Bearing Sensor Anomaly Detection.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "source": [
    "# Bearing Failure Anomaly Detection"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "source": [
    "# import libraries\r\n",
    "import os\r\n",
    "import pandas as pd\r\n",
    "import numpy as np\r\n",
    "import seaborn as sns\r\n",
    "sns.set(color_codes=True)\r\n",
    "import matplotlib.pyplot as plt\r\n",
    "%matplotlib inline\r\n",
    "\r\n",
    "from numpy.random import seed\r\n",
    "from tensorflow.random import set_seed\r\n",
    "import tensorflow as tf\r\n",
    "\r\n",
    "from tensorflow.keras.layers import Input, Dropout, Dense, LSTM, TimeDistributed, RepeatVector, BatchNormalization\r\n",
    "from tensorflow.keras.callbacks import EarlyStopping, ModelCheckpoint\r\n",
    "from tensorflow.keras.models import Model, load_model\r\n",
    "from tensorflow.keras import regularizers\r\n",
    "\r\n",
    "from sklearn.preprocessing import MinMaxScaler\r\n",
    "from sklearn.decomposition import PCA\r\n",
    "from sklearn.svm import OneClassSVM\r\n",
    "from sklearn.ensemble import IsolationForest"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "source": [
    "# set random seed\r\n",
    "seed(10)\r\n",
    "set_seed(10)"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "# Data loading and pre-processing"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "source": [
    "# load, average and merge sensor samples\r\n",
    "merged_data = pd.read_csv('./data/Averaged_BearingTest_Dataset.csv',index_col=0)"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "source": [
    "# transform data file index to datetime and sort in chronological order\r\n",
    "merged_data.index = pd.to_datetime(merged_data.index)\r\n",
    "merged_data = merged_data.sort_index()\r\n",
    "print(\"Dataset shape:\", merged_data.shape)\r\n",
    "merged_data.head()"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Dataset shape: (982, 4)\n"
     ]
    },
    {
     "output_type": "execute_result",
     "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>Bearing 1</th>\n",
       "      <th>Bearing 2</th>\n",
       "      <th>Bearing 3</th>\n",
       "      <th>Bearing 4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2004-02-12 10:52:39</th>\n",
       "      <td>0.060236</td>\n",
       "      <td>0.074227</td>\n",
       "      <td>0.083926</td>\n",
       "      <td>0.044443</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-02-12 11:02:39</th>\n",
       "      <td>0.061455</td>\n",
       "      <td>0.073844</td>\n",
       "      <td>0.084457</td>\n",
       "      <td>0.045081</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-02-12 11:12:39</th>\n",
       "      <td>0.061361</td>\n",
       "      <td>0.075609</td>\n",
       "      <td>0.082837</td>\n",
       "      <td>0.045118</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-02-12 11:22:39</th>\n",
       "      <td>0.061665</td>\n",
       "      <td>0.073279</td>\n",
       "      <td>0.084879</td>\n",
       "      <td>0.044172</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-02-12 11:32:39</th>\n",
       "      <td>0.061944</td>\n",
       "      <td>0.074593</td>\n",
       "      <td>0.082626</td>\n",
       "      <td>0.044659</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     Bearing 1  Bearing 2  Bearing 3  Bearing 4\n",
       "2004-02-12 10:52:39   0.060236   0.074227   0.083926   0.044443\n",
       "2004-02-12 11:02:39   0.061455   0.073844   0.084457   0.045081\n",
       "2004-02-12 11:12:39   0.061361   0.075609   0.082837   0.045118\n",
       "2004-02-12 11:22:39   0.061665   0.073279   0.084879   0.044172\n",
       "2004-02-12 11:32:39   0.061944   0.074593   0.082626   0.044659"
      ]
     },
     "metadata": {},
     "execution_count": 4
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "# Define train/test data"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "source": [
    "train = merged_data['2004-02-12 10:52:39': '2004-02-15 12:52:39']\r\n",
    "test = merged_data['2004-02-15 12:52:39':]\r\n",
    "print(\"Training dataset shape:\", train.shape)\r\n",
    "print(\"Test dataset shape:\", test.shape)"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Training dataset shape: (445, 4)\n",
      "Test dataset shape: (538, 4)\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "source": [
    "fig, ax = plt.subplots(figsize=(14, 6), dpi=80)\r\n",
    "ax.plot(train['Bearing 1'], label='Bearing 1', color='blue', animated = True, linewidth=1)\r\n",
    "ax.plot(train['Bearing 2'], label='Bearing 2', color='red', animated = True, linewidth=1)\r\n",
    "ax.plot(train['Bearing 3'], label='Bearing 3', color='green', animated = True, linewidth=1)\r\n",
    "ax.plot(train['Bearing 4'], label='Bearing 4', color='black', animated = True, linewidth=1)\r\n",
    "plt.legend(loc='lower left')\r\n",
    "ax.set_title('Bearing Sensor Training Data', fontsize=16)\r\n",
    "plt.show()"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1120x480 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "Let’s get a different perspective of the data by transforming the signal from the time domain to the frequency domain using a discrete Fourier transform."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "source": [
    "# transforming data from the time domain to the frequency domain using fast Fourier transform\r\n",
    "train_fft = np.fft.fft(train)\r\n",
    "test_fft = np.fft.fft(test)"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "source": [
    "# frequencies of the healthy sensor signal\r\n",
    "fig, ax = plt.subplots(figsize=(14, 6), dpi=80)\r\n",
    "ax.plot(train_fft[:,0].real, label='Bearing 1', color='blue', animated = True, linewidth=1)\r\n",
    "ax.plot(train_fft[:,1].imag, label='Bearing 2', color='red', animated = True, linewidth=1)\r\n",
    "ax.plot(train_fft[:,2].real, label='Bearing 3', color='green', animated = True, linewidth=1)\r\n",
    "ax.plot(train_fft[:,3].real, label='Bearing 4', color='black', animated = True, linewidth=1)\r\n",
    "plt.legend(loc='lower left')\r\n",
    "ax.set_title('Bearing Sensor Training Frequency Data', fontsize=16)\r\n",
    "plt.show()"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1120x480 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "source": [
    "# frequencies of the degrading sensor signal\r\n",
    "fig, ax = plt.subplots(figsize=(14, 6), dpi=80)\r\n",
    "ax.plot(test_fft[:,0].real, label='Bearing 1', color='blue', animated = True, linewidth=1)\r\n",
    "ax.plot(test_fft[:,1].imag, label='Bearing 2', color='red', animated = True, linewidth=1)\r\n",
    "ax.plot(test_fft[:,2].real, label='Bearing 3', color='green', animated = True, linewidth=1)\r\n",
    "ax.plot(test_fft[:,3].real, label='Bearing 4', color='black', animated = True, linewidth=1)\r\n",
    "plt.legend(loc='lower left')\r\n",
    "ax.set_title('Bearing Sensor Test Frequency Data', fontsize=16)\r\n",
    "plt.show()"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1120x480 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "source": [
    "# normalize the data\r\n",
    "scaler = MinMaxScaler()\r\n",
    "X_train = scaler.fit_transform(train)\r\n",
    "X_test = scaler.transform(test)"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "# Model"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Machine learning (ML)"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Isolation Forest"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "source": [
    "clf = IsolationForest(random_state = 42)\r\n",
    "clf.fit(X_train)\r\n",
    "pred = clf.predict(X_train)"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "source": [
    "pca = PCA(n_components=2)\r\n",
    "pca.fit(X_train)\r\n",
    "printcipalComponents = pca.transform(X_train)\r\n",
    "principalDf = pd.DataFrame(data=printcipalComponents, columns = ['principal component1', 'principal component2'])"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "source": [
    "fig = plt.figure(figsize = (8, 8))\r\n",
    "ax = fig.add_subplot(1, 1, 1)\r\n",
    "ax.set_xlabel('Principal Component 1', fontsize = 15)\r\n",
    "ax.set_ylabel('Principal Component 2', fontsize = 15)\r\n",
    "ax.set_title('2 component PCA', fontsize=20)\r\n",
    "\r\n",
    "targets = [1, -1]\r\n",
    "colors = ['g', 'b']\r\n",
    "for target, color in zip(targets,colors):\r\n",
    "    indicesToKeep = pred == target\r\n",
    "    ax.scatter(principalDf.loc[indicesToKeep, 'principal component1']\r\n",
    "               , principalDf.loc[indicesToKeep, 'principal component2']\r\n",
    "               , c = color\r\n",
    "               , s = 50)\r\n",
    "ax.legend(targets)\r\n",
    "ax.grid()"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "source": [
    "pred = clf.predict(X_test)"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "source": [
    "pca = PCA(n_components=2)\r\n",
    "pca.fit(X_test)\r\n",
    "printcipalComponents = pca.transform(X_test)\r\n",
    "principalDf = pd.DataFrame(data=printcipalComponents, columns = ['principal component1', 'principal component2'])"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "source": [
    "fig = plt.figure(figsize = (8, 8))\r\n",
    "ax = fig.add_subplot(1, 1, 1)\r\n",
    "ax.set_xlabel('Principal Component 1', fontsize = 15)\r\n",
    "ax.set_ylabel('Principal Component 2', fontsize = 15)\r\n",
    "ax.set_title('2 component PCA', fontsize=20)\r\n",
    "\r\n",
    "targets = [-1, 1]\r\n",
    "colors = ['b', 'g']\r\n",
    "for target, color in zip(targets,colors):\r\n",
    "    indicesToKeep = pred == target\r\n",
    "    ax.scatter(principalDf.loc[indicesToKeep, 'principal component1']\r\n",
    "               , principalDf.loc[indicesToKeep, 'principal component2']\r\n",
    "               , c = color\r\n",
    "               , s = 50)\r\n",
    "ax.legend(targets)\r\n",
    "ax.grid()"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### One Class SVM"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "source": [
    "clf = OneClassSVM(gamma='auto').fit(X_train)\r\n",
    "pred = clf.predict(X_train)\r\n",
    "pred_score = clf.score_samples(X_train)"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "source": [
    "pca = PCA(n_components=2)\r\n",
    "pca.fit(X_train)\r\n",
    "printcipalComponents = pca.transform(X_train)\r\n",
    "principalDf = pd.DataFrame(data=printcipalComponents, columns = ['principal component1', 'principal component2'])"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "source": [
    "fig = plt.figure(figsize = (8, 8))\r\n",
    "ax = fig.add_subplot(1, 1, 1)\r\n",
    "ax.set_xlabel('Principal Component 1', fontsize = 15)\r\n",
    "ax.set_ylabel('Principal Component 2', fontsize = 15)\r\n",
    "ax.set_title('2 component PCA', fontsize=20)\r\n",
    "\r\n",
    "targets = [1, -1]\r\n",
    "colors = ['g', 'b']\r\n",
    "for target, color in zip(targets,colors):\r\n",
    "    indicesToKeep = pred == target\r\n",
    "    ax.scatter(principalDf.loc[indicesToKeep, 'principal component1']\r\n",
    "               , principalDf.loc[indicesToKeep, 'principal component2']\r\n",
    "               , c = color\r\n",
    "               , s = 50)\r\n",
    "ax.legend(targets)\r\n",
    "ax.grid()"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "source": [
    "pred = clf.predict(X_test)"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "source": [
    "pca = PCA(n_components=2)\r\n",
    "pca.fit(X_test)\r\n",
    "printcipalComponents = pca.transform(X_test)\r\n",
    "principalDf = pd.DataFrame(data=printcipalComponents, columns = ['principal component1', 'principal component2'])"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "source": [
    "fig = plt.figure(figsize = (8, 8))\r\n",
    "ax = fig.add_subplot(1, 1, 1)\r\n",
    "ax.set_xlabel('Principal Component 1', fontsize = 15)\r\n",
    "ax.set_ylabel('Principal Component 2', fontsize = 15)\r\n",
    "ax.set_title('2 component PCA', fontsize=20)\r\n",
    "\r\n",
    "targets = [-1, 1]\r\n",
    "colors = ['b', 'g']\r\n",
    "for target, color in zip(targets,colors):\r\n",
    "    indicesToKeep = pred == target\r\n",
    "    ax.scatter(principalDf.loc[indicesToKeep, 'principal component1']\r\n",
    "               , principalDf.loc[indicesToKeep, 'principal component2']\r\n",
    "               , c = color\r\n",
    "               , s = 50)\r\n",
    "ax.legend(targets)\r\n",
    "ax.grid()"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Autoencoder"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### ANN Autoencoder"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "source": [
    "# define the autoencoder network model\r\n",
    "def ann_autoencoder_model(X):\r\n",
    "    input_dim = X.shape[1]\r\n",
    "    inputs = Input(shape=(X.shape[1],))\r\n",
    "    encode = Dense(input_dim//3*2, activation='relu',kernel_regularizer=regularizers.l2(0.01))(inputs)\r\n",
    "    encode = BatchNormalization()(encode)\r\n",
    "    encode = Dense(input_dim//3, activation='relu',kernel_regularizer=regularizers.l2(0.01))(encode)\r\n",
    "    encode = BatchNormalization()(encode)\r\n",
    "\r\n",
    "    decode = Dense(input_dim//3, activation='relu')(encode)\r\n",
    "    decode = BatchNormalization()(decode)\r\n",
    "    decode = Dense(input_dim//3*2, activation='relu')(decode)\r\n",
    "    decode = BatchNormalization()(decode)\r\n",
    "    output = Dense(input_dim, activation='sigmoid')(decode)\r\n",
    "    model = Model(inputs=inputs, outputs=output)\r\n",
    "    return model"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "source": [
    "# create the autoencoder model\r\n",
    "model = ann_autoencoder_model(X_train)\r\n",
    "model.compile(optimizer='adam', loss='mse',metrics=['mae'])\r\n",
    "model.summary()"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Model: \"model\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         [(None, 4)]               0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 2)                 10        \n",
      "_________________________________________________________________\n",
      "batch_normalization (BatchNo (None, 2)                 8         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 1)                 3         \n",
      "_________________________________________________________________\n",
      "batch_normalization_1 (Batch (None, 1)                 4         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 1)                 2         \n",
      "_________________________________________________________________\n",
      "batch_normalization_2 (Batch (None, 1)                 4         \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 2)                 4         \n",
      "_________________________________________________________________\n",
      "batch_normalization_3 (Batch (None, 2)                 8         \n",
      "_________________________________________________________________\n",
      "dense_4 (Dense)              (None, 4)                 12        \n",
      "=================================================================\n",
      "Total params: 55\n",
      "Trainable params: 43\n",
      "Non-trainable params: 12\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "source": [
    "# fit the model to the data\r\n",
    "nb_epochs = 100\r\n",
    "batch_size = 10\r\n",
    "\r\n",
    "es = EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=4)\r\n",
    "mc = ModelCheckpoint('./model/ann_autoencoder.h5', monitor='val_loss', mode='min', verbose=1, save_best_only=True)\r\n",
    "\r\n",
    "history = model.fit(X_train, X_train, epochs=nb_epochs, batch_size=batch_size,\r\n",
    "                    callbacks=[es, mc],validation_split=0.05)"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Epoch 1/100\n",
      "43/43 [==============================] - 2s 19ms/step - loss: 0.0977 - mae: 0.1945 - val_loss: 0.0810 - val_mae: 0.1751\n",
      "\n",
      "Epoch 00001: val_loss improved from inf to 0.08096, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 2/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0914 - mae: 0.1865 - val_loss: 0.0793 - val_mae: 0.1748\n",
      "\n",
      "Epoch 00002: val_loss improved from 0.08096 to 0.07926, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 3/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0799 - mae: 0.1688 - val_loss: 0.0768 - val_mae: 0.1718\n",
      "\n",
      "Epoch 00003: val_loss improved from 0.07926 to 0.07683, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 4/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0724 - mae: 0.1557 - val_loss: 0.0744 - val_mae: 0.1687\n",
      "\n",
      "Epoch 00004: val_loss improved from 0.07683 to 0.07439, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 5/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0706 - mae: 0.1536 - val_loss: 0.0724 - val_mae: 0.1663\n",
      "\n",
      "Epoch 00005: val_loss improved from 0.07439 to 0.07243, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 6/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0666 - mae: 0.1457 - val_loss: 0.0708 - val_mae: 0.1640\n",
      "\n",
      "Epoch 00006: val_loss improved from 0.07243 to 0.07078, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 7/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0670 - mae: 0.1505 - val_loss: 0.0690 - val_mae: 0.1616\n",
      "\n",
      "Epoch 00007: val_loss improved from 0.07078 to 0.06901, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 8/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0643 - mae: 0.1428 - val_loss: 0.0674 - val_mae: 0.1596\n",
      "\n",
      "Epoch 00008: val_loss improved from 0.06901 to 0.06742, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 9/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0622 - mae: 0.1432 - val_loss: 0.0660 - val_mae: 0.1590\n",
      "\n",
      "Epoch 00009: val_loss improved from 0.06742 to 0.06602, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 10/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0608 - mae: 0.1430 - val_loss: 0.0639 - val_mae: 0.1570\n",
      "\n",
      "Epoch 00010: val_loss improved from 0.06602 to 0.06392, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 11/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0591 - mae: 0.1391 - val_loss: 0.0620 - val_mae: 0.1562\n",
      "\n",
      "Epoch 00011: val_loss improved from 0.06392 to 0.06200, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 12/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0563 - mae: 0.1361 - val_loss: 0.0600 - val_mae: 0.1545\n",
      "\n",
