Time series forecasting - Darts models-Copy1.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "2c02fd89",
   "metadata": {},
   "source": [
    "# Time series forecasting using the darts library"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f5a62844",
   "metadata": {},
   "source": [
    "In this notebook we will forecast the price of bitcoin using temporal fusion transformer, temporal convolution networks and the NBEATS model. Here all implementation is done using the darts library, in easy to use library which works a lot like scikit learn.\n",
    "\n",
    "For more details, I encourage you to see the documentation - https://unit8co.github.io/darts/README.html"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b0268d5d",
   "metadata": {},
   "source": [
    "**Team:**\n",
    "* Sameer Mahajan\n",
    "* Geethika Marru\n",
    "* Harsh Deokuliar\n",
    "\n",
    "**Course:** IE-575 – Predictive Analytics (Spring II, 2022)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "4f11095e",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-04-27 17:43:19 prophet.plot ERROR: Importing plotly failed. Interactive plots will not work.\n"
     ]
    }
   ],
   "source": [
    "#importing necessary libraries\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "from darts import TimeSeries\n",
    "from darts.dataprocessing.transformers import Scaler\n",
    "from darts.models import TFTModel,NBEATSModel, TCNModel,Prophet, NHiTS,RNNModel,TransformerModel\n",
    "from darts.metrics import mape, rmse, r2_score,mae\n",
    "from darts.utils.statistics import check_seasonality, plot_acf,extract_trend_and_seasonality\n",
    "from darts.utils.timeseries_generation import datetime_attribute_timeseries\n",
    "from darts.utils.utils import ModelMode, SeasonalityMode, TrendMode\n",
    "\n",
    "from darts.datasets import SunspotsDataset, AirPassengersDataset\n",
    "\n",
    "import sys\n",
    "import time\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "import logging\n",
    "logging.disable(logging.CRITICAL)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "93f3979d",
   "metadata": {},
   "outputs": [],
   "source": [
    "#loading the bitcoin prices dataset with sentiment scores\n",
    "df = pd.read_csv(\"final_price_dataset.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "b3193e54",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date</th>\n",
       "      <th>btc_inflation_rate</th>\n",
       "      <th>mining_difficulty</th>\n",
       "      <th>hash_rate</th>\n",
       "      <th>trans_count</th>\n",
       "      <th>Price</th>\n",
       "      <th>Vol.</th>\n",
       "      <th>Change %</th>\n",
       "      <th>total_tweets</th>\n",
       "      <th>Google_trends</th>\n",
       "      <th>new_senti</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2018-01-01</td>\n",
       "      <td>0.043336</td>\n",
       "      <td>8.290000e+21</td>\n",
       "      <td>1.485130e+19</td>\n",
       "      <td>241601</td>\n",
       "      <td>13436.40113</td>\n",
       "      <td>78430</td>\n",
       "      <td>-2.93</td>\n",
       "      <td>62044.0</td>\n",
       "      <td>44</td>\n",
       "      <td>0.189724</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2018-01-02</td>\n",
       "      <td>0.047594</td>\n",
       "      <td>8.290000e+21</td>\n",
       "      <td>1.644330e+19</td>\n",
       "      <td>340809</td>\n",
       "      <td>14732.45336</td>\n",
       "      <td>137730</td>\n",
       "      <td>9.74</td>\n",
       "      <td>77723.0</td>\n",
       "      <td>54</td>\n",
       "      <td>1.243548</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2018-01-03</td>\n",
       "      <td>0.043609</td>\n",
       "      <td>8.290000e+21</td>\n",
       "      <td>1.497180e+19</td>\n",
       "      <td>395806</td>\n",
       "      <td>15068.48685</td>\n",
       "      <td>106540</td>\n",
       "      <td>2.73</td>\n",
       "      <td>79086.0</td>\n",
       "      <td>55</td>\n",
       "      <td>0.791414</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2018-01-04</td>\n",
       "      <td>0.046728</td>\n",
       "      <td>8.290000e+21</td>\n",
       "      <td>1.613350e+19</td>\n",
       "      <td>424840</td>\n",
       "      <td>15150.04439</td>\n",
       "      <td>110970</td>\n",
       "      <td>0.15</td>\n",
       "      <td>74534.0</td>\n",
       "      <td>51</td>\n",
       "      <td>1.228296</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2018-01-05</td>\n",
       "      <td>0.039635</td>\n",
       "      <td>8.290000e+21</td>\n",
       "      <td>1.372410e+19</td>\n",
       "      <td>342564</td>\n",
       "      <td>16929.99358</td>\n",
       "      <td>141960</td>\n",
       "      <td>11.69</td>\n",
       "      <td>76404.0</td>\n",
       "      <td>50</td>\n",
       "      <td>1.762493</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Date  btc_inflation_rate  mining_difficulty     hash_rate  \\\n",
       "0  2018-01-01            0.043336       8.290000e+21  1.485130e+19   \n",
       "1  2018-01-02            0.047594       8.290000e+21  1.644330e+19   \n",
       "2  2018-01-03            0.043609       8.290000e+21  1.497180e+19   \n",
       "3  2018-01-04            0.046728       8.290000e+21  1.613350e+19   \n",
       "4  2018-01-05            0.039635       8.290000e+21  1.372410e+19   \n",
       "\n",
       "   trans_count        Price    Vol.  Change %  total_tweets  Google_trends  \\\n",
       "0       241601  13436.40113   78430     -2.93       62044.0             44   \n",
       "1       340809  14732.45336  137730      9.74       77723.0             54   \n",
       "2       395806  15068.48685  106540      2.73       79086.0             55   \n",
       "3       424840  15150.04439  110970      0.15       74534.0             51   \n",
       "4       342564  16929.99358  141960     11.69       76404.0             50   \n",
       "\n",
       "   new_senti  \n",
       "0   0.189724  \n",
       "1   1.243548  \n",
       "2   0.791414  \n",
       "3   1.228296  \n",
       "4   1.762493  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.drop(\"Unnamed: 0\",axis=1,inplace=True)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c4d0bec1",
   "metadata": {},
   "source": [
    "Here is the data description of the attributes used.\n",
    "\n",
    "1. Bitcoin inflation rate: The percentage of new coins issued, divided by the current supply.\n",
    "\n",
    "2. Mining Difficulty: The current estimated number of hashes required to mine a block.\n",
    "\n",
    "3. Hash rate: The average estimated number of hashes per second produced by the miners in the network.\n",
    "\n",
    "4. Transaction count: Bitcoin transactions per day.\n",
    "\n",
    "5. Vol. : Trading Volume of bitcoin\n",
    "\n",
    "6. Change %: % change since last day\n",
    "\n",
    "7. Total tweets: No of tweets per day related to bitcoin\n",
    "\n",
    "8. Google trends: Google search trend score for bitcoin\n",
    "\n",
    "9. new_senti : Newly generated sentiment scores."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "3948adda",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 692 entries, 0 to 691\n",
      "Data columns (total 11 columns):\n",
      " #   Column              Non-Null Count  Dtype  \n",
      "---  ------              --------------  -----  \n",
      " 0   Date                692 non-null    object \n",
      " 1   btc_inflation_rate  692 non-null    float64\n",
      " 2   mining_difficulty   692 non-null    float64\n",
      " 3   hash_rate           692 non-null    float64\n",
      " 4   trans_count         692 non-null    int64  \n",
      " 5   Price               692 non-null    float64\n",
      " 6   Vol.                692 non-null    int64  \n",
      " 7   Change %            692 non-null    float64\n",
      " 8   total_tweets        692 non-null    float64\n",
      " 9   Google_trends       692 non-null    int64  \n",
      " 10  new_senti           692 non-null    float64\n",
      "dtypes: float64(7), int64(3), object(1)\n",
      "memory usage: 59.6+ KB\n"
     ]
    },
    {
     "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>btc_inflation_rate</th>\n",
       "      <th>mining_difficulty</th>\n",
       "      <th>hash_rate</th>\n",
       "      <th>trans_count</th>\n",
       "      <th>Price</th>\n",
       "      <th>Vol.</th>\n",
       "      <th>Change %</th>\n",
       "      <th>total_tweets</th>\n",
       "      <th>Google_trends</th>\n",
       "      <th>new_senti</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2018-01-01</th>\n",
       "      <td>0.043336</td>\n",
       "      <td>8.290000e+21</td>\n",
       "      <td>1.485130e+19</td>\n",
       "      <td>241601</td>\n",
       "      <td>13436.40113</td>\n",
       "      <td>78430</td>\n",
       "      <td>-2.93</td>\n",
       "      <td>62044.0</td>\n",
       "      <td>44</td>\n",
       "      <td>0.189724</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-02</th>\n",
       "      <td>0.047594</td>\n",
       "      <td>8.290000e+21</td>\n",
       "      <td>1.644330e+19</td>\n",
       "      <td>340809</td>\n",
       "      <td>14732.45336</td>\n",
       "      <td>137730</td>\n",
       "      <td>9.74</td>\n",
       "      <td>77723.0</td>\n",
       "      <td>54</td>\n",
       "      <td>1.243548</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-03</th>\n",
       "      <td>0.043609</td>\n",
       "      <td>8.290000e+21</td>\n",
       "      <td>1.497180e+19</td>\n",
       "      <td>395806</td>\n",
       "      <td>15068.48685</td>\n",
       "      <td>106540</td>\n",
       "      <td>2.73</td>\n",
       "      <td>79086.0</td>\n",
       "      <td>55</td>\n",
       "      <td>0.791414</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-04</th>\n",
       "      <td>0.046728</td>\n",
       "      <td>8.290000e+21</td>\n",
       "      <td>1.613350e+19</td>\n",
       "      <td>424840</td>\n",
       "      <td>15150.04439</td>\n",
       "      <td>110970</td>\n",
       "      <td>0.15</td>\n",
       "      <td>74534.0</td>\n",
       "      <td>51</td>\n",
       "      <td>1.228296</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-05</th>\n",
       "      <td>0.039635</td>\n",
       "      <td>8.290000e+21</td>\n",
       "      <td>1.372410e+19</td>\n",
       "      <td>342564</td>\n",
       "      <td>16929.99358</td>\n",
       "      <td>141960</td>\n",
       "      <td>11.69</td>\n",
       "      <td>76404.0</td>\n",
       "      <td>50</td>\n",
       "      <td>1.762493</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            btc_inflation_rate  mining_difficulty     hash_rate  trans_count  \\\n",
       "Date                                                                           \n",
       "2018-01-01            0.043336       8.290000e+21  1.485130e+19       241601   \n",
       "2018-01-02            0.047594       8.290000e+21  1.644330e+19       340809   \n",
       "2018-01-03            0.043609       8.290000e+21  1.497180e+19       395806   \n",
       "2018-01-04            0.046728       8.290000e+21  1.613350e+19       424840   \n",
       "2018-01-05            0.039635       8.290000e+21  1.372410e+19       342564   \n",
       "\n",
       "                  Price    Vol.  Change %  total_tweets  Google_trends  \\\n",
       "Date                                                                     \n",
       "2018-01-01  13436.40113   78430     -2.93       62044.0             44   \n",
       "2018-01-02  14732.45336  137730      9.74       77723.0             54   \n",
       "2018-01-03  15068.48685  106540      2.73       79086.0             55   \n",
       "2018-01-04  15150.04439  110970      0.15       74534.0             51   \n",
       "2018-01-05  16929.99358  141960     11.69       76404.0             50   \n",
       "\n",
       "            new_senti  \n",
       "Date                   \n",
       "2018-01-01   0.189724  \n",
       "2018-01-02   1.243548  \n",
       "2018-01-03   0.791414  \n",
       "2018-01-04   1.228296  \n",
       "2018-01-05   1.762493  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.info()\n",
    "df['Date'] = pd.to_datetime(df[\"Date\"])\n",
    "#df.drop(\"Unnamed: 0\",axis=1,inplace=True)\n",
    "df.set_index(\"Date\",inplace=True)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "8b62a0e8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Creating a timeseries object from the darts library. It will be used for the timeseries analysis for state-of-the-art models.\n",
    "plt.figure(figsize=(10,6))\n",
    "series = TimeSeries.from_series(df[\"Price\"]) \n",
    "series.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "68c67c58",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#creating train test split\n",
    "plt.figure(figsize=(10,6))\n",
    "train, test = series.split_before(0.80)\n",
    "train.plot()\n",
    "test.plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9613f950",
   "metadata": {},
   "source": [
    "# Data Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "e975077a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#using the scalar wrapper to normalize our time series data\n",
    "plt.figure(figsize=(10,6))\n",
    "scaler = Scaler()\n",
    "train_scaled = scaler.fit_transform(train)\n",
    "\n",
    "train_scaled.plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1db5efdf",
   "metadata": {},
   "source": [
    "Next we will create the input tensor with all the attributes for time series analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "7a9bea02",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "components: Index(['Price'], dtype='object', name='component')\n",
      "duration: 691 days 00:00:00\n",
      "frequency: <Day>\n",
      "frequency: D\n",
      "has date time index? (or else, it must have an integer index): True\n",
      "deterministic: True\n",
      "univariate: True\n"
     ]
    }
   ],
   "source": [
    "# create time series object for target variable\n",
    "ts_P = TimeSeries.from_series(df[\"Price\"]) \n",
    "\n",
    "# check attributes of the time series\n",
    "print(\"components:\", ts_P.components)\n",
    "print(\"duration:\",ts_P.duration)\n",
    "print(\"frequency:\",ts_P.freq)\n",
    "print(\"frequency:\",ts_P.freq_str)\n",
    "print(\"has date time index? (or else, it must have an integer index):\",ts_P.has_datetime_index)\n",
    "print(\"deterministic:\",ts_P.is_deterministic)\n",
    "print(\"univariate:\",ts_P.is_univariate)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "163b6894",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "components (columns) of feature time series: Index(['btc_inflation_rate', 'mining_difficulty', 'hash_rate', 'trans_count',\n",
      "       'Vol.', 'Change %', 'total_tweets', 'Google_trends', 'new_senti'],\n",
      "      dtype='object', name='component')\n",
      "duration: 691 days 00:00:00\n",
      "frequency: <Day>\n",
      "frequency: D\n",
      "has date time index? (or else, it must have an integer index): True\n",
      "deterministic: True\n",
      "univariate: False\n"
     ]
    }
   ],
   "source": [
    "#adding all the covariates (external attributes other than price) in  a timeseries object\n",
    "df2 = df.drop(['Price'],axis=1)\n",
    "ts_covF = TimeSeries.from_dataframe(df2)\n",
    "\n",
    "# check attributes of the time series\n",
    "print(\"components (columns) of feature time series:\", ts_covF.components)\n",
    "print(\"duration:\",ts_covF.duration)\n",
    "print(\"frequency:\",ts_covF.freq)\n",
    "print(\"frequency:\",ts_covF.freq_str)\n",
    "print(\"has date time index? (or else, it must have an integer index):\",ts_covF.has_datetime_index)\n",
    "print(\"deterministic:\",ts_covF.is_deterministic)\n",
    "print(\"univariate:\",ts_covF.is_univariate)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "9b52d0e7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'numpy.ndarray'>\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(692, 9, 1)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#checking shape of our model.\n",
    "ar_covF = ts_covF.all_values()\n",
    "print(type(ar_covF))\n",
    "ar_covF.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "4cac8015",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pandas.core.frame.DataFrame"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# example: operating with time series objects:\n",
    "# we can also create a pandas series or dataframe from a time series object\n",
    "df_covF = ts_covF.pd_dataframe()\n",
    "type(df_covF)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "2c54e2b9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "training start: 2018-01-01 00:00:00\n",
      "training end: 2019-07-07 00:00:00\n",
      "training duration: 552 days 00:00:00\n",
      "test start: 2019-07-08 00:00:00\n",
      "test end: 2019-11-23 00:00:00\n",
      "test duration: 138 days 00:00:00\n",
      "first and last row of scaled price time series:\n"
     ]
    },
    {
     "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>component</th>\n",
       "      <th>Price</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2018-01-01</th>\n",
       "      <td>0.74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-11-23</th>\n",
       "      <td>0.30</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "component   Price\n",
       "Date             \n",
       "2018-01-01   0.74\n",
       "2019-11-23   0.30"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# train/test split and scaling of target variable\n",
    "#code contains normalized values of prices which will be fed into the model\n",
    "ts_train, ts_test = ts_P.split_after(0.8)\n",
    "print(\"training start:\", ts_train.start_time())\n",
    "print(\"training end:\", ts_train.end_time())\n",
    "print(\"training duration:\",ts_train.duration)\n",
    "print(\"test start:\", ts_test.start_time())\n",
    "print(\"test end:\", ts_test.end_time())\n",
    "print(\"test duration:\", ts_test.duration)\n",
    "\n",
    "#creating a scalar obejct to normalize the data\n",
    "scalerP = Scaler()\n",
    "scalerP.fit_transform(ts_train)\n",
    "ts_ttrain = scalerP.transform(ts_train)\n",
    "ts_ttest = scalerP.transform(ts_test)    \n",
    "ts_t = scalerP.transform(ts_P)\n",
    "\n",
    "# make sure data are of type float\n",
    "ts_t = ts_t.astype(np.float32)\n",
    "ts_ttrain = ts_ttrain.astype(np.float32)\n",
    "ts_ttest = ts_ttest.astype(np.float32)\n",
    "\n",
    "print(\"first and last row of scaled price time series:\")\n",
    "pd.options.display.float_format = '{:,.2f}'.format\n",
    "ts_t.pd_dataframe().iloc[[0,-1]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "1150d1d1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "first and last row of scaled feature covariates:\n"
     ]
    },
    {
     "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>component</th>\n",
       "      <th>btc_inflation_rate</th>\n",
       "      <th>mining_difficulty</th>\n",
       "      <th>hash_rate</th>\n",
       "      <th>trans_count</th>\n",
       "      <th>Vol.</th>\n",
       "      <th>Change %</th>\n",
       "      <th>total_tweets</th>\n",
       "      <th>Google_trends</th>\n",
       "      <th>new_senti</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2018-01-01</th>\n",
       "      <td>0.61</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.02</td>\n",
       "      <td>0.34</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.41</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.40</td>\n",
       "      <td>0.68</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-11-23</th>\n",
       "      <td>0.66</td>\n",
       "      <td>1.84</td>\n",
       "      <td>1.63</td>\n",
       "      <td>0.54</td>\n",
       "      <td>0.05</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.03</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.68</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "component   btc_inflation_rate  mining_difficulty  hash_rate  trans_count  \\\n",
       "Date                                                                        \n",
       "2018-01-01                0.61               0.00       0.02         0.34   \n",
       "2019-11-23                0.66               1.84       1.63         0.54   \n",
       "\n",
       "component   Vol.  Change %  total_tweets  Google_trends  new_senti  \n",
       "Date                                                                \n",
       "2018-01-01  0.00      0.41          0.50           0.40       0.68  \n",
       "2019-11-23  0.05      0.52          0.03           0.15       0.68  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# train/test split and scaling of feature covariates\n",
    "covF_train, covF_test = ts_covF.split_after(0.8)\n",
    "\n",
    "scalerF = Scaler()\n",
    "scalerF.fit_transform(covF_train)\n",
    "covF_ttrain = scalerF.transform(covF_train) \n",
    "covF_ttest = scalerF.transform(covF_test)   \n",
    "covF_t = scalerF.transform(ts_covF)  \n",
    "\n",
    "# make sure data are of type float\n",
    "covF_ttrain = covF_ttrain.astype(np.float32)\n",
    "covF_ttest = covF_ttest.astype(np.float32)\n",
    "\n",
    "pd.options.display.float_format = '{:.2f}'.format\n",
    "print(\"first and last row of scaled feature covariates:\")\n",
    "covF_t = covF_t.astype(np.float32)\n",
    "covF_t.pd_dataframe().iloc[[0,-1]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "450aa399",
   "metadata": {},
   "outputs": [],
   "source": [
    "#checking for seasonality. We dont have to worry about it as the model will intrinsicly learn it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "bede7798",
   "metadata": {},
   "outputs": [],
   "source": [
    "MSEAS = 90          # max seasonality to check: days\n",
    "ALPHA = 0.05        # significance level for seasonality test\n",
    "FIGSIZE = (9, 6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "b6488865",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "seasonal? True\n",
      "There is seasonality of order 38 days\n"
     ]
    }
   ],
   "source": [
    "# check for seasonality, via ACF\n",
    "mseas=MSEAS\n",
    "for m in range(2, MSEAS):\n",
    "    is_seasonal, mseas = check_seasonality(series, m=m, alpha=ALPHA, max_lag=MSEAS)\n",
    "    if is_seasonal:\n",
    "        break\n",
    "\n",
    "print(\"seasonal? \" + str(is_seasonal))\n",
    "if is_seasonal:\n",
    "    print('There is seasonality of order {} days'.format(mseas))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a895568b",
   "metadata": {},
   "source": [
