preprocess.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Extracting LSTM features and Feature Maps from the main data\n",
    "To better understand the code, check out `./prep_files/preprocessing_synced_for_MTL.ipynb`. That notebook explains the thought process behind this code on a sample dataset whilst this goes straight to the point."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import src.preprocess\n",
    "from src.preprocess import LSTM_featureExtractor, Prep_data_for_CNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.read_csv('./data/concat_users.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "User   Position\n",
       "User1  Bag         304434\n",
       "       Hand        304438\n",
       "       Hips        304434\n",
       "       Torso       304430\n",
       "User2  Bag         372171\n",
       "       Hand        372170\n",
       "       Hips        372172\n",
       "       Torso       372177\n",
       "User3  Bag         347627\n",
       "       Hand        347627\n",
       "       Hips        347627\n",
       "       Torso       347626\n",
       "Name: Segment, dtype: int64"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.groupby(['User', 'Position'])['Segment'].count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "sub_segmented_data = src.preprocess.create_sub_segments(data, window_size=112, overlap=0.5)\n",
    "sub_segmented_data.to_csv('./data/sub_segmented_data.csv', index=False, header=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "del data, sub_segmented_data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-----"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "subbed_data = pd.read_csv('./data/sub_segmented_data.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 8092672 entries, 0 to 8092671\n",
      "Data columns (total 18 columns):\n",
      " #   Column          Dtype  \n",
      "---  ------          -----  \n",
      " 0   time_ms         float64\n",
      " 1   acceleration_x  float64\n",
      " 2   acceleration_y  float64\n",
      " 3   acceleration_z  float64\n",
      " 4   gyroscope_x     float64\n",
      " 5   gyroscope_y     float64\n",
      " 6   gyroscope_z     float64\n",
      " 7   orientation_w   float64\n",
      " 8   orientation_x   float64\n",
      " 9   orientation_y   float64\n",
      " 10  orientation_z   float64\n",
      " 11  Date            int64  \n",
      " 12  Position        object \n",
      " 13  Coarse_label    float64\n",
      " 14  Fine_label      float64\n",
      " 15  Segment         int64  \n",
      " 16  User            object \n",
      " 17  Sub_Segment     int64  \n",
      "dtypes: float64(13), int64(3), object(2)\n",
      "memory usage: 1.1+ GB\n"
     ]
    }
   ],
   "source": [
    "subbed_data.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Feature Extraction for LSTM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "lstm_featExtract = LSTM_featureExtractor(subbed_data, num_features=6, window=112)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "lstm_features, lstm_labels = lstm_featExtract.feature_extractor(start=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.save('./data/lstm_features_labels/full_lstm_features.npy', lstm_features)\n",
    "np.save('./data/lstm_features_labels/full_lstm_labels.npy', lstm_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(72256, 112, 6)\n",
      "(72256,)\n"
     ]
    }
   ],
   "source": [
    "print(lstm_features.shape)  # Should be (num_samples, window, num_features)\n",
    "print(lstm_labels.shape)    # Should be (num_samples,)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "del lstm_features, lstm_labels"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Feature Extraction for CNN-RNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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></th>\n",
       "      <th>Sub_Segment</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>User</th>\n",
       "      <th>Segment</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">User1</th>\n",
       "      <th>5</th>\n",
       "      <td>35728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>33040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>18928</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>11088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">User2</th>\n",
       "      <th>64</th>\n",
       "      <td>67088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66</th>\n",
       "      <td>14672</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>70</th>\n",
       "      <td>26880</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>73</th>\n",
       "      <td>27440</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75</th>\n",
       "      <td>17584</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>User3</th>\n",
       "      <th>22</th>\n",
       "      <td>50288</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Sub_Segment\n",
       "User  Segment             \n",
       "User1 5              35728\n",
       "      7              33040\n",
       "      9              18928\n",
       "      12             11088\n",
       "User2 64             67088\n",
       "      66             14672\n",
       "      70             26880\n",
       "      73             27440\n",
       "      75             17584\n",
       "User3 22             50288"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "subbed_data.query(\"Coarse_label == 5.0 & Position=='Hand'\").groupby(['User','Segment'])[['Sub_Segment']].count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This data has 3 drivers whilst my sample had 2."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "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>Segment</th>\n",
       "      <th>Sub_Segment</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>User</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>User1</th>\n",
       "      <td>4</td>\n",
       "      <td>882</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>User2</th>\n",
       "      <td>5</td>\n",
       "      <td>1372</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>User3</th>\n",
       "      <td>1</td>\n",
       "      <td>449</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Segment  Sub_Segment\n",
       "User                       \n",
       "User1        4          882\n",
       "User2        5         1372\n",
       "User3        1          449"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "subbed_data.query(\"Coarse_label == 5.0\").groupby('User')[['Segment','Sub_Segment']].nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "windowed_df = Prep_data_for_CNN().get_window_df_from_dir('./data/sub_segmented_data.csv',\n",
    "                                                         prep_for_FMAPextract=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "time it took: `304m 0.5s`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "windowed_df.to_csv('./data/windowed_df.csv', index=False, header=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I should have set `prep_for_FMAPextract=True` because I need the full windowed_df to be able to split it into train and test without leaking data into the SimpleCNN. I'll do it manually."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "36304"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(windowed_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "windowed_df['user'] = windowed_df['user_act'].values\n",
    "del windowed_df['user_act']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "windowed_df = Prep_data_for_CNN().prep_for_FMAPextract(subbed_data, windowed_df)\n",
    "\n",
    "# change other labels to `Not_driving`  (a dummy label)\n",
    "windowed_df['user_act'] = windowed_df['user'].values\n",
    "idx_users = windowed_df.query('coarse_label == 5.0').index\n",
    "df_with_drivers = windowed_df.index.isin(idx_users)\n",
    "windowed_df.loc[~df_with_drivers, 'user'] = 'Not_driving'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "windowed_df.to_csv('./data/full_windowed_df.csv', index=False, header=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "72256"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(windowed_df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "----\n",
    "\n",
    "This `full_windowed_df` is made up of duplicates and interpolated data. As a result, I'll need to take 80:20 of the duplicates (original and corr dupe) as well as 80:20 of interpolated data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def converter(instr):\n",
    "    return np.fromstring(instr[1:-1],sep=' ')\n",
    "\n",
    "windowed_df = pd.read_csv(\n",
    "    './data/full_windowed_df.csv',\n",
    "    converters={\n",
    "        'long_acc_window': converter,\n",
    "        'tranv_acc_window': converter,\n",
    "        'ang_vel_window': converter\n",
    "    }\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['user', 'coarse_label', 'segment', 'position', 'window_index',\n",
       "       'long_acc_window', 'tranv_acc_window', 'ang_vel_window', 'Interp_Flag',\n",
       "       'Dupl_Flag', 'Train_Flag', 'user_act'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "windowed_df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "idx_non_dupe = windowed_df.query(\"Dupl_Flag==0 & Interp_Flag==0\").index\n",
    "idx_interp = windowed_df.query(\"Dupl_Flag==0 & Interp_Flag==1\").index"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For each non_dupe, it's corresponding duplicate is the next idx"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Get training and test index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "tr_idx_nondupe, test_idx_nondupe = train_test_split(idx_non_dupe,\n",
    "                                                    test_size=0.2,\n",
    "                                                    stratify=windowed_df.loc[idx_non_dupe, 'user_act'].values,\n",
    "                                                    random_state=42)\n",
    "\n",
    "tr_idx_interp, test_idx_interp = train_test_split(idx_interp,\n",
    "                                                  test_size=0.2,\n",
    "                                                  stratify=windowed_df.loc[idx_interp, 'user_act'].values,\n",
    "                                                  random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# get the idx for the duplicates\n",
    "tr_idx_dupe = tr_idx_nondupe+1\n",
    "test_idx_dupe = test_idx_nondupe+1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_idx = np.concatenate([tr_idx_nondupe, tr_idx_dupe, tr_idx_interp], axis=None)\n",
    "test_idx = np.concatenate([test_idx_nondupe, test_idx_dupe, test_idx_interp], axis=None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "TR_tr_idx_nondupe, VAL_tr_idx_nondupe = train_test_split(tr_idx_nondupe,\n",
    "                                                    test_size=0.2,\n",
    "                                                    stratify=windowed_df.loc[tr_idx_nondupe, 'user_act'].values,\n",
    "                                                    random_state=42)\n",
    "\n",
    "TR_tr_idx_interp, VAL_tr_idx_interp = train_test_split(tr_idx_interp,\n",
    "                                                  test_size=0.2,\n",
    "                                                  stratify=windowed_df.loc[tr_idx_interp, 'user_act'].values,\n",
    "                                                  random_state=42)\n",
    "\n",
    "# get the idx for the duplicates\n",
    "TR_tr_idx_dupe = TR_tr_idx_nondupe+1\n",
    "VAL_tr_idx_dupe = VAL_tr_idx_nondupe+1\n",
    "\n",
    "# get new training and validation set\n",
    "TR_train_idx = np.concatenate([TR_tr_idx_nondupe, TR_tr_idx_dupe, TR_tr_idx_interp], axis=None)\n",
    "VAL_train_idx = np.concatenate([VAL_tr_idx_nondupe, VAL_tr_idx_dupe, VAL_tr_idx_interp], axis=None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# ## subsetting only the training data\n",
    "# windowed_df = windowed_df.loc[train_idx, :]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Begin hyperparam tuning for SimpleCNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_tr, y_tr, uniq = Prep_data_for_CNN().prep_input_for_CNN(windowed_df.loc[TR_train_idx, :])#.sample(1000))\n",
    "X_val, y_val, _ = Prep_data_for_CNN().prep_input_for_CNN(windowed_df.loc[VAL_train_idx, :])#.sample(500))\n",
    "X_test, y_test, _ = Prep_data_for_CNN().prep_input_for_CNN(windowed_df.loc[test_idx, :])#.sample(500))\n",
    "\n",
    "# normalizing X\n",
    "mean = X_tr.mean(axis=(0, 2), keepdims=True)\n",
    "std = X_tr.std(axis=(0, 2), keepdims=True)\n",
    "\n",
    "X_tr = (X_tr - mean) / std\n",
    "X_val = (X_val - mean) / std\n",
    "X_test = (X_test - mean) / std"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array(['Not_driving', 'User1', 'User2', 'User3'], dtype='<U11'),\n",
       " array([0, 1, 2, 3], dtype=int64))"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "uniq, np.unique(y_tr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.utils.class_weight import compute_class_weight\n",
    "\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "from torch.utils.data import DataLoader"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(46240, 3, 224)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_tr.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "class_weights = torch.tensor(compute_class_weight('balanced', classes=np.unique(y_tr), y=y_tr), dtype=torch.float32)\n",
    "class_weights = class_weights.to(torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([ 0.2936,  5.0925,  3.3057, 10.5571], device='cuda:0')"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "class_weights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from src.dataset import SignalsDataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create datasets for each feature\n",
    "feature_indices = {'long_acc': 0, 'tranv_acc': 1, 'ang_vel': 2}\n",
    "\n",
    "## dataset for extracting feature maps from data\n",
    "train_datasets = {feature: SignalsDataset(X_tr, y_tr, idx) for feature, idx in feature_indices.items()}\n",
    "valid_datasets = {feature: SignalsDataset(X_val, y_val, idx) for feature, idx in feature_indices.items()}\n",
    "test_datasets = {feature: SignalsDataset(X_test, y_test, idx) for feature, idx in feature_indices.items()}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<link rel=\"stylesheet\" type=\"text/css\" href=\"./custom.css\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from src.engine import simpleCNN_engine\n",
    "from src.model_simpleCNN import SimpleCNN\n",
    "from src.dataset import SignalsDataset\n",
    "from src.preprocess import Prep_data_for_CNN\n",
    "from src.hyperparam import RayTuning\n",
    "from ray import tune, train\n",
    "\n",
    "from IPython.display import display, HTML\n",
    "\n",
    "# Load custom CSS file\n",
    "css = HTML('<link rel=\"stylesheet\" type=\"text/css\" href=\"./custom.css\">')  # Ensure the path is correct\n",
    "display(css)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SimpleCNN(\n",
       "  (conv1): Conv1d(1, 224, kernel_size=(3,), stride=(1,), padding=(1,))\n",
       "  (batch_norm1): BatchNorm1d(224, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "  (conv2): Conv2d(224, 224, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "  (batch_norm2): BatchNorm2d(224, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "  (maxpool): MaxPool2d(kernel_size=(1, 2), stride=(1, 2), padding=0, dilation=1, ceil_mode=False)\n",
       "  (flatten): Flatten(start_dim=1, end_dim=-1)\n",
       "  (fc1): Linear(in_features=25088, out_features=4, bias=True)\n",
       "  (dropout): Dropout(p=0.7, inplace=False)\n",
       ")"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "SimpleCNN()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Hyperparam Tuning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Hyperparam Tuning - Long Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Enhanced Hyperparameter search space\n",
    "config = {\n",
    "    \"optimizer\": tune.choice([\"adam\", \"adamw\"]),\n",
    "    \"lr\": tune.loguniform(1e-4, 1e-2),\n",
    "    \"scheduler\": tune.choice([\"exp\", \"ReduceLROnPlateau\"]),\n",
    "    \"gamma\": tune.uniform(0.1, 0.9),  # for exponential scheduler\n",
    "    \"patience\": tune.choice([5, 10, 15]),  # for ReduceLROnPlateau\n",
    "    \"epochs\": tune.choice([30]),\n",
    "    \"batch_size\": tune.choice([32, 64])\n",
    "}\n",
    "\n",
    "criterion = nn.CrossEntropyLoss(weight=class_weights)\n",
    "save_dir = 'SimpleCNN_long'\n",
    "model = SimpleCNN\n",
    "modelType = 'SimpleCNN'\n",
    "engine = simpleCNN_engine"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "mytuner = RayTuning(config, save_dir, criterion, model, modelType, engine)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div class=\"tuneStatus\">\n",
       "  <div style=\"display: flex;flex-direction: row\">\n",
       "    <div style=\"display: flex;flex-direction: column;\">\n",
       "      <h3>Tune Status</h3>\n",
       "      <table>\n",
       "<tbody>\n",
       "<tr><td>Current time:</td><td>2024-07-26 04:06:51</td></tr>\n",
       "<tr><td>Running for: </td><td>03:30:47.64        </td></tr>\n",
       "<tr><td>Memory:      </td><td>10.4/13.9 GiB      </td></tr>\n",
       "</tbody>\n",
       "</table>\n",
       "    </div>\n",
       "    <div class=\"vDivider\"></div>\n",
       "    <div class=\"systemInfo\">\n",
       "      <h3>System Info</h3>\n",
       "      Using AsyncHyperBand: num_stopped=25<br>Bracket: Iter 8.000: 84.1823056300268 | Iter 4.000: 84.1823056300268 | Iter 2.000: 84.1823056300268 | Iter 1.000: 84.1823056300268<br>Logical resource usage: 2.0/16 CPUs, 1.0/1 GPUs (0.0/1.0 accelerator_type:G)\n",
       "    </div>\n",
       "    \n",
       "  </div>\n",
       "  <div class=\"hDivider\"></div>\n",
       "  <div class=\"trialStatus\">\n",
       "    <h3>Trial Status</h3>\n",
       "    <table>\n",
       "<thead>\n",
       "<tr><th>Trial name             </th><th>status    </th><th>loc            </th><th style=\"text-align: right;\">  batch_size</th><th style=\"text-align: right;\">  epochs</th><th style=\"text-align: right;\">   gamma</th><th style=\"text-align: right;\">         lr</th><th>optimizer  </th><th style=\"text-align: right;\">  patience</th><th>scheduler        </th><th style=\"text-align: right;\">  iter</th><th style=\"text-align: right;\">  total time (s)</th><th style=\"text-align: right;\">   loss</th><th style=\"text-align: right;\">  accuracy</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr><td>train_model_a8304_00000</td><td>TERMINATED</td><td>127.0.0.1:13880</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.818537</td><td style=\"text-align: right;\">0.00648272 </td><td>adam       </td><td style=\"text-align: right;\">        15</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        838.642 </td><td style=\"text-align: right;\">1.38369</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_a8304_00001</td><td>TERMINATED</td><td>127.0.0.1:17696</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.542466</td><td style=\"text-align: right;\">0.00379149 </td><td>adam       </td><td style=\"text-align: right;\">         5</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        705.346 </td><td style=\"text-align: right;\">1.38889</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_a8304_00002</td><td>TERMINATED</td><td>127.0.0.1:16352</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.261052</td><td style=\"text-align: right;\">0.000270069</td><td>adam       </td><td style=\"text-align: right;\">        10</td><td>exp              </td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         69.5704</td><td style=\"text-align: right;\">1.3605 </td><td style=\"text-align: right;\">  30.1219 </td></tr>\n",
       "<tr><td>train_model_a8304_00003</td><td>TERMINATED</td><td>127.0.0.1:19116</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.434606</td><td style=\"text-align: right;\">0.00312242 </td><td>adamw      </td><td style=\"text-align: right;\">         5</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        712.762 </td><td style=\"text-align: right;\">1.38683</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_a8304_00004</td><td>TERMINATED</td><td>127.0.0.1:2828 </td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.675692</td><td style=\"text-align: right;\">0.000770331</td><td>adamw      </td><td style=\"text-align: right;\">        15</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        830.25  </td><td style=\"text-align: right;\">1.38352</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_a8304_00005</td><td>TERMINATED</td><td>127.0.0.1:5864 </td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.727   </td><td style=\"text-align: right;\">0.000830383</td><td>adam       </td><td style=\"text-align: right;\">         5</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        808.91  </td><td style=\"text-align: right;\">1.38624</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_a8304_00006</td><td>TERMINATED</td><td>127.0.0.1:12568</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.492348</td><td style=\"text-align: right;\">0.000880474</td><td>adamw      </td><td style=\"text-align: right;\">         5</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        710.236 </td><td style=\"text-align: right;\">1.38808</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_a8304_00007</td><td>TERMINATED</td><td>127.0.0.1:24364</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.661274</td><td style=\"text-align: right;\">0.00713931 </td><td>adam       </td><td style=\"text-align: right;\">         5</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        701.398 </td><td style=\"text-align: right;\">1.38901</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_a8304_00008</td><td>TERMINATED</td><td>127.0.0.1:26784</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.860435</td><td style=\"text-align: right;\">0.000264009</td><td>adam       </td><td style=\"text-align: right;\">         5</td><td>exp              </td><td style=\"text-align: right;\">     2</td><td style=\"text-align: right;\">        162.43  </td><td style=\"text-align: right;\">1.35454</td><td style=\"text-align: right;\">  38.6491 </td></tr>\n",
       "<tr><td>train_model_a8304_00009</td><td>TERMINATED</td><td>127.0.0.1:5272 </td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.270687</td><td style=\"text-align: right;\">0.00174542 </td><td>adam       </td><td style=\"text-align: right;\">         5</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        700.999 </td><td style=\"text-align: right;\">1.38812</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_a8304_00010</td><td>TERMINATED</td><td>127.0.0.1:15884</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.729498</td><td style=\"text-align: right;\">0.00203688 </td><td>adamw      </td><td style=\"text-align: right;\">        15</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        712.606 </td><td style=\"text-align: right;\">1.38763</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_a8304_00011</td><td>TERMINATED</td><td>127.0.0.1:18976</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.50977 </td><td style=\"text-align: right;\">0.00296031 </td><td>adam       </td><td style=\"text-align: right;\">         5</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        703.559 </td><td style=\"text-align: right;\">1.38881</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_a8304_00012</td><td>TERMINATED</td><td>127.0.0.1:8412 </td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.12455 </td><td style=\"text-align: right;\">0.00364408 </td><td>adamw      </td><td style=\"text-align: right;\">        15</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        828.198 </td><td style=\"text-align: right;\">1.38584</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_a8304_00013</td><td>TERMINATED</td><td>127.0.0.1:6852 </td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.877347</td><td style=\"text-align: right;\">0.000118718</td><td>adamw      </td><td style=\"text-align: right;\">        10</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         82.1694</td><td style=\"text-align: right;\">1.37272</td><td style=\"text-align: right;\">  36.4957 </td></tr>\n",
       "<tr><td>train_model_a8304_00014</td><td>TERMINATED</td><td>127.0.0.1:20196</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.330572</td><td style=\"text-align: right;\">0.00013025 </td><td>adamw      </td><td style=\"text-align: right;\">         5</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         71.7683</td><td style=\"text-align: right;\">1.31715</td><td style=\"text-align: right;\">  45.9483 </td></tr>\n",
       "<tr><td>train_model_a8304_00015</td><td>TERMINATED</td><td>127.0.0.1:29672</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.667359</td><td style=\"text-align: right;\">0.000106861</td><td>adamw      </td><td style=\"text-align: right;\">         5</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         83.0429</td><td style=\"text-align: right;\">1.30506</td><td style=\"text-align: right;\">  50.774  </td></tr>\n",
       "<tr><td>train_model_a8304_00016</td><td>TERMINATED</td><td>127.0.0.1:18844</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.778246</td><td style=\"text-align: right;\">0.000525593</td><td>adamw      </td><td style=\"text-align: right;\">        15</td><td>exp              </td><td style=\"text-align: right;\">     4</td><td style=\"text-align: right;\">        284.905 </td><td style=\"text-align: right;\">1.36218</td><td style=\"text-align: right;\">  37.006  </td></tr>\n",
       "<tr><td>train_model_a8304_00017</td><td>TERMINATED</td><td>127.0.0.1:31140</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.58823 </td><td style=\"text-align: right;\">0.00540978 </td><td>adamw      </td><td style=\"text-align: right;\">        10</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        713.338 </td><td style=\"text-align: right;\">1.38791</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_a8304_00018</td><td>TERMINATED</td><td>127.0.0.1:15616</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.771765</td><td style=\"text-align: right;\">0.00508974 </td><td>adam       </td><td style=\"text-align: right;\">        15</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        809.58  </td><td style=\"text-align: right;\">1.38577</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_a8304_00019</td><td>TERMINATED</td><td>127.0.0.1:11996</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.701096</td><td style=\"text-align: right;\">0.00049464 </td><td>adam       </td><td style=\"text-align: right;\">        10</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         81.5897</td><td style=\"text-align: right;\">1.38674</td><td style=\"text-align: right;\">  80.9306 </td></tr>\n",
       "<tr><td>train_model_a8304_00020</td><td>TERMINATED</td><td>127.0.0.1:17812</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.160771</td><td style=\"text-align: right;\">0.00839899 </td><td>adam       </td><td style=\"text-align: right;\">        10</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">     4</td><td style=\"text-align: right;\">        281.258 </td><td style=\"text-align: right;\">1.38833</td><td style=\"text-align: right;\">   5.17167</td></tr>\n",
       "<tr><td>train_model_a8304_00021</td><td>TERMINATED</td><td>127.0.0.1:28288</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.752537</td><td style=\"text-align: right;\">0.00258126 </td><td>adam       </td><td style=\"text-align: right;\">         5</td><td>exp              </td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         69.6695</td><td style=\"text-align: right;\">1.38746</td><td style=\"text-align: right;\">   7.94776</td></tr>\n",
       "<tr><td>train_model_a8304_00022</td><td>TERMINATED</td><td>127.0.0.1:25360</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.348775</td><td style=\"text-align: right;\">0.00278254 </td><td>adamw      </td><td style=\"text-align: right;\">        15</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        716.131 </td><td style=\"text-align: right;\">1.38633</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_a8304_00023</td><td>TERMINATED</td><td>127.0.0.1:28028</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.355649</td><td style=\"text-align: right;\">0.000222178</td><td>adamw      </td><td style=\"text-align: right;\">        15</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         72.1455</td><td style=\"text-align: right;\">1.31338</td><td style=\"text-align: right;\">  44.5732 </td></tr>\n",
       "<tr><td>train_model_a8304_00024</td><td>TERMINATED</td><td>127.0.0.1:15796</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.414981</td><td style=\"text-align: right;\">0.0080418  </td><td>adamw      </td><td style=\"text-align: right;\">        15</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        713.849 </td><td style=\"text-align: right;\">1.39155</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "</tbody>\n",
       "</table>\n",
       "  </div>\n",
       "</div>\n",
       "<style>\n",
       ".tuneStatus {\n",
       "  color: var(--jp-ui-font-color1);\n",
       "}\n",
       ".tuneStatus .systemInfo {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       ".tuneStatus td {\n",
       "  white-space: nowrap;\n",
       "}\n",
       ".tuneStatus .trialStatus {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       ".tuneStatus h3 {\n",
       "  font-weight: bold;\n",
       "}\n",
       ".tuneStatus .hDivider {\n",
       "  border-bottom-width: var(--jp-border-width);\n",
       "  border-bottom-color: var(--jp-border-color0);\n",
       "  border-bottom-style: solid;\n",
       "}\n",
       ".tuneStatus .vDivider {\n",
       "  border-left-width: var(--jp-border-width);\n",
       "  border-left-color: var(--jp-border-color0);\n",
       "  border-left-style: solid;\n",
       "  margin: 0.5em 1em 0.5em 1em;\n",