      "Epoch 00012: val_loss improved from 0.06200 to 0.06004, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 13/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0551 - mae: 0.1358 - val_loss: 0.0581 - val_mae: 0.1526\n",
      "\n",
      "Epoch 00013: val_loss improved from 0.06004 to 0.05811, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 14/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0553 - mae: 0.1365 - val_loss: 0.0567 - val_mae: 0.1509\n",
      "\n",
      "Epoch 00014: val_loss improved from 0.05811 to 0.05672, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 15/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0534 - mae: 0.1368 - val_loss: 0.0557 - val_mae: 0.1505\n",
      "\n",
      "Epoch 00015: val_loss improved from 0.05672 to 0.05565, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 16/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0511 - mae: 0.1337 - val_loss: 0.0547 - val_mae: 0.1505\n",
      "\n",
      "Epoch 00016: val_loss improved from 0.05565 to 0.05466, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 17/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0490 - mae: 0.1308 - val_loss: 0.0532 - val_mae: 0.1495\n",
      "\n",
      "Epoch 00017: val_loss improved from 0.05466 to 0.05319, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 18/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0491 - mae: 0.1350 - val_loss: 0.0518 - val_mae: 0.1492\n",
      "\n",
      "Epoch 00018: val_loss improved from 0.05319 to 0.05175, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 19/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0477 - mae: 0.1321 - val_loss: 0.0506 - val_mae: 0.1488\n",
      "\n",
      "Epoch 00019: val_loss improved from 0.05175 to 0.05056, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 20/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0465 - mae: 0.1337 - val_loss: 0.0497 - val_mae: 0.1491\n",
      "\n",
      "Epoch 00020: val_loss improved from 0.05056 to 0.04966, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 21/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0451 - mae: 0.1303 - val_loss: 0.0495 - val_mae: 0.1508\n",
      "\n",
      "Epoch 00021: val_loss improved from 0.04966 to 0.04947, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 22/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0445 - mae: 0.1335 - val_loss: 0.0477 - val_mae: 0.1490\n",
      "\n",
      "Epoch 00022: val_loss improved from 0.04947 to 0.04771, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 23/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0428 - mae: 0.1306 - val_loss: 0.0462 - val_mae: 0.1468\n",
      "\n",
      "Epoch 00023: val_loss improved from 0.04771 to 0.04624, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 24/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0430 - mae: 0.1325 - val_loss: 0.0456 - val_mae: 0.1476\n",
      "\n",
      "Epoch 00024: val_loss improved from 0.04624 to 0.04556, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 25/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0417 - mae: 0.1329 - val_loss: 0.0442 - val_mae: 0.1459\n",
      "\n",
      "Epoch 00025: val_loss improved from 0.04556 to 0.04419, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 26/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0414 - mae: 0.1343 - val_loss: 0.0440 - val_mae: 0.1483\n",
      "\n",
      "Epoch 00026: val_loss improved from 0.04419 to 0.04400, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 27/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0395 - mae: 0.1307 - val_loss: 0.0429 - val_mae: 0.1472\n",
      "\n",
      "Epoch 00027: val_loss improved from 0.04400 to 0.04286, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 28/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0394 - mae: 0.1316 - val_loss: 0.0409 - val_mae: 0.1431\n",
      "\n",
      "Epoch 00028: val_loss improved from 0.04286 to 0.04093, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 29/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0374 - mae: 0.1293 - val_loss: 0.0400 - val_mae: 0.1420\n",
      "\n",
      "Epoch 00029: val_loss improved from 0.04093 to 0.04000, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 30/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0375 - mae: 0.1306 - val_loss: 0.0400 - val_mae: 0.1453\n",
      "\n",
      "Epoch 00030: val_loss improved from 0.04000 to 0.03998, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 31/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0347 - mae: 0.1265 - val_loss: 0.0379 - val_mae: 0.1380\n",
      "\n",
      "Epoch 00031: val_loss improved from 0.03998 to 0.03794, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 32/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0348 - mae: 0.1267 - val_loss: 0.0369 - val_mae: 0.1366\n",
      "\n",
      "Epoch 00032: val_loss improved from 0.03794 to 0.03690, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 33/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0340 - mae: 0.1247 - val_loss: 0.0361 - val_mae: 0.1354\n",
      "\n",
      "Epoch 00033: val_loss improved from 0.03690 to 0.03608, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 34/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0332 - mae: 0.1260 - val_loss: 0.0352 - val_mae: 0.1354\n",
      "\n",
      "Epoch 00034: val_loss improved from 0.03608 to 0.03522, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 35/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0333 - mae: 0.1264 - val_loss: 0.0341 - val_mae: 0.1347\n",
      "\n",
      "Epoch 00035: val_loss improved from 0.03522 to 0.03412, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 36/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0323 - mae: 0.1243 - val_loss: 0.0331 - val_mae: 0.1330\n",
      "\n",
      "Epoch 00036: val_loss improved from 0.03412 to 0.03308, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 37/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0296 - mae: 0.1213 - val_loss: 0.0325 - val_mae: 0.1334\n",
      "\n",
      "Epoch 00037: val_loss improved from 0.03308 to 0.03254, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 38/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0294 - mae: 0.1216 - val_loss: 0.0325 - val_mae: 0.1336\n",
      "\n",
      "Epoch 00038: val_loss improved from 0.03254 to 0.03245, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 39/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0294 - mae: 0.1230 - val_loss: 0.0306 - val_mae: 0.1294\n",
      "\n",
      "Epoch 00039: val_loss improved from 0.03245 to 0.03061, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 40/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0279 - mae: 0.1183 - val_loss: 0.0300 - val_mae: 0.1280\n",
      "\n",
      "Epoch 00040: val_loss improved from 0.03061 to 0.03001, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 41/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0275 - mae: 0.1186 - val_loss: 0.0295 - val_mae: 0.1291\n",
      "\n",
      "Epoch 00041: val_loss improved from 0.03001 to 0.02946, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 42/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0272 - mae: 0.1196 - val_loss: 0.0287 - val_mae: 0.1279\n",
      "\n",
      "Epoch 00042: val_loss improved from 0.02946 to 0.02870, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 43/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0266 - mae: 0.1188 - val_loss: 0.0278 - val_mae: 0.1267\n",
      "\n",
      "Epoch 00043: val_loss improved from 0.02870 to 0.02780, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 44/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0251 - mae: 0.1149 - val_loss: 0.0268 - val_mae: 0.1244\n",
      "\n",
      "Epoch 00044: val_loss improved from 0.02780 to 0.02678, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 45/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0249 - mae: 0.1148 - val_loss: 0.0259 - val_mae: 0.1230\n",
      "\n",
      "Epoch 00045: val_loss improved from 0.02678 to 0.02594, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 46/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0232 - mae: 0.1113 - val_loss: 0.0254 - val_mae: 0.1233\n",
      "\n",
      "Epoch 00046: val_loss improved from 0.02594 to 0.02543, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 47/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0221 - mae: 0.1101 - val_loss: 0.0244 - val_mae: 0.1207\n",
      "\n",
      "Epoch 00047: val_loss improved from 0.02543 to 0.02438, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 48/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0225 - mae: 0.1101 - val_loss: 0.0239 - val_mae: 0.1199\n",
      "\n",
      "Epoch 00048: val_loss improved from 0.02438 to 0.02389, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 49/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0222 - mae: 0.1096 - val_loss: 0.0228 - val_mae: 0.1173\n",
      "\n",
      "Epoch 00049: val_loss improved from 0.02389 to 0.02279, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 50/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0227 - mae: 0.1126 - val_loss: 0.0219 - val_mae: 0.1152\n",
      "\n",
      "Epoch 00050: val_loss improved from 0.02279 to 0.02191, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 51/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0207 - mae: 0.1073 - val_loss: 0.0218 - val_mae: 0.1154\n",
      "\n",
      "Epoch 00051: val_loss improved from 0.02191 to 0.02182, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 52/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0204 - mae: 0.1072 - val_loss: 0.0209 - val_mae: 0.1128\n",
      "\n",
      "Epoch 00052: val_loss improved from 0.02182 to 0.02095, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 53/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0203 - mae: 0.1089 - val_loss: 0.0207 - val_mae: 0.1122\n",
      "\n",
      "Epoch 00053: val_loss improved from 0.02095 to 0.02074, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 54/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0203 - mae: 0.1107 - val_loss: 0.0213 - val_mae: 0.1117\n",
      "\n",
      "Epoch 00054: val_loss did not improve from 0.02074\n",
      "Epoch 55/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0189 - mae: 0.1044 - val_loss: 0.0204 - val_mae: 0.1102\n",
      "\n",
      "Epoch 00055: val_loss improved from 0.02074 to 0.02041, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 56/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0183 - mae: 0.1029 - val_loss: 0.0191 - val_mae: 0.1087\n",
      "\n",
      "Epoch 00056: val_loss improved from 0.02041 to 0.01914, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 57/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0197 - mae: 0.1059 - val_loss: 0.0480 - val_mae: 0.1876\n",
      "\n",
      "Epoch 00057: val_loss did not improve from 0.01914\n",
      "Epoch 58/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0195 - mae: 0.1063 - val_loss: 0.0263 - val_mae: 0.1325\n",
      "\n",
      "Epoch 00058: val_loss did not improve from 0.01914\n",
      "Epoch 59/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0196 - mae: 0.1073 - val_loss: 0.0184 - val_mae: 0.1067\n",
      "\n",
      "Epoch 00059: val_loss improved from 0.01914 to 0.01838, saving model to ./model\\ann_autoencoder.h5\n",
      "Epoch 60/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0186 - mae: 0.1051 - val_loss: 0.0196 - val_mae: 0.1054\n",
      "\n",
      "Epoch 00060: val_loss did not improve from 0.01838\n",
      "Epoch 61/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0191 - mae: 0.1068 - val_loss: 0.0210 - val_mae: 0.1072\n",
      "\n",
      "Epoch 00061: val_loss did not improve from 0.01838\n",
      "Epoch 62/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0189 - mae: 0.1035 - val_loss: 0.0220 - val_mae: 0.1097\n",
      "\n",
      "Epoch 00062: val_loss did not improve from 0.01838\n",
      "Epoch 63/100\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0200 - mae: 0.1095 - val_loss: 0.0208 - val_mae: 0.1071\n",
      "\n",
      "Epoch 00063: val_loss did not improve from 0.01838\n",
      "Epoch 00063: early stopping\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "source": [
    "# plot the training losses\r\n",
    "fig, ax = plt.subplots(figsize=(14, 6), dpi=80)\r\n",
    "ax.plot(history.history['loss'], 'b', label='Train', linewidth=2)\r\n",
    "ax.plot(history.history['val_loss'], 'r', label='Validation', linewidth=2)\r\n",
    "ax.set_title('Model loss', fontsize=16)\r\n",
    "ax.set_ylabel('Loss (mae)')\r\n",
    "ax.set_xlabel('Epoch')\r\n",
    "ax.legend(loc='upper right')\r\n",
    "plt.show()"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1120x480 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "source": [
    "# load best model\r\n",
    "model = load_model('./model/ann_autoencoder.h5')"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "source": [
    "# plot the loss distribution of the training set\r\n",
    "X_pred = model.predict(X_train)\r\n",
    "X_pred = pd.DataFrame(X_pred, columns=train.columns)\r\n",
    "X_pred.index = train.index\r\n",
    "\r\n",
    "scored = pd.DataFrame(index=train.index)\r\n",
    "scored['Loss_mae'] = np.mean(np.abs(X_pred-X_train), axis = 1)\r\n",
    "plt.figure(figsize=(16,9), dpi=80)\r\n",
    "plt.title('Loss Distribution', fontsize=16)\r\n",
    "sns.distplot(scored['Loss_mae'], bins = 20, kde= True, color = 'blue');\r\n",
    "plt.xlim([0.0,.5])"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "C:\\Users\\user\\anaconda3\\envs\\py38tf2\\lib\\site-packages\\seaborn\\distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
      "  warnings.warn(msg, FutureWarning)\n"
     ]
    },
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "(0.0, 0.5)"
      ]
     },
     "metadata": {},
     "execution_count": 28
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1280x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "source": [
    "# calculate the loss on the test set\r\n",
    "X_pred = model.predict(X_test)\r\n",
    "X_pred = pd.DataFrame(X_pred, columns=test.columns)\r\n",
    "X_pred.index = test.index\r\n",
    "\r\n",
    "scored = pd.DataFrame(index=test.index)\r\n",
    "Xtest = X_test\r\n",
    "scored['Loss_mae'] = np.mean(np.abs(X_pred-Xtest), axis = 1)\r\n",
    "scored['Threshold'] = 0.275\r\n",
    "scored['Anomaly'] = scored['Loss_mae'] > scored['Threshold']\r\n",
    "scored.head()"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "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>Loss_mae</th>\n",
       "      <th>Threshold</th>\n",
       "      <th>Anomaly</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2004-02-15 12:52:39</th>\n",
       "      <td>0.111532</td>\n",
       "      <td>0.275</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-02-15 13:02:39</th>\n",
       "      <td>0.203659</td>\n",
       "      <td>0.275</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-02-15 13:12:39</th>\n",
       "      <td>0.093592</td>\n",
       "      <td>0.275</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-02-15 13:22:39</th>\n",
       "      <td>0.034371</td>\n",
       "      <td>0.275</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-02-15 13:32:39</th>\n",
       "      <td>0.071713</td>\n",
       "      <td>0.275</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     Loss_mae  Threshold  Anomaly\n",
       "2004-02-15 12:52:39  0.111532      0.275    False\n",
       "2004-02-15 13:02:39  0.203659      0.275    False\n",
       "2004-02-15 13:12:39  0.093592      0.275    False\n",
       "2004-02-15 13:22:39  0.034371      0.275    False\n",
       "2004-02-15 13:32:39  0.071713      0.275    False"
      ]
     },
     "metadata": {},
     "execution_count": 29
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "source": [
    "# calculate the same metrics for the training set \r\n",
    "# and merge all data in a single dataframe for plotting\r\n",
    "X_pred_train = model.predict(X_train)\r\n",
    "X_pred_train = pd.DataFrame(X_pred_train, columns=train.columns)\r\n",
    "X_pred_train.index = train.index\r\n",
    "\r\n",
    "scored_train = pd.DataFrame(index=train.index)\r\n",
    "scored_train['Loss_mae'] = np.mean(np.abs(X_pred_train-X_train), axis = 1)\r\n",
    "scored_train['Threshold'] = 0.275\r\n",
    "scored_train['Anomaly'] = scored_train['Loss_mae'] > scored_train['Threshold']\r\n",
    "scored = pd.concat([scored_train, scored])"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "source": [
    "# plot bearing failure time plot\r\n",
    "scored.plot(logy=True,  figsize=(16,9), ylim=[1e-2,1e2], color=['blue','red'])"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "metadata": {},
     "execution_count": 31
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1152x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### LSTM Autoencoder"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "source": [
    "# reshape inputs for LSTM [samples, timesteps, features]\r\n",
    "X_train = X_train.reshape(X_train.shape[0],  X_train.shape[1], 1)\r\n",
    "print(\"Training data shape:\", X_train.shape)\r\n",
    "X_test = X_test.reshape(X_test.shape[0], X_test.shape[1], 1)\r\n",
    "print(\"Test data shape:\", X_test.shape)"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Training data shape: (445, 4, 1)\n",
      "Test data shape: (538, 4, 1)\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "source": [
    "# define the autoencoder network model\r\n",
    "def lstm_autoencoder_model(X):\r\n",
    "    hidden_size = X.shape[1] * X.shape[2] // 2\r\n",
    "\r\n",
    "    inputs = Input(shape=(X.shape[1], X.shape[2]))\r\n",
    "    L1 = LSTM(hidden_size, return_sequences=True, kernel_regularizer=regularizers.l2(0.00))(inputs)\r\n",
    "    L2 = LSTM(hidden_size, return_sequences=False)(L1)\r\n",
    "    L3 = RepeatVector(X.shape[1])(L2)\r\n",
    "    L4 = LSTM(hidden_size, return_sequences=True)(L3)\r\n",
    "    L5 = LSTM(hidden_size, return_sequences=True)(L4)\r\n",
    "    output = TimeDistributed(Dense(X.shape[2]))(L5)    \r\n",
    "    model = Model(inputs=inputs, outputs=output)\r\n",
    "    return model"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "source": [
    "# create the autoencoder model\r\n",
    "model = lstm_autoencoder_model(X_train)\r\n",
    "model.compile(optimizer='adam', loss='mse',metrics=['mae'])\r\n",
    "model.summary()"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Model: \"model_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_2 (InputLayer)         [(None, 4, 1)]            0         \n",
      "_________________________________________________________________\n",