    "Our data seems to have a seasonliy of 38 days."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "2cec6aee",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABBwAAAFcCAYAAABr+63OAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAABryklEQVR4nO3dd3QU5dvG8e+k0EvohN4RkCZDF6VjAQTpoBQRRJoFsaCg8EN6E0EpIigqSBEpgiCgNKUMIB0EpAYiPRhaQrLvH1nyolLDJrM7e33O2RMyMzt7rbeT2dx55hnD5XIhIiIiIiIiIuJJAXYHEBERERERERHnUcNBRERERERERDxODQcRERERERER8Tg1HERERERERETE49RwEBERERERERGPU8NBRERERERERDxODQcRERH5D8Mw8hmG4TIMI8juLCIiIuKb1HAQERHxAYZhHDYMo7bdOUS1EBERuVdqOIiIiDiARiKIiIiIt1HDQURExMsZhjEdyAMsNAwj0jCMN2+65KGjYRhHgZXubV8wDGOPYRjnDcNYahhG3pv24zIMo4thGPvd68cbhmG41wUahjHCMIwzhmH8CTx9l0y5DcP4zjCM04ZhnDUMY5x7eYBhGO8ZhnHEMIxThmF8aRhGeve6G5k7GIZxzJ2hi2EY5Q3D2G4YxoUb+3Fv394wjHWGYXxsGEaEYRh7DcOoddP6HIZhLDAM45xhGAcMw+h007oPDMOY5X79vw3D2GUYhvmv58515z9kGEbPe3nurWqRkJqKiIj4AzUcREREvJzL5XoeOAo0cLlcaVwu17CbVj8OFAPqGYbRCOgDPAtkAdYAM/61u/pAeaA00Byo517eyb2uLGACTW+XxzCMQGARcATIB+QEZrpXt3c/agAFgDTAuH/toiJQGGgBjAHeBWoDJYDmhmE8/q9t/wQyA+8D3xmGkdG9bgZwHMjhzjvo5oYE0NCdKwRYcCOHYRgBwEJgmzt7LeBVwzDq3e25d6mFiIiI3EQNBxEREd/2gcvluuRyua4ALwGDXS7XHpfLdR0YBJS5eZQDMMTlcl1wuVxHgZ+BMu7lzYExLpfrmMvlOgcMvsNrViDul/ze7te+6nK51rrXtQFGuVyuP10uVyTwDtDyX5d8/M/9nGXAJWCGy+U65XK5wohrkpS9adtT7lzRLpfrW2Af8LRhGLmBR4G33Pv6HfgMeP6m5651uVyLXS5XDDCduCYLxDVcsrhcrgEulyvK5XL9CUwGWt7Dc0VEROQeqeEgIiLi247d9O+8wEfuSxMuAOcAg7i/4t8QftO/LxM3AgHiGgg37+vIHV4zN3DE3dT4txz/eu4RIAjIdtOyv27695VbfJ/mpu/DXC6X61/7y+F+nHO5XH//a92d3msKd+MjL5Djxn8n93+rPv/KeLvnioiIyD3SiVNERMQ3uO5h+THgQ5fL9XUC9n+SuEbCDXnusO0xII9hGEG3aDqcIO4X+pv3c524pkKuBOTKaRiGcVPTIQ9xlzicADIahpH2pqZDHiDsHvZ5DDjkcrkKJyAP3L4WIiIichONcBAREfENfxE3J8KdTADeMQyjBIBhGOkNw2h2j/ufBfQ0DCOXYRgZgLfvsO1G4hoUQwzDSG0YRgrDMKq6180AXjMMI79hGGmIu6zj29uMhrgXWd25gt3vpRiw2OVyHQN+BQa7X78U0BG4l2bLRuCiYRhvGYaR0j1h5sOGYZS/x0z3UgsRERG/p4aDiIiIbxgMvOe+BOCNW23gcrnmAUOBmYZhXAR2Ak/e4/4nA0uJm0hxC/Dd7TZ0z2vQAChE3ASKx4mbABLgc+LmPFgNHAKuAj3uMcOtbCBugskzwIdAU5fLdda9rhVxk1aeAOYB77tcrp/utsOb8pdxZzxD3PwP6e8x011rISIiImD887JIEREREe9gGEZ74EWXy/Wo3VlERETk/mmEg4iIiIiIiIh4nBoOIiIiIiIiIuJxuqRCRERERERERDxOIxxERERERERExOPUcBARERERERERjwuyO8A98MlrPsLDw8mePbvdMSQRqLbOpvr6B9XZ+VRj51JtnU81dj7V2HGM263QCIdEEhMTY3cESSSqrbOpvv5BdXY+1di5VFvnU42dTzX2H2o4iIiIiIiIiIjHqeEgIiIiIiIiIh6nhoOIiIiIiIiIeJwaDiIiIiIiIiLicWo4iIiIiIiIiIjHqeEgIiIiIiIiIh6nhoOIiIiIiIiIeJwaDiIiIiIiIiJ+7scff6Ro0aIUKlSIIUOGeGSfajgkkpiYGLsjiIiIiIiIiNxVTEwM3bp1Y8mSJezevZsZM2awe/fuB96vGg4e5nK56N+/Py+88ALR0dF2xxERERERERG5o40bN1KoUCEKFChAsmTJaNmyJfPnz3/g/arh4GHh4eF8/PHHrFixgk6dOuFyueyOJCIiIiIiIj7AMIxEedxNWFgYuXPnjv8+V65chIWFPfD7CbrbBqZppgWWAyWASsAhYB6QAogBOliWddg0zeLARPc++1qWtdw0zTTAdCALsMiyrCHufQ537+uo+/lRD/xOvERoaCg//PADNWrU4IsvviB79uweu/5FRERERERExNNu9Yfye2lU3M29jHC4AtQH5ri/v05ck+AxYDDQ2718EPACUA8Y4F7WCfjBsqxHgeqmaeYyTbMsEGpZVjVgN9D0gd+Fl6lYsSKTJk0iKCiIoUOHMnr0aLsjiYiIiIiIiJdzuVyJ8ribXLlycezYsfjvjx8/To4cOR74/dy14WBZ1nXLsk7f9P01y7JujK2IJq4BAXFNhP2WZV0EzpqmmRmoDCxzr/+JuFENNy/7EajywO/CC9WoUYOpU6cC8Prrr/P111/bnEhERERERETkv8qXL8/+/fs5dOgQUVFRzJw5k4YNGz7wfu96ScXtmKYZDPQDXnQvunm8RQSQEQgBLt5i2Yl/Lfv3vjsDnQG6d+9OnTp1EhrTNtHR0dSoUYN+/foxYMAA2rdvj8vlokaNGh7Z/8iRI1m8eDGzZs0iU6ZMHtmn3Jvo6GiPXM8k3kn19Q+qs/Opxs6l2jqfaux8qrF3+uCDD6hVqxaxsbG0aNGCkJCQe6pTzpw5b7suwQ0HYBIwwbKsg+7vY29aFwKcA84D6YAL7mWHgUD3spu3+wfLsia59w/gk7MuhoWFkTNnTvr378+VK1cYPnw4L730EitXrqRixYoPvP8bl2lMnDiR8ePHP/D+5N7dqK04k+rrH1Rn51ONnUu1dT7V2PlUY+/Utm1b2rZt69F9JuguFaZpvgccsizr25sWh5umWdg0zXRARsuyzgDrgbru9bWB3/61rB6wLkHJfcjQoUNp164dly9f5umnn2bfvn0e2/eKFSs8ti8RERERERERT7mnhoNpmouJaxJMNk2zLfA+UNM0zV9M0xzs3qwP8Dmw1L0eYDJQ3zTNdcAvlmWFWZa1FThpmuYaoDgw13NvxzsZhsHkyZN56qmnOHv2LHXr1vXYEKJ9+/YRERHhkX2JiIiIiIiIeIpxLzNW2szrA97KrYYJXbp0idq1a7N+/XoefvhhVq9eTYYMGe5739evXyc4ODj++5kzZ9KiRYsHziz3RkPAnE319Q+qs/Opxs6l2jqfaux8qrHj3Pb+mQm6pEISJnXq1CxatIhixYqxc+dOGjZsyJUrV+57PxcvXvzH90uWLPFURBERERERERGPUMMhiWXKlImlS5eSK1cu1q5dS8uWLbl+/frdn3iTf19CsWLFinu6t6qIiIiIiIhIUlHDwQa5c+dm6dKlZMiQgQULFtClS5f7ahhcuHABgJIlS5I1a1aOHz/OH3/8kUhpRURERERERO6fGg42KV68OD/88AMpU6ZkypQpvPfee/f83BsNh4wZM1KrVi0Ali9fnhgxRURERERERBJEDQcbVa5cmdmzZxMYGMigQYMYO3bsPT3vRsMhffr0ajiIiIiIiIiIV1LDwWZPP/00U6ZMAeDVV19l5syZd33OjTkcQkJCqF27NgA///zzfc8FISIiIiIiIpJY1HDwAu3atWPo0KG4XC7atm3LTz/9dMftb4xwCAkJIW/evBQqVIiIiAg2b96cBGlFRERERERE7k4NBy/Ru3dvXn/9daKjo2ncuDGWZd1225sbDkD8KAddViEiIiIiIiLeQg0HL2EYBsOHD6dNmzZcunSJJ5988rZ3nrh5DgdQw0FERERERES8jxoOXiQgIIDPP/+cevXqcebMGerVq8fJkyf/s92/RzjUqFEDwzD49ddfuXTpUhImFhEREREREbk1NRy8TLJkyZgzZw4VKlTg8OHDPPHEE/ENhhtunjQS4m6PWbZsWaKioli/fn0SJxYRERERERH5LzUcvFCaNGn44YcfKFq0KNu3b+eZZ57h6tWr8ev/PcIB4NFHHwVg3bp1SRlVRERERERE5JbUcPBSmTNnZunSpeTIkYPVq1fTunVrYmJigP/O4QBQtWpVAH799dckzyoiIiIiIiLyb2o4eLG8efOydOlSQkJCmDdvHl27dsXlct1yhMONhsNvv/0W35gQERERERERsYsaDl7u4YcfZuHChaRIkYJJkybxwQcf/GcOB4CcOXOSN29eLl68yK5du2xKKyIiIiIiIhJHDQcf8Oijj/Ltt98SEBDAgAEDOH/+PPDPSyrg/0c5aB4HERERERERsZsaDj6iYcOGTJo0Kf771KlTExQU9I9tqlSpAqA7VYiIiIiIiIjt1HDwIR07duTDDz8EIDQ09D/rK1WqBKjhICIiIiIiIvYLuvsm4k3eeecdChUqRN68ef+zrlSpUqRIkYI//viDs2fPkilTJhsSioiIiIiIiGiEg88xDIPmzZtTsWLF/6wLDg6mXLlyAGzcuDGpo4mIiIiIiIjEU8PBYXRZhYiIiIiIiHgDNRwcRg0HERERERER8QZqODjMjYbDxo0biY2NtTmNiIiIiIiI+Cs1HBwmV65c5MyZkwsXLvDHH3/YHUdERERERET8lBoODqTLKkRERERERMRuajg4kBoOIiIiIuI0ly9fZvbs2XTq1IknnniCp59+mjfffJPdu3fbHU1EbiPI7gDieWo4iIiIiIhT7N27l08//ZQvv/ySCxcu/GPd4sWLGTFiBE2aNOG9996jdOnS9oQUkVvSCAcHeuSRRwgKCmLHjh1ERkbaHUdERERE5L6dOXOGDh06ULx4ccaOHcuFCxcoX748Q4cOZdGiRcydO5cuXboQHBzMnDlzKFOmDM888wybNm2yO7qIuGmEgwOlSpWKcuXKsWHDBn7++WcaNGhgdyQRERERkXu2efNmGjduzLFjxwgODqZdu3Z07dqVsmXL/mO7Z599lr59+zJixAgmTJjAggULWLBgAR07dmTixIkEBgbecv8XL15k69atnDlzhjRp0vDHH38QHh5OmjRpaN++PaGhoUnxNkUcTw0Hh3r66afZsGEDCxcuVMNBRERERHzGwYMHadSoEefPn6dSpUpMnz6dQoUK3Xb7HDlyMGrUKN5++21GjRrF2LFjmTJlCpGRkXz22WekSZPmH9vPmTOHzp07c/78+Vvub8CAAYwcOZKuXbt69H2J+CPD5XLZneFuvD7grYSFhZEzZ07bXn/r1q088sgjhIaGcvz4cQICdPWMp9hdW0lcqq9/UJ2dTzV2LtXW2S5evEipUqU4cuQI9evXZ+7cuSRLluy+9rFu3TqeeOIJIiMjyZ8/P02bNqV27drkzp2b4cOHM3XqVABKlSpF/vz5uXjxIvnz5yd//vxs2bKFefPmATBy5Ehef/11j79H0XHsQMbtVmiEg0OVKVOGnDlzEhYWxtatWylXrpzdkURERERE7qh3794cOXKE0qVLM2PGjPtuNgBUrVqVdevW8fzzz7N9+3aGDx/O8OHD49enSJGCESNG0LVrVwzjv78nTZo0iZdeeolevXqRM2dOWrRo8UDvScSf6c/eDmUYBvXr1wfiho2JiIiIiHizZcuWMWnSJJIlS8ZXX331n0sh7kepUqXYtGkTixcv5p133qFUqVLkzZuXJk2aYFkW3bp1u2WzAaBz586MHDkSgPbt27Nx48YE5xDxd2o4OFjr1q0B+PLLL4mJibE5jYiIiIjIrUVERNCxY0cAXn/9dR5++OEH3meyZMl48sknGTRoENu2bePw4cPMmTOHEiVK3PW5r732Gi+++CJXr16lYcOGHDly5IHziPgjNRwcrFq1ahQqVIgTJ06wbNkyu+OIiIiIiNzS66+/zvHjx6lQoQJdunSxOw6GYTB+/HiqV6/OX3/9RZUqVZLsdpsRERFs376dnTt3EhERkSSvKZJY1HBwMMMw6NChAwBTpkyxOY2IiIiIyH8tXryYzz//nOTJkzNt2jSCgrxjmrlkyZLx3XffUa1aNU6cOEGlSpXo0KEDu3fv/s+2LpeLB5mM/+rVq3z88ceUK1eOkJAQSpcuTcmSJcmYMSMVK1ZkypQpXLt27UHejogt1HBwuHbt2hEUFMT333/P0aNH7Y4jIiIiIhLvzJkzdOrUCYCBAwdSrFgxmxP9U4YMGfjpp5949dVXCQgIYNq0aZQoUYIiRYrwwgsvMGLECHr27Em2bNkICAggU6ZM1KlThz59+rB27dq7NiFcLhdz5syhcOHC9OzZky1btpAiRQqKFy9OsWLFCAwMZOPGjbz44os88sgjWJaVRO9cxDPUcHC4nDlz0qxZM2JiYhg/frzdcUREREREAIiNjaVdu3acOHGCqlWr8tprr9kd6ZaSJ0/O6NGj2b17N126dCFNmjTs37+fqVOn0rt3bz7++GNOnz4NwLlz51i+fDmDBw+mWrVqPPzww3z88cdcuHDhP/s9ePAgTz/9NM2aNeP48eOUKlWKWbNmcf78eXbt2sXu3bs5f/4806dPp0iRIuzevZtKlSrRr18/rl+/nsT/FUQSxniQoT9JxOsD3oo33Vt248aNVKxYkZCQEI4cOUK6dOnsjuTTvKm24nmqr39QnZ1PNXYu1dYZXC4XvXr1YvTo0WTIkIHff/+dPHnyAN5f4+joaLZt28batWvjP1vXr1+fcuXKceLECSzLYt26dUyfPp2//voLgJQpU1K+fHny5s1LtmzZOHToEN9//z0xMTGkT5+eIUOG0LlzZwICbv334CtXrvDuu+8yZswYXC4Xjz32GLNmzSJbtmxJ+dY9xttrLPft1rd8QQ2HRONtB9Fjjz3GmjVr6Nu3LwMGDLA7jk/zttqKZ6m+/kF1dj7V2LlUW2fo378/H3zwAcHBwSxcuJB69erFr3NKjaOjo5k/fz4TJkxgxYoV/1kfGBjIc889x9ChQ++5cfDLL7/QqlUrwsPDKViwIMuWLaNAgQKejp7onFJjiaeGQ1LztoNo7dq1VKtWjdSpU3PgwAGyZ89udySf5W21Fc9Sff2D6ux8qrFzqba+b/78+TRq1IiAgADmzJlD48aN/7HeiTU+duwY+/bt48iRI5w+fZrMmTNTq1Yt8ufPf9/7Cg8Pp379+mzevJns2bOzbNkySpYsmQipE48Ta+znbttw8I4pYCXRPfroozRo0ICFCxfSq1cvvv76a7sjiYiIiIifCQsLo127dgAMHjz4P80Gp8qdOze5c+f2yL6yZ8/OypUradSoET///DOPPfYYX3zxBQ0bNvTI/kU8SZNG+pHRo0eTMmVKvvnmGxYtWmR3HBERERHxIy6Xiy5duhAREUH9+vXp3bu33ZF8Vrp06Vi8eDHPPvssFy5c4JlnnuHVV1/VrTPF66jh4EcKFizIwIEDAeJ/2IuIiIiIJIXZs2ezaNEi0qdPz8SJEzGM247ClnuQIkUKZs+ezYgRIwgKCuKjjz6icuXK7Nmzx+5oIvHUcPAzr7zyChUrViQsLIw333zT7jgiIiIi4gcuX77MG2+8AcDQoUPJkSOHzYmcISAggF69erFu3Try58/P1q1bKVOmDK+++irr1q0jPDxct9AUW6nh4GcCAwOZMmUKwcHBTJo0iXXr1tkdSUREREQcbsSIERw7doyyZcvy4osv2h3HcSpUqMDvv//OCy+8QFRUFB999BGPPvoooaGhpEuXjnr16rFmzRq7Y4ofUsPBD5UoUYK33noLgG7duqnrKSIiIiKJJjIykjFjxgAwatQoAgMD7Q3kUOnSpWPKlCls3bqVzp07U6ZMGbJkycKVK1dYtmwZjz32GO+88w4+cJdCcRA1HPzUO++8Q968edm2bRufffaZ3XFERERExKEmT57M+fPnqVKlCtWrV7c7juOVKVOGiRMnsnXrVk6dOsVff/1Fv379CAoKYsiQIbz++utqOkiSUcPBT6VKlYoRI0YA0L9/fy5fvmxzIhERERFxmqioKEaNGgXA22+/bXMa/5Q1a1b69+/PvHnzSJYsGWPGjGHKlCl2xxI/oYaDH2vSpAmmaRIeHh5/IhARERER8ZQZM2Zw/PhxihcvztNPP213HL9Wv379+JHNPXr0YOfOnTYnEn8QdLcNTNNMCywHSgCVLMvaaZpmC+AV4CrQzrKsY6ZpFgcmuvfZ17Ks5aZppgGmA1mARZZlDXHvczhQCTgKdLAsKyoR3pvchWEYDB06lFq1ajFgwACefPJJypUrZ3csEREREXGA2NhYhg0bBsCbb75JQID+1mm3559/np9//pmpU6fSvn171q9fT1DQXX8lFEmweznqrwD1gTkApmkGA68D1YG+7gfAIOAFoB4wwL2sE/CDZVmPAtVN08xlmmZZINSyrGrAbqCpZ96KJETNmjXp3r070dHRPPvss2zatMnuSCIiIiLiAN9//z27d+8mV65ctGrVyu444vbRRx+RJ08eNm/ezPDhw+2OIw5314aDZVnXLcs6fdOiwsAuy7KiLMtaB5R0Lw+1LGu/ZVkXgbOmaWYGKgPL3Ot/Im5Uw83LfgSqeOB9yAMYPnw4FSpU4OjRo1SpUkWTSIqIiIjIA7l+/TrvvfceEDd3Q7JkyWxOJDekTZuWSZMmAdC3b19WrVplcyJxsoSMnwkBLt70/Y372hg3LYsAMv5r25uXnfjXsn8wTbMz0Bmge/fu1KlTJwEx7RUdHU1YWJjdMe7ZN998w8CBA5k2bRqdOnVi27ZtGvp2G75WW7k/qq9/UJ2dTzV2LtXWN3z77bfs2bOHvHnz8tRTT91XzVTjxPfwww/z8ssv8+mnn9KkSRMWLFhA3rx5k+z1VWNnyZkz523XJaThcB5Id9P3Me6vsTctCwHO3bTtBfeyw8Q1KNL9a7t/sCxrEjDJ/a1P3rMlLCzsjv/hvdHUqVOpXLkyXbt2Zdy4cZw+fZpp06aRIkUKu6N5FV+srdw71dc/qM7Opxo7l2rr/WJiYpgwYQIAAwcOJH/+/Pf1fNU4aYwdO5aDBw+ybNky2rVrx7p168iaNWuSvLZq7D8S8ufrA0Bx0zSTmaZZFdjuXh5ummZh0zTTARktyzoDrAfqutfXBn7717J6wLoEpxeP69y5Mz/88ANp06bl22+/pXbt2pw5c+a22+seviIiIiJys/nz57N//37y5ctHy5Yt7Y4jtxEUFMTs2bMpW7YsBw4coH79+kRGRtodSxzmnkY4mKa5GCgDFAU+BUYDq4i7S0Vb92Z9gM/d++znXjYZ+Mo0zQ7AQsuywoAw0zRPmqa5hri7VGimEi9Tr1491q5dy9NPP826deuoXLkyw4cPJ1OmTPz555/8+eefHDx4kPXr13Pw4EFSpkzJCy+8wKBBg0iXLt3dX+AuIiMj2blzJ2fPniVVqlQUL16cbNmyeeCdiYiIiEhicrlcDB06FIBevXrpDgheLl26dCxevJgqVaqwadMmmjVrxoIFCwgODrY7mjiE4QN/ofb6gLfihGFCJ06coH79+mzduvWeti9cuDArVqwgd+7cd93W5XKxb98+duzYET+C4ty5c6xYsYJVq1YRG/v/V+gYhsFrr73Ghx9+6BWXdzihtnJ7qq9/UJ2dTzV2LtXWu61evZrHH3+cTJkyceTIEVKnTn3f+1CNk97+/fupUqUKZ86coW3btkybNg3DMO7+xARSjR3ntv+zqOUot5UjRw5Wr17NmDFjWL58OVevXqVAgQLxj5IlS1K2bFl2797N888/z/bt23nssceYOXMmFStWvOU+XS4Xc+fO5Z133uHAgQO33CYoKIiSJUsSGhrKxYsX2bBhA6NGjcKyLJYsWUKqVKkS822LiIiISALduM1i9+7dE9RsEHsULlyYxYsXU716db788kty5MjB4MGD7Y4lDqARDonE37p258+f54knnmDjxo0EBATQqVMnunbtSnBwMGvXruXo0aNcvHiRn3/+mR07dgCQOXNmqlSpQvbs2QEICQmhWLFiNGrUiJCQkPh9b9y4kcaNG3PixAmeeOIJfvjhB1vvnuFvtfU3qq9/UJ2dTzV2LtXWe+3cuZOSJUuSIkUKjh49SpYsWRK0H9XYPj/++CMNGjTg+vXrfPTRR/Ts2TNRXkc1dhyNcJDElSFDBlatWkW/fv0YOXIkEydOZOLEibfcNnv27Lz//vt06tSJwMDAW25zswoVKrBy5UqqVavGjz/+yKeffkq3bt08/RZERERE5AH0798fgBdffDHBzQax1xNPPMGUKVNo164dr776KtmzZ6d58+Z2xxIfphEOicSfu3Z79uxhyJAhrFmzhqtXr/L4449TpEgRQkJCyJ07Nw0bNiRZsmT3vd958+bx7LPPkjp1anbt2pWk9wq+mT/X1h+ovv5BdXY+1di5VFvvtG3bNsqUKUPy5Mn5888/yZEjR4L3pRrbb+jQobz99tskS5aMH3/8kRo1anh0/6qx42iEgySdYsWK8cUXX3h8v40bN6ZZs2bMnj2b119/nblz53r8NURERETk/rhcLt544w0AXn755QdqNoh3ePPNNzlx4gRjx46lUaNGrF69mtKlS9sdS3yQfRfCiyTA6NGjSZ06Nd999x3Lli2zO46IiIiI3/v2229Zvnw5GTJkoE+fPnbHEQ8wDIPRo0fTvHlzLl68yBNPPMHevXvtjiU+SA0H8Sk5c+akX79+QFzn9ebbZ4qIiIhI0jp16hSvvPIKEDcMX3M3OEdAQABffvklNWrUIDw8nCpVqrBq1Sq7Y4mPUcNBfE7Pnj3JlSsX27ZtY/bs2XbHEREREfFLLpeL9u3bc+rUKapXr07Hjh3tjiQeljx5chYuXEjDhg05f/48derUYfr06XbHEh+ihoP4nBQpUsSPcnj//fc1ykFERETEBp999hlLliwhY8aMTJ8+3dbblkviuXE58yuvvEJ0dDRt27bl9ddfJzo62u5o4gP0U0F8Uvv27cmTJw/79u1j+fLldscRERER8SsnT56kd+/eAIwbN45cuXLZnEgSU2BgIGPGjGH8+PEEBQUxevRo6tSpw19//WV3NPFyajiITwoODqZLly4AfPLJJzanEREREfEvr7zyChERETz55JO0bNnS7jiSRLp27covv/xC9uzZWbVqFeXKlWPjxo12xxIvpoaD+KyOHTsSHBzMwoULOXr0qN1xRERERPzCwoULmT17NqlTp+bTTz/FMAy7I0kSqlq1Klu2bKFq1aqEhYVRq1Yttm3bZncs8VJqOIjPypo1K02bNiU2NpZJkybZHUdERETE8S5evEjXrl0BGDhwIHnz5rU5kdghNDSUlStX0qJFCyIjI6lfvz7h4eF2xxIvpIaD+LQbJ7zJkycTFRVlcxoRERERZ3v33Xc5fvw45cuXp0ePHnbHERslS5aMadOmUbVqVY4fP07z5s01kaT8hxoO4tOqVq1KyZIlOXXqFN99953dcUREREQc67fffmP8+PEEBgYyefJkAgMD7Y4kNkuRIgVz5swhR44crFmzhjfffNPuSOJl1HAQn2YYBt26dQNg8ODBukWmiIiISCKIioqiU6dOuFwuevfuTenSpe2OJF4ie/bszJ49m+DgYMaMGcPMmTPtjiReRA0H8Xnt2rUjd+7cbN++nW+//dbuOCIiIiKOM2zYMHbt2kWhQoXo16+f3XHEy1SpUoXRo0cDcRO779y50+ZE4i3UcBCflyJFCt5//30A+vXrp2vHRERERDxo3759/O9//wNg4sSJpEyZ0uZE4o26du3K888/z+XLl3n22WeJiIiwO5J4ATUcxBHatWtHkSJFOHDgANOmTbM7joiIiIhj9OjRg6ioKDp06EDNmjXtjiNeyjAMJkyYQOnSpdm/fz/t27fH5XLZHUtspoaDOEJQUFB8571///5cuXLF5kQiIiIivu/333/np59+Im3atIwYMcLuOOLlUqVKxdy5c0mfPj3ff/89w4cPtzuS2EwNB3GMpk2bUqZMGcLCwhg6dKjdcURERER83s3X5WfMmNHmNOILChYsyJdffgnAO++8wy+//GJvILGVGg7iGAEBAYwdOxaAIUOGcODAAZsTiYiIiPiukydPMmPGDAICAujZs6fdccSHNGzYkHfeeYfY2FhatGjBiRMn7I4kNlHDQRylWrVqtGvXjmvXrtG9e3ddNyYiIiKSQJ988gnR0dE0atSI/Pnz2x1HfMyAAQOoWbMmp06donnz5prY3U+p4SCOM2zYMEJCQli6dCnfffed3XFEREREfM6VK1eYMGECAK+99prNacQXBQUFMWPGDHLkyMG6det4++237Y4kNlDDQRwna9asDB48GIBXXnmFv//+2+ZEIiIiIr7lq6++4syZM5imSdWqVe2OIz4qa9aszJ49m6CgIEaNGsXs2bPtjiRJTA0HcaROnTpRvnx5wsLCGDBggN1xRERERHyGy+VizJgxQNzoBsMw7A0kPq1KlSqMHDkSgA4dOrBz506bE0lSUsNBHCkwMJBPP/0UwzAYPXo0+/fvtzuSiIiIiE9YtmwZu3fvJmfOnDRr1szuOOIAPXr0oE2bNly6dIlGjRpx4cIFuyNJElHDQRyrXLlydOjQgZiYGEaNGmV3nCQRExPDqlWrGDp0KG+88QbTp0/n9OnTdscSERERH3LjVpjdu3cnODjY5jTiBIZhMGnSJMqWLcvBgwfp0aMHMTExdseSJGD4wCz+Xh/wVsLCwsiZM6fdMfzenj17KF68OClSpODYsWNkzpz5gffpjbWNiopi4sSJDBo0iPDw8H+sCwwMpEaNGtStW5fmzZuTN29em1L6Bm+sr3ie6ux8qrFzqbaJa/fu3ZQoUYJUqVJx7NgxMmbMmOQZVGPnOnz4MKZpcvbsWd59910GDhxodyTxjNted6URDuJoxYoV46mnnuLq1at8+umndsdJFMeOHaNatWr07NmT8PBwChUqRPfu3enfvz9PPPEEAMuXL+fNN98kX758NG7cmB07dticWkRERLzRjbkb2rVrZ0uzQZwtX758fPvttwQEBPDhhx/qjnJ+QA0HcbxevXoBMG7cOK5evWpzGs9asWIFjzzyCBs3biRPnjzMnz+fP/74g48//ph+/fqxZMkSwsPDmTFjBi1btiR58uR8//33mKbJ+PHj8YERTiIiIpJEzp07x/Tp04G4O32JJIZatWrx3nvvAXGNrd27d9ucSBKTGg7ieDVq1KBMmTKcOnWKr7/+2u44HuFyuRg5ciR169blzJkz1K1bly1bttCwYcP/zCSdOXNmWrZsyYwZMzh8+DCdOnUiKiqK7t2789JLLxEdHW3TuxARERFvMnPmTK5evUrdunUpWrSo3XHEwTp16kSrVq2IjIykadOmXLlyJckzxMTEcPDgQXbs2MHmzZuxLIvt27ezd+9ezp07pz/MeYgaDuJ4hmHEj3IYOXIksbGxNid6MFeuXOH555/njTfeIDY2lj59+rB48WIyZcp01+dmz56dSZMmMWPGDFKkSMHkyZN56qmnNFOwiIiIMG3aNADat29vaw5xPsMw+OyzzyhWrBh79uyhT58+SfK6sbGxLF26lCZNmpA+fXoKFSpEqVKlME2T8uXLU7p0aYoVK0amTJkIDQ3lhRdeYNOmTUmSzak0aWQi0WQ33iU6Opr8+fMTFhbGDz/8wFNPPZXgfdlZ28jISOrUqcP69etJnTo1X3zxBU2aNEnQvjZs2EDDhg05deoUDz30EPPnz6dIkSIeTux7dOz6B9XZ+VRj51JtE8eNySLTpUtHeHg4KVOmtC2Laux8N2q8efNmKlWqxPXr11mxYgU1a9ZMlNc7cOAAX331FV9++SWHDh2KX54rVy5CQkJIliwZEPc7w9WrVzl58iSRkZHx21WvXp0PPviAxx9/PFHyOYAmjRT/FhwcHH8t4siRI21OkzBRUVE0bdqU9evXkzdvXn777bcENxsAKlasyIYNG3j44YfZu3cvFStWZOPGjR5MLCIiIr5i8uTJALRo0cLWZoP4l3LlytG3b18gbmSNp0bdulwu/vzzT8aPH0/lypUpXLgw/fv359ChQ+TNm5eBAwdy7Ngxjh07Fn9JxebNm9m+fTt//PEHFy9eZMeOHfTu3Zt06dLxyy+/ULNmTTZv3uyRfP5EDQfxG506dSJNmjSsXLmS33//3e449yU2NpaOHTuydOlSsmTJwvLlyylZsuQD7zdfvnz8+uuvNGzYkAsXLlC3bl0sy/JAYhEREfEVkZGRTJ06FYAuXbrYnEb8TZ8+fahQoQLHjh2jZ8+eD7Qvl8vFV199Rbly5ShYsCDdu3ePHxnctm1bfvrpJw4ePMi7775Lrly5brsfwzB4+OGHGTZsGEePHqVt27bExsby8ssvc+DAASIiIjTHwz0KsjuASFIJCQnhxRdfZMyYMYwcOTJ+FmZvd/36dbp27cpXX31F6tSpWbx4MYUKFfLY/tOmTcucOXNo1aoVc+fOpU6dOixfvpxy5cp57DVERETEe3311VdERERQpUoVHnnkEbvjiJ8JCgpi+vTplClThunTp/Pkk0/SqlWr+97PoUOH6NSpEytWrADiPvvXrFmTZ599lkaNGpE6deoE5UufPj3jxo1jxYoVbNq0icKFCwNxI6gzZ85MlixZ7umRMWNGAgL87+/9msMhkejaM+90+PBhChYsSEBAAIcOHbpjZ/N2krK2kZGRtGzZkh9++IEUKVIwf/586tatmyivFR0dTYsWLZg3bx5p0qRhxowZ1K9fP1Fey5vp2PUPqrPzqcbOpdp6VnR0NCVKlGD//v3xt9G2m2rsfLeq8fjx4+nevXv8hO9vv/32PU2KHhsby/jx43n77be5fPkymTNnZsiQIbRp04YUKVJ4LPNvv/3Gm2++SVhYGKdPn/7HHA/3IiAggEyZMt21SWGaJunTp/dY7iRy2zkc1HBIJPpB6b1atGjBrFmzeOeddxg0aNB9Pz+pavvXX39Rv359LMsiY8aMLFy4kCpVqiTqa0ZFRdGhQwe++eYbDMNg2LBh9OrV6z+32nQyHbv+QXV2PtXYuVRbz/rss8/o1KkThQsXZteuXQQHB9sdSTX2A7eqscvlYsCAAQwYMIDY2FhSpkxJx44def3118mfP/8t97Njxw5efvll1q1bB8R9zv/444/JkiVLor+HK1eucObMGU6fPn3bx83rz58/f0/7Xb9+PRUrVkzk9B6nhkNS0w9K77VmzRoee+wxsmfPztGjR+/7xJoUtT148CB16tTh0KFD5M+fnyVLliTZ/bBdLhcffvhh/AQ+Xbt25eOPP/abIWA6dv2D6ux8qrFzqbaec+nSJR566CGOHz/uNaMbQDX2B3eq8fr16/nf//7H4sWLgbiRAc8++yxdu3alevXqGIZBREQE//vf/xgzZgwxMTFkz56dTz75hMaNGyfl27gv0dHRnD179q6NiS+//JLcuXPbHfd+qeGQ1PSD0nu5XC6KFy/O3r17mTdvHo0aNbqv5yd2bSMjI6lYsSK7d+/GNE0WLVpEtmzZEu31bmfOnDk899xzXLt2jTZt2jB16lSv+KtHYtOx6x9UZ+dTjZ1LtfWcN954g5EjR1K2bFksy/KaPy6oxs53LzXeuXMnI0aM4Ouvv+b69esAPPTQQ+TPn5/Vq1dz6dIlDMOga9euDBw4kJCQkCRILreh22KK3GAYBp06dQL+/xZQ3sLlctG5c2d2795NsWLFWLlypS3NBoCmTZuyZMkS0qRJw9dff82zzz7LlStXbMkiIiIinrV+/XpGjx5NQEAAkyZN8ppmg8gNDz/8MNOmTePw4cP069eP0NBQ9u7dy5IlS7h06RI1a9Zk06ZNjBs3Ts0GL6afLOKX2rZtS7Jkyfjxxx85duyY3XHiTZs2jRkzZpAmTRq+++470qZNa2ueGjVqsGLFCjJmzMiiRYt48sknuXjxoq2ZRERE5MEcP36cxo0bExsby2uvvYZpmnZHErmtnDlz0r9/f44cOcJPP/3EggUL2LNnDytWrNBd1XyAGg7ilzJnzhx/ov3888/tjgPAH3/8QY8ePQD45JNPeOihh2xOFKdChQqsXr2aHDlysGrVKmrWrMmZM2fsjiUiIiIJcOnSJRo2bEh4eDjVq1dn8ODBdkcSuSfBwcHUrl2bBg0aeM3nZLk7NRzEb924rGLKlCnExMTYmuXatWu0atWKS5cu0bp1a5577jlb8/xbiRIlWLt2LQUKFGDz5s089thjHD9+3O5YIiIich9iY2Np164dW7dupWDBgsyZM8cv5mcSEfuo4SB+q0aNGhQoUIBjx46xbNkyW7O8++67bNmyhfz58/Ppp5965W0o8+fPz9q1a3n44YfZs2cP1apV00gHERERH/L+++8zd+5c0qdPz6JFi8iUKZPdkUTE4dRwEL8VEBDAiy++CNg7eeSyZcsYOXIkgYGBfPPNN6RLl862LHcTGhrKqlWrKFeuHIcPH6Zdu3bExsbaHUtERETu4ptvvmHgwIEEBgYya9YsDUkXkSShhoP4tfbt2xMYGMjChQsJDw9P8tc/evRo/OUTAwYMoFKlSkme4X5lzJiR7777jowZM7J48WImTJhgdyQRERG5gw0bNvDCCy8AMGbMGOrWrWtzIhHxF2o4iF8LDQ2lQYMGXL9+nU8++SRJXzsyMpJnn32W06dPU7duXd56660kff0HkSdPHiZOnAjAe++9x9mzZ21OJCIiIrdy/vx5mjVrxrVr1+jSpQvdunWzO5KI+BE1HMTv9erVC4DRo0dz+vTpJHnNy5cv06BBAzZv3kz+/PmZMWMGgYGBSfLantKkSRNq167N+fPn6devn91xRERE5F9cLhedO3fm2LFjVKhQgbFjx3rlPFEi4lxBCXmSaZoBwFSggHtRe8AEXgGuAu0syzpmmmZxYKL7dfpalrXcNM00wHQgC7DIsqwhD/YWRB7Mo48+ytNPP80PP/zA//73P8aOHZuor3f16lUaNWrEL7/8QmhoKMuWLSNjxoyJ+pqJwTAMPvroI0qVKsWECRPo3LkzpUuXtjuWiIiIuC1YsIA5c+aQNm1aZsyYoTtSiEiSS+gIhzJAcsuyqgEDgO7A60B1oK/7ATAIeAGo594OoBPwg2VZjwLVTdPMlcAMIh4zaNAgAgICGDduHD///HOivc7+/fupU6cOP/30E1mzZmXlypUUKlQo0V4vsRUvXpzu3bsTGxtLz549cblcdkcSERER4NKlS/Ts2ROADz/8kAIFCtzlGSIinpfQhsNxANM0DSAEOA3ssiwryrKsdUBJ93ahlmXttyzrInDWNM3MQGXgxj0IfwK8f5Y8cbxSpUrRt29fXC4Xzz33HHv27PHIfl0uF9u2bWPgwIFUrFiRIkWKsHbtWrJly8by5csdMUP0Bx98QObMmVm9ejXff/+93XFERESEuCbD0aNHKVu2LC+//LLdcUTETyXokgrgDBAL7AGSA22ArDetv3Ex+s0XiUUAGYlrUFz817J/ME2zM9AZoHv37tSpUyeBMe0THR1NWFiY3THkPnTo0IEff/yRDRs2ULlyZUaOHMkTTzzxn+3uVNuoqCh27drF5s2b2bRpE5Zl8ddff8WvT548Oc888wx9+/YlQ4YMjvl/5NVXX+W9997j9ddfp0yZMiRLlszuSAmmY9c/qM7Opxo7l2p7dwcOHGDEiBFA3B8Gbv4s4gtUY+dTjZ0lZ86ct12X0IZDPeCKZVkPmab5CPAWcOmm9THur7E3LQsBzgHngXTABfeyw//euWVZk4BJ7m99cox2WFjYHf/Di3dauXIlbdq04fvvv+fFF1+kdevWjB07lkyZMsVvc6vanjx5kv79+/Ptt99y4cKFf6zLnj079evXp379+tSuXZvUqVMnxVtJUm+++SZffvklf/zxB4sWLaJHjx52R0owHbv+QXV2PtXYuVTbO7sxWjM6OppOnTrRsGFDuyPdN9XY+VRj//Egd6k47/56AcgMFDdNM5lpmlWB7e514aZpFjZNMx2Q0bKsM8B64MbNf2sDvz1ABhGPSpUqFXPnzmXs2LGkSpWKb775hhIlStzxUoG5c+dSsmRJJk6cyIULFyhatCgdO3ZkypQp7Nmzh7CwMCZPnswzzzzjyGYDQHBwMEOHDgWgf//+RERE2JxIRETEP02bNo1ffvmFzJkzM3jwYLvjiIifS2jDYRmQwzTNVcA3xE0SORpYBQx0PwD6AJ8DS4H33csmA/VN01wH/GJZlsbSiFcJCAigR48ebN++nccee4y//vqLxo0b07p1a06ePAlAZGQkc+bMoXHjxjRt2pSzZ89Sp04dduzYwd69e/nss8944YUXeOihhwgI8I+7zz7zzDNUq1aNs2fPMnDgwLs/QURERDzq1KlTvPHGGwCMGTPmHyM0RUTsYPjArPJeH/BWNEzIGWJjYxk/fjxvv/02ly9fJjAwkFy5chEeHs61a9cASJkyJcOGDaNr165+01y4nU2bNlGxYkUCAgLYvHmzT94mU8euf1CdnU81di7V9vaee+45vv76a+rWrcuPP/6IYRh3f5IXUo2dTzV2nNv+sPHv345E7uLm0Q7PPvsshmFw5MgRrl27RuXKlRk2bBh79+6le/fuft9sAChfvjxdu3YlJiaGTp06ERMTc/cniYiIyAObP38+X3/9NSlSpOCTTz7x2WaDiDhLQieNFPErBQsWZO7cuZw7d44dO3ZQunRpQkJC7I7llQYNGsT333/Ppk2b+Pjjj3n11VftjiQiIuJoBw4coF27dkDc7TALFixocyIRkTj6k6zIfciYMSOFChVSs+EO0qVLxyeffALAe++9x5EjR2xOJCIi4lxXrlyhadOmRERE0LhxY1577TW7I4mIxFPDQUQ8rmHDhjRr1oxLly7x8ssv4wNzxYiIiPgcl8tF165d2bZtG4UKFWLq1Km6lEJEvIoaDiKSKMaOHUtISAhLlixhxowZdscRERFxnClTpjBt2jRSpkzJ3LlzSZ8+vd2RRET+QQ0HEUkU2bNnZ/jw4QC89NJL7Ny50+ZEIiIizrFlyxa6d+8OwMSJEylVqpTNiURE/ksNBxFJNB07dqR169ZERkbSoEEDjh8/bnckERERn3fu3DmaNGnCtWvX6NKlC88//7zdkUREbkkNBxFJNIZh8Nlnn1GxYkUOHz5MzZo1OXnypN2xREREfJbL5aJz584cPnwY0zQZM2aM3ZFERG5LDQcRSVQpU6ZkyZIllClThv3791OrVi1OnTpldywRERGf9M033zB37lzSpk3LrFmzSJ48ud2RRERuSw0HEUl0GTJk4KeffqJEiRLs2bOHp556isjISLtjiYiI+JTjx4/TrVs3AMaMGUP+/PltTiQicmdqOIhIksicOTPLly+nQIECbN68mRYtWhATE2N3LBEREZ/gcrno2LEjERER1K9fnw4dOtgdSUTkrtRwEJEkkz17dn788UcyZcrE4sWL6devn92RREREfMKECRNYtmwZmTJlYvLkyRiGYXckEZG7UsNBRJJU4cKFmTVrFgEBAQwaNIh58+bZHUlERMSrHThwgDfeeAOATz/9lOzZs9ucSETk3qjhICJJrmbNmgwdOhSAtm3bsmfPHpsTiYiIeKeYmBjat2/P5cuXadWqFc2aNbM7kojIPVPDQURs0atXL1q0aEFkZCSNGzfm4sWLdkcSERHxOiNHjmTdunWEhoYybtw4u+OIiNwXNRxExBaGYTBlyhRKlizJvn37eP755zWJpIiIyE127dpF3759AZgyZQoZM2a0OZGIyP1Rw0FEbJM6dWrmzZtHSEgICxYsoHfv3nZHEhER8QrR0dG0a9eOqKgoOnXqxJNPPml3JBGR+6aGg4jYqmDBgsybN4/g4GBGjx6t4aIiIiLAsGHD2Lx5M3ny5GHEiBF2xxERSZAguwOIiFSvXp3PPvuMdu3a8corr1CgQAGeeuopu2N51JYtW5g7dy67du3i77//JiQkhEyZMpE3b17KlClDmTJlyJEjh25zJiIi7Nixg/79+wPw+eefky5dOpsTiYgkjBoOIuIV2rZty59//kn//v157rnn2LJlC/ny5bM71gPbt28fb7zxBosWLbrrtlmyZKFSpUo89dRTPP300+TOnfu220ZFRXH69GkMwyBTpkwkT57ck7FFRMQmLpeLnj17Eh0dzcsvv0ytWrXsjiQikmBqOIiI1+jXrx+bN29m0aJFNG/enDVr1vjsL9JXrlzhww8/ZNiwYURHR5MmTRratWtH9erVSZ8+PRcuXODMmTMcOHCArVu3snXrVk6fPs3ChQtZuHAhACVLlqRcuXIEBQVx5swZTp06xalTpzh9+jQRERHxrxUUFMQjjzxCgwYNaNq0KQ899JBdb1tERB7QokWL+OWXX8iYMSODBg2yO46IyANRw0FEvEZAQABffPEFjzzyCJs2baJ3796MHTvW7lj3bePGjbRq1Yo///wTgBdffJEPP/yQrFmz3vY5LpeLw4cPs2LFCn744Qd++uknduzYwY4dO265fWBgIJkzZ8YwDE6dOsXGjRvZuHEjffv2pUSJEjRr1oymTZtSvHhxXaYhIuIjrl+/zptvvgnENeFDQkLsDSQi8oAMl8tld4a78fqAtxIWFkbOnDntjiGJQLVNfBs3buTRRx8lOjqaWbNm0axZsyR77Qet7+TJk+nWrRvR0dGULFmSCRMmUKVKlfvez7Vr11i7di2HDh3i+vXrZMmShSxZspA1a1ayZs1KSEgIAQFx8/7+/fffrFq1irlz5/L9999z4cKF+P2EhoZSuXJlGjVqRPPmzX12xIin6Th2PtXYuZxc2wkTJvDyyy9TsGBBdu/eTbJkyeyOZAsn11jiqMaOc9u/bqnhkEh0EDmXaps0xo0bR48ePUibNi3r16+nePHiSfK6Ca1vTEwMvXr14qOPPgKge/fujBw5Msk/LEZFRfHzzz8zZ84c5s2bx9mzZ+PXFS5cmLFjx/LEE08kaSZvpOPY+VRj53Jqbf/++28KFSrEqVOnmD17Nk2bNrU7km2cWmP5f6qx49y24aDbYoqIV+rWrRvNmzfn77//plq1avz88892R7qtK1eu0LBhQz766COCg4P5/PPP+fjjj235y1SyZMmoV68ekydP5tSpU+zbt49x48bx0EMPsX//fp588kmeeeaZ+Ms9RETEOwwbNoxTp05RuXJlmjRpYnccERGPUMNBRLySYRh8/vnnNGjQgHPnzlGzZk1efvnlf0yW6A2ioqJo2rQpixcvJlOmTCxfvpwOHTrYHQuImxOjSJEidOvWjW3btjFixAjSpEnDggULKF68ON27d+fIkSN2xxQR8XvHjx9n5MiRAIwcOVJz74iIY6jhICJeK3Xq1MybN4++ffsSFBTEhAkTKF68OBMmTODKlSt2xyMiIoIGDRrENxtWr17NY489ZnesW0qWLBm9evXijz/+4LnnnuPatWuMHz+eQoUK8eKLL3LmzBm7I4qI+K2+ffty5coVmjZtSuXKle2OIyLiMWo4iIhXCwwMZMCAAWzdupVKlSpx4sQJXn75ZfLly8fAgQM5fvy4LbkOHz5M1apVWbZsGVmyZGHp0qVJNs/EgwgNDWX69Ons3LmT5557DpfLxZQpUyhWrBgrV660O56IiN/Ztm0bX3zxBcHBwQwePNjuOCIiHqWGg4j4hIcffpi1a9cyc+ZMypYty6lTp+jbty958uShTp06fPXVV1y6dClJsmzYsIGKFSuya9cuihcvzoYNGyhXrlySvLanlChRgunTp7N7925q1KjBmTNnqFevHl988YXd0URE/IbL5eKNN97A5XLRtWtXChUqZHckERGPUsNBRHxGYGAgLVq0YPPmzSxbtoxmzZoRHBzM8uXLef7558mePTsdOnTAsqxEyzB79myqV6/OqVOnqF27NuvWrSN//vyJ9nqJrUiRIixfvpxevXpx/fp12rdvT9++fYmNjbU7moiI4y1dupTly5eTPn16+vbta3ccERGPU8NBRHyOYRjUqVOHWbNmER4ezqeffkrlypWJjIxk2rRplC9fnqeeeooDBw547DVdLheDBw+mefPmXL16lU6dOrF48WJCQkI89hp2CQgIYMSIEYwfP56AgAAGDhxI/fr1OXfunN3RREQcKyYmht69ewPw3nvvkSlTJpsTiYh4nhoOIuLTMmTIQJcuXfj111/Zt28fvXr1Il26dCxZsoRSpUoxbdq0B36N69ev8+KLL9KnTx8Mw2D48OFMnDiR4ODgB38DXqRr164sWbKETJkysWTJEh555BE2btxodywREUeaOXMmO3fuJG/evHTv3t3uOCIiicJwuVx2Z7gbrw94K2FhYeTMmdPuGJIIVFvvd+rUKV577TW++eYbANq2bcv48eNJkybNXZ/77/rGxMTQoUMHpk+fTsqUKfn6669p3LhxomX3BkeOHKFZs2Zs2rSJwMBAevXqxTvvvPNAozmio6M5dOgQUVFRxMbGcvHiRY4dO/aPx19//YVhGGTLlo28efOSL18+8uXLR9myZcmbN6/n3iA6jv2BauxcTqjt9evXKV68OPv37+fzzz/3mtspewsn1FjuTDV2nNvey1cNh0Sig8i5VFvfMW3aNLp27cqVK1d46KGHmDVrFiVLlrzjc26u77Vr12jbti2zZs0iderULF26lKpVqyZFdNtdu3aNd999l1GjRuFyuUifPj2tWrWibdu2VKpU6Y73iL9+/TpHjhxh06ZNbNy4kY0bN7Jly5YHupVp7ty5qVatWvyjWLFiBAQkfJCejmPnU42dywm1nTZtGh06dKBgwYLs3buXoKAguyN5FSfUWO5MNXYcNRySmg4i51Jtfcvu3btp1qwZu3fvJkWKFIwbN44XXnjhtr8wh4WFkSNHDpYuXcpbb73F9u3bSZs2LYsWLeKxxx5L4vT2W79+Pe+88w6//PJL/LKsWbNStGhRYmNjiY6OJioqiqtXrxIREUFERASXL1++5b7y5s1LmjRpCAgIIFWqVOTOnfsfj+zZswNw8uRJjhw5wuHDhzlw4AAbNmzgwoUL/9hXtmzZeOaZZ2jUqBG1a9e+78tbdBw7n2rsXL5e2+joaIoWLcqhQ4f48ssvef755+2O5HV8vcZyd6qx46jhkNR0EDmXaut7Ll++TI8ePfj8888BaNOmDZ9++ilp06b9z7ZLlixh5MiRrFixAoA8efIwf/58ypQpk5SRvc6OHTuYPn06X331FSdPnrzjtgEBAWTNmhXTNKlQoQIVKlSgfPnyZMyYMUGvHRsby65du1izZg1r1qxh9erVnDhxIn59vnz56NevH+3atbvnUQ86jp1PNXYuX6/t5MmT6dy5M0WLFmXXrl0EBgbaHcnr+HqN5e5UY8dRwyGp6SByLtXWd02fPp0uXbpw+fJlihQpwvTp06lQoQKXLl1i/vz5TJ8+nR9//BGAkJAQ+vTpQ/fu3UmZMqXNyb2Hy+Xi0KFDHD58mODgYIKDg0mWLBnJkycnffr0pE+fnjRp0tzxkgtPZNi2bRvff/8933zzDfv37wegYsWKvP/++9StWzf+A3xsbCzh4eEcOnSIP//8k6NHjxIREUF4eHj8NoZhxD9iY2M5f/48586dIzIyEsMwyJUrFyVLloxvnNwYiSHeTT+rncuXa3vt2jWKFCnC0aNHmTFjBi1btrQ7klfy5RrLvVGNHUcNh6Smg8i5VFvftnfvXpo1a8bOnTuBuKH558+fJyoqCoDkyZPTvXt3+vTpk+C/yEvSiYmJYcaMGfTu3Zvw8HAA0qZNS758+YiKiuLIkSNcvXrVo6+ZK1cuChQoQNasWcmWLRtZs2YlT5485MuXjzx58pA1a1ZSp06dqE0XuTv9rHYuX67tJ598Qrdu3ShRogTbtm3T6Ibb8OUay71RjR1HDYekpoPIuVRb33flyhXee+89pk2bxrlz5wCoXLkyzz33HNWqVbvrxJLifS5evMiECROYOHEif/755z/WZc6cmQIFCpA/f37y5s1LxowZiYmJiR+p4HK54h+GYZAhQwYyZsxI2rRpuX79OocPH2br1q1s3LgRy7KIjIy8a56UKVOSJUuW+EfWrFnJkiULoaGhFCpUiMcee4wMGTIkyn8LiaOf1c7lq7W9cuUKhQsXJiwsjNmzZ9O0aVO7I3ktX62x3DvV2HHUcEhqOoicS7V1jujoaMLCwsiSJQupU6cGVF8nCA8P59SpUwQEBJA3b95bztWR0DrHxMRw4MABTpw4wV9//cWpU6cIDw/n6NGjHDp0iGPHjnH69Om7jqoICgqicePGDBkyhAIFCtx3Drk7HcvO5au1/fDDD3nvvfcoU6YMmzdvfqA77Tidr9ZY7p1q7Di3bTjoHjwi4reCg4PJly+f3THEw7Jnz55o8ywEBgZStGhRihYtetttXC4XkZGRnD59+h+PU6dOceLECbZv386aNWuYPXs28+fP59VXX+Xdd98lXbp0iZJZROwXHh7O4MGDARgxYoSaDSLiN9RwEBER8SDDMEibNi1p06a97eiFsLAw+vTpw5dffsmwYcOYPn06gwYNok2bNvd9i08R8X6vv/46ly5don79+tSqVcvuOCIiSUbtVRERkSSWM2dOvvjiCzZu3EjFihU5efIkHTp0oGjRosydOxcfuNxRRO7RkiVLmDFjBilTpuSjjz6yO46ISJJSw0FERMQm5cuX59dff+XLL7+kaNGiHDp0iKZNm9KzZ081HUQcIDIyki5dugDQv39/zdkiIn5HDQcREREbBQQE8Pzzz7Nz507Gjx9P8uTJGTduHF27diU2NtbueCLyAN577z2OHj1K2bJlee211+yOIyKS5NRwEBER8QJBQUF07dqV+fPnkyJFCiZMmMBLL72kpoOIj9q4cSNjx44lMDCQzz77jKAgTZ0mIv5HDQcREREvUq9ePRYuXEjKlCn57LPP6NOnj92RROQ+RUdH8+KLL+JyuXjttdd45JFH7I4kImILNRxERES8TO3atZk3bx5BQUEMHTqUzz//3O5IInIfRowYwY4dO8ifPz/9+/e3O46IiG3UcBAREfFC9erVY8KECQB0796dXbt22ZxIRO7F/v3745sMEydOJFWqVDYnEhGxT4IvJjNNszrQ172PUUAK4BXgKtDOsqxjpmkWBya6t+lrWdZy0zTTANOBLMAiy7KGPNhbEBERcaaOHTuyZs0avvjiC1q0aMHGjRv1y4uIF3O5XHTt2pVr167x/PPPU6dOHbsjiYjYKkEjHEzTTAH0Ap60LOtxYDHwOlCduCZEX/emg4AXgHrAAPeyTsAPlmU9ClQ3TTNXgtOLiIg43Lhx4yhatCi7du3i1VdftTuOiNzBvHnzWL58ORkyZGDkyJF2xxERsV1CL6moAlwBFpqmOQ8oD+yyLCvKsqx1QEn3dqGWZe23LOsicNY0zcxAZWCZe/1PQKWExxcREXG2NGnSMGvWLJInT87kyZNZsGCB3ZFE5BYuX77M66+/DsD//vc/smTJYnMiERH7JbThkA3IDzQAJgEfABdvWh/o/mrctCwCyAiE3LTtjWUiIiJyG6VKlWLIkLgrEDt37syZM2dsTiQi/zZs2DCOHDlCqVKleOmll+yOIyLiFRI6h8MFYK1lWVGmaa4EvgSO37Q+xv315puHhwDngPNAOvc+QoDD/965aZqdgc4QN1GWL17/Fh0dTVhYmN0xJBGots6m+voHX6xzkyZN+Pbbb1m/fj0vvPACn376qd2RvJov1ljujTfW9ujRo/FNwX79+vHXX3/ZnMi3eWONxbNUY2fJmTPnbdcltOGwEXjV/e+yxF0iUdw0zWTEXV6x3b0u3DTNwsBfQEbLss6YprkeqAt8BtQGOv5755ZlTSJu5ASAK4EZbRUWFnbH//Diu1RbZ1N9/YOv1vnrr7+mVKlSLFy4kOeee47mzZvbHclr+WqN5e68rbaxsbG0adOGa9eu0bp1a5o0aWJ3JJ/nbTUWz1ON/UeCLqmwLOsssMA0zdXAEOImiRwNrAIGuh8AfYDPgaXA++5lk4H6pmmuA36xLEutLRERkXtQoEABRowYAUDXrl0JDw+3OZGIjBgxglWrVpEtWzY++ugju+OIiHgVw+Xy+gEEXh/wVtS1cy7V1tlUX//gy3V2uVzUq1ePn376iYYNG/L9999jGMbdn+hnfLnGcmfeVNuFCxfyzDPP4HK5WLBgAQ0aNLA7kiN4U40lcajGjnPbDyIJnTRSREREbGAYBlOmTCFdunQsWLCA6dOn2x1JxC/t37+f1q1b43K5GDBggJoNIiK3oIaDiIiIj8mdO3f80O2ePXty7NgxmxOJ+Jdr167RsmVLIiMjadasGe+9957dkUREvJIaDiIiIj6oXbt2NGjQgIiICFq2bEl0dLTdkUT8xrBhw9iyZQsFChRg8uTJuqxJROQ21HAQERHxQTcurciZMye//vorb7/9tt2RRPzCoUOHGDRoEABTpkwhffr0NicSEfFeajiIiIj4qCxZsjBr1iyCgoIYNWoU8+bNszuSiOO98sorXL16ldatW1O9enW744iIeDU1HERERHxYlSpVGDp0KAAdOnTgyJEjNicSca6FCxeycOFC0qVLF3+LWhERuT01HERERHzca6+9xjPPPENERARt2rTh+vXrdkcScZzLly/Ts2dPAAYMGEBoaKjNiUREvJ8aDiIiIj7OMAw+++wzcuTIwbp16+KvLxcRzxk8eDCHDx+mdOnSdOvWze44IiI+QQ0HERERB8icOTNffvklhmHQv39/1q1bZ3ckEcfYs2cPw4YNA2D8+PEEBQXZnEhExDeo4SAiIuIQtWrV4s033yQ2NpY2bdoQERFhdyQRn3f16lVatWpFVFQUHTt2pGrVqnZHEhHxGWo4iIiIOMiAAQMoV64cR44coUuXLrhcLrsjifi0Xr16sW3bNgoWLMjo0aPtjiMi4lPUcBAREXGQZMmS8c0335A6dWpmzpzJ9OnT7Y4k4rMmT57MJ598QrJkyZgxYwZp06a1O5KIiE9Rw0FERMRhihQpwscffwxAt27d+PPPP21OJOJ7Fi5cyMsvvwzApEmTKF++vM2JRER8jxoOIiIiDtS+fXuaNWtGZGQknTp10qUVIvfhl19+oVmzZsTExNCnTx/atWtndyQREZ+khoOIiIgDGYbB+PHjyZQpEytXrmTq1Kl2RxLxCZs2baJBgwZcu3aNLl26MHDgQLsjiYj4LDUcREREHCpLlix89NFHQNzEdydPnrQ5kYh327VrF0888QSRkZG0bt2a8ePHYxiG3bFERHyWGg4iIiIO1rp1a5566ikuXLhA9+7d7Y4j4rX27dtH7dq1OXfuHA0aNGDatGkEBOijsojIg9BPUREREQczDIMJEyaQJk0avvvuO5YsWWJ3JBGv88cff1CjRg3Cw8OpVasWs2bNIjg42O5YIiI+Tw0HERERh8udOzcffPABAD169ODq1av2BhLxIvv376dGjRqcPHmSGjVqsGDBAlKkSGF3LBERR1DDQURExA/07NmTEiVKcPDgQYYPH253HBGvsG/fPqpXr86JEyd4/PHHWbhwIalSpbI7loiIY6jhICIi4geCg4MZP348AIMGDeLQoUM2JxKx1549e+KbDdWrV2fRokWkTp3a7lgiIo6ihoOIiIifePzxx2nTpg1Xr17l1VdftTuOiG127dpF9erV4+ds+OGHH0iTJo3dsUREHEcNBxERET8yfPhw0qZNy4IFC1i0aJHdcUSS3Pbt26levTqnTp2ibt26uoxCRCQRqeEgIiLiR0JDQxkwYAAQN6/DlStXbE4kknR+//13atasyZkzZ3jiiSeYP38+KVOmtDuWiIhjqeEgIiLiZ7p3707JkiU5dOgQQ4cOtTuOSJKYN28ejz76KGfPnuXpp59m3rx5uhuFiEgiU8NBRETEzwQFBcVPIDlkyBAOHjxocyKRxBMbG8sHH3zAs88+y6VLl2jdujVz585Vs0FEJAmo4SAiIuKHqlWrRtu2bbl27Ro9e/bE5XLZHUnE444fP06dOnXo378/AQEBDB8+nK+++orkyZPbHU1ExC+o4SAiIuKnhg0bRrp06Vi8eDELFiywO46IR82aNYuSJUuycuVKsmTJwuLFi3njjTcwDMPuaCIifkMNBxERET+VLVs2Bg4cCMArr7zC5cuXbU4k8uAiIiJo27YtLVq04MKFCzz99NPs2LGDevXq2R1NRMTvqOEgIiLix15++WVKly7NkSNHGDx4sN1xRB7ImjVrKF26NNOnTydlypR8+umnLFy4kGzZstkdTUTEL6nhICIi4seCgoL45JNPgLhLLPbv329zIpH7FxMTQ9++falevTpHjhyhXLlybN26lS5duugSChERG6nhICIi4ueqVKlChw4diIqKokePHppAUnzKhQsXaNCgQfzlQe+++y6//fYbRYsWtTmZiIio4SAiIiIMGTKEkJAQli5dyg8//GB3HJF78scff1CpUiWWLFlCpkyZWL58OQMHDiQ4ONjuaCIighoOIiIiAmTNmpX3338fgLfeeovr16/bnEjkzn788UcqVKjAvn37KFmyJJs2baJGjRp2xxIRkZuo4SAiIiJA3ASS+fPnZ/fu3UydOtXuOCK35HK5GDFiBE8//TQRERE0btyYX3/9lfz589sdTURE/kUNBxEREQEgefLk8Xeq6NevH5cuXbI5kcg/Xb16lVdeeYXevXsTGxvL+++/z5w5c0iTJo3d0URE5BbUcBAREZF4zZs3p3z58oSHhzNq1Ci744jEO3bsGNWqVeO7774jVapUzJkzhw8++ICAAH2cFRHxVvoJLSIiIvEMw2D48OFA3G0y//rrL5sTicCqVasoV64clmWRK1cufv31V5o0aWJ3LBERuQs1HEREROQfHn/8cRo0aEBkZCT9+/e3O474MZfLxdixY6lVqxanT5+mdu3aLFmyhNKlS9sdTURE7oEaDiIiIvIfQ4YMISAggEmTJrF3716744gfOn/+PC1btuSVV14hJiaG3r17s2TJEjJkyGB3NBERuUdqOIiIiMh/FC9enBdffJGYmBjeeecdu+OIn/n5558pVaoUs2bNInXq1MycOZNhw4YRFBRkdzQREbkPajiIiIjILX3wwQekSpWK77//ntWrV9sdR/zAtWvXePPNN6lVqxbHjx+nUqVKbNu2jRYtWtgdTUREEkANBxEREbml0NBQevfuDUDXrl2JioqyOZE42e7du6lUqRLDhw/HMAzef/991qxZQ8GCBe2OJiIiCaSGg4iIiNzWW2+9RaFChdi1axdDhw61O444kMvlYvz48ZQrV47ff/+dAgUKsHbtWj744ANdQiEi4uPUcBAREZHbSpkyJZMmTQJg4MCBmkBSPCoyMpKWLVvSvXt3rl69SocOHfj999+pXLmy3dFERMQD1HAQERGRO6pRowYdOnQgKiqKzp07Exsba3ckcYC9e/dSoUIFZs2aRdq0aZk1axaff/45adOmtTuaiIh4iBoOIiIiclcjRowga9asrFmzhpEjR9odR3zcd999R4UKFdizZw/Fixdn06ZNNGvWzO5YIiLiYWo4iIiIyF1lzJiRKVOmANCnTx8sy7I5kfii69ev89Zbb9GkSRP+/vtvmjdvzoYNGyhatKjd0UREJBGo4SAiIiL3pH79+vTo0YPr16/TqlUr/v77b7sjiQ8JDw+nTp06DBs2jMDAQEaNGsXMmTNJkyaN3dFERCSRqOEgIiIi92zYsGGUKlWKAwcO0KVLF1wul92RxAesWrWKsmXL8ssvv5A9e3ZWrlzJa6+9hmEYdkcTEZFE9ED3GjJNsxUw1rKsLKZptgBeAa4C7SzLOmaaZnFgovt1+lqWtdw0zTTAdCALsMiyrCEP9hZEREQkqaRIkYIZM2ZQoUIFvvnmG0qWLMnbb79tdyzxUleuXKFfv36MGjWK2NhYqlevzowZM8iePbvd0UREJAkkeISDaZoBQFPgmGmawcDrQHWgr/sBMAh4AagHDHAv6wT8YFnWo0B10zRzJTSDiIiIJL3ixYvz1VdfAfDOO+8wduxYmxOJN1q7di1ly5ZlxIgRQNzcHz/99JOaDSIifuRBLqloDcwBYoHCwC7LsqIsy1oHlHRvE2pZ1n7Lsi4CZ03TzAxUBpa51/8EVHqADCIiImKDRo0axTcaXnnlFQYNGmRzIvEWYWFhPPfcc1SrVo19+/ZRvHhxfvvtNz788EOCgh5ocK2IiPiYBP3UN00zEGgONAJ6ASHAxZs2CXR/vfnCvAgg47+2vbHs3/vvDHQG6N69O3Xq1ElITFtFR0cTFhZmdwxJBKqts6m+/kF19oxnn32Wq1ev8tZbb/Huu+9y8uRJ3nrrLa+4Ll81Tlxnz57l2LFjnDt3Lv5x9uxZDhw4wOrVq7ly5QrJkyfn5Zdfpnv37qRIkcJj9VBtnU81dj7V2Fly5sx523UJbTM/B8yyLCvWNE2A80C6m9bHuL/G3rQsBDh307YX3MsO/3vnlmVNAia5v/XJ2ajCwsLu+B9efJdq62yqr39QnT2nd+/e5MqVi+eff55x48ZhGAZjxowhIMDeealVY8+5cuUKW7duZcOGDfGPw4cP3/E5jRo1YtSoUeTPn9/jeVRb51ONnU819h8JbTgUB8qapvkccZdTdAaKm6aZDCgPbHdvF26aZmHgLyCjZVlnTNNcD9QFPgNqAx0f5A2IiIiIvVq1akWqVKlo3rw5H3/8MZGRkUyePJnAwMC7P1m8SmxsLPv37/9Hc2Hbtm1cv379H9ulTp2aIkWKkCVLFrJkyULmzJnJkiULuXLl4vHHHydfvnz2vAEREfEqCWo4WJb11o1/m6ZpWZb1mvsuFauIu0tFW/fqPsDn7tfp5142GfjKNM0OwELLsjSWRkRExMc988wzLFy4kEaNGjF16lTOnDnDN998Q5o0aeyOJndw7do1fv75Z3799Vc2bNjAxo0buXDhwj+2MQyDhx9+mIoVK8Y/SpQooYaSiIjcleED98/2+oC3omFCzqXaOpvq6x9U58Szdu1annnmGc6dO0e2bNl45ZVXaNOmDXny5EnSHKrx7V2+fJkff/yRuXPnsnDhQv7+++9/rA8NDf1Hc8E0TdKmTWtT2v9SbZ1PNXY+1dhxbjt5k6YKFhEREY959NFH+e2332jTpg2WZdGnTx/69OnDo48+SuvWrWnQoAE5c+ZMkoklL126xK5du9iyZQs7d+7k+PHjBAQEkCpVKjJkyEDFihWpVq0aefPmTfQsdrp+/TqbN29mxYoVrFy5knXr1nH16tX49aVLl6ZOnTrxDYZcuXJ5xcSfIiLi+zTCIZGoa+dcqq2zqb7+QXVOfC6Xi6VLlzJ16lQWLlzIlStX4telSZOGFClSEBwcTGxs3PzSoaGh5MyZkxw5cvzj641/Z86c+T+/BF+6dImTJ0/+43HixAkOHjzIli1bOHz4MPfyOSdfvnzUq1ePunXrUqtWLdKnT+/Z/xhJLDY2lh07drBy5UpWrlzJqlWr/jOKoXz58jRt2pQmTZpQsGBBm5ImjI5f51ONnU81dpzbdqnVcEgkOoicS7V1NtXXP6jOSevvv/9m/vz5zJw5k3Xr1v1njoB7kSxZMkJDQ0mXLh3Xrl3j5MmT//kl+t+CgoIoVqwYpUuXpmzZsuTJk4eAgAAuX75MWFgYa9euZe3atf/IExgYSKVKlahXrx716tWjTJkyJEuW7L7zJqXIyEj++OMPNm7cyMqVK/n55585c+bMP7YpXLgwNWvWpFatWlSvXp0sWbLYlPbB6fh1PtXY+VRjx1HDIanpIHIu1dbZVF//oDrbx+VyceHCBaKjo4mOjiYwMJCYmBhOnjxJWFgYJ06c+MfXG/8+f/78f/aVPHlyQkND4x85cuQgNDSUvHnzki1bNh577LG7NgtiYmLYsmULS5cuZenSpfz222/ExMTErw8KCqJw4cKUKlWKqlWrUq1aNUqWLJnkEya6XC7Cw8PZs2cPe/fu/cfj2LFj/9k+V65c1KpVi5o1a1KjRg1y586dpHkTk45f51ONnU81dhw1HJKaDiLnUm2dTfX1D6qz77l8+TInT54kMjKS4OBgQkNDCQkJue1cAwmtcUREBD///DNLly5l5cqV7N+//z+XZaRLlw7TNClcuDAFChSgQIEC5MuXj9y5c5MlSxYCAgIS9B6vXr3Kzp07OXz4MEePHv3HY//+/Vy8ePGWz0uWLBmFCxfm4YcfpkaNGtSsWZNChQo5dh4GHb/Opxo7n2rsOJo0UkRERHxXqlSpkmSugfTp09OoUSMaNWoExDU69u7di2VZ8ZdgHDp0KH5+hH8LDg7moYceip+AsWrVqjz00EP/+eU/KiqKnTt3YllW/GPHjh1cv379ttkyZMhAsWLFeOihh+K/PvTQQ+TLl4+gIH2kExER76Ozk4iIiMhtpEqVikceeYRHHnmEzp07A3F/mfv99985dOgQBw8e5M8//+Tw4cOEhYVx9uxZduzYwY4dO/jss88AyJQpE6ZpkiVLFk6dOsXx48c5cOAAUVFR/3gtwzAoXrw4RYoUIU+ePP945M+fnyxZsjh21IKIiDiTGg4iIiIi9+HG3TNu5dKlS/z+++9s2LCB9evXs3btWk6ePMnSpUv/s22RIkUwTTP+UbZsWdKkSZPY8UVERJKMGg4iIiIiHpI6dWqqVq1K1apVgbjJHg8dOsTu3bs5ffo02bJlI2fOnOTLl8/nb78pIiJyN2o4iIiIiCQSwzDiJ5YUERHxNwmbRllERERERERE5A7UcBARERERERERj1PDQUREREREREQ8Tg0HEREREREREfE4NRxERERERERExOPUcBARERERERERj1PDQUREREREREQ8Tg0HEREREREREfE4NRxERERERERExOPUcBARERERERERjzNcLpfdGURERERERETEYTTCQUREREREREQ8Tg0HEREREREREfE4NRxERERERERExOPUcBARERERERERj1PDQUREREREREQ8Tg0HEREREREREfE4NRw8wDRNw+4MInJ/TNN82P1VPwdFfJzOwyK+R+dhEf9guFwuuzP4JNM0KwEhlmX9aHcW8SzTNNMBPwHdLMuy7M4jnmWaZnmgD1AUeMyyrDM2R5JEYJpmNaAlMBPYbllWhGmahmVZOuk5hM7DzqXzsLPpPOwfdB6WG9RRvA83/oJimmYnYCHQwP2BR3zcTbU1gGtAaqCmaZohduYSz7ipvgOB4cAy4EsgnZ25JHGYppkPeA3YAjwODALQhxzfp/Owc+k87Gw6D/sXnYflZmo43J+s7q9LgMeAXwDTNM00tiUST8kM8T8IswCHgRJAWQ3VdYQbx+7XlmVVtyzrU6ACcR9oNZzTeUKBAMuypliWNRAoZppmdVCtHUDnYefSedjZdB72LzoPSzwV/B6YplndNM2VwBDTNOsD5y3L2gOcBNIS96FHfNBNtR1mmmZ90zTTWJZ1HJgH/AzUw30yFN9zi2P3yE2rLaAdgGVZsXbkE88wTbOjaZqLTNPMBWBZ1m/AadM0a7k3GQO86V6nWvsgnYedS+dhZ9N52D/oPCx3oobDvWlK3PCv4UAZoLN7+WbgBFDQNM2UpmkGgSav8jH/rm070zRTAKZlWdOI68ivNk3zCdsSyoP4d3073bRuDRDurrf4KPe13rWBYKCGaZrJTNMMJO7679qmaSazLGsBcbU27cwqD0TnYefSedjZdB52OJ2H5W7UcLiDmz6wHAFSWZa1m7iJTx4xTbOEZVlXgKVAJuA74F1NhuIb7lDbKkAu4KhpmrOAAsQdJ1ttCSoJcof6mqZpFnevSwU8bFnWVTsyyoNz/7y9aFlWK+AloCZQyLKsGGAtccduF9M0UwLXgb32pZWE0HnYuXQedjadh/2DzsNyL9Rw+JcbQ4Eg7jpC93VGp4EQ0zQzWZZ1gLgD6FH3ZqWAhsAm4EN9yPFe91jbVcATwFFgtWVZ1YGXLcv6y47Mcu/u49it5t5sGVDUNM38SZ9WEuoWdTbc/z4M7AIamaaZyrKsE8QN4SwNfA+cAS4ldV65f6Zp3rim29B52FnusbY6D/uo+zh2dR72Ybeos87DckdBdgfwFqZpPgl0A66apjkV+NmyrMuWZcWaprkJeAaoCCwGLvL/zZpwoL77oBIvdJ+1jQSuW5Y168bz3dehiZdKwLF748ToMk2ztmVZOvn5gDvU+eZfLqcDHwCF3cM5DxD3F5fUlmVFJHVmuT/u67vbACdM0xxJ3PwM6Dzs++6ztjoP+5gEHLs6D/ugO9RZ52G5IzUcAPcB8RIwibgPLo2B1e51hYibRfcQUMs0zdxAHWAlgGVZ2+3ILPcmAbWtCyy3J63crwQeuytuPF8fcnzDXepcmLghnLMty/rLNM2jwALi/kr6tmVZFwF9yPFypmkGEzd53DigKHH1XgxscNdY52EflYDa6jzsQxJ47Oo87GPuUuciQA10Hpbb8NtLKkzTTG7+/72dcwM7gDBgH3GT2mQ0TbMC0A+4bFnWDOAT4q4r/MGyrE+SPLTckwes7SL3rZrES3ng2FV9fcB91Pld4JJlWefc14hWBQZbltVWf/H2bv+qcRHgpGVZq4BpQB7gSTNuhvMBQKTOw77jAWur87CX88Cxq/r6gPuo8/vA3zoPy+34ZcPBNM2WwG7gQ4i/5mgl8BzwO3HXmz0P1AI6WJb1rXu7g0A/y7KmJn1quReqrbOpvv7hPuvc0bKsr9zbXQEaWZY1IelTy/24RY13ATlN02xhWVYUcTVOSdx1/O0ty5rt3k7HspdTbZ1N9fUP91nnDpZlfePeTudh+Q+/aziYcbfeSYX7tjymaT4FYFnWz8QN4fvYsqy2xN37GcuyYkz3bbbc32syKi+l2jqb6usfElrnmyatum5LcLlnt6hxffeq0cRd97uEuGu8cwIZLcu65r6sBtCx7M1UW2dTff1DAuocZZpmoM7DcjuGy+X8Y980zbzAW8APwDbLso67lz8NtATauSe2eRMIJO4vaT2Ac5Zl9bQpttwD1dbZVF//oDo7311q3Iq4Gse4l6W3LCvCNM2BwFT3X0bFS6m2zqb6+gfVWRKT40c4mKaZjLhri44B2YDxN61eSdyQoM7u778mbmKbfsB6fZD1bqqts6m+/kF1dr57qPEp4AX3tgZQyDTNZUCEPsh6N9XW2VRf/6A6S2Jz7AgH0zSbAFmJu8fvFCvuPs6YpjkF2GNZ1gj3QZOXuOuTNgDLLMvaa5pmMvf1SeKFVFtnU339g+rsfAms8WLiJgcNtCwr0pbgcleqrbOpvv5BdZak4rgRDqZpZjZN83ugKXG3bakFhJum2cG9SX+gqWmaWdzXkqUDKgGNABeAPsh6J9XW2VRf/6A6O98D1LgxEGBZ1hV9kPVOqq2zqb7+QXWWpOa4hgMQC3xmWVYrYCBQHJgLPGyaZmHLso4CS4F6ZtyEcuWAXpZl1bQsa59tqeVeqLbOpvr6B9XZ+RJa4xqWZf1hW2q5F6qts6m+/kF1liQVdPdNfM554oYGYVnWGdM0swIXgf1AL6ALkJG4CVGuA7o9j+9QbZ1N9fUPqrPzqcbOpdo6m+rrH1RnSVJOnsPBANID31iW9ZR72STi7hmbjLhbvfxt6RY9Pke1dTbV1z+ozs6nGjuXautsqq9/UJ0lqThxhMPNgoC1pmmWA54ApgB/WJZ13t5Y4gGqrbOpvv5BdXY+1di5VFtnU339g+osic6xIxwATNOsBywk7pYuX1uWNd3mSOIhqq2zqb7+QXV2PtXYuVRbZ1N9/YPqLEnB6SMczgJ9gLGa1dxxVFtnU339g+rsfKqxc6m2zqb6+gfVWRKd0xsOmy3LsuwOIYlCtXU21dc/qM7Opxo7l2rrbKqvf1CdJdE5+pIKEREREREREbFHgN0BRERERERERMR51HAQEREREREREY9Tw0FEREREREREPE4NBxERERERERHxOKffpUJEREQSkWma+YBD7m+vAaeB1cB7lmUdusPzUgFvAocty5qWyDFFRETEBhrhICIiIp6wFegCrARaA7+appn1DtunAt4H2id+NBEREbGDRjiIiIiIJ5xwj1SYZprmNaAT8JJpmqWA2kBK4E/gXcuy5gE37v3+uGmaLqA/MMj9aAWkBn4CulqWdTpJ34mIiIh4hEY4iIiIiKctcX8tDWwi7tKJd9zLvjRNMwXQx/39HuIaDHPc2/QCFgJjgCeBT5MmsoiIiHiaRjiIiIiIpxnury6gOHENhWQ3rc8HLHP/+5RlWTMBTNOc6l720k3b1k28mCIiIpKY1HAQERERT6vn/rodGACsIG7EQhfgaSAFcc2IfzOA60B9IMa9TKMxRUREfJQaDiIiIuIJOUzTbA88DrQDwoGj7nWpiBvVUPWm7S8CsUAh0zTbAGuJu5SinPv5y4kbHZGf/x8NISIiIj5EfzUQERERTygLTCRugshvgMrAV8BMoCTwLLD0xsaWZUUDw4EQ93bVgMHuZdWAccTN4bAqqd6AiIiIeJbhct1qRKOIiIiIiIiISMJphIOIiIiIiIiIeJwaDiIiIiIiIiLicWo4iIiIiIiIiIjHqeEgIiIiIiIiIh6nhoOIiIiIiIiIeJwaDiIiIiIiIiLicWo4iIiIiIiIiIjHqeEgIiIiIiIiIh73f7rpeZYOmkTyAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1296x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ETS analysis: try to discern trend and seasonal components\n",
    "# showing the trend and seasonality trends seperately\n",
    "ts_trend, ts_seas = extract_trend_and_seasonality(ts=series, freq=mseas)\n",
    "\n",
    "plt.figure(100, figsize=(18, 5))\n",
    "ts_trend.plot()\n",
    "plt.title(\"trend component\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "86acec7a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#seasonal component plot\n",
    "plt.figure(100, figsize=(18, 5))\n",
    "ts_seas.plot()\n",
    "plt.title(\"seasonal component\");"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3799bf29",
   "metadata": {},
   "source": [
    "Next we have normalized all the features and now we will proceed to train and evaluate some models"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b90b0a9b",
   "metadata": {},
   "source": [
    "# TFT MODEL"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "54790434",
   "metadata": {},
   "source": [
    "First we train the Temporal fusion transformer(TFT). We check the forecast for both the 7 days and 30 days window"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "af59add4",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Using a 30 day window or lag\n",
    "model = TFTModel(   input_chunk_length=30,\n",
    "                    output_chunk_length=1,\n",
    "                    hidden_size=64,\n",
    "                    lstm_layers=2,  #LSTMLAYERS,\n",
    "                    num_attention_heads=4,\n",
    "                    dropout=0.1,\n",
    "                    batch_size=64,\n",
    "                    n_epochs=300,\n",
    "                    #likelihood=QuantileRegression(quantiles=QUANTILES), \n",
    "                    random_state=42, \n",
    "                    force_reset=True,\n",
    "                 torch_device_str =\"cuda\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "f6e31e25",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "96f1f677aa8d4d7497be36ee95cb87e5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<darts.models.forecasting.tft_model.TFTModel at 0x1eee97ca908>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# training\n",
    "model.fit(  ts_ttrain, \n",
    "            future_covariates=covF_ttrain,\n",
    "            verbose=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7220895c",
   "metadata": {},
   "source": [
    "Now we proceed to test our model using the historical_forecasts function below"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "93be8d24",
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture\n",
    "#Testing our model using historical_forecasts on ts_ttest \n",
    "preds = model.historical_forecasts(\n",
    "    ts_ttest, start=0.3, forecast_horizon=1,future_covariates=covF_ttest,retrain=False\n",
    ")\n",
    "\n",
    "# scale back:\n",
    "preds = scaler.inverse_transform(preds)\n",
    "actual = scaler.inverse_transform(ts_ttest)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b2a9bc33",
   "metadata": {},
   "source": [
    "Our model has been evaluated on the test data. Let us see the plots of the actual and predicted values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "b6bb8c7d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'TFT (window of 30 days), MAPE : 1.81%)')"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plotting the actual and predicted price values\n",
    "\n",
    "#plt.title(\"Naive Forecast (MAPE: {:.2f}%)\".format(mape(ts_test, ts_predN)))\n",
    "actual[41:].plot(label=\"actual\")\n",
    "preds.plot(label=\"predicted\")\n",
    "plt.title(\"TFT (window of 30 days), MAPE : {:.2f}%)\".format(mape(actual,preds)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "a12e4592",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MAPE: 1.814260520040989\n",
      "rmse: 222.79033\n",
      "mae: 168.94402\n"
     ]
    }
   ],
   "source": [
    "#metrics of time series forecast\n",
    "print(\"MAPE:\",mape(actual,preds))\n",
    "print(\"rmse:\", rmse(actual,preds))\n",
    "print(\"mae:\",mae(actual,preds))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "689dcce7",
   "metadata": {},
   "source": [
    "We get an MAPE of 1.56% on the TFT model with a sliding window of 30 days. Next we will check a sliding window of 7 days."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "a654198d",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Using a 7 day window or lag\n",
    "model = TFTModel(   input_chunk_length=7,\n",
    "                    output_chunk_length=1,\n",
    "                    hidden_size=64,\n",
    "                    lstm_layers=2,  #LSTMLAYERS,\n",
    "                    num_attention_heads=4,\n",
    "                    dropout=0.1,\n",
    "                    batch_size=64,\n",
    "                    n_epochs=300,\n",
    "                    #likelihood=QuantileRegression(quantiles=QUANTILES), \n",
    "                    random_state=42, \n",
    "                    force_reset=True,\n",
    "                 torch_device_str =\"cuda\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "63527c76",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5fc99146c172487bbedcca79990b37dc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<darts.models.forecasting.tft_model.TFTModel at 0x1eee9955788>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# training\n",
    "model.fit(  ts_ttrain, \n",
    "            future_covariates=covF_ttrain, \n",
    "            verbose=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "cec67ac0",
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture\n",
    "#Testing our model using historical_forecasts on ts_ttest \n",
    "preds = model.historical_forecasts(\n",
    "    ts_ttest, start=0.1, forecast_horizon=1,future_covariates=covF_ttest,retrain=False\n",
    ")\n",
    "\n",
    "# scale back:\n",
    "preds = scaler.inverse_transform(preds)\n",
    "actual = scaler.inverse_transform(ts_ttest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "993d907e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'TFT (window of 7 days), MAPE : 1.93%)')"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYoAAAEsCAYAAADdO/TjAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAABtTElEQVR4nO2dd5wURdrHv7WZDeyyATaQs+RQCCgoKopwivnM8VTu9dQznJ7hDK/vGU49c7hDxXSeARWziIgEEcRSEZCcYQOwbGbzTr9/VM/O7DJsYmemd6nv57Ofmamu7v51zWw/XfU89ZSwLAuDwWAwGA5FSLAFGAwGg8HZGENhMBgMhgYxhsJgMBgMDWIMhcFgMBgaxBgKg8FgMDSIMRQGg8FgaBBjKAwACCFmCCGeauY+E4UQGw7jnJYQom9L9z9chBB/F0LkCiFy/HDs+4UQ/2nt43odv4sQYp0QItJf5zD4RggRKYRYL4ToHGwtgcIYCgcjhCjx+nMJIcq8Pl9s34yq6tW7t95nSwhxwOvzRB/niQD+BjzWHH2WZS2xLGtAa11vIBFCdANuBQZZlpXqY/vF9dqx1G7L0YFXezCWZe0BvgWubeo+QojtQohKIURyvfKV9rX1rFd+v11+dL3yK4QQNXa7FNn7n2Zvm2T/Vkvq/Y1vyXUKIf5PCLFaCFEthLi/kboJQojXhRB77b/7623/Vgixz9b8qxDiDK9tw4UQv9kPDjd7lYcLIX6wfy8AWJZVAcwC/tqSa2qLGEPhYCzLinX/ATuB073K3rKrvetdz7KsB+rtBzDcq2yJj1OdAay3LCszENflEHoA+y3L2utro2VZb9Vrx+uArcDPgRTZCG8BM5q5zzbgQvcHIcRQoEP9SkIIAVwK5AGX+zjOMrtdEoBXgPeEEIn2tqx6v8lYy7KWNVOnm83A7cDnTaj7JBAN9ASOBi4VQlzptf3PQJplWR3RBvY/Qog0e9vDwF+A4cDfhBDuh4dbgA8sy9pV71z/BS4/Unp0xlAYAKYCi9wf7KeyW+33GfZT5XX2575CiDyhmSSE2O2133YhxF+EEKuEEIVCiHeFEFFe228TQmQLIbKEEFd5CxBCxAsh3rCf+HYIIf4mhAixt+1wP8kLIS6x9QyyP18thPjI10Ud6phCiMnA10C6/bT7WhPa6HLgDesQqQyEEL2EEIuEEMVCiK+B+k/ts4UQOXa7LBZCDLbLxwgh9gghwrzqniOEWGm/P1oIoeyn4D1CiCe8DvsD0FsI0aMJ+t28CVxW/7p81JsIpKNvrhcI3es8CMuyXOin6w5A72boaBKWZb1uWdaXQHETqp8OPGpZVqllWdvRBqz2d2ZZ1irLsqrdH4FwwN1T6AUssB+WNgHdhRDdgXPQBqi+rt1APjCuRRfWxjCGwgAwFPD2NSwCJtnvj0c/SR9vfz4OWHKoGybwe+BU9D/eMOAKACHEqegntpOBfsDkevs9C8SjbzbHo29m7qdBbz3H+dBTa+SackzLsuajjaP7yfeKQ+yPrb2HfR5fN1Q3/wV+QhuI/+Pgp/Av0dfdGd0reQvAsqwfgf3odnFzCfqGDvA08LT9FNwHeM9dyb7pbUY/BTeV5UBHIcRRQohQ4HzAly/lcuBT4F3782m+DmYbuKuBEvQNtlkIIV4QQrzQ3P0aOmS990Pqne8zIUQ52sguBJS9aQ1wihCiK7pHsgV4BrjdsqyqQ5xrHc1r+zaLMRRtn98LIQq8/tJbcIwE6j6xLQIm2k/0xwGPAsfa247n0DdmgGcsy8qyLCsPfaMZ4dYJvGpZ1hrLsg4A97t38Lph3WlZVrH9NPhP9NCHW4/bMExEDxO4P/vU04RjNofL0MZxm6+N9pPnGOAey7IqLMtajL72WizLmmXrqEBf+3AhRLy9+XW0ccAevpmCNjwAVUBfIUSyZVkllmUtr3f6YvT31xzcvYqTgfVAnSFHIUQ0cB7wX/sm+T4HG75xQogCIAc9lHWWZVmF9rb0er/JAiFEjC8hlmVdZ1nWdc3UfyjmAncIIeKEDpK4Cj0U5X2+04A4YBrwld0jAv0Q8z/AJ8DN6N97MbBVCPGx3Vs8r975WtL2bRJjKNo+71mWleD1l9WCY+Sj/3kAsCxrC/oJcQT6xvwZkCWEGEDjhsI7gqgUcPtJ0gHvcd4dXu+TgYh6ZTuADPu923ClAqHop9xjhXa+xgMrfeho7JjN4TL0zfxQpAP5tgH0PhegjZYQ4hEhxBYhRBGw3Usj6Cf604UQsWiDusSyrGx72x+A/sB6IcSPwnYaexEHFDTzet4ELkL39nz1ks4CqoEv7M9vAVOFECledZbbv7dky7LG2b00N1n1fpMJ9drGX9wIlKF7Nh8DbwO761eyLKvKHs6aIoSYbpftsCxrmmVZo+x9H0Abj8fRv7fpwBNefhhoWdu3SYyhMACsQt+MvFkEnAtE2OO2i9A3zE74vjE3Rjae8WCA7l7vc9FPzj3qbc8EsCxrM9ro3AgstiyrGG2QrgW+83oq9KbBYzYVIcSxaEPwfgPVsoFO9Z6ava/vInTAwGS0YevpPjyA3b7L0DfoS/EMO2FZ1ibLsi5ED1n9A3jffR572Kcv8GtzrsmyrB1op/Y04EMfVS5HG/idQocOz0aP51/oo65jsCwrz7Ksiy3LSrUsazD6/raigV3C0MN59bkXeNmOLBsKKLu3tBvd3m6Ooplt31YxhsIA+snx+Hpli4DrgcX254XADegbc00LzvEecIUQYpA9tHGfe4N9vPeAB+1hgx7oaBPvsXO3HndvZmG9z3Vo4jGbwuXoqJdDOlPtG68C/lcIESGEmIB2rLqJAyrQvoho4CEfh3kDHd0zFJjjLrSd9ym2MSywi93tfzSw3T6/OzS1qesG/AE4sf6TvhAiAzgJ7ZMYYf8NRxspX9FPfkXo8NQo9L0qTAgRZQ8r+qrbRwiRZPfgpqIfJP5ubxsohJgqhOhgH/MSfPi37CCJScCLdtE24EQhRBe0j2mnXS8DSET7fNo9xlAYQI+nD6zn31iEvsG5DcV36JvcYlqA3dV/CliAdsAuqFflBuAA2lH9HXqMflYDeup/9kVjx2wQ+wb1exoednJzETAWHU56H3WHdN5AD0VlAmvxfXOZg+79zKl38z4V+E0IUYJ2bF9gWVa5ve1i4F9edbuheyaNYlnWFsuylI9NlwIrLcuaZ1lWjvsP7dgdJoQY4mOf+rijybz/zvFVUQjxLyHEv3xts3kJPZx0IXC3/f5Se9+Jdru4GQ2sRvsOHgYutizrN/ep0L6hvcA+dDTX+ZZl1Q93fh74s9fD0J3onuxvwEN2W4D+vl+3fU7tHnHo4BXDkYQQ4lr05LObgq3lSEUIsQWYUW+8/1B1O6ON5Ui34RBCvAzMtizrK/8qPbIReu7Er8Bxh5qH094whsJgcAD2E/c/gP6H8LkYDEEjrPEqBoPBnwghFgKDgEuNkTA4EdOjMBgMBkODGGe2wWAwGBrEGAqDwWAwNEi7NBQ5OTkWOumXo/+cotMpOtqCRqfocLI2p+hoKzqdooMGaJeGoqamJfPBAo9TdDpFR0M4RaNTdPjCKdqcoqMxnKLTKToaol0aCoPBYDC0Hq0eHiuljAPmA4OBcUqpNVLKxYALnaTtWrtsEPBvW8M9Sqn5UspYdJ6bFOAzpdQj9jEfQ+d93wlcqZSqbG3dBoPBYPCNP3oUZeg8Md5J1E5SSk1CT4d3LzP4EDoN8BR0pkaAa4DPlVITgElSyq5SypFAmlJqIjr9wbl+0GwwGAyGQ9DqhkIpVa2U2levzL3wR0d0LhbQN/9NSqkiYL+UMhkYD8yzt3+N7kV4l80FjmltzQaDwWA4NAGZmS2lTAE+QqdePsMu9l6JqhCdiTEBKPJRllWvzNc5rsVeaH7GjBlMmzatteT7jaqqKjIzg79MtVN0NIRTNDpFhy+cos0pOhrDKTqdoiMj49BLtQTEUNg9jGOllEejh5xORfss3CSgs27mo3sdBXbZdvRCNR3r1fN1jpnATIDMzEyroYt2CpmZmQ1+OUeajoZwikan6PCFU7Q5RUdjOEWnU3Q0hN+jnqSUYVJK93kK0WmfAXKklP2klB2BRKVULjr98in29snolMneZVOApf7W3BaorrZ46j2LddsbDH82GAyGw8YvPQop5RfoBU8GoPPlXyWldKF7EX+yq92FXhsgDL2iFOjc8/+RUl4JfKqUygQypZTZUsol6Kinx/yhua0x/ye4+TmLaePg80dF4zsYDIaAsnDhQiIiIjjmmJa7VWNjYykpKWm8op/xi6FQStV3EBy08ItSai16PWbvshLgTB91b2tNfe2B7fbyKet3BleHwWDwzcKFC4mNjT0sQ+EUzIS7NkpWrh5y2rEHqqrN8JPBECjOPPNMRo8ezeDBg5k5cyYAc+fOZdSoUQwfPpyTTjqJ7du3869//Ysnn3ySESNGsGTJEq644gref98zayA2NhaAAwcOcNJJJzFq1CiGDh3Kxx9/HJTragizHkUbJStXv9bUwM490MfZvjCDoVURwj/DrU1ZdmHWrFkkJiZSVlbGmDFjOOOMM7jmmmtYvHgxvXr1Ii8vj8TERP74xz8SGxvLX/7yFwBeeeUVn8eLjIxkzpw5dOzYkdzcXMaNG8f06dP9do0twRiKNorbUABsyTSGwmAIFM888wxz5swBYNeuXcycOZPjjjuOXr16AZCY6DOC/5BYlsVdd93F4sWLCQkJITMzkz179pCamtrq2luKGXpqo2Tt97zfHPwQbIMhoFiWddh/u3fvPqisMRYuXMj8+fNZtmwZv/76KyNHjmT48OFNevoPCwvD5XLV6q+s1JmI5syZw759+/jpp59YuXIlXbp0oby8/PAaqJUxhqKNUrdHYXwUBkMgKCwspFOnTkRHR7N+/XqWL19ORUUFixYtYtu2bQDk5empXnFxcRQXF9fu27NnT3766ScAPv74Y6qqdMKKoqIiOnfuTHh4ON9++y