       "}\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[36m(train_model pid=13880)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00000/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=13880)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00000/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=13880)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00000/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=13880)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00000/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=13880)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00000/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=13880)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00000/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=13880)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00000/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=13880)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00000/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=13880)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00000/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=13880)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00000/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00001/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00001/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00001/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00001/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00001/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00001/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00001/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00001/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00001/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00001/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=16352)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00002/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=19116)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00003/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=19116)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00003/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=19116)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00003/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=19116)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00003/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=19116)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00003/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=19116)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00003/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=19116)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00003/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=19116)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00003/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=19116)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00003/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=19116)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00003/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=2828)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00004/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=2828)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00004/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=2828)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00004/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=2828)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00004/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=2828)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00004/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=2828)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00004/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=2828)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00004/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=2828)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00004/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=2828)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00004/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=2828)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00004/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=5864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00005/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=5864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00005/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=5864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00005/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=5864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00005/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=5864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00005/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=5864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00005/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=5864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00005/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=5864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00005/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=5864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00005/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=5864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00005/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=12568)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00006/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=12568)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00006/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=12568)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00006/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=12568)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00006/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=12568)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00006/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=12568)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00006/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=12568)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00006/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=12568)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00006/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=12568)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00006/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=12568)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00006/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=24364)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00007/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=24364)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00007/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=24364)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00007/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=24364)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00007/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=24364)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00007/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=24364)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00007/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=24364)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00007/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=24364)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00007/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=24364)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00007/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=24364)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00007/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=26784)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00008/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=26784)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00008/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=5272)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00009/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=5272)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00009/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=5272)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00009/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=5272)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00009/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=5272)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00009/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=5272)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00009/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=5272)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00009/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=5272)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00009/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=5272)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00009/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=5272)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00009/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=15884)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00010/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=15884)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00010/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=15884)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00010/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=15884)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00010/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=15884)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00010/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=15884)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00010/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=15884)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00010/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=15884)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00010/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=15884)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00010/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=15884)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00010/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=18976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00011/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=18976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00011/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=18976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00011/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=18976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00011/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=18976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00011/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=18976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00011/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=18976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00011/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=18976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00011/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=18976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00011/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=18976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00011/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=8412)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00012/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=8412)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00012/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=8412)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00012/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=8412)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00012/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=8412)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00012/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=8412)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00012/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=8412)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00012/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=8412)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00012/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=8412)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00012/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=8412)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00012/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=6852)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00013/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=20196)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00014/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=29672)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00015/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=18844)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00016/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=18844)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00016/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=18844)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00016/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=18844)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00016/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=31140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00017/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=31140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00017/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=31140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00017/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=31140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00017/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=31140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00017/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=31140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00017/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=31140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00017/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=31140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00017/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=31140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00017/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=31140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00017/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=15616)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00018/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=15616)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00018/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=15616)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00018/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=15616)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00018/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=15616)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00018/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=15616)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00018/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=15616)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00018/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=15616)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00018/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=15616)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00018/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=15616)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00018/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=11996)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00019/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=17812)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00020/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=17812)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00020/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=17812)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00020/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=17812)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00020/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=28288)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00021/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=25360)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00022/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=25360)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00022/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=25360)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00022/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=25360)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00022/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=25360)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00022/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=25360)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00022/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=25360)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00022/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=25360)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00022/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=25360)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00022/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=25360)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00022/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=28028)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00023/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=15796)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00024/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=15796)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00024/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=15796)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00024/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=15796)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00024/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=15796)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00024/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=15796)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00024/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=15796)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00024/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=15796)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00024/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=15796)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00024/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=15796)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long/trial_a8304_00024/checkpoint_000009)\n",
      "2024-07-26 04:06:51,358\tINFO tune.py:1009 -- Wrote the latest version of all result files and experiment state to 'c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long' in 0.0406s.\n",
      "2024-07-26 04:06:51,388\tINFO tune.py:1041 -- Total run time: 12647.72 seconds (12647.60 seconds for the tuning loop).\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best trial config: {'optimizer': 'adam', 'lr': 0.006482722864364993, 'scheduler': 'ReduceLROnPlateau', 'gamma': 0.8185367503936773, 'patience': 15, 'epochs': 30, 'batch_size': 32}\n",
      "Best trial final validation loss: 1.3836872604011832\n",
      "Best trial final validation accuracy: 84.1823%\n",
      "Test Accuracy: 0.8532484605272261\n",
      "Best trial test set accuracy: 0.8532%\n"
     ]
    }
   ],
   "source": [
    "mytuner.main(train_datasets['long_acc'], valid_datasets['long_acc'], \n",
    "             test_datasets['long_acc'], num_samples=25)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Hyperparam Tuning - Tranv Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Enhanced Hyperparameter search space\n",
    "config = {\n",
    "    \"optimizer\": tune.choice([\"adam\", \"adamw\"]),\n",
    "    \"lr\": tune.loguniform(1e-4, 1e-2),\n",
    "    \"scheduler\": tune.choice([\"exp\", \"ReduceLROnPlateau\"]),\n",
    "    \"gamma\": tune.uniform(0.1, 0.9),  # for exponential scheduler\n",
    "    \"patience\": tune.choice([5, 10, 15]),  # for ReduceLROnPlateau\n",
    "    \"epochs\": tune.choice([30]),\n",
    "    \"batch_size\": tune.choice([32, 64])\n",
    "}\n",
    "\n",
    "criterion = nn.CrossEntropyLoss(weight=class_weights)\n",
    "save_dir = 'SimpleCNN_tranv'\n",
    "model = SimpleCNN\n",
    "modelType = 'SimpleCNN'\n",
    "engine = simpleCNN_engine"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "mytuner = RayTuning(config, save_dir, criterion, model, modelType, engine)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div class=\"tuneStatus\">\n",
       "  <div style=\"display: flex;flex-direction: row\">\n",
       "    <div style=\"display: flex;flex-direction: column;\">\n",
       "      <h3>Tune Status</h3>\n",
       "      <table>\n",
       "<tbody>\n",
       "<tr><td>Current time:</td><td>2024-07-26 08:08:26</td></tr>\n",
       "<tr><td>Running for: </td><td>04:01:13.86        </td></tr>\n",
       "<tr><td>Memory:      </td><td>10.9/13.9 GiB      </td></tr>\n",
       "</tbody>\n",
       "</table>\n",
       "    </div>\n",
       "    <div class=\"vDivider\"></div>\n",
       "    <div class=\"systemInfo\">\n",
       "      <h3>System Info</h3>\n",
       "      Using AsyncHyperBand: num_stopped=25<br>Bracket: Iter 8.000: 84.1823056300268 | Iter 4.000: 84.1823056300268 | Iter 2.000: 84.1823056300268 | Iter 1.000: 84.1823056300268<br>Logical resource usage: 2.0/16 CPUs, 1.0/1 GPUs (0.0/1.0 accelerator_type:G)\n",
       "    </div>\n",
       "    \n",
       "  </div>\n",
       "  <div class=\"hDivider\"></div>\n",
       "  <div class=\"trialStatus\">\n",
       "    <h3>Trial Status</h3>\n",
       "    <table>\n",
       "<thead>\n",
       "<tr><th>Trial name             </th><th>status    </th><th>loc            </th><th style=\"text-align: right;\">  batch_size</th><th style=\"text-align: right;\">  epochs</th><th style=\"text-align: right;\">   gamma</th><th style=\"text-align: right;\">         lr</th><th>optimizer  </th><th style=\"text-align: right;\">  patience</th><th>scheduler        </th><th style=\"text-align: right;\">  iter</th><th style=\"text-align: right;\">  total time (s)</th><th style=\"text-align: right;\">   loss</th><th style=\"text-align: right;\">  accuracy</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr><td>train_model_27755_00000</td><td>TERMINATED</td><td>127.0.0.1:21988</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.528737</td><td style=\"text-align: right;\">0.00158318 </td><td>adam       </td><td style=\"text-align: right;\">         5</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        806.16  </td><td style=\"text-align: right;\">1.38444</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_27755_00001</td><td>TERMINATED</td><td>127.0.0.1:9064 </td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.771186</td><td style=\"text-align: right;\">0.000529405</td><td>adamw      </td><td style=\"text-align: right;\">         5</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        828.746 </td><td style=\"text-align: right;\">1.38566</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_27755_00002</td><td>TERMINATED</td><td>127.0.0.1:27976</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.483143</td><td style=\"text-align: right;\">0.00254284 </td><td>adamw      </td><td style=\"text-align: right;\">         5</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         71.9858</td><td style=\"text-align: right;\">1.38646</td><td style=\"text-align: right;\">   5.17167</td></tr>\n",
       "<tr><td>train_model_27755_00003</td><td>TERMINATED</td><td>127.0.0.1:18604</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.862524</td><td style=\"text-align: right;\">0.00171393 </td><td>adam       </td><td style=\"text-align: right;\">        15</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        809.232 </td><td style=\"text-align: right;\">1.38344</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_27755_00004</td><td>TERMINATED</td><td>127.0.0.1:17696</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.359897</td><td style=\"text-align: right;\">0.00811564 </td><td>adam       </td><td style=\"text-align: right;\">        10</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        810.891 </td><td style=\"text-align: right;\">1.38612</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_27755_00005</td><td>TERMINATED</td><td>127.0.0.1:29976</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.574084</td><td style=\"text-align: right;\">0.000685096</td><td>adam       </td><td style=\"text-align: right;\">        10</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        701.323 </td><td style=\"text-align: right;\">1.38782</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_27755_00006</td><td>TERMINATED</td><td>127.0.0.1:30656</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.66486 </td><td style=\"text-align: right;\">0.00732549 </td><td>adamw      </td><td style=\"text-align: right;\">        15</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        829.401 </td><td style=\"text-align: right;\">1.38361</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_27755_00007</td><td>TERMINATED</td><td>127.0.0.1:19824</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.740568</td><td style=\"text-align: right;\">0.000760507</td><td>adam       </td><td style=\"text-align: right;\">        15</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        704.21  </td><td style=\"text-align: right;\">1.38806</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_27755_00008</td><td>TERMINATED</td><td>127.0.0.1:16424</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.197586</td><td style=\"text-align: right;\">0.00104904 </td><td>adam       </td><td style=\"text-align: right;\">        10</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        801.032 </td><td style=\"text-align: right;\">1.38525</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_27755_00009</td><td>TERMINATED</td><td>127.0.0.1:5856 </td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.537672</td><td style=\"text-align: right;\">0.000156785</td><td>adamw      </td><td style=\"text-align: right;\">         5</td><td>exp              </td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         79.0356</td><td style=\"text-align: right;\">1.32574</td><td style=\"text-align: right;\">  49.1568 </td></tr>\n",
       "<tr><td>train_model_27755_00010</td><td>TERMINATED</td><td>127.0.0.1:24788</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.387067</td><td style=\"text-align: right;\">0.000716382</td><td>adamw      </td><td style=\"text-align: right;\">        15</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        685.636 </td><td style=\"text-align: right;\">1.38766</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_27755_00011</td><td>TERMINATED</td><td>127.0.0.1:13180</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.307673</td><td style=\"text-align: right;\">0.000349233</td><td>adamw      </td><td style=\"text-align: right;\">        10</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">     4</td><td style=\"text-align: right;\">        318.439 </td><td style=\"text-align: right;\">1.37035</td><td style=\"text-align: right;\">  61.6881 </td></tr>\n",
       "<tr><td>train_model_27755_00012</td><td>TERMINATED</td><td>127.0.0.1:3868 </td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.26508 </td><td style=\"text-align: right;\">0.000113466</td><td>adam       </td><td style=\"text-align: right;\">        10</td><td>exp              </td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         78.1023</td><td style=\"text-align: right;\">1.3264 </td><td style=\"text-align: right;\">  48.0931 </td></tr>\n",
       "<tr><td>train_model_27755_00013</td><td>TERMINATED</td><td>127.0.0.1:10548</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.597227</td><td style=\"text-align: right;\">0.00176478 </td><td>adam       </td><td style=\"text-align: right;\">        15</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        676.49  </td><td style=\"text-align: right;\">1.38746</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_27755_00014</td><td>TERMINATED</td><td>127.0.0.1:28972</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.159832</td><td style=\"text-align: right;\">0.00631054 </td><td>adam       </td><td style=\"text-align: right;\">        15</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        676.912 </td><td style=\"text-align: right;\">1.38704</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_27755_00015</td><td>TERMINATED</td><td>127.0.0.1:22812</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.702361</td><td style=\"text-align: right;\">0.00126435 </td><td>adamw      </td><td style=\"text-align: right;\">        15</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        800.353 </td><td style=\"text-align: right;\">1.38354</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_27755_00016</td><td>TERMINATED</td><td>127.0.0.1:28748</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.320806</td><td style=\"text-align: right;\">0.000146893</td><td>adamw      </td><td style=\"text-align: right;\">        15</td><td>exp              </td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         68.9882</td><td style=\"text-align: right;\">1.3543 </td><td style=\"text-align: right;\">  25.8324 </td></tr>\n",
       "<tr><td>train_model_27755_00017</td><td>TERMINATED</td><td>127.0.0.1:18100</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.52119 </td><td style=\"text-align: right;\">0.00665138 </td><td>adam       </td><td style=\"text-align: right;\">        10</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        675.702 </td><td style=\"text-align: right;\">1.38958</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_27755_00018</td><td>TERMINATED</td><td>127.0.0.1:11100</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.500503</td><td style=\"text-align: right;\">0.00184062 </td><td>adam       </td><td style=\"text-align: right;\">        15</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        678.423 </td><td style=\"text-align: right;\">1.38754</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_27755_00019</td><td>TERMINATED</td><td>127.0.0.1:21752</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.211871</td><td style=\"text-align: right;\">0.00809945 </td><td>adam       </td><td style=\"text-align: right;\">        10</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        781.172 </td><td style=\"text-align: right;\">1.38749</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_27755_00020</td><td>TERMINATED</td><td>127.0.0.1:16548</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.873226</td><td style=\"text-align: right;\">0.00502119 </td><td>adamw      </td><td style=\"text-align: right;\">         5</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        803.017 </td><td style=\"text-align: right;\">1.38428</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_27755_00021</td><td>TERMINATED</td><td>127.0.0.1:344  </td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.113677</td><td style=\"text-align: right;\">0.000658181</td><td>adam       </td><td style=\"text-align: right;\">        15</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        783.406 </td><td style=\"text-align: right;\">1.38687</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_27755_00022</td><td>TERMINATED</td><td>127.0.0.1:18392</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.551679</td><td style=\"text-align: right;\">0.00701793 </td><td>adam       </td><td style=\"text-align: right;\">        15</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        677.79  </td><td style=\"text-align: right;\">1.38786</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_27755_00023</td><td>TERMINATED</td><td>127.0.0.1:12076</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.159838</td><td style=\"text-align: right;\">0.000129452</td><td>adam       </td><td style=\"text-align: right;\">        10</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         68.2402</td><td style=\"text-align: right;\">1.31998</td><td style=\"text-align: right;\">  51.5351 </td></tr>\n",
       "<tr><td>train_model_27755_00024</td><td>TERMINATED</td><td>127.0.0.1:15172</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.836819</td><td style=\"text-align: right;\">0.000122222</td><td>adamw      </td><td style=\"text-align: right;\">        10</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         80.5245</td><td style=\"text-align: right;\">1.30556</td><td style=\"text-align: right;\">  59.9758 </td></tr>\n",
       "</tbody>\n",
       "</table>\n",
       "  </div>\n",
       "</div>\n",
       "<style>\n",
       ".tuneStatus {\n",
       "  color: var(--jp-ui-font-color1);\n",
       "}\n",
       ".tuneStatus .systemInfo {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       ".tuneStatus td {\n",
       "  white-space: nowrap;\n",
       "}\n",
       ".tuneStatus .trialStatus {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       ".tuneStatus h3 {\n",
       "  font-weight: bold;\n",
       "}\n",
       ".tuneStatus .hDivider {\n",
       "  border-bottom-width: var(--jp-border-width);\n",
       "  border-bottom-color: var(--jp-border-color0);\n",
       "  border-bottom-style: solid;\n",
       "}\n",
       ".tuneStatus .vDivider {\n",
       "  border-left-width: var(--jp-border-width);\n",
       "  border-left-color: var(--jp-border-color0);\n",
       "  border-left-style: solid;\n",
       "  margin: 0.5em 1em 0.5em 1em;\n",