      "lstm (LSTM)                  (None, 4, 2)              32        \n",
      "_________________________________________________________________\n",
      "lstm_1 (LSTM)                (None, 2)                 40        \n",
      "_________________________________________________________________\n",
      "repeat_vector (RepeatVector) (None, 4, 2)              0         \n",
      "_________________________________________________________________\n",
      "lstm_2 (LSTM)                (None, 4, 2)              40        \n",
      "_________________________________________________________________\n",
      "lstm_3 (LSTM)                (None, 4, 2)              40        \n",
      "_________________________________________________________________\n",
      "time_distributed (TimeDistri (None, 4, 1)              3         \n",
      "=================================================================\n",
      "Total params: 155\n",
      "Trainable params: 155\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "source": [
    "# fit the model to the data\r\n",
    "nb_epochs = 100\r\n",
    "batch_size = 10\r\n",
    "\r\n",
    "es = EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=4)\r\n",
    "mc = ModelCheckpoint('./model/lstm_autoencoder.h5', monitor='val_loss', mode='min', verbose=1, save_best_only=True)\r\n",
    "\r\n",
    "history = model.fit(X_train, X_train, epochs=nb_epochs, batch_size=batch_size,\r\n",
    "                    callbacks=[es, mc], validation_split=0.05)"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Epoch 1/100\n",
      "43/43 [==============================] - 4s 23ms/step - loss: 0.2384 - mae: 0.4568 - val_loss: 0.1240 - val_mae: 0.3109\n",
      "\n",
      "Epoch 00001: val_loss improved from inf to 0.12398, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 2/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.1756 - mae: 0.3796 - val_loss: 0.0763 - val_mae: 0.2214\n",
      "\n",
      "Epoch 00002: val_loss improved from 0.12398 to 0.07625, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 3/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.1129 - mae: 0.2867 - val_loss: 0.0512 - val_mae: 0.1796\n",
      "\n",
      "Epoch 00003: val_loss improved from 0.07625 to 0.05116, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 4/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0621 - mae: 0.2067 - val_loss: 0.0570 - val_mae: 0.2014\n",
      "\n",
      "Epoch 00004: val_loss did not improve from 0.05116\n",
      "Epoch 5/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0558 - mae: 0.1946 - val_loss: 0.0555 - val_mae: 0.2007\n",
      "\n",
      "Epoch 00005: val_loss did not improve from 0.05116\n",
      "Epoch 6/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0507 - mae: 0.1865 - val_loss: 0.0528 - val_mae: 0.1963\n",
      "\n",
      "Epoch 00006: val_loss did not improve from 0.05116\n",
      "Epoch 7/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0478 - mae: 0.1788 - val_loss: 0.0507 - val_mae: 0.1930\n",
      "\n",
      "Epoch 00007: val_loss improved from 0.05116 to 0.05074, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 8/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0459 - mae: 0.1771 - val_loss: 0.0494 - val_mae: 0.1909\n",
      "\n",
      "Epoch 00008: val_loss improved from 0.05074 to 0.04936, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 9/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0428 - mae: 0.1699 - val_loss: 0.0480 - val_mae: 0.1884\n",
      "\n",
      "Epoch 00009: val_loss improved from 0.04936 to 0.04797, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 10/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0420 - mae: 0.1685 - val_loss: 0.0454 - val_mae: 0.1832\n",
      "\n",
      "Epoch 00010: val_loss improved from 0.04797 to 0.04544, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 11/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0391 - mae: 0.1610 - val_loss: 0.0453 - val_mae: 0.1832\n",
      "\n",
      "Epoch 00011: val_loss improved from 0.04544 to 0.04534, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 12/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0380 - mae: 0.1593 - val_loss: 0.0432 - val_mae: 0.1783\n",
      "\n",
      "Epoch 00012: val_loss improved from 0.04534 to 0.04324, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 13/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0357 - mae: 0.1544 - val_loss: 0.0413 - val_mae: 0.1738\n",
      "\n",
      "Epoch 00013: val_loss improved from 0.04324 to 0.04130, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 14/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0363 - mae: 0.1554 - val_loss: 0.0402 - val_mae: 0.1711\n",
      "\n",
      "Epoch 00014: val_loss improved from 0.04130 to 0.04023, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 15/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0339 - mae: 0.1498 - val_loss: 0.0396 - val_mae: 0.1696\n",
      "\n",
      "Epoch 00015: val_loss improved from 0.04023 to 0.03958, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 16/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0324 - mae: 0.1465 - val_loss: 0.0381 - val_mae: 0.1663\n",
      "\n",
      "Epoch 00016: val_loss improved from 0.03958 to 0.03814, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 17/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0304 - mae: 0.1413 - val_loss: 0.0370 - val_mae: 0.1636\n",
      "\n",
      "Epoch 00017: val_loss improved from 0.03814 to 0.03698, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 18/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0303 - mae: 0.1411 - val_loss: 0.0363 - val_mae: 0.1619\n",
      "\n",
      "Epoch 00018: val_loss improved from 0.03698 to 0.03631, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 19/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0299 - mae: 0.1386 - val_loss: 0.0347 - val_mae: 0.1577\n",
      "\n",
      "Epoch 00019: val_loss improved from 0.03631 to 0.03470, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 20/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0290 - mae: 0.1377 - val_loss: 0.0347 - val_mae: 0.1573\n",
      "\n",
      "Epoch 00020: val_loss improved from 0.03470 to 0.03465, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 21/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0278 - mae: 0.1347 - val_loss: 0.0347 - val_mae: 0.1577\n",
      "\n",
      "Epoch 00021: val_loss did not improve from 0.03465\n",
      "Epoch 22/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0279 - mae: 0.1353 - val_loss: 0.0343 - val_mae: 0.1567\n",
      "\n",
      "Epoch 00022: val_loss improved from 0.03465 to 0.03428, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 23/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0269 - mae: 0.1328 - val_loss: 0.0324 - val_mae: 0.1518\n",
      "\n",
      "Epoch 00023: val_loss improved from 0.03428 to 0.03239, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 24/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0269 - mae: 0.1315 - val_loss: 0.0324 - val_mae: 0.1520\n",
      "\n",
      "Epoch 00024: val_loss did not improve from 0.03239\n",
      "Epoch 25/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0257 - mae: 0.1294 - val_loss: 0.0321 - val_mae: 0.1516\n",
      "\n",
      "Epoch 00025: val_loss improved from 0.03239 to 0.03214, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 26/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0245 - mae: 0.1262 - val_loss: 0.0290 - val_mae: 0.1434\n",
      "\n",
      "Epoch 00026: val_loss improved from 0.03214 to 0.02902, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 27/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0226 - mae: 0.1201 - val_loss: 0.0283 - val_mae: 0.1418\n",
      "\n",
      "Epoch 00027: val_loss improved from 0.02902 to 0.02829, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 28/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0214 - mae: 0.1175 - val_loss: 0.0276 - val_mae: 0.1402\n",
      "\n",
      "Epoch 00028: val_loss improved from 0.02829 to 0.02758, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 29/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0184 - mae: 0.1083 - val_loss: 0.0250 - val_mae: 0.1321\n",
      "\n",
      "Epoch 00029: val_loss improved from 0.02758 to 0.02497, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 30/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0180 - mae: 0.1065 - val_loss: 0.0238 - val_mae: 0.1268\n",
      "\n",
      "Epoch 00030: val_loss improved from 0.02497 to 0.02380, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 31/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0164 - mae: 0.1016 - val_loss: 0.0234 - val_mae: 0.1253\n",
      "\n",
      "Epoch 00031: val_loss improved from 0.02380 to 0.02336, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 32/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0167 - mae: 0.1024 - val_loss: 0.0228 - val_mae: 0.1222\n",
      "\n",
      "Epoch 00032: val_loss improved from 0.02336 to 0.02277, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 33/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0164 - mae: 0.1016 - val_loss: 0.0224 - val_mae: 0.1212\n",
      "\n",
      "Epoch 00033: val_loss improved from 0.02277 to 0.02242, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 34/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0169 - mae: 0.1041 - val_loss: 0.0222 - val_mae: 0.1204\n",
      "\n",
      "Epoch 00034: val_loss improved from 0.02242 to 0.02219, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 35/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0169 - mae: 0.1031 - val_loss: 0.0220 - val_mae: 0.1195\n",
      "\n",
      "Epoch 00035: val_loss improved from 0.02219 to 0.02196, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 36/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0173 - mae: 0.1038 - val_loss: 0.0218 - val_mae: 0.1191\n",
      "\n",
      "Epoch 00036: val_loss improved from 0.02196 to 0.02179, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 37/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0151 - mae: 0.0976 - val_loss: 0.0216 - val_mae: 0.1180\n",
      "\n",
      "Epoch 00037: val_loss improved from 0.02179 to 0.02161, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 38/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0164 - mae: 0.1016 - val_loss: 0.0214 - val_mae: 0.1173\n",
      "\n",
      "Epoch 00038: val_loss improved from 0.02161 to 0.02143, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 39/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0164 - mae: 0.1011 - val_loss: 0.0212 - val_mae: 0.1169\n",
      "\n",
      "Epoch 00039: val_loss improved from 0.02143 to 0.02118, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 40/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0165 - mae: 0.1004 - val_loss: 0.0211 - val_mae: 0.1163\n",
      "\n",
      "Epoch 00040: val_loss improved from 0.02118 to 0.02108, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 41/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0154 - mae: 0.0986 - val_loss: 0.0209 - val_mae: 0.1156\n",
      "\n",
      "Epoch 00041: val_loss improved from 0.02108 to 0.02086, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 42/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0165 - mae: 0.1014 - val_loss: 0.0208 - val_mae: 0.1159\n",
      "\n",
      "Epoch 00042: val_loss improved from 0.02086 to 0.02083, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 43/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0163 - mae: 0.1015 - val_loss: 0.0207 - val_mae: 0.1153\n",
      "\n",
      "Epoch 00043: val_loss improved from 0.02083 to 0.02070, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 44/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0155 - mae: 0.0986 - val_loss: 0.0206 - val_mae: 0.1146\n",
      "\n",
      "Epoch 00044: val_loss improved from 0.02070 to 0.02058, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 45/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0158 - mae: 0.1004 - val_loss: 0.0204 - val_mae: 0.1144\n",
      "\n",
      "Epoch 00045: val_loss improved from 0.02058 to 0.02040, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 46/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0153 - mae: 0.0989 - val_loss: 0.0203 - val_mae: 0.1143\n",
      "\n",
      "Epoch 00046: val_loss improved from 0.02040 to 0.02035, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 47/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0149 - mae: 0.0973 - val_loss: 0.0202 - val_mae: 0.1138\n",
      "\n",
      "Epoch 00047: val_loss improved from 0.02035 to 0.02025, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 48/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0146 - mae: 0.0966 - val_loss: 0.0201 - val_mae: 0.1139\n",
      "\n",
      "Epoch 00048: val_loss improved from 0.02025 to 0.02005, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 49/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0153 - mae: 0.0978 - val_loss: 0.0200 - val_mae: 0.1136\n",
      "\n",
      "Epoch 00049: val_loss improved from 0.02005 to 0.02001, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 50/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0162 - mae: 0.0994 - val_loss: 0.0199 - val_mae: 0.1138\n",
      "\n",
      "Epoch 00050: val_loss improved from 0.02001 to 0.01992, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 51/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0158 - mae: 0.1004 - val_loss: 0.0198 - val_mae: 0.1131\n",
      "\n",
      "Epoch 00051: val_loss improved from 0.01992 to 0.01981, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 52/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0156 - mae: 0.1002 - val_loss: 0.0197 - val_mae: 0.1132\n",
      "\n",
      "Epoch 00052: val_loss improved from 0.01981 to 0.01973, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 53/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0155 - mae: 0.0991 - val_loss: 0.0196 - val_mae: 0.1128\n",
      "\n",
      "Epoch 00053: val_loss improved from 0.01973 to 0.01955, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 54/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0153 - mae: 0.0984 - val_loss: 0.0195 - val_mae: 0.1122\n",
      "\n",
      "Epoch 00054: val_loss improved from 0.01955 to 0.01951, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 55/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0152 - mae: 0.0966 - val_loss: 0.0194 - val_mae: 0.1122\n",
      "\n",
      "Epoch 00055: val_loss improved from 0.01951 to 0.01945, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 56/100\n",
      "43/43 [==============================] - 0s 10ms/step - loss: 0.0148 - mae: 0.0972 - val_loss: 0.0192 - val_mae: 0.1116\n",
      "\n",
      "Epoch 00056: val_loss improved from 0.01945 to 0.01915, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 57/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0146 - mae: 0.0962 - val_loss: 0.0192 - val_mae: 0.1116\n",
      "\n",
      "Epoch 00057: val_loss did not improve from 0.01915\n",
      "Epoch 58/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0154 - mae: 0.0985 - val_loss: 0.0190 - val_mae: 0.1112\n",
      "\n",
      "Epoch 00058: val_loss improved from 0.01915 to 0.01896, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 59/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0158 - mae: 0.0992 - val_loss: 0.0190 - val_mae: 0.1115\n",
      "\n",
      "Epoch 00059: val_loss did not improve from 0.01896\n",
      "Epoch 60/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0149 - mae: 0.0967 - val_loss: 0.0189 - val_mae: 0.1110\n",
      "\n",
      "Epoch 00060: val_loss improved from 0.01896 to 0.01893, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 61/100\n",
      "43/43 [==============================] - 0s 10ms/step - loss: 0.0156 - mae: 0.1000 - val_loss: 0.0188 - val_mae: 0.1108\n",
      "\n",
      "Epoch 00061: val_loss improved from 0.01893 to 0.01877, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 62/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0151 - mae: 0.0973 - val_loss: 0.0188 - val_mae: 0.1106\n",
      "\n",
      "Epoch 00062: val_loss did not improve from 0.01877\n",
      "Epoch 63/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0157 - mae: 0.0988 - val_loss: 0.0187 - val_mae: 0.1106\n",
      "\n",
      "Epoch 00063: val_loss improved from 0.01877 to 0.01869, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 64/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0152 - mae: 0.0977 - val_loss: 0.0185 - val_mae: 0.1103\n",
      "\n",
      "Epoch 00064: val_loss improved from 0.01869 to 0.01850, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 65/100\n",
      "43/43 [==============================] - 0s 10ms/step - loss: 0.0149 - mae: 0.0970 - val_loss: 0.0184 - val_mae: 0.1102\n",
      "\n",
      "Epoch 00065: val_loss improved from 0.01850 to 0.01842, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 66/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0145 - mae: 0.0961 - val_loss: 0.0183 - val_mae: 0.1100\n",
      "\n",
      "Epoch 00066: val_loss improved from 0.01842 to 0.01834, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 67/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0151 - mae: 0.0971 - val_loss: 0.0181 - val_mae: 0.1091\n",
      "\n",
      "Epoch 00067: val_loss improved from 0.01834 to 0.01813, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 68/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0150 - mae: 0.0983 - val_loss: 0.0183 - val_mae: 0.1093\n",
      "\n",
      "Epoch 00068: val_loss did not improve from 0.01813\n",
      "Epoch 69/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0147 - mae: 0.0972 - val_loss: 0.0181 - val_mae: 0.1092\n",
      "\n",
      "Epoch 00069: val_loss did not improve from 0.01813\n",
      "Epoch 70/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0156 - mae: 0.0998 - val_loss: 0.0180 - val_mae: 0.1086\n",
      "\n",
      "Epoch 00070: val_loss improved from 0.01813 to 0.01799, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 71/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0146 - mae: 0.0961 - val_loss: 0.0181 - val_mae: 0.1089\n",
      "\n",
      "Epoch 00071: val_loss did not improve from 0.01799\n",
      "Epoch 72/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0147 - mae: 0.0977 - val_loss: 0.0179 - val_mae: 0.1085\n",
      "\n",
      "Epoch 00072: val_loss improved from 0.01799 to 0.01786, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 73/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0150 - mae: 0.0990 - val_loss: 0.0177 - val_mae: 0.1081\n",
