnnn8+bb77JqlWrMAyDlStX8tZbb1FaWsrBgwdP+prp06cTHh5e+UhMTHT7Z5Gm4b3lexj9/CK+2pCO3QbXD27DwofO5c6RHVRA82SlhfDbp/DGOWYBzT8Mrn0Pxv1dBTQREfEMjbUyZ4XYnubqyvlZkLO3cd5TRESkGbD8X5gntsYZhnHKdrlHH32UjIwMhgwZgmEYxMbGMmnSJJ555hkcjpMXOaZOncqUKVMqn+fm5qqQJtX8mnKIR+f8hsuA4R1b8n8XdadbvAe1gUhVRbnmypmbv4Dt849NzB97FlzzDrTsYG0+ERFpmkoLzdb1gFrcI5QWQmb5isbxfRokVjW+ARDT3VzAZv8acxVpERERqTfLimhRUVE4HI5qo84yMzOrjU6rEBgYyFtvvcUbb7zBgQMHiI+P51//+hehoaFERUWd9DX+/v74+/u7Pb80HYfyS7jv/TW4DLiibyueu6a35j3zRPnZsPVrs3C2awE4S479LKw19LoaRv4ZfAOtyygiIk3bhk/Muca6XAi9f2fOceaoPp1IFQc2gqsMgqIgvHXj5ARo1a+8iLYaul/SeO8rIiLShFlWRPPz86N///7Mnz+fyy+/vPL4/PnzufTSS0/7Wl9fX1q3Nm9CPvjgAy6++GLsmsBa6sAwDB78eB0ZuUW0jw7m75f1VAHtVAzDvBHf/AUc3gPB0RAaCyGxEBJ3bD8oyn0Tyufuhy1fmatt7vkZDNexn7XsCN0ugW4TzPYY/bmJiEhDS10GZUWw8VPzERQFPa+EXteaRauT/V10fCtnY/5dldAXViUfe38RERGpN0vbOadMmcJNN93EgAEDGDp0KP/6179ITU3lrrvuAsxWzH379vHuu+8CsG3bNn799VcGDx7M4cOHef755/ntt9945513rPwY4sVm/ZTCj1sy8fOx88r1/Qj2t7zD2bO4nOY/GDZ/AZu/hNwazKtic0BITPkj7ti+vRa/W2cJ7P4J9q6oejyu17HCWXQXFc5ERKRxXfoKDLod1n8EGz425xz79Q3z0bKjWUzrdQ20SDr2msaeD61CQr9j76/FBURERNzC0orBtddeS3Z2NtOmTSM9PZ2ePXvy9ddf07ZtWwDS09NJTU2tPN/pdPLcc8+xdetWfH19GTVqFEuXLiUpKcmiTyDebE3qYZ7+ZgsAj0/QHGiVyoohZTFsngtbvoaC4xbt8A2GzuOgVX8oyIa8TDiaYW7zMiD/IBhOOJpuPlhXzzA2SBxsFs26XVz1HyUiIiKNzWYzi2EJfWHs32HXQlj/gflFU/YOWPD/zEfiELOY1uNy64poMd3A4Q9FOXBol+YLFRERcQObYRiG1SEaU25uLuHh4eTk5BAWpqJJc5VTUMqFLy1h35FCLuoVz8zr+jbvNs6SfNjxvTnibNt3UJx77GcBEdD1IrOQ1f7c08855iw1C2l5GccV2A6UF9dcp37dycR0ha4XQ2hcXT6RiIjb6R7C81n2Z1R81Cykrf8QUhYd+zvP7mvOh4YBU7ZAWHzjZQL492jYtxKunAVnXdW47y0iIuJFanoPod41aXYMw+DPs9ez70ghbSKDmH7FWc2vgOZyQdZmSFli3uzvXABlhcd+HhJrFrC6TYCks888aXIFh6/5D4TG/keCiIiIlfxDoc915iM3HX77xCyoZWwwfx7Wypq/G1v1M4to+9eoiCYiIuIGKqJJs/OfX/bw7cYMfB02Zl7fl7CAGhaIvJlhQPZOs2C2e4lZPDu+TRMgom152+Ql0Hqg+xYHEBERaU7C4mHYvebjwCbY9g20GWZNlooW0n2rrXl/ERGRJkZFNGlWftuXw5NfbgZg6vhu9GodYW2ghnQk1ZzbLGWxWTQ7ur/qz30Coe1QaHcOdDjPnLS/uY3IExERaUix3c2HVSoWF0hfZy4WZHdYl0VERKQJUBFNmo284jL++L/VlDhdjOkWyy3Dk6yO5H656bDoH7BrARzeXfVnDj9zkv6kEWbhrFV/8PGzJKaIiIg0gqhO5qJApflwcJu52ICIiIjUmYpo0iwYhsFfP93A7uwCEsID+OfVvZrePGh5WfDOxebqYAA2h1koa3cOtBthFtBOtyiAiIiINC12ByT0gT0/m/OiqYgmIiJSLyqiSbPw0co05q7bj8Nu4+Xr+xIR1MRGYBUehv9cbhbQwhPhoufNVk3/UKuTiYiIiJUS+ppFtH2roc/1VqcRERHxaiqiSZO3NeMoj8/dCMCD47rQv22kxYncrDgP3rsaDmwwV9Wc+Dm07GB1KhEREfEEFYsL7F9jbQ4REZEmQMvvSZNWUFLGPf9bTVGpi5Gdo7nznPZWR3Kv0iL44DrYuwICW8BNc1RAExERkWMqimgZG6CsxNosIiIiXk5FNGnSHv98Izsy84gN8+f5a3pjtzehedCcpfDxJHP1Tb8QuHG2tSuAiYiIiOeJbA8B4eAshqzNVqcRERHxaiqiSZP16eq9fLxqL3YbvPi7vrQM8bc6kvu4nPDZXbDtG/AJgOs/NBcREBERETmezXZsNNq+1dZmERER8XIqokmTlJ5TyKNzfgPg/tGdGdK+pcWJ3Mgw4MsH4LdPwO4L1/4Xks62OpWIiIh4qoR+5lbzoomIiNSLimjSJD39zRbyS5z0axPBH8/raHUc9zEMmPd/sPodsNnhyn9Dp7FWpxIRERFPVrm4gEaiiYiI1IeKaNLkrNx9iM/X7sdmg2mX9sTRlOZBW/QMLJtp7l/yMvS43No8IiIizUxxmZOCkjKrY9ROq/KRaJmbobTQ2iwiIiJeTEU0aVKcLoO/fbERgN8NTKRnq3CLE7nRsldg4VPm/gX/gL43WptHRESkmXn2uy30fmIe/1ueanWU2glrBcHR4CqDjN+sTiMiIuK1VESTJuXjlWn8ti+X0AAfHhzXxeo47rPqHfjur+b+ef8HQ+6yNo+IiEgzFBrgS1Gpi+Uph6yOUjs2m+ZFExERcQMV0aTJyCks5ZnvtgIweUznprMa52+z4Yv7zf3h98OIB63NIyIizd7ixYuZMGECCQkJ2Gw25syZc9rzFy5ciM1mq/bYsmVLlfNmz55N9+7d8ff3p3v37nz22WcN+Clqr2Khol9TDuFyGRanqSXNiyYiIlJvKqJJk/Hi99s5lF9Cx5gQJg5ta3Wc+jEMOLgDls6ET+8ADBjwexjzhPltsoiIiIXy8/Pp3bs3M2fOrNXrtm7dSnp6euWjU6dOlT9btmwZ1157LTfddBPr1q3jpptu4pprrmH58uXujl9nPRPCCPZzkFNYyuaMXKvj1E4rjUQTERGpLx+rA4i4w47Mo7y7bDcAj13cHV+HF9aHD++BlMWwe4m5PZp+7Ge9roULn1MBTUREPML48eMZP358rV8XExNDRETESX82Y8YMxo4dy9SpUwGYOnUqixYtYsaMGbz//vv1ies2Pg47A5IiWbQti+W7DtEjwYvmXq0YiZa1FYqPgn+otXlERES8kBdWGkSqMgyDJ77YRJnLYGz3WM7pHG11pJrJTYf1H8Hn98CMXvBiL5j7R1j/oVlAc/hD0ggY9//g0lfBrv+5ioiId+vbty/x8fGMHj2aBQsWVPnZsmXLGDduXJVj559/PkuXLj3l9YqLi8nNza3yaGgVLZ2/7Mpu8Pdyq5AYCGsNGJC+3uo0IiIiXkkj0cTrfb85kyXbD+LnsPN/F3WzOs7p7VoIm+aaI82yt1f9md0HWvWHdueYxbPEQeAbaElMERERd4qPj+df//oX/fv3p7i4mP/85z+MHj2ahQsXcs455wCQkZFBbGxsldfFxsaSkZFxyutOnz6dJ554okGzn2hw+0gAft1tzotmt3vRKPGEPpC715wXLWm41WlERES8jopo4tWKSp38/ctNANw2oh1tWwZbnOg0snfCu5cB5RMR2+wQ39ssmLUbCW2GgH+IlQlFREQaRJcuXejS5diq2UOHDiUtLY1//vOflUU0ANsJ0xYYhlHt2PGmTp3KlClTKp/n5uaSmJjoxuTVndUqnCA/B0cKStl64Cjd4sMa9P3cqlU/2PKl5kUTERGpIxXRxKvN+imF1EMFxIb5c8+ojlbHOb2NnwIGxJ4Fo/4KbYdBYITVqURERCwxZMgQ/vvf/1Y+j4uLqzbqLDMzs9rotOP5+/vj79+4q3H7ls+LtnhbFr/syvauIlrFvGj7tEKniIhIXWiSJfFaGTlFvLJgBwB/Gd+VYH8PrwlvnGNuB98JXS9UAU1ERJq1NWvWEB8fX/l86NChzJ8/v8o58+bNY9iwYY0d7YwGtzNbOr1uXrSKItrhFCg8bG0WERERL+ThVQeRU/vHt1soKHHSr00El/VpZXWc08vaBgd+M+c963qR1WlERETqJS8vjx07dlQ+T0lJYe3atURGRtKmTRumTp3Kvn37ePfddwFz5c2kpCR69OhBSUkJ//3vf5k9ezazZ8+uvMb999/POeecwz/+8Q8uvfRSPv/8c77//nt++umnRv98Z1KxuMCvKV42L1pgC2jRziyi7V8DHc6zOpGIiIhX0Ug08Uqr9hziszX7sNngiUt6nna+FI+waY65bX8uBEVamURERKTeVq5cSd++fenb1xzZNGXKFPr27ctjjz0GQHp6OqmpqZXnl5SU8OCDD9KrVy9GjBjBTz/9xFdffcUVV1xRec6wYcP44IMPePvtt+nVqxfJycl8+OGHDB48uHE/XA30ah1OoK+DwwWlbMs8anWc2mnVz9xqXjQREZFa00g08Toul8Hf5pqLCVzTP5GzWodbnKgGKlo5e1xuaQwRERF3OPfcczEM45Q/T05OrvL84Ycf5uGHHz7jda+66iquuuqq+sZrcOa8aC1Ysv0gv+zMpmucl82L9ttsSFkCwx8Au75TFxERqSn9rSle5+NVaWzYl0Oovw8PXdDlzC+wWtZWyNwIdl+1coqIiDQRFS2dy1MOWZyklpJGmNtdC+C/l8PRjNOfLyIiIpVURBOvkltUyrPfbQXg/jGdiApp3BW56qRiFFqHUeZcJCIiIuL1KhYXWJ5y6LSj8jxOQh+49BXwDYJdC+G14bB9/pleJSIiIqiIJl7mpe+3czCvhA7RwUwcmmR1nJrZ+Jm5VSuniIhIk9GrdQQBvnYO5ZewPTPP6ji10/dGuGMhxJ4FBQfhvavgu0egrNjqZCIiIh5NRTTxGjsy80heuhuAxyb0wM/HC/7zzdwMWZvNVs4uF1qdRkRERNzEz8fOgLbmaLRfdmVbnKYOorvAbd/DoDvN58tmwqyxkL3T2lwiIiIezAuqECJgGAbTvtxEmctgTLcYRnaOtjpSzVS0cnYcDYERViYRERERN6to6fTKIhqAbwBc+Axc9wEERkL6OnjjHFj3gdXJREREPJKKaOIVFm3LYvG2LPwcdv7vou5Wx6kZw1Arp4iISBM2pEP54gK7vGxetBN1GQ9/+Bnang0lefDZnfDpHVB81OpkIiIiHkVFNPEKry40WwsmDm1LUlSwxWlqKHMzHNwKDj/z5lRERESalF6twwnwtZOdX8IOb5sX7URhCXDzXBj1f2BzwPoP4fURsG+11clEREQ8hopo4vHWph3h15RD+Dps3DaivdVxam7THHPbYTQEhFsaRURERNzP38dBvzbmytte29J5PLsDRj4Et3wN4YlwOAVmjYOlL4PLZXU6ERERy6mIJh7vX4vNUWiX9G5FXHiAxWlqSK2cIiIizcKQ9mZL5y8phyxO4kZthsBdS6DbJeAqhXn/Bx/dpEKaiIg0eyqiiUfbk53Pt79lAHDHOV40Ci1zExzcBg5/tXKKiIg0YRVFtOW7sr17XrQTBbaAa96Fi2eYU1Ns+RL2q7VTRESaNxXRxKO9uSQFlwHndommS1yo1XFqrmIUWscxEBBmbRYRERFpML0Tw/H3sXMwr4SdWV4+L9qJbDYYcAt0OM98nvartXlEREQspiKaeKxD+SV8vCoN8LJRaGrlFBERaTaqzovWhFo6j5c4yNymLbc2h4iIiMVURBOP9e6y3RSVujirVThDy1slvMKB3yB7R3kr5wVWpxEREZEGVjkvWlNYXOBkWpcX0fausDaHiIiIxVREE49UWOLk3WV7AHMUms1mszhRLVSMQus0Fvy9qAVVRERE6mRw+0jAHInWpOZFq9CqH9gckLsPcvZanUZERMQyKqKJR/pk9V4O5ZfQukUg43vGWR2n5tTKKSIi0uz0SYzAz8fOwbxidh3MtzqO+/kFQ1xPc1/zoomISDOmIpp4HKfL4M0luwC47ex2+Di86D/TjA1waBf4BEBntXKKiIg0BwG+Dvq1iQCacEtn4mBzq5ZOERFpxryoOiHNxbyNGezJLiA80JdrBiZaHad2qrRyhlibRURERBrN4HYV86I10cUFWmtxARERERXRxKMYhsEbi81RaDcNaUuQn4/FiWpBrZwiIiLNVsXiAst3ZTfNedEqVuhMXw+lhdZmERERsYiKaOJRVuw+zNq0I/j52Ll5WJLVcWonfR0cTgGfQOh0vtVpREREpBH1bWPOi5Z5tJiUpjgvWkQbCIkFVynsX2t1GhEREUtYXkR79dVXadeuHQEBAfTv358lS5ac9vz33nuP3r17ExQURHx8PLfccgvZ2U107olm6F+LdwJwZb9WRIf6W5ymlipGoXUep1ZOERGRZibA10GfxAigibZ02mzHRqPt1eICIiLSPFlaRPvwww+ZPHkyjzzyCGvWrGHEiBGMHz+e1NTUk57/008/MXHiRG699VY2btzIxx9/zIoVK7jtttsaObk0hB2ZR/l+cyY2G9w2or3VcWpHrZwiIiLNXmVLZ0oT/YK3cl40FdFERKR5srSI9vzzz3Prrbdy22230a1bN2bMmEFiYiKvvfbaSc//5ZdfSEpK4r777qNdu3acffbZ3HnnnaxcubKRk0tD+PfiFADGdIulQ7SXjeTavwaO7AHfIOg0zuo0IiIiYoEh7SMBc4XOpjkvWvkKnWm/ml8gioiINDOWFdFKSkpYtWoV48ZVLTiMGzeOpUuXnvQ1w4YNY+/evXz99dcYhsGBAwf45JNPuOiii075PsXFxeTm5lZ5iOfJzC3iszX7ALjzHC8bhQbHtXKeD37B1mYRERERS/Rr0wI/h50DucXszi6wOo77xfcGuy/kZ8Lh3VanERERaXSWFdEOHjyI0+kkNja2yvHY2FgyMjJO+pphw4bx3nvvce211+Ln50dcXBwRERG8/PLLp3yf6dOnEx4eXvlITEx06+cQ93hn2W5KnC76tYlgQFKk1XFqxzBg0xxzX62cIiIizdbx86It39UEWzp9AyChj7m/d4WlUURERKxg+cICNputynPDMKodq7Bp0ybuu+8+HnvsMVatWsW3335LSkoKd9111ymvP3XqVHJyciofaWlpbs0v9ZdfXMZ/fzHnwbvjnA7WBVn4D3iuG/z4/6DwSM1ft381HEk1Wzk7jm2weCIiIuL5jm/pbJIq50Vbbm0OERERC1hWRIuKisLhcFQbdZaZmVltdFqF6dOnM3z4cB566CF69erF+eefz6uvvspbb71Fenr6SV/j7+9PWFhYlYd4lg9XpJFTWEq7qGDGdj/5n32DK86Dn2fA0f2w+Bl4sRcsehaKj575tZWtnBeAX1CDxhQRERHPNrh8cYFfdh1qovOiaXEBERFpviwrovn5+dG/f3/mz59f5fj8+fMZNmzYSV9TUFCA3V41ssPhAGiaNynNQJnTxayfzAUFbhvRDof95KMQG9zmL6C0AMJaQXQ3KMqBBU/CjF7w84tQcop5TQwDNs4x99XKKSIi0uz1a9MCX4eNjNwiUg81wXnRKopoB34zv4QUERFpRixt55wyZQpvvvkmb731Fps3b+aBBx4gNTW1sj1z6tSpTJw4sfL8CRMm8Omnn/Laa6+xa9cufv75Z+677z4GDRpEQkKCVR9D6uGrDensO1JIy2A/ruzX2rog6z8wt/0nwR9+hivehMgOUHgI5j8GL/aGX16D0qKqr9u3CnLSwDcYOqmVU0REpLkL9Ds2L1qTbOkMS4DwRDBc5pQWIiIizYilRbRrr72WGTNmMG3aNPr06cPixYv5+uuvadu2LQDp6emkpqZWnj9p0iSef/55Zs6cSc+ePbn66qvp0qULn376qVUfQerBMAz+tXgXADcPSyLA12FNkNz9sGuRuX/W1WB3QK+r4Z5f4dJXIaKtuQrVt3+Bl/rCillQVmKeX9HK2WU8+AZak19EREQ8yuB2Zkvn8l2HLE7SQFoPNLeaF01ERJoZH6sD3H333dx9990n/VlycnK1Y/feey/33ntvA6eSxrB0ZzYb9+cS4GvnxiFtrQuy4WPAgMQhENnu2HGHD/S9wSysrX0PFj8Lufvgqynm/GnnPKxWThEREalmSPuWzFywg192ZZ920SyvlTgINn4KaVqhU0REmhfLV+eU5uuN8lFo1wxIJDLYz7og6z8yt72vPfnPffxgwC1w72oY/wyExJqrcc79I+TuBb8Q6Dim8fKKiIiIR+vXNgJfh439OUWkHSq0Oo77VcyLtvdXc35YERGRZkJFNLHE5vRcFm/Lwm6D285ub12QjN/MiXEdfmceTeYbAIPvhPvWwrgnIchs1aDHZebPRERERIAgPx96tY4A4JeUJjgvWuxZ4BMAhYche4fVaURERBqNimhiiTcW7QRgfM942rQMsi5IxYICnc+HwBY1e41fEAy7F+5fD9d/ZI5OExERETnOkPaRQBNdXMDHDxL6mfuaF01ERJoRFdGk0e3MymPuuv0A3DWyg3VBXE7Y8Im53+t3tX+9f4hZfPMLdm8uERER8XpD2jfxxQUSKxYX+NXaHCIiIo1IRTRpdC/9sB2XAWO6xXBW63DrgqQsgqPp5gi0TuOsyyEiIiJNTv+2LfCx29h3pJC0QwVWx3G/xMHmVkU0ERFpRlREk0a1I/No5Si0yWM6Wxtm3YfmtscVZluCiIiIiJuY86KZXxY2yZbO1uWLC2RtgcIjlkYRERFpLCqiSaN68YcdGAaM6x5Lz1YWjkIryYfNX5j7vU6xKqeIiIhIPVS0dP7SFFs6Q6KhRTvAgH0rrU4jIiLSKFREk0az7cBRvlzvIaPQtnwFpfnmzV/FMu0iIiIibjS4vIi2dOdBDMOwOE0DqLiHSlthbQ4REZFGoiKaNJoXv9+OYcAFPeLonhBmbZh15aty9roWbDZrs4iIiEiTNLhdJIG+DtJziti4P9fqOO5XWUTTCp0iItI8qIgmjWJLRi5fbUgHYPLYTtaGOZoBuxaY+72usTaLiIiINFkBvg7O6RwFwLxNByxO0wAq5kXbt8pc9VxERKSJUxFNGsWM+dsBuOiseLrGWTwKbcMnYLjMG7+WHazNIiIiIk3auO5xAMzbmGFxkgYQ0x38QqA411xgQEREpIlTEU0a3Mb9OXy7MQObDe4fY/EoNID15a2cvbWggIiIiDSs87rG4LDb2JJxlLRDBVbHcS+HD7TqZ+6n/WptFhERkUagIpo0uBe/PzYKrXNsqLVhDmyCjA1g94UeV1ibRURERJq8FsF+DExqATTRls7EweZ2rxYXEBGRpk9FNGlQv+3LYd6mA9hsMNmTRqF1GgdBkdZmERERkWahSbd0ttbiAiIi0nyoiCYNasb32wC4pHcCHWMsHoXmcsL6j819tXKKiIhIIxnbPRaAFbsPcSi/xOI0btZ6gLnN3gH52dZmERERaWAqokmDWb/3CN9vzsRug/tGe8AotN0/wdH9EBAOnS+wOo2IiIg0E4mRQXSLD8NlwI9bMq2O415BkRDV2dxXS6eIiDRxKqJJg5lRPhfaZX1a0SE6xOI0wPoPzW2Py8HH39osIiIi0qyMKx+N1qRbOvdqcQEREWnaVESTBrE27Qg/bsnEYbdxryeMQispgE2fm/u9fmdtFhERES+3ePFiJkyYQEJCAjabjTlz5pz2/E8//ZSxY8cSHR1NWFgYQ4cO5bvvvqtyTnJyMjabrdqjqKioAT9J4xnXwyyiLd6eRWGJ0+I0bpZYMS+aimgiItK0qYgmDeKF+eZcaJf1aUW7qGCL0wBbv4aSPIhoC22GWJ1GRETEq+Xn59O7d29mzpxZo/MXL17M2LFj+frrr1m1ahWjRo1iwoQJrFmzpsp5YWFhpKenV3kEBAQ0xEdodN3jw2gVEUhRqYufdhy0Oo57VRTR9q0CZ5m1WURERBqQj9UBpOlZtecwi7Zl4bDbuG90R6vjmNaVr8rZ61qw2azNIiIi4uXGjx/P+PHja3z+jBkzqjx/6qmn+Pzzz/niiy/o27dv5XGbzUZcXJy7YnoUm83G2O6xJC/dzbyNGZWLDTQJUV3APxyKc+DAb5DQx+pEIiIiDUIj0cTtKlbkvLJfK9q29IBRaHmZsPNHc7+XVuUUERGxmsvl4ujRo0RGRlY5npeXR9u2bWndujUXX3xxtZFqJyouLiY3N7fKw5NVtHT+sCUTp8uwOI0b2e3HVunU4gIiItKEqYgmbrVy9yGWbD+Ij93Gved5wFxoABs+AcMJrfpDlIeMjBMREWnGnnvuOfLz87nmmmsqj3Xt2pXk5GTmzp3L+++/T0BAAMOHD2f79u2nvM706dMJDw+vfCQmJjZG/DoblBRJeKAvh/JLWLXnsNVx3CtxsLlNW25tDhERkQakIpq41Qvlo9Cu6t+axMggi9OUq1iVUwsKiIiIWO7999/nb3/7Gx9++CExMTGVx4cMGcKNN95I7969GTFiBB999BGdO3fm5ZdfPuW1pk6dSk5OTuUjLS2tMT5Cnfk47Izuan7mJrdKZ+JAc6vFBUREpAlTEU3c5teUQ/y8Ixtfh417RnnIiK+srZC+Fuw+0PNKq9OIiIg0ax9++CG33norH330EWPGjDntuXa7nYEDB552JJq/vz9hYWFVHp6uoqVz/uYDGEYTaulsNQCwwZE9cPSA1WlEREQahIpo4jYVK3JePSDRc0ahVSwo0HEsBLe0NouIiEgz9v777zNp0iT+97//cdFFF53xfMMwWLt2LfHx8Y2QrvGM6BSNn4+dPdkFbDuQZ3Uc9wkIg5ju5v5ejUYTEZGmSUU0cYtlO7NZtsvDRqG5XLDhY3O/txYUEBERcZe8vDzWrl3L2rVrAUhJSWHt2rWkpqYCZpvlxIkTK89///33mThxIs899xxDhgwhIyODjIwMcnJyKs954okn+O6779i1axdr167l1ltvZe3atdx1112N+tkaWrC/DyM6RgFNsaVzkLnVvGgiItJEqYgmblExF9rvBrahVUSgxWnK7fkZctLMJdc7j7c6jYiISJOxcuVK+vbtS9++fQGYMmUKffv25bHHHgMgPT29sqAG8MYbb1BWVsY999xDfHx85eP++++vPOfIkSPccccddOvWjXHjxrFv3z4WL17MoEGDGvfDNYLjWzqblMoimlboFBGRpsnH6gDi/XZl5fFryiF87DbuHtXB6jjHrC9v5exxKfgGWJtFRESkCTn33HNPO59XcnJylecLFy484zVfeOEFXnjhhXom8w7ndY3FZtvA+r057D9SSIKnfAFZXxUrdO5fA2Ul4ONnbR4RERE300g0qbf5m8xvUYd2aEl8uIfcBJYWwqa55n4vtXKKiIiI54gO9ad/mxYAfN+URqNFtoegluAshoz1VqcRERFxOxXRpN7mlRfRxnaPtTjJcbZ9C8W5EJ4IbYZZnUZERESkisqWzk1NqIhms0FrzYsmIiJNl4poUi9ZR4tZnXoYgDHdPKiItuETc3vWVWDXf+YiIiLiWcZ2jwPMxZlyCkstTuNGiQPNbZpW6BQRkaZH1QWplx82H8AwoFfrcM+Zz6PwMGyfZ+6fdY21WUREREROol1UMJ1iQihzGSzcmml1HPepmBdNRTQREWmCVESTeqls5fSkUWibPgdnCcT0gNjuVqcREREROamKls55TamlM6Ev2BxwdD/k7LU6jYiIiFupiCZ1ll9cxk87DgIwrkecxWmOs/5jc9vramtziIiIiJxGRUvnwi2ZFJc5LU7jJn7BEHeWua950UREpIlREU3qbPG2LErKXLSJDKJzbIjVcUw5e2HPT+Z+z6uszSIiIiJyGr1ahRMb5k9+iZNlO7OtjuM+rfqb2/R11uYQERFxMxXRpM4qVpMa1z0Wm81mcZpyv802t22HQ0SitVlERERETsNut1Wubt6kWjrjeprbAxutzSEiIuJmKqJJnZQ6XfywxZwE1yNbOc9SK6eIiIh4voqWzvmbDuByGRancZNYFdFERKRpUhFN6mRFyiFyCkuJDPajf9sWVscxZW6GAxvA7gvdL7U6jYiIiMgZDW3fklB/H7KOFrNu7xGr47hHTDdzezQd8ptQm6qIiDR7KqJJnVS0HIzuGoPD7iGtnOs/MredxkJQpLVZRERERGrAz8fOuV1jgCbU0ukfCi2SzP1MjUYTEZGmQ0U0qTXDMI7Nh+YprZwuF2z4xNxXK6eIiIh4kcp50TZmWJzEjdTSKSIiTZCKaFJrm9Jz2XekkABfO2d3jLI6jmnvr5CTCn4h0GW81WlEREREauzcLtH4OmzszMpnZ1ae1XHcI7aHuT3wm7U5RERE3EhFNKm1eRvNUWjndIom0M9hcZpyFa2c3SaAb6C1WURERERqISzAl6EdzC8m5zeVls7KIppGoomISNOhIprUWsV8HRWtB5ZzlsLGz8x9tXKKiIiIF2pyLZ0V7ZyZm8HltDaLiIiIm6iIJrWSdqiAzem52G0wupuHFNF2/ACFhyA4BtqNtDqNiIiISK2NLb+vWpN2hMyjRRancYMWSeAbBGVFcGiX1WlERETcQkU0qZWKFoOBSZFEBvtZnKbchvJWzp5XgsPH2iwiIiIidRAXHkDvxAgMA37YnGl1nPqzOyCmm7mvedFERKSJUBFNamXeJrPFwGNaOYvzYMvX5n4vtXKKiIiI9xpXfn+ledFEREQ8k+VFtFdffZV27doREBBA//79WbJkySnPnTRpEjabrdqjR48ejZi4+TqcX8KK3YcBGNc9zuI05bZ8BWWFENkeEvpZnUZERESkziqKaD/tOEhecZnFadygYl40FdFERKSJsLSI9uGHHzJ58mQeeeQR1qxZw4gRIxg/fjypqaknPf/FF18kPT298pGWlkZkZCRXX60RSI3hxy2ZOF0GXeNCadMyyOo4popWzrOuAZvN2iwiIiIi9dAxJoT2UcGUlLmYv6kJLDBQORJN7ZwiItI0WFpEe/7557n11lu57bbb6NatGzNmzCAxMZHXXnvtpOeHh4cTFxdX+Vi5ciWHDx/mlltuaeTkzVNFa8E4T2nlzMuCnQvM/V7XWJtFREREpJ5sNhuX9mkFwKer91mcxg1iupvbI6lQlGNtFhERETewrIhWUlLCqlWrGDduXJXj48aNY+nSpTW6xqxZsxgzZgxt27Y95TnFxcXk5uZWeUjtFZU6WbQtC4BxPTyklXPjZ2A4zTbOlh2sTiMiIiJSb5f1TQDg5x0Hycz18lU6gyIhzCwKkrnZ2iwiIiJuYFkR7eDBgzidTmJjq45qio2NJSPjzMPX09PT+eabb7jttttOe9706dMJDw+vfCQmJtYrd3P10/aDFJY6SQgPoEdCmNVxTBWtnBqFJiIiIk1E25bB9G/bApcBc9fttzpO/amlU0REmhDLFxawnTCPlWEY1Y6dTHJyMhEREVx22WWnPW/q1Knk5ORUPtLS0uoTt9mqaOUc2z22Rn8+De7QLti7Amx26HGF1WlERERE3Obyvk2opVMrdIqISBNiWREtKioKh8NRbdRZZmZmtdFpJzIMg7feeoubbroJPz+/057r7+9PWFhYlYfUjtNl8P3m8vnQPKWVc8Mn5rbdSAj1kDnaRERERNzgorPi8XXY2JSey9aMo1bHqR+t0CkiIk2IZUU0Pz8/+vfvz/z586scnz9/PsOGDTvtaxctWsSOHTu49dZbGzKilFuTepjs/BLCAnwY1C7S6jhgGLBerZwiIiLSNLUI9mNUlxgAPlvj5aPRKkeibQKXy9osIiIi9WRpO+eUKVN48803eeutt9i8eTMPPPAAqamp3HXXXYDZijlx4sRqr5s1axaDBw+mZ8+ejR25WZpX3sp5XtcYfB2WdwBD+jrI3g4+AdD1YqvTiIiIiLjdFf3Mls7P1+7D5TIsTlMPLTuCww9KjkJOqtVpRERE6sXHyje/9tpryc7OZtq0aaSnp9OzZ0++/vrrytU209PTSU2t+pdtTk4Os2fP5sUXX7QicrNjGAbzNpott2O7e0or58fmtvMFEKD2XBEREWl6RnWNISzAh/ScIn7Zlc2wjlFWR6obhy9Ed4GMDWZLZ4skqxOJiIjUmaVFNIC7776bu++++6Q/S05OrnYsPDycgoKCBk4lFXZk5rE7uwA/h52RXaKtjgMu57H50NTKKSIiIk2Uv4+Di3ol8P6vqXy6Zp/3FtHAnBetoojW9SKr04iIiNSZB/TmiSeraOUc3rElIf6W11xh9xLIy4CACOg41uo0IiIiIg2moqXz298yKCxxWpymHirnRfvN2hwiIiL1pCKanJbHtXKuL2/l7HEZ+Jx+ZVYRERERbzagbQtatwgkr7iM+eUrpXulyiKaVugUERHvpiKanFJGThHr9uZgs8GY7jFWx4HSItg819w/S62cIiIi0rTZbDYu72uORvts9V6L09RDbPliYNk7oUTTsoiIiPdSEU1OqeIbz76JEcSEBlicBtj+HRTnQlhraDPU6jQiIiIiDa6iiLZ4+0EO5hVbnKaOQmIgOBowIGuz1WlERETqTEU0OaX55fOheU4r50fm9qwrwa7/dEVERKTpax8dQu/ECJwugy/W7bc6Tt2ppVNERJoAVSLkpHKLSlm28yAA43rEWpwGKDwM2+eZ+2rlFBERkWbk8j4JAHy2Zp/FSeqhoqVTRTQREfFiKqLJSS