       "}\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[36m(train_model pid=21988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00000/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=21988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00000/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=21988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00000/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=21988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00000/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=21988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00000/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=21988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00000/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=21988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00000/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=21988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00000/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=21988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00000/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=21988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00000/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=9064)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00001/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=9064)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00001/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=9064)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00001/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=9064)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00001/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=9064)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00001/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=9064)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00001/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=9064)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00001/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=9064)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00001/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=9064)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00001/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=9064)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00001/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=27976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00002/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=18604)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00003/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=18604)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00003/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=18604)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00003/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=18604)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00003/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=18604)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00003/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=18604)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00003/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=18604)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00003/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=18604)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00003/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=18604)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00003/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=18604)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00003/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00004/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00004/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00004/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00004/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00004/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00004/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00004/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00004/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00004/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=17696)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00004/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=29976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00005/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=29976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00005/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=29976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00005/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=29976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00005/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=29976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00005/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=29976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00005/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=29976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00005/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=29976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00005/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=29976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00005/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=29976)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00005/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=30656)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00006/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=30656)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00006/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=30656)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00006/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=30656)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00006/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=30656)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00006/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=30656)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00006/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=30656)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00006/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=30656)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00006/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=30656)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00006/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=30656)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00006/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=19824)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00007/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=19824)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00007/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=19824)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00007/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=19824)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00007/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=19824)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00007/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=19824)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00007/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=19824)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00007/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=19824)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00007/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=19824)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00007/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=19824)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00007/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=16424)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00008/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=16424)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00008/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=16424)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00008/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=16424)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00008/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=16424)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00008/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=16424)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00008/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=16424)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00008/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=16424)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00008/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=16424)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00008/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=16424)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00008/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=5856)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00009/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=24788)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00010/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=24788)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00010/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=24788)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00010/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=24788)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00010/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=24788)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00010/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=24788)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00010/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=24788)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00010/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=24788)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00010/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=24788)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00010/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=24788)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00010/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=13180)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00011/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=13180)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00011/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=13180)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00011/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=13180)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00011/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=3868)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00012/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=10548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00013/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=10548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00013/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=10548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00013/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=10548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00013/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=10548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00013/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=10548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00013/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=10548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00013/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=10548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00013/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=10548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00013/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=10548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00013/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=28972)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00014/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=28972)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00014/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=28972)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00014/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=28972)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00014/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=28972)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00014/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=28972)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00014/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=28972)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00014/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=28972)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00014/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=28972)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00014/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=28972)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00014/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=22812)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00015/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=22812)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00015/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=22812)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00015/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=22812)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00015/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=22812)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00015/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=22812)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00015/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=22812)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00015/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=22812)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00015/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=22812)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00015/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=22812)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00015/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=28748)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00016/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00017/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00017/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00017/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00017/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00017/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00017/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00017/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00017/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00017/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00017/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=11100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00018/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=11100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00018/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=11100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00018/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=11100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00018/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=11100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00018/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=11100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00018/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=11100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00018/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=11100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00018/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=11100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00018/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=11100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00018/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=21752)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00019/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=21752)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00019/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=21752)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00019/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=21752)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00019/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=21752)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00019/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=21752)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00019/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=21752)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00019/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=21752)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00019/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=21752)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00019/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=21752)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00019/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=16548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00020/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=16548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00020/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=16548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00020/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=16548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00020/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=16548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00020/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=16548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00020/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=16548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00020/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=16548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00020/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=16548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00020/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=16548)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00020/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=344)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00021/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=344)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00021/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=344)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00021/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=344)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00021/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=344)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00021/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=344)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00021/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=344)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00021/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=344)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00021/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=344)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00021/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=344)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00021/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=18392)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00022/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=18392)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00022/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=18392)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00022/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=18392)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00022/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=18392)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00022/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=18392)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00022/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=18392)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00022/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=18392)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00022/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=18392)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00022/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=18392)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00022/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=12076)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00023/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=15172)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv/trial_27755_00024/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=22600)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00000/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=22600)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00000/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=22600)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00000/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=22600)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00000/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=22600)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00000/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=22600)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00000/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=22600)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00000/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=22600)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00000/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=22600)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00000/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=22600)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00000/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=18580)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00001/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=24932)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00002/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00003/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00003/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00003/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00003/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00003/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00003/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00003/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00003/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00003/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=18100)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00003/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=12796)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00004/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=15988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00005/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=15988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00005/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=15988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00005/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=15988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00005/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=15988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00005/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=15988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00005/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=15988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00005/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=15988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00005/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=15988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00005/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=15988)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00005/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00006/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00006/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00006/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00006/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00006/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00006/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00006/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00006/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00006/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=864)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00006/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=9656)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00007/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=23500)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00008/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=23500)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00008/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=23500)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00008/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=23500)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00008/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=23500)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00008/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=23500)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00008/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=23500)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00008/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=23500)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00008/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=23500)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00008/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=23500)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00008/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=11328)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00009/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=11328)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00009/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=11328)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00009/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=11328)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00009/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=11328)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00009/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=11328)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00009/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=11328)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00009/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=11328)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00009/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=11328)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00009/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=11328)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00009/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=10148)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00010/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=18132)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00011/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=18132)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00011/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=18132)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00011/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=18132)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00011/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=18132)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00011/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=18132)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00011/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=18132)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00011/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=18132)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00011/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=18132)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00011/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=18132)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00011/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=18980)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00012/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=17700)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00013/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=29780)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00014/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00015/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00015/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00015/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00015/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00015/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00015/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00015/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00015/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00015/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00015/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=14876)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00016/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=13080)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00017/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=16920)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00018/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=16920)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00018/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=16920)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00018/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=16920)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00018/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=16920)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00018/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=16920)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00018/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=16920)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00018/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=16920)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00018/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=16920)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00018/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=16920)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00018/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=27236)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00019/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=27236)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00019/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=27236)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00019/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=27236)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00019/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=27236)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00019/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=27236)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00019/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=27236)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00019/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=27236)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00019/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=20532)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00020/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=20532)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00020/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=20532)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00020/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=20532)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00020/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=20532)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00020/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=20532)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00020/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=20532)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00020/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=20532)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00020/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=20532)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00020/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=20532)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00020/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=29140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00021/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=29140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00021/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=29140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00021/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=29140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00021/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=29140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00021/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=29140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00021/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=29140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00021/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=29140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00021/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=29140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00021/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=29140)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00021/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00022/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00022/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00022/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00022/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00022/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00022/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00022/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00022/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00022/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=4764)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00022/checkpoint_000009)\n",
      "\u001b[36m(train_model pid=16248)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00023/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=13860)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00024/checkpoint_000000)\n",
      "\u001b[36m(train_model pid=13860)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00024/checkpoint_000001)\n",
      "\u001b[36m(train_model pid=13860)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00024/checkpoint_000002)\n",
      "\u001b[36m(train_model pid=13860)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00024/checkpoint_000003)\n",
      "\u001b[36m(train_model pid=13860)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00024/checkpoint_000004)\n",
      "\u001b[36m(train_model pid=13860)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00024/checkpoint_000005)\n",
      "\u001b[36m(train_model pid=13860)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00024/checkpoint_000006)\n",
      "\u001b[36m(train_model pid=13860)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00024/checkpoint_000007)\n",
      "\u001b[36m(train_model pid=13860)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00024/checkpoint_000008)\n",
      "\u001b[36m(train_model pid=13860)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel/trial_e68b4_00024/checkpoint_000009)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-07-26 08:08:26,464\tINFO tune.py:1009 -- Wrote the latest version of all result files and experiment state to 'c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_tranv' in 0.0289s.\n",
      "2024-07-26 08:08:26,495\tINFO tune.py:1041 -- Total run time: 14473.90 seconds (14473.83 seconds for the tuning loop).\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best trial config: {'optimizer': 'adam', 'lr': 0.0015831811326078972, 'scheduler': 'exp', 'gamma': 0.5287373400905854, 'patience': 5, 'epochs': 30, 'batch_size': 32}\n",
      "Best trial final validation loss: 1.3844412874780307\n",