      "\n",
      "Epoch 00073: val_loss improved from 0.01786 to 0.01772, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 74/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0143 - mae: 0.0965 - val_loss: 0.0177 - val_mae: 0.1082\n",
      "\n",
      "Epoch 00074: val_loss improved from 0.01772 to 0.01765, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 75/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0145 - mae: 0.0958 - val_loss: 0.0176 - val_mae: 0.1076\n",
      "\n",
      "Epoch 00075: val_loss improved from 0.01765 to 0.01763, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 76/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0148 - mae: 0.0973 - val_loss: 0.0176 - val_mae: 0.1079\n",
      "\n",
      "Epoch 00076: val_loss improved from 0.01763 to 0.01762, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 77/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0144 - mae: 0.0957 - val_loss: 0.0175 - val_mae: 0.1079\n",
      "\n",
      "Epoch 00077: val_loss improved from 0.01762 to 0.01755, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 78/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0146 - mae: 0.0973 - val_loss: 0.0177 - val_mae: 0.1080\n",
      "\n",
      "Epoch 00078: val_loss did not improve from 0.01755\n",
      "Epoch 79/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0145 - mae: 0.0969 - val_loss: 0.0174 - val_mae: 0.1077\n",
      "\n",
      "Epoch 00079: val_loss improved from 0.01755 to 0.01742, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 80/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0134 - mae: 0.0919 - val_loss: 0.0174 - val_mae: 0.1075\n",
      "\n",
      "Epoch 00080: val_loss improved from 0.01742 to 0.01735, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 81/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0146 - mae: 0.0968 - val_loss: 0.0173 - val_mae: 0.1079\n",
      "\n",
      "Epoch 00081: val_loss improved from 0.01735 to 0.01732, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 82/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0138 - mae: 0.0935 - val_loss: 0.0173 - val_mae: 0.1075\n",
      "\n",
      "Epoch 00082: val_loss did not improve from 0.01732\n",
      "Epoch 83/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0146 - mae: 0.0979 - val_loss: 0.0174 - val_mae: 0.1074\n",
      "\n",
      "Epoch 00083: val_loss did not improve from 0.01732\n",
      "Epoch 84/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0149 - mae: 0.0996 - val_loss: 0.0173 - val_mae: 0.1072\n",
      "\n",
      "Epoch 00084: val_loss improved from 0.01732 to 0.01728, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 85/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0141 - mae: 0.0951 - val_loss: 0.0173 - val_mae: 0.1073\n",
      "\n",
      "Epoch 00085: val_loss improved from 0.01728 to 0.01725, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 86/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0145 - mae: 0.0968 - val_loss: 0.0172 - val_mae: 0.1077\n",
      "\n",
      "Epoch 00086: val_loss improved from 0.01725 to 0.01723, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 87/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0135 - mae: 0.0942 - val_loss: 0.0169 - val_mae: 0.1068\n",
      "\n",
      "Epoch 00087: val_loss improved from 0.01723 to 0.01690, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 88/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0147 - mae: 0.0965 - val_loss: 0.0170 - val_mae: 0.1075\n",
      "\n",
      "Epoch 00088: val_loss did not improve from 0.01690\n",
      "Epoch 89/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0139 - mae: 0.0939 - val_loss: 0.0167 - val_mae: 0.1064\n",
      "\n",
      "Epoch 00089: val_loss improved from 0.01690 to 0.01670, saving model to ./model\\lstm_autoencoder.h5\n",
      "Epoch 90/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0135 - mae: 0.0934 - val_loss: 0.0168 - val_mae: 0.1069\n",
      "\n",
      "Epoch 00090: val_loss did not improve from 0.01670\n",
      "Epoch 91/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0141 - mae: 0.0942 - val_loss: 0.0168 - val_mae: 0.1072\n",
      "\n",
      "Epoch 00091: val_loss did not improve from 0.01670\n",
      "Epoch 92/100\n",
      "43/43 [==============================] - 0s 9ms/step - loss: 0.0136 - mae: 0.0943 - val_loss: 0.0171 - val_mae: 0.1069\n",
      "\n",
      "Epoch 00092: val_loss did not improve from 0.01670\n",
      "Epoch 93/100\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0137 - mae: 0.0934 - val_loss: 0.0167 - val_mae: 0.1067\n",
      "\n",
      "Epoch 00093: val_loss did not improve from 0.01670\n",
      "Epoch 00093: early stopping\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "source": [
    "# plot the training losses\r\n",
    "fig, ax = plt.subplots(figsize=(14, 6), dpi=80)\r\n",
    "ax.plot(history.history['loss'], 'b', label='Train', linewidth=2)\r\n",
    "ax.plot(history.history['val_loss'], 'r', label='Validation', linewidth=2)\r\n",
    "ax.set_title('Model loss', fontsize=16)\r\n",
    "ax.set_ylabel('Loss (mae)')\r\n",
    "ax.set_xlabel('Epoch')\r\n",
    "ax.legend(loc='upper right')\r\n",
    "plt.show()"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1120x480 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "source": [
    "# load best model\r\n",
    "model = load_model('./model/lstm_autoencoder.h5')"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "source": [
    "# plot the loss distribution of the training set\r\n",
    "X_pred = model.predict(X_train)\r\n",
    "X_pred = X_pred.reshape(X_pred.shape[0], X_pred.shape[1])\r\n",
    "X_pred = pd.DataFrame(X_pred, columns=train.columns)\r\n",
    "X_pred.index = train.index\r\n",
    "\r\n",
    "scored = pd.DataFrame(index=train.index)\r\n",
    "Xtrain = X_train.reshape(X_train.shape[0], X_train.shape[1])\r\n",
    "scored['Loss_mae'] = np.mean(np.abs(X_pred-Xtrain), axis = 1)\r\n",
    "plt.figure(figsize=(16,9), dpi=80)\r\n",
    "plt.title('Loss Distribution', fontsize=16)\r\n",
    "sns.distplot(scored['Loss_mae'], bins = 20, kde= True, color = 'blue');\r\n",
    "plt.xlim([0.0,.5])"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "C:\\Users\\user\\anaconda3\\envs\\py38tf2\\lib\\site-packages\\seaborn\\distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
      "  warnings.warn(msg, FutureWarning)\n"
     ]
    },
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "(0.0, 0.5)"
      ]
     },
     "metadata": {},
     "execution_count": 38
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1280x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "source": [
    "# calculate the loss on the test set\r\n",
    "X_pred = model.predict(X_test)\r\n",
    "X_pred = X_pred.reshape(X_pred.shape[0], X_pred.shape[1])\r\n",
    "X_pred = pd.DataFrame(X_pred, columns=test.columns)\r\n",
    "X_pred.index = test.index\r\n",
    "\r\n",
    "scored = pd.DataFrame(index=test.index)\r\n",
    "Xtest = X_test.reshape(X_test.shape[0], X_test.shape[1])\r\n",
    "scored['Loss_mae'] = np.mean(np.abs(X_pred-Xtest), axis = 1)\r\n",
    "scored['Threshold'] = 0.25\r\n",
    "scored['Anomaly'] = scored['Loss_mae'] > scored['Threshold']\r\n",
    "scored.head()"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "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>Loss_mae</th>\n",
       "      <th>Threshold</th>\n",
       "      <th>Anomaly</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2004-02-15 12:52:39</th>\n",
       "      <td>0.106398</td>\n",
       "      <td>0.25</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-02-15 13:02:39</th>\n",
       "      <td>0.212795</td>\n",
       "      <td>0.25</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-02-15 13:12:39</th>\n",
       "      <td>0.081269</td>\n",
       "      <td>0.25</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-02-15 13:22:39</th>\n",
       "      <td>0.014543</td>\n",
       "      <td>0.25</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-02-15 13:32:39</th>\n",
       "      <td>0.073532</td>\n",
       "      <td>0.25</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     Loss_mae  Threshold  Anomaly\n",
       "2004-02-15 12:52:39  0.106398       0.25    False\n",
       "2004-02-15 13:02:39  0.212795       0.25    False\n",
       "2004-02-15 13:12:39  0.081269       0.25    False\n",
       "2004-02-15 13:22:39  0.014543       0.25    False\n",
       "2004-02-15 13:32:39  0.073532       0.25    False"
      ]
     },
     "metadata": {},
     "execution_count": 39
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "source": [
    "# calculate the same metrics for the training set \r\n",
    "# and merge all data in a single dataframe for plotting\r\n",
    "X_pred_train = model.predict(X_train)\r\n",
    "X_pred_train = X_pred_train.reshape(X_pred_train.shape[0], X_pred_train.shape[1])\r\n",
    "X_pred_train = pd.DataFrame(X_pred_train, columns=train.columns)\r\n",
    "X_pred_train.index = train.index\r\n",
    "\r\n",
    "scored_train = pd.DataFrame(index=train.index)\r\n",
    "scored_train['Loss_mae'] = np.mean(np.abs(X_pred_train-Xtrain), axis = 1)\r\n",
    "scored_train['Threshold'] = 0.25\r\n",
    "scored_train['Anomaly'] = scored_train['Loss_mae'] > scored_train['Threshold']\r\n",
    "scored = pd.concat([scored_train, scored])"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "Having calculated the loss distribution and the anomaly threshold, we can visualize the model output in the time leading up to the bearing failure."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "source": [
    "# plot bearing failure time plot\r\n",
    "scored.plot(logy=True,  figsize=(16,9), ylim=[1e-2,1e2], color=['blue','red'])"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "metadata": {},
     "execution_count": 41
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1152x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Unsupervised log message anomaly detection\r\n",
    "2 autoencoder + Isolation Forest"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "source": [
    "# reset train, test data\r\n",
    "scaler = MinMaxScaler()\r\n",
    "X_train = scaler.fit_transform(train)\r\n",
    "X_test = scaler.transform(test)"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "source": [
    "# 1st Autoencoder\r\n",
    "autoencoder1 = ann_autoencoder_model(X_train)\r\n",
    "autoencoder1.compile(optimizer='adam', loss='mse', metrics=['mae'])\r\n",
    "\r\n",
    "early_stopping = EarlyStopping(monitor='loss', mode='min')\r\n",
    "history1 = autoencoder1.fit(X_train, X_train,\r\n",
    "        epochs=200,\r\n",
    "        batch_size=10,\r\n",
    "        shuffle=True,\r\n",
    "        validation_split=0.05,\r\n",
    "        callbacks = [early_stopping])\r\n",
    "print('========== First Autoencoder Finish ==========')\r\n",
    "\r\n",
    "# Isolation Forest\r\n",
    "f1_pre = autoencoder1.predict(X_train)\r\n",
    "\r\n",
    "clf = IsolationForest(random_state = 42, contamination = 0.3)\r\n",
    "clf.fit(f1_pre)\r\n",
    "pred = clf.predict(f1_pre)\r\n",
    "\r\n",
    "p1 = f1_pre[pred==1]\r\n",
    "p2 = f1_pre[pred==-1]\r\n",
    "\r\n",
    "# 2st Autoencoder\r\n",
    "autoencoder2 = ann_autoencoder_model(p1)\r\n",
    "autoencoder2.compile(optimizer='adam', loss='mse', metrics=['mae'])\r\n",
    "\r\n",
    "history2 = autoencoder2.fit(p1, p1,\r\n",
    "        epochs=200,\r\n",
    "        batch_size=10,\r\n",
    "        shuffle=True,\r\n",
    "        validation_split=0.05,\r\n",
    "        callbacks = [early_stopping])\r\n",
    "print('========== Second Autoencoder Finish ==========')"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Epoch 1/200\n",
      "43/43 [==============================] - 1s 10ms/step - loss: 0.0603 - mae: 0.1406 - val_loss: 0.0678 - val_mae: 0.1707\n",
      "Epoch 2/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0579 - mae: 0.1426 - val_loss: 0.0628 - val_mae: 0.1675\n",
      "Epoch 3/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0540 - mae: 0.1419 - val_loss: 0.0586 - val_mae: 0.1648\n",
      "Epoch 4/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0506 - mae: 0.1406 - val_loss: 0.0548 - val_mae: 0.1625\n",
      "Epoch 5/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0485 - mae: 0.1414 - val_loss: 0.0519 - val_mae: 0.1611\n",
      "Epoch 6/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0440 - mae: 0.1373 - val_loss: 0.0492 - val_mae: 0.1597\n",
      "Epoch 7/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0431 - mae: 0.1391 - val_loss: 0.0469 - val_mae: 0.1584\n",
      "Epoch 8/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0425 - mae: 0.1402 - val_loss: 0.0449 - val_mae: 0.1573\n",
      "Epoch 9/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0406 - mae: 0.1398 - val_loss: 0.0434 - val_mae: 0.1573\n",
      "Epoch 10/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0394 - mae: 0.1423 - val_loss: 0.0417 - val_mae: 0.1559\n",
      "Epoch 11/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0384 - mae: 0.1399 - val_loss: 0.0406 - val_mae: 0.1559\n",
      "Epoch 12/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0360 - mae: 0.1381 - val_loss: 0.0394 - val_mae: 0.1555\n",
      "Epoch 13/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0349 - mae: 0.1376 - val_loss: 0.0383 - val_mae: 0.1546\n",
      "Epoch 14/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0361 - mae: 0.1413 - val_loss: 0.0374 - val_mae: 0.1543\n",
      "Epoch 15/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0345 - mae: 0.1407 - val_loss: 0.0369 - val_mae: 0.1547\n",
      "Epoch 16/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0330 - mae: 0.1390 - val_loss: 0.0363 - val_mae: 0.1548\n",
      "Epoch 17/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0314 - mae: 0.1348 - val_loss: 0.0355 - val_mae: 0.1537\n",
      "Epoch 18/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0316 - mae: 0.1384 - val_loss: 0.0351 - val_mae: 0.1540\n",
      "Epoch 19/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0317 - mae: 0.1370 - val_loss: 0.0348 - val_mae: 0.1540\n",
      "Epoch 20/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0314 - mae: 0.1391 - val_loss: 0.0346 - val_mae: 0.1542\n",
      "Epoch 21/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0305 - mae: 0.1365 - val_loss: 0.0345 - val_mae: 0.1547\n",
      "Epoch 22/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0312 - mae: 0.1410 - val_loss: 0.0342 - val_mae: 0.1545\n",
      "Epoch 23/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0306 - mae: 0.1391 - val_loss: 0.0340 - val_mae: 0.1545\n",
      "Epoch 24/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0314 - mae: 0.1401 - val_loss: 0.0338 - val_mae: 0.1544\n",
      "Epoch 25/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0308 - mae: 0.1395 - val_loss: 0.0336 - val_mae: 0.1544\n",
      "Epoch 26/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0308 - mae: 0.1408 - val_loss: 0.0336 - val_mae: 0.1545\n",
      "Epoch 27/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0305 - mae: 0.1397 - val_loss: 0.0333 - val_mae: 0.1540\n",
      "Epoch 28/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0305 - mae: 0.1382 - val_loss: 0.0330 - val_mae: 0.1535\n",
      "Epoch 29/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0294 - mae: 0.1376 - val_loss: 0.0332 - val_mae: 0.1541\n",
      "Epoch 30/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0304 - mae: 0.1395 - val_loss: 0.0331 - val_mae: 0.1540\n",
      "Epoch 31/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0288 - mae: 0.1361 - val_loss: 0.0330 - val_mae: 0.1539\n",
      "Epoch 32/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0300 - mae: 0.1386 - val_loss: 0.0330 - val_mae: 0.1539\n",
      "Epoch 33/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0294 - mae: 0.1366 - val_loss: 0.0330 - val_mae: 0.1540\n",
      "Epoch 34/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0300 - mae: 0.1387 - val_loss: 0.0331 - val_mae: 0.1544\n",
      "Epoch 35/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0312 - mae: 0.1418 - val_loss: 0.0330 - val_mae: 0.1540\n",
      "Epoch 36/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0302 - mae: 0.1381 - val_loss: 0.0328 - val_mae: 0.1537\n",
      "Epoch 37/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0284 - mae: 0.1370 - val_loss: 0.0332 - val_mae: 0.1547\n",
      "Epoch 38/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0286 - mae: 0.1358 - val_loss: 0.0331 - val_mae: 0.1545\n",
      "Epoch 39/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0304 - mae: 0.1402 - val_loss: 0.0330 - val_mae: 0.1543\n",
      "Epoch 40/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0298 - mae: 0.1392 - val_loss: 0.0329 - val_mae: 0.1541\n",
      "Epoch 41/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0295 - mae: 0.1383 - val_loss: 0.0327 - val_mae: 0.1535\n",
      "Epoch 42/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0303 - mae: 0.1394 - val_loss: 0.0329 - val_mae: 0.1540\n",
      "Epoch 43/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0293 - mae: 0.1374 - val_loss: 0.0330 - val_mae: 0.1542\n",
      "Epoch 44/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0300 - mae: 0.1394 - val_loss: 0.0328 - val_mae: 0.1537\n",
      "========== First Autoencoder Finish ==========\n",
      "Epoch 1/200\n",
      "43/43 [==============================] - 1s 10ms/step - loss: 0.0604 - mae: 0.0403 - val_loss: 0.0555 - val_mae: 0.0302\n",
      "Epoch 2/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0540 - mae: 0.0279 - val_loss: 0.0498 - val_mae: 0.0205\n",
      "Epoch 3/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0485 - mae: 0.0185 - val_loss: 0.0447 - val_mae: 0.0124\n",
      "Epoch 4/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0436 - mae: 0.0106 - val_loss: 0.0404 - val_mae: 0.0062\n",
      "Epoch 5/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0394 - mae: 0.0051 - val_loss: 0.0366 - val_mae: 0.0025\n",
      "Epoch 6/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0357 - mae: 0.0020 - val_loss: 0.0331 - val_mae: 8.6791e-04\n",
      "Epoch 7/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0323 - mae: 6.5897e-04 - val_loss: 0.0299 - val_mae: 2.3870e-04\n",
      "Epoch 8/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0292 - mae: 1.7731e-04 - val_loss: 0.0271 - val_mae: 5.0068e-05\n",
      "Epoch 9/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0264 - mae: 3.5679e-05 - val_loss: 0.0245 - val_mae: 1.0096e-05\n",