3cmkWp06B9dDAdokOsjgPbvgNnCcR0h7ieVqcRERERaTQTeifgY7exfm8OOzLzrI5TNxqJJiIiTYCKaHJSFa2c4zyllTN9nbltf66lMUREREQaW8sQf0Z2jgZgjreORqssov0GhmFtFhERkTqqUxFt4cKFbo4hnqSkzMXCLZmAh7RygnnDBcduwERERESakcv7la/SuWYfLpcXFqGiuoDNYU7RcTTd6jQiIiJ1Uqci2gUXXECHDh148sknSUtLc3cmsdiyXdkcLS4jOtSfPq0jrI5jfluZoSKaiIiINF9jusUS6u/DviOFrNh9yOo4tecbAFGdzH21dIqIiJeqUxFt//793H///Xz66ae0a9eO888/n48++oiSkhJ35xMLfL7WbBM4v0csdrvN4jRA3gEoPAQ2O0R3tTqNiIiISKML8HUw/ixzmo05a5tAS6eIiIgXqlMRLTIykvvuu4/Vq1ezcuVKunTpwj333EN8fDz33Xcf69atc3dOaSQFJWV891sGAJf3bWVxmnIV31a27Ai+gdZmEREREbHI5X1bA/Dl+nSKSp0Wp6kDLS4gIiJert4LC/Tp04e//OUv3HPPPeTn5/PWW2/Rv39/RowYwcaN+gvS28zfdID8EieJkYH0a9PC6jimihutmO7W5hARERGx0OB2kSSEB3C0qIwfy+ev9Sqx5Susq4gmIiJeqs5FtNLSUj755BMuvPBC2rZty3fffcfMmTM5cOAAKSkpJCYmcvXVV7szqzSCz9fuB+CyPq2w2TyglROO3WhV3HiJiIiINEN2u41LyzsFPl3thS2dFSPRDm6DsmJrs4iIiNRBnYpo9957L/Hx8dx111107tyZNWvWsGzZMm677TaCg4NJTEzk6aefZsuWLe7OKw0oO6+YRduyALi0j4e0csJxRTQtKiAiIiLN2xXlRbSFWzM5lO9l8xGHtYKAcHCVmYU0ERERL1OnItqmTZt4+eWX2b9/PzNmzKBnz+ojhBISEliwYEG9A0rj+XJ9Ok6XQa/W4XSMCbE6jslZClnlxVgV0URERKSZ6xQbSo+EMMpcBl+t3291nNqx2dTSKSIiXq1ORbQffviB6667Dj8/v1Oe4+Pjw8iRI+scTBrfZ2vMtoDLPGkU2sHt4CoFv1CIaGN1GhERERHLVSz+9OkaL27p1AqdIiLihepURJs+fTpvvfVWteNvvfUW//jHP+odShrf7oP5rE07gt0GF/eOtzrOMce3cnrKHG0iIiLN3OLFi5kwYQIJCQnYbDbmzJlzxtcsWrSI/v37ExAQQPv27Xn99dernTN79my6d++Ov78/3bt357PPPmuA9N7vkj4J2G2wJvUIuw/mWx2ndrRCp4iIeLE6FdHeeOMNunbtWu14jx49TnpDJJ5vzlrzm8yzO0UTExpgcZrjVHxLqVZOERERj5Gfn0/v3r2ZOXNmjc5PSUnhwgsvZMSIEaxZs4a//vWv3HfffcyePbvynGXLlnHttddy0003sW7dOm666SauueYali9f3lAfw2vFhAYwolM0cKyTwGuonVNERLyYT11elJGRQXx89dFK0dHRpKen1zuUNC7DMJhTfgN2ed8Ei9OcQIsKiIiIeJzx48czfvz4Gp//+uuv06ZNG2bMmAFAt27dWLlyJf/85z+58sorAZgxYwZjx45l6tSpAEydOpVFixYxY8YM3n//fbd/Bm93ed9WLNqWxZy1+5g8ppPnrKp+JtFdARvkHYC8LAiJtjqRiIhIjdVpJFpiYiI///xzteM///wzCQkeVoSRM1qbdoTd2QUE+joY1z3O6jhVVRbRqi9eISIiIt5h2bJljBs3rsqx888/n5UrV1JaWnrac5YuXXrK6xYXF5Obm1vl0VyM6xFLkJ+DPdkFrE49bHWcmvMPgch25n6mRqOJiIh3qVMR7bbbbmPy5Mm8/fbb7Nmzhz179vDWW2/xwAMPcPvtt7s7ozSwilFo43rEEuxfp8GJDaPgEBwtX3Uqppu1WURERKTOMjIyiI2NrXIsNjaWsrIyDh48eNpzMjIyTnnd6dOnEx4eXvlITEx0f3gPFeTnwwU9zS8/va+lU/OiiYiId6pTEe3hhx/m1ltv5e6776Z9+/a0b9+ee++9l/vuu69yCL54h1Kniy/Xmy24l/X1oFU5ATI3mduINhAQZm0WERERqZcT2w0Nw6h2/GTnnK5NcerUqeTk5FQ+0tLS3JjY81Ws0vnl+nRKylwWp6mFynnRNlmbQ0REpJbqNOzIZrPxj3/8g0cffZTNmzcTGBhIp06d8Pf3d3c+aWA/bT9Idn4JLYP9GNExyuo4VamVU0REpEmIi4urNqIsMzMTHx8fWrZsedpzThyddjx/f/9mff85rEMUsWH+HMgtZsHWTM7v4WHTcpxKTHdzW7GAlIiIiJeo00i0CiEhIQwcOJCePXs26xsYb1Yx/H9C7wR8HPX6z8H9tDKniIhIkzB06FDmz59f5di8efMYMGAAvr6+pz1n2LBhjZbT2zjsNi7rY45Ge295qsVpaqHi3i5rCzjLrM0iIiJSC3WeAGvFihV8/PHHpKamUlJSUuVnn376ab2DScPLKy5j3ibzG1+Pa+UErcwpIiLiofLy8tixY0fl85SUFNauXUtkZCRt2rRh6tSp7Nu3j3fffReAu+66i5kzZzJlyhRuv/12li1bxqxZs6qsunn//fdzzjnn8I9//INLL72Uzz//nO+//56ffvqp0T+fN7lhcFv+tWQXi7dlsTMrjw7RIVZHOrMW7cA3CEoL4NAuiO5sdSIREZEaqdPQow8++IDhw4ezadMmPvvsM0pLS9m0aRM//vgj4eHh7s4oDWTexgyKSl20iwqmd2sP+3NzOSFzs7mvdk4RERGPsnLlSvr27Uvfvn0BmDJlCn379uWxxx4DID09ndTUYyOj2rVrx9dff83ChQvp06cPf//733nppZe48sorK88ZNmwYH3zwAW+//Ta9evUiOTmZDz/8kMGDBzfuh/MybVoGMbqr2fL67tLd1oapKbtdLZ0iIuKV6jQS7amnnuKFF17gnnvuITQ0lBdffJF27dpx5513Eh8f7+6M0kAqWjkv69PqtJP2WuLwbvPbSZ8AiGxvdRoRERE5zrnnnlu5MMDJJCcnVzs2cuRIVq9efdrrXnXVVVx11VX1jdfs3DI8ie83H+CTVXv50/ldCAvwtTrSmcX2gH0rzc6DnldYnUZERKRG6jQSbefOnVx00UWAOaFrfn4+NpuNBx54gH/9619uDSgNIzO3iJ93mEvKX9Y3weI0J1HxrWRMN7A7rM0iIiLSRLzzzjt89dVXlc8ffvhhIiIiGDZsGHv27LEwmdTHsA4t6RQTQn6Jk09W7rU6Ts1UrtC50docIiIitVCnIlpkZCRHjx4FoFWrVvz2m1nwOHLkCAUFBe5LJw1m7rr9uAzo1yaCti2DrY5TneZDExERcbunnnqKwMBAAJYtW8bMmTN55plniIqK4oEHHrA4ndSVzWbj5mFJALy7bDcu16lHCXqMins8FdFERMSL1KmINmLEiMrVk6655hruv/9+br/9dq677jpGjx7t1oDSMD5fux/w0AUF4LgimuZDExERcZe0tDQ6duwIwJw5c7jqqqu44447mD59OkuWLLE4ndTHFf1aERrgw+7sAhZty7I6zpnFls+JlpMKRTnWZhEREamhOhXRZs6cye9+9zsApk6dyoMPPsiBAwe44oormDVrllsDivvtyDzKhn05+NhtXHSWh85hV9HOqZFoIiIibhMSEkJ2djYA8+bNY8yYMQAEBARQWFhoZTSppyA/H343MBGAt71hgYHAFhDW2tw/sMnaLCIiIjVU6yJaWVkZX3zxBXa7+VK73c7DDz/M3Llzef7552nRooXbQ4p7zVljjkIb2TmaliH+Fqc5ieI8c2EBgBgV0URERNxl7Nix3Hbbbdx2221s27atco7bjRs3kpSUZG04qbebhiRhs8HibVnszMqzOs6ZVbZ0aoVOERHxDrUuovn4+PCHP/yB4uLihsgjDcwwDOasLV+V01NbOTM3m9uQOAhuaW0WERGRJuSVV15h6NChZGVlMXv2bFq2NP+eXbVqFdddd53F6aS+2rQMYnTXWADe9YbRaJoXTUREvIxPXV40ePBg1qxZQ9u2bd2dRxrYqj2H2Xu4kGA/B2O6xVod5+TUyikiItIgIiIimDlzZrXjTzzxhAVppCHcMjyJ7zcf4JNVe/nT+V0IC/C1OtKpqYgmIiJepk5FtLvvvps//elP7N27l/79+xMcXHV1x169erklnLjfZ2vMUWgX9Iwn0M9hcZpT0MqcIiIiDeLbb78lJCSEs88+GzBHpv373/+me/fuvPLKK5qWowkY1qElnWJC2J6Zxycr9/L7s9tZHenUKhaQytwELhfY6zRds4iISKOp099U1157LSkpKdx3330MHz6cPn360Ldv38qteKaSMhdfrk8H4HJPbeUErcwpIiLSQB566CFyc3MB2LBhA3/605+48MIL2bVrF1OmTLE4nbiDzWbj5mFJALy7bDcul2FtoNNp2REcflCSB0f2WJ1GRETkjOo0Ei0lJcXdOaQRLNyaSU5hKTGh/gzt4KFzjRmGRqKJiIg0kJSUFLp37w7A7Nmzufjii3nqqadYvXo1F154ocXpxF2u6NeKf3y7hd3ZBSzalsWorjFWRzo5hw9Ed4WM9eb9X6QHj5oTERGhjkU0zYXmnT5fa67KeUnvBBx2m8VpTiFnLxTngN0HojpbnUZERKRJ8fPzo6CgAIDvv/+eiRMnAhAZGVk5Qk28X5CfD78bmMi/l6Tw9tLdnltEA7PzoKKI1u1iq9OIiIicVp2KaO++++5pf15xQyaeI7eolPmbDwAevConHBuFFtUFfPyszSIiItLEnH322UyZMoXhw4fz66+/8uGHHwKwbds2WrdubXE6caeJQ5N486cUFm/LYkdmHh1jQqyOdHKViwv8Zm0OERGRGqhTEe3++++v8ry0tJSCggL8/PwICgpSEc0Dfbshg5IyF51iQuiREGZ1nFPTypwiIiINZubMmdx999188sknvPbaa7RqZX6x9s0333DBBRdYnE7cKTEyiNFdY/l+8wHeXbabaZd66FyzWqFTRES8SJ2KaIcPH652bPv27fzhD3/goYceqncocb+KVTkv69sKm81DWzlB86GJiIg0oDZt2vDll19WO/7CCy9YkEYa2i3Dk/h+8wFmr9rLg+d3ISzA1+pI1VUsJHVoF5Tkg1+wtXlEREROo05FtJPp1KkTTz/9NDfeeCNbtmxx12XFDdJzCvklJRuAS/skWJzmDDI3mVutzCkiItIgnE4nc+bMYfPmzdhsNrp168all16Kw+GwOpq42bAOLekUE8L2zDw+WbmX35/tgRP3h0RDcAzkZ0LmFmjd3+pEIiIip2R358UcDgf79+935yXFDeau3Y9hwKCkSFq3CLI6zqmVFsHB7eZ+bHdrs4iIiDRBO3bsoFu3bkycOJFPP/2UTz75hJtuuokePXqwc+dOq+OJm9lsNiYNTwLgnWW7cbkMawOdiuZFExERL1GnkWhz586t8twwDNLT05k5cybDhw93SzBxny/XpwNwaV8PH4V2cCsYTghsAaHxVqcRERFpcu677z46dOjAL7/8QmRkJADZ2dnceOON3HfffXz11VcWJxR3u7xvK/7xzRb2ZBewcFsm53WNtTpSdbE9YNcCzYsmIiIer05FtMsuu6zKc5vNRnR0NOeddx7PPfecO3KJm+w9XMCGfTnYbXB+jzir45xe5XxoPcGT520TERHxUosWLapSQANo2bIlTz/9tL4IbaKC/Hy4dmAi/16SQvLSPR5aRCufxiN9nbU5REREzqBO7Zwul6vKw+l0kpGRwf/+9z/i42s3gujVV1+lXbt2BAQE0L9/f5YsWXLa84uLi3nkkUdo27Yt/v7+dOjQgbfeeqsuH6NZ+G7jAQAGJkUSFeJvcZoz0KICIiIiDcrf35+jR49WO56Xl4efn58FiaQxTByahM0Gi7dlsSMzz+o41bUZbG73rYJiD8wnIiJSzq1zotXWhx9+yOTJk3nkkUdYs2YNI0aMYPz48aSmpp7yNddccw0//PADs2bNYuvWrbz//vt07dq1EVN7l29/M1s5L+jp4aPQ4Ng8GCqiiYiINIiLL76YO+64g+XLl2MYBoZh8Msvv3DXXXdxySWXWB1PGkhiZBBjupkj0N5dttvaMCcT2R4i2oKrFPb8bHUaERGRU6pTEe2qq67i6aefrnb82Wef5eqrr67xdZ5//nluvfVWbrvtNrp168aMGTNITEzktddeO+n53377LYsWLeLrr79mzJgxJCUlMWjQIIYNG3bK9yguLiY3N7fKo7nIPFrEyj2HAS9o5QSNRBMREWlgL730Eh06dGDo0KEEBAQQEBDAsGHD6NixIzNmzLA6njSgW4YlAfDJqr3kFpVaG+ZkOowytzt/tDaHiIjIadSpiLZo0SIuuuiiascvuOACFi9eXKNrlJSUsGrVKsaNG1fl+Lhx41i6dOlJXzN37lwGDBjAM888Q6tWrejcuTMPPvgghYWFp3yf6dOnEx4eXvlITEysUb6mYP6mAxgG9E6MICEi0Oo4p5eXCflZgA2iu1mdRkREpEmKiIjg888/Z9u2bXzyySd8/PHHbNu2jc8++4yIiAir40kDGtqhJZ1jQygocfLJyr1Wx6muw3nmducCa3OIiIicRp0WFjjVvBm+vr41Hul18OBBnE4nsbFVJzeNjY0lIyPjpK/ZtWsXP/30EwEBAXz22WccPHiQu+++m0OHDp1yXrSpU6cyZcqUyue5ubnNppD27W/m7/ECrxiFVt7K2bID+AVZm0VERKQJOf4+6GQWLlxYuf/88883cBqxis1m4+ZhSTzy2W+8s2w3k4YlYbd70EJO7c4Bm91crT1nH4S3sjqRiIhINXUqovXs2ZMPP/yQxx57rMrxDz74gO7du9fqWrYTVmE0DKPasQoulwubzcZ7771HeHg4YN7sXXXVVbzyyisEBlYfbeXv74+/v4dPqN8AcgpKWbYzG4Dze3jgKkwnUiuniIhIg1izZk2NzjvV/Zc0HZf3bcU/vtnCnuwCFm7L9KyVOgNbQEI/2LcSdi2AvjdanUhERKSaOhXRHn30Ua688kp27tzJeeeZQ69/+OEH3n//fT7++OMaXSMqKgqHw1Ft1FlmZma10WkV4uPjadWqVWUBDaBbt24YhsHevXvp1KlTXT5Ok/T95gOUuQy6xIbSPjrE6jhndmCTua1Y4lxERETcYsECtceJKcjPh98NasO/Fu/inaV7PKuIBua8aPtWmi2dKqKJiIgHqtOcaJdccglz5sxhx44d3H333fzpT39i7969fP/991x22WU1uoafnx/9+/dn/vz5VY7Pnz//lAsFDB8+nP3795OXd2zp623btmG322ndunVdPkqT9e3G8lZOb1iVE461c8bUbiSjiIiIiNTcdYPaAPDzjoPkFHjYAgPtyxcX2LUAXC5rs4iIiJxEnYpoABdddBE///wz+fn5HDx4kB9//JGRI0fW6hpTpkzhzTff5K233mLz5s088MADpKamctdddwHmfGYTJ06sPP/666+nZcuW3HLLLWzatInFixfz0EMP8fvf//6krZzNVX5xGYu3ZQFeUkRzlkHWFnNf7ZwiIiIiDaZdVDBdYkMpcxks2JppdZyqWg8EvxAoyIYDG6xOIyIiUk2dimgrVqxg+fLl1Y4vX76clStX1vg61157LTNmzGDatGn06dOHxYsX8/XXX9O2bVsA0tPTSU1NrTw/JCSE+fPnc+TIEQYMGMANN9zAhAkTeOmll+ryMZqshVuzKC5z0bZlEF3jQq2Oc2bZO8BZYt40RbS1Oo2IiIhIkzaufL7ceZtOvpiXZXz8IOlsc3/nj9ZmEREROYk6FdHuuece0tLSqh3ft28f99xzT62udffdd7N7926Ki4tZtWoV55xzTuXPkpOTq6wYBdC1a1fmz59PQUEBaWlpPPfccxqFdoLjWzm9YpLg41s57XUeHCkiIiIiNTCuu9mpsHBrFkWlTovTnKCDOd8yOzWXn4iIeJ46VSw2bdpEv379qh3v27cvmzZtqncoqbuiUic/bj4AwAU9vKCVE7Qyp4iIiEgj6tkqjITwAApKnPy846DVcaqqmBct9RcoKbA2i4iIyAnqVETz9/fnwIED1Y6np6fj41OnBT/FTX7ecZD8EidxYQH0bh1hdZyaURFNREREpNHYbDbGlX/ZOm9j9Xt6S0V1grBW4CyG1KVWpxEREamiTkW0sWPHMnXqVHJyciqPHTlyhL/+9a+MHTvWbeGk9r797Vgrp93uBa2ccFwRrae1OURERESaiXHdzXnRvt98AKfLsDjNcWw26FA+Gk0tnSIi4mHqVER77rnnSEtLo23btowaNYpRo0bRrl07MjIyeO6559ydUWqozOlifnkr5/ne0spZeBhy95r7sd2tzSIiIiLSTAxsF0l4oC/Z+SWsTj1sdZyq2quIJiIinqlORbRWrVqxfv16nnnmGbp3707//v158cUX2bBhA4mJie7OKDW0POUQRwpKiQz2Y2BSC6vj1MyB8jn0wttAQLi1WURERESaCV+HndFdYwCYt9HDVulsf665zdwIRz2s3VRERJq1Oi+FGBwczNlnn82ECRM455xziIiI4JtvvmHu3LnuzCe1UNHKObZbLD4OL1nlUvOhiYiIiFhiXA+zpXPepgMYhge1dAZHQXxvc3/XQkujiIiIHK9OqwDs2rWLyy+/nA0bNmCz2TAMA5vt2PxbTqeHLZXdDLhcBt+Vf4t4wVle0soJ5jeMoCKaiIiISCM7p3M0/j529mQXsPXAUbrGhVkd6Zj2oyB9Hez8EXpfa3UaERERoI4j0e6//37atWvHgQMHCAoK4rfffmPRokUMGDCAhQsXujmi1MSatMNkHi0m1N+HYR1aWh2n5ipHomk+NBEREZHGFOTnw4hO0YAHrtLZ4Txzu2sBeNIoORERadbqVERbtmwZ06ZNIzo6GrvdjsPh4Oyzz2b69Oncd9997s4oNVDRynletxj8fRwWp6khl+vYnGhamVNERESk0R1r6fSwedHaDAGfQMg7AJmbrE4jIiIC1LGI5nQ6CQkJASAqKor9+/cD0LZtW7Zu3eq+dFIjhmHwbXkr5/ieXtTKeWQ3lOaDwx8iO1idRkRERKTZGd01BrsNftuXy74jhVbHOcbHH9oOM/e1SqeIiHiIOhXRevbsyfr16wEYPHgwzzzzDD///DPTpk2jffv2bg0oZ7Zxfy5phwoJ8LVzTudoq+PUXEUrZ0xXcNRpej4RERERqYeWIf4MSIoEYL6nrdJ5fEuniIiIB6hTEe3//u//cLlcADz55JPs2bOHESNG8PXXX/PSSy+5NaCcWcWCAiM7RxPk50XFqMr50NTKKSIiImKVcd2PrdLpUTqMMre7f4bSImuziIiIUMfVOc8///zK/fbt27Np0yYOHTpEixYtqqzSKY2jYj608T3jLU5SSwd+M7damVNERETEMuf3iOPJrzazPOUQh/NLaBHsZ3UkU0x3CIk150VLWw7tR1qdSEREmrk6jUQ7mcjISBXQLLAjM4/tmXn4OmyM6hpjdZzaqRyJpiKaiIiIiFUSI4PoFh+G02Xw45ZMq+McY7NB+/LRaDt/tDaLiIgIbiyiiTUqWjmHdYgiPNDX4jS1UJwHh1LMfbVzioiIiFjqWEunp82LVl5E07xoIiLiAVRE83IVrZwXeNOqnABZWwDDHKIfHGV1GhEREZFmbVwPs4i2aFsWhSVOi9Mcp/255jZ9PeQftDSKiIiIimhebO/hAjbsy8Fug7Hl3x56DbVyioiIiHiM7vFhtIoIpKjUxU87PKhYFRoHMT0AA3YttDqNiIg0cyqiebGKUWgDkyKJCvG3OE0tVRTRYrpbm0NEREREsNlslaPRKqYL8Rhq6RQREQ+hIpoXq7jB8bpWToB9q8xt3FnW5hAREZE6e/XVV2nXrh0BAQH079+fJUuWnPLcSZMmYbPZqj169Dg2Kj05Ofmk5xQVFTXGx2n2xnU37yl/2HyAMqfL4jTHqSii7VwAhmFtFhERadZURPNSmUeLWLnnMGAuS+5VCg/D/tXmftIIa7OIiIhInXz44YdMnjyZRx55hDVr1jBixAjGjx9PamrqSc9/8cUXSU9Pr3ykpaURGRnJ1VdfXeW8sLCwKuelp6cTEBDQGB+p2RuY1IIWQb4cLiitvM/0CG2GgcMPcvfBwe1WpxERkWZMRTQvNW/jAQwDeidGkBARaHWc2klZAoYLojpDeCur04iIiEgdPP/889x6663cdtttdOvWjRkzZpCYmMhrr7120vPDw8OJi4urfKxcuZLDhw9zyy23VDnPZrNVOS8uzsu+LPRiPg47o7uVr9K58YDFaY7jFwRthpr7aukUERELqYjmpSpbOb1tFBocu/lpP8raHCIiIlInJSUlrFq1inHjxlU5Pm7cOJYuXVqja8yaNYsxY8bQtm3bKsfz8vJo27YtrVu35uKLL2bNmjWnvEZxcTG5ublVHlI/48oXq5q3KQPDk1onK1s6f7Q2h4iINGsqonmhIwUlLNuZDXjpfGg7K4po51oaQ0REROrm4MGDOJ1OYmOrrg4eGxtLRsaZJ6VPT0/nm2++4bbbbqtyvGvXriQnJzN37lzef/99AgICGD58ONu3n7yFb/r06YSHh1c+EhMT6/6hBIARnaIJ8LWz93Ahm9OPWh3nmIovX3f/BGUl1mYREZFmS0U0L/TD5kzKXAZd40JpFxVsdZzaObwbDqeAzQFJZ1udRkREROrBZrNVeW4YRrVjJ5OcnExERASXXXZZleNDhgzhxhtvpHfv3owYMYKPPvqIzp078/LLL5/0OlOnTiUnJ6fykZaWVufPIqZAPwfndIoGPGyVzrheENQSSvJg7wqr04iISDOlIpoX+uY384bG6xYUANi10Ny2HggBYZZGERERkbqJiorC4XBUG3WWmZlZbXTaiQzD4K233uKmm27Cz8/vtOfa7XYGDhx4ypFo/v7+hIWFVXlI/Y0rv8ect8mD5kWz2491MWheNBERsYiKaF6mqNTJku1ZgJe3cnbQfGgiIiLeys/Pj/79+zN//vwqx+fPn8+wYcNO+9pFixaxY8cObr311jO+j2EYrF27lvj4+HrlldoZ3TUGh93G5vRc0g4VWB3nmA7nmdudKqKJiIg1VETzMit3H6a4zEVcWABd40KtjlM7LiekLDL3taiAiIiIV5syZQpvvvkmb731Fps3b+aBBx4gNTWVu+66CzBbLSdOnFjtdbNmzWLw4MH07Nmz2s+eeOIJvvvuO3bt2sXatWu59dZbWbt2beU1pXG0CPZjUFIk4GGj0SruH/evhsLD1mYREZFmycfqAFI7P+04CMDwjlE1mnPEo6SvM294/MOgVX+r04iIiEg9XHvttWRnZzNt2jTS09Pp2bMnX3/9deVqm+np6aSmplZ5TU5ODrNnz+bFF1886TWPHDnCHXfcQUZGBuHh4fTt25fFixczaNCgBv88UtW4HrEs25XNvI0Z3Hp2O6vjmMJbQVRnOLgNUhZD90utTiQiIs2MimheZulOs4h2dqeWFiepg4r5K5JGgEP/6YmIiHi7u+++m7vvvvukP0tOTq52LDw8nIKCU7cHvvDCC7zwwgvuiif1MLZ7LE98sYkVuw9xKL+EyODTz1/XaDqcZxbRdv6oIpqIiDQ6tXN6kSMFJWzYlwPAsA5RFqepA82HJiIiIuIVWrcIokdCGC4Dvt/sgS2dmhdNREQsoCKaF1m2MxvDgE4xIcSGBVgdp3ZKCiBtublfsbKSiIiIiHiscd3LV+nc6EFFtKSzwe4LR/bAoV1WpxERkWZGRTQvcvx8aF4ndSk4SyCsNbTsaHUaERERETmD83vGArBkexYFJWUWpynnHwKJ5XPk7fzR2iwiItLsqIjmRX4uL6Kd7Y1FtMpWznPB2xZEEBEREWmGusSG0iYyiOIyF4u3HbQ6zjFq6RQREYuoiOYl9h4uYHd2AQ67jcHtI62OU3u7Fprb9poPTURERMQb2Gw2xnU3R6PN25RhcZrjdDjP3KYsBmeptVlERKRZURHNSyzdkQ1A79bhhAb4WpymlvIy4cBv5r7mQxMRERHxGuN6mPOi/bA5k1Kny+I05RL6QHA0FOfC1m+sTiMiIs2Iimhe4idvbuWsGIUW1wuCvTC/iIiISDPVv20LokP9ySksZfaqvVbHMdkd0G+iub/i39ZmERGRZkVFNC/gchmV86F55aIClfOhqZVTRERExJs47DbuGtkBgBd/2E5RqdPiROX63wI2u9nSmbXV6jQiItJMqIjmBbYeOEp2fgmBvg76tmlhdZzaMQzYVV5E03xoIiIiIl7nhsFtSAgPID2niP/+ssfqOKaIROg83txfMcvaLCIi0myoiOYFKkahDWoXiZ+Pl/2RZW2Fo+ngEwBthlqdRkRERERqKcDXwf1jOgHwyoIdHC3ykMn8B91mbte9D8V51mYREZFmwcsqMs3Tz149H1r5KLQ2Q8A3wNosIiIiIlInV/ZrTfvoYA4XlDLrpxSr45janQuRHcwFBtZ/aHUaERFpBlRE83AlZS6WpxwCvHQ+tIpFBdTKKSIiIuK1fBx2/jS2CwBvLknhUH6JxYkAux0Glo9GWzHLnEZERESkAamI5uHWph2hoMRJy2A/usaFWh2ndpylsPsnc1+LCoiIiIh4tfE94+jZKoy84jJeXbDD6jimPteDTyBkboTUZVanERGRJk5FNA9X0co5rGMUdrvN4jS1tHcFlORBUBTEnmV1GhERERGpB7vdxkPnd4X/396dx0dVnX8c/8ySTPaEEEgChBD2HSQsAqIoiKJ1bQVti1pBpVULoj8rWitiW7RVS62CWndbAVtwqdBKVFYBWQSEsMoWloQAgSRkz8z9/XGTgbAlIctNZr7v1+u+5s6dO3eekwvJyZNzzgO8v2ofh04UWBwREBwFPW8z91f/3dJQRETE9ymJ1sCVJ9EGt2tqcSQXYVd5Vc4rzOH2IiIiItKoXd4hhv5J0RSXenj5q51Wh2Pqd6/5uPUzyM2wNhYREfFpymw0YLmFJazffwJorOuhlSfRNJVTRERExBfYbDZ+c625Ntq/1h1g95EGUBUzvickDABPKXz3vtXRiIiID7M8iTZjxgySkpIICgoiOTmZZcuWnffcxYsXY7PZztq2bdtWjxHXn9V7snB7DBKbhpAQHWJ1ONVTcAIOrjP3tR6aiIiIiM9IToxmWOfmuD0GL6XssDocU/lotLXvgLvU2lhERMRnWZpEmzNnDhMnTuTJJ59k/fr1DBkyhJEjR5KWlnbB923fvp309HTv1qFDh3qKuH4tL5/K2RhHoe1dBoYHmnaAyFZWRyMiIiIitejRazphs8Hn36ez+WC21eFA1xvNdXhzD8H2+VZHIyIiPsrSJNpLL73E2LFjGTduHF26dGH69OkkJCQwc+bMC76vefPmxMXFeTeHw1FPEdevFT8cA+CyxphE866HNtTSMERERESk9nWJj+DGXi0AeGHhdoujAZwuSL7L3F/zprWxiIiIz7IsiVZcXMy6desYMWJEheMjRoxgxYoVF3zvJZdcQnx8PMOGDWPRokUXPLeoqIicnJwKW2OQmVvI9sO52GwwsG0jLCqwe7H5qKmcIiIiIj7p4eEdcdptLN5+hNV7sqwOB5J/ATY77FkKRxpAYk9ERHyOZUm0o0eP4na7iY2NrXA8NjaWjIxzV9WJj4/njTfeYO7cucybN49OnToxbNgwli5det7PmTZtGpGRkd4tISGhVttRV8pHoXVrEUGT0ECLo6mmE2mQtQtsDmhzmdXRiIiIiEgdaBMTyqh+Zt/6T//bhmEY1gYUlQAdR5r7Go0mIiJ1wPLCAjabrcJzwzDOOlauU6dO3HvvvfTp04eBAwcyY8YMrr/+el544YXzXn/y5MlkZ2d7t/3799dq/HWlUa+HVj6Vs1VfCIq0NhYRERERqTO/vqoDLqedtfuOs3j7EavDgf7jzMcNs6CoAVQOFRERn2JZEi0mJgaHw3HWqLPMzMyzRqddyKWXXsrOnTvP+7rL5SIiIqLC1tAZhsE3ZUm0Rrke2u7y9dA0lVNERETEl8VFBnH3oDYA/OmL7Xg8Fo9GSxoKTdtDcS58P8faWERExOdYlkQLDAwkOTmZlJSUCsdTUlIYNGhQla+zfv164uPjazs8S+05mkd6diGBDjt9E6OtDqd6PB7YvcTc13poIiIiIj5v/BXtCHc52Zqew+eb0q0Nxm6HvmPN/TVvgtVTTEVExKdYOp1z0qRJvPnmm7z99tts3bqVhx9+mLS0NMaPHw+YUzHvvPNO7/nTp0/nk08+YefOnaSmpjJ58mTmzp3Lgw8+aFUT6kT5KLTkxCYEBzayyqMZG6EgCwLDoWWy1dGIiIiISB1rEhrIfZe3BeClhdspcXusDaj3TyEgBDK3QNpKa2MRERGf4rTyw0ePHs2xY8eYOnUq6enpdO/enQULFpCYmAhAeno6aWlp3vOLi4t59NFHOXjwIMHBwXTr1o358+dz3XXXWdWEOlG+HtplHRrhVM7y9dCShoAjwNpYRERERKRe3HNZEu+u2MveY/n8a+0BfjqgtXXBBEdBj9vgu/dg9d8hseqzXERERC7EZlheRqd+5eTkEBkZSXZ2doNcH83tMbhk6kJyCkv55IHB9E6Isjqk6nnvBrOs+Mg/w4D7rI5GRESk1jT0PoToHlnt7eV7mPr5FuIiglj8f0MJCrBwRkX69/D6ELA74eFUCI+zLhYREWnwqtqHsLw6p1S0+WA2OYWlhAc56dGykVW2LM6HtFXmftuhloYiIiIiIvXrZ5e2pmVUMBk5hXywcp+1wcT3hIQB4CmFde9ZG4uIiPgMJdEamPKpnAPbNsVht1kcTTWlrQR3MUS0hJgOVkcjIiIiIvXI5XQwYbjZB5yx+AdyC0usDajfvebjunfAbXEsIiLiE5REa2C+aczroe0uWw+t7ZVga2QJQBERERGpsVsvaUm7ZqEczy/hvRV7rQ2m640Q2gxy02H7AmtjERERn6AkWgNSUOxm7d7jAAxu3wiTaLsWm4/trrQ0DBERERGxhtNh56GrzNFo763cR1Gp28JgXNDnTnN/9d+ti0NERHyGkmgNyNp9WRS7PcRHBtE2JtTqcKrnZCYc3mTuJ11hbSwiIiIiYpnre8YTHxnEkdwiPt1wyNpgkn8BNjvsXQZHtlsbi4iINHpKojUg3/xwDDBHodka23TI3UvMx7geENbM2lhERERExDIBDjt3D2oDwJvLdmMYhnXBRCVAp+vM/TVvnvucwhzI3Ao7U2Dt2/DVszDvfvjgVtjxRf3FKiIiDZ7T6gDklPL10Aa3b2pxJBfh9PXQRERERMSv3d6/NS9/tZMdh0+yZMcRhnZqbl0w/cbCts9hwywIj4Psg5B94NRWlH3+9x7fAx1GaL1fEREBlERrMI7nFbP5kPkDfHC7RrYemscDu8qSaFoPTURERMTvRQYHMLpfa97+Zg9vLttjbRItaSg0bQ/HfoCvpp77nKAoiGx1aotoCUueh6zdkPE9xPeqx4BFRKShUhKtgVi5+xiGAR1jw2geEWR1ONWTOg9yD0FgOLQeaHU0IiIiItIA/GJwG95dsYflPxxly6EcuraIsCYQux2ufxFWvgohTc9IlrWCyJbgCj/7fYe+g63/gdRPlEQTERFASbQG49RUzkY2Cq20GL5+1twf/GsICLY2HhERERFpEBKiQxjZI57536fz5vLdvDSqt3XBtB1qbtXR7ZayJNrHMOx3mtIpIiIqLNBQlCfRLmtsSbTv3oPjeyG0OVz6K6ujEREREZEG5L4hbQH4z8ZDHM4ptDiaaupwDTiDzXXR0jdaHY2IiDQASqI1APuz8tl7LB+H3Ub/pGirw6m6olxzrQiAob8BV5i18YiIiIhIg9IrIYr+baIpcRu8u2Kv1eFUjysMOo4w97d8YmkoIiLSMCiJ1gCs2GWOQuudEEV4UIDF0VTDylch7whEt4U+d1kdjYiIiIg0QOOGJAHwz1X7yCsqtTiaaup6s/mY+jEYhqWhiIiI9ZREawCW/3AMaGTroZ08Aiv+Zu5f9RQ4GlHyT0RERETqzfAusSTFhJJTWMq/1u63Opzq6Vg+pXMvpG+wOhoREbGYkmgW83gMVjTG9dCW/hmKT0KLS079hU5ERERE5Ax2u417LjNHo739zV7cnkY0oisw1EykgTkaTURE/JqSaBbbfjiXY3nFBAc46J0QZXU4VZO1B9a+be4Pf8YsGy4iIiIich4/6dOKJiEBpGXlszA1w+pwqqfbLeZj6iea0iki4ueU/bDYd2nHAUhObEKgs5HcjkV/AE8JtLsK2l5hdTQiIiIi0sAFBzr4+aWJAPx92W6Lo6mmDiMgIARO7IND662ORkRELNRIsja+K/VQDgA9WkVaHEkVpW+ETf8y94dPsTQUEREREWk8xgxMJNBh57u0E6zbl2V1OFUXGKIpnSIiAiiJZrnUg9kAdG/RSJJoX04xH3vcBvG9LA1FRERERBqP5uFB3HxJCwD+vnSPxdFUU/mUzi2faEqniIgfUxLNQiVuD1szcgHo3jLC4miqYPdi2PU12APgyietjkZEREREGplxQ9oC8MWWDPYdy7M4mmpofzUEhMKJNDj0ndXRiIiIRZREs9APmScpLvUQ7nKS0CTE6nAuzOOBlKfN/b73QHSStfGIiIiI5WbMmEFSUhJBQUEkJyezbNmy8567ePFibDbbWdu2bdsqnDd37ly6du2Ky+Wia9eufPyxps/5ko6x4VzRsRmGAW8vb0Sj0TSlU0REUBLNUuXroXVtEYHdbrM4mkps+QTSN0BgGFz+f1ZHIyIiIhabM2cOEydO5Mknn2T9+vUMGTKEkSNHkpaWdsH3bd++nfT0dO/WoUMH72srV65k9OjRjBkzho0bNzJmzBhGjRrFt99+W9fNkXp0b9lotI/WHuBEfrHF0VSDt0rnp5rSKSLip5REs9Dm8vXQWjbw9dDcJfD1s+b+oF9DWDNr4xERERHLvfTSS4wdO5Zx48bRpUsXpk+fTkJCAjNnzrzg+5o3b05cXJx3czgc3temT5/O1VdfzeTJk+ncuTOTJ09m2LBhTJ8+/ZzXKioqIicnp8ImDd/g9k3pHBdOQYmbf3574aRrg9KhbEpndhoc1JROERF/pCSahVIPmUm0bi0a+Hpo370HWbshtBkMfMDqaERERMRixcXFrFu3jhEjRlQ4PmLECFasWHHB915yySXEx8czbNgwFi1aVOG1lStXnnXNa6655rzXnDZtGpGRkd4tISHhIloj9c1ms3lHo723Yi/FpR6LI6qigGDodK25nzrP2lhERMQSSqJZxOMx2FI2nbNBj0QrOgmLnzf3r/gNuMKsjUdEREQsd/ToUdxuN7GxsRWOx8bGkpGRcc73xMfH88YbbzB37lzmzZtHp06dGDZsGEuXLvWek5GRUa1rTp48mezsbO+2f//+GrZM6ssNvVoQG+EiM7eIzzYesjqcqvNO6fxEUzpFRPyQ0+oA/NXeY3nkFbsJCrDTNibU6nDOb9UMyMuEJknQ5y6roxEREZEGxGaruKarYRhnHSvXqVMnOnXq5H0+cOBA9u/fzwsvvMDll19+Udd0uVy4XK6LDV8sFOi0c9egNvzpf9t5c9luftyn5Xnvc4PSfri5RnDOATiwFhL6WR2RiIjUI41Es8jmslFoXeIjcDoa6G3IOwrfvGzuD3sKnIHWxiMiIiINQkxMDA6H46wRYpmZmWeNJLuQSy+9lJ07d3qfx8XF1fia0nj8rH8iIYEOtmXksvyHo1aHUzUBwdBppLm/5RNLQxERkfrXQLM3vi/1YCNYD23pC1CcC/G9oOstVkcjIiIiDURgYCDJycmkpKRUOJ6SksKgQYOqfJ3169cTHx/vfT5w4MCzrrlw4cJqXVMaj8iQAEb1Ndex+/uyPRZHUw1dbzYfUz8BTyNZz01ERGqFpnNaZHNZUYHuLRroemjH98KaN8394c+AXflWEREROWXSpEmMGTOGvn37MnDgQN544w3S0tIYP348YK5XdvDgQd5//33ArLzZpk0bunXrRnFxMf/4xz+YO3cuc+fO9V5zwoQJXH755Tz//PPcdNNNfPrpp3z55ZcsX77ckjZK3btncBLvr9zL0h1H2J6RS6e4cKtDqtzpUzoProWE/lZHJCIi9USZEQsYhkFqQy8q8PUfwFMCba+EdldaHY2IiIg0MKNHj2b69OlMnTqV3r17s3TpUhYsWEBiYiIA6enppKWlec8vLi7m0UcfpWfPngwZMoTly5czf/58br31Vu85gwYNYvbs2bzzzjv07NmTd999lzlz5jBgwIB6b5/Uj9ZNQ7imWxwAry3ZZXE0VRQQBJ2uM/dTP7E0FBERqV82w/CvsjI5OTlERkaSnZ1NRIQ1UykPHM/nsucXEeCwsfmZa3A5HZbEcV6HU2HmYMCA+5ZAi95WRyQiImK5htCHkAvTPWqcNu4/wU2vfgPAnPsuZUDbphZHVAXbFsDsOyCiJUzcrFkbIiKNXFX7EPpub4HNB81RaB2ahze8BBrA+n8CBnS5QQk0EREREalTvRKiuKN/awAmf7yJolK3xRFVQburIDAccg7CgTVWRyMiIvVESTQLpJavh9ayAf6F1OOB1Hnmfq87rI1FRERERPzC49d2JibMxe4jecxY1AimdQYEQefyKZ0fWxuLiIjUGyXRLNCg10NLWwm56eCKNBdNFRERERGpY5EhAUy5sSsAMxfv4ofMXIsjqoJuZdXrt3yqKp0iIn5CSTQLbD5ojkTr1qIBjkQrH4XW5UfgdFkbi4iIiIj4jet7xHNV5+YUuz08MW8zHk8DX7q53VXgioDcQ3BgtdXRiIhIPVASrZ5l5hSSmVuEzQZd4htYEs1deqrCUPdbL3iqiIiIiEhtstlsTL2pG8EBDlbvzeKjtfutDunCnK7TqnRqSqeIiD9QEq2elU/lbNcsjJBAp8XRnGHvUsg/CiFNIekKq6MRERERET/TqkkIj4zoCMAfF2wlM7fQ4ogqoSmdIiJ+RUm0euYtKtAQp3Junms+drkRHAHWxiIiIiIifunuQW3o0TKSnMJSpv5ni9XhXFi7K821hHPTYf+3VkcjIiJ1TEm0erb5oDkSrVuLBlZUoLQYtv7H3O/+Y2tjERERERG/5XTYmXZrD+w2+Pz7dBZty7Q6pPNzulSlU0TEjyiJVs82l41E69aygY1E2/U1FGZDWBwkDrI6GhERERHxY91bRjL2siQAfvvJZvKKSi2O6AI0pVNExG8oiVaPTuQXc+B4AdAAR6KVT+XsdgvYHdbGIiIiIiJ+7+GrO9IyKpiDJwr4S8oOq8M5v7ZlUzpPZsD+VVZHIyIidUhJtHq0payoQOvoECKDG9CaY8X5sH2Bua+qnCIiIiLSAIQEOvn9Ld0BePubPWw+mG1xROfhDIQuPzL3NaVTRMSnKYlWj7xTORtaUYGdC6H4JES2hlb9rI5GRERERASAKzs154ZeLfAY8Pi87yl1N9Dpkl1vNh83zoavfw9HGvDIORERuWhKotWj8qIC3Vs2sKmcqfPMx+63gM1mbSwiIiIiIqf53Y+6EhHkZPPBHN5dsdfqcM6t7VCI7Q5FObD0z/BqP3j9clj5KuRmWB2diIjUEiXR6lFqQxyJVpQLO74w91WVU0REREQamGbhLp64rgsALy7cwf6sfIsjOgdnIIxNgR+/BR2uAbsT0jfCF0/AS13g/Ztg/T+hMMfqSEVEpAaURKsneUWl7D6aBzSwogLb/wulhdC0PcT1tDoaEREREZGzjOqbQP+kaApK3Pzu080YhmF1SGcLDIEeP4GffQSPbIfrXoCEAWB4YPdi+PRX8EIH+Ogu2DYfSoutjlhERKrJ8iTajBkzSEpKIigoiOTkZJYtW1al933zzTc4nU569+5dtwHWkq3pORgGxEa4aBbusjqcU7xVOW/VVE4RERERaZDsdht/vKUHgQ47i7YfYf6mdKtDurDQGOh/L4xdCBM2wlW/hZhO5h+vt3wCs39qJtQ+nwQnj1gdrYiIVJGlSbQ5c+YwceJEnnzySdavX8+QIUMYOXIkaWlpF3xfdnY2d955J8OGDaunSGuuvJpQ94Y0Cq3gOPzwlbmvqZwiIiIi0oC1bx7Gr65sB8CUz7aQnV9icURV1KQNXP5/8MC3cP9SGPgghMdD4QlY+xa8Nhh2fW11lCIiUgWWJtFeeuklxo4dy7hx4+jSpQvTp08nISGBmTNnXvB9999/Pz/96U8ZOHBgPUVac6mHzPUPujWkogJbPwdPCTTvBs07Wx2NiIiIiMgF/XJoO9o1C+XoySKe+99Wq8OpHpsN4nvBNX+Ah1NhzCfQrAucPAwf3AIpvwN3I0kMioj4KcuSaMXFxaxbt44RI0ZUOD5ixAhWrFhx3ve988477Nq1i6effrpKn1NUVEROTk6FzQqby5NoDamoQPlUzu63WhuHiIiIiEgVuJwOpt1qruM7a/V+vt522OKILpLdAe2uhHu/hr73mMe++Su8NQKydlsbm4iInJdlSbSjR4/idruJjY2tcDw2NpaMjHOXgd65cyePP/44//znP3E6nVX6nGnTphEZGendEhISahx7dRWWuNl5OBeA7g1lJNrJI7BnibmvJJqIiIiINBL9k6K5e1AbAB75aCPp2QXWBlQTgSHwo7/AqA8gKAoOfQevXQ7ff2R1ZCIicg6WFxawnbGYvWEYZx0DcLvd/PSnP+WZZ56hY8eOVb7+5MmTyc7O9m779++vcczVteNwLqUegyYhAbSIDKr3zz+nLZ+YlYJaXALRba2ORkRERESkyiZf15nuLSM4nl/ChFkbKHV7rA6pZrreCOOXQ+tBUJwL8+6Fj8dDUa7VkYmIyGksS6LFxMTgcDjOGnWWmZl51ug0gNzcXNauXcuDDz6I0+nE6XQydepUNm7ciNPp5Ouvz70Yp8vlIiIiosJW38rXQ+veMvKcCUJLbJ5nPqqggIiIiIg0Mi6ng1fu6EOYy8nqvVm8/NVOq0OquagEuOs/MHQy2OywcRa8fjkc/M7qyEREpIxlSbTAwECSk5NJSUmpcDwlJYVBgwaddX5ERASbNm1iw4YN3m38+PF06tSJDRs2MGDAgPoKvdrKK3N2bSjroWUfhLSV5n63W6yNRURERETkIrSJCeUPt3QH4G+LfmDFD0ctjqgWOJww9HG4ewFEtDLXR3trBHzzMnga+Wg7EREfYOl0zkmTJvHmm2/y9ttvs3XrVh5++GHS0tIYP348YE7FvPPOO81A7Xa6d+9eYWvevDlBQUF0796d0NBQK5tyQeVFBbq3aCDroW35BDAg4VKIbGV1NCIiIiIiF+Wm3i25vV8ChgET5mzgSG6R1SHVjsSB8Mvl0OVG8JRAylPwzx9DbiMtpCAi4iMsTaKNHj2a6dOnM3XqVHr37s3SpUtZsGABiYmJAKSnp5OWlmZliDVW6vawLf3UdM4GwVuVU1M5RURERKRxe/qGbnSMDeNIbhGTPtqAx2NYHVLtCG4Co96HH00HZzDs+hpeGwyb/l03o9IyNsE/fgIvdYPdS2r/+iIiPsBmGIaP/JSpmpycHCIjI8nOzq6X9dG2Z+RyzfSlhLmcfP/0COx2i9dEy9oDL/c211mYtA3Cz15/TkRERM5W330IqT7dI/+143AuN76ynMISD49d24lfDW1vdUi1K3Mb/PseyEw1n8f2gKt+Cx2vgZquuZx9AL7+g7kGG2W/GjpcMOo96DSyZtcWEWkkqtqHsLw6p6/zrocWH2F9Ag0g9WPzsc0QJdBERERExCd0jA3nmRu7AfDiwh2s25dlcUS1rHlnuPcrGPoEuCLg8CaYNRreuvriR40VZsOXU+BvybDxQ8Aw10vuOBLcRTDn5+aoNxER8VISrY5tPmQm0bq1bCB/DVVVThERERHxQaP6JnBT7xa4PQa/nrWBE/nFVodUuwKCYehvYMJGGDzRnOJ5YA28fyO8dwPsX1O165QWw6rX4K+9YflfoLQQEgfDuK/gtndh9AfQYxR4SmHuOFj3bt21SUSkkVESrY6lHmxARQWObDf/amV3QpcbrI5GRERERKTW2Gw2/nBLD9o0DeHgiQIe+/f3+OTKNSHRcPUzMGED9L8P7AGwZym8NRw+vN1c2+xcDMP8g/qr/eB/v4GCLIjpBHfMhrvnQ6u+5nmOALjldeh7D2DAfybAilfqq3UiIg2akmh1yOMx2FJWVKBBjEQrH4XW7irzh6+IiIiIiA8Jczl55ad9CHTYWbjlMO+t2Gt1SHUnPA6u+zM8tA56/9xc83jHf+G1y+Bfv4CjO0+du/cbeHMY/PsXcHwvhMXCDX+FX64w1z07c101ux2ufwkGTzCfL3wSFv3RTMTVhMcNuxdD2rc1v5aIiAWcVgfgy/Zl5XOyqBSX0077ZmHWBmMYqsopIiIiIj6ve8tInriuM1P+s4U/LthG3zbRdG/ZAGaF1JUmiXDzq3DZRDPRlTrP3LZ8Ar1+ao44277APDcg1EyMDXoQAkMvfF2bDYY/Y67B9vWzsOR5KMqFa/5Y/WIGHrf5B/2lf4aj281jsT1gwP3Q4zYICKpuq0VELKGRaHWovKhA5/gInA6Lv9SHN8OxnWalnU7XWRuLiIiIiEgdumtQG0Z0jaXY7eHBD7/jZFGp1SHVvZgOcNs7MH65WRzA8MCGf5gJNJsD+o41p4AO/U3lCbRyNhtc/iiM/LP5fNUM+OwhMylWFe4S2PAhvNIP5o0zE2iuSAgIMZeZ+exB+EtX+OpZyEm/qGaLiNQnjUSrQ+VFBbq3aAhTOctGoXW4GoIaQDwiIiIiInXEZrPxp5/0JPXl5ew9ls8T8zbx19t7Y6vuCKrGKK4H/HS2WWhg2QtmwmroZGjW8eKvOeA+cIXBpw/A+g+g+CTc8gY4A899fmkxbJwFy16EE/vMY8FNYOCD0P9eM8H33fuw+u+Qvd+M85vp0PVmGDAeEvpdfKwiInVISbQ6tOVQ2XpoVhcV0FROEREREfEzUSGBvHxHb0a9vorPNh5icPumjO7X2uqw6k9CP/jpnNq7Xu+fmiPY/j0WUj+GopNmJc+A4FPnlBaZSbbl083kGEBIDAx6CPqNBVf4qXMHT4BLH4Dt881qoWkrYPO/za1lMgz4JXS96fyJOhGpnGHAyUzIPwoxHc3CIVIjSqLVEcMwvNM5u1tdVODgOjiRZq6B0PEaa2MREREREaknyYnRPDqiE8//bxtPf5ZK1/hIerTy4fXR6lrXm+COUJjzc/ghBf7xE7hjlvmL+Xfvm8mz3EPmuWGxZqIs+e7zTx91OM1rdr0J0jfCt6/Dpn+Zv7/MGwcLfwv9xkHfX0BoTH21UmrK44bcDPN30KjWENnS6oh8W3G+OeLz+N5zbPugtMA8L7YH/PhNaN7ZslB9gc3wybrP55eTk0NkZCTZ2dlERNRdcuvgiQIGP/c1TruNzc9cQ1CAo84+q1L/ewJWvQrdfwI/ecu6OERERBqx+upDyMXTPZJz8XgM7n53DUt3HCHc5eTde/qRnKhK9TWybwX8cxQU50LzrpB/DE4eNl8LbwGXPQx9xlQcpVZVJ4/AundgzZunrukMMhNygydUfT23ulSYDVm7zbicQWY7yx8dgdUvvNDYeNyQm24mybzbvlP72QfBU2Kea3OYv4N2u8XamH3J3m/MpPXxPWairPz/yfnY7Oa/y9JC89/pNX8w10j09X+n1VTVPoRGotWR8lFoHWLDrU2gAexcaD52vdHaOERERERE6pndbuPVn17C2PfWsnpPFmPeWs2bd/ZlUHuNbLpoiYPg7v/AB7dC5hbzWGRrGPIw9P4ZOF0Xf+2wZnDFYzB4ollhdNUMOLTerA763Qdw9TNmRU8rEgCHt8DqN+D7OVCSf56TbBWTauWPgWEQ2xVa9oVWfaFpB7A3ojp/B9aZa9dlboHsA+CppFiHzQEh0ZB3xJwCbHi0tFBNGQasmgkLnzS/nqdzRUJ0G2hyji0yAfKz4JNfwq6vYP4j8MNXcOMrENq0vlvR6GkkWh15KWUHL3+1k58kt+KF23rV2edUKvsA/KWbmX1+bLe5oKeIiIhUm0Y5NXy6R3IhBcVu7vtgLct2HiXQaee1n/fhqs6xVofVuB3ZDstegjaDodcddbPekmHAlk8h5SlzlBNAq35w7fPQKrn2P+9M7lLY8V9zquneZaeOh8QABpQUmtPlzkxqVMYVCS0vOZVUa9nXTCA2NDmH4Mtn4PvZFY/bnRDZypyuGdUaohLLpm4mmI/h8Wai89MHYeOH5u+jt7wBPW+zph2NXWkRfP4wbPin+bz7T6DLj04lyqrye77HA9++Bl8+De5iCIuDW16DdlfWZeSNRlX7EEqi1ZGx767hq22ZTLmhK3cPTqqzz6nU+n+YVXRaJsO9X1sXh4iISCOnBE3Dp3sklSkqdfPgh+tJ2XIYp93GX2+/hOt7xlsdllRFSSGsfMVM2pXkmcd63QHDnoaIOriH+Vnw3Xuw5q1TRRJsDuh8PQy4HxIHnxoNZxhmUqKkwJwyd67HguPmiLqD6+DQhlPrVJ0uqnXFpFpsN7MqqhVKCmDFK7D8pVOj7nrdAZeMgSaJZpLMXoUZVx43fPZr2PAPM5F282vQa3Tdxu5rcjPMdQgPrDG/hiP+AJf+8uJHY6Z/D3PHwdHt5vOBD8Kw39VsBKkP0HROi20+VF5UwOKFS3ctMh/bKrssIiIiIv7N5XQw42d9eOSjjXy28RAPzfqOwpJe/Di5ldWhSWUCguDyR83pol89AxtnmduWz8xppAMfvLg12M6UvtGcsrnp32YCDCA42iyQ0G+sOfrqTDabmYCoLAnR4yfmo7vEnBZ5YK2ZVDuwFo7uOLWmWOq8U+8Ji4XodtC0bCvfj25bO+09k2GY1VdTnobs8pF//eHa5y5u5J/dATf+zZy6+t378PH9YLjNaq+NhbvUbIcVU4gProPZPzPXoAuKgtverfnIsfiecN9ic1ro2rfN5PSepfDjt6BZx1oI2rdpJFodOJJbRL8/fInNBpunXEOoy6JcpccDL3Qwy9nePR/aXGZNHCIiIj5Ao5waPt0jqSq3x+CJeZuYs9YcYfTszd0Zc2mixVFJtRxYB/97HA6sNp9HtoYRU6HrzdVPdrhLYOtn8O0bsH/VqePxvaD//eZaXgFBtRb6ORVmmyPVyhNrB9dVvmB8RCto2vZUYq1ZZ2jR5+LXuTq0Hv43GdJWll2/JQx/xkz+1TSB5PHAgkfMpA02M7HWZ0zNrlmXSgog9ROzyMX+b80RYAEhZWvchZgJzIDT9p3BZceCzeIXSVdAhxFmBdqLtXEOfPYQuIvMe3v7h+Z9rk3b5ptTbguyzDZcO81MGPth0QGNRLNQatkotKSYUOsSaACHN5sJtIBQ868HIiIiIiKCw25j2q09CA508O6KvTz1yWYKi93ce3lbq0OTqmqVDGMXmiPGviwbNfWvu81pltf8wUyq5R8zfx/KP3balgV5ZxzLO3JqyqLdCV1vMpNnCf3rL5kQFAlth5pbuYITkLULju0ue9xV9viDmXTLOWBue5ZWvFaTJHM5n/JpofE9LzxKLjcDvnq2bL0tw0ymXDYRBv0aAkNqp312O1z/kpmMWvMmfPaguY5c8l21c/3acnQnrH3H/FoUnjh13PBA8Ulzq4pvXzOTkH3uMpOFES2qHoPHbf6bXvE383nHkXDrGxBUB38c6ny9mXj9+H7YswQ+nwg/fGkmOUNUxfhclESrA6mHcgDo3sLiqZy7y6ZythkMzkBrYxERERERaUDsdhtP39CVkEAHMxbv4g8LtpJXXMqEYR2w+eEojEbJZjMXqu98HXzzV3Pb9w28MbT61wptDn1/Acm/qJs11i5GcJSZDGt5xjRKwzCTgWcm1jI2w7GdcHyPuW3+t3m+PQDiepxKqrXqa04HLS2CVa+a68yVJ4d63AbDp5x72mpN2Wxw3Qvm2nKrX4f//Nqc2tn3ntr/rOooLYZtn5uj5E4vHhGZYCb5eowyk5Al+ebafCUF5n5pYdmxgtO2fHME4ea5kHMQFv/RrCzbaaQ5HThp6IWrshYcN6uZ7vrKfD7kUbjyybqt5BoRD2M+gZV/M5Op2z6Hg9/BVU9C8y7mvxUVKPRSEq0ObD5Yvh6axdMItB6aiIiIiMh52Ww2Hru2M6EuJ3/+YjvTv9xJQbGbx0d2ViKtMQkMhSufMBe9//JpM4EB5uiukBgIaXraFg2hZx5rai7qXxfVReuCzWZO2Qxtao6WO13BcTMBUr7W2sG15mi7Q9+ZG2+Y5wU3Macm5qabz1smm+uenXm9uoh95PPmGmOrZpgVJz1u6H9v3X7uuRzfC+veg/UfmKMRwRwp1+EaM6HafnjViiecy4jfw9b/mIm5fd+Yialtn5ujBPv+wlzbLzSm4nuO7IBZt5tJUWcw3DwDut9aoyZWmd0OgyeY01DnjjWTsp8+cOr1oCgzmRbdFqKTTu03SYKw5n41/VNrotWBy/+0iLSsfP45bgCD28dU/oa6UFIIzyea2fFfrTIzyCIiInLRtN5Ww6d7JDXx1vI9PPv5FgB+fmlrpt7YHbvdf34x9CnFeeAIbDxJsbpkGHBiX8UiBukbzXW2wKyyOXyKOdqqLkc7nSuulKdOTVkc+Sez6mldc5fCzoVmcuuHL4GydEhYHPS509yiEmr3MzO3mWurbZgFReaAGxyB5rThvvdA64FmTHPHQVGOOQLu9g/NabhWKM6DpS9A2irI2g0nMy58fkComVCLSjCTs0GRlW+B4fX7760KqtqHUBKtlmXnl9Br6kIANvzuaqJCLJpGuXsxvH+T+c3gkW1+lRkWERGpC0rQNHy6R1JTs1an8cTHmzAMuLVPS/704544HQ3rFz2RGistNtfPzk03Rx65wqyJwzDgyynwzXTz+TXTYOCvaufapUXmVNejO87YfoCSvFPntb3STGR1Gln3SdfifHOU5Nq3y0YFloluC1l7AANaD4JR70NYs7qNpTqK88xRe1m7y7Y9px6z9+NNRFaLDVwRpwo0OCt7DDLPdUXA4F/XcgNNKixgkdR0M7PcqkmwdQk0OG0q51Al0ERERKROzJgxgz//+c+kp6fTrVs3pk+fzpAhQ8557rx585g5cyYbNmygqKiIbt26MWXKFK655hrvOe+++y6/+MUvznpvQUEBQUF1XBlPBLijf2tCAh1M+mgj8747SGGJm+mjLyHQqUSa+BBnILTsY3UU5u+pw6eYUyaXvQhfTDZHyHW79dTr2M5+PP01wwMn0s5Olh3fa752LsHRcMnPzSqUtV3t8kICQ8wiA33GmJVQ174Dm/5lJqTAXI9v5J8a3nrmgaEQ283czlRaZH79s3ZD9gFzJF1h9oW30kLAMEfllY/Mq6qgqDpLolWVkmi1LPVgNYsKFJ00F1Ts/KPane9cXlSgndZDExERkdo3Z84cJk6cyIwZMxg8eDCvv/46I0eOZMuWLbRu3fqs85cuXcrVV1/NH//4R6KionjnnXe44YYb+Pbbb7nkkku850VERLB9+/YK71UCTerTTb1bEhTg4KEP17NgUwb5xWuZ+bNkggMvcm0kETk/mw2uesosNrD0T+bItC+n1M61XREQ09HcmnU8td8kCRwWp0JaXAI3XgIjnoXUj80pjt1usTami+F0QUwHc6uqksJTybbyYg2lBVV7vFCV2XqiJFotSz1kZlK7tajiFILUeeaQzt2LzURabWSd845B+vfm/uklkkVERERqyUsvvcTYsWMZN24cANOnT+eLL75g5syZTJs27azzp0+fXuH5H//4Rz799FP+85//VEii2Ww24uLi6jR2kcpc0y2ON+/qy30frGXx9iPc9fZq3rq7L+FBWmNLpNbZbGYlyMAQs8JqSdlIJcM47ZFzHCs7HtGqLJFzRrIsLLbhz8oKijRHxPmTgCBzC2tudSQXRUm0WnZt93giggMY1L5p1d6w/1vzMf+YubBh5+tqHsSexYABzbtCuDqhIiIiUruKi4tZt24djz/+eIXjI0aMYMWKFVW6hsfjITc3l+jo6ArHT548SWJiIm63m969e/Pss89WSLKdrqioiKKiIu/znJycarZE5Pwu79iMD8YO4J531rB6bxY/e/Nb3v1Ff6JDG9hUKxFfcdnD5ibSgGlyfy27tnscU2/qTnJidOUnA+xffWp/46zaCcK7HpqmcoqIiEjtO3r0KG63m9jY2ArHY2NjyciopIpXmRdffJG8vDxGjRrlPda5c2feffddPvvsM2bNmkVQUBCDBw9m586d57zGtGnTiIyM9G4JCbVcUU38Xr820cy671KiQwP5/kA2o19fyeGcQqvDEhERiyiJZqX8LHPRw3I7/mceqwnDMKeGgtZDExERkTplO2OajGEYZx07l1mzZjFlyhTmzJlD8+anpnNceuml/PznP6dXr14MGTKEjz76iI4dO/K3v/3tnNeZPHky2dnZ3m3//v01a5DIOXRvGclH919KXEQQOzNPcttrK9mflW91WCIiYgEl0ax0YI352LQDxPYAd7G5qGBNZO02y8zaAyBxUM1jFBERETlDTEwMDofjrFFnmZmZZ41OO9OcOXMYO3YsH330EcOHD7/guXa7nX79+p13JJrL5SIiIqLCJlIX2jcP51/jB9I6OoS0rHx+8toKfsjMtTosERGpZ0qiWal8PbSEAdDrdnN/4+yaXXPX16euGRhas2uJiIiInENgYCDJycmkpKRUOJ6SksKgQef/I96sWbO4++67+fDDD7n++usr/RzDMNiwYQPx8fE1jlmkphKiQ/jX+IF0aB7G4ZwiRr2+is0Hs60OS0RE6pGSaFYqXw8toT/0uA1sdjiwGo7tuvhreqdyDq1pdCIiIiLnNWnSJN58803efvtttm7dysMPP0xaWhrjx48HzKmWd955p/f8WbNmceedd/Liiy9y6aWXkpGRQUZGBtnZp5IQzzzzDF988QW7d+9mw4YNjB07lg0bNnivKWK12Igg5tw/kB4tI8nKK+aON1axZm8Nl2MREZFGQ0k0q7hL4eA6cz9hAITHQrth5vOLHY3mLoU9S839tlfVPEYRERGR8xg9ejTTp09n6tSp9O7dm6VLl7JgwQISExMBSE9PJy0tzXv+66+/TmlpKQ888ADx8fHebcKECd5zTpw4wX333UeXLl0YMWIEBw8eZOnSpfTv37/e2ydyPtGhgXx47wD6J0WTW1TKmLe+ZcmOI1aHJSIi9cBmGIZhdRD1KScnh8jISLKzs61dN+PQBnjjCgiKhMf2gt0Om/4Nc8dCVGv49UbzWHXsXw1vXQ1BUfDYbrA76iBwERER/9Rg+hByXrpHUp8Kit2M/8c6luw4QoDDxt/uuIRru2vqsYhIY1TVPoRGolmlfCpnq/6nkmWdrwdXBJxIg7SV1b/mrkXmY9LlSqCJiIiIiNSh4EAHf7+zL9f3iKfEbfCrf37Hh9+m4fH41RgFERG/oiSaVbxFBU6bnhAQDF1vMvc3zqr+Nb3roV1Zo9BERERERKRygU47L99xCbclt8JjwBMfb+L6vy3nf5vTlUwTEfFBSqJZ5fSiAqfrdYf5mPoJlBRU/XpFuWZRAoC2SqKJiIiIiNQHh93G8z/uyaMjOhLmcrI1PYfx//iO615epmSaiIiPURLNCjmHIDvNrMbZMrnia60HmmuiFefCtvlVv+beb8BTCk3aQHRSrYYrIiIiIiLnZ7fbePCqDiz/zZU8dFV7wlxOtmXkepNp/92kZJqIiC9QEs0K5aPQYruBK7zia3Y79Lzd3K9Olc7dZeuhtR1a4/BERERERKT6okICeWREJ5b/5kp+fVV7wsuSab/853eM/OsyFiiZJiLSqCmJZgXvVM4B5369V1kSbddXkHu4atcsLyqgqZwiIiIiIpaKCglk0ohOLP/NVfx6WAfCXU62H87lV2XJtPnfK5kmItIYKYlmBW9RgfMk0Zq2M6t2Gh7Y9K/Kr5d9EI5uB2xmZU4REREREbFcZEgAk67uyPLfXMWEYR0IDzKTaQ98+B3X/nUpn39/SMk0EZFGREm0+lZSCOkbzf0ziwqcrlc1pnSWV+VscQmERNcoPBERERERqV2RIQE8XJZMmzjcTKbtOHySBz9cz22vr2TzwWyrQxQRkSpQEq2+pW8ATwmExUJU4vnP63YLOALh8CbI2HTha5Yn0dppKqeIiIiISEMVGRzAxOGnkmkhgQ7W7TvODa8s54mPN3E8r9jqEEVE5AKURKtv3qmc/cFmO/95IdHQ8Vpz/0Kj0QzjVBJN66GJiIiIiDR45cm0rx8Zyk29W2AY8OG3aQx9YTEfrNqHW1M8RUQaJCXR6lt5UYFWF5jKWa7XHebjpn+Bu/Tc5xxOhbxMCAi58PRQERERERFpUOIig/jr7Zcw575L6RwXTnZBCU99spkb/racNXuzrA5PRETOoCRafTKMyosKnK79cAhpCicPnxptdqbdZVU5EweB01UrYYqIiIiISP0Z0LYpnz90GVNv6kZEkJMt6Tnc9tpKJs5ez+GcQqvDExGRMkqi1afjeyDviLnWWXyvys93BkL3n5j7G2ed+5xdZUk0TeUUEREREWm0nA47dw5sw6JHh3JH/wRsNvhkwyGuemExry/ZRXGpx+oQRUT8npJo9al8Kmd8bwgIqtp7yqt0bvscCnMqvlZSCPtWmPsqKiAiIiIi0ug1DXMx7daefPrAYHonRJFX7Gbaf7dx7V+XsmTHEavDExHxa5Yn0WbMmEFSUhJBQUEkJyezbNmy8567fPlyBg8eTNOmTQkODqZz58785S9/qcdoa+j0ogJV1eISiOkEpYWw5dOzr1daYFb6bN619uIUERERERFL9WwVxbxfDuLPP+lJTFggu4/kcdfbq7l2+lJe+GI769OO41EBAhGReuW08sPnzJnDxIkTmTFjBoMHD+b1119n5MiRbNmyhdatW591fmhoKA8++CA9e/YkNDSU5cuXc//99xMaGsp9991nQQuqqXwkWlXWQytns5mj0b56xqzS2WfMqde8VTmHXrjSp4iIiIiINDp2u43b+iZwTfc4/vrlTt5bsZdtGblsy8jllUU/EBMWyJWdmjOsSyxDOsQQ6rL01zsREZ9nMwzDsj9fDBgwgD59+jBz5kzvsS5dunDzzTczbdq0Kl3j1ltvJTQ0lA8++KBK5+fk5BAZGUl2djYREREXFfdFKcyB51oDBjyyHcLjqv7e7APwl+7meyd8D00SzeNvDIVD6+Hm16D3HXUQtIiIiJSzrA8hVaZ7JL4uK6+Yxdsz+WprJkt2HOFkUan3tUCnnYFtmzK8S3Ou6hJLy6hgCyMVEWlcqtqHsOxPFcXFxaxbt47HH3+8wvERI0awYsWKKl1j/fr1rFixgt///vfnPaeoqIiioiLv85ycnPOeW6cOrgMMiEqsXgINILIVJF0Oe5bA9x/BFf8H+VlwaIP5etuhtRysiIiIiIg0NNGhgdzapxW39mlFcamH1Xuy+HLrYb7adpj9WQUs2XGEJTuO8NSnqXSOC2d4l1h+1CueznFKKouI1AbL1kQ7evQobreb2NjYCsdjY2PJyMi44HtbtWqFy+Wib9++PPDAA4wbN+68506bNo3IyEjvlpCQUCvxV9vFTOU8Xa+ykWYbZ4FhmAk1DGjWBSLiayVEERERERFpHAKddi7rEMOUG7ux9P+uJOXhy/nNtZ3pm9gEuw3vlM9rpy/jF++sZs3eLKtDFhFp9CyfNG87Yy0vwzDOOnamZcuWcfLkSVatWsXjjz9O+/btueOOc09nnDx5MpMmTfI+z8nJsSaRdjFFBU7X5QaYPwmydsGBtbBrkXlco9BERERERPyazWajQ2w4HWLD+eXQdt5pn1+kZpCy5TCLth9h0fYj9G8TzS+vbMfQjs0q/Z1LRETOZlkSLSYmBofDcdaos8zMzLNGp50pKSkJgB49enD48GGmTJly3iSay+XC5XLVTtAXy+OBA2vM/YtNornCoMuN8P1s2Pgh7C5LorW7snZiFBERERERn3D6tM+9R/N4felu5q47wOq9Wax+J4uu8RH8cmg7rusRj8OuZJqISFVZNp0zMDCQ5ORkUlJSKhxPSUlh0KBBVb6OYRgV1jxrkI5sg6IcCAiF5t0u/jq9bjcfN8yCE2lgD4DEwbUTo4iIiIiI+Jw2MaFMu7UHy35zJfcOSSIk0MGW9BwemrWe4S8tYfbqNIpK3VaHKSLSKFg6nXPSpEmMGTOGvn37MnDgQN544w3S0tIYP348YE7FPHjwIO+//z4Ar776Kq1bt6Zz584ALF++nBdeeIGHHnrIsjZUSflUzlbJ4KjBlzzpcghvAbmHzOcJ/c0RaiIiIiIiIhcQGxHEk9d35VdD2/Peyr28881e9hzN4/F5m5j+5U7GDUnijv6tCXVZvuKPiEiDZel3yNGjR3Ps2DGmTp1Keno63bt3Z8GCBSQmJgKQnp5OWlqa93yPx8PkyZPZs2cPTqeTdu3a8dxzz3H//fdb1YSqqWlRgXJ2B/S8Db75q/m8raZyioiIiIhI1TUJDWTi8I7cO6Qts1an8fdlu8nIKeT387fyyqIfuPPSRK7uGke3FhHYNdVTRKQCm2EYhtVB1KecnBwiIyPJzs4mIqKeSj2/3McsCPCzf0OHq2t2rcNbYOZAc3/cV9Cqb83jExERkUpZ0oeQatE9Eqm+olI38747yGtLdrHvWL73eFRIAIPbxXBZhxguax9DQnSIhVGKiNStqvYhNFa3ruUdMxNoUDsJr9iuMHgiFJ6AFn1qfj0REREREfFbLqeDO/q35rbkVizYnMFnGw6xavcxTuSXMH9TOvM3pQOQ2DSEy9qbCbVB7WKIDAmwOHIRkfqnJFpdO1A2lbNZZwhuUjvXvPqZ2rmOiIiIiIgI4HTYubFXC27s1YISt4eN+0+w/IejLN95lPX7T7DvWD77jqXxz2/TsNugR6soLmvflMHtY2jfPIyYUJemf4qIz1MSra55iwr0szYOERERERGRKghw2OnbJpq+baKZOLwjuYUlfLs7y0yq/XCUHzJPsnH/CTbuP8Gri8xZN067jdiIIOIjg4iPCiY+Moi4sudxkUG0iAomJsyFQ4k2EWnElESra7VVVEBERERERMQC4UEBDO8ay/CusQCkZxfwzQ/HWL7zCKv3ZJGRU0ipx+DgiQIOniiAfcfPeR2H3UZsuIvWTUPoGBt+2hZGVEhgfTZJROSiKIlWl9wlcHCdua8kmoiIiIiI+ID4yGB+ktyKnyS3AqDE7eFIbhHp2YVkZBeSnl1gPuYUkn7C3D+cW4TbY3Aou5BD2YWs2p1V4ZrNw10Vkmodyh7Dg7T2mog0HEqi1aWM76G00FwLrWl7q6MRERERERGpdQEOOy2igmkRFXzec9weg6Mnizh0ooDdR/LYkZnLjoxcdhw+ycETBWTmFpGZW8TyH45WeF+LyCC6tohgcPsYhnRoRrtmodhsmhIqItZQEq0ulU/lbNUf7HZrYxEREREREbGIo2zNtNiIIC5pXbHg2smiUnYezmXHYTOptqNs/3BOkXfk2pdbMwFoGRXMkA4xXN6xGYNVJVRE6pmSaHWpvKhAQn9r4xAREREREWmgwlxOLmnd5KzkWnZ+CdsP57I+7ThLdx5hzZ7jHDxRwOw1+5m9Zj92G/RKiGJIh2Zc3iGG3glROB0avCAidUdJtLqkogIiIiIiIiIXJTIkgP5J0fRPiub+K9pRUOxm1Z5jLNtxlKU7j/BD5knWp51gfdoJXv5qJ+EuJ4PaN2Vw+xjCg5wYBuYGGIaBAWCAgeE97jEMAGLCXLRrFkZi0xAClIgTkfNQEq2uZB+AnINgc0DLPlZHIyIiIiIi0qgFBzq4slNzruzUHIBDJwpYvvMoS3Ye4ZsfjnIiv4QvUg/zRerhi/4Mp91Gm5hQ2jULpX3zMNo1C/M+hrr067OIv9N3gbpSPgotrgcEhlobi4iIiIiIiI9pERXMqH4JjOqXgNtjsOlgNst2HOG7tOOUeswRZjabDRtgs1H2eOo52LDZzNFqh3MK2XXkJPnFbn7IPMkPmSfPSsbFRwZ5E2oJ0SGEu5yEupyEuhyEB5XtBzoJKzse6NSINhFfoyRaXfFO5dR6aCIiIiIiInXJYbfROyGK3glRF30Nj8cgI6fQm0TbdeTU49GTxaRnF5KeXciynUcrvxgQ6LAT6nIQFmQm1xx2Gw67DZvNhsNmxmy3mZvDbsNuN4/bbea+3Qa2skSfrWyfcyQDy5+7nHaiQwOJDg2kaVgg0aEumpY9jw4NJCjAcdFfGxExKYlWV7xFBbQemoiIiIiISENnt9toERVMi6hgLu/YrMJrJ/KL2XXkJLsy8/jhyEkOnSggr6iUvCI3J4tKySsu5WRhKSeLSikq9QBQ7PZQnO/heH6JFc05S2igg+jTkmtRIQEE2O3Y7WYSzl6ewLOZiTv7acfKX28e7iKpWRhJTUNp2SQYh91mdbNE6pWSaHWhOB8yvjf3NRJNRERERESkUYsKCSQ5MZrkxOhKzy1xe8gvcnOyuJS8IjOxll/kptTjwWMYuD1mQQOPx8BtGLg9xjmPe8yKCGWFEU4VRygvigDmMcqOFZS4ycor5lheMVl5RRw7WUxWnrmVegzyit3kZRWwP6ugVr4mgQ47CdHBJMWEkhQTSpuyx6SYUGLDg7ArwSY+SEm0unBoPXhKITweIhOsjkZERERERETqSYDDTmSInciQAKtDAcxEW05haVlC7VRy7URBCW6PgVGWsPOUPZrPTx0zDHOqa6nH4NCJAvYey2PvsXyKSz3sOpLHriN5Z31mUICdNk1DaRbuOk9Mp+1z6onTbqdVk2DaNDWTcm2ahpAQHdJopqIahkFhiYcAhw2nqrz6JCXR6oJ3Kmf/8hUrRUREREREROqdzWYjMjiAyOAAkmJqp+id22OQnl3AnqN57D2ax56j+ew5epK9x/JJy8qnsMTDtoxctmXk1vizbDaIjwiiTUwoiU3NxFpiU3PEW+voEAIcNopKPRSXeih2eygq8VDsdlNU6jl1/LR9t2Gctp6c7bR15YAKz8115zyGQW5hKTmFJeQUlJJbWOLdzyksOe01c7+8qEVIoIMwl5OwICfhLifhQQHe52EuJ+FB5hbmCjDXznM5CfEWpnCUFa1wEhLgqNKoPsMwKCr1UFDsJr/ETUFxKfnFbvKLza9FVHAAsRFBxIQFKsFXA0qi1QVvUQGthyYiIiIiIiK+xWG30apJCK2ahDCkQ8X140rcHg4cL2DP0ZOcyC+pMK7ETEuV7Z8jL1RU4iEtK5+9x/LYdyyfvUfzyC0q5VB2IYeyC1mx61hdNanWlSewMnOLanytkMCypFrZY6DTTkGxm4IS8zMKit3kF5fiMSq/lt0GMWEu4iKDaB4eRFyki9jwIGIjg4iNCCIuIojYCBeRwQHYNCjoLEqi1TbDgANKoomIiIiIiIj/CXDYvWuj1ZRhGGTlFbO3LKG2r2wq6d5jeew5mkduYWmF8+02cDkdBDrtuJx2Ass27zGHWUjBu66cYU4nLX9+1rpzhoHNZiM8yElEcAARQQFEePdPHTv99VCXgxK3QW7ZKLWTRaeKTuQWlpB72vOThaXkFJpr5+WVraGXV+T2Pi9PipUn5I5U8esW6LATHOggJNBBcKADl9PB8bxijpwswu0xyMwtKkvuZZ/3GhFBTjrHR9A5LpzOcRF0jg+nU2w4oS7/TiP5d+vrQtZuyD8GDhfE9bQ6GhEREREREZFGyWaz0TTMRdMwF8mJTSq8ZhgGOQWlGBhmssxhb1DTFKNDA2v0/vL11U4WlZJfbCbd8orc5BWXUlzqITjgVJIsJNB5aj/Acd6vg9tjcOxkEYdzijicU0hGTiGZZY/lxw7nFHI8v4ScwlJW78li9Z6sCtdIbBpC57hwOsVF0CUunM7xEbSODvGbSq1KotW28vXQWvYBZ83+04iIiIiIiIjI2Ww2W4Mp3lAXbDYbwWWJMTh3gYbqcthtNI8IonlEED2IPO95hSVu9hzNY1tGDtvSc9makcu29Bwyc4vYdyyffcfy+SL1sPf84AAHraNDql2RNSjATnCAg+AAB0FlCcDgwLLnpyUJg8rOCQtycmWn5hfd/tqgJFptK0+itepnbRwiIiIiIiIiItUUFOCgS3wEXeIj4JJTx7Pyir2JtW0ZOWzLyGV7Ri4FJW62H655EYnKRAQ5+X7KNXX+OReiJFpti0o0p3G2Hmh1JCIiIiIiIiIitSI6NJBB7WIY1C7Ge8ztMdh7LI9DJwqqdS2PAUUlZnGE8iIJBSVuCsvWf/M+P+314ABHbTep2pREq21DJpmbiIiIiI+bMWMGf/7zn0lPT6dbt25Mnz6dIUOGnPf8JUuWMGnSJFJTU2nRogWPPfYY48ePr3DO3Llzeeqpp9i1axft2rXjD3/4A7fccktdN0VEREQugsNuo12zMNo1C7M6lHrRcFbdExEREZFGY86cOUycOJEnn3yS9evXM2TIEEaOHElaWto5z9+zZw/XXXcdQ4YMYf369TzxxBP8+te/Zu7cud5zVq5cyejRoxkzZgwbN25kzJgxjBo1im+//ba+miUiIiJyXjbDMAyrg6hPOTk5REZGkp2dTUREhNXhiIiISCOhPkRFAwYMoE+fPsycOdN7rEuXLtx8881MmzbtrPN/85vf8Nlnn7F161bvsfHjx7Nx40ZWrlwJwOjRo8nJyeG///2v95xrr72WJk2aMGvWrEpj0j0SERGRi1HVPoRGoomIiIhItRQXF7Nu3TpGjBhR4fiIESNYsWLFOd+zcuXKs86/5pprWLt2LSUlJRc853zXLCoqIicnp8ImIiIiUleURBMRERGRajl69Chut5vY2NgKx2NjY8nIyDjnezIyMs55fmlpKUePHr3gOee75rRp04iMjPRuCQkJF9skERERkUopiSYiIiIiF8Vms1V4bhjGWccqO//M49W55uTJk8nOzvZu+/fvr1b8IiIiItWh6pwiIiIiUi0xMTE4HI6zRohlZmaeNZKsXFxc3DnPdzqdNG3a9ILnnO+aLpcLl8t1sc0QERERqRaNRBMRERGRagkMDCQ5OZmUlJQKx1NSUhg0aNA53zNw4MCzzl+4cCF9+/YlICDgguec75oiIiIi9Ukj0URERESk2iZNmsSYMWPo27cvAwcO5I033iAtLY3x48cD5lTLgwcP8v777wNmJc5XXnmFSZMmce+997Jy5UreeuutClU3J0yYwOWXX87zzz/PTTfdxKeffsqXX37J8uXLLWmjiIiIyOmURBMRERGRahs9ejTHjh1j6tSppKen0717dxYsWEBiYiIA6enppKWlec9PSkpiwYIFPPzww7z66qu0aNGCl19+mR//+MfecwYNGsTs2bP57W9/y1NPPUW7du2YM2cOAwYMqPf2iYiIiJzJZpSv6OoncnJyiIyMJDs7m4iICKvDERERkUZCfYiGT/dIRERELkZV+xBaE01ERERERERERKQSSqKJiIiIiIiIiIhUQkk0ERERERERERGRSiiJJiIiIiIiIiIiUgkl0URERERERERERCqhJJqIiIiIiIiIiEgllEQTERERERERERGphJJoIiIiIiIiIiIilXBaHUB9MwwDgJycHIsjERERkcakvO9Q3peQhkf9PBEREbkYVe3n+V0SLTc3F4CEhASLIxEREZHGKDc3l8jISKvDkHNQP09ERERqorJ+ns3wsz+nejweDh06RHh4ODabrdavn5OTQ0JCAvv37yciIqLWr99Qqd3+025/bDP4Z7v9sc2gdvtTu6vbZsMwyM3NpUWLFtjtWhGjIarrfh7o/4q/tBn8s93+2Gbwz3b7Y5tB7fandtdVP8/vRqLZ7XZatWpV558TERHhN/84T6d2+w9/bDP4Z7v9sc2gdvuT6rRZI9Aatvrq54H+r/gTf2y3P7YZ/LPd/thmULv9SW338/RnVBERERERERERkUooiSYiIiIiIiIiIlIJJdFqmcvl4umnn8blclkdSr1Su/2n3f7YZvDPdvtjm0Ht9qd2+2Obpeb88d+NP7YZ/LPd/thm8M92+2ObQe32p3bXVZv9rrCAiIiIiIiIiIhIdWkkmoiIiIiIiIiISCWURBMREREREREREamEkmgiIiIiIiIiIiKVUBJNRERERERERESkEkqi1bIZM2aQlJREUFAQycnJLFu2zOqQ6tSUKVOw2WwVtri4OKvDqlVLly7lhhtuoEWLFthsNj755JMKrxuGwZQpU2jRogXBwcEMHTqU1NRUa4KtRZW1++677z7r3l966aXWBFtLpk2bRr9+/QgPD6d58+bcfPPNbN++vcI5vna/q9JmX7zXM2fOpGfPnkRERBAREcHAgQP573//633d1+5zucra7Yv3+kzTpk3DZrMxceJE7zFfvd9S+9TPUz/PV74/qJ/nH/088M++nvp56ufVZT9PSbRaNGfOHCZOnMiTTz7J+vXrGTJkCCNHjiQtLc3q0OpUt27dSE9P926bNm2yOqRalZeXR69evXjllVfO+fqf/vQnXnrpJV555RXWrFlDXFwcV199Nbm5ufUcae2qrN0A1157bYV7v2DBgnqMsPYtWbKEBx54gFWrVpGSkkJpaSkjRowgLy/Pe46v3e+qtBl87163atWK5557jrVr17J27VquuuoqbrrpJu8PVF+7z+Uqazf43r0+3Zo1a3jjjTfo2bNnheO+er+ldqmfp36eL31/UD/PP/p54J99PfXz1M87Xa3fb0NqTf/+/Y3x48dXONa5c2fj8ccftyiiuvf0008bvXr1sjqMegMYH3/8sfe5x+Mx4uLijOeee857rLCw0IiMjDRee+01CyKsG2e22zAM46677jJuuukmS+KpL5mZmQZgLFmyxDAM/7jfZ7bZMPzjXhuGYTRp0sR48803/eI+n6683Ybh2/c6NzfX6NChg5GSkmJcccUVxoQJEwzD8I//11I71M/zfernneLLPw/K+WM/zzD8t6+nfp5v3+f67OdpJFotKS4uZt26dYwYMaLC8REjRrBixQqLoqofO3fupEWLFiQlJXH77beze/duq0OqN3v27CEjI6PCfXe5XFxxxRU+f98BFi9eTPPmzenYsSP33nsvmZmZVodUq7KzswGIjo4G/ON+n9nmcr58r91uN7NnzyYvL4+BAwf6xX2Gs9tdzlfv9QMPPMD111/P8OHDKxz3l/stNaN+nvp55fzp+4Ov/jwo54/9PPC/vp76eern1fb9dtYoUvE6evQobreb2NjYCsdjY2PJyMiwKKq6N2DAAN5//306duzI4cOH+f3vf8+gQYNITU2ladOmVodX58rv7bnu+759+6wIqd6MHDmS2267jcTERPbs2cNTTz3FVVddxbp163C5XFaHV2OGYTBp0iQuu+wyunfvDvj+/T5Xm8F37/WmTZsYOHAghYWFhIWF8fHHH9O1a1fvD1Rfvc/nazf47r2ePXs23333HWvWrDnrNV//fy21Q/089fNO5w/fH3z150E5f+zngX/19dTPUz8P6ub/tZJotcxms1V4bhjGWcd8yciRI737PXr0YODAgbRr14733nuPSZMmWRhZ/fK3+w4wevRo73737t3p27cviYmJzJ8/n1tvvdXCyGrHgw8+yPfff8/y5cvPes1X7/f52uyr97pTp05s2LCBEydOMHfuXO666y6WLFnifd1X7/P52t21a1efvNf79+9nwoQJLFy4kKCgoPOe56v3W2qXv/07UT/P5G/3HXz3Z385f+zngX/19dTPUz/vdLV5vzWds5bExMTgcDjO+mtkZmbmWVlPXxYaGkqPHj3YuXOn1aHUi/IKVf5+3wHi4+NJTEz0iXv/0EMP8dlnn7Fo0SJatWrlPe7L9/t8bT4XX7nXgYGBtG/fnr59+zJt2jR69erFX//6V5++z3D+dp+LL9zrdevWkZmZSXJyMk6nE6fTyZIlS3j55ZdxOp3ee+qr91tqh/p5JvXzTP5238E3fh6U88d+HvhfX0/9PPXz6qqfpyRaLQkMDCQ5OZmUlJQKx1NSUhg0aJBFUdW/oqIitm7dSnx8vNWh1IukpCTi4uIq3Pfi4mKWLFniV/cd4NixY+zfv79R33vDMHjwwQeZN28eX3/9NUlJSRVe98X7XVmbz8UX7vW5GIZBUVGRT97nCylv97n4wr0eNmwYmzZtYsOGDd6tb9++/OxnP2PDhg20bdvWr+63XBz180zq5/nv9wdf+Hngj/08UF+vnPp5Z/OF+2xJP++iyhHIOc2ePdsICAgw3nrrLWPLli3GxIkTjdDQUGPv3r1Wh1ZnHnnkEWPx4sXG7t27jVWrVhk/+tGPjPDwcJ9qc25urrF+/Xpj/fr1BmC89NJLxvr16419+/YZhmEYzz33nBEZGWnMmzfP2LRpk3HHHXcY8fHxRk5OjsWR18yF2p2bm2s88sgjxooVK4w9e/YYixYtMgYOHGi0bNmyUbf7l7/8pREZGWksXrzYSE9P9275+fnec3ztflfWZl+915MnTzaWLl1q7Nmzx/j++++NJ554wrDb7cbChQsNw/C9+1zuQu321Xt9LqdXbTIM373fUrvUz1M/z5e+P6if5x/9PMPwz76e+nnq59VlP09JtFr26quvGomJiUZgYKDRp0+fCqWDfdHo0aON+Ph4IyAgwGjRooVx6623GqmpqVaHVasWLVpkAGdtd911l2EYZtncp59+2oiLizNcLpdx+eWXG5s2bbI26FpwoXbn5+cbI0aMMJo1a2YEBAQYrVu3Nu666y4jLS3N6rBr5FztBYx33nnHe46v3e/K2uyr9/qee+7xfq9u1qyZMWzYMG/HyjB87z6Xu1C7ffVen8uZnStfvd9S+9TPUz/PV74/qJ/nH/08w/DPvp76eern1WU/z2YYhnFxY9hERERERERERET8g9ZEExERERERERERqYSSaCIiIiIiIiIiIpVQEk1ERERERERERKQSSqKJiIiIiIiIiIhUQkk0ERERERERERGRSiiJJiIiIiIiIiIiUgkl0URERERERERERCqhJJqIiIiIiIiIiEgllEQTEamhxYsXY7PZOHHihNWhiIiIiEgtUj9PRE6nJJqIiIiIiIiIiEgllEQTERERERERERGphJJoItLoGYbBn/70J9q2bUtwcDC9evXi3//+N3BqCP78+fPp1asXQUFBDBgwgE2bNlW4xty5c+nWrRsul4s2bdrw4osvVni9qKiIxx57jISEBFwuFx06dOCtt96qcM66devo27cvISEhDBo0iO3bt9dtw0VERER8nPp5ItKQKIkmIo3eb3/7W9555x1mzpxJamoqDz/8MD//+c9ZsmSJ95z/+7//44UXXmDNmjU0b96cG2+8kZKSEsDsFI0aNYrbb7+dTZs2MWXKFJ566ineffdd7/vvvPNOZs+ezcsvv8zWrVt57bXXCAsLqxDHk08+yYsvvsjatWtxOp3cc8899dJ+EREREV+lfp6INCQ2wzAMq4MQEblYeXl5xMTE8PXXXzNw4EDv8XHjxpGfn899993HlVdeyezZsxk9ejQAWVlZtGrVinfffZdRo0bxs5/9jCNHjrBw4ULv+x977DHmz59PamoqO3bsoFOnTqSkpDB8+PCzYli8eDFXXnklX375JcOGDQNgwYIFXH/99RQUFBAUFFTHXwURERER36N+nog0NBqJJiKN2pYtWygsLOTqq68mLCzMu73//vvs2rXLe97pHa/o6Gg6derE1q1bAdi6dSuDBw+ucN3Bgwezc+dO3G43GzZswOFwcMUVV1wwlp49e3r34+PjAcjMzKxxG0VERET8kfp5ItLQOK0OQESkJjweDwDz58+nZcuWFV5zuVwVOlhnstlsgLnWRvl+udMH6QYHB1cploCAgLOuXR6fiIiIiFSP+nki0tBoJJqINGpdu3bF5XKRlpZG+/btK2wJCQne81atWuXdP378ODt27KBz587eayxfvrzCdVesWEHHjh1xOBz06NEDj8dTYe0NEREREalb6ueJSEOjkWgi0qiFh4fz6KOP8vDDD+PxeLjsssvIyclhxYoVhIWFkZiYCMDUqVNp2rQpsbGxPPnkk8TExHDzzTcD8Mgjj9CvXz+effZZRo8ezcqVK3nllVeYMWMGAG3atOGuu+7innvu4eWXX6ZXr17s27ePzMxMRo0aZVXTRURERHya+nki0tAoiSYijd6zzz5L8+bNmTZtGrt37yYqKoo+ffrwxBNPeIfZP/fcc0yYMIGdO3fSq1cvPvvsMwIDAwHo06cPH330Eb/73e949tlniY+PZ+rUqdx9993ez5g5cyZPPPEEv/rVrzh27BitW7fmiSeesKK5IiIiIn5D/TwRaUhUnVNEfFp5RaXjx48TFRVldTgiIiIiUkvUzxOR+qY10URERERERERERCqhJJqIiIiIiIiIiEglNJ1TRERERERERESkEhqJJiIiIiIiIiIiUgkl0URERERERERERCqhJJqIiIiIiIiIiEgllEQTERERERERERGphJJoIiIiIiIiIiIilVASTUREREREREREpBJKoomIiIiIiIiIiFRCSTQREREREREREZFK/D91kWL16BUdGgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1500x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for s in final_res:\n",
    "    print(s)\n",
    "    plot_stats([final_res[s][0]])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bb3e8f7f",
   "metadata": {},
   "source": [
    "# Final result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "3a0c4738",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>S loss</th>\n",
       "      <th>S acc</th>\n",
       "      <th>F loss</th>\n",
       "      <th>F acc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>emodb</th>\n",
       "      <td>0.379157</td>\n",
       "      <td>0.888889</td>\n",
       "      <td>0.312610</td>\n",
       "      <td>0.906810</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>emovo</th>\n",
       "      <td>0.384948</td>\n",
       "      <td>0.873354</td>\n",
       "      <td>0.344954</td>\n",
       "      <td>0.895643</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ravdess</th>\n",
       "      <td>0.357362</td>\n",
       "      <td>0.873255</td>\n",
       "      <td>0.281702</td>\n",
       "      <td>0.912898</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           S loss     S acc    F loss     F acc\n",
       "emodb    0.379157  0.888889  0.312610  0.906810\n",
       "emovo    0.384948  0.873354  0.344954  0.895643\n",
       "ravdess  0.357362  0.873255  0.281702  0.912898"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.concat([start_df, final_df], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 236,
   "id": "c115901e",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "\n",
    "with open('/datasets/nicolas_facchinetti/param.p', 'wb') as parameters:\n",
    "    pickle.dump({d: param_found[d].get_best_hyperparameters()[0].get_config()['values']\n",
    "                 for d in param_found}, parameters)"
   ]
  }
 ],
 "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": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}