07duwI8FU1jjEUbZDKKot9BZ7PW7IOWdVgMLQip556KtXV1QwbNox77rmHcePGkZKSwsyZMzn77LMZPnw4559/PgCnn346c+bMqXVmX3PNNSxatIijjz6aH374gZiYGADOPvtslFJIKXnrrbcYOHBgMC/RJ8ZH0QbZU29u+hYz9GQwBITIyEi+/PJLn9umTp1a53P//v1ZtWpVnbLly5fXvn/44YcB7dNYtmyZz2M6YQ4FmB5Fm8Ttn+hrO7C3ZDUtWsNgMBhagjEUbRC3f+KoHpAUD2UVkL2/4X0MBoOhpRhD0QZxG4r0ZOiTrt+b4SeDweAvjKFog2Tt18NM6cmizvCTwWAw+APjzG6DePcoqu112Tfvtqi7xIfBYDC0DsZQtEFqDUUShIcKwDI9CoPB4DfM0FMbxO24Tk/2pO4wPgqDoW2xcOFCTjvtNAA++eQTHnnkkUPWLSgo4IUXXmj2Oe6//34ef/zxFmt0YwxFGyTLy1AYH4XB4Cxqamqavc/06dO54447Drm9pYaitTCGoo1RUWmxvxDCQiE5HrokQkwHyCuC/GIzl8Jg8Cfbt29n4MCBXH755QwbNoxzzz2X0tJSevbsyQMPPMCECROYPXs28+bNY/z48YwaNYrzzjuvduLc3LlzGThwIBMmTODDDz+sPe5rr73G9ddfD8CePXs466yzGD58OMOHD+f777/njjvuYMuWLYwYMYLbbtPL8zz22GOMGTOGYcOGcd9999Ue68EHH2TAgAFMnjyZDRs2tMp1Gx9FG8M97JSaCCEh2nmdkWyxcRfk7IdOcUEUZzAECHGcqxWOkoZedNODtbjxZ+cNGzbwyiuvcOyxx3LVVVfVPulHRUXx3XffkZuby9lnn838+fOJiYnhH//4B0888QS3334711xzDQsWLKBv3761qT7qc+ONN3L88cczZ84campqKCkp4ZFHHmHNmjWsXLkSgHnz5rFp0yZWrFiBZVlMnz6dxYsXExMTwzvvvMMvv/xCdXU1o0aNYvTo0YfVSmAMRZvDO+LJTVy0fj3grISTBkO7pFu3bhx77LEAXHLJJTzzzDMAtTf+5cuXs3bt2to6lZWVjB8/nvXr19OrVy/69etXu6974SNvFixYwBtvvAFAaGgo8fHx5Ofn16kzb9485s2bx8iRIwGd6mPTpk0UFxdz1llnER2tbwrTp09vlWs2hqKN4e2fcBMTpV9LygKvx2AIBk158m+MzMxMMjKav5BL/ZTi7s/uJH+WZXHyySfz9ttv16m3cuXKVluMyLIs7rzzTmbMmFGn/KmnnvLLgkfGR9HG8A6NdRPbQb8aQ2Ew+J+dO3fWJvF7++23mTBhQp3t48aNY+nSpWzevBmA0tJSNm7cyMCBA9m2bRtbtmyp3dcXJ510Ei+++CKgHeNFRUUHpSyfMmUKs2bNqvV9ZGZmsnfvXo477jjmzJlDWVkZxcXFfPrpp61yzcZQtDHca2WnJ3ueGoyhMBgCx1FHHcXrr7/OsGHDyMvL43/+53/qbE9JSeG1117jwgsvZNiwYYwbN47169cTFRXFzJkz+d3vfseECRPo0aOHz+M//fTTfPvttwwdOpTRo0fz22+/kZSUxLHHHsuQIUO47bbbOOWUU7jooosYP348Q4cO5dxzz6W4uJhRo0Zx/vnnM2LECM455xwmTpzYOhfdGitFOe1v9+7dVlugJTov+3uNxcQaa9bnrtqyqx7WZS996mpgz9bVEWicotEpOnzhFG1O0dEYLdG5bds2a/DgwUHX4ScOeU81PYo2RpZX1JOb2h5FaeD1GAyG9o8xFG2M3EL92rmTpyzWRD0ZDAGhZ8+erFmzJtgyAo4xFG0M9xKoKQmespgo7a8oKTMT7gwGQ+tjDEUbwrKs2h5Fcryn3DizDQaDPzGGog1xoAwqKqFDJERHmagng8EQGFp9wp2UMg6YDwwGxgHbgDlAFFADXKmU2i6lHAT829Zwj1JqvpQyFngTSAE+U0o9Yh/zMftYO+39K1tbd1vA17ATGENhMBj8iz96FGXAacD79udq9M39OOBh4Da7/CHgKmAK8IBddg3wuVJqAjBJStlVSjkSSFNKTQTWAuf6QXObwNewExhDYTAY/EurGwqlVLVSap/X5wqllHu1hCq04QB989+klCoC9kspk4HxwDx7+9foXoR32VzgmNbW3FbwNhQul4vVq1dTXl5eaygOGENhMBj8QMByPUkpw4F7gavtIu+EJIVAIpAAFPkoy6pX5uv41wLXAsyYMYNp06a1nng/UVVVRWZm01cc2ritA5DA1o0r6Nr1bLKzs5kwYQJ3PDgb6Ex+URWZmbl+1xEMnKLRKTp84RRtTtHRGE7R6RQdDeW9CmRSwJnAv5RSW+zP3vl9E4A8IB/oCBTYZduBULvMu95BKKVm2ucgMzPTakmyr0DTWFKyDTstXvnc4s5LBJ3iBDUhFmCxee0yyM4G4LvvvmPZd18DF1NRHd6iJGctTY4WSJyi0Sk6fOEUbU7R0RhO0ekUHQ0RkKgnKeXfgG1KqXe9inOklP2klB2BRKVULrAcOMXePhlYVq9sCrA0EJqdwJPvWTz2Nrzzjf6cW2jPk6jK5eOPP65d+OSf/7gfMD4Kg8HgH/zSo5BSfgGMAAZIKV8E7gOWSilPBJYppe4E7gJm2RrutXd9CfiPlPJK4FPbt5EppcyWUi5BRz095g/NTmSvnYJ+W7YFCLJzq4EwQmrymDp1KuHh4VxyySX85+1PAGMoDAaDf/CLoVBK1XcQvOGjzlpgYr2yEuBMH3Vvq192JJCvMwizc49+3ZFVCnQkNTmM8PBwAJ555hm+nPs1+4EDZTqBlz/y0RsMhiMXM+HOwWTvqwBgW1aN/VlPH+mZ4VnvtFOnThw9ZhTUlOGyBOVH5AwTg8HgT4yhcDC7cw4AsH67TgubW6h7CgN61w386tu3L7h0XTP8ZDAYWhtjKBxMebWeIFFUHkNVtUVxeQQAQwak1qnXp08fqNHjVCbVuMFgaG2MoXAolVUWNdgz6Qhh5x6oqNFr8o4cWndlrDqGwvQoDAZDK2MMhUPJL677ecGKPBAhULWfwYMG1NnWt29fYygMBoPfMIbCodQ3FB98tQ2AEFc+KSkpdbb16tWr1kdRWFKNwWAwtCbGUDiUzL11l6v7YY0OZ4oOLz0o/DUyMpIOEToyatvOfRgMBkNrYgyFQ9mRWVjnc0F1bwASYmt81u8YEwoYQ2EwGFofYygcyq5sPZQkXPYYVERnALok+v7KOnXUE/B2ZeX7X5zBYDiiMIbCoWTt1V7p+LAddcq7donyWT85UUdIZebonsgj/7H420suLMuso20wGA6PQGaPNTSDPfurAOgSk0NB8ZDa8j7d433WT02JhW2wJ/cAZeUu7pypywf3srhwsknpYTAYWo7pUTiU3AIdvZSaUEpEqCcvx8A+ST7rZ6RqA7I/v5xP5v5YW379P0spKTW9CoPB0HKMoXAoebZronNiGOlJHgd2ekqEz/rdu+q0HgXFVfz7tU89xzkQzQOvG0NhMBhajjEUDqWoVEcxpSVHMqBnh9ry+utl15Yn6Do1Vge+/X4jAKGVW8Fy8cS7LjbsNMbCYDC0DGMoHMqBikgA0rtE072LpzwlwXd997rZhMZCeBoAEwaXwt7/UuMK4cNFxlAYDIaWYQyFQymv1tFNPdLj6N7Z44w+VI+ijqGI1MsqnjTxKOLDtgKwVG3zm1aDwdC+MYbCoVRasQD06pZANz2FgvAwiIv2Xd9jKGKISx4IQPcuYUyeOAiATTtK/CnXYDC0Y4yhcCAulwuX0F2Hvj2TaoeeUhI45Op1sbYBiejQiR59xwKQkQKD+mknd26Rbye4wWAwNIYxFA5kT+4BCImAmlKSOsUwtDdER8GIvofex92jyOg+kAp00sD0ZBg5SFuZ4opYf8s2GAztFDPhzmZPnsXj71hER8H/XhVc+7llRx4QQ6hVCMSSnCDYOfvQw04AMfaE7eJSKLOnXWQkQ8Ko7gBUiWRqalyEhppnA4PB0DzMXcOmvBIefwde/SLYSmDbrgIAIkIO1JYlxQsiwg89w9rdo9hfBAfKdA+kYwykd4lH1BRCSCRrN2b7U7bBYGinGENh444myi1suF4g2JmlHc8dwssbqemhQyQIAe7UThnJHn9GVIhOFPjjr7sPW1tVtWXyRxkMRxjGUNhER0FUBJRVQGl5cG+EmXZCwJjIykZqeggJEbXDT6D9E27iO+iFtFetzz0sXaXlFoMutZhyqzEUBsORhDEUNkIIx/Qq9uRWABAf42rWfrGeCdxkeBmKLp30cTZsr7dsXjNZ8DNszoRFv+KYXsV7CyxiTnHx3Spn6DEY2iPGUHiRnKBfcwuCqQL2FejcTp3imreft6Hw7lF0T9VrVezIbnoPxRdfLNc348oq7QdxAi98ZFFaDl8rYygMBn9hDIUXTulR5BXpHkBKQvOC0mLq9Cg8ju/+PXVobE5ey79uy7L4YrmXxsPrnLQKxaUWS1fr93vNek0Gg99o9fBYKWUcMB8YDIxTSq2RUr4KnAb8r1LqObveIODftoZ7lFLzpZSxwJtACvCZUuoRu+5jwDhgJ3ClUurwHo0PQTANxYadFtc/ZfHwtYLCAzohYJfk8GYd41A9iqEDdGrywrJoXC4XISHNNxjrdsCOHM/n/YXUyUEVDL79GartxLp78oKrxWBoz/ijR1GGNgrve5XdBdxWr95DwFXAFOABu+wa4HOl1ARgkpSyq5RyJJCmlJoIrAXO9YNmILiG4vW5FvMV/OUFi5JybSAyOjcwccIHhzIUvTP0BldYF3bt2tUifV8sq/t5f1GLDtOqfLXCM9y0tyB4OgyG9k6rGwqlVLVSal+9Ml8B/GlKqU1KqSJgv5QyGRgPzLO3f43uRXiXzQWOaW3NbpLj9XBNbmHgx7u32S20aCUUo/MzdU9v3mzqQzmza41GRBqbNm1qkT63f6KDTmpLngMMxTzP+kymR2Ew+JFgzsz2nj1WCCQCCUCRj7KsemUHIaW8FrgWYMaMGUybNq3ZgsKIBuLZkXmAzEz/3wmrqqrIzMwEYMOOJEDnY6oJ7wpAdERl7famEGLFA7oX4qrIwr2rVQGQBhHprFjxLkcdddQhdfiiuFSwZFUXQgScNKKcz37owJadBWRmljZZ2+FSX+OOPaFszuxMVIRFeaVgT56LzEz/TyhsrK2CiVO0OUVHYzhFp1N0ZGRkHHJbMA2Fd+xnApAH5AMdgQK7bDsQapd51zsIpdRMYCZAZmam1dBFH4q+3S3AoqwmhoyMZoYctYDMzMzaLydz/8GhsMMHdSUjo+nrXXdO0sdIioc+vepef1RYKeXVUezKPnDQD8Jbhy8+XGRRXWMxYRgM7deBz36AGpFARkanJms7XOpr/ORH/V39brzgk6VQXBZCUnI6UZH+XR+8sbYKJk7R5hQdjeEUnU7R0RDBjHrKkVL2k1J2BBKVUrnAcuAUe/tkYFm9sinAUn8JClZ47IEyi735Oo14glhTW97c8Fj3hLt0H8tqJ8Vp///aLQXN1vfxt9qLfVRaNkkd9Y04r7jlw3PfrbJYsfbwhvfc/olTxwo62/Zqjx35VFZhsW67CZc1GFoLvxgKKeUX6Jv7S1LKy6SUD6Od2ddLKf9pV7sLmAV8Bdxnl70EnCalXAosVEplKqV+AbKllEuAQcAH/tAMwXNmb7ejiVITyilY+/fa8oRmJnyNjdY38YyUg7d1TdFf9dbdTU8L4mb5L3rkb9FXs0i0+3b7W9hGBcUWk2+xmPIXi5qalt3MLcvi21/0+yljoHOCfu8Okb31eYtBl1k8/rYxFgZDa+CXoSellC8HwZ316qwFJtYrKwHO9HG8+hFTfiFYhsLtyK4o3AD7PyI9djtHj+xJaGjzhlHibGe2rx5Fj/QIftgI+4ubvy5FbrE+8MZV35CbfR7Qv8VRTz+sg4pK/bd7H/RIbf4xcvZD0QE9xNati6BLojYI7h7Fj+v1620vWvRIhfNO8O9wlMHQ3jET7rxIsp+WcwsDm6LCbSj27vyByIgQ1CsdmPNg87+a6RNg6lj4w+8OvjH2SNMGoqwmnsrK5k1DKa5M0G8qdvH5x28ALY96WrbG066bWpij0N1evfXS4Af1KLZ5+bQvfdDi+9WmZ2EwHA7GUHgRFSmI7QBV1Xpdh0CxLdu+kZVv47LLLiMtLa1Fx+mdLvjisRCOGXqwoaidqR2RTm5u05MD1tRYVAkdXxtSnc2ibz4CWj6PYtlvnvctNRRbbUPQy26mLnYc3J58PVt7f6FO8PjHM3TP5fZ/BdZQrFhr8etmY5wM7QdjKOoRjOGn2ifg8u2cfvrpfjmHZy5FOvv27WuwrjcbtheDCIOqvVx5+UVQpY1MSwyFy2XxwzrP5027W3Yz3VbPUHRO0EZwb75Vu61nKjwyQyAErFgH5RWBuXFXVlmceLPFCX+2qK42xsLQPjCGoh7BiHzamuXpURxzjH/mE6a5/RYRac0yFD+t3gNAFPu49dZboUpHJ+cX6xt/c1i3AwpLPJ9bPvSkz9s7XRsI7x7FNnvGTa80iI8VDOmle4g/bWzZuZrLvgKdMDG/uOXXZzA4DWMo6hHoHoVlWWzJ1AmL+nYLIynJhye6FWhpj2L1xgIAOkYVMnDgQKKiwqG6EJcLCg80vG99ltvDTgP06qwtH3ryMgaAJzw27+DexjFD9Ov3nqhjv+L9u1m5OTDnNBj8jTEU9Qi0ocgvhgPlYVBTwnHjB/ntPF3cc+MiOrN3b9MNxaadOpy2c3wlQgi6desGVfuB5ofILvtN9wQuOVn3BLZm0aLhmfrGwH1te/M9vY1eafocxwzRr9+vCcwwkPfv5pdNZujJ0D4whqIegTYUHv/ENiZOnOC388R0EISHVkJIFLtzmu5g2LFHz/bu1ll/7tatG1RrQ9HcyCe3I/uk0XquR1U17NzbvGNUVlns3gchIZ7std4T7hrqUbQkkq2swuLsu128+FHT9vUeslzZsrRaBoPjMIaiHp3i9I0xUIkBPf6J7Rx77LF+PVd8lF5taFdO08Nj9xTosNq+3XQ2wK5du0K1jkOt79AuOmCxJdN3uxUUW6zdDhHhMKo/9NPprJo9/LRzD7hcEB9VxM03XY9SipQEvS23UK/ABx5D0ScDUhJ0b2NLC9LpLPgZ5iyBu19qmnO6/tCTU1YCNBgOB2MobNasWcOoUaP47+vPAoHrUag1+um8g8ihb9++fj1XYlw1ANm5NU3ep6BUTw8f3E93teoMPdUzFNc+ZnHUpb7TZ7ijnUb1h8gIQV87tc2mZmY9d/cY8rN+5oUXXmDMmDEcM/5oosNLsSxYv9MeekrX9YQQHDtUv/9+jXbAL15pUVLatBu4sifv5Rfr6CnQIbjXP+nyOT/D+3ezrwCy9zfv+gwGJ2IMhU16ejorV65kyzq9jFugop7UGh1u2qdrKEL4dwZx5076xra38OCv/fonXfS70MXOPXVvfmWW9oKPGqzzguihJx35VH/o6eeNejjpa3XwuX+xo47G2olr+3XV19rcEFm3I5vybaSnp9OpUyeUUpQWbgfAsgTxsdApztOWbj/F0tUWMx63OP5Gi782cW6F2uCpN9fOL/XsB/D8HPj9/RYHyuoep35P1Aw/GdoDxlDYJCYmIqWkulwnXgpUj2LTzioA5BD/RDt5k5asV87LL667ct7ufaG8+LEetrn6H1btcElefglWaApYNQwf4GUoansUdW+K7hQay30k/Nu5V5f1sbPhtnToyXty4nXXXcfu3bt5//33SU/2ZKPJSPIMrblcrlo/xawv4OXP9PvPljU+LGRZVm2PAuDLH8Cy4NUv9H6Z++CRt+oeY1+BfnX7TX4xhsLQDjCGwouTTz65dkJZIAxFTU0NOQV6/YiTju3t9/N1T9V+hqLyDnXK31oQjcvOcv61gpc+1e/V6iwQIYS59hERoX8q3s5s76insgqLIjtcdpmPUNTdttO6q52wsOWGwn5Tvp0TTjiB6OhozjnnHCaM7V9bJ6xKj2dt3bqVlJQUnn7wcsLDLKprdIbe6Cjt69jWyPIVWbmQkwcdYyAyQg9DfbEiis2ZEG8nbHzsHdie7TEW7t/N5NH6daWZoW1oBxhD4cUpp5xS+7QcCEPx9MwvqArvjXCVcvpk/4XGuumZrg1EWXUcNTXaT1FZZfH2Al1+wzm63q3PW2zPtli5TrdFTFh+7TG6du0KVbYz26uNvFeY254DOfvr3iB32xG5bkPRJ8NTt6oZIbIbdugeWJTIZsyYMbXl7kl3AHnZPwPw1FNPkZeXx+x336RzxBpCQ+HtewWnHq3rLfi54XO5exNHHwWTRuj3d8/SCcH+dBZcfLJOEXLbiwcbipPH6J6TGXoytAeMofBi/PjxREfoeQP7C61mzzxuDrm5uTz3vo5COmHoPuLjwhvZ4/BJS7K/7vAu7N+vjcCcxZBbFMqQXvD0jYJzJ0FJmb75rdtSDEBirCfxVUJCApFhuuuwN7+6tjyn3nJSP6yt+3mX3aNwh9l2iBR06wzVNbAjp+nXsCVLd33GjehMeLinzdxpPAAyNy9h69atvPrqqwBERUWR+fV4/jD075wzSXDiKF13wc/6+y0tt3hjrsXsby0W/mJRYK+14fZPyAEwday9TG6RHr674lTBIzMEHSLh/YWQnavrug3FxGE6wmtzpnZ+GwxtGWMovIiIiOCESROgugCXJSgoaXyflnLnXfdQmXABAA9c18N/J/Ki9qk7onPt7OwX7PkB150lEELw9I2CiHD4YBH8tEVHOqUnegyCEILUJH2DztlfVVtef81q9+Q60HmWcgshLNQzdg/NH34qOmBRUh4JNWVMOWGE72sDrLKtXHjhhZSUlHD88cfz0UcfER5azcxn72P27NmcOErXW/Cz9kPc8pzF5Q9Z/P4+naNp+JVVlFVYqA26nhwgmDrOc/yJw6BfN0HXzoKh9ojh1mx9LLehSEuCwT31+1VbmnZ9BoNTMYaiHoHwU6xevZqXP8yGiC70TquojcrxN7Wzs8O7sG/fPjbuslj8K8REubjkFKiuriY9WXD5qdppu3qfHmjvlV63t9O1ix6q2u8V4VPbo6jUA//LvXoUmXay2owUCAnxXKt7+Kk2kqkRan0KFds58cQT6mxzpxoHoHwbK1asAODGG29kypQpPPXUUwD86U9/Iik6l9REbdzmLIaXP4fQUJh2dCWiKpude8M458bF/GQbitEDtFHrbYfcXuWVxt29nsaOHJ3jqaJS+0CiowQj++lt831EgRkMbQljKOqh/RT6zrY33z9DBp999hl0+QMAfzo7yu9hsW48PQptKH605wWMG1DM7049nu7du7Njxw5uu0AQEgLun8fA3jF1jtOzqx6nLyz1RBrt3mv3LvZrT/iP6z3pOdzDTrs2f8/s2bNr93EvsJST17R2/nG1/l5Cq3YyatQo39cGULkDgO7duzN9+nQA/vjHP3LCCSewb98+/vznG2t7FVc8bFFTA3+YBr8f+i7W+ssB+HKNZF8BRIWW8Ozjt/Hkk09yxYQVXH7iHi6a7DlVD3t2+PYcT8STe3b/RXaqkn99bFFZZYafDG0XYyjqMXDgQKJC9ZjTjyt3+OUcC5ZugMRTCRU1XDrFL6fwSVw0hIpKCI1hV1Z+7eQ0tfg/LFmyhOzsbB5++GH6dROcNdEzKW/YgMQ6x+ndPRGsGsqrImod0avW29agdA1R1m5Ky2H1Vl3kdmS7Snfy0EMP1R6nS6K+ke7x+MobZOHy7QB0Ta4mLKzu4owZdtLDrikw/ugRgO49uOuFhITw8ssvEx0dzTvvvEOnkJWAXnckOgruv1LwySefQME3pEf9CqG611S+bylPPPFPbr31Vu69YTxvPdidOR++V3veHl30NezI8Qw7uQ3FiaNgSC/d23rv26Zdo8HgRIyhqIcQgu5dtMP0+5/3tPrxLcti2YZEEKEcO6iQlITA9CZAX1tshHZE78gqZ9WmCgAKc1YwePBghBDMmjWLXbt2kVw2q3a/0UPqLsLdvXu32jQe7kl3G7fZd8nKPZTvWwR4cjttz7Z9HJW7WblyJStXrgQg1bY/OU2cvexe77t318iDtnXrInjxVsErfxW88MIL3HPPPdxwww116vTu3ZsHH3wQgB+/frS2/JbfQ6fYCr766isAXr+/CyEh2gBOOTaJRx55hOuuu47jjjuO6upqLr744tqeUe3Q0x4OMhRCCG46T3+/T75nmXQehjaLMRQ+6N9N/0Ovb2Z6iaawefNmDoQMA+BkGfjmj4/WxmH33ip+WqeNRr+MGpYtW8YFF1xAVVUV1113Ha8+eyNsv4cLJuwmPbmuTu9Jd25DkZWrjUFkSAEULQM8E+9WrLSTLFVor/Xrr78OeAxFU3sUOXlaR8903+t+//EMwSlHC0aMGMEDDzxAhw4dDqpz5ZVXEh4ejvpuNscMrqR/N7jtQsG3337LgQMHGDFiBJOPSedm+wZ/3w1j+Otf/8rzzz/PwoULufHGG6mpqeHCCy/kySefJLWTnty3I8djKMJFEZs36xzjF52sDcfPG2Hp6qZdp8HgNIyh8MGoo/Rsqsy8mEZqNp9ly5ZBx/EAyAFVjdRufZI76iGlnP2QXaCv8/wzRhIXF8fdd9+NEILPPvuMyspKrjgpi7cf6n7QMepMuiuC0tJSisp1W11+4RQo/hHw5EZavUnfQUccpceH/vOf/1BZWVnrV6gfWnso8g7oG3//HrHNvGoP8fHxTJ48GZfLxZWj/8PaNwQdYwSffqp9K26fxmPXCfI+F4z3CjQQQnDbbbdx5513UlNTwy233MIZU0YAsGOPVeuj+OKj1xg2bBg///wzHSIFM/Qh+d/XrGbNGTEYnIIxFD44foweTyisTG31Yy9Ysho69CM8tIqjulc3vkMr47457ypMw0UEVGRx0vF64trgwYM599xzAZ3S5NFHH/V5DN2j0N2A3EKL77//HsL1BInLLzwVDqwCVwUbdkJBsYtM20dxw4yzGTx4MLm5uXz55Ze1UVg5eU3LsnqgSo/pDB1weOlOzj77bAA+/PBDQkMFlmVp/wQeQyGEqJMvyo0QggcffJA5c+YwdOhQsnevh6p8yioEb3+sZ/BZlfsoKytj+vTpZGdn86ezBB1jdPTTufdYAVuW1WBoLYyh8MGxsifUHMAVmkTW3rJWPfbCn/XQz5DupYSFtuqhm0RGinbuZhbrCQARNdvqZK19+OGHOf7443n11VdJSUnxeYy4uDjC0T2KdVsP8NX8pRAaS5ioYtyYQaSnJcOBXwF4/6vtVAltRE45fhBXXHEFAK+99hoxHQSxHaCyqu4Sqb6orq6hOkQfZ/TQwzPg06dPJyQkhPnz51NYWMgvv/xCZmYm6enpB0VT+UIIwZlnnsnKlSt59913CXfp+F61Tvu2pp08lgkTJpCZmcmZZ55JQkw5X/9T0CkOPlkKp91hoqAMbQtjKHwQGRlOpGsnAPOXHuyoqKhs2T95cXExO/J1MP7ksa0/rNUUuqdFAeCKHqk/p5TXCc/t06cPCxcurH2yPhRJUfrmuOK3cuYv1smdkjpWERISwqmnngrFevLAS++uhYguCKpJSxJccsklAHzxxRdUVVV5HNqNDD9t25kLoTHgKqVLUlTzLroenTt3ZuLEiVRVVfHWW29xyy23AHDaaac1K1Q5JCSE3//+95x8/ABdEKvzmV956el88MEH9OjRgxUrVvDss89y9CDBomcEXRLhm5/goyWHdQkGQ0AxhuIQJMfoO9eylXVDcpatsYidYnHNoy5qappnMFasWAFxeorvcSP8n7LDF7272QYqRJ9fDmrZeH/3JN0+Xy7JZtU6nYOjm5108Iwzzqg1FCu26Kf/pLhKQkMFqamp9OzZk8rKSrZs2UKqPYrUmEN7zaYCACLJbZV5J+7hpz/96U8sWrSI1NRUbrvtthYdq0+GHaortJM9OV4boxdffBGA5557jurqaob2Efz5XK19/k+mR9FafL/aYviVLpYFaLnbIxFjKA5Br1QdzbJ6S93V4D79XmchffkzuPpRq1nGYsl3yyFW+wPG+T8HoE+6dq5roKYc17NFx5l+ok47UhHaD1e4nmKdkaLH0k4//XSee0RPXCNWD+W4V8gDGDRIX/zatWs9fopGQmQ3btd5p+Iim7n+6iE466yzat8PHDiQZcuWtXjhKPdcCjfu8NgpU6bQv39/du3axZw5cwA4yR7Z+uanFp3K4IM5SyxWbdH/mwb/YAzFIRjaVz8dbt9TNxTz542e9699qWf2lpY37Qf69bJ9ENqB9IQikgM4f8Kb2jQeNpOObtl4/9133EDPLtUQEsm0Cx8BPOGuQgj+eNkkYryiU3ukehwy3oaiyUNPWdpgJ8c1fRnXhujWrRu33nor55xzDkuXLqVnz54tPlaPek3oNhQhISHceOONADz99NMAlOUuJSqsnK1ZddOTG1qO+7fTmJ/L0HJa3VBIKeOklD9IKUuklEPssvOllN9LKRdIKbvZZYOklEuklMuklJPtslgp5Rwp5XdSyju8jvmYXfctKaXvIPpW5pgRekwk94AnwsayrFpD8dqdgugo+M88GPkHix98LNZTn3WZCQCMDVJvAuqmuggVlXTv0vJjjR6oh1zWZnc/+NihgtGeJSJq04tDfUPhnp3dcPtl7dPb05Nb7+b6+OOP8/7775OYmNh45QboWc9QJMV73l9++eXEx8ezdOlSLr/8ck6YdBzlOXMB06toLdwJKf2ZxPNIxx89ijLgNOB9ACllOHALMAm4x/4DeAi4CpgCPGCXXQN8rpSaAEySUnaVUo4E0pRSE4G1wLl+0HwQk4+1h1ZCulJRocNYM/fpfD6d4uCyU+G75wSDe8HGXXDMdRZzfzj0Tay8vJz86oEAnDKu5fMADpeEWAhBz9/ISCytk6SvuYzoq/fdbqcJd9/03YwZ6HnfNcWzrc7QUxN7FHsL9fNBj7SAPCc0C+8eRUIshId5rjU2NpY//EHn9XrjjTd0GHDBN4DxU7QWbv9W4YHg6mjPtLqhUEpVK6X2eRX1A35TSlUqpZYC9lL3pCmlNimlioD9UspkYDwwz97+NTCuXtlc4JjW1uyL1JQYQqtzICSK75SOgPLOJiqEYGR/gZopuOZ0cLngmfc9//hPvGsRN8XFr/YKZ5s3b4GOxwIwcXgQ4mJthBB0tm/oY4clHNaxRvav+7n+sNaYgZ4bpnsdCoCjjtILZ69fv56UBB1S2piPoqBMj2MN6BU8I3sokuOhQ6TnfX1uuOEGOnXqRLdu3bTD3DYU7jTnhsPDbShMj8J/hDVe5bBJALw9kO67pPfjZyGQWK+ud1lWvbKDkFJeC1wLMGPGDKZNm3bYwuPCcigglc++2cTAnpEs/CkWiKNfWgmZmcW19W44TTDr8y58rWDN+iyioywefKMzJWWh/OvDYv52cTGffP0bRJ5LOAXEh5eRmQlVVVVkZmYets7m0jk+iZw8SO9UTGZmSYt1dIkNATxjV6GufWRmemabd+sUCmgLESHqbktLSyM7O5t9mWuAoezeW0WmOx+5D0qrO0EkpCVaQWkzN4dqq4ykFDZnhdExupLMzLpWLzw8nO+++47o6Gj27dvHY489hqjKYm9+Ot8s39tqEy+D9XsKpo4aF+zLTwUEuQUN/4bqcyS2V0NkZGQcclsgDEU+0NHrszstqcurLAHI86pbYJdtRxuWjvXqHYRSaiYwEyAzM9Nq6KKbSrfkbRTkwqbMEDIyMticoyWPHxJJRobnkjIy4KTRLub9CMs2pZEQC3n2Kmk/bIglI6Mjv2zRT8Ld43fStesIbJ0Nfjn+okeai1XbYOyQjmRkxLdYR3q6RUqCJ3XFkP4pZGSIOtvTky325sPYYSl07uTZNnToULKzswmt0Vln9xeHH1JDYWEhrjA9vjPx6N5kpAcvBuNQbdWnq4vNWZCeEuFzu7usR48edO7cmb1530CXS1mzqzOTx7dOYEOwfk/B1LE338Jl98pKKw79G/LFoXTe+LSLqAh49H8C8ztzyvfWEIFoic3AICllhJTyWGCVXZ4jpewnpewIJCqlcoHlwCn29snAsnplU4ClAdAMwMAe+h9402796nZkX/n74Rx33HH8/LNn0eULTtJ13l1g8dKnnuGEXzZBboHFr9v1mMSI3q0T3nk4/O0ywU3nwRkTDu84QghGeEWU1h96EkLwycOCz/4h6hgJ8PgpsnfoGdx78znk0rNbt+2ACG0o0pODEy3WGO51KXwNPXkjhNBrfRcsAGDJKjP0dDh4r6zYGkNPhSUWz34Aj72tV1Q0aPxiKKSUX6Bv7i8BFwJPAouAv9t/AHcBs4CvgPvsspeA06SUS4GFSqlMpdQvQLaUcgkwCPjAH5p9cZzUwyYb93bnP7OXkJUL1BRSU7KRJUuWIKXkzjvvBODMiRAeBt/+AvN/gsgIkLYzd8HPsLOoFwAnjg7ORDtvjh4kePKGEKKjDv+m617FLaYDxEYffLzRAwRTjj643G0oNm5YTUKsXjs77xA2dPX6PSDCCKeQyAhnGor+3bSurr6zntRhzJgxUKHXOvHXKopHCt4TNYtLafYk2PpkeY1c/bbtsA7VrvDL0JNSypeD4N16ddYCE+uVlQBn+jhey6bMHibXnNeX+17PIY++XP5IAcQAxb9wwQXnk5GRwVNPPcUjjzzCDTfcQHp6OlOOtvjse73vOcfBsD4Ctd7ilc8tKkiF6gKmTEgLxqX4jRH9BGDVzodoKm6H9tq1a0kdoJ8Gc/IgOeHgur9t0o+NcZHFQKeDKziAq0/T0U4XnNR43TFjxkCNzlZb0rqpxI446q/VXlwKCXEtP16Wl3tpzTYYP6Tlx2pPmAl3DRAZIfj0n10QuHDFSACSo3byyiuv8PjjjzN+vE4XvmaNznV0wYmep91rThdM1rswT2fdRhR/T88e3QJ3AQHg+BF6hbjmzjR3G4p169Y1ui7Fpp36bpocV9FClf4nPlZw47kHD7H5QkoJNXqcpKTMDG8cDvV/M4c7/OTdo1iz1Xw3boyhaIRjhoZw03mez3fcMIXo6GgAhgzRjxu//aaXcps+QQ89jB2kb6Aj+kKilxs/JWIdoaHBC431B+nJgpyPBG/c3bwhoaSkJLp06cKBAweIjSwFDh0iuzNHR0ulJbWPf9zOnTuTnqp/GPlFgU81357YU2+99cOdS5Fdr0dh0BhD0QQevDaUo3poH8Q5kz2zqwYPHgx4DEVctGDTfwWLnxUIIQgNFZzolbV6QNregOoOFHHRokUT99x+itAaPe3mUD2KnDxtXHsdYmW7togcoTPOFpe2D+MXLFq/R+H5Poyh8GAMRRPoEClY9qJgzeuCnmmeG6K7R+EeegKIihREhHvqTB5tv68pQQ4wze3NsGF6SdjiPL1saE7ewTfN9evX1/7zDurnTP9ESxh3tJ53Wl4VZibdHQZuQ+Fe2+Vw8z159yj25uvwW4MxFE0mPlbURra4cfco1q5de8h/9unHQkRICex9h6MGtiw7aXvl5JNPBmDrBh0B4CuNxx133IEVrtfwOKp3x4MrtFFGDB8KNaVACKXlwVbTdnE7s/vY0xAOd+jJ7cwOt8N8TOSTxhiKwyAlJYXOnTtTXFzMrl0HL3AEkJYsGFU5FTbPoF+/fgFW6GxOOOEEIiMj2b7xB0CvsX3lwy7O+ZuL7FyLxYsX8/EncxEx2vGdkRxMta1LUlKSl0M7yGLaMO4eRf+u+rW1nNnH2NFOZvhJE4iZ2e2awYMHs3fvXtasWUP37t191tm0USeJ6t+/v8/tRyrR0dEcf/zxzFuaDcCGnfoPYNNuF2Fr74d+/8aK7E63lGoG9Qz+HJTWIiEhAWqKgc4Ul9bNvGtoGi6X5TEUdjDh4Qw9WZZVO/R0shQsWmmxeqtF3WxDRyamR3GY1Hdo1yc/P5/9+/cTExNDWlr7mkPRGkydOhUO/EqfmG857wR49I/VdInLZ/VWwS/iVehyKdGRFq/cmu/YyXYtQRsK06M4HPKKoKZGZ+x1hyUXHsZs6sISKKuA2A4wXv9bs2Zrayht+xhDcZj4cmh7869//QuAAQMGtMoSnu2NqVOnAhYFP57PG3dW8tm/T2bPNyOhYjdE6sfEN+4OYVCP9hVGGh8fb/coTORTS3H3JrokamMBhzf05PZPpCfDkN76/ZptJsMvGENx2Hj3KGpqavjrX//K7bffzu7du/nwww+56667ALj33nuDKdOx9O/fn549e7J//35+97vfsXjxYtISq3nyD+uZMMzimT8LzpnU/gxsZGQkIeiuxL48481uCW5Hdkx4Cf98VP9/uZ3Zu/davDDHorq66Tf5bC9D0bmToHMnPdN7V/uMam8WxkdxmHhHPj388MM8+uijADz11FOEhGg7/Mgjj3DGGWcETaOTEUIwdepUXnzxRRYsWEBERARz5sxh7Nix3BRscX4mIrSScmDv/lIgOthy2hzuHkXenvVsX/sjDIECO/v/vbMsXv0CoqMEV0xt2vHcjuw0e1HLIb1gQT6o9RzWSpDtAdOjOEw6depEeno6ZWVl3HOPXrxvypQpVFdXU1FRwZVXXsntt98eZJXORg8/aV544QXGjh0bRDWBIzJMzzg3PYqWUbsE6r4NUF0AeHoU7qCInzY0vUfhNhTptqE41l5i7cpHLL5YdmQPPxlD0Qq4/RQAN998M3PnzmXt2rW89dZb/Pvf/za+iUY45ZRTOOecc3jggQdqlw09EugQoZdmyc03hqIluNdZL9izHqp1Gt79hdqXtXOPrrNqS9OPl73fvS67/n+942LBeSdA0QE47Q6LmZ8cucbCDD21AkOHDmXevHmMGDGChx9+GICBAwcycODARvY0gB6vf//994MtI+BER7mgEvKKKoMtpU1Sm76jai/UaEORW1BNVXU4Wft1WOuqLdoZ3ZSHNbcz2z30FB0lePd+PQR13yyLW5+3uPhkiOlw5D34mR5FK3DjjTdy0003MWfOHCIjI4Mtx9BGiNPLgFNQVNNwRYNPalOMV+4hOkIP45WUhbB7L7hc+mZeUNJ0Z7R76Om/rz3OVVddRVlZGUII7r1CMH6wDmP+cHErX0QbwRiKVqB79+48+eST9OzZM9hSDG2IuBh9Mys6YAxFS/D0KPZw7dWXgquKGiuC73/ZV6deU4ef3FFPn30wk1dffZWpU6dSVKRX07piqv6uXv3i4OGnG592MfIPLkracZizMRQGQ5BIiNWZ7MySmy2jNsV4ZQ4333xT7fDTc68tqlPvx7WNr2NiWZZnLYpKnSlg0aJFTJgwgenTp/PsvRMJC6nk219ge7bn+3K5LGZ9ASs3wfK1h31JjsUYCoMhSCR01C7CkvIjb8z7cKmp8aTb6N0tlu7duxMVpoMClq/S65uEVGuP9vxle3wew7IsHn/8cW655Rb25lVRXgnUFIHrAPPmzaN3796sXr2aTz/9lDW/LqU6ZzYAr8/1HGN7DhywZ9av3NT61+kUjDPbYAgSifHan1VWYZ7XmsuefKhxCajcw4Rj9FKSCbGCnANAjI5rHdNnLz/s6MKabQe3b3V1NbfccguzZ+ub/96SJOBOqMjihBNO4OSTT+aHH37g448/JiEhgV27dnHz/74OnS/m3x+Vc8/lUYSECFZ7DWv9uqX95oUyhsJgCBIpnaIAKKs0/4bNxR3+SsUujjnmGAA6J0dpQxGtsw1ffnoXfni2hqKqVAqLK4iP04a5vLyc888/n08++YTo6Giqqqp4a/a3MPROqMzm6quvBiA5OblOuPbmzVt5fsVOsvO789XyEqYeE8dqr1xQ7blHYR5lDIYgkZKkw54qqttPVtxAURvJVLG7du367unxuixEG4TjxnQh0rUTRCivzdYL15eXl3PmmWfyySefkJCQwIIFC3jmmWcgQifsjGA/Z599ts9zPvHE46SE6LVTXpuzHcDOLqtZvxPKK9qnv8kYCoMhSHRJ0mk7qlwmpLq5bMvSE+tE1e7aJXWTE+r2zHp0gX7p2l/x3KxFzJkzhzPPPJOvFvxM+Mivuf5/FzF27FhmzJhBz1HXAzDiqE5ERUX5PGdERATHDNHnVZu0kXdHVEWEQ3UNrN3eqpfpGIyhMBiCRFpnnfK02uoQZCVtj3XbdJrY+KhiwsK0gYiP8WxP7Aix0YLJ43WSps3ZHTj77LP56quviO5zE1WxJ/LoR4P5cZ3FF8th+4ExRIbX8MajExs87wmjdO9vR346xaUWmzIhJASm2llnVm5u5Qt1CMZQGAxBIj1VD5W4hDEUzWXLLj2bvUuCJ/TVnWocdG8C4KRxKfrz4LOYMGECQ4cOJWPETQBUVgvOvdfiuif0cNFD14YxoIfv3oSbo0f2hNJ11FhRvPW1Xg+jX1cYN0g7sX/d3D6HnowXzWAIEqkpHcFyQUg0lZUuIiLMc1tT2b1P35h7pHraLD5WAPpG7c72On4wREbAzsIefP3fxbgsGHiJRXws9OpSycotEQCM6g83ntP4eQcMGACF70H0UTz/oT7X0N4wwl7l2PQoDAZDqxIeHgYuPYSSva8oyGraFvuKtF+nX3dPb8x76Mndo0iKF1w8GSwLnv3QYva3uvyMY2Hmzfkkx0NoKMy8TRAW1nhoa2JiIh2tXwHPetpDewuG99Hvf93SPhc6MobCYAgiIZaerZW9pzjIStoOFZUWJZVxYNVwVJ+E2vI6Q0+pnpv+n89zp9+AN+fpm/h5JwjSk1z89LLgl5cFowc0ff7DwPS6KUKG9obUJOicYFFYAr+syz/Enm2XgAw9SSlDgTeADGA7cA1wNvBnoBy4XCm1S0o5CPi3resepdR8KWUs8CaQAnymlHokEJoNhkAQJsqoBLL3HcYankcYmbWpNrLo3at7bXm8Dx8FwLA+ghNHWSz4GTbugo4xcLKE3H3QvYuAZi5KNOKoJFb8sg2iegHaUAghSIjYyV56cOff3+ar//6phVfnTALVozgb2KqUmgSstT/fAkwC7rH/AB4CrgKmAA/YZdcAnyulJgCTpJRdA6TZYPA7EaHaKbs390CQlXiwLIsFP1kUFDtzCKV2DkX5Trp39xiKuj2KuvvcdJ6nx3DGBIiMaPkM6oEDB0LhEgCio6B3ui4v27ccALW+/SV5DJSh6A2stN//DMwAflNKVSqllgL2WlKkKaU2KaWKgP1SymRgPDDP3v41MC5Amg0GvxMZZhuK/aVBVuLhm5/gpJv1+gtOZOcedzLA3XUMhbePov7Spb8bD30z9PvzTzy8NBvehmJwTwgJERQVFZG5Tq+pksdE9uxpXwttByrqaR26l/ABMNku8/behdqv3t9gIZAIJHjVdZcdhJTyWuBagBkzZjBt2rTW0O1XqqqqyMzMDLYMx+hoCKdobG0d4SE6vHP7rv2HfdzW0rZ0ZTQQz/LfqsisHefxsH1PKGEh0DXF95Ozv7+rH1cJIJUI9lJUVFSbCrz8gC7vEOmi8sAeMuvZ3pk3hbFySzjDu5WRmdlynQkJCZD7HpGpp/E/v5tIZmYFn3/+Oa7cT6EqF2KH8/zrnzDj4jFNOp5TftsZGRmH3BYoQ/EZcIKUcgGwBsgHOnptd//iXF5lCUCeV90Cu2y7rxMopWYCMwEyMzOthi7aKWRmZjb45RxpOhrCKRpbW0dcdDY5lVBZE3HYx20tbSWV+t9wW044aWnphIR4nt+qqi2GXmsRFQGZHwqfK8f5+7vK2q8TPSXHldU5Twbw96stUhND6dr14PNnZMAJXsuxt1RnamoqkWFVVPx6NuecUEjHjsn88MMPYFURWfQhFUnX8umKJB64vWnHdspvuyECMvSklHIppW5WSp2Ivvk/BQySUkZIKY8FVtlVc6SU/aSUHYFEpVQusBw4xd4+GVgWCM0GQyCItaM784urgyvEi912UE9Zhee9m8x9kF+sF/lxp/kONNuy9HNlRtLBPZq7LxP84TT/ZnANDQ2lf//+AGzYsAHLsvjiiy8A+J+z9Re6es/QdpX3KSCGQkqZKqVcKKWcD5QrpZYATwKLgL/bfwB3AbOAr4D77LKXgNOklEuBhUqp4PfRDIZWwr3KXWGJcxyg3sZhw86623Z6Le2waXdg9NQnO0+PVPdKD9584YEDBwKwfv16Vq5cSU5ODhkZGdz/lzOh5BdqREdmf9v4gklthYC0tFIqBx3h5F32LvBuvbK1wMR6ZSXAmf5VaDAEh/gYfdMrPuBqpGbg2O3lh92wE072Gmrf6bVt0244fkTAZNWSd0B7rQf2immkpv9wGwqlFOvXrwdg2rRpxMfHkxH2NZmM5LnZJVx6ahTV1dUUFhZSWFhIWloaHTq0vZQtZsKdwRBEOrlXuStzxoI3LpeFt/964+66wyfePYrNuwM/tFJSalFREw2ucgb16xzw87txG4pnnnmGhx56CICpU6cCcOqoPHBV8OOmTvzhpheJiYkhOTmZPn36MHbsWMrLy4Omu6UYQ2EwBJFO8Tob6QGHrHK3Nx+qvNwlBw89eYxDMIaevNeh6NGje4N1/cmpp57KxIkT6dGjB6mpqRxzzDFMmTIFgMmTRkHWs1gIZi0cTmVlFfGJ3Qgd9F9W50/h//7v/4Kmu6WYpIAGQxBxr3JXXhnaSM3A4PZPxEVDcSls2FV3e7B9FB5DsZPu3QcFXoBNYmIiixcv9rltwoQJsPMoSLkIOo5jxgPr+HlXP35cD3Q6i0ceT+ecc85h1KhRgRV9GDjjMcZgOEJJTtTj1eVVdVe5W7fdYs3WwA/tuA3FMUMgLFQbhjKv6B1vH8XmTD1UFUi2ZekJiqIqh9TU1EZqB4euXbty7lmn0j/iFQD+/bVtJABCInB1PJkrr7ySysrK4IlsJsZQGAxBxNcqd1XVFhNvsJhwvVXnJh0I3I7sXmk6NYVl1e05uJ/oO0Tq8Nmsg+fj+ZWdWTrhXlyHGkJCnHv7mj17NusX38txw/XnPhlw03n6fUyPi1i1ahXvvfde8AQ2E+e2tMFwBNDZNhQ1Vgeqq7Vz4LdtsL8QCktg9ZbA6tm9TxumrimC/t10mdtPUVhiUXQAYjrAiL66bHOAg9Uzc3SW3YQ4ZwzVNYQQgnfvF/zjj4Ilzwr+dJYOWHDFnwIijA8++CDICpuOMRQGQxDpaM+jIDSWwsJCAH7a4Nn+86bA6nH3GLqmwADbUGy0/RRu/0T3znpVNwi8n2Jvrs7LkRgf3khNZ5CaJLj9IkFasqBvV8GgnlBWFQnxxzF37lxKStpG1mBjKAyGIOKemU1obG08/k8bPcNNP28M8NCT7aNY/dNcCrJXALBuh54M6PZPdO8C/bpqA7cpwCGyhQf0+RLi2uat64wJ+jV10LWUl5czd+7c4ApqIm2ztQ2GdkJctP0mNI4HHtCZ9ev0KDYGVo/bUPzzoZt45dnbAZj92SpKS0trexTdOkPfIPUoiu1Ef/GxbfPWNf1YbWDLY3Uo7Ycffli7rby8nA8++IDrr7+eVatW+dw/WLTN1jYY2gnePYp58+Yx/5tF/Orll1i9FSqrAvPUblmWJ31HxW76ZujZ4pWhPVm4cFHtHIruXUTt0NNm21D8utkiO9f/Og+U61tWp45tY+ipPkcfBV0SoaCsI8SM5LPPPqOsrIwHHniAtLQ0zj33XJ5//nn++te/BltqHYyhMBiCSGSEIDwMCIkAEcGt97xCRaVeO6F/N6isgrXbA6Mlt1CfL4xicJXy2EO3ERVWBmEJfPi58umj2JwJHyy0GHGVxel3+t9QlFZoJ3ZSG/FR1CckRHDeJP0+fug/KC4uZurUqdx3330UFBQwbPhwiOrNwkXfUVHhnFxRxlAYDEHG3atI6tyTVVv1DXD0ABilE5TWGYryJ+7QWKtce69HjRrJ0f21s/XzH6Lq+Cg6xgg6d4LySrjo/6xanXvy/Gss3PNNUhIjG6npXO68RNAhEgrDT4K48WzevJmk5M7c8/RKQkf/DGM2UZ56D99//32wpdZiDIXBEGTchuLKa/8CsXq27ugBglH99Xh2Qw7tj5ZYbMvybM/OtZh6m4v5qvk3bPewU03pDhITE+nWrRt/Oq8TADmuE9merZ3a7tXj3CvGVVaBe0rDwl+afdpmUVGjDURKYttLrOcmPVlw6/n6fUjfxxg4+hwyfreD/3t/KL+4o9y6XMncrxYETWN9jKEwGIKMHKBfVxWc5TEU/T09ikOFyK5Ya3HW3RZX/cNjFN6cB3N/gBuesrCsphmLvfkWlVVWnTxKI0eORAjB9AnhhFECsSPZvU8P+3RN0dUG2KmWRg+Auy/V77/9xb89iiqXNhBpKcHLHNsa3HahICUBXLHj2RT7Lqu2RtA1BZ6/WdA9uQTCk/jw26JGjxMojKEwGILMP/8kiI6CeSsTIU7n9B7YrazWUPy6GaqrD74Br9mmX39Y69m+Ym0VAOt3wrc/N37unXssup1rMfgyy3OTr9SGAiAqUjCqx/ba+l0StV8F4PYLBbf8Hj59WDB1rC5buLI5V958atCGIrVz2zYUHWME912h26zGJbj6NPjtDcF1Zwmu/J3uNW0uHM3+/UFaHaoexlAYDEGmV7rgwavtiXciBMo2sWvbajrFCXql6VQZ9ZPzAWy1h5zKKmDdDvj111/5ZIEnXvWFjxp/ul+0Ug8dbc6E9xfahRUeQwFwzRmxte+7e2X2HthD8M/rQ6Aqhzf/dRORYVVs2Ak5ef65rdTUWFgh2kBkdIn3yzkCyYzp8OyfBe/cvZ+Xbg+pnXx5yZQIXSHxDL78alEQFXowhsJgcAA3nANj3clQS37ml1/0YL+7V3H3SxY/rK07nLQ127P//f/8jOlnX0pVWE9wVYKrijmLXWTua9hYqPV6e1qSV2E9Q3HFmT0JqdYnqyzezNVXX80111zDe++9x9tvv83QoUN58flnqNq/EIDl6yKaff1NobjUvpbqYjp1avuGIixMcP05gglD6iYH7NtV0DU+C8LieO2TPYfYO7AYQ2EwOIDQUMFb9wiGZ2yE3U/UGoqzJuqnzI+/g3F/tLjsQS9DkeXZ/8N5O6iJGgJAevw+2P8RLiuEmZ82YijsiKpX7xQ89IdS2Ps2Hap+ql0TGiAsLIQBnbSeX5d/wCuvvMLLL7/M+eefz0UXXcT+/fvp0KEDrrxvAPh+rX8ikvbsLwNAuEoID2+b4bFN5dzj9RyWZZt7NNnX5E+MoTAYHEKfDMFTM3KgRNUaiotPEWx8S/CXC3Sd2Qs9/ojNuz3rbIcnjueSGf8EYMqELnSq0KsMPzu77JAT9qqrrdoomzED4eiuP8CGSxg+dAChoXWT7v3zxo5EZt3LqC6LefDBB3nkkUeYNGkSffr04bnnnmP27NlQqIdJlq31z008a49OCBjCAb8c30ncfEkGWC5Koybx8qvvNr6DnzELFxkMDmLEiBEArF69murqarKzs3nl+efpGBNDUswN7D/QkU27oWtni/1FoXqYKSSCkLgRFFp6ic1xg8MZfetkrp+1mnyG8vQ7Bdx2aaeDzrV2h/ZvZCSVs3HtKlas0LmdvIed3Eydchzlm4+rU+Y9e9jlctGny61sqS5mW04cmfssMlKat7xrboFFh0iI6eB7v5x92kCEi7a3lGhz6Z4aSr/ULDbtSeXmvy/gjNNOJKFTClNvs+iRCrPuCOwzvulRGAwOIiEhgZ49e1JeXs6GDRu45ppr+Mc//sG9997L/p36if2n9ZVsybR7CeWbyUgspaJKMO8nvVreqP5wzTVXk17zGgB3