      "Epoch 10/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0239 - mae: 9.0643e-06 - val_loss: 0.0221 - val_mae: 9.7677e-06\n",
      "Epoch 11/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0215 - mae: 8.8218e-06 - val_loss: 0.0199 - val_mae: 6.6981e-06\n",
      "Epoch 12/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0194 - mae: 7.0769e-06 - val_loss: 0.0180 - val_mae: 7.0333e-06\n",
      "Epoch 13/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0175 - mae: 7.0048e-06 - val_loss: 0.0162 - val_mae: 6.1095e-06\n",
      "Epoch 14/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0158 - mae: 7.4426e-06 - val_loss: 0.0146 - val_mae: 6.1020e-06\n",
      "Epoch 15/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0143 - mae: 7.5226e-06 - val_loss: 0.0132 - val_mae: 6.1244e-06\n",
      "Epoch 16/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0128 - mae: 7.4186e-06 - val_loss: 0.0119 - val_mae: 7.0706e-06\n",
      "Epoch 17/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0116 - mae: 7.2520e-06 - val_loss: 0.0107 - val_mae: 6.1542e-06\n",
      "Epoch 18/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0104 - mae: 7.3102e-06 - val_loss: 0.0096 - val_mae: 7.0855e-06\n",
      "Epoch 19/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0093 - mae: 7.1799e-06 - val_loss: 0.0086 - val_mae: 6.1095e-06\n",
      "Epoch 20/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0084 - mae: 7.1075e-06 - val_loss: 0.0077 - val_mae: 7.1228e-06\n",
      "Epoch 21/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0075 - mae: 7.0745e-06 - val_loss: 0.0069 - val_mae: 6.0499e-06\n",
      "Epoch 22/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0068 - mae: 6.9500e-06 - val_loss: 0.0062 - val_mae: 7.0781e-06\n",
      "Epoch 23/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0061 - mae: 7.8716e-06 - val_loss: 0.0056 - val_mae: 6.1393e-06\n",
      "Epoch 24/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0054 - mae: 7.4078e-06 - val_loss: 0.0050 - val_mae: 6.0722e-06\n",
      "Epoch 25/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0048 - mae: 7.7640e-06 - val_loss: 0.0044 - val_mae: 7.1377e-06\n",
      "Epoch 26/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0043 - mae: 7.4776e-06 - val_loss: 0.0040 - val_mae: 7.2122e-06\n",
      "Epoch 27/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0039 - mae: 7.6674e-06 - val_loss: 0.0035 - val_mae: 7.0557e-06\n",
      "Epoch 28/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0034 - mae: 7.6066e-06 - val_loss: 0.0031 - val_mae: 6.2212e-06\n",
      "Epoch 29/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0030 - mae: 7.2450e-06 - val_loss: 0.0028 - val_mae: 6.2063e-06\n",
      "Epoch 30/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0027 - mae: 7.5294e-06 - val_loss: 0.0025 - val_mae: 6.1169e-06\n",
      "Epoch 31/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0024 - mae: 7.3761e-06 - val_loss: 0.0022 - val_mae: 6.1989e-06\n",
      "Epoch 32/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0021 - mae: 7.8627e-06 - val_loss: 0.0019 - val_mae: 6.1542e-06\n",
      "Epoch 33/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0019 - mae: 7.5573e-06 - val_loss: 0.0017 - val_mae: 6.1393e-06\n",
      "Epoch 34/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0016 - mae: 7.5500e-06 - val_loss: 0.0015 - val_mae: 6.1542e-06\n",
      "Epoch 35/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 0.0014 - mae: 7.8337e-06 - val_loss: 0.0013 - val_mae: 6.2659e-06\n",
      "Epoch 36/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0012 - mae: 7.1213e-06 - val_loss: 0.0011 - val_mae: 6.9588e-06\n",
      "Epoch 37/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 0.0011 - mae: 7.4148e-06 - val_loss: 9.7166e-04 - val_mae: 6.3106e-06\n",
      "Epoch 38/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 9.3807e-04 - mae: 7.0576e-06 - val_loss: 8.4128e-04 - val_mae: 1.6689e-05\n",
      "Epoch 39/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 8.1156e-04 - mae: 8.3357e-06 - val_loss: 7.2602e-04 - val_mae: 6.1914e-06\n",
      "Epoch 40/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 6.9981e-04 - mae: 7.2666e-06 - val_loss: 6.2445e-04 - val_mae: 6.1244e-06\n",
      "Epoch 41/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 6.0140e-04 - mae: 7.4505e-06 - val_loss: 5.3522e-04 - val_mae: 6.2510e-06\n",
      "Epoch 42/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 5.1502e-04 - mae: 7.6516e-06 - val_loss: 4.5710e-04 - val_mae: 7.0781e-06\n",
      "Epoch 43/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 4.3946e-04 - mae: 7.7483e-06 - val_loss: 3.8894e-04 - val_mae: 1.6659e-05\n",
      "Epoch 44/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 3.7359e-04 - mae: 8.1532e-06 - val_loss: 3.2968e-04 - val_mae: 7.7561e-06\n",
      "Epoch 45/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 3.1638e-04 - mae: 7.6622e-06 - val_loss: 2.7837e-04 - val_mae: 6.2808e-06\n",
      "Epoch 46/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 2.6688e-04 - mae: 8.1360e-06 - val_loss: 2.3409e-04 - val_mae: 1.6682e-05\n",
      "Epoch 47/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 2.2421e-04 - mae: 8.2654e-06 - val_loss: 1.9605e-04 - val_mae: 6.2659e-06\n",
      "Epoch 48/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 1.8759e-04 - mae: 7.3741e-06 - val_loss: 1.6350e-04 - val_mae: 6.9439e-06\n",
      "Epoch 49/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 1.5628e-04 - mae: 8.1182e-06 - val_loss: 1.3576e-04 - val_mae: 7.1079e-06\n",
      "Epoch 50/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 1.2963e-04 - mae: 7.4631e-06 - val_loss: 1.1223e-04 - val_mae: 6.4000e-06\n",
      "Epoch 51/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 1.0704e-04 - mae: 7.6869e-06 - val_loss: 9.2358e-05 - val_mae: 7.8902e-06\n",
      "Epoch 52/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 8.7992e-05 - mae: 7.8852e-06 - val_loss: 7.5651e-05 - val_mae: 7.0930e-06\n",
      "Epoch 53/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 7.1993e-05 - mae: 7.5070e-06 - val_loss: 6.1672e-05 - val_mae: 6.2585e-06\n",
      "Epoch 54/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 5.8622e-05 - mae: 7.2526e-06 - val_loss: 5.0032e-05 - val_mae: 6.3851e-06\n",
      "Epoch 55/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 4.7502e-05 - mae: 7.2579e-06 - val_loss: 4.0387e-05 - val_mae: 8.9034e-06\n",
      "Epoch 56/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 3.8299e-05 - mae: 8.5784e-06 - val_loss: 3.2437e-05 - val_mae: 7.5996e-06\n",
      "Epoch 57/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 3.0721e-05 - mae: 8.2191e-06 - val_loss: 2.5916e-05 - val_mae: 8.1882e-06\n",
      "Epoch 58/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 2.4515e-05 - mae: 8.2347e-06 - val_loss: 2.0597e-05 - val_mae: 8.2925e-06\n",
      "Epoch 59/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 1.9458e-05 - mae: 8.4270e-06 - val_loss: 1.6281e-05 - val_mae: 1.6697e-05\n",
      "Epoch 60/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 1.5360e-05 - mae: 8.5670e-06 - val_loss: 1.2798e-05 - val_mae: 7.0482e-06\n",
      "Epoch 61/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 1.2058e-05 - mae: 7.3263e-06 - val_loss: 1.0003e-05 - val_mae: 7.0706e-06\n",
      "Epoch 62/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 9.4124e-06 - mae: 7.7908e-06 - val_loss: 7.7738e-06 - val_mae: 6.2361e-06\n",
      "Epoch 63/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 7.3046e-06 - mae: 7.4262e-06 - val_loss: 6.0058e-06 - val_mae: 6.1393e-06\n",
      "Epoch 64/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 5.6353e-06 - mae: 7.4800e-06 - val_loss: 4.6120e-06 - val_mae: 6.2212e-06\n",
      "Epoch 65/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 4.3213e-06 - mae: 8.4655e-06 - val_loss: 3.5200e-06 - val_mae: 8.0466e-06\n",
      "Epoch 66/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 3.2932e-06 - mae: 8.3476e-06 - val_loss: 2.6697e-06 - val_mae: 7.2420e-06\n",
      "Epoch 67/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 2.4939e-06 - mae: 7.6790e-06 - val_loss: 2.0118e-06 - val_mae: 8.9854e-06\n",
      "Epoch 68/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 1.8765e-06 - mae: 8.6232e-06 - val_loss: 1.5062e-06 - val_mae: 8.0243e-06\n",
      "Epoch 69/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 1.4027e-06 - mae: 8.0634e-06 - val_loss: 1.1200e-06 - val_mae: 6.1989e-06\n",
      "Epoch 70/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 1.0415e-06 - mae: 7.3410e-06 - val_loss: 8.2733e-07 - val_mae: 7.0781e-06\n",
      "Epoch 71/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 7.6802e-07 - mae: 8.2741e-06 - val_loss: 6.0678e-07 - val_mae: 6.6981e-06\n",
      "Epoch 72/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 5.6241e-07 - mae: 8.6497e-06 - val_loss: 4.4190e-07 - val_mae: 7.0557e-06\n",
      "Epoch 73/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 4.0891e-07 - mae: 8.2300e-06 - val_loss: 3.1955e-07 - val_mae: 7.4506e-06\n",
      "Epoch 74/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 2.9512e-07 - mae: 7.2420e-06 - val_loss: 2.2961e-07 - val_mae: 1.6905e-05\n",
      "Epoch 75/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 2.1145e-07 - mae: 8.0341e-06 - val_loss: 1.6336e-07 - val_mae: 7.2271e-06\n",
      "Epoch 76/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 1.5039e-07 - mae: 9.7414e-06 - val_loss: 1.1575e-07 - val_mae: 1.6905e-05\n",
      "Epoch 77/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 1.0610e-07 - mae: 8.3664e-06 - val_loss: 8.1268e-08 - val_mae: 1.7047e-05\n",
      "Epoch 78/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 7.4283e-08 - mae: 8.3035e-06 - val_loss: 5.6636e-08 - val_mae: 1.6972e-05\n",
      "Epoch 79/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 5.1588e-08 - mae: 8.0166e-06 - val_loss: 3.8839e-08 - val_mae: 6.3553e-06\n",
      "Epoch 80/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 3.5533e-08 - mae: 7.6272e-06 - val_loss: 2.6578e-08 - val_mae: 6.3032e-06\n",
      "Epoch 81/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 2.4281e-08 - mae: 7.4557e-06 - val_loss: 1.8041e-08 - val_mae: 6.5118e-06\n",
      "Epoch 82/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 1.6478e-08 - mae: 7.8436e-06 - val_loss: 1.2211e-08 - val_mae: 8.3894e-06\n",
      "Epoch 83/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 1.1102e-08 - mae: 8.2562e-06 - val_loss: 8.1760e-09 - val_mae: 8.2329e-06\n",
      "Epoch 84/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 7.3983e-09 - mae: 7.3826e-06 - val_loss: 5.4359e-09 - val_mae: 7.4580e-06\n",
      "Epoch 85/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 4.9215e-09 - mae: 7.7752e-06 - val_loss: 3.5970e-09 - val_mae: 9.1940e-06\n",
      "Epoch 86/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 3.2532e-09 - mae: 8.1070e-06 - val_loss: 2.3739e-09 - val_mae: 7.2345e-06\n",
      "Epoch 87/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 2.1626e-09 - mae: 8.2831e-06 - val_loss: 1.5083e-09 - val_mae: 6.5863e-06\n",
      "Epoch 88/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 1.4374e-09 - mae: 8.5264e-06 - val_loss: 1.0467e-09 - val_mae: 8.0615e-06\n",
      "Epoch 89/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 9.5932e-10 - mae: 8.8236e-06 - val_loss: 6.3728e-10 - val_mae: 6.0424e-06\n",
      "Epoch 90/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 6.5011e-10 - mae: 8.2344e-06 - val_loss: 4.2696e-10 - val_mae: 6.6161e-06\n",
      "Epoch 91/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 4.4590e-10 - mae: 8.0022e-06 - val_loss: 5.9471e-10 - val_mae: 1.6958e-05\n",
      "Epoch 92/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 3.7057e-10 - mae: 9.8775e-06 - val_loss: 5.5222e-10 - val_mae: 1.7434e-05\n",
      "Epoch 93/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 2.7455e-10 - mae: 8.9529e-06 - val_loss: 1.5895e-10 - val_mae: 6.6161e-06\n",
      "Epoch 94/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 2.1181e-10 - mae: 8.5061e-06 - val_loss: 4.7235e-10 - val_mae: 1.7762e-05\n",
      "Epoch 95/200\n",
      "43/43 [==============================] - 0s 7ms/step - loss: 2.0211e-10 - mae: 9.0255e-06 - val_loss: 1.6229e-10 - val_mae: 7.6070e-06\n",
      "Epoch 96/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 1.7965e-10 - mae: 8.7955e-06 - val_loss: 9.7186e-11 - val_mae: 6.9141e-06\n",
      "Epoch 97/200\n",
      "43/43 [==============================] - 0s 8ms/step - loss: 1.6478e-10 - mae: 8.9067e-06 - val_loss: 1.7519e-10 - val_mae: 9.2909e-06\n",
      "========== Second Autoencoder Finish ==========\n"
     ]
    }
   ],
   "metadata": {
    "tags": []
   }
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "source": [
    "# plot the training losses\r\n",
    "fig, ax = plt.subplots(figsize=(14, 6), dpi=80)\r\n",
    "ax.plot(history1.history['loss'], 'b', label='1st autoencoder', linewidth=2)\r\n",
    "ax.plot(history2.history['loss'], 'r', label='2nd autoencoder', linewidth=2)\r\n",
    "ax.set_title('Model loss', fontsize=16)\r\n",
    "ax.set_ylabel('Loss (mae)')\r\n",
    "ax.set_xlabel('Epoch')\r\n",
    "ax.legend(loc='upper right')\r\n",
    "plt.show()"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1120x480 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "source": [
    "# plot the loss distribution of the training set\r\n",
    "X_pred = autoencoder2.predict(autoencoder1.predict(X_train))\r\n",
    "X_pred = pd.DataFrame(X_pred, columns=train.columns)\r\n",
    "X_pred.index = train.index\r\n",
    "\r\n",
    "scored = pd.DataFrame(index=train.index)\r\n",
    "scored['Loss_mae'] = np.mean(np.abs(X_pred-X_train), axis = 1)\r\n",
    "plt.figure(figsize=(16,9), dpi=80)\r\n",
    "plt.title('Loss Distribution', fontsize=16)\r\n",
    "sns.distplot(scored['Loss_mae'], bins = 20, kde= True, color = 'blue');\r\n",
    "plt.xlim([0.0,.5])"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "C:\\Users\\user\\anaconda3\\envs\\py38tf2\\lib\\site-packages\\seaborn\\distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
      "  warnings.warn(msg, FutureWarning)\n"
     ]
    },
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "(0.0, 0.5)"
      ]
     },
     "metadata": {},
     "execution_count": 45
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1280x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "source": [
    "# calculate the loss on the test set\r\n",
    "X_pred = autoencoder2.predict(autoencoder1.predict(X_test))\r\n",
    "X_pred = pd.DataFrame(X_pred, columns=test.columns)\r\n",
    "X_pred.index = test.index\r\n",
    "\r\n",
    "scored = pd.DataFrame(index=test.index)\r\n",
    "Xtest = X_test\r\n",
    "scored['Loss_mae'] = np.mean(np.abs(X_pred-X_test), axis = 1)\r\n",
    "scored['Threshold'] = 0.375\r\n",
    "scored['Anomaly'] = scored['Loss_mae'] > scored['Threshold']\r\n",
    "scored.head()"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "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>Loss_mae</th>\n",
       "      <th>Threshold</th>\n",
       "      <th>Anomaly</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2004-02-15 12:52:39</th>\n",
       "      <td>0.164771</td>\n",
       "      <td>0.375</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-02-15 13:02:39</th>\n",
       "      <td>0.203675</td>\n",
       "      <td>0.375</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-02-15 13:12:39</th>\n",
       "      <td>0.162726</td>\n",
       "      <td>0.375</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-02-15 13:22:39</th>\n",
       "      <td>0.181921</td>\n",
       "      <td>0.375</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-02-15 13:32:39</th>\n",
       "      <td>0.091424</td>\n",
       "      <td>0.375</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     Loss_mae  Threshold  Anomaly\n",
       "2004-02-15 12:52:39  0.164771      0.375    False\n",
       "2004-02-15 13:02:39  0.203675      0.375    False\n",
       "2004-02-15 13:12:39  0.162726      0.375    False\n",
       "2004-02-15 13:22:39  0.181921      0.375    False\n",
       "2004-02-15 13:32:39  0.091424      0.375    False"
      ]
     },
     "metadata": {},
     "execution_count": 46
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "source": [
    "# calculate the same metrics for the training set \r\n",
    "# and merge all data in a single dataframe for plotting\r\n",
    "X_pred_train = autoencoder2.predict(autoencoder1.predict(X_train))\r\n",
    "X_pred_train = pd.DataFrame(X_pred_train, columns=train.columns)\r\n",
    "X_pred_train.index = train.index\r\n",
    "\r\n",
    "scored_train = pd.DataFrame(index=train.index)\r\n",
    "scored_train['Loss_mae'] = np.mean(np.abs(X_pred_train-X_train), axis = 1)\r\n",
    "scored_train['Threshold'] = 0.375\r\n",
    "scored_train['Anomaly'] = scored_train['Loss_mae'] > scored_train['Threshold']\r\n",
    "scored = pd.concat([scored_train, scored])"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "source": [
    "# plot bearing failure time plot\r\n",
    "scored.plot(logy=True,  figsize=(16,9), ylim=[1e-2,1e2], color=['blue','red'])"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "metadata": {},
     "execution_count": 48