      "Best trial final validation accuracy: 84.1823%\n",
      "Test Accuracy: 0.8532484605272261\n",
      "Best trial test set accuracy: 0.8532%\n"
     ]
    }
   ],
   "source": [
    "mytuner.main(train_datasets['tranv_acc'], valid_datasets['tranv_acc'], \n",
    "             test_datasets['tranv_acc'], num_samples=25)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Hyperparam Tuning - Ang. Vel Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Enhanced Hyperparameter search space\n",
    "config = {\n",
    "    \"optimizer\": tune.choice([\"adam\", \"adamw\"]),\n",
    "    \"lr\": tune.loguniform(1e-4, 1e-2),\n",
    "    \"scheduler\": tune.choice([\"exp\", \"ReduceLROnPlateau\"]),\n",
    "    \"gamma\": tune.uniform(0.1, 0.9),  # for exponential scheduler\n",
    "    \"patience\": tune.choice([5, 10, 15]),  # for ReduceLROnPlateau\n",
    "    \"epochs\": tune.choice([30]),\n",
    "    \"batch_size\": tune.choice([32, 64])\n",
    "}\n",
    "\n",
    "criterion = nn.CrossEntropyLoss(weight=class_weights)\n",
    "save_dir = 'SimpleCNN_ang_vel'\n",
    "model = SimpleCNN\n",
    "modelType = 'SimpleCNN'\n",
    "engine = simpleCNN_engine"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "mytuner = RayTuning(config, save_dir, criterion, model, modelType, engine)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div class=\"tuneStatus\">\n",
       "  <div style=\"display: flex;flex-direction: row\">\n",
       "    <div style=\"display: flex;flex-direction: column;\">\n",
       "      <h3>Tune Status</h3>\n",
       "      <table>\n",
       "<tbody>\n",
       "<tr><td>Current time:</td><td>2024-07-26 11:22:28</td></tr>\n",
       "<tr><td>Running for: </td><td>03:13:41.68        </td></tr>\n",
       "<tr><td>Memory:      </td><td>9.7/13.9 GiB       </td></tr>\n",
       "</tbody>\n",
       "</table>\n",
       "    </div>\n",
       "    <div class=\"vDivider\"></div>\n",
       "    <div class=\"systemInfo\">\n",
       "      <h3>System Info</h3>\n",
       "      Using AsyncHyperBand: num_stopped=25<br>Bracket: Iter 8.000: 84.1823056300268 | Iter 4.000: 84.1823056300268 | Iter 2.000: 84.1823056300268 | Iter 1.000: 84.1823056300268<br>Logical resource usage: 2.0/16 CPUs, 1.0/1 GPUs (0.0/1.0 accelerator_type:G)\n",
       "    </div>\n",
       "    \n",
       "  </div>\n",
       "  <div class=\"hDivider\"></div>\n",
       "  <div class=\"trialStatus\">\n",
       "    <h3>Trial Status</h3>\n",
       "    <table>\n",
       "<thead>\n",
       "<tr><th>Trial name             </th><th>status    </th><th>loc            </th><th style=\"text-align: right;\">  batch_size</th><th style=\"text-align: right;\">  epochs</th><th style=\"text-align: right;\">   gamma</th><th style=\"text-align: right;\">         lr</th><th>optimizer  </th><th style=\"text-align: right;\">  patience</th><th>scheduler        </th><th style=\"text-align: right;\">  iter</th><th style=\"text-align: right;\">  total time (s)</th><th style=\"text-align: right;\">   loss</th><th style=\"text-align: right;\">  accuracy</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr><td>train_model_e68b4_00000</td><td>TERMINATED</td><td>127.0.0.1:22600</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.122708</td><td style=\"text-align: right;\">0.00294518 </td><td>adam       </td><td style=\"text-align: right;\">         5</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        782.417 </td><td style=\"text-align: right;\">1.38678</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_e68b4_00001</td><td>TERMINATED</td><td>127.0.0.1:18580</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.607312</td><td style=\"text-align: right;\">0.000276186</td><td>adamw      </td><td style=\"text-align: right;\">        10</td><td>exp              </td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         69.7181</td><td style=\"text-align: right;\">1.35348</td><td style=\"text-align: right;\">  47.0207 </td></tr>\n",
       "<tr><td>train_model_e68b4_00002</td><td>TERMINATED</td><td>127.0.0.1:24932</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.554379</td><td style=\"text-align: right;\">0.000348329</td><td>adam       </td><td style=\"text-align: right;\">         5</td><td>exp              </td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         78.7623</td><td style=\"text-align: right;\">1.36478</td><td style=\"text-align: right;\">  25.3913 </td></tr>\n",
       "<tr><td>train_model_e68b4_00003</td><td>TERMINATED</td><td>127.0.0.1:18100</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.417497</td><td style=\"text-align: right;\">0.000609574</td><td>adamw      </td><td style=\"text-align: right;\">        10</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        808.306 </td><td style=\"text-align: right;\">1.38553</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_e68b4_00004</td><td>TERMINATED</td><td>127.0.0.1:12796</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.541492</td><td style=\"text-align: right;\">0.000106497</td><td>adam       </td><td style=\"text-align: right;\">         5</td><td>exp              </td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         69.0398</td><td style=\"text-align: right;\">1.34828</td><td style=\"text-align: right;\">  55.5738 </td></tr>\n",
       "<tr><td>train_model_e68b4_00005</td><td>TERMINATED</td><td>127.0.0.1:15988</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.781498</td><td style=\"text-align: right;\">0.00102957 </td><td>adamw      </td><td style=\"text-align: right;\">         5</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        808.335 </td><td style=\"text-align: right;\">1.38423</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_e68b4_00006</td><td>TERMINATED</td><td>127.0.0.1:864  </td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.621413</td><td style=\"text-align: right;\">0.000864981</td><td>adam       </td><td style=\"text-align: right;\">        15</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        684.38  </td><td style=\"text-align: right;\">1.38712</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_e68b4_00007</td><td>TERMINATED</td><td>127.0.0.1:9656 </td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.597032</td><td style=\"text-align: right;\">0.000312471</td><td>adam       </td><td style=\"text-align: right;\">         5</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         79.2085</td><td style=\"text-align: right;\">1.35023</td><td style=\"text-align: right;\">  38.2427 </td></tr>\n",
       "<tr><td>train_model_e68b4_00008</td><td>TERMINATED</td><td>127.0.0.1:23500</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.129619</td><td style=\"text-align: right;\">0.00356949 </td><td>adam       </td><td style=\"text-align: right;\">        15</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        690.173 </td><td style=\"text-align: right;\">1.38702</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_e68b4_00009</td><td>TERMINATED</td><td>127.0.0.1:11328</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.770556</td><td style=\"text-align: right;\">0.00107333 </td><td>adamw      </td><td style=\"text-align: right;\">        10</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        817.818 </td><td style=\"text-align: right;\">1.38434</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_e68b4_00010</td><td>TERMINATED</td><td>127.0.0.1:10148</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.736762</td><td style=\"text-align: right;\">0.000152187</td><td>adamw      </td><td style=\"text-align: right;\">         5</td><td>exp              </td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         66.6627</td><td style=\"text-align: right;\">1.35215</td><td style=\"text-align: right;\">  36.6254 </td></tr>\n",
       "<tr><td>train_model_e68b4_00011</td><td>TERMINATED</td><td>127.0.0.1:18132</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.366736</td><td style=\"text-align: right;\">0.00721537 </td><td>adam       </td><td style=\"text-align: right;\">        10</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        812.803 </td><td style=\"text-align: right;\">1.38607</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_e68b4_00012</td><td>TERMINATED</td><td>127.0.0.1:18980</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.795434</td><td style=\"text-align: right;\">0.000239143</td><td>adam       </td><td style=\"text-align: right;\">         5</td><td>exp              </td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         70.9889</td><td style=\"text-align: right;\">1.3585 </td><td style=\"text-align: right;\">  47.7904 </td></tr>\n",
       "<tr><td>train_model_e68b4_00013</td><td>TERMINATED</td><td>127.0.0.1:17700</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.472953</td><td style=\"text-align: right;\">0.000221048</td><td>adam       </td><td style=\"text-align: right;\">        10</td><td>exp              </td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         75.6058</td><td style=\"text-align: right;\">1.36045</td><td style=\"text-align: right;\">  34.1001 </td></tr>\n",
       "<tr><td>train_model_e68b4_00014</td><td>TERMINATED</td><td>127.0.0.1:29780</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.708872</td><td style=\"text-align: right;\">0.000266896</td><td>adam       </td><td style=\"text-align: right;\">         5</td><td>exp              </td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         72.473 </td><td style=\"text-align: right;\">1.35977</td><td style=\"text-align: right;\">  47.4617 </td></tr>\n",
       "<tr><td>train_model_e68b4_00015</td><td>TERMINATED</td><td>127.0.0.1:4764 </td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.127397</td><td style=\"text-align: right;\">0.00751575 </td><td>adamw      </td><td style=\"text-align: right;\">        10</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        848.595 </td><td style=\"text-align: right;\">1.38482</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_e68b4_00016</td><td>TERMINATED</td><td>127.0.0.1:14876</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.544549</td><td style=\"text-align: right;\">0.000406395</td><td>adamw      </td><td style=\"text-align: right;\">        10</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         91.1313</td><td style=\"text-align: right;\">1.36889</td><td style=\"text-align: right;\">  32.4137 </td></tr>\n",
       "<tr><td>train_model_e68b4_00017</td><td>TERMINATED</td><td>127.0.0.1:13080</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.521269</td><td style=\"text-align: right;\">0.00414551 </td><td>adam       </td><td style=\"text-align: right;\">         5</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         74.3608</td><td style=\"text-align: right;\">1.387  </td><td style=\"text-align: right;\">   7.94776</td></tr>\n",
       "<tr><td>train_model_e68b4_00018</td><td>TERMINATED</td><td>127.0.0.1:16920</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.603962</td><td style=\"text-align: right;\">0.00121684 </td><td>adamw      </td><td style=\"text-align: right;\">        15</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        753.419 </td><td style=\"text-align: right;\">1.38812</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_e68b4_00019</td><td>TERMINATED</td><td>127.0.0.1:27236</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.641946</td><td style=\"text-align: right;\">0.00818439 </td><td>adam       </td><td style=\"text-align: right;\">         5</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">     8</td><td style=\"text-align: right;\">        587.838 </td><td style=\"text-align: right;\">1.38775</td><td style=\"text-align: right;\">   7.94776</td></tr>\n",
       "<tr><td>train_model_e68b4_00020</td><td>TERMINATED</td><td>127.0.0.1:20532</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.418415</td><td style=\"text-align: right;\">0.00909001 </td><td>adam       </td><td style=\"text-align: right;\">        10</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        804.216 </td><td style=\"text-align: right;\">1.38581</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_e68b4_00021</td><td>TERMINATED</td><td>127.0.0.1:29140</td><td style=\"text-align: right;\">          32</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.871732</td><td style=\"text-align: right;\">0.00528139 </td><td>adamw      </td><td style=\"text-align: right;\">         5</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        819.107 </td><td style=\"text-align: right;\">1.38472</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_e68b4_00022</td><td>TERMINATED</td><td>127.0.0.1:4764 </td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.822356</td><td style=\"text-align: right;\">0.000677725</td><td>adamw      </td><td style=\"text-align: right;\">        15</td><td>exp              </td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        702.995 </td><td style=\"text-align: right;\">1.38719</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "<tr><td>train_model_e68b4_00023</td><td>TERMINATED</td><td>127.0.0.1:16248</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.224919</td><td style=\"text-align: right;\">0.000723663</td><td>adam       </td><td style=\"text-align: right;\">        15</td><td>exp              </td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         69.059 </td><td style=\"text-align: right;\">1.36515</td><td style=\"text-align: right;\">  25.8497 </td></tr>\n",
       "<tr><td>train_model_e68b4_00024</td><td>TERMINATED</td><td>127.0.0.1:13860</td><td style=\"text-align: right;\">          64</td><td style=\"text-align: right;\">      30</td><td style=\"text-align: right;\">0.185824</td><td style=\"text-align: right;\">0.00098898 </td><td>adam       </td><td style=\"text-align: right;\">        10</td><td>ReduceLROnPlateau</td><td style=\"text-align: right;\">    10</td><td style=\"text-align: right;\">        701.145 </td><td style=\"text-align: right;\">1.38812</td><td style=\"text-align: right;\">  84.1823 </td></tr>\n",
       "</tbody>\n",
       "</table>\n",
       "  </div>\n",
       "</div>\n",
       "<style>\n",
       ".tuneStatus {\n",
       "  color: var(--jp-ui-font-color1);\n",
       "}\n",
       ".tuneStatus .systemInfo {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       ".tuneStatus td {\n",
       "  white-space: nowrap;\n",
       "}\n",
       ".tuneStatus .trialStatus {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       ".tuneStatus h3 {\n",
       "  font-weight: bold;\n",
       "}\n",
       ".tuneStatus .hDivider {\n",
       "  border-bottom-width: var(--jp-border-width);\n",
       "  border-bottom-color: var(--jp-border-color0);\n",
       "  border-bottom-style: solid;\n",
       "}\n",
       ".tuneStatus .vDivider {\n",
       "  border-left-width: var(--jp-border-width);\n",
       "  border-left-color: var(--jp-border-color0);\n",
       "  border-left-style: solid;\n",
       "  margin: 0.5em 1em 0.5em 1em;\n",
       "}\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-07-26 11:22:28,268\tINFO tune.py:1009 -- Wrote the latest version of all result files and experiment state to 'c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_ang_vel' in 0.0357s.\n",
      "2024-07-26 11:22:28,293\tINFO tune.py:1041 -- Total run time: 11621.73 seconds (11621.64 seconds for the tuning loop).\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best trial config: {'optimizer': 'adam', 'lr': 0.0029451848340113214, 'scheduler': 'ReduceLROnPlateau', 'gamma': 0.12270840309802439, 'patience': 5, 'epochs': 30, 'batch_size': 32}\n",
      "Best trial final validation loss: 1.386783880752753\n",
      "Best trial final validation accuracy: 84.1823%\n",
      "Test Accuracy: 0.8532484605272261\n",
      "Best trial test set accuracy: 0.8532%\n"
     ]
    }
   ],
   "source": [
    "mytuner.main(train_datasets['ang_vel'], valid_datasets['ang_vel'], \n",
    "             test_datasets['ang_vel'], num_samples=25)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To view tensorboard, run below in terminal;\n",
    "\n",
    "`tensorboard --logdir=\"c:/Users/LEGION/Documents/Msc Data Science - Uni of Exeter/ECMM451 - Data Science Research Project (2023)/workspace/ray_results/SimpleCNN_long\"`\n",
    "\n",
    "-----\n",
    "\n",
    "#### Model Training\n",
    "Time to train the model with the given optimal hyperparameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "from src.model_simpleCNN import SimpleCNN\n",
    "from src.engine import simpleCNN_engine"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.random.seed(42)\n",
    "\n",
    "batch_size = 32\n",
    "train_dl_long = DataLoader(train_datasets['long_acc'], batch_size, shuffle=True, num_workers=4)\n",
    "train_dl_tranv = DataLoader(train_datasets['tranv_acc'], batch_size, shuffle=True, num_workers=4)\n",
    "train_dl_angvel = DataLoader(train_datasets['ang_vel'], batch_size, shuffle=True, num_workers=4)\n",
    "\n",
    "valid_dl_long = DataLoader(valid_datasets['long_acc'], batch_size, shuffle=True, num_workers=4)\n",
    "valid_dl_tranv = DataLoader(valid_datasets['tranv_acc'], batch_size, shuffle=True, num_workers=4)\n",
    "valid_dl_angvel = DataLoader(valid_datasets['ang_vel'], batch_size, shuffle=True, num_workers=4)\n",
    "\n",
    "test_dl_long = DataLoader(test_datasets['long_acc'], batch_size, shuffle=False, num_workers=2)\n",
    "test_dl_tranv = DataLoader(test_datasets['tranv_acc'], batch_size, shuffle=False, num_workers=2)\n",
    "test_dl_angvel = DataLoader(test_datasets['ang_vel'], batch_size, shuffle=False, num_workers=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Longitudinal acceleration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "criterion = nn.CrossEntropyLoss(weight=class_weights)\n",
    "model_long = SimpleCNN()\n",
    "optimizer_long = optim.Adam(model_long.parameters(), lr=0.00011113828101205411, weight_decay=1e-4)\n",
    "scheduler_long = optim.lr_scheduler.ExponentialLR(optimizer_long, gamma=0.8492356384312725)\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "\n",
    "model_long = model_long.to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "engine_long = simpleCNN_engine(model_long, optimizer_long, scheduler_long, criterion, device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   Epoch [1/20], Train Loss: 1.7608, Train Accuracy: 31.34%, Val Loss: 1.4588, Val Accuracy: 24.67%\n",
      "   Epoch [2/20], Train Loss: 1.5188, Train Accuracy: 38.94%, Val Loss: 1.4012, Val Accuracy: 57.27%\n",
      "   Epoch [3/20], Train Loss: 1.3438, Train Accuracy: 52.37%, Val Loss: 1.3145, Val Accuracy: 70.79%\n",
      "   Epoch [4/20], Train Loss: 1.2767, Train Accuracy: 59.19%, Val Loss: 1.3164, Val Accuracy: 70.98%\n",
      "   Epoch [5/20], Train Loss: 1.2277, Train Accuracy: 62.01%, Val Loss: 1.2572, Val Accuracy: 63.90%\n",
      "   Epoch [6/20], Train Loss: 1.1954, Train Accuracy: 63.55%, Val Loss: 1.3131, Val Accuracy: 69.81%\n",
      "   Epoch [7/20], Train Loss: 1.1597, Train Accuracy: 64.54%, Val Loss: 1.2886, Val Accuracy: 66.38%\n",
      "   Epoch [8/20], Train Loss: 1.1213, Train Accuracy: 65.35%, Val Loss: 1.2994, Val Accuracy: 72.02%\n",
      "   Epoch [9/20], Train Loss: 1.1175, Train Accuracy: 65.89%, Val Loss: 1.2766, Val Accuracy: 66.33%\n",
      "   Epoch [10/20], Train Loss: 1.0935, Train Accuracy: 65.97%, Val Loss: 1.2818, Val Accuracy: 65.55%\n",
      "   Epoch [11/20], Train Loss: 1.0799, Train Accuracy: 66.17%, Val Loss: 1.2670, Val Accuracy: 67.71%\n",
      "   Epoch [12/20], Train Loss: 1.0522, Train Accuracy: 67.36%, Val Loss: 1.2786, Val Accuracy: 66.47%\n",
      "   Epoch [13/20], Train Loss: 1.0463, Train Accuracy: 67.29%, Val Loss: 1.2521, Val Accuracy: 68.11%\n",
      "   Epoch [14/20], Train Loss: 1.0363, Train Accuracy: 67.37%, Val Loss: 1.2608, Val Accuracy: 69.83%\n",
      "   Epoch [15/20], Train Loss: 1.0268, Train Accuracy: 67.63%, Val Loss: 1.3044, Val Accuracy: 66.19%\n",
      "   Epoch [16/20], Train Loss: 1.0264, Train Accuracy: 67.32%, Val Loss: 1.3032, Val Accuracy: 67.47%\n",
      "   Epoch [17/20], Train Loss: 1.0209, Train Accuracy: 67.93%, Val Loss: 1.2852, Val Accuracy: 69.17%\n",
      "   Epoch [18/20], Train Loss: 1.0158, Train Accuracy: 67.89%, Val Loss: 1.2568, Val Accuracy: 69.65%\n",
      "   Epoch [19/20], Train Loss: 1.0040, Train Accuracy: 68.02%, Val Loss: 1.2789, Val Accuracy: 69.40%\n",
      "   Epoch [20/20], Train Loss: 1.0009, Train Accuracy: 68.81%, Val Loss: 1.2812, Val Accuracy: 68.75%\n"
     ]
    }
   ],
   "source": [
    "long_hist = engine_long.train_validation(train_dl_long, valid_dl_long, epochs=20, save_path='./model_checkpoint/run2_simpleCNN_long_chkpt.pth')\n",
    "\n",
    "np.save('./model_checkpoint/run2_long_history.npy', np.array(long_hist))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "from src.plot import plot_history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_history(long_hist)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Tranversal Acceleration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "criterion = nn.CrossEntropyLoss(weight=class_weights)\n",
    "model_tranv = SimpleCNN()\n",
    "optimizer_tranv = optim.Adam(model_tranv.parameters(), lr=0.00015558427999575448, weight_decay=1e-4)\n",
    "scheduler_tranv = optim.lr_scheduler.ExponentialLR(optimizer_tranv, gamma=0.6560277884599833)\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "\n",
    "model_tranv = model_tranv.to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "engine_tranv = simpleCNN_engine(model_tranv, optimizer_tranv, scheduler_tranv, criterion, device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   Epoch [1/20], Train Loss: 1.9712, Train Accuracy: 31.83%, Val Loss: 2.0513, Val Accuracy: 34.11%\n",
      "   Epoch [2/20], Train Loss: 1.6318, Train Accuracy: 37.51%, Val Loss: 1.4291, Val Accuracy: 43.28%\n",
      "   Epoch [3/20], Train Loss: 1.4013, Train Accuracy: 49.68%, Val Loss: 1.2625, Val Accuracy: 56.28%\n",
      "   Epoch [4/20], Train Loss: 1.2955, Train Accuracy: 58.05%, Val Loss: 1.2406, Val Accuracy: 68.10%\n",
      "   Epoch [5/20], Train Loss: 1.2303, Train Accuracy: 60.61%, Val Loss: 1.2289, Val Accuracy: 68.73%\n",
      "   Epoch [6/20], Train Loss: 1.1915, Train Accuracy: 62.63%, Val Loss: 1.2135, Val Accuracy: 66.71%\n",
      "   Epoch [7/20], Train Loss: 1.1780, Train Accuracy: 63.60%, Val Loss: 1.2136, Val Accuracy: 64.92%\n",
      "   Epoch [8/20], Train Loss: 1.1538, Train Accuracy: 64.29%, Val Loss: 1.2101, Val Accuracy: 65.75%\n",
      "   Epoch [9/20], Train Loss: 1.1471, Train Accuracy: 64.57%, Val Loss: 1.2034, Val Accuracy: 66.08%\n",
      "   Epoch [10/20], Train Loss: 1.1454, Train Accuracy: 64.70%, Val Loss: 1.2133, Val Accuracy: 67.19%\n",
      "   Epoch [11/20], Train Loss: 1.1348, Train Accuracy: 64.92%, Val Loss: 1.2190, Val Accuracy: 70.17%\n",
      "   Epoch [12/20], Train Loss: 1.1184, Train Accuracy: 65.46%, Val Loss: 1.2026, Val Accuracy: 65.67%\n",
      "   Epoch [13/20], Train Loss: 1.1233, Train Accuracy: 65.03%, Val Loss: 1.2063, Val Accuracy: 67.77%\n",
      "   Epoch [14/20], Train Loss: 1.1233, Train Accuracy: 65.48%, Val Loss: 1.2110, Val Accuracy: 67.88%\n",
      "   Epoch [15/20], Train Loss: 1.1330, Train Accuracy: 65.69%, Val Loss: 1.2076, Val Accuracy: 69.38%\n",
      "   Epoch [16/20], Train Loss: 1.1376, Train Accuracy: 65.50%, Val Loss: 1.2109, Val Accuracy: 67.95%\n",
      "   Epoch [17/20], Train Loss: 1.1234, Train Accuracy: 65.59%, Val Loss: 1.2042, Val Accuracy: 66.22%\n",
      "   Epoch [18/20], Train Loss: 1.1366, Train Accuracy: 65.34%, Val Loss: 1.2151, Val Accuracy: 67.88%\n",
      "   Epoch [19/20], Train Loss: 1.1268, Train Accuracy: 65.47%, Val Loss: 1.2076, Val Accuracy: 69.07%\n",
      "   Epoch [20/20], Train Loss: 1.1296, Train Accuracy: 65.21%, Val Loss: 1.2011, Val Accuracy: 66.57%\n"
     ]
    }
   ],
   "source": [
    "tranv_hist = engine_tranv.train_validation(train_dl_tranv, valid_dl_tranv, epochs=20, save_path='./model_checkpoint/run2_simpleCNN_tranv_chkpt.pth')\n",
    "\n",
    "np.save('./model_checkpoint/run2_tranv_history.npy', np.array(tranv_hist))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_history(tranv_hist)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Angular Velocity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "criterion = nn.CrossEntropyLoss(weight=class_weights)\n",
    "model_ang = SimpleCNN()\n",
    "optimizer_ang = optim.Adam(model_ang.parameters(), lr=0.00013662142756319453, weight_decay=1e-4)\n",
    "scheduler_ang = optim.lr_scheduler.ExponentialLR(optimizer_ang, gamma=0.4740572036017916)\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "\n",
    "model_ang = model_ang.to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "engine_ang = simpleCNN_engine(model_ang, optimizer_ang, scheduler_ang, criterion, device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   Epoch [1/20], Train Loss: 1.5027, Train Accuracy: 54.33%, Val Loss: 1.3446, Val Accuracy: 67.18%\n",
      "   Epoch [2/20], Train Loss: 1.3517, Train Accuracy: 58.30%, Val Loss: 1.2751, Val Accuracy: 69.12%\n",
      "   Epoch [3/20], Train Loss: 1.2339, Train Accuracy: 60.66%, Val Loss: 1.2363, Val Accuracy: 57.67%\n",
      "   Epoch [4/20], Train Loss: 1.1628, Train Accuracy: 62.59%, Val Loss: 1.2235, Val Accuracy: 63.56%\n",
      "   Epoch [5/20], Train Loss: 1.1432, Train Accuracy: 63.07%, Val Loss: 1.2136, Val Accuracy: 63.40%\n",
      "   Epoch [6/20], Train Loss: 1.1287, Train Accuracy: 63.04%, Val Loss: 1.2025, Val Accuracy: 66.25%\n",
      "   Epoch [7/20], Train Loss: 1.1232, Train Accuracy: 63.28%, Val Loss: 1.1972, Val Accuracy: 61.84%\n",
      "   Epoch [8/20], Train Loss: 1.1144, Train Accuracy: 63.63%, Val Loss: 1.2024, Val Accuracy: 62.72%\n",
      "   Epoch [9/20], Train Loss: 1.1186, Train Accuracy: 63.63%, Val Loss: 1.1975, Val Accuracy: 63.81%\n",
      "   Epoch [10/20], Train Loss: 1.1182, Train Accuracy: 63.57%, Val Loss: 1.1920, Val Accuracy: 61.86%\n",
      "   Epoch [11/20], Train Loss: 1.1230, Train Accuracy: 63.52%, Val Loss: 1.2049, Val Accuracy: 63.07%\n",
      "   Epoch [12/20], Train Loss: 1.1168, Train Accuracy: 63.61%, Val Loss: 1.2008, Val Accuracy: 61.32%\n",
      "   Epoch [13/20], Train Loss: 1.1118, Train Accuracy: 63.64%, Val Loss: 1.1865, Val Accuracy: 62.92%\n",
      "   Epoch [14/20], Train Loss: 1.1173, Train Accuracy: 63.69%, Val Loss: 1.1960, Val Accuracy: 63.45%\n",
      "   Epoch [15/20], Train Loss: 1.1269, Train Accuracy: 63.32%, Val Loss: 1.1983, Val Accuracy: 61.36%\n",
      "   Epoch [16/20], Train Loss: 1.1146, Train Accuracy: 63.50%, Val Loss: 1.2095, Val Accuracy: 61.64%\n",
      "   Epoch [17/20], Train Loss: 1.1179, Train Accuracy: 63.40%, Val Loss: 1.1988, Val Accuracy: 62.64%\n",
      "   Epoch [18/20], Train Loss: 1.1154, Train Accuracy: 63.66%, Val Loss: 1.2095, Val Accuracy: 64.37%\n",
      "   Epoch [19/20], Train Loss: 1.1062, Train Accuracy: 63.66%, Val Loss: 1.1977, Val Accuracy: 61.07%\n",
      "   Epoch [20/20], Train Loss: 1.1188, Train Accuracy: 63.42%, Val Loss: 1.2035, Val Accuracy: 65.07%\n"
     ]
    }
   ],
   "source": [
    "ang_hist = engine_ang.train_validation(train_dl_angvel, valid_dl_angvel, epochs=20, save_path='./model_checkpoint/run2_simpleCNN_angvel_chkpt.pth')\n",
    "\n",
    "np.save('./model_checkpoint/run2_angvel_history.npy', np.array(ang_hist))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA/gAAAF5CAYAAAArqIHDAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAADYqklEQVR4nOzdd3hUVfrA8e+k9wAJaRBCKAFCBELvgoAUKaKiNEEUUFdkBVeB3VVB/YmICriiLKiggqCIIKwCSi8CkS6EGkILISGkB1Ln/v64mQlDejItyft5nnlm5s6dM+9Myp33nnPeo1EURUEIIYQQQgghhBBVmo2lAxBCCCGEEEIIIUTlSYIvhBBCCCGEEEJUA5LgCyGEEEIIIYQQ1YAk+EIIIYQQQgghRDUgCb4QQgghhBBCCFENSIIvhBBCCCGEEEJUA5LgCyGEEEIIIYQQ1YAk+EIIIYQQQgghRDVgZ+kAqhKtVsuNGzdwd3dHo9FYOhwhhBACRVFIS0sjICAAGxs5b19ZcqwXQghhbcpzrJcEvxxu3LhBYGCgpcMQQgghCrl27Rr169e3dBhVnhzrhRBCWKuyHOslwS8Hd3d3QP1gPTw8LByNEEIIAampqQQGBuqPUaJy5FgvhBDC2pTnWC8Jfjnohup5eHjIQV8IIYRVkeHkxiHHeiGEENaqLMd6mawnhBBCCCGEEEJUA5LgCyGEEEIIIYQQ1YAk+EIIIYQQQgghRDUgCb4QQgghhBBCCFENSIIvhBBCCCGEEEJUA5LgCyGEEEIIIYQQ1YAsk2cBeVqFiOhE4tMy8XF3omNwHWxtZHkjYb0URSEnJwetVmvpUISo1mxtbbG3t7d0GKIqURTIywY7R0tHIqqwnJwc8vLyLB2GEDWGKY/3kuCb2ZZTsczZFElsSqZ+m7+nE28NCWVAmL8FIxOisOzsbOLj47lz544c+IUwE0dHR7y9vWUNdlEyRYGo7bDjXUiJgck7wbO+paMSVUxqaioJCQlkZWVZOhQhahxTHe+tMsHfs2cP8+fP58iRI8TGxrJ+/XoeffTRYvfftWsXvXv3LrQ9NjYWPz8//f3Fixczf/58bt68SevWrfnPf/5Dx44dTfEWirTlVCwvrjyKct/2mymZvLjyKJ+PbStJvrAad+7c4dq1a9ja2lK7dm2cnZ2xtbVFo5HRJkKYgm6kTEpKCjExMQCS5IvC7k3sbxxDnW2phYwESfBFuaSmphITE4Obmxve3t7Y29vLMV4IMzD18d4qE/yMjAxat27Ns88+y2OPPVbm5507d87gw/Hx8dHf/v7775k+fTpLliyhU6dOLFy4kP79+3Pu3DmD/UwlT6swZ1NkoeQeQAE0wJxNkfQL9ZPh+sIqJCQkYG9vT1BQELa2tpYOR4gawdnZGXd3d65fv05CQoIk+KLA/Ym9Rvd/WaZOiYpJSEjAzc2N+vXrS2IvhJmZ8nhvlUX2Bg4cyLvvvsvw4cPL9TwfHx/8/Pz0Fxubgrf38ccfM2nSJCZMmEBoaChLlizBxcWFr776ytjhFykiOtFgWP79FCA2JZOI6ESzxCNESXJzc8nIyKBOnTqS3AthZhqNBk9PT7KyssjJybF0OMIaRO2EZb1h5eMQe1Ldpsi0KVFxOTk5ZGVl4enpKcm9EBZiquO9VSb4FdWmTRv8/f3p168f+/fv12/Pzs7myJEj9O3bV7/NxsaGvn37cuDAAbPEFp9WfHJfkf2EMKXc3FxAnRskhDA/XeEdqX0hANg8I384PpLYC6PQ/W+Rop5CWJYpjvfVIsH39/dnyZIlrFu3jnXr1hEYGEivXr04evQooA5BysvLw9fX1+B5vr6+3Lx5s9h2s7KySE1NNbhUlI+7k1H3E8Ic5Ky+EJYhf3vCwMB5UKeRpaMQ1ZD8rxHCskzxN2iVc/DLq1mzZjRr1kx/v2vXrkRFRbFgwQK+/fbbCrc7d+5c5syZY4wQ6RhcB39PJ26mZBY5D18D+HmqS+YJIYQQQug17E6hPhmNDSgy/14IIYShatGDX5SOHTty8eJFALy9vbG1tSUuLs5gn7i4OIMq+/ebNWsWKSkp+su1a9cqHI+tjYa3hoQCajJflLeGhEqBPSGEEEIYOvo1JF4EFy9wD1C36Xv0q+1XOSGEEBVQbY8Kx48fx99fXXLOwcGBdu3asX37dv3jWq2W7du306VLl2LbcHR0xMPDw+BSGQPC/Pl8bFv8PA2H4dd2sZcl8oQQQghRWFYa7Hpfvf3gTPBuqt7u+TqMXQcBrcHNB1zrWi5GIYQQVsMqE/z09HSOHz/O8ePHAYiOjub48eNcvXoVUHvWx40bp99/4cKF/Pzzz1y8eJFTp07xyiuvsGPHDl566SX9PtOnT2fZsmV8/fXXnDlzhhdffJGMjAwmTJhg1vc2IMyffTMeYvWkznRprA7HH9kxUJJ7IayYRqMp16Vhw4ZGj6Fhw4Ymnyu5a9cuNBoNzzzzjElfRwhRDvsXQcYtqNMY2k8At/x6Qhnx0KQvTNoJr5wCz3qWjVOIKqqmHONFzWGVc/APHz5M79699fenT58OwPjx41mxYgWxsbH6ZB/UKvmvvvoqMTExuLi40KpVK7Zt22bQxlNPPcWtW7d48803uXnzJm3atGHLli2FCu+Zg62Nhi6NvbgY78+BqETOxKaZPQYhRNmNHz++0LZ9+/YRFRVF69atadOmjcFj3t7eZopMCFGtpcXBH5+qt/vNAVt7tbceID1/2qFGA3ay4okQFSXHeFHdWGWC36tXLxSlqFJ0qhUrVhjcf/3113n99ddLbXfKlClMmTKlsuEZTWiAJwCnb1S8Or8QwvTu/58D8MwzzxAVFcWjjz7K7NmzTR7D9u3bZU10IWoaNx8Y/jmc2wzNBxdsA0iPt1xcQlQjcowX1Y1VJvg1RQt/dzQauJWWRXxapiyRJ2q0PK1CRHSi/m+hY3AdKTp5j8aNG1s6BCGEuWk00HK4etHRDdFPjyv6OUJYITnGl0yO8cKYrHIOfk3h4mBHsLcrAJHSiy9qsC2nYuk+bwejlh3k72uOM2rZQbrP28GWU7GWDq3cVqxYgUajYfbs2Zw/f56RI0fi6+uLjY0NGzZsAODixYvMnj2bLl264Ofnh4ODA/Xr12fcuHGcP3++yHaLmp93+fJlNBoNvXr14u7du8ycOZOgoCAcHR1p0qQJ8+bNK3E0VHnk5ubyn//8h3bt2uHm5oabmxsdO3bk888/Jy8vr9D+6enpzJ07l9atW+Pp6YmbmxuNGzdmxIgRbN261WDfW7duMXPmTEJDQ3Fzc8PT05OQkBDGjRtHRESEUeIXosrJvlP0dq+mam9+UHfzxiNEBckx3jqO8ZmZmXz55ZcMGzaMRo0a4ezsTK1atejZsydr1qwp9nmKorB69Wr69euHl5cXTk5ONGzYkCeffNKggLnOtWvXmDp1KiEhITg7O1OnTh3at2/PnDlzSE2VfMccpAffwloGeHLpVgaRsan0auZj6XCEMLstp2J5ceVR7j9E3UzJ5MWVR6vsChPnzp2jQ4cOeHl50bt3b5KSkrC3twfgiy++4IMPPiAsLIwOHTrg6OhIZGQk3377LT///DN79+6lVatWZX6t7OxsHn74YSIjI+nVqxcZGRns3r2bmTNnkpaWxrvvvlup95KXl8ewYcP49ddf8fDwoF+/fiiKwo4dO/jb3/7G77//zo8//oiNjY1+/759+3Lo0CG8vb3p1asXTk5OXL9+nV9//RVXV1f69+8PQFpaGp06dSI6OprAwED69euHnZ0dV69eZc2aNTRq1IiOHTtWKn4hqpyoHbBuEvR5A9o9Y/hY/XYwcpVFwhKivOQYbz3H+MuXLzNx4kQCAgJo1qwZHTt25ObNm/zxxx/s3buXs2fPFpqOkJeXx6hRo1i7di0ODg5069YNX19frl27xi+//EJ2djZ9+vTR7793716GDh1KcnIyDRs2ZMiQIdy9e1ff9rBhwwrVNBDGJwm+hYX6e7DpxA2Zhy+qFEVRuJtTuNe2vPK0Cm9tPF3owA+gABpg9sZIujXxrtRQPmd7W7NXp12zZg1Tpkxh4cKF2NraGjz26KOP8vzzzxMcHGywffny5Tz77LP6lUDK6sCBAzz44INER0frl/M8fPgwnTt3ZsGCBcycORM3N7cKv5eFCxfy66+/0rJlS7Zv364vThobG0vv3r1Zv349n332mb7GyZ49ezh06BAdOnRgz549ODkVTD9KTU3lwoUL+vs//vgj0dHRDB06lPXr1+tPEoDasx8XJ8OQRQ2jzYPf3oQ7CRB/1tLRiBqoqh3jwfzH+ap4jK9bty6///47ffr0MfisoqOjeeihh3jnnXd45plnDFYJmDt3LmvXriU0NJT//e9/Bu8pJSVFv+IZQGJiIo8//jjJycnMnz+f6dOnGxzTDxw4QEBAQJnft6g4SfAtrGWA+od6RhJ8UYXczckj9M2tpe9YSQpwMzWTB2b/Vql2It/uj4uDef/d1a1bl3nz5hU68AN07ty5yOdMmDCBL7/8kl27dpGSkoKnp2eZXsvGxob//ve/+gM/QPv27Rk4cCD/+9//OHz4ML169arQ+wD45JNPAPj4448NVh7x9/dn/vz5DB06lEWLFukT/Fu3bgHQrVs3g+QewMPDg3bt2unv6/Z96KGHDL4IgPoZ1q0ra3uLGubk9xD3Fzh6woPFFBBWFMhMAQdXtbK+EEZU1Y7xYP7jfFU8xnt5edG3b99C24ODg/nXv/7FpEmT2LRpEy+//DKgjhz46KOPAPjqq68KnbDw9PTkwQcf1N//4osvuHXrFgMGDOAf//hHodfp0qVLmd6vqDxJ8C0sND/Bj76dQUZWLq6O8iMRojro27cvLi4uxT6enp7Opk2bOH78OImJifrqubGxsSiKQlRUFG3bti3TawUFBdGsWbNC20NCQvRtVtTVq1e5evUqdevW5eGHHy70+ODBg6lVqxYXL17k5s2b+Pn50aZNG2xsbFi+fDmhoaE89thjeHl5Fdm+LtmfP38+vr6+PPLII7i7u1c4XiGqtJy7sCN/uG2P6eBSp+j9FrWG5CsweTcEtDFbeEIIVVU+xu/bt49du3YRExNDZmYmiqLo27h3hN3hw4dJTk6mdevWdOrUqdR2t23bBsDzzz9frniE8Uk2aWHebo74ejgSl5rFmdhU2jcs5mAuhBVxtrcl8u3+lW4nIjqRZ5b/Wep+KyZ0oGNwxf82nO0Ln2E3tQYNGhT72I4dOxg5cqS+97ooaWlpZX6t+vXrF7ldlyhnZWWVua373bhxA1C/YBRFo9EQFBREcnIyMTEx+Pn5ERISwgcffMCsWbOYPHkyL7zwAmFhYfTp04dnnnnGYO5hnz59mDZtGgsXLmTUqFHY2dnRtm1b+vXrx7PPPkujRo0qHLsQVc7BzyA1BjwDodMLxe/nlN/zJ0vlCROoasd4MP9xvioe41NSUnjsscdKnB5wb1zXrl0Dyl7hv7z7C9ORKvpWINRf7cWPjJVh+qJq0Gg0uDjYVfrSo2ld/D2dKG7WnAbw93SiR9O6lXodc8+/BwoNTddJT0/nySefJCEhgTfffJPIyEgyMjLQarUoisKoUaMAylUZ9/6h7eZW1Of76quvEhUVxSeffMIjjzzC1atXWbBgAW3atGHRokUG+3788cecOXOGefPm0bt3b06dOsX//d//0bx5c9atW2eutyGEZWUkwN4F6u2H3gD7EpbOlaXyhAlVtWO8JY7zVfEYP2PGDHbs2MGDDz7Irl27SEhIIDc3F0VR9KvbGGvlHWFZkuBbgZYB6pn40zGS4IuaxdZGw1tDQgEKfQHQ3X9rSGi1Wit379693L59m8cff5w5c+bQokULXFxc9F9OLl26ZOEIDekK4ly5cqXYfXSP1atXz2B7YGAgL7/8Mhs3buTWrVt8++232Nra8vrrr5OUlGSwb7NmzXj99df57bffuH37NvPnzycnJ4cXX3zRyO9ICCt1cRtkp4N/a3hgRMn7SoIvqgA5xlvXMX79+vXY2tqyceNGHnzwQby8vPQ1BIqKKzAwEICoqKgytV/e/YXpSIJvBXSF9qQHX9REA8L8+XxsW/w8Dc+G+3k6Vdnlc0qiS2yLGnJ38eJFjh49au6QStSgQQMaNGjArVu3ilzv9pdffiEpKYkmTZrg5+dXbDt2dnaMHTuWDh06kJ2dbTDP735OTk784x//wN/fn1u3bhEfL8OQRYGYmBjGjh2Ll5cXzs7OPPDAAxw+fFj/eHp6OlOmTKF+/fo4OzsTGhrKkiVLLBhxGbUeCc/vhsELobQeO7f8ZXVliL6wcnKML2DpY3xSUhIeHh4Gxfp0fvjhh0Lb2rVrR61atThx4gQRERGltq8r4Ld06dLKBysqRRJ8K6ArtHfuZho5eVoLRyOE+Q0I82ffjIdYPakzi0a2YfWkzuyb8VC1O/BDQVGcn376yWB+XnJyMs8995y+EI810VXUnT59ukHMN2/e5LXXXgPg73//u377zp072bZtG1qt4f+z6Ohozpw5g0aj0X/52bBhAwcPHiz0mkeOHCEuLg43Nzdq1apl7LckqqikpCS6deuGvb09mzdvJjIyko8++ojatWvr95k+fTpbtmxh5cqVnDlzhldeeYUpU6awceNGC0ZeRv6toV4ZCm9JD76oQuQYbx3H+JCQEJKSkvj+++8Nti9YsICdO3cW2t/R0ZFp06YB8NxzzxUayZeSksLu3bv19ydOnIi3tzebN29m4cKFhYb7Hzx40OCEfUREBM2bN6dPnz6Vfm/CkBTZswKBtV1wd7QjLSuXqFvpNPcrfGZNiOrO1kZDl8ZFV1qvTtq3b0+/fv34/fffCQkJ0S9ts2vXLry9vRk2bBg///yzZYO8z7Rp09ixYwebN2+madOmPPTQQyiKwvbt20lLS+PRRx/lb3/7m37/EydOMG3aNOrWrUu7du3w8vLi1q1b7N69m6ysLF5++WX90P9du3axaNEi6tWrR3h4OB4eHty4cYO9e/ei1WqZM2cODg4OlnrrwsrMmzePwMBAli9frt92/9JNf/zxB+PHj9f/bU2ePJn//ve/REREMHToUHOGWzZxkWDnCF7lKEyl68HPKL6IlxDWRI7xlj/Gz5o1i7FjxzJy5EgWL15M/fr1OXHiBGfPnmXatGksWLCg0HP++c9/cuzYMTZs2EBISAg9evTAx8eHa9eucfToUfr166dfKq9OnTqsXbuWoUOHMm3aND755BM6dOjA3bt3OXPmDBcvXuTYsWP4+Kj/v+7cucO5c+fIzMw06+dQE0gPvhWwsdHQIr/QnszDF6L6+/nnn/nXv/5F3bp12bx5M0eOHGHkyJEcPHjQKnurdXP2Fi1aRKNGjdi6dSu//fYbzZo1Y/Hixfz4448GRYAGDx7Mv//9b0JCQjhx4gRr167l9OnTdO/enXXr1hkU2XvmmWd49dVXCQgIICIignXr1hEdHc2gQYPYtm0b06dPt8RbFlZq48aNtG/fnhEjRuDj40N4eDjLli0z2Kdr165s3LiRmJgYFEVh586dnD9/vshlHkGtQJ2ammpwMRtFgU1/h8Wd4NRPZX+eVxNoPhiCe5ouNiFEhVjrMX7MmDH88ssvdO7cmePHj7N582YCAgLYsWNHsSc/7ezsWLduHStWrKBz584cPnyYn376ievXrzN48GBeeeUVg/179erFiRMneOGFF1AUhQ0bNrB//348PT15++23pcK+mWgUKZdYZqmpqXh6epKSklLk/JXKmL3xNCv+uMxz3YN5Y3CoUdsWorwyMzOJjo4mODi42EqxQgjTKc/foCmPTdZG91lMnz6dESNG8Oeff/L3v/+dJUuWMH78eEBN2CdPnsw333yDnZ0dNjY2LFu2jHHjxhXZ5uzZs5kzZ06h7Wb5PE9vgLXjwd4Vph4F9+LrWAhhTHKcF8I6lPVvsTzHehmibyV08/BP30ixcCRCCCGEddJqtbRv35733nsPgPDwcE6dOmWQ4P/nP//h4MGDbNy4kaCgIPbs2cNLL71EQECAvgjUvWbNmmUwUiQ1NVVfDdqkcrNh22z1dteXJbkXQghhFJLgWwl9Jf0bqSiKYpF1u4UQQghr5u/vT2io4Si3Fi1asG7dOgDu3r3LP//5T9avX88jjzwCQKtWrTh+/DgffvhhkQm+o6Mjjo6Opg/+foe/hKRotWBe15fL/3xFgbtJ4OAGdlKnQgghhErm4FuJpj7u2NtqSM3M5XrSXUuHI4QQQlidbt26ce7cOYNt58+fJygoCICcnBxycnIMakKAWkfi/lUdLOpuMuyep97uNQsc3crfxqft4YNgiD1uzMiEEEJUcZLgWwkHOxua+rgDcPqGFNoTQggh7jdt2jQOHjzIe++9x8WLF/nuu+9YunQpL730EgAeHh48+OCDvPbaa+zatYvo6GhWrFjBN998w/Dhwy0c/T32faz2vtdtDuFPV6wN5/ylAWWpPCGEEPeQBN+K6Ifpx0qCL4QQQtyvQ4cOrF+/ntWrVxMWFsY777zDwoULGTNmjH6fNWvW0KFDB8aMGUNoaCjvv/8+//d//8cLL7xgwcjv4+KlDq3v9zbYVnC2pGv+UnmS4AshhLiHzMG3IqEBHnAEIqXQnhBCCFGkwYMHM3jw4GIf9/PzY/ny5WaMqAK6/V3tudf1wleEmy7BjzdOTEIIIaoFSfCtSMsAT0AttCeEEEKIasylTuWe7+arXkuCL4QQ4h4yRN+KtPBX5+DfSMkkKSPbwtEIIYQQwmgUBX75B1zaZZz2pAdfCCFEESTBtyLuTvYEebkAMg9fCCGEqFYu/AZ/LoNVTxonKdf34MscfCGEEAUkwbcyof5qob3TMg9fCCGEqB7ycuH3N9XbHSeDk2fl26zTCFoMgcYPVb4tIYQQ1YYk+FZGV0lflsoTQgghqonjK+HWWbVyfvRuWBAGKdcr16ZvKDy1Eh76l3FiFEIIUS1YZYK/Z88ehgwZQkBAABqNhg0bNpT5ufv378fOzo42bdoYbJ89ezYajcbg0rx5c+MGbgShuqXyJMEXQgghqr7MNPgtv/c+Ox1unoKMeMhIsGxcQgghqiWrTPAzMjJo3bo1ixcvLtfzkpOTGTduHH369Cny8ZYtWxIbG6u/7Nu3zxjhGpWukn7UrXTuZudZOBohhBBCVIiiwMVtsLg9ZN077U5r3Ne4kwi5WcZrUwghRJVmlQn+wIEDeffddxk+fHi5nvfCCy8wevRounTpUuTjdnZ2+Pn56S/e3t7GCNeofNwd8XZzQKvAubg0S4cjhABGjx6NRqPhnXfeKXXfiIgINBoNvr6+5Obmlvu1nnnmGTQaDbt27TLY3qtXLzQaDZcvXy5zWytWrECj0TB79uxyx1FeFYnPHMz5GQihF7UTlvWGlY9D2k3Tvc7nXeGDYLgWYbrXEKKak2N86az1GC+KZpUJfkUsX76cS5cu8dZbbxW7z4ULFwgICKBRo0aMGTOGq1evlthmVlYWqampBhdT02g0tJBCe0JYlaeffhqAVatWlbrvypUrARg1ahR2dnYmjcucNBoNDRs2tHQYQlQNm2fAjWOmfx3nOuq1VNIXosLkGC/H+OqmWiT4Fy5cYObMmaxcubLYP7ZOnTqxYsUKtmzZwueff050dDQ9evQgLa34XvK5c+fi6empvwQGBprqLRjQDdOXefhCWIeHH34YX19fzp07x59//lnsfrm5uXz//fdAwRcGY/nmm284c+YM9erVM2q7xmLt8QlhVgPnQUC4eltja7rXcfNRrzNume41hKjm5BhfOmuPTxiq8gl+Xl4eo0ePZs6cOYSEhBS738CBAxkxYgStWrWif//+/PrrryQnJ/PDDz8U+5xZs2aRkpKiv1y7ds0Ub6GQUKmkL4RVsbW1ZdSoUUDB2fui/Pbbb8THx9OiRQvatWtn1BgaNGhA8+bNsbe3N2q7xmLt8QlhVo17w6SdMHYd+LdSt5ki0dcl+NKDL0SFyTG+dNYenzBU5RP8tLQ0Dh8+zJQpU7Czs8POzo63336bEydOYGdnx44dO4p8Xq1atQgJCeHixYvFtu3o6IiHh4fBxRx0S+WdvZlKnlYxy2sKYZUUxWqKR40dOxaA77//nry8ogtg6ob36fZNTk7mP//5D/379ycoKAhHR0e8vLwYMGAAv//+e7lev6T5b/v376dv3764u7tTq1Yt+vfvz6FDh4pt6+LFi8yePZsuXbrg5+eHg4MD9evXZ9y4cZw/f95gX90cP4ArV64YrETSq1evMsUXGRnJmDFj8Pf3x8HBgXr16jFu3DjOnTtXaN9du3ah0Wh45plnSExM5MUXX8Tf3x9HR0fCwsL46quvyvaBlcGdO3d45513CAsLw9nZGU9PT3r27MmaNWuK3P/WrVvMnDmT0NBQ3Nzc8PT0JCQkhHHjxhERYTgH+sqVK7z44ouEhITg4uJCnTp1aNmyJc8//3yR71tUQxoNNOlbONE35lcvfYIfb7w2hTAXOcbryTG+Ysf4iv4McnJyWLJkCd27d6dWrVo4OzvTpEkTJkyYwJEjRwrtf+bMGZ577jkaNmyIo6MjPj4+dOvWjQ8//LBCtRhMrcpPHvHw8OCvv/4y2PbZZ5+xY8cOfvzxR4KDg4t8Xnp6OlFRUUYfYmMMDb1ccba35W5OHtEJ6TTxcbd0SEKYl6JA1HbY8S6kxMDkneBZ36IhtWvXjhYtWnDmzBl+//13BgwYYPB4RkYGP//8MxqNhjFjxgBw8OBBpk6dSsOGDWnWrBldunTh6tWr/Pbbb/z222988cUXPPvss5WK63//+x/Dhw8nNzeXjh070qhRI06cOEHPnj155plninzOF198wQcffEBYWBgdOnTA0dGRyMhIvv32W37++Wf27t1Lq1ZqMtKkSRPGjx/P119/jaurK0888YS+nbIsNbp9+3aGDBnC3bt3CQ8Pp1evXpw9e5Zvv/2W9evX8+uvv9KjR49Cz0tOTqZLly6kp6fTo0cPEhIS2LNnD8899xxarZaJEydW7APLl5aWRu/evTly5Ah169Zl8ODBZGRksGPHDvbu3cuBAwdYtGiRwf6dOnUiOjqawMBA+vXrh52dHVevXmXNmjU0atSIjh07AnDt2jXatm1LYmIiTZs2ZdCgQeTl5XHlyhWWLVtGly5daNasWaXiF1WILtFv3Kfg/1pqDLjWrXzbbr7qtfTgi6pEjvFlJsf4klXkZ5CRkcGgQYPYs2cPrq6u+iT/8uXLrFq1Ck9PT4MRGmvXruXpp58mKyuLFi1aMHz4cFJSUjh9+jSvvfYaEydOpFatWmWK12wUK5SWlqYcO3ZMOXbsmAIoH3/8sXLs2DHlypUriqIoysyZM5Wnn3662Oe/9dZbSuvWrQ22vfrqq8quXbuU6OhoZf/+/Urfvn0Vb29vJT4+vsxxpaSkKICSkpJSofdVHsMX71OCZvxP2XDsuslfS4j73b17V4mMjFTu3r1r3hfWahXlwu+K8t8HFeUtD0V5q5Z6HXPMvHEU47333lMAZcyYMYUe++abbxRAefDBB/XbLl26pBw4cKDQvkePHlVq1aqleHh4KGlpaQaPjR8/XgGUnTt3Gmx/8MEHFUCJjo7Wb0tNTVXq1q2rAMpXX32l367VapUZM2YogAIob731lkFbBw4cUC5dulQorq+++koBlN69exd6DFCCgoIKbS8pvvT0dMXX11cBlE8//dRg/48//lgBlPr16xv8nu3cuVMf98iRI5XMzEz9Y+vXr1cApUGDBsXGcb/ly5cX+RlMmTJF/15TU1P128+cOaP4+PgogLJp0yb9dt1nM3ToUCUvL8+grfj4eOWvv/7S33/zzTcVQJkyZUqheK5cuaJcvHix1LjL8zdozmNTTWDyz1OrVZSczNL3K4vzv6n/Iz/vZpz2RI1hkeO8HOPlGG/kY3xFfgbPPfecAig9e/YslAfevHlTOXjwoP7++fPnFScnJ8XOzk5ZtWqVwb5arVbZunWrwXuoiLL+LZbn2GSVQ/QPHz5MeHg44eFqgZrp06cTHh7Om2++CUBsbGypFfDvd/36dUaNGkWzZs148skn8fLy4uDBg9Sta4Qz6Cagm4cvhfaEVcvOKP6Sk1n2fbPvqOtF65aVij2Z/6T89aJz797T7t372r1TcrtGNGbMGDQaDRs2bCAjI8PgMd28Pd3QPYDg4GA6d+5cqJ3w8HBeeuklUlNT2blzZ4Xj+fHHH7l16xY9e/ZkwoQJ+u265X7q1y+6R6Rz585Fjm6aMGEC3bp1Y9euXaSkVH4Vjx9++IG4uDi6dOnCSy+9ZPDYtGnTaNeuHdevX2fdunWFnuvh4cGnn36Ko6Ojftujjz5KWFgYV69erdRSPRkZGXz55ZfY2Njw2Wef4e5eMEqqefPm/Pvf/wYw6MG/dUstYvbQQw9hY2N46Kxbty5hYWGF9u3bt2+h127QoAGNGzeucOyiGtBowM6x9P3KonYwtBgKTfsbpz0h7iXHeD05xhdmDcf48v4Mbty4wYoVK3B0dOSbb74plAf6+vrSqVMn/f0FCxaQmZnJxIkTGT16tMG+Go2Ghx9+2OA9WAurHKLfq1cvFKX4uecrVqwo8fmzZ88utCZkcXMqrZWukr4U2hNW7b2A4h9r+jCMWVtwf34TyCnmYOzgBtnpBUWolPvmv311z1C5gHCYvKvg/uJOkFLMCb+6zeGl4ueplVeDBg3o2bMnu3fvZsOGDfphenFxcWzfvh0nJydGjBhh8Jy8vDy2b9/OH3/8QWxsLFlZ6nzDCxcuGFxXxN69ewEYOXJkocfs7e154oknWLhwYZHPTU9PZ9OmTRw/fpzExERycnIA9QSqoihERUXRtm3bCsd2b3y6z+l+Y8eO5ciRI+zdu7fQPu3atcPLy6vQc0JCQjh16hSxsbEVXtLnyJEj3L17l/bt2xc5BPHpp59m6tSp7N+/H61Wi42NjX643vz58/H19eWRRx4xODFwf+wA//znP7G1taVv3744OTlVKFYhSuTdBJ761tJRiOpKjvGAHONLi8/Sx/jy/Ax27dpFXl4egwcPJigoqNS2t23bBsDzzz9fplishVUm+KKg0F5kbCqKougLYAhRLWWnq9f3H/St0NNPP83u3btZuXKl/oC1evVq8vLyeOyxx/D09NTve/36dQYPHsyJEyeKba+kpTpLc+PGDYBiD1LFHRx37NjByJEj9T3Nxo7r/viKi0O3PSYmptBjxfVM6JJq3QHcFHHVqlULT09PUlJSSEpKwsvLiz59+jBt2jQWLlyoX/+4bdu29OvXj2effZZGjRrpn//MM8/w22+/8cMPPzBkyBCcnJzo0KEDAwYM4Nlnn8XPz6/CsQshRJUjx/gKkWN86cr7M9CtiFbWkXTl3d9aSIJvpUJ83bG10ZCYkc3N1Ez8PZ0tHZIQhf3zRvGP3b8k1GvFrFgRvQd2vQ+xx9XnFPUF4Nkt4Kdbauq+mUUvHUKdzlVkEMXHV0FPPPEEU6ZMYdu2bcTHx+Pj46Mfund/0c6JEydy4sQJHn/8cV5//XWaNWuGu7s7NjY2LF26lOeff77E0UqmkJ6ezpNPPkliYiJvvvkmI0eOJCgoCGdnZzQaDaNHj2b16tVmiaukE5f3D4M3t6Ji+/jjj3n++ef5+eef2bZtG/v37yciIoIPPviA1atX8/jjjwPqkkvff/89M2fO5Oeff2bHjh0cOnSIvXv38v7777Nlyxa6du1q7rckqitFgbtJYO8C9jJSRBiRHOPlGF8J5jjGW+PPwBpY5Rx8AU72tjSp6wbIPHxhxRxci7/c/0WzuP2aDVSH45W0XrSd8z3t3neyy8GlhDhcjP6WPT09GTp0KLm5uaxevZqzZ89y5MgRvL29DaruZmRk8Pvvv+Pr68v3339Px44d8fT01B/ULl26VOlY/P39AXVpm6IUtX3v3r3cvn2bxx9/nDlz5tCiRQtcXFz0B2JjxKUTEBBQYny6OXb16tUz2muWRWlxpaSkkJycjLOzM7Vr1zZ4rFmzZrz++uv89ttv3L59m/nz55OTk8OLL75YqJ3w8HBmz57Nnj17uHXrFtOmTSMtLY1XXnnF6O9J1GD/7QkfBMOV/ZaORFQ3coyXY3wJLH2Mr8jPIDAwEICoqKgyvUZ597cWkuBbMV2hPZmHL6o9c6wXbUS6IjurVq3Sr4v71FNPYW9vr98nJSUFrVaLv78/traGX2ZycnJYv359pePQLT3zww8/FHosNze3yMI2SUlJQNHD4y5evMjRo0eLfC17e/tyr/Wqi2/16tVFPq7rFSlqCR1TateuHc7Ozhw5cqTI+ZG6uLp161ZiL4OTkxP/+Mc/8Pf359atW8THF78WuYeHB3PnzkWj0XDq1KnKvwkhdFzy57GmF//7J4RFyTG+QuQYX7KK/Ax69eqFra0tW7du1Q+/L4muWO7SpUuNE7SZWOdflgDumYcvCb6oKe7/EhDQGtx8jLNetBENGDAAb29v/vzzT5YsWQIUHrrn4+ODp6cnp06dYv/+gp61vLw8ZsyYwfnz5ysdx4gRI/Dy8mLXrl18/fXX+u2KovDWW28VudpISEgIAD/99JPB/Lzk5GSee+45fSGe+wUEBBAXF0dycnKZ43vyySfx9fVl3759hQ6On3zyCYcPH6ZevXr6oe3m4urqyrPPPotWq+Wll14yqJZ8/vx53n33XQCmTp2q375hwwYOHjxYqK0jR44QFxeHm5ubfh3cb7/9tsgkfvPmzSiKou8REMIo3HzV6/Q4y8YhRGnkGF8ucowvWUV+BgEBAYwbN47MzEzGjx/P7du3DR6Pj4/n0KGCwo2vvPIKTk5OLFu2jO+//95gX0VR+P333w3qBaxfv57mzZszbtw4Y73NCpEE34qF+uf34MdWfikLIaqUe78EvHIKPM07hLs09vb2+qq2CQkJNG3a1GBZFQA7Oztef/11cnNzefDBB3n44YcZOXIkTZo0YcmSJYWWlKkId3d3vvzyS2xtbXnmmWfo3Lkzo0ePJiwsjPnz5zNp0qRCz2nfvj39+vXj6tWrhISEMHz4cIYPH05wcDA3btxg2LBhRb6Wbshi27ZtGTt2LBMnTmT+/Pklxufq6sqqVatwdnbm+eefp3379owePZq2bdvy97//HTc3N1avXm2RCvNz586lXbt2/P777zRq1Ignn3ySRx55hNatW3Pz5k2mTp3KkCFD9Pvv2rWLLl26UL9+fYYMGcKYMWPo3bs3nTp1QqvVMmfOHBwcHABYt24dDzzwAE2aNGH48OGMHj2aLl268Nhjj2FjY6M/gSCEUbj5qNcZxRfUEqVQFMiteOFOUU5yjC8TOcaXrKI/g0WLFtG1a1d27txJUFAQgwYNYuTIkXTp0oXAwECDEQkhISEsX74cjUbDyJEjadmyJaNGjWLQoEEEBQXx8MMPc/duwdKOKSkpnDt3rtzLuRubJPhWTDdE/1riXVLuFn3GTYhqzZjrRRvZvWfz710X917//Oc/+frrr2nVqhX79+9n27ZttG7dmoMHD9K+fXujxDFs2DB27txJ7969OXXqFL/88gv+/v7s3r272EJuP//8M//617+oW7cumzdv5siRI4wcOZKDBw/qe6HvN3fuXKZMmUJubi7ff/89X375Jb/88kup8fXp04c///yTUaNGcf36dX788Udu3rzJ2LFjOXz4sNmH5+u4u7uze/du5syZg7e3Nxs3bmTv3r20b9+e7777jkWLFhns/8wzz/Dqq68SEBBAREQE69atIzo6mkGDBrFt2zamT5+u33f69Om89NJLuLu7s3fvXtavX098fDxPPfUUhw4dKrTMkhCVokvwpQe//BSlYH32BWGQct3SEdUscowvlRzjS1aRn4G7uzs7d+5k0aJFtGzZkr1797Jx40Zu3brFmDFjCvW+jxw5ksOHDzN27FhSUlJYt24dR44coUGDBnz00Ue4ubmZ9D1WhEapiaUFKyg1NVW/dJKHh4dZXrPb+zuISb7Lmsmd6dyo8HqRQphCZmYm0dHRBAcHy/rdQlhAef4GLXFsqs6q3Od58gf4aRIE94TxmywdTdWgKBC1HXa8CzeOofZ3aWHybghoY+HgzEOO80JYh7L+LZbn2CQ9+FZOCu0JIYQQolj6Hnwpsleqe3vsVz4OsSfzH9BaNCwhhDAmO0sHIErWMsCD3yPjpNCeEEIIIQqrFQShw6BOI0tHYt2idsL2OWqPvW6ZtqLWZBdCiCpOEnwrpy+0d0MK7QkhhBDiPnWC4clvLB2F9ds8AxLOqbclsRdCVGMyRN/KtaznCcDF+HSycuWAJIQQomaLiYlh7NixeHl54ezszAMPPMDhw4cN9jlz5gxDhw7F09MTV1dXOnToYPGqxsLCBs6DgHD1tka+/gohqi/5D2flAjyd8HS2J1ercCEu3dLhCCGEEBaTlJREt27dsLe3Z/PmzURGRvLRRx9Ru3Zt/T5RUVF0796d5s2bs2vXLk6ePMkbb7xRvQuJabWQcRty7pa+b03VuDf0y1+iUtHNuddYLBwhhDAVGaJv5TQaDS0DPPgj6janb6QQlt+jL4QQQtQ08+bNIzAwkOXLl+u3BQcHG+zzr3/9i0GDBvHBBx/otzVu3NhsMVrEl/0g5jCMWgPNBlo6Gut1YrXhfSdPyExGX0VfCCGqAenBrwJ08/Cl0J4QQoiabOPGjbRv354RI0bg4+NDeHg4y5Yt0z+u1Wr55ZdfCAkJoX///vj4+NCpUyc2bNhQbJtZWVmkpqYaXKocV2/1Oj3OsnFYs6x0OL1evd3nLfXaxg7GrIOA1upqBK51LRefEEIYiST4VUDLerJUnrAMRVEsHYIQNZL87RXt0qVLfP755zRt2pStW7fy4osvMnXqVL7++msA4uPjSU9P5/3332fAgAH89ttvDB8+nMcee4zdu3cX2ebcuXPx9PTUXwIDA835loxDlsor3ZlNkJOhrjbQ+W9g6wh3EqB2EEzaCa+cAs96lo7S7OR/jRCWZYq/QUnwq4BQf3VY/pnYVLRa+UcsTM/OTp29k5WVZeFIhKiZcnJyALC1tbVwJNZFq9XStm1b3nvvPcLDw5k8eTKTJk1iyZIl+scBhg0bxrRp02jTpg0zZ85k8ODB+n3uN2vWLFJSUvSXa9eume39GI2br3otCX7xjq9Sr9uMBnsnqN9BvX9lP2g0YOdoudgsQPe/Rfe/RghhGaY43kuCXwU0ruuKo50NGdl5XEm8Y+lwRA1gZ2eHq6sriYmJ5OXJ6g1CmJOiKKSkpODo6Ii9vb2lw7Eq/v7+hIaGGmxr0aKFvkK+t7c3dnZ2Je5zP0dHRzw8PAwuVY4+wZch+kVKugKX9wIaaDVS3RbUVb2+8ofFwrIke3t7HB0dSUlJkV58ISzEVMd7KbJXBdjZ2tDcz50T11OIvJFKsLerpUMSNYC3tzfXrl0jOjoaT09PnJ2dsbW1RaORqsNCmIKiKOTk5JCSkkJ6ejr16tW84cKl6datG+fOnTPYdv78eYKCggBwcHCgQ4cOJe5TLenmjksPftFuXwAXL/B7AGrlT8HQJfhXD1guLgvz9vYmJiaG69ev4+npib29vRzjhTADUx/vJcGvIkIDPDhxPYXTN1J4pJW/pcMRNYCLiwvBwcHEx8eTlJREQkKCpUMSokZwdHSkXr16VbMn2cSmTZtG165dee+993jyySeJiIhg6dKlLF26VL/Pa6+9xlNPPUXPnj3p3bs3W7ZsYdOmTezatctygZua9OCXrElfmH5WnXOvE9gJxvwIgR0tF5eF6f7HJCQkEBMTY+FohKh5THW8lwS/iggN8ASuERkrhfaE+Tg4OFC/fn39mUbd/FYhhGnY2trKsPwSdOjQgfXr1zNr1izefvttgoODWbhwIWPGjNHvM3z4cJYsWcLcuXOZOnUqzZo1Y926dXTv3t2CkZtYrUAIfRRqNbB0JNbLzgE8AgruO7hA036Wi8dK6Kal5OTkyJQ8IczIlMd7jSITb8osNTUVT09PUlJSzN6zcuRKEo9//gd13R358199zfraQgghrJclj03VkXye1cyt8+DVBGyk7JQQouoqz7FJ/ttVES383dFo4FZaFvFpmZYORwghhBDCumWlw9JesKg1pN4o/HhGAmybDWsnmDsyIYQwGUnwqwgXBzt9cb3IGzJMXwghhBD30GrVhDVbVtvRO7MRcjLA1g7ci6hfZGMH+xbC6Z8gTeoXCCGqB6tM8Pfs2cOQIUMICAhAo9GwYcOGMj93//792NnZ0aZNm0KPLV68mIYNG+Lk5ESnTp2IiIgwXtBm0DLAE4DTkuALIYQQ4l4rHoH5jeHi75aOxHoc/069bjNaXev+fs61wC9MvX21Zi6XJ4Sofqwywc/IyKB169YsXry4XM9LTk5m3Lhx9OnTp9Bj33//PdOnT+ett97i6NGjtG7dmv79+xMfX3WWlAn1V+dbSKE9IYQQQhhw9VKvZak8VdJluLwX0ECrkcXvF9RNvb4iCb4QonqwygR/4MCBvPvuuwwfPrxcz3vhhRcYPXo0Xbp0KfTYxx9/zKRJk5gwYQKhoaEsWbIEFxcXvvrqK2OFbXItA/ITfOnBF0IIIcS9ZKk8QyfWqNeNHlRXGShOUFf1WhJ8IUQ1YZUJfkUsX76cS5cu8dZbbxV6LDs7myNHjtC3b0H1eRsbG/r27cuBAweKbTMrK4vU1FSDiyWF5if4l29nkJ6Va9FYhBBCCGFFJMEvoNXC8VXq7TZjSt63QX6CH3ca7iSaNi4hRM2mKJCbZfKXqRYJ/oULF5g5cyYrV67Ezs6u0OMJCQnk5eXh6+trsN3X15ebN28W2+7cuXPx9PTUXwIDSzgDbAbebo74ejiiKHBWhukLIYQQQsfNR71Ov2XZOKzBtUOQfBUc3KH54JL3dasL3iGAAlcPmiU8IUQNoyhwcRss6w0LwiDluklfrson+Hl5eYwePZo5c+YQEhJi1LZnzZpFSkqK/nLt2jWjtl8RukJ7Mg9fCCGEEHrSg1+gQWd49jcYNB8cXErfP6gruHjBXenBF0IY0b2J/crH4cYJyIhXVzwxocLd3VVMWloahw8f5tixY0yZMgUArVaLoijY2dnx22+/0b17d2xtbYmLMzzoxcXF4efnV2zbjo6OODo6mjT+8gr192DH2XhOx0iCL4QQQoh8rroefCmyh0YDDTqpl7J4+P9g8MKiK+0LIUR5KQpEbYcd78KNY6CxzX9Aa5aXr/IJvoeHB3/99ZfBts8++4wdO3bw448/EhwcjIODA+3atWP79u08+uijgHoSYPv27fqTAlWFrtDe6dgUC0cihBBCCKvhWR9aDgePepaOpOpxdLN0BEKI6iJqJ2yfY5jYK3lmDcEqE/z09HQuXryovx8dHc3x48epU6cODRo0YNasWcTExPDNN99gY2NDWFiYwfN9fHxwcnIy2D59+nTGjx9P+/bt6dixIwsXLiQjI4MJEyaY7X0Zg67Q3vmb6eTkabG3rfKzLIQQQghRWe6+MGKFpaOwvDVj1HoE3aeXXD2/KIoCeTlg52Ca2IQQ1d/mGZBwTr1t5sRexyqzw8OHDxMeHk54eDigJufh4eG8+eabAMTGxnL16tVytfnUU0/x4Ycf8uabb9KmTRuOHz/Oli1bChXes3aBtV1wd7QjO0/Lxfh0S4cjhBBCCGEdki7D2f/B4eXlH27/55fwcSjsnmeS0IQQNcTAeRCg5rAFQ/PNyyp78Hv16oWiKMU+vmLFihKfP3v2bGbPnl1o+5QpU6rckPz72dhoaOHvQcTlRCJvpNLC38PSIQkhhBDCGmjz1KXe7J1r5rDz46vV60a91CkL5WFrD2k34MofRg9LCFGDNO6t/g+6fw6+GXvzrbIHX5RMN0z/9A0ptCeEEEKIfN8Ohw+bwLnNlo7E/LRaOPGdervNmPI/P6ibeh1zGHIyjReXEKLmUbTQsCdM2glj14F/q/wHzJN6S4JfBekS/EgptCeEEEIIHde66nVNXCrvyn5IvgqOHtD8kfI/v04jcPODvGw1yRdCiIq6dgjmNYQNf4MmfQsS/YDWao0Q3f9qE7HKIfqiZLpK+pE3UlEUBY0s6yKEEEIIt/y6QjUxwT+e33vfcjg4uJT/+RoNBHWF0z+pw/QbdjdufEKImiNqJ+RkqCcMQf3/0qQvNO6jbrMz7TLs0oNfBTX1ccfeVkNqZi7Xk+5aOhwhhBBCWAM3H/U6Pd6ycZhbVjpE/qzeDh9b8XaCuqrXV/ZXPiYhRM11aZd63aiX4XaNxuTJPUiCXyU52NnQ1McdkHn4QgghhMhXU3vw87Kh8wsQ3BPqd6h4O7p5+Nci1OXyhBCivDJTIOaIevv+BN9MZIh+FdUywIPI2FQib6QwIMzP0uEIIYQQwtJ0PfgZtywbh7m51IE+b1a+nbrNoWEP8GsF2RngXKvybQohapbL+9SK+V5NoFagRUKQBL+KCg3wgCMQGSs9+EIIIYSg5vbgG4uNDTzzP0tHIYSoCEUxy/z2UhU3PN+MJMGvoloGeAIyRF8IIYQQ+TwCoOVj4O6nftmtCUV4T68HO2e1gJWtfK0VosZRlII151NiYPJO8KxvuXgkwRcV1cJfnYMfm5JJYkY2dVwdLByREEIIISzKpQ6MWG7pKMxHq4Xf3oSUq/DEVxD2uHHazb4D1/9UK+nb2BqnTSGEcd2b2N84hlpaTgsZCZZL8LVaaD0SLu1Wp/tYiBTZq6LcnewJ8lKXgYmUXnwhhBBC1DRX9qnJvaMHNBtknDa1WljQEr4ZCnGnjdOmEMJ4FAUuboNlvWHl4xB7Mv8BrUXDAtRpPj1ehfEbLVrDQxL8KizU3wOAyNgUC0cihBBCCKugzVOXyctKs3Qkpnf8O/U67DGwdzZOmzY2UK+tevvKH8ZpUwhhHFE7Cyf2Sp5lY7JCkuBXYS0D1ARf5uELIYQQAoDvnoIPm0LkRktHYlpZaRD5s3q7zRjjth3UVb2+st+47QohKmfzjPzh+FhfYq8ocOonSLf8KiaS4FdhukJ7MkRfCCGEEAC41lWvq3sl/cifIeeOuhRV/Q7GbTuom3p95Q/1S7sQwjoMnAf+bdTbGiurjxEfCT9OgEWtIS/HoqFIgl+Fheb34EfdSudutpWdxRJCCCFMICYmhrFjx+Ll5YWzszMPPPAAhw8fLnLfF154AY1Gw8KFC80bpCW5+ajX6fGWjcPUdMPz24w2/moBAeFg5wR3EiDhgnHbFkJUXOPe0LQ/eAapJ/esia56foPOYGtv0VAkwa/CfNwd8XZzQKvA2ZvSiy+EEKJ6S0pKolu3btjb27N582YiIyP56KOPqF27dqF9169fz8GDBwkICLBApBbk5qteV+ce/Ow7kJkKGhtoNdL47ds5FowKkGH6QliPhIuwfwGkXIFeM6HLy/c8aOG0VpfgN+5t0TDA4p+EqAyNRkMLfaE9SfCFEEJUb/PmzSMwMJDly5fTsWNHgoODefjhh2ncuLHBfjExMbz88susWrUKe3vL9qSYXU3owXdwgRf3wctHwLOeaV7j3mH6QgjLUxT4ZTrkZUOTvtByOPR8tWCovk9z9f+fbpqSOeVmw+X8k4GNepn/9e8jCX4Vp5uHL4X2hBBCWIOvv/6azMxMk7S9ceNG2rdvz4gRI/Dx8SE8PJxly5YZ7KPVann66ad57bXXaNmypUnisGo1oQdfp04j07XdYggMmAfdp5nuNSxFUSA3y9JRCFE+f/0I0bvV6TODPlSn5jjXhkYPqo+HPQ6vnDLdSb+SXP8TcjLAxRt8LH/ckQS/itPNw5dCe0IIIazBhAkTCAgI4OWXX+bEiRNGbfvSpUt8/vnnNG3alK1bt/Liiy8ydepUvv76a/0+8+bNw87OjqlTp5apzaysLFJTUw0uVZouwc+opj34qbHmWQLQLww6vwC+oaZ/LXO5d/3wBWGQct3SEQlRNneTYOss9XbP16BOcMFjLYaq19F71Ok1lqAbnt+ol7rUpoVZPgJRKbql8s7eTCVPK5VehRBCWNbEiRPJzc1l8eLFtG3bls6dO/Pll1+SkZFR6ba1Wi1t27blvffeIzw8nMmTJzNp0iSWLFkCwJEjR1i0aBErVqxAU8bCa3PnzsXT01N/CQwMrHScFuXup/ZktRkDWq2lozG+7XPgwxA4tsrSkVQd9yb2Kx+HGyfUE0AZCZaOTIiy2f4OZNwC72bQ9b6Tt6HDYNxGGPuTZWIDwwTfCkiCX8U19HLF2d6WzBwt0Qnplg5HCCFEDbd06VJiY2NZunQpHTp0ICIigsmTJxMQEMALL7xQbMX7svD39yc01LBHtUWLFly9ehWAvXv3Eh8fT4MGDbCzs8POzo4rV67w6quv0rBhwyLbnDVrFikpKfrLtWvXKhyfVXDygCe+ggFzraInyaiy0gqWx/MOMf3rpcfDsZXqpSq6P7GPPZn/QDU88WNCeVqFA1G3+fl4DAeibhulQ80UbVZbeTmQcF69PfhjsHMw/PxiFfIa9gRbu8q9TGV+Jk9+A8P/C00frlQMxlK5T0JYnK2Nhhb+7hy9mszpG6k08XG3dEhCCCFqOFdXVyZOnMjEiRM5deoUS5cuZdWqVSxdupRly5bRqlUrnn/+eUaPHo2Hh0eZ2+3WrRvnzp0z2Hb+/HmCgoIAePrpp+nbt6/B4/379+fpp59mwoQJRbbp6OiIo6OFhnWK8tEl915NoX5707/ejePw80tQpzGEjzX966EmGRHRicSnZeLj7kTH4DrY2lRgGcConepohxvHULBBA6DUjCWVjfYZAltOxTJnUySxKQV1Rfw9nXhrSCgDwvytps1qzdYexm+CqwcgqGvJn19Lvwotm1npn4mHP7Q2wYoeFVTNTu3WTFJoTwghhLUKCwvjk08+4caNG6xcuZKePXty4sQJXnrpJQICAnjuuec4cuRImdqaNm0aBw8e5L333uPixYt89913LF26lJdeegkALy8vwsLCDC729vb4+fnRrFkzU75N66LNg7Q4dSk5SzNmQbfj36nXbUZX6Et8uTXoBGggMQrSbpr85baciqX7vB2MWnaQv685zqhlB+k+bwdbTsWWv7HNM+DGMQA0Ru6xt+beZ2N+hltOxfLiyqMGSR/AzZRMXlx51GrarBE0Gn1yX9TnF5dyhxtrXuHu/FD1f185VMefiST41YCu0N4fFxOs8p+tEEIIkZOTQ1paGmlpaoE0RVHIycnRL3n3xBNPkJycXGIbHTp0YP369axevZqwsDDeeecdFi5cyJgxY8zwDqqQ75+Gj0Lgr7WWi8HYBd0SL6lr0mtszNdT5uQJfg+ot028XJ6xk4w/W8zgojbAmCECRj4Jkc9YJwyM+RnmaRXmbIqkqEh02+ZsiiwUa26elvSsXG6lZXEt8Q7n49I4cS2Zg5dus/1MHDN/+qvcbVqCVZzESYmBrf/Sn6gs6WeixYZwm4s437mB9symMr9ERX/OBja+DPs/gbvJZX5dU5Mh+tVAyt0cAE7dSOXva44DMtRHCCGEdTh48CDLli3jhx9+4M6dO9ja2vLYY4/xwgsv0KNHD3766SfmzZvH+vXrcXFx4ZtvvimxvcGDBzN48OAyv/7ly5cr+Q6qIDcf9TrjlvlfW1EgajvseDe/B9kG0KoF3TzrV7zdE2vU60a9wcP4iWuxgrrBzZPqyYWwx0zyEmVJMmb+9BeZOdpCAxcURbefor+v1Sps3HuDzzS3AUhVnPHQ3CVXscFOU9Cbn5CejVtOHo52NmUqSqlLoO+PU5dAfz62bbm/dxpruHppn6EGeGvjaVr4e5CTp5CZk5d/0ZKZk8dd3f1cLZnZeVyISyt0ouD+NmNTMun6/nYUBf3zc/IqnggrKNxOSSMiOpEujb0q3E5lWc0Ugi0z4MwmSL4CT60kIjqxxJ/J5rwOhNtc5I//Lee5jfXQaECDJv8aNBpN/nX+bY3uhEzxU1d0P+difyYpMXD0G/XEY9txlX7LxmKVCf6ePXuYP38+R44cITY2lvXr1/Poo48Wu/++ffuYMWMGZ8+e5c6dOwQFBfH8888zbVrB2qWzZ89mzpw5Bs9r1qwZZ8+eNdXbMIstp2KZt7nwe6jMP1shhBCiMpKSkvj2229ZtmwZkZGRKIpCYGAgM2bMYOLEifj5+en3HTVqFCNGjCA8PJxff/3VglFXI7oEP718Q1Ur5f7EXmOb/4ARhocrSkGC32a0ej8v2zxLYgV1hUOfm7QHPyL6domJC0DynRxe+f54mdprrbnItw5zcdNksS+vJc/l/INONmd51W4trTWX0CoabDQK45dHcFq5hb2tBncne9wc7XB3ssu/tsfdqeC+q6Md/90dVWICPWdTJP1C/co83728JwwURSH1bi63M7JIupPN7fRsEjOyuZ2RzemYlFIT8rjULB6cv6tMsZVVXGrx00+c7W1xdrDF2d4WJ3sbsnK1XE+6W2R0PW1O8qrdWgI0t1n8xzIaevfE39PZqLGWhSlO4kAF6iKc2wJnNqFobNlffyLbNp7m98iS/59t0Xbkn6ymsyYS59wUkjFeXbKPfzvHsPB6tAuqTYive0Hsuur5AW3BuZbBc4xZC6K8rDLBz8jIoHXr1jz77LM89ljpZ0tdXV2ZMmUKrVq1wtXVlX379vH888/j6urK5MmT9fu1bNmSbdu26e/b2Vnl2y+zspytLO8/WyGEEKIyxo4dy08//URWVhYajYaBAwfywgsvMGjQIGyKqepuZ2dHhw4dDNazF5WgT/DjzfN69xR00yf29xd0u3EUXOuqve/lnT+v0cCEX+HE92Dvog77T4mByTsrNyqgLIK6qtfxkXAnEVzqGKVZRVE4fSOVzadiWXu4bNMXmvq4UdddPamh+wjVPknDj/Tl2I/xyL3LQW0LJub8gywc2aNtzdWcuqyxfxcvTRq3FVduK+oUz5w8hcQMNVmu8PtB7els8cYWPJztcHGww8XBFlfH/GsHO1wcbfW3next+Wp/dImjFl75/jht9l8m6U4OtzOySb6TTW4lh4rb2WhwdbTDyd4GJ3s1+Xa0t8XJzgZnB1uc7NRkPDUzhx1nSx8BM2doKO0b1tEn80526nVRoyIORN1m1LKDBu9Ul9i3tik48fJn5AVWnM6lY8M6DGkTwKAwP7zcTH8yy1R5RXlGBNxMyeTwhet02/p3agP/zRnI+5sygNKXWb2q+BKpDSLU5go7Bt8hI3SoGruijo5Qr0GrKPkjXxSOX03mHz+eLKlZAP68ksSfV5IAcHO0o01gLdo2qMWYmC34QqHl8Sw9CsIqM9yBAwcycODAMu8fHh5OeHi4/n7Dhg356aef2Lt3r0GCb2dnZ9BrUNWVNlSl1GElQgghhJF99913+Pn58eyzzzJ58mQaNGhQpucNHz5cXw1fVJKbr3ptrh78zTMgIX91g+Iqtf8vf1SlvQvUaaReur0C9dup27V56jDXopJ/RYFbZ+HsJtjxNkYb9l8Wrt7q2tsJ5yD2ODR+yODh8vTSKYrC8WvJbD51k82nYrmWWFRvbvHeHhZWpu9zh85+xZcrZ/BR7ggyKUgMYxQfsnDAXpPHdzl9+HjiIB6o70l6Vi5pmbpLjv5+ev79tKxcTt9IJSI6sdTXzs7TkpCeDVT8ZIFOZo6Wg0W8ppujHXVcHQwuWTl5bDpZ+hz7b5/rVKbPME+r0H3eDm6mZBaZ8GoAP08nxnZuWOZkt2NwHfw9nbiZcpce9yT2uYp64tNGo75SS38PTt+AiMuJRFxOZPbG03Rr4s3Q1gE83NIXDyf7IuOtTG9xVm4e64/GlCmvePPnv2jfsA513Zyo6+5IXXdHajnbY1PE65U2IuDtYWE42tsQEZ1IRHQiVxPv8LrdGgbb3eS64s2i3Mdo5utOx+A6dGhYm3d/OcOttKxifyb77LsQmneFOle2UKf7s6W+72BvNz76/XyJP+farg6M7hjI8WspHLuaRHpWLvsuJrDv4i2edtwNGnj1SG3sk0/StkFtMrJzebuIEyXmHF1tlQl+ZR07dow//viDd99912D7hQsXCAgIwMnJiS5dujB37twyf/EwuQoMN4tPK3k4V3n3E0IIISpr7dq1DBs2rNyj5IYMGcKQIUNMFFUNY+4Ef+C8gh784njUUyvR59yBuFPqpeOkgsdPrIYts9TE36sJeDWG2o0gMxmOr1LnwRtz2H95PPGlGv99vfdl6aXL0yocuZLE5lOxbDl102BfJ3sbejfz4eFQX97fcpb41OITFz9PNWkr1j2jC9qHNOAV10ncvS9Zy8GOD3JHstjhEybZb8bRZx62Tva4O9nj71nyR1C497lon4xsQ4ifOxlZedzJzi24zs7jTpZ6fTdbPWHwR9TtUtsb1yWIfqG+1HF1wMvVkVou9jjZ2xbaL0+rcPhKUqkJeYmf4T1sbTS8NSSUF1ceRQMGberS2LeGhJYriba10fBJx2Qcdr9La5tL+voJ99ZFAPjgiVZMc23G/07EsvHEDf6KSWHP+VvsOX8Lh/U29G5Wl6Gt69GnhQ9O9rbl6i3WahWuJd3h7M00zt9M42ycen0pIaPMhfRWHbrGqkPXDLbZ2WjwdnPE292Bum5q0u/l5sCqg1dLHKXxxs+nDLY3s7nOZLtfALjV/W3+6DqE2q4O+scd7GxK/Jm06DMWflsDl3aqxfmcSl6GtSw/5/eGhxn8PZ+PS+PIlSRizx+m7qUU7iiObEqsT3biNdb8ee3+lzB4z+YaXV2tEvz69etz69YtcnNzmT17NhMnTtQ/1qlTJ1asWEGzZs2IjY1lzpw59OjRg1OnTuHuXvQcjaysLLKyCubWpKaaYLmZe+eslXO4mY+7k1H3E0IIISrr8ccft3QI4t4h+opi+iXlGvdWh6j++QX8+g91m8bWsDd/5Hfg2xKSrqjLzt2OAp+WBY/fjoKsVLWXPPZ40a9jqXXcdZX071Faz+Tf+zQlISOLLafiSEgv+C7p6mDLQy18GRTmx4PN6uLioH4Vd3awrXgyeescrHgEOv8NekwvMWn5VduJ49rGtLGJgj3vw+AFZfoICnqfS06gH2kVUKbE5UDU7TIl+APD/MvU426KhHxAmD+fj21bKHn2q+hQ69xsOpyZBzaX1LhKCMXf05lJPRsxqWcjohMy2HTiBhtP3OBifDpbT8ex9XQcrg62tAzwIOJyUqHn30zJ5IX830N3JzvOx6Vx7mYa5+PSuZtT9N+Ri70Nd3JKP3nWrbEXaOBWWha30rJIupNDrlbhZmomN1PL36kY4utG3xa+dAyuQ/eIFdhF5UGzRwjvV3h1lNJ+Jj3C/OFiL6jbHHIzgZIT/LK0ee/P2dZGQwt/D1r4e4DyP7gEdsHdWNyxC0evJrHzbDxnb6YV+1rmGl1drRL8vXv3kp6ezsGDB5k5cyZNmjRh1KhRAAZD/lu1akWnTp0ICgrihx9+4Lnnniuyvblz5xYqzGc0RqgyW9Z/tmU9WymEEEJU1tGjR1m5ciWjRo2iQ4cORe4TERHBmjVrGDduHG3atDFvgDWBmy+EPaFea3PBtvCQXqPTaMDBTb3t0xLsHAy/34Aah3cT9XK/B2dAqyfVRD8xCvZ+BJkppo+7AspS9X7h9gv6bR5OdvQN9WVQmD/dm3oX2QNd4WQy4QJ8PURdMeH0ejXJt3cqoT1nsjrPht1Pw5GvodOLUDek1Pds7ATaFN9hjZ6Q57fZL9SvcsXSbp2HPfPVaR4D3lenmdw4pu/RLU2wtytT+zTl5YeacPZmGhtP3GDTiRtcT7pbZHIPBT+fRff8Huo42NnQ1MeNZn7uNPN1p5mfO839PPB2c6DHBztL/Zl881wng/efk6fldnq2mvCnZ+oT/4joRPZcSCj1/b3UuwnD2tRT7wR8Ctvfhl4zi92/1J/JuJ9Lfc1yt1mUO4lg54RDyEP0C/WlX6gvzf3c9SualcTUo6urVYIfHBwMwAMPPEBcXByzZ8/WJ/j3q1WrFiEhIVy8eLHY9mbNmsX06dP191NTUwkMDKxckEasMmuKs5VCCCFEZXz66ad89913zJo1q9h9goOD+eyzz0hLS2PZsmVmjK6GsHdWh5WbW1z+cNvgHmoio/u+kxqjFtgrib0T+LRQLwB+rfTD/hWNDRrFzMPy73doKZzZCD1eJYJWpVa9B+jdrC7juzaka2NvHOyKLjB5r3InGbej1OQ+PQ58w9TExt6pbO3dHATnflU/45GryvQRGDOBLv07rMLsQU3K/R3WKAl5EbFWqLc1/oya2J/6Cf07dHCFSTshajuaX/4BSdFlbk6jKeg9fr1/M749eIU3fz5d6vM6NKxN18beakLv505QHRfsbIv+faxIXmFva4OfpxN+nk5AwVyPA1G3y5TgG4w0dveFRxeX+pwK/0yM2WafN6Dna6DN0W+yltHV1SrBv5dWqzUYXn+/9PR0oqKiePrpp4vdx9HREUdHI1atLEuV2XIyxdlKIYQQoqL27t1L27ZtqVu3+ISubt26tG3blt27d5sxMmFyugTft6Xao9+kLzTuU7El7Rr3ZsudZmzZ+B0Tslbpi5HdP1+5vCpSjExRFO5cPozr5b3sywzmX8WPwDXwaHg9ejXzKVd8ZU4yki7D10MhLRbqtlCT+yIq/BfbXt85cH6rWhMhJ9PgxEBJjJlAF/0dVmGY+1lmu62n1m/x0LD8KyWYIvkrl7jTsPsDiPwZfZrcfLCaDAa0Ue836QtTjsCnbdWfJWAw2qUUGo0GT+eyjcwZ2zmooIe8FMbMK8o1SuPmKfALK3PbpcrLhat/gJMn+Lc2Xrv3s3cCCv52rGV0tVUm+Onp6QY969HR0Rw/fpw6derQoEEDZs2aRUxMDN988w0AixcvpkGDBjRv3hyAPXv28OGHHzJ16lR9G//4xz8YMmQIQUFB3Lhxg7feegtbW9tie/hNoixVZitA98/22wOXmb0pEg8nO/a+3rvYs3NCCCGEqcTExBQ7NP9eQUFBnDxZ+vJEooLycuFOAtg5FVqf2WTi8nsTfe/5oq7RVGi9+i2nYnlx1TEUWrCBdwqWE7tnHfeKtFmWYmQZWbmcvJ7CsWtJHLuazLGryfS+68l8e3C8cZAr2QPK9Hom66VLvgorhkDqdfAOgfEb1Wr/5VE3BP52AOo2K/fLGzOB1p8wuHQbJWo7rc5/itvtk5BkxpUSjCnmCCy7Z6WFFkPVxN6/VeF9bW3VkS6rR4Kto/qzzIgrfbRLPlP1FhvrJE6ZRxpfOwDLB6qf1YgVYFN4Cku57fw/2PcxPDACHv+i8u3dr5iTYtYyutoqE/zDhw/Tu3dv/X3dMPnx48ezYsUKYmNjuXr1qv5xrVbLrFmziI6Oxs7OjsaNGzNv3jyef/55/T7Xr19n1KhR3L59m7p169K9e3cOHjxYYg+D0d1bZVZjA0YcbmZro2FUpwa8t/ksqZm5XL59hyY+bkZrXwghhCgLR0dHkpOTS90vNTUVW1sjfJETRVv3HERugAHzoPMLpn+9tDh1HrjGRi1wVQmF57hr2KNtzZ7sVvpEv74mkVou3pT1N6ikongvrDzKuC5B5GoVjl1N5tzNVO4vKH7EJhSAtraXWPR4M97depmE9EpUva+MqJ2QchXqNIbxmwqKKpZXBZJ7o1MUbC9tp4sRpq6a1b2rX92zggEBbcG/DdQJVhN735YlNkPIAPU5N46qvfw9ppX5hJgpe4uNdRKn1BEBzb3gv8PVjS51jJPcg/q57vtYHaWSm1Whk4wl+noIZKfD0P9A/fYGD1nD6GqrTPB79eqFohR/ZnbFihUG919++WVefvnlEttcs2aNMUKrHF2V2ajtsP1diC1hOZkKcLSzJTywFofy15KUBF8IIYS5tWzZkn379pGYmEidOkV/sUxMTGTPnj2EhRlxSKYwZO6l8hQtdHoB7iaDg0ulmoqITixmjntBou9ALmHfXcXfMx57Ww0OdjbqxdYWezsNjrbqfXtbG+xsNSzadqHEonjfHLhisD3A04nwBrUJb1CL8Aa1aOnvAf+Zh21aLMO8Y3F8tKXleunajVcToUa9wd2v8u1lJMCh/6oJqZ1D6fsby71TV3WfnKVWSiire2tpJV1Wa0Xc/AteOQmO7uqIlWe3lnnKAxoN9HsbEi9Bm9HlKohpLb3FpSlxRMC+BXDrDLh4Q5+3jPei9TuAmx+k34RLuyHkYeO1nZkCMYfV/3m6/7P3MUUtiPKwygS/Wrt3Ttr6F+Hk6vwHyj7vpiSdGnnlJ/i3Gd2pQaXbE0IIIcpj7Nix/O1vf+OJJ57gm2++oX59w+G1MTExjB8/njt37jBmTOFlkISR6Hp1M+LN83oe/upIRSM4G1vassQasrHn6NWiK4hX1JBW/jzSyp82gbXzC4bdJ6grnFoHV/5gQK8e5u2lS48HW4eC6RbhY43TrlYLX/WH2xfBxcs8oz107p26WuTpFytSaPWrfNG71VEr0Xuh+SB1W1mTe53gHuqlAqyht7gsihwRkHQFduX/z+j/f0XWkKgwGxtoMVhduvPMz8ZN8C/vU5N7ryZQq/ji65asBSEJvqVoNNBvDvz1g3q2sm4I3E0s87yb4nTKH4ZzKDoRRVHQmHrtWyGEEOIeEydOZPXq1ezatYuQkBAGDBhA48aNAYiKimLr1q3cvXuXbt268cILZkwmahpdgp9upgTfCE5cS2bp3kv8ejK2TPtP6hFM/douZOdqyc7TGlzn6O7naolOyODYteRS2+sb6ltyQhTULT/B3w+YsZcuI0EdEmzvDE+vB+faxmvbxga6TIH/vQK750Hrkear2TBwHmybDbHHzfN6FVFcYq8TMlBNTr0aG+f1crMh965aHK6MLN1bXCGKAr++pr7Xhj2g1VPGf40WQ9UE/+yvMDgXbI2U9l7apV436mWc9kxAEnxLcveF0KHquqWBnWHQB5WeIxLeoBZ2NhpiUzK5nnSXwDqVGyYnhBBClIednR2bN29m6tSpfP3112zYsMHgcVtbWyZMmMCiRYuws5OvISZj7iH6sSegTiN1mHI5aLUKO87Gs3TvJSKiE/XbHexsyM4temSjbm7xzIEtypTEHIi6zahlB0vdr9RiZEHdwNHToKfR5L10dxLVavm3zoK7vzoFwpgJPkD403Dwc7U3ff9C6DvbuO0XRzd19cLv8Nu/IOG8eV63rAxWvyrm96zXTOMl9xe2wS/T1JG+gxeU66kWXzmgvM7+Dy5sBRt7eOSj4j/fygjqBs511A7UK/uMl5BH7VSvrTjBlzLrltb+OfX6r7VqEYhKcnGw44H66lm/Q/ccKIUQQghzcXFx4YsvvuDq1ausWrWK999/n/fff59Vq1Zx9epVvvzyS9zcpE6MSZmzBz83W60cPrc+pMSU6SmZOXmsibhKvwW7mfjNYSKiE7Gz0fBY23ps/nsPPhnZBg0Fc4l1KjK3WFeMrLi9NajV9EstRla3GcyIVit9m8OdRPVzjT+tDgN/7Au1eJux2dqpo0pBTfRTrhv/Ne4Xc1TtxdVo1OHTL0VAz3/cs4MV9D5vnlHQa19CbTCjsXdSV0g4+q06fL06c3AFj0Do8pLpij3a2kHzR9TbuqS8slKuw+0L6t9jw4pNqzAHOXVuaQ27q5UeG3ZXf1mMoFOwF8euJhMRfZsn2lWhpUWEEEJUK/7+/uZdjlYU0Pfgx6vzrG1M2KeTcA60uShOnhy85Uh8dEyxw4STMrJZefAKXx+4TEJ6NgDujnaM7tyAZ7o2xN/TGYAW/h5Gm1tstGJkGs09Vd5NSFHgzCZYPxly7uZv04KjCU+KhQyAoO5qT+eO/4Phn5vutS5sg++eVJcwG/apWlhOo4He/4brR+HSDrV4oJJX6amrlWKw+pWt6QsANuwOwQ+q8/r3zFc/m+pIUdTfZ9c6cHwVdJxkuqUQu74M7Z+FgHDjtHdpt3od0NZ8U1kqQBJ8S9NoYPT3Rm2yU3AdluyOMhjqJoQQQogaxLUuhD2hJkp52WBjojXZAeJOA3A8qx6jvjik33zvGvNXbmfw5b5ofjh8jcwcdeh9gKcTz3YP5qkOgbg7Fa4ebsy5xUYvRnbv0mjGopvvve1tuHnCuG2XRqOBh99WRwycWA1d/gZ+Dxj/dW4chx/GFSTLNvekIhoNPPKhWvCv6cMFy9BZSuPecOe2GuP+heZJ9B/6N3y5G45/B92nGW/4vzVQFLi4HXbq6hnkFxjPSDBdgm/s0QFejaHNmNKXP7QwSfCroXYNa6PRwOXbd4hLzcTXw4QHdSGEEKIId+7cYefOnVy4cIG0tLQil7/VaDS88cYbFoiuBrC1hye+NMtLRZ86SDBwIsfwS7pujfnwwFocv56sH+XcMsCDyT0bMegBf+xtSx5ZYMy5xUY5YZB8DVYMUufCz7hsvHW7DeZ7W2gGbb120GokOHmo8/2NLekKrBoBORlqT/XQ/xSee+3VuCCptWRyD3D1IKybqJ4ke2G/urz1jvuSU2ML7AhN+sHF39Ve/OFLjP8a5laoUKHuZ26Cz6+0OCo7179BZ/Vi5SqV4N+5c4eEhAS8vLxwdXXVb09KSmLevHmcOnWKBg0a8Oqrr+or6Ipi5GRC5AZ17s2Dr1eqKQ8ne0L9PTh9I5VD0YkMbR1gnBiFEEKIMlixYgXTpk0jNbVgubP7V3bR3ZcEv2rL0yrEXzxKMHBWMVyeV3dKR1fBvlezukzu0Ygujb0stspPpU8YeATAnSTIToO4U+Df2jiB3btknGLmxOdew5eYpuDZnURY+bi6bKNvGDz1Ldg5lPyc9HiI3gMPPGH8eEqTcxc2/A1Q1KWtXb0KlrnWJaupMaaZQtB7lprgn/weuk9XV9qqqu49caVP7M28JGJ6vLpaQ9xpmLzLNL/fVqZSpwjfeecdgoODOXv2rH5bVlYWnTt3Zv78+fz6668sWbKELl26EBtbtiVPaqyEc7D+edj9gVEK4nQKVg9eEdG3K92WEEIIUVbbtm3jueeeQ6PR8M9//pMuXboA8N///pfXXnuNJk2aoCgKU6ZM4auvvrJwtNVcXi6kxsJd464Xf6+I6EQaaS8DcEbboNj9Pni8FSsmdKRrE++qvYSvjW1BD96VP4zXbsgAqN1QvW2Oef7Fuf9nY4zicjmZsHqUWpzMox6MWVv6MnCpN+A/7eGnyfopIGa1411IjFJHMvT/v4LtGo2a6E/aCa+cAs96xn/teu2g2SD1RM/ZTcZv35zuLVRo7sRex8FNXbEs9njllmSMOaK+F60FT8CVUaUS/B07dtC4cWPatWun37Zy5UouXLhA79692bp1K1OnTiUhIYEFC8q33EON498a6ncAbQ4c/abSzekqwco8fCGEEOb00UcfodFo2LlzJ++88w5NmzYFYNKkSbz//vucPn2aV155ha+++srg+4MwgZ//Bh83N8r3iuKk3LpOXU0KWkXDeaX4ebSO9tVo4aagrur1lf2VbysvF7b+C/5YpJ6MGfIJ+LdSH7Nkon/zL/h2uDofv7JiDqvJkaMnjPlRHQVRGo8AaNRTne/+y6vmqWKvcy0CDn6m3h6yqOhiahqNaacQ9J0NEzZDj1dN9xrmMHAeeFi44LeDCzTtp94+U4kTJrveh6W9IOK/RgnLlCr13/bq1av6A7fOxo0b0Wg0LF++nH79+rFw4UJCQkLYvHlzpQKtEXRL5h1ZAdrKFfDo0FBdI/V8XDqJGdmVDEwIIYQomz///JPOnTvTunXRQ5ft7Oz48MMP8fHx4a233jJzdDWMGZbKq+Phzls54/k8bwh3Kb7mT6lrzFclQd3U6yt/VC7xvJukVpM/kF8tvdtUdU36STth7LqCRN8Sq1pf3A5RO9SebF0l/4pq2B3G/ggjV4FvaNmfN+B9sHeFqweMc6KhLHRD8xUttB4FIf3N87r3q9us4ERSVaUo6smS1PxlF13z/x9Z4sRVi6HqdeTGiv3N5mbD5fwTela8PJ5Opf5jJCUlUatWLf19RVHYt28frVq1IjAwUL+9devWXLt2rTIvVTO0HA7OtSHlGlz4rVJNebk50tRHXU5FevGFEEKYS3p6Og0aFAzVdnRUe7nS0tL022xsbOjUqRN79+41e3w1yr1L5ZlIu+bB/OY2jPm5I4t8vMxrzFclAeFg56xWWE84X7E2bp2HZfnzue2cYcQKtYK6jY3hMPCx6yCgtXqyxpxLxnV6ATwD1XnmByu4ZF5OwWoFNOoFweVMjDzrQ68Z6u3f3lDn8ZvavoXqVAI3Pxgw1/SvVxbp8WqBwqrmWgTsek+93ftf8Oo5y524avow2DqoP9tbZ0vf/37X/1SLQ7rWBZ9ynKSykEp9sn5+fkRHR+vvHzlyhKSkJB588EGD/ar0XCtzsneC8LHq7T8rX/lWhukLIYQwNz8/PxITC447/v5qNe7z5w0TocTERO7erWTPoCiZPsGPM9lL6NaYL0q51pivSuwcILCDersiw/TP/wZf9FHneHsGwnO/qZ089zPHfO/i2DvBQ/kFMPctgIxy1nQ6vQEWd4Rb5yoXR+e/Qd3mcCcBdrxTubbKotPz8MAIGLxA7XSztFPrYFFr2DLT0pGUX4NO6lSDAe+rBcRtbCx34srJAxo/pN6O3Fj+51/apV4HP6i+DytXqQjbtGlDREQEGzZsIC0tjXfeeQeNRsPgwYMN9rtw4QIBAVLJvUzaP6teX9wGidEl71uKTo3yC+1dlkJ7QgghzKN58+ZcuHBBf79r164oisIHH3ygXyrvjz/+YMeOHTRrZuQ1ioUh3RdnE/bgc/YXQpTL2JFb6CE/Tyc+H9u2/GvMVwVhj0OHSWpF+PK6egCyUqFBVzXZ0fdoFsPU872L88AI8Gulxrrng7I/78oBtThe8hU4vqpyMdjawyMfqbcPL4frRyrXXmlc6sDjX0DzQaZ9nbLyawW5mXDuV7WOgbXLzTYs6tl9GnR+0XAfS5240g3Tr8g8fF2C36iXsaIxqUotk/f666/zv//9j8cffxxQh+i3adOGhx56SL9PXFwcJ06cYNSoUZWLtKao00j9pc/Nguz0SjXVsaHagx95I5XUzBw8nOyNEaEQQghRrEceeYS///3vRERE0LFjR/r06UOrVq348ccfqVevHv7+/pw6dQqtVssrr7xi6XCrN1P34Odmww/jaaTNwU+ziFYtH+DpLg0rvsZ8VdLumYo/96F/q0Xk2o4vfak4S7KxgYffgW+GwZ9fQMfJBWvUF+fWeVg9EvKyoNkj0McIdTYadoc2Y8DRA7ybVL69otz8Sz1ZY22jjr2bQqun1BoEO+eqtQysVfYd+GGcOnVl/EZwdC95f3OfuGo2UB1eH9JfrXVmU8ZaAJkpBSdXqkiCX6ke/K5du7J+/Xq6d+9O8+bNGTt2LBs3bsTmnqELq1evxt3dnQEDBlQ62Bpj5HfwzP/A74FKNePn6USQlwtaBY5cNt0SOUIIIYTOuHHj2Lx5M76+anJpY2PDL7/8Qr9+/YiPj+fYsWO4uLjw7rvvMnbsWAtHW83pEvy7iWoybmy3L4A2h1TFmTibuswY2Jwujb0Y1qYeXRp7Vd/kvrxSb8CmvxfMSbexhY6TrDu512nUS+140ubCoSUl75sWp651n5kM9dqrPeFlTaJKM2wxDHy/9OX1KiLmCPy3J6wZrXawWZsHX1cL0138XZ3Xbo0yU9Wf/cXfIf6MerE2LnXgbwegzxvl+728vF9dzcGrCdQKLH1/K1CpHnyAIUOGMGTIkGIff+WVV+QMfXkZ8WxWx4Z1uHL7DoeiE+nd3Mdo7QohhBBF8fT0pH9/w8rT9erVY8uWLdy5c4eUlBR8fHywta3YF/+YmBhmzJjB5s2buXPnDk2aNGH58uW0b9+enJwc/v3vf/Prr79y6dIlPD096du3L++//37NnCroXFvt/XOtqy7Di3ETSu3NU9gAZ5UGjOsSTJCXq1Hbt3q5WRBzVO2p9CtmqP61P+H7MeooClsHGDTfvDEaQ7931CJl7SYUv09WOnw3AlKuqqNRR3+vLk9mLPf2rCuKWu3eGO3nZhVUzXdwtcxUiNLUaQRtRsOxb9VVDcZXYA65KWXchpWPqWvMO3rAmLUQ2NHSURlPkz7wzC+QlVb6vlbC+qsE1GTp8XD020o1oZ+HHy3z8IUQQpje9OnTeeedoothubi44O/vX+HkPikpiW7dumFvb8/mzZuJjIzko48+onZttRjWnTt3OHr0KG+88QZHjx7lp59+4ty5cwwdOrTC76dKs7GBx5ZC//9TkxcjO3fiAACXbIJ4+SETDZ22Zrveh+UDiq8yf/w7WDFITe59QqHLS+aNz1h8Q9XicyWNONg2G2JPgIu3WkDN1ds0sSRcgBWD4Zfpxmlv9zy1qrprXRhYjjoD5vbg62BjD9G74fI+S0dTIPUGLB+oJvcu3uoI5AadLR1VyXLuwtlf1REnZWHnqE4TaTbQtHEZUaUS/Li4OPbs2UNcnOEHFBUVxciRIwkLC2PQoEEcOHCgUkHWSFlpsKgNbJwCsScr3Eyn/Er6J6+ncDc7z0jBCSGEEEX79NNPOXmy4setksybN4/AwECWL19Ox44dCQ4O5uGHH6ZxY3VesKenJ7///jtPPvkkzZo1o3Pnznz66accOXKEq1evmiSmmiojK5fk6GMA1GvekVouVWC4ubHp1im/v5J+bg5sngkbXoS8bGg+WK2UX7uh2UM0urycootA9/6nWqV89A9qj7OpZKaqn/eJ1ZVPdGOOqsvigVo138WKl3Ks1QDajlNHgVjL8PfES/BVf0g4Bx71YMJm8G9t6ahKt3oUrBkFp9dbOhKTqVSC//7779O7d29SUlL021JTU+nevTtr164lMjKSLVu20LdvX4OKuqIMHN0h5GH19uGKL5lXv7Yz/p5O5GoVjl2VefhCCCFMq379+mi1WpO0vXHjRtq3b8+IESPw8fEhPDycZcuWlficlJQUNBoNtWrVMklMVi8vB1Jjjb6G+LK9l2isvQxAp87lXN+8ugjsCBobSIpWezIVRV0ebn4wHMrv1e/5Ojz5bekFx6qCuNOwuBOsGqH+XoH6ni9uU4do3zwF7r6mjaF+O2ifP1Xgl38UxFFeuVnw80vq3Oqwx6FF8dONrUavmTD1uFq/wRooivo51mkEz26BuiGWjqhsmubnV2fKMNXh7K/w6+tw5Q/TxmRklUrwd+3aRWhoKCEhBT/QFStWEBcXx6hRozh37hwff/wxd+/e5aOPPqp0sDVO++fU65Nr1QqOFaDRaOiY34t/MNq4B3chhBDifo8++ii7d+8mLc348xUvXbrE559/TtOmTdm6dSsvvvgiU6dO5euvvy5y/8zMTGbMmMGoUaPw8PAocp+srCxSU1MNLtXK/16Bj5tXqrPgfvGpmazdfQwfTTIKGhz8K7BUXHXg5FlQEPnQUljWG9aOL5ir23c2PPSvKrFudpl41lcL6N2+AEe+hq3/hIUPqMXVbpyAjHjISDB9HA+9AS5ecOtM8dMjSrP7A4iPVIeVD6widRHcfMy3pFxZeDWGcT/DhC3qCIOqokX+cu5X/ih9CdHT6yHiv3Bxu+njMqJK/ceJiYmhUSPDYTi//PILdnZ2LFy4kKZNm/LKK6/QunVrdu/eXalAa6SG3cG7GeRkwInvK9xMp2CZhy+EEMI85syZQ4MGDRg0aBDHjh0zattarZa2bdvy3nvvER4ezuTJk5k0aRJLlhSu7p2Tk8OTTz6Joih8/nnxScDcuXPx9PTUXwIDq0aV5DLTL5VXyhfZcvj49/Mk5DjwtscctWico5vR2q5SFAU8839f9i8oPKWyUW/zx2RKTp7Qc4Z6e/PrcGAxpFzLf9A0o3aK5FJHLfwHah2ElOvlb6NRL6gVBIM/Blcvo4ZnFjFH4fph879u9F44/1vBfZ8Wph+1YWy1GkBAOKDA2V+K309R4NIu9XYVWR5Pp1IJflpaGi4uBRUs8/LyOHDgAO3atcPbu6C4RvPmzbl+vQJ/fDWdRgMd8nvxD3+p/qJVgK4H/9jVZLJyZR6+EEII0xk2bBiOjo7s37+f9u3bU79+fbp27cpDDz1U6NKnT59yte3v709oaKjBthYtWhSaX69L7q9cucLvv/9ebO89wKxZs0hJSdFfrl27Vuy+VZI+wS9jQalSnL2Zyg+Hr5GFA4MeG4fGWoYLm1vUTrXH/uz/CrYp1fw7VtROOPGdetvS77X1KAjsrHaCbZlV/ucH94CXIiB0mPFjM7Wj36q/e7/+o8K5QYWc/RW+fQx+eBpuGPfkrdm1yC+8WtIw/fhIdVSKvUuVWxWgUsvkBQQEcPbsWf39ffv2kZ6eTq9evQz2y83NxcGhBhZfMYbWI9XKpLfOqkVFGnYvdxON67ri5erA7Yxs/rqeQvuGVlxERAghRJW2a9cu/W1FUbhx4wY3btwocl+NpnzrpHfr1o1z584ZbDt//jxBQUH6+7rk/sKFC+zcuRMvr5J75xwdHXF0tMKlsYzFLX+J3PRbRmlu7q9n0SowMMyvZn+f2DxDLS5Wk1jTe7axgUc+Utevv31Rncrq5Fn687LSCuoh2DuZNkZTaTYQ7F3VJPvcZmg+yLSvpyiwbY46SgWgYW+o28K0r2lqocNg+xyI3gN3k9QlRe+n670P6mqdyyeWoFI9+F26dOHkyZMsXLiQv/76i3//+99oNBqGDDEsVHHmzBnq1bOiOSNViZMnPDAC7JzgVsX+qd47D/+QzMMXQghhQtHR0WW+XLp0qVxtT5s2jYMHD/Lee+9x8eJFvvvuO5YuXcpLL6nLj+Xk5PDEE09w+PBhVq1aRV5eHjdv3uTmzZtkZ2eb4u1aPyP24O85f4vd529hb6vh3YA/1PmpmdWsZkFZDZyXP8wXtdBeTWBt79kvTJ0D/vyesiX3sSdgQUu1XoI5e76NzdVbXbYQYOd7YKKipvoCigsfKEjuAR76d9U9OaLj1Rh8WoI2V03yixK1U72uglNtKvXXOWvWLBwdHXn11Vdp06YN+/fvp1evXnTt2lW/z+XLl4mMjKRTp05lbnfPnj0MGTKEgIAANBoNGzZsKHH/ffv20a1bN7y8vHB2dqZ58+YsWLCg0H6LFy+mYcOGODk50alTJyIiIsock0X1mgXTzxQM16+ATpLgCyGEMIOgoKByXcqjQ4cOrF+/ntWrVxMWFsY777zDwoULGTNmDKDWBtq4cSPXr1+nTZs2+Pv76y9//FG1qiAbjauuB79yc/DztArv/aouzzW+U3289r8Na5+BuzX0e0Xj3jBpp7rmu25pMI2tZWMyNWt8z8E9wNa+9P1ys2HDS2pP/+W9po/L1Lq+DA7uEPdX2arBl4cusV/WWy2gmHLftCWbSg0Atx6PfARTjhQ9TSM3u2D5yyo2/x4qOUS/ZcuW7Nu3j0WLFpGQkEC7du147bXXDPbZunUrrVu35tFHHy1zuxkZGbRu3Zpnn32Wxx57rNT9XV1dmTJlCq1atcLV1ZV9+/bx/PPP4+rqyuTJkwH4/vvvmT59OkuWLKFTp04sXLiQ/v37c+7cOXx8fMr1vs3Ow7/STXTML7R35HIiuXla7Gyt4MyrEEIIUU6DBw9m8ODBRT7WsGFDlKrcM2cKuiH62WmQnQEOrhVqZt3R65y9mYaHkx0vtwaOZqsJRq3ynaSpVjQaaNIXGveBqO2w4938uck2mLXonDlZ63vOzYI//gMth6u9s/fb97GaDDvXgUc+Vt9HVeZSB7r8DXbPg11z1WX+bIxwsiVqpzp0/cYx6xilYUpBXYp/LOWaulJDriv4hBa/n5XSKFZ+JNRoNKxfv75cJwgAHnvsMVxdXfn2228B6NSpEx06dODTTz8F1Eq8gYGBvPzyy8ycObNMbaampuLp6UlKSkqJBXtM6uYpqNscbMt3biZPqxD+9m+kZuaycUo3WtWvZZr4hBBCmJVVHJvucX/Bu9I0aGBdyytZ2+dZaYqirvftXBsenAFO5X9Pd7Jz6f3hLuJSs/jXoBZMqnUEfpqoFjl7bqsJgq6iFKUg6U2NgUm7rGtZM1Owlve8cSoc/Vo9+TDmR8ME/uZfsLSXOhz7ia/Ude+rg7vJsKiVOirh8S/hgScq1k5eDsQcUYeqH1isLoVYksm7IaBNxV6rKlEUuHNbnRJhBcpzbKomYywMHTt2jD/++IN3330XgOzsbI4cOcKsWQVVNm1sbOjbty8HDhwotp2srCyysrL09y2+Nu53I+H8Zhi5utwFNWxtNHRoWIftZ+OJiE6UBF8IIYRJNGzYsMzF8zQaDbm5uSaOqIbTaODRzyrVxBd7o4lLzaJ+bWfGdQ2CnV+rD/iFGSHAauTe3u287CpXmKtCrOU9d50KJ1arQ8vPbILQ/CrpeTmw4UU1uW82CFqWPjK4ynCuBV1eVlfaKup/rqIU/TPR5qknPaL3QPRuuHJAXY0A1F57/9ZqvQKNreVXSzC12BOwZ746Gmn4fcupajRWk9yXl1ES/Li4OL766iv27t1LTEwMAPXq1aNnz55MmDABX1/zrI9Yv359bt26RW5uLrNnz2bixIkAJCQkkJeXVygOX19fg1UA7jd37lzmzJlj0pjLxbsJnAf+/KJCFTM7BqsJ/sFLiUzs0cj48QkhhKjxevbsWWSCr9VquXbtGlevXkWr1dKlSxdZYacKiE/LZMnuKABmDGiOo50txJ1SH/RtacHIrJhGUzOS+3tZ+j17N4Fur8CeD2DLTGj8EDi6wd6P1WRWYwvXDqmjDDzrWy5OY+vyEnSdAvbOBdvuHVWREgOTdqjvWfd/ef0L8NcPhu24eEFwT/USNgKuHyqYflGdE31trnpCyN4VBn+sfo55uepnZYwpDxZS6QR/3bp1PPvss6SnpxvMe/vrr7/YunUr77//Pl9++SWPP2764TB79+4lPT2dgwcPMnPmTJo0acKoUaMq3N6sWbOYPn26/n5qaiqBgYHGCLVi2j+rzi+K2g6Jl6BO+ZL0To3Uefh/Xk5Eq1Wwsani84+EEEJYnXuXySvK+fPnmThxIoqisHnzZvMEVdPlZkPGLTUBK2eP1ILfL3AnO482gbUY3Cq/JlDcafXaV3rwhRXpMR1Ofg/JV2DX+9C4FxxZrj6maNXh1hkJ1SvBd3ApuH1vYn/jGKABFPjvgzDx94LaBIEd1eX1GnaD4AfVpN4nVF16UMca6yyYQkBb8KgPqdchagc0fwQubFVHfbQZAwPmWjrCCqlU9YTDhw8zatQoMjIyGD58OOvXr+fYsWMcP36cDRs28Nhjj5Gens7o0aM5fPiwsWIuVnBwMA888ACTJk1i2rRpzJ49GwBvb29sbW2JizNcIiYuLg4/P79i23N0dMTDw8PgYlF1Gql/bACHl5f76S0DPHBxsCXlbg7n49OMHJwQQghRupCQEH766SciIyN56623LB1OzbBlJiwIhUP/LdfTzsel8f2fak2Ffz3SQh2ZkXEb0mLVHapg8SlRjdk7w4B56u0D/1ErwOtXj7DqkmOVoyhw/jdYEKa+5xvHdA+oV3dvq0PxdcLHwozLMPp7tVCfX5hhcq+jm36hWzkhoLVatNO1rqnfkfloNGqBQoDI/NUIonaqdQ3yqu7SqpVK8OfOnUteXh5r167lxx9/ZNiwYbRu3ZpWrVoxdOhQ1q5dy9q1a8nJyeH99983VsxlotVq9fPnHRwcaNeuHdu3bzd4fPv27XTpUkIFRWvUQZ12wLGVkJNZrqfa29rQLqg2ABGyXJ4QQggL8fb2plOnTqxZs8bSodQMbvlTFNPjSt7vPnN/PYNWgQEt/ejQUF1uF+fa8PJRGP2DOgRaCGsRtRP2zDPcVl2HlutE7VSXs/tuhNoLXZTBC6D16IL79s7lK9Z9b6L/yqnqVzhSV6/h3GZ1tNOlXer9Krg8nk6lhujv27ePrl27Mnz48GL3GT58ON26dWPv3rKvOZmens7Fixf196Ojozl+/Dh16tShQYMGzJo1i5iYGL755htAXd++QYMGNG/eHIA9e/bw4YcfMnXqVH0b06dPZ/z48bRv356OHTuycOFCMjIymDBhQnnftmWF9C8YShK5AVqPLNfTOzasw94LCRyKTmRcl4YmCVEIIYQojaIohUbWCRPRLZWn780s3b4LCew8dws7Gw0zBjYveMDGRh3qW9RSZEJY0uYZkHDO0lGYV1nec0BbsHeq/GtZus6CqQR2AlcfyIhXCzXevqAWG2zYo/hChVauUj34KSkpZVrepkGDBqSkpJS53cOHDxMeHk54eDigJufh4eG8+eabAMTGxhosw6PVapk1axZt2rShffv2LF68mHnz5vH222/r93nqqaf48MMPefPNN2nTpg3Hjx9ny5YtZisAaDQ2ttD+GfX26Q3lfrpuHv6hS4myVrAQQgiLOHbsGLt37yYoqAavoW5O5ezBz9Mq/N+vZwAY2zmIYG9XU0UmhPEMnAcBau6ApuoWSCuXmviejc3GFloMVm//8qp67R8OMYfV0RELwiClmNERVqpSPfh+fn4cO3as1P2OHz9e4lz3+/Xq1avE5HPFihUG919++WVefvnlUtudMmUKU6ZMKXMcVqvtePBqqhaCKKdW9T1xsLMhIT2L6IQMGtWV4XVCCCGM596T6/dLT0/n/PnzbN68mdzcXJ5//nkzRlaD6RP8svXgrz8Ww5nYVNyd7Jjap6nhg9tmg6OHWoDKvYp1kojqrXFvdVj1vYXhqnMFeKiZ79kUWgyF2JNqUg/qSOmVj6MvLFjFijNWKsHv378/X3zxBf/85z955513sLU1PHOkKApvvPEGZ8+eZdKkSZUKVNzDzQdaPlqhpzrZ29ImsBYR0YlERCdKgi+EEMKoZs+ejUajKfFEvYuLS6GVaoQJ6YboZ8SrQ06LWjM7393sPD7cqg75ndK7CXVc71nKMC8XDixWh6y2LH56phAWo5svXhMqwOvUxPdsbI16gTYPVj+lLp2XcSv/gar5+VUqwX/jjTf46aefmDdvHqtXr+bJJ5+kYcOGAFy5coW1a9dy+fJlvLy8+Pe//22MeMX9tHnqL2I55oZ0Cq6jT/BHdix9ioUQQghRVsuXF7/Ki4ODA/7+/nTo0AFXVxn2bTa6qtd52ZCZrBbKK8aX+y5xMzWTerWcGd+1oeGDty+obTi4QS2ZXiGsWFFJb2pM9aoAf7+a+J6NIWonbJ9zz9KCqMsqVmGVSvDr16/Pjh07GDNmDKdOnWL+/PnqEiqgP3P/wAMPsGrVKurXrzrDGqqMYyvVdT67ToVOk8v8tE7BXvyHixySSvpCCCGMbPz48ZYOQdzP3kmd3ufkqfbgF+NWWhaf74oC4PUBzXCyv29Ob9xp9fr+NbOFsFb3Jr1VsFhahdTE91wZBoUKq0d9skol+KAm8CdPnmTXrl3s3buXGzduABAQEECPHj3o1atXZV9CFCf7DqRcg8NfQsdJJQ65u1fboFrY2WiISb7L9aQ71K/tYuJAhRBCCGFRQz8pdZeF286TkZ1H6/qeDGkVUHiHuFPqtV+YkYMTwsSqawX4ktTE91wRA+cV9OBXk/oFlU7wdXr16lVsMv/VV19x/fp1fRV8YSStn1KL3dw6C1f2Q8PuZXqai4MdYfU8OX4tmYjoREnwhRBCGM3Ro0dZuXIlo0aNokOHDkXuExERwZo1axg3bhxt2rQxb4CiSBfi0ljz5zUA/jmoBTY2RXQa3MxP8H1bmjEyIYQwoWpYqNAs46uWLVvGnDlzzPFSNYuTJ7Qaod7+88tyPbVTcB0AImSYvhBCCCP69NNP+eyzz/Q1eYoSHBzMZ599xuLFi80XWE2Xm60u9ZR+S78pT6twIOo2Px+PYcaPJ8nTKjwc6qtfUrcQ3RB93wfMELAQQpiJblrDpJ0wdh34t8p/oGpORaqaUYsC7Z9Tr89shLT71rdVFMjNKvJpHSXBF0IIYQJ79+6lbdu21K1bfGGnunXr0rZtW3bv3m3GyGq4bW/BgpZw4FMAtpyKpfu8HYxadpC/rznO0WvJAHRtUkxyn5miVuEH8A01Q8BCCGFm9yf6Aa3VVUiqWKFCSfCrOv9WUL+jWkn/2DfqNkWBi9tgWW9YEKaesb9P+4Z10GjgUkIG8amZZg5aCCFEdRUTE1Ni771OUFCQvm6PMAPdF9T0eLaciuXFlUeJTSl8/J+zMZItp2ILP9/JE/4ZC387BI7uJg5WCCEs6N5E/5VT4FnP0hGViyT41UGHier1n8vh/G9qYr/ycbhxQj3bnpFQ6Cmezva08PMAIOKy9OILIYQwDkdHR5KTk0vdLzU1FVtb21L3E0bi5guAkh7HnE2RJdaKnrMpkjxtEXvYOYBPc9PEJ4QQ1qaKFiqUBL86aDEUmj4Mjm7w3QiIPZn/QMlrOMowfSGEEMbWsmVL9u3bR2Ji8ceWxMRE9uzZQ2ioDPU2m/wE/07ijSJ77nUUIDYlU74bCCFEFSUJflUXtRNWDIQLv8Hti+q2MlZ9lEJ7QgghjG3s2LGkp6fzxBNPcP164SliMTExPPnkk9y5c4cxY8ZYIMIays0HANs78WXaPT7tvpMA61+An6dA4iVjRyaEEMKIjLZMnrCQzTMg4Zx6u5zLOeh68M/eTCMpI5varg7Gjk4IIUQNM3HiRFavXs2uXbsICQlhwIABNG7cGICoqCi2bt3K3bt36datGy+88IKFo61B8nvwHbOTsEGLtpQ+Hh93p4I7eblw6ifIy4Lu00wZpRBCiEoqV4Ivc+Ws0MB5sH1OhdZs9HJzpImPGxfj0/nzciIPt/QzYaBCCCFqAjs7OzZv3szUqVP5+uuv2bBhg8Hjtra2TJgwgUWLFmFnJ/0MZuPqDRobNIqW5h7ZRKY6FbmbBvDzdNJ3AgCQGKUm9/auUDvYPPEKIYSokHIdWRWlpJIsJdNoNBV+rihB497QqBdEbYcd76qJfjl0DK7Dxfh0IqIlwRdCCGEcLi4ufPHFF7zzzjvs2rWLa9euARAYGEivXr3w9/e3cIQ1kI2turSuvTOverXgubXRhXbRfVN7a0gotjb3fG+LO6Ve+4aCjczuFEIIa1auBF+rLblom7AQ3VIOjfvAhd/hxwmQnV6mp3YKrsN3h65KJX0hhBBG5+/vz6hRoywdhtB55EMA6t9MAwon+H6eTrw1JJQBYfedgLmpS/BbmjhAIYQQlSVj46oTjQZCHoaX/oTPOkJWGtg7F6x9WwTdELxTMSmkZ+Xi5ii/EkIIISpOq9WSnp6Os7Mz9vb2Re6Tk5PD3bt3cXNzw0Z6hM3ui71qobwBLX0Z3zWY+LRMfNzVYfkGPfc6cafVa98wM0YphBCiIuSoWh15BsATy9XbOXfhxtFid/X3dKZBHRe0ChyWXnwhhBCVtGDBAmrXrs3u3buL3Wf37t3Url2b//znP2aMTJCbRcL1i+w/HgnA5Acb06WxF8Pa1KNLY6+ik3u4Z4i+JPhCCGHtJMGvrpr2g64vq7d/fgnuFJ+8d5Tl8oQQQhjJ+vXrCQwMpG/fvsXu07dvX+rXr8+6devMGJlg5//h/UU7ntP8TLug2rRtULv05+TcBVt7QKPOwRdCCGHVJMGvzh56Exr1hsELwKVOsbtJgi+EEMJYLly4QMuWpc/VDgsL48KFC+VuPyYmhrFjx+Ll5YWzszMPPPAAhw8f1j+uKApvvvkm/v7+ODs707dv3wq9TnWU7axO2aurSWZSj0Zle5K9M/z9BMy6Dk6eJoxOCCGEMUiCX53ZOcDT6yHs8RJ36xzsBcCJ68lk5pR9mT0hhBDifikpKXh6lp4Ienp6kpSUVK62k5KS6NatG/b29mzevJnIyEg++ugjatcu6In+4IMP+OSTT1iyZAmHDh3C1dWV/v37k5mZWe73Ut0cjFPr7ATap9Ev1Ld8T3Z0M0FEQgghjE0qqlV39y5PmB4PyVehfnuDXQLrOOPn4cTN1EyOXk2ia2NvMwcphBCiuvD39+fkyZOl7nfy5El8fHzK1fa8efMIDAxk+fLl+m3BwQXrsiuKwsKFC/n3v//NsGHDAPjmm2/w9fVlw4YNjBw5slyvV53kaRV+PJ9NT6CRc0bx8+2FEEJUadKDX1PERcLn3WD1SEiLM3hIo9HIMH0hhBBG8dBDD3HmzBm+//77Yvf54YcfiIyMpHfv3uVqe+PGjbRv354RI0bg4+NDeHg4y5Yt0z8eHR3NzZs3Deb/e3p60qlTJw4cOFBkm1lZWaSmphpcqqOtp29yOtUZAI+8chzrvx0OywdBbOknbYQQQlieJPg1RZ1gcPOBjFvw0yTQGg7FlwRfCCGEMbz22ms4ODgwbtw4pkyZwsmTJ8nIyCAjI4OTJ08yZcoUnn76aRwcHHjttdfK1falS5f4/PPPadq0KVu3buXFF19k6tSpfP311wDcvHkTAF9fw+Hnvr6++sfuN3fuXDw9PfWXwMDACrxr66YoCkv3XOKWok6d0GSmQE4Zpixo8+DKAbiyH+xdTBylEEIIY5AEv6awd1aXzrN3gejdsO9jg4c75Sf4R68mkZ2rtUSEQgghqoHmzZvzzTffYGtry+eff054eDgeHh54eHgQHh7OZ599hq2tLV9//TVhYeVbdk2r1dK2bVvee+89wsPDmTx5MpMmTWLJkiUVjnfWrFmkpKToL9euXatwW9bqyJUkjl9LJtPOHcXWQd2Ycav0JyZegty76neHOsGl7y+EEMLirDLB37NnD0OGDCEgIACNRsOGDRtK3P+nn36iX79+1K1bFw8PD7p06cLWrVsN9pk9ezYajcbg0rx5cxO+CytUNwQGfaje3vmeelY+XxMfN+q4OpCZo+WvmBQLBSiEEKI6GDFiBCdPnuT555+nSZMmODo64ujoSJMmTXjxxRc5ceIETz31FFpt+U4o+/v7ExpquFRbixYtuHr1KgB+fn4AxMUZTkWLi4vTP3Y/R0dH/QkI3aW6WbrnEgCPhddH0/456PZ30CX6Jbn5l3rt0wJsbE0YoRBCCGOxygQ/IyOD1q1bs3jx4jLtv2fPHvr168evv/7KkSNH6N27N0OGDOHYsWMG+7Vs2ZLY2Fj9Zd++faYI37q1GQ2tngJFC+uegzvqkHyNRkPHhmov/qHo25aMUAghRDXQpEkTPvvsM86dO8edO3e4c+cO586dY/HixaSnpzN9+nTq169frja7devGuXPnDLadP3+eoKAgQC245+fnx/bt2/WPp6amcujQIbp06VL5N1UFRSdk8PsZ9YTHxB7BMPB96Pc2uJehin7cafXat3wjLYQQQliOVVbRHzhwIAMHDizz/gsXLjS4/9577/Hzzz+zadMmwsPD9dvt7OyKPYNfY2g08MhHcP0wJEapPfmPqL36HYPrsOX0TSKiE/lbL8uGKYQQonq5du0aq1atYuXKlZw5cwZFUdBoylfJfdq0aXTt2pX33nuPJ598koiICJYuXcrSpUsB9WT1K6+8wrvvvkvTpk0JDg7mjTfeICAggEcffdQE78r6fbnvEooCDzX3oYmPe/meLAm+EEJUOVaZ4FeWVqslLS2NOnXqGGy/cOECAQEBODk50aVLF+bOnUuDBg2KbScrK4usrCz9/WpTWdfRHUasgEP/hb5v6TfrCu0dvpxEnlaRJXSEEEJUSlpaGmvXrmXlypXs2bMHRVFQFIV69erx1FNPMWrUqHK116FDB9avX8+sWbN4++23CQ4OZuHChYwZM0a/z+uvv05GRgaTJ08mOTmZ7t27s2XLFpycnIz99qxeYkY2Px65DsCkHo3UjblZkB4HNvbg4V9yA3Gn1Gs/SfCFEKKqqJYJ/ocffkh6ejpPPvmkflunTp1YsWIFzZo1IzY2ljlz5tCjRw9OnTqFu3vRZ7Tnzp3LnDlzzBW2efm3gkcNp0C08PfA3cmOtMxczsSmElbP00LBCSGEqKry8vLYsmUL3377LZs2bSIzMxNFUQC1h33Xrl306NGj3L33OoMHD2bw4MHFPq7RaHj77bd5++23K9R+dbLy4BUyc7SE1fOgc6P8To89H8KeD6D9czD44+KfnJcD3iGQlw0+ocXvJ4QQwqpY5Rz8yvjuu++YM2cOP/zwAz4+PvrtAwcOZMSIEbRq1Yr+/fvz66+/kpyczA8//FBsWzWhsi4AigJHVmCbnUYH/Tx8WS5PCCFE2f35559MnTqVgIAAhg4dyg8//EBubi5Dhw5l7dq1dOjQAYCePXtWOLkXZZeZk8c3By4Dau+9/jN3y/9ulBFfcgO29vD0T/CP8+Bcy2RxCiGEMK5q1YO/Zs0aJk6cyNq1/9/enYc1deX/A3/fsEUQguygLOKGguJSRVzaulWspW4zWm2rra3T6dhFHTvqTEd07NSZ6WKdab92+Y2lHevaVu2qdd+VulBFrQKCCwQQkFVZc39/XBKIJCSBhITwfj1PHsjNuZdz7k04+dyzbcPYsWObTOvp6YmePXsiLS1Nbxr1rL927/tFwOn1QOZRDAlbiv2/5uHUtQI8N4JL4hARUdPeeOMNfPHFF7h69aqmpX7YsGF46qmnMH36dM1wufvnyyHL2nEuC/llVejs2QGP9m3QFV8d4JcZCPCJiKhNspsW/E2bNuHZZ5/Fpk2bMHHiRIPpy8rKkJ6ejsBAA+PP2oN+TwCCA3BhGybUSDMP/5xZCJVKtHLGiIjI1i1fvhxXr15FQEAAVq1ahWvXruHo0aP4/e9/32guHGodKpWIT45IS+M9OzwMTg4Nvu51rJs9vyxXx54NVN+zUO6IiMiSbDLALysrQ3JyMpKTkwEAGRkZSE5O1qxzu2zZMsyePVuTfuPGjZg9ezbeeecdxMTEICcnBzk5OSgurl/PffHixTh06BAyMzNx/PhxTJkyBQ4ODiZP8GOXQmKA0X+Rfj2ZgCgnJe7crUba7TIrZ4yIiNoCURSRk5OD3bt3Y8+ePSgqKrJ2ltq1A1fykH67HO4ujpgxOFj7xYYt+GITN/I/fRR4JwLIPGa5jBIRkdnZZIB/+vRpDBgwQLPE3aJFizBgwAAsX74cAKBUKjXBPgB8/PHHqKmpwfz58xEYGKh5vPrqq5o0t27dwsyZM9GrVy9Mnz4d3t7eOHnyJHx9fVu3cLZq+EIgfBSEmntY5/IfuKCq8Th8UZRm3yUiIqpz6tQpzJ8/H97e3pqW+8DAQEybNg1ff/01qqurrZ3Fdkfdej8rJgTuciftF93qAvzqu0CVnhv5qlog7zJQqqy/IUBERG2CIIpN3b6lhkpKSqBQKFBcXAwPDw9rZ8f8yvKAdcOB8jx8UTMGJ/q8jvdnDZQC+/R9wP43gOIs4HcHAEUXa+eWiIhgO3VTTU0NfvzxR2zYsEEze74gCOjUqROmTZuGAwcOID09HbW1tVbLozFs5Xw21/lbRXj8/WNwlAk4smQUAhUdGid6s7MU3L98FvDu1vj1/FTg/QcAxw7An7MAmYPlM05ERHqZUjfZZAs+WUlHP2DqRxAhYLrDQdy4eh7Hd29B2fsPAhumAdm/SLPuludbO6dERGRjHB0dER8fjy1btiAnJweffPIJRo4ciTt37uCTTz5Beno6AGDp0qWaIXhkfp8cyQAAxEcH6Q7uAeCBucDwBYCTq+7Xc1Okn369GdwTEbUxDPBJW7fRuBj5J/y9ehZWqdZi2InfQZ5fV9FDZdWsERFR2+Dh4YHnnnsOBw8eRGZmJv7+978jIiICoijirbfewqBBg9C7d2+sWrXK2lm1K7fu3MUPF5QAgOdHNrESziOrgHErAQ89Ew3nXpR+BkSZOYdERGRpDPBJy8/7vkLthW1Y4fw/RAqZAABHgYE9ERE1T3BwMJYtW4aLFy/i9OnTeOWVV+Dn54crV65gxYoV1s6eXfn0WCZqVSKGd/dGZJCi+QfKqbux788An4iorWGATxq1KhE+R5cjWiZNzqMvsK/ltA1ERNQMAwcOxJo1a5CVlYXvv/8eTzzxhLWzZDeK71Vjc5I0AfG8keFNJ66pBIpuSPPq6KJuwWeAT0TU5jDAJ42kjEK8Xvk0flFJXwxqRN1vj/Idi4Frh5peXoeIiEgPmUyGCRMm4IsvvrB2VuzG5qQbKK+qRS9/dzzU08AKQcf+DbzXFzi4uvFrKhXQczwQEgv497FMZomIyGIcrZ0Bsh15pRU4puqLSVVReFB2Hn903IZo4RpqRJlWa77H7dPA549Ld/anf657Bl4iIiJqFVU1Knx6LBMA8NzIrhAEoekd1EvfleU1fk0mAya+bd4MEhFRq2ELPmn4ucvrfhNwWBWNSVWrMLtqCS6KYQAAlSh9YcgPmSAtnVOWp71cnsq2lz4iIiKyR99fyEZOSQV83V0wqX+Q4R06+ks/y3ItmzEiImp1bMEnjSFdvRCokCOnuAJS53sp0D9c1U/Tot9FKESnqW8Dzq5A/lXA0UXaWaUCPnoICB4MDP0D4NPD8B8URaC2qv4YREREZBJRFPHxYWlpvGeGhcHF0Yhl7Zpqwb9zHXDzAZzdzJhLIiJqLWzBJw0HmYCEeGm8nXbnPinQn1y1CqenHIaDZxfA1QsIGVqfJOMQkHsBOL0eeP8BYOMMIOOw7nH6ogik7QU+GQWsiQKKb1myWERERHbrWFoBLitL0MHJAU/GhBi3k7oFvzxPukHf0JdzgTc7A1d+NG9GiYioVTDAJy1xUYFY99RABCjkWttlAvCfWQMxPjpU947hDwPPfA/0ehSAAFzdBXwWD3w4EkjeBNRUaQf2G6YB2b9IXy7K8y1eLiIiInv0yRFp5ZsZg4Ph6eps3E5udZPwqWqAe3fqt6tUQN4lACLgxfl1iIjaInbRp0biogIxrk8AkjIKoSy+h5XfXkLxvWrImpq0RxCAsBHSIz8NOLUOSN4oterv+D1Qfhu4+DWQfQ4Q1N0HdS/DR0RERIZdySnFoau3IROAucO7Gr+jozPQwQu4VyiNw3fzlrbfyQCq7wKOcsDLwFJ7RERkk9iCTzo5yATEdvPG1IFd8PRQqdV+46kbxu3s0x2Y+A6w8CIw4GnARQHs+SugPC+9LnIyPiIiopZSt97HRQUgxNvVtJ0feBYYsRCQe9Rvy02Rfvr1BhzYBkRE1BYxwCeDnhgSDEEAjqblIzO/3PgdXb2Am0lAZbH0XF9gn3kMqKlseUaJiIjaibySCuxMzgIAzBvZjNb2McuBsSu0V8PJqQvw/SNbnkEiIrIKBvhkUJdOrniopzReb9PPRrbiq034JxA0QPpd0DOz709/Bt7uAeycL90QICIioiYlHs9Eda2IB0I7YUBIJ/McNPei9NM/yjzHIyKiVscAn4wya4g0M++Xp2+hqsaEsfPdRgHzDgBPfQUE9pO23R/ou/oAFcXAuQ3AtUP121W1jWf3bYoosicAERHZvfLKGnxRN2xu3oPNHCtfXSEtiVd0s35b7gXpJwN8IqI2iwE+GWV0hB8CPOQoKK/C7os5pu0sCED3sY0DffXb78lt0gz8g54F+k6r3+/SDmBtP2DPckD5i+4l9wAuu0dERO1CrUrEifQCLN+ZguJ71QjzdsXY3v7NO1jSR1Idu3+V9FwUgaF/APo/yS76RERtGGdQIaM4OsgwfXAw/r0vFRtP3UB8dJDpB1EH+t3GAOn7gP1vACVZ0nq8is7SDPwNXf4OKL4JHFsrPbx7AH1/A0RNA3x6SF9G1MfJPgfphoFKWnav4ZhCIiKiNm5XihIrv70EZXGFZltheRX2XMpBXFSg6QfsWHdjoCxX+ikIwNAXzZBTIiKyJgb4ZLQnBgfj/f2pOHGtAOm3y9DNt2PzDtQw0K+tAhxddKeb/H9A5GTgwpfA1d1AQSpwcLX08OoKuHhILftcdo/IukSx6c+ytdl6/ogM2JWixIsbzuL+fmylFTV4ccNZrHtqoOlBfkc/6WfZbbPkkYiIbAO76JPRgjw7YFQv6QvBJmOXzGuKIDT9hdupA9BnEjDjf8BracCUj4CggdJrhRn1s/1y2T0i67D14TG2nj8iI9SqRKz89lKj4B6AZtvKby+hVqVnGJs+97fg3zojTbJXU9XcrBIRkQ1ggE8mmRVTN9ne2VuoqG7FwFruAUQ/AVQ1WKZPX2BfmNE6eSJqrxoGzhumAdm/AOV50vAYW2Dr+SMyQVJGoVa3/PuJAJTFFUjKKDTtwG51Lfh3C4DaamDXUmDdMODyN83PLBERWR0DfDLJw738EKSQo+huNXalmDjZnjkYs+zel88An4wBziQClaWtlTNqLVwtwXruD5yV5+tesJHhMbaeP6JmyCvVH9w3J52Gq1ddPSoCZXlcIo+IyE4wwCeTOMgEPFG3ZN5Gc3TTN5Uxy+4JDkDWaeDbV4G3ewJXf2r9fJL5sbu1dV3cIbXuNQycbWl4zNU9wL8H2G7+DOGNK6OsWLECgiBoPSIiIjSv5+Tk4Omnn0ZAQADc3NwwcOBAfPXVV1bMccv5ucvNmk5D5gC4+Uq/30oCqssBBxfAu7uJOSQiIlvCSfbIZDMGB2PtvlQkZRYiNbcUPfzdWzcDumbjbziL/lNfATkXgLOfA3cygMDo+n3zfgVcvYGOvob/TluYmKst5LGluFqC9ahUwLUDwNnPgEs767frHR5zDQjq3ypZ01CeB5K/AJI+BsS6lnp9+buT2fr5M6Th+7s4C/jdAb6vDYiMjMTevXs1zx0d67/KzJ49G0VFRfjmm2/g4+ODjRs3Yvr06Th9+jQGDBhgjey22JCuXghUyJFTXKFzHL4AIEAhx5CuXqYffNAzUh1Ski0994sAHPjVkIioLWMLPpnM30OOMRHS2L2NSVZoxVdTB/rqFv2gaGlWYJ+ewPBXgJd+Bl48Drg3WCP4h8XAuxHAlqeB1D2ASkcg0BZaittCHluqtbpbs+W0seIs4NC/gLXRwIap9cG9k6v0U+/wmGeBjx4CTn0M3LtjufyVFwAn/g9YNwL4aCRw6kMpuJfVBSaCnqpt2xwg8THg4nZpzLE1cZ6AZnN0dERAQIDm4ePjo3nt+PHjePnllzFkyBCEh4fj9ddfh6enJ86cOWPFHLeMg0xAQnwfvcE9ACTE94GDTNCRwoBRy4CxCUBFifSc3fOJiNo8mwzwDx8+jPj4eAQFBUEQBOzYsaPJ9F9//TXGjRsHX19feHh4IDY2Frt3726U7oMPPkBYWBjkcjliYmKQlJRkoRLYP/Vke1+daeXJ9nRpGOgvSAEUneu3+/aqT1ddAVTfA1Q10iRCX/wGeK8vsP/vUsteW/jC3RbyaA7pB+q7g2efk7aZu7t1e7hJ0hx7EoD3ooADfweKbwByBTDkBeD3x4A/ZxsYHuMIKJOBH18D8lObnwdDN11uJQG7lwG5FwAHZyByCvDkl8Cfc+ryF607fxCAzCPAtmeka35+W/Pz2FycJ6DFUlNTERQUhPDwcDz55JO4caP+RvOwYcOwZcsWFBYWQqVSYfPmzaioqMDDDz+s93iVlZUoKSnRetiauKhAjOvj32h7gELevCXy7pdbtyoNA3wiojbPJvthlZeXIzo6GnPnzsXUqVMNpj98+DDGjRuHN998E56envj0008RHx+PU6dOabrkbdmyBYsWLcKHH36ImJgYvPfeexg/fjyuXLkCPz8/SxfJ7jzYwxddOnXArTv38P15JaYNsoEupQaX3ZMD8/ZJEwmd/R9wfjNQkgUc/pf0cPOTgmVNUGCBL9zN7VIvikDqXmBfgpR/S+XRVrr8f/sKUGRk75D8VOC7hYBXV8CrG+AVXvfoCji7NU7PLv/a17nwmjRsRa6QXvPuLrWGh44ABs0BesdLS1aqNTU85umvgbzLwPVjQJfB9fvsfwOoqQD6Pwn49W46X/d3V68oBs59AXgEAcNeqs9D97FAzzggapo0WZgx+Zu1Vbo5cOYzoCxH+/1RXSGdD8GIVlBTPycqFVCQJg0lSP4CKL8NTdtrW5onwAbExMQgMTERvXr1glKpxMqVKzFy5EikpKTA3d0dW7duxYwZM+Dt7Q1HR0e4urpi+/bt6N5d/7jy1atXY+XKla1YiubJLroHAJg/qht6+rvDz13qlt+slnu16grps3DlR+m5f6QZckpERNYkiKJo4sKprUsQBGzfvh2TJ082ab/IyEjMmDEDy5cvByB9KRg8eDDef/99AIBKpUJwcDBefvllLF261KhjlpSUQKFQoLi4GB4eHiblxx59cCANb+2+goEhnvj6D8OtnR3TVVdIgf2pj4GqUumLva6Pw0NLgYhHAd+I5ge+zRlne+OkNJdA6k9A5lGg+q7+tFP/HxA2HHAPNC5AMUf+zKm2BihMr+9xkbYf2PQEUFspdbcWddzI+N0haTz15e+ALU/qPq57IDB2hbTEoigCl7+VWqZv/yrdJGkYXKmPZ2vMedNFfZ33rZICe7/ewM1TQNw/gaG/l9JUlQMlSsDHiIm2Gr5vSrKAeQfre9CoVd8D3u4FVBZLz4MGAv1nAX1/A3To1Pg42ecgBb8i4NOjvieARxdgwXlpYjBTy3t//mqqgKs/AhGP1R/vwJtS1/3Bz0vvF/UND33HM+Zzcu4LIHmj1Kuhqsy4PDfjfdie66aioiKEhobi3XffxXPPPYeXX34ZSUlJePPNN+Hj44MdO3ZgzZo1OHLkCPr27avzGJWVlaisrO8xUlJSguDgYJs6nwVllRj0hjTvwM9/GQtfdzPdhD25Tloez8kNGLkQeOA57RtmRERkE0yp622yBb+lVCoVSktL4eUlVVJVVVU4c+YMli1bpkkjk8kwduxYnDhxwlrZbPN++0AXrNlzFWdvFOHXnBJEBNjGFyGjOcml4LCqbik9ffe6Dv1DesgcpfH9/lHAsJfruyk3xVBrcW2N1LKXmwJUFEnBhdo3LwP5V40ry94EKYBxUUiTJPlGSMGbbwTg10d7HgJT8mdplaXAuQ3SeOrqu8DCFKm1uPto4E/XgJsn6/N2f0Cu1nkgMOVjKWAtTJd+FqRL57NUKQXG6QeAfSvru/sDtt9yas6bLupj7X4duH25fvvNUwAE6bypObsZF9wD2hNe6rsJIXMCpqyTAt2ru4Dss9Jj95+BXo9K1+/i9rprrB41VvdZzE+V9u81QWr9N5W+/Dk6A30m1acTReDCl9J5+PFPwN4VQL/p0ucxoG8Tn5Pb0pCf7HN1j2Rg8rr6mxwlWcD1o3V/swPQKQy4my/tp+/GFZnE09MTPXv2RFpaGtLT0/H+++8jJSUFkZFSS3R0dDSOHDmCDz74AB9++KHOY7i4uMDFxbYnKj2WXgAAiAhwN19wD0jz1gDSTaUHX5N+t5WeXERE1Cx2GeC//fbbKCsrw/Tp0wEA+fn5qK2thb+/dpDj7++PX3/9Ve9xdN3Vp3p+7nKM6+OPH1NysPHUDfxtUhscuzfhn/WBn74AMnCANBt/RRGQd0l6DH6u/vWUr6VZxv2jpGDAPwrw7gFcP6IdnALQdKk/sBoozQZuX5FaqQGpBWXQXEBWF+SEj5K6mjt3lJb9u5OpP48yJ+m1ymIpaLt5qv61oAHA7w7WPz+/FegYIAUax/+tO3+WVpINnPoIOP1pfcuuq7d0PtStly4dm+5ureYRBETPaPw37hZKwb5XOLA+Dsi/YuFCmYk5brqoaqWu7R063Xes+wyYDTy4GOgU2rI8NzU8xsERiJgoPcpuAxe2Sd3Uc1OASzukh/r9d3/AO+wVYMTClrcoGhq+IwjSZ+T8FuDn/yf18DiTKD18IqTP6J2Mxp+Tz+Klm1QNZZ+rD/AjHgM8OkufQZ+e0rm4//rq+0yTUcrKypCeno6nn34ad+9KvZxk6v+hdRwcHKBSte2bKUdTbwMARnT3MZDSRB3rvheV5Vq/JxcREZmF3QX4GzduxMqVK7Fz584Wj61vK+PyrGlWTAh+TMnB9rNZWDohAq7Obewt1W0UEP5w01+449+TJu0qyZLGv+dc0B6nePMUcO2g9GhEzzjb1F31vzu5Af59pGNW35UCWwB49F/1aQwFBdM/k1rs81Ol4CTvcv3PhnmtrQa2/75xfiwVYNzfElSYARz8B5DypdTyCUhjvmPnA9Eztcd6q+laFrEkq379Zn1cveoDQ2Nu5FzcLp0rB6fml7cldF1jAJpgsviWlO/Og+r3OfMZcOtn6WbG3YL6h3oG+6D+9x3rPoOfa3lwb4qOvkDsH6THe/2AouvSdn3vv/vH11uS3AMYMk9qtb9+TOqyf/0YkP8r9H6OK0ulm2v+kVIQr36o+feRHg0ZWuaTmrR48WLEx8cjNDQU2dnZSEhIgIODA2bOnAlPT090794dL7zwAt5++214e3tjx44d2LNnD7777jtrZ73ZRFHE0VRpItURPSwU4BekSStS5FxAu5yXhIjIjrSxaKxpmzdvxvPPP49t27Zh7Nixmu0+Pj5wcHBAbm6uVvrc3FwEBAToPd6yZcuwaNEizXP1uDyqN7ybD0K8XHGj8C6++0WJ6YPb4Pkx5gu3IEhfdBRdgJ7jtfcf9KzUDT43BchJkbqVa1oi9XT7HzQX6D5GCgw8Q+tb7VuSR0cXICBKejTUcOhBRYk0NKGq3OBpgShK3ecD+ko3D0zprqmvJaimUprcEABCh0tDHXqMN1x+wLju4PoYcyPn4g5gzHLjj2lOWkMI9ASTW56UXlteUD9u/NoB6caEPpZagcAc4tcavuliDYIAhI24b2UKPZ/jKR9JM/g3pytzc29ctXO3bt3CzJkzUVBQAF9fX4wYMQInT56Er6903n744QcsXboU8fHxKCsrQ/fu3fHZZ5/h0UcftXLOm+9afjmyiyvg7CBDTFdv8x1YFKWbwGo5dTPp80YTEVGbZjcB/qZNmzB37lxs3rwZEydO1HrN2dkZgwYNwr59+zST9alUKuzbtw8vvfSS3mO2hXF51iaTCZgVE4J//Pgrvki60TYDfLXmfuH2i5AeaukHgJ/+IrX2qycLu9+gOc2b0K05eWw46Z6bNzDjiwaBVRPjgO9kAN/UfT5kTlIZA6KluQcCo6WhCOreBmr6JktTtwT5RUiT3nV9ULsl2tRzYK6AqmEeY16oD5yr7wG/bAKiZ0k3RCxFFKWeFj8uaTCEQE8w6dhBagGvLAU6eErb+kyWbhK5etc/OnhJP3MuAAdsuBu4MTddrMmYXh8tmXhTrSU3rtqhzZs3N/l6jx498NVXX7VSblqHuvV+UGgndHA2YZLJpuial0Tf/x4iImpTbDLALysrQ1pamuZ5RkYGkpOT4eXlhZCQECxbtgxZWVn4/PPPAUjd8ufMmYO1a9ciJiYGOTk5AIAOHTpAoZBmQl60aBHmzJmDBx54AEOGDMF7772H8vJyPPvss61fQDvzm0Fd8M5PV/DLzSJczC5GZJCO2afbkpZ+4e42Sloz3JKBi6VbswEpyO36EKD8RZp/IOeC9Eiue33kH+tbvO8VA6f/n9QKnnMemlZo9RfGstv1xx2x0MTCmpm+myQNJ107kyjNLH3oLWDkImDgbPMFXqIonaOLO6Tx54XXgPj/AGfWN33TZe6uxjeFIifr/zvu/lIvEVsNoAHb7q7e2jcgmnvjiuzeEUt0z9e6qUhERPbEJgP806dPY9SoUZrn6m7yc+bMQWJiIpRKJW7cqF8j++OPP0ZNTQ3mz5+P+fPna7ar0wPAjBkzcPv2bSxfvhw5OTno378/du3a1WjiPTKdT0cXjI8MwHfnldh46gb+PkX3UkRtTku+cLdW4GL21uwG+fOPBOZ8IwWkxTcB5XkpMFX+Iv0eULeKQPoB4Ic/SjPXa9zXEuTs2ozCWVhTN0k6eAHuQdJEiD8sBo68a3ygr2sGak1Qv10K7O9k1L/mKAdc3IB5B8wfTNpyAN2QrXZXbyvnj+xWda0KJ69JM+iPNGeAb0wPFSIiapMEUdS3Nhjdrz2vNWzI8fR8zPrkFNycHXDqL2PR0cUm7x1ZjzHrhVtTc/InilIA9P4Qwy1BtrrGfFNqKoGzn0vBfWm2tM2jsxToP/Cc9tAHoOkZqNP2ARum1qd1lAM9xknjt3uM1x7qoG8WfXOcQ1t/H6rZ6jJdNnr+WDeZly2dz9OZhfjNhyfg6eqEM6+Pg4NMMLyTsQxN3toW/28TEdkpU+omRmFkFrHh3gj3ccO1/HJ8k5yNWTEh1s6SbbH1cbbNyZ86wLXXliBHF2lW9YGz6wP9kiypx8Lg5+vT6Zt34OSHwPg3pDRhI6TZqoNjpG719wf1DVmyNdvW34dqttpdva2cP7Ib6u75w7v5mDe4B9hDhYjITjHAJ7MQBAEzh4Tg7z9cxsak6wzw9bHVwEWtOfmz9cnSWur+QD9kqLRdFIFftkgT2RXfRKN5B1K+Ah5ZVX9OF16S1kE3liWDSVt/H9o6nj9qJcfSLLQ8XkO2OkSGiIiahQE+mc20QV3w1u4rSMkqwflbRejXxdPaWaLW0h5agtSBPmDcDNRD5kmT5anXoDcluG+IwSRRu1RaUY1zN4sAACO6WzDAV2MPFSIiu2DE4tNExvFyc8aEvgEAgI2nbhhITXZJ/QVx3gHgqa+AoGigo5/9tQT9uOS+4F6HbqPrl90jIjLRyWuFqFWJCPN2RbBXK05UypuKRERtGgN8MqtZQ6Su+d/8ko3Simor54aspmGgvyDFJiYiM6sJ/wSCBki/Cwziicj8jqZKy4tatHs+ERHZHQb4ZFZDunqhu19H3K2qxY7kbGtnh6zNXluCuo2q76UQWLdcIAN9IjKjI+rx993trAcUERFZFAN8Miv1ZHuA1E2fqzCS3bp/OII60Oe/VSJqoeyie7h2uxwyAYjt5m3t7BARURvCb6JkdtMGdoaLowyXlSVIrpsgiMhutZd5B4io1RytWx6vXxdPKDo4WTk3RETUljDAJ7PzdHXGxH6BAIAvONketRf2Pu8AEbUadff8kRx/T0REJmKATxbxZIzUTf+789kovsfJ9qgdsdd5B4ioVahUIo5pxt8zwCciItMwwCeLGBjSCb383VFRrcL2s7esnR0iIqI24ZKyBIXlVXB1dsCAkE7Wzg4REbUxDPDJIgRBwKy6VvyNSZxsj4iIyBhH61rvh4Z7w9mRX9OIiMg0rDnIYiYP6Ay5kwxXc8tw5voda2eHiIjI5qkn2GP3fCIiag4G+GQxig5OiO8XBEBaMo+IiIj0q6iuRVJmIQBOsEdERM3DAJ8sSt1N/7sLShTdrbJyboiIiGzX6cw7qKpRwd/DBd39Olo7O0RE1AYxwCeL6h/sid6BHqiqUeGdn65iZ3IWTqQXoFbFMflEREQNHUm7DQAY0d0XgiBYOTdERNQWOVo7A2TfBEFAdLACl5Ul+N/J6/jfyesAgECFHAnxfRAXFWjlHBIREdkG9fh7ds8nIqLmYgs+WdSuFCW2JN1stD2nuAIvbjiLXSlKK+SKiIjIthSUVeJidgkAYDgn2CMiomZigE8WU6sSsfLbS9DVGV+9beW3l9hdn4iI2r1j6QUAgIgAd/i6u1g5N0RE1FYxwCeLScoohLK4Qu/rIgBlcQWSMgpbL1NEREQ26GiqNP6e3fOJiKglGOCTxeSV6g/um5OOiIjIHomiqBl/z+75RETUEgzwyWL83OVGpfPtyK6IRETUfl3LL0d2cQWcHWSI6ept7ewQEVEbxgCfLGZIVy8EKuQwtNDP/x1Mg7L4XqvkiYiIyNaoW+8HhXZCB2cHK+eGiIjaMgb4ZDEOMgEJ8X0AoFGQr37uKBNwNK0Aj6w5jO3nbkEUOeEeERG1L0fqAvwRHH9PREQtxACfLCouKhDrnhqIAIV2d/0AhRwfPjUQuxY8iOhgT5RW1GDhll/w4oazKCirtFJuiYjIlq1YsQKCIGg9IiIitNKcOHECo0ePhpubGzw8PPDggw/i3j3b7SVWXavCyWvSDPqcYI+IiFrKJgP8w4cPIz4+HkFBQRAEATt27GgyvVKpxKxZs9CzZ0/IZDIsWLCgUZrExMRGXwrkcuPGiFPLxEUF4uiS0dg0byjWPtEfm+YNxdEloxEXFYjufh3x1e9j8cdxPeEoE7DrYg7Gv3cYey7lWjvbRERkgyIjI6FUKjWPo0ePal47ceIE4uLi8MgjjyApKQk///wzXnrpJchkNvl1BwDwy80ilFXWwNPVCZFBCmtnh4iI2jhHa2dAl/LyckRHR2Pu3LmYOnWqwfSVlZXw9fXF66+/jjVr1uhN5+HhgStXrmieC4Kh0eFkLg4yAbHddE8c5Oggw8tjemBUhB8WbU3G1dwyzPv8NH47qAuWx/eBu9yplXNLRES2ytHREQEBATpfW7hwIV555RUsXbpUs61Xr16tlbVmUXfPH97NBw4yfi8hIqKWsclb2hMmTMAbb7yBKVOmGJU+LCwMa9euxezZs6FQ6L/7LQgCAgICNA9/f39zZZnMIKqzAt+8NAIvPBgOQQC2nbmFuPeO4Hh6vrWzRkRENiI1NRVBQUEIDw/Hk08+iRs3bgAA8vLycOrUKfj5+WHYsGHw9/fHQw89pNXCr0tlZSVKSkq0Hq3pWBrH3xMRkfnYZIBvKWVlZQgNDUVwcDAmTZqEixcvNpne2pV+eyR3csCyR3tjy+9iEezVAVlF9zDrk1NY+e1FVFTXWjt7RERkRTExMUhMTMSuXbuwbt06ZGRkYOTIkSgtLcW1a9cASOP0582bh127dmHgwIEYM2YMUlNT9R5z9erVUCgUmkdwcHBrFQelFdU4d7MIADCiOwN8IiJquXYT4Pfq1Qvr16/Hzp07sWHDBqhUKgwbNgy3bt3Su481K/32bkhXL/z46oOYOSQEAPDpsUxM/PcR/FL3RahWJeJEegF2JmfhRHoBalWcfZ+IyN5NmDABv/3tb9GvXz+MHz8eP/zwA4qKirB161aoVCoAwAsvvIBnn30WAwYMwJo1azT1vz7Lli1DcXGx5nHz5s3WKg5OXitErUpEmLcrgr1cW+3vEhGR/bLJMfiWEBsbi9jYWM3zYcOGoXfv3vjoo4+watUqnfssW7YMixYt0jwvKSlhkN+KOro4YvXUvnikjz+WfHUe6bfLMXXdcYzv44+zN4qQU1KhSRuokCMhvg/iogKtmGMiImpNnp6e6NmzJ9LS0jB69GgAQJ8+fbTS9O7dW9ONXxcXFxe4uLhYNJ/6HE29DYDd84mIyHzaTQv+/ZycnDBgwACkpaXpTePi4gIPDw+tB7W+URF++Gnhg4iPDkKtSsQPKTlawT0A5BRX4MUNZ7ErRWmlXBIRUWsrKytDeno6AgMDERYWhqCgIK3JdAHg6tWrCA0NtVIOm3ZEPf6+u6+Vc0JERPai3Qb4tbW1uHDhAgID2eLbFni6OuO9Gf3h2UH3jPrqDvorv73E7vpERHZq8eLFOHToEDIzM3H8+HFMmTIFDg4OmDlzJgRBwGuvvYZ///vf+PLLL5GWloa//vWv+PXXX/Hcc89ZO+uNZBfdw7Xb5ZAJ0LvKDBERkalssot+WVmZVst6RkYGkpOT4eXlhZCQECxbtgxZWVn4/PPPNWmSk5M1+96+fRvJyclwdnbWdNX729/+hqFDh6J79+4oKirCW2+9hevXr+P5559v1bJR8yVlFKLoXrXe10UAyuIKJGUU8ssSEZEdunXrFmbOnImCggL4+vpixIgROHnyJHx9pRbwBQsWoKKiAgsXLkRhYSGio6OxZ88edOvWzco5b+xo3fJ40cGeUOi5eU1ERGQqmwzwT58+jVGjRmmeq8fBz5kzB4mJiVAqlY3G0w0YMEDz+5kzZ7Bx40aEhoYiMzMTAHDnzh3MmzcPOTk56NSpEwYNGoTjx483GqtHtiuvtMJwIgA5xfcsnBMiIrKGzZs3G0yzdOlSLF26tBVy0zL13fM5/p6IiMzHJgP8hx9+GKKov5t1YmJio21NpQeANWvWYM2aNS3NGlmRn7vcqHT/2PUrakVgcv8gODq021EoRERko1QqEccY4BMRkQUw+qE2Y0hXLwQq5BCaSCMTgNySSize9gvGvnsIX525hZpaVavlkYiIyJBLyhIUllfB1dkBA0I6WTs7RERkRxjgU5vhIBOQEC8Nqbg/yBfqHmtm9MeyCRHwcnNGZsFd/HHbL3hkzWFsP3eLk+8REZFNOFrXej803BvOjvwqRkRE5sNahdqUuKhArHtqIAIU2t31AxRyrHtqICb174wXHuqGI38ahSVxEejk6oRr+eVYuOUXjFtzCDuTsxjoExGRVakn2GP3fCIiMjdBNDR4nTRKSkqgUChQXFwMDw8Pa2enXatViUjKKEReaQX83OUY0tULDrLGnffLKmvw2fFMfHLkGoruSjPwd/friFfG9MDEvoGafYw9HhGRrWHdZF6WPp8V1bXot/InVNWosGfhg+jh7272v0FERPbFlLrJJifZIzLEQSYYtRReRxdHzB/VHbNjQ+sC/Qyk5ZXhlU3n8J99qXh1bA/IIGDV95egLK6fpT9QIUdCfB/ERQVashhEFsebV0S25XTmHVTVqODv4YLufh2tnR0iIrIzDPCpXXCXO+Gl0T0we1gYEo9l4v8duYbUvDK8tPGczvQ5xRV4ccNZrHtqIIN8arN2pSix8lvevCKyJUfSbgMARnT3hSDwZhsREZkXx+BTu+Ihd8IrY3rgyJLReGVMd70z8qvHraz89hLH7FObtCtFiRc3nNUK7oH6m1e7UpRWyhlR+6Yefz+yB8ffExGR+THAp3ZJ0cEJseE+aCp0FwEoiyuQeDwDJRXVJv+NWpWIE+kF2JmchRPpBbxR0Aw8h81TqxKx8ttLOt/fvHlFZD0FZZW4mF0CABjOCfaIiMgC2EWf2q280grDiQCs+u4yVn13GaHerogKUiCyswcigxSICvKAd0cXnfuwa3TL8Rw2X1JGYaOW+4bUN6+SMgqNmsuCiMzjWHoBACAiwB2+7rrrDyIiopZggE/tlp+73HAiAD5uzsgvr8L1gru4XnAX31+o79ocqJAjMkiByCAPRHVWIKqzB5JvFOEPX5xt1HrKcf3GU3cv5zk0TU2tCievFeKDg6lGpTf2JhcRmcfRVGn8PbvnExGRpTDAp3ZrSFcvBCrkyCmu0NmVWQAQoJDj6JLRKLlXjYvZJUjJLkZKVjEuZZfgWn45lMUVUBZXYO/l3Pr9BOjtGi1A6ho9rk+A1WcyN/fs6uY6nqHu5bZ0Dm1BTa0KpzIK8d15JXZfzEFheZXR+567UYQxvf3R0cV6VUF7nOW/PZaZAFEUNePvR/TwtXJuiIjIXjHAp3bLQSYgIb4PXtxwFgK0g3L1V+2E+D5wkAno5OaMET18MKJBq0tpRTUuK0txMbsYKVkluJhdjKu5pWhqWLOtdI02d/f3lh6vViXiekE5ruaWYu+lPCO7lxcgtptprWC2HlgZm7+aWhWSMgrx3QUldqfkoKBBUN/J1QmPRAZgz6Vc3CmvanKeicTjmfjyzC389oEumBMbhjAfNwuUSr/2OAyjPZaZJNfyy5FdXAFnBxmGhHlZOztERGSnBFEUOcuSkUpKSqBQKFBcXAwPDw9rZ4fMxJxfuL88cxOLt503mO6hnr544aFwDA7zgpND6851qa/7uzqMNLX7uynHU6lEZBXdw5WcUlzNK8XVnFJczS1D2u0yVNWoTCqHq7MDBod5YUCIJ/oHSw9PV+cm82nLgZWh/NWqRJzKKMD3dS31+WXaQX1cVAAm9g3C0HAvODrINNcF0H3zasbgYCRlFuLa7XJpuwCM6uWHOcPCMLK7D2QWvvFh7vehmi3fxLFUmVk3mZelzudnxzOR8M1FxIZ7Y9PvhprtuEREZP9MqZsY4JuAX6Lsl7mCghPpBZj5yUmj03vIHTEqwg9je/vjoV6+8JA7WTSPtSoRI/65X28LecNhCcYc29DxAKmMj/TxR2peGVLzynC3qlZnOrmTDD383NHJ1QmH67qxmqqrjxsGBHuif13QHxHgAWdHmcUCK3NpKn8igAd7+uBSdinyyyo1r3m6OiEuMgAT+wViaLi3zhtFhm4aqFQijqblI/F4Jvb/mqdJE+7rhmeGhWHqwC6Nuu/b4vvQ2PJak6XKDLBuMjdLnc/nPzuNvZdz8dr4Xpg/qrvZjktERPaPAb6F8EsUGaL+Et/UuH5PVyeMjvDDgSu3tcZLO8oEDA33xtjefhjT2x/BXq6a11oSuIiiiDt3q5GRX469l3Kx7lC6wXK4OMrg7CiDo0yAQ93DUSaDTAY4ymR1zwXcq6rF9cK7Bo/XkLODDOG+bugV4I6e/upHRwR3coVMJhh1DgMUcvzfkwNxIasY524UIflmETLyy3WWIzLIA1dySlGu58ZCSwIrczDmJomap6sTxveRgvrYbrqDel3HNyYgz8gvx+cnMrHt9C2UVdYAANxdHPHbB4IxOzYUYT5uZgug91/OxdzPThtM9+FTA42ea8HWb+IYe/Nv07yhJg/fYd1kXpY4n9W1Kgz42x6UVdbgm5eGo18XT7Mcl4iI2gcG+BbCL1FkDENdo9WBRq1KxLkbd7Dnci72Xc5DWl6Z1nEiAtwxro8/XJ0d8a9dvxoMXIruViEjvxyZBeXIyL+L6wXlyMwvR0Z+OUoqaixRVKPFRQVgUnQQevi7I8zbFY4GAlNjz2FDRXerkHyzSBPwJ98sQvG9aqPz2JzAqqWt2feqavHlmZv4686LBtMunRCB50Z0tfiQjrLKGnx15hY+O56Ja/n13fejgjxwIaukUfr7r0l1rQq5JRXILqpAdtE9ZBffk37WPc8quodSE96PggB4uTrDp6MLvDs6w7ujC3w61j13k356ujrhxS/O4nZppe5jwLo3ccoqa7B8Zwq+PptlMO3aJ/pjUv/OJh2fdZN5WeJ8ns4sxG8+PAFPVyeceX2czQwbISKitoEBvoXwSxQZqzktnRn55dh3ORd7LuXi58zCJifra8jJQYCrswOK7zUdNAUq5Ojk6oxLysZB2v3WzuiPvl0UUIkialQiampFze+1DR4Xsorxjx9/NXi85gTPLW0tFkURGfnlWH8sAxtO3jCYPkghx+CuXogI8EBEgDsiAt0R4CGHIOj+Im5q/orvVeNStjQZ48XsEqRkFSP9dpnR17k5gV9LqFQiDqfeRuLxTBy8cttgeicHAd5uzsgrrTS6TK2tOe/DlrhRcBeJxzOx7fRNlFYad1ODLfjWZ4nzuWbPVazdl4qJfQPxwZMDzXJMIiJqP0ypmziLPpEFxEUFYlyfAJNad7v6uOH5keF4fmQ47pRX4eDVPGxJuomTGYVN/q3qWlET3Pu5uyDMxw1dvd2knz6uCPNxQ6iXGzo4Oxjd/f2x6CCjWpiGhnvjs+OZBo83pKvpM0Y35xxq/W1BQLhvR0zsG2RUgJ9dXIGdydnYiWzNNg+5oxTwB7qjV4A7IuqGFRxLy9fZHTynuAIvbjiLf0zrCz8POS7VBfIXs0twQ89QBg+5o1E9LPzc5QbTmJNMJuDhXn54uJcfvj57C4u2/tJk+upaETklUgu6s4MMAQo5gjzlCPLsgM6eHRCkfijk8POQI+69wwbfNwcXP4ySihrkl1WioKwKBeWVuF1aiYLyKuTX/Swoq8SNwru4c9dwb40LWcUWD/BFUcSpjEKsP5qBPZdzob6FHu7jivyyKpRW1Jj9s0K271iaenk801b+ICIiMhUDfCILcZAJzQ4mOrk5Y8qALpAJgsEAHwBeG98LzwwLg5uB9cxNWRrQGOY+nq7jtzQgG9LVC4EKeZPBpK+7C/4+OQpX88rwa04pruSU4NptaWhDUmYhkjK1r4GDAJ3HUm9b8tUFnXnp0qkDIoM8EBmkQFRn6ae3mzNG/uuARW6SmIux12/RuJ54YkgwfNxcDM7Cb8z7xsXJAb5ODvB1d2nyWMaOb3/zh8vYevom4iIDEBcVgMggD709NExVUV2Lb3/JxvpjmbjcoJfMgz19MXd4GB7s4YufLuVY7LNCtqu0ohrnbhYBAEZ0Z4BPRESWxQCfyIYZ22o7MKSTweBeLS4qEOueGtioe3lAM2cbN/fxzM2YmxB/mxSJcZEBGBdZ/1plTS3S88pxJbcEv+aU4ldlKa7klCKnpAK1RnRBV3f5jwzyQFSQAn2CPPQu42fJmyTmYOz7cHCYl9Fpzfm+MXQTB5AmXKxVqZCWV4b389Lw/oE0dPbsgLgoKdgfGNJJ5zk2NM9CXmkFvjh5A1+cuq5ZulDuJMO0gV3w7PAwdPdzt0iZqW2oVYn47HgmalUi/D1cEOTZwdpZIiIiO8cx+CbgOEdqbcZ2qW/O5GHmXi/cltcfB8y3hNqmpOtY9nWKwXSmjplvC0u82fL70JiJGYd198GBX/OwKyUHB6/cxr3q+pUVfDq6YFwff8RFBSA23FuzvKK+a9KlkyvWH8vAd78oUVWr0rw2Z1gYnhgcrPdmjjnLrMa6ybzMdT5t+TNNRERtCyfZsxB+iSJraM6M8qSbOQIrSy53Zss3SdrC+9CUgOpeVS0Op97G7pQc7LmcqzWzv7vcEb0DPZBkxPAYABgY4om5I7pifGSAxVc50IV1k3mZ43za+rKNRETUtjDAtxB+iSJrYUuQ7bBka7atawvvw+bcJKmqUeHktQLsupiDny7mIr9M93J793s8OhBzR4Sjf7CnGXLefKybzKul51P9P6Lh56Qhe/4fQURElsEA30L4JYqsyZZbd9ubttCabSn2/j6sVYn434lMrPj2ksG0rb3snj6sm8yrpefTkr18iIiofTKlbmr9voRGOHz4MOLj4xEUFARBELBjx44m0yuVSsyaNQs9e/aETCbDggULdKbbtm0bIiIiIJfL0bdvX/zwww/mzzyRhahnlJ/UvzNiu3nbVVDV1qgnSwtQaE8oF6CQ23VwD9j/+9BBJqCTm/7x8w3llepuoaX2zdj3Bd8/RERkCTY5i355eTmio6Mxd+5cTJ061WD6yspK+Pr64vXXX8eaNWt0pjl+/DhmzpyJ1atX47HHHsPGjRsxefJknD17FlFRUeYuAhHZubioQIzrE2DXrdntlbErARibjtoXvn+IiMiabLIFf8KECXjjjTcwZcoUo9KHhYVh7dq1mD17NhQKhc40a9euRVxcHF577TX07t0bq1atwsCBA/H++++bM+tE1I7Ye2t2e6Vedk/f1RQgzT0wpKtXa2aLAKxYsQKCIGg9IiIiGqUTRRETJkwwqhegufH9Q0RE1mSTAb4lnDhxAmPHjtXaNn78eJw4cULvPpWVlSgpKdF6EBGRfXOQCUiI7wMAjYI09fOE+D68oWMlkZGRUCqVmsfRo0cbpXnvvfcgCNa5Pnz/EBGRNbWbAD8nJwf+/v5a2/z9/ZGTk6N3n9WrV0OhUGgewcHBls4mERHZgPY8z4Ktc3R0REBAgObh4+Oj9XpycjLeeecdrF+/3ko55PuHiIisxybH4NuKZcuWYdGiRZrnJSUlDPKJiNoJzrNgm1JTUxEUFAS5XI7Y2FisXr0aISEhAIC7d+9i1qxZ+OCDDxAQEGDU8SorK1FZWb80orl66/H9Q0RE1tBuAvyAgADk5uZqbcvNzW3yC4CLiwtcXFwsnTUiIrJR6nkWyDbExMQgMTERvXr1glKpxMqVKzFy5EikpKTA3d0dCxcuxLBhwzBp0iSjj7l69WqsXLnSIvnl+4eIiFpbuwnwY2NjsW/fPq0l9Pbs2YPY2FjrZYqIiIiMNmHCBM3v/fr1Q0xMDEJDQ7F161b4+vpi//79OHfunEnHZG89IiKyJzYZ4JeVlSEtLU3zPCMjA8nJyfDy8kJISAiWLVuGrKwsfP7555o0ycnJmn1v376N5ORkODs7o08faaKbV199FQ899BDeeecdTJw4EZs3b8bp06fx8ccft2rZiIiIyDw8PT3Rs2dPpKWl4cKFC0hPT4enp6dWmmnTpmHkyJE4ePCgzmOwtx4REdkTQRRF0dqZuN/BgwcxatSoRtvnzJmDxMREPPPMM8jMzNSqrHXNlhsaGorMzEzN823btuH1119HZmYmevTogX/961949NFHjc5XSUkJFAoFiouL4eHhYVKZiIiILKE9101lZWUICQnBihUrMH36dOTn52u93rdvX6xduxbx8fHo2rWrUcdsz+eTiIhskyl1k00G+LaKlT4REdma9lQ3LV68GPHx8QgNDUV2djYSEhKQnJyMS5cuwdfXt1F6QRCwfft2TJ482ei/0Z7OJxERtQ2m1E022UWfiIiI6H63bt3CzJkzUVBQAF9fX4wYMQInT57UGdwTERG1RwzwiYiIqE3YvHmzSenZSZGIiNobBvgmUH9RMNcauURERC2lrpMYzJoH63oiIrI1ptT1DPBNUFpaCgBcPoeIiGxOaWkpFAqFtbPR5rGuJyIiW2VMXc9J9kygUqmQnZ0Nd3d3nbP2txXqNX5v3rzZpicQspdyAPZTFnspB8Cy2CJ7KQdg3rKIoojS0lIEBQVBJpOZKYftF+t622MvZbGXcgD2UxZ7KQdgP2Wxl3IA1qvr2YJvAplMhi5dulg7G2bj4eHR5j84gP2UA7CfsthLOQCWxRbZSzkA85WFLffmw7redtlLWeylHID9lMVeygHYT1nspRxA69f1vNVPREREREREZAcY4BMRERERERHZAQb47ZCLiwsSEhLg4uJi7ay0iL2UA7CfsthLOQCWxRbZSzkA+yoL2SZ7eo/ZS1nspRyA/ZTFXsoB2E9Z7KUcgPXKwkn2iIiIiIiIiOwAW/CJiIiIiIiI7AADfCIiIiIiIiI7wACfiIiIiIiIyA4wwCciIiIiIiKyAwzw7czq1asxePBguLu7w8/PD5MnT8aVK1ea3CcxMRGCIGg95HJ5K+VYtxUrVjTKU0RERJP7bNu2DREREZDL5ejbty9++OGHVspt08LCwhqVRRAEzJ8/X2d6W7oehw8fRnx8PIKCgiAIAnbs2KH1uiiKWL58OQIDA9GhQweMHTsWqampBo/7wQcfICwsDHK5HDExMUhKSrJQCSRNlaO6uhpLlixB37594ebmhqCgIMyePRvZ2dlNHrM571FzMHRNnnnmmUb5iouLM3hcW7omAHR+ZgRBwFtvvaX3mNa4Jsb8z62oqMD8+fPh7e2Njh07Ytq0acjNzW3yuM39bFH7YC91PWA/9T3r+sZsqV5hXS9p7WsCsL63Rn3PAN/OHDp0CPPnz8fJkyexZ88eVFdX45FHHkF5eXmT+3l4eECpVGoe169fb6Uc6xcZGamVp6NHj+pNe/z4ccycORPPPfcczp07h8mTJ2Py5MlISUlpxRzr9vPPP2uVY8+ePQCA3/72t3r3sZXrUV5ejujoaHzwwQc6X//Xv/6Ff//73/jwww9x6tQpuLm5Yfz48aioqNB7zC1btmDRokVISEjA2bNnER0djfHjxyMvL89SxWiyHHfv3sXZs2fx17/+FWfPnsXXX3+NK1eu4PHHHzd4XFPeo+Zi6JoAQFxcnFa+Nm3a1OQxbe2aANDKv1KpxPr16yEIAqZNm9bkcVv7mhjzP3fhwoX49ttvsW3bNhw6dAjZ2dmYOnVqk8dtzmeL2g97qusB+6jvWddrs7V6hXW9da4JwPreKvW9SHYtLy9PBCAeOnRIb5pPP/1UVCgUrZcpIyQkJIjR0dFGp58+fbo4ceJErW0xMTHiCy+8YOactdyrr74qduvWTVSpVDpft8XrIYqiCEDcvn275rlKpRIDAgLEt956S7OtqKhIdHFxETdt2qT3OEOGDBHnz5+veV5bWysGBQWJq1evtki+73d/OXRJSkoSAYjXr1/Xm8bU96gl6CrLnDlzxEmTJpl0nLZwTSZNmiSOHj26yTS2cE3u/59bVFQkOjk5idu2bdOkuXz5sghAPHHihM5jNPezRe1XW63rRdF+63vW9bZfr7Cub91rIoqs7+9nqfqeLfh2rri4GADg5eXVZLqysjKEhoYiODgYkyZNwsWLF1sje01KTU1FUFAQwsPD8eSTT+LGjRt60544cQJjx47V2jZ+/HicOHHC0tk0SVVVFTZs2IC5c+dCEAS96WzxetwvIyMDOTk5WuddoVAgJiZG73mvqqrCmTNntPaRyWQYO3asTV2r4uJiCIIAT0/PJtOZ8h5tTQcPHoSfnx969eqFF198EQUFBXrTtoVrkpubi++//x7PPfecwbTWvib3/889c+YMqqurtc5vREQEQkJC9J7f5ny2qH1ry3U9YH/1Pet6269XANb1tnhNWN+bp75ngG/HVCoVFixYgOHDhyMqKkpvul69emH9+vXYuXMnNmzYAJVKhWHDhuHWrVutmFttMTExSExMxK5du7Bu3TpkZGRg5MiRKC0t1Zk+JycH/v7+Wtv8/f2Rk5PTGtk12o4dO1BUVIRnnnlGbxpbvB66qM+tKec9Pz8ftbW1Nn2tKioqsGTJEsycORMeHh5605n6Hm0tcXFx+Pzzz7Fv3z7885//xKFDhzBhwgTU1tbqTN8Wrslnn30Gd3d3g93crH1NdP3PzcnJgbOzc6MvkE2d3+Z8tqj9ast1PWCf9T3retuvV1jXS2zpmgCs743ZxxiOzd6TbN78+fORkpJicExKbGwsYmNjNc+HDRuG3r1746OPPsKqVassnU2dJkyYoPm9X79+iImJQWhoKLZu3WrUXT1b9d///hcTJkxAUFCQ3jS2eD3ai+rqakyfPh2iKGLdunVNprXV9+gTTzyh+b1v377o168funXrhoMHD2LMmDFWy1dLrF+/Hk8++aTBCaisfU2M/Z9LZE5tua4HrP+5tQTW9baNdb3tYn1vHmzBt1MvvfQSvvvuOxw4cABdunQxaV8nJycMGDAAaWlpFsqd6Tw9PdGzZ0+9eQoICGg0S2Vubi4CAgJaI3tGuX79Ovbu3Yvnn3/epP1s8XoA0JxbU867j48PHBwcbPJaqSv869evY8+ePU3e0dfF0HvUWsLDw+Hj46M3X7Z8TQDgyJEjuHLlismfG6B1r4m+/7kBAQGoqqpCUVGRVvqmzm9zPlvUPtlbXQ+0/fqedb1t1yus623vmqixvjduH2MwwLczoijipZdewvbt27F//3507drV5GPU1tbiwoULCAwMtEAOm6esrAzp6el68xQbG4t9+/ZpbduzZ4/W3XFr+/TTT+Hn54eJEyeatJ8tXg8A6Nq1KwICArTOe0lJCU6dOqX3vDs7O2PQoEFa+6hUKuzbt8+q10pd4aempmLv3r3w9vY2+RiG3qPWcuvWLRQUFOjNl61eE7X//ve/GDRoEKKjo03etzWuiaH/uYMGDYKTk5PW+b1y5Qpu3Lih9/w257NF7Yu91vVA26/vWdfbbr3Cut72rklDrO8lZqnvmz09H9mkF198UVQoFOLBgwdFpVKpedy9e1eT5umnnxaXLl2qeb5y5Upx9+7dYnp6unjmzBnxiSeeEOVyuXjx4kVrFEEURVH84x//KB48eFDMyMgQjx07Jo4dO1b08fER8/LyRFFsXIZjx46Jjo6O4ttvvy1evnxZTEhIEJ2cnMQLFy5YqwhaamtrxZCQEHHJkiWNXrPl61FaWiqeO3dOPHfunAhAfPfdd8Vz585pZpz9xz/+IXp6eoo7d+4Uz58/L06aNEns2rWreO/ePc0xRo8eLf7nP//RPN+8ebPo4uIiJiYmipcuXRJ/97vfiZ6enmJOTo5VylFVVSU+/vjjYpcuXcTk5GStz01lZaXechh6j1qjLKWlpeLixYvFEydOiBkZGeLevXvFgQMHij169BArKir0lsXWrolacXGx6OrqKq5bt07nMWzhmhjzP/f3v/+9GBISIu7fv188ffq0GBsbK8bGxmodp1evXuLXX3+teW7MZ4vaL3up60XRvup71vW2W6+wrrfONTFUFjXW9+at7xng2xkAOh+ffvqpJs1DDz0kzpkzR/N8wYIFYkhIiOjs7Cz6+/uLjz76qHj27NnWz3wDM2bMEAMDA0VnZ2exc+fO4owZM8S0tDTN6/eXQRRFcevWrWLPnj1FZ2dnMTIyUvz+++9bOdf67d69WwQgXrlypdFrtnw9Dhw4oPP9pM6vSqUS//rXv4r+/v6ii4uLOGbMmEZlDA0NFRMSErS2/ec//9GUcciQIeLJkyetVo6MjAy9n5sDBw7oLYeh96g1ynL37l3xkUceEX19fUUnJycxNDRUnDdvXqPK29avidpHH30kdujQQSwqKtJ5DFu4Jsb8z7137574hz/8QezUqZPo6uoqTpkyRVQqlY2O03AfYz5b1H7ZS10vivZV37OuT9DaZkv1Cut6SWtfE0NlUWN9b976Xqj7Q0RERERERETUhnEMPhEREREREZEdYIBPREREREREZAcY4BMRERERERHZAQb4RERERERERHaAAT4RERERERGRHWCAT0RERERERGQHGOATERERERER2QEG+EQEQRAMPp555hlrZ9OgFStWQBAEJCYmWjsrRERENoV1PVH74GjtDBCR7ZgzZ47e10aMGNGKOSEiIiJLYF1PZN8Y4BORBu+GExER2TfW9UT2jV30iYiIiIiIiOwAA3wiahZBEBAWFoaqqiokJCSgW7dukMvlCA8Px/Lly1FRUaFzv4KCArz22mvo0aMH5HI5vLy8EBcXh59++knv3yooKMBf/vIX9O3bF25ubvDw8EDfvn3xpz/9CUqlUuc+Fy5cwOOPP45OnTrBzc0NDz30EI4fP26WshMREbUHrOuJ2h4G+ETUbKIoYtq0aXjrrbfQp08fTJw4EYWFhVi1ahUee+wx1NbWaqXPysrCkCFD8Pbbb6OqqgqTJ0/GgAEDsHfvXowfPx5r1qxp9DcuX76M/v37480330R+fj7Gjx+PsWPHQhRFvPXWWzh16lSjfU6fPo2hQ4ciMzMT48ePR48ePXD48GGMGTMGKSkpFjsfRERE9oZ1PVEbIxJRuwdANPXfgXqfLl26iOnp6ZrteXl5YlRUlAhAXLNmjdY+jz32mAhAnDVrllhZWanZfuTIEdHV1VV0cHAQz507p9leXV0t9urVSwQgLliwQGsfURTFlJQUMS0tTfM8ISFBk6+1a9dqpV2wYIEIQHz66adNKicREZE9YF1P1D4wwCciTUXZ1GP79u069/n4448bHe/HH38UAYjdunXTbEtPTxcBiB07dhQLCgoa7bNo0SIRgPj8889rtm3ZskUEIEZGRoo1NTUGy6Gu9IcPH97otfz8fBGAGBoaavA4RERE9oZ1PVH7wFn0iUijqaVzQkJCdG5/4oknGm2Li4tDp06dkJ6eDqVSicDAQBw9elTzmpeXV6N9nn76abz77rs4cuSIZtvevXsBAM8//zwcHByMLscjjzzSaJu3tze8vLz0juMjIiJqD1jXE9k3BvhEpGHq0jmdOnWCu7u7ztdCQ0Nx584dZGdnIzAwENnZ2QCAsLAwnenV27OysjTbbt68CQDo1q2bSfnq0qWLzu3u7u4oLCw06VhERET2hHU9kX3jJHtEZBMEQTDbsWQy/msjIiKyNazriSyPnwwiarY7d+6gtLRU52s3btwAAAQFBWn9vH79us70mZmZAIDOnTtrtgUHBwMA0tPTzZJfIiIiMg3reqK2hQE+EbXI1q1bG2376aefUFhYiPDwcAQGBgIARowYAQDYtWsXioqKGu2zYcMGAMDIkSM128aOHQsA+O9//wuVSmXurBMREZERWNcTtR0M8ImoRVauXKm5Iw8A+fn5eO211wAA8+fP12wPDw/HxIkTUVpaildffRXV1dWa106cOIF169bBwcFBa5+pU6eiZ8+eSElJwZ/+9CetfQDg4sWLuHbtmoVKRkRERADreqK2hJPsEZHGM888o/e1kJAQ/O1vf2u0rV+/foiMjMSYMWPg5OSE/fv3o6ioCKNGjcIrr7yilf6jjz7CyJEj8fnnn+PQoUOIjY3F7du3cfDgQdTW1uKdd95B//79NekdHR3x1VdfYdy4cXjnnXewceNGxMbGQhRFpKamIiUlBdu3b0d4eLg5TwMREZHdYl1PZOesvU4fEVkfjFgbNzo6utE+oaGhYkVFhfjnP/9ZDAsLE52dncXQ0FDxL3/5i3j37l2dfys/P1/84x//KHbr1k10dnYWPT09xUceeUTcvXu33vzl5uaKixcvFnv27CnK5XJRoVCIffv2FZcsWSIqlUpNOvXauJ9++qnO44SGhor8t0dERO0R63qi9kEQRVFsvdsJRGQvBEFAaGioVpc9IiIish+s64naHo7BJyIiIiIiIrIDDPCJiIiIiIiI7AADfCIiIiIiIiI7wDH4RERERERERHaALfhEREREREREdoABPhEREREREZEdYIBPREREREREZAcY4BMRERERERHZAQb4RERERERERHaAAT4RERERERGRHWCAT0RERERERGQHGOATERERERER2QEG+ERERERERER24P8DGbZHmP9fnVkAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 1200x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_history(ang_hist)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Testing the model on the test set drawn from the windowed_df[training]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Agreement accuracy among models: 0.7217186743236699\n"
     ]
    }
   ],
   "source": [
    "## I can use any engine for this\n",
    "engine_long.test_all(model_long, model_tranv, model_ang, \n",
    "                     test_dl_long, test_dl_tranv, test_dl_angvel)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Seeing as the original data was split to get training, validation and test set (64:16:20), this means that the models were trained on just 64% of the training data instead of 70%, leaving 6% for val. Now that we're done testing, we can train the validation set."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Endgame Training\n",
    "Longitudinal Acceleration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "criterion = nn.CrossEntropyLoss(weight=class_weights)\n",
    "model_long = SimpleCNN()\n",
    "optimizer_long = optim.Adam(model_long.parameters(), lr=0.00011113828101205411, weight_decay=1e-4)\n",
    "scheduler_long = optim.lr_scheduler.ExponentialLR(optimizer_long, gamma=0.8492356384312725)\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "\n",
    "model_long = model_long.to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded states from checkpoint with loss: 0.6895003318786621\n"
     ]
    }
   ],
   "source": [
    "# Load the checkpoint\n",
    "checkpoint = torch.load('./model_checkpoint/run2_simpleCNN_long_chkpt.pth')\n",
    "\n",
    "# Load the different states\n",
    "model_long.load_state_dict(checkpoint['model_state_dict'])\n",
    "optimizer_long.load_state_dict(checkpoint['optimizer_state_dict'])\n",
    "scheduler_long.load_state_dict(checkpoint['scheduler_state_dict'])\n",
    "\n",
    "# Load the loss\n",
    "loss = checkpoint['loss']\n",
    "\n",
    "print(f'Loaded states from checkpoint with loss: {loss}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "engine_long = simpleCNN_engine(model_long, optimizer_long, scheduler_long, criterion, device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   Epoch [1/20], Train Loss: 1.4268, Train Accuracy: 62.78%\n",
      "   Epoch [2/20], Train Loss: 1.3788, Train Accuracy: 63.16%\n",
      "   Epoch [3/20], Train Loss: 1.3547, Train Accuracy: 62.35%\n",
      "   Epoch [4/20], Train Loss: 1.3365, Train Accuracy: 62.53%\n",
      "   Epoch [5/20], Train Loss: 1.3130, Train Accuracy: 62.28%\n",
      "   Epoch [6/20], Train Loss: 1.3204, Train Accuracy: 62.07%\n",
      "   Epoch [7/20], Train Loss: 1.3304, Train Accuracy: 62.72%\n",
      "   Epoch [8/20], Train Loss: 1.3084, Train Accuracy: 63.13%\n",
      "   Epoch [9/20], Train Loss: 1.2825, Train Accuracy: 62.80%\n",
      "   Epoch [10/20], Train Loss: 1.2799, Train Accuracy: 62.42%\n",
      "   Epoch [11/20], Train Loss: 1.2841, Train Accuracy: 63.10%\n",
      "   Epoch [12/20], Train Loss: 1.2914, Train Accuracy: 62.71%\n",
      "   Epoch [13/20], Train Loss: 1.2764, Train Accuracy: 63.16%\n",
      "   Epoch [14/20], Train Loss: 1.2812, Train Accuracy: 62.77%\n",
      "   Epoch [15/20], Train Loss: 1.3214, Train Accuracy: 63.08%\n",
      "   Epoch [16/20], Train Loss: 1.2525, Train Accuracy: 63.18%\n",
      "   Epoch [17/20], Train Loss: 1.3050, Train Accuracy: 63.41%\n",
      "   Epoch [18/20], Train Loss: 1.2644, Train Accuracy: 62.67%\n",
      "   Epoch [19/20], Train Loss: 1.2932, Train Accuracy: 63.22%\n",
      "   Epoch [20/20], Train Loss: 1.2750, Train Accuracy: 62.38%\n"
     ]
    }
   ],
   "source": [
    "long_endgame_hist = engine_long.train_endgame(valid_dl_long, epochs=20,\n",
    "                                              save_path='./model_checkpoint/endgame_2_simpleCNN_long_chkpt.pth')\n",
    "np.save('./model_checkpoint/endgame_2_long_history.npy', np.array(long_endgame_hist))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Tranversal Acceleration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "criterion = nn.CrossEntropyLoss(weight=class_weights)\n",
    "model_tranv = SimpleCNN()\n",
    "optimizer_tranv = optim.Adam(model_tranv.parameters(), lr=0.00015558427999575448, weight_decay=1e-4)\n",
    "scheduler_tranv = optim.lr_scheduler.ExponentialLR(optimizer_tranv, gamma=0.6560277884599833)\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "\n",
    "model_tranv = model_tranv.to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded states from checkpoint with loss: 1.0944734811782837\n"
     ]
    }
   ],
   "source": [
    "# Load the checkpoint\n",
    "checkpoint = torch.load('./model_checkpoint/run2_simpleCNN_tranv_chkpt.pth')\n",
    "\n",
    "# Load the different states\n",
    "model_tranv.load_state_dict(checkpoint['model_state_dict'])\n",
    "optimizer_tranv.load_state_dict(checkpoint['optimizer_state_dict'])\n",
    "scheduler_tranv.load_state_dict(checkpoint['scheduler_state_dict'])\n",
    "\n",
    "# Load the loss\n",
    "loss = checkpoint['loss']\n",
    "\n",
    "print(f'Loaded states from checkpoint with loss: {loss}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "engine_tranv = simpleCNN_engine(model_tranv, optimizer_tranv, scheduler_tranv, criterion, device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   Epoch [1/20], Train Loss: 1.3552, Train Accuracy: 63.54%\n",
      "   Epoch [2/20], Train Loss: 1.3456, Train Accuracy: 63.56%\n",
      "   Epoch [3/20], Train Loss: 1.3653, Train Accuracy: 63.45%\n",
      "   Epoch [4/20], Train Loss: 1.3478, Train Accuracy: 63.61%\n",
      "   Epoch [5/20], Train Loss: 1.3345, Train Accuracy: 63.42%\n",
      "   Epoch [6/20], Train Loss: 1.3461, Train Accuracy: 63.13%\n",
      "   Epoch [7/20], Train Loss: 1.3283, Train Accuracy: 63.15%\n",
      "   Epoch [8/20], Train Loss: 1.3304, Train Accuracy: 63.45%\n",
      "   Epoch [9/20], Train Loss: 1.3592, Train Accuracy: 63.35%\n",
      "   Epoch [10/20], Train Loss: 1.3656, Train Accuracy: 63.09%\n",
      "   Epoch [11/20], Train Loss: 1.3457, Train Accuracy: 63.45%\n",
      "   Epoch [12/20], Train Loss: 1.3653, Train Accuracy: 63.24%\n",
      "   Epoch [13/20], Train Loss: 1.3599, Train Accuracy: 63.12%\n",
      "   Epoch [14/20], Train Loss: 1.3675, Train Accuracy: 63.49%\n",
      "   Epoch [15/20], Train Loss: 1.3756, Train Accuracy: 63.70%\n",
      "   Epoch [16/20], Train Loss: 1.3352, Train Accuracy: 63.62%\n",
      "   Epoch [17/20], Train Loss: 1.3611, Train Accuracy: 63.72%\n",
      "   Epoch [18/20], Train Loss: 1.3468, Train Accuracy: 63.33%\n",
      "   Epoch [19/20], Train Loss: 1.3420, Train Accuracy: 63.76%\n",
      "   Epoch [20/20], Train Loss: 1.3637, Train Accuracy: 63.39%\n"
     ]
    }
   ],
   "source": [
    "tranv_endgame_hist = engine_tranv.train_endgame(valid_dl_tranv, epochs=20,\n",
    "                                              save_path='./model_checkpoint/endgame_2_simpleCNN_tranv_chkpt.pth')\n",
    "np.save('./model_checkpoint/endgame_2_tranv_history.npy', np.array(tranv_endgame_hist))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Angular Velocity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "criterion = nn.CrossEntropyLoss(weight=class_weights)\n",
    "model_ang = SimpleCNN()\n",
    "optimizer_ang = optim.Adam(model_ang.parameters(), lr=0.00013662142756319453, weight_decay=1e-4)\n",
    "scheduler_ang = optim.lr_scheduler.ExponentialLR(optimizer_ang, gamma=0.4740572036017916)\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "\n",
    "model_ang = model_ang.to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded states from checkpoint with loss: 1.460357427597046\n"
     ]
    }
   ],
   "source": [
    "# Load the checkpoint\n",
    "checkpoint = torch.load('./model_checkpoint/run2_simpleCNN_angvel_chkpt.pth')\n",
    "\n",
    "# Load the different states\n",
    "model_ang.load_state_dict(checkpoint['model_state_dict'])\n",
    "optimizer_ang.load_state_dict(checkpoint['optimizer_state_dict'])\n",
    "scheduler_ang.load_state_dict(checkpoint['scheduler_state_dict'])\n",
    "\n",
    "# Load the loss\n",
    "loss = checkpoint['loss']\n",
    "\n",
    "print(f'Loaded states from checkpoint with loss: {loss}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "engine_ang = simpleCNN_engine(model_ang, optimizer_ang, scheduler_ang, criterion, device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   Epoch [1/20], Train Loss: 1.3782, Train Accuracy: 61.72%\n",
      "   Epoch [2/20], Train Loss: 1.3834, Train Accuracy: 61.45%\n",
      "   Epoch [3/20], Train Loss: 1.3971, Train Accuracy: 61.76%\n",
      "   Epoch [4/20], Train Loss: 1.3570, Train Accuracy: 61.91%\n",
      "   Epoch [5/20], Train Loss: 1.3784, Train Accuracy: 61.52%\n",
      "   Epoch [6/20], Train Loss: 1.3999, Train Accuracy: 61.62%\n",
      "   Epoch [7/20], Train Loss: 1.3696, Train Accuracy: 61.90%\n",
      "   Epoch [8/20], Train Loss: 1.4101, Train Accuracy: 61.23%\n",
      "   Epoch [9/20], Train Loss: 1.3516, Train Accuracy: 62.17%\n",
      "   Epoch [10/20], Train Loss: 1.3763, Train Accuracy: 61.78%\n",
      "   Epoch [11/20], Train Loss: 1.3662, Train Accuracy: 61.96%\n",
      "   Epoch [12/20], Train Loss: 1.3631, Train Accuracy: 61.52%\n",
      "   Epoch [13/20], Train Loss: 1.3584, Train Accuracy: 61.71%\n",
      "   Epoch [14/20], Train Loss: 1.3772, Train Accuracy: 61.72%\n",
      "   Epoch [15/20], Train Loss: 1.3667, Train Accuracy: 61.84%\n",
      "   Epoch [16/20], Train Loss: 1.4011, Train Accuracy: 61.39%\n",
      "   Epoch [17/20], Train Loss: 1.4129, Train Accuracy: 61.40%\n",
      "   Epoch [18/20], Train Loss: 1.3591, Train Accuracy: 60.80%\n",
      "   Epoch [19/20], Train Loss: 1.3703, Train Accuracy: 61.49%\n",
      "   Epoch [20/20], Train Loss: 1.3534, Train Accuracy: 62.16%\n"
     ]
    }
   ],
   "source": [
    "ang_endgame_hist = engine_ang.train_endgame(valid_dl_angvel, epochs=20,\n",
    "                                              save_path='./model_checkpoint/endgame_2_simpleCNN_angvel_chkpt.pth')\n",
    "np.save('./model_checkpoint/endgame_2_angvel_history.npy', np.array(tranv_endgame_hist))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plotting only Long_hist and Tranv_hist as I've lost the first ang_hist data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "from src.plot import plot_endgame_history\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "long_hist = np.load('./model_checkpoint/run2_long_history.npy')\n",
    "tranv_hist = np.load('./model_checkpoint/run2_tranv_history.npy')\n",
    "ang_hist = np.load('./model_checkpoint/run2_angvel_history.npy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_endgame_history(long_endgame_hist, long_hist)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(np.arange(len(long_endgame_hist[0])), long_endgame_hist[0], '-o', label='Train loss (test + val)')\n",
    "plt.xlabel('Epoch', size=15)\n",
    "plt.ylabel('Loss', size=15)\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_endgame_history(tranv_endgame_hist, tranv_hist)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkYAAAG2CAYAAACap0noAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAACD1klEQVR4nO3deXxU9bk/8M+ZPZnse0ISEkJCCIEAIotoBUUJasSltVKoeq1L/aHWqq1SUUBbkWorbS8XW28Leqm4VEBwAVcEBGWNEgIhhJCEZJKQfbLMZJbz+2PmnEz2TGY5yzzv14tXm8nJmTNOZvLM9/ssDMuyLAghhBBCCBRCXwAhhBBCiFhQYEQIIYQQ4kSBESGEEEKIEwVGhBBCCCFOFBgRQgghhDhRYEQIIYQQ4kSBESGEEEKIk0roC5ASu92OmpoahIaGgmEYoS+HEEIIISPAsiyMRiOSkpKgUAy9JkSBkRtqamqQkpIi9GUQQgghZBSqqqqQnJw85DEUGLkhNDQUgOM/bFhYmMBXQwghhJCRaGtrQ0pKCv93fCgUGLmB2z4LCwujwIgQQgiRmJGkwVDyNSGEEEKIEwVGhBBCCCFOFBgRQgghhDhRjpEP2Gw2WCwWoS+DEEGp1WoolUqhL4MQQtxCgZEXsSyL2tpatLS0CH0phIhCREQEEhISqO8XIUQyKDDyIi4oiouLQ3BwMP0xIAGLZVl0dnaivr4eAJCYmCjwFRFCyMhQYOQlNpuND4qio6OFvhxCBBcUFAQAqK+vR1xcHG2rEUIkgZKvvYTLKQoODhb4SggRD+71QDl3hBCpoMDIy2j7jJAe9HoghEgNbaURQgiRDJudxeHyJtQbTYgL1WFmehSUCgrAifdQYEQIIUQSdhcZsGZXMQytJv62xHAdVhXkID+XEvyJd9BWmgjZ7CwOlTXig8JqHCprhM3OCn1JbktLS8P69esFP4cnSkpKkJCQAKPRKNg1iMnq1asxdepU/uunn34ajzzyiHAXRALK7iIDHtpyvFdQBAC1rSY8tOU4dhcZBLoyIjcUGInM7iIDrlz3JZa8/i1+9XYhlrz+La5c96XPXvQMwwz5b/Xq1aM675EjR/DAAw9492L9bMWKFXjkkUf4acybN29GRESEV+9j7969YBhGkr2vnnzySbzxxhs4f/680JdCZM5mZ7FmVzEG+ojI3bZmV7EkP0QS8aHASESE+ERkMBj4f+vXr0dYWFiv25588kn+WJZlYbVaR3Te2NhYSVfoVVZW4sMPP8Q999wj9KUMaPPmzZg3b56g1xATE4OFCxdi48aNgl4Hkb/D5U393hddsQAMrSYcLm/y30UR2aLAyIdYlkVnt3VE/4wmC1btPDXkJ6LVO4thNFmGPRfLjvxTU0JCAv8vPDwcDMPwX585cwahoaH45JNPcNlll0Gr1eLAgQMoKyvD4sWLER8fj5CQEFx++eX4/PPPe5237zYYwzD43//9X9x6660IDg5GZmYmdu7c6dZ/z8rKSixevBghISEICwvDHXfcgbq6Ov7733//PebPn4/Q0FCEhYXhsssuw9GjRwEAFRUVKCgoQGRkJPR6PSZNmoSPP/540Pt69913kZeXhzFjxgBwrOz813/9F1pbW/utppnNZjz55JMYM2YM9Ho9Zs2ahb179/LnGuy+L1y4gPnz5wMAIiMjwTCMTwKxtrY2BAUF4ZNPPul1+/bt2xEaGorOzk4AwFNPPYWsrCwEBwdj3LhxePbZZ4ctsy8oKMDbb7/t9WsmxFW9cfCgaDTHETIUSr72oS6LDTnP7fHKuVgAtW0mTF796bDHFj+/EMEa7z21Tz/9NF555RWMGzcOkZGRqKqqwg033IA//OEP0Gq1ePPNN1FQUICSkhKkpqYOep41a9bgj3/8I15++WX87W9/w9KlS1FRUYGoqKhhr8Fut/NB0ddffw2r1Yrly5fjpz/9KR+ELF26FNOmTcPGjRuhVCpRWFgItVoNAFi+fDm6u7uxb98+6PV6FBcXIyQkZND7279/P2bMmMF/fcUVV2D9+vV47rnnUFJSAgD8zz/88MMoLi7G22+/jaSkJGzfvh35+fk4efIkMjMzB73vlJQUvP/++7j99ttRUlKCsLAwvimiN4WFheGmm27CW2+9hUWLFvG3//vf/8Ytt9zCr+yFhoZi8+bNSEpKwsmTJ3H//fcjNDQUv/3tbwc998yZM3Hx4kVcuHABaWlpXr92QgAgLlTn1eMIGQoFRmRYzz//PK677jr+66ioKOTl5fFfv/DCC9i+fTt27tyJhx9+eNDz3HPPPViyZAkA4MUXX8Rf//pXHD58GPn5+cNewxdffIGTJ0+ivLwcKSkpAIA333wTkyZNwpEjR3D55ZejsrISv/nNb5CdnQ0AyMzM5H++srISt99+OyZPngwAGDdu3JD3V1FR0Ssw0mg0vVbUXM+7adMmVFZWIikpCYAj92b37t3YtGkTXnzxxSHvmwsK4+LivJ6/5Grp0qX4+c9/js7OTgQHB6OtrQ0fffQRtm/fzh+zcuVK/v+npaXhySefxNtvvz1kYMQ95oqKCgqMiM/MTI9CYrhu0O00BkBCuKN0nxBPUWDkQ0FqJYqfXziiYw+XN+GeTUeGPW7zf10+7Is/SO3d0QuuAQIAtLe3Y/Xq1fjoo49gMBhgtVrR1dWFysrKIc8zZcoU/v/r9XqEhYXxs7SGc/r0aaSkpPBBEQDk5OQgIiICp0+fxuWXX47HH38c9913H/7v//4PCxYswE9+8hNkZGQAAB599FE89NBD+PTTT7FgwQLcfvvtva6nr66uLuh0w3/6PHnyJGw2G7Kysnrdbjab+dEw7t73QCorK5GTk8N/bbVaYbFYeq16/e53v8Pvfve7AX/+hhtugFqtxs6dO3HnnXfi/fffR1hYGBYsWMAf88477+Cvf/0rysrK0N7eDqvVirCwsCGvi1vh4rbjCPEFpYLBqoIc/HLL8X7f4zoYrSrIoX5GxCsox8iHGIZBsEY1on9XZcYiMVyHwV7WDBz9Oq7KjB32XN7uNqzX63t9/eSTT2L79u148cUXsX//fhQWFmLy5Mno7u4e8jzcthb/mBgGdrvda9e5evVqnDp1CjfeeCO+/PJL5OTk8Csi9913H86fP4+f//znOHnyJGbMmIG//e1vg54rJiYGzc3Nw95ne3s7lEoljh07hsLCQv7f6dOn8Ze//GVU9z2QpKSkXud//vnnMWPGjF63/fKXvxz05zUaDX784x/jrbfeAgC89dZb+OlPfwqVyvHZ6NChQ1i6dCluuOEGfPjhhzhx4gSeeeaZYZ/TpiZHsmtsbKxbj4cQd81Ii4JygL9YCeE6bFw2nfoYEa+hwEgkuE9EAPoFR2L7RPTNN9/gnnvuwa233orJkycjISEBFy5c8Ol9Tpw4EVVVVaiqquJvKy4uRktLS6+VlKysLPz617/Gp59+ittuuw2bNm3iv5eSkoJf/vKX2LZtG5544gm8/vrrg97ftGnTUFxc3Os2jUYDm83W7zibzYb6+nqMHz++1z/XLbfB7luj0QBAv/P2pVKpep07Li4OQUFBvW4bLldr6dKl2L17N06dOoUvv/wSS5cu5b938OBBjB07Fs888wxmzJiBzMxMVFRUDHk+ACgqKoJarcakSZOGPZYQT7x7tAo2OzAlORxzMxyrsUtnpeDAU9dQUES8igIjEcnPTcTGZdOREN57C0dsn4gyMzOxbds2FBYW4vvvv8fPfvYzr678DGTBggWYPHkyli5diuPHj+Pw4cO46667cPXVV2PGjBno6urCww8/jL1796KiogLffPMNjhw5gokTJwIAHnvsMezZswfl5eU4fvw4vvrqK/57A1m4cCEOHTrUK2BJS0tDe3s7vvjiCzQ0NKCzsxNZWVlYunQp7rrrLmzbtg3l5eU4fPgw1q5di48++mjY+x47diwYhsGHH36IS5cuob293Wf/DX/0ox8hISEBS5cuRXp6OmbNmsV/LzMzE5WVlXj77bdRVlaGv/71r73yjwazf/9+XHXVVT5JGieEY7OzeOs7x1b9z2ePxYQExxZviE4tig+LRF4oMBKZ/NxEHHjqGmy9fzb+cudUbL1/tug+Ef35z39GZGQkrrjiChQUFGDhwoWYPn26T++TYRh88MEHiIyMxI9+9CMsWLAA48aNwzvvvAMAUCqVaGxsxF133YWsrCzccccdWLRoEdasWQPAsSKzfPlyTJw4Efn5+cjKysL//M//DHp/ixYtgkql6tWG4IorrsAvf/lL/PSnP0VsbCz++Mc/AgA2bdqEu+66C0888QQmTJiAW265BUeOHOEr9Ia67zFjxmDNmjV4+umnER8fP2TyuqcYhsGSJUvw/fff91otAoCbb74Zv/71r/Hwww9j6tSpOHjwIJ599tlhz/n222/j/vvv99UlEwIA2Hf2Ei42dyE8SI2CvCREhzhWWpvah97qJWQ0GNadpjcBrq2tDeHh4Whtbe2XlGoymVBeXo709PQRJe0S8duwYQN27tyJPXu803JBbj755BM88cQT+OGHH/hcpb7odUG84Rebj+CLM/X4xZXpePamHGw9XIkV207i2uw4/POey4W+PCIBQ/397ouq0ggZxIMPPoiWlhYYjUZ+LAjp0dHRgU2bNg0aFBHiDRebO/FliaN6deksxypstN6xYtTQQStGxPvoHY2QQahUKjzzzDNCX4Zo/fjHPxb6EkgA2Hq4EiwLzB0fjXGxjvYU/FZah1nISyMyRTlGhBBCRKnbasc7RxyVqMtmjeVvj9JrAVCOEfENCoy8jFK2COlBrwfiiT2natHQ3o24UC0W5MTzt0c5t9I6um0wWYZudUGIuygw8hKueSF1ACakB/d66Nvck5CR2PKto5fWnTNToXbp7himU0GtdJTpN1GeEfEyyjHyEqVSiYiICH7ERXBwsNc7UBMiFSzLorOzE/X19YiIiIBS6d0xNUT+SuuM+K68CUoFgyUzU3p9j2EYRAZrUG80o6mjG0kR1EeLeA8FRl7EdToe6fwvQuQuIiKiVwdwQkbq386GjtdmxyExvH/gEx2iRb3RjIZ2SsAm3kWBkRcxDIPExETExcXBYrEIfTmECEqtVtNKERmVzm4r3j92EQCwbPbYAY/hSvZpK414GwVGPqBUKukPAiGEjNLOwhoYzVaMjQ7GleNjBjwmigIj4iOUfE0IIUQ0WJbFlu8cSddLZ6VCMcgsNC4waqTAiHgZBUaEEEJE4/uLrSiqboNGpcBPLksZ9Dh+K416GREvo8CIEEKIaHAl+jdNTkSkM/gZSFQIrRgR36DAiBBCiCi0dHZj1/c1AIClgyRdc6Kd3a8baSwI8TIKjAghhIjCf45dhNlqx8TEMExPjRjy2J55abRiRLyLAiNCCCGCY1kWbzl7Fy2bnTpsg9woyjEiPkKBESGEEMEdLGvE+YYOhGhVuGXqmGGP55KvjWYrzFaal0a8hwIjQgghguOSrm+dNgZ67fAt9sJ0aiidpfzNHdRQl3gPBUaEEEIEVddmwqfFdQAG73Tdl0LhmJcGUAI28S4KjAghhAjq7cNVsNlZXJ4WiQkJoSP+uRiuZJ/yjIgXUWBECCFEMFabHVsPc0nXI1st4tBYEOILFBgRQggRzBdn6lHbZkK0XoP83AS3fpbGghBfoMCIEEKIYLik65/MSIFW5d7wbX4sCOUYES+iwIgQQoggLjR0YH9pAxjGMTDWXVHO7te0lUa8iQIjQgghgnjLmVt0dVYsUqKC3f75KEq+Jj5AgREhhBC/M1lseO9oFQBg2Sz3kq45MZRjRHyAAiNCCCF+9/FJA5o7LRgTEYT52XGjOgdVpRFfoMCIEEKI33FJ10tmpvAdrN0VzW+lUfI18R5RBkb79u1DQUEBkpKSwDAMduzYMeTxBw4cwNy5cxEdHY2goCBkZ2fj1Vdf7XVMWloaGIbp92/58uU+fCSEEEL6Kq5pw/HKFqgUDO64PGXU5+GSr9tMVlhsdm9dHglwww+kEUBHRwfy8vJw77334rbbbhv2eL1ej4cffhhTpkyBXq/HgQMH8OCDD0Kv1+OBBx4AABw5cgQ2W8+gwaKiIlx33XX4yU9+4rPHQQghpL8t3zlWixbmJiAuVDfq80QEqaFgADsLNHd0Iy5s9OcihCPKwGjRokVYtGjRiI+fNm0apk2bxn+dlpaGbdu2Yf/+/XxgFBsb2+tnXnrpJWRkZODqq6/2zkUTQggZltFkwY4T1QBGn3TN4ealNXZ0o5ECI+IlotxK89SJEydw8ODBQYOe7u5ubNmyBffeey8YZvC9bbPZjLa2tl7/CCGEjN6OE9Xo7LZhfFwIZo+L8vh80VSyT7xMVoFRcnIytFotZsyYgeXLl+O+++4b8LgdO3agpaUF99xzz5DnW7t2LcLDw/l/KSmj3wsnhJBAx7Istnzr6F20dFbqkB9MR6pnLAglYBPvkFVgtH//fhw9ehSvvfYa1q9fj61btw543D//+U8sWrQISUlJQ55vxYoVaG1t5f9VVVX54rIJISQgHK1oRkmdEUFqJW6bnuyVc0ZT92viZaLMMRqt9PR0AMDkyZNRV1eH1atXY8mSJb2OqaiowOeff45t27YNez6tVgutVuuTayWEkEDDlejfnJeE8CC1V85JvYyIt8lqxciV3W6H2dx/aXXTpk2Ii4vDjTfeKMBVEUJIYGpsN+OTk7UAgGWzPUu6dhVF3a+Jl4lyxai9vR3nzp3jvy4vL0dhYSGioqKQmpqKFStWoLq6Gm+++SYAYMOGDUhNTUV2djYARx+kV155BY8++miv89rtdmzatAl33303VCpRPnRCCJGl945dRLfNjrzkcExODvfaebnk6yZKviZeIsro4OjRo5g/fz7/9eOPPw4AuPvuu7F582YYDAZUVlby37fb7VixYgXKy8uhUqmQkZGBdevW4cEHH+x13s8//xyVlZW49957/fNACCGEwG5n8dZ3zqRrL64WAT05RpR8TbyFYVmWFfoipKKtrQ3h4eFobW1FWFiY0JdDCCGSsLekHvdsOoIwnQrf/W4BgjRKr537UFkjlrz+LcbF6vHlE/O8dl4iL+78/ZZtjhEhhBBx4Er0f3xZileDIsBlK41yjIiXUGBECCHEZ6pbuvDlmToAwNLZqV4/P5d83dJpgZXmpREvoMCIEEKIz7x9uBJ2FrgiIxoZsSFeP39ksAZcn8jmTovXz08CDwVGhBBCfMJis+PtI47GuN4s0XeldM5LA2g7jXgHBUaEEEJ84tNTdbhkNCMuVIvrcuJ9dj98L6N2qkwjnqPAiBBCiE9wna7vvDwFaqXv/txQk0fiTaLsY0QIIUSabHYWh8ubcLK6FYfON4IBcOdM7yddu4qmsSDEiygwIoQQ4hW7iwxYs6sYhlYTf5tGpcAPF1uQFBHks/ulFSPiTbSVRgghxGO7iwx4aMvxXkERAJitdjy05Th2Fxl8dt89K0aUY0Q8R4ERIYQQj9jsLNbsKsZQYxTW7CqGze6bQQvRIY6xILSVRryBAiNCCCEeOVze1G+lyBULwNBqwuHyJp/cP7eV1kCDZIkXUGBECCHEI/XGwYOi0RznLkq+Jt5EgREhhBCPxIXqvHqcu6JoXhrxIgqMCCGEeGRmehQSwwcPehgAieE6zEyP8sn9c1tpzZ3dPstjIoGDAiNCCCEeUSoYrCrIGfB7zjFmWFWQA6WCGfAYT3EjQVgWaOmkVSPiGQqMCCGEeGzu+BjoVP3/pCSE67Bx2XTk5yb67L7VSgUigtUAaDuNeI4aPBJCCPHYO0eqYLLaMT5WjxduyUW90Yy4UMf2ma9WilxF6TVo6bSgob0bmb4by0YCAAVGhBBCPGKzs9h88AIA4BdXjcOcjBi/X0O0XoPzlzpoxYh4jLbSCCGEeOTTU7W42NyFyGA1bp02RpBriKLu18RLaMWIECI73CDTeqPJr9s5geqfB8oBAEtnjYVOrRTkGqL0ju7XNC+NeIoCI0KIrAw0yDQxXIdVBTk+TQAOVN9XteBoRTPUSgZ3zRkr2HVQk0fiLbSVRjxis7M4VNaIDwqrcaiskXqIEEENNsi0ttXk80Gm3iSl19W/vnGsFhVMSUJcmG8aOI5EtLPJYyONBSEeohUjMmr0yZyIyVCDTFk4+ums2VWM63ISRL2tJqXXVW2rCR/94Ag2770yXdBr4XKMGinHiHiIVozIqMjlkzmRD6EHmXqD1F5Xbxy6AKudxaz0KOSOCRf0WqKdOUa0lUY8RYERcdtwn8wBxydzMS//E/kRepCpp6T2uurstuKt7yoBCL9aBLhWpVFgRDxDgRFxmxw+mRP5EXqQqaek9rradrwarV0WpEYFY8FE4TsqcjlGzZ0W2EUSPBJposCIuE3qn8yJPAk9yNRTUnpd2e0sn3T9X3PTRJGzxc1Ls9lZtHZZBL4aImUUGBG3Sf2TOZEnoQeZekpKr6uvz17C+UsdCNWq8JMZKUJfDgBAo1IgTOeoJ6IEbOIJCoyI27hP5oP9eRH7J3MiXzmJ4Rgo7okP0/p8kKmnpPS64ho63jkzBSFa8RQ3R4c4mzxSyT7xAAVGxG1S/2RO5Gv9F2dhZ4Erx0dj6/2zER7kmLj+1yXiDoqAoV9XHDG8rs7UtuHAuQYoGOCuOWmCXktflIBNvIECIzIq+bmJ2LhsOoL6tP+PD9OJ/pM5kaezdUZsP1ENAPhtfjbmZERjSrKjhPz8pXYhL23E8nMT8befTRvwe4unJonidfUv52pRfm4CUqKCBb6a3np6GVFgREaPAiMyavm5iciKD+l1299+Nk0Ub94k8Pz507NgWSB/UgKmJEcAALLiQwEAZ+ukERgBwATnNetUCvzlp1Px0NUZAIDdp2pR2dgp5KWhod2MHYU1AIBfiKBEvy8aC0K8gQIjMmosy+JcveMPTnJkEADgTK1RyEsiAeqHiy3YfaoWDAM8fn0WfzsXuJfWS+f3sqTOca3ZiWFYPG0Mfps/AXPGRcNksePZD4rAssKVov/720p0W+3IS4nA9NRIwa5jMLSVJm1iGYUjnqw5IjnVLV3o6LZBrWRw4+RE/H3feRTXtAp9WSQAvfLpWQDArVPH8KtEAJDJrxhJJzA66/xwwa0cMQyD39+ai0Xr9+Prs5fw0UkDbpqS5PfrMltt+L9vKwA4VosYRnw5hFzydUM7VaVJjZhG4dCKERk17o9NRmwIv3VxqqZNwCsigejb843Yd/YSVAoGjy3I6vW9zDjHilFdm1kyvW24FaOshJ4ALyM2BA/Nc2yprdlVjDaT/x/LzsIaNLSbkRiuw6LcBL/f/0jQVpo0iW0UDgVGZNS4vI3M+FBMSgoD4NhKs9jsQl4WCSAsy+KVPSUAHKXjqdG9k4FDdWokOZs+lkpk1Yh7XfXN33toXgbSY/S4ZDTzj9lfWJblS/TvmpMGtVKcfzpoK016xDgKR5y/3UQSuBWjCfEhSI0KRohWhW6rHWUSqQAi0rf37CUcrWiGVqXAI9dkDngMt51WIoHAyGSxoaKxA0DPVhpHp1biD7fkAgD+79sKFFa1+O26Dp1vxJlaI4LUSvxsZqrf7tddVJUmPWIchUOBERk1LjDKjA+FQsEgJ9GxanSqmrbTiO/Z7T2rRXdfkYb4sIE7QvMJ2BKoTDtX3w47C0QEqxEbqu33/SvGx+C2aWPAssDvtp2E1U+rs1yJ/o8vS0Z4sNov9zka/Ly0jm5Bk9TJyIlxFA4FRmRUbPaeijQu2TXHuZ1GeUbEHz4pqsWpmjaEaFX4pbOkfSBSSsDmrjErPnTQ5Obf3TgR4UFqFBvasPngBZ9fU3lDB744Uw/AMRdNzLgVI6udRVuXVeCrISMhxlE4FBiRUalq6oTJYodWpUCqs8lb7hhHM71TVJlGfMxqs+NPnzlWi+67Kp3/gzgQKfUyKqnrXZE2kJgQLVYsygYA/Pmzs6hu6fLpNW36phwsC1yTHYdxsSHD/4CAtColQp0jShpoXpokiHEUDgVGZFS4T7bj40L4EQVcAnaxoY2WsYlPbT9RjfOXOhAZrB620SBXmdbQbkazyHNPSgdJvO7rjhkpmDE2Ep3dNqzeecpn19PaacF7Ry8CEGdDx4FEhVACtpSIccQUBUZkVM4O8Ml2fFwINCoFjCYrqpp8+ymWBC6z1Yb1n5cCcFRqheqGznnRa1UYE+FoQCr27bSS2p6ttKEoFAxevG0yVAoGnxXXYc+pWp9cz9Yjleiy2JCdEIorMqJ9ch/exidg0yBZycjPTcQrP5nS7/aEcGFGTFFgREbFtVSfo1Yq+ECJttOIr7x9uArVLV2ID9OOeIjpBGdPoLP14t1OM5os/LbYcIERd8wDPxoHAFi98xTazd7NqbHY7HjDmcN0r0gbOg6EehlJU1qMY5U0MliNv9w5FVvvn40DT10jyIgpCozIqPArRgm9l/wnUQI28aHObiv+9uU5AMAj12RC12eI8WAy+co08a4YlTqDtrhQLSKHyJly9cg1mUiJCoKh1YRXPzvr1evZXVQLQ6sJMSEa3Jzn/07bo9XTy4hyjKSEG/Q8MTEMi6eOwZyMaL9un7miwIi4zWKz4/wlR6+VzLjen2x7AiNaMSLe98bBCjS0m5EaFYw7ZqSM+Oey4sRfmVZaN7JtNFdBGiVeWOzobbTpm3IUVXvvdcc1dFw6a+yIA1AxiNI72hxQLyNpOd/g+JsyLlYv8JVQYERGoaKxA902O/QaJZ+7wclJ4irTaMWIeFdrlwWvfV0GAHhsQSY0qpG/fXHBhph7GZXU9m5/MVLzJsThpimJsLPAM9tPeqVD8LGKZhRWtUCjVGDZ7LEen8+fYkIox0iKypwrphkiqHykwIi4jcsvGu9s7OhqYmIoGAaoN5pxyUhL2cR7/nf/ebR2WZAZF4LFU8e49bPj40LAMI5VhEaRDhgdbHt6JJ67KQehWhW+v9iKLc5Br57gGjounpo0YKNJMaOxINLUs2JEgRGRoJLanlEgfQVrVBgX41gKpe004i0N7WZ+a+eJ6ye4nXsQpFEiJdLRb0us/YxKRrGVxokL0+G3+RMAAC/vKUFd2+i7BF9s7sQnzqGd90qkRN8VjQWRHqvNzo/C4f5+CIkCI+K20vqh38An0XYa8bKNe8vQ2W3DlORwLJwUP6pz8KNB6sWXZ9Tc0c2vsGaOIjACgJ/NGou8lAi0m614flfxqK/lzUMVsLPAFRnRmOgc8yMl0c4cI0q+lo6q5i5YbCy0KkW/9AwhUGBE3DZcrxW+0SMFRsQLalq68H/O7aHfLJww6rJxfphsrfgCI24bbUxEEEKcnZvdpVQwePHWXCgVDD46acBXzjEe7ugwW7H1cCUA6TR07Mu1wSM1mpUGriItPUbfLz1DCBQYEbeYrTZcaOwEMJIVI9pKI57725el6LbaMSs9CleOjxn1ecQ8TLYnv2h0q0WcSUnhuNc5z+zZD4rQ1W1z6+ffO1oFo8mKcTF6zJ8Q59G1CIXrY2SxsTB6ubcT8Q2uylkMidcABUbETeUNHbDZWYTpVIgPGzgpk1sxutDYCaPJ4s/LIzJT3tCBd50jKTxZLQJ6WkucrTeKbiXBk/yivh5bkIWkcB0uNnfhL1+UjvjnbHYWm5wNHf9rbpooPrmPhk6thF7jaC8gtso0m53FobJGfFBYjUNljV6pIJSDsktcRZrw+UUABUbETa7baIP9kYrUa5AU7piEfNogvm0LIh3rPz8Lm53FNdlxmJHm2RDJ8XEhUDBAS6cFl0RWmXZ2hDPSRkKvVWGNs7fR/+4/P+Ktwy9O16GisRNhOhVuvyzZ4+sQUs92mnie591FBly57kssef1b/OrtQix5/Vtcue5L7HYmugcybsVIDBVpAAVGxE38kMthlvxzaDuNeOi0oQ07v68BADxxfZbH59OplUiNclSmiWk7jWVZfivNGytGAHBdTjyuz4mH1c7id9tPwj6ClYl/feOo+lsyKxXBmtHlOYkF3+RRJCtGu4sMeGjLcRhae1cL1raa8NCW4wEfHJ1vcLwexdDcEaDAiLiJX/KPGzqyp9EgxFN/+vQsWBa4cUoin7fmKS4BW0wdsC8ZzWjptEDBOFa1vGX1zZOg1yhxrKIZbx+pGvLYUzWt+PZ8E5QKBnePcP6cmIlpXprNzmLNrmIMFJpyt63ZVRyw22qtnRY0OANYWjEikjTSsQUUGBFPHK9sxuen66BggMev83y1iDOBD4zEs2LEfdhIi9Z7dfRGUkQQHr/e0dvopU9OD9lwlesRdcPkRCSJoFzaU2LqZXS4vKnfSpErFoCh1YTD5U3+uygRKXOuFsWHaUddkeltFBiREevqtqGiyVmRNsxW2qQxjk/4pXVGmK3uVcYQ8sqeEgDAjy9L9mqlihiHyQ7X/sITd88Zi0lJYWgzWfGHjwbubVTfZsIu55alVEv0+xLTilG9cWTNNkd6nNzw+UUx4lgtAigwIm4ou9QOlnV8GosJGXpMQFK4DhHBaljtLM7WiufTORG/b8414GBZIzRKBR69NtOr585y2UoTS2VaqRcTr/tSKRV48dbJYBhgR2ENDpQ29Dvm/76tgMXG4rKxkZiaEuH1axBCND8vTfjk67hQnVePkxu+Ii1OHPlFAAVGxA09n2yHfwNnGMZlO40SsMnIsCyLl52rRT+blYpk5xgPbxkXq4dSwaDNZEW9SGb58Xl7HvYwGkxeSgTucg6CXbnjJEyWnhVck8WGf3/naOh471x5rBYBLsnXIlgxmpkehcRwHQZrfsAASAzXYWa6Z1WXUsU1d6QVIyJJZ4cZBdIXjQYh7vr8dD0Kq1oQpFZi+fzxXj+/VqXE2GhuZprw22l2O8tv603wwVYa54mFExAfpsWFxk78z1fn+Nt3nKhGU0c3xkQEjXrUihiJaStNqWCwqiBnyGNWFeS4Pf9PLnpK9WnFaEj79u1DQUEBkpKSwDAMduzYMeTxBw4cwNy5cxEdHY2goCBkZ2fj1Vdf7XdcdXU1li1bxh83efJkHD161EePQn7OupkLQStGxB12O4s/fepYLfqvuWk+m+qeFSeeBOzqli50dNugVjJI8+HwzDCdGqsKJgEA/mfvOfzn2EV8cKIa//2VowHkPVekQaUU5Z+DUYkSUWAEAPm5idi4bDofsHFCtCpsXDYd+bmJAl2ZsKw2Oy40iqvrNQCIIwW8j46ODuTl5eHee+/FbbfdNuzxer0eDz/8MKZMmQK9Xo8DBw7gwQcfhF6vxwMPPAAAaG5uxty5czF//nx88skniI2NRWlpKSIjI339cGSjpwmde4HRaYMRNjsbsJ+IyMjs+qEGZ2qNCNWp8OCPMnx2P1nxIdh9qifQFxI30HZcTAjUPg5MFuUmIDcpDEU1bXjyve/52xn05OTIhWtVGsuyHnVM95b83ESYLHY89k4hf9ukpNCADYoA4KLIhsdyRBkYLVq0CIsWLRrx8dOmTcO0adP4r9PS0rBt2zbs37+fD4zWrVuHlJQUbNq0iT8uPV0+e+q+1m62orqlC8DIk0TTY0IQpFaiy2JDeUOHV3u0EHmx2Ox49bOzAIBfXp2B8GC1z+6L72VUL3xgVFI7soap3rDnVC2KBtjWZgE88e73CNYoZfNHmgv0uq12dHTbRFMGzr2HTkkOxw8XW/H9xVZYbHafB8VixTV2FMvwWI4sn40TJ07g4MGDuPrqq/nbdu7ciRkzZuAnP/kJ4uLiMG3aNLz++utDnsdsNqOtra3Xv0DF5UHEhWoRETyyT5dKBYPsRMcbPm2nkaH859hFXGjsREyIBvdckebT++JWPM/VtQtemcYPj/VBRZorrsngUOTUZDBYo0KQmpuXJo4kewC42OxodzIvKxaRwWqYLPaAzsEsqxffNhogs8AoOTkZWq0WM2bMwPLly3Hffffx3zt//jw2btyIzMxM7NmzBw899BAeffRRvPHGG4Oeb+3atQgPD+f/paSk+ONhiNJop39z22nFAfziJ0MzWWz4q3PY6f+bNx56H3+6T4/RQ6VgYDRbh2y85w++7GHkKhCbDIqpySOnqsmxYpQSFYzLxjrSOI5ekM9/c3eJbRQIR1aB0f79+3H06FG89tprWL9+PbZu3cp/z263Y/r06XjxxRcxbdo0PPDAA7j//vvx2muvDXq+FStWoLW1lf9XVTV0W3054/KLuAnlI0WVaWQw3KTxFdtOwtBqQmKYFj+blerz+9WoFHyis5CVaTY7i3OX3MvbG61AbDLIbac1iWReGgBUOVeMUqKC+aHIRwI4MCoTYUUaINIco9HicoYmT56Muro6rF69GkuWLAEAJCYmIiend8nkxIkT8f777w96Pq1WC63WN5UxUtOzYuTekqdrZZpYkiCJ8HYXGbBmV3GvVYxOix17S+r9kueSFR+Cc/XtKK1rx7wJcT6/v4FUNHag22qHTq1ASpR3+zX1FYhNBsVWmWazs6hp6VkxUisd74XHKpoD9r2R62FEW2l+YrfbYTb37C3PnTsXJSUlvY45e/Ysxo4d6+9LkyQuMMp085NtVnwolAoGzZ0WwbctiDgMNmm8rcvit0njmXHCD5PlX1NxoT6v2AzEJoNi20qrbTPBYmOhVjJICNMhd0w4NCoFGtq7caGxU+jL87vWrp7hsek+bFUxGqIMjNrb21FYWIjCwkIAQHl5OQoLC1FZ6ejQumLFCtx111388Rs2bMCuXbtQWlqK0tJS/POf/8Qrr7yCZcuW8cf8+te/xrfffosXX3wR586dw1tvvYV//OMfWL58uV8fmxS1dlpQ1+YIMjPdrCzTqZX8z9B2GhHLpHF+NEi9cL2M+Io0H2+jAb2bDPYNjriv5dZksKfJoziSr6uccyaTIoKgVDDQqpTIS3akGgTidhq3WhQXqkWozndVqKMhysDo6NGjvUrwH3/8cUybNg3PPfccAMBgMPBBEuBYHVqxYgWmTp2KGTNmYMOGDVi3bh2ef/55/pjLL78c27dvx9atW5Gbm4sXXngB69evx9KlS/374CSIK2seExE0ql/gHGr0SJzEkgTMbQmfE3BmGve6cnd7erS4JoMJ4b23yxLCdbJsMhjtnOfYKJIcIy4wSnEZc8PlGR270CzINQmJyy8S2zYaINIco3nz5g35ZrV58+ZeXz/yyCN45JFHhj3vTTfdhJtuusnTyws4Pdtoo/sFnpQUjm3Hq2nFiIgmCXhstB5qJYOObhuqW7q8PpNtJLgGk+5uT3siPzcR1+Uk4HB5E+qNJsSFOrbP5LRSxBHbVlpVc09+EWeGszLtSEXgrhiJLfEaEGlgRMSFewMf7SwnKtknHLEkAauVCoyLCUFJnRGlde1+D4zMVkfTU8C3M9IGolQwmJMR7df7FIKY5qUBwEVuxSiqp8MzV7J//lIHGtvN/CpXIOiZkSa+FSNRbqURceFL9Uf5Bs5tpVW3dKFZJG9SRBhiSgLmVkCFSMAub+iA1c4iVKtCYrh8KsHERGxVaXypvksQHhGs4ScJHKsIrO20Mr4iTXwrRhQYkWGd9XD6d5hOjVTn8nGxgVaNAtlQk8b9nQTMJ2ALMEyWnzuYEBqQZdr+EK135hh1mAXvcA70bu7o6rKxjg8BRwMoMLLZWVQ4K/HEmGNEgREZUkO7GY0d3WAYeDTrbBIlYBOn/NxE/OmOvH63+zsJOEvAFaOzfMdr8f1RkAuuwaPJYkdnt03QazFbbahz5s2lRPYelnp5mjPPKIAq0y42d6LbZodGpUCSiIbHcijHiAyJ+6ORGhWMII1y1OeZlBSGT4pqKQGbAOjpiK7XKvHirZMFSQLmtobP1bfDbmf9OsSypM4/o0ACWbBGCa1KAbPVjqaObp+PmhlKdXMXWNZxTdwWH+dyZ2VaUXUrTBYbdOrRv89KBbeNNi5GL8rEf1oxIkMqHeUokL5oNAhxxXUAHhulx+KpYzAnI9rvb5Bjo4KhUSrQZbHhorNiyF883Z4mw2MYhk/AFroyrdKlVL/v1mlyZBDiw7Sw2Fh8X9UiwNX533mRjgLhUGBEhlQyylEgfXFbaecvtaNL4GVtIrxqZ2CUFCFc4rFKqeDfmP25ndbVbeP/UGa5OZSZuCcqRBxNHntK9ftvGzEMgxkBlmfEz0iLEedWMgVGZEilXlryjwvTISZECzsLnK6lVaNAZ2jlAiNh8wt6OmD7LzA6V98OlnVUTcUEUHm2EKL04mjyyJXqD9YWYkaA5RnxFWlxtGJEJIZlWZTUei8XoicBmwKjQFfT4khEFT4wcnxiLfVjZVpPfpE4Py3LiVh6GfGl+oMMC+byjI5VNMPu43E4YnCeVoyIVNUbzWgzWaFUMF7ZC+5p9EiVaYGuZytN2MAoM97/w2Qpv8h/xJJjxJfqRw78+56dEIpgjRJGk9Wvq5dCcAyPdWxtUo4RkRxutSgtOhhaleeVEpSATTj8VprAzQ2zXCrTfD24lsOvwlJ+kc9xOUZCb6UNt2KkUiowPZXbTpN3npGYh8dyKDAigzrr5ZJibsXoTK0RFpvdK+ck0mOzs6htFcdWWmpUMF/SzQ359DVv5e2R4fVspQmXfG00WdDSaQEweGAE9OQZHZN5npHYK9IACozIELwdGKVGBSNEq0K31c4n35HA09BuhsXGQqlgEBcqbPKxUsHwjUv9sZ3WZrKgxhkUZnnYAoMMj0u+FjLHiNtGiwxWI2SIXkpcZZrsV4wauOGx4swvAigwIkPgxxZ4KTBSKBjkJDoTsKtpOy1QcT2M4kO1UCmFfwvifr9L630frHOrRQlhOoQHi3MbQU6iRJBjNNw2GmdqagSUCgbVLV38a0SOyuodK0ZiHAXCEf5diYgSy7L8m7inPYxc5VBlWsATS0Uax5/DZEtqe2akEd8TQ1VaVVP/4bEDCdGq+A+Ocu5n1LNiRFtpRGKqW7rQ0W2DWslgbLT3foFpZhqpEUlFGofb0vLHMNmeijTxflqWE25eWme3TbDGslxX9eQBmjv2Jfc8I5udxYUG5/BYkZbqAxQYkUFwfV3GxYRA7cXtDq4yrdjQJoqJ18T/akTS3JHDbaWV1bfD6uOiAC4wyqTEa78I0aqgcb5/NQqUgD3SFSNA/nlGrsNjxwzSukAMKDAiA+Kb0Hl5yT8zPgQapQJGk5VPSiSBpUYE40BcJUcGIUitRLfNjgofV6ZRDyP/YhiGzzMSajttpDlGQM+K0ZnaNhhNFp9elxC4irT0aHEOj+VQYEQGxFekxXl3uVOtVCDLmbNE22mBic8xChfHJ0aFS2VaqQ/zjBrbzWhw9tPJpK00vxEyAZtl2WGbO7qKD9MhNSoYdhY4Udni46vzP64aWcz5RQAFRmQQZ320YgQAkxKp0WMgE1uOEeCagO27PCPu3KlRwQjWDF62TbyLyzNqEqDJY2NHN7osNjAMRrx1NGOsY9XoqAzzjLjhsWKuSAMoMCIDsNlZnKv3bqm+q0ljKAE7UJksNv6Tu1i20gCXYbI+XDE6SzPSBCHkVhqXX5QQphvx9IAZafLNMzpPK0ZEqqqaOmGy2KFVKZA6gn1xd9Ew2cBlcDY3DNYoER4knj4+/hgmW0IdrwUR7Wzy2CBA8nVVM7eNNvL30cudeUYnqpplNyHgfAPX9VrcHw4oMBIBm53FobJGfFBYjUNljX6b2TQY7pPt+LgQnyTIZSeEgWEcQ2ovGYVr1U/8z3UbjWHEk3yZ6SzZP9/Q7rM/Rmdrub5gFBj5k5BbadyK0UhK9TkZsSEID1LDZLGjWEYfHttMFv79XuwrRrTRLbDdRQas2VXMf5IGgMRwHVYV5CA/N1GQa/J15Yxeq0J6jB7nL3XgVE0r5k2I88n9EPHhAqNEgYfH9jUmIgjBGiU6u22oaOzAeC+P62BZllaMBCLkVtrF5pGX6nMUCgYzxkbiizP1OHKhCXkpET66Ov/iKtJiQ7UIE+nwWA6tGAlod5EBD2053isoAoDaVhMe2nIcu4sMglwXlyTqy14rXD8j2k4LLFxF2hgRJV4Djj9GmfG+a/RY12aG0WSFUsGI/tOy3AhZlcZXpLmZksDlGR2VUZ4Rn18UI/7ffwqMBGKzs1izqxgDbZpxt63ZVSzIttpZH4wC6YvLM5LTUjEZnhgr0jhZPhwmy60WpUUHjzgJl3iHkGNB+B5GbjYz5PoZHa1olk0jXG7FKMPLLWB8gQIjgRwub+q3UuSKhSNR9XC5f0s2LTY7/wuc6cPp3zQaJDCJreu1K36YrA9WjCi/SDhCbaXZ7Cyqm0e3YjR5TDg0SgUa2s2oaPRt01F/KaMVIzKceuPgQdFojvOWisYOdNvs0GuUPt3u4LbSLjR2yrLDKxkYv2IkshwjwLfDZCm/SDjRIY6qtHazFSaL/+alGVq7YLWzUCsZxIe59/uuUysxJdnxHnlEJv2MzkukhxFAgZFg4kJH9kIZ6XHewuVXjI8PhcKHLduj9Bo+Afe0wfdTzYnwWJbt6Xot4hWj8oYOdFu9W5lWSoGRYMJ0KqiVjvcyf64acflFYyKCRlXdy+UZHauQfp6Rzc6ivJECIzKMmelRSAzXYbCXCwNH5c7M9Ch/XpZfp3/Tdlpgaem0oMv5iT1BhCtGieE6hGpVsNpZlDv7rXiD3c7yHzgoMPI/hmEQGez/7TR3ZqQNhOuALYcVo+rmLnRbxT88lkOBkUCUCgarCnIAoF9wxH29qiDH74P2zvrxk20OVaYFFC6/KCZEA51afAnIDMNgvA+20y42d6HLYoNGqUBatPcbppLhCVGZdpHrYeRGqb6ry5yBUdmlDjS2S7vfW1mD44NBWnSwqIfHcigwElB+biI2Lpve79NzeJAaG5dNF6SPkT9K9TnUATuwiHkbjZMVxyVgey8w4vKLMuJCoFLSW64Q+CaPfux+zXe9dqO5o6tIvQaZzgouqW+nSSm/CPAwMOrs7ERlZSU6OnovOzc3N+Ppp5/GTTfdhP/3//4fysrKPLpIOcvPTcSBp67B1vtnY+GkBADA1JRwQYIis9XGbyH4qrmjKy4wKq0zwmz1X1IkEUZP4rV4AyNfDJP15/Y0GViUcyxIox+7X3Ndr91p7tgXV7Yv9cCoTCIz0jgeBUYvvPAC0tPTcebMGf42s9mM2bNn4+WXX8bHH3+M1157DXPmzIHBIEyzQilQKhjMyYjGbxZOAAAcONcoSM+N8oYO2OwsQnUqxIdpfX5/YyKCEB6khtXO+nRGFREHbistUUTDY/vih8nWe2/FiAuM/LEKSwYWLcBWmqc5RgAwYyw3UFbaeUY9zR2l8eHAo8Doyy+/REZGBi677DL+ti1btqC0tBTz58/Hnj178Oijj6KhoQGvvvqqxxcrd+PjQjApKQxWO4uPT/o/kCyp7RkF4o85VgzDUAJ2ABFr12tXXGBU0djptVVM19cVEQbf5NFPK0Ymiw11bY5tO3ebO7q63FmZdrK61a+tBrxNSs0dAQ8Do8rKSmRmZva6befOnWAYBps2bcJ1112H9evXIysrC5988olHFxoobs5LAgDs/L7G7/dd6sf8Ig7lGQUOMXe95sSHaRGqU8FmZ/k3c09YXRqmUnNH4USF+HfFqNr5ux6sUfKJ36OREhWE2FAtLDYWP1yU5odHo8mCeokMj+V4FBg1NzcjIiKC/5plWRw4cABTpkxBSkoKf3teXh6qqqo8uauAUeAMjI5caOL/kPhLiQC5EDQzLXAYJBAYMQzTs53mhQTsC42d6LbZEezjhqlkaD1jQfyTfO2aX+TJ6jvDMLg8Tdpl+9wHg5gQ8Q+P5XgUGCUkJKC8vJz/+tixY2hubsbVV1/d6zh/bMvIRVJEEGamRYFlgQ9/8O+qkRBN6LgVo9OGNkHmwhH/sNrsqG1zVqWJsIeRK2+OBuHzi+JCfNowlQyNS772V+6mpxVprrg8o6NSDYycpfoZElktAjwMjKZOnYrDhw9jx44dMBqNeOGFF8AwDG666aZex5WWliIpKcmjCw0kBVP9v53W1W1DhfNTTpYfl/zHxYZAp1ags9uGC43ea6pHxKXOaIadBdRKBjEhvk/s90SWF3sZcflF1NhRWP7uY+RpDyNXl7t0wLZL8MNjWb3jfX2cREr1AQ8Do9/+9rcAgNtvvx0RERHYtWsX8vLycM011/DH1NXV4fvvv++VoE2GduPkRKgUDIqq2/gyR18ru9QOlnW8gfjzD5dSwSA7gfKM5I7bFk4MDxL9ygm/YlTvvRUjyi8SVowzx8hosvqlNYg3KtI4ExNDEaxRos1k9crvpL8F3IrRFVdcge3bt+PKK69EdnY2li1bhp07d0Kh6Dnt1q1bERoaivz8fI8vNlBE6TW4MjMGALCz0D+rRj2fbP0f1VNlmvz1BEbi3kYDenoZVTR2eFwJ5M9O8mRwYTo133G5ucP3Q6u5OWmpXgiMVEoFpqVGAJBmnpHUmjsCXuh8XVBQgK+//hqnTp3Cm2++ieTk5F7ff+yxx9Dc3IylS5d6elcBhatO2/V9DVjW98unXN8WId7AuQTsYloxki0plOpzYkO0iAhWw87CoxVbk8WGC43O7WkKjASlUPTMS2v0QwJ2z4qRd37fpZpnZLOzON/AbaUFyIoR8Z3rJyVAq1LgfEOHX7aYzgqYC+Fasu+PIJD4nxRK9TkMw/CjQTzJMzp/ydEwNcxPDVPJ0Hoq03ybZ2Q0WdDS6ViV8qTrtSuuA/ZRiXXArmlxDo9VKrySb+UvHgVGdXV12LdvH+rq6nrdXlZWhjvvvBO5ubm44YYbcOjQIY8uMhCFaFVYMDEeAPBBYbXP70/I6d8TEkKhVDBo6ujmK5eIvBgk0PXalTdGg7jmF1FlrvCi/BQYcdtoUXoN9FqVV845LTUSCsYxkJh7LUkBt+KaFiON4bEcjwKjl156CfPnz0dra09uSFtbG6688kq89957KC4uxu7du7FgwQKUlpZ6fLGBhutp9OEPBp9WI7SbrXxDMiFyjHRqJcY7959PVdN2mhxVS2CArKuekv3RrxiVUH6RqPBNHn3c/ZrfRvOg43VfIVoVcpwr60cvSGfVqMyZXySVUSAcjwKjvXv3IicnB1lZWfxtmzdvRl1dHZYsWYKSkhL8+c9/RldXF/70pz95fLGBZt6EWIRqVTC0mnyadMe9+ceFahERPPourZ6gDtjyxm2lSSHHCPDOilEpVaSJSozePzlGXHPHZC8kXrvi8oykNFD2vMSGx3I8Coyqq6sxbty4Xrd99NFHUKlUWL9+PTIzM/HYY48hLy8PX3/9tUcXGoh0aiXycxMAAB/4sKcR18hOyDfwHKpMk60OsxWtXY6cCylUpQE9qzxVzZ3o6h5dZVoJ39yRAiMx8FeTx4tcc0cv59TMkGAHbClWpAEeBkZGoxHBwT1Pvs1mw6FDh3DZZZchJiaGvz07OxsXL1705K4C1s3OZo+fnDTAYrP75D7E8AZOo0Hki8uJCNWpECqRkQAxIVpE6TVgWeDcKHrHdJitfK6JENvTpD+/baU1ebcijcOtGJ02tKHdbPXquX2lLBBXjJKSknDmzBn+6wMHDqC9vR3z5s3rdZzVaoVGI8wWjdTNGReNmBANmjstOFDa4JP76Om1ItwbOLdiVN3ShZZO/3SnJf5RLaFSfVeZcaPvgM014osJ0SJa5J2+A4W/qtIqXeakeVNCuA7JkUGws8CJSvFvp/UeHiutDwceBUZz5szBDz/8gPXr1+PkyZNYuXIlGIZBQUFBr+NOnz6NMWPGeHShgUqlVOCmKY5VI19Vp/GBkYBbaeFBav4TFvUzkhcpleq74ofJ1rsfGPVUpEnrD4Kc+aMqjWXZnq00L+cYAT3jQY5IIAG7vKFneGx4kDRWijkeBUYrVqyAVqvFE088galTp+Kbb77BvHnzcMUVV/DHXLhwAcXFxZg1a5bHFxuouOq0T4vrRp3vMJjWTgvq2hxRPfcJWSiTEmk7TY4MEup67Yr7oDCaYbJcXzDKLxKPaD/MS2to70aXxQaGAZJ80JqC72ckgTwjqW6jAR4GRpMmTcKBAwewbNky5OfnY+XKldixY0evY/bs2YO8vDzccsstntxVQJueGoHkyCB0dtvwxZm64X/ADdyn4TERQYLnf9BoEHmSWqk+J8uDrbQSqkgTHW5Ls7XL4rN8Ta5UPyFMB61K6fXzcytGJypbfPYYvKUn8Vp6gZHH3aemT5+ON954Y9DvP/jgg3jwwQc9vZuAxjAMbs5Lwv/sLcMHhTX81po3cG/6mSJIEJ00hkr25UhqpfocbivtYnMXOsxWt5r10Yw08YkIUkPBAHYWaO7oRlyY91d0qnyUX8QZHxuCMJ0KbSYrThvaMCU5wif34w1SrUgDaCSIZHDVaV+XXOJLn72BW/KfIII3cK4yrexSu9e3DIlwuKo0qa0YReo1iHGuMrhTmea6PU0VaeLRe16ab7bTuPyiZC9XpHEUCgYzJJJnFLBbaZy6ujqsXbsWN9xwA/Ly8pCXl4cbbrgBL730Ur9xIWR0shPCkBUfgm6bHXuKar12Xq6BXaYIAqO4UC1iQjSws8CZWlo1kgO7nUVNq2MrTWo5RkBPYOPOdhq3PZ0UrhN8e5r05usEbF+vGAHSyDOy21k++VpqXa8BLwRG77//PrKysrBy5Urs3r0bJ0+exMmTJ7F7924888wzmDBhAt5//31vXGvAWzzVUdn3wffeq07jq2dEEBgxDIMc6mckK40d3ei22sEwjnJjqeEr09wIjEpqha/yJAOL8nECNj8OxAcVaRyun9HRimbRDt2ubumCmR8eK62VYsDDwOjo0aNYsmQJOjo6cOutt2L79u04ceIECgsLsWPHDtx2221ob2/Hz372Mxw9etRb1xywCpy5RYfKGlFv9HzYakO7GY0d3WAYYLzAFWkcGg0iL9w2WnyoDmql9HbuRzMaREwfNkhv0XyTR9+MBeGaenpzTlpfU5LDoVEqcMlo5nsmiQ23jTY2OhgqCb7uPbritWvXwmaz4b333sN//vMfLF68GHl5eZgyZQpuvvlmvPfee3jvvfdgsVjw0ksveeuaA1ZqdDCmpkTAzgIf/WDw+HzcG3hqVDCCNN6voBgNLjAqpso0WeASrxN9ULrsD6MZJkuJ1+IV7cOxIDY7y/+++3LFSKdWYnKyY2VdrHlGXOK1FPOLAA8DowMHDuCKK67ArbfeOugxt956K+bOnYv9+/d7clfEafFUrtmj57PTuP4sYuq1wiVgn6k1wiryclQyPKmW6nOynK+NmlYTjKbhix5Ylu3ZSqPASHR8uZVmaO2C1c5CrWQQ74OKN1czxjryjI5ViDPP6HyD42+LFCvSAA8Do9bWVqSmpg57XGpqKlpbaQXAG26ckggFAxRWtaCy0bNl1BIRducdGxWMEK0KZqsdZc5PHUS6pFqqzwkPViMu1LHKUDqCyrSG9m40d1pEtT1NenBbaU0+mJfGbaONiQiCUsF4/fyuxF6ZVlbPrRhJ8zXgUWCUkJCAEydODHtcYWEhEhISRnzeffv2oaCgAElJSWAYpl/TyL4OHDiAuXPnIjo6GkFBQcjOzsarr77a65jVq1eDYZhe/7Kzs0d8TWIRF6rDnIxoAMCuHzxbNSoV4ZK/QsEgJ5EaPcoFl2MkxYo0jjvbadw22lgRbU+THr6sSvNH4jXnMueK0bn6dp/PfhsNbsUoILfSFi5ciJKSEvzud7+Dzda/7wzLsli5ciXOnDmD/Pz8EZ+3o6MDeXl52LBhw4iO1+v1ePjhh7Fv3z6cPn0aK1euxMqVK/GPf/yj13GTJk2CwWDg/x04cGDE1yQmi/Oc1WkezE4T85J/DiVgy4bUt9IA9xKwKb9I3Hq20ryffH2xyX+BUZRew3eUPlYhrlWjdrOV7+OVIcFSfcDDztfPPvsstm3bhnXr1mHr1q244447kJaWBgCoqKjAe++9hwsXLiA6OhorV64c8XkXLVqERYsWjfj4adOmYdq0afzXaWlp2LZtG/bv348HHniAv12lUrm1ciVWC3MTsHJHEc7WteNMbRuyE8LcPke90Yw2kxVKBSO6qJ5LwC6qphUjqZP6VhrgXsn+WRoFImpc8rUvcoyquOGxPuxh5OrytCiUXerA0YomXJcT75f7HInzzoq0mBANwoOl2cfLo8AoOTkZX375JZYuXYqioiK8/PLLYBjH3irXX2Hy5Mn497//jeTkZM+vdoROnDiBgwcP4ve//32v20tLS5GUlASdToc5c+Zg7dq1Q+ZImc1mmM09nyza2sSxghEepMbVE2LxWXEddhbWIDvf/cCIX/KPDvbJTB9PcAnYxYY2sCzL/04RaTFbbbhkdLx+pLxi1LOVNvyKEbcKK4aGqaQ/LseopdMCq83u1VJyvrmjj7pe9zUjLQpvH6nCUZHlGfEVaRJdLQK8MCtt8uTJ+OGHH7B3717s378fNTWOvJekpCRcddVVmDdvnqd3MWLJycm4dOkSrFYrVq9ejfvuu4//3qxZs7B582ZMmDABBoMBa9aswVVXXYWioiKEhg78JrZ27VqsWbPGX5fvlsVTkxyB0fc1+M3CCW4HDyUiGgXSV2Z8CDRKBYwmK6qaupAa7Z9PYMS76lodQZFWpUCkRD85Aj1babVtJrR2WRAeNPBjYVmW324T4+uKAJHBGjAMwLJAc6cFsc7Eem/gc4z8tmLkyDM6ebEVJosNOrU4PuByK0YZceLaiXCHx4ERZ968eYMGQf/6179w8eJFPPfcc966uwHt378f7e3t+Pbbb/H0009j/PjxWLJkCQD02pqbMmUKZs2ahbFjx+Ldd9/FL37xiwHPt2LFCjz++OP8121tbUhJSfHpYxipa7PjodcocbG5C8crW/hkvJEqFdEokL7USgWyEkJQVN2GUzWtFBhJVLXLNpqUV/3CdGokhutgaDXhXL0Rlzk7D/dV02pCu9kKlYJBeox0/yjImVLBICJIjeZOC5o6ur0WGJksNj6vxh85RoCj/1xMiBYN7WacrG7F5WkD/176W5kMVoz80pLy9ddf98vKS3p6OiZPnoz7778fv/71r7F69epBj42IiEBWVhbOnTs36DFarRZhYWG9/olFkEbJ7yvvHEUSdonIu/NOSqTRIFIn1eGxA8nk84wG307jtqfHxeqhUUmv22+g8EUCNvchQK9R+m11lGEYftXoiIjmpkl5eCxHtq9eu93eKz+or/b2dpSVlSExMdGPV+Vd3Oy0j04a3GqGyLKsS6m+OKP6SWOoZF/q+K7XEi7V52Q5exJxW9ADOUv5RZLgi+7XVS4Vaf5cHeX6GYklz8h1eKxUmzsCXtxK86b29vZeKznl5eUoLCxEVFQUUlNTsWLFClRXV+PNN98EAGzYsAGpqal8X6J9+/bhlVdewaOPPsqf48knn0RBQQHGjh2LmpoarFq1Ckqlkt9qk6IrM2MQGaxGQ3s3Dp1vxFWZsSP6ueqWLnR026BWMkgT6ZI/zUyTPjmU6nP4BOz6wQMjsa/CEoeeeWleDIycFWnJfsov4vR0wG6G3c5C4ePGksPhhseqlYwkh8dyRBkYHT16FPPnz+e/5vJ87r77bmzevBkGgwGVlZX89+12O1asWIHy8nKoVCpkZGRg3bp1ePDBB/ljLl68iCVLlqCxsRGxsbG48sor8e233yI2dmTBhBiplQosmpyIt76rxAeFNSMOjLj8onExIaId7JmdEAaGcbQVuGQ0ezVJkviHHEr1OSPpZUQ9jKTBF2NB/F2RxslJCkOQWonWLgvOXWoX/HfvvHO1aGy0XpLDYzmiDIzmzZvHl/sPZPPmzb2+fuSRR/DII48Mec63337bG5cmOovzkvDWd5XYU1SL39+SO6LKBO6TbZaIe63otSqkx+hx/lIHTtW0Yt6EOKEvibiJ73ot0QGyrrjtsUtGM1o6uxERrOn1fZudxTnnyBDqYSRu0Xz3a+/lGPGBkZ9XjNRKBaalRuBgWSOOXGgSPjDiKtIknF8EyDjHKFBcnhaFxHAdjGYr9pZcGtHP8J9sRT7LietnRNtp0sOyLKqb5ZN8HaJV8StfA60aVTV1wmSxQ6tSINVPVUlkdHwxFsSf40D64rbTxJBn1JN4Le6/LcOhwEjiFAoGN01xJJDv/H5k1WlnJbBiBPTkGRVTYCQ5bSYrOrodY4KSwqUfGAGu22n984y4VdjxcSE+HyBKPBMV4ux+7c0cI+cAWX9vpQEuCdgVwlem9TR3DKAVI6VSOap/hw8f9tX1E/RUp31xuh5Gk2XIY12X/IVedh1OTwI2VaZJDbeNFqXXyGaY6lDDZM+KuGEq6S3ayytGbSYLWrsc77v+3koDgGmpEVAwjuCsttXk9/t3xQVGGSLfjRiOW4ERy7Kj/kd8Z1JSGMbF6GG22vFZcd2Qx0ppyZ/bSrvQ2Il3jlTiUFkjbHb6XZICOZXqczLjBk/APst92BD5KixxqUrzUmDE5RdF6TXQa/2fthuqU2NiouNDpJCrRu1mK2rbHIGZVIfHctwKjOx2+6j/2Ww2Xz2GgMcwDArykgAAHxTWDHnsWQkt+R8ubwR3iU+9fxJLXv8WV677EruLDMJeGBmWnEr1OUOV7NOKkXRwOUbNnd1e+aDFb6MJWJ4uhjyjcudqUbReusNjOZRjJBM3T3UERgfONaCxffBqi7MS6bWyu8iAh7YcR9/3rdpWEx7acpyCI5EzyKhUn8PlGDW0d/fahum22vmk00yRNkwlPSKdFYUsC7R0er5qdNGZeJ0s4Aq8GPKMzjdwFWnSfw1QYCQTGbEhyB0TBpudxcdFtYMed1bEM9I4NjuLNbuKMdBnOe62NbuKaVtNxOS4lRasUfHJta4J2BcaO2C1s9BrlLIKBOVKrVTwg4C9kWckVKm+qxnO0SDFNW1oN1sFuYayeumPAuFQYCQjNzu304aanXZW5KNAAOBweRMMQyQRsgAMrSYcLhe+CoMMrEaGW2kAkBXXPwHbtcpTysNyA0m0F5s8cl2vhahI4ySGB2FMRBDsLFBY2SLINZQ5mztSYEREpSAvCQwDHLnQzA81dGW12fmqATFXpNUbR1ZZMdLjiP9xv39yC4wGGiZL+UXS481eRmJYMQIg+EBZviKNttKImCSGB+Fy517zh9/3T8K+0NiJbpsdwSJf8o8LHdn2y0iPI/5ls7Ooc1aniPn3bDS4ldYSlxUj7v+LeXua9NYzL82z7tcsy+Kic8VI6CpfIfOMHMNj5dHcEaDASHZuHqI67azLG7jQwwaHMjPd0c17sCtk4MhdmZke5c/LIiN0yWiG1c5CpWBkN+POtZcR14aEWz2iFSPpiNI7mzx6uGLU0N6NLosNDCP86iiXZ3SisgVWm92v913T2gWTxTE8VsjqPG+hwEhmbpicCJWCQbGhjW/kyJHKKBClgsGqghwA6BcccV+vKsgRfbuBQMVto8WH6WT3HGXEhoBhgOZOCxrau2Gy2HCh0bk9nSDu1xXp4a0mj9wokMQwHTQqYf+cZsWFIkSrRGe3Da99fd6vfd+4bTSpD4/lSP8RkF6i9BpclRkDANjZZzuNL9WXQBO6/NxEbFw2HQl9qpriw3TYuGw68nMTBboyMhyu67XcttEAIEij5LdMSuuMOFffDpYFIoPViA2R1+qYnEV5Kfmayy8SslSf82lxLSw2RyD0yqclfu37xs9Ik/goEA4FRjLE9TTaWVjdq+u4FEr1XeXnJuLAU9dg6/2z+U94f7g1l4IikeNL9SPkmQOWGcclYBt7bU9TRZp0cDlGTR7OS+Pyi4ROvOb6vpmtvbfQ/NX3jZ+RJoP8IoACI1m6LicBOrUCFxo7cbLaMWfMbLWh3FlOKaVcCKWCwZyMaFw7MQ4AcFigigsycnIt1edwCdhn69v5xGspvaaI96rS+Io0AUv1xdD3rae5I60YEZEK0apw7cR4AMBOZxJ2eUMHbHYWoToV4sOkt+Q/e1w0AODb8xQYiV2NTEv1Oa4J2FypPs1Ik5ZoPvnas6o0LsdIyBUjMfR9K6unFSMiAVx12q4famCzsyhx6bUixSV/LjAqqm6F0WQR+GpGxmZncaisER8UVgfUANwaZ45Rkoy6Xrvixn6crWunijSJ4rbSmjstsHvwuuTnpAmYYyR037cO1+GxMlkx8v8oYOIX8ybEIlSnQl2bGYfLm1AqsfyivpIigjA2OhgVjZ04eqEZ87PjhL6kIe0uMmDNruJen+QSw3VYVZAj+xwpuW+lZcSGQMEArV0WtHY5gnQxd5In/XHz0mx2Fq1dFkQ6t9bcYbOz/OqokFtpQvd941I0ovUaRAS7/99RjGjFSKa0KiUW5SYAcFSn9eRCSPcNfHY6t53WKPCVDI1LhOy7vB0IA3C7um183oZcAyOdWtmrmV9EkBqhOmlPEw80GpUCoTrHusBoK9MMrV2w2llolArEC9hsdri+b4Bv+77xFWkyWS0CKDCStZvzxgBwVKcddSYtj5fwHvDsDMcLW8yBkRgSIYXEleqHaFUI08lzQXp3kaFX0NvSZfFbWTTxHk97GXHbaGMigwRtmDtU3zfOk9dn+aynWBlXkRYj3b8tfVFgJGNtXRYoGKCj24bmTseS/+PvfS/ZN3Auz+hkdSvaRJpnJIZESCFx22iJ4TpJ5rINR+iyaOI9PZVpo0vA5nsYiaDT82B937hgaO/ZBp/d93nnilFGHK0YEZHbXWTA8reOo+/CxCWjWbJv4InhQUiLDoadBb8CJjZCJ0IKTc4VaYG+Gig30c6GnA2j7GXEV6SJoLkj0Lvv21/unIqt98/Gew/OgVLBYNf3NfigsNon90srRkQS5PwGLvayfaETIYXGV6TJMDAK9NVAufF8K034Uv2+uL5vi6eOwZyMaEwfG4lHr8kEAKzcUcR/cPGW3sNjacWIiJic38B7AiNx5hkF+gBcfsVIhqX6gb4aKDeeNnmsaha+Im0kls/PQF5KBIwmK55873uP2hP0ZWgz9QyPFcnKmTdQYCRDcn4Dd+1nJMY8I9dEyMHIeQCunEv1A301UG48nZcmxhWjgaiUCrx6Rx6C1EocLGvEpoMXvHbuMueg8tSoYKhlMDyWI59HQnhyfgNPCNchPUYPOwscEemKV35uIl68Lbff7ZHBatkPwJXzVlqgrwbKDT8vbRTJ1yaLDfVGx89JYaVkXGwInrlxIgBg3e4z/Iw/T53nS/Xlk18EUGAkS3J/A589Tvxl+1zh7Pg4PWY5/zsvmz1W1kERy7IuydfSC7qHM1RZNPe1nFcD5SaKGwsyiuRrbnisXqNEZLA0elgtnZWK+RNi0W2147G3C9Hdp7JyNM47mztmUGBExE7ub+BiT8AGgK/O1ANw9JLKdzbaPFPrnU9pYtXcaYHJ4niz7Vs2LBeDlUUnhOtkvxooN9EebKW5VqRJpS0FwzBY9+MpiAxWo9jQhvWfn/X4nHJs7gjQSBDZ4t7A+46lSJDBWAouMDpV04rWLgvCg8T1ic1steGbc46+IfMnxKGz2woAKK5pE/KyfI5bLYoN1UKrUgp8Nb6Tn5uI63IScLi8CfVGE+JCHauvUv2gEaj4eWkd3WBZ1q0A5yLfw0j822iu4kJ1WHvbZPxyy3G89nUZ5mfH4fK00e8cnL/ErRhRYEQkQq5v4PFhOoyL0eN8QweOlDdhQU680JfUy9ELzejotiEmRItJSWFodwZG1S1daOnsls08ob7k3MOoL64smkgXl3xttbNo67Ii3I0tMalUpA0kPzcRP74sGf85dhGPv1uIjx+9alQjbTrMVv5Dt5x6GAG0lSZ7fftaSD0o4swScdk+t402b0IsFAoGYTo1/wZabJDvqpGcS/WJ/GhVSoRouXlp7iVgcxVpqRJIvB7IqoIcjIkIQlVTF174sHhU5+CGx0bpNaMawitmFBgRSeITsMtFGBiVOAKj+RPi+NtyEsMAyHs7raZVvqX6RJ5G28uIzzGS2FYaJ1Snxp/vyAPDAO8evYg9p2rdPgefXxQjr200gAIjIlFz+DyjNrR2iaefUWVjJ8oudUCpYHBlZgx/e05iOIAAWTGiwIhIxGh7GXEDZKVQqj+YWeOi8cCPxgEAVmw7iUtG91bNevKL5LWNBlBgRCQqLkyHcbF6sCxE1cF771nHatFlYyN7JYVPSgqAFSPaSiMSE+NMwHanZL/NZOE/jIlhgKwnHr8uC9kJoWjq6MbT7/8Alh15V2y5VqQBFBgRCRPjeBAuv8h1Gw0AcpyB0bn6dpitNr9flz/Iues1kaeerbSRr5Zw+UXReg30WmnXL2lVSqy/cyo0SgW+OFOPt49UjfhnuRUjuTV3BCgwIhImtsDIZLHhYJnjWuZnx/b6XmK4DhHBaljtLErr2oW4PJ+y2Oz8iBkKjIhU8E0e3dhK47bRkiW8jeYqOyEMv82fAAB44cNiXHAmVQ/FMTxWnqX6AAVGRMJmOztKFxva0NopfJ7RofONMFvtSAzXYUJ8aK/vMQwj6wTsujYT7CygUSr4xnmEiF30KJKvL/KJ1/L5AHDv3HTMGReNzm4bfv1uIay2obtiG9pM6LLYoFLIa3gshwIjIllxYTpkOPOMvhNBddpebhstO27AZnF8YCTDBGxuGy0xQgeFTFpCEPkbTVUaPzxWRgGBQsHglTvyEKpV4URlCzbuLRvyeG5GWmq0vIbHcuT3iEhAEct4EJZl8VXJJQD984s4OTJOwO5JvJbPp2gif1HO5OsGN5Kv+eaOEi3VH8yYiCA8f8skAMBfvijFDxdbBj1WzhVpAAVGROLEkmd0vqEDlU2d0CgVuGKQjsh8YGRog90+8uoPKahppVJ9Ij0xzhyj0SRfS7Hr9XBumToGN05OhNXO4tfvFKKre+BCETlXpAEUGBGJm+Vs9Hi6tg0tne4Pg/QWrhpt1rioQStVMmJDoFEp0G628tO55aKnhxGV6hPp4FaMmpzz0obDsqzkmzsOhWEY/P6WXMSFalF2qQPrdp8Z8Dh+xUhmo0A4FBgRSYsLdc0zEm47ba9zG23eINtoAKBWKvik7GJDq1+uy1+oVJ9IEZd8bbGxMJqtwx5/qd0Mk8UOhpHv73qkXoOXf5IHANh88AL2nb3U7xguxygjjlaMCBElbpinUNtpHWYrn/w9f0LskMdyCdinZJZnRF2viRTp1EoEa5QAgKYR5BlxpfqJYTpoVPL983l1VizumjMWAPCb/3zfazW+s9vKj/+R2/BYjnyfWRIwhE7A/uZcAyw2FmOjg5E+zNwguSZgU9drIlXujAXhSvXl0sNoKCsWTcS4WD3q2sx4ZkcRv9XIbaNFBqtlNzyWQ4ERkbxZ6Y7A6IxAeUau1WgDlem7ck3Alot2sxVtJsc2RCKtGBGJ4bbTGtuHT8DmE69lmF/UV5BGifU/nQqVgsFHPxjwQWENAEehCSDfijSAAiMiA7GhWoyPCwHL+n/ViGVZ7C1xJF7PG2YbDQCyExw5RoZWk9sTvcXK4FwtCg9SI0TiIxJI4IkO4SrTRr6VJseKtIFMSY7Ao9dmAgCe/aAIlU2d+Nr5fheiVcIms+paDgVGRBbmCFS2X1JnhKHVBJ1awW/pDSVUp8bYaMenzdMyWTWqpvwiImHubKXJuSJtMP9vXgampkTAaLJiwZ++xvvHqwEAe8824Mp1X2J3kUHgK/Q+CoyILAjVz+irM45ttCsyYqBTK0f0M5NklmfEV6RRfhGRIHfGgvCBUQDkGHFUSgVunTYGANDdZ1RIbasJD205LrvgiAIjIgtcP6MztUY0+3GL6ivnsvJw1Wiu5DYaxEDNHYmEjXQsiNVm5z8EBMpWGgDY7Cxe+3rgESHcRtqaXcWy2lajwIjIQkyIFplxjmRAf81Na+2y4FhFM4Ch+xf1xSVgn6qRRy8j2kojUjbSrTRDqwk2OwuNUoH40MBZHT1c3gSDszx/ICwc/20OC9hHztsoMCKy0dPPyD8v0AOlDbDZWYyPC3FraT0nMRwAUHapAybLwC33pYS6XhMpiw4ZWVUat42WHBkUUIOS642DB0WjOU4KKDAisuHvPKPRbKMBQHyYFlF6DWx2FmfrjL64NL+irtdEyqL1I6tKu+isSAuEHkau4ka4OjbS46SAAiMiGzPTe/KMfF0Kb7ez/BiQ+W5sowGOeUR8npHEE7Dtdha1rRQYEely3Uobal5aT0VaYP2ez0yPQmK4DoOtkTEAEsN1/PuvHFBgRGQjJkSLrHhnnpGPV41O1bShod0MvUaJGWnuvyHIpdFjQ4cZ3TY7FAwQH6oV+nIIcRu3ldZttaNjkGnygEtzxwBbMVIqGKwqyAGAfsER9/WqghwoZbS9SIERkRV/9TPittGuzIwZ1cwkuZTsc9to8WE6qJT0dkKkJ1ijgk7t+N0dal5aVbOzuWMA9TDi5OcmYuOy6Ujo05IjIVyHjcumIz83UaAr8w1qU0tkZfa4aLxxqMLnCdg9+UXubaNxuK2004Y22O2sZJM5DVSRRmQgWq9FdUsXGjvMSI0eOPDpWTEKzN/1/NxEXJeTgMPlTag3mhAX6tg+k9NKEYcCIyIr3D53SZ0Rje1mvt2/NzV1dKOwqgWAe2X6rtJj9NCqFOjotqGiqXPY4bNiRaX6RA6i9BpHYDTIipHJYkO90VG1FogrRhylguGrf+WM1r6JrESHaDEh3jGPzFd9NfadvQSWBSYmhvVbWh4plVLBz02T8nYadb0mcsDlGQ1WtHHRuY0WolUhIljtt+siwqDAiMgO94nmkI/yjEZbpt9XTwK2dBs9UtdrIgfDNXl07WHEMPLbOiK9UWBEZGe2czyILxKwbXYWX591lulnj24bjSOHkv0a2kojMtAzL23gJo8XA7QiLVCJMjDat28fCgoKkJSUBIZhsGPHjiGPP3DgAObOnYvo6GgEBQUhOzsbr7766qDHv/TSS2AYBo899ph3L5yIwsx0x4rR2bp2NAzTzdZdhVUtaOm0IEynwrSUCI/OJYeS/WrnVloibaURCYtyNnkcfMUocCvSApEoA6OOjg7k5eVhw4YNIzper9fj4Ycfxr59+3D69GmsXLkSK1euxD/+8Y9+xx45cgR///vfMWXKFG9fNhGJKL2Gz9/xdp7RXuc22o+yYj0uT89OCAPDAHVtZq8HcP5gttr46x5DK0ZEwqKHGSQb6BVpgUaUgdGiRYvw+9//HrfeeuuIjp82bRqWLFmCSZMmIS0tDcuWLcPChQuxf//+Xse1t7dj6dKleP311xEZGemLSyciwY0HOVTm3e00T8v0Xem1KqRHO6rRTktw1YjreB2kVlJCKpE0PsdokKq0Si4wohWjgCDKwMhTJ06cwMGDB3H11Vf3un358uW48cYbsWDBghGdx2w2o62trdc/Ig2+mJtW32ZCUbXjd+BqDxOvOROd22mnJJhnxJXqJ0boKCGVSNpwVWmB2vU6UMkqMEpOToZWq8WMGTOwfPly3Hffffz33n77bRw/fhxr164d8fnWrl2L8PBw/l9KSoovLpv4wCxnP6PSeu/lGe11Jl3nJYcjxkv9kaScgM2V6tM2GpG6aD7HqP97RWuXBW0mKwBHVRqRP1kFRvv378fRo0fx2muvYf369di6dSsAoKqqCr/61a/w73//GzrdyJNEV6xYgdbWVv5fVVWVry6deFmkS57Rd17qgs3lF422qeNApJyAzXe9Dqc/FkTaopwrRiaLHZ3d1l7f41aLovUa6LXUEzkQyOpZTk9PBwBMnjwZdXV1WL16NZYsWYJjx46hvr4e06dP54+12WzYt28f/vu//xtmsxlKpbLf+bRaLbRaGowpVbPHReNMrRGHzjfgximezfKx2OzYf7YBgOdl+q4mOVeMzl9qR1e3DUGa/r+HYlVDPYyITOg1SmhUCnRb7Whs70ZwVM+fxotcDyPaRgsYsloxcmW322E2O5ZFr732Wpw8eRKFhYX8vxkzZmDp0qUoLCwcMCgi0sc1evTG3LRjFc0wmq2I1mswZUy4x+fjxIZqEROigZ11jDGREr5UP4JK9Ym0MQwzaGVaVRNXqk8fAAKFKFeM2tvbce7cOf7r8vJyFBYWIioqCqmpqVixYgWqq6vx5ptvAgA2bNiA1NRUZGdnA3D0QXrllVfw6KOPAgBCQ0ORm5vb6z70ej2io6P73U7kY1Z6FBgGOFffjktGM2JDR7/6x1WjXZ0V69WBrwzDICcpHPvOXkJxTRumetgbyZ+4rTTKMSJyEB2igaHV1C/PiOt6TYnXgUOUgdHRo0cxf/58/uvHH38cAHD33Xdj8+bNMBgMqKys5L9vt9uxYsUKlJeXQ6VSISMjA+vWrcODDz7o92sn4hERrEF2QhhOG9rwXXkjbpqSNOpz7T3jSLye58VtNE5OYpgjMJLQaBCWZanrNZEVvslje98VIyrVDzSiDIzmzZsHlmUH/f7mzZt7ff3II4/gkUceces+9u7dO4orI1Ize1wUThvacKhs9IFRdUsXSuqMUDDAjzJjvHyFPQnYUirZb+uyoqPbBoC6XhN5GHQrjet6Tc0dA4Zsc4wIAYA5XuhnxFWjTU+NRESwxivX5Yor2T9jMMJmH/wDgZhwPYyi9Rro1JSjR6QvaoDAiGVZPvk6lbbSAgYFRkTWZjrzjMoudaDeaBrVOb46452hsYNJj9FDp1agy2LDhcYOn9yHtxmoIo3IDN/92iUwutRuhslih4Kh3/VAQoERkbWIYA0mJjhWZEbTz8hsteGbc44y/Xle6nbdl1LBIDtBWo0ee/KLaBuNyMNAW2lcRVpieBDUHs5GJNJBzzSRPX5u2ii20w6XN6HLYkNcqJbf8vIFqTV65Ev1qbkjkYnoEC75uqcqje9hRKX6AYUCIyJ7Pf2M3A+M+G20CXE+nQc2KUlaK0bcVhqV6hO5GGgrjWakBSYKjIjszUxz5Bmdv9SB+jb38oy4xOv52b7ZRuPwM9MksmJEpfpEbobaSqNS/cBCgRGRvfBgNR94fFs+8jyjCw0dON/QAZWCwdzx3i/Td5WdEAYFA1wymkedJO5PNdT1msgMNy+ts9sGk8XRiqKnuSN9AAgkFBiRgMDnGZWNfDuNWy26PC0KoTq1T66LE6RRIj1GD0D822k2O4ta58obbaURuQjVqqBWOrbLue006nodmCgwIgGB62f0nRt5Rl+VcGX6vt1G4+QkOWawiX07rd5ogs3OQq1kEBtCQ5aJPDAM09PLqL0bVpudXxmlrbTAQoERCQiXO/sZnW/oQN0I8oy6um18Fdv8Cb7pX9QXn2ck8hUjLr8oPkzn1blxhAgt2jkWpKHDDEOr4wOARqVAnAdzFon0UGBEAkJ4kJqv/BpJddqh8w3ottoxJiII4+NCfH15AFwq00S+YsSV6lPiNZGb6JCeFSNuGy05Iog+AAQYCoxIwJidzpXtD5+A3dPtOtanZfquJjpXjMobOtDZbfXLfY6GoYVK9Yk8uY4FueisSEum/KKAQ4ERCRizRzg3jWVZfMWV6ftpGw0AYkO1iAvVgmWBM7VGv92vu6jrNZEr115GfOI1NXcMOBQYkYBxeXoUFIxjRaa2dfA8o7JL7bjY3AWNSsE3h/QXrgP2KRHnGVHXayJXPb2MzKik5o4BiwIjEjAceUaOyq/vygdfNeK20WaPi0awRuWXa+NIIQGbul4TuYpyJl83dXT3dL2mirSAQ4ERCSizx0UBGHo7rWcbzT9l+q6kMDONul4TueKSrxvau1HV7Ox6Tc0dAw4FRiSgDNfo0Wiy4MgFR3K2P/OLONyK0RlDG6w2u9/vfzhd3TY0d1oAUNdrIj/cVlpNSxcuGR3DZGnFKPBQYEQCCpdndKGxk98ScvXNuUZYbCzSY/RIc3ai9qe0aD2CNUqYrXZcaOzw+/0Pp8b53yxUq0KYj7uBE+JvXPJ1vTMoCtGqEBFMv+eBhgIjElDCdGrkjnHmGQ1Qts+NAZknwDYaACgUDF+2L8YEbNpGI3LGNXjkJEcG+a1dBxEPCoxIwBmsbF+oMv2++ARsEeYZcYERbaMROQoLUkHl0syRKtICEwVGJOBwCdiH+gRGpw1G1LWZEaRWYmZ6lBCXBsAlAVuEK0bU9ZrIGcMwiHRupwFAKgVGAYkCIxJwLk9z5BlVNHbyKyBATzXa3PHR0KmVQl1er5J9lmUFu46BUNdrInfRLoERNXcMTBQYkYATqlNj8pj+/Yx68ouE20YDgAkJoVAwju67XBKoWHDJ19T1mshVlL4n2bqj2wqbXVwfTojvUWBEAhKfZ1TmSMBu7bTgWEUzAOESrzk6tRIZsY7BtWLbTquhrtdExnYXGXC8soX/+uU9Z3Hlui+xu8gg3EURv6PAiAQkvp+RM89oX+kl2FkgKz4EySLoWzJJhI0eWZbltx5pK43Ize4iAx7achwmS+/+YbWtJjy05TgFRwGEAiMSkGakRUKpYFDZ1Inqli5RVKO5EmMCdlNHN8xWOxgGiA+jrTQiHzY7izW7ijHQphl325pdxbStFiAoMCIBKdSln9GhskZ8XeKYjyZ0fhEnJ9FxbWJaMeK20WJDtNCo6K2DyMfh8iYYhhgszQIwtJpwuLx/7zMiP/TuRgIWV7b/l8/PorGjGzq1AtNSI4S9KKeJiaEAgPKGDrSbrQJfjUNP4jVtoxF5qTcOHhSN5jgibRQYkYDFNXLjhkWaLHbMf2WvKHIJokO0SHBuV50RyaoR5RcRuYoLHdnW8EiPI9JGgREJSLuLDPifr8r63S6mRMsckSVg812vw+mPA5GXmelRSAzXYbDhHwwcv/dCNn4l/kOBEQk4Ukm0dG30KAY11PWayJRSwWBVQQ4A9AuOuK9XFeRAqaC5aYGAAiMScKSSaCm2kn3KMSJylp+biI3LpiOhz4poQrgOG5dNR35uokBXRvxNJfQFEOJvUkm05LbSztQaYbXZoVIK+zmGcoyI3OXnJuK6nAQcLm9CvdGEuFDH9hmtFAUWCoxIwJFKomVKZDBCtCq0m60439CBrPhQwa6l22rnx5Mk0jgQImNKBYM5GdFCXwYREG2lkYAjlURLhYLhy/ZP1bQKei11bSawLKBRKXoN2SSEELmhwIgEHCklWoolAdt1G41hhP/vQgghvkKBEQlIUkm0FEvJPpd4TaX6hBC5oxwjErCkkGjJjwapaQPLsoKt1lCpPiEkUFBgRAKa2BMtM+NDoFIwaO60oLbNhMRwYQITbiuNAiNCiNzRVhohIqZTKzE+LgSAsHlGPTlGtJVGCJE3CowIETkxJGBzW2lCrVgRQoi/UGBEiMhxCdinhAyMqOs1ISRAUGBEiMjxK0YCVaa1mSwwmqwAgCTaSiOEyBwFRoSI3ERnYFTZ1Ik2k8Xv929wbqNFBKsRrKF6DUKIvFFgRIjIReo1SHL2DzpjMPr9/vltNMovIoQEAAqMCJGAnCSun5H/R4NQqT4hJJBQYESIBAjZAbsnMKL8IkKI/FFgRIgECJmAbaCu14SQAEKBESESMMm5YnS2th0Wm92v911NW2mEkABCgREhEpAcGYRQrQrdNjvO1bf79b655Gvqek0ICQQUGBEiAQzDYGKS/ztg2+0salup6zUhJHBQYESIRAiRZ9TQbobFxkKpYBAXqvXb/RJCiFAoMCJEIiYJsGLE5RclhOmgUtLbBSFE/uidjhCJcC3ZZ1nWL/fZMzyW8osIIYGBAiNCJCIzLhRqJYPWLgtqnHk/vmag4bGEkABDgREhEqFRKTA+LhSA/7bTqFSfEBJoKDAiREK4BOxTfhgNYrOzKKp23I/ZaoPN7p/tO0IIERIFRoRISI6fErB3Fxlw5bovceRCMwBg0zcXcOW6L7G7yODT+yWEEKFRYESIhPijZH93kQEPbTkOQ588ptpWEx7acpyCI0KIrFFgRIiEcCtGF5u70Npl8fr5bXYWa3YVY6BNM+62NbuKaVuNECJbogyM9u3bh4KCAiQlJYFhGOzYsWPI4w8cOIC5c+ciOjoaQUFByM7OxquvvtrrmI0bN2LKlCkICwtDWFgY5syZg08++cSHj4IQ7wsPUiM50pEIfdoHq0aHy5v6rRS5YgEYWk04XN7k9fsmhBAxEGVg1NHRgby8PGzYsGFEx+v1ejz88MPYt28fTp8+jZUrV2LlypX4xz/+wR+TnJyMl156CceOHcPRo0dxzTXXYPHixTh16pSvHgYhPsFvp/kgz6jeOLI2ACM9jhBCpEYl9AUMZNGiRVi0aNGIj582bRqmTZvGf52WloZt27Zh//79eOCBBwAABQUFvX7mD3/4AzZu3Ihvv/0WkyZN8s6FE+IHOUlh+LS4zut5RizL4kJDx4iOjQulho+EEHkSZWDkqRMnTuDgwYP4/e9/P+D3bTYb3nvvPXR0dGDOnDmDnsdsNsNsNvNft7X5bxQDIYPpKdn33u9jVVMnnvugCF+VXBryOAZAQrgOM9OjvHbfhBAiJrIKjJKTk3Hp0iVYrVasXr0a9913X6/vnzx5EnPmzIHJZEJISAi2b9+OnJycQc+3du1arFmzxteXTYhbuATsc/VGdFvt0KhGvyNusdnx+v7z+OsXpTBZ7NAoFViQE49PTjoqz1xTrBnn/64qyIFSwfQ7FyGEyIEoc4xGa//+/Th69Chee+01rF+/Hlu3bu31/QkTJqCwsBDfffcdHnroIdx9990oLi4e9HwrVqxAa2sr/6+qqsrXD4GQYY2JCEJ4kBoWG4vSeuOoz3PkQhNu/Ot+/HF3CUwWO2aPi8LHv7oK/7N0OjYum46EPvPREsJ12LhsOvJzEz19CIQQIlqyWjFKT08HAEyePBl1dXVYvXo1lixZwn9fo9Fg/PjxAIDLLrsMR44cwV/+8hf8/e9/H/B8Wq0WWq3W9xdOiBsYhkFOYhgOnW9EcU0bJiWFu/XzzR3deOmTM3jnqCPQj9Zr8MyNE3HrtDFgGMdKUH5uIq7LScDh8ibUG02IC3Vsn9FKESFE7mQVGLmy2+298oNGewwhYpST5AyM3EjAZlkW7x+vxosfn0ZTRzcAYMnMFDyVn42IYE2/45UKBnMyor12zYQQIgWiDIza29tx7tw5/uvy8nIUFhYiKioKqampWLFiBaqrq/Hmm28CADZs2IDU1FRkZ2cDcPRBeuWVV/Doo4/y51ixYgUWLVqE1NRUGI1GvPXWW9i7dy/27Nnj3wdHiBe4W7J/rt6IZ7YX4Ttn/6EJ8aH4w625mJFGSdSEEOJKlIHR0aNHMX/+fP7rxx9/HABw9913Y/PmzTAYDKisrOS/b7fbsWLFCpSXl0OlUiEjIwPr1q3Dgw8+yB9TX1+Pu+66CwaDAeHh4ZgyZQr27NmD6667zn8PjBAv4WemGdrAsiy/BdaXyWLDf395Dn/fVwaLjYVOrcBjC7LwiyvToVbKKsWQEEK8gmFZlnr7j1BbWxvCw8PR2tqKsLAwoS+HBLBuqx25q/ag22bH/t/OR0pUcL9jvj57Cc/uKEJlUycA4NrsOKy+edKAxxJCiJy58/dblCtGhJChaVQKjI/To9hgxOZvLmBBTjyfHF3fZsLzHxbjwx8cJfcJYTqsvjkHCyclDLqyRAghxIECI0IkaHeRAeUNjpWgf35Tjn9+U46EMB3mTYjFRz8YYDRboWCAe65Ix+PXZyFESy91QggZCXq3JERidhcZ8NCW4+i7B17bZsLbRxwl+HnJ4fjDrZORO8a9Un5CCAl0FBgRIiE2O4s1u4r7BUWuwoNUeO+XV3jUEZsQQgIVvXMSIiGHy5tgaB16sn1rlxXHKpr9dEWEECIvFBgRIiH1xqGDInePI4QQ0hsFRoRISFyobviD3DiOEEJIbxQYESIhM9OjkBiuw2BF9wyAxHDHXDNCCCHuo8CIEAlRKhisKsgBgH7BEff1qoIcGvZKCCGjRIERIRKTn5uIjcumIyG893ZZQrgOG5dNR35uokBXRggh0kfl+oRIUH5uIq7LScDh8ibUG02IC9Xxna8JIYSMHgVGhEiUUsFgTka00JdBCCGyQltphBBCCCFOFBgRQgghhDhRYEQIIYQQ4kSBESGEEEKIEwVGhBBCCCFOFBgRQgghhDhRYEQIIYQQ4kSBESGEEEKIEwVGhBBCCCFO1PnaDSzLAgDa2toEvhJCCCGEjBT3d5v7Oz4UCozcYDQaAQApKSkCXwkhhBBC3GU0GhEeHj7kMQw7kvCJAADsdjtqamoQGhoKhvHusM62tjakpKSgqqoKYWFhXj232NBjla9Aerz0WOUrkB5voDxWlmVhNBqRlJQEhWLoLCJaMXKDQqFAcnKyT+8jLCxM1r+cruixylcgPV56rPIVSI83EB7rcCtFHEq+JoQQQghxosCIEEIIIcSJAiOR0Gq1WLVqFbRardCX4nP0WOUrkB4vPVb5CqTHG0iPdaQo+ZoQQgghxIlWjAghhBBCnCgwIoQQQghxosCIEEIIIcSJAiNCCCGEECcKjPxow4YNSEtLg06nw6xZs3D48OEhj3/vvfeQnZ0NnU6HyZMn4+OPP/bTlY7e2rVrcfnllyM0NBRxcXG45ZZbUFJSMuTPbN68GQzD9Pqn0+n8dMWjt3r16n7XnZ2dPeTPSPE55aSlpfV7vAzDYPny5QMeL6Xndd++fSgoKEBSUhIYhsGOHTt6fZ9lWTz33HNITExEUFAQFixYgNLS0mHP6+5r3l+GerwWiwVPPfUUJk+eDL1ej6SkJNx1112oqakZ8pyjeT34w3DP7T333NPvuvPz84c9rxif2+Ee60CvX4Zh8PLLLw96TrE+r75EgZGfvPPOO3j88cexatUqHD9+HHl5eVi4cCHq6+sHPP7gwYNYsmQJfvGLX+DEiRO45ZZbcMstt6CoqMjPV+6er7/+GsuXL8e3336Lzz77DBaLBddffz06OjqG/LmwsDAYDAb+X0VFhZ+u2DOTJk3qdd0HDhwY9FipPqecI0eO9Hqsn332GQDgJz/5yaA/I5XntaOjA3l5ediwYcOA3//jH/+Iv/71r3jttdfw3XffQa/XY+HChTCZTIOe093XvD8N9Xg7Oztx/PhxPPvsszh+/Di2bduGkpIS3HzzzcOe153Xg78M99wCQH5+fq/r3rp165DnFOtzO9xjdX2MBoMB//rXv8AwDG6//fYhzyvG59WnWOIXM2fOZJcvX85/bbPZ2KSkJHbt2rUDHn/HHXewN954Y6/bZs2axT744IM+vU5vq6+vZwGwX3/99aDHbNq0iQ0PD/ffRXnJqlWr2Ly8vBEfL5fnlPOrX/2KzcjIYO12+4Dfl+rzCoDdvn07/7XdbmcTEhLYl19+mb+tpaWF1Wq17NatWwc9j7uveaH0fbwDOXz4MAuAraioGPQYd18PQhjosd59993s4sWL3TqPFJ7bkTyvixcvZq+55pohj5HC8+pttGLkB93d3Th27BgWLFjA36ZQKLBgwQIcOnRowJ85dOhQr+MBYOHChYMeL1atra0AgKioqCGPa29vx9ixY5GSkoLFixfj1KlT/rg8j5WWliIpKQnjxo3D0qVLUVlZOeixcnlOAcfv9JYtW3DvvfcOOVBZqs+rq/LyctTW1vZ67sLDwzFr1qxBn7vRvObFrLW1FQzDICIiYsjj3Hk9iMnevXsRFxeHCRMm4KGHHkJjY+Ogx8rlua2rq8NHH32EX/ziF8MeK9XndbQoMPKDhoYG2Gw2xMfH97o9Pj4etbW1A/5MbW2tW8eLkd1ux2OPPYa5c+ciNzd30OMmTJiAf/3rX/jggw+wZcsW2O12XHHFFbh48aIfr9Z9s2bNwubNm7F7925s3LgR5eXluOqqq2A0Ggc8Xg7PKWfHjh1oaWnBPffcM+gxUn1e++KeH3eeu9G85sXKZDLhqaeewpIlS4YcMuru60Es8vPz8eabb+KLL77AunXr8PXXX2PRokWw2WwDHi+X5/aNN95AaGgobrvttiGPk+rz6gmV0BdA5Gv58uUoKioadj96zpw5mDNnDv/1FVdcgYkTJ+Lvf/87XnjhBV9f5qgtWrSI//9TpkzBrFmzMHbsWLz77rsj+hQmZf/85z+xaNEiJCUlDXqMVJ9X0sNiseCOO+4Ay7LYuHHjkMdK9fVw55138v9/8uTJmDJlCjIyMrB3715ce+21Al6Zb/3rX//C0qVLhy2IkOrz6glaMfKDmJgYKJVK1NXV9bq9rq4OCQkJA/5MQkKCW8eLzcMPP4wPP/wQX331FZKTk936WbVajWnTpuHcuXM+ujrfiIiIQFZW1qDXLfXnlFNRUYHPP/8c9913n1s/J9XnlXt+3HnuRvOaFxsuKKqoqMBnn3025GrRQIZ7PYjVuHHjEBMTM+h1y+G53b9/P0pKStx+DQPSfV7dQYGRH2g0Glx22WX44osv+Nvsdju++OKLXp+oXc2ZM6fX8QDw2WefDXq8WLAsi4cffhjbt2/Hl19+ifT0dLfPYbPZcPLkSSQmJvrgCn2nvb0dZWVlg163VJ/TvjZt2oS4uDjceOONbv2cVJ/X9PR0JCQk9Hru2tra8N133w363I3mNS8mXFBUWlqKzz//HNHR0W6fY7jXg1hdvHgRjY2Ng1631J9bwLHie9lllyEvL8/tn5Xq8+oWobO/A8Xbb7/NarVadvPmzWxxcTH7wAMPsBEREWxtbS3Lsiz785//nH366af547/55htWpVKxr7zyCnv69Gl21apVrFqtZk+ePCnUQxiRhx56iA0PD2f37t3LGgwG/l9nZyd/TN/HumbNGnbPnj1sWVkZe+zYMfbOO+9kdTode+rUKSEewog98cQT7N69e9ny8nL2m2++YRcsWMDGxMSw9fX1LMvK5zl1ZbPZ2NTUVPapp57q9z0pP69Go5E9ceIEe+LECRYA++c//5k9ceIEX4X10ksvsREREewHH3zA/vDDD+zixYvZ9PR0tquriz/HNddcw/7tb3/jvx7uNS+koR5vd3c3e/PNN7PJyclsYWFhr9ex2Wzmz9H38Q73ehDKUI/VaDSyTz75JHvo0CG2vLyc/fzzz9np06ezmZmZrMlk4s8hled2uN9jlmXZ1tZWNjg4mN24ceOA55DK8+pLFBj50d/+9jc2NTWV1Wg07MyZM9lvv/2W/97VV1/N3n333b2Of/fdd9msrCxWo9GwkyZNYj/66CM/X7H7AAz4b9OmTfwxfR/rY489xv93iY+PZ2+44Qb2+PHj/r94N/30pz9lExMTWY1Gw44ZM4b96U9/yp47d47/vlyeU1d79uxhAbAlJSX9vifl5/Wrr74a8PeWezx2u5199tln2fj4eFar1bLXXnttv/8GY8eOZVetWtXrtqFe80Ia6vGWl5cP+jr+6quv+HP0fbzDvR6EMtRj7ezsZK+//no2NjaWVavV7NixY9n777+/X4Ajled2uN9jlmXZv//972xQUBDb0tIy4Dmk8rz6EsOyLOvTJSlCCCGEEImgHCNCCCGEECcKjAghhBBCnCgwIoQQQghxosCIEEIIIcSJAiNCCCGEECcKjAghhBBCnCgwIoQQQghxosCIEEIIIcSJAiNCiOgwDDPsv3vuuUfoyxzW6tWrwTAMNm/eLPSlEEJGSCX0BRBCyGDuvvvuQb935ZVX+vFKCCGBggIjQoho0UoLIcTfaCuNEEIIIcSJAiNCiCwwDIO0tDR0d3dj1apVyMjIgE6nw7hx4/Dcc8/BZDIN+HONjY34zW9+g8zMTOh0OkRFRSE/Px+ffvrpoPfV2NiIZ555BpMnT4Zer0dYWBgmT56M3/72tzAYDAP+zMmTJ3HzzTcjMjISer0eV199NQ4ePOiVx04I8R4KjAghssGyLG6//Xa8/PLLyMnJwY033oimpia88MILuOmmm2Cz2XodX11djZkzZ+KVV15Bd3c3brnlFkybNg2ff/45Fi5ciFdffbXffZw+fRpTp07Fiy++iIaGBixcuBALFiwAy7J4+eWX8d133/X7maNHj2L27Nm4cOECFi5ciMzMTOzbtw/XXnstioqKfPbfgxAyCiwhhIgMANbdtyfuZ5KTk9mysjL+9vr6ejY3N5cFwL766qu9fuamm25iAbA/+9nPWLPZzN++f/9+Njg4mFUqleyJEyf42y0WCzthwgQWAPvYY4/1+hmWZdmioiL23Llz/NerVq3ir+svf/lLr2Mfe+wxFgD785//3K3HSQjxLQqMCCGiwwUTQ/3bvn37gD/zj3/8o9/5PvnkExYAm5GRwd9WVlbGAmBDQkLYxsbGfj/z+OOPswDY++67j7/tnXfeYQGwkyZNYq1W67CPgwuM5s6d2+97DQ0NLAB27Nixw56HEOI/VJVGCBGtocr1U1NTB7z9zjvv7Hdbfn4+IiMjUVZWBoPBgMTERBw4cID/XlRUVL+f+fnPf44///nP2L9/P3/b559/DgC47777oFQqR/w4rr/++n63RUdHIyoqatCcJEKIMCgwIoSIlrvl+pGRkQgNDR3we2PHjkVzczNqamqQmJiImpoaAEBaWtqAx3O3V1dX87dVVVUBADIyMty6ruTk5AFvDw0NRVNTk1vnIoT4FiVfE0LIABiG8dq5FAp6qyVEKujVSgiRjebmZhiNxgG/V1lZCQBISkrq9b8VFRUDHn/hwgUAwJgxY/jbUlJSAABlZWVeuV5CiPhQYEQIkZV33323322ffvopmpqaMG7cOCQmJgLoGSmye/dutLS09PuZLVu2AACuuuoq/rYFCxYAAP75z3/Cbrd7+9IJISJAgREhRFbWrFnDr/YAQENDA37zm98AAJYvX87fPm7cONx4440wGo341a9+BYvFwn/v0KFD2LhxI5RKZa+fue2225CVlYWioiL89re/7fUzAHDq1CmcP3/eR4+MEOIPlHxNCBGte+65Z9Dvpaam4vnnn+9325QpUzBp0iRce+21UKvV+PLLL9HS0oL58+fj0Ucf7XX83//+d1x11VV488038fXXX2POnDm4dOkS9u7dC5vNhj/96U+YOnUqf7xKpcL777+P6667Dn/605/w1ltvYc6cOWBZFqWlpSgqKsL27dsxbtw4b/5nIIT4EQVGhBDReuONNwb9Xl5eXr/AiGEY/Oc//8Hzzz+Pt956i69AW758OZ555hmoVL3f8saMGYMjR45g7dq12LFjB7Zt24bg4GBce+21eOKJJwYss8/NzcX333+Pl19+GTt37sTHH38MrVaL1NRUPPXUU5g9e7Z3HjwhRBAMy7Ks0BdBCCGeYhgGY8eO7bWNRggh7qIcI0IIIYQQJwqMCCGEEEKcKDAihBBCCHGi5GtCiCxQuiQhxBtoxYgQQgghxIkCI0IIIYQQJwqMCCGEEEKcKDAihBBCCHGiwIgQQgghxIkCI0IIIYQQJwqMCCGEEEKcKDAihBBCCHH6/4ScIQF4rUg2AAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(np.arange(len(tranv_endgame_hist[0])), tranv_endgame_hist[0], '-o', label='Train loss (test + val)')\n",
    "plt.xlabel('Epoch', size=15)\n",
    "plt.ylabel('Loss', size=15)\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_endgame_history(ang_endgame_hist, ang_hist)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(np.arange(len(ang_endgame_hist[0])), ang_endgame_hist[0], '-o', label='Train loss (test + val)')\n",
    "plt.xlabel('Epoch', size=15)\n",
    "plt.ylabel('Loss', size=15)\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Agreement accuracy among models: 0.6756382757904933\n"
     ]
    }
   ],
   "source": [
    "## I can use any engine for this\n",
    "engine_long.test_all(model_long, model_tranv, model_ang, \n",
    "                     test_dl_long, test_dl_tranv, test_dl_angvel)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Feature Maps Extraction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "def converter(instr):\n",
    "    return np.fromstring(instr[1:-1],sep=' ')\n",
    "\n",
    "windowed_df = pd.read_csv(\n",
    "    './data/full_windowed_df.csv',\n",
    "    converters={\n",
    "        'long_acc_window': converter,\n",
    "        'tranv_acc_window': converter,\n",
    "        'ang_vel_window': converter\n",
    "    }\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Not_driving' 'User1' 'User2' 'User3']\n",
      "['User1' 'User2' 'User3']\n"
     ]
    }
   ],
   "source": [
    "print(windowed_df.user.unique())\n",
    "print(windowed_df.query('coarse_label==5.0').user.unique())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "recall `tr_idx_nondupe` and `tr_idx_interp` created above. I'll use it to split my data to get training and validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "TR_tr_idx_nondupe, VAL_tr_idx_nondupe = train_test_split(tr_idx_nondupe,\n",
    "                                                    test_size=0.2,\n",
    "                                                    stratify=windowed_df.loc[tr_idx_nondupe, 'user_act'].values,\n",
    "                                                    random_state=42)\n",
    "\n",
    "TR_tr_idx_interp, VAL_tr_idx_interp = train_test_split(tr_idx_interp,\n",
    "                                                  test_size=0.2,\n",
    "                                                  stratify=windowed_df.loc[tr_idx_interp, 'user_act'].values,\n",
    "                                                  random_state=42)\n",
    "\n",
    "# get the idx for the duplicates\n",
    "TR_tr_idx_dupe = TR_tr_idx_nondupe+1\n",
    "VAL_tr_idx_dupe = VAL_tr_idx_nondupe+1\n",
    "\n",
    "# get new training and validation set\n",
    "TR_train_idx = np.concatenate([TR_tr_idx_nondupe, TR_tr_idx_dupe, TR_tr_idx_interp], axis=None)\n",
    "VAL_train_idx = np.concatenate([VAL_tr_idx_nondupe, VAL_tr_idx_dupe, VAL_tr_idx_interp], axis=None)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now I have `train_idx` from that I have `TR_train_idx` (training), `VAL_train_idx` (validation) and `test_idx` (testing)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(14453, 46240, 11563)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(test_idx), len(TR_train_idx), len(VAL_train_idx)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_tr, y_tr, _ = Prep_data_for_CNN().prep_input_for_CNN(windowed_df.loc[TR_train_idx, :])\n",
    "X_val, y_val, _ = Prep_data_for_CNN().prep_input_for_CNN(windowed_df.loc[VAL_train_idx, :])\n",
    "X_test, y_test, _ = Prep_data_for_CNN().prep_input_for_CNN(windowed_df.loc[test_idx, :])\n",
    "\n",
    "\n",
    "# Create datasets for each feature\n",
    "feature_indices = {'long_acc': 0, 'tranv_acc': 1, 'ang_vel': 2}\n",
    "\n",
    "## dataset for extracting feature maps from data\n",
    "TR_all_dataset = {feature: SignalsDataset((X_tr-mean)/std, y_tr, idx) for feature, idx in feature_indices.items()}\n",
    "VAL_all_dataset = {feature: SignalsDataset((X_val-mean)/std, y_val, idx) for feature, idx in feature_indices.items()}\n",
    "TEST_all_dataset = {feature: SignalsDataset((X_test-mean)/std, y_test, idx) for feature, idx in feature_indices.items()}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<All keys matched successfully>"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load current trained model state\n",
    "# Load the checkpoint\n",
    "checkpoint_l = torch.load('./model_checkpoint/endgame_2_simpleCNN_long_chkpt.pth')\n",
    "checkpoint_t = torch.load('./model_checkpoint/endgame_2_simpleCNN_tranv_chkpt.pth')\n",
    "checkpoint_a = torch.load('./model_checkpoint/endgame_2_simpleCNN_angvel_chkpt.pth')\n",
    "\n",
    "# Load the different states\n",
    "model_long.load_state_dict(checkpoint_l['model_state_dict'])\n",
    "model_tranv.load_state_dict(checkpoint_t['model_state_dict'])\n",
    "model_ang.load_state_dict(checkpoint_a['model_state_dict'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "batch_size = 1\n",
    "\n",
    "# dataloaders\n",
    "TR_dataloaders = {\n",
    "      'long': DataLoader(TR_all_dataset['long_acc'], batch_size, shuffle=False),\n",
    "      'tranv': DataLoader(TR_all_dataset['tranv_acc'], batch_size, shuffle=False),\n",
    "      'angvel': DataLoader(TR_all_dataset['ang_vel'], batch_size, shuffle=False)\n",
    "}\n",
    "VAL_dataloaders = {\n",
    "      'long': DataLoader(VAL_all_dataset['long_acc'], batch_size, shuffle=False),\n",
    "      'tranv': DataLoader(VAL_all_dataset['tranv_acc'], batch_size, shuffle=False),\n",
    "      'angvel': DataLoader(VAL_all_dataset['ang_vel'], batch_size, shuffle=False)\n",
    "}\n",
    "TEST_dataloaders = {\n",
    "      'long': DataLoader(TEST_all_dataset['long_acc'], batch_size, shuffle=False),\n",
    "      'tranv': DataLoader(TEST_all_dataset['tranv_acc'], batch_size, shuffle=False),\n",
    "      'angvel': DataLoader(TEST_all_dataset['ang_vel'], batch_size, shuffle=False)\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "# models\n",
    "models = {\n",
    "      'long': model_long,\n",
    "      'tranv': model_tranv,\n",
    "      'angvel': model_ang\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "from src.preprocess import FeatureMaps_extractor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Extracting Feature Maps train: 100%|██████████| 46240/46240 [06:54<00:00, 111.48it/s]\n",
      "Extracting Feature Maps valid: 100%|██████████| 11563/11563 [01:38<00:00, 117.09it/s]\n",
      "Extracting Feature Maps test: 100%|██████████| 14453/14453 [02:04<00:00, 116.31it/s]\n"
     ]
    }
   ],
   "source": [
    "idx_lists = [TR_train_idx, VAL_train_idx, test_idx]\n",
    "dataloaders = [TR_dataloaders, VAL_dataloaders, TEST_dataloaders]\n",
    "splits = ['train', 'valid', 'test']\n",
    "base_dir = './data/feature_maps_labels/'\n",
    "\n",
    "for i in range(len(idx_lists)):\n",
    "      feature_extractor = FeatureMaps_extractor(base_dir, device, idx_lists[i], splits[i])\n",
    "      feature_extractor.feature_map_extractor(models, dataloaders[i], splits[i])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Done. Total file size = `40.7 GB (43,803,578,368 bytes)`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "test = np.load('./data/feature_maps_labels/train/10002.npy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3, 224, 224)"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[-8.16360414e-02,  4.78573710e-01, -8.40168476e-01, ...,\n",
       "         -2.22032249e-01,  7.07064494e-02, -2.57724762e-01],\n",
       "        [ 2.97466904e-01, -1.28953016e+00, -1.27021897e+00, ...,\n",
       "          3.82092521e-02,  6.04807734e-02,  1.69129163e-01],\n",
       "        [ 1.09014976e+00,  2.16618037e+00, -2.97740608e-01, ...,\n",
       "          2.55739037e-02, -1.75157875e-01,  2.15569377e-01],\n",
       "        ...,\n",
       "        [-4.22323376e-01,  1.24451947e+00, -1.09783101e+00, ...,\n",
       "          1.49238437e-01,  1.30221337e-01,  1.06817521e-01],\n",
       "        [-1.18651462e+00, -4.73963797e-01, -1.02510417e+00, ...,\n",
       "         -1.68773413e-01, -3.15741271e-01, -1.33159816e-01],\n",
       "        [ 1.09627768e-01, -1.01654992e-01, -1.94163769e-01, ...,\n",
       "         -3.40037465e-01, -1.70686439e-01, -9.19869393e-02]],\n",
       "\n",
       "       [[ 1.73653513e-01, -1.37913615e-01,  4.26242203e-01, ...,\n",
       "         -5.18541932e-01, -3.88835579e-01, -2.23628968e-01],\n",
       "        [-4.26551282e-01, -5.58969200e-01, -7.41269648e-01, ...,\n",
       "         -7.97137082e-01, -4.26398844e-01, -5.01707733e-01],\n",
       "        [-3.93656433e-01, -7.95119286e-01, -8.23817015e-01, ...,\n",
       "          1.97273806e-01, -2.91518003e-01, -1.20897964e-01],\n",
       "        ...,\n",
       "        [ 6.81656003e-02, -4.32582684e-02, -5.83969355e-01, ...,\n",
       "         -6.95284486e-01, -1.09969333e-01, -2.60462135e-01],\n",
       "        [ 1.45901009e-01,  1.85425848e-01, -4.61474717e-01, ...,\n",
       "         -2.78627902e-01, -3.58017474e-01, -5.92643321e-01],\n",
       "        [-3.10838491e-01, -4.20158893e-01, -4.45628762e-01, ...,\n",
       "         -5.23022711e-01, -2.74795920e-01, -5.29100478e-01]],\n",
       "\n",
       "       [[-1.54293925e-01, -4.27797407e-01, -2.60588139e-01, ...,\n",
       "         -1.13586009e-01, -4.90005612e-01, -6.72349781e-02],\n",
       "        [ 4.49351687e-03, -2.10294217e-01,  1.72965765e-01, ...,\n",
       "         -1.76572539e-02, -9.50427055e-02, -1.39461011e-01],\n",
       "        [-2.58360744e-01, -4.79433592e-03,  2.14619897e-02, ...,\n",
       "         -1.05833024e-01,  9.72270444e-02, -9.24224183e-02],\n",
       "        ...,\n",
       "        [-1.01951025e-01, -3.21350098e-01, -1.20222017e-01, ...,\n",
       "         -2.15151966e-01, -2.61496335e-01, -4.09979187e-02],\n",
       "        [-3.38012502e-02,  3.65547746e-01, -1.53137684e-01, ...,\n",
       "         -2.06174791e-01,  3.16832334e-01, -2.43218541e-01],\n",
       "        [ 4.04048264e-02, -1.61713851e-03, -6.62322119e-02, ...,\n",
       "         -8.61311425e-03, -3.94777022e-02,  2.51169857e-02]]],\n",
       "      dtype=float32)"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Split lstm_features into training and test too."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "from src.preprocess import Save_Lstm_Feats_n_Labs\n",
    "from tqdm import tqdm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "lstm_features = np.load('./data/lstm_features_labels/full_lstm_features.npy')\n",
    "lstm_labels = np.load('./data/lstm_features_labels/full_lstm_labels.npy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(72256, 112, 6)"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lstm_features.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1., 2., 3., 4., 5., 6., 7., 8.])"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.unique(lstm_labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "~~We are interested in knowing whether the signal is taken in a car or not. This means our positive class is `5.0` and the rest are negative.~~ I have decided to do a full on transport mode classification. So all I'll do is shift the labels by 1 so that start from 0."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "#lstm_labels_mapped = np.where(lstm_labels == 5, 1.0, 0.0)\n",
    "lstm_labels_mapped = lstm_labels - 1.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0., 1., 2., 3., 4., 5., 6., 7.])"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.unique(lstm_labels_mapped)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Saving Lstm Feat and Labels: train: 100%|██████████| 46240/46240 [00:43<00:00, 1066.23it/s]\n",
      "Saving Lstm Feat and Labels: valid: 100%|██████████| 11563/11563 [00:10<00:00, 1094.36it/s]\n",
      "Saving Lstm Feat and Labels: test: 100%|██████████| 14453/14453 [00:13<00:00, 1089.38it/s]\n"
     ]
    }
   ],
   "source": [
    "idx_lists = [TR_train_idx, VAL_train_idx, test_idx]\n",
    "splits = ['train', 'valid', 'test']\n",
    "base_dir = './data/lstm_features_labels/'\n",
    "\n",
    "lstm_saver = Save_Lstm_Feats_n_Labs(base_dir)\n",
    "for i in range(len(idx_lists)): \n",
    "      for j in tqdm(range(len(idx_lists[i])), desc=f\"Saving Lstm Feat and Labels: {splits[i]}\"):\n",
    "            idx = idx_lists[i][j]\n",
    "            lstm_saver.save_combined_feature_map(lstm_features[idx], lstm_labels_mapped[idx],\n",
    "                                                 splits[i], idx)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "torch-gpu",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}