PF/I1/MXHnQuZS+mk7n+E8aPH89dd90F+DYUjRESEsKNN1wHRd8B2kneFIqLi3niiScYecojpJxWStdpmw451OIeeooIdc6MZX9y/fl6YaYDUVO54YYb+Pg7WPAzvPGV7yg4f2IMhcHgMNw36qeeeoqvvvqKjh07ctddd9EpUkc0vfHhSj6b/xsAUWQzcaTOLFhRJQgLhaG9ISIiAvXlnUSzE1dEd6Zc+iYPP/wwLper9jxqg32zKVbExurIpvDwcCZNmtQi3VdccQUR5bpX8tG3TVvR6P77H+DWp4tYWX47hHSgoKYvr7//q8+6uXm6JxEV3nZWhjsczp0kEMKCxCm898GXPDRLz4isqYGsAEfNGkNhMDgMt6F4+eWXAfjLX/7Cgw8+yFUXjAVgscrjg8+1D2NI3xjkAM9QzZDennkOaWmd+ffdelEJq+sd3HXXPZx11lkUFBQAnh4FJT8xa9Ys9u7dy7Zt2+jXr1+LdHfs2JHpx+shrnnLyxqtb1kWr82LhR73InCRFKVXQXrw5W0+6+cWaAPRIaLG5/b2RnqyYMJQASFRkH4jK3ck127bGeBgKGMoDAaH4T30k5KSwk033QTAFefqWdsVof1YuV4vcnTyxD7IgZ59R3uClQC4cHKIXqmuQx9CRy3hE9WT4yefT3mFi1832z2KA79w0kknkZKSQkZGxmFpf/DOs8BVTmF1V35ZvbPBuqtWrSIv7FQAXvprCK/erSfTbS4YzYaNmw+qn1+kVwCMiXIdtK29ct4J9kNAj3vrlBtDYTAc4Xgbir/97W/ExcUBMKC7ICzEBR16Q8wwAMYN78zIfiDs+4k7P5Sb0FDBUzcIIiOgJmYs9HmSVeH/5U8PbaKyWkDpBo4eNYDExMRW0d6/bw+SIrcCcM9jXwCwdu1avvvuu4PqvvneNxA3hlBRyYUnCU47LonYsFyI7MpfH/z0oPoFxbonERcd/HDRQHHO8fZ3K3TcUXrsdgB2GENhMBzZpKenc8oppzB27FhmzJhRWx4eJhjUy/6X7XgMAL3TdTqIIb10ce2kPS+mjhPkzBH8915Bn6TtEN6JWd/aw0slilNOOaVV9Z8yVvtMvlp+gJtuuomhQ4cyceJEFi5cWKfe7Pk6iml073yiowRCCC6crLd9tqIjubl1/RxF9up2HaObF03VlklPFhw71P5wYDWFm18CqF0bJFAEJDxWShkCvAr0touuACTwZ6AcuFwptUtKOQj4t63rHqXUfCllLPAmkAJ8ppR6JBCaDYZgIYTgq6++8rltaG9YtQUQep5DrzRd/ubfBN/9nMfoAUk+90uI0zfhk4Z1JP2Yf1CTboe2Fv/EKaec06r6px3bkbe/g+rocTz99F8gVkLnizjrrgpGj3Vxz+WC3km72VkyHCLhstM8vZk/X5jCS3MtahLO4MuvFnDpxb+v3eZe3S6h45H1fHv9WYLvVll0E2+za/86SG2/Q08jgEil1ETgAeB64BZgEnCP/QfwEHAVMMWuB3AN8LlSagIwSUrZNUCaDQbHMbS352m6S6JnzsHwvoIzj218fkHnzslcMWkLrD0P9r1HdOnHjBs3rlU1yv6VhAgLYiU9h11A5NHfQ8afKQg7mW9+ggv/1+KNd76BhJMAOPt4T8qPwb0EKR2yIDyRBT/XfWo+UK6NY6e49j0ruz7nnyQo/Vrw1yu7QcUOoP0ait0AUkoBJAD7gN+UUpVKqaWAu3OVppTapJQqAvZLKZOB8cA8e/vXQOv+qg2GNsTQ3p73vdNadowbbrgB9n8I6y9k8nHDWj0dRly0xfC+AkLCyU19i4qqUEb12AnrLyGiXJG9H/7+8XEQ2oFeybmkJdcdShrUVQ85rdtRd8CjtFIbiuQE/+SScjIdIgUXX3wxokqHSLfLoScgF3AB64BI4GLAKw8l7nwB3r+YQiARbViK6pUdhJTyWuBagBkzZjBt2rRWku4/qqqqyMzMDLYMx+hoCKdoDLaOlJgQQK8clJpQRmZmQe22pmpLTk5m3LhxLF++nLFjx7b69VRVVTGyzwF+2RRDSRkM61XJO/eGcs6va1i95lIY+TPloT0BmDbOddD5e6QcgE2wfV9MnW1llWEQCuGh5a2iOdjfZUt09OqWwNaaEopKY1m7MYv4mNYzGA1FvAXKUEwBypRSA6WUo4C/Qp2ELe7AaO+4twQgD8gHOgIFdtl2XydQSs0EZgJkZmZahxvmFwgyMzMPOxyxPeloCKdoDLaO9HSL+FiLwhIY0rcDGRmedRmao+2dd97hvffe489//jMREa37hJ6Zmcnvjo1l1lyL5Hj45B+R9Ejtxssvv8wFF1xAtfUSu7gRgGvO7kJGRt0exQlHl/HG91BYmVbneqpcuRAK/Xuntsp3EOzvsiU65OjRbN28A2IGUx2adlDb+YtA5nrKt18LgGSgh5QyAhgDrLK35Ugp+wF7gESlVK6UcjlwCvAyMBn4QwA1GwyOQgjB0N4W362CXmktv0n06NGD2267rRWV1eWMCfDE9YLJo6FHqtZ59NFHs3XrVmpqPFlwh/U5eN/jx+gl9MpDeuFyuQix11mttnSKki5tfHW7w2HYsGG899suiBnMzj2+288fBMpQzAMulVIuQg893QJ0Axaho54us+vdBcyydblnmLwE/EdKeSXwqVIq+H1FgyGI/OV8QUyUxRkTgq3k0ISGCm7+/aG3vXXvoY1cr27xiKocrPDO/Lo+n5GDdCRXDTrsNqNLXKvrbSsMHz4cXtMTGQPp0A6IoVBK1QAX+dj0br16a4GJ9cpKgDP9Js5gaGOcMVFwxsT2PZcgWuzgAJ1Z8tN+Rg5KoqqqCitEG4guSR2CrC54DBs2DCq+B2B7jkVdt67/OLICkg0GQ5sgJUZHPv20Tk+eyM8vhFCduNC9ZOiRSLdu3YgO1W2zcUfg0q0bQ2EwGBxHjxRtIDbs0kYhe28RiBCEq5TQ0CPXUAgh6NtNryC4aWfjiRdbC2MoDAaD4ziqh3Z278zVjuvsvXp1u1BKg6bJKYw4Ss8QyMwNbaRm62EMhcFgcByjj9KO69zSzliWxZ5cbSDCQ9r/6naNMX50d7BqKCqPpSpACxgZQ2EwGBzH8KNSoSqPKiuWnP2wJ/fIWt2uIUaNGAqVWViEkLkvMOc0hsJgMDiOXr16QulaANbtgH357tXtqoKoyhkMGTIEKnYBsHl3YNrDGAqDweA4kpKSCK3UixepdaXkFegbYnRkdTBlOYLo6GjiIvT85WW/ZAXknMZQGAwGxyGEIKmDnlGm1paSZ69uFxt15Cxa1BBdk/QQ3I+r8xup2ToYQ2EwGBxJ92SdDu7XLSEUlBx5q9s1xNC+OkR23fbA9LCMoTAYDI5kWM8SqCljY3YncgoTAOgYY25ZAMeN0Tnms/IDk/fKtLrBYHAk/XqnQr5e6W9r8dEAJMQFbu6Ak/ndSf0BKLW6Ul7u/0gwYygMBoMj6dmzJ+R+AECl1RGAxPhAJrx2Lj0zOhLqyofQWL5Zstbv5zOGwmAwOJIRI0ZA3mfg8jwxpx7BKcbrk9hhPwDzlmzz+7mMoTAYDI5k4MCBrPn1e0b29MwqG9g3+AsNOYXeadqRvSIAkU/GUBgMBscyePBgbrqkW+3njqZDUcvIo3Ta9Y27/B8JZgyFwWBwNKcfC+G2ayLuyF2K4iAmjtZrp+eVJVNYWMj8+fO56qqrqKysbPVzGUNhMBgcTac4wYzp0CcDhvQOthrnMLh3uH7ToR+zZs3i7LPP5tVXX+XVV19t9XOZEAKDweB4nr3JPNPWp9ZdE9WHW265Dajh/PPP55prrmn1cxlDYTAYDG2QmA6CTtEl5JfGQFQPRo05jpMv/jdCtP7CTsZMGwwGQxtlYA89AbHrwGmUdp/J1Y+F8crnrX8eYygMBoOhjTKsXxQArl5PsX5nKP27wYUntf55jKEwGAyGNkr/bnqYKStXEBYK/71XENPBDD0ZDAaDwaZfV8/7B/4gGD2g9Y0EGGe2wWAwtFnGDoL4WBg/GG6/0H/nMYbCYDAY2iidOwly5kBEOISE+Kc3AcZQGAwGQ5smKtJ/BsKN8VEYDAaDoUEC0qOQUh4NPGp/TAO+AJYDfwbKgcuVUruklIOAf9u67lFKzZdSxgJvAinAZ0qpRwKh2WAwGAyagPQolFIrlFKTlFKTgCXAR8AtwCTgHvsP4CHgKmAK8IBddg3wuVJqAjBJSunl5zcYDAaDvwno0JOUMgwYB+wDflNKVSqllgJD7SppSqlNSqkiYL+UMhkYD8yzt39t728wGAyGABFoZ/aJwCIgASjyKncvhOvtlSkEEuvVdZcdhJTyWuBagBkzZjBt2rTW0uw3qqqqyMzMDLYMx+hoCKdodIoOXzhFm1N0NIZTdDpFR0bGoReFCrShOA94C8gHOnqV19ivLq+yBCDPq26BXbbd14GVUjOBmQCZmZlWQxftFDIzMxv8co40HQ3hFI1O0eELp2hzio7GcIpOp+hoiIAZCnvYaTwwA92DGCSljADGAKvsajlSyn7AHiBRKZUrpVwOnAK8DEwG/hAozQaDwWAAYVn+X0YPQEp5MnCWUuo6+/P5wE3oqKfLfEQ93auU+tqOevoPOurpUxP1ZDAYDIElYIbCYDAYDG0TM+HOYDAYDA1iDIXBYDAYGsQYCoPBYDA0iDEUBoPBYGgQYygMBoPB0CDGUBgMBoOhQdq8oZBS+j8ZezvCtFfTMW3VOKaNmkdbba82aSiklEdJKR+TUnZUSjl2IoiUso/X+6D9QNpCe5m2ahzTRs3DtFfr0aYMhZQyVEp5L3p9ivl2llnHIaUUUsq7gE1Syvvs4oD/UNtCe5m2ahzTRs3DtFfr06YMBZAKdACeB0KllJfYaT+cRiiwAhgOTJZSpiulXFLKQLd3W2gv01aNY9qoeZj2amUcbyiklFOklLcDKKUy0QsfnYNOKd4Z+IeUcnQQJQIgpTxVSvmmlPI6IEopNV8ptRr4Evhfu5rfu51tob1MWzVJm2mjZmDay7842lBIKaehv+TjpZSX2MXfA3crpc5USj2BXszoRLt+UMYhpZQdgCvRKdS7AA+6tSilHgKOklKOVkpZdhZdf+lwfHuZtmqSNtNGzdNp2svPONJQeDXgz8BJwM3A6VLKeKVUAbDGq85SYABAEB1F/YAypdRc4O/o9TNO9dJ4D/rHex0wzI862kJ7mbZqHNNGzcO0l59xlKGQUg6x34YAKKVylFIHgG3AWuA6e7sFhEkpLwX+BSwN8BOfsF/dOlcBqVLK05VSVcAc4DyvH0IYcBwwBFjvRx2ObC9vrcFqq/o4sa2c0kZeOhzXRvU12jqD1l71dDi2vQ4XRxgKKeUYKeUc4D0pZbJSqsa7i2h/+e8Ag6WUSUopF9ADGAnMUEq9GgjrLKU8Xkr5DnCLl4Ms2t78KHp9DZRSnwBdpJQn2NtigGOUUtcppUr9pCPCvd0J7SWlnCilfEZKOd5+srKC1FZ1dNhljmgrW9vz9mtCkNuovg5HtFE9neOklKfamiypo5si7c2BbC9fOtzLOTumvVqToK1HIaUUdiP/HZgAvAvEA+8ppbbadVKBaK/PfwVuAOYqpa4OsM5h6B/jU+ju6wil1EV2nXSgBHgM2AS8ju4CP66U2hRAHUFtLy+NRwP3Ae8BaUBvpdS1dp1AtlVDOtKADkFsq57AE8DntrY0pdSf7G1+b6Mm6ghqG9XTeQ3wEPq7fFMptdxrWyDbqyEdQb9f+Ytg9ig6269vKaUmKaVeBI5GW39si70cGGV/HoMe/3s+wI2eYr8moP9p5gKvAquklJFSymOAn2ztDwDV6B9qTiv/SJuiQxHc9nJrjAMKlFKvK70i4SQp5WQp5QgC21YN6fiR4LZVGhCilHpFKfV3tMP1BCnlcALTRk3REew28mYe+oFyoZYjY21dpwA/ELj2akiHE+5XfiHgPQop5STgXmAn8D6wwN0dlHqSTLJS6hYpZTL6B7zX3pYBHLCdQ4HWOVsp9bmU8mUgB93F/RA99vg/6Bv3fq99o5RS5cHUEcj28vGdrkOvbf6NUuobKeV/vDRGKaVyvfb1V1s1WUcg2kpK+QfgLOB/lFK77LKZwLu2tunANcD56KdSf7VRi3QE4f/vIJ12+XhgErDK/l+oc5+w6/i1vZqiI9Dt5W+CYSieQ3dzdwBnA0VKqWfsbROBscBz7i9aShmmlKoOqMiDdf4e2KGUelVK+SCwXCn1qZTyS+C/Sqk3pfap1LT22GMLdITbY6QBw0vjTmA6UIOO7vg/IA9YBXQC/qWU2mCP57r82FZN1RGQ35aUsiN6LfhE9Prv79razkGPW9+nlKqUUr4KvKCU+tEfbdRCHQH///OlUylVaW+LBC5ARza9AlQqpaqllOFAtb/bqwk6Itx12hMBG3qSHi//DvSTylq0w2eM9MxWjAaGeD8NBOFH6kvnf9AzPHuibzRD7TpZ6LFRlFKt/SNtqY6AGQkfGn8DZqNnxO5C/yP9FXgOPRSwwdbYqgb1MHQEwkgIpVSRUupCYAY6hr6vUqoG+A79P/hHqecCVKF7QX5poxbqCPj/ny+d7u1KqQrgCyAJ+AC4296nKhDt1QQd7c5IgJ8NhZTS7W9wOxdDgH1AgtTRAJvRP9KJ9i7zgAFSyl7+1HUYOo9FO8mGSCkXAPnAx+1NRytoXAJMsbvhw9GOP/eTWKuEBTpFRxO1ucNNtwO/AWdKKaOVUlnooIThwEdALnCgveloBZ1nuH0BNiOBM9A+lAdby0A4RYcT8cssRSnlacDFQJaU8p9ANoDSYZw/oht3LNoiF2En7LK/nMlKxyL7nWbqLADClFJZUsq/ABXefon2oKMVNRbieQjJBa62/9kOe5KRU3Q0U5v3ud4E7gf62cNLm9FPrDFKqcL2pKOVdfaxb9ybge3AaUqnyGg3OpxMqxsKe6zwcnQ3fwD6x/cF8IOUsh86OmEbcJKUshtwMvCNe/8AGomW6Fxga8xqbzr8oPFbW+Pi9qajBdr6AyeggxH2SCl3Ap8Ai4A7lM4q2lpGwhE6/KRzMXC7Umpje9PhdFpl6Enq8MwE+2N/IFsptQh4DegOTJVSnoQOXytRSr0NvAB0BT5XOjTW77SCzhfakw4/a3y+Pek4TG33AcVKqTzbB3As8LBS6rLWMPZO0REAnZcqpbLbi462xGEbCinlBejp6g8C2I7EDCnl+bZjZx861e5O4Aql1Gy73hbgXqXUq4eroS3pdIqOtqDRKTpaQduVSqn/2vXKgDOVUv9qTzraik6n6GhrHJahkFJGoSOVrrE/n2ZvehI99vkl2v+QASQqpSpk3anuAXH+OEWnU3S0BY1O0dFK2iqlXsTG7YtrlUgip+hoKzqdoqMt0ux5FFLKHuhQw8+BX5VSu+3y3wEXApcrHXKH1Dl+CqVO0/Gq/aQXEJyi0yk62oJGp+hwsjan6GgrOp2io63TrB6F1InC7kPHpndBr9zkZgGwF7jKriuAvlLKeUBhgH+kjtDpFB1tQaNTdDhZm1N0tBWdTtHRHmhSj0JKeQ46N9M84BWl1CS7/BVgnVLqcbuhe6DH/n5ARw5kAqFKqRL/yHemTqfoaAsanaLDydqcoqOt6HSKjvZEg4ZC6vwlLwNlwB60E+hE4Cul00h0R09iOl0ptU/qzKZz0I6gGSpA4WNO0ekUHW1Bo1N0OFmbU3S0FZ1O0dEeaWzoyQW8rPQ09r8Dg9DT1YdIKfsppXYCXwFTpM51NBq4VSl1QoAb3Sk6naKjLWh0ig4na3OKjrai0yk62h2NTbjLR3ffUErlSik7o2dSbwJuBf6ITpj1q9IRAQEJdXWwTqfoaAsanaLDydqcoqOt6HSKjnZHU30UAr2o0H+VUtPsspnoeOMIdLhZsQpyrhOn6HSKjrag0Sk6nKzNKToawyk6naKjPdGcFB5hwHdSytHAqejUuhuVUvl+UdZynKLTKToawikanaLDF07R5hQdjeEUnU7R0S5o8jwKKeUU4FN0WNlbSqk3/SmspThFp1N0NIRTNDpFhy+cos0pOhrDKTqdoqO90JwexX7gLuAZ5eyc607R6RQdDeEUjU7R4QunaHOKjsZwik6n6GgXNMdQ/KSUUn5T0no4RadTdDSEUzQ6RYcvnKLNKToawyk6naKjXRDwpVANBoPB0LYI2FKoBoPBYGibGENhMBgMhgYxhsJgMBgMDeKXNbMNhiMBKWU0cDuwXSn1mpTyCvRs39uUUo8HVZzB0IqYHoXB0HKi0Wmsr7A/L0KvcfBpsAQZDP7A9CgMhpbjDr88XkppATvQqatvAzZIKbcDycC/gauBb4B/oWcJh6GX2pwr9boJD6GNTAzwNXCdUmpfAK/FYDgkpkdhMLScu+zXdeibvK/hphggClgGnAXMBB5Dr5fwiF3nTnTSuk+Bp4CpwIv+Em0wNBdjKAyGljPPft2rlHoH8LXgjQu4GZ3uGuBNpdQzQBbQyy5zr908Az2UFQOc4hfFBkMLMENPBkPLacps1TKlVKWUssr+XGi/1gCh9nsBVKMNRo1dZh7iDI7B/BgNhpZThO4x9JVSXoz2T7SET9EPbZcD3dHZTme0ikKDoRUwhsJgaCFKqSq0vyEB+A+e3kBzedg+zkTgObSPYlErSDQYWgWT68lgMBgMDWJ6FAaDwWBoEGMoDAaDwdAgxlAYDAaDoUGMoTAYDAZDgxhDYTAYDIYGMYbCYDAYDA1iDIXBYDAYGsQYCoPBYDA0yP8DqnCR9PkbOr8AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plotting the actual and predicted price values\n",
    "actual[13:].plot(label=\"actual\")\n",
    "preds.plot(label=\"predicted\")\n",
    "plt.title(\"TFT (window of 7 days), MAPE : {:.2f}%)\".format(mape(actual,preds)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "0a7ac58b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MAPE: 1.934490166604519\n",
      "rmse: 260.9608\n",
      "mae: 186.95518\n"
     ]
    }
   ],
   "source": [
    "#metrics of time series forecast\n",
    "print(\"MAPE:\",mape(actual,preds))\n",
    "print(\"rmse:\", rmse(actual,preds))\n",
    "print(\"mae:\",mae(actual,preds))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aabb15d3",
   "metadata": {},
   "source": [
    "We get an MAPE of 1.96% on the TFT model with a sliding window of 7 days."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "51c30cb0",
   "metadata": {},
   "source": [
    "# TCN Model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "48b6da18",
   "metadata": {},
   "source": [
    "We next train the Temporal convolution network (TCN) on a sliding window of 30 days and 7 days"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "12adcd2f",
   "metadata": {},
   "outputs": [],
   "source": [
    "from darts.models import TCNModel\n",
    "#30 day window\n",
    "EPOCH = 300    \n",
    "FC_N = 30               # forecast periods\n",
    "FC_STRIDE = 10\n",
    "DROP = 0.1\n",
    "LEARN = 0.001\n",
    "\n",
    "model_TCN = TCNModel(\n",
    "        input_chunk_length=30,\n",
    "        output_chunk_length=1, \n",
    "        n_epochs=EPOCH,\n",
    "        dropout=0.2,\n",
    "        dilation_base=3,\n",
    "        weight_norm=True,\n",
    "        batch_size=16,\n",
    "        kernel_size=5,\n",
    "        num_filters=4,\n",
    "        optimizer_kwargs={'lr': LEARN},\n",
    "        log_tensorboard=True,\n",
    "        force_reset=True,\n",
    "        random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "edf97d59",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "db5df66847e04c85a0ea6137245b2264",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<darts.models.forecasting.tcn_model.TCNModel at 0x170d9467b88>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# training: load a saved model or (re)train\n",
    "# training\n",
    "model_TCN.fit(  ts_ttrain, \n",
    "            past_covariates=covF_ttrain, \n",
    "            verbose=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "658e809d",
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture\n",
    "#Testing our model using historical_forecasts on ts_ttest \n",
    "preds = model_TCN.historical_forecasts(\n",
    "    ts_ttest, start=0.3, forecast_horizon=1,past_covariates=covF_t,retrain=False\n",
    ")\n",
    "\n",
    "# scale back:\n",
    "preds = scaler.inverse_transform(preds)\n",
    "actual = scaler.inverse_transform(ts_ttest)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "975dc6f8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'TCN (window of 30 days), MAPE : 2.43%)')"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plotting the actual and predicted price values\n",
    "actual[41:].plot(label=\"actual\")\n",
    "preds.plot(label=\"predicted\")\n",
    "plt.title(\"TCN (window of 30 days), MAPE : {:.2f}%)\".format(mape(actual,preds)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "8da4c879",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MAPE: 2.42711640894413\n",
      "rmse: 298.4291\n",
      "mae: 215.51251\n"
     ]
    }
   ],
   "source": [
    "#metrics of time series forecast\n",
    "print(\"MAPE:\",mape(actual,preds))\n",
    "print(\"rmse:\", rmse(actual,preds))\n",
    "print(\"mae:\",mae(actual,preds))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ece3b1ea",
   "metadata": {},
   "source": [
    "The TCN model gives an MAPE of 2.43% for a 30 sliding window.Lets check the 7 day window now."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "d8c06b59",
   "metadata": {},
   "outputs": [],
   "source": [
    "from darts.models import TCNModel\n",
    "#30 day window\n",
    "EPOCH = 300    \n",
    "FC_N = 30               # forecast periods\n",
    "FC_STRIDE = 10\n",
    "DROP = 0.1\n",
    "LEARN = 0.001\n",
    "\n",
    "model_TCN = TCNModel(\n",
    "        input_chunk_length=7,\n",
    "        output_chunk_length=1, \n",
    "        n_epochs=EPOCH,\n",
    "        dropout=DROP,\n",
    "        dilation_base=2,\n",
    "        weight_norm=True,\n",
    "        batch_size=16,\n",
    "        kernel_size=5,\n",
    "        num_filters=3,\n",
    "        optimizer_kwargs={'lr': LEARN},\n",
    "        log_tensorboard=True,\n",
    "        force_reset=True,\n",
    "        random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "f458559b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "85de64ad2cb84683a7ad830368d47bf6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<darts.models.forecasting.tcn_model.TCNModel at 0x170da188608>"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# training: load a saved model or (re)train\n",
    "# training\n",
    "model_TCN.fit(  ts_ttrain, \n",
    "            past_covariates=covF_ttrain, \n",
    "            verbose=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "fa66c914",
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture\n",
    "#Testing our model using historical_forecasts on ts_ttest \n",
    "preds = model_TCN.historical_forecasts(\n",
    "    ts_ttest, start=0.1, forecast_horizon=1,past_covariates=covF_ttest,retrain=False\n",
    ")\n",
    "\n",
    "# scale back:\n",
    "preds = scaler.inverse_transform(preds)\n",
    "actual = scaler.inverse_transform(ts_ttest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "d32fec90",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'TCN (window of 7 days), MAPE : 3.11%)')"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plotting the actual and predicted price values\n",
    "actual[13:].plot(label=\"actual\")\n",
    "preds.plot(label=\"predicted\")\n",
    "plt.title(\"TCN (window of 7 days), MAPE : {:.2f}%)\".format(mape(actual,preds)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "7cb3b1f0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MAPE: 3.111855499446392\n",
      "rmse: 365.04944\n",
      "mae: 287.0434\n"
     ]
    }
   ],
   "source": [
    "#metrics of time series forecast\n",
    "print(\"MAPE:\",mape(actual,preds))\n",
    "print(\"rmse:\", rmse(actual,preds))\n",
    "print(\"mae:\",mae(actual,preds))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "53cddc82",
   "metadata": {},
   "source": [
    "The TCN model gives an MAPE of 3.11% for a 7 sliding window."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "24073380",
   "metadata": {},
   "source": [
    "# NBEATS MODEL"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "806fb730",
   "metadata": {},
   "source": [
    "Next we train the NBeats model on both the 7 and 30 day windows."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "15db7ec0",
   "metadata": {},
   "outputs": [],
   "source": [
    "#NBEATSModel - 30 day window\n",
    "NBEATS_model = NBEATSModel(    input_chunk_length=30,\n",
    "                        output_chunk_length=1,\n",
    "                        num_stacks=32,\n",
    "                        layer_widths=16,\n",
    "                        n_epochs=100,\n",
    "                        log_tensorboard=True,\n",
    "                        generic_architecture=True, \n",
    "                        random_state=42,\n",
    "                        force_reset=True,\n",
    "                        save_checkpoints=True,\n",
    "                        torch_device_str=\"cuda\"\n",
    "                    )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7e109bd0",
   "metadata": {},
   "source": [
    "Lets train the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "115b593f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "787c10cfdab8485bad55f8ecbde8f2a5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<darts.models.forecasting.nbeats.NBEATSModel at 0x17084c0c248>"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# training: load a saved model or (re)train\n",
    "NBEATS_model.fit(  series=ts_ttrain, \n",
    "            past_covariates=covF_ttrain,\n",
    "                verbose=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1fa968fa",
   "metadata": {},
   "source": [
    "Now that the model has been trained lets test the forecast on the test data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "0e20a2c9",
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture\n",
    "#Testing our model using historical_forecasts on ts_ttest \n",
    "preds = NBEATS_model.historical_forecasts(\n",
    "    ts_ttest, start=0.3, forecast_horizon=1,past_covariates=covF_ttest,retrain=False\n",
    ")\n",
    "\n",
    "# scale back:\n",
    "preds = scaler.inverse_transform(preds)\n",
    "actual = scaler.inverse_transform(ts_ttest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "85381778",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'N-BEATS (window of 30 days), MAPE : 14.45%)')"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plotting the actual and predicted price values\n",
    "actual[41:].plot(label=\"actual\")\n",
    "preds.plot(label=\"predicted\")\n",
    "plt.title(\"N-BEATS (window of 30 days), MAPE : {:.2f}%)\".format(mape(actual,preds)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "808c3efb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MAPE: 14.445382356643677\n",
      "rmse: 1408.7769\n",
      "mae: 1241.7488\n"
     ]
    }
   ],
   "source": [
    "#metrics of time series forecast\n",
    "print(\"MAPE:\",mape(actual,preds))\n",
    "print(\"rmse:\", rmse(actual,preds))\n",
    "print(\"mae:\",mae(actual,preds))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "156c2899",
   "metadata": {},
   "source": [
    "We get a MAPE of 14.45% which isnt great.Lets check the performance for the 7-day window."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "1d02d526",
   "metadata": {},
   "outputs": [],
   "source": [
    "#NBEATSModel - 7 day window\n",
    "NBEATS_model = NBEATSModel(    input_chunk_length=7,\n",
    "                        output_chunk_length=1, \n",
    "                        num_stacks=32,\n",
    "                        layer_widths=16,\n",
    "                        n_epochs=100,\n",
    "                        nr_epochs_val_period=1,  \n",
    "                        log_tensorboard=True,\n",
    "                        generic_architecture=True, \n",
    "                        random_state=42,\n",
    "                        force_reset=True,\n",
    "                        save_checkpoints=True,\n",
    "                        torch_device_str=\"cuda\"\n",
    "                    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "6ed966c2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d4d36333310f4bfe8c23189db08b0172",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<darts.models.forecasting.nbeats.NBEATSModel at 0x1716ad6b9c8>"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# training: load a saved model or (re)train\n",
    "NBEATS_model.fit(  series=ts_ttrain, \n",
    "            past_covariates=covF_ttrain,\n",
    "                verbose=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "91ab10b4",
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture\n",
    "#Testing our model using historical_forecasts on ts_ttest \n",
    "preds = NBEATS_model.historical_forecasts(\n",
    "    ts_ttest, start=0.3, forecast_horizon=1,past_covariates=covF_ttest,retrain=False\n",
    ")\n",
    "\n",
    "# scale back:\n",
    "preds = scaler.inverse_transform(preds)\n",
    "actual = scaler.inverse_transform(ts_ttest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "3e9c7700",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'N-BEATS (window of 7 days), MAPE : 3.14%)')"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEsCAYAAAAsMK9uAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAABxTklEQVR4nO2dd3xUxfbAv5MOJKGGFnrvdUSQ3sSKiFiwYXliw/f0qe/ZeM9n+/mevSuiYgUVRBQUEemdUZAOoYdAgFASAklImd8fc3ezKRtCskl2Yb6fz352d+7ce8/cu3vPnHNmzgitNRaLxWKxAARVtAAWi8Vi8R+sUrBYLBaLG6sULBaLxeLGKgWLxWKxuLFKwWKxWCxurFKwWCwWixurFCxnhRBiqRCi61nu874QYnwJz3ebEGJJSfb1BUKIOkKIRUKIE0KIV8rg+LuFEEN8fVyP4/9VCPFiWR3f4h0hxHAhxJSKluNssUrBxzh/8oNCiCoeZX8RQiwoYh8thDgphEgVQiQJISYLIap5bF8ghEh3trteP+Y7RlMhRI4Q4l2PMs/6OUKINI/vNwkhqgkhPhZCJDoPvW1CiH8WIeeVwAmt9ZqzuSZa63u01s+ezT5+xFggCYjWWj+cf6MQ4ud81/m0EGJ9+YvplQnAzUKI2sWpLIQY4Pwev8tX3tkpX5CvXAghdgohNhVyLM/fbZIQ4jshRD1n2yTnWnleuz9L0kAhxEAhxHohxHEhxBEhxHQhRGwR9Z916mcJIZ4uot4nTptbeJQ96rRlgxCig0d5byHE9577a61/ADoIITqVpF0VhVUKZUMI8Lez3Kez1joSaAZUB57Ot32c1jrS43Vlvu23AseAG4QQ4QCe9YG9wJUeZV8CrwGRQFugKjAc2FGEjPcAn59luwKdxsAm7WWWp9b60nzXeRnwbblKWARa63TgZ8zvo7gcBi4SQtT0KBsDbCukbj+gNtBMCHFBIdvHOdelFVAN85tz8b98v+nOZyGjJ5uAYVrrakB9IA54r4j624F/ALO8VRBC9AGa5yurB9yJ+Y++D7zolIcArwAPFnKoyZiORcBglULZ8BLwiGdvv7horVOAH4B2Z7nrrcBTQCaQX2F44wLgK631Ma11jtZ6i9Z6amEVhRBhwCBgofM9wrE8ajnfn3J6XtHO9+eEEK87nycJIZ5zPg8QQuwTQjwshDgkhDgghLjd4zw1hRA/CCFShBCrKPjHvEgIsVoIkey8X+SUD/TsoQsh5jr7u74vEUKM8NI2b8echHkY/sPpyRbp5hFCNAH6UoTiFELcIoTY4/Ron8y3rYcQYrnT4z0ghHjbue4IId4R+dxXQogfhRAPOp//KYRIcCy+rUKIwR5VFwCXFyV7Pk4D3wM3OMcOBq4Dviyk7hhgBvCT87lQtNZHgWlAB291SorW+qDWer9HUTbQooj6n2qtfwZOFLbdeci/BYzLt6kRsMb5j87FKAcwyuAHrfXuQg63gLO79hWOVQplg8L8GB452x2FENWBEcCKs9inL9AAmAJ8Q/F7hSuA54UQtwshWp6hbksgR2u9D9w90NVAf2d7P2AP0Nvj+0Ivx6qLsUxiMT2vd5x2A7wDpAP1gDucFwBCiBqY3t2bQE3gVWCW06NdDrQQQtRy/tQdgAZCiCghRCWgO7A4vyBFHVNrfRvmQejq0c49wzW6FVistd5V2EYhRDtMD/YWTI+2Jua+ucgGHgJqAb2AwcB9zrZPgdFCiCDnWLWc7ZOFEK0xD7ALtNZRwDBgt8dxNwNn2wv/jNzf0TBgI+D54EUIURkYhblGX2Ks1LDCDubIew1wVq5Hj/2PO713b9sbCSGOA2mY/93/SnIeh4eARVrrdfnKtwMdnc7eEGCjEKIhRnm+7OVYm4Emrs5SIGCVQtnxL+ABIURMMev/4fyokzA9kg/ybX/T+WO4Xp4++jHAz1rrY8BXwKWieD7kBzB/5nHAJiHEdiHEpV7qVqNgz2oh0N95CHfCPFj7CyEiMFZIgYewQybwjNY6U2v9E5AKtHZ6pNcA/9Jan9Rab8A8DF1cDsRprT/XWmdprScDWzBusXSMMu4HSGAdsASjpHo6+x0pRBavx/Qie1HcCkwqYvsoYKbWepHWOgMYD+S4Nmqtf9dar3Dk2I35DfR3tq0CkjGKAMyDaIHW+iBGmYQD7YQQoVrr3VprTzfgCYwSLjZa62VADUfh3IpREvkZCWQAc4CZGLdp/l7xm87v+k/gAPB3j22P5PtNf4oXtNbVtNZeBxxorfc67qNaGIt5yxmaWCjOQ/5uzP83/zmOAM8D8zDtfAR4A/gncLUQYqEQYoYQwlPRu/4z1UoiT0VglUIZ4TzQZgKPeZYLIZ4QuYG19z02dXN+1BGY3uRi5+Hq4q/OH8P1Gu8crxJwLY5pr7Vejokf3FgMGdO01i9orbtjeq3fAN86vef8HAOi8pUtBAYA3YD1wK+Yh1hPYLvWOsnLqY9orbM8vp/CxDZiMA+WeI9tezw+18/33bXdFVR0yeOyUhY48vTHu9VypmMWC6cXWxco1P3mcS5327TWJwG3ohJCtBJCzBQm8J8CvIB5yLn4FLjZ+XwzjptKa70d48J4GjgkhJgihKjvsV8URqGcLZ9jOgwDgemFbB8DfOMosQzgOwq6kFy/21it9U1a68Me217O95v26n4qLo6b6lNghtNZOVtex3RYCr1eWuvJWutuWutLMdZoBsb6eRnTkfiWvFaD6z9zvASyVAhWKZQt/wbuwuMB4zyEXYG1e/LvoLXOBCYCTSme//VqIBp413mYJDrnO5vAoiuW8QJQxTl3fuIwg008H5bLgNaODAu11pswVs7leH8IF8VhIAto6FHWyOPzfkzgl3zbE5zP+ZXCQs6sFM50zOIyBvhOa51aRJ0DeLTNcb94BnPfw/RwW2qto4EnAOGx/QvgKiFEZ8zggO9dG7TWX2mt+zht0cB/PfZri+mpny2fY9xXP2mtT3lucHrDgzAjm1y/u1HAZY6rqCIJwQS/S+KyGQy85NEmgOVCiDydLKcz9gLwMMa1Gu/8h1ZjrGYXbYHdzraAwCqFMsTpwX0N/LW4+zgulNsxvtGdxdhlDPAx0BHo4rx6A12EEB3PcK7xQogLhBBhjlXyN0yPZmshbcnEBNf6e5SdAn4H7if3obsMY36ftVLQWmdjeptPCyEqOz54z97jT0ArIcSNQogQIcT1mID8TI9ztwZ6AKu01hsxD8kLgUVeTnumY54RD2tt0hmqTgWuEEL0cXzvz5D3PxgFpACpQog2wL2eOzvxnNWYh/U0rXWac/7WQohBwow6S8f8drI9du2PGYHkkneSMEH0InFiI/2BJwvZfAtmNFJrcn93rYB9wOgzHduXCCFGOtcgyHHXvooJCB/1Uj/U+b0HASHCDJoIdja3wsRfujgvMBZAfkvpKWCSE+Dei3F/1sFYVZ7/2zzXPiDQWtuXD1+YAN8Qj+8NMX/UBUXso4GTGN+6q7cxzGP7AucYqR6v3zEWQRbQsZBj/oQxzwuVyyl7CtjgnPOoc56LipDzckzswrPs/zAPoXDn+zinPXU86kwCnnM+DwD2ebtmGBfSTEemVcCzwBKPun2ctic7733yHWs5MN/j+1Rg8xnumddjespexP6jMS4nUYzfxxjMQ+QI5mHr2fZ+GEshFROPecaz7U6dm53rO9CjrJNzrU4493EmUN/ZFoF5UHvej9+Au7zIV+D+eGz7C87v2JHzgULq/ANQHr/bv3g51iTMKCfP33RSEdctFejrZdsDwC7MfygRM+Ciscf294H3851b53vdVsR/s0W+staY/2iIR9mjmHjgJjz+jxi3aucz/S786SUcwS2WYiHM7OIH9FlOYLP4BiFEP4wbqYnWOqcY9R8AGmqt/+F8D8O4kjppY/1ZyghhJnveorW+rqJlORusUrBYAgQhRCimF/yn1vqZipbHcm5iYwoWSwAghGiLiffUw4yQsVjKBGspWCwWi8WNtRQsFovF4sYqBYvFYrG4CXSlkH9YmQZ0YmJioeWB/gq0dgWavOdKOwJN3nOxTQEgr1cCXSkUSnZ29pkrBSCB1q5Ak9cbgdaOQJO3OARamwJNXk9KkhvEjZQyCjPLtT3QUym1QUr5CXAF8B+l1NtOvXaY5F4hwHil1FwpZSRmZmYMMFMp9aJT9yVM7py9wO1KqdOlkdFisVgsxae0lkIaRgF4JgF7AjO7z5MXMCmQh2FmaYLJCTRLKdUHGCClbCCl7ArUU0r1xcwMHFVK+SwWi8VyFpRKKSilspRSh/OVHSikaj2lVJxSKgU4IqV05Yuf42z/FWMdeJbNBi4qjXwWi8ViOTtK5T46CzwzPSYDNTD5xVMKKdufrywPUsqxOMvbjRs3jqFDhxY4WWZmJgkJZ5vk0v8JtHYFmrzeCLR2BJq8xSHQ2uTv8sbGes8MX15KwTNHSzVM0q5jmNS2x52y3UAwueluXfXyoJSagFmMHLxE0RMSEopsdKASaO0KNHm9EWjtCDR5i0OgtSnQ5PWkvEYfJUopW0opo4EaSqkkzFKQFzvbh2CyW3qWDQOWlpN8FovFYsEHloKU8idM3vHWUsr3MItKDAeCpZRNlVIPY4LPHzvncy1z9yHwhZTyduBHpVQCkCClPCClXIwZffRSaeUrDp999hkdOnSgW7du5XE6i8Vi8VsCPfdRqd1HmzZton379tSvX589e/YQElJeHrWzJ9BM0kCT1xuB1o5Ak7c4+HubFixYQFhYGBddZMbGlETeyMhIUlOLWrjPpwhvG87JyWtnw65duwDYv38/c+fOrWBpLBZLILJgwQKWLVtW0WL4hPNeKezfv9/9edKkSRUniMVi8TtGjBhB9+7dad++PRMmmPEts2fPplu3bnTu3JnBgweze/du3n//fV577TW6dOnC4sWLeeihh5g6NXf6VmRkJACpqakMHjyYbt260bFjR2bMmFEh7SoK//WVlBOeSuH777/n2LFjVK9evQIlslgsngjh1dNRKorjOv/444+pUaMGaWlpXHDBBVx11VXcddddLFq0iKZNm3L06FFq1KjBPffcQ2RkJI888ggAb731VqHHi4iIYPr06URHR5OUlETPnj0ZPnx4mbWxJJz3lsKBA2aunRCCjIwMvv766wqWyGKx+AtvvvkmnTt3pmfPnsTHxzNhwgT69etH06ZNAahRo8BUqiLRWvPEE0/QqVMnhgwZQkJCAgcPHiwL0UvMea8UXJbCyJEjAetCslj8Da01+/bt8/kC9WdiwYIFzJ07l+XLl/Pnn3/StWtXOnfuXKxefXBwMDk5OW75T582Kdy+/PJLDh8+zO+//87atWupU6cO6enppbtAPsYqBUcp3H///URFRbFy5Uo2b95cwVJZLJaKJjk5merVq1O5cmW2bNnCihUryMjIYOHChe4BKkePmvm1UVFRnDhxwr1vw4YN+f333wGYMWMGmZmZ7mPWrl2b0NBQ5s+fz549e8q5VWfGKgVHKbRo0YLrr78egE8//bQiRbJYLH7AJZdcQlZWFp06dWL8+PH07NmTmJgYJkyYwMiRI+ncubP7mXHllVcyffp0d6D5xhtvZOHChfTo0YOVK1dSpUoVAG666SaUUkgp+fLLL2nTpk1FNrFQzut5CtnZ2YSFhZGTk0NGRgarV6+mT58+NGzYkL179/pc2NLi72O18xNo8noj0NoRaPIWh0BrUwDIa+cpFMahQ4fIyckhJiaGsLAwevXqRXBwMPHx8WRkZFS0eBaLxVLunNdKweU6ql+/PgBBQUHUqVMHwO9GBFgsFkt5YJUCUK9ePXeZVQoWi+V85rxWCq45Ci5LAaBu3bqAVQoWi+X85LxWCvndR5BrKSQmJlaITBaLxVKRWKWAtRQsFovFhVUKFB5TCBRLYfr06UybNo3s7OyKFsVisXhhwYIFXHHFFQD88MMPvPjii17rHj9+nHffffesz/H000/z8ssvl1hGF+e1Ugj0mMKOHTsYOXIko0aNomPHjnzzzTfuqfUWi6XsKUlnbPjw4Tz22GNet5dUKfiK81opBHpMYcWKFe7Pmzdv5vrrr6d169Y89dRTrF27tlj5XSwWS+Hs3r2bNm3aMGbMGDp16sSoUaM4deoUTZo04ZlnnqFPnz58++23zJkzh169etGtWzeuvfZa90I5s2fPpk2bNvTp04fvvvvOfdxJkyYxbtw4wHQ+r776ajp37kznzp1ZtmwZjz32GDt27KBLly48+uijALz00ktccMEFdOrUiX//+9/uYz3//PO0bt2aIUOGsHXrVp+0+7xNnZ2VlcXBgwcRQrgVAQSWpbB69WoAxo8fT4MGDXjuuefYvn07zz//PM8//zxdu3Zl8eLF7in2FksgIvrlAPUA31rBetGZ+8Rbt27lo48+onfv3txxxx3uHnxERARLliwhKSmJkSNHMnfuXKpUqcJ///tfXn31VW666Sbuuusu5s2blyeFTn7++te/0r9/f6ZPn052djapqam8+OKLbNiwgbVr1wIwZ84c4uLiWLVqFVprhg8fzqJFi6hSpQpTpkxhzZo1ZGVl0a1bN7p3717q63LeWgoHDx5Ea01MTAyhoaHu8kCyFFxKoXfv3owdO5adO3cyd+5c7rnnHqpWrcqaNWuYP39+BUtpsQQuDRs2pHfv3gDcfPPNLFmyBMD9kF+xYgWbNm2id+/edOnShU8//ZQ9e/awY8cOmjZtSsuWLRFCcPPNNxd6/Hnz5nHvvfcCJrNq1apVC9SZM2cOc+bMoWvXrnTr1o0tW7YQFxfH4sWLufrqq6lcuTLR0dEMHz7cJ20ulaUgpYwC5gLtgZ5KqQ1SyuuBvwHpwBilVLyUcpJT5yQwSyn1kpQyEvgciAFmKqVedI75EtAT2AvcrpQ6XRoZvVFYPAGgevXqhIaGkpKSQlpaGpUqVSqL05earKws1qxZA4CUEoCQkBAGDx7M4MGDiYmJ4dlnn2XevHnuAJfFEojoRUEVlksof5ps13eX9a21ZujQoUyePDlPvTlz5vhs4RytNY8//jh33313nvLXX3+9TBbnKa2lkAZcAUwFkFKGAn8HBgDjnZeL25VSA5RSLznf78IoiD7AACllAyllV6CeUqovsAkYVUr5vFJYPAHI407yZxfSxo0bSUtLo1mzZtSsWbPA9kGDBgGmJ+LJunXrGDNmDDt37iwXOS2WQGbv3r0sX74cgMmTJ9OnT58823v27MnSpUvZvn07AKdOnWLbtm00b96cXbt2sWPHDve+hTF48GDee+89wAStU1JSCqThHjZsGB9//LE7VpGQkMChQ4fo168f06dPJy0tjRMnTvDjjz/6pM2lUgpKqSyl1GGPopbARqXUaaXUUqCjU66BD6WUv0opOztlvYA5zudfMdaBZ9ls4KLSyFcUhQ1HdREIcQWX6+iCCy4odHvPnj2JiIjgzz//JCkpyV3+xBNP8NlnnzFixAhOnTpVLrJaLIFK27Zt+fTTT+nUqRNHjx51u3pcxMTEMGnSJEaPHk2nTp3o2bMnW7ZsISIiggkTJnD55ZfTp08fGjduXOjx33jjDebPn0/Hjh3p3r07GzdupGbNmvTu3ZsOHTrw6KOPcvHFF3PjjTfSq1cvOnbsyKhRozhx4gTdunXj+uuvp0uXLlxzzTX07dvXJ232daC5GpDi8T3YeX9EKXVEStkGmIRRAJ51k4EaTtn+fGV5kFKOBcYCjBs3jqFDhxYQIjMzk4SEhCIFdUXqIyMjC9R1+fU2btxIgwYNijxOeeLZrgULFgDQqlUrr22VUrJkyRKmTZvGFVdcwbFjx/jll18AWL9+PbfdcQ+9r3qdvh0zqFPd90NZi3MfAoFAa0egyVscKqJNiYmJ5OTkMH58rsPj2LFjLF26lIyMDLc8rVu35vvvv8+zb2ZmJh07duS3337LU56QkMDQoUMZOnSoe//8w08TEhJ46aWX8nwfNWoUo0aNKlDvtttu47bbbitQfiaKcsX5WikcA6I9vmcDKKWOOO9bpJRIKYM96h7HKIPdGCXi2r8acDT/CZRSE4AJztcSr6fgMsXatGlToG6TJk0A47f3p5zonu3atGkTAEOGDPEq46WXXsqSJUtYu3Ytd999NzNnziQrK4sOHTqwfft2vv2jH9/ur0ZsDPz0P0Gn5r71TwZATvliEWjtCDR5i0NFtCkzM5OQkJASnTeQ74GvRx9tB9pJKcOklL2BdQBSymjnvTYQrpTKBlYAFzv7DQGW5ysbBiz1sXxuvMUUwP9HIKWnp7N+/XqCgoLo1q2b13r54wouv+YjjzzC3Y99B3XvACDhMPS5XzNXnd28huMnNKs327kQlnOTJk2asGHDhooWo9wptVKQUv6EeZB/CIwGXgMWAs85L4AvpJRLgBnAw07Zh8AVUsqlwAKlVIJSag1wQEq5GGgHTCutfN4I5JjC2rVrycrKom3btkRGRnqtJ6UkKiqKbdu2sXLlShYtWkR4eDiDh13NtHXDTKU9/yZGzOfEKbj0Uc173xdvUXOAq5/S9Lhbs2qTVQwWy7lCqd1HSqnLCin+Ol+dAgNolVKpwIhCyh8trUzFwduQVDizpaCU4r333uOVV16hWrVqZSajN84UZHYREhJCv379mDVrFvfddx9aa6644gqe/TySfYehW8ssdm54n8OLnmfUo3FMXdGU+17VzFoOE/8BdWsWdCc9++yzrFy5kjsf+pgFa2oB8KuCHu18306LxVL+nJeT1zIzMzl06FCB2cwuzmQpPP3003z88cd89NFHZSqnN4qjFNIyNLNXaqq3ugfq3sUfh4ZAs1fYX/U1JvwIoSHw6ZOhPPnEPwHNtl9G8sV4qBYJs5ZDh9s0c1bltQDS09N5/vnnmTVrFjf+Y627fOl6aylYLOcK56VScM1mrl27NiEhBY2loiwFrbX7obx0qfeQR1paGjfddBOffvqpj6TO5UxKIT1D02Os5tJHNV+oy6Dl+9D0/yD2QZbHmeDXM3cIOjQTjBs3joYNG7Ju3Tr0wa9YP0kwVMKRZLjleU1WVu4Df9WqVWbt6srtSa8yBHLMvMJlG3JISUm1yfgslnOA81IpFBVkPn5CE1nVu6UQHx/PoUOHAFiyZIlX//tPP/3EV199xSOPPOLTtNYpKSls3bqV0NBQOnXqVGidZz7VbNgFsTEwagCEH/sc9r2CrPkNnz8lWPG+4LGbjWsoIiKCZ555BoCnnnqKmKqn+eUVQauGcOgYLPoz97gLFy4EoMXAD0xB4oeQvpfkk0FUrXch4eHhdOzYkZtuuolXXnklzwQci8USGJyXSsFbPCHjtKbtLZquY6MIrdGL1NRUTp48maeOq5cOcPjwYfdMxvy4HqBJSUl5spmWlnXr1qG1plOnToSHhxfYvjZO87/JIAR8+x/Bt88EcUvPJQTteYzXHm7AzRcLLmyXN1Zwyy230KFDB/bs2cN7772HEILrBpptX8/LVXoLFiyA8IbsSr2Q4CD4+3UQqdcDEFZrIFlZWWzYsMGtDG+77TYyMzN91naLxVL2nJdKISYmhtGjRzNgwIA85Zv3QOJROHwcstrOhmqDC1gLnkoBvLuQXEoBYMaMGT6RG2Dfvn0ANG3atMC2rCzNHS9qsrPhgZHQq4N5+L/55pvExcUVmKLvIjg4mGeffRbA7e66bqDZd9oic9zTp0+b6f6xD5GdE8QNg+GVFx7ghcdMXqUbxr7JyZMnWbVqFX97+idE71RW7mzOgw8+6LO2WyyWsue8VAoXXXSRuzfryTqTpoSIMNBBkdB+Jl/OyZuPz6UU+vXrBxSuFI4ePcr69evd33/44Qefye6ycgobSvvK17AmDhrXhefvyrUGKlWqRLNmzYo87rBhwwgPD2ft2rUkJSXRoRm0bWxiC/P+MO1O03UR9e8B4B83muP37mD2X7oBKleuzAUXXMCKA8PQQZWgyTO8++77TJw40RdNz8PaOE2Ny3N473sb5LZYfMl5qRS88ed284B5/GZB0/AfICiMZ79uzpFkU56Tk4NSCsDdA3al0vXEFWvo1asX1atXZ+vWrT5bAMObUsjM0jz3mZHz/YcFkZXPbnZypUqV3CmC58+fn8eF9M18bVxHTV9Bi3BuHYZ79nOn5lClEuxIgINHzWS2lZucg0Y0hZrDue+++5g7d26xZVFbNO98p0k95f2BP+FHzbET8MNSqxQsFl9ilYIH65zEoV1awOAmP0HKSjKzg1nrhA3i4uJISUmhfv36XH755URERLBlTzbD/p7O5t25DyeX62jw4MFcfvnlgO+sBW9KYf0OSE2DFrFwyYUlS1cxePBgAHe+lusGmeN8twimzkuDWlcREZrJi3fnHj8kRHBhW/N52QZ46ztzHRrWNmX1e/yXzMxMhg4dyl133cXRowUyl5CRkcGoUaPo1asXb3x9kovu04x7XdPqJs2knzU5OXkf/Dk5mumLzOf4QyVqqsVi8YJVCh643EedmkPdunXg5No85Z5DQcPCwujRowfUH8ccFcZr3+Y+uBYtMk+s/v37uxe+KGulsHKzeb+wFJPI8iuFdk0EHZrCsROwNt3MKXz0+gzq1cqrdHo7uXCnL9Z8Pc8EuX98URBZKYf9qS34y4PvEhoaysSJE2ndujVTpkxBa80j7+Rw7fgc+l3/FdN+2sCKwzfy4DuVyMyC5rFw4Ajc/n+aC+/RHDyae31XbDSxH7BKwWLxNVYpOBw8qjl0DKKrGJ983bp14eQ6ANbvNA+k/PMDevfuDZEm99BSJ4SQkpLCH3/8QUhICL169WLYsGGEhoaybNkyDh8+TGnxqhScVBP5RxadDd27dyc6Oprt27ezZ88eAK53rAWCowjN2sVTt0cV2O8iJ6D9+S9wOhOu6AWdWwhuGGBSc2fUupt169bRv39/kpKSGD16NCNvfYFXvoapC2HV0TEgN0H9+yEng79dspZtXwq+eEoQGwNqC/xnUq5S+G5R7ueUk5By0rqQLBZfcd4qBa01aRm5DxNPK8E90/nkhjzb8iuFnr36QJUuAGzaDcdOaJYuXUpOTg5SSqpUqUJ0dDSDBg0iJyeHWbNmlVpu70rBvLtcOSUhJCTEPSLLZS1cOzB3++CmswgLLah0erU31oGLB64xX+645BRBQTDlN6haqzXz58/nvffeIzw8nO+XOSvaHZsDh78hKiKDmMhjsG4gU968jKNHj3DTxYI5rwiEgI9mwf4kk5dp+mJHXicxu7UWLBbfcV4qhe8Xa5pdr3nGo/f5pxM36OQM0jGWgun+b9wF6emZBZa/rNngIgiu7D7Gio25riPX6CTAZy6k9PR0jh07RkhISJ7V1pJTNVv2QlgodG5RqlMUcCG1biRoGfop7H6SW6+IKXSfqpHGzWTqw2Bn7fBGtbO5qjdkZsHb0zVCCO655x6WLltBcN0bTaXd43nxtj0k/xJB4szq9OsazsGDB7n//vsB48K6pr+xQF6eolm3A3buhzo1oK8zd2+vf+YttFgCkvNSKVSLhN2JpgfrmpG8znERuUbV1KlTB7KTCc7aT/ppmL1wO+np6TRv3pwaNczaPzsO5l1ke+l67Q4y9+/f313uCjbPmzev2BlIC8Plfqpbty5BQbm3bvUW8961JYSHlW5NBJdScMm6detWElaMg/gX87QpP8N6mPeHrhUEBeXK8PAN5vNb03CP4jquO5MdXJtq4Yd57rGr+cc//oEQgqCgID755BOqVKnCN9984x7u++Qt5hjvz4APfjDHGNHHuPnAWgoWiy85L5VC305Qv5ZRDC63i8tF5Oppu5Li6VST5+HnxWbSmGe+od+3Og/4lJUAfDlzF6tXryYoKMg9vBOgUaNGREZGkpyczLFjx0ostyu9hjfXUY9SuI5ctGvXjrp165KYmMicOXMYOnQop06dYsSIEYWmBXHxnzsES94RjM2XD7d3R8HFF8CJU/C/r8z1+upX8/7ADTE8+eQTeRYfb9asmdtK+OKLLwDo0lJweS9Iy4D3vjf1RvYT7hFO8YdsTMFi8RXnpVIIDhZcb9afYfJcTWaWZtNu893lBomMjKRSpUrkpKwF4JuZZnhPHqWwzbxXTzeL1+xOqk1WlqZLly7uJT3BxCiaN28O4F7IuyS4Zld7DTK3Lf3KaUII9+I8w4cPJz4+nl69erkf0N6oHCHo3VHkecC7eO4vjrXwHexJ1ExzhpOOHly4vDfddBMA33zzDadPm8mDLmsBoGokDOgKjeqYMmspWCy+47xUCpD7QPpmvgkSu4ZBuiZ9CSG47LLL4JQJNh8/3RDAvTh2drZmTZw51qalrxFb4yQEV6Fag/7cfffdec6VmaVJrPkhtHiXnTt3lljmwiwFrbVPhqN64nIhnT59mvbt2zNz5kyqVKlS4uNd0FZwVR/T07/in5rkVOPqatukcKXQsWNH2rdvz9GjR5kzZw5gUna4YhVXXgRhoZ6WQolFs1gs+ThvlYJsY5RA4lF4c6oTT8iXCeLbb79lwaw3AajdbDAzZ850Wwpb4+FUupmkVbdmMEN6mIfmM6/PZezYsXmOM2MJHMzsDvXuZuEfaSWW2RVT8FQKexJNNtOaVU17fMEll1xCpUqVaNKkCb/88os7hlIanrnTjCLasMt8v3GId6tGCOG2Fr788kt3+Vt/M0Fnl9XgUgo20Gyx+I7zVikIIbjBcSFNmm3eO7cQBepc1LUWoSFwKCWSfgNzF5n73cla0b21eXeN1V+2oaB/2xUcBZi9oeTdeZf7yBXvAFjlWAk92lCo66Yk1K9fny1btrBu3TqfLT7eqXmuy04IuGFw0fVHjx4NmGSCrhTcrRtp/nPDZlo2MOs2uJTCvsOUKoBvsVhyOW+VAsBop7fqWhumU/OCdUJDBO2amM8bd+WW/7HNPIS6t3ISwzmzepeuJw87EjRzFYSFZENOJrtOdGXX/pI9wApzH/li0lphNGrUiKioghPVSsN/7hBEVzEjhxrULlreJk2a0Lt3b9LS0pgxYwYnT55kxIgRdOjQgQ8+MOs5RFUWVIuE9NOQlOxTUS2W85bzWim0byro6OEyKkwpeJav84gR57cU2jY2AdD4QxB/MPeh/+GP5vPlPU7C4clAMK9/60Ol4LIUfDDyqKxp1VCQME3wzX+Kp8BcLqQPPviAQYMG8eOPPwJ5kxC64wrWhWSx+IRSKQUpZZSUcqWUMlVK2cEpu15KuUxKOU9K2dApayelXCylXC6lHOKURUopp0spl0gpH/M45ktO3S+llGGlka84uKyFyErQtGA2agA6NjN11u1wZUvNDTK7lEJQkKBXe/N5+UbzfjpT88nP5vODN1SG/a8D8NEszdGUs1cM+ZVCZpZ2K6dAUApgAvkhIcVTCtdeey0hISEsWbKEVatWuUd0bdy40V3HHVewwWaLxSeU1lJIA64ApgJIKUOBvwMDgPHOC+AF4A5gGPCMU3YXMEsp1QcYIKVsIKXsCtRTSvUFNgGjSinfGbl5qJnMdmlP8ky68sRlKax3Bg5tizcZSRvEQO3qufv0di1qM02zJ1Hz/WITBO7YDPp2DqVRjWNw7FdOpgs+OMvJzdnZ2SQlJeWm4MCMmko/bQLMNav61n3kD9SqVYtLL70UgG7durnTjGzZsoWsrCwAOwLJYvExpVIKSqkspZRnlreWwEal1Gml1FLA8bRTTykVp5RKAY5IKWsBvYA5zvZfgZ75ymYDF5VGvuLQsI5g71TBV+O9P1RdLqZ1O0xAM7/ryMUNg40Lael6aHer5p/vG2vg7uEid67CvlcBeGuaJiur+NbCoUOHyMnJoVatWoSGhgK4rZVurYp9mIDj7bff5q233mLhwoW0bNmShg0bcvr0afd8j9y5CjbQbLH4ghAfH68akOLx3UlZhucTNxmoka+uZ9n+fGV5kFKOBcYCjBs3jqFDhxYQIjMzk4SEhLMSPKWIbVpD9cg6HEsNosUNmRw8FgQE0bLeCRISUt31Kgn47b9B/OfzaH5cUYndiRARphnYIZGEBG16+MenUDUihQNHoln8+yFaNcgqlnyuldxq1arlbtviNdFAFZrXTiEh4WQRe1cMJbkP+QkODubqq68mOTmZ5ORkmjdvTnx8PIsWLSIyMpIqoZWAamzbnUZCwnGfyJ0fX7SjPAk0eYtDoLXJ3+UtalShr5XCMSDa43u2857jUVYNOOpR97hTthujRKLz1cuDUmoCMMH5Wmj3MCEhwWdDKV0M6JrD9MWw84C5ZFUj4eZLo4mNzZv/KDYWfvgf/LJK8/xnmpH9gmjXyqSH6NixI1OmTCGm0l6S0ztwIKU2A2OL5/ZZu3YtYEYFudoWt99c1v6yKrGx1XzQSt9SFvehe/fuLFiwgMTERGJjY+ncWgOapNRKxMaWfIJdUZRFO8qSQJO3OARamwJNXk98rRS2A+2cAPEFwDqnPFFK2RI4CNRQSiVJKVcAFwMTgSHAnUBt4GHgM0z8oeACyBXEZ08K1sRBjWioXR1qRJl0Gd4Y1kMwrEfe7a51kkMyNgEdWLtdc+PQ4imF/Cmzc3K0e0W4ri3PsjEBTPv2JprvCjbbCWwWi28ptVKQUv4EdAFaA+8BrwELgXTgVqfaE8DHzvn+5ZR9CHwhpbwd+FEplQAkSCkPSCkXA3uBl0orn6+IrCzo27l0x3DlP0pPWgGVr2NtXPH3za8Udu43Sebq1YQ6Nc69ILM3OnToAOQqhQZONu/9R0zqkaIUtcViOTOlVgpKqcsKKf46X51NQN98ZanAiEKO92hpZfJXXJbC4d1zoR2s3W4C18WZiZxfKbiCzOeTlQDQtq0Ze7t161YyMzMJDwulTg3NwaNm+c4GtStYQIslwDmvJ6+VNzVq1CA6OpqTRzYSXTmHw8ch8Ujx9k1MTAQ8lYIJp5xvSiEyMpImTZqQmZnJ9u3Gf2aHpVosvsMqhXJECOG2FprXNaOFXHGBM+HdUjj/3CUF4gqOC8kqBYul9FilUM644gq1q5iH/Nkqhbp166K15g9nLYeu5/AcBW/YYLPFUnZYpVDOuCyFyto81f/cfuZJV1rrPJbCgSNmpnTVSO+pOc5l8isFO4HNYvEdVimUMy6lkJP8B1A8S+HYsWOcPn2aqKgoKleunCfI7Kt02YGEN0vBuo8sltJjlUI541IKxxKWERJs8iidTCu6h+uyEmrXNk+/NS7X0XkWZHbRtm1bhBBs27aN06dPu4elJiRVrFwWy7mAVQrljCumsGvnFto2Nik0Nuwqep8CSsE98uj8sxIAKleuTNOmTcnKyiIuLo5a1Uz5kaJylVgslmJhlUI506hRI4KCgti3bx8dm5ksIGeaxLZrl9EauUrBlJ+vlgLkdSHVcNYCOmqVgsVSaqxSKGdCQ0Np06YNWmtqhscDsLaIYLPWmokTJwJw4YUXcvyEZtcBiAiDNo3KRWS/xFMpVIs0ZcdTzaxmi8VScqxSqAD69OkDQPoRsz5AUcHmpUuXsmrVKmrWrMmoUdfywhfmodepOcVerOZcpE2bNgBs376dkBCzLKfWkOx/yWItloDCKoUKoHfv3gDs3WyWl1y/03sP95VXXgHg3nvv5YOfY3hpMoQEw9O3n78KAaB69eoApKQYn1ENJ7eudSFZLKXDKoUKwGUpqOWzaRADJ9Ng4+6C9eLi4pgxYwZhYWGIho/yytQogoLgy/GCS3ue30ohKsoEEk6cOAFYpWCx+AqrFCqApk2bUrduXY4cOULP1skAvD+joKXw2muvobVGXvkBz34RiRCaTx4TXDfo/FYIANHRRgu4lYITbLYjkCyW0mGVQgUghHC7kDrWmAfApJ/hSHKuYjhy5AiTJk2C6pex4vAtALxwewq3XmIVAuRaCtZ9ZLH4FqsUKgiXC2n3pllcciGkZcB735tty5Yt49JLLyUtuB3BHb4lRwvGj4Fbhp6qOIH9jALuI9ew1BMVJZHFcm5glUIF4bIUlixZwsPXm97/m1OzuPb6W+jduzerNxwnqOMssolgzCXwnzusheBJfqVQ01kV1VoK/kdmlibxiB0qHChYpVBBdOnShcqVKxMXF0fHhofp2ExzODmYqQtDCa13HeEXriUnJIahEj78hzgvcxwVRZUqVRBCcOrUKbKzs6kRZa7P0RT78PE3xk/UxF6jWb3Z3ptAwCqFCiI0NJQLL7wQgGXLltKu6mwARItXyWwxmYzsCK7pD9OeE4Sex/MRvCGEyGMtuGIKNtDsfyxYk0NODvy80nudfYc0da7KYfzEnPITzFIoVilUIK64wqeffsr3H46GjAR0UDQhwfDaOMG3zwiiKluF4I3ClIJ1H/kf6+OMi+/X5d6XGZyz2qSD/2Z+eUll8Uap12jOj5QyGPgMiAV2A3cBm4AEp8rzSqlfpZTtgA8cGcYrpeZKKSOBz4EYYKZS6kVfy+dPuOIKM2bMAGBAw28Ja/wg/7pN0LujVQZnIo9SsIFmvyQ9Q3MqywR8/tzp/XHjWldke4LZJyLc/v4rirKwFEYCO5VSAzDKYCSQrJQa4Lx+deq9ANwBDAOeccruAmYppfoAA6SUDcpAPr+hV69eBAWZW1C7dm2mfjCGX14JsgqhmHgqBRto9k/i4jPcn09kRHPwaOFxhT93mPecHNgaXx6SWbxRFkqhGbDW+fwH0BeIlFIulFJ+JaWs4Wyrp5SKU0qlAEeklLWAXsAcZ/uvQM8ykM9viI6OpkePHgC888471KxZs4IlCixcE9hSUlJy3UfWUvArFq7Ynef771sL1tFa86dH/q9NuwvWsZQfPncfAZsxvf9pwBCgGtBbKXVESnkr8DTwV8CzO5wM1HDqpuQry4OUciwwFmDcuHEMHTq0gACZmZkkJCQUKPdH3njjDfbs2UOvXr3OKHMgtQvKXt6QEPPz3b17N61a7wfqcSxFEx+fQJAPuzv2upec+ct3AbkLic9ceIDOjfJaCwlJQRxPreP+vmJdCv3apuap409tKg7+Lm9sbKzXbWWhFGYCA6WU84ANQKJSyhVh+hb4i/PZc5hBNeAocAyIBo47ZbvzH1wpNQGY4Hwt1BZNSEgostH+RGxsrNtaOBOB1C4oe3ld60uEhYXRuFEs0VVySDkpiKxWn+pRvnPB2etecvYcXGc+pG2HSi1Yuz2I2Ni6eer8sVsDmqAg4z7aeySK2Niqeer4U5uKQ6DJ64nP3UdKqRyl1ENKqUGYB/0PUspwZ3M/wGUoJkopW0opo4EaSqkkYAVwsbN9CLDc1/JZzh28zmq2cQW/Ye9Bo5xr6oUAbNobkWd7UlISa52VBAd1M2XWfVSx+FwpSCnrSikXSCnnAunAVmCZlHIR8Ajwb6fqE8DHwC8eZR8CV0gplwILlFL+a39ZKpz8+Y9ssNm/0Fpz9KS5R9cMjoasEySnR3PomFECP//8M7Vr1+bbn0yg4bqBgqCg3BFIZ0NaWhoLFiwgOzvbt404D/G5+0gplQgMyFfcvZB6mzBBaM+yVGCEr2WynJt4y5Rqg83+we7du8kONQMIr760KxNmr4Gq/fh9K1zaE95991201mxNiIBg6NEWmteHuH1mBFLnFsU/1/jx43nllVe44447ePw/HzLudfjnjYKB3fK6EVNPaSpHQFCQHeHnDTt5zRKw2DUV/Bv1xzoIiwWy6XdhM4LTTXxhyZ9pHDlyhNmzZ0NQZU4HNSQ0RNO2MbRvavY9GxdSTk4OU6ZMAeDjjz/m9idW8MsqGPUvTcLhXItjxUaTbqP7XZqk4zblhjesUrAELN6UwpHkipLI4smSVbtABBEVdoLKlUJoUtOMN1n4+wmmTp1KVlYWVOkIIoj6VZMJCxW0b2L23bir+A/t1atXk5CQQHR0NEIIlqxz5cGC0f/JYsPGzWzdc5rhj2tSTsLaOBj6sLZ5srxglYIlYLHps/2b3zcaJVC/RiYAF7QNBmDjnnC++uors63lpQCEZ5m4Qrsm5oFe2EqE+UnL0Ow7pJk+fToAt912G//3v7chqjvoLIKyk1i8LpiOV3yPHHOAw8dNMLtVQ6MYhj2iSU61iiE/VilYAhbPyWsANavaTKn+xNbdZjZzy0ZhAAzq2QCyUzmeFs2i5ZuIiIigW7/bATgab0Ynncl9tG3bNmbPmcfb0zRNr9c0uV7z5Q9xAIwcOZIu/e4FEQInfidnk1mciob/JDWnIQ1qJPPdc4LfXhM0qw9qi3ExWfJilYIlYLGWgv+SkpJC0onKAHRqZZS3lF0hdY2pENmd4cOHczTDjOVP2vMb8fHxtG6I1xFIv//+O50GPc2lTzXmgTc0B49CdjbsyxxCTEwMffr0YdGfZp+brmzItlVv8/frnGOcPsip1UPIyjhKg9qCea8LKoXDXGU7EfmxSsESsNhAs/+ybt06iGgCQNP65jHTvn17gk4qU6H5W1w09G7W73RGAZ38k3nz5hERLmhev2AOpK3b99Hnji1kNPsCIpoScnobz96ebjbWGsmVw68mODiYhWtN0Q2XxtKyZUtevCeISY/DhREPc3T/Hzz88MMANK4r6OBYJRt3leGFCECsUrAELFYp+C9//vknhDcCoIkzgTksLIz2VWfDid+hUnP+Obk/J05BVPhJyDzMb7/9BhR0Ic1bdYxLHqtJetXRBJFB46xXyFrZnoRVjxCevQtCY2jW9U7SMjSrNoMQ0Kej2Tc0RDDm0iA+//A/hIeH8+mnn/LrryYnZ4dmps4GqxTyYJWCJWDJH1NwuY/sQjsVz9q1a92WQhOPrBY9uzeFdf1oEb2CjExjJXRsZjLezJs3D621ewTSmjjNkxMyGfJwFTJDmxGRHcfSt7OZOeFSQkKCeP/998jY9zkA25O7sWIjZGZBlxZQLV+ak5YtW/Lvf5s5sk899RQAHZqaOut3WveRJ1YpWAKWyMhIAFJTU9FaW0vBj1i3fhOEm4lrDWvnlv/973/nptHXMPedBrw6ThBZCcZcHkVMTAwJCQls27bNPQLppcnwwhfBaBFCleMfs+HLKHp2iqRDhw489thj5oBJ3wIwY2kQc383D/d+nQuX6Z577gFg06ZNaK3d7qMNO33c+ADHKgVLwBISEkKlSpXIycnh1KlTedJn5+TY3l9FsvtAJogQ6lTPJjwst9fepk0bvvjiCxo3bsRD1wmSfxaMHR7EoEGDAGMtuNxHAKTFEbZlGN+93IjmTeq7i5988klat24NpzbRsGYKx07A29+Zbf27FD5buXr16lStWpXU1FSOHDmSx32kdeG/l8Qjmje+1ZzOPH9+T1YpWAIaz7hCaIggqrIJUp44VcGCncfk5OSQdKIKAE3rFZ1OwpVuYvDgwYBJV/H+K+Po03wDxD8Pf3Tl87fvon379nn2i4iIYO7cuUyePJm/XOW4EU+abX07eT9fkyZNAJOCo15NqB4Fx07AAS8rhd77qubBtzSf/VJkM84prFKwBDT5k+JZF1LFk5SURE6oCTI3q1+8R8yIESNo164dR44c4f3332XJpM6w+188+fjfue666wrdp0GDBtxwww3cMDhX8bRvCrWqeVdETZsaM2TXrl0IIejoshYKcSEdSdbMcvI0r9lmLQWLJSDwlhTPBpsrjgMHDkCEUQqN656hskNMTAwbNmzgjz/+4PHHH6dDhw7cfvvtPPPMM2fct1VDQZeW5rO3eIILl6Wwa5cZcuSOKxQyAumb+SZwDbBpT3FacW5QFovsWCzlhh2W6n8cOHAAwpsA0KRu8bORCiHo2rUrXbt25YUXXjirc/5jtGDc65oxlxR9PpelsHv3bgA6NBOAdkYg5d33izm51sH5NJfBKgVLQJNfKdS0azVXOImJiW5LoUkxLYXSMnqIYPSQMysgT/cReLcUdu7XLNsAVSqZ74ePw+HjmpgiXFPnCtZ9ZAlozhRTyM7WXkeWWMqGAwcOQKjRBvVqVrAw+fAMNEPuRLmNu/KOWPtijnm/ui8emVvLR8aKxioFS0BTIKbgoRQyszR9x2k6jNFkZVnFUF4cOHAAgo2yjq5SwcLkw1MpmLktgvq1IC0Ddh0wdbTWbtfRLReLEq3xEMhYpWAJaAomxXMypZ7QfPgjLN9o/syuP7yl7Dlw4ACEGO0cVbmChclHVFQUNWvWJD09nYMHDwIUmMS2arNZ/a1uDZNqu70rnfdZrPEQyFilYAlovAWadx+Apz/J/RPv2F/uop23HEhMdFsK/qYUoGBcoWO+SWyPvLIdgGv6ZRISkmspFGeNh3MBnweapZTBwGdALLAbuAsYCfwNSAfGKKXipZTtgA8cGcYrpeZKKSOBz4EYYKZS6kVfy2c5t/AWaP5+CXiGEnZapVBu7E88BvWCCQvJITTE/8ayNGnSBKUUu3btolevXnlGII2fCEvimoPOpknEPOBS2jUx+9mYQskZCexUSg0ANjnf/w4MAMY7L4AXgDuAYYBrMPJdwCylVB9ggJSyQRnIZzmH8BZodimEq/qY9x0J54fpX9ForUk8lApAVGX/HKlTYFiqYwl8twie/xzQWbDlZpL2moV/GtY2Fk9SMhw6du7/jspCKTQD1jqf/wDuBjYqpU4rpZYCTlJb6iml4pRSKcARKWUtoBfgxP35FehZBvJZziG8BZoBRg2Am4aaB5N1H5UPJ06cIO20sQ6iq/i3UnC5j9o2Num2M7MgSOTAlpsg6RuT6RUzf8JlLZwPweaysO02Y3r/04AhTpnnVKJg593zF5MM1ACqedR1leVBSjkWGAswbtw4hg4dWkCAzMxMEhISStwAfyXQ2lUe8mZkmCUfDx8+TEJCAtlpAiHqEBIEf7vqMKlpAohhy55MEhKSSnQOe92Lz44dO9zxhIjQkl/z/PiyTa7sulu3bnUfs1PTmmzcE0q/mA+Yt2gqYFZ627dvH0IImtSuyspNlVm69jgta585sZa//2ZiY2O9bisLpTATGCilnAdsAI4BHv03sp33HI+yasBRj7rHnbLd+Q+ulJoATHC+FmrLJSQkFNnoQCXQ2lUe8rp6fZmZmcTGxhIbCx8/pqkZDX261+X4CQ1o4g+HUr9+fYQ4+96rve7FZ/v27W6lUKNqqM/k8GWbLrjgAgD279/vPub8tzQpJ2Hk5R+76yUlJREcHEy9evXo0V7z9QJNwrGqxMZWL1d5yxufu4+UUjlKqYeUUoMwD/rXgXZSyjApZW9gnVM1UUrZUkoZDdRQSiUBK4CLne1DgOW+ls9ybpE/pgBw26WCK3ubh3+1KEGNaDiVDgePVoiI5xWecxT8ceQRQOPGjQHYu3cv2dmmj1o9SlCjSipr164lJCQEKSUAa9aYNaWLCjannNTc/VIOz36qSTkZ+DEHnysFKWVdKeUCKeVcIF0ptRh4DVgIPOe8AJ4APgZ+Af7tlH0IXCGlXAosUEr5r/1l8QvyxxQKo7mTht/GFcoeM0fBv5VCpUqVqFu3LpmZmezfn/ujWLFiBTk5OXTr1o3evXsDuUrBNas5f0whPUMz4gnNhB/hXx9pmt+gee0bTfrpcmhIGeFz95FSKhEz0siz7Gvg63xlm4C++cpSgRG+lsly7pJ/SGphNI+F1VtgRwL07ui1msUH5LEUKlWwMEXQpEkTEhMT2bVrFw0bNgRg8eLFAPTt25eOHc0PxRVsblDbzM52jUCqXV2Qna256VnN/DVmolvzWFi6Hv7+tubbdjVY9n6FNK3U2MlrloCmOEqhmdtSCHzT3t9JTEyEYBPI9VdLAQoOSwVYsmQJAH369KFr165ArqXgOQJpw07Yk6i5+2XNd4ugaiT88opg8duCmf8VRFeB5ZvC2b4vMH9vVilYAprw8HBCQkI4ffq0eyRSfprXN/EFO4Gt7AmEmAIUHJaamZnJihUrAOjduzdt27YlLCyMHTt2kJycDEA7E4pg2COaJtdpPpoFEWHw4/8JOjUXCCG4vJfgkh6m3i+ryrdNvsIqBUtAI4Q4o7XQ3BkEssNGqMqcvErBP+cpQMFsqX/88QdpaWm0adOGmJgYQkND6dChAwDr1pmxMQO6mvZkZUNMNRjcHWb9V9C3c952XnKh+f7LKmspWCwVwpmCzYESaNZac81TOfS+L4dZywMz5XegWgqeriMXXbp0AXJdSDdfDBs/ExycITj0QxBzXwtiUPeCiu9iM+KVeWvgdGbg3UP/S0xisZwlZ7IU6teC8DA4dAxOnNJ+24Nds82kWgC44p+a3h3hpXuhVwf/lDc/GRkZHD16FFEnGo1/KwWXpbB27VruueceFi0yF95TKbjiCq5gc1paGivnf83evXtJSkoiJSWFe+65h169euU5dmyMoHXD02yND2XpehjYrezb40usUrAEPIXNVfAkKEjQtK5my16TQrtT8/KUrvjMdGbldG0Jew+ZkSyDHtRs+xIa1vF/xeBKRR1eqRbp+Pfoo0aNGlG1alWSk5P54IMPAOOK7Nevn7uOZ7D5xIkTXHbZZW6LwsXixYvZsmULYWFhecoHdMpga3wov6zSDOyW994dOqZ5abJm7XaY9LggNsa/7q11H1kCnuIOSwX/jivMWm5cDf+5Q7BzimBQN0g/nWs9+DsHDphFK0IizIxff7YUwsLCWLp0KZ988gnvvvsur7zyCj/88IPbrQTQqVMnhBBs3LiRYcOGsWTJEmJjY3nqqad4/fXXadOmDbt27XIrFU/6dzKDHmZ7BJuPJGv+8V4OTa/XvDwF5ip493v/cy9ZS8ES8JzVBDY/VQoHj2pWbTajWQZ3h8oRgr9cAfP+0ExbqPnbtf7VmywMl1IIDq0KOf6tFADat29P+/btvW6PioqiRYsWxMXFsXz5cho1asT8+fNp1swswNCkSRNGjBjBs88+y2233ebunAD0aHOaSuHw53ZIPKIJDoY+92u2xZvtsg2oLTB9ETx/V5k286yxloIl4CneXAVXtlT/65kB/LzSvA/sahQCwOW9ICwUlqw3SsPfcSkFHQCB5uLSrZsJCDRp0oSFCxe6FQLA8OHD6dWrF4cPH+bVV1/Ns19EGAzoYj5PX2xiRNvizYI+qycIlr0rqBYJm/fA1r3+dW+tUrAEPGeKKUCu+8hf5yrMXGYeDFdclGsRRFcRDJVmbYjvF1eUZMUnMTERgBxhtMG5oBSefPJJxo0bx8KFC93BaRdCCF580awD9vLLL3Po0KE824f1MPfyr28YK7BJXfjlZYFsIwgNEVxxkak33c/cg1YpWAKeYsUU/HhY6ulMzZzV5vPleQeycE1/82D5blHe3uTJtMJ7l0dTNNk5hW4qc1yWQmZOBHBuKIWOHTvy1ltv0ahRo0K39+vXj8suu4zU1FReeOGFPNuGOZPYsrKhVlUz67lerVylf3Vf8/n7JdZSsFh8SnFiCk3rmfc9iZCV5V9/wsXr4MQpswJY47p5YwfDe0NwMMz7A46dMHMX7n4ph+qXa76Yk7cd0xdpal+leeGrKCoCoxQEGVmhAET68egjX/Lss88C8NVXX5GTk6uRWzcyI90iK8FP/xO0apj33g7rYdxMKzdBwmH/+U1apWAJeIpjKUSEC2JjTK9tz8Hykqx4uEYd5bcSAGpWFQzoYuT+cSk8/YnJyJmZBXe/rNmyx+y7J1Fzx4ua7Gz4dlElsrPL/yFjJq6ZvEdVKpmhwOcDXbt2JTY2lsOHD7N+/Xp3uRCCpe8Idn8juKBtwWtRpZJwWxMzlhTYXGFYpWAJeIqjFMCM/wczFLC0fPij5j0fDSecucy8e8YTPHG5kJ78UPPMJAgKgos6mDUirn9ak3pKc+MzmuNmaWSOnghm6fpCD1WmmGR4/p8h1dcIIRg8eDAAv/32W55tkZUFNat6V44uF9L0xdZSsFh8RnECzQAj+5k/4NSFef+AOTkatUUXu3cdf1Az9iXNfa9qEo+U7s+8fIMmbh9Uj4Ke7QqvM6KPWUN432Hz/d2HBLNfFrSIhXU7oPMdmmUbIDYGbr/M1KkIP3VSUlJApLgoC7wphTNxpeMeXLDGuAf9AasULAFPcZXCVX0gJBjmr4Gk47l/wH99pLlgrObzOcU739fzcj8vXue93pnIydH89U0jx70jICSk8B5lvVqCfp3N5ydugbuvEkRVFnz9tCA0xIyoCgqCL8cL7rzcCV4uplxzJ2VmZpKRkUFQWFXg/FUKixYtIjMzs9j71YgW9O+c6x70B6xSsAQ8zZubvBWrVq3iyJEjXuvViBYM7g7Z2bk+3ORUzZvTzOeVm4r3EJ38W269RX+W/MH7yU9mAlP1KqeoefJDfv/9d68PlC+eMrn6n/tLruLo1lrw5t8E4WHwwl2C/l0EPdtBTNVsdh2A9TtLLFoesrM1f27XZBYRoE9NNb6rSpF1gPNPKcTGxtK6dWtSU1NZterscma73IOev6uKxCoFS8DTuHFjLrnkEtLT05k4cWKRdUcNyOtC+vBHM/IHYHsxZjtvi9f8sS33e0ktheMnNI9PMDIc+/0uHn7oHqSUREdHc8899xSo36C2ydUvRF5r4p6rBMdmCf55kykPDhYM7W5SLJR2bsPS9Zpxr+VQf6Smyx2aK/7pXTG44jmVqsQA559SABgyZAjg3YV0/Phx9u3bV6D8uoHGgp2zmlK7I32BVQqWc4K//vWvALzzzjtkZWV5rTeij/HhzlVmlvDrU3P/hNsL/l8LMHmueb9+EISGGJ/+8RL4gp/+RHP4OLStfxgOT6F+/fq0atWK9PR0PvjggwIToYqiUni+oY4yHYDvSxG8/PRnTZ/7Ne9MN9llg4LMQ+uB1wtP6e1SCmGVawLnp1LwFleIi4vj/vvvJzY2lhYtWrB9+/Y822tVE1zWE3JyYMrZhSTKBKsULOcEw4YNo2XLlsTHxzNjxgyv9WpVyx3iectzmoTD0KaReejtPQQZp70/SLXWTHFM/NsvE/Roa2YbL91wdrLGxWvenm7OObCxkfWmm25i69atDBgwAICVK1ee3UE96N0+g8hKsCbODFUtCe9MN/vdcRn8/qFgydvGTfXBD/DKlIL1Xe6j0IgawPk1+sjFgAEDCAoKYvny5Zw6dYrTp09zxx130Lp1a959911OnTpFRkYGc+fOLbDvLcOMYv98zjloKUgpg6SUn0opFzuv5lLKOCnlAuc11KnXztm+XEo5xCmLlFJOl1IukVI+5mvZLOcuQUFBPPDAAwC89dZbRdZ1uZB+dYam/uNGQeM6pqe2O9H7fn9uhy17zezUwd2gbydTvmht0YokP98uMHGNm4ZCcoLJcdCmTRsAevbsCeBeGrIkRITBpeYwJRr/vnO/ZvUWM9fgrQcF3VoLenUQfPaEuW7/eF/zQ77RTS5LITQAMqSWFdWrV6dbt25kZmayaNEiRo4cySeffEJYWBh33nknDz74IFD4vb2il1nr+Y9tsGl3xSqGsrAUugDhSqm+wDPAOCBZKTXAef3q1HsBuAMY5tQDuAuYpZTqAwyQUjYoA/ks5yhjxowhKiqKhQsX8ueff3qtd3VfM8QToF5NuHEItHByIxUVV5g81/xZrx1oRgr1c5ZhXOQlrvDudE3EEM283/P+yX9eab6P6CPYsmUL4Ful4Do2wIwSDE39dr55H947NzkfwHWDBM/eKdAa/u/LwpVCUGg1wL+X4ixLXC6kBx54gFmzZlGzZk2WLVvGxIkTufnmmwFYvnx5gf0iwgXXDjCfP//l3FMK+wCklAKoBhwGIqWUC6WUX0kpazj16iml4pRSKcARKWUtoBfgGhj4K9CzDOSznKNER0dz++23A0VbC3VqCPp3MZ//eo0gPEzQwul+eIsrxB/UTHb8vaMHmwfeRR2MclFb4FR63j/y3oOaR97VnM6EN6flbjt2wswpCAmGwd21Wym0bt0agAsvvBAwI6mys7OL3fb8uJaEXL7x7NN6fD3P1L9uYMEH+5hLzPvuA3nLXe6joFCTcuR8tBQgVymkpaVRp04dFixY4M602qlTJyIiIti2bVuho+Ruudhc7y9/NcOVK4qyWE8hCcgBNgPhQG/gA6XUESnlrcDTwF8Bz19cMlADo0RS8pXlQUo5FhgLMG7cOIYOHVpAgMzMTBIS/DRxfikItHZVhLyjRo3izTff5Ntvv+Xpp58uMFrHxdM3BTOnTQTX9TlJQgLERFYBolm7NZWEhNz5DnsOBvPW91FMW5xDZragcZ0smtQ4jKtZ7RvXYsPuUGYuTKJ3h9Pu/e55rRppGcax/tNyzcat+6kWqflheQQ5OdXp2S6DbZs3cPLkSWrWrEl6err7WjVs2JD4+Hjmz59P27Ztz/oaZGZmwsn9NKkbw+7EEH5bcYgOTb0H3z3ZeSCYNXG1iaqUQ8cGB8l/+3KyIUjU5eAx2LUngTDnCRIfbxYKyNKmzVkZx0hISDtr2YtqUyD89ps2bUqdOnUIDg5mypQpVK9ePY/cnTp1YtWqVfz0008MGjQo7741IbZWDPGHQpj2WxIXtTud//A+IzY21uu2slAKw4A0pVQbKWU34FWl1A3Otm+BvzifPXM5VgOOAseAaOC4U7Y7/8GVUhOACc7XQtVpQkJCkY0OVAKtXRUhb/369alevTrHjh0jODiYevXqFVovNhb69wDzM4Pu7TSgSUyuQmysmQy3a79m2OOa1DQTFB49BJ69M5SGHm0aLHPYsBs276/FdU6wcPZKzc+rNFUqQasGsCZOsDyuHn+5QrBym/nZX90/guTkZADatWuX5zr17t2bKVOmsGvXLvcwx7PBdd37dMphdyLsTKrNsD7Fc+dM+s1chxH9gmjetPB7V79WjpldHVqf2PquobDB5j2sGgCNYqsTG1ugT1diAum3v337dhITE2nRokWBbf369WPVqlXExcVxyy23FNg+5tIcXvgcJv1ak1FDCg5BLg/KavTRMef9OFBbShnufO8HuMZjJUopW0opo4EaSqkkYAVwsbN9CFDQ+WaxFIEQwu2fd7lmikNh7qMflkJqGnRveZrNnwu++lcQzWPz/kn7dnLiCs4ktozTmgdeN5//fZtg3EhnYtJcTU6Odi+mc2lPCsQTXPgqrtCznTn3io15+06/rNJ8PKvwoaXfzPfuOnLRoLZ5d6XdgNyYQraoApy/7iOAyMhIKlUqfPhVr14m66G3e3vfCEHVSPhpBUycWWYiFklZKIU5QH0p5ULgK+A/wDIp5SLgEeDfTr0ngI+BXzzKPgSukFIuBRYopfzfXrT4HSVRCs3qmfjA7kTcE7QWrzPvoweeKpD22EVfJ/3E0g0w5KEcutyh2Z4AbRvD30bByH5m9bT5a2DWcjPmv1EdaNekHJSCs9Lkik25ZafSNdeM19z5X820hXnrb9mjWbfDjIJxxSQKo6GjFOI9plK4YgrZmIfh+TgktTi47u3KlSvzpNl2ERsjeO/v5rf24FuauPjyjy343H2klMoGbsxX3L2QepuAvvnKUoERvpbJcn5REqUQES5oEKOJPwR7D0Kz+to9W7lHG+PbXbBggXuSmYva1QVdWmrWxsFvv5uyoCB49++CsFBBWChc1lPz/WIY51gQl/U0Fo03pdClSxfCwsLYtGkTx48fp1q1aiW5DHRqboanbos3i8bXrCqYsxpOOq7+ca9rBnUz6T8AJvxo5Lu6L4SFercUGppJy3mUgstSOJcW2CkL6tev744Zbd68udA1okcPEfy4TDN5LtzyvGbJ297zYpUFdvKa5ZzD9ZDdunXrWe3nOSw1bp/p1depAY1iTvO3v/2NgQMH0q9fP9LT0/Ps98MLgm/+I/jlZcGqDwT7pgoGdM39E984xHze66zjcFlP892bUggPD3ePWFm9enWRMmdnZ7N69epCcyaFhgikc+hVm827awW38DA4eBQefsd8f2WK5rVvjLV0x2VFP4Aa1jbb4w/l9mJdSuF0dhhglUJRnMmFBPDOQ4KGtc0CPP+bXF6SGaxSsJxzlMRSgLxxhUXONIde7bIZO/Yu3nzzTQAOHjzIlCl5p/Q2rCO4dqDg4h5mMRXPJRcBrrgodxWysFAY1M08RBMSEggPD6dx48YFZCmOCyklJYUrr7ySHj168H//93+F1nGl416xSXM6U7szcU59xsxQnvQzjHk+h0fedXJBPSro27lopVBYTMHlPkrPtErhTLjubWHzFVxUjxK867iRXEOEywurFCznHE2bNiU0NJQ9e/Zw6tSpYu/Xwgkib0/QLHYCx7/Pe505c+ZQvXp17r//fgDefPPNs0pLXSlccHU/87l/Z7PilsuKadWqlXvkjieeSiEhIYEXX3yRsWPH8t1335GWlsbevXvp06cPP//8MwCTJk0qVKbcYLPJ2X88Fdo3NQv6/Od2s+2zX0zdDx4R3HnFmd0UhcUUXJZCeqbxSFul4J3ixowu6mDedx0o3zToZTEk1WKpUEJDQ2nevDlbtmxh27ZtdOnSpVj7ebqP1m49DYQRv+FLGjVqxJw5c2jcuDHffPMNa9asYenSpfTp06fYMv1jtGDDTs2jo4t2HblwPTjmzJlDo0aN3EHJDz/8kKioKMLCwjhy5AitWrUiJSWFXbt2sXLlSvd+7uM4LuuVm3Mzw17T35Q9fD18t8i4lt57WDB2ePH81l6VggjhdFYQQUFQKbzwfS1m+c7Q0FA2bdpEcnIyVatWLbRe9SiIrgIpJ+FIMtSqVj7yWUvBck5SmmGpi9acZt+RMMhK5sKOVfjhhx9o3bo1ERER3H333QC88cYbZyVPh2aCPz4KYugFxVMKjRo1omHDhmRlZREcHMyoUaN49tln6d69OydOnODIkSMMHDiQFStWcOONZlzHV199VeA4sTGCBjGQnJprEbhWoAsJESx8U7Dra8E9VxU/kFmnupmRffg4pGcYRZOamppn1bWKGF8fKERERNCtWze01ixd6n1lHSEEzeqbz7sOeK3mc6xSsJyTlEQpNHf+gClpxi9et/JO5s+bS61atdx17r33XkJCQpg+fTp79+4tsXxnUgpCCGbMmMEnn3zCgQMH+Pbbb3nqqadQShEXF8e0adOYPXs21atXZ/To0QB88803haYNd1kLGaehWX0zKslFRLigSb2ze4AHBwvqO5fEFVc4ceLEebk+c0m59NJLAfjvf/9bpGuoaV3zvnN/eUhlsErBck5SEqVQpZKgcshx9/cHxnQtMAmpfv36XHvttWRnZ/Pee++VWL4zKQUwbobbbruNmjVr5ilv0aIFI0eOJCzMKK/u3bvTsmVLDh48yPz58wscxxVXADNvwhe9+Ib5gs15lIKNJ5yRBx98kBo1arBo0SJ++eUXr/WaOhPyraVgsZSSwpTCN998wzXXXMN9993H888/z9SpU/P00rTWZJ3IHcbaz8soHNeCPu+++y5//PFHseTJzs5m9uzZbN++naysLOLi4gDyzHkoKUIIt7UweXLB8Ys9PYbCu1xHpcUzrpCdnU1aWhqEnN/J8M6GqlWr8vjjjwPw+OOPFzqRDaCZk0Zk14FyHIGktQ7kV6Hs27fP26aAJtDaVZHyHjt2TAO6UqVKOjs7W6ekpOjo6GiT2Mfj9cUXX7j3UUppWk7U9M3W4YOzdXpGjta6YDtycnL01VdfrQEdHR2tFy9efEZ5Jk6c6D5nkyZNNKAbNmzos/Zu3rzZLc/27dvzbEtLz9ENRmbrDrdm6+zsnCKPk5GRoWfPnq3vvvtufeGFF+pVq1YVWu+Rd7I1fbP185/l6OPHj5trXX+4pm+2Hvxgts/a5eJc/O2fOnVKx8bGakBPnjy50DqzluVo+mbrIQ/5/Jp6fa5aS8FyTlKtWjXq1q1LWloa8fHxfPrpp6SkpNCtWzfeeOMNrr/+eoA8LqAffvgB0k1qrh5tIDys8F61EIIpU6Zw7bXXkpKSwsUXX1ykCwBwr7YVEhLC7t27AejYsWNpm+mmTZs2dO3alZSUlAIupIhwwcbPBMvfEwQFebcUpk6dSu3atbnkkkv44IMPWLlyJc8++2yhdT0nsLmGo0acx+szl4RKlSrx9NNPA/DUU08VOgHRuo8sFh/iWqNg06ZN7vUVHn/8cf76178yceJEoqKiWLp0KRs2mPU0f/zxRzjyAzUj0884GicsLIzJkydzxx13kJaWxpVXXsl3333ntb5rotLKlStZtGgR//3vf3n11Vd90Uw3rlFIhckRXUUQeYaFb15//XWSk5Np27Ytjz32GMHBwfz888+FrhftGVNwTVwLr2yiz1YpFJ/bbruNVq1asWPHDj777LMC25s4SmFPImRnl5MLqSgzIgBehRJopmZxCbR2VbS8d999twb0sGHD3O6azMxM9/b77rtPA3rcuHE6Pj5eA7py5co6LS0tz3GKakd2drZ+8MEHNaCDg4P1559/XqDO/v37NaCjoqJ0VlaW7xqYj4SEBB0UFKRDQ0P14cOHz2rf1NRUHRoaqoOCgvSxY8e01lpfccUVGtBvvPFGgfqrNhm3Rufbs/WqVavM9b3wv5q+2fq+V6z76Gzk/fjjjzWgBw0aVOj2eiOMq273gaJdf2eJdR9Zzj9cwWaXa+f+++8nJCR3vqZrzsFnn33G119/DcDFF19MREREsc8RFBTEq6++yvjx48nOzubWW2/l/fffz1PHZSX06NGj0NnLvqJ+/foMGzaMzMxMvvzyy7Pad/ny5WRmZtK1a1d3Ar5bb70VgE8//bRAfc9As8t9FBJ+/q7PXBpGjBhBSEgICxcuJCkpqcB211yF8hqWapWC5ZzFc7hnpUqVuOuuu/Js79SpE7169SIlJYV//9tkbx8+fPhZn0cIwTPPPMP//vc/tNbce++9eSYluZSCKxFaWeJajvSTTz45q/1ccYgBAwa4y6688kqqVavGH3/84XaxuahdHUJD4GgKHDlmUokEhZmZuefr+swlpXr16gwePJjs7GxmzJhRYHt5xxWsUrCcs3gqhZtvvpkaNQquBHbPPfcAcPLkSYQQXH755SU+36OPPspDDz0EmFxELspTKQwfPpxq1arx559/smbNmmLvt2DBAgAGDhzoLouIiHAH5D///PM89YOCzExpgD2JZjhlkB2SWmKuueYaAKZNm1ZgW65SKJ+YglUKlnOWRo0aERVlJlQ98MADhda59tprqV7duD0uvPBCateuXapz/uUvZrXZqVOnkpGRwenTp1FKARTIS1QWhIeHc/XVVwO51kJOTg5ffvmlewRUfk6ePMmqVasICgoqkM/J5UL64osvyM7OzrPNpRTc2VKtUigxI0aMICgoiLlz53L8+PE821xzFaz7yGIpJUFBQUybNo2pU6d6Hf5ZqVIl94P82muvLfU527VrR+fOnTl+/DizZ8/mzz//JCMjg9atWxdqqZQFrt79l19+yeHDhxk5ciQ333wzl112WZ5F5F0sXbqUrKwsunXrViA5W69evWjRogX79+/nt99+y7PNFVc4cNTEabRdirPExMTE0L9/fzIzM80oOA+s+8hi8SFDhw51m+beeP7555k9ezZ/+9vffHJOzwR15ek6ctGhQwe6dOnC0aNHadOmjdtPnZmZ6V4XwpPCXEcuhBDuBeanTp2aZ5tLKRxKNilRc6xSKBXeXEgupbDTKgWLpXwIDQ1l2LBhPhsZdMMNNwBmMtyvv/4KlK9SgNyA89GjR2nfvj1ffPEFAO+//z4pKSl56hYWZPbkggvMgs2uSXcuXBPYjpwwo7Wy7PrMpeLqq69GCMHs2bPdI7oAYmuZoP7Bo2aN7bLG5+spSCmDgE+AZk7RbYAE/gakA2OUUvFSynbAB44M45VSc6WUkcDnQAwwUyn1oq/ls1jKmkaNGtGnTx+WLFnCzJkzgfJXCrfeeivTp0+nWbNmvPHGG0RGRvLBBx+wePFiJk6cyN///nfATDxbvXo1wcHBXteHqFfPdFUPHMjbVXXFFI6nmbiNXZ+5dNSvX5+LLrqIpUuX8tNPP7ndgMHBgsZ1NNsTjAupfdOylaMsLIUuQLhSqi/wDDAO+DswABjvvABeAO4Ahjn1AO4CZiml+gADpJQNykA+i6XMcbmQAKKiomjXrl25nr9atWrMnz+fjz76iMjISMCMjgIzc9mVUmHp0qVkZ2fTvXt3oqOjCz1W/fpmoPz+/XkjnS73UVJafahzBydPO5aCVQolZuTIkQDuFfVclGdcoSyUwj4AKaUAqgGHgY1KqdNKqaWAK+JXTykVp5RKAY5IKWsBvYA5zvZfgbIfrmGxlAHXXnute6LchRdeWKaT1orL5ZdfTps2bYiPj+frr78mNTXVnRLDm+sIoFatWoSEhHD06FEyMjLc5c3qmzWn03JqQKsPSUmzlkJpcbnq8qd8L8/FdspiOc4kIAfYDIQDNwGe4/xc/w7PGS7JQA2MEknJV5YHKeVYYCzAuHHjGDp0aAEBMjMzCx1lEegEWrsCTV5vlLQd/fr1Y968ebRv375cr0NR8t555508+uijjB07ljFjxrhTNnfs2LFIGWNiYjhw4ABr1qyhYcOG7vJp/wrlviemEX8kiibtLkW2DSftRDIJqeXXJn+kpPK6rLotW7awb98+99oXNapUAaJZt+0ECQknijhC8YiNjfW6rSyUwjAgTSnVRkrZDfgncNJju2uws2cC8WrAUeAYEA0cd8p25z+4UmoCMMH5WmjUJSEhochGByqB1q5Ak9cbJW3HW2+9xf/+9z8ef/xx6tSpUwaSFU5R8j7wwAO8+eabxMfHExwcjJSSYcOGMXr06CKtmYYNG3LgwAG01nmOHRsL/06fSPzWP/j2i9VIKYFIXzcp4H5LJZW3fv36REVFkZycTEREhHvVvy6tTeb1wyciiY0t3M3nK8pCKYB5uIN5uNcCGkspw4ALgHXOtkQpZUvgIFBDKZUkpVwBXAxMBIYAd5aRfBZLmdOuXbs8M5v9gfDwcJYvX87u3bvp2rUrlSsXz9fjCjbnjytAbu4j10RBS8kRQtCqVSt+//13tm3b5lYK5Zn/qCyUwhzgFinlQoz76O9AQ2AhZvTRrU69J4CPHRn+5ZR9CHwhpbwd+FEpFTj2osUSIMTGxp51L9ZbsBlyU2e7XB+W0uGpFC666CIAmju3a3uCSaEdHFx2+aV8rhSUUtnAjYVs+jpfvU1A33xlqcAIX8tksVhKh7dhqWAtBV/jWqLVtWQrQPUoQWyMJuEw7NgPrRp627v02MlrFovljHizFHJyctyWQpUqVcpdrnMRl1LYtm1bnvJOzsyv9TvK9vxWKVgsljPizVI4dcqkza5cubJfDLs9F/CqFJqb93U7y3ZWs1UKFovljHizFKzryPe0bNkSMO4j15BhgE7NTRxhnbUULBZLRePNUrBKwfdUrVqV2rVrk5aWlmeug9tSsErBYrFUNDExMQQHB3PkyJE8s5rtyKOyoTAXUutGJjHezv1w4lTZuZCsUrBYLGckKCjIbS0kJia6y62lUDYUphRCQwTtmpjPG3aW3bmtUrBYLMWisAlsVimUDWcKNq+3SsFisVQ0hQWbrfuobPCmFDo2cwWbrfvIYrFUMIUFm62lUDaccVhqGQabrVKwWCzFojBLwSqFsqF58+YIIdi1a5d77QvIncC2bidoXTbWglUKFoulWBRmKVj3UdkQERFB48aNyc7OZteuXe7yujWhVlVIToX4Q2VzbqsULBZLsbCWQvnimsTm6UISQpS5C8kqBYvFUixsTKF8qagRSFYpWCyWYmFHH5Uv3pVC2Y5AskrBYrEUi8JmNVtLoexwKYWtW7fmKbfuI4vF4hcEBQVRt25dIHdWs1UKZUfbtm0B2LRpU57ydk0gKAi2xkN6hu+tBasULBZLsck/q9m6j8qORo0aERkZyaFDhzh0KHeoUaVwQb/OcLGEYyd8f96yWqPZYrGcg7jiCq5gs7UUyg4hBB06dGDFihVs3LiR2rVru7fNf6Ps+vPWUrBYLMUmf7DZKoWypUOHDgBs2LCh3M5plYLFYik2+YelWvdR2dK+fXsANm7cWG7n9Ln7SErZA/if87Ue8BNwBeBaLeJ5pdSvUsp2wAeODOOVUnOllJHA50AMMFMp9aKv5bNYLCXH01LQWltLoYypCEvB50pBKbUKGAAgpZwIfA/0VUoNyFf1BeAO4CAwG5gL3AXMUkpNlFLOllJ+oZTa52sZLRZLyXBZCjNnzuTGG28kJyeHiIgIQkJseLIs8FQKWmuEEGV+zjK7k1LKEKAnMBaIlFIuxFgL45RSR4F6Sqk4p+4RKWUtoBfwiHOIX539p+Y77ljnmIwbN46hQ4cWOHdmZmaeZezOFQKtXYEmrzcCrR1lKW/dunWpUqUKSUlJTJkyBYBq1aqV+fU5X++B1ppq1apx/PhxlFJuS620xMbGet1Wlup9ELBQKZUjpeytlDoipbwVeBr4K+Cp8pKBGkA1ICVfWR6UUhOACc7XQgfpJiQkFNnoQCXQ2hVo8noj0NpRlvLGxsaSkJDAhg0b2LJlC3FxcQwcOLDMr8/5fA86derEokWLOHr0KBdccIFPjlkUZakUrgW+BFBKHXHKvgX+4nzO8ahbDTgKHAOigeNO2e4ylM9isZSAqlWr0rt3b3r37l3RopwXdOjQgUWLFrFhwwaGDRtW5ucrk9FHjuuoF7BIShkmpQx3NvUDtjufE6WULaWU0UANpVQSsAK42Nk+BFheFvJZLBZLoFDeweayshQGAosc11F14Ccp5UkgAxNcBngC+NiR4V9O2YfAF1LK24EflVKB40S0WCyWMqC8h6WWiVJQSv2KCRSjlDoIdC+kziagb76yVGBEWchksVgsgYinUsjJySEoKIj58+dTs2ZNOnXq5PPz2clrFovF4sfUrFmTevXqcerUKXbv3k1iYiLXX389PXr04Pfff/f5+axSsFgsFj/HFVdYv349Y8aM4fDhw/Tu3ZuuXbv6/FxWKVgsFouf43Ih/fOf/2TOnDnUrFmTzz77jKAg3z/CrVKwWCwWP8dlKbgW3Pnoo4/KbN6GVQoWi8Xi57iUAsC9997LVVddVWbnsglLLBaLxc/p1KkTDRo0oG7durz88stlei6rFCwWi8XPqVSpEjt37gQgNDS0TM9llYLFYrEEAGWtDFzYmILFYrFY3FilYLFYLBY3VilYLBaLxY1VChaLxWJxY5WCxWKxWNxYpWCxWCwWN1YpWCwWi8WN0LrQZY4tFovFch5iLQWLxWKxuLFKwWKxWCxurFKwWCwWixurFCwWi8XixioFi8VisbixSsFisVgsbqxSsFgsFoubgFcKUkpR0TJY7H2oKOx1r3jOtXsQkEpBStlWSvmSlDJaKXXOzL6TUjb3+Oz3P7Rz5T7Y617x2HvgPwSUUpBSBksp/wV8DsxVSqVUtEy+QEoppJRPAHFSyn87xX77xzhX7oO97hWPvQf+R0ApBaAuUAl4BwiWUt4spWxXwTL5gmBgFdAZGCKlrK+UypFS+uv9OVfug73uFY+9B36Gv154N1LKYVLKfwAopRKAxcA1wFigNvBfKWX3ChSxREgpL5FSfi6lvA+IUErNVUqtB34G/uNU8xuz9Fy5D/a6Vzz2Hvg3fq0UpJSXYX4k/aWUNzvFy4AnlVIjlFKvAr8Cg5z6fmt2eiKlrATcDnwJ1AGed8mulHoBaCul7K6U0lLKkAoUFTh37oO97hWPvQf+j18rBeAPYDDwEHCllLKqUuo4sMHj4i8FWgMEUMCnJZCmlJoNPAdEA5d4tGk85s9yH9CpgmT05Fy5D/a6Vzz2Hvg5fqUUXBfZ5U9USiUqpU4Cu4BNwH1OVQ2ESClvAd4HlgaChvboEa0D6kopr1RKZQLTgWs9flAhQD+gA7ClQoT1INDvQ6Bddw95A/q6u/CULRDuQT55z4l7cDb4xXoKUsr+wL3AamCyUmq/lDJMKXXao05rTC/ib0qpI1LKFpgb9IVS6o8KEfwMSCn7AtcCk4FNSqlkKWVlpdQpKeUgjAk62Kk7C3hZKTVfSjkC2K2UWusH8gbcfXDacQMwBVivlDoeANc9v7wBd909kVL2BKo5FoHrQRumlMrw03tQmLxBSqlsjzoBdQ9KSoUpBSmlcPyGnYD/Aa9jzMUuSqkbnTp1gcpKqZ3O938CDwCzlVJ/qRDBz4BHu3oA/wa+AeoBzZRSY5069YFU4CUgDvgUY0q/rJSK80N56wGVAuE+SCmbAK8CszDtqKeUut/Z5jfXvZjyBsx190RKeRfwAua39LlSaoXHNn+8B0XJG3DPoNJSke6jGOe9GuaHPxv4BFgnpQyXUl4EKKAbgJTyAoxv7x0/vxmudkUBx5VSnyqlXgQGSCmHSCm7AL8DPYBngCzMHyOxgv4UxZF3NYFzH+phengfKaWewwQuB0opO+Nf191FUfIG0nX3ZA7QB1gASCllJObDxcBK/O8eFCXvCgLzHpSYcrcUpJQDgH8Be4FvlVKzpJQTgUTgQeA7jL/uXoyyOOLsFwucdII8fke+dk0FNgN3Ar8ppX6TUn5BbrsilFJJHvtGKKXSA0Fef7sPUso7gauBe5VS8U7ZBOBrpx3DgbuA6zE9voq+7iWS19+uuyeFtckp7wUMANY5//NaGAV4yKOOX9yD4sjrz/fAl1SEUngbYyrvAa4D9iilPpFSPg+sUEr9KKX8GfhKKfW5lDLUCUj5NR7t2gsMB7IxoxKeBY4C64DqwPtKqa1SymAgp6JGK5RA3hClVFZFyOoNKWU08AFQA/gC+BrTjmuArsC/lVKnpZSfAO8qpVZX5HUvobx+d909KaxNrliIlDIcEyuJBj4CTiulsqSUoUCWv9yDYsibJ75zrlNu7iOPyPweTA9oE+amDHH8qtWBjk6d/Ri/I/6uEApp10bgW8wMzXjMj+yfwNsYM3krgFIqu4L+FCWV168eTE4sJEUpNRq4GzNOvIUTGFyC+W3fI824+EyMJVSh172E8vrVdffEW5tc25VSGcBPQE1gGvCks0+mP92DYsh73igEKGOlIKWs4ry7gplBwGGgmpSyplJqO+YP0RsTbOogpZwHHANmlKVspaGY7VoMDHNMz86YIJarR1KuQ9cCTV5vFNIO19DN3cBGYIQ0o4z2YwYudAa+B5KAk1be0lOMNl3l8sk7dAWuwsRHni9vZRBo8voDZTJjUEp5BXATsF9K+QpwAECZnCarMRf9QoxWPg6EKDMM9REgwxVH8DfOsl3J5CrdJOAvzg+x3Ca4BJq83iiiHZ5yfQ48DbR0XETbMb3BKkqpZCtv6TjLNjV3Hr7bgd3AFcqkh7DyBgA+VwqOv3AMxv3QGvND/wlYKaVsiRl5sAsYLKVsCAwF5gE4PSa/pITtmg+glFpk5S0ZZ2hHK2AgZsDCQSnlXuAHYCHwmDIZLMtbIQSUvMWhhG1aBPxDKbXNyhtY+MR9JM0Q0mrO11bAAaXUQmAS0Ai4VEo5GDMMLVUpNRl4F2gAzFJKvesLOXyND9r1jpX37DmLdvwbOKGUOur44nsD/6eUurU8OxiBJm9x8EGbblFKHbDyBh6lVgpSyhsw07+fB3ACl7FSyuudAM1hTKrZvcBtSqlvnXo7gH8ppT4prQxlQaC1K9Dk9cZZtuN2pdRXTr00YIRS6n0rb+kItDYFmrz+TqmUgpQyAqiMGVft8uMBvIbxlf6MWTAjFqihzBT3YNf+Fe2r9kagtSvQ5PVGCdpxWppFT1zBw3IdqRNo8haHQGtToMkbCJz1PAUpZWPMkMVZwJ9KqX1O+eXAaGCMcvKFSJNRMFlK+RzwidMr9UsCrV2BJq83Aq0dgSZvcQi0NgWavIHGWVkKUsowjE8uHpML3dMHPQ84BNzh1BVACynlHCDZn29GoLUr0OT1RqC1I9DkLQ6B1qZAkzcQKZalIKW8BrPC0BzgI6XUAKf8I2CzUupl5wY0xvj1VmKi/QlAsFIqtWzELx2B1q5Ak9cbgdaOQJO3OARamwJN3kCmSKUgTe6PiUAacBATzBkE/KJMaopGmElOVyqlDkuT8XQ6JqBzt/LT4V2B1q5Ak9cbgdaOQJO3OARamwJN3nOBM7mPcoCJykwLfw5oh5n+3UFK2VIptRf4BRgmzdJ53YGHlVID/fxmBFq7Ak1ebwRaOwJN3uIQaG0KNHkDnjNNXjuGMddQSiVJKWsDKZg86A8D92ASS/3pRPH9YlhjMQi0dgWavN4ItHYEmrzFIdDaFGjyBjzFjSkIoComc+llTtkEzNjfMMxwsBPKT4Y2FpdAa1egyeuNQGtHoMlbHAKtTYEmbyBzNmkuQoAlUsruwCWY1LLblFLHykSy8iPQ2hVo8noj0NoRaPIWh0BrU6DJG5AUe56ClHIY8CNm2NeXSqnPy1Kw8iLQ2hVo8noj0NoRaPIWh0BrU6DJG6icjaVwBHgCeFOdW/nFA61dgSavNwKtHYEmb3EItDYFmrwBydkohd+VUqrMJKk4Aq1dgSavNwKtHYEmb3EItDYFmrwBSbkvx2mxWCwW/6XcluO0WCwWi/9jlYLFYrFY3FilYLFYLBY3ZbJGs8VyriGlrAz8A9itlJokpbwNM3v2UaXUyxUqnMXiQ6ylYLEUj8qYlM23Od8XYnL3/1hRAlksZYG1FCyW4uEaCtlfSqmBPZg0zY8CW6WUu4FawAfAX4DfgPcxs25DMMtAznbWA3gBo1CqAL8C9ymlDpdjWywWr1hLwWIpHk8475sxD/TCXEZVgAhgOXA1MAF4CbMOwItOnccxidx+BF4HLgXeKyuhLZazxSoFi6V4zHHeDymlpgCFLdqSAzyESe0M8LlS6k1gP9DUKXOtIXw3xh1VBbi4TCS2WEqAdR9ZLMWjOLM805RZGD7T+Z7svGcDwc5nAWRhlEO2U2Y7Zxa/wf4YLZbikYKxBFpIKW/CxBNKwo+YztgYoBEm2+fdPpHQYvEBVilYLMVAKZWJiQ9UA74gt5d/tvyfc5y+wNuYmMJCH4hosfgEm/vIYrFYLG6spWCxWCwWN1YpWCwWi8WNVQoWi8VicWOVgsVisVjcWKVgsVgsFjdWKVgsFovFjVUKFovFYnFjlYLFYrFY3Pw/QtRxjwGKzUcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plotting the actual and predicted price values\n",
    "actual[41:].plot(label=\"actual\")\n",
    "preds.plot(label=\"predicted\")\n",
    "plt.title(\"N-BEATS (window of 7 days), MAPE : {:.2f}%)\".format(mape(actual,preds)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "ec60dfce",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MAPE: 3.14338319003582\n",
      "rmse: 364.63235\n",
      "mae: 274.1947\n"
     ]
    }
   ],
   "source": [
    "#metrics of time series forecast\n",
    "print(\"MAPE:\",mape(actual,preds))\n",
    "print(\"rmse:\", rmse(actual,preds))\n",
    "print(\"mae:\",mae(actual,preds))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e3408478",
   "metadata": {},
   "source": [
    "The 7-day window has a MAPE of 3.14% which is much better compared to the 30-day window."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cf812e6f",
   "metadata": {},
   "source": [
    "# GridSearch"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cd2c98e6",
   "metadata": {},
   "source": [
    "Below we find the optimal TCN model with grid search"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "id": "11e2c939",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d2ff443146434d2682603bd3263cb518",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "98343795139c478e8f67e09e9d7a2843",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3a6ff3f8636b4cdeb6d9b369b539c808",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "675c1c5ab85f4c229bca19b304f6a2d4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "eee63e9d6d064363a9559b4fa9ef4649",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "44a98cd80b3048f8880e172ad3dcfb29",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a4365132bc30470db898c4f40bbaceff",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "96a8fa49fad2439daf09ed029f1e9ca6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4faf6ad218dd461f83eb5518579f618c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4690d99d81fe49c08a142feb461dc294",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e21ce3de56a34b078fbd05cd79fd04d7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0e104d3876254d868b710298641d6525",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2222fe230b8b4efdb390cce1974cbb4b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ceb27afc1a8b44a0af87f1827b8e0f7d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e77c93beeb384722ae7e923220d46215",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1b1a07c3c19c4da2aef76d4a3f80ddca",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5530c43a26f74875941d55abe6b28212",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0d05ba5de32a4efaa06e9c48acffc659",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e299868e31cb44d18a2ceb8127a6df31",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "61505ce55bea4adbad165e26a59143c0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1ebd6e659fb84391b5e43a800d2ee4d5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "de53768e77994f42b956d62ef975e662",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "585eb3538af849fb84d2221f6d94f769",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "260f5189d9cd4f0da2140b3e6ca3250e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4d98062dcc444c4dae2c4067d671f5db",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2d8e3fe3dae94ce7ae5982116e9eb31b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "63a7eb2508894ddea371695fb4d36b8c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7a79568daa5145d4848bfd746893bee6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d5731657d748449dab1e05ef781be892",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bf45b3a6d5124a71aa15cd066e57300e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "00a10ff2df4e400a924c09c4120ec67d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0bf76534fc8547c79c6a9c3708eac5d8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3202867556284fe39a587e4e940203bd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b407894ea573404794db82ab52191f68",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5d5fedd3a7e04d2ea1cd6bf19b8dd800",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "51ea738c5d614dad960067dc322c1206",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d064b7ee98f24164a6e07a574f240c81",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e8e23e01db834facba27bc41b560ef53",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "75948dd0375d4c56a21bc3f649d9074b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d7165b1c93e14ae1ab706f1a82070cbf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f8ccab1ec2d04365b78c6edf86287ea3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "57fb9941654d44cf82cf451e11c373c3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a0e3b75481914ea8994aa3a7320f888c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3c5fa4ef2e5e4d29920ae84d81022dcd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a78afdd4dfa2428ca3805a94c8e30210",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3c584199b8c44af393f8854e1a44deee",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "73f5ea13027442f4817b19154c8d5c0f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5016b13948b2485bacd489e177f2857a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5cb36485a8524f068ccc2b324af7da2b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d99dc47633584529896d6334019f9dd6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bb234b4da0794931ac02dbd3cabfc409",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e3698cb15e724a9086c5f4fa45f2f2f9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "15281ab971384914aa5a58d45c1db94b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "706ebb0a15a9434ca135abfa9f160386",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f9484b7c7e9343d18560fca5742affc9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e5cb1bc51d164a68bef67baa36a226a7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "09dadd1ec0794e1c8262e1fcf85187d0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ffc9f52e08864b1cb07af15043d9214d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "47c45f49b99e4b3f826d1c1a9af3db40",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ca2828c7ddba47509731abab97da5930",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4bea1ff672574899b1be59d48307a100",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9740fcca0be2443a92a31e1d8484b9b1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5ec5e4f19f7e4c6092576b93acd202cb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "79effdf1b33e4ab1aa5cd3f027256951",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "92f090f0262348c0b467da77cbc8aa2d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1cea056ec8884d43a9a0852df6a0502e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e99085c165784e3b9ef56d820c1a38a3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c06074c86ddb4e9b92263fef3344156a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0d7f5febb3ef4e3eb26b41052433650e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c0cff8010f3544ab92c478a100e06bca",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0ffd9ef974c74811996f4abc891166d3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8b6c05ea74c6422fb9300d86b75ec546",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2c5ad5e336df4bd898dc903c04ca1407",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "01eefa5809854705bee461259bcd43d9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fa9c52cb9f7840099f994cffbd0ca13f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b539a735e57444eda3ca62a031e6e06f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8725ea7aadda48cdb8251acda05a3fff",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0964011d7cdb4813967318c12ae36315",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "991b2aef4e5a40d7ae951421b008ef95",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "23afeae4a63946a6b9641ea5749e26cd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "514541c26df84bec9dd6bee7b140543e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0f20149ff02e4cc49755f32f60fd9331",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7bc3c646fdb54497a2bcfb5be68ca2a9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e771b3f9bc224dd3a1b19c68920dc64d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f7f9f7e213bd45409f1622122ccf164d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1d1e296c02f54674bd3d0b82ff9d72fe",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "df72127fa15e4e9fbff6c4560862f106",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e1facd2cee9b437f8b0791648a89a069",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e04a7e3fa4ee4c9f967d312daae32558",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d3353f9938c74fa6b8d7367a0787d4c0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d247023618df43a5a3fcb29a6b1eee29",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5a3d2af0772e4c6db1e8b7cda8534323",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "48e88458a3764a0ebf1233c9e82646a0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c6fcd68ecff646888d2f424c3f1e4740",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dfeb19b58488496f9f24d5af22802e62",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b5b47a68e5714d63acef20ac9f0581bc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dc5982917af24e5c9ff6f3b59e61b20c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "126f8809957a48818487c2f1ec12f831",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b421f0bbaa9944d1a6ac938d1baacddf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ca5bb1fc199d4a85b2d04983af3d1462",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "492d8966d1b645789fdf7a1f00d12e5e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d25c4adc5f374c25b35735b6e9e25c26",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3a168a09d62c4f058b36f11ec93a721f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "89e1f46a01874f0c95d74b42c9cdce0d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a8d9b75ca214483f81b9491733d93c3d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7b723e6d4be04727b02ae1df20a02e18",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0fdfb0ec65384d58a322f9ccb2c14227",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "048810ddf6894b859c05d0b011d0f8f1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0ce8fc3c4f7d47ef84d89f66291d1200",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "17ceb863f92d4c8fa765e62d7ac292f1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7afc6b79a4e945e29f84c7abb6f31ddc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2b060fe4841947f582a2c8f197802063",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f9bc891f6e1e433b8850b4724a007f47",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4b624a0986e24a1593abe3deb8c9ec58",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "205b5cca27444b90a88afc30b0ce3bb2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c35f0edb12f449399debcebc0f6f83a4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6cd3dd41581d47a4b426edf9c40c150e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e1aa5c99840947e09c18a258a26c141e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fa652f38372c4b3a99710a2a8f08273c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a4f552fcfa574a569104790389c19ad3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "907bcfc4004e4888a62ca8479599d473",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6e6b9bb0a1214c4a9b371adecf1ef450",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8577c80fca7e49a99a2a589a771471f9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5fb41739c447411d9e121f020686c114",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f9fd92f7042c443e9e813f5f264b97d6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "454283b80c6f4edab2d9e4afabe57af9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1155ae5e5a994a8eb79837ff72677407",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b9a913c7ce56433e8f20c69399b04c43",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4c8babccb963457f8398924a4dc490ce",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9c0317cbe4e847fb84c38e856bf40dc4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5c6dbd44adaf4129b5daebab52e91163",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a7f643941ec74f76b6112b95685afc3c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e9d4073fb6434565af16783b57b2de6b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a9886b2b35904c92a6d67b5a210b62b3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2c55bf904f284ea7ac766e94d55fd096",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "eeebbe6511c9414d83d9082f6d53cace",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c9ac346edc3349e1bfb2b7625e460841",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d894cb051381467792a4456b5410a623",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "81245524a47249afa46d80a9cfd6409f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "43826901aa364b3bb856a01d0a610da8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d357997b39f34e50b9304586d897e81a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "47a6d23ac58a484f9cf572851c53738e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f909f0c80c1e40ab8e054a960d3d6005",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2a7d155d4602458982cf7514a86fe54d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "43aa513743c54f37bb8869f372853509",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4cbc5ae0a16e43f39c966415a13da2ba",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "60030d5ce658459eb3a1e528d4646184",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6076bab88212455185a4050f386fef3a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "239d2f8dbff247e3aa9e0e21db962082",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "943b756ab6444a91bb77db8d96b10242",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d1f61a4cf95d447e8435666b891c4eb3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "48256fef974b44abb571908522cd2e96",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4a4d7f587189418a9ca348e0988be994",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "956e9a526006455f9963936c5917c70b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5050a0cefc474fc097037e2ea0165974",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a73f3e050cf744d99257dbc0ccbe367e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "53207cc78ebe46d3a826d75f62237b94",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "01b837f196334a23b285d5d1e3ac3b2e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ffa099b7c2794c4d99d93419155125b0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "972182db4a544bf3a990f1a1d1669a7e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "68dd2c20850c4fbeb818da6aa9e438d7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9eb6933daa4d4465a4aa2527a2c02db5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e90a221321e6480da8395cf5fbab1e20",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "630d3592d03c48b2a86a0932d0f40335",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7f34cf57122348469f3207df7adf2155",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2850764ca60b4f24917085824f42ab57",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7a03f83562744e8dbdf37f72140bb54e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6bc2279ebdf84e17b14c33d18feb12de",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6139847bd4774818bd3e9e81e76d6a66",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e4383e3ea2db4e1f8b8df50fc59291a8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "221a43d8d39c48b1a575a82909e30ca2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "79ba4d7caa3645258d615be8305df5b2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e655a5c65e6e4f3dbc0340935c161485",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8c51771785a241509f5710740f4d87ce",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c0efed14343c4ecd92c885b73186d2eb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4a3b7789217d451cacaaa83b3ad8cc96",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4f8b5945c7ef41719d19687a9beac6e4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "706aed67a70743548a736622a6fc821f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cee72fe350d841e69dda53e4b6b2d427",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4b9f17282661441f84f7dbe9dd1dbc0d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "72b6edf3125d4068b19c4f131dafeee6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bb5ca21d40c54323bbf561a994fcd3e8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cb784602afa242a2a9206f800aedaf5d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "df6d462ddf3c4148b9d89161c68eae7d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "eaaf881e08f34939803bb5b2c79cab16",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bc1bb411d15d444cbb7f161499ae82d7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8678bd3ea1134cd6821fcd7429dff0f4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "54090ff551824368be4474b56d8864c7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "558673a7ad0c46beb4f714ca6d06bea1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "85b89e40d4814f3ba68182c9556f1961",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d57a54cd33284d5bad9dbfe802cab648",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a48ced4a17844000934e1ea49caf5fb0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "88d08273902b45518b610747b9a46591",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "669888668c294b8aa8efd61bc35e0389",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5c0dc314e04243ae8101f2f282db716b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a4c76828a88e4f1986d95f7213fc2d0f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0721929bf7884a2285372f513bb6dc55",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "96f4c92ef98642f4bd099e4b678fa0fc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6ffa273fd8624856913dfca8aaff3a67",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d821d4a70ecf4c849cf9b41a05b255c1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "33503859edba4dbe8b94c3aeac62de44",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "95c5e8a1e4834af5a68704989f97a899",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "df34b34d15d8463b8a0ffe51db4019d4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c81d272b506d47a7adf5eff123d7ffbe",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b0918b86a4cf42e7b572df597e4fb0aa",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4fd26104b3dd445287805f86d513166e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "04fd1631a79b44a29ad9a79633e6d7dd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1b4dd21da2ef4b548dc66dfff91af690",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cb3ad7bddc6d400bb23bca2786ef2b66",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a27c1155d11049dd9349d1e964ae7f6f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8fc7c20b3a8b44b09402abb158e1aaf9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7948db4a61024103923bcbe3b8304492",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "32eca241bcd343d48868a119a60575b2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "66ed942131c847798fe0d8ce1d136052",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "136f942ed6854133a61404b30aa3e299",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e26fd28bfaac4d84a402744861e8ef2d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 17it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4bf0a10306e74eba82841ac442b71cec",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ebb4a1d443aa48d6862dcdc23a237253",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b9674ee509f04b7fb9c371639e495704",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1531db9eb89c4c4c876aee4f2b5f7e15",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e2d6cd19dc034bc7b37b4f873b1910a3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a03c6f658be94585a7c07cf717c31918",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "40f009013dd34016bc3bfabff3a0b377",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a20e801dc5ff4574aac3e05dd603b7f7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d957e2ee35744b1bb748735ae78c7bda",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "539e32685ab94ce8bfc0bcb37cc720b6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c3e01e48ddb24bffaf3e6efcbfb4f52f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "24b719b6492a4fb3bc43fdf5599e73f0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "251b027b012c40f8830053eaaebb4db1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c5b892983a204cfaa97f666192484079",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ab5388fdba0f48c994b130dc22e8f14c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f5b2a0a2e6714a51a00644daff928def",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "85ac15378ce349a0930e194330825ae6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "de6e2b56f8d14e5691b8bf081f1defd5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2976d415ed01440b9f1fdddaac51841c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "806beabd30f04b7e97245e777b55b110",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3f1ba00401b54f5cbe044fc8594b2da3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "231f2e933fcf49ecb2375f5a47d0a08b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "79d9b6b6e736423096d1bae34dc778d8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3c532cf1e4a04ac68ceae7018f5571f5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5a24934317aa4c358027fc299f807f09",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0c0b398cb4144e33b06806c75c83c7aa",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "79aa3d38f2b04fdb9739acdcd8057ba3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "178884bbd35a47b48fbfe86ebab40a49",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "52bbe5b1a0314937a873ce919fd7c62a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4b2e0610225b4d199f0393342d666aec",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c0f9d15788d545b8b41483692cf657a5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bcdd2ecabd8b46c4a8e445f826703be6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4f7247e5dc6843f6b3eea4fff36148fe",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a7d9414c6b3b490885a0e8dcb1c9e951",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "37aa65296c874503ae53ea2047d95de5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "aa8e149e45924aaa957379ded470f503",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "61e8cff5af264458b3ae9944c730d101",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8c15d11e74e0484f932144b48644d996",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ba96ba7cf7b742c090cd66c92b2b98cb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "36bda49ab3e24f0d9e2d1fbdfa7a8aef",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "61cdb17690e14a65bcb7b337cd7c9a59",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fe9505d60c074259b242f3ba602c8551",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c4b7d56ac0cb4e90b579b39d6cc4dc97",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8c0977c8a4ff419dbb082e8872d5ffe1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "95a35199fb274b4ea0111f8106eccd65",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "50dd007d446a401cb6beb239a42b7ef2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "16023fd8ee844417a9e28007633bbbaf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "902ca5c9b0da4d5aa48adf073ef38a1a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "72c46f89a4884c80ac7aaafcb3e4af96",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "317175ea923f4a3eaee2b6a208f98aa1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2ba9960c96ef42d9ad673de68ed7648d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4b80d0e88e5641c8b86657cf4f2b2d21",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "93699154c153458683f52a88ed5c9ec3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f961c4abed0640fd8c6d0402713fcd9a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7039836619d840f496bc5ab7163e3206",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d3947834d317485dae387f19683052aa",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "310406b6fd574172a83ca1918bf271bc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e65efbed41794d859a18c30a075dca25",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "669cf492b32e421bbe32fcd7b68403c3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9652f37671a24656b69efd70d347fd68",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "773be2776c6e4ad8a62a7832fb02dab6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "eeb1a5f3b4cb420bb16e58c55d9c4ecc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8df545782486450bba8a37f1f1aaaa7b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9d1e18d9a0794002a0e38cd720e2e782",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ca537d4657084081b8ae86abf2715665",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "04e9fcf981c44c14b3cfbb1b3e7989d5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bafb6b9965d643eb86f72d5bd71d056a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b13a8116f7f24a86803e9f82d93f2020",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "26730de168ae4a5aa7c08b09096b1549",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f20794f7a52a43f2a56ba234c5956b25",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6298edd8590046cfbccaf476be0cf1d2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "445d3669b52b4ddbbab3789510eec68c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ef271eb85b564d7c9934f26f7e3d38be",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1fdbb0a12a074315886a9b2a5b258681",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "530b2915129d45fa959a5ab1a2da1466",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1c648b2af3d241548afdda7e1fe4e713",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b4a46296ad824036bc67ccfc64fe967c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "199f66cbd9e84643b6a60267d6d72eaa",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "74ef95686b4c424f99d45b6cc32513d9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ab3e6c88ec58404884112694f7d1451c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1a745e1c8e944f57a06b8f74cfa18b68",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7281d858f7ad44019f5ad9a320f406de",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5467cfb227384aee8830aed64bfe7a04",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6c19931b030c4879a148030b9bcf2bdd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2fa8d1cf2e664a73bf9a6e64a1282b34",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "86137db2dd494467b52f93daf8b89f17",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f7c9c5d7fea442eb9dfe4f993897a5eb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "27a601bc81c745e58fa4c1fa7a2fe29e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8e6dfe0f8c894e1b87770bc01cf9b627",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9a6dd79cf0744753b22628dea48e17be",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "61140466b4e94d998fb8889d954f34b7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "858620a7416f49cbbef7261d92a42f82",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "eb6bb174945144209325fe3e821e99de",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e0a440d047cc43a597dae2a8a768adec",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3b5952a91d8e4157aefa831b9f21c656",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "12de56a9d40d4bf5bc55b4bf20719aca",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "66184f4e3c524a748688a87226e77eae",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b0432b19bb9547da8180600688266756",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "de1e799b60234912a86d59eef050b7f0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "021272e31acc41e1bb9c919b5a608e91",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ca44cc2167ba424d9fcaa642ddfdccd7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "03704e82b98f4b479d7c1f2670b83ebb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c55578e8d6a54dfba702f047d2eea696",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7f9d58fc3a0f4aae8622a5aece00f708",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8ec653b61c7249b7bac69be32a3afa59",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "09aecda12945473f88385352017febfa",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b23c41f707d44218be378866e96015ac",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d944a5712e424c04b90e7c531754e6a4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 16it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "parameters = {\n",
    "    \"input_chunk_length\" : [7,30,60],\n",
    "    \"output_chunk_length\" : [1], \n",
    "    \"dilation_base\" : [2,3,4],\n",
    "    \"kernel_size\" : [3,4,5],\n",
    "    \"num_filters\" : [2,3,4],\n",
    "    \"dropout\" : [0.1,0.2],\n",
    "    \"torch_device_str\" : [\"cuda\"]\n",
    "    \n",
    "}\n",
    "\n",
    "best_model = model_TCN.gridsearch(parameters=parameters,series=ts_ttrain,val_series=ts_ttest,past_covariates=covF_t,metric=mape,reduction=np.mean)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "id": "ec9ddb52",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(<darts.models.forecasting.tcn_model.TCNModel at 0x171d467bcc8>,\n",
       " {'input_chunk_length': 30,\n",
       "  'output_chunk_length': 1,\n",
       "  'dilation_base': 3,\n",
       "  'kernel_size': 5,\n",
       "  'num_filters': 4,\n",
       "  'dropout': 0.1,\n",
       "  'torch_device_str': 'cuda'},\n",
       " 14.766673743724823)"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e0218d78",
   "metadata": {},
   "source": [
    "The best model has a window of 30 days , dilation base of 3 , kernel size of 5 , 4 filters and a dropout of 0.1. Lets next train a TCN model with these paramters and evalutate its metrics."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4977f198",
   "metadata": {},
   "source": [
    "Grid search for the NBEATS model is performed below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "6d5364ab",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4c94d086f9024e39baa354c45ab9091e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2c69e5ca95264c4093da47634324f74b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Predicting: 18it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "parameters = {\n",
    "    \"input_chunk_length\" : [7,30,60],\n",
    "    \"output_chunk_length\" : [1,7], \n",
    "    \"num_stacks\" : [16,32,64],\n",
    "    \"num_blocks\" : [1,2,3],\n",
    "    \"num_layers\" : [4],\n",
    "    \"layer_widths\" : [16,32,64,128,256],\n",
    "    \"torch_device_str\" : [\"cuda\"]    \n",
    "}\n",
    "\n",
    "best_model = NBEATSModel.gridsearch(parameters=parameters,series=ts_ttrain,val_series=ts_ttest,past_covariates=covF_t,metric=mape,reduction=np.mean)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "3b073dca",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(<darts.models.forecasting.nbeats.NBEATSModel at 0x17187c48a48>,\n",
       " {'input_chunk_length': 7,\n",
       "  'output_chunk_length': 1,\n",
       "  'num_stacks': 32,\n",
       "  'num_blocks': 1,\n",
       "  'num_layers': 4,\n",
       "  'layer_widths': 16,\n",
       "  'torch_device_str': 'cuda'},\n",
       " 26.609021425247192)"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dc1b3bcb",
   "metadata": {},
   "source": [
    "the best NBEATS-model has the above parameters. This model has already been trained and hence we will not train it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "id": "d03d0cb3",
   "metadata": {},
   "outputs": [],
   "source": [
    "#retraining the TCN model with optimal parameters\n",
    "\n",
    "model_TCN_grid_search = TCNModel(\n",
    "        input_chunk_length=30,\n",
    "        output_chunk_length=1, \n",
    "        n_epochs=300,\n",
    "        dropout=0.1,\n",
    "        dilation_base=3,\n",
    "        weight_norm=True,\n",
    "        batch_size=16,\n",
    "        kernel_size=5,\n",
    "        num_filters=4,\n",
    "        optimizer_kwargs={'lr': LEARN},\n",
    "        log_tensorboard=True,\n",
    "        force_reset=True,\n",
    "        random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "id": "1f558327",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5db2ad57446b4952a9da15f222dc4729",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<darts.models.forecasting.tcn_model.TCNModel at 0x171e199dc88>"
      ]
     },
     "execution_count": 108,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# training: load a saved model or (re)train\n",
    "# training\n",
    "model_TCN_grid_search.fit(  ts_ttrain, \n",
    "            past_covariates=covF_ttrain, \n",
    "            verbose=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "id": "c2d9d5ea",
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture\n",
    "#Testing our model using historical_forecasts on ts_ttest \n",
    "preds = model_TCN_grid_search.historical_forecasts(\n",
    "    ts_ttest, start=0.3, forecast_horizon=1,past_covariates=covF_ttest,retrain=False\n",
    ")\n",
    "\n",
    "# scale back:\n",
    "preds = scaler.inverse_transform(preds)\n",
    "actual = scaler.inverse_transform(ts_ttest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "id": "ad79a1a8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'N-BEATS (window of 7 days), MAPE : 5.41%)')"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plotting the actual and predicted price values\n",
    "actual.plot(label=\"actual\")\n",
    "preds.plot(label=\"predicted\")\n",
    "plt.title(\"N-BEATS (window of 7 days), MAPE : {:.2f}%)\".format(mape(actual,preds)))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7723300b",
   "metadata": {},
   "source": [
    "### Issues / Improvements\n",
    "1. A more rigerous grid search for time series forecasting of these models"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "efca3fe6",
   "metadata": {},
   "source": [
    "#Credits\n",
    "\n",
    "1. https://unit8co.github.io/darts/README.html\n",
    "2. https://towardsdatascience.com/n-beats-unleashed-deep-forecasting-using-neural-basis-expansion-analysis-in-python-343dd6307010"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "affc5e06",
   "metadata": {},
   "source": [
    "In summary the state-of-the-art models do a good job of forecasting the price of bitcoin."
   ]
  }
 ],
 "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}