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA6sAAAH9CAYAAAD8qupGAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAAC3tUlEQVR4nOzdd3gc1dUG8Hdm+6q5SXKv2IsxmGIDBmN6AFMTIPSQ0JNQ8tFDJ6EFQgmEBAiYAAaHHmpoxlTTXMA2xcK4d8myZattn++Pq7tzZ3Z2tasu+f09jx9Ju7MzszOz63vmnHuvZhgGiIiIiIiIiLoSvbN3gIiIiIiIiMiOwSoRERERERF1OQxWiYiIiIiIqMthsEpERERERERdDoNVIiIiIiIi6nIYrBIREREREVGXw2CViIiIiIiIuhx3R2wkFAqNA/BHAFsBbKioqLi1I7ZLRERERERE3VNHZVb7AbiioqLiIgD7dtA2iYiIiIiIqJvSDMPosI2FQqGrAayvqKh4qsM2SkRERERERN1OR5UB+wHcC+DVioqKdzpim0RERERERNR9dVQZ8E0AJgD4VSgUeqyDtklERERERETdVKvKgEOhUDGAzwAcXVFRsaLpsdMAXA/AC+C+ioqKf7TBfhIREREREdF2pMVlwKFQaG8AjwIYozw2CMBtEFnUCIDPQqHQBxUVFd+3cDM+AHsCWA8g0dJ9JSIiIiIioi7HBWAAgDkQ8aNFa/qsngfgQgDTlccOBTCroqJiMwCEQqEXAZwI4M8t3MaeAD5pxT4SERERERFR1zYFwKf2B1scrFZUVJwLAKFQSH14IEQWVFoPYK+WbkOua8uWeiSTHTdqMeWnb99CVFfXdfZuUDvgue3ZeH57Lp7bnovntufiue25eG4z03UNvXsXANYYMqWtRwPWHB5LtmJ9CQBIJg0Gq10cz0/PxXPbs/H89lw8tz0Xz23PxXPbc/HcNsuxy2dbjwa8FkB/5e8BANa18TaIiIiIiIioh2vrzOpMADeHQqFSAPUATgBwfhtvg4iIiIiIiHq4Ng1WKyoq1oZCoesAfAAxdc1jFRUVX7XlNoiIiIiIiNpSIhHHli1ViMejbb7uykodyWRrekb2DG63F717l8Llyj0EbXWwWlFRMdz29wwAM1q7XiIiIiIioo6wZUsV/P4gCgr6Q9OchuFpObdbRzy+fQerhmGgvn4btmypQr9+A3J+XVv3WSUiIiIiIupW4vEoCgqK2zxQJUHTNBQUFOeduWawSkRERERE2z0Gqu2rJceXwSoRERERERF1OQxWiYiIiIiIqMthsEpERERERNSFzJ8/FxddxBlAGawSERERERFRl9Om86wSERERERF1Z88958Z//uNps/VpmgbDMAAAp54aw8knx1u8rqeeehzvvvsWdF3HnntOwu9/fwnC4UbcfPN1qK6uBgCcffZ52G+/A/Dss0/jrbfehK5rGDt2HK666rqM650/fy6eeupxGAawbt0aHHjgISgoKMAnn3wEwzBw9933o0+fvnjppefw9tv/QzjcCF3X8ac/3YHhw0fghx++wwMP3ItIJIySkl648sprMXDgoBa/T4mZVSIiIiIioi7u888/xaeffoxp06bj8cefwdq1q/HKKy/h448/RP/+A/H440/jxhtvwYIF3yAej+Ppp5/AtGnTMW3a09B1HVVVlVnX//333+Haa2/E9OnP45VXXkSvXr0xbdp07LDDaMyc+S7q6+vw8ccf4cEHH8H06c9jypQD8d//voBYLIa//OVW3HTTbXj88Wdwyiln4M47b2uT98zMKhERERERUZOTT463Kvtp53briMeTrV7PvHlzceihh8Pn8wMAjjrqWLz11pv43e8uxiOP/AObNlVin332w29+cw7cbjd23nk8zj33TEyZcgCOP/6XKC0ty7r+kSNHoby8PwCgpKQXJk7cCwBQXt4ftbXbUFBQiJtvvhUzZ76L1atX4csvP8Po0SGsXr0S69atwR//eFlqXfX19a1+vwCDVSIiIiIioi7PMJK2v4FEIo4hQ4ZixowX8cUXn2P27I/x7LNP45lnXsQdd9yD775bhC+++AyXX34JbrzxFuy++4SM63e7raGhy+Wy/L1x4wZcfPEFOOGEkzBp0r7o06cvliypQCKRxMCBg/DEEzMAAIlEAlu2bG6T98wyYCIiIiIioi5ujz32xMyZ7yASCSMej+N//3sNe+wxES+99BymTXsEBx98KC6//I/YsmULampqcPrpJ2LkyB1w7rm/xZ577o2lS5e0avuLF3+PwYOH4OSTT8dOO+2ML774DMlkAsOGDce2bduwYMHXAIA333wNN9+cuX9sPphZJSIiIiIi6mIWLvwGP/vZlNTfhx02Ffvuux/OOedMJBJx7L33PjjhhJMRiYRx883X4cwzT4bb7cbZZ5+P3r1747jjjsd5550Jn8+P8vL+OPLIY1q1P3vuOQn//e+LOOOMX8Lj8WCnnXbGsmVL4fV6ccstf8H999+NaDSKYLAA11//p9a+fQCAJkem6qKGA1heXV2HZLJL7+d2rbS0CFVVtZ29G9QOeG57Np7fnovntufiue25eG4714YNK9G//7B2WXdb9VntCezHWdc19O1bCAAjAKywL8/MKhERERERUQ+3YMHXuO++vzo+d/fd96Nfv9IO3qPmMVglIiIiIiLq4XbddffUIEjdBQdYIiIiIiIioi6HwSoRERERERF1OQxWiYiIiIiIqMthn1UiIiIiIqIu4p577sSiRQsQj8ewZs1qDB8+EgBQW7sNRx55DM4554I229b69etw8cUX4MUXX8/5NSeeeAz+/vdHMGDAQMvjF110Ps4++3zsscfENts/BqtERERERERdxOWXXw3ADCTloEjTpj3SmbvVKRisEhERERERdQM//PAdfvvbs1FVVZnKsv7vf6/jrbfewNatNZg8eX/88pen4K9/vR0bN26Eruu44IILseeee2Pu3K/wz38+AE3TUFRUhJtvvh0AEIlEcNNN12DZsqUoKirGHXfcjZKSXpg9+xM8+uhDMIwkBg4chCuvvBZ9+vRN7Us0GsWdd96CxYt/QP/+A7F1a02bv18Gq0RERERERE18z82A/z9Pt9n6NA0wDPF7+NQzEDn5tBava/PmzXj44cfR0NCAE088GqeeegYAoKqqEk8//QLcbjduuukaHHXUsdhvvwOwadMm/P735+CJJ2bgySen4corr8HYsePwwgvP4scfF2PIkKGoqdmCk08+HTvttDOuv/4qzJz5Lg4++FD89a+346GHpmHAgIGYMeMp3HvvXbj11jtT+/Lii88BAJ555kWsXr0Kv/71qS0/SBkwWCUiIiIiIuoGJk3aF16vF16vFyUlvbBt2zYAwJgxO8LtFqHd3LlfYeXKlXjsMVE2HI/HsXbtGuy33/649torMWXKAZgy5QDsueckrF+/Dv36lWKnnXYGAIwYMQpbt9bg+++/w9ix41L9Uo899nhMn/6EZV+++WYejj32eADAkCFDscsu49v8/TJYJSIiIiIiahI5+bRWZT/t3G4d8XiyTdblcrlSv2uaBqMpZevz+VKPJxJJPPDAQyguLgEAbNpUhd69+2D06BAmT94fn332Cf75zwdw4IHf4bDDplrWCQCGYcAwkmmPJRIJ295oluXs62kLnLqGiIiIiIioh5gwYSJefvkFAMDy5cvw61+fgkgkjPPO+zUaGupx0kmn4aSTTsOPPy7OuI6ddtoZ33+/COvXrwMAvPbay9hjjwmWZSZO3AvvvfcOkskkNmxYj0WLFrb5e2FmlYiIiIiIqIe49NKrcNddt+HXvz4FhmHg+uv/jGCwABdccCFuu+1PcLlc8Pl8uPLKazKuo0+fvrjyyutw7bVXIBaLo3///vjjH2+0LHP88b/E8uVLcfrpJ6J//wEYOXJUm78XTaaOu6jhAJZXV9chmezS+7ldKy0tQlVVbWfvBrUDntuejee35+K57bl4bnsuntvOtWHDSvTvP6xd1t2WZcDdnf0467qGvn0LAWAEgBX25VkGTERERERERF0Og1UiIiIiIiLqchisEhERERERUZfDYJWIiIiIiLZ7XXwsn26vJceXwSoREREREW3X3G4v6uu3MWBtJ4ZhoL5+G9xub16v49Q1RERERES0XevduxRbtlShrq6mzdet6zqSSY4G7HZ70bt3aX6vaad9ISIiIiIi6hZcLjf69RvQLuvmtEQtxzJgIiIiIiIi6nIYrBIREREREVGXw2CViIiIiIiIuhwGq0RERERERNTlMFglIiIiIiKiLofBKhEREREREXU5DFaJiIiIiIioy2GwSkRERERERF0Og1UiIiIiIiLqchisEhERERERUZfDYJWIiIiIiIi6HAarRERERERE1OUwWCUiIiIiIuoh1q/XOnsX2gyDVSIiIiIioh5gwQIdu+5aiB9+6BlhXs94F0RERERERNu5tWtFeLduXc/IrjJYJSIiIiIi6gHq6+VPBqtERERERETURcggta6uk3ekjTBYJSIiIiIi6gFkkFpXx8wqERERERERdREys8oyYCIiIiIiIupUkQhQUyN+ZxkwERERERERdQlnnx3AmDFFAMwBllgGTERERERERJ3qvffcAIDNm1kGTERERERERF3M0qU6y4CJiIiIiIioaxHBqvidZcBERERERETUacJh8/dly/RUkPr11y68/76rk/aq7TBYJSIiIiIi6obWrzczqF984cKmTeLv2loNp54a7KzdajPuzt4BIiIiIiIiyt+6dSL3uM8+cXz+ec8L7ZhZJSIiIiIi6obWrhWZ1HvvDWPffeOdvDdtj8EqERERERFRN1JTI/qrbt0qgtXevQ3sskuyc3eqHTBYJSIiIiIi6kbGjCnCaacF0NgogtVgEBgzhsEqERERERERdRI5qNKnn7rR0ABomgGfDwiFEgAAXTdSyyYSnbKLbYbBKhERERERUTcxf745JU19vYZgENA0YM89k7jjjjAWLqzH9ddHAACxWGftZdvoeUNGERERERER9VBff23mG1et0hAIiEyqpgHnnCOiU49HPBaLAX5/x+9jW2FmlYiIiIiIqJv49lszs7pwoQtBh+lUPR7xMxrV0p/sRhisEhERERERdRMbN2rYeWfRGXXtWh0FBUbaMjJY7e5lwAxWiYiIiIiIuonKSs0yTU0gkL6M12uWAXdnDFaJiIiIiIi6sI0bNey9dwG++05HdbWGgQOTCAZFQCp/qphZJSIiIiIionb37bc6li/X8d57biSTGkpLDfTqJYPV9OW9XvGTfVaJiIiIiIio3axdK8K2hQvFz7IyAyUlIliVowGrmFklIiIiIiKidrdunciQypGAy8qSKC7OnFmVU9dEox2zf+2FwSoREREREVEXJjOrK1aomVXxXPbMKsuAiYiIiIiIqJ3IzKpUWmoomdX0YFX2WWUZMBEREREREbUbmVkFgL59kygoAIqKmi8DZrBKRERERERE7SIctmZWDzggAcDMnvr9mTOr7LNKRERERERE7eK663wIhzWUlycBABMnimBVZk+TyfR+qW63+Mk+q0RERERERNQuPvjAjWOPjeE//2nEPvvEceKJorY32/Q0Xm/PGA3Y3dk7QERERERERM62bdNQXm5g552TePXVxtTj2YJVzrNKRERERERE7cYwgNpaczAlVUmJeEz2T1WZowGzDJiIiIiIiIjawObNwC67FOC119yorwcMQ0NhYXqweuaZMVxxRQS//W16ra/MrHb3MmAGq0RERERERF3ETz/p2LhRx7nnBrBpk8iMFhenL+fzAVddFXWcukb2WWUZMBEREREREbWJzZvN0t2vvnIBcC4DzkaOBhyNsgyYiIiIiIiI2oAarC5bJsK14uL8glWzz2qb7VanYLBKRERERETURWzaZIZoy5eL3wsL81uHywVomsFglYiIiIiIiNqGU2Y13zJgTRPZVQarRERERERE1CaqqzUMHJhEMGi0uAwYECMCs88qERERERERtYnNmzX07WugXz8DtbUi2Mw3swqIYJWZVSIiIiIiImoT1dUa+vQxUFpqBqj59lkFAI+HfVaJiIiIiIiojVRXi8xqaWkSAFBQYMDlyn89os9q9y4Ddnf2DhAREREREZEgg9WCAvF3QUH+JcCA7LPahjvWCRisEhERERERdbL333fhp5901NVp6NXLgN5UA9unT8uCVa/XQCTShjvYCRisEhERERERdaBPP3Xhhx90nHee2an0mWc8mDVLhGfFxQZOPz2GYcOSGDky2aJtbNigo6LCheefj+Okk+Jtst8djcEqERERERFRBzr++CAAWILVbds0NDSIPqbFxQYKC4ETT2x5kCkzsxMnJlq+o52MAywRERERERF1kIQSO553nh9vvinyh9u2mYMhtWT0X7uXX27Axx/XY+TIlpURdwUdGqyGQqGRoVBobkduk4iIiIiIqD00NgJ1dfm95ocfzBDs1Vc9OOusAABrsFpc3PoAc9y4JHbcsWUlxF1FhwWroVCoF4ALAOR5OomIiIiIiLqegw4qwMiRRXm9Zv5853lotm0zf2+LYLUn6LBgtaKioqaiouJqMFglIiIiIqIeYNmy/MOpNWuc5z5VM6tFRQxWAfZZJSIiIiIiahUjj9hy48b0ECwcBqJRNVhti73q/hisEhERERER5UkNUPPpt7pxo4YRI8y+pD6fga1brdlWZlaFVgeroVCoOBQKfRsKhYYrj50WCoW+D4VCP4VCoQvV5SsqKo5u7TaJiIiIiIg605Yt5u+bNzuX9tbWAi+8YH1s40YNo0ebwWokouHmm32pv91uA4FAm+5qt9WqeVZDodDeAB4FMEZ5bBCA2wBMABAB8FkoFPqgoqLi+5Zup2/fNhi7mdpVaSlrFXoqntuejee35+K57bl4bnsuntvuZcMG9a9ClJamL/PHPwKPPw7MnVuECRPEY1VVwOTJ1uVeesmT+r2kRENZGa8FoJXBKoDzAFwIYLry2KEAZlVUVGwGgFAo9CKAEwH8uaUbqa6uQzLJVHhXVVpahKqq2s7eDWoHPLc9G89vz8Vz23Px3PZcPLfdz/ffuwAEAQBLlzZg2LCE5XnDAH78MQDAjVmzwhg6NIZYDKiqKkRxcRSAL22dAFBQkERVVX377nwXoeta1sRkq8qAKyoqzq2oqPjE9vBAAOuVv9cDGNya7RAREREREXUlGzeapb/2MuBwGCgvL8Knn4rc4Jw5YrqaqioNhqGhvNzAzJn1uPbaSNp6OW2NqT0GWHIq2O7es9ESEREREREp1q0zQ6nqag1vveVGpCn2XLvWGhJ99ZUIVmWA279/EuPHJzF5ctyynNttMFhVtEewuhZAf+XvAQDWtcN2iIiIiIiIOsW33+oYOjQJTTNwww1+/PrXAbzxhsikqoGs2w2sXKmjvl5kVgGgtFQEpAUF1nXusEMSZWUMVqXW9ll1MhPAzaFQqBRAPYATAJzfDtshIiIiIiLqFF9/7cKkSQmsWmUOjiTLgdXM6jXXALfcAixbpqO6Wjzer58ISMeOTeIvfwnjgAPi+P57F3bdNQGfc1fW7VKbZ1YrKirWArgOwAcAvgEwo6Ki4qu23g4REREREVFn2LhRw7p1OnbfPYE+fcwej5WVIhiVmdVVq2px0kniuSVLdFRVicf79hXBqqYBZ58dw6hRBo45Jo6hQw2UlzOzKrVJZrWiomK47e8ZAGa0xbqJiIiIiIi6ki+/FH1Qd9stiXfeaUBhIXDQQUFUVopgdN06Df36JeH3A6WlgK4bWLJER0ODhkDASCv/JWftUQZMRERERETUYz39tAcDByYxYUIC7qaIqrzcsGRWBw4UGVK/Hxg61MBPP+nw+cysKjWvPQZYIiIiIiIi6pHWrtXw4YdunHFGLBWoAkBZmQhW//1vD2bOdGPgQLM8eMyYJBYv1rFpk5bqr0rNY7BKRERERESUIzkNzWGHWaedKStLYuNGDfff7wUAHHOM+fzOOyfw00861q3TmFnNA4NVIiIiIiKiHM2f74Lfb2Ds2KTlcZFZ1bFunY577w3jl780g9Vx45JIJDQsXuxisJoH9lklIiIiIiJqxhdfuLBqlYa5c8UUMx6P9Xl1ftSDDrJmXXfeOZH6ncFq7hisEhERERERNVm4UMcVV/hRXm7gyScboTfVot53nxdffeVCPC6mm7GbPDmBvfeOY/LkBAYNsgakw4YZ8PsNhMPss5oPBqtERERERERNXn/djW++Ef1Sp0wJ4v77w5g4MYlvv9VRXy9G+50wIZH2uh13TOL11xsd16nrwPPPN+Kxxzw49NC44zKUjsEqERERERFRk4oKHaNGJeHxGFi82IXbb/fhoYfCqKoyh/vZY4/0YLU5kyYlMGlS/q/bnnGAJSIiIiIioiYVFS7stFMCr73WgLPOiuKzz1z48ENX6vmysmRamS+1DwarRERERES0Xdq82fp3YyOwYoWGUCiJXr2A3/wmhmRSw333+QAAu+2WwMEHJ6BpHb+v2yMGq0REREREtN1ZuFDHjjsW4cUXRc9IwwBuucUHwxDBKiD6ofbrl8SyZaI0+M03G3DvveHO3O3tCoNVIiIiIiLa7ixaJEp7H3zQCwBYsEDHY495sffecey3n+hbqmnAxIni9732EtPVuDnqT4dhsEpERERERNudlStFLe/ixTo2bwbee88NTTPwxBNhy1yoe+4psqx77cXBkToa7wsQEREREVGPtnkz0KsXUnOmAsBPP4k/kkkNO+5YBEBkUdVAFQCOPDKGV19145BDOOVMR2NmlYiIiIiIeqzNm4EddyzCVVf5LI8vXarj8MPjOOaYGABgypQ4Lr44mvb6UaMMzJzZgP79OQJwR2NmlYiIiIiIeqwZMzwAgKee8uK3v41i1Soda9fqWLZMx0EHxXDFFRFcc00EO+zAYLSrYbBKREREREQ90uOPe3DrrT6EQgmsWqXjn//0YsYMD5JJ0V/10EPjKCwEA9UuimXARERERETUI730kgdjxiTx4ouNOOmkGJ57zgxUL7ggmhr1l7omBqtERERERNTjGAawZImOvfdOoLzcwOmnxxCLiUD13nvDuOWWSCfvITWHwSoREREREfU41dUaamo07LCDmHpm/Phk6rlx45hR7Q4YrBIRERERUY8jp6YZPVoEqboOlJeL38eMSWZ8HXUdHGCJiIiIiIh6nCVLRLAqM6sA8PbbDfj2Wx0FBZ21V5QPBqtERERERNTjLFmiw+83MHiwOdLvoEEGBg1iCXB3wTJgIiIiIiLqcX76SceoUUnojHi6LZ46IiIiIiLqcZYs0VP9Val7YrBKREREREQ9SjgMrFqlWfqrUvfDYJWIiIiIiHqUZct0GIbGzGo3x2CViIiIiIi6tKef9uCxxzw5Lz97tgsAEAoxWO3OOBowERERERF1aZdd5gcAnHturNllIxHgwQe9mDQpjrFjGax2Z8ysEhERERFRl2WYM89g40at2eV//FHH+vU6zjorBq35xakLY7BKRERERERd1oYNZsS5cGHz4cu6dWL5YcOYVe3uGKwSEREREVFGkYjIaB57bADff9+68CESAd56y51ThhQQger11/tSf3/zjavZ16xfL/Zx4ECjmSWpq2OwSkREREREjpJJ4Jhjgthll0J88YUb//ynN+vy8+bpaGzM/Py55wbw618H8Nvf+i3lvXbbtgFTpwYxfnwhXn9dDKzk8xlYvLj58GX9eg0ul4HSUgar3R2DVSIiIiIicvTOO25LNnP+fDN8ePddFyZOLMC334rHNm7UMHVqAS691J9xfTLYnD3bjX33LcDTT4tANBoF7r/fiy1bxHK33ebDvHnWLOrkyQksW5ZLGbCO8nIDruaTsNTFMVglIiIiIqKUWAy4914vNm7U8OKLbvTvn8Szzzbgiisi+OknF1as0LB1K3DGGUGsWqXjjTfEBCMyEH3zzcwTjlRXazjnnChuuSWMXr0MXHaZH/Pn63j1VTduu82HadO8mD9fxxNPeHDeeVFUVtbiu+/q8MYb9dhhhySWL9ezZmQBkVkdMIBZ1Z6AU9cQEREREVHKO++48Ze/+FBTo6GiQsduuyVw8MEJDBhg4O67ffjySxd0JeX1+ecihVlRIR6MRDQkEkjLbEYiQF2dhrIyAxdcEMPpp8cwYUIh7r3Xh/p6scxDD3nx7LMe9O9v4I9/jAAASktFSe+iRUk0NGiorNRQXp45GF2/XuP8qj0EM6tERERE1GKRCFBT4/xcY6Po80hd3+rVGmprgYce8uDsswMAgBdfdGP5ch2jR4uTGAolUVxsYM4cF958040BA5K44IIo5s934d//9uD6683y3xdeSM+Jbd4sBlXq00cEmoWFwAUXRPHuu27Mnu3GHnskUFurYdUqHfffH0ZRkfX1I0aI/XjjDXfG7OpHH7mwapXOwZV6CAarRERERNRi117rw5gxRfjpJ+vorq+95sawYUV45BFPJ+0Z5erNN92YMKEQBx1UgJtuEgHnhAkJbNqkIxbTUsGqrgMTJybw6qsevPuuG8ceG8fhh8cRiWi4+mozUN1ttwQuuSSQNhjSpk3iGunb1wwkzz03ipISA8OGJfHf/zbgv/9twAsvNODAAxNp+zlypNiPa67x45lnnK+r66/3obzcwNlnR1txRKirYLBKRERERC1iGMD06WJ02COPLMCsWS68/LIbK1dquOYaMd3ITTf58bOfBfHUU56sfQ2XLNFx6aU+y5ya1L6iUeDKK3046yyRSV21SscNN0SwalUtXn21AX36iOBwzBgzPX7IIXFs3aohHhd9TydPTmD33c3A8uqrI5gxoxEej5EWUFZXi3Pbr595IRQXA2+80YBXX21AICAGUTrggPRAFQCGDTNw001huFwG/vxns3RYSiaBFSt0HHNMHCNHMrPaE7DPKhERERG1yLJlIvj43e+ieOcdN049NQDDMIPNPn2S2LxZx4IFLixY4MKqVRquv15kvJJJWPo9PvaYB88848WsWW48/XQjQqEkZs5044gj4pblqO3cfLMPTz4pbjZccEEUW7dqOO+8KPxNSdLjj4/jiSc8qcwqAJxzTgyRCGAYGoYPFwHhvfeG8fXXLpx+egxa0+mfOlW8dty4BE45JQ4gvQxYyrV/qaYBF14Yw9ixSZxyShBz5rgsGdgNGzREIhqGDWPteU/Bjz4RERERtcjnn4u8x5lnRnH77WEYhoYJE8zg4fLLRWB68skxHH98DI884kVtLTBtmgfjxxdg5UozsP3gAzfGj09A04BTTw3g+OMD+M1vApg9m/OPtAfDAF55xY2f/zyGjRtrccstETzwQDgVqALAtddG8NprDZa+o7oOXHRRDBdfbJbZjhuXxBlnmIEqANxxRwTjxiVxxx2+VEZdZlbVMuCW2GuvBFwuAw8/7MWiRWY4s3Kl+H34cAarPQUzq0RERETUIjJTNnCggVGjEvj443qMHp3EgAEiujn77BiKiw0cfXQc33+v4+WXPRg1yox87rvPi/vui2DBAh0rVui4444wJkxIYOrUICorRTN1yRIdU6Y4l4VSyyxdquHaa/3YtEnH5MlRS5CpKiwEJk5sWeBXWmrgV7+K4dJL/fjySxfeftuNZ58V57RXr9YFq4WFwIABBmbNcuPLL1144IEw7r7bi8MPFxlcZlZ7DgarRERERNQi0abkmldUkmLHHUWQ8PHH9di8WYPLBZx8sgggJk5MYsyYBDZv1nD++TGsXKnhmWe8eO89N2prNZSXJ3H00XGUlxt4440GLF+u45JL/HjhBQ/69jVw3HHxzniLPdJDD3nxwQciDNhrr/a7EXDQQeKcHXts0PK4fUqblrj88iheecWNRYt0nHOO6HP7ww8u6LqBwYPZX7WnYLBKRERERC0SjQIul5EWfMigVaVpwNtvN8DrFcFtLAZMmJDE3Lk6CguBc86JpubOnDAhiQkTkvjXv5KYN8+Fiy/2w+9vxK67JtG/vwHDQMZsIDWvqso8eO05H+nAgQamTo3hrbc8OO64GI45Jo6GhrZZ9+mni3lab73Viwce8KUeD4WS8HAA6h6DwSoRERERtUg0qsHna345qbDQ/N3jAc44I4Yzzsi8fCAggtdwWMOvfhXEhAkJHHZYHH/7mxdLl9bBzZZs3las0LBwoQtHHBHDX/8aaffBqx55JIx//zuB44+Pp25GtKXTTovh4Ye9uP32CMaNS2CHHVgC3JPwI05ERERELRKNmiXA7WHkyCQ+/9z8e948F+bNE2ncykoNAwey3DMfS5bomDy5AADw61/H2iV4tPP7gd/9LtZu6x850sA339Sjb1+D2fYeiKMBExEREVGLRCKAx9N+Ac8tt0Twxhv1OOWUGPr1s2bM1q1jZJKLujrg9dfduPtubypQBYBdduk5g1b168dAtadiZpWIiIiIWiTfMuB8FRYCe+2VxF57hVFXB/zmNwH88IOOqiod69frePVVHRMnJjBoEDOsmZx+eiA1xRAAXHJJBEccEceECSyXpa6PwSoRERERtUgs1r5lwKrCQuDFFxuxZQsQChXhxx913HmnD717G6ioqOuYnehGamqApUt1S6D6zDMN+NnPek5GlXo+BqtERERE1CKRCOD1dmxWs1cvMfDSF1+IvqtbtmhoaACCweyv296MGWPOZ3vIIXEsWKBj//0ZqFL3wj6rRERERNQi0ajWYZlVSdPElChffmnOl3PRRX7EbdOwbt3asfvVlVRXWztw/utfjfj88/p2Ldkmag8MVomIiIioRURmteO3O3BgEuGwCMjOPDOKN97w4NRTA7j2WhGN3XefF6NHF+GDD1yYP3/7a+5+/bX5nvv3T6KoCCgp6cQdImohlgETERERUYtEo4DP1/GDG40Zk8QnnwCaZuCGGyJ46ikvPvrIjY8/NhCPA088ISLos84KoHdvA/Pn1283o8UuX67ht78NpP4eMYIDKVH3tf3daiIiIiKiNhGLafB4On67Rx4pan4NQ7NkDA1DSwWqANDQoGHtWn27yq7edJMP27Zp8PvFTQQGq9SdbT+fXCIiIiJqU5EIOqUf5D77iIGCgkERkB1xRAwA4HIZ6N8/iZtuCluW/9e/vPjlLwNYvbpnp1dXrdLwzjtunHtuFAsW1KG42MD48QxWqftiGTARERERtUg02vGjAQOA2w3MnFmPoiKx7X/+M4yNGyMoLhYBbDAI/OUvPkQiGlwuA//9r0j/PvSQF7ffHunw/e0oc+a4YBgazjgjht69gTlz6lBc3Nl7RdRyzKwSERERUYtEIh0/GrA0fnwSI0aIYLWwEBg1ykBpqYGCAjFicP/+4rljjjGHCX7sMS/+9KeeOyRuXZ3IHPfpI957796Ay5XtFURdG4NVIiIiImqRWKxzRgPOxYABovz1nHNi6Ns3iT/9KYx+/ZJ49FEPEj10utH6evGzoKDjs91E7YFlwERERETUIp1VBpyLgQPFfo0bl8APP4gorrgYuPRSP9as0TBsWNfc79aQmdVgsJN3hKiNMLNKRERERC0SiWidMsBSLvbYI4Fx4xIoLDQfGzlSZFuXLu2ZTeD6eg3BoMHSX+oxeuYnlYiIiIjaXVfOrJ5/fgwffNBgeUwGq8uX6zC65m63Sl0dS4CpZ2GwSkRERER5MwwZrHb2nuSurMxAQYGBOXNcGDeuAM880wmTxLaj+noNBQWdvRdEbYfBKhERERHlLR4HDKPzRgNuCU0T2dWXX/Zg0yYdl17qx4YNPWfu1fp6DYWFzKxSz8FglYiIiIjyFmmarrSrlgFnctJJMcvfc+b0nA6e9fUsA6aehcEqEREREeUtGhU/u1NmFQDOOiuGPfdM4IYbRLS9ZImOZ5914913u3/QWlenWQaUIuruOHUNEREREeUtFhPls90tWPV6gTffFAMvPfGEB2+/7caiRTomTkzgsMMaO3nvWqe+Hhg6lJlV6jmYWSUiIiKivMkyYJ+v+wZHo0cn8c03LiQSGpYv7/7NYpFZ7b7ng8iu+38qiYiIiKjDddcyYNWYMWIqm6FDk6is1FFX18k71EocDZh6GgarRERERJS3SKR7lgGrfvGLGE45JYarrxZp4hUrum/T2DDEPKvMrFJPwj6rRERERJS3WNOgut1tNGDV7rsnsfvuYSxaJILU5ct17LxzspP3qmXCYSCZZGaVepbue/uIiIiIiDpNT8isSsOHiwC1O/dbrasT54NT11BP0n0/kURERETUaWSfVZ+vc/ejLRQVAaNGJfHpp913+pr6evGTwSr1JAxWiYiIiChvtbU9K5M3dWoMn37qQk2N+Pv773WsXq116j4tXqxj8eLszfUtW4Dnn3fjk09E777S0p5xPogA9lklIiIiohaorBSBXHl5zwiOjjwyjgcf9GHWLDd+/vM4DjxQdP5cu7YWHk/n7NP++4t9qKysdXx+9WoNd97pw/PPmzu4ww7ds88tkRNmVomIiIgob5WVGjTNQN++PSNY3W23JAoLDXzxhQsLF5pN5OnTOydS3bYt+/OvvOLGhAmFeO89M/fk8xkYMqRnnA8igMEqEREREbVAZaWGvn2NTss6tjW3G5gwIYEnnvDisMNERnOHHRJ44AFvqn9uR5o71+w/a9jiT8MA/vlPMbLVli0a3G6xgKYBOlv31IPwciYiIiKivFVWaj2uf+TuuydSvx9/fAy33hrBunU6nnzSjMijUeCRRzzNZj5b6oUX3Kis1PDFF2awWmurAp4zR8c335jPH3tsHAAQDnduH1uitsZglYiIiIjyVlWlo6ysZwWrv/hFHLvtlsDnn9fh4YfDOOigBA48MI7rrvPj2GMDSCSA999344Yb/DjppGBaxrO1Kis1XHhhAL/6VcBS3ltZqSEaBZYs0bF+vYYbbvCjuNjAwIGif+ohh8Thdhu4+eZw2+4QUSfjAEtERERElLfKSg0jR/aswXzGjk3i3XcbUn9rGnD33WGceWYAX3zhxgcfuDBnjsj1zJ/vwo8/6giF2u4YrFsnMqNffy2ypgcfHMesWW5UVup48kk3HnnEiyFDkli9WsdVV0XwzTcurFunY8SIJNatq2uz/SDqKphZJSIiIqK8GIYIVsvLe1aw6mToUAPvvtuAfv2SeOopD+bMcaF3b5FSnTMnv3lZDQOoqNAz9oFdt87aND/zzBgA4Oc/D+KRR0Qf1dWrdZx5ZhSXXx7FqFHJ1D4S9UQMVomIiIgoLzU1QCTS8/qsZuL1isDx7bc9+PJLN046KYY+fZKYOze/pvSbb7oxZUoB9tijADU1QFWVhlNPDeDGG31YutQa/I4cmcSkSXHH9Rx/fByaBpx6agxXXBHZbs4DbX9YBkxEREREeZk9WzQhd9ml52dWpfPPj+Lee30AgLPOimLFCh0zZngRCAC//W0Ua9bo2HffBLSmMY42bNBw9dU+PPBAGCUl4rGPPhLBaGWljjFjilLrfv994OGHAcCbemzKlDh69UrfD6/XSA0EteOOSey4YycMVUzUQZhZJSIiIqK8/O9/bvTpk8SkSYnmF+4h+vQBZs2qxzff1GHkSAO/+U0Ue+6ZwLRpXpx0UhC/+EUQ774rgtHLL/fhuOOCeOstD555xhxJ+PPPXRg9Ov2YXXRRJPW7phm46qoIbrwxAl0HTjghlnruqKNiuOeeMAKBdnyjRF0Ig1UiIiKibm7jRg3HHx9IDdDT3j780IVDDknAvZ3V6O28cxIDB4qS20MOSeDNNxuw++4JLF8umtTvvOPGpk0apk/3ph5raBDnZONGDT/+6MLJJ6eX9l55ZRTz5onfDUPDFVdEUdSUeH3ooTBee00M+nTZZVHH1xP1VAxWiYiIiLqhZBJobBS/P/qoB59+6sb06SKLd+WVPrzxRvtEko2NwKZNOkaP3n5KgLMZPNg8DjNnurFggbV5fffdXkyZEsT48QUAgKlT49hhBzO72rdvEoEAsOuu4u9hw9KP66RJCWzcWLtdlV0TAQxWiYiIiLqlO+/0Yq+9ChAOA9XVInsXCABLl2p48kkvzj67dbWiiQRQX5/++IYNYlv9+zNwApDKtALAhg06nn7aY3k+mdRQUeGCYWiYMiWO0aOTePvtBtx0U9jyepcL+PjjerzxRgOcaB2TNCfqUhisEhEREXUDhjLga10dMG2aFxs36pg5042VK0WTrqpKw5tvimBJTmuSi+XLNTz7rDUTe9VVPowYUYSrr/bh+efN5zZsENvq358j0ALAoEHiOA8bloSmGanjr7rppjAef7wR//ynCFCLi4Hdd09aXg+IAZPKy3lciSQGq0RERERdgGEADzzgRVWVmUK77z4vFizQMWOGG/37F6KhKen2xhtubNumwe838OKLbixaJAb2WbdOw6uvisDS68096Pnb33y45JIAli/XkGyKnaZPFyPT/vvfXvz731589pkLRxwRTE2vMmAAgyoAGDRIHIddd02gb1/xezBoPTYHHpjA0UfHLYGoLB/mcSTKbDvrFk9ERETUNX37rY5bb/Vh9mwXnnuuERs3arjjDh9Wr9Ywf74oI50xAzjuOODDD93o1y+JI4+MY8YMD+JxEeC++64bkYj4vbJS/Kyo0LF6tYZDD00gFgPeftuNo4+Op8pKV63SMGuWCED//W8v/vtfNyZPto5Y+803Oh5+2IP5812YP18GqywDBoCBA5NNPw389a8RzJ7twoQJCfzudwFMmRLHPvsksNNO6cdq4EADQ4cmMWHC9jOiMlG+mFklIiIi6gLCokIUn3/uwrRpHsydK4LC7793YfhwEexMmyYysB9/7ML++ydwwAGJVKA6ZUockYgGn8/AeedFUV2tIxYDpkwpwGmnBWEYwF13eXHOOQF88IFY96JFOiZOLMTGjTpcLgMPPyxKi19+2VrKmkhoePtt62NFRSAAQ4YY0DQDw4cncdRRcdx+ewSDB4ts6X77JXDFFVHH/qZuNzB3bj1OOomj+xJlwmCViIiIqAuoqRERTTis4Zpr/DjrLDFA0uLFeqqf6DffAAsX6ti0SccBB8QxeXIcmmZg7NhEqo/qscfGMWaM+F0OvAQAmzZpeP99UVS3dKlY35tvir/79k3imWcaMWZMAuecE7Xslzo67aWXqvOBttlb79bKygy8/noDTj3VnA91xIgkvF4DO+zA7DNRa7AMmIiIiKgL2LLFOfpraNBSpbfhMPD3v3vhchn42c8S6NMHOPvsGHbdNYFt28Trzz03mgpuZSkwAJx0UgDffSfW8913Olat0vDf/3qw775xvPKKmAPn4IMbYBgiiN2wQcedd4YxaVICug48/LAH558fw4MPei0j4BKw117WoLSszMC8efUoK+NxImoNBqtEREREXYDMrALAOedEUVenIRwGXn1VlN9OmhTHF1+48dprHhx8cBz9+olA6I47RLYzHgcOOyyO4cMNzJsn1rN2rVlE9913LgwdmoSmAQsXunD88UGsW6fh+uvNjCAgMqbXXhvBJZcEcOSR5qBA990ntlNRUcesag44qi9R67EMmIiIiKgL2LzZjAAHDjTw97+HU1OdAKL/o6SWnEpuNzB8uAiQZEZvwQJrU+/55xswdWoc337rwqpVOh56KIxjjknvM3nKKXGsX1/rGHAVFgIFBXm+OSKiFmCwSkRERNQFyMzqL34Rw5lnin6jHo/oTwrA0v/RKcBUDRxowO83MGuWtYhu5EgDv/pVDC6XgdJSMZpwJi5Xi94GEVGbYRkwERERURdQU6NhxIgkHnkkbHl82DAD1dVASYmBuXOBZLIOejPpBpcL2HHHJL75RkSct98exo47imB39OgkFi+uQ2OjBo8n21qIiDpXhwSroVBoCIC7ANQC+KiiouKZjtguERERUXexebOGPn3Sy26vuiqCU04JYqedkhg/Hqiqyq0v5LhxCXzzjQsFBSKb6vOZz5WUiOCXiKgr66gy4PMA3FNRUXE+gF910DaJiIiIuo2aGg29eqUHkAcfnEBlZS0GDMgvuBw2TCy/zz4JS6BKRNRddFSw2h/A2qbfE9kWJCIiIurpNmzQUF9vfWzLFudgtaUOOUT0R/3DH6LNLElE1DV1VLC6GsDADt4mERERUZtLJoHGxpa/3jCAww8P4pZbrOnOzZs19O7ddsHqLrsksWFDLfbem3kCIuqeOipwfAzAZaFQaBqAxztom0RERERt7s47vRg2rAgNDS17/apVGtav1/HJJ+Zwu1u2ALW1GoYMSWZ5Zf6aG4iJiKgra9UAS6FQqBjAZwCOrqioWNH02GkArgfgBXBfRUXFPyoqKtYDOL2V+0pERETU6aZPF0Po1tRoCAbzz4QuWCCC1CVLXKiu1tC3r4Hly0VUOWJE2warRETdWYuD1VAotDeARwGMUR4bBOA2ABMARAB8FgqFPqioqPi+NTvZt29ha15OHaC0tKizd4HaCc9tz8bz23Px3LYfr1f8dLkKUVqa/rxhAC++CBx8MNC3b/rzP/5o/j5zZiEuugiorhZ/T5wYdFyniue25+K57bl4blumNZnV8wBcCGC68tihAGZVVFRsBoBQKPQigBMB/LkV20F1dR2SSQ6v3lWVlhahqqq2s3eD2gHPbc/G89tz8dy2LcMANM382+UqAKBj2bIGlJen9wf99lsdJ51UAABYtqwWhYXAhx+68O9/e3DXXRHMnu3H2LEavF7g4otdCIXq8c03bmiaF4WFdaiqyrwvPLc9F89tz8Vzm5mua1kTky3uyVBRUXFuRUXFJ7aHBwJYr/y9HsDglm6DiIiIqDNNm+bByJGFlpF7XU1dTbds0Rxfs3mz+fiSJToWLtRx0klBvPWWB7NmuTBvngtTpiTw9NNilKYvv3Rh+XIdgwYZ8Pvb7a0QEXU7bd3t3ulbm50viIiIqNtJJIBrrvGjvl7DypVmk8njEdVeNTXi7+++03HQQUFUVopm0LZtZnNo3TodV13lR1GReM3rr3sQDmuYNCmB8nIDZWVJfP21C1995cKoUWwyERGp2jpYXQsxp6o0AMC6Nt4GERERUbv46CMXbrtNdEqdPdscrXfdOjMAdTd1opKZ1ddec+O771z43e/8+POfvdi61Vz23nu9mD/fhTvvDKO0NIkPPhDrnDRJlA/vsksSL7/swapVOi64gPOhEhGp2jpYnQngkFAoVBoKhYIATgDwdhtvg4iIiKhdvPSSBw884EUsZg1WV6zQEYmI342mYTRkUPrppyJ6/eQTNx580IcHH/SmXrdokQu77JLACSfEMXSogURCw5gxCfTrJ1YyerTIph51VAyHHML5UImIVG0arFZUVKwFcB2ADwB8A2BGRUXFV225DSIiIqL2smGDBsPQsGGDhtmzXdh11wRcLgPXXuvHz34WBADU1YkgdcsWDfX1wNdfW5tTS5eKv/v1E4HoHnskoGnAsGHib5lVBYCzzori8ssjeOihsGUQJyIiauU8qwBQUVEx3Pb3DAAzWrteIiIioo62YYOIGCdMEKNTXnhhFKtXa9i8WcPixS6sXq2lgtWZM93QNCAe13DuuVF8/bULwaCBTz5xo6DAgN4Uw+64owhShw4VP/fZxwxWR4wwcPXVLP8lInLS1mXARERERN3Whg1m08jrNfDrX0exebP52H/+40Ft0wwUa9fqeOIJL4YOTeK22yJ4660GDBwoyntLSoxUmfCYMSJI3XnnJHw+A5Mns9yXiCgXDFaJiIiIADQ2AjU1Zi3u0083Yvhwc573vfeO45FHvEgkzGX69k3i7rvNEt7SUhGYFhcbiETEg6GQeOyYY+JYsKAO/ftz7ngioly0ugyYiIiIqDtZuVJDeXn6nKbr11s7je6yiwgyX3mlAQsW6Bg7NomTTgpallm0qD41OjAAlJaagehjjzXiySc9qcc0DejTpw3fCBFRD8dglYiIiLq1REL883qbXzYWA/bcsxBHHhnDE0+EAZij+27caC0469tXPLHvvgnsu28CCaV69667wjj66LglUAXMYDUc1nDssXEce2y8ZW+KiIgYrBIREVH3dtddXrz9thsffdTQ7LJVVSJ7+r//eQCIYPXYYwNobNSwcKGYqua886KpwZBULpcY4XfTJh1lZUZq+hmVGay29N0QEZHEYJWIiIi6hU2bNMcA8f333ViyRIdhAOvWaejf38CmTRo2bdIwbpw16Ny40Vrqm0wCX34pmkO6buD002P4058iaRlTaezYJD75RE9lY+3KysQTsr8qERG1HAdYIiIioi5v6VIN48YVYM4ca9Olvh747jsd8biGN95wY/fdC/H8825ceaUPBx1UgC++cFmWV4NVwzD/Lioy8M47DbjnnsyBKgD84Q9impmxY51H9JWZ1Ugk77dIREQ2DFaJiIioy1u2TIdhaFi2zNp0WbjQlRqd98ILxYhJS5fqqK4Wy118sd8SOKr9Utev17Bqlfj70Ucbseuu6aW/dvvvn0BlZS1GjnROrfbubSAYNPDnPzNaJSJqLZYBExERUZcn+5pu3ix+PvqoBzvvnMScOWbmNBwWz9XVadA0EUyuXKnjP//x4De/iQGwZla//tqF+fNFsDpkSNtMJ+NyAStW1LXJuoiItncMVomIiKjL27RJBJXV1Rq2bgWuu05kUadOjUHXDSSTZhBaWalh61YNRx8dwzffuPDFFy5LsOpyGUgmgbPOCqReM3hw81lVIiLqWCwDJiIioi5Pzax++qm41+7xGJg3z4WJExNpy9bUaOjVy8DYsUn88IPZ3Kms1BEKJbHbbtbgNBAAERF1MQxWiYiIqMuTweqKFTqmT/cAAHRdBJ+HHWYGq+PGJVBZqWPrVg0lJUAolMDSpTpiMTGg0urVGsrLDRx0kDn/6XHHxTr2zRARUU5YBkxERETtxjCA2lqguLh169m0SQSrn37qTvVHldPD7LtvHIAPALDbbgk8/7wHsZiGkhIDAwcmEY1qWL5cx1dfufD99y6cckoYRx0Vx6ZNGm65JcKsKhFRF8XMKhEREbWb1193Y/z4QtTUtG49MlgFgL32SuCqq8zRdkeNMkt6hw83EIuJZUtKDOy4o3iuokLHjBkejB+fwPnnxzBkiIG772agSkTUlTFYJSIionbxxRcuzJnjQkODyGw++KAnbd7TXMkyYADYaadkaj5TAOjd21yurMwMXHv1MjB6dBKaZmDBAh0LF+qYMiUBna0fIqJugWXARERE1OaqqzUce2ww9feqVTpuv92HE06IY9KkRJZXpqurE+uTQqEkyspEsOpyiZ+/+lUUug6Ul5tBbEmJgUAAGDHCwAsveBCNapgwIb9tExFR5+G9RSIiImpzcj5Uad48F+JxDevXa2nLJhLAvHk6DGWq0yVLdPz85wFs2wZceKEfmgb07i0WGDw4mcqgFhSI5e+5J4K//jWC3XYzg9FevcTyoVAC69eLJs+eezJYJSLqLhisEhERUasZBvDCC25Eo+Jve7D61Vei/HfDBg2NjcCKFebzb73lxtSpBbjzTi9iMRG83nGHF5995sZ777nx2WdunHZaDG+/XY8zzojigAMSqcBVBqRSnz7m7/K5HXYQge3uuycsmVciIuraGKwSERFRq732mhsXXhjAgw96AaQHq/Pni2D1xx9dGDasCHvtVYi6OvHcsmWiOXLvvT4cd1wQkycXwCtWgw0bNGzdqmHYMAMjRhi4994IfD5g2DADp58exRNPNKbtS0GBCEjlCMSjR4tg9bzzom36nomIqH2xzyoRERFl1NAAnHpqALfeGsEuuyQzLrdxowhOKyvFzy1bml93ZaWGwkLDkmWdO1cEtTKA/fJL8feAAdZtu1zAffdF4OSddxrw4otu9O0rgtZf/jKOHXaox8SJmfefiIi6HmZWiYiIKKOvv3bh88/duOYaX9bl4nHxc8YMDw4/PIjq6uabGJs2afjwQxfmzcs8QvBnn4n76gMH5l6+O2ZMEtdeG4XWFAO7XGCgSkTUDTFYJSIiomZFo9ay3hUrNPz61/5UKW9jo3g+HNbw9deu1EBKV1wRwdtv1zuu84or/DjppCB++MGF8eOdBz7atk2sx55ZJSKino/BKhEREWVUUyOCxYit4vauu3x46y0P3nhDZD7VqWUA4PvvdfTvn8RVV0Wxxx5JHH10DPvvL9KvJSUiS7p4sZlRHTPGDEYHDUoPTPv358BIRETbGwarRERElNGWLTJYtQajxcUieKyp0fDqq2689ZZ1GIzvv3elRuwFgMcfD+O55xpx/vlRvPZaQ+rxQw6Jp37KgZH23ltkWXfd1cy2Bs0pW4mIaDvBYJWIiIgyksFq1DaQrgwsN23ScN55Aaxda21S1NRo6NPHmg11uYBbb41g7Fgzc3rUUXFUVNTi+OPjGDhQPC5LgnfemXOiEhFtzzgaMBEREWVUUyN+1tZaM6vy7+XLM9/3tgerToYMSaJ3b/H7gAEGqquTGDRIvM7vB77/vi6tBJmIiLYPDFaJiIgoI9lndetWDY2NQCCA1N+AOX+qk1xG8B0yxMyyHnFEHCNG6DjqqDiuvDKC88+PoqSkFTtPRETdGoNVIiIiymjzZjOjunGjhuHDRQAqy4Pt5b+qqVPjza5/8GAzoD333Fjq9yuvjDotTkRE2xH2WSUiIqI0jzziwTnn+FOZVcAMXNev11BZaS0LPvTQOAYPto7iKwdKciL7pXq9bbXHRETU0zCzSkRERGluuMEPANhppwSKigzU1mpobNSQSAC77loIACgrS6KyUtz3vv32MC67zI81a3Rce20Ee++dgCtzhTBefbUBDQ1a5gWIiGi7x8wqERERZVRVpaVG6W1osM6nus8+Zua0vNzAaaeJMt5TTolZnnNSUACUlnLuVCIiyozBKhEREVkYSgxZVaWnBkpqbLSW/44aZZb9BgLAiSfGUVlZi/79GYQSEVHrMVglIiIiCzV7CgAjRpiZVXtfVSIiovbCPqtERERksWpVpmBVQ2Wl+XhVlYYPPqhPm4OViIioLTBYJSIiIovVq62FVyNHimC1sRGoqxPPTZkSx8UXR1NT2RAREbU1BqtERERksWyZNViVAWljo4aaGg1FRQZeeqmxM3aNiIi2I+yzSkRE7e7f//bgzTd5f7S7WLRIx4AB5uBJAwcm4fUaaGwUfVbLyphNJSKi9sdglYiI2t3jj3vw3HMMVruLRYtcmDjRnHqmoAAIBoH6eg1r1ugoK0tmeTUREVHbYMuhFXzPzYD/P0939m50Po8LJbHs8+lRN8Vz27N14Pl9cqUL/o0GSn7OIKdDtOLcJuLAEytdGGYkcVnTPe2SnyfwZoML0cfFMgMH8lx2Gn4v91w8tz1XJ5/b8KlnIHLyaZ22/dZgZpWIiNqdYQBJVo52C/UNYmTfggLr43pTiyEYNDB8GANVIiJqf5phdOnWw3AAy6ur65BkK6fLKi0tQlVVbWfvBrUDntuerSPP7+67F2DYsCReeYWD8nSE1pzb6dM9uPxyP+bNq0M8Dng8wODBBg4+OIhvv3XhkEPi+M9/eB47C7+Xey6e256L5zYzXdfQt28hAIwAsML+PMuAiYiozX3xhQsjRiRRXi5uNMbjQCzGuTi7g6VLdfh8BgYNMlLZVAAIBMTPvn1585iIiDoGy4CJiKjNHXtsEEccEUz9nUiIgJW6vmXLNIwYkbQEqgAQCIgglcEqERF1FAarRETULtauNf+LicU0xGKduDOUs2XLdIwYkd4nVfYaYrBKREQdhcEqERG1qYTDgIfxODOr2bzwghuzZ7s6ezeQSAArVugYOTI9IG1oGnipXz8OrkRERB2DfVaJiLohGRC6Oj++SeOUQU0k2Gc1mwsvFB1CKys7dwCOTZs0RKMahgxJD0gbm8ZU6tePmVUiIuoYzKwSEXVDhx0WxIABRZ29G46cgtVYzPlxAurrO3sPTNGo+On3Z86ssgyYiIg6CoPVVjrooCAOPzzY/IJZbN4MPP88k9xElLtFi7pgSrVJNGrNoBoGkEhoPbIM2DCAWbNcSLaiMnblyq7zX7G8oeB2+C+poUH8ZLBKREQdpev8D9lNffedC19/3fJGo2EAu+5aiIsuCmDDBpbIEVHb+e47vVNuhNmDUlmyLLN2Pckbb7hxyilBPPqop8XrkMFqMNj5QWA8Lv4f8ji8nYEDxf716dP5+0lERNsHBqudbNEiHZGIaByEw9bnPvjAhfvv93bCXhFRTzB9ugfXXefv8O3ag1KZrZOBUE+ydq14T6tXt/y/05UrxToGDOj8IDBbZnX69EZMm9aI4uKO3SciItp+sfa0k23aZDbe7NmIk08W5cV/+IPZ8qurE/2Gyso6v1FDRF1bZ/UTtW9TZlZ7Yp9VOZ2L1oo4fMUKEegWFHT+97r8f8jjSd+X8nIDxxzTA2u5iYioy2JmtY04TdWQi9pas4WTy0iZt93mw6mnBlq2MSLarkSj7TO36dy5etZBgWSfVZfLQCQCLF8u/qvpqX1WW6uyUhwvmZH+29+8KCvrnMGz5PXiVAZMRETU0RistoLaCJwzx9Wihti2bWqw2vzylZVaqmFDRJSNyKxqbRJQSXV1wDHHBPHss5mjGfldpuvAJZf4ccghBZbHe5K2yKzKG5WyS8jtt/sAiGPd0eS+MFglIqKugMFqK6gNiWOPDeLuu/PvX1qrTKmXS0MuEtFSDZrWePppD8rKinKeMsEwgCef9GDVKgbKRN2F2Ve07dZZW6shkdCwZUvm7wI1WH33XbO3STKp5TVq7rJlGmbN6rqjHgNtFayKnzKzGgiIlbbVjcm5c3XMnJnbcWRmlYiIuhIGq61QV2dtSMydm3+jyppZbb5hEg4DkUjem0kjB27KtTH0/fc6rrzSj2uv7fjBWojIKtdMqQw82jKj2dgofsppTJyYZcAiYHXap1xMmlSIU05p3dRg7c0wxHttTbAqbybIYLW4WAar5sFbs0bDJ5+0LHA/8sgCnHZabsdR7ovb3fn9Z4mIiBistoI9WC0qMhCN5j7B+wcfuCwBbqbsh2EAH37oQkODCFTDYeCUUwJ44IGWjxTsatpsMgn86U++ZhtBb78tsiPl5a2YTJCI2oR95PBM5A2wtgxWGxo0y0/n7Yqfut66YLU7aIvMqhmsipWUlKRnVvfbrwAnnNC2gfvEiQW44Qaf5TFmVomIqCthsNoKagkvAHi9wHHHBTFiRG4DY5x8chAff2yWyGVqxN12mxcnnRTEyy97EIloMAwNs2a5ceutPnzxRcvutOu6aAw1Nmr4xz+8zTaC3ntP7CcngyfqfDK7CWTPsspMXS5VG7mSgXIuwarLJQZZUmW6KZdItE3VSGfRNPE+167V0v5vqKrSUFZWhGefdR6AX54fM7MqfqrBarbj3VKrVul45BHrTU+5L05T1xAREXU0BqutYM+sbt6sYd68lvevytSIe+ABX+p5e2NuxYqWNWBktsP+HjKRjaa2bPTm46efNLz7btfuu0bUUdTAJdtI5PI7pS36rK5Zo+H9911obJSZ1czLZs+sanj2WTf23996g+z3v/djyJCWj4BrGEBZWVGrKk5aul3AzKzuvnshjj7a+t6WLxdPPvWU877J8yO/34NBsdKNG9O/b+1z2NrdcYcXv/yl84jxhgHMmaMjmcx8c9Scuib7doiIiDoCg9VWqK+3NiSqq82/7dkOwwBOPDHgGHDJhokaCDplSxIJIBy2blM2HPMlG5DZBklRyYZpZ009se++hTjjjI7puxaNAtOne/IaCIaoI6mf+2xltbKstLkAJxcHHVSAU08NKn1WM393qFPX2IPVeBz44QcXFi92Wfb9v//NHB05fR+uXSuyla++KlKAMtC7/fbOC1bl7z/8YP2el8cgUxZcHodkUkM8bp4vtc+qpGbVnfzwg45ly5z/a58504WjjirAI494sHWr8/kzy4BZRUNERJ2PwWor2Eu91GDV3qBobAQ+/tidyryqQV+vXkbaY0590pwyq7n2XbOTjafNm3MLVmXjuKf1N3Pyz396cfnlfjz/fNeog2vLaUeoZ1C/XzoqsyqDG/ldkK1vvppZddnuz0Wj5v43F3jZ16f65hux4pdeclvWZQ+O28Nzz7kxdaq4eaYGq/b/EyR1jAAn6jkU4xKIY+yUWW2uHLi2Vst4c2LJEnFwvvvOhZoa52WYWSUioq6EwWoryBLa99+vx9lnRy3Bam2ttUFhH5TEqc+Z2iCzZ20B0YiwB6dqptUwgDfecOcUUMoGnbrPmSSTPTdYTSbFMVMbkfKY1NR0/jQ9f/yjD+XlmUsjv/tOR1lZEZYu7fx9pY6jBizZM6tymba7PnLLrIqfTqMBx+NQSonT1+EUfDsFX7LaI9hUcCG/C+3BcXtYsMCFr78Wb0wNVjPd/JPLZApW1XMYjZo3JTdtSl9fcwH+tm1axmtCTnsWj6dX1UQiwJYtZlacwSoREXUFDFZbQQarO+6YxNChydR/8oB1ShrAbFg5DU4iGybWYDV9e/G4llYGrGZaf/hBx9lnB/DBB8231mSDLpfMqhogZ8vidEfTp3tw9tkB/Oc/ZstM9j3rChnNxx8XJY2ZMmPTp4v9njWra2SBqWOo/UXj8cyfYXndtOVNJnkjLXufVbFMpj6r2aa/cdpXp+tffocWFMjB4sTjHRGs1tZqSCY1JBLWYDXTDS55PDIHq+brolEtFczL/2PU7yKnG5kqEaw6LyP/v4jH0/f1nHMCCIWKlKlrsm6GiIioQzBYbYW6OsDvN+DxpI+Say8Hk40PGWyqjTT5mNrozJRZtZcBq33X6urkz+YDUHsZcLY59az943pWBm/pUnEgtmwxH5PHpiv1WZXn1k5mR/r06QKRNbXIypVaagCeTOyfffUzma3EV95AyxSsJpNiWqx8bszI75dcRwN2mrpG7r9Tn3un9+P0vZMps9peZcDLlmmpslz5/R6N5pZZNfukOq9bfc9qZlXetFRvGDaXWa2tzXy+5TmLx9O/+999V0SncjA99lklIqKugMFqKxiGhn79xH/o9mA1U2bVqYRO3sFWGxhOGYdwGEgktLTHJFnide+9XowZU5h139OD1czLWrM4WVfb7to6gJSNwULlcHWlzKqcDiPTDQhZsiznZaTuxTCAPfcsxN57Z/+8HnNM0DJSbq6fSfmdkil4eest4KSTgvj229z/K5CBWkvnWRVlwOJ3p8DLaV+dHpOBrhygrr0zq5MmFWKXXcR5kt084nHr90SmzKosY85WBizfRzRqBt7yc6/+f5LtuBuG2LdM51t+37/1lhsXXyxGDA40DRzcu7fY/uLF4oQxs0pERF0Bg9VWuPDCKP7zH9FCKrS1Ne19VjNlVseOTWD6dLEOtdHplFl1ekwtC5YNoh9/dDXb31I2sGSwk61hkuvIox2hrbcvj6ksJQTMOWiTyczHMJkE/vEPT8aMZ1vxiVmLMgarsvHZFQJryt/Chbl9BdunxMr1Mymfy1QqvHy5+JnrQGuAGThlKwM2RwPOnlm94QY/XnjB7fha+2vs5PZlsGX2WRUfhnnzdHz4YesiV8MAvv8+/RzJz2M0qlm6RmQaXb25zGoiYQarkYiWyqw2NIjXqN/n2YLVcFhkoRMJDcmk2P/Nm83n5fe9YZjr8HrFdgcNEjtXUSGOGfusEhFRV8BgtRX69TMQCon/4NVgBzCzDxs2aHjzTXdaZlU21v7ylwh23lm0dpobYMkpYFEzq/b+rJJhiOXWrdOwdq1YRjaGcsmsqtmPzs6stvX25XGWQSFgZlazeestN/70Jz/+/Gdf8wu3grdpFo5MQbE8fz2tL/H24s03W5a+UgNFe7WFSpbPZhoddt068dN+cy0buWwkomW87szMqpG6+SPF42af1XnzXGnHQH7G1Rsw2fqsyjJW+V0og+N77vHhxhtb9/n86CMXDjywIG0AM/n9/sMPeqo0OJk0g1UZMEsyAM/WZ7WgQC4rvnN9PgOGoaGhAdi61Vw2WxmwmoG99lrgqqt82HFHMyPvdFNC7lv//mKf5bQ3DFaJiKgrYLDaRgoLncuATz45gLPOCqQaMeYomGK5YNBIBYpq9sMpa2EfdKmkxLA0XOwN0g0bNNTVAY884sHQoUU466wALrvMD8ApWO0efVbbPlgVP9VGtwxWs5Ucy8Z4PhmplpBZj0yZVXldZRtkx0kkAlxzjc+SdaGOp851ae+Pnk2ufVbNzKrz82vXip/yZkgsBtxwgw9VVZmvJ/V7KFN2VW5X09LLctXMKpAeKMvXqsfDKdsqty0rIGTwKrdXU6O1+vMpX79mjfW/Svn9fsYZATzzjLijlEiYn8dEQrMcc/meEgkRfNo/d/G4ecOzrk6DYWipriX19ZrlOsmWWVXHSrjzTuDJJ61zzjqN/i6DfPU70O02crppR0RE1N4YrLYReVdckg2w1avFIZZ33+2jAQeD5h1sNdjMJbPaq5dhyabaG7vjxxdiv/0K8NRTYgNLl+qpRpbs3yobQdnuonelPqv5BmXNkcdZzWrn0mdVNojb+njU1FjnVmyuDFg2+pvry2sYSGXVATE35bRpXtx+e/tmhru6Zcs0bNjQea1y9frJJ7BSg47c+qw6r9ueWf3xRx2PPOLFRx+5sGmThpNPDmDaNOuXgxpcZhqZVg3O7EGP2mcVEN9BaqAk349aNZItsypfK5eXn83aWrFu++f4pZfcWLhQx8qVGu64w5v1cy6/k+3nxmmQqXhcs5QBq9+bap/V444LWrKdgOyzKn6X39FmsGq9qZEts9pchtwpWBVlw9b/P5hVJSKiroLBahvJlFn1+cTjGzaIQ20PVgMBAy6XGEhHbZDZB2gC0jOrffoYlgaGDEBV69bpqQalOqKofb7W7AMsidcXFxtdIFg1f08kMpc35kpmlNTGsiwjzNaIlSNltnXwfPPNPpx1VkDZjvjpVAbcXJmk6vXX3dh998JUHz6ZqerJfV2nTfOk5sLMZNKkQowfn31wo/akXj9Oc2pmsmZN88GqYTSfWbUHq/J7IRzWcP31PnzwgRuvvmr9clADokyZVXl9JRLZ+6wC4rtOXY8MrJ3646vSg1Xxt9xeba2GSERLC+5+97sADj20AK+/7sZ99/ksN3Gk887z49JLfan3IUfIBcSxdMpuqmXA6v6J/TdvKi1Y4Gp6XryvuXN1xGJaqs/qtm3iNXLwvro6rdkBlr77TseWLc7/b6jUmwtHHRXDpZeK/xDCYev/HxxciYiIugoGq20kELD+LcuxZGZMZm/CYQ3/+587VY4bDIpyK4/Hmt1T+yhJ9uxaSUn2zKokB+doaDBHibT3b82lz2pxsdHpAyypje4TTwxg8GCRoVi8WG9Rv00zkE9v5GUL5MzS7fy3mc3mzdbSRXmzo65Ow+9/78f8+eZHVr150dx7X7VKrPPhh82SRaD9pvloiW+/1fHTT20X/N9+uw/PPdd+KaItW4CysiI8+2zLW/bq58kp62UnM+hr1uipvqCZPpNiDtDsfVZlGbDaDxUQwcv8+SKokqPESvlkVpNJpzJgawC5bZt1PfK1TgGsSl7/9kys3J4M3NSspGrTJjltVfrzr77qwTPPeFP7snat+UHJ1H88HrcGi+rn02mApcWLdYwfX4AjjxRlObIM+JJLxH8mahmweiwaG8WNjTPOCKTKiQ86qADHHBNsNlhVDR5soLRU9vfVLNcIp60hIqKuogs1Vbs3e6nbjBlePP+8G34Rk6aC1cZG4De/MSNbWfrldlsbZE6j+aoNOp/PQDBo5DTAkprFkHf47YGtvc/qxIkFmD7dY3l9UZHR5pnEfKkN89mzRZCweLGO/fcvwN13ezO8KjOnMmCnbdnJRmdLgvd333WlZcmlaNTa103e7FixQseLL3rwq1+Z145686K5YFWet48+cqVGCQXab5qPlrjySj9uvtnfZuuLx7Ofn5YOShWJiABr40bx9XnPPdZS6oULdYweXWjJxuWyD7kEq2bwpGH4cKNpHdkDRvvvAPC3v3lx113e1E0xeXNNfp9UV2tYsUK8P/tnXu0Xman/pAx8nDKr4bC1D+q2bZolAHS6oeZ0U0gGZvKzKL+nXC6xXfn5yDRCrzze2Y67/H5Yt85cJlOprdim9bU//qjjnXdcljJgGfy//roHmzebB8felUQGq3V11lHLGxo0LFyo4913xU3P008X3wk//ujKa3Ty3r2N1P9PDzzgtbwvZlaJiKirYLDaDsaNEy3Qb791pZUBNzZq6NfPvL0ugxGPx9ogc8oGqAGO3y/+WQPR5vctFhMNJqcBS6RIBFi1Ssfll/ub9lk8bs+sjhhRiN//vu2Ci1wkEkBFhY7hw83SzfXrxXv56qv8Iy+nMmB5HrJPCWItQczVt9/qOOOMIK67zrmvaCxmvQ7kTRCngaDUGxXZRoQF1FGDNWzbZq6nvYLVeBy4/35vxqDcieib13b7EItlHxCspdv6xS+C2GGHotQxtAely5bp2Lo1t76wsRhQVia+D3INVhsagOpqHcOHJ1OPZVpWsgect9/uw913m9egDFTk94ksVQXSv1fq6jT06mX2p3Qit+cUrNqzf/G4lspyiv22liQ77YO6z/LzIr+nNM0aNGbKrMrjnSmYBcxpa9TMarZgtb5eQ3m5OC/19RqOOCKIX/0qaOlb3r+/eP7LL60fPvuI8moZcH29KBMWwap5bv/3Pw/ee8+MLPPJrPbubaT+f3roIW/q5gTAPqtERNR1MFhtB5dfHkWvXgaiUaRlVsNhpEqvADMY8XisgaBTZlXNpAUCBgIBM7M6a5YLc+c2H3lEo87lwmqjXjbGNM1Aba05x2NxcfpcsC++2LGtmnhcw4oVmiWj09JSVsMwgzz1fclA3img37YNqKoyy6nzHR1Z9kuUA2/ZxePWfZHbUUcZte+neDz7dtXrafNmrd3LgF980Y3bbvPhnntyz3ZHIlrWkU7zYRjiWlGDnB9+0LFsmbn+bEFKNvJzZp+GSpLbzOXmUTwuMmgul5FTsBqNmoGTDFYzlaKr10dz+2KWAYu/ZV/fAQOSaesPh7VUwJUpEFQzq/ZyeqdgT95wAtSy3uyZVfNGk7WPayxm3caWLRqqqzX8618ey77I4/3jj3rGUbEXLRLHIZ/Manm5GcjL7+zvvtNTy8h9sH9fyyobyV4GHAwaCAZFNjvTzcZ8gtVevczMqh0zq0RE1FUwWG0HgYC4Yx2NmlOPmANsaGl9wADROLBnVu39htTGWyAgAmH52CmnBDFrVvMtjFhMcwxWrYM7mft00UV+vPKKJ/W+OnuApVgsfSApGXA1NyKuXaY+cfI9Ll2qW/qIAsDEiYUYN65QGe20+e0kk2bppDz2vgyD8NrLgOV2ZLCpbk89j82dFzUwq67WUg389sqsyv3NlsG3i0atZaUrV4pyx5YwbyaYjx1wQAEmTTIz8rkEh9nYg1Rz25rlZzbxuAafT2S5chkNOBbTUoMrNResqo9nuz7KypIOZcA6XC4Do0YlHTO3ck7OTAG/2kfTvm35/aJSg0H5WrVfq9OxlEGj/EzI5aNRa9C2dStw7bU+XH+935LNlDeO7rnHh3HjzOtCDeyXLBHLqyN0Z+6zqqGuTksFqw0NGsaMETs3Z44r9T4y3ZCRAyxJMli97DI/Zs50IxgUy8iBmZzkE6z6fOI7nYiIqCtjsNoOfD7xLxLRHIMBp1F7xQBL1j6rffpkbkj4/UZTsJrfvonMqtjOgAFmdKc2SGU2wOMBliwxLxF7v9psnnvOjbKyoqzTLLREIpH+nnMZvddJpj6f8ljMnOnGEUdYO5LJIMwpGMrk9dfd2HXXQtTVmcGbvIlhJ/pZpmdPZVCgBuRqsNpcoL5li1lmuHmzllre5Wqfxqq8xjK9T+fXWG8g7LlnIQ49tMBx2Y0bNey+e0HGAZnkeckWLLc0syqp17YawOSTWY3FxOcqEMjc51zlnFnNnt2U28lkzJhkWhkwABQWAl6v82e+rEyc1+Yzq5pDsJr+GtlNAlDLetMDWPVveY3Zp66xB6tbtmip59T5Y9WbFWrFgtMIx3IuVyBzZlX2xTXLgIEBA6zdQCKRzFPP2MuA1e//det0FBQYCATEccl0beXTZ9XnM9JummmaHOU89/UQERG1JwarbUg2/P1+A16vYQkMVU6NZLdbjNgqsyZbt5qTwtvpuigH8/nEaMD5BGnRqNlY2m03M0JTGyeyMeZ2i4asZO9Xm82f/iRaQbkMMtMceyDpdEyB/DKr11zjw29/a9bA5ZqFMvcjvXw4kzVrRMalqsrMamcqv4vFrO9XNkplkKxuTz0OuWRWd9ghmfrdDFab3/+WkO/Tm8eYV9Fo7mXAr7/uxtq1OqZNc96APB7ZAsZ85jV1ogYtan+/XPo8q8u63UbTZ7n55WMxcT3puoFBg7IHFtYMvfN7LSgQAZW8caPuQ0GBkTZKuVRYKAZ4k9elYQC33+5NfX+pfVbt1QfNlQHLGwzqvtj3QQ0o7VPXRKOaZRComhoNJSXid/WcZ7rWMo1wnG3/ATNwHzJEdNH4xz+8adn7xsZsmVXr3/asZ0GByLauXatl/A7MJ7O6yy5J+P3OAXJrpwQjIiJqKwxW25Ds5+P3i0Z6JOLcP1QGq7fdZrbGPB4DX3zhxh57FGL9eg01NcgYrJaUiIBYTpeTacoaJ7GYlmrU7babGd2p2RnZ4PF4jFQjT/4tG43NBciywenU9zZfasMpHk8vY3aaFqI533zjSpXmifWq68ulHDP9dZnIY7Ftm9nIzFQGLAYFSt+OvGaswar5e3PlyFu2aBg1yhzIp737rMpzlikod2LPrGYjj0umvnVmKa7Iwj7/fPqCMnDJpxRSve7Va1sNcMzMai5lwOImkN+f2+c4GtWwZo2OAQOM1H7n0mc1U+A8aJAYOM0+dQ0gg1Xn6ao8HjF1lhykavFiHX/7mw/nnhto2rZ4PJFIz/w6BVTqHLO1tWJKoL//3bwRYd8H9Xjby4DtfVZraswBoXIp/W4uWHUqYxaPi9eVlRm4994wfvzRhW+/Tb8bVF+voago/ZorKjJw1llRDBsmPqf279hg0MDkyXEsWeLC8uXOH9xswWpJiVjh/vvHUVlZi7Ky9D6rvXqJn/n2xSciImovDFbbkMxSiT6rouGXKVg95pgYzjvPbIGpje5ddy2EYVgzq2r/1eJio2k0YDlHXuZ9OuSQ9JasLBVTs6ZqY1BmJVwu67rVfrXNBUeysdrafoH2fYvH09+vPMZqqZ6TqioNkycHsWyZhpoa6yAlTv1Ec9mnXKbykftXU2OWI2Yqj43FNMRiZrbczDKZfZ7N9TqXMdoZhtj24MFJeL2GZYAl+5RLbUUtA161SsO33+ooKyvCtGnOA3IZhniNeP/Nr182pjPNB6lmVl980Y2LLgqkLSOD1XwGk1EDFbUE1lpyay0VzyYeF10F7CN7ZxKLiWlrBg9Opr5vchsN2HmZgQNFkFRbK86BNbOa3j1B8noN9OplZlblTQ/53aH26c6lDFj9npDn5YcfzEDPvg/OwaoZcMttDBqUxBdfuFBYmD1YVW9YZBvBWteNZjOrhYWG5bvVyciR6c97PMCdd0bw4Yf1ePDBRowfn8SRR5onsaDAwEEHiTf77rvOF21trRmY28n/T9QgOD1YZWaViIi6FgarbUg2Hv1+pAZYUgMKteQqYGs7O00VoAaraqPi6qsjOO+8aOqxbH3d9t8/js8+s3Zkko0tdUAPpzJgj0dkKwYNSmLRorqmYFWW2pnLOzWWZYC2eLHe6oDVHlTaS+BkIzVTtverr3S8/74Lr73mxpIlLjz0kDdt2hK1MWx/P06BeT5lwHJ/t25tvgxYri+XeVzVc5Dt5kFdnThvvXsb6N3bwJYtZkCY76BUUnW1hpkzM9cQqzdpJk4sxMEHi76nf/ubc9mu+j7tffqc9lEeJ/vnZs4cHccfH0hlaGMxLWOmTGar82mYq9eymll1yobn2mdVZFZzKwOORMRI0oMGGakgW577WbNcuPxyM2WfbZ5VOX3WoEGib2o8rmHLFmsgKcuAna5xmVm1B6vycyE/s/YyYJfLeSAp9biq50t2rbDvgxpQyu8k9fjJbfzhD1EsXuzCBx+Ig6VmcFXqCO3ZMqs+n7Wvu0p+pxQWGmn9T+2cglX5/0dBAXDSSXFoGvDEE2Hsu69488EgEAol0auXgaVLnf/rrq1FxrEO5P8n6v8XcuoaSQ7+15L5o4mIiNoDg9U2JBuPPp8sA9YsDahzzzVbr/aRH52yO2qDRwa6breBE0+M4+CDE6nHsg1i5HabI3dKMhj1+4Fvv63Db34TtWVWzYyTmKbCQHm50TTAkljGafRgyTAAwxDruPlmP3bbzXmQnGzuvNOLG24QDW97hsierZbHOFOwevTRBTj11KBl1GB7ebLTPKuS81Q/6cu+8YbbsTEsg5atW83+apkzq9af2QIe9drKFjTLoKxPHwN9+ogpUmQA3dJG6eOPe3D66QFL30CV3G97RixTXzv1fdr79DkFcXK/7cHq738fwKefulNlkqKPdvZgNRLJvd+3GlSpmdXXXnOnAvF8Bt+SfVblyN6zZrmy7ks4rGH9epFZle9dHuNTTgli+nRv6vXWGzDWYyDL0HfbzQxQjjkmiMcfN28miMyq4XgNer2wZFbt71mtwFCvzQEDjNT0QfPn12HZMnEBqQGi+vvo0SKok/uwdKmGv/zF6zg4mhqEvfSSG4WFBk46SeyQnH7LPvftU081oKREjHJuGMCCBXoqEHbK2ns84vvR6fMrr4eCAvEvmxEjnDKrmbp9GE3rNaBpcCwhlrZtyzwwn7xBoX6e7N0RZGaVZcBERNRVMFhtQ6edJhpGwaAoA5aZ1cMPj+Odd+oxdarZarNnVt3u9AaGHEkSMDNxauNCriNbZtXlSh/kRja4CwsNlJWJf4ZhDrojAxBZFiiDYo/HUMor1QyTdf32ACZTgOIkGhXZsXvu8eGRR7xNrzefF2XAzpnV5rKEsuS1vl5LK9/N1mc1W7AqG9F1dcDZZwcc+0bKxmFNTeYsn3298bjMSmVePtM8q2vWaJZjIW8+FBaK7MrmzWaGN5cyZlU8LqYB+fxzFwxDw7p1zl8h8pzbA51Mwbd6jdj7rToFq5kyq/YSzlgs880ctZwz1+yqmhXcssV8fMYML26/3QfDyD5Pr50IVkWGa8ECV1PAmXnu4rVrRan04MFGKhOX6eZKtsxqIgH88pcxXHaZGaDIaVqkggIjbUotSWRWzQDNPqCV/NswrKMBDxmShGFoCAbFAFGFhSJ7qt5QUEe0lcGq/Ey+9ZYb997rswzIJK99NSu8ZIkLd98dTvueXb/eer1OnpzAz38eQzQKfPaZCz/7WUFqehuZiVRvLHq9Rsbspfws5ZJZlf1Sd9/d/OA6VdcASI0bIANg+6BIksslSpQzjXXQr594XP0+s++n07RqREREnYnBahu64YYIli2rRTAIZTRgYPToBHbfPWlpONkHdbEHSO+9V48DDzRbebJcS22oyMeylQ+63emNIBmsFhcbqWXEPoifshHf0CACQRkoqyODWueEta4/UwCTi//9z42jj7YOi2nPENmDRxmMrFmj4aWXMndAlFkgp0GfsvVZdQq2zbkgrX1KGxu1poDFXNYcYMkMxDIFiWp5cXOZOacBllas0LDHHoW45x7zDoW8Pvx+AyUlhmWgp3ynqPj2Wx2PPebFp5+K46zOj6mSQWP6sXRer3q87AG9080YeZzsfW7la2V2NhrVMn4+1GB1zRoNf/6zt9njoQarTtO2VFdreWZWtdQAS5I68JedzNz37m2kMnH2EnB5jWUr1U8kxPWg68jYx7GgwEhNXWPfhr3PqhmkmuW/kvr5GTxYbGvo0GTq3NlvpqlZUzNYle8j/TMsp7NauVJLZQ8B4NBDRSmtmgVV50v1ekWwLG4sali9Wjz344/i+MuAVK1MkZnVbNOKFRYalu96XU9ftrAQePvtevznP+admUx9p2VmVQbNmboQJJPiuu/b1/munQxW1c9TYSHw00+1qT69DFaJiKirYbDahnRd/OcPiAZQOCwCK5kNlY0FIH2aAjX7EwgYGD8+aRmp1SmzKh/Llrl0u0XpmNo3STZ4zWDV2k9JZijCYc2SWXW7Rabk5JMDlvkK7cFqa6ar2bZNS5UQS/a+mfYsmGx8bdqk43e/C6RleiWZjXGaOkg2sleu1NKedy5DNYMiNTiNRoHbbvNi8OCi1GPqAEsymMrUx9Sc9qT5gYacgtWNG8VFI/voieXMsm/ZB9HMrGbfhp19v9UMl0oGHPbsYqZMsXqM7WXATpnRTNPDyJsHarBsLwM++ugAvvtOt1QA/P73ATz4oC+VVctEDaScbnrceacXFRVmCbLd4sU6vv7aOs2NmGdVnVPTrBS4/36v5bMmj43bnX6Tyb5MtmqBZNLsI5k5WEWq9N++DY9HvK6+XsPs2S589pnb8p7VbavHYcgQEUgNHWoNAFXqzQo53ZK9kkF+RktKDCQSYj7oRELD+PFmoFZUlL5+9Vro21d8N3o8Yr3ye3HVKi31PACUlSVT84/qurgGsgersqJFLKP2h5WCQQN77JFEnz7mY5kyq/KmhPzuzzSSuPzebK7Pqv2GUXGxuW0Gq0RE1NXkMQ4m5cPrFeVshmFmJtXyLHufVbWBXloqGlHqnXbZQFHvqsvfs/VZlQ1SOZUOYGZW7Y052cCUDfLGxvTMKiACITWjdf75wDvvmOtrzUiSTgGadeoapzJg6/KrV+vo1Ss9uyADK/sAL4WFZnnzsccG00oFIxEN776r44wzzDsM6n6Gw+axjUY1zJghDtQ777hxzDHx1HNbt2qOmWmn9/r++y7ssYd4D2LwnfTASB1xVwbb8saDun41s+pyWaf/yTdYtY/kmimLLq+hXPvEqkGtvQxYBBiG7TFkXb8MeNR5haWvvnLjhhtEhkzXDSSTGn76SbyP5kYGtlYUaAgErCWsTz7pVZZNP2f77y9qOSsra1P7L+ZZNZeRx3TOHBduu81nCW7l94THow6wZN2OWMbIOnVNIqGlvhsyBSgFBSIQjMXSrxPZZxUAfvGL9M+F+t7V/RgyRLxGlsGKdRkANPh8BiIRLXWjQdMM7Ltvoun6tq5fBqvFxeLalzcZdt01gVmz3E2vF8v6fIZj+b28eSirYKqrxXFeuVL8lN/Xfr84RnIU7dpaYODA7JlVQFxL0ajYzsaN1mXs5clA5jmPzb7J8rXZA8pMwap83Om7RAxklXkkYSIios7CzGo78fnMKQ7kHXb1zrm9saI2qOWdeLXxIhsoaoZUPpatz6ps0KqlcNXVos+YfM7M0GhoaDDLEBMJDXV15nbkyJwAUg1CAPjpJ+Djj82/nfrqTZ0axHfficvtwgv9qcGT7JyCp/TRgK3P2zNnsrEJWDN2K1aIx+0ZsaIic4AVp6xwJAK8/LI17aGut7FR7acI7LSTSD++8IK76fXiua1bzcyq0/s0DDPwuOSSAB5/XGxTNn7tolFRYujziUzZK6+4MXOmO7X++npg82bz+Pj94lyL7LQMJvPLgttHQs2cWTX3MRfqcvbz6XQzRh7HTFUFMqiLxTTHAZYaGjTU1mqpgEUtfXcye7YLK1dqluNVU6OlqhOc5PLeZWZVvQklM6vyGKrlyvJYZM+smutWt6NKJJrPrAaDMrOqWcpnAREs77FHenmAvK4y3QQZPFhmVtVg1dweYF5jb7/dgIEDjaagz5qpl8FqUZGBd99149prxQEMhcR6ZSZUXb+dDEa9XjEllPzsixt0Zr9Tr9cM9ESwKs6704BI6o0HGSzLa0wdGMkp4Mw0wJL9PDc3d3GmPqvy8SuuSK/Fb+7GBRERUWdhsNpO5Dyr8ndJNkjsjRU1s1pWJhpcarAqGyj5ZlbNYNV8rLraOim9OgXGTTf5LGWwW7akZ1adt2OuTzZUTz7ZbEXPm+fCrbeKA/HCC57U4El2To1c62jAWlqAYn//sowPsGYCly0Tl7u9r6HMrEYizmWqb77pxqpV1o+KGgA1NmqWMmCZud6wwVoOKkYDzu19AmZQbS8ZB4DzzvPj3nt98PnM+W/PPz+Ae+7xpdZ/6KEF2HHHIiWzKm44qHPVtldmVQZYTjcunLap3oBIH2DJKdgUPzNnVtG0fefPR12d+Myppfni8fRtGQbwm98EcPfdPsu+NzRoqf6ETr7+2oXHHvOkBXoq2WdVvQnV2Khh61bzxon62VXLgDMNsGT21xV/BwLpI/omEmYwJUf9tZNT1wDAPvsUWp7zeoFdd01a+oha35fjwxg3Lolhw5LYZ5/0gYV8PgMul5E6d/J70+tNL/uuqREj8qo3/Q47LJ6qWJFVHvL1TmTwJrevjhRcUGCkvve8XgO9e4vfk0lxbRcWOn8fFhSYx1V245DXmDromVNmNdP3q9x/eQwyDbBkf192gYDI6F9wQfqHRv4f4LRfREREnYnBajtRM5lq40L2ac2WWZX9tNSSRLkOtVErH8s2wqxaBixVV1szQrKhGouZDeRLLxXRgyhjtg7EpBo/PtH0WrXsT/z8v/+L4G9/M9+YPQvlPH+m9b2MGVOI//7X3HAslt6H1J45u+suH2bNEm9cnVbHKXDy+cy5JDNl1e67z4e5c601eur5qqvTLGXAso+h3E/nzGr6vtgDCrk/9sxqMgm8+qqnaf9FAGrvSxqLaam5GGWwZ5YBO/ctzIV6PMeOTWQcYMnss5r+3EsvufHXv1ojCGsZcPN9hs3MqvmYOuWLmVl1DnarqqylnuZ6xc/Nm83+uZs2adi6VcOqVVra8SouTt83aeZMkfHLVEUg90/0WbU+vn69nroRoAaS8prweNDUVcBwCFbFT/l4MGikXW9qn1VNg2PQ7fNl60cpgrGf/cy587V9n8aOTeD66yPo18/AnDn1lr6l8jvN6xXrldeOfNztNhz7rBYUmAGhz2fg6acbU99z6mfGPpeoJINI+bx646WwEEqwitSgRbGYhro6cbPP6dio3/Xy+MrtqJ8Fp8xqphJ0mcE2uwU4LydlKgNWK2PSnxM/5f9bapk2ERFRZ2Kw2k7U4FDNrMpGlL3PqszoffJJPa68UrRq1MaF0+AastGiTvVgJzOeaoNNBKvqMuKnHIxmwoSEZR7AbJlVmT21z4Uq16tmOFwu4PvvzUtu7dr0IMIpuzh9unkwE4nM86xKDQ0aTjklCMNwHgRHFQjIQYean1ZGpQbINTVaKtiqrzf7vsmAyeyziqxlwPbH5LL2YFV9v16vCEDtwapzn1WnMuDs79NOZkzXr6/F+PHJtEwrIAIhp2BSuvjiAP76V2sAly2zmqmMF7DuvzptjZl51RxvQsjqAfvgN3V1YlCxHXcswnXX+fDwwx7su6/oZ7p6tZ52jtQKhZ//PIZQKD14k3O+OkkkzKlrVBs3aqkSa7UMWL5v+d0gr12VXEYet4KC9PMsglVzm07lny6Xc6krYAY1f/pT2LEcOJHQLKPg/uIXcVxyiXNdtPxekaXNZrBqPm8fDVgEq0bq+01+h8l1qecl18yqWtK+006J1DnxeIArrogiFEqkrncRrKYfGzWQlMdXXmPqDT3nzKrzsZ4yJYGFC+tw9NHxpm1kz6yWlZnPqyPKZ6uMkcGqzwd8+mk93n23PvPCREREHYjBajtRg0qnYDXT3fGhQ5Op59Q77fJOvLUMuPnMqlMZcCJhLQM2B1jS0NgotqWWnprBqtOoluKndU5HOQiMNZv64oseHHhgQervJUvSL7/mMn11delBpVMwAwAVFbpjMKXy+825JO0ZvWzUzGpNjRlsqdkZGSTKjIoYDVj8rgaXy5aJx+39R+X7LCiwPGwJ7ERpb3r5sn0gIECcVxGsmtO5ZJvH1YkogRQBcjBoOB4zdTRmp7JaJ9Zgtfk+qzIAVUvC1RsT6jqyXQP2MuD6ejO4ffxxL2680Z86fuvXa2lZ2kDAsGQGnT7Xa9ZkvjEgB1iyv27DBi010JccpVa8L/FTfmZlplz8bliWkUFu374GPvzQjWXLzPWofVYBoFev9H3T9eyZVfk62UdbFY9bv/cyDR4EmN9NXq/RlFm1Pi6CVWumvqZGBKv2yhH5fddcGXBpaRI/+1nc8rx6zUyYYJ1qbPz4JA46yHyfRUXO61UzprIcuLTUvPEnR0N2yqxmO0bq9DnNleqq1QLPP29+eLINHqYG/WPGJFNlz0RERJ2NwWo7cSrXBcwyYCPDzXHrnfn0x63rFT/V4G30aGvDUb1jrnLqsyozq4GAdZ9lFtipsSOfcyoD9nis27H78cf8g9Xbb/fhu+8yl+TusUcCL70kWuvvvedOBRr9+zuXtQUCZj/O+jySCeGw2V9xyxazz6rMFg8ZkkwFNvJnPK6lRiKW79MwgMMOK8Ajj3jTghm5P3KgF3Xbkstl9llVqeuSpd0ysBVlwC3LrG7bZpaQB4POpdOyry5gzQr6fAaOOMLcoPoZyDYasFMZr7zm1f1X+1qr5zKfYLWuLvOgTYmEORenJDPzgHh/TiWn1dU66uvT98MwxDUhBlgSr5MDA23cqKdKrGUfaMDaZxUwpyICzJJYeWy2bhUl/PvtJxZ4+GG1QkGzTI3lnFltvh+l/XexbvFPfVzt056+LjOD6fEYqSlY5DGxZlbFTxGsmt9v8kaaU+m82iVD+sc/whg3Lpnx+T32SKS2L28sqd/HhYWG4/eh0/e3eo39978NuP32sCWYlrJlPq3byJ5Ztd/cymX9cl8zTalFRETUWRistpNMmdUpU0TD0d5HbNq0Rhx/fMwyJYw1WJXlvOZjot+YOSDJbbeF8bvfWaMPp9GAAWvGUx1gSWZW1bv3TpleSQZSTmXAHo+RdbRUOeCRKt/RaQFrZnXUqCSmTEnA7TawbZuZVRwwwPn9+/1qn9X8MqtyICy1DFhmw4YOTaaC6GjUDEJkJlM2CsNhEcRs3Jg+r2qmzKpaBhyJiCDF3shUA7eqKg0ul9FUZmlYSqmdbg588okrrU+ptG2beQMiGBRTjdi3rQ5Uo5aoDxhg4KmnwvjjHyNp27ZnVtVA1p5ZNQzrAEqSOgq0ei6zjZbtNMCSenztN1uWL9ctn6VAwEj9nSmzCgBr1+qW/r6Aec48HvN1xcXiMzV3rp4avVoNVtXRgAFxo0VeN/L7or5ew5VX+vD55y4UFRm44YYoRowwS7ZlX3H1+2XixAT23996MYh5eZ0/v+rj9mC1sVF8H6jHKVtWz8ysWgMqM7NqpAWrALJmVkePTh9tWKXuj/r8yJHidePHJ1LftfJ4qWXNPp9zkKsGkvJmgNqHdOhQA+ee63yHKPdg1fr38cfHcM01DvX2aevP/F08ZYq4GDl1DRERdTUMVtuJtc+q2QC48soo3n+/HmPHWjN9xxwTx8MPWztfOjWo1MaQpomGiwxqdN1acqauw95gU+/sqwMsOWVWsw2w5FQGLAM3kVlNf43k1Jcv3wF/AGvwpk43EYloqQa6PbM6YID4OxiU2Ubnvo2Zt6mhTx8Dum6gpkZL65s5eLCBWExLBYb2gXxkP0NZJltbmz54T+Y+q2rwosHtNtIGq1KD/spKzXLDQZ3+x+l4n3BCMK1PqVRbq6XOqcyq24NJOfrt0KFJS2bVPDfip3rM7JlVa59ba7B55JHBVL9g9XVyyiWxjtxuPKjnpX//ZFOwar52yJAk/vCHCI49Vlzgy5frlpsHgYD52fL50ks0Zd/vNWu0tMyq/MyomVWXy0D//gbeftsDjwc49dRYKtOovi/5mfV40m9ULFum4cknvZg/35W6KVZUZE6l5ZQp/L//i+KFF8SJ/PnPY7j88gimTo3nlFm1Z5Pr68W1nKlKxE5uQ507VqzXfF5+XtTzHQik91Xdf/8E/vGPRtx4YyRtPapMwerVV0ewZk0tiovNcyJvnKjvweNxHmBJPf8yWJXfkRMnZk9bNjfHr2TPrO6xRwK77tp8SjTbObj55gg+/bQegwYxWCUioq6FwWo7sY4GbD7ucgG77JLbSIu6LjJybrehlBpalwkEzMyqyyVG53z5ZTPqMgdYsr7OKbMq5qTMnFl17rNqjnRZVSVGwjUzq+kjAKtk5kjVkmBVDUxkIOHziX3atk3st32ETNkHTPZZTSSs5dSnnhrDM89kjl4bGkQjt1cvw1IGLMmsa2OjCJrtA/nI9ykzj3V16Vll+Vx6GbD5e2Ojtd+ik8pKPdVHTtdlsCrLaDMHdU59ReUck4DZCLf3IZZlwIMHOwer8lpUj5kMXEtKDDQ2WrPM9v2YN89sdavrmDvXheHDk037lPFtWahTr5SUiM+SGkQPGWLguuuiuPvucNO+aJbgrG9fQwlW08uA5YBLq1en9582ByIz5+Z0ucx9v+yyiGWAHMAsc1XnSI7FRCZa7rcc6Vi8J/FTBKvid6dgFRA3v1aurMXDD4dx9dXRrGXA6uP275a6OjGKuLUM2Hk9QHoZsHxMVpl4POZ5Vq+LQMBIZTvlOjQN+OUv45Z9csqAqoNLqc8XFRmWmw+Amlm1vp9cRwP2+QzMn1+H55/PfjcsW+bTug3769QB+DKvI1vm1uMRfVWJiIi6Ggar7URtLGUL2Joj+yTKxp69Yej3m9k5t1s0qNQSuFzKgM0BlkRjPBi0jpIqAx2nxo4MWOJxDf/3f35ceqk/1aBUyxudrFql4euvrZdgazOrsrHo9Yq5JevrRd82e8ZLlgXLPoexmDXo9XoNHHhg5mxFOCzmx+zVC02ZVWsgIvsAyjlY1RE6ATNgkOdu27b0MmAZSKYPsGQtcXUaDVhlz6wahjrAUubXbd2qYdMm4Oyz/di8Gan9lKWx8rqwZ6Q3bNDQu7eBoiJrn1V5rchgQM2SyUCrVy8xaJM1WLUeW7VBHo2KYKKxEViwQMfkyfGmfcots6pe54WFSMusDh4sPkvqta/+3r9/MvW3Uxnw8OFi/ZWVmmV+X9Ff1VyfvD5dLrPs97jj4mndBeSxMMuAzamIZAZWnU5Ifs6LiozUtSbPue7w7R8IpAdlTrL1WZVBubUMOFufVfFTBKvid+uo50ZqkCnr9C9qn9WMq1duBDiXJav7L8cUEMuLn07BfaZgVf2eUbt0DB5sWNbtJFvmU2W/xrxe8zNhr8JQ5Zq5JSIi6koYrLYTtbFl7y+ZD7dbNGJkQ8teAubzmZktuYy9UQWkNyjV4EkuE42KRn4gYKC8PD0zbJa1pQey0ahoJFdXa6kRTjVNNNiefLIxlWmUdtghAcPQcPjhBVi50mzVZcr02af6UanBhdrYjUQ0NDSIUUPtr+/bV0w9IeYeTR9gKRrVsjbuGhtFY7x3b+fMqhxNs6FBrEs93v36JVOBijx3tbXpwapkb4DapypyGg1YVV+fPleuDBTnz3fhllu8SCSA88/346uv9FQGdMsWDf/6F/DGG57U4Dy1tWYAJINoe2C4YYPWFMRZ99vMrGYuA+7d20BDg/U6sE9N5PcD554bxSGHxBGNapg8uQDDhhUhGtWw447Jpn3KeDhSAgHDMrBQMGik9VkdNEgOwmM+pl4X5eVmn1Wfz0j7fBYWiufDYaQym4AIgOR7dLmspfbPPdeIm28OY8CA9FJTswwYTT8NS1k3YJ2CRQa7BQXmjQOzz2rz30tOWUm5XcmezZOjMuc6GrAZ7KsVJOY6p06NY9EiF+bP1y3XhayKUNeRbf3qCOeZg1VrNwvAuY+v2y32136+1b9l32z1uzSblg6w5PGYmflMgyvJ5YiIiLobBqvtRG1sOfWZypXMrMqGjP2uut9vpBrmMphUsxj2AUgk2QhXl5eBk8wOyAyabBzJgEwdhMPvN0tLa2tF0BaLaZaG19SpcYRC9mDV/FvNvuUasKnUMlE1kywyqyIIURuqgGjEBwJmZjWRsAZdkYg1M/L++9a60kRCBLO9ehlpmVUx9Y/YX5llUoPV0lJD6bMqHtu4UbOM1qqylwGvXWv92DqNBmwnrxunoOHvf/ehulrDK694cOqpwVQwWlOjoW9fsYzMbDn1WbUHhhs36k1BnPXxXDKrJSXNZ1ZjMXHO5DleutQ8HvI455JZ3WmnpOXzVFjoXAYMWIMb9fOlvk+fL/3zGQyKaWnCYWuf1VjMOsCS/I7QdWC//RL4/e/FAcg0R6haBiwGRjPXvXmzeTysmVXxWKYy4GzbscuWWZXTN+U6wJI6n6lTZvX002MoKTFw772+LJnVzN8Pcv3qDSs1e6y+Vv2ekcs4B6sisB440MCeeyawww7ioKrn/4QT4qisrM0aQALA0UfHmtaZfTnJqQxY9qtlZpWIiHqaDgtWQ6HQyFAoNLejttfZMjUy8yX6jZkZBHsWIxAwg0y1NFBSgze18Th4cHoZsMz8yEbdAQeICEg21mSDWA1WPR7RWI9GtVSwGo+nZwnsDay99kqk+gvKwET023Q+DtlK6NSgx+zzJoIOkVlNn9ewpMTAsGFJDB2ahMslAmw1s2ofMMmpn7HaZ1VdPhAw582UpZ9q38h+/Yy0MuDKSh2vvOKcWikuFj/leVEz0YBoVDsdN2v2S/zMVI4p918Eo2ZmVW67qkpPDcBljgYsnlMDpWRSBI9DhiQt18AJJ8Rwyikxy75YM6viGi0okJlV8zl7n9VoVHwOxHVnfU6O7ptLn9Xx4xPwesV0Tw880IjCQvFZkjcejjkmhkMOEZ8BTTOPpz2zai0DTv98+v0is6qWAcfj6gBLRmrd9gAyUx9EeR7lSND261VSg1VzgCUzo9ucTN9j1j6r1n2U0/uoAWe2QMnMEpvvy16ae/HFUbz7rhvffmvudCCQPhpwtveQW2bV/F0Gq7K82hrgivcXDBp4880GHHJIerCaq4ceCmPhwjrHsmwn9u8ytbvFhAliP154oQFffWV9HYNVIiLqjjokWA2FQr0AXACgrplFewwZGNqzYvlyu41UdhVw6rNqHWBJ/SlfDwD77pvA0Ueb6TenMmCZ+ZGZ1fvvD+NPfwpjwgQRaMnySLV0UmS4RMN72zYROEaj6ZkOewOruBj45z/FCqNRDTU1wMiRRXjtNeeALdt8rSo1MxONahkzq716Gfjf/xpw+eXRVHbKmllNz8z99a/pozX37m00TV1jPh4MmkGLLIkMBMR7kBldM0BvPgMoA46SEgOFhQZWrbJ+bF0uIy37CIgpM+T1J49/pgarGkjJzGlNjRkMbtqkpW5mmAMspWdWKyp0bN2qYc89E5abIxddFE1Nj+FUBhyJiAF5gkFzNFlJHkNABMPxuJa67tQbFYA4r16vYRlBFzCnDlIde2wcmgbMnt2AU06Jo7BQZB9lcHz99RFL8CKvLfUY9u1rDqokBliybqOgQNy4aGwUg49JM2e6U9MLietP/p4eiDixZla1jNPzyKxeUZE4VuvWafjPf8SLcwmOMmdWnStHfD4Dq1fraY9nC4yd+qzag355o0OVa59VM1h17kOrBqvq94wcvMmpbNrtFgH0tddGmpZ13u9c+HzmgG+5SO+zamDMmCRefLEBt98u9ueAAxLYc0/rcrmWGRMREXUl7RKshkKh80Oh0IfyHwB/RUXF1diOglXZULNPWZIv+wBL6X2k1D6r4jG1gSkfO/5469Q41mkYxE8zWJXZGOB3v4ulGmJTp8YxaVIc110XsbzW4xGvTSTMzKq9kZteumaWT8ZiwEcfZb/tn628TWVmWoxUH9yCgvQ+ryUlIrAQJdZiHsdsmVUA+PWvY3jpJTMy83gMFBWJ+VztZcAy4JdBoN9voFcvMSqxnBuzri69/6kTGRx6PGJAH3U+0RtuiMDtTu/XCYjAzZxP1np87NRgVb5myxaz/2ZVlVnGas+sqkH+V1+JDey1V8LSOFavW6cyYJktDQSMVD9fadMmDcuWiRFvZfDs9ZrnWFVSkl5+LB63/v3OO/WYPNk6upTMPsr34zSQDWBt9Ou69QaJ/TWi1FxkVtWbDBdcEMDPflaQWt+oUUmccEIMjzxiPZGZMqtq4ByPm+fffpNIZvDl5+f444P485/FTuaSWc08z6r5u3q8+/c3sGZNywZYEn1A09ep7r9KZFbTM7F2Zhmw+Zj63jMF3gcckMBRR8Vw663htNe43cCkSQkceqg4wDKgtQ/k1h7s/wfI79r99084djmRFTK5DuBERETUlbRLYVBFRcW/APyrPdbdXchMl33KknyJwZUyT0EjSgwzB6u5lH7JdcsywUwNrpIS4LXXGi0DxcgM1+bN4rWRiAg07I1Hp9I12RBubAReey37jqoZjy++qMP06V784x/pLVSz4Sv6CdbXi+yW/T2ppcxyRFV7n1Un6vGU5XeGYS0hlgEKAGzdau5PcbFojHs8wOLFLowcWYTf/tYWbdlompF67x6PKDv9/ntxoh99tBHHHRfHhx8GHINVObjO+vXZ58oFrNlL2bivqdHQp494TGRWZbAqHjMzq+Zr5851oV+/JEaMsA4OZJ1KRPy0ZlaRyqyqU9f065fE99+7MGlSIS6+OIL/+79o0zoMeL3pg1L17i3L3a2ZxoEDk6ipMVvrQ4emfy7HjEkiFtOwaJEIKp0GsgG0tEa/es3Zr3NZEh4Oa6nyWDt5Q+qhh9JPYnN9VuXgYPL877xzEl9/be6gPD4y2Fu2zAyYcwtWnR/PNM/qgAFJrFkjj1/6/jqvy7y+Bw5Mpq0TEJ8pTbNmzP3+3PqsyvegVrlkKgNW+f3Av/+d6QafdXvypklLMqv5sh/L5rqcPPlkIzZs0Cx98ImIiLoLDrDUTkaNEo2u//u/DFFPjmRWNdvUNZLMMuQy9YRKNuZkds3e4Lazzxvr8ZjTbUSjWk6ZVZEZE79ffrkfr7+evUZNHaRk6ND0+Sx32SWBX/0qiuOPN/tFyqlrRBmwdXk1WFXnWR0xIomCAgOXXWYGkX36mP1N1VJAEaxaB1ICnPus+nzAL34Rx89/HrMcmzlzskcMfr+a0ROjNMvyUXkMXC7nsuVevYARI5Kp9chlzXWb72XbNvPxykpxAamZ1YYGLXWOZebVaeqazZs1DBwoRoLOVioKWPubylGb7X1W1ZG0//53XyqolteP+r5dLjE9iFOAdeONEYwda2ZSnTJ9sr/fZ5+50/ZZbhMQ5+Hvf2/EP//Z2PR45jJgWRLe2AisWaNb+i5L2QItp0BE1w1lMDVx7cobVjvvbM0Wy/JSeYNBlctowOqxnD3bvCNjzUyavw8YYJYBq4/nWgYsbyLYR7fWNKSV8qt9VnOZusZaBmw+n2t5bLbvVVm23pI+q/myl283t//BIDByZPsH0URERO0hr8xqKBQqBvAZgKMrKipWND12GoDrAXgB3FdRUfGPTK+vqKg4uuW72r2UlhqorKxtfsFmyKxqpr5cTpPQA+I18Xh6Fuj++xsxdqy1wSwbczU14mdzpWxOd/ZlIBOLiX/2KS/swaI6wvGGDTr69hVZLTXoU6kBtNudHkj072/gnnvMGwOiRFRTyoDN/bj00ihGjLAGnbGYhoYGoLw8iS+/NEf0WbasNmMjVR3FVS2j9fnM/b3vPl/TYwYuuUREZxdeaJ6UefOyB6s+n9mYd7ut02DIx12u9EGIAJFNGzUqibffdu53FwiYGTk1syoHcBKjHJvre+UV8ebNPqvicbW/rMyQAtZGtFMZsBpoyvlwg0GRqZZZ3IEDDSxaZK5n8WI9tW6v17BkVktKZJCcfiwOPjiB1atjuPJKV9NxSF9m5EhRqi2zj/bAQx3A7OSTzU61ambVqa+23w8sX64jHNawyy5JbNpkXW/2/pbpQYa6vLhRYWap1YHAnniiEYcfLvbTqc93Ln1W1e2r8zerWTr1s6j2vcx1NGB16pohQ8Q2Nm5M/x4oKDBSXR4AcX5ymbpG7kemMuBMpdZ2TgPXSbJsPdd1tYY9Q5ppeiEiIqKeIOdgNRQK7Q3gUQBjlMcGAbgNwAQAEQCfhUKhDyoqKr5vy53s27eZ2dR7MDHCKNCnj4ggy8uDKC01n5fzeQJA377mc7IvW//+hZZlLrkkfRuysdnQIFp8gwZZt5FNaWkRPB4xWiwgAhBN88DvF8+Z+2Z/XQDl5ebfO+6oY+VKa4ZPVVhoRiClpUUYMsT+vNuyveJipOZO7dfPi0GDvE37oeHWW30AzBZ2UZHMTrlRVmbdb/txUP8uKfGirEz83thofpQKCtwYPLjQ9jrzmGYb2dguGNQwYEBh0+8u7LCD2WLu31+sMxh0Llvu08eDKVOAv/4VWLHCg9JST6qsV+ynhi1bxO+xmBmZVVeLc7l1qycVzA4eDMyYIY7hsGEFKC0V02WIgMeH0lJxPA1DvL/S0iLLdTd4cFEqqBswQPz0+wOpYxKNikxwWZl8fyKyGDnS+hXV2Cge79vXj61bre+7Tx8dpaVFjtkt+/4MHFjk2L9vzz2B994Tn7vycms6Ut7EKSiwXmsya9m/fzDtpsHgwQUoKQFWrRJ/jx3rwpw51mXUz62d040jt1tLbV8O1uX1iuM0ZYr55n/9a/PF9s8LAPTuLZ5X34udWt5u/VyYv6uf41GjzM9p797m7/36ZX6P8posKfFh/Hjxe2WlnrZfxcVAZaX594ABgdRnqbhYXN9O5HdP377m8/37F6a2q1ZtZDsWvXqZv/fvX2h5PzKQ7ds3kPN3Z0tNngwceKA477NnA2VlBRm3me39UPfGc9tz8dz2XDy3LZNPZvU8ABcCmK48diiAWRUVFZsBIBQKvQjgRAB/brM9BFBdXYdkcvu8e6xponHe2BgBEEQ4XI+qKjPDYRg+iKQ2UFvbgKoqUQbochUC0FBTU9vsHJyiHLOo6bUuhMN1qKpq7niLD1xVVS283qJU38xkEti6NQ5d11BVZdaHJpMeAGZDuqGhAbW1SQCitanrcbjdOjJVpkej0dT7rKqqxVFHAS+/7MLxxweb1h9DVZXZv8ww/Ni61YV4XIemRRCJxAAUIhZLoqqq3rZuL+JxL2pqDAwZkrCsx27bNh1AQdPrIohGkwACqK4Wx07sSxz19Y2pYwQA5eXmMY3HzXPWHK83iW3b6pvWFUcwGAMggozGRnEtxOP+pmDZno2KYvjwKIBCrFol3ndDgzv1eq83CXm81683j680Z46BXXYR5blHHBHDY4+J52Mx870Eg4XYtCmGqioRNdbXB1FSYqCqqhHRqBfypkBNTW0qIyRGQC7Epk2NqKqKNz0fRFmZgWRSvL/Vq8MA/OjdOwL1xsLSpeLvSKQRsZhuea64OIGqqga4XEHIcwEAQ4emv/ctW2ods6t9+/oBeODzGaiqso5+petivclkHFVVZlSaTIrz2dBQj/p6PbUNAGhsrIOu+wCIQGnAAOv7EcfM/Nzayc8mYFZLuN3mviWTAYTDGn76KQbAD4+nDkAhJk2y7mM8bl635nYbAQRQVZW5AkRUOojPqFiuSPldaGgw151IiPMmfjffa22t9XvL+h7Fd0MsFkFhoficRqNI2y+fz3peI5EGRKNuAF4kEtHUNZhp/ZpmXuM1NbWpwafEz/T3ZadePzU11oqL2lpx3TQ2mtd0e3r+eeDWW72YPduHujrnY1taWpT1/VD3xXPbc/Hc9lw8t5npupY1MZlzsFpRUXEuAIRCIfXhgQDWK3+vB7BXfrtI2cjBVyZPTuD66yPYbTdroyRTGbDTNDaZyBI62RfS3jcs19dL9fVaDqMBW8s1vd7sJXT29QUCwG67mQ18+z54vUaqtDUYNAdYSjjEBG43UqPzNjfqsLodMYem+F0tA3a5DMv7XbKk1jIabT6jcgYChqWsVs4jCqhzp4rpXOz8ftHH9bzzojjmmPQRQdVrRy0DlhoaNHz4oTgvQ4ea150sAwbEsVX7rEYi5nlUyxOdykatZcCizFNee7IkXQ64I1VWitfIMmCVHFBKnqOhQ5PYZ59Eqvxafe+ZSmD79DH7n9qZ5diG4+Neb/q5FX1Wzb+dpijJdj3Yr7faWmupsQxgX3zRg1AogSFDDPz4Y23a582pDLg1Ayyp1Ay1U7m32M/Mr1evb3VKrWzbAXIfDdhpGjGn78rm2OdZVcly9I6cHsacn3b7vJFLRETbh9aOBuzUwdD59jm1iBwNOBhEqtGtyjSXoWhQpweNTjRNNDLNOUHza/zYG4r19da5WJ3W6fFYR4v1eg34fJmHq3QKLpym35HkPKuAKPOT+5gpWDUMDdu2WUsCnaiBiuiPaB1JWe6Xuj/2aVPkORk8WIyc6nYbWL26DgMGpJeHyNGD5evUYFVuO1Nj2+8X5/a228yMU6aRotVgGwB23TWBb7/V8e23GgYMAAYPdh4sSc6LKkWj6X1Wd9rJetCdBliqq9NQWGgOhCX7Lg8caL1uZLAqRwNWyUGz5Pb9fgN//7uZJTf7NxoZR0aVU00ZDh8Bp3lW1cf9fiPtOg0GDUsA17dvfgMsqfsZDIqpdexB97JlOhoaNNx8cxiaZi1XlZxuwuTSZ1XuW/apZ5wDVPU6yd5n1by5oWnA738fxZ57pn9Q7fuQ62jA5jyrue1PJtkGWIrF5E2Ujgsc1ZskREREPVVrRwNeC6C/8vcAAOtauU5SiJGAc2soqoPn5JNZBUTDUo4omu9cgenBqpYWPNrXKefJVLefrdHlcgFvvlmP//3PLOG1DnbknO0CRCNfzjl6yy3ppYJyXxsbNccMlH0/JDWzKo+dfZls6+jTx8CrrzbgnXcaMr7G7UZqZF2v1zmzmi1YTV+f+Xo1ILNnVktKjFSWy+tFauAbuz59DHzzjQuTJwexYIGOSMSctmj5cvH1cuSR1rLIbAMs2eenle/32msj8HoNVFVZRwNW7btvouk552l6nKZ2spPBpBqAS2bgZn3czCSnnwv1GvH7DcebIblm42SwpS7vdptTB+2wQ+b7hMGgmPpFlct3Qz6j7dp/Vz/fma4f9TVyGzffHMFRR6WX0jpNh5XL/pnBavYBn44+Opb+oCLbAEudkVnNZXApIiKi7q61mdWZAG4OhUKlAOoBnADg/FbvFaXIMuBMMk0P4XZbp7hojpybUtPSp99ojr2x1NCQ/ti4cQmMHp3AkiWu1P5ZM6vZ5ygMhRLYc09rgzfb9BNqGWdBgcji/vSTtQ+ipAb5zZUBq9tUM6vW9Zm/779/eqNbrsPvN7DPPs59Fe375vGIQFNm/oDmg1W17NFp35LK4ZR9jtV9LCoyUnO0qplV1VFHxXHLLWJHnn3Wg2jUPPYnnxzD3LkunHOONQgwy4DFT8NQy4CNpv0xM6hyVO1HH/WkptWxXy9PPtmIqVPFsTbLI53fe7YgTZYBqzcfpEzrNUezNUukTzwxhiuvjMDlMqsKiooMx8Ai1xtK8nxmCpqyfU/ouhj4Sp0jOZepa4JBYOzYBK64IvN8wGpQqp4T9XtEHdzKLtNxtXPqCiBl+56T16P6ebAvv3p1bbPZ1q4WrMrPSkfM7UpERNRZWhWsVlRUrA2FQtcB+ABi5IrHKioqvmqTPSMAwDnnRLMGj5n6hTUX5NrJ7E8gkD41QnPsGY+6Oi0t0zlqlIHZsxvQv38hkkmtaZoP83mfL72sU3rjjfq0QBUQ++lyGUgk0jO56t/N9cFVj5PTfJSZlrX3R7Qvs3JlrWPjVWY37a8988wonnrKejBlA1keL3X7spHqlHk//fQoTj01PVOUOVjV0paTI636fGYQZ3fSSTHccYcX8biGtWs1Sxnwzjsn8dZbDWmvEaXtRqoMOBwGkkkNwaB5rmQZsHr8evc2sH69nLrGsIxs7NRXMj2zau3T6iTT+7Su1zmL7/MZmDw5gfvuC+O442Kp4yfPczDofK5yz6zK/Xfue9nceoqKjLRy9eboOvDRR+nnUJUps5rrTS8Z7DY3BYtTZlVew9mCVacyYLtc9lXdhv07Ul43zd3saksnnhjDkCHJtG4GREREPUnewWpFRcVw298zAMxoqx0iqxNPzD6ypNrAUxtT+QarMvtgnw81F/ZGcjicHjxKXq8ITtxuw7J/Pl/mAZb22itzCaHLJfqh2l+rNmydMoyZ9r+5xqa1DNg5syoDkkzl1HId9ufvuiuCP/85guHDzYhZnlOv10g7n9kyq/feG3G86aCuQy0DTg9WzZJVrzfzDYzycgPvvNOAu+7yYeFCl6UMOBufzywDliW3BQVGqgxbzttr7/v7449a6vVqllk9ljKYs98wMedJzXyO1TJru0xlwJMmJXDEESI41TTg9NOtNwnkNeLzOWdWs5X5q2SwZRjmychWXWBnv7ZzrbpQ9eplpJWMW0t/1f0R2+vTJ/swBk7z8jpx6rMqr+HswWrmQbPykS24v/feMA4+OIHx4ztuyIaSEuCww7JXZhAREXV3rS0Dpk6WrQw4n1FnZYMz1/6qn35an8qMOQUnmQJlj0cEqzIA8noNRKNa02jAue+vJAdRsg/opDZMm8usylFkgVzKgK0BklNmtbkgQB3ox/46+77Kczh4sIHBg60NYXnc7edZ1zMPIJQps6r2H5X7IgNHGWw98USj4+Bbu+ySxL77xvHOO+Jg5BIUeL1igKWNGzX84hfioissNFLB4vr1zplVyeOxnjd1v+Rxsc9nm0uf1dwyq9bHJ01KYNKkzEGDvEa8Xutre/c2sGWL1nQN51KSK5ZRBwmz9lPPvg571UA+3w/SggV1lusGcM72A+a5ay6gynWgIPvzXq95DWd7LzvvnMSZZ0ax116tC+yyHd/evYFf/Sp7n1ciIiLKH4PVbi5TGbDIXOZez2sGq7llH8aMMVusslEqS3Lt+2XdjihFVPupyX6O2UYDbo49yFC3ny0AAYABA8zn8ysDNmxljwYikeZHYM61jx5gBr6vvJI+CJMMSO2PZ2u4q8G2PFfm+oxU1s7ttgZZQPpASaqhQ51HhM1E3KQAnnzSg59+EjtcUCC2WVRkYMMGs9xXUoNTr9ewjYxsrltOwWTvD5xLsFpcnPk5+bp8R5KVnymfz3rOL7ooiltu8aG0NL/Mqhos5lMGbL8R05JgtbmbWernIRRK4rnnGjB5cvYgMVPG2s7+vKaZgXu2G0TBIHD33c5zsOajJceLiIiIWqe1owFTJ8s0GrCu5zaAiiSzYfmOBCz2QfxUg8JMUzjY5wZUsyrN9VnLJltmtflg1Wz9N1cynG2AJZld23vv3AZNyiXokQ1kn89cftCgpOMyTvtopz6nBn+ANRMp+qzmXj6plo/nkiGXZcBqKa889mrwli2zqo6sq56HL74QB+TAA63BdS7BZi7lpPkGq2bQb1huFpx3XhQbN9ZajkE28hirwWquAywBbROsZhMIGJZrxe0GDjoo0ezNi513TuLss6PYZ5/cuzxI8oaLrrd/X9GWlE0TERFR6/C/325ObQja5wHMr8+q+JnvHKuAGVComa5MWR57VlENWltSBizZg9VM5dFOysvVzGo+wao1ozd1ahzz59fh+OOzN7ozlQE7cbrh8P779fjgg3plGevz2RrV6rJ33x3G9ddHUiPY2kdLlccil/Oili/nctPB5xOZVXVf5fb79Us27athOY9qcO3zWfvRqudBZoDHjbMG9XJbzfURveOOMJ57Ln1QIfs1mysZSNvLgLP1BXYij7FaBpxPn1V71UBbBnivvNKA2bPrbXOr5rZ+vx/4y18ijvPDqpy+z3IpA24rzKwSERF1PAar3Zy1cWj9vWXBav77IAMKNUOUeYCl9Iyq2L7zlDm5Dvhkz57KddnnlnSiZujs/Rzt7AMsqX/36pV5ihdVPuWkToFnnz7WQMwe0OaaWS0rM3DJJdFUMKUeB3U04Fyo5ym3MmAxdU1dnRmtye3LzOqIEUnL8bVnVjNt/8EHw/jpp9q0YyeDp+aO+znnxHDQQenZcbPcPfvrM7GXAeebqXPqs2q9QdVcn9X2y6zuu28CgwdbP8NtHdzJ6+r66yNYskTMwZPLaMBthcEqERFRx2Ow2s1lmsZCTA+S+3raYjRgNbOabYAl9adcTkzNYt32Qw814ssv65EL+zyOMijO1gfRSXMDLKmZMHtg36tXbsdOjmCaKaBfurQWf/97I4DczmGmKVqcWPs4ygGUxN/qe3e7jdTf0cxTbKaoWdlcy4CjUQ11ytS39jLgHXawZkbtfVZVamZVzHeavs2W9jmV5LHLdy5NOQiY04jO+XDqs6oGqM2tuz2DVUktA27rOUcPO0xkzKdMiaema2GwSkRE1LNxgKVuTg0MrFPXOE+T0dx6WpNZVYPVTKWgou+lOVqtOq2ErltrInfZJWkp0c3GHijKfbL3y2xOPqXI9pGAc91WrGnQ0EzBRVGRuf+5NMLty+Q6wJK8PszMqvXGhwxW4/Hma1XVrGwuZcBer5GWWZXXnsySjxhhXY+aWbVnb3O51nMZYCkX+b5enp++fY28S4hV8vyoA2PlUwasniOgvYJV8/fWHme7Aw9MYO1a69zFHRmsss8qERFRx+N/v92c2nBrTRmwDFha0mfVqQw4W2bVOhejuY70gYJy3xf7a2Wjubk+qHb59CG0L5trZlUGf9kCF1nq2bLMauZlnUaPVW9UyLJpXTfLgHPJrOY7wFLv3gY2b9ZQWyuOxRFHxFI3O+S8q/aBh9SbAXLfjzoq9+lC5HtvaZBmjr6c3zV1xBFx/PGPEdx4Y6SVmdX0MmD7dFXZ2D8L7RF8qfvQHsGwPSDPZTTgtpLveSciIqLWY2a1m8s0GnC+ZcAygGxuTlInsgGpBheZ+i3aSyHNPqv5jWrbHNmIzTXb+cUXdVi1qnUtXlma2ByZWc2WCcsnY9TS0YDNzKr8Kc5NLGbNrMZyiAetAyw1v3xpqYF580QZcCiUwFNPhVPPyfLf8eOt/UZlZtXtNlLH5dFHw4hEwsiFWcbbsqBDBqtGni93u4HLLhMRf6QVM6g0N8BSZ/ZZldQbOPncbGopeS5YBkxERNQzMVjt5jINaOJ259tnVfxsu8yq83pEZlXNkJllwG0ZrG7bJlrNcqTb5owcaWDkyOxTzjSnrcqAAVECDYisXHPyGQ1Y3aZcTmbV5fQ4sZj4Ka+JXIJVdR9ymeqmtNRAdbWGrVu1tIGczjwzht12S2DXXe19VsVP+9zCuV4nuQ6wlIkZrLZ8PuDW9ON0mromn3lWOyJYVbV1GbCTfCoQWovBKhERUcdjsNrNqQ1UtTF11llRS3/A5phlwC3fB3V+zEzZNfuIqOqowPay2lwbu04B6UEHxTF1agx//nMrUll5yjUQiTfFn9ne39ixSaxcWZvT+bCXJ+ZaBiztsksSn3wCDB+eTAWwum7eSIjF8gvOcs2sJpMaVq3SMWyYNSjVNKQFqoBZZp3L+p20tgxYTvWSb2bVuo78X+NyGUgkNMcBltRz39znZdddk5g8OY7Zs91pr20PHRHcySy8fe7h9pBPFwEiIiJqGwxWuzk1i6U2Dg8/PL8soQwA2iqzmnmeVcOxn60oA7ZPwdL8vixblj5FCSBKJp98Mrfy0I4mg7/mgotcbxzY15PtuDk9d/PNEVx9dQR+PzBtmjiZLpcZHA4enF8gkMsAS2VlYpmVK3XsvHNu16rHIwYZamkZb0tH85VaWgbcWm63yCA69VnNZ4Cl8nIDL7/ciPJyMeFqe5fOtvVowE5+//sYJkxIYp99WlcVkQtmVomIiDoeg9VuTs0ytaYxJYPelmRWx44VAc3AgWZQs9NOzo3HwYMNbN5sLmdmVltWBpzPXKBdhQw22qpPnz0zmX00YOfH5XkX+6TB5TIwapSBl18Gdt01v6A/lwGW1Cx8PuewVy+jxRkuc+qaju2z2lryfMrMqjo6c74DGqnHrr2Dr44I7nQdeQWqr77akFNZuxMGq0RERB2PwWo311bBqhxkpyWZ1UMOAebPt86Husceztm4G2+MWMoYZZbM603P9HREn7e28PnndWhszD2COumkGP71Ly8OP7z5/qi52G8/a2M923XQXDbNPr3LL34BVFXltz+5lOmWlZkXQT4jNvfqZaChoWXRqnzvrR0NuKODVbnfTnMgq+8l3yC+vYOvrjjVS2sysBwNmIiIqON1k3CAMvn/9u482q6yvOP49yaXBEOAoBAmC5bpSYwgDihQEzSwFEdEoYuC80BtHepYRaxAga6KIqUFHBAqxqAuAm20dZaUIghSq9aKPFQU0ATRFhRIIHP/2Pvc3MR7ucM5OWefN9/PWlnknH3PvS/3l33OfvY7tXOxOlw7+6yOZLRCc8vnt9ZqwN20//4bgfFfyB5yyAZ+/esHO/bzt+yZnEzP6pbH2yk0xrvAUktr1eHxmDVr49Cc34na1LM6ude3fifdLla3334jDz00MOKw2nYKTnsKJ6aJxbckSaXrk3JAW1trnuFIvTcTcdllD3PQQeOf47hpn9WNfdOzunTpqnHvqdotS5euYsmSQRYtmjburWse7Xg7hcx4elaHF9gTGQZ8yCEbWL58cndlWj1j7a4GvGHrr+WzmWuueZirrx7crMBvaWcoucXXxFjcS5LUfQ0tB9RtrWHAk9lndbgXv3hi3V6tInn6dDj++LVcc80gM2fCsmVTG3sx3Y3FXCbqiCPWM3PmRhYtmvaov7fxDgPe2sXqwACccsoaFi+exg47jL/gOuusya/u3G7Paku7PasHHrieo44a/7+hOXM2cPrpa4Yev/3tm34H7fWsNuuGS9NZrEqS1H0WqwI2reQ7fEXfbhjeszprFnzpSw/z/vdP5/rrvTKcqE29opPPsNVT14nFusZy/vmrWbhwPQsWdGbu7lg2zcft7QJLN9ywatKv3XL4eDuFt8XXxPj7kiSp+yxWBcCRR65n2bKVExrC2wmbitVNz82YsbHtHt5t0ZaLI01GJwre8e6DOmXKxHvi29Hu72fOnOrc6PY58mics9o9TR3pIUlSySxWBVS9RvPmdf8ivFWstoYhA5x66lqOPbZ7RUwpWj2G7VxUd2IY8Hi2rumFdocBH3fcOvbbbyVPelIZxarF18RY3EuS1H0Wq+qpxz1uAzvvvHGzAmL27I3Mnu18uolqFR+d6Vmd+Gu//vWVfPWrg40tgjrx+zn44OYUqtDMYcDLlq3k7rsb+o+gDc7xlSSp+yxW1VOve93arg4FLVknVvJtp2f10EM3cOiha8b+wh4aHNzY2FWmJ6Od1YC3VrE6b96GnozS2NrsWZUkqfvKu/2tvjJjBuy7rz0WnbBpmGs7BUz7Cyw12fHHr+PII8u5OeKc1e5p6ogBSZJKVlAfg7Rta11Mt3NR3Yne2Sa7+OJHet2Ejmonp4HJbVe7zSr1nJAkqcm8VywVotWj2ol5jM7P6w8lDWluuk03gzw3JEnqFotVqRCd6BUtvWe1NObUPa1z44lPLG8+riRJTeV9eakQ4x0GfMklD7PXXiP3Dlms9pd25idrYqZOhSVLVhW5eJQkSU1lsVqAH/3oIVau7HUr1Gvj3Uf0hBNGX2Co9AWWSuMw4O5asGB9r5sgSdI2xUudAuy+u70rGt4rOvl/D63v4cqn/cGbCpIkqWRekkqFaGeP1Jbx9s6qGSxWJUlSySxWpUIMDFQrlXZi701XA+4P3lSQJEkls1iVCjJ1ansFjAss9RcXWJIkSSXzvrxUkMHBdocBV8WPc1b7w2RyuvrqVdx6qwFLkqTms1iVCnLiiWtZsGD01X7H0ip0HV7aH1o5DQyMv4d1/vz1zJ/vqraSJKn5vCSVCvKRj6xu6/WdWKRJ3bOpWO1tOyRJkrYGx4JJGuLWNf2ldVPBYlWSJJXIS1JJQ1wNuL+05hhbrEqSpBJZrEoa0ip+nLPaH+xZlSRJJbNYlTTErWv6i8WqJEkqmcWqpCGt4sc5q/2hlZPFqiRJKpGXpJKG2LPaX6ZPh3nz1nPJJY/0uimSJEkd58w0SUNaxapzVvvDlCmwbNmqXjdDkiRpq7BnVdKQ1irArgYsSZKkXrNYlTTEfVYlSZLUFF6SShrinFVJkiQ1hcWqpCGtItU5q5IkSeo1i1VJQ+bO3cDcueuZNcs5q5IkSeot+08kDTn88PVcd52ry0qSJKn37FmVJEmSJDWOxaokSZIkqXEsViVJkiRJjWOxKkmSJElqHItVSZIkSVLjWKxKkiRJkhrHYlWSJEmS1DgWq5IkSZKkxrFYlSRJkiQ1jsWqJEmSJKlxLFYlSZIkSY1jsSpJkiRJahyLVUmSJElS41isSpIkSZIax2JVkiