Model backlog/Train/167-tweet-train-5fold-roberta-balanced-finetune.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "execution_count": null,
   "metadata": {},
   "source": [
    "## Dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_kg_hide-input": true,
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5"
   },
   "outputs": [],
   "source": [
    "import json, warnings, shutil\n",
    "from tweet_utility_scripts import *\n",
    "from tweet_utility_preprocess_roberta_scripts_aux import *\n",
    "from transformers import TFRobertaModel, RobertaConfig\n",
    "from tokenizers import ByteLevelBPETokenizer\n",
    "from tensorflow.keras.models import Model\n",
    "from tensorflow.keras import optimizers, metrics, losses, layers\n",
    "from scripts_step_lr_schedulers import *\n",
    "\n",
    "SEED = 0\n",
    "seed_everything(SEED)\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "execution_count": null,
   "metadata": {},
   "source": [
    "# Load data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "_kg_hide-input": true,
    "_kg_hide-output": true
   },
   "outputs": [],
   "source": [
    "# Unzip files\n",
    "!tar -xf /kaggle/input/tweet-dataset-5fold-roberta-base-96-complete/fold_1.tar.gz\n",
    "!tar -xf /kaggle/input/tweet-dataset-5fold-roberta-base-96-complete/fold_2.tar.gz\n",
    "!tar -xf /kaggle/input/tweet-dataset-5fold-roberta-base-96-complete/fold_3.tar.gz\n",
    "!tar -xf /kaggle/input/tweet-dataset-5fold-roberta-base-96-complete/fold_4.tar.gz\n",
    "!tar -xf /kaggle/input/tweet-dataset-5fold-roberta-base-96-complete/fold_5.tar.gz\n",
    "\n",
    "# balanced files\n",
    "!tar -xf /kaggle/input/tweet-dataset-5fold-roberta-base-96-complete/balanced_fold_1.tar.gz\n",
    "!tar -xf /kaggle/input/tweet-dataset-5fold-roberta-base-96-complete/balanced_fold_2.tar.gz\n",
    "!tar -xf /kaggle/input/tweet-dataset-5fold-roberta-base-96-complete/balanced_fold_3.tar.gz\n",
    "!tar -xf /kaggle/input/tweet-dataset-5fold-roberta-base-96-complete/balanced_fold_4.tar.gz\n",
    "!tar -xf /kaggle/input/tweet-dataset-5fold-roberta-base-96-complete/balanced_fold_5.tar.gz"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "_cell_guid": "79c7e3d0-c299-4dcb-8224-4455121ee9b0",
    "_kg_hide-input": true,
    "_uuid": "d629ff2d2480ee46fbb7e2d37f6b5fab8052498a"
   },
   "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>textID</th>\n",
       "      <th>text</th>\n",
       "      <th>selected_text</th>\n",
       "      <th>sentiment</th>\n",
       "      <th>jaccard</th>\n",
       "      <th>text_len</th>\n",
       "      <th>text_wordCnt</th>\n",
       "      <th>text_tokenCnt</th>\n",
       "      <th>selected_text_len</th>\n",
       "      <th>selected_text_wordCnt</th>\n",
       "      <th>selected_text_tokenCnt</th>\n",
       "      <th>fold_1</th>\n",
       "      <th>fold_2</th>\n",
       "      <th>fold_3</th>\n",
       "      <th>fold_4</th>\n",
       "      <th>fold_5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>cb774db0d1</td>\n",
       "      <td>i`d have responded, if i were going</td>\n",
       "      <td>i`d have responded, if i were going</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>35</td>\n",
       "      <td>7</td>\n",
       "      <td>10</td>\n",
       "      <td>35</td>\n",
       "      <td>7</td>\n",
       "      <td>10</td>\n",
       "      <td>train</td>\n",
       "      <td>train</td>\n",
       "      <td>train</td>\n",
       "      <td>validation</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>549e992a42</td>\n",
       "      <td>sooo sad i will miss you here in san diego!!!</td>\n",
       "      <td>sooo sad</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>45</td>\n",
       "      <td>10</td>\n",
       "      <td>13</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>validation</td>\n",
       "      <td>train</td>\n",
       "      <td>train</td>\n",
       "      <td>train</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>088c60f138</td>\n",
       "      <td>my boss is bullying me...</td>\n",
       "      <td>bullying me</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>25</td>\n",
       "      <td>5</td>\n",
       "      <td>6</td>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>train</td>\n",
       "      <td>validation</td>\n",
       "      <td>train</td>\n",
       "      <td>train</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>9642c003ef</td>\n",
       "      <td>what interview! leave me alone</td>\n",
       "      <td>leave me alone</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>30</td>\n",
       "      <td>5</td>\n",
       "      <td>6</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>train</td>\n",
       "      <td>validation</td>\n",
       "      <td>train</td>\n",
       "      <td>train</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>358bd9e861</td>\n",
       "      <td>sons of ****, why couldn`t they put them on th...</td>\n",
       "      <td>sons of ****,</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.214286</td>\n",
       "      <td>74</td>\n",
       "      <td>14</td>\n",
       "      <td>17</td>\n",
       "      <td>13</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>train</td>\n",
       "      <td>train</td>\n",
       "      <td>validation</td>\n",
       "      <td>train</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       textID                                               text  \\\n",
       "0  cb774db0d1                i`d have responded, if i were going   \n",
       "1  549e992a42      sooo sad i will miss you here in san diego!!!   \n",
       "2  088c60f138                          my boss is bullying me...   \n",
       "3  9642c003ef                     what interview! leave me alone   \n",
       "4  358bd9e861  sons of ****, why couldn`t they put them on th...   \n",
       "\n",
       "                         selected_text sentiment   jaccard  text_len  \\\n",
       "0  i`d have responded, if i were going   neutral  1.000000        35   \n",
       "1                             sooo sad  negative  0.200000        45   \n",
       "2                          bullying me  negative  0.166667        25   \n",
       "3                       leave me alone  negative  0.600000        30   \n",
       "4                        sons of ****,  negative  0.214286        74   \n",
       "\n",
       "   text_wordCnt  text_tokenCnt  selected_text_len  selected_text_wordCnt  \\\n",
       "0             7             10                 35                      7   \n",
       "1            10             13                  8                      2   \n",
       "2             5              6                 11                      2   \n",
       "3             5              6                 14                      3   \n",
       "4            14             17                 13                      3   \n",
       "\n",
       "   selected_text_tokenCnt      fold_1      fold_2      fold_3      fold_4  \\\n",
       "0                      10       train       train       train  validation   \n",
       "1                       3  validation       train       train       train   \n",
       "2                       2       train  validation       train       train   \n",
       "3                       3       train  validation       train       train   \n",
       "4                       4       train       train  validation       train   \n",
       "\n",
       "  fold_5  \n",
       "0  train  \n",
       "1  train  \n",
       "2  train  \n",
       "3  train  \n",
       "4  train  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "database_base_path = '/kaggle/input/tweet-dataset-5fold-roberta-base-96-complete/'\n",
    "k_fold = pd.read_csv(database_base_path + '5-fold.csv')\n",
    "display(k_fold.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "execution_count": null,
   "metadata": {},
   "source": [
    "# Model parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "vocab_path = database_base_path + 'vocab.json'\n",
    "merges_path = database_base_path + 'merges.txt'\n",
    "base_path = '/kaggle/input/qa-transformers/roberta/'\n",
    "\n",
    "config = {\n",
    "  'MAX_LEN': 96,\n",
    "  'BATCH_SIZE': 32,\n",
    "  'EPOCHS': 2,\n",
    "  'LEARNING_RATE': 3e-5,\n",
    "  'ES_PATIENCE': 2,\n",
    "  'question_size': 4,\n",
    "  'N_FOLDS': 5,\n",
    "  'base_model_path': base_path + 'roberta-base-tf_model.h5',\n",
    "  'config_path': base_path + 'roberta-base-config.json'\n",
    "}\n",
    "\n",
    "with open('config.json', 'w') as json_file:\n",
    "    json.dump(json.loads(json.dumps(config)), json_file)"
   ]
  },
  {
   "cell_type": "markdown",
   "execution_count": null,
   "metadata": {},
   "source": [
    "# Tokenizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "_kg_hide-output": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['./vocab.json', './merges.txt']"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tokenizer = ByteLevelBPETokenizer(vocab_file=vocab_path, merges_file=merges_path, \n",
    "                                  lowercase=True, add_prefix_space=True)\n",
    "tokenizer.save('./')"
   ]
  },
  {
   "cell_type": "markdown",
   "execution_count": null,
   "metadata": {},
   "source": [
    "## Learning rate schedule"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "_kg_hide-input": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Learning rate schedule: 3e-05 to 3e-05 to 2.04e-06\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "lr_min = 1e-6\n",
    "lr_max = config['LEARNING_RATE']\n",
    "train_size = len(k_fold[k_fold['fold_1'] == 'train'])\n",
    "step_size = train_size // config['BATCH_SIZE']\n",
    "total_steps = config['EPOCHS'] * step_size\n",
    "decay = .998\n",
    "\n",
    "rng = [i for i in range(0, total_steps, config['BATCH_SIZE'])]\n",
    "y = [exponential_schedule_with_warmup(tf.cast(x, tf.float32), warmup_steps=1, lr_start=lr_max, lr_max=lr_max, lr_min=lr_min, decay=decay) for x in rng]\n",
    "\n",
    "sns.set(style=\"whitegrid\")\n",
    "fig, ax = plt.subplots(figsize=(20, 6))\n",
    "plt.plot(rng, y)\n",
    "print(\"Learning rate schedule: {:.3g} to {:.3g} to {:.3g}\".format(y[0], max(y), y[-1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "execution_count": null,
   "metadata": {},
   "source": [
    "# Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "module_config = RobertaConfig.from_pretrained(config['config_path'], output_hidden_states=False)\n",
    "\n",
    "def model_fn(MAX_LEN):\n",
    "    input_ids = layers.Input(shape=(MAX_LEN,), dtype=tf.int32, name='input_ids')\n",
    "    attention_mask = layers.Input(shape=(MAX_LEN,), dtype=tf.int32, name='attention_mask')\n",
    "    \n",
    "    base_model = TFRobertaModel.from_pretrained(config['base_model_path'], config=module_config, name=\"base_model\")\n",
    "    last_hidden_state, _ = base_model({'input_ids': input_ids, 'attention_mask': attention_mask})\n",
    "\n",
    "    x_start = layers.Dropout(.1)(last_hidden_state)\n",
    "    x_start = layers.Dense(1)(x_start)\n",
    "    x_start = layers.Flatten()(x_start)\n",
    "    y_start = layers.Activation('softmax', name='y_start')(x_start)\n",
    "\n",
    "    x_end = layers.Dropout(.1)(last_hidden_state)\n",
    "    x_end = layers.Dense(1)(x_end)\n",
    "    x_end = layers.Flatten()(x_end)\n",
    "    y_end = layers.Activation('softmax', name='y_end')(x_end)\n",
    "    \n",
    "    model = Model(inputs=[input_ids, attention_mask], outputs=[y_start, y_end])\n",
    "    \n",
    "    return model"
   ]
  },
  {
   "cell_type": "markdown",
   "execution_count": null,
   "metadata": {},
   "source": [
    "# Train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "_kg_hide-input": true,
    "_kg_hide-output": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "FOLD: 1\n",
      "Train for 687 steps, validate for 171 steps\n",
      "\n",
      "EPOCH 1/2\n",
      "time: 281.4s loss: 4.4199 loss_start: 2.2118 loss_end: 2.2081 val_loss: 1.8919 val_loss_start: 0.9844 val_loss_end: 0.9075\n",
      "Saved model weights at \"model_fold_1.h5\"\n",
      "\n",
      "EPOCH 2/2\n",
      "time: 266.5s loss: 4.0319 loss_start: 2.0342 loss_end: 1.9976 val_loss: 1.8470 val_loss_start: 0.9704 val_loss_end: 0.8766\n",
      "Training finished saved model weights at \"last_model_fold_1.h5\"\n",
      "Train for 583 steps, validate for 171 steps\n",
      "\n",
      "EPOCH 1/2\n",
      "time: 265.6s loss: 4.7964 loss_start: 2.4273 loss_end: 2.3690 val_loss: 1.8586 val_loss_start: 0.9627 val_loss_end: 0.8959\n",
      "Saved model weights at \"model_fold_1.h5\"\n",
      "\n",
      "EPOCH 2/2\n",
      "time: 266.8s loss: 4.7831 loss_start: 2.4190 loss_end: 2.3641 val_loss: 1.8323 val_loss_start: 0.9546 val_loss_end: 0.8778\n",
      "Training finished saved model weights at \"last_model_fold_1.h5\"\n",
      "\n",
      "FOLD: 2\n",
      "Train for 687 steps, validate for 171 steps\n",
      "\n",
      "EPOCH 1/2\n",
      "time: 281.6s loss: 4.4631 loss_start: 2.2228 loss_end: 2.2403 val_loss: 2.0377 val_loss_start: 1.0384 val_loss_end: 0.9993\n",
      "Saved model weights at \"model_fold_2.h5\"\n",
      "\n",
      "EPOCH 2/2\n",
      "time: 267.4s loss: 4.0305 loss_start: 2.0342 loss_end: 1.9963 val_loss: 1.9440 val_loss_start: 0.9898 val_loss_end: 0.9542\n",
      "Training finished saved model weights at \"last_model_fold_2.h5\"\n",
      "Train for 583 steps, validate for 171 steps\n",
      "\n",
      "EPOCH 1/2\n",
      "time: 266.5s loss: 4.8134 loss_start: 2.4364 loss_end: 2.3770 val_loss: 1.9814 val_loss_start: 1.0302 val_loss_end: 0.9512\n",
      "Saved model weights at \"model_fold_2.h5\"\n",
      "\n",
      "EPOCH 2/2\n",
      "time: 267.7s loss: 4.7971 loss_start: 2.4362 loss_end: 2.3609 val_loss: 1.9607 val_loss_start: 0.9987 val_loss_end: 0.9621\n",
      "Training finished saved model weights at \"last_model_fold_2.h5\"\n",
      "\n",
      "FOLD: 3\n",
      "Train for 687 steps, validate for 171 steps\n",
      "\n",
      "EPOCH 1/2\n",
      "time: 280.7s loss: 4.4839 loss_start: 2.2277 loss_end: 2.2562 val_loss: 2.0063 val_loss_start: 0.9829 val_loss_end: 1.0234\n",
      "Saved model weights at \"model_fold_3.h5\"\n",
      "\n",
      "EPOCH 2/2\n",
      "time: 266.7s loss: 4.0987 loss_start: 2.0606 loss_end: 2.0381 val_loss: 1.9263 val_loss_start: 0.9704 val_loss_end: 0.9559\n",
      "Training finished saved model weights at \"last_model_fold_3.h5\"\n",
      "Train for 583 steps, validate for 171 steps\n",
      "\n",
      "EPOCH 1/2\n",
      "time: 265.8s loss: 4.8800 loss_start: 2.4696 loss_end: 2.4104 val_loss: 1.8959 val_loss_start: 0.9542 val_loss_end: 0.9417\n",
      "Saved model weights at \"model_fold_3.h5\"\n",
      "\n",
      "EPOCH 2/2\n",
      "time: 266.7s loss: 4.8707 loss_start: 2.4603 loss_end: 2.4104 val_loss: 1.9190 val_loss_start: 0.9724 val_loss_end: 0.9465\n",
      "Training finished saved model weights at \"last_model_fold_3.h5\"\n",
      "\n",
      "FOLD: 4\n",
      "Train for 687 steps, validate for 171 steps\n",
      "\n",
      "EPOCH 1/2\n",
      "time: 281.6s loss: 4.4745 loss_start: 2.2088 loss_end: 2.2657 val_loss: 1.9341 val_loss_start: 1.0059 val_loss_end: 0.9282\n",
      "Saved model weights at \"model_fold_4.h5\"\n",
      "\n",
      "EPOCH 2/2\n",
      "time: 267.6s loss: 4.0771 loss_start: 2.0412 loss_end: 2.0358 val_loss: 1.9318 val_loss_start: 0.9963 val_loss_end: 0.9355\n",
      "Training finished saved model weights at \"last_model_fold_4.h5\"\n",
      "Train for 583 steps, validate for 171 steps\n",
      "\n",
      "EPOCH 1/2\n",
      "time: 266.6s loss: 4.8670 loss_start: 2.4445 loss_end: 2.4225 val_loss: 1.9250 val_loss_start: 1.0000 val_loss_end: 0.9250\n",
      "Saved model weights at \"model_fold_4.h5\"\n",
      "\n",
      "EPOCH 2/2\n",
      "time: 268.0s loss: 4.8526 loss_start: 2.4371 loss_end: 2.4155 val_loss: 1.9074 val_loss_start: 0.9980 val_loss_end: 0.9094\n",
      "Training finished saved model weights at \"last_model_fold_4.h5\"\n",
      "\n",
      "FOLD: 5\n",
      "Train for 687 steps, validate for 171 steps\n",
      "\n",
      "EPOCH 1/2\n",
      "time: 282.1s loss: 4.4716 loss_start: 2.2305 loss_end: 2.2411 val_loss: 1.9674 val_loss_start: 1.0033 val_loss_end: 0.9641\n",
      "Saved model weights at \"model_fold_5.h5\"\n",
      "\n",
      "EPOCH 2/2\n",
      "time: 267.5s loss: 4.0358 loss_start: 2.0317 loss_end: 2.0041 val_loss: 1.9070 val_loss_start: 0.9731 val_loss_end: 0.9339\n",
      "Training finished saved model weights at \"last_model_fold_5.h5\"\n",
      "Train for 583 steps, validate for 171 steps\n",
      "\n",
      "EPOCH 1/2\n",
      "time: 266.4s loss: 4.8156 loss_start: 2.4353 loss_end: 2.3803 val_loss: 1.9195 val_loss_start: 0.9872 val_loss_end: 0.9323\n",
      "Saved model weights at \"model_fold_5.h5\"\n",
      "\n",
      "EPOCH 2/2\n",
      "time: 267.9s loss: 4.7981 loss_start: 2.4252 loss_end: 2.3730 val_loss: 1.8842 val_loss_start: 0.9652 val_loss_end: 0.9190\n",
      "Training finished saved model weights at \"last_model_fold_5.h5\"\n"
     ]
    }
   ],
   "source": [
    "AUTO = tf.data.experimental.AUTOTUNE\n",
    "strategy = tf.distribute.get_strategy()\n",
    "k_fold_best = k_fold.copy()\n",
    "history_list = []\n",
    "\n",
    "for n_fold in range(config['N_FOLDS']):\n",
    "    n_fold +=1\n",
    "    print('\\nFOLD: %d' % (n_fold))\n",
    "    # Load data\n",
    "    base_data_path = 'fold_%d/' % (n_fold)\n",
    "    x_train = np.load(base_data_path + 'x_train.npy')\n",
    "    y_train = np.load(base_data_path + 'y_train.npy')\n",
    "    x_valid = np.load(base_data_path + 'x_valid.npy')\n",
    "    y_valid = np.load(base_data_path + 'y_valid.npy')\n",
    "    \n",
    "    step_size = x_train.shape[1] // config['BATCH_SIZE']\n",
    "    valid_step_size = x_valid.shape[1] // config['BATCH_SIZE']\n",
    "\n",
    "    # Build TF datasets\n",
    "    train_dist_ds = strategy.experimental_distribute_dataset(get_training_dataset(x_train, y_train, config['BATCH_SIZE'], AUTO, seed=SEED))\n",
    "    valid_dist_ds = strategy.experimental_distribute_dataset(get_validation_dataset(x_valid, y_valid, config['BATCH_SIZE'], AUTO, repeated=True, seed=SEED))\n",
    "    train_data_iter = iter(train_dist_ds)\n",
    "    valid_data_iter = iter(valid_dist_ds)\n",
    "    \n",
    "    # Step functions\n",
    "    @tf.function\n",
    "    def train_step(data_iter):\n",
    "        def train_step_fn(x, y):\n",
    "            with tf.GradientTape() as tape:\n",
    "                probabilities = model(x, training=True)\n",
    "                loss_start = loss_fn(y['y_start'], probabilities[0], label_smoothing=0.2)\n",
    "                loss_end = loss_fn(y['y_end'], probabilities[1], label_smoothing=0.2)\n",
    "                loss = tf.math.add(loss_start, loss_end)\n",
    "            grads = tape.gradient(loss, model.trainable_variables)\n",
    "            optimizer.apply_gradients(zip(grads, model.trainable_variables))\n",
    "            # update metrics\n",
    "            train_loss.update_state(loss)\n",
    "            train_loss_start.update_state(loss_start)\n",
    "            train_loss_end.update_state(loss_end)\n",
    "        for _ in tf.range(step_size):\n",
    "            strategy.experimental_run_v2(train_step_fn, next(data_iter))\n",
    "\n",
    "    @tf.function\n",
    "    def valid_step(data_iter):\n",
    "        def valid_step_fn(x, y):\n",
    "            probabilities = model(x, training=False)\n",
    "            loss_start = loss_fn(y['y_start'], probabilities[0])\n",
    "            loss_end = loss_fn(y['y_end'], probabilities[1])\n",
    "            loss = tf.math.add(loss_start, loss_end)\n",
    "            # update metrics\n",
    "            valid_loss.update_state(loss)\n",
    "            valid_loss_start.update_state(loss_start)\n",
    "            valid_loss_end.update_state(loss_end)\n",
    "        for _ in tf.range(valid_step_size):\n",
    "            strategy.experimental_run_v2(valid_step_fn, next(data_iter))\n",
    "\n",
    "    # Train model\n",
    "    model_path = 'model_fold_%d.h5' % (n_fold)\n",
    "    \n",
    "    model = model_fn(config['MAX_LEN'])\n",
    "    optimizer = optimizers.Adam(learning_rate=lambda: exponential_schedule_with_warmup(tf.cast(optimizer.iterations, tf.float32), \n",
    "                                                                                       warmup_steps=1, lr_start=lr_max, \n",
    "                                                                                       lr_max=lr_max, lr_min=lr_min, decay=decay))\n",
    "    loss_fn = losses.categorical_crossentropy\n",
    "    train_loss = metrics.Sum()\n",
    "    valid_loss = metrics.Sum()\n",
    "    train_loss_start = metrics.Sum()\n",
    "    valid_loss_start = metrics.Sum()\n",
    "    train_loss_end = metrics.Sum()\n",
    "    valid_loss_end = metrics.Sum()\n",
    "        \n",
    "    metrics_dict = {'loss': train_loss, 'loss_start': train_loss_start, 'loss_end': train_loss_end, \n",
    "                    'val_loss': valid_loss, 'val_loss_start': valid_loss_start, 'val_loss_end': valid_loss_end}\n",
    "        \n",
    "    history = custom_fit(model, metrics_dict, train_step, valid_step, train_data_iter, valid_data_iter, \n",
    "                         step_size, valid_step_size, config['BATCH_SIZE'], config['EPOCHS'], config['ES_PATIENCE'], \n",
    "                         model_path, save_last=True)\n",
    "    \n",
    "    #################### Fine-tune on balanced set ####################\n",
    "    # Load data\n",
    "    x_train_fine = np.load('balanced_' + base_data_path + 'x_train.npy')\n",
    "    y_train_fine = np.load('balanced_' + base_data_path + 'y_train.npy')\n",
    "    \n",
    "    step_size_fine = x_train_fine.shape[1] // config['BATCH_SIZE']\n",
    "    \n",
    "    # Build TF datasets\n",
    "    train_dist_ds_fine = strategy.experimental_distribute_dataset(get_training_dataset(x_train_fine, y_train_fine, config['BATCH_SIZE'], AUTO, seed=SEED))\n",
    "    train_data_iter_fine = iter(train_dist_ds_fine)\n",
    "\n",
    "    history_fine = custom_fit(model, metrics_dict, train_step, valid_step, train_data_iter_fine, valid_data_iter, \n",
    "                              step_size_fine, valid_step_size, config['BATCH_SIZE'], config['EPOCHS'], config['ES_PATIENCE'], \n",
    "                              model_path, save_last=True)\n",
    "    # Join history\n",
    "    for key in history_fine.keys():\n",
    "        history[key] += history_fine[key]\n",
    "    history_list.append(history)\n",
    "    \n",
    "    # Make predictions (last model)\n",
    "    x_train = np.load(base_data_path + 'x_train.npy')\n",
    "    x_valid = np.load(base_data_path + 'x_valid.npy')\n",
    "    train_preds = model.predict(get_test_dataset(x_train, config['BATCH_SIZE']))\n",
    "    valid_preds = model.predict(get_test_dataset(x_valid, config['BATCH_SIZE']))\n",
    "    \n",
    "    k_fold.loc[k_fold['fold_%d' % (n_fold)] == 'train', 'start_fold_%d' % (n_fold)] = train_preds[0].argmax(axis=-1)\n",
    "    k_fold.loc[k_fold['fold_%d' % (n_fold)] == 'train', 'end_fold_%d' % (n_fold)] = train_preds[1].argmax(axis=-1)\n",
    "    k_fold.loc[k_fold['fold_%d' % (n_fold)] == 'validation', 'start_fold_%d' % (n_fold)] = valid_preds[0].argmax(axis=-1)\n",
    "    k_fold.loc[k_fold['fold_%d' % (n_fold)] == 'validation', 'end_fold_%d' % (n_fold)] = valid_preds[1].argmax(axis=-1)\n",
    "    \n",
    "    k_fold['end_fold_%d' % (n_fold)] = k_fold['end_fold_%d' % (n_fold)].astype(int)\n",
    "    k_fold['start_fold_%d' % (n_fold)] = k_fold['start_fold_%d' % (n_fold)].astype(int)\n",
    "    k_fold['end_fold_%d' % (n_fold)].clip(0, k_fold['text_len'], inplace=True)\n",
    "    k_fold['start_fold_%d' % (n_fold)].clip(0, k_fold['end_fold_%d' % (n_fold)], inplace=True)\n",
    "    k_fold['prediction_fold_%d' % (n_fold)] = k_fold.apply(lambda x: decode(x['start_fold_%d' % (n_fold)], x['end_fold_%d' % (n_fold)], x['text'], config['question_size'], tokenizer), axis=1)\n",
    "    k_fold['prediction_fold_%d' % (n_fold)].fillna(k_fold[\"text\"], inplace=True)\n",
    "    k_fold['jaccard_fold_%d' % (n_fold)] = k_fold.apply(lambda x: jaccard(x['selected_text'], x['prediction_fold_%d' % (n_fold)]), axis=1)\n",
    "\n",
    "    # Make predictions (best model)\n",
    "    model.load_weights(model_path)\n",
    "    train_preds = model.predict(get_test_dataset(x_train, config['BATCH_SIZE']))\n",
    "    valid_preds = model.predict(get_test_dataset(x_valid, config['BATCH_SIZE']))\n",
    "    \n",
    "    k_fold_best.loc[k_fold_best['fold_%d' % (n_fold)] == 'train', 'start_fold_%d' % (n_fold)] = train_preds[0].argmax(axis=-1)\n",
    "    k_fold_best.loc[k_fold_best['fold_%d' % (n_fold)] == 'train', 'end_fold_%d' % (n_fold)] = train_preds[1].argmax(axis=-1)\n",
    "    k_fold_best.loc[k_fold_best['fold_%d' % (n_fold)] == 'validation', 'start_fold_%d' % (n_fold)] = valid_preds[0].argmax(axis=-1)\n",
    "    k_fold_best.loc[k_fold_best['fold_%d' % (n_fold)] == 'validation', 'end_fold_%d' % (n_fold)] = valid_preds[1].argmax(axis=-1)\n",
    "    \n",
    "    k_fold_best['end_fold_%d' % (n_fold)] = k_fold_best['end_fold_%d' % (n_fold)].astype(int)\n",
    "    k_fold_best['start_fold_%d' % (n_fold)] = k_fold_best['start_fold_%d' % (n_fold)].astype(int)\n",
    "    k_fold_best['end_fold_%d' % (n_fold)].clip(0, k_fold_best['text_len'], inplace=True)\n",
    "    k_fold_best['start_fold_%d' % (n_fold)].clip(0, k_fold_best['end_fold_%d' % (n_fold)], inplace=True)\n",
    "    k_fold_best['prediction_fold_%d' % (n_fold)] = k_fold_best.apply(lambda x: decode(x['start_fold_%d' % (n_fold)], x['end_fold_%d' % (n_fold)], x['text'], config['question_size'], tokenizer), axis=1)\n",
    "    k_fold_best['prediction_fold_%d' % (n_fold)].fillna(k_fold_best[\"text\"], inplace=True)\n",
    "    k_fold_best['jaccard_fold_%d' % (n_fold)] = k_fold_best.apply(lambda x: jaccard(x['selected_text'], x['prediction_fold_%d' % (n_fold)]), axis=1)\n",
    "    \n",
    "    ### Delete data dir\n",
    "    shutil.rmtree(base_data_path)\n",
    "    shutil.rmtree('balanced_' + base_data_path)"
   ]
  },
  {
   "cell_type": "markdown",
   "execution_count": null,
   "metadata": {},
   "source": [
    "# Model loss graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "_kg_hide-input": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fold: 1\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fold: 2\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fold: 3\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIUAAAN0CAYAAAAqJZifAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOzde3xU9Z3/8feZyeQ+uZE7uahbwUsCwSBURasRRasIuqW1rY+HbhVUqFiVqm1VClYpVqUtaHmI9dJudSmtoixoXcS6XvrYtfl5aaxVViwRAgEMCcmQZJKZ8/sjycnckpmEhBkyr+djt5k53++c8zmTL0nm7fd8j2GapikAAAAAAADEFVu0CwAAAAAAAMDRRygEAAAAAAAQhwiFAAAAAAAA4hChEAAAAAAAQBwiFAIAAAAAAIhDhEIAAAAAAABxiFAIAACgV01Njd5+++1olwEAAHBUEAoBAAAAAADEIUIhAAAAAACAOEQoBAAAEMDtduu+++7TjBkzNGPGDN13331yu92SpKamJl1//fWaOnWqpk2bpm9961vyer2SpMcee0xnn322pkyZolmzZukvf/lLNE8DAABgUAnRLgAAACDW/OpXv9L777+vF154QYZhaOHChXr00Uf1ve99T08++aQKCgqswOf999+XYRjasWOHfve73+kPf/iDCgoKtGvXLissAgAAiEXMFAIAAAiwadMmLVq0SOPGjVNOTo4WLVqkF198UZKUkJCg/fv3q6GhQQ6HQ1OnTpVhGLLb7XK73fr000/V1dWlkpISlZWVRflMAAAABkYoBAAAEGDfvn0qLi62nhcXF2vfvn2SpGuvvVbl5eX6zne+o/PPP1+PPfaYJKm8vFw//OEPtXr1ap155pm65ZZb1NjYGJX6AQAAIkEoBAAAECA/P18NDQ3W8z179ig/P1+SlJ6erjvvvFOvvvqq1q5dqyeffNK6lGz27Nl69tln9dprr8kwDD344INRqR8AACAShEIAAAABLrnkEv3qV79SU1OTmpqa9Mgjj2j27NmSpNdee007d+6UaZpKT0+X3W6XzWbTjh079Je//EVut1uJiYlKSkqS3W6P8pkAAAAMjIWmAQAAAixcuFAul0uXXXaZJOmiiy7SwoULJUk7d+7Uvffeq6amJmVkZOib3/ympk+frn/84x966KGH9Omnn8rhcGjKlClavnx5NE8DAABgUIZpmma0iwAAAAAAAMDRxeVjAAAAAAAAcYhQCAAAAAAAIA4RCgEAAAAAAMShiBaarqmpse6iIUlLlizR2Wef7dfH4/HoJz/5id544w0ZhqEFCxZo3rx5I18xAAAAAAAAjljEdx/75S9/qQkTJgzYvmnTJtXX1+uVV15Rc3Oz5s6dqzPOOEMlJSVh922apnX7VsMwIi0JAAAAAAAAwzRil49t2bJF8+bNk81mU05OjmbOnKmXX345ote63W7V1dXJ7XaPVDlR9eGHH0a7BGBQjFHEOsYoYh1jFLGOMYpYxxhFrIuXMRrxTKElS5bINE1VV1fr1ltvVUZGhl/7nj17VFxcbD0vKirS3r17R67SY0hHR0e0SwAGxRhFrGOMItYxRhHrGKOIdYxRxLp4GaOGaZpmuE579uxRUVGR3G637rvvPrlcLj344IN+fWbPnq377rtPkyZNkiStW7dOjY2Nuuuuu8IW0dnZqbq6umGeAgAAAAAAAEKprq4esC2imUJFRUWSpMTERH3rW9/SjTfeGLJPQ0ODFQoFzhyKREVFhbWY9bGstrZ20DcdiDbGKGIdYxSxjjGKWMcYRaxjjCLWxcsYDbum0OHDh9Xa2iqpZ0HoLVu26OSTTw7qd9FFF2nDhg3yer1qamrS1q1bNWvWrJGvGAAAAAAAAEcs7EyhL774QjfddJM8Ho+8Xq/+5V/+RUuXLpUkzZ8/X4sXL1ZlZaXmzJmj999/XxdeeKEkadGiRSotLR3d6gEAAAAAADAsYUOh0tJSbdy4MWTbunXrrMd2u13Lli0bucoAAAAAAAAwakbslvQAAAAAAAA4dhAKAQAAAAAAxCFCIQAAAAAAgDgU0S3pAQAAAMSerm6vdu9vU/3eQ6rf26r6xlbV7z2k/QcPy/7HPb29jJ7/Nfoe9T5R/3Ojv8HvufVVhm9z+NcFHifE9sB9DPkYYY5t9HcY/DgRvC8D1zbAMcIcO6DZv32gYwzx/AY8RgTvS+AxAt7q4GMMo7YDB5r01v+963+coJqNEMcOU1ukNQ+wnwHbjYFfM9Axwv87GG5tRsT/DoZfW/ifEUf6bzTif28R/YwI83NgGD8f9rd0KR4QCgEAAAAxrtvj1Z4DLu3sC3/2tqq+8ZAa9rvk8ZqSJJvNUHFumsqLMlSaY6iwsEBmT5NMWQ/6Hsnsb/T9Ym3vez5Qu9Vshtne+8qAZpmmGXyMEPVGVtvAxwi9feDXDXiMIR87uN3v+xCitkiP0d/u39/aV6TvS+DrfGsNsW1kaut50OnuUv0X+yI6TtAxwr4m/PkNVluo10V+DIwVdpt0UU20qxh9hEIAAABAjPB4Te39wmXN/Nm5t2fmz+79ber29HzqNAypcFyaygqcOqOyWGUFTpUVOlWSny5Hgl2SVFtbq+rqimieCjConjFaHe0yRk1g6Df8wCo4IB04gB38GOGC28gD4GG8JlzQGya4HOw4gec35NA4sH/v19312xUPCIUAAACAo8zrNdXYdLgn/Gls7Q2ADmnXvjZ1dXutfgU5qSordGrqyQUqL8pQWYFTJQVOJTnsUaweQDgDXRLnc/EUYtzhpviIS+LjLAEAAIAo8HpN7W9uD1rzp76xTe4uj9UvLztFZQVOVU3IV1mBU+VFTpXmO5WcxJ/rAIDRw28ZAAAA4AiZpqkvWjqsGT99a/583tiq9s7+8CcnI1nlhU5dfMZxKivsueyrrMCp1GRHFKsHAMQrQiEAAAAgQqZp6mBrZ9CaP583tsrV0W31y3ImqazAqZnTyq01f8oKnEpPTYxi9QAA+CMUAgAAAEJobu1UfWP/3b76ZgC1tfffpjgjLVFlhU595bQSlRVmqLzQqbLCDGWkEf4AAGIfoRAAAADi2iGX22/B575Lv1ra3Faf9BSHygqdmlE13pr5U16YoSxnUhQrBwDgyBAKAQAAIC642rv6Z/z0Lfi8t1UHWzutPilJCSordGr6qUXWJV9lhU7lZCRbdxMCAGCsIBQCAADAmHK4o0uf9836aWzVzj09IdAXLR1Wn+REu0oLnDrtpHyVF2b0BkAZys0i/AEAxA9CIQAAAByTOjq79fk+30u+emYB7T/YbvVJdNhVWpCuSV/KVVlv+FNemKG8rBTZbIQ/AID4RigEAACAmNbZ5dGuxtagNX8amw7LNHv6JNhtKi1I1ynHjVPZl53W7d4LctJkJ/wBACAkQiEAAADEhK5uj3bvd6l+7yHrVu/1e1u19wuXvFb4Y6g4L10nlmbr/NPLrDV/isalyW63RfcEAAA4xhAKAQAA4Kjq9njVsL/NmvnTd6v3hgMueXvTH5vNUHFumo4rztA5U0pUXtSz6HNxXroSCH8AABgRhEIAAAAYFR6PV3u+cFnr/fQFQA3729Tt6Ql/DEMqGpemskKnzpxUrLICp8qLMjQ+L02OBHuUzwAAgLGNUAgAAABHxOs11dh02Jrx07fmz659berq9lr9CnJSVV6YoWmnFFq3ey8pcCrJQfgDAEA0EAoBAAAgIl6vqf3N7f5r/jS26vPGNrm7PFa/vOwUlRU4VTUhX+W9Cz6X5juVnMSfngAAxBJ+MwMAAMCPaZo60Nyh+sZDfmv+fN7Yqg53f/gzLjNZZQVOXXzGcb23eneqtMCp1GRHFKsHAACRIhQCAACIU6ZpqulQh9+aP32zfw53dFv9spxJKi906oLp5dbdvsoKM5SeQvgDAMCxjFAIAAAgDjS3dvav+dPYf7v3tvYuq09GWqLKCp06r7rUWvOnrDBDGWmJUawcAACMFkIhAACAMeSQy23N9tm555A1A+iQy231SU9xqKzQqRlV4601f8oKMpTlTIpi5QAA4GgjFAIAADgGtbV3WbN9fAOg5tZOq09qcoLKCpz6ckWRNfOnvChD2c4kGYYRxeoBAEAsIBQCAACIYYc7unzW+2m17vzVdKjD6pOcaFdpgVNTTyroXe+nZ+ZPblYy4Q8AABgQoRAAAEAM6Ojs1uf7WrVzj8+aP42t2n+w3eqT6LCrtCBdk0/MVXlhhrXgc15Wimw2wh8AADA0QwqF1qxZo9WrV2vTpk2aMGGCX9vq1av1zDPPKD8/X5J02mmnaenSpSNXKQAAwBjQ2eXRrsb+u331Lf687+BhmWZPH0eCTSX56TrluHEq+7Kzd92fDOXnpMpO+AMAAEZIxKHQhx9+qPfee0/FxcUD9pk7d67uuOOOESkMAADgWNbV7dGufW1+d/vaubdVjV+45O0NfxLshsbnpWtCWbZmTiuzbvdeNC5NdrstuicAAADGvIhCIbfbreXLl+vBBx/U1VdfPdo1AQAAHDO6PV7t3t/Wv+ZPY8/Mn4YDLnl70x+bzVBxbppOKM7UuaeVWIs+F+elK4HwBwAARElEodAvfvELXXbZZSotLR203+bNm/Xmm28qLy9PN910k6ZMmTIiRR5LzL553wAAYEzxeLza84VL9Xtb9Ze/HdLWD99RfWOrGva3qdvTG/4YUuG4NJUVOnXmpGLrsq/xeWlyJNijfAYAAAD+DDNMivHuu+9q1apVevrpp2UYhmpqarR27dqgNYX279+vrKwsORwOvfXWW1qyZIm2bNmi7OzssEV0dnaqrq7uyM4kRvxm2359fsCttCSbUpNsSku2Ky3Z53GI7YkJ/BdCAABihddr6qDLo/0tXdrX0qV9zV3a39KtA4e65PH298tOtysv06H8TIfyMhOUn+VQrtMhRwJr/gAAgNhRXV09YFvYmULvvPOOduzYofPPP1+StHfvXl177bVasWKFZsyYYfXLy8uzHp911lkqKirS9u3bNW3atIgLraioUFJSUsT9Y5Enea+2vv2hktOy1NLmVourU7ua3GppO6yubm/I1yQl2pWZnqTMtMSer+mJykzr/ZqepMz0JGX4tCUnctM4HJna2tpBfzAA0cYYxdHg9Zrad/Cwz+3ee9b82dXYKrfP7+y87BSVF+XorCqndbv3Aw3/pzOmnx7F6oHB8XMUsY4xilgXL2M0bLqwYMECLViwwHo+0EyhxsZGFRQUSJI++ugj7d69W8cff/wIlxv7pp1aKHvH7qDBY5qm2ju7dcjlVnNbpw61udXS1tnz2NXzuMXl1sHWDv1zzyG1tHUOHiKl+QdGWb2BUUZAmJSZlqjkJEIkAED8Mk1TB5o7rLt89a3583ljqzrcHqvfuMxklRU4Nems460Fn0sLnEpNdgTt89A+ZvkCAIBj3xGlBfPnz9fixYtVWVmphx9+WB9++KFsNpscDoceeOABv9lD8c4wDKUmO5Sa7FDhuLSw/X1DpJa2zp5ZR73BUc/znsfNvSHSobZOv/+q6asvRMpIT1KWz6yjLEIkAMAYYpqmmg519N7mvWfmT31jT/hzuKPb6pftTFJZoVMXTC/vWfOnIEOlhU6lpwSHPwAAAGPZkD/9b9u2zXq8bt066/HKlStHpiJIGl6I1OH2+AVGLa2hQ6Sdew+ppXXgECnRYe8JjPwuaQu4vM1nhhIhEgDgaDJNU81tnT53+2rVzj09AZCrvcvql5GWqPLCDJ1XXWrd7ausMEMZaYlRrB4AACB28Gl+jDAMQylJCUpJShh2iHSorVPNvTOS+i5za2nrVH1ja9gQyZptFBQiBa+LlJxol2GwCCcAILy+30P1PjN/du5pVetht9UnPcWh8qIMnVM13lrzp6wgQ1nOY3udQgAAgNFGKBSnjiRE6l8XqeeytsB1keobW9XS5pa7yxNyX1aINNC6SAHhEiESAIx9be1d1kLP9dbaP61qbu20+qQmJ6iswKkzKot6b/XeM/Mn25nE7wkAAIBhIBRCRIYaIklSR2e3f2DUtzZSwCVtYUOkBJsynUkDr4tEiAQAx4zDHV3WzB9r4ee9rWo61GH1SUmyq7TAqaknFais0KnywgyVFTo1LjOZn+8AAAAjiFAIoyY5KUGFQwyRWgIDpBCLa+9qbFVzmBApY4DAKDMt0QqY+mYopSQl8CEDAEZYR2d3/2Vfjf0B0IHmdqtPosOusoJ0VU3Is+72VV6YodysFNls/FwGAAAYbYRCiBnJSQlKTkpQQU5qRP1DhUiHXMHrIu1qbFWLy61O9+Ah0sDrIiUpIz3RmqFEiAQA/Tq7PPo8YM2f+r2tamw6bPVxJNhUkp+uihPG+S34nJ+TKjvhDwAAQNQQCuGYdSQh0iGXW82tnTrkCl4XKVyI5Eiw9d+FLS34bmx9IVJfGyESgLGgq9ujXfvagtb82fuFS6bZ0yfBbmh8XromlGVr5rSy3nV/MlSYkyq73RbdEwAAAEAQQiHEjSMNkfzXReq/vG3X/ja1tHUOHiJZl635zzryv6SNEAlA9HV1e9VwoM1a66fvsq89X7jk9fakPzabofF5aTqhOFPnnlZiXfZVlJumBMIfAACAYwahEDCAIYdI7m4dCrwbm7Uukn+IdKitUx1hQiRrUe2AGUl+l7cRIgEYJo/Hq4YDLr/bve/c26qG/W3y9IU/hlQ4Lk3lRRmaMbnYutvX+Lw0ORLsUT4DAAAAHClCIWCEJCcmKDknQflDDJF8A6P+dZH6H4cLkRLsNv9FtX1mHWWk9Sy47bsuUkoS/+yBeOLxmmr8wtVz2Vdj/92+du1rU7fHK0kyDKkgJ1VlBRn6ckWhteZPSX66Eh2EPwAAAGMVnw6BKBnJEKl/XaRONRxwqSVMiJSSaCjv9T8HzTrK6A2TfGcopSYzEwk4Fni9pvYdPNx/yVfvDKBdja1yd3utfvnZKSorzNBpE/NVXuRUWUGGSgrSlZzInwQAAADxhr8AgWPEUEOkzi5P8J3ZWnu+frqzQY7kZLW09YRIh1ydau8cOEQKWlTbeu6z4LaTEAk4GkzT1P7mdv81fxp7wh/fMHhcZrLKCzM06Uu51u3eSwucSk12RLF6AAAAxBJCIWCMSnLYlZ+dqvzs4BCptrZD1dXVftv6QqRDPrOOfGcktbh62oYSImUEzDrqXxep/zEhEhCaaZpqOtTRe7cv/zt+tXd2W/2ynUkqK3Tqwunlvbd7z1BpoVPpKYQ/AAAAGByhEABJg4dIofiGSD2XtIUOkfZ+0XM528AhktG79pFPeORMDFgXqfdxepLSCJEwxpimqea2TtXvadVOnzV/6htb5WrvsvplpieqrCBDNVNLrVu9lxU65UxNjGL1AAAAOJYRCgEYlqGGSO4uT8i7sbX03q2tuc03RHL7zYTw1RciBc46smYl+a2LRIiE2NLS1tmz1s+eQ9pp3fWrVa2H3VYfZ6pDZYUZOqdqfO/dvnpm/2Q5k6JYOQAAAMYiQiEAR0Wiw6687BTlZadE1N8KkVyB6yL1hEh9odKQQiSfdZCC1kXqXXCbEAkjoe2wu/duXz0BUN+iz81tnVaf1OQElRdm6MxJRdaaP2WFGcp2JjEGAQAAcFQQCgGISUcSIg22LtLe+khCpP5ZSAOui5SepMy0RKWlOPgAH8cOd3T1Lvbse7v3Q2o61B/+pCTZVVrg1NSTC6y7fZUVOjUuM5mxAwAAgKgiFAIwJgwnRPK9bG2gdZEa6w+rua0zbIjkty5Sb2BkzUpKS1KWkxDpWNbe2a3PG3sCH2sG0N5WHWhut/okOuwqK0hX1YR8lRU4VV6UobICp3KzUmSz8T0HAABA7CEUAhCXEh125WalKDdraCGSb2AUal2kxvrDanF16nBH6BDJbjOsoChw1lFGepKyfNqy0pMIkY6yzi5Pb/jjHwDtazps9XEk2FSa71TFCeN61/vpCYDys1MJfwAAAHBMIRQCgAgMNUTq6vYEzToKXhepU9vrm8OGSBk+s478L2nzn5HUsyaSg2AiAu4uj3bvb+u93Xv/Hb/2Nrlkmj19EuyGxuel66SybF04rcxa86dwXJrsvMcAAAAYAwiFAGAUOBKOIERyuXWorVPNvSGS74yk7Qeb1dIWWYiUkZbof0lbHIZIXd1eNexv61n3x2fNnz0HXPL2hj82m6HxeWk6YXymzqsusW71XpSbpgS7LbonAAAAAIwiQiEAiAHDCZEOudxqbu0PkVp8L2/rC5F2NetQW6dcA4RINpthBUUZAYFRf4DUGzA5YzdE8ni8ajjg6r/sq/cSsIb9bfL0pj82QyrKTVNZYYZmTO653Xt5YYaK89LlSCD8AQAAQPwhFAKAY5Ajwa5xmSkalzm0EKml785sIUKkQy63/i+CEMmagZTmEx45fddF6m9LTxnZEMnjNdX4hcv/sq/GVu3a16Zuj1eSZBhSYU6aygqd+nJFYe/t3jNUkp+uRId9xGoBAAAAjnWEQgAQB44kROqbddQXIvXNUDrkcuvTXT2Xs4ULkXxnHfktqh0wI6kvRPKapvZ+4eq93Xv/mj+79rXK3e219p+fnaKywgxVn5Tfu+hzhkoK0pWcyK83AAAAIBz+agYABBl6iOT1X//IFXBnNt8QyeWWq70r5H5sNkMZqYlytbvV5dltbc/NTFZZYYYmnZir8sL+mT+pyY4ROV8AAAAgHhEKAQCOmCPBNqwQKXBdpOa+EOngFzp90r+ovDBDpQVOpaUQ/gAAAAAjjVAIAHDUhQuRamtrVV193NEtCgAAAIgz3G4FAAAAAAAgDg0pFFqzZo0mTpyoTz75JKjN4/Fo2bJlmjlzpi644AJt2LBhxIoEAAAAAADAyIr48rEPP/xQ7733noqLi0O2b9q0SfX19XrllVfU3NysuXPn6owzzlBJScmIFQsAAAAAAICREdFMIbfbreXLl2vp0qUyDCNkny1btmjevHmy2WzKycnRzJkz9fLLL49osQAAAAAAABgZEYVCv/jFL3TZZZeptLR0wD579uzxm0VUVFSkvXv3HnmFAAAAAAAAGHFhLx9799139be//U1LliwZ9WLq6upG/RhHS21tbbRLAAbFGEWsY4wi1jFGEesYo4h1jFHEurEyRqurqwdsCxsKvfPOO9qxY4fOP/98SdLevXt17bXXasWKFZoxY4bVr6ioSA0NDZo0aZKk4JlDkaioqFBSUtKQXhOLem6lPPCbDkQbYxSxjjGKWMcYRaxjjCLWMUYR6+JljIa9fGzBggV68803tW3bNm3btk2FhYX69a9/7RcISdJFF12kDRs2yOv1qqmpSVu3btWsWbNGrXAAAAAAAAAM35BuSR9o/vz5+tvf/iZJmjNnjkpKSnThhRfq61//uhYtWjToGkQAAAAAAACInohvSd9n27Zt1uN169ZZj+12u5YtWzYyVQEAAAAAAGBUHdFMIQAAAAAAABybCIUAAAAAAADiEKEQAAAAAABAHCIUAgAAAAAAiEOEQgAAAAAAAHGIUAgAAAAAACAOEQoBAAAAAADEIUIhAAAAAACAOEQoBAAAAAAAEIcIhQAAAAAAAOIQoRAAAAAAAEAcIhQCAAAAAACIQ4RCAAAAAAAAcYhQCAAAAAAAIA4lRLsAAAAAAABiSUtLiw4cOCC32x3tUhAlCQkJ+uijj6JdxqASExOVm5urzMzMYe+DUAgAAAAAgF4dHR1qbGxUSUmJUlJSZBhGtEtCFLhcLqWlpUW7jAGZpqn29nbt2rVLSUlJSk5OHtZ+uHwMAAAAAIBe+/fvV15enlJTUwmEELMMw1Bqaqpyc3O1f//+Ye+HUAgAAAAAgF4dHR1KT0+PdhlARJxOpzo6Oob9ekIhAAAAAAB6dXd3KyGBlVZwbEhISFB3d/ewX08oBAAAAACADy4bw7HiSMcqoRAAAAAAAEAcYk4cAAAAAABj0MSJE8P2GT9+vLZt2zbsYzz33HP6wQ9+oFdffVUlJSXD3k+f1atXa82aNfr444+PeF8Ij1AIAAAAAIAxaP369X7Pv/vd72rixIm66aabrG2JiYlHdIxzzz1X69evV35+/hHtB9FBKAQAAAAAwBhUVVXl9zwxMVHZ2dlB2315PB6ZphnxYts5OTnKyck5ojoRPawpBAAAAABAnJo4caJWrVqlxx57TDU1NaqoqNAnn3yizs5O3X///br00ks1ZcoUnXXWWbrhhhv06aef+r3+ueee08SJE7Vr1y5rW01NjZYsWaLNmzfr4osvVlVVla644gr99a9/HVaNbW1tWr58uWbMmKGKigrNmjVLTz31lEzTtPq4XC7de++9Ovfcc1VRUaEzzzxT11xzjV+9Tz/9tC6++GJNmjRJp59+uq644gr913/917BqGiuYKQQAAAAAQBx77rnnVFpaqjvuuEMpKSnKz8+X2+2Wy+XSjTfeqLy8PLW0tOiZZ57RN77xDb300kvKy8sbdJ+1tbX67LPPdPPNNyspKUm/+MUvdMMNN2jbtm3KyMiIuDav16sFCxbo73//uxYvXqwJEyboz3/+s1asWKGmpibdeuutkqQVK1Zo27ZtuuWWW3TcccepublZ/+///T+1trZKkl588UWtXLlSCxcu1NSpU9XZ2amPP/5Yzc3Nw3/jxgBCIQAAAAAABrHtr/X6r/+tj2oNF0wrU83UslHZt2maeuKJJ5ScnOy3/b777rMeezwezZgxQ2eeeaY2b96sa665ZtB9trW1aePGjcrMzJQk5ebm6mtf+5pef/11zZ49O+LaXn/9ddXW1mrFihW64oorJEkzZsxQe3u7nnjiCV1zzTXKycnRe++9p9mzZ2vevHnWay+44ALr8XvvvaeJEyfqu9/9rrXtK1/5SsR1jFVcPgYAAAAAQBw7++yzgwIhSdqyZYvmzZunqVOn6pRTTlFVVZUOHz6sHTt2hN1nVVWVFQhJ/XdC27Nnz5Bqe+edd2Sz2XTppZf6bb/sssvU1dWl9957T5JUWVmp559/XmvXrtXf/vY3eTwev/6VlZX66KOPdO+99+rtt99We3v7kOoYqyKaKbRw4ULt2rVLNptNqampuvvuu3XyySf79Vm9erWeeeYZa8Xx0047TUuXLh35igEAAAAAOM/3Z5UAACAASURBVIpqpo7eLJ1YEOrOYX2XYl1++eX67ne/q+zsbBmGoQULFsjtdofdp28gJPXf5ayzs3NItbW0tCgzMzPoLmm5ublWuyTdddddys3N1R//+EetWrVKWVlZmjNnjm655RalpKRo7ty56uzs1B/+8Ac988wzSkhI0Fe+8hXdeeedKikpGVJNY0lEodDKlSvldDolSVu3btUPf/hDPf/880H95s6dqzvuuGNkKwQAAAAAAEfV5s2bVV5erp/+9KfWtq6uLiuEOVoyMzPV0tIit9vtFwwdOHBAkpSVlSVJSktL02233abbbrtNu3fv1p/+9Cc99NBDcjgc+v73vy/DMHTllVfqyiuvVEtLi9566y399Kc/1S233KINGzYc1XOKJRFdPtYXCEk91wUahjFqBQEAAAAAgOjq6OiQ3W732/bCCy8EXZY12qZNmyav16uXX37Zb/umTZvkcDg0efLkoNeMHz9e3/nOdzRhwgRt3749qD0zM1Nf/epXdfHFF4dsjycRLzT9ox/9SG+99ZZM09Tjjz8ess/mzZv15ptvKi8vTzfddJOmTJkyYoUCAAAAAICj4+yzz9bWrVt1//3367zzzlNdXZ1++9vfDunOYSPhnHPOUXV1tZYuXaqmpiadeOKJev3117VhwwZdf/31ysnJkSR94xvfUE1NjSZMmKDU1FS98847+sc//qG5c+dKku6++26lpaWpqqpK48aN0z//+U+98MILOuuss47q+cSaiEOhvlXHN27cqAceeEDr1q3za7/yyit1ww03yOFw6K233tLChQu1ZcsWZWdnR1xMXV1dxH1jXW1tbbRLAAbFGEWsY4wi1jFGEesYo4h1sTpGExIS5HK5ol3GqPB6veru7g46v66urqBtl1xyierr6/XCCy9o/fr1OuWUU7Rq1SrddtttfvvoWyOovb3d2jbQcSRZt7ofSN96Rb59Vq1apTVr1mjdunVqbm5WcXGxbr31Vn3729+2+k2ePFmbN2/WY489pu7ubpWUlOi2227T1772NblcLp166ql68cUXtXHjRrW1tSkvL08XX3yxbrjhhgHrOVbGgdvtHvTfU3V19YBthmma5lAPOGnSJL3++uuDBj5XXHGF7rzzTk2bNi3s/jo7O1VXV6eKigolJSUNtZyYU1tbO+ibDkQbYxSxjjGKWMcYRaxjjCLWxfIY/eijj4JurIT443K5lJaWFu0yInIkYzbsmkIul8vvlnHbtm1TZmamtZhTn8bGRr+Cdu/ereOPP35YRQEAAAAAAGB0hb18rL29XTfffLPa29tls9mUmZmptWvXyjAMzZ8/X4sXL1ZlZaUefvhhffjhh7LZbHI4HHrggQeUl5d3NM4BAAAAAAAAQxQ2FMrNzdXvf//7kG2+6wqtXLly5KoCAAAAAADAqIrolvQAAAAAAAAYWwiFAAAAAAAA4hChEAAAAAAAQBwiFAIAAAAAAIhDhEIAAAAAAABxiFAIAAAAAAAgDhEKAQAAAAAAxCFCIQAAAAAAxqAbb7xR06ZNk9vtDtne1tamqqoq3XnnnRHvs6amxq//c889p4kTJ2rXrl2Dvm7Xrl2aOHGinnvuuYiP1eepp57SK6+8ErR99erVmjhx4pD3d6QC34NjGaEQAAAAAABj0OWXX66Wlha99tprIdv/9Kc/qb29XXPnzh32Mc4991ytX79e+fn5w95HOL/5zW9ChkLz5s3T+vXrR+248YBQCAAAAACAMejcc89VVlaWNm7cGLJ948aNKi4u1vTp04d9jJycHFVVVSkxMXHY+xiuwsJCVVVVHfXjjiWEQgAAAAAAjEGJiYm65JJL9MYbb6ipqcmvraGhQe+8844uu+wyGYahN998U/Pnz9eMGTM0efJkXXrppXriiSfk8XgGPUaoy8fa29v14x//WNOnT9eUKVN0ww03aO/evUGv/eCDD7R48WKdc845mjRpkmbNmqWHH35YHR0dVp+amhrt3r1bmzZt0sSJEzVx4kTr0q1Ql4+1tbVp+fLlmjFjhioqKjRr1iw99dRTMk3T6vM///M/mjhxol599VUtX75c06dP15e//GUtWbJEhw4divwNDjiXa665RlOmTFFVVZWuvvpqffDBB0F9/u3f/k3Tp0/X5MmTdf755+vHP/6x1b5//37dcccdVu0zZszQ9ddfry+++GJYNUUiYdT2DAAAAAAAomru3Ln63e9+p5deeknf/va3re0vvviiTNO0Lh37/PPPdcYZZ+iqq65SUlKS6urqtHr1ajU1NWnJkiVDOuY999yjl156SYsWLVJlZaXeeuutkPvYs2ePTjrpJF1++eVKS0vT9u3b9eijj+rzzz/XqlWrJElr1qzRggULNHHiRN10002SemYnheL1erVgwQL9/e9/1+LFizVhwgT9+c9/1ooVK9TU1KRbb73Vr/99992n8847Tw899JA+++wz/exnP5PdbtfKlSuHdL7/+Mc/dNVVV+lLX/qSVqxYIcMw9Nhjj+mqq67S73//e5100klyuVy67rrrVFlZqRUrVigtLU27d+/Wu+++a+3n9ttvV0NDg26//XYVFRXpwIED+stf/qL29vYh1TMUhEIAAAAAAAyi9YM/q/X9bVGtwTm5Rs5J5w75dZMmTdKXvvQlbdy40S8UeuGFFzRlyhQdf/zxkqRvfvObVptpmpo6daq6urr0xBNP6NZbb5XNFtmFRjt27NB//ud/6pZbbtGCBQskSTNmzNDhw4f1H//xH359Z82a5XfM0047Tenp6brjjjt0zz33KDs7W6eccooSExOVnZ0d9lKx119/XbW1tVqxYoWuuOIK69jt7e164okndM011/gFSqeffrruvvtuq99nn32mDRs26Kc//WlE59rn0UcfVWJiop566illZGRIks466yzV1NRozZo1WrNmjXbs2KGWlhZ9//vf10knnWS9tq9OSXrvvfd0yy236LLLLrO2XXzxxUOqZai4fAwAAAAAgDFszpw5+uCDD/TZZ59J6rmMaceOHZozZ47VZ9++fbrnnnt03nnnqaKiQqeeeqp+/vOf69ChQ0O6fOmDDz6Q1+sNCjMuueSSoL5tbW362c9+ppkzZ6qyslKnnnqqbr/9dpmmqZ07dw75PN955x3ZbDZdeumlftsvu+wydXV16b333vPb/pWvfMXv+YQJE+R2u3XgwIEhH/fcc8+1AiFJSk9PV01Njd555x1J0nHHHaeMjAwtXbpUL7zwgvbs2RO0n4qKCv3617/W008/rY8//tjvkrfRwkwhAAAAAAAG4Zx07rBm6cSKOXPmaNWqVdq4caNuueUWbdy4UYmJifrqV78qqeeyqxtvvFH79u3TTTfdpBNOOEFJSUnaunWr1q5dq87OzoiPtW/fPknSuHHj/LYHPpekH/zgB3r77be1ePFinXzyyUpJSdEHH3yg5cuXD+mYfVpaWpSZmRm06HVubq7V7isrK8vved/rOjs7lZqaOqTj5uXlBW3Pzc21jul0OvWb3/xGjz76qJYtWyaXy6UTTzxRN910kzVj6uc//7nWrFmjxx9/XPfff7/y8vJ05ZVXauHChRHP1BoqZgoBAAAAADCGFRQU6Mwzz9SmTZvkdru1ZcsW1dTUKDMzU5JUX1+vuro6LVmyRF//+tc1depUVVZWym63D/lYfbemD5xdFPi8s7NTr776qq699lpdffXVmjZtmiorK5WcnDzMs5QyMzPV0tIit9vtt71v5k9gCDRSMjMzQ84uOnDggPUeS9LJJ5+s1atX63//93+1fv16lZaW6nvf+54++eQTST3B2dKlS/XGG2/opZde0hVXXKHVq1cHXXY3kgiFAAAAAAAY4+bOnavdu3fr4Ycf1sGDB60FpiVZd/tyOBzWtq6uLm3atGnIx5k0aZJsNpteeuklv+2bN2/2e+52u+XxeJSQ4H8B0/PPPx+0T4fDEdHMoWnTpsnr9erll1/2275p0yY5HA5Nnjw50tMYktNPP12vv/662trarG1tbW167bXXNG3atKD+CQkJqqqq0s033yyv16tPP/00qM8JJ5ygW2+9VZmZmdq+ffuo1C1x+RgAAAAAAGPeBRdcoPT0dD311FMaN26czj77bKvthBNO0Pjx47Vq1SrZbDYlJCTo6aefHtZxTjjhBF166aX65S9/Ka/Xa9197L//+7/9+jmdTlVVVenJJ59Ufn6+srOz9cc//lGNjY1B+/zSl76kv/71r3rttdeUm5ur7OxslZSUBPU755xzVF1draVLl6qpqUknnniiXn/9dW3YsEHXX3/9gHctO1ILFy7Un//8Z11zzTWaP3++DMPQunXr1N7erkWLFkmSXnvtNa1fv14zZ85USUmJ2tvb9dvf/lZpaWmaMmWKWltbdc0112j27Nk64YQT5HA49Oqrr6qlpUVnnXXWqNQtEQoBAAAAADDmJScn66KLLtIf/vAHXXrppX4zdBITE/XII49o+fLluuOOO5SZmal//dd/VXFxse66664hH2v58uVKTU3VE088oa6uLk2fPl0PPvigvvWtb/n1e+ihh/TjH/9Yy5YtU3Jysi6++GL96Ec/0vXXX+/X79Zbb9Xdd9+t733ve+ro6NDll18e8g5hNptNjz32mB5++GE9/vjjam5u1vjx4/WDH/xAV1999ZDPI1InnXSSfvvb32rVqlW68847ZZqmJk+erH//93+37jRWXl6u5ORkPfroo9q/f7/S0tJUWVmpJ598UoWFhXK73Tr11FO1YcMGNTQ0yDAMHX/88XrwwQc1c+bMUavdMI/GctZhdHZ2qq6uThUVFUpKSop2OUestrZW1dXV0S4DGBBjFLGOMYpYxxhFrGOMItbF8hj96KOPdPLJJ0e7DESZy+VSWlpatMuIyJGMWdYUAgAAAAAAiEOEQgAAAAAAAHGIUAgAAAAAACAOEQoBAAAAAADEIUIhAAAAAACAOEQoBAAAAACAjxi4STcQkSMdq4RCAAAAAAD0cjgcam9vj3YZQETa29vlcDiG/fqESDotXLhQu3btks1mU2pqqu6++26dfPLJfn08Ho9+8pOf6I033pBhGFqwYIHmzZs37MIAAAAAADja8vPztXv3bo0fP14pKSkyDCPaJQFBTNNUe3u7du/erYKCgmHvJ6JQaOXKlXI6nZKkrVu36oc//KGef/55vz6bNm1SfX29XnnlFTU3N2vu3Lk644wzVFJSMuziAAAAAAA4mjIyMiRJDQ0N6urqinI1iBa3263ExMRolzEoh8OhgoICa8wOR0ShUF8gJEltbW0hk9ItW7Zo3rx5stlsysnJ0cyZM/Xyyy/ruuuuG3ZxAAAAAAAcbRkZGUf0QRvHvtraWk2ePDnaZYy6iEIhSfrRj36kt956S6Zp6vHHHw9q37Nnj4qLi63nRUVF2rt378hUCQAAAAAAgBFlmENcqnrjxo3avHmz1q1b57d99uzZuu+++zRp0iRJ0rp169TY2Ki77ror7D47OztVV1c3lDIAAAAAAAAQRnV19YBtEc8U6jN37lzdc889OnjwoLKzs63tRUVFamhosEKhwJlDkaioqFBSUtJQS4o5tbW1g77pQLQxRhHrGKOIdYxRxDrGKGIdYxSxLl7GaNhb0rtcLu3Zs8d6vm3bNmVmZiorK8uv30UXXaQNGzbI6/WqqalJW7du1axZs0a+YgAAAAAAAByxsDOF2tvbdfPNN6u9vV02m02ZmZlau3atDMPQ/PnztXjxYlVWVmrOnDl6//33deGFF0qSFi1apNLS0lE/AQAAAAAAAAxd2FAoNzdXv//970O2+a4rZLfbtWzZspGrDAAAAAAAAKMm7OVjAAAAAAAAGHsIhQAAAAAAAOIQoRAAAAAAAEAcIhQCAAAAAACIQ4RCAAAAAAAAcYhQCAAAAAAAIA4RCgEAAAAAAMQhQiEAAAAAAIA4RCgEAAAAAAAQhwiFAAAAAAAA4hChEAAAAAAAQBwiFAIAAAAAAIhDhEIAAAAAAABxiFAIAAAAAAAgDhEKAQAAAAAAxCFCIQAAAAAAgDhEKAQAAAAAABCHCIUAAAAAAADiEKEQAAAAAABAHCIUAgAAAAAAiEOEQgAAAAAAAHGIUAgAAAAAACAOEQoBAAAAAADEIUIhAAAAAACAOEQoBAAAAAAAEIcIhQAAAAAAAOIQoRAAAAAAAEAcIhQCAAAAAACIQwnhOhw8eFC333676uvrlZiYqPLyci1fvlw5OTl+/VavXq1nnnlG+fn5kqTTTjtNS5cuHZ2qAQAAAAAAcETChkKGYei6667T9OnTJUkrV67Ugw8+qPvvvz+o79y5c3XHHXeMfJUAAAAAAAAYUWEvH8vKyrICIUmqqqpSQ0PDqBYFAAAAAACA0TWkNYW8Xq+effZZ1dTUhGzfvHmzZs+ere985zt69913R6RAAAAAAAAAjDzDNE0z0s7Lli1TY2Oj1qxZI5vNP0/av3+/srKy5HA49NZbb2nJkiXasmWLsrOzw+63s7NTdXV1Q68eAAAAAAAAA6qurh6wLeyaQn1WrlypnTt3au3atUGBkCTl5eVZj8866ywVFRVp+/btmjZtWsSFVlRUKCkpKeL+saq2tnbQNx2INsYoYh1jFLGOMYpYxxhFrGOMItbFyxiN6PKxVatWqa6uTo888ogSExND9mlsbLQef/TRR9q9e7eOP/74kakSAAAAAAAAIyrsTKHt27dr7dq1Ou6443TllVdKkkpKSvTII49o/vz5Wrx4sSorK/Xwww/rww8/lM1mk8Ph0AMPPOA3ewgAAAAAAACxI2wodOKJJ+rjjz8O2bZu3Trr8cqVK0euKgAAAAAAAIyqId19DAAAAAAAAGMDoRAAAAAAAEAcIhQCAAAAAACIQ4RCAAAAAAAAcYhQCAAAAAAAIA4RCgEAAAAAAMQhQiEAAAAAAIA4RCgEAAAAAAAQhwiFAAAAAAAA4hChEAAAAAAAQBwiFAIAAAAAAIhDhEIAAAAAAABxiFAIAAAAAAAgDhEKAQAAAAAAxCFCIQAAAAAAgDhEKAQAAAAAABCHCIUAAAAAAADiEKEQAAAAAABAHCIUAgAAAAAAiEOEQgAAAAAAAHGIUAgAAAAAACAOEQoBAAAAAADEIUIhAAAAAACAOEQoBAAAAAAAEIcIhQAAAAAAAOIQoRAAAAAAAEAcIhQCAAAAAACIQ4RCAAAAAAAAcSghXIeDBw/q9ttvV319vRITE1VeXq7ly5crJyfHr5/H49FPfvITvfHGGzIMQwsWLNC8efNGrXAAAAAAAAAMX9iZQoZh6LrrrtOf/vQnbdq0SaWlpXrwwQeD+m3atEn19fV65ZVXtH79eq1evVq7du0alaIBAAAAAABwZMLOFMrKytL06dOt51VVVXr22WeD+m3ZskXz5s2TzWZTTk6OZs6cqZdfflnXXXfdyFYc4/Y8e6+yPvtAn21NkGx2GXa7DJtdMnoey9b7vPer9dgeYpvNLtlsMuwJ1uutbbYE66ths0sBbYbNJvV9tScM2GbYe+s0+vr5tNnsAfvoP64MmwzDiPbbDQAAAAAAhilsKOTL6/Xq2WefVU1NTVDbnj17VFxcbD0vKirS3r17h1RMXV3dkPrHIkfmv8h+fKoM0yt5PZLp7XlseiWv1+e5R/KaMjweyezu2d7b3vf/ht9rPMH76HssMyrnaho94ZBs9v7Hhq3nsS3wsb33uSHZ7P1tgX0Nu2QYMnv7BO5Phs2nzZDZu1//ffQ8t9oMm0yf1/s/twfv36/msRt81dbWRrsEYFCMUcQ6xihiHWMUsY4xilg3VsZodXX1gG1DCoXuvfdepaam6qqrrjriokKpqKhQUlLSqOz76KlWbW3toG/6SDN7AyjT4+n56vXK9HZL1teeNtPr8dnW89X0eqTB2rxemZ6e0Mr0dMv0eiXfNm/vNr+23noGaOt5bd8x3TK7vQE19nz16+fxSFEJv4yIZnL1z7Dy7Rc4W6tvptVAbb779ZnJZdj8ZoP5zhKTLXxbf+39be998IGmnFbd32aw5jxiy9H+OQoMFWMUsY4xiljHGEWsi5cxGnEotHLlSu3cuVNr166VzRb8AbKoqEgNDQ2aNGmSpOCZQxg9PZd+2WTYHdEuZVSZptcKsPyDpd5tnuBtg7VZjz0DBFF9bebQ9tH31dvtlrwdvaHaYMGcf3B2NGRL+uerPht6Z0cFh16+ly0O0maECM7sdhnGQJdH+gRkQfsPuCzSCHV5pM/+rH2HCOYC28bwzC8AAAAAGKqIQqFVq1aprq5Ojz32mBITE0P2ueiii7RhwwZdeOGFam5u1tatW/W73/1uRItFfDMMm5Rgk6ExHn4NNAvL9ATM1upts2aKhZqt1TfTyr/t8/qdKiku8psp1h9UBYRpvccNaut97O3qHDSsGyiYiwrDd0ZW7ywtI3Aml/96Wj3rbgW29YVOIWZmhQjQQq0nNlib31pjRqjgLExoR/gFAAAAIAJhQ6Ht27dr7dq1Ou6443TllVdKkkpKSvTII49o/vz5Wrx4sSorKzVnzhy9//77uvDCCyVJixYtUmlp6ehWD4xB1gf7UQy//s9Wq6woToU0TbPnksRBZlpFOpPLNH0ucwy5D69PqNUdsI9Ql0X6z+jqC9+83R0B+w0V2gXXGBUBlxyGDJ0CZnKFDp36Lo8MdflkYKgWcFlkUKjmE7T1hWiDtNlcTeo6uLd3XS+jJ+jqu8yxd02xnm1GcB/Dp498+vj2AwAAABA+FDrxxBP18ccfh2xbt26d9dhut2vZsmUjVxmAMavnw3tvwDCG+YVfQWtm+cy+CpppNVhbqJlcYS6BHCxU8/rvo+eyR9+1yULtw+cSyN41x0ZapqTP3xjx3fbyD4qMgLDJL3DSEPv4BFL+QZXv89E6tm/4ZettHvjYMmwyQu1XgX0CngfsN7ieMH1C7re3j/zPw78eW+/DwP36vO8h6xnsexOwX/keO8z77umS2d0Ver8AgDHLNE1JpmSafRsk05TZt81qlySz9+9Bn229fzf5b+/bl7d3P+rf3tunv7+Ct/e+NuQxB2vzqz34mD1PvcHHDDiXoPfAbz8hztWq3eeY8u3vDVmPGXjMAc6nt0PI994MOJfBjpns6pJYUwgAMFx+4dcY/mlrmmbQDCtrFpYZvJh90IL1nu6gts927NBxx5X3/5HR98vb+kPAG/RHwKB9/P5Y8Yb4I8vb/8dC4B8Wpk8fDdbHG6KeQfYbeGyvR2bgH5Vhj+0NOPfQ5+T3h9FAfUIcW1G6u+WxIFvSZ/81QKNPoDT8MC5En4CQrCd/svk/DxvGhe8j+YRvviFZ4GvCBnaD7Heogd2Qjz3IfgcKUw3f2YWh9zuk741vHxka0vdmgGD0aIWO/j93Aj/sSQr48NT/c0nW64J/5vX0H8qHvf59D3JMnxpDfkBU6A97ffWaoY4Z6gPvcD7c+x4z0g+8AwYBATWF+8Ab9KE8gmMGhQTB7/2g74EC2gb5wG6appxtrdr9tz+EfQ/89xOmnlEOCawxhxjU97NYwT/TJevnevDvA/n8jvL5mStDjoTUo34W0TCGP6YAAI4GwzB6LiGzj9yvFHdHqpyVY/+/zMS6AcMvSb7hWnBA5x9IDRhuDRhU+bxGgccOrGcI+/Xrc2RB4K5duzS+uHiQYw8/jOs51kB9vLI+yIZ6f0z1vjbge2Oakrdb3oBjB9c4hO9N4Pd8KOOCD1dHICDIGuDS2kyPR/98zTZgSBDq3xjfkxhnfa8Dg075feD1DxQ14Afe0B+Q+/cZtL33OIZ1nMHqMUJsNyRbz1ebYci0O2RLTOmtI8IP7AH19Ae3/sccqJ4jDQn62gyf98//XAO+T77vQcD30Ag4l3D1DHTMnv+zBbUZAecSbtwM+P0OfG/93vsI3oOQ+x5o/A33mL31jrCxcjv6cAiFAABASH4fKvq2Ra2a2PJpba2y42BK+WgLG+pZ/xV/4LCp/3KLgUO9oIAsMOgaILgKmj0XIgAbuE+IwC5UGDdIaBY02zEwGB0ksNu3r1FZBQXq+/BkBHwQC/5Q6vshLfIPe/37CW6LKCQIc8ywH3hDfTgMe8wQH3iDApFB3oMBjxnph/vQxwz6nozyB95oa6it1UR+jgJRRygEAACAqOi7zLaHndBxBO2srVUuH7gBAGHYol0AAAAAAAAAjj5CIQAAAAAAgDhEKAQAAAAAABCHCIUAAAAAAADiEKEQAAAAAABAHCIUAgAAAAAAiEOEQgAAAAAAAHGIUAgAAAAAACAOJUS7AEkyTVOS5Ha7o1zJyOns7Ix2CcCgGKOIdYxRxDrGKGIdYxSxjjGKWDeWxmhiYqIMwwjabph9iUwUtba26pNPPol2GQAAAAAAAGNORUWFkpKSgrbHRCjk9XrlcrnkcDhCJlcAAAAAAAAYnpieKQQAAAAAAICji4WmAQAAAAAA4hChEAAAAAAAQBwiFAIAAAAAAIhDhEIAAAAAAABxiFAIAAAAAAAgDhEKAQAAAAAAxCFCIQAAAAAAgDhEKAQAAAAAABCHCIUAAAAAAADiEKEQAAAAAABAHCIUAgAAAAAAiEOEQgAAAAAAAHGIUAgAAAAAACAOEQoBAAAAAADEIUIhAAAAAACAOEQoBAAAAAAAEIcIhQAAAAAAAOIQoRAAAAAAAEAcIhQCAAAAAACIQ4RCAAAAAAAAcYhQCAAAAAAAIA4RCgEAAAAAAMQhQiEAAAAAAIA4RCgEAAAAAAAQhwiFAAAAAAAA4hChEAAAAAAAQBwiFAIAAAAAAIhDhEIAAAAAAABxiFAIAAAAAAAgDhEKAQAAAAAAxCFCIQAAcMyrqanR22+/He0yAAAAjimEQgAAAP+fvTuPj6q6/z/+nsnMZF+BkBAgFIWoEAgSSFWgEnBlU1tcqYgFRBBaaFGwIAIixSq0BpCCBXdLQREpLvxUtFqt2lTFfEWkAolAQsCQfZlJZn5/sBAywQAAIABJREFUJJlksgcmZMi8no/yILn33HvP3BxF3j3nc9zk6NGjiouLU3l5+Vndh5ALAACcC4RCAAAAHuJswyQAAIDWIBQCAAAdhtVq1YoVKzRs2DANGzZMK1askNVqlSTl5OTonnvuUWJiooYOHarbb79ddrtdkrRx40YNHz5cgwYN0jXXXKNPPvmkyefs27dPN910ky699FJdfvnlWrlypSRp0qRJkqQhQ4Zo0KBB+uKLL5SRkaE777xTSUlJSkpK0m9/+1vl5+c775WcnKyNGzdq3LhxSkhI0Lx583T8+HHNmDFDgwYN0qZNm9riVQEAAMjU3h0AAABwl6eeekpfffWVdu7cKYPBoJkzZ2r9+vX6zW9+oy1btqhr167OwOerr76SwWDQoUOH9OKLL2r79u3q2rWrjh496gyLGrNixQrdeeeduuGGG1RUVKSDBw9Kkl544QWNGjVKn3/+uUymyv/MSk9P1z333KMhQ4aosLBQs2fPVkpKin7/+98777d7925t3LhR4eHh8vPz05dffqlHHnlEl19+eRu9KQAAAGYKAQCADmTXrl2aNWuWOnXqpIiICM2aNUuvv/66JMlkMunkyZM6fvy4zGazEhMTZTAY5OPjI6vVqu+//142m03du3dXz549m3yOyWRSRkaGcnJyFBgYqISEhEbbxsbG6oorrpDFYlFERISmTJmizz//3KXNL3/5S0VHR8vPz+/sXwIAAEALEQoBAIAOIzs7W926dXN+361bN2VnZ0uSfvWrXyk2NlZ33323Ro0apY0bN0qqDG0efPBBpaSk6PLLL9fcuXN14sSJJp+zYsUKHTlyRNddd51+/vOfa+/evY22/fHHHzV37lwNHz5cl156qebPn6/Tp0+7tImOjj7TjwwAAHDGCIUAAECHERkZqePHjzu/z8zMVGRkpCQpKChICxYs0LvvvqsNGzZoy5YtzqVk48aN08svv6y9e/fKYDDo8ccfb/I5vXr10urVq/XJJ59o2rRpmjNnjoqLi2UwGOq1feKJJ2QwGPT666/rv//9r/74xz/K4XC4tGnoOgAAgLZGKAQAADqMMWPG6KmnnlJOTo5ycnK0bt06jRs3TpK0d+9epaeny+FwKCgoSD4+PjIajTp06JA++eQTWa1WWSwW+fr6ysfHp8nn7Ny5Uzk5OTIajQoJCZEk+fj4KCIiQkajUT/88IOzbVFRkQICAhQSEqITJ07o6aefbvZzdO7c2eUeAAAAbYFQCAAAdBgzZ85U//79NX78eI0fP179+vXTzJkzJVUWfJ4yZYoGDRqkW265RbfddpuSkpJktVr1xBNPKCkpScOGDVNOTo7mzp3b5HM+/PBDjRkzRoMGDdKKFSu0Zs0a+fr6yt/fXzNmzNBtt92mxMREffnll7rvvvv0zTffKDExUdOnT9fVV1/d7OeYPn26nnrqKSUmJuqvf/2rW94NAABAXQZH3fnLAAAAAAAA6PCYKQQAAAAAAOCFTO3dAQAAAE80depUpaam1jt+zz33aMaMGe3QIwAAAPdi+RgAAAAAAIAX8ojlYw6HQ2VlZfW2ZwUAAAAAAEDb8IhQyGq1Ki0tTVartb274hb/93//195dAJrEGIWnY4zC0zFG4ekYo/B0jFF4Om8Zox4RCnU0paWl7d0FoEmMUXg6xig8HWMUno4xCk/HGIWn85YxSigEAAAAAADghQiFAAAAAAAAvBChEAAAAAAAgBciFAIAAAAAAPBChEIAAAAAAABeiFAIAAAAAADACxEKAQAAAAAAeCFTe3cAAAAAgPsUlth0Ms+mH04USJIMBslgMMggSQbJIIMMBlWdq/y6+pjL8SbbV97YoEaucTnX8DWG6gsBAO2GUAgAAAA4D1ltFTqaXagjmfnKyMrXkcx8pWcV6FRuSWWD3Sfat4Mt1HDIVCdIqvq6qZBJtYItQ2Wa1ez9XEKqBq4xGl2vrQ6yXNo3e69a19S5viZQa+CaRp7pck1D/VEjoV6z96vztdTw/eoEibXbN3+vmnd5PDNfB3MONB4+NvBzdf05ut6vsZ+r68+vqfs18PNrZizV/Tk11L7xezXwc2hqnLXgZ1R7XDT1Xl3eXRPnav8s67/XBsZFA+9IBoOMtcYqPA+hEAAAAODBKuwOnfixSOlZ+TqSWaD0rHylZ+br+Kki2e0OSZLJx6geXYPUv3cnxUaHqPB0lnr3/okcDskhSQ6HHJW/SXJUHq/9dd1zkhyOyq9rX+tQ5cnm71XZrwavafR+ta5xub6p+7l+XdWNZu5X9bnqXtPAvWvu53Dep+bapu5V/9q61zjkkMPeyPW13qXLz66RZzb1s2j6fg1cX+8z1z3neo3d0dy9ah+v4+v8Bg6iI/PkELhuWOhntGnw4HPzXtoToRAAAADgARwOh3LyS5WeVaD0zHxn+JNxolBWW4Wkyr+wREUEKjY6WFcM7KbYqBD1ig5RdOdAmXxqyoWmpuZr8KDu7fVRgEZVB2upqam69NJLGw0kGwz3qr5whopNhYXONk2cqxsCNhjQ1Q8YGw78ao7bG71XddBaN6hrPmBsKrhrMvhrJNxrMrxrgyDY3uTPyPVn7DEhsLVA3oBQCAAAADjHikpslaFPVQBUvQSsoNjmbBMe7KvY6BBdd1kv9YoOVs+oEPXsGiw/X/4THuev6uVIRqNBPj7sewTPlZqa2t5dOCf4EwUAAABoI7byCv1wotA56yc9q0BHMvNr6v5I8vc1KTYqWJcPqJn50zMqWKFBvu3YcwCANyAUAgAAAM5Shd2hEzlFLsFPRla+jp2sXffHoO6Rwerfu5N6RgWrV3SIYqNC1CXcnyKsAIB2QSgEAAAAtJDD4dDpgrJ6O35lZBXUq/vTMypYl8dXzv6JjQ5Wty5BLnV/AABob82GQqdPn9b999+vjIwMWSwWxcbGatmyZYqIiGiw/aFDh3TjjTfq9ttv1wMPPOD2DgMAAADnQlGJTRlZBTqSla+MzHwdycpXemaBCoqtzjbhwb6Kjaqs+xMbFazYaOr+AADOH83+aWUwGDR16lQlJSVJklatWqXHH39cjz76aL22FRUVWrJkiUaPHu3+ngIAAABtwFZeoaPZhZWzfqpm/qRn5evk6Ybq/kQ7Z/7ERoVQ9wcAcF5rNhQKCwtzBkKSlJCQoJdffrnBths3btSVV16p4uJiFRcXu6+XAAAAwFmy2x3KyilSemZBrcLPDdf9uaRXJ8VeVjnzpxd1fwAAHVSr5rXa7Xa9/PLLSk5Ornfu22+/1UcffaTnnntO69evd1sHAQAAgNaorvtTHfqkZ1YuAfvhRIHKrBXOdlGdAhQbFaLL4rupV1SIekYHK4a6PwAAL2JwOByOljZeunSpTpw4obVr18porPnD0maz6fbbb9fKlSt14YUXKiUlRcXFxS2uKVRWVqa0tLTW9x4AAABerdRmV3auTdl5NmXn2nQi16bsvHKVlNmdbQL9jIoMNatrmFmRYWZFhprVJdQkXzPhDwCg4xs8eHCj51o8U2jVqlVKT0/Xhg0bXAIhSTp58qQyMjI0ffp0SVJ+fr4cDocKCwu1fPnyFne0f//+8vU9/9dlp6amNvnSgfbGGIWnY4zC0zFGz73quj/pmTU7ftWv++OjnlEhGtE7xGXLd2+s+8MYhadjjMLTecsYbVEotGbNGqWlpWnjxo2yWCz1znfr1k2ffvqp8/vWzhQCAAAApMq6Pydyiutt+X7sZGGTdX9io0IUSd0fAABapdlQ6ODBg9qwYYN69eqlW2+9VZLUvXt3rVu3TtOmTdOcOXMUHx/f5h0FAABAx+FwOJRbUFYz66eq/k9Go3V/op1bvlP3BwAA92g2FOrTp48OHDjQ4LlNmzY1eHz27Nln1ysAAAB0GMWlNmVkFVQFQFWFnzPzVVBsdbYJC/JVbHSwrkmKrdzxKzpEPboGy9+3VfuiAACAVuBPWQAAALiFrdyuo9nVs34KnEvAshuo+3NZfLRio4MVG1W59Css2Pvq/gAA0N4IhQAAANAq1XV/Kmf91ARAx08WqqKq7o+P0aDukUG6qFeErvlp5cyf2OgQdQnzl9FI3R8AADwBoRAAAAAaVF33Jz0rX0cyC5yFn+vW/ekaEaBe0SH6af8o545f3boEyWyi7g8AAJ6MUAgAAADOuj/p1cFP1eyf/CLXuj89o2rq/sRGBatnVAh1fwAAOE/xJzgAAIAXcdb9qbXjV3qma90fP4uPYqNC9NP+NTt+UfcHAICOh1AIAACgA7LbHco+XVxvx68G6/7EVtb9qQ6AIsMDqPsDAIAXIBQCAAA4z50uKHUWfK6e/ZORVaDSWnV/IiMC1Cuqsu5PbFRl4Wfq/gAA4N0IhQAAAM4TxaU2ZZwoqBcA5RXW1P0JDbIoNipEVyXFVm73Hh2snl2DFeBnbseeAwAAT0QoBAAA4GFs5XYdO1lYVfA5v2oJWIGyc4qdbarr/gy9pGbHr9ho6v4AAICWIxQCAABoJ9V1f9Iz83UkK18ZmQU6kpWvY9mudX9iIoN0Uc9wXZ3UU72qwh/q/gAAgLNFKAQAAHAOnC4odYY+TdX9iY0KVlK/KOfMnxjq/gAAgDZCKAQAAOBGNXV/Cpzbvdet+xMSaFGv6Oq6P5U7flH3BwAAnGuEQgAAAGeguu5Peu0t37PyXer++Fp8FBsVrKGXRCk2OkS9okLUMzpY4cF+7dhzAACASoRCAAAATahd96d6x6/9h08o52//qFf3J66q7k/1lu/U/QEAAJ6MUAgAAKBKbkGZc+ZP5c5fBco4ka+Sslp1f8L9FR5o0pWJP1HPqvAnpkugzCafduw5AABA6xEKAQAAr1NSVl611XuBc8v3jKwC5RaWOdtU1/0ZNaSnc8v3nlGVdX9SU1M1ePAl7fgJAAAAzh6hEAAA6LBs5XYdP1moI7Xq/qRn5etEnbo/PbsGa8glXatm/lQWfg4L8pXBwNIvAADQcREKAQCA81513Z+MrIJaAVC+jp0sVHlFZd0fo9GgmC5B6tszXFcN7anYqtk/XSOo+wMAALwToZCbHTqWp29+KJEl9JSCAy0KDjArKMAiXzN1BgAAcIe8wrLK4KdW4eeG6v70jArRkEuinFu+d48Mou4PAABALYRCbpby9y/0v6N5+vuH/3I5bjEZFRRgUUigRUEBZgUHWBTkX/l77fAouOpccEBlO1+zD1PXAQBeqbruT3XwU738q3bdn+CAmro/1Tt+Vdf9AQAAQNOaDYVOnz6t+++/XxkZGbJYLIqNjdWyZcsUERHh0m7dunV644035OPjI5PJpLlz52r48OFt1nFPteLeK/TeR/9Rj9gLVVBsVUGxTYW1fs8vsqqwxKZjJwurvrepvMLe6P3MJmOtwKgmNGooQAqu1cbXQpgEADg/lFfYdSy70GXHryOZDdf9Sby4a9Wyr2D1ig5RWDB1fwAAAM5Us6GQwWDQ1KlTlZSUJElatWqVHn/8cT366KMu7QYMGKC7775b/v7++vbbbzVp0iR99NFH8vPza5uee6gAP7Oiwy0a2KdLi9o7HA6VWSsqQ6MSa2WQVGSrCpSsKiyu+bqg2KbMU0X6rjhXBcVW2cobD5NMPjVhUkhgzawkZ3hUNTsp2N/1mB9hEgCgjdjtDp3MLam35fvR7IJ6dX/69AjTVUN7Ord8p+4PAACA+zUbCoWFhTkDIUlKSEjQyy+/XK9d7VlBcXFxcjgcys3NVVRUlJu62jEZDAb5+Zrk52tSl3D/Vl1bai13hkZ1w6Pq2UnVx07kFOt/R3NVUGyT1VbR6D1NPgbnLKQg/zrL3erMRqo9e8nf10SYBABwctb9qbXjV0aWa92fLuH+io0K0eCLIiu3fKfuDwAAwDllcDgcjpY2ttvtuvvuu5WcnKw777yz0XY7duzQc889px07drTovmVlZUpLS2tpN3CWbOUOlVjtlb/K7M6vi8vqHyupdcxW0fhQMRokf1+j/C21fvm6/h5Q53t/i1G+ZgNhEgCcx8psdp3MK1d2nk3ZuTadyLUpO8+motKa2az+FqMiw0zqGmZWZJhZkaGVv/wsxnbsOQAAgHcYPHhwo+daVWh6+fLlCggI0KRJkxpt89lnn+nPf/6zNm/e3JpbS5L69+8vX1/fVl/naVJTU5t86ecrq61ChSU2FRTVnZFUMyup9oylnNM2FRaXqNTa+Mwko9HgnJXkstytelaSf9WMpEDXekoBviaWEZyFjjpG0XEwRj1PeYVdx04Wuuz4lZ6Vr6wfa+r+WMw+6hkVrMviuyo2OlixUZWzf8I7YN0fxig8HWMUno4xCk/nLWO0xaHQqlWrlJ6erg0bNshobPj/2fviiy80f/58rV+/Xr1793ZbJ+EZLGYfRZh9FBHSujpRtvIKFRbblF8rNHJd3lZzLCe/VOlZ+SostrosMajLaDRU1UmqWcYWFGBWSK0i3EEBlqrvzc5lbgF+ZsIkAGiCw+FQ9umSqmVfNUu/6tf9CdSF3cOcu37FRgera0SgfPh3LAAAwHmjRaHQmjVrlJaWpo0bN8pisTTYZt++fZo7d66efPJJ9evXz62dxPnNbPJReIiPwlsdJtlVWFIZJOUX1QRJlQW5a2YsFRbbdLqgVBknClRQZFVJWXmj9zQapED/2rOOahXerp6V1MCMpQA/M3/RAdDh5BWW1dvxKyOrwOXfo7Xr/sRGVxZ9jukSJIuZuj8AAADnu2ZDoYMHD2rDhg3q1auXbr31VklS9+7dtW7dOk2bNk1z5sxRfHy8li5dqtLSUj300EPOax977DHFxcW1Xe/RoZlNRoUH+yk8uHVhUnmFvZkC3DWzk/IKy3Q0u1CFxVYVlTYeJhkMUqCf2bmMLSjAouDqcCnQtQh37RlLgf6ESQDaX2lZuTJOVC75OpKVr4zMAh3JylduQZmzTXCAWbHRIUpO7FEZ/kSFqGdUsAL9ze3YcwAAALSlZkOhPn366MCBAw2e27Rpk/PrV155xX29As6CyceosGBfhQW3rj5VeYVdRSUNh0nOY1Wzk/KLrDp+slAFxTYVldiavG9g1TK32qFRQ8vdXM75m+XjQwFWAK1TXfenOvSprvtzIqdY1dtKVNf9qd7xq3rL945Y9wcAAABNa1WhaaAjM/kYFRrkq9Cg1oVJFRV2FZbYKotwFzew3M2lfpJVmaeKVFBsVVGpTU3t/RfoZ6opsu3fwHK3WjOWggLMlcvdCJMAr+BwOHTydElN8OOs+1Oo8orKXb+MRoO6dQ7UBd3DlJzYU72igxUbHULdHwAAADgRCgFnyedMwyS7Q0UlrkvaCoutdQpy14RJWTnFKiy2qrCk6TApoDpMamB2UvWxE8dL5B/xo8s5E2ES4JGq6/5UBz/Vu3/VrvvTOcxfvaJr6v7ERoWoeyR1fwAAANA0QiGgnfgYDQoJrCxo3Rp2u0PFpa67uTlnJBVZVVBnCVx2TnHVMjer7LXCpJf/+ZHLff19TQ3u2la7XlLNbm/V5ywymwiTAHeoXfenesv3unV/gvxd6/7ERlVu+07dHwAAAJwJQiHgPGM0GhRUFci0ht3uUHFZuQqLrfosdZ+697ygKjyqCpKKXGcsncwtcZ63NzEzyd/Xx2UZW+XStqoAqVYx7rozlswmZjDAO5VX2HX8ZKFz5k/1jl9ZOUU1dX9MRmfdn8rt3isDoIgQP+r+AAAAwG0IhQAvYTQaFORvVpC/WTGdLLr0osgWXWe3O1RSVt7ALm4NL3dLz8x3trM3kSb5WXwaLLLd0Iyk2udZDoPzRXXdn7pbvrvU/TFI3boEqXdMqEYm9lBsVLB6RYeoayfq/gAAAKDtEQoBaJLRaFCgv1mB/mZFdQps8XUOR3WYVLXEragqPCqx1inIXXn+hxMFlW2LrKpoIkyymH0U4gyNXItsVy9pC647YynAIl/CJLShvMIyZ+jTaN2fUD/FVtX9qd7xi7o/AAAAaE+EQgDahMFgUICfWQF+ZnWNCGjxddVhUu3ZR/nFVpdd3GoX4D6aXegs1l1e0USYZDLWW8YW5F8VKDUxO8nX7MNyHThV1/3JyMrXkVqFn083UPdn5ODuzi3fY6NDFETdHwAAAHgYQiEAHqV2mBTZyjCp1FrhEhrVDZRq7/R27GRlmJRfZHMu5WmI2WR0Kazd0NeuM5Yqj/taCJPOZxUVdh0/VVQ588c5+6fhuj+D4iLVq2rHr9ho6v4AAADg/EEoBKBDMBgM8vc1yd/XpMjwll/ncDhUZqtQQZFNhSU1oVF14W1nwFRSudwt81SRvivOVUGxVbbyxsMkk4/Rtci2v9k1PKoOleoU6PYjTDqnHA6HTuaWuOz4lZ6Vrx9OuNb9ie5cVfdncPfKos/RIYqi7g8AAADOc4RCALyawWCQn8UkP4tJXcL9W3VtZZhkrbOkzebcta32crcTOcX639FcFRTbZLVVNHpPk4+h0SLbQQFmhQTUXe5W+bW/r4kwqRn5RVZn6FNd+Dk9K1/Fpa51f3pGh2hQ30jnjl89ugZT9wcAAAAdEqEQAJwhX7OPfMP81Tms9WFSYQMBUuVytspZSdVhUvbpYn1/1KqCEpvKrI2HST5GQ71d3Gq+rpqJ5F8nYAq0dMgwqdRarh9OFDhn/1QGQPnKya+p+xPob1av6BBdeWnVzJ+oygAoKMDSjj0HAAAAzi1CIQA4x3zNPvIN9Ven0NaFSVZbhTMwqpyhZKsJlEqsLru5ncot0eHMPBUUWVXaRJhkNBoqZx35u+7a5gyU/CuP1a2nFOjX/mGSS92fWjt+Zf3oWvenR1SwEvpGKrZqxy/q/gAAAACVCIUA4DxhMfsowuyjiBC/Vl1nK69odnlbdUHuH/NKlZ6Zr4Jiq0rKmg6TKuskuS5jC66zvC2kzoylAD+zjK2sw1Nd96fulu8N1f35SbcQjby0u3pGVwZA1P0BAAAAGkcoBAAdnNnko/AQH4W3Okyyq7CkVr2koppZSQW1jhUW23S6oFQZJwpUWGx1qdFTl9EgBfrXDpCqAqPAyllJ1YHS/u8K9e/DXzlrANW+Z6dQP8U66/4EKzYqRN27BsuXuj8AAABAqxAKAQAaZDYZFR7sp/Dg1oVJ5RX2ml3bim0qKKm73K1mdlJeYZmOZheqsNiqojphUqB/kXpFh+hnl3av2fKduj8AAACA2xAKAQDcyuRjVFiwr8KCfVt1XUWFXYUlNuUXWXXwu280cthQ6v4AAAAAbYhQCADgEXx8jAoN8lVokK+yj7Z/IWsAAACgozO2dwcAAAAAAABw7hEKAQAAAAAAeKFml4+dPn1a999/vzIyMmSxWBQbG6tly5YpIiLCpV1FRYUeeeQRffjhhzIYDJo+fbomTpzYZh0HAAAAAADAmWt2ppDBYNDUqVP19ttva9euXerRo4cef/zxeu127dqljIwM7dmzR1u3blVKSoqOHj3aJp0GAAAAAADA2Wk2FAoLC1NSUpLz+4SEBB0/frxeuzfeeEMTJ06U0WhURESERo8erbfeesu9vQUAAAAAAIBbtKqmkN1u18svv6zk5OR65zIzM9WtWzfn99HR0crKyjr7HgIAAAAAAMDtWrUl/fLlyxUQEKBJkya1SWfS0tLa5L7tITU1tb27ADSJMQpPxxiFp2OMwtMxRuHpGKPwdB1ljA4ePLjRcy0OhVatWqX09HRt2LBBRmP9CUbR0dE6fvy4BgwYIKn+zKGW6N+/v3x9fVt1jSdKTU1t8qUD7Y0xCk/HGIWnY4zC0zFG4ekYo/B03jJGW7R8bM2aNUpLS9O6detksVgabHPttddq27ZtstvtysnJ0TvvvKNrrrnGrZ0FAAAAAACAezQbCh08eFAbNmxQdna2br31Vk2YMEGzZs2SJE2bNk1ff/21JGnChAnq3r27rr76at18882aNWuWevTo0ba9BwAAAAAAwBlpdvlYnz59dODAgQbPbdq0yfm1j4+Pli5d6r6eAQAAAAAAoM20avcxAAAAAAAAdAyEQgAAAAAAAF6IUAgAAAAAAMALEQoBAAAAAAB4IUIhAAAAAAAAL0QoBAAAAAAA4IUIhQAAAAAAALwQoRAAAAAAAIAXIhQCAAAAAADwQoRCAAAAAAAAXsjU3h0AAAAAAHRc+fn5ys7Ols1mcx4zmUzav39/O/YKaNr5MEbNZrMiIyMVEhJyxvcgFAIAAAAAtIn8/HydOHFCMTEx8vf3l8FgkCQVFRUpMDCwnXsHNM7Tx6jD4VBJSYmOHTsmSWccDLF8DAAAAADQJrKzsxUTE6OAgABnIATg7BkMBgUEBCgmJkbZ2dlnfB9CIQAAAABAm7DZbPL392/vbgAdlr+/v8vSzNYiFAIAAAAAtBlmCAFt52z/+SIUAgAAAAAA8EKEQgAAAAAAAF6IUAgAAAAAgGbExcU1+ys5OfmsnvHqq68qLi5OR48edUufU1JSFBcX55Z7nWuffvqpUlJSZLfb3X7vd955R1u2bHH7fc9HbEkPAAAAAEAztm7d6vL9fffdp7i4OM2ePdt5zGKxnNUzrrzySm3dulWRkZFndZ+O4LPPPtPatWt17733ymh073yWd955Rx9//LGmTJni1vuejwiFAAAAAABoRkJCgsv3FotF4eHh9Y7XVlFRIYfDIZOpZX/1joiIUERExFn1E42zWq1nHdx1NM2OzFWrVuntt9/WsWPHtGvXLvXt27demx9//FELFy5UZmambDabfvrTn2rRokUtHvgAAAAAAJzv4uLiNGPGDAUGBupvf/ubMjMz9corr+iCCy7QE088oY8//ljHjh1TQECA4uPjNX/+fF1wwQXO61999VUtXLhQ7777rrp37y5JSk5O1qWXXqqRI0dq7dq1yszMVO/evfXggw8qMTGx1X0sLCzU6tWrtWfPHuXm5iomJka33XabJk+e7NzJqqioSKtXr9a7776rU6dOKSQkRH379tXixYud/X322Wf1t7/9TceOHZOvr68Z1b9qAAAgAElEQVR69Oihe++9V1dddVWL+rFv3z6tWbNG33zzjUpLS9W5c2cNHz5cDz/8sFJSUrR27VpJUr9+/ZzXHDhwQJL05JNPau/evcrIyJDJZFJcXJzmzZvnEtB9+umnuvPOO5WSkqJ//vOfeuedd1ReXq7Ro0drx44dzp+XJMXExOi9995r9bvsCJpNbUaNGqU777xTd9xxR6NtNmzYoAsuuEAbN26UzWbT7bffrj179uj66693a2cBAAAAAPBkr776qnr06KEHHnhA/v7+ioyMlNVqVVFRke6991516dJFeXl5eumll3TLLbfozTffVJcuXZq8Z2pqqg4fPqxf//rX8vX11Z///GfNmDFD7733nkJCQlrcN7vdrunTp+ubb77RnDlz1LdvX73//vtauXKlcnJyNG/ePEnSypUr9d5772nu3Lnq1auXcnNz9d///lcFBQWSpNdff12rVq3SzJkzlZiYqLKyMh04cEC5ubkt6kdRUZGmTp2q+Ph4rVy5UoGBgTp27Ji++OILSdLEiROVlZWl7du366WXXpKPj4/L9SdOnNDkyZMVFRWlkpISvf7665o0aZK2b9+uiy66yKXt8uXLNWLECD322GOyWq3q27evcnJy9PXXX+upp56SdPbL/s5nzYZCLUkeDQaDioqKZLfbZbVaZbPZ1LVrV7d0EAAAAADQcbz3nwy99cnhen/RP5euGtpTyYk92+TeDodDmzdvlp+fn8vxFStWOL+uqKjQsGHDdPnll2v37t266667mrxnYWGhXnvtNYWGhkqSOnfurF/84hf64IMPNG7cuBb37YMPPlBqaqpWrlypm266SZI0bNgwlZSUaPPmzbrrrrsUERGhL7/8UuPGjdPEiROd19aeAfTll18qLi5O9913n/PYz372sxb349ChQ8rLy9P8+fNdQpzqPkVFRSkqKkqSNHDgwHqrkOq+y+HDh2vMmDHavn27Fi1a5NJ2wIABLu2lymV6ZrO5yaV/3sIt1Zpmzpypw4cPa9iwYc5fgwcPdsetAQAAAAA4bwwfPrxeICRJb7zxhiZOnKjExERdcsklSkhIUHFxsQ4dOtTsPRMSEpyBkFSz7CkzM7NVffv8889lNBo1duxYl+Pjx4+XzWbTl19+KUmKj4/Xjh07tGHDBn399deqqKhwaR8fH6/9+/dr+fLl+vjjj1VSUtKqfvTq1UshISFasmSJdu7c2erP8fHHH+uXv/ylkpKSdMkll6hfv346cuSIDh8+XK9tS5ezeSu3FP156623FBcXp2effVZFRUWaNm2a3nrrLV177bWtuk9aWpo7uuMRUlNT27sLQJMYo/B0jFF4OsYoPB1jFJ7AZDKpqKjI5VjSxZ2UdHGndupRjbr9ai273a7y8vJ69wkLC6t37IMPPtDcuXM1btw4TZ06VWFhYTIajZo9e7aKi4ud7cvKyiRJJSUlzmN2u12BgYEN9regoKDJz2G1WiXVfNYff/xRISEhstlsstlsznaBgYGSpOzsbBUVFWnevHkKDQ3Vtm3btGbNGoWGhmrMmDGaNWuW/P39ddVVVyk/P187d+7USy+9JJPJpCuuuEK//e1v1a1bt2bfndFo1F/+8hdt2rRJS5cuVVFRkS644ALNmDFDo0aNqtf32jOF9u/fr2nTpumyyy7T4sWL1blzZ/n4+GjZsmUu7620tFSSFBwcXO8dlZeXy+FwNDsGznaMnCtWq7XJf+c3NWnHLaHQCy+8oEcffVRGo1HBwcFKTk7Wp59+2upQqH///vL19XVHl9pVamoqM6Xg0Rij8HSMUXg6xig8HWMUnmL//v3OwKG2oqKiBo+fT4xGo0wmU73PYbFY6h179913FRsbq8cff9x5zGazKT8/3+Ue1X8f9vf3dx5r7DmNPavueakm9OnUqZPy8/NlNptd6uhUhx9du3ZVYGCgAgMDtWDBAi1YsEDHjh3T22+/rSeeeEIBAQGaP3++JGny5MmaPHmy8vLy9K9//Ut/+MMf9OCDD2rbtm0teHvSpZdeqqeeekrl5eVKS0vTX/7yFz3wwAPauXOn+vbt69L32qHQhx9+KJPJpKeeekpms9l5vLCwUGFhYc7PWj1bq/a7rGYymWQwGJp8d+fTGLVYLBo4cOAZXeuW5WPdu3fXP//5T0mVCdUnn3yiPn36uOPWAAAAAACc10pLS+vVUNq5c2e9ZVltbejQobLb7Xrrrbdcju/atUtms7nBYCEmJkZ33323+vbtq4MHD9Y7Hxoaquuvv17XXXddg+ebYzKZlJCQoF//+tey2+36/vvvJdUEWtUzfqqVlJTIaDQ6d0qTpE8++UTHjx9v8TMtFotzVpa3a3am0COPPKI9e/bo1KlTmjJlisLCwrR7925NmzZNc+bMUXx8vB588EEtWbJE48aNU0VFhZKSknTzzTefi/4DAAAAAODRhg8frnfeeUePPvqoRo4cqbS0ND3//POt2jnMHUaMGKHBgwdryZIlysnJUZ8+ffTBBx9o27ZtuueeexQRESFJuuWWW5ScnKy+ffsqICBAn3/+ub799lvdcMMNkqTFixcrMDBQCQkJ6tSpk44cOaKdO3fqiiuuaFE/9u7dq61bt2r06NHq3r27SkpK9PzzzyswMFCDBg2SJF1wwQWSpC1btmjEiBEyGo2Kj4/X8OHD9eyzz2rBggX6+c9/rsOHD2v9+vWt2uzqggsuUG5url566SXniqXqOk3eptlQaNGiRfWqd0vSpk2bnF/37NlTW7ZscW/PAAAAAADoAG6++WZlZmbqlVde0datWxUfH68NGza47N51LhiNRm3cuFGrV6/W008/rdzcXMXExGjhwoWaPHmys11iYqLefPNNbdy4URUVFerRo4cWLlyoO++8U1Ll0q9XX31VO3fuVEFBgSIjIzV+/HjNmTOnRf2IjY2Vn5+f1q9fr5MnTyowMFDx8fHasmWLc9exkSNH6vbbb9dLL72kdevWyeFw6MCBAxo+fLgWLVqkLVu2aM+ePerTp48ee+wx5/byLTFx4kR99dVXWrNmjfLz8xUTE6P33nuvFW+y4zA4HA5He3eirKxMaWlp1BQCzhHGKDwdYxSejjEKT8cYhafYv3+/Lr744nrHz6d6LfBO59MYbeyfs5ZwS00hAAAAAAAAnF/csvsYAAAAAACAw+FotoB27d3E0L74SQAAAAAAALfYsWOHFi5c2GSbAwcOnKPeoDmEQgAAAAAAwC1Gjhyp7du3t3c30EKEQgAAAAAAwC3Cw8MVHh7e3t1AC1FoGgAAAAAAwAsRCgEAAAAAAHghQiEAAAAAAAAvRCgEAAAAAADghQiFAAAAAAAAvBChEAAAAAAAgBciFAIAAAAAoBn33nuvhg4dKqvV2uD5wsJCJSQkaMGCBS2+Z3Jyskv7V199VXFxcTp69GiT1x09elRxcXF69dVXW/ysas8884z27NlT73hKSori4uJafb+zVfcdnE8ae5fukJKSok8++aRN7l0boRAAAAAAAM248cYblZeXp7179zZ4/u2331ZJSYluuOGGM37GlVdeqa1btyoyMvKM79Gc5557rsEgY+LEidq6dWubPbcjauxdusPatWv173//u03uXRuhEAAAAAAAzbjyyisVFham1157rcHzr732mrp166akpKQzfkZERIQSEhJksVjO+B5nKioqSgkJCef8uXDV2Ey0tkIoBAAAAABAMywWi8aMGaMPP/xQOTk5LueOHz+uzz//XOPHj5fBYNBHH32kadOmadiwYRo4cKDGjh2rzZs3q6KioslnNLR8rKSkRA8//LCSkpI0aNAgzZgxQ1lZWfWu3bdvn+bMmaMRI0ZowIABuuaaa7R69WqVlpY62yQnJ+vYsWPatWuX4uLiFBcX51y61dDyscLCQi1btkzDhg1T//79dc011+iZZ56Rw+Fwtvn0008VFxend999V8uWLVNSUpJ++tOf6ne/+53y8/Nb/oLrfJa77rpLgwYNUkJCgiZPnqx9+/bVazNlyhQlJSVp4MCBGjVqlB5++GHn+ZMnT+qBBx5w9n3YsGG655579OOPP7a4H88++6yuu+46DRgwQEOGDNFNN92k//f//p+kpt9lenq65s+fr+TkZA0YMECjRo3SkiVLlJeX53L/BQsWaMSIEfriiy906623asCAAXrsscecP4cNGzY4752SknImr7JZpja5KwAAAAAAHcwNN9ygF198UW+++abuuOMO5/HXX39dDofDuXTshx9+0GWXXaZJkybJ19dXaWlpSklJUU5Ojn73u9+16pkPPfSQ3nzzTc2aNUvx8fH617/+1eA9MjMzddFFF+nGG29UYGCgDh48qPXr1+uHH37QmjVrJFUuSZo+fbri4uI0e/ZsSZWzkxpit9s1ffp0ffPNN5ozZ4769u2r999/XytXrlROTo7mzZvn0n7FihUaOXKknnjiCR0+fFh//OMf5ePjo1WrVrXq83777beaNGmSLrzwQq1cuVIGg0EbN27UpEmT9Pe//10XXXSRioqKNHXqVMXHx2vlypUKDAzUsWPH9MUXXzjvc//99+v48eO6//77FR0drVOnTumTTz5RSUlJi/rxxhtvaNWqVZo5c6YSExNVVlamAwcOKDc3t9l3mZ2draioKD344IMKDQ3VDz/8oL/85S+aPn16vSV6BQUFmjdvnu6++27NnTtXfn5+Gjt2rG655RbddNNNuuWWWyRVzuRqC4RCAAAAAIBzpmDf+8r97/9Tno9Pu/UheGCyggdc2errBgwYoAsvvFCvvfaaSyi0c+dODRo0SD/5yU8kSbfddpvznMPhUGJiomw2mzZv3qx58+bJaGzZop1Dhw7pH//4h+bOnavp06dLkoYNG6bi4mL97W9/c2l7zTXXuDzz0ksvVVBQkB544AE99NBDCg8P1yWXXCKLxaLw8PBml4p98MEHSk1N1cqVK3XTTTc5n11SUqLNmzfrrrvucgmUhgwZosWLFzvbHT58WNu2bdMf/vAHGQyGFn1eSVq/fr0sFoueeeYZhYSESJKuuOIKJScna+3atVq7dq0OHTqkvLw8zZ8/XxdddJHz2up+StKXX36puXPnavz48c5j1113XYv7sW/fPsXFxem+++5zHvvZz37m/LqpdzlkyBANGTLE+f2gQYPUs2dP3XHHHfrmm290ySWXOM8VFxfrj3/8o0aPHl2vD5GRkW2+pI/lYwAAAAAAtNCECRO0b98+HT58WFJleHDo0CFNmDDB2SY7O1sPPfSQRo4cqf79+6tfv37605/+pPz8/FYtX9q3b5/sdnu9MGPMmDH12hYWFjrDhfj4ePXr10/333+/HA6H0tPTW/05P//8cxmNRo0dO9bl+Pjx42Wz2fTll1+6HK8dmEhS3759ZbVaderUqVY/98orr3QGQpIUFBSk5ORkff7555KkXr16KSQkREuWLNHOnTuVmZlZ7z79+/fXX//6Vz377LM6cOCAy5K3lujXr5/279+v5cuX6+OPP27xDCOpsi7Qhg0bdO2112rAgAHq16+fM0SsHjfVTCaTRo4c2aq+uRMzhQAAAAAA50zwgCtlvGCIAgMD27srZ2TChAlas2aNXnvtNc2dO1evvfaaLBaLrr/+ekmVy67uvfdeZWdna/bs2erdu7d8fX31zjvvaMOGDSorK2vxs7KzsyVJnTp1cjle93tJWrhwoT7++GPNmTNHF198sfz9/bVv3z4tW7asVc+slpeXp9DQ0HpFrzt37uw8X1tYWJjL99XXtfbZeXl56tKlS73jnTt3dj4zODhYzz33nNavX6+lS5eqqKhIffr00ezZs50zpv70pz9p7dq1evrpp/Xoo4+qS5cuuvXWWzVz5swWzdSqDsO2b9+ul156SSaTST/72c+0YMECde/evclrV69erRdeeEEzZ87UoEGDFBgYqBMnTui+++6r9z4iIiLk046z5pp9E6tWrVJycrLi4uL03XffNdrujTfe0Lhx4zR27FiNGzeu1WkgAAAAAACermvXrrr88su1a9cuWa1WvfHGG0pOTlZoaKgkKSMjQ2lpafrd736nm2++WYmJiYqPjz+jv/hXb01fd3ZR3e/Lysr07rvv6le/+pUmT56soUOHKj4+Xn5+fmf4KaXQ0FDl5eXV2w2r+u/6dUMgdwkNDW0wTzh16pTzHUvSxRdfrJSUFH322WfaunWrevTood/85jfO3KJTp05asmSJPvzwQ7355pu66aablJKSUm/ZXWMMBoNuvfVWbd++Xf/+97+1atUq7du3T3Pnzm322t27d2vChAmaOXOmLrvsMg0YMEDBwcGNPqc9NRsKjRo1Si+++KJiYmIabfP1119r7dq12rx5s/7xj3/opZdeavQDAwAAAABwPrvhhht07NgxrV69WqdPn3YWmJbk3O3LbDY7j9lsNu3atavVzxkwYICMRqPefPNNl+O7d+92+d5qtaqiokImk+tioB07dtS7p9lsbtHsnaFDh8put+utt95yOb5r1y6ZzWYNHDiwpR+jVYYMGaIPPvhAhYWFzmOFhYXau3evhg4dWq+9yWRSQkKCfv3rX8tut+v777+v16Z3796aN2+eQkNDdfDgwVb3KTQ0VNdff72uu+46l+sbe5elpaX1fhavvvpqq57Z0p/T2Wp2+VhiYmKzN3nmmWd09913O6d4EQgBAAAAADqqq666SkFBQXrmmWfUqVMnDR8+3Hmud+/eiomJ0Zo1a2Q0GmUymfTss8+e0XN69+6tsWPH6sknn5TdbnfuPvbPf/7TpV1wcLASEhK0ZcsWRUZGKjw8XK+88opOnDhR754XXnih/vOf/2jv3r3q3LmzwsPDG1wONWLECA0ePFhLlixRTk6O+vTpow8++EDbtm3TPffc0+iuZWdr5syZev/993XXXXdp2rRpMhgM2rRpk0pKSjRr1ixJ0t69e7V161aNHj1a3bt3V0lJiZ5//nkFBgZq0KBBKigo0F133aVx48apd+/eMpvNevfdd5WXl6crrriiRf1Yvny5wsLClJCQoE6dOunIkSPauXOny/WNvcvhw4frtddeU9++fRUbG6s9e/a47IzWEhdeeKHef/99DR8+XCEhIYqMjFTXrl1bdY+WcEtNoe+//17du3fXHXfcoeLiYl111VW69957230aFAAAAAAA7ubn56drr71W27dv19ixY11mhVgsFq1bt07Lli3TAw88oNDQUP385z9Xt27dtGjRolY/a9myZQoICNDmzZtls9mUlJSkxx9/XLfffrtLuyeeeEIPP/ywli5dKj8/P1133XX6/e9/r3vuucel3bx587R48WL95je/UWlpqW688Ub94Q9/qPdco9GojRs3avXq1Xr66aeVm5urmJgYLVy4UJMnT27152ipiy66SM8//7zWrFmjBQsWyOFwaODAgXrhhRecO43FxsbKz89P69ev18mTJxUYGKj4+Hht2bJFUVFRslqt6tevn7Zt26bjx4/LYDDoJz/5iR5//PEGd/lqSEJCgnbv3q2dO3eqoKBAkZGRGj9+vObMmeNs09i7XLRokRwOh/70pz9JqgzYnnjiCU2cOLHF72Hx4sVasWKFZsyYIavVqvvuu0+zZ89uxZtsGYOjhSW4k5OTtWHDBvXt27feuXHjxikmJkZPPvmkrFarpk6dqltvvdVlCl1TysrKlJaW1rqeAwAAAAA8mslk0oUXXtje3QA6tP/9738qLy9v9PzgwYMbPeeWmULdunXTtddeK4vFIovFolGjRmnfvn0tDoWq9e/fX76+vu7oUrtKTU1t8qUD7Y0xCk/HGIWnY4zC0zFG4Sn279/f4C5jRUVF5+3uY/AO59MYtVgsZ1zjqfl92Fpg7Nix+uijj+RwOGSz2fTvf//bOa0LAAAAAADAU1RUVKi8vLzRX3a7vb27eM40O1PokUce0Z49e3Tq1ClNmTJFYWFh2r17t6ZNm6Y5c+YoPj5eY8aMUVpamq6//noZjUYNGzZMv/jFL85F/wEAAAAAAFrsrrvu0meffdbo+RtvvFGLFy8+hz1qP82GQosWLWqwGNamTZucXxuNRi1cuFALFy50b+8AAAAAAADcaOnSpSoqKmr0fHh4+DnsTftyS00hAAAAAACA80Hv3r2bbdNUaNSRuKWmEAAAAAAAAM4vhEIAAAAAgDbjcDjauwtAh3W2/3wRCgEAAAAA2oTZbFZJSUl7dwPosEpKSmQ2m8/4ekIhAAAAAECbiIyM1LFjx1RcXMyMIcCNHA6HiouLdezYMUVGRp7xfSg0DQAAAABoEyEhIZKk48ePy2azOY9brVZZLJb26hbQrPNhjJrNZnXt2tX5z9mZIBQCAAAAALSZkJCQen9pTU1N1cCBA9upR0DzvGWMsnwMAAAAAADACxEKAQAAAAAAeCFCIQAAAAAAAC9EKAQAAAAAAOCFCIUAAAAAAAC8EKEQAAAAAACAFyIUAgAAAAAA8EKEQgAAAAAAAF6IUAgAAAAAAMALEQoBAAAAAAB4IUIhAAAAAAAAL0QoBAAAAAAA4IUIhQAAAAAAALxQs6HQqlWrlJycrLi4OH333XdNtj106JAGDhyoVatWua2DAAAAAAAAcL9mQ6FRo0bpxRdfVExMTJPtKioqtGTJEo0ePdptnQMAAAAAAEDbMDXXIDExsUU32rhxo6688koVFxeruLj4rDsGAAAAAACAttNsKNQS3377rT766CM999xzWr9+/RnfJy0tzR3d8Qipqant3QWgSYxReDrGKDwdYxSejjEKT8cYhafrKGN08ODBjZ4761DIZrNp8eLFWrlypXx8fM7qXv3795evr+/ZdqndpaamNvnSgfbGGIWnY4zC0zFG4ekYo/B0jFF4Om8Zo2cdCp08eVIZGRmaPn26JCk/P18Oh0OFhYVavnz5WXcQAAAAAAAA7nfWoVC3bt306aefOr9PSUlRcXGxHnjggbO9NQAAAAAAANpIs7uPPfLIIxoxYoSysrI0ZcoUjRkzRpI0bdo0ff31123eQQAAAAAAALhfszOFFi1apEWLFtU7vmnTpgbbz549++x7BQAAAAAAgDbV7EwhAAAAAAAAdDyEQgAAAAAAAF6IUAgAAAAAAMALEQoBAAAAAAB4IUIhAAAAAAAAL0QoBAAAAAAA4IUIhQAAAAAAALwQoRAAAAAAAIAXIhQCAAAAAADwQoRCAAAAAAAAXohQCAAAAAAAwAsRCgEAAAAAAHghQiEAAAAAAAAvRCgEAAAAAADghQiFAAAAAAAAvBChEAAAAAAAgBciFAIAAAAAAPBChEIAAAAAAABeiFAIAAAAAADACxEKAQAAAAAAeKFmQ6FVq1YpOTlZcXFx+u677xpss27dOo0ZM0bjx4/XTTfdpA8//NDtHQUAAAAAAID7mJprMGrUKN1555264447Gm0zYMAA3X333fL399e3336rSZMm6aOPPpKfn59bOwsAAAAAAAD3aDYUSkxMbPYmw4cPd34dFxcnh8Oh3NxcRUVFnV3vAAAAAAAA0CYMDofD0ZKGycnJ2rBhg/r27dtkux07dui5557Tjh07WtyJsrIypaWltbg9AAAAAAAAmjd48OBGzzU7U6g1PvvsM/35z3/W5s2bz+j6/v37y9fX151dahepqalNvnSgvTFG4ekYo/B0jFF4OsYoPB1jFJ7OW8ao20KhL774QvPnz9f69evVu3dvd90WAAAAAAAAbcAtW9Lv27dPc+fO1ZNPPql+/fq545YAAAAAAABoQ82GQo888ohGjBihrKwsTZkyRWPGjJEkTZs2TV9//bUkaenSpSotLdVDDz2kCRMmaMKECTpw4EDb9hwAAAAAAABnrNnlY4sWLdKiRYvqHd+0aZPz61deecW9vQIAAAAAAECbcsvyMQAAAAAAAJxfCIUAAAAAAAC8EKEQAAAAAACAFyIUAgAAAAAA8EKEQgAAAAAAAF6IUAgAAAAAAMALEQoBAAAAAAB4IUIhAAAAAAAAL0QoBAAAAAAA4IUIhQAAAAAAALwQoRAAAAAAAIAXIhQCAAAAAADwQoRCAAAAAAAAXohQCAAAAAAAwAsRCgEAAAAAAHghQiEAAAAAAAAvRCgEAAAAAADghQiFAAAAAAAAvBChEAAAAAAAgBdqNhRatWqVkpOTFRcXp++++67BNhUVFVq6dKlGjx6tq666Stu2bXN7RwEAAAAAAOA+zYZCo0aN0osvvqiYmJhG2+zatUsZGRnas2ePtm7dqpSUFB09etStHQUAAAAAAID7NBsKJSYmKjo6usk2b7zxhiZOnCij0aiIiAiNHj1ab731lts6CQAAAAAAAPdyS02hzMxMdevWzfl9dHS0srKy3HFrAAAAAAAAtAFTe3egtrS0tPbugtukpqa2dxeAJjFG4ekYo/B0jFF4OsYoPB1jFJ6uo4zRwYMHN3rOLaFQdHS0jh8/rgEDBkiqP3Oopfr37y9fX193dKndlOed1P/9+wP17dtXMhhqfskgQ62va/9ukCSDUTJU/S7Vb+u8R02b6msMMtQ61sTz6j7L5Xit59S+Hh1Sampqk/9iANobYxSejjEKT8cYhadjjMLTecsYdUsodO2112rbtm26+uqrlZubq3feeUcvvviiO2593sn6+6MKzs5Q5n/auyfuUh0SqU4I5RogyVAdIhnqh1tqRdsGg7CWhWMtC8KablvZD9eArrK/9UM3l3vU+jwNtjU09Dla07Z2f5v5zC1oazmWrgJLYaPPqvxfQ++hkbYNhpJ1wscmf/b1g8rGPkfls1rwHhp5VpPvFwAAAAC8SLOh0COPPKI9e/bo1KlTmjJlisLCwrR7925NmzZNc+bMUXx8vCZMmKCvvvpKV199tSRp1qxZ6tGjR5t33hNF3fqQ/u/f76tvXF/J4aj8JcnhsEsOSXLUOu6Qo1YbVbVxqLm2Nccdzu+rrq98WK3jNb+3qG0Tz3J+jqrrWtS2zsamIckAACAASURBVOdwPqvBtnXeQwPPqvc5qt+Vw171qFpt7XY5HA456n1me9WjGvjMdY7VPK+mX41/5la0bbBf1e+m7QVKOvn1OXnUeaalgVcTgWJr2tYJwpxhVYvDMddn1QvC1Iq2LgFdI6HbmQaVLp+tbtuG34PvseP/n717j2+yvvs//k7SHHpIT5S29ACKYtksUgRh06pYmKCCp5863XSi4zRgbHh2U1E8IPPANhAZbIinOW88cyvq3JDN6e2ht047FblBBcqhQHo+JGmS3x+lV5umR2hJaF7Px4NH2ivfXPlc6ReavPkeVOHZcbBt23OaW4LhkHrMwc9pXE/rEPfgOYMC7lavVUi42dVzmNQ6pA4Ki41QubPnCD5nUO2tzwkAAKJC0+cHf6vPUP5WnyP8LZ8v2nzd9Nmj47bGZ5t2vg5+nq7uDxz8zNr2ebpXc9M1dnVN3TxXe9ekQ7umtueOazRLUTBSyBQwPsWGj9vtVklJSb+YPiZFzzAz9I2QcMoIkFqHS2o/CGs35Gtu2/K4kpIS5Z/43XbDsXaDsNbP1WHY2EXg1e519LBtR9fW+jp60rZ1mNc2vOsq5Gtz/vauo71wtqOfZ0sIGhymhp6j9c+jk7ZtrzkofGz/dWj5Rdl1Pwt6LbpqG3LNrepB93USPAUdU6swqm0I2GokZGh41xw+tXNOUzfOqTYBXrvnbBO0mUySQkM+U6saunNOmcwhIV+XwWF7QWHb52kvODx4nVu3bdNxxx/fxTlbh4DmNq9bS/DX9jVp/5w9+3mGvs6MiIw2vB89evX3QKD56507dygnO7uD+8MfCHT2PE3vpQ7ltW117qD/OA++H1L770fMob8nQ37nBbdv97Eyd+vclY1mfWfawnC/EH0uohaaBqBWH3ZaHevl5/DHlcqa2vN1v4DeFmj95qfVG6OPP/5YowoKunjzpDZvgtXmTWF7b9a6eEPW0XOEhFv+Nt83H1PQG9DQ0LR1SNcmiDPeaLY9FnzOppeq1ZtHhb75DH3z2uacanPOQOg5g96kqp1z+v3yt30DG/Qmuu2b5q7O2frnGXys7YjLQJtr1REaZdlagqS9/3vEn/Yw9SDk6yh86yLkC36zHTrqrdvBYav2IUFaO226CjSbQr52zilTO8/T/jmD6jgKRi2aa13yHCg9hA+tRyYQ6Ojf66gJBDr7XRIl4iS5Nnd0b5gDAbNF5h48T8j96qiOTupUB+fq4jXo3uvR+j9eDvGauvPatq1Dbe/v5uMj5D8ydveTRaa7QigEAAibpjcAluBjkmSxymxzhKUmHH3aHbEYNPqu61Aw9FgHgdbBc37x+ecaPnx4q7Cq7fO0/2Ev5JytpkG3HGv7gTL4nOo05GsnwGvnnGonTOzWOdX2WDfP2Xq6d+vXwN8of0e1q3X79n6e3QgOOwof2wkj+pskSTv/Ge4qDlV/DQTaXoOiLBAIDgc+/uTfGnXyyREbCADRglAIAAAc1Y7ECMu2fKXlcmQd38fPgiOpq1Cwz0YtqqV9S/jYwXMEhXydjDQJSF9/vU3HDh0aoYFAN6ZMov+LsclsPfqXDgGOdoRCAAAAiHrB4aKlz4PFvuZxxynhRNYUAgB0zhzuAgAAAAAAAHDkEQoBAAAAAABEIUIhAAAAAACAKEQoBAAAAAAAEIUIhQAAAAAAAKIQoRAAAAAAAEAUIhQCAAAAAACIQoRCAAAAAAAAUSgm3AVIUiAQkCR5PJ4wV9J73G53uEsAOkUfRaSjjyLS0UcR6eijiHT0UUS6/tRHbTabTCZTyHFToDmRCaPq6mp99dVX4S4DAAAAAACg38nPz5fdbg85HhGhkN/vV21traxWa7vJFQAAAAAAAA5NRI8UAgAAAAAAwJHFQtMAAAAAAABRiFAIAAAAAAAgChEKAQAAAAAARCFCIQAAAAAAgChEKAQAAAAAABCFCIUAAAAAAACiEKEQAAAAAABAFCIUAgAAAAAAiEKEQgAAAAAAAFGIUAgAAAAAACAKEQoBAAAAAABEIUIhAAAAAACAKEQoBAAAAAAAEIUIhQAAAAAAAKIQoRAAAAAAAEAUIhQCAAAAAACIQoRCAAAAAAAAUYhQCAAAAAAAIAoRCgEAAAAAAEQhQiEAAAAAAIAoRCgEAAAAAAAQhQiFAAAAAAAAohChEAAAAAAAQBQiFAIAAAAAAIhChEIAAAAAAABRiFAIAAAAAAAgChEKAQAAAAAARCFCIQAAAAAAgChEKAQAAAAAABCFCIUAAAAAAACiEKEQAADoV4qKivTuu++Gu4xD8v777+uMM84IdxkAACBKEAoBAAAAAABEIUIhAAAAAACAKEQoBAAA+iWPx6N7771XhYWFKiws1L333iuPxyNJcrlcmjVrlsaMGaOxY8fqRz/6kfx+vyRp1apVOv300zVq1ChNmjRJ7733XqfP4/f7tWrVKk2cOFHjxo3TL37xC1VUVEiSdu7cqby8PL344osaP368xo0bp0cffdR4bENDg2655RadcsopOvfcc/XZZ5/10asBAAAQKibcBQAAAPSFRx99VP/+97/18ssvy2Qyac6cOVqxYoV++ctf6rHHHlNGRoYR+Pz73/+WyWTStm3b9PTTT+u5555TRkaGdu7caYRFHXniiSf01ltv6amnnlJqaqruueceLVq0SA8//LDRpri4WK+//rq++eYbXXLJJTr77LN13HHHafny5dq+fbv++te/qr6+XjNmzOjT1wQAAKA1RgoBAIB+af369Zo7d64GDBig1NRUzZ07V6+88ookKSYmRvv27dOuXbtktVo1ZswYmUwmWSwWeTwebd26VV6vVzk5ORo8eHCnz/Pss89qwYIFyszMlM1m07x58/TGG2+osbHRaDNv3jw5HA4NHz5cw4cP15dffilJ2rBhg2bPnq3k5GQNGjRIV111Vd+9IAAAAG0wUggAAPRLZWVlysrKMr7PyspSWVmZJOmnP/2pli9frmuvvVaS9MMf/lAzZ87UkCFD9Ktf/UrLli3T//3f/6mwsFC33HKLMjIyOnyeXbt2ae7cuTKbW/6vzWw268CBA8b3aWlpxtexsbGqq6szahw0aFBQjQAAAEcKI4UAAEC/lJ6erl27dhnf7969W+np6ZKkhIQE3XLLLfrb3/6mlStX6rHHHjOmkk2dOlXPPPOMNm7cKJPJpAcffLDT58nMzNTq1av10UcfGX8+++yzToOkZgMHDtTu3buDagQAADhSCIUAAEC/dN555+nRRx+Vy+WSy+XSI488oqlTp0qSNm7cqG+//VaBQEAJCQmyWCwym83atm2b3nvvPXk8HtlsNtntdlkslk6f54orrtBvf/tblZaWSmpaxPqtt97qVo3nnHOOVq1apcrKSu3Zs0dPPvnk4V00AABADzB9DAAA9Etz5sxRbW2tzj//fEnS5MmTNWfOHEnSt99+q7vvvlsul0uJiYm64oorNG7cOH355Zd66KGHtHXrVlmtVo0aNUqLFi3q9Hl+8pOfKBAI6Nprr1VZWZkGDBigc889VxMnTuyyxnnz5mnhwoWaMGGC0tPTdfHFF+uJJ544/IsHAADoBlMgEAiEuwgAAAAAAAAcWUwfAwAAAAAAiEJMHwMAAOjC9OnTVVxcHHJ81qxZmj17dhgqAgAAOHwRMX0sEAgYCzqaTKZwlwMAAAAAANDvRcT0MY/Ho5KSEnk8nnCX0iv+85//hLsEoFP0UUQ6+igiHX0UkY4+ikhHH0Wki5Y+GhGhUH/T0NAQ7hKATtFHEenoo4h09FFEOvooIh19FJEuWvoooRAAAAAAAEAUIhQCAAAAAACIQoRCAAAAAAAAUYhQCAAAAAAAIAoRCgEAAAAAAEQhQiEAAAAAAIAoRCgEAAAAAAAQhboMhcrLyzVjxgxNmjRJU6dO1bx58+RyuTpsv23bNo0cOVJLlizp1UIBAAAAAADQe2K6amAymTR9+nSNGzdOkrRkyRI9+OCDuu+++0La+nw+LVy4UBMnTuz9SgEAAAB0KBAIyFXVoO17qvXJtlpV+LeHuySgQ998Qx9FZCvf16DR4S7iCOgyFEpOTjYCIUkqKCjQM888027bVatWafz48aqrq1NdXV3vVQkAAABAkuT3B7Svol479lZr+55q7Syr1va91dqxt1p1DY0tDf+nPHxFAt1BH0UEs8WYdMm54a6i73UZCrXm9/v1zDPPqKioKOS+L7/8Uu+8846eeOIJrVixotcKBAAAAKKRzx/Q3gO1RuBj/CmrkdvjM9qlOO3KzXDqrNG5ys1wanCGU3t2btWIEflhrB7oXElJifLz6aOIXP/31efhLuGIMAUCgUB3G991113au3evli9fLrO5ZTkir9erH/3oR1q8eLGOP/54LVu2THV1dbr55pu7dV63262SkpKeVw8AAAAc5Rp9AblqGrWv0qt9lY3aX9Vy6/O3tEuMs2hgUowGJlqbbpOsSku0Ks7O3jEAgI6NHt3xRLhujxRasmSJvv32W61cuTIoEJKkffv2afv27Zo5c6YkqaqqSoFAQDU1Nbr77ru7XWh+fr7sdnu320eq4uLiTl90INzoo4h09FFEOvooDoXH61Ppvhpt39M84qfpdte+Wvn8Tf9PazJJGalxGpyVqtMKnE0jfzKdyklPUJzD2u3noo8i0tFHEemipY92KxRaunSpSkpKtGrVKtlstpD7s7Ky9P777xvf93SkEAAAANBf1LsbtbOsebpXTdPaP3urtfdArQ5mPzKbTRo0IF6DM536Xv4gDc5oCoCy0xPksPVohQcAAA5Zl79xtmzZopUrV+qYY47R5ZdfLknKycnRI488ohkzZmj+/PkaMWJEnxcKAAAARJKaeq927q0OWfOnrLzeaBNjMSl7YIKGZidp/Mk5xpo/WQPjZY2xhLF6AAC6EQoNGzZMmzdvbve+1atXt3v85z//+eFVBQAAAESIyhp30CLPO/Y0BUGuqgajjS3GrJwMp75zzACd/b0EY+RP5oB4xVhY8wcAEJkYmwoAAICoFwgEVF7tNgKfHWUtI38qazxGu1i7RbkZThWcMLAp+MlsGvkzMCVOFrMpjFcAAEDPEQoBAAAgagQCAe2rqG813atlzZ/aeq/RLj7WqsEZTo07cZAx5Ss3w6m0ZIdMJsIfAED/QCgEAACAfsfnD6jMVWcEPs0h0M6yatW7fUa75AS7cjISdMaobCP4GZzhVLLTTvgDAOj3CIUAAABw1Gr0+bV7f23IyJ+dZdXyNPqNdgOSHMpNd2ri2CFG8JOTnqCkBHsYqwcAILwIhQAAABDxvI0+le6r1Y49Tev9NI/+2bWvRo2+gNEuPSVWuRlOnTQszVjzJzfdqfhYaxirBwAgMhEKAQAAIGI0eBq1s6wmaIv3HXurtftAnfz+pvDHbJIyBsRrcIZTY7+baYz8yU5PUKydt7cAAHQXvzUBAABwxNU1eIOmfDWP/Ckrr1Pg4MAfi9mkrIHxGjIoUYUFLWv+ZA9MkM1qCe8FAADQDxAKAQAAoM9U13m0fc/B8Kesumn6195q7a9sMNpYY8zKHpigvMEpmjh2sDHyZ1BavGIs5jBWDwBA/0YoBAAAgMMSCARUUeNuCn72VGtHWcs27xXVbqOd3WZRbnqCRhyf1rLNe6ZTGanxspjZ6QsAgCONUAgAAADdEggEdKCyIWiL9+Y/1XVeo12cI0a5GU6d8p0M5bba5j0tOVZmwh8AACIGoRAAAACC+P0BlZXXBW3x3jzyp97daLRzxtk0ONOp00ZmKzcjwVjzJzXRIZOJ8AcAgEhHKAQAABClfD6/9rjqtH1PtXa22uZ9x94aebw+o12K067cDKcmjMlt2uL94MifpAR7GKsHAACHi1AIAACgn/M2+rVrf03IyJ+dZTVq9PmNdmnJsRqc4VT+91ut+ZORoIQ4WxirBwAAfYVQCAAAoJ9we30qLWva3n3n3paRP7v218rvb9rn3WSSMlLjlJvh1Ojh6caaPznpCYpzWMN8BQAA4EgiFAIAADjK1Lsb2yz03DT6Z4+rVoGm7Edms0mDBsRrcKZTp56UZYz8yRoYL4eNt4AAAIBQqNd9vLlMH26pkce2W6mJdqU4HUpJtMsaYwl3aQAA4ChTU+89uMV7y0LPO/ZWa195vdEmxmJS9sAEHZeTpLNG5xhr/mSlxfP+AwAAdIpQqJf919++UsnWCr364QdBxxNirUpJdCjFaVdqokPJB29TnPag4/GxVnbrAAAgylTWuEOmfO3YWy1XldtoY4sxKyfDqROPHaDc7x1c7DnTqczUOFks5jBWDwAAjlaEQr3sntmn6R//+lC5x56g8qoGlVe7jVtXVYMqqt364huXyqsa5Gn0hzzeGmNuCooOjjBqug0Nk5KddsXwBhAAgKNGIBCQq6ohaLpXcwBUVesx2sXaLcrNcGpUXrqxxXtuhlMDU+JkMfMfRwAAoPcQCvUyi9mkxDiLjs9J7rRdIBBQXUOjyqsbVF7lVnl1g1xVblVUN8h1METavb9Wn3/tCnqj2FpivC1kpFGy0xE0bS3F6VCcI4bRRwAAHCF+f0D7K+qbRv6UVWv7npaRP7UNjUa7hFircjOc+v6IQUbwk5vuVFqyg9/bAADgiCAUChOTyaT4WKviY63KSXd22tbb6FdlTctII1fQCKSmUGnXvhq5qtxB28o2s1ktnU5ba75NTrAz/BwAgG7y+QPa66o9uOZPy8ifnXur1eDxGe2SE+zKzXDqzJNzmkb+ZDaFP8lOO+EPAAAIK0Kho4A1xqy05FilJcd22i4QCKi23tsmNAoOk3aW1eiz/9uvmnpvyONNJikp3t5q2lrwbeswKdZO1wEARIdGn1+799cao322763W5q/L5Pqv/5a31VTwAUkO5WY4dfa4IUHbvCcl2MNYPQAAQMe6/GRfXl6um266Sdu3b5fNZtOQIUO0aNEipaamBrV75JFH9Nprr8lisSgmJkYLFizQ6aef3meFI5TJZFJCnE0JcTYNzuy8rbfR1860Nbcx8shV3aDte6pUXu2Wzx8IebzDZgkZadQUGjWFSM2jkhLj7ax/AAA4Kni8PpXuqwlZ82fXvpqg34XpqXFKirPo1IIhGpyRcDD8cSo+1hrG6gEAAHquy1DIZDJp+vTpGjdunCRpyZIlevDBB3XfffcFtTvppJN07bXXKjY2Vl9++aWuvPJKvfPOO3I4HH1TOQ6LNcai9NQ4pafGddrO7w+ous4TMl3NVd2gioO33+yq0sfVDaprtU5CM7PZpOQE28HgqCVESnXalZzoUGrzSKREh+xWts0FAPS9BnejdpbVtGzzfnDNnz0HatWc/ZhNUuaAeOVmOPW9/ExjvZ+c9AQ57DEqLi7W6NEnhvdCAAAADlOXoVBycrIRCElSQUGBnnnmmZB2rUcF5eXlKRAIqKKiQpmZXQxZQUQzm01KSrArKcGuYwYldtq2wdOoimp3q9CoQa5W09jKqxu0rbRCFdVutTP4SHGOmJaRRk6Hkg/ett2FzRlnk5nRRwCALtTWe7WjrHmb95aRP2WuOqONxWxS1sAEHZuVpDNGNa35k5ORoOyBCbLxnxUAAKCf69HCMH6/X88884yKioo6bffSSy9p8ODBBEJRxmGLUeaAGGUOiO+0nc8fUFWtu2XR7OZpa63WP9qys0IV1Q2qd/tCHm8xm5oWxm490qjNbaqzafoab+gBoP+rqvUY6/203ub9QGWD0cYaY1ZOeoKGD0nR2WMHG2v+DEqLVwybLAAAgChlCgQC7YzZaN9dd92lvXv3avny5TKb238D9cEHH+imm27SmjVrNHTo0G6d1+12q6SkpLtlIIq4vX7VNPhVU+9TTYNP1fU+1dT7VdPgU029T9UHv65tCN11TZIcNpMSHBY5Yy1KcJiVEGtRQqxFTodFCbEHv3dYFGszsQMMAESwQCCgmga/9ld5ta+yUfsqD95WeYN+B1hjTBqYGKOBSVYNTIpRWqJVA5OsSom3MMoUAABEpdGjR3d4X7dHCi1ZskTffvutVq5c2WEg9PHHH+vGG2/UihUruh0ItZafny+7/ejfoaNpnYGOX3T0Pp/Pr8paT9BOa83rHzXf7qtu0OZd9fJ4Q0cfxVjMQSOMUtssot28JlKy0y5rzNH/P8r0UUQ6+mj0CgQC2l/R0DTqp82aP613zox3xCg3w6lTj3FqcKbTWPMnLTn2iIQ/9FFEOvooIh19FJEuWvpot0KhpUuXqqSkRKtWrZLNZmu3zaeffqoFCxbo97//vU48kYUXcWRZLGalJjqUmtj5wuaBQED17saDoVGr9Y6qWqav7XXV6YtvXKqq9bR7DmecTakhU9ZadmBrDpXiHDGMPgKADvj9AZWV14VM+dqxt0b17paNC5xxNg3OdOr0guyDU76advtKTXTwbywAAMBh6jIU2rJli1auXKljjjlGl19+uSQpJydHjzzyiGbMmKH58+drxIgRuuuuu9TQ0KA77rjDeOxvfvMb5eXl9V31QA+ZTCbFOayKc1iVk+7stG2jz2+MOmoZfRS8C9uubTVyVbnV6AudvmazWppGGbUacdQUGjmCQqXkBLssrGcBoJ/y+fzafaDW2OK9OQDaWVYTNHIzNdGu3AynJpySe3CxZ6cGZziVlHD0jyAGAACIVF2GQsOGDdPmzZvbvW/16tXG188//3zvVQVEgBiLWWnJsUpLju20XSAQUG29N2T0UeswaWdZjT77v/1BUx+amUxSUrzdGGHU3vS15uNxDmtfXS4AHBZvo0+79tU2Tfna0zLyp3RfbVBwPjAlVrkZTp10fJox5Ss3I0EJce2PRAYAAEDf6dHuYwBCmUwmJcTZlBBn0+AuNtzzNvqMdY5cVW5VHLxtvf7R9j1VKq92y+cPXQPeYbO0TFtrFRi1XvcoJdGuxHi7LCyoCqAPuL0+lZbVtJru1bTmz+4DtfIf/HfLZJIyU+OVm+HUmO9kGDt95aQnEG4DAABEEEIh4AiyxliUnhqn9NS4Ttv5/QFV13nambbWcvvNrip9Ut2g2obGkMebzSYlJ9gOTlULXTS7bJ9b2Qdqley0y2HjnwEAoeoavNpZVhOy5s9eV52a9y01m03KSovX4EynCkdmGVO+stMTZLdawnsBAAAA6BKfBoEIZDablJRgV1KCXUMGJXbatsHTqIpqd6ud1lqmr5VXNx3bVlqpihq38b/4krTmr29JkuIcMSEjjVIOrnvUOlRyxtnYzhnoh2rqPNqxt83In73V2l9Rb7SJsZiVk56gYbkpKhoz2FjsOSstoV/syAgAABCtCIWAo5zDFqPMATHKHBDfaTufP6DqWo/Kqxv0QXGJBmTkNoVIrUYfbd1ZofLqBtW7fSGPt5hNSm5nnaOQ6WtOu2yMEAAiSiAQUGWNx9jivfWaP+XVbqOdzWpRbkaC8o8bcHCtn6bt3jNT41gQHwAAoB8iFAKiRHOok+y0y7XbodGjB3fYtt7dGLTOUetbV3WD9lfUa8uOClXWuI1pJK3Fx1pbdlhrNfooJdGuVKdDyYlNYVJCrJUtpYFeFAgE5KpqaDXdq8ZY86e6zmO0i7XHaHCGU6OHZxijfnIznEpPiWNEIAAAQBQhFAIQItYeo1h7grLSEjpt5/P5VVnrCdpprW2ItHm7S64qd9DW081iLOaDgVFzaORQqtOu5IO3zaOPkp12pqgArfj9Ae2rqDeme7Ve86eu1TpjCbFWDc506tSTBhnBz+AMpwYkOQhkAQAAQCgE4NBZLGalJjatO9SZQCCgendjh4tmu6oatNdVpy++camq1tPuOZxxtuCRRm1GIDWHSvGOGD7sot/w+QPae6A2KPTZsbdaO8pq5Pa0BK3JTrsGZzg1/uQcDc5wKjezKQBKTrDz9wEAAAAdIhQC0OdMJpPiHFbFOazKSXd22rbR529aOLvViCNXm0W0d22rUXm1W95Gf8jjbTHmNiONghfRbg6TkhPsrJGCiNHo82v3/tqW4Ofgmj+l+2qC+nlakkO5GU5NGjfEGPmTm+FUYrwtjNUDAADgaEUoBCCixFjMSkuOVVpybKftAoGAauu9wTutBd02aGdZjUq27ld1nTfk8SaTlBRvNxbLDl00u+V4nMPaV5eLKOPx+lS6ryZk5M+ufbXytdodMCM1TrkZTo3KS9fgVmv+0BcBAADQmwiFAByVTCaTEuJsSoizKTej89FH3kZfh9PWmtdC2r63WhXVDWr0ha6c7bBZQhbMTnE6lJpoV7LTYYRJiQl2WVikF5Ia3I3aWRa6zfveA7Vqzn7MJmlQWrxyM5z6Xn7Lmj85AxPksPPrGQAAAH2Pd50A+j1rjEXpKXFKT4nrtJ3fH1BNvdcYaeSqcqsiaPqaW9/uqdInXzWottVivs3MJikpIXSkUWpi6C5sDhv//PYHtfXepm3e9wSP/CkrrzfaxFhMyhqYoKHZSTpzVMuaP1lp8bJZLWGsHgAAANGOTyUAcJDZbFJivE2J8TYNGZTYaVu316fy1ruutRp9VH5wTaSvd1WposYtvz909FGsPSZkpFFKO9PXnHE2tgiPAJU17uCRPwdDIFdVg9HGFmNWTrpTw49J1dnjWtb7GZQWrxjWrwIAAEAEIhQCgENgt1qUOSBemQPiO23n8wdUXesxRhq5Do5Caj2NbevOCpVXN6je7Qt5vMVsUrLTHrJgdvP0teZd11KcdkadHKZAIKCKarcR/LQe+VNZ07IrnsNmUW6GUwUnDDS2eM/NcCo9NY7pgwAAADiqEAoBQB9qDnWSnXYdm9V523p3Y9Cua61vXdUNOlDRoC07KlRZ41YgdPCR4mOt7UxbC92FzRlnjeptygOBgPZV1Gvn3tA1f2rrWxYlj3fEaHBmosZ+N1ODM1tG/qQlxTJ6CwAAAP0CoRAARIhYe4xi7QnKSkvotJ3P51dlrSdo4WxXdYMqDoZH5VVufbW9XK4qtzze0NFHMRbTwWlrwSONTGpL3gAAIABJREFU2q6FlOJ0yBpz9E578vkD2lde1xT8tFrzZ2dZddCorKSEpsXKzyjIbhn5k+lUitMe1eEZAAAA+j9CIQA4ylgsZqUmNq1F1JlAIHBw9NHBndaM0KglTNrrqtOX37qCpke15oyzhow0alkLqSVUinfEhC1A8fn82n2gtmXK154aI/zxNPqNdqmJDuVmJGjCKYONKV+5GU4lJdjDUjcAAAAQboRCANBPmUwmxTmsinNYlT2w89FHjT6/KmtaLZTdauHs5sW0P//apfKqBnlbBS3NbDFmJbe761pTmJR6cC2k5AS7LIe46LK30add+2pD1vzZta9Gjb6W+XTpKbHKyXDqpGFpxsifnAynEmKth/S8AAAAQH9FKAQAUIzFrAFJsRqQFNtpu0AgoNp6b/BOa0G3DdpZVqOSrftVXecNebzJJCXG25pGGLWZsmYsop3o0G6XR28X7wha82f3gTpjJzeTScocEK/cdKdO+U6GseZPTrpTsXZ+tQEAAADdwTtnAEC3mUwmJcTZlBDXtA5PZ7yNvnZCo+DRRzvKalRR3RA00qdFmSxmk7IGxmtwZqIKRx5c8yfTqayBCbKz2xoAAABwWAiFAAB9whpjUXpKnNJT4jpt5/cHVFPvNUYauarc2rnjG43//kgNSks4qhe7BgAAACIZoRAAIKzMZpMS421KjLdpyKBESVKxqUyDMxPDXBkAAADQv3UZCpWXl+umm27S9u3bZbPZNGTIEC1atEipqalB7Xw+n+655x7985//lMlk0syZM3XppZf2WeEAAAAAAAA4dF2OyTeZTJo+fbreeOMNrV+/Xrm5uXrwwQdD2q1fv17bt2/Xm2++qWeffVbLli3Tzp07+6RoAAAAAAAAHJ4uQ6Hk5GSNGzfO+L6goEC7du0Kaffaa6/p0ksvldlsVmpqqiZOnKjXX3+9d6sFAAAAAABAr+jRmkJ+v1/PPPOMioqKQu7bvXu3srKyjO8HDRqkPXv29KiYkpKSHrWPZMXFxeEuAegUfRSRjj6KSEcfRaSjjyLS0UcR6fpLHx09enSH9/UoFLr77rsVFxenK6+88rCLak9+fr7sdnufnPtIKi4u7vRFB8KNPopIRx9FpKOPItLRRxHp6KOIdNHSR7u9z++SJUv07bff6re//a3M5tCHDRo0KGha2e7du5WZmdk7VQIAAAAAAKBXdSsUWrp0qUpKSvTII4/IZrO122by5Mlat26d/H6/XC6X3nrrLU2aNKlXiwUAAAAAAEDv6DIU2rJli1auXKmysjJdfvnluuCCCzR37lxJ0owZM/TZZ59Jki644ALl5OTo7LPP1mWXXaa5c+cqNze3b6sHAAAAAADAIelyTaFhw4Zp8+bN7d63evVq42uLxaK77rqr9yoDAAAAAABAn+n2mkIAAAAAAADoPwiFAAAAAAAAohChEAAAAAAAQBQiFAIAAAAAAIhChEIAAAAAAABRiFAIAAAAAAAgChEKAQAAAAAARCFCIQAAAAAAgChEKAQAAAAAABCFCIUAAAAAAACiEKEQAAAAAABAFCIUAgAAAAAAiEKEQgAAAAAAAFGIUAgAAAAAACAKEQoBAAAAAABEIUIhAAAAAACAKBQT7gIAAAAAANGpqqpKZWVl8nq94S4FCBITE6Mvvvgi3GV0ymq1Kj09XYmJiYd8DkIhAAAAAMARV1VVpb179yo7O1uxsbEymUzhLgkw1NbWKj4+PtxldCgQCKi+vl6lpaWSdMjBENPHAAAAAABHXFlZmbKzsxUXF0cgBPSQyWRSXFycsrOzVVZWdsjnIRQCAAAAABxxXq9XsbGx4S4DOKrFxsYe1vTLLkOhJUuWqKioSHl5efrqq6/abXPgwAHNnDlTU6dO1eTJk3XnnXeqsbHxkIsCAAAAAPR/jBACDs/h/h3qMhSaMGGCnn76aWVnZ3fYZuXKlTruuOO0fv16rV+/Xv/5z3/05ptvHlZhAAAAAAAA6DtdLjQ9ZsyYLk9iMplUW1srv98vj8cjr9erjIyMXikQAAAAAAAAva9X1hSaM2eOvv76axUWFhp/Ro8e3RunBgAAAAAgouXl5XX5p6io6LCe44UXXlBeXp527tzZKzUvW7ZMeXl5vXKuo9VVV12lq666KtxlhFWvbEn/+uuvKy8vT48//rhqa2s1Y8YMvf7665o8eXKPzlNSUtIb5USE4uLicJcAdIo+ikhHH0Wko48i0tFHEek8Ho9qa2vDXUavWLt2bdD3N9xwg4YNG6ZZs2YZx2w222Fd79ixY7V27VrFx8f3yuvm8Xgkqd/8DA6Fz+eT1PFrcLS8Nh6Pp9N/8zsbtNMrodBTTz2l++67T2azWU6nU0VFRXr//fd7HArl5+fLbrf3RklhVVxczEgpRDT6KCIdfRSRjj6KSEcfRaQrLi6WzWZTfHx8uEvpFd///veDvrfb7UpLSws53prP51MgEFBMTPc+lsfHxysnJ+ew6mzNZrMZ541WFotFUvuvQW1t7VHz2thsNo0cOfKQHtsr08dycnL0j3/8Q1JTQvXee+9p2LBhvXFqAAAAAACOenl5eVq6dKlWrVqloqIi5efn66uvvpLb7dZ9992nKVOmaNSoUTrttNM0e/Zsbd26Nejx7U0fKyoq0g033KBXX31V55xzjgoKCnTxxRfro48+OqQaa2pqtGjRIhUWFio/P1+TJk3S2rVrFQgEjDa1tbW6++67NX78eOXn5+vUU0/VtGnTgup9/PHHdc455+ikk07SKaecoosvvlh//etfe1TLBx98oKuvvlqjRo1SQUGBfvrTn4bsiH7VVVfpiiuu0LvvvquLLrpII0eO1JQpU/TWW2+FnO/VV1/V5MmTlZ+fr/POO6/H9fRXXUaS99xzj958803t379f11xzjZKTk/Xqq69qxowZmj9/vkaMGKFf/epXWrhwoaZOnSqfz6dx48bpsssuOxL1AwAAAABwVHjhhReUm5urm2++WbGxsUpPTzem0f3sZz/TwIEDVVlZqT//+c/64Q9/qA0bNmjgwIGdnrO4uFhff/21fvGLX8hut+t3v/udZs+erb///e9KTEzsdm1+v18zZ87U559/rvnz5+uEE07Q22+/rcWLF8vlcum6666TJC1evFh///vftWDBAh1zzDGqqKjQ//7v/6q6ulqS9Morr2jJkiWaM2eOxowZI7fbrc2bN6uioqLbtbz99tuaM2eOzjzzTD3wwAOSpD/+8Y/68Y9/rFdeeUWDBg0y2u7YsUP33nuvZs6cqZSUFD322GOaP3++NmzYoCFDhkiS3n33XV1//fUaP368brnlFrlcLt17771qbGzUscce2+26+qMuQ6HbbrtNt912W8jx1atXG18PHjxYjz32WO9WBgAAAACIKn//aLv++sH2sNbwg7GDVTRmcJ+cOxAIaM2aNXI4HEHH7733XuNrn8+nwsJCnXrqqXr11Vc1bdq0Ts9ZU1Ojl156SUlJSZKktLQ0XXLJJdq0aZOmTp3a7do2bdqk4uJiLV68WBdffLEkqbCwUPX19VqzZo2mTZum1NRUffLJJ5o6daouvfRS47E/+MEPjK8/+eQT5eXlad68ecaxM888s9t1SE2vxymnnKJHH33UOPa9731PEyZM0Jo1a/TrX//aOF5eXq6nnnpKxxxzjCTpxBNPVGFhoTZs2KDZs2dLkn7/+99r6NChWrFihczmpglTxx13nC677LKoD4V6ZfoYAAAAAADo3Omnnx4SCEnSa6+9pksvvVRjxozRd7/7XRUUFKiurk7btm3r8pwFBQVGICTJ2FFs9+7dPartww8/lNls1pQpU4KOn3/++fJ6vfrkk08kSSNGjNCLL76olStX6rPPPjMWa242YsQIffHFF7r77rv17rvvqr6+vkd1fPPNN9q+fbumTp2qxsZG44/D4dCoUaNCpsYNGTLECIQkacCAARowYIB27dolqSlkKykp0aRJk4xASJJGjhyp7OzsHtXWH/XKQtMAAAAAAByuojF9N0onEqSnp4cca56KddFFF2nevHlKSUmRyWTSzJkzjR3COtM6EJJaFpB2u909qq2yslJJSUnG45ulpaUZ90tNs4nS0tL0/PPPa+nSpUpOTtYFF1ygBQsWKDY2VhdeeKHcbreee+45/fnPf1ZMTIzOPPNM3XLLLd1aKPvAgQOSpF//+tdBI4KaZWVlBX3f9vqlpteg+bUrLy+X1+s1rqO9a4tmhEIAAAAAAITJq6++qiFDhuj+++83jnm9XiOEOVKSkpJUWVkpj8cTFAzt379fkpScnCypaaeu66+/Xtdff71KS0v1xhtv6KGHHpLVatWNN94ok8mkyy+/XJdffrkqKyv1r3/9S/fff78WLFigdevWdVlH8/Ncf/317e7eZrVae3RdKSkpslqtxnW0tn///qgfLcT0MQAAAAAAwqShocHYGr3Zyy+/HDItq6+NHTtWfr9fr7/+etDx9evXy2q1trvleXZ2tq699lqdcMIJ2rJlS8j9SUlJOvfcc3XOOee0e397hg4dquzsbG3ZskUjRowI+TN8+PAeXZfFYlF+fr7eeOMN+f1+4/i///1vlZaW9uhc/REjhQAAAAAACJPTTz9db731lu677z6dddZZKikp0ZNPPtmjncN6wxlnnKHRo0dr4cKFcrlcGjZsmDZt2qR169Zp1qxZSk1NlST98Ic/VFFRkU444QTFxcXpww8/1JdffqkLL7xQknT77bcrPj5eBQUFGjBggL755hu9/PLLOu2007pVh8lk0sKFCzVnzhx5vV6dc845SklJ0f79+/Xxxx8rKytL11xzTY+ubf78+br22ms1Z84cXX755XK5XFq2bFmXO7tFA0IhAAAAAADC5LLLLtPu3bv1/PPP69lnn9WIESO0cuXKoN27jgSz2axVq1bp4Ycf1h//+EdVVFQoOztbt956q66++mqj3ZgxY7RhwwatWrVKPp9Pubm5uvXWW/WTn/xEknTyySfrhRde0Msvv6zq6mqlp6fr/PPP1/z587tdy5lnnqmnnnpKK1eu1G233aaGhgYNHDhQI0eO1Lnnntvjazv11FP14IMPatmyZZo3b56GDBmiX/3qV3riiSd6fK7+xhQIBALhLsLtdqukpET5+fmy2+3hLuewFRcXa/To0eEuA+gQfRSRjj6KSEcfRaSjjyLSFRcXKy4uTt/5znfCXQrQrtraWsXHx4e7jG754osvDvnvEmsKAQAAAAAARCGmjwEAAAAAgD4XCAS6XEA7JoaY4kji1QYAAAAAAH3uxRdf1K233tppm82bNx+haiARCgEAAAAAgCPgrLPO0nPPPRfuMtAKoRAAAAAAAOhzKSkpSklJCXcZaIWFpgEAAAAAAKIQoRAAAAAAAEAUIhQCAAAAAACIQoRCAAAAAAAAUYhQCAAAAAAAIAoRCgEAAAAAAEQhQiEAAAAAAA7Dz372M40dO1Yej6fd+2tqalRQUKBbbrml2+csKioKav/CCy8oLy9PO3fu7PRxO3fuVF5enl544YVuP1eztWvX6s033ww5vmzZMuXl5fX4fIer7WsQbQ7nZ9ldhEIAAAAAAByGiy66SJWVldq4cWO797/xxhuqr6/XhRdeeMjPMX78eD377LNKT08/5HN05Yknnmg3FLr00kv17LPP9tnzInwIhQAAAAAAOAzjx49XcnKyXnrppXbvf+mll5SVlaVx48Yd8nOkpqaqoKBANpvtkM9xqDIzM1VQUHDEnxd9r8tQaMmSJSoqKlJeXp6++uqrDtu99tprmjp1qqZMmaKpU6dq//79vVooAAAAAACRyGaz6bzzztM///lPuVyuoPt27dqlDz/8UOeff75MJpPeeecdzZgxQ4WFhRo5cqSmTJmiNWvWyOfzdfoc7U0fq6+v15133qlx48Zp1KhRmj17tvbs2RPy2E8//VTz58/XGWecoZNOOkmTJk3Sww8/rIaGBqNNUVGRSktLtX79euXl5SkvL8+YutXe9LGamhotWrRIhYWFys/P16RJk7R27VoFAgGjzfvvv6+8vDz97W9/06JFizRu3Dh973vf0w033KCqqqruv8BtrmXatGkaNWqUCgoKdPXVV+vTTz8NaXPNNddo3LhxGjlypCZMmKA777zTuH/fvn26+eabjdoLCws1a9YsHThwoNt1NDY26g9/+IMmT55snOP++++X2+022jRP//rLX/6i3/3udyosLNSYMWPa/Tl192fZ22K6ajBhwgT95Cc/0Y9//OMO23z22Wdavny5Hn/8cQ0cOFDV1dVhSS8BAAAAAAiHCy+8UE8//bQ2bNgQ9Pn5lVdeUSAQMKaO7dixQ9///vd15ZVXym63q6SkRMuWLZPL5dINN9zQo+e84447tGHDBs2dO1cjRozQv/71r3bPsXv3bg0fPlwXXXSR4uPjtWXLFq1YsUI7duzQ0qVLJUnLly/XzJkzlZeXp5///OeSmkYntcfv92vmzJn6/PPPNX/+fJ1wwgl6++23tXjxYrlcLl133XVB7e+9916dddZZeuihh/T111/rgQcekMVi0ZIlS3p0vV9++aWuvPJKHX/88Vq8eLFMJpNWrVqlK6+8Uv/1X/+l4cOHq7a2VtOnT9eIESO0ePFixcfHq7S0VB9//LFxnptuukm7du3STTfdpEGDBmn//v167733VF9f3+1abrzxRm3cuFHTp0/XySefrK1bt+p3v/udSktLtWzZsqC2q1at0qhRo3TvvffK5XLp/vvv1w033KCnnnrKaNPdn2Vv6zIUGjNmTJcnWbt2ra699loNHDhQkuR0Og+/MgAAAABAVKn+9G1V//vvYa3BObJIzpPG9/hxJ510ko4//ni99NJLQaHQyy+/rFGjRunYY4+VJF1xxRXGfYFAQGPGjJHX69WaNWt03XXXyWzu3iov27Zt03//939rwYIFmjlzpiSpsLBQdXV1+stf/hLUdtKkSUHPefLJJyshIUE333yz7rjjDqWkpOi73/2ubDabUlJSupwqtmnTJhUXF2vx4sW6+OKLjeeur6/XmjVrNG3atKBA6ZRTTtHtt99utPv666+1bt063X///TKZTN26XklasWKFbDab1q5dq8TEREnSaaedpqKiIi1fvlzLly/Xtm3bVFlZqRtvvFHDhw83HttcpyR98sknWrBggc4//3zj2DnnnNPtOj766CO99tprWrJkiRH2nXrqqUpKStKNN96oL774Qt/5zneM9llZWXrooYeM710ul37zm99o7969ysjI6NHPsrf1yppCW7du1Y4dO/TjH/9YF110kVasWBE0ZAwAAAAAgP7uggsu0Keffqqvv/5aUtM0pm3btumCCy4w2pSVlemOO+7QWWedpfz8fJ144on67W9/q6qqqh5NX/r000/l9/tDwozzzjsvpG1NTY0eeOABTZw4USNGjNCJJ56om266SYFAQN9++22Pr/PDDz+U2WzWlClTgo6ff/758nq9+uSTT4KOn3nmmUHfn3DCCfJ4PD1edubDDz/U+PHjjUBIkhISElRUVKQPP/xQknTMMccoMTFRCxcu1Msvv6zdu3eHnCc/P19/+tOf9Pjjj2vz5s09zi/++c9/ymq16uyzz1ZjY6Pxp7Cw0KiztfauX5JRW09+lr2ty5FC3eHz+bR582Y99thj8ng8mj59urKysnq8snpJSUlvlBMRiouLw10C0Cn6KCIdfRSRjj6KSEcfRaTzeDyqra0NOmY+7hQlHXdKmCpq0bau7vrBD36gpUuXat26dZo7d67WrVsnm82m8ePHq7a2Vn6/X7NmzdK+ffs0a9YsHXPMMXI4HNq4caP+9Kc/qaKiQnFxcZKapmg1NjYatTSvVVNfX6/a2lqVlpZKkmJjY4PqbX682+02jt9444364IMPNHv2bOXl5Sk2NlYlJSW6//77VVlZabRr+5zNPB5P0Oty4MABJSYmyuv1yuv1Gu3i4+MlNQVftbW1xppFDocj6JzNIUzr621P23oqKyuVnJwcUl9SUpJxHWazWX/4wx+0evVq3XXXXaqtrdVxxx2n2bNna8KECZKk++67z2hz3333KS0tTZdccommT58eNFKro36wd+9eeb1ejRo1qt37m6+/eTpa25+R3+83rqenP8v2eDyeTv/NHz16dIf39UoolJWVpcmTJ8tms8lms2nChAn69NNPexwK5efny26390ZJYVVcXNzpiw6EG30UkY4+ikhHH0Wko48i0hUXF8tmsxkhQn9x7LHH6tRTT9Xrr7+uX/7yl/rrX/+qoqIiDRo0SJL0zTff6PPPP9dvfvOboNFD7733nqSmUKD5NTGbzYqJiTG+b/6s3NwmOztbUlNIlJaWZpyrrq7OaB8fHy+3261NmzZp3rx5mjFjhtFu+/btkpoCm46es1nzmsHNxwcMGKCqqipZrdag9YSbg4uMjAzFx8fL4XCEPEd719KRtvU0hz9tH1NZWamkpCTj+Mknn6xHH31UjY2NKikp0R/+8AfdfPPNevnll3XCCScoPj5e99xzj6SmaXgvvfSSVq5cqYyMDP3oRz8yrqWj2tLS0mS32/X000+3e396erri4+MVGxtrXG/rc7V9Xbr7s+yIzWbTyJEjO7y/M70yfWzKlCl65513FAgE5PV69T//8z9Bc/cAAAAAAIgGF154oUpLS/Xwww+rvLw8aLBE88gZq9VqHPN6vVq/fn2Pn+ekk06S2WzWhg0bgo6/+uqrQd97PB75fD7FxASPCXnxxRdDzmm1WoN2z+rI2LFj5ff79frrrwcdX79+vaxW6yEHFF055ZRTtGnTJtXU1BjHampqtHHjRo0dOzakfUxMjAoKCvSLX/xCfr9fW7duDWkzdOhQXXfddUpKStKWLVu6Vcfpp58ut9utmpoajRgxIuRPRkZGj66ruz/LvtDlSKF77rlHb775pvbv369rrrlGycnJevXVVzVjxgzNnz9fI0aM0HnnnaeSkhKde+65MpvNKiws1CWXXNLnxQMAAAAAEEl+8IMfKCEhQWvXrtWAAQN0+umnG/cNHTpU2dnZWrp0qTEK5vHHHz+k5xk6dKimTJmi3//+9/L7/caOVf/4xz+C2jmdThUUFOixxx5Tenq6UlJS9Pzzz2vv3r0h5zz++OP10UcfaePGjUpLS1NKSopycnJC2p1xxhkaPXq0Fi5cKJfLpWHDhmnTpk1at26dZs2a1eGuZYdrzpw5evvttzVt2jTNmDFDJpNJq1evVn19vebOnStJ2rhxo5599llNnDhROTk5qq+v15NPPqn4+HiNGjVK1dXVmjZtmqZOnaqhQ4fKarXqb3/7myorK3Xaaad1q45x48ZpypQpmj9/vqZNm2aEOqWlpdq0aZNuuOEGY2Hx7ujuz7IvdBkK3XbbbbrttttCjq9evdr42mw269Zbb9Wtt97au9UBAAAAAHAUcTgcmjx5sp577jlNmTIlaISOzWbTI488okWLFunmm29WUlKS/t//+3/Kyspq93N3VxYtWqS4uDitWbNGXq9X48aN04MPPmhMgWr20EMP6c4779Rdd90lh8Ohc845R7/+9a81a9asoHbXXXedbr/9dv3yl79UQ0ODLrroIt1///0hz2s2m7Vq1So9/PDD+uMf/6iKigplZ2fr1ltv1dVXX93j6+iu4cOH68knn9TSpUt1yy23KBAIaOTIkXrqqaeM2UpDhgyRw+HQihUrtG/fPsXHx2vEiBF67LHHlJmZKY/HoxNPPFHr1q3Trl27ZDKZdOyxx+rBBx/UxIkTu13LAw88oCeffFLPP/+8Vq5cKZvNpuzsbBUWFgZNAeuu7v4se5spEAHbhLndbpWUlLCmEHCE0EcR6eijiHT0UUQ6+igiXXFxseLi4oK27QYiSWdrCkWaL7744pD/LvXKmkIAAAAAAAA4uvTK7mMAAAAAAABHO5/Pp0AgoMbGRjU2Nobcbzabg7atP9oRCgEAAAAAAEiaNm2aPvjggw7v72idpaMVoRAAAAAAAICku+66S7W1tWpoaJDD4Qi5PyUlJQxV9R1CIQAAAAAAADVtDy8dXQtNH47+MxEOAAAAAAAA3UYoBAAAAAAIi0AgEO4SgKPa4f4dIhQCAAAAABxxVqtV9fX14S4DOKrV19fLarUe8uMJhQAAAAAAR1x6erpKS0tVV1fHiCGghwKBgOrq6lRaWqr09PRDPg8LTQMAAAAAjrjExERJ0q5du+T1esNcDRDM4/HIZrOFu4xOWa1WZWRkGH+XDgWhEAAAAAAgLBITEw/rAy3QV4qLizVy5Mhwl9HnmD4GAAAAAAAQhQiFAAAAAAAAohChEAAAAAAAQBQiFAIAAAAAAIhChEIAAAAAAABRiFAIAAAAAAAgChEKAQAAAAAARCFCIQAAAAAAgChEKAQAAAAAABCFugyFlixZoqKiIuXl5emrr77qtO22bds0cuRILVmypNcKBAAAAAAAQO/rMhSaMGGCnn76aWVnZ3fazufzaeHChZo4cWKvFQcAAAAAAIC+EdNVgzFjxnTrRKtWrdL48eNVV1enurq6wy4MAAAAAAAAfafLUKg7vvzyS73zzjt64okntGLFikM+T0lJSW+UExGKi4vDXQLQKfooIh19FJGOPopIRx9FpKOPItL1lz46evToDu877FDI6/Xq9ttv1+LFi2WxWA7rXPn5+bLb7YdbUtgVFxd3+qID4UYfRaSjjyLS0UcR6eijiHT0UUS6aOmjhx0K7du3T9u3b9fMmTMlSVVVVQoEAqqpqdHdd9992AUCAAAAAACg9x12KJSVlaX333/f+H7ZsmWqq6vTzTfffLinBgAAAAAAQB/pcvexe+65R2eccYb27Nmja665Ruedd54kacaMGfrss8/6vEAAAAAAAAD0vi5HCt1222267bbbQo6vXr263fY///nPD78qAAAAAAAA9KkuRwoBAAAAAACg/yEUAgAAAAAAiEKEQgC4G4HMAAAW6UlEQVQAAAAAAFGIUAgAAAAAACAKEQoBAAAAAABEIUIhAAAAAACAKEQoBAAAAAAAEIUIhQAAAAAAAKIQoRAAAAAAAEAUIhQCAAAAAACIQoRCAAAAAAAAUYhQCAAAAAAAIAoRCgEAAAAAAEQhQiEAAAAAAIAoRCgEAAAAAAAQhQiFAAAAAAAAohChEAAAAAAAQBQiFAIAAAAAAIhChEIAAAAAAABRiFAIAAAAAAD8//buPziK+v7j+GvvRwJYSIRiCEOF0CkMw0GhYNGBqkMYqgOS1GlGBiytrWgnDnSwQwItFaGp0yidRmmcdrT+QafVGaaxNEQ0g61joHwVU2uKWKWOiiUJGCARws/c7fePu1xu7/ZyFzjYI/t8zDB7t/vZ3ffufWbhXux+Di6UMhSqrq7W/PnzNXnyZH3wwQe2bWpra7Vo0SItWbJEd999t5qamjJeKAAAAAAAADLHl6pBcXGxVqxYoeXLlydtM336dH3/+9/X0KFD9Z///Ef33nuv9uzZoyFDhmS0WAAAAAAAAGRGylBo9uzZKTfyjW98I/p68uTJMk1TnZ2dGjNmzOVVBwAAAAAAgCsi42MK/eUvf9GNN95IIAQAAAAAAJDFDNM0zXQazp8/X7/97W81adKkpG3efPNNVVRU6LnnntPEiRPTLuL8+fM6cOBA2u0BAAAAAACQ2qxZs5IuS/n4WLrefvttrV27Vk8//fSAAqFYgUBAubm5mSrJMc3Nzf2edMBp9FFkO/oosh19FNmOPopsRx9FtnNLH83I42MtLS1as2aNnnrqKU2dOjUTmwQAAAAAAMAVlDIUqqqq0q233qr29nbdd999WrRokSRp5cqV+ve//y1J2rRpk86dO6dHHnlEJSUlKikp0fvvv39lKwcAAAAAAMAlS/n42IYNG7Rhw4aE+c8880z09Z///OfMVgUAAAAAAIArKuO/PgYAAAAAAIDsRygEAAAAAADgQoRCAAAAAAAALkQoBAAAAAAA4EKEQgAAAAAAAC5EKAQAAAAAAOBChEIAAAAAAAAuRCgEAAAAAADgQoRCAAAAAAAALkQoBAAAAAAA4EKEQgAAAAAAAC5EKAQAAAAAAOBChEIAAAAAAAAuRCgEAAAAAADgQoRCAAAAAAAALkQoBAAAAAAA4EKEQgAAAAAAAC5EKAQAAAAAAOBChEIAAAAAAAAuRCgEAAAAAADgQj6nCxhsznz4tnI/+j91hY7K8OXI8PnCU69fhs/fN/XlyPDGLev9Y5DVAQAAAACAKytlKFRdXa1XXnlFR44cUX19vSZNmpTQJhgMqqqqSk1NTTIMQw888IDKysquSMHZrnPfixr2ybs6/v5lbMTjiwmTItPYQMk2YIoJlbz+vkDKaxNMxbaJbj/c1uPzS73LDSNj5wUAAAAAAGSXlKFQcXGxVqxYoeXLlydtU19fr8OHD6uxsVGdnZ0qLS3VLbfconHjxmW02GtB4fJH9c8339CMaVNl9lyUGbwYnvZclBm8EPO6d/4FmcEe6zRhvdi24dehM+f65ttsKyMigZHH55e8ftup4cuRvD55fDl9y3yRZZZAKlm4FbfMEnb5wwEZ4RQAAAAAABmXMhSaPXt2yo289NJLKisrk8fj0ciRI7VgwQK9/PLLuv/++zNS5LXEMDySL0feYSMcq8E0TSnUExcqXZDZ00/4FGkT6rkoBS9apz3haWyoFV12/kx4mU24pWBPRo4n6Z1SlkfwbEKnZG3TvtsqZpnHm5FjAQAAAAAgW2RkTKG2tjaNHTs2+r6wsFDt7e2Z2DQugWEYkjccbCjXuTpMMxQJiGLvlEoWSFnvhEq8UyqybuzdVpFloQvn+trYBF4KBS//YAxP6mAq+rieTbAUfaQvxWOACevF3W1FOAUAAAAAyJCsGmj6wIEDTpeQMc3NzU6XcA0xJOVE/kTe+pS53mmGpFBQRrAnPA31SKEeGaFgZNojBYMyzB4pGDs/Znl/618ISmdPJbaPnQZ7ZMi8/EMxPJLHK9Pjkzw+mR6v5I1MPbFTX0w7r0yvz7I81+PTgU/eCi/39rNewvLwVDzSh6uA6yiyHX0U2Y4+imxHH0W2Gyx9dNasWUmXZeRrd2FhoVpbWzV9+nRJiXcOpSsQCCg318FbWzKkubm535MOdzJDwbjH+Gzuhko1vpTtenbjSp2XedFuXKqLUgbCKXl9Nnc6+VPcKWX/q3t9Y1TlSD6fPN7INMXg6PIy3tRgxnUU2Y4+imxHH0W2o48i27mlj2YkFLrjjju0fft2LVy4UJ2dndq9e7f++Mc/ZmLTwKBheLwycoZGb4hyQu94U/98a79mBKamHNA8ncf4+safCo8j1RtMhSLjTdmNUWUGL2bkePr/1b00xqBKNo5U7Pp2g6PHzJPHSzgFAAAA4JqUMhSqqqpSY2OjOjo6dN999yk/P18NDQ1auXKlVq9erWnTpqmkpETvvPOOFi5cKEl66KGH9KUvfemKFw9gYHrHm5IvV97r8hyro3e8KfVYAyPr3VI2YVU0kErjbqueiwpdPC/z7Gn7X/vrCT8GePmM1MFUql/dS3pXVXp3Wxk+P+NNAQAAABiwlKHQhg0btGHDhoT5zzzzTPS11+vVpk2bMlsZgEHLMDzhR8B8OfI4WIdphhIGLrcNpmx/xS/VY3wxg6GfPyuz53Obu60iv9Rnhi7/YCLntC+Q6uduqP4GPo8Pt1I8xmf9pT5/+BcYAQAAAFwTsmqgaQC4mgzDI8OfK/mdHcssPN5UfPhkFzYlWZ5wl5V9cBU61534S4CZHm/K47P+Yl46v7pnEzrltrWp8+IRyVBf0GR4IoOcG+G73mL/JMzzyFBcG8PomxeZGoYnPLh9zD6MuO2G2/VtN9outqbeNnH7tM7zxBxPqtoVbd+7TyNmX5b1AAAAgEtEKAQADguPN+WVcoY4VkN4vKmgTcDU32N8/YRWdoOlR6ahM2eTjGfVN97UMEkn3nfsdFxbLKFZTHikuHAqPuSSYQmeBhZypRuixYZekX0p/TqjwV1sbTahYGIAmBisWY7nUkPBmPc5hz/V58bxuLCvv3MYaaeYfdrWnjjPLkhM+rnGh5dGknMX8/mnrj3JPIJJAMBVYJq9/3Fp9s6IbxCzrHfSX9tIw+gia9vo7IwMNZH9CIUAAOEvdl6fDK9Pyh3qWB2macoMXtTbzc2aOWOGZIbCfzGbofBf1KYZ+YeBGX0vmTLNUPhv8N5H8cxQuF1Mm+i6lvVMyzrhZiHLOn3r2bSLqc2Mral3PZnR9tGa4tpZ5kXrTH08lveRdvE12Z6vhHmhyL+BYvZpczxm3PFb9ml7vvq2H4qrM7rPmHXja7I7x8k//ySfq12dsX3iMlwnqePgZW1ikOjvrjebUDD+zrdU4aVdyKVU4ZU1KE0eCtoFbeneoRcX9qURrCXWmnhHY8rjiT0Pcecw/nhyWj/Wqdxu2y9L8V+AonPT+LIU09iyPPw2Sdt+95debbH7MwdYW/hw4tums9/eY0u9/b4dxX2pHMj+ouum39Z6/mNXSl5b8v3ZLbevpd9zkmY/+0JXl9oO7Yo0S+88Wc+rw/06/rO1+ywG0ncstaZuazknA+rX4fkD6teR5Ve8X0ffx9frjLycYdJNc5wu44ojFAIAZA3DMPrGm3IwnII7DCwU7JtnmqZa3nlH06dPUzRY6zd4652XGKylEwraBVrW4C1mnm2wlvxYLAGdbGrq73iioWeSuvoLL3v3aVu73fmyq91mXnwoGApJCimU6hwqMai0bD/J8aX1uSY7h1f4S891kj5ruaK7GAQid7rF3/EWfW/0NUmnbXRibWv027Z3P9btGvFt+6ltYPuzP46E/fVTW/x2jfg60lzHuHBGoTM29fRXi6XW+LaRc9IblkaWG5ZNxC6zqXtA+7PZhu15Ta9t3+eX+vzZ1Z7wOaSxjYH1696VBtCv47adVr+OTuL7dRrnMVXftD3O5J/7Jx2fyw0IhQAAgCuF76zo/eW+8NRI3tzCHPIF+YaPvCJ1wR1S3vVmF3LZBWuyC51COnDggAKBQGRv6X/JtH5JHNgX00v58mwMoG3sttP6ghi3bR53zC7Nzc2aNGuW02UASV1sbna6hKuCUAgAAAC4yvoeFYuZl8Hth65rk3/k2AxuEQAwGPHbwQAAAAAAAC5EKAQAAAAAAOBChEIAAAAAAAAuRCgEAAAAAADgQoRCAAAAAAAALkQoBAAAAAAA4EKEQgAAAAAAAC7kc7oASTJNU5J04cIFhyvJnPPnzztdAtAv+iiyHX0U2Y4+imxHH0W2o48i2w2mPpqTkyPDMBLmG2ZvIuOgU6dO6YMPPnC6DAAAAAAAgEEnEAgoNzc3YX5WhEKhUEjd3d3y+/22yRUAAAAAAAAuTVbfKQQAAAAAAICri4GmAQAAAAAAXIhQCAAAAAAAwIUIhQAAAAAAAFyIUAgAAAAAAMCFCIUAAAAAAABciFAIAAAAAADAhQiFAAAAAAAAXMjndAHXqo8++kjr1q1TZ2en8vPzVV1drQkTJljaBINBVVVVqampSYZh6IEHHlBZWZkzBcN10umjW7du1Z/+9CfdcMMNkqSvfe1r2rhxowPVwm2qq6v1yiuv6MiRI6qvr9ekSZMS2nANhdPS6adcR+GUkydPqqKiQocPH1ZOTo7Gjx+vzZs3a+TIkZZ2XEvhlHT7KNdROK28vFz/+9//5PF4NGzYMP3sZz/TlClTLG0G87WUUOgSbdy4UcuWLVNJSYl27NihRx55RNu2bbO0qa+v1+HDh9XY2KjOzk6Vlpbqlltu0bhx4xyqGm6STh+VpNLSUlVWVjpQIdysuLhYK1as0PLly5O24RoKp6XTTyWuo3CGYRi6//77NWfOHEnhEHPLli167LHHLO24lsIp6fZRiesonFVdXa3hw4dLknbv3q2f/OQnevHFFy1tBvO1lMfHLsHx48d18OBBLV68WJK0ePFiHTx4UCdOnLC0e+mll1RWViaPx6ORI0dqwYIFevnll50oGS6Tbh8FnDJ79mwVFhb224ZrKJyWTj8FnJKfnx/9si1JM2bMUGtra0I7rqVwSrp9FHBabyAkSadPn5ZhGAltBvO1lDuFLkFbW5sKCgrk9XolSV6vVzfccIPa2tost0O2tbVp7Nix0feFhYVqb2+/6vXCfdLto5LU0NCgPXv2aPTo0Vq1apVmzpzpRMlAAq6huFZwHYXTQqGQnn/+ec2fPz9hGddSZIP++qjEdRTO++lPf6q9e/fKNE09++yzCcsH87WUO4UAF1u6dKleffVV1dfX6wc/+IHKy8t18uRJp8sCgGsG11Fkg5///OcaNmyY7r33XqdLAWz110e5jiIb/OIXv9Brr72mNWvW6PHHH3e6nKuKUOgSFBYW6ujRowoGg5LCg04dO3Ys4RbzwsJCyy2SbW1tGjNmzFWtFe6Ubh8dPXq0/H6/JGnu3LkqLCzUoUOHrnq9gB2uobgWcB2F06qrq/XJJ5+opqZGHk/iP+25lsJpqfoo11Fkk9LSUr3xxhsJweRgvpYSCl2CUaNGacqUKdq5c6ckaefOnZoyZUrCYzl33HGHtm/frlAopBMnTmj37t365je/6UTJcJl0++jRo0ejr9977z0dOXJERUVFV7VWIBmuobgWcB2Fk37961/rwIEDqq2tVU5Ojm0brqVwUjp9lOsonNTd3a22trbo+7/97W/Ky8tTfn6+pd1gvpYapmmaThdxLfrwww+1bt06ff755xoxYoSqq6s1ceJErVy5UqtXr9a0adMUDAa1efNm7d27V5K0cuVK3XPPPQ5XDrdIp49WVlbq3Xfflcfjkd/v1+rVq3Xbbbc5XTpcoKqqSo2Njero6ND111+v/Px8NTQ0cA1FVkmnn3IdhVMOHTqkxYsXa8KECRoyZIgkady4caqtreVaiqyQbh/lOgondXR0qLy8XGfPnpXH41FeXp4qKys1depU11xLCYUAAAAAAABciMfHAAAAAAAAXIhQCAAAAAAAwIUIhQAAAAAAAFyIUAgAAAAAAMCFCIUAAAAAAABcyOd0AQAAAJlQV1en9evX2y4bPny43nrrratcUdi6dev0j3/8Q6+//roj+wcAAEiGUAgAAAwqTz75pMaMGWOZ5/V6HaoGAAAgexEKAQCAQWXKlCkaP36802UAAABkPcYUAgAArlFXV6fJkydr//79Ki8v18yZMzVnzhxt2rRJ586ds7Q9duyYKioqNGfOHAUCAd11113asWNHwjY//fRTrV27VnPnzlUgEFBxcbGqqqoS2h08eFDLli3TV7/6VS1cuFDPP/+8Zflnn32myspKzZs3T4FAQPPmzdODDz6o48ePZ/YkAAAARHCnEAAAGFSCwaB6enos8zwejzyevv8LW7t2re68804tW7ZMLS0tevrpp3X27Fn98pe/lCSdOXNG3/nOd9TV1aWHH35YY8aM0V//+ldVVFTo3LlzuueeeySFA6GysjINHTpUq1at0vjx49Xe3q49e/ZY9n/69Gn9+Mc/1ne/+1099NBDqqur06OPPqqioiLdfPPNkqSKigq1traqoqJChYWF6ujo0L59+3T27NkreboAAICLEQoBAIBB5c4770yYd/vtt+t3v/td9P2tt96qyspKSdK8efNkGIaeeuopPfjggyoqKlJdXZ0+/vhjbdu2TXPmzJEk3XbbbTp+/Lhqamr07W9/W16vV1u3btX58+e1Y8cOFRQURLf/rW99y7L/7u5ubdy4MRoA3XTTTdqzZ48aGhqi8/71r39pzZo1WrJkSb/HAgAAkCmEQgAAYFCpra21BDSSNGLECMv7+LBl0aJFqqmpUUtLi4qKirR//34VFBREA6FeS5Ys0fr16/Xf//5XkydP1t69e3X77bcn7C/e0KFDo+GPJOXk5GjChAlqbW2NzgsEAvr9738v0zR18803a9KkSTIMY0DHDgAAMBCEQgAAYFD5yle+knKg6S9+8YuW96NGjZIkHT16VJLU1dWl0aNHJ12vq6tLktTZ2ZnwS2d24kMpKRwMXbhwIfq+pqZGv/nNb/Tss8/qscce0+jRo7V06VKVl5dbHn0DAADIFP6FAQAAXKejo8Pyvncw5947fvLy8hLaxK6Xn58vSbr++uujQdLlGjVqlDZu3Kimpibt2rVLd999t7Zu3aoXXnghI9sHAACIRygEAABcZ9euXZb3DQ0N8ng8mj59uiTp61//utrb29Xc3Gxpt3PnTo0aNUpf/vKXJUlz587V3//+dx07diyj9U2cOFEPP/yw8vLydOjQoYxuGwAAoBePjwEAgEHlvffe08mTJxPmBwKB6OvXX39d1dXVmjdvnlpaWlRbW6vS0lIVFRVJCg8UvW3bNq1atUpr1qxRQUGB6uvrtXfvXm3evFler1eStGrVKr322mtaunSpfvjDH+rGG2/U0aNH1dTUpC1btqRd86lTp/S9731Pd911lyZOnCi/369XX31VXV1dmjt37mWeEQAAAHuEQgAAYFD50Y9+ZDt/37590ddPPPGEnnvuOb3wwgvy+/0qKyuL/hqZJA0bNkx/+MMf9MQTT2jLli3q7u5WUVGRHn/8cZWUlETbjRs3Ttu3b1dNTY1+9atfqbu7WwUFBSouLh5Qzbm5uZo6daq2b9+u1tZWGYahoqIibdmyRQsWLBjgGQAAAEiPYZqm6XQRAAAAV0NdXZ3Wr1+vxsbGlINRAwAADHaMKQQAAAAAAOBChEIAAAAAAAAuxONjAAAAAAAALsSdQgAAAAAAAC5EKAQAAAAAAOBChEIAAAAAAAAuRCgEAAAAAADgQoRCAAAAAAAALkQoBAAAAAAA4EL/Dy9spncdYu1/AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1440x1080 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fold: 4\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fold: 5\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set(style=\"whitegrid\")\n",
    "for n_fold in range(config['N_FOLDS']):\n",
    "    print('Fold: %d' % (n_fold+1))\n",
    "    plot_metrics(history_list[n_fold])"
   ]
  },
  {
   "cell_type": "markdown",
   "execution_count": null,
   "metadata": {},
   "source": [
    "# Model evaluation (last model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "_kg_hide-input": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col0 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col1 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col2 {\n",
       "            color:  red;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col3 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col4 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col5 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col6 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col7 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col8 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col9 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col10 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col11 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col12 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col0 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col1 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col2 {\n",
       "            color:  red;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col3 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col4 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col5 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col6 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col7 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col8 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col9 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col10 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col11 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col12 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col0 {\n",
       "            color:  green;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col1 {\n",
       "            color:  green;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col2 {\n",
       "            color:  red;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col3 {\n",
       "            color:  green;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col4 {\n",
       "            color:  green;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col5 {\n",
       "            color:  green;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col6 {\n",
       "            color:  green;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col7 {\n",
       "            color:  green;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col8 {\n",
       "            color:  green;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col9 {\n",
       "            color:  green;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col10 {\n",
       "            color:  green;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col11 {\n",
       "            color:  green;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col12 {\n",
       "            color:  green;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col0 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col1 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col2 {\n",
       "            color:  red;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col3 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col4 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col5 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col6 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col7 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col8 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col9 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col10 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col11 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col12 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col0 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col1 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col2 {\n",
       "            color:  red;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col3 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col4 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col5 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col6 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col7 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col8 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col9 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col10 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col11 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col12 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col0 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col1 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col2 {\n",
       "            color:  red;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col3 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col4 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col5 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col6 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col7 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col8 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col9 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col10 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col11 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col12 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col0 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col1 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col2 {\n",
       "            color:  red;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col3 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col4 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col5 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col6 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col7 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col8 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col9 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col10 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col11 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col12 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col0 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col1 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col2 {\n",
       "            color:  red;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col3 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col4 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col5 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col6 {\n",
       "            color:  orange;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col7 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col8 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col9 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col10 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col11 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col12 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col0 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col1 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col2 {\n",
       "            color:  red;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col3 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col4 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col5 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col6 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col7 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col8 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col9 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col10 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col11 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col12 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col0 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col1 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col2 {\n",
       "            color:  red;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col3 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col4 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col5 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col6 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col7 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col8 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col9 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col10 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col11 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col12 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col0 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col1 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col2 {\n",
       "            color:  red;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col3 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col4 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col5 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col6 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col7 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col8 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col9 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col10 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col11 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col12 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col0 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col1 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col2 {\n",
       "            color:  red;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col3 {\n",
       "            color:  green;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col4 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col5 {\n",
       "            color:  green;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col6 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col7 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col8 {\n",
       "            color:  green;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col9 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col10 {\n",
       "            color:  green;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col11 {\n",
       "            color:  black;\n",
       "        }    #T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col12 {\n",
       "            color:  green;\n",
       "        }</style><table id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >Train</th>        <th class=\"col_heading level0 col1\" >Valid</th>        <th class=\"col_heading level0 col2\" >Var</th>        <th class=\"col_heading level0 col3\" >Train_fold_1</th>        <th class=\"col_heading level0 col4\" >Valid_fold_1</th>        <th class=\"col_heading level0 col5\" >Train_fold_2</th>        <th class=\"col_heading level0 col6\" >Valid_fold_2</th>        <th class=\"col_heading level0 col7\" >Train_fold_3</th>        <th class=\"col_heading level0 col8\" >Valid_fold_3</th>        <th class=\"col_heading level0 col9\" >Train_fold_4</th>        <th class=\"col_heading level0 col10\" >Valid_fold_4</th>        <th class=\"col_heading level0 col11\" >Train_fold_5</th>        <th class=\"col_heading level0 col12\" >Valid_fold_5</th>    </tr>    <tr>        <th class=\"index_name level0\" >Metric/Jaccard</th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202level0_row0\" class=\"row_heading level0 row0\" >Overall</th>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col0\" class=\"data row0 col0\" >0.727435</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col1\" class=\"data row0 col1\" >0.704075</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col2\" class=\"data row0 col2\" >0.023360</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col3\" class=\"data row0 col3\" >0.732597</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col4\" class=\"data row0 col4\" >0.709649</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col5\" class=\"data row0 col5\" >0.727963</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col6\" class=\"data row0 col6\" >0.698772</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col7\" class=\"data row0 col7\" >0.721124</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col8\" class=\"data row0 col8\" >0.704331</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col9\" class=\"data row0 col9\" >0.728299</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col10\" class=\"data row0 col10\" >0.699828</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col11\" class=\"data row0 col11\" >0.727193</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row0_col12\" class=\"data row0 col12\" >0.707798</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202level0_row1\" class=\"row_heading level0 row1\" >Sentiment negative</th>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col0\" class=\"data row1 col0\" >0.562697</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col1\" class=\"data row1 col1\" >0.522338</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col2\" class=\"data row1 col2\" >0.040359</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col3\" class=\"data row1 col3\" >0.576491</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col4\" class=\"data row1 col4\" >0.518469</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col5\" class=\"data row1 col5\" >0.561975</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col6\" class=\"data row1 col6\" >0.520604</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col7\" class=\"data row1 col7\" >0.552463</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col8\" class=\"data row1 col8\" >0.521848</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col9\" class=\"data row1 col9\" >0.561584</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col10\" class=\"data row1 col10\" >0.522694</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col11\" class=\"data row1 col11\" >0.560972</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row1_col12\" class=\"data row1 col12\" >0.528075</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202level0_row2\" class=\"row_heading level0 row2\" >Sentiment neutral</th>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col0\" class=\"data row2 col0\" >0.973897</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col1\" class=\"data row2 col1\" >0.972530</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col2\" class=\"data row2 col2\" >0.001367</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col3\" class=\"data row2 col3\" >0.974242</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col4\" class=\"data row2 col4\" >0.973539</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col5\" class=\"data row2 col5\" >0.974403</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col6\" class=\"data row2 col6\" >0.973450</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col7\" class=\"data row2 col7\" >0.971582</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col8\" class=\"data row2 col8\" >0.970589</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col9\" class=\"data row2 col9\" >0.974867</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col10\" class=\"data row2 col10\" >0.970516</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col11\" class=\"data row2 col11\" >0.974392</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row2_col12\" class=\"data row2 col12\" >0.974557</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202level0_row3\" class=\"row_heading level0 row3\" >Sentiment positive</th>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col0\" class=\"data row3 col0\" >0.557534</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col1\" class=\"data row3 col1\" >0.521099</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col2\" class=\"data row3 col2\" >0.036435</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col3\" class=\"data row3 col3\" >0.561145</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col4\" class=\"data row3 col4\" >0.540992</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col5\" class=\"data row3 col5\" >0.559261</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col6\" class=\"data row3 col6\" >0.504334</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col7\" class=\"data row3 col7\" >0.549579</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col8\" class=\"data row3 col8\" >0.524981</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col9\" class=\"data row3 col9\" >0.560027</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col10\" class=\"data row3 col10\" >0.509891</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col11\" class=\"data row3 col11\" >0.557658</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row3_col12\" class=\"data row3 col12\" >0.525295</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202level0_row4\" class=\"row_heading level0 row4\" >0 <= text wordCnt < 3</th>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col0\" class=\"data row4 col0\" >0.858669</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col1\" class=\"data row4 col1\" >0.849416</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col2\" class=\"data row4 col2\" >0.009254</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col3\" class=\"data row4 col3\" >0.860431</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col4\" class=\"data row4 col4\" >0.809701</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col5\" class=\"data row4 col5\" >0.857357</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col6\" class=\"data row4 col6\" >0.863135</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col7\" class=\"data row4 col7\" >0.852907</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col8\" class=\"data row4 col8\" >0.860892</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col9\" class=\"data row4 col9\" >0.868485</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col10\" class=\"data row4 col10\" >0.836538</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col11\" class=\"data row4 col11\" >0.854167</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row4_col12\" class=\"data row4 col12\" >0.876812</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202level0_row5\" class=\"row_heading level0 row5\" >3 <= text wordCnt < 10</th>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col0\" class=\"data row5 col0\" >0.783808</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col1\" class=\"data row5 col1\" >0.764802</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col2\" class=\"data row5 col2\" >0.019006</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col3\" class=\"data row5 col3\" >0.789486</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col4\" class=\"data row5 col4\" >0.758005</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col5\" class=\"data row5 col5\" >0.781952</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col6\" class=\"data row5 col6\" >0.767883</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col7\" class=\"data row5 col7\" >0.778450</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col8\" class=\"data row5 col8\" >0.770040</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col9\" class=\"data row5 col9\" >0.785605</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col10\" class=\"data row5 col10\" >0.756443</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col11\" class=\"data row5 col11\" >0.783548</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row5_col12\" class=\"data row5 col12\" >0.771640</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202level0_row6\" class=\"row_heading level0 row6\" >10 <= text wordCnt < 30</th>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col0\" class=\"data row6 col0\" >0.692851</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col1\" class=\"data row6 col1\" >0.666583</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col2\" class=\"data row6 col2\" >0.026267</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col3\" class=\"data row6 col3\" >0.698183</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col4\" class=\"data row6 col4\" >0.679244</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col5\" class=\"data row6 col5\" >0.694128</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col6\" class=\"data row6 col6\" >0.658758</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col7\" class=\"data row6 col7\" >0.685373</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col8\" class=\"data row6 col8\" >0.665855</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col9\" class=\"data row6 col9\" >0.693477</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col10\" class=\"data row6 col10\" >0.663384</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col11\" class=\"data row6 col11\" >0.693091</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row6_col12\" class=\"data row6 col12\" >0.665675</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202level0_row7\" class=\"row_heading level0 row7\" >30 <= text wordCnt < 100</th>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col0\" class=\"data row7 col0\" >0.651725</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col1\" class=\"data row7 col1\" >0.638358</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col2\" class=\"data row7 col2\" >0.013367</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col3\" class=\"data row7 col3\" >0.645364</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col4\" class=\"data row7 col4\" >0.695229</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col5\" class=\"data row7 col5\" >0.693150</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col6\" class=\"data row7 col6\" >0.485683</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col7\" class=\"data row7 col7\" >0.672265</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col8\" class=\"data row7 col8\" >0.561538</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col9\" class=\"data row7 col9\" >0.627974</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col10\" class=\"data row7 col10\" >0.633724</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col11\" class=\"data row7 col11\" >0.619872</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row7_col12\" class=\"data row7 col12\" >0.815618</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202level0_row8\" class=\"row_heading level0 row8\" >0 <= selected text wordCnt < 3</th>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col0\" class=\"data row8 col0\" >0.662523</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col1\" class=\"data row8 col1\" >0.622074</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col2\" class=\"data row8 col2\" >0.040449</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col3\" class=\"data row8 col3\" >0.672189</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col4\" class=\"data row8 col4\" >0.630021</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col5\" class=\"data row8 col5\" >0.657219</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col6\" class=\"data row8 col6\" >0.605825</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col7\" class=\"data row8 col7\" >0.639666</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col8\" class=\"data row8 col8\" >0.602732</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col9\" class=\"data row8 col9\" >0.674516</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col10\" class=\"data row8 col10\" >0.626353</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col11\" class=\"data row8 col11\" >0.669028</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row8_col12\" class=\"data row8 col12\" >0.645441</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202level0_row9\" class=\"row_heading level0 row9\" >3 <= selected text wordCnt < 10</th>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col0\" class=\"data row9 col0\" >0.685977</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col1\" class=\"data row9 col1\" >0.665825</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col2\" class=\"data row9 col2\" >0.020152</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col3\" class=\"data row9 col3\" >0.690364</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col4\" class=\"data row9 col4\" >0.668893</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col5\" class=\"data row9 col5\" >0.689889</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col6\" class=\"data row9 col6\" >0.666695</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col7\" class=\"data row9 col7\" >0.692531</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col8\" class=\"data row9 col8\" >0.679165</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col9\" class=\"data row9 col9\" >0.677904</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col10\" class=\"data row9 col10\" >0.655758</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col11\" class=\"data row9 col11\" >0.679197</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row9_col12\" class=\"data row9 col12\" >0.658615</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202level0_row10\" class=\"row_heading level0 row10\" >10 <= selected text wordCnt < 30</th>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col0\" class=\"data row10 col0\" >0.858459</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col1\" class=\"data row10 col1\" >0.851662</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col2\" class=\"data row10 col2\" >0.006798</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col3\" class=\"data row10 col3\" >0.859798</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col4\" class=\"data row10 col4\" >0.855060</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col5\" class=\"data row10 col5\" >0.861627</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col6\" class=\"data row10 col6\" >0.851384</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col7\" class=\"data row10 col7\" >0.855930</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col8\" class=\"data row10 col8\" >0.858599</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col9\" class=\"data row10 col9\" >0.856155</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col10\" class=\"data row10 col10\" >0.849333</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col11\" class=\"data row10 col11\" >0.858788</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row10_col12\" class=\"data row10 col12\" >0.843932</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202level0_row11\" class=\"row_heading level0 row11\" >30 <= selected text wordCnt < 100</th>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col0\" class=\"data row11 col0\" >0.884550</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col1\" class=\"data row11 col1\" >0.888328</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col2\" class=\"data row11 col2\" >-0.003778</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col3\" class=\"data row11 col3\" >0.908367</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col4\" class=\"data row11 col4\" >0.829779</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col5\" class=\"data row11 col5\" >0.912773</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col6\" class=\"data row11 col6\" >0.796481</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col7\" class=\"data row11 col7\" >0.865497</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col8\" class=\"data row11 col8\" >0.909091</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col9\" class=\"data row11 col9\" >0.882206</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col10\" class=\"data row11 col10\" >0.912308</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col11\" class=\"data row11 col11\" >0.853906</td>\n",
       "                        <td id=\"T_11a04af0_9b13_11ea_89fb_0242ac130202row11_col12\" class=\"data row11 col12\" >0.993980</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7fbfbb21ad90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "display(evaluate_model_kfold(k_fold, config['N_FOLDS']).style.applymap(color_map))"
   ]
  },
  {
   "cell_type": "markdown",
   "execution_count": null,
   "metadata": {},
   "source": [
    "# Model evaluation (best model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "_kg_hide-input": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col0 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col1 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col2 {\n",
       "            color:  red;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col3 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col4 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col5 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col6 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col7 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col8 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col9 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col10 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col11 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col12 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col0 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col1 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col2 {\n",
       "            color:  red;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col3 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col4 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col5 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col6 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col7 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col8 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col9 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col10 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col11 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col12 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col0 {\n",
       "            color:  green;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col1 {\n",
       "            color:  green;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col2 {\n",
       "            color:  red;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col3 {\n",
       "            color:  green;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col4 {\n",
       "            color:  green;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col5 {\n",
       "            color:  green;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col6 {\n",
       "            color:  green;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col7 {\n",
       "            color:  green;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col8 {\n",
       "            color:  green;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col9 {\n",
       "            color:  green;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col10 {\n",
       "            color:  green;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col11 {\n",
       "            color:  green;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col12 {\n",
       "            color:  green;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col0 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col1 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col2 {\n",
       "            color:  red;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col3 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col4 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col5 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col6 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col7 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col8 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col9 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col10 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col11 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col12 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col0 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col1 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col2 {\n",
       "            color:  red;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col3 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col4 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col5 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col6 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col7 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col8 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col9 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col10 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col11 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col12 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col0 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col1 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col2 {\n",
       "            color:  red;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col3 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col4 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col5 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col6 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col7 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col8 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col9 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col10 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col11 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col12 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col0 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col1 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col2 {\n",
       "            color:  red;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col3 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col4 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col5 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col6 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col7 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col8 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col9 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col10 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col11 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col12 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col0 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col1 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col2 {\n",
       "            color:  red;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col3 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col4 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col5 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col6 {\n",
       "            color:  orange;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col7 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col8 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col9 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col10 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col11 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col12 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col0 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col1 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col2 {\n",
       "            color:  red;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col3 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col4 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col5 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col6 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col7 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col8 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col9 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col10 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col11 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col12 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col0 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col1 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col2 {\n",
       "            color:  red;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col3 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col4 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col5 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col6 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col7 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col8 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col9 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col10 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col11 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col12 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col0 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col1 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col2 {\n",
       "            color:  red;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col3 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col4 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col5 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col6 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col7 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col8 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col9 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col10 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col11 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col12 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col0 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col1 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col2 {\n",
       "            color:  red;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col3 {\n",
       "            color:  green;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col4 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col5 {\n",
       "            color:  green;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col6 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col7 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col8 {\n",
       "            color:  green;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col9 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col10 {\n",
       "            color:  green;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col11 {\n",
       "            color:  black;\n",
       "        }    #T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col12 {\n",
       "            color:  green;\n",
       "        }</style><table id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >Train</th>        <th class=\"col_heading level0 col1\" >Valid</th>        <th class=\"col_heading level0 col2\" >Var</th>        <th class=\"col_heading level0 col3\" >Train_fold_1</th>        <th class=\"col_heading level0 col4\" >Valid_fold_1</th>        <th class=\"col_heading level0 col5\" >Train_fold_2</th>        <th class=\"col_heading level0 col6\" >Valid_fold_2</th>        <th class=\"col_heading level0 col7\" >Train_fold_3</th>        <th class=\"col_heading level0 col8\" >Valid_fold_3</th>        <th class=\"col_heading level0 col9\" >Train_fold_4</th>        <th class=\"col_heading level0 col10\" >Valid_fold_4</th>        <th class=\"col_heading level0 col11\" >Train_fold_5</th>        <th class=\"col_heading level0 col12\" >Valid_fold_5</th>    </tr>    <tr>        <th class=\"index_name level0\" >Metric/Jaccard</th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202level0_row0\" class=\"row_heading level0 row0\" >Overall</th>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col0\" class=\"data row0 col0\" >0.725405</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col1\" class=\"data row0 col1\" >0.702960</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col2\" class=\"data row0 col2\" >0.022446</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col3\" class=\"data row0 col3\" >0.728912</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col4\" class=\"data row0 col4\" >0.706852</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col5\" class=\"data row0 col5\" >0.724942</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col6\" class=\"data row0 col6\" >0.698109</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col7\" class=\"data row0 col7\" >0.721287</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col8\" class=\"data row0 col8\" >0.705776</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col9\" class=\"data row0 col9\" >0.726654</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col10\" class=\"data row0 col10\" >0.698889</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col11\" class=\"data row0 col11\" >0.725230</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row0_col12\" class=\"data row0 col12\" >0.705172</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202level0_row1\" class=\"row_heading level0 row1\" >Sentiment negative</th>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col0\" class=\"data row1 col0\" >0.558537</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col1\" class=\"data row1 col1\" >0.519918</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col2\" class=\"data row1 col2\" >0.038619</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col3\" class=\"data row1 col3\" >0.568729</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col4\" class=\"data row1 col4\" >0.513151</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col5\" class=\"data row1 col5\" >0.555500</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col6\" class=\"data row1 col6\" >0.514153</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col7\" class=\"data row1 col7\" >0.553001</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col8\" class=\"data row1 col8\" >0.524660</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col9\" class=\"data row1 col9\" >0.557200</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col10\" class=\"data row1 col10\" >0.523034</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col11\" class=\"data row1 col11\" >0.558253</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row1_col12\" class=\"data row1 col12\" >0.524591</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202level0_row2\" class=\"row_heading level0 row2\" >Sentiment neutral</th>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col0\" class=\"data row2 col0\" >0.973858</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col1\" class=\"data row2 col1\" >0.972797</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col2\" class=\"data row2 col2\" >0.001060</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col3\" class=\"data row2 col3\" >0.974117</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col4\" class=\"data row2 col4\" >0.973879</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col5\" class=\"data row2 col5\" >0.974460</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col6\" class=\"data row2 col6\" >0.973842</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col7\" class=\"data row2 col7\" >0.971861</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col8\" class=\"data row2 col8\" >0.971087</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col9\" class=\"data row2 col9\" >0.974490</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col10\" class=\"data row2 col10\" >0.970587</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col11\" class=\"data row2 col11\" >0.974360</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row2_col12\" class=\"data row2 col12\" >0.974590</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202level0_row3\" class=\"row_heading level0 row3\" >Sentiment positive</th>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col0\" class=\"data row3 col0\" >0.554857</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col1\" class=\"data row3 col1\" >0.519374</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col2\" class=\"data row3 col2\" >0.035483</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col3\" class=\"data row3 col3\" >0.556545</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col4\" class=\"data row3 col4\" >0.536414</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col5\" class=\"data row3 col5\" >0.555385</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col6\" class=\"data row3 col6\" >0.507553</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col7\" class=\"data row3 col7\" >0.549253</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col8\" class=\"data row3 col8\" >0.526411</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col9\" class=\"data row3 col9\" >0.559224</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col10\" class=\"data row3 col10\" >0.506488</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col11\" class=\"data row3 col11\" >0.553879</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row3_col12\" class=\"data row3 col12\" >0.520005</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202level0_row4\" class=\"row_heading level0 row4\" >0 <= text wordCnt < 3</th>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col0\" class=\"data row4 col0\" >0.859026</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col1\" class=\"data row4 col1\" >0.852936</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col2\" class=\"data row4 col2\" >0.006090</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col3\" class=\"data row4 col3\" >0.859557</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col4\" class=\"data row4 col4\" >0.817164</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col5\" class=\"data row4 col5\" >0.858258</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col6\" class=\"data row4 col6\" >0.866446</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col7\" class=\"data row4 col7\" >0.854634</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col8\" class=\"data row4 col8\" >0.860892</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col9\" class=\"data row4 col9\" >0.869394</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col10\" class=\"data row4 col10\" >0.839744</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col11\" class=\"data row4 col11\" >0.853286</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row4_col12\" class=\"data row4 col12\" >0.880435</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202level0_row5\" class=\"row_heading level0 row5\" >3 <= text wordCnt < 10</th>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col0\" class=\"data row5 col0\" >0.782412</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col1\" class=\"data row5 col1\" >0.763823</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col2\" class=\"data row5 col2\" >0.018589</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col3\" class=\"data row5 col3\" >0.786823</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col4\" class=\"data row5 col4\" >0.757730</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col5\" class=\"data row5 col5\" >0.778563</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col6\" class=\"data row5 col6\" >0.764677</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col7\" class=\"data row5 col7\" >0.779603</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col8\" class=\"data row5 col8\" >0.771953</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col9\" class=\"data row5 col9\" >0.784377</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col10\" class=\"data row5 col10\" >0.755627</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col11\" class=\"data row5 col11\" >0.782694</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row5_col12\" class=\"data row5 col12\" >0.769127</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202level0_row6\" class=\"row_heading level0 row6\" >10 <= text wordCnt < 30</th>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col0\" class=\"data row6 col0\" >0.690400</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col1\" class=\"data row6 col1\" >0.665241</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col2\" class=\"data row6 col2\" >0.025160</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col3\" class=\"data row6 col3\" >0.693825</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col4\" class=\"data row6 col4\" >0.674819</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col5\" class=\"data row6 col5\" >0.691214</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col6\" class=\"data row6 col6\" >0.659258</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col7\" class=\"data row6 col7\" >0.685011</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col8\" class=\"data row6 col8\" >0.667136</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col9\" class=\"data row6 col9\" >0.691542</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col10\" class=\"data row6 col10\" >0.661965</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col11\" class=\"data row6 col11\" >0.690411</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row6_col12\" class=\"data row6 col12\" >0.663025</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202level0_row7\" class=\"row_heading level0 row7\" >30 <= text wordCnt < 100</th>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col0\" class=\"data row7 col0\" >0.648452</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col1\" class=\"data row7 col1\" >0.634015</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col2\" class=\"data row7 col2\" >0.014437</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col3\" class=\"data row7 col3\" >0.643254</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col4\" class=\"data row7 col4\" >0.677685</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col5\" class=\"data row7 col5\" >0.683504</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col6\" class=\"data row7 col6\" >0.483656</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col7\" class=\"data row7 col7\" >0.666038</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col8\" class=\"data row7 col8\" >0.559848</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col9\" class=\"data row7 col9\" >0.623545</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col10\" class=\"data row7 col10\" >0.660510</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col11\" class=\"data row7 col11\" >0.625917</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row7_col12\" class=\"data row7 col12\" >0.788374</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202level0_row8\" class=\"row_heading level0 row8\" >0 <= selected text wordCnt < 3</th>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col0\" class=\"data row8 col0\" >0.657574</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col1\" class=\"data row8 col1\" >0.617156</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col2\" class=\"data row8 col2\" >0.040418</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col3\" class=\"data row8 col3\" >0.674756</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col4\" class=\"data row8 col4\" >0.632257</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col5\" class=\"data row8 col5\" >0.651354</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col6\" class=\"data row8 col6\" >0.603315</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col7\" class=\"data row8 col7\" >0.659297</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col8\" class=\"data row8 col8\" >0.623980</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col9\" class=\"data row8 col9\" >0.647748</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col10\" class=\"data row8 col10\" >0.599855</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col11\" class=\"data row8 col11\" >0.654717</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row8_col12\" class=\"data row8 col12\" >0.626375</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202level0_row9\" class=\"row_heading level0 row9\" >3 <= selected text wordCnt < 10</th>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col0\" class=\"data row9 col0\" >0.685903</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col1\" class=\"data row9 col1\" >0.666729</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col2\" class=\"data row9 col2\" >0.019174</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col3\" class=\"data row9 col3\" >0.681051</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col4\" class=\"data row9 col4\" >0.661342</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col5\" class=\"data row9 col5\" >0.688392</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col6\" class=\"data row9 col6\" >0.666611</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col7\" class=\"data row9 col7\" >0.675802</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col8\" class=\"data row9 col8\" >0.661878</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col9\" class=\"data row9 col9\" >0.696961</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col10\" class=\"data row9 col10\" >0.676176</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col11\" class=\"data row9 col11\" >0.687309</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row9_col12\" class=\"data row9 col12\" >0.667639</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202level0_row10\" class=\"row_heading level0 row10\" >10 <= selected text wordCnt < 30</th>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col0\" class=\"data row10 col0\" >0.857545</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col1\" class=\"data row10 col1\" >0.852132</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col2\" class=\"data row10 col2\" >0.005413</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col3\" class=\"data row10 col3\" >0.855809</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col4\" class=\"data row10 col4\" >0.851995</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col5\" class=\"data row10 col5\" >0.860091</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col6\" class=\"data row10 col6\" >0.852238</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col7\" class=\"data row10 col7\" >0.854410</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col8\" class=\"data row10 col8\" >0.859350</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col9\" class=\"data row10 col9\" >0.858558</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col10\" class=\"data row10 col10\" >0.850896</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col11\" class=\"data row10 col11\" >0.858856</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row10_col12\" class=\"data row10 col12\" >0.846181</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202level0_row11\" class=\"row_heading level0 row11\" >30 <= selected text wordCnt < 100</th>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col0\" class=\"data row11 col0\" >0.884776</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col1\" class=\"data row11 col1\" >0.885736</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col2\" class=\"data row11 col2\" >-0.000960</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col3\" class=\"data row11 col3\" >0.908367</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col4\" class=\"data row11 col4\" >0.829779</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col5\" class=\"data row11 col5\" >0.912773</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col6\" class=\"data row11 col6\" >0.789678</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col7\" class=\"data row11 col7\" >0.863193</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col8\" class=\"data row11 col8\" >0.909091</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col9\" class=\"data row11 col9\" >0.884900</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col10\" class=\"data row11 col10\" >0.912308</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col11\" class=\"data row11 col11\" >0.854647</td>\n",
       "                        <td id=\"T_158a5e26_9b13_11ea_89fb_0242ac130202row11_col12\" class=\"data row11 col12\" >0.987826</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7fc41ba35b10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "display(evaluate_model_kfold(k_fold_best, config['N_FOLDS']).style.applymap(color_map))"
   ]
  },
  {
   "cell_type": "markdown",
   "execution_count": null,
   "metadata": {},
   "source": [
    "# Visualize predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "_kg_hide-input": true
   },
   "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>text</th>\n",
       "      <th>selected_text</th>\n",
       "      <th>sentiment</th>\n",
       "      <th>jaccard</th>\n",
       "      <th>text_tokenCnt</th>\n",
       "      <th>selected_text_tokenCnt</th>\n",
       "      <th>prediction_fold_1</th>\n",
       "      <th>jaccard_fold_1</th>\n",
       "      <th>prediction_fold_2</th>\n",
       "      <th>jaccard_fold_2</th>\n",
       "      <th>prediction_fold_3</th>\n",
       "      <th>jaccard_fold_3</th>\n",
       "      <th>prediction_fold_4</th>\n",
       "      <th>jaccard_fold_4</th>\n",
       "      <th>prediction_fold_5</th>\n",
       "      <th>jaccard_fold_5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>i`d have responded, if i were going</td>\n",
       "      <td>i`d have responded, if i were going</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>i`d have responded, if i were going</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>i`d have responded, if i were going</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>i`d have responded, if i were going</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>i`d have responded, if i were going</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>i`d have responded, if i were going</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>sooo sad i will miss you here in san diego!!!</td>\n",
       "      <td>sooo sad</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>13</td>\n",
       "      <td>3</td>\n",
       "      <td>sooo sad</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>sad</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>sad</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>sad</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>sad</td>\n",
       "      <td>0.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>my boss is bullying me...</td>\n",
       "      <td>bullying me</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>6</td>\n",
       "      <td>2</td>\n",
       "      <td>bullying me...</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>my boss is bullying me...</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>bullying me...</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>bullying me...</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>bullying me...</td>\n",
       "      <td>0.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>what interview! leave me alone</td>\n",
       "      <td>leave me alone</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>leave me alone</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>leave me alone</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>leave me alone</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>leave me alone</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>leave me alone</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>sons of ****, why couldn`t they put them on th...</td>\n",
       "      <td>sons of ****,</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.214286</td>\n",
       "      <td>17</td>\n",
       "      <td>4</td>\n",
       "      <td>****,</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>sons of ****,</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>****,</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>****,</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>****,</td>\n",
       "      <td>0.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>http://www.dothebouncy.com/smf - some shameles...</td>\n",
       "      <td>http://www.dothebouncy.com/smf - some shameles...</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>http://www.dothebouncy.com/smf - some shameles...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>http://www.dothebouncy.com/smf - some shameles...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>http://www.dothebouncy.com/smf - some shameles...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>http://www.dothebouncy.com/smf - some shameles...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>some shameless plugging for the best rangers f...</td>\n",
       "      <td>0.833333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2am feedings for the baby are fun when he is a...</td>\n",
       "      <td>fun</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.071429</td>\n",
       "      <td>17</td>\n",
       "      <td>1</td>\n",
       "      <td>fun</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>fun</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>fun</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>fun</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>fun</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>soooo high</td>\n",
       "      <td>soooo high</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>soooo high</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>soooo high</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>soooo high</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>soooo high</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>soooo high</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>both of you</td>\n",
       "      <td>both of you</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>both of you</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>both of you</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>both of you</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>both of you</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>both of you</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>journey!? wow... u just became cooler.  hehe.....</td>\n",
       "      <td>wow... u just became cooler.</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>19</td>\n",
       "      <td>7</td>\n",
       "      <td>wow... u just became cooler.</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>wow...</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>wow...</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>wow...</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>wow...</td>\n",
       "      <td>0.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>as much as i love to be hopeful, i reckon the ...</td>\n",
       "      <td>as much as i love to be hopeful, i reckon the ...</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>as much as i love to be hopeful, i reckon the ...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>as much as i love to be hopeful, i reckon the ...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>as much as i love to be hopeful, i reckon the ...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>as much as i love to be hopeful, i reckon the ...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>as much as i love to be hopeful, i reckon the ...</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>i really really like the song love story by ta...</td>\n",
       "      <td>like</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>12</td>\n",
       "      <td>1</td>\n",
       "      <td>i really really like</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>i really really like</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>i really really like</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>like</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>i really really like</td>\n",
       "      <td>0.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>my sharpie is running dangerously low on ink</td>\n",
       "      <td>dangerously</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.125000</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>dangerously low</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>dangerously low</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>dangerously low on ink</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>dangerously low on ink</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>dangerously low on ink</td>\n",
       "      <td>0.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>i want to go to music tonight but i lost my vo...</td>\n",
       "      <td>lost</td>\n",
       "      <td>negative</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>13</td>\n",
       "      <td>1</td>\n",
       "      <td>lost my voice.</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>lost</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>lost</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>lost my voice.</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>lost</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>test test from the lg env2</td>\n",
       "      <td>test test from the lg env2</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>test test from the lg env2</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>test test from the lg env2</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>test test from the lg env2</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>test test from the lg env2</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>test test from the lg env2</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 text  \\\n",
       "0                 i`d have responded, if i were going   \n",
       "1       sooo sad i will miss you here in san diego!!!   \n",
       "2                           my boss is bullying me...   \n",
       "3                      what interview! leave me alone   \n",
       "4   sons of ****, why couldn`t they put them on th...   \n",
       "5   http://www.dothebouncy.com/smf - some shameles...   \n",
       "6   2am feedings for the baby are fun when he is a...   \n",
       "7                                          soooo high   \n",
       "8                                         both of you   \n",
       "9   journey!? wow... u just became cooler.  hehe.....   \n",
       "10  as much as i love to be hopeful, i reckon the ...   \n",
       "11  i really really like the song love story by ta...   \n",
       "12       my sharpie is running dangerously low on ink   \n",
       "13  i want to go to music tonight but i lost my vo...   \n",
       "14                         test test from the lg env2   \n",
       "\n",
       "                                        selected_text sentiment   jaccard  \\\n",
       "0                 i`d have responded, if i were going   neutral  1.000000   \n",
       "1                                            sooo sad  negative  0.200000   \n",
       "2                                         bullying me  negative  0.166667   \n",
       "3                                      leave me alone  negative  0.600000   \n",
       "4                                       sons of ****,  negative  0.214286   \n",
       "5   http://www.dothebouncy.com/smf - some shameles...   neutral  1.000000   \n",
       "6                                                 fun  positive  0.071429   \n",
       "7                                          soooo high   neutral  1.000000   \n",
       "8                                         both of you   neutral  1.000000   \n",
       "9                        wow... u just became cooler.  positive  0.500000   \n",
       "10  as much as i love to be hopeful, i reckon the ...   neutral  1.000000   \n",
       "11                                               like  positive  0.100000   \n",
       "12                                        dangerously  negative  0.125000   \n",
       "13                                               lost  negative  0.100000   \n",
       "14                         test test from the lg env2   neutral  1.000000   \n",
       "\n",
       "    text_tokenCnt  selected_text_tokenCnt  \\\n",
       "0              10                      10   \n",
       "1              13                       3   \n",
       "2               6                       2   \n",
       "3               6                       3   \n",
       "4              17                       4   \n",
       "5              27                      27   \n",
       "6              17                       1   \n",
       "7               3                       3   \n",
       "8               3                       3   \n",
       "9              19                       7   \n",
       "10             27                      27   \n",
       "11             12                       1   \n",
       "12              9                       1   \n",
       "13             13                       1   \n",
       "14              8                       8   \n",
       "\n",
       "                                    prediction_fold_1  jaccard_fold_1  \\\n",
       "0                 i`d have responded, if i were going        1.000000   \n",
       "1                                            sooo sad        1.000000   \n",
       "2                                      bullying me...        0.333333   \n",
       "3                                      leave me alone        1.000000   \n",
       "4                                               ****,        0.333333   \n",
       "5   http://www.dothebouncy.com/smf - some shameles...        1.000000   \n",
       "6                                                 fun        1.000000   \n",
       "7                                          soooo high        1.000000   \n",
       "8                                         both of you        1.000000   \n",
       "9                        wow... u just became cooler.        1.000000   \n",
       "10  as much as i love to be hopeful, i reckon the ...        1.000000   \n",
       "11                               i really really like        0.333333   \n",
       "12                                    dangerously low        0.500000   \n",
       "13                                     lost my voice.        0.333333   \n",
       "14                         test test from the lg env2        1.000000   \n",
       "\n",
       "                                    prediction_fold_2  jaccard_fold_2  \\\n",
       "0                 i`d have responded, if i were going        1.000000   \n",
       "1                                                 sad        0.500000   \n",
       "2                           my boss is bullying me...        0.166667   \n",
       "3                                      leave me alone        1.000000   \n",
       "4                                       sons of ****,        1.000000   \n",
       "5   http://www.dothebouncy.com/smf - some shameles...        1.000000   \n",
       "6                                                 fun        1.000000   \n",
       "7                                          soooo high        1.000000   \n",
       "8                                         both of you        1.000000   \n",
       "9                                              wow...        0.200000   \n",
       "10  as much as i love to be hopeful, i reckon the ...        1.000000   \n",
       "11                               i really really like        0.333333   \n",
       "12                                    dangerously low        0.500000   \n",
       "13                                               lost        1.000000   \n",
       "14                         test test from the lg env2        1.000000   \n",
       "\n",
       "                                    prediction_fold_3  jaccard_fold_3  \\\n",
       "0                 i`d have responded, if i were going        1.000000   \n",
       "1                                                 sad        0.500000   \n",
       "2                                      bullying me...        0.333333   \n",
       "3                                      leave me alone        1.000000   \n",
       "4                                               ****,        0.333333   \n",
       "5   http://www.dothebouncy.com/smf - some shameles...        1.000000   \n",
       "6                                                 fun        1.000000   \n",
       "7                                          soooo high        1.000000   \n",
       "8                                         both of you        1.000000   \n",
       "9                                              wow...        0.200000   \n",
       "10  as much as i love to be hopeful, i reckon the ...        1.000000   \n",
       "11                               i really really like        0.333333   \n",
       "12                             dangerously low on ink        0.250000   \n",
       "13                                               lost        1.000000   \n",
       "14                         test test from the lg env2        1.000000   \n",
       "\n",
       "                                    prediction_fold_4  jaccard_fold_4  \\\n",
       "0                 i`d have responded, if i were going        1.000000   \n",
       "1                                                 sad        0.500000   \n",
       "2                                      bullying me...        0.333333   \n",
       "3                                      leave me alone        1.000000   \n",
       "4                                               ****,        0.333333   \n",
       "5   http://www.dothebouncy.com/smf - some shameles...        1.000000   \n",
       "6                                                 fun        1.000000   \n",
       "7                                          soooo high        1.000000   \n",
       "8                                         both of you        1.000000   \n",
       "9                                              wow...        0.200000   \n",
       "10  as much as i love to be hopeful, i reckon the ...        1.000000   \n",
       "11                                               like        1.000000   \n",
       "12                             dangerously low on ink        0.250000   \n",
       "13                                     lost my voice.        0.333333   \n",
       "14                         test test from the lg env2        1.000000   \n",
       "\n",
       "                                    prediction_fold_5  jaccard_fold_5  \n",
       "0                 i`d have responded, if i were going        1.000000  \n",
       "1                                                 sad        0.500000  \n",
       "2                                      bullying me...        0.333333  \n",
       "3                                      leave me alone        1.000000  \n",
       "4                                               ****,        0.333333  \n",
       "5   some shameless plugging for the best rangers f...        0.833333  \n",
       "6                                                 fun        1.000000  \n",
       "7                                          soooo high        1.000000  \n",
       "8                                         both of you        1.000000  \n",
       "9                                              wow...        0.200000  \n",
       "10  as much as i love to be hopeful, i reckon the ...        1.000000  \n",
       "11                               i really really like        0.333333  \n",
       "12                             dangerously low on ink        0.250000  \n",
       "13                                               lost        1.000000  \n",
       "14                         test test from the lg env2        1.000000  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "display(k_fold[[c for c in k_fold.columns if not (c.startswith('textID') or \n",
    "                                                  c.startswith('text_len') or \n",
    "                                                  c.startswith('selected_text_len') or \n",
    "                                                  c.startswith('text_wordCnt') or \n",
    "                                                  c.startswith('selected_text_wordCnt') or \n",
    "                                                  c.startswith('fold_') or \n",
    "                                                  c.startswith('start_fold_') or \n",
    "                                                  c.startswith('end_fold_'))]].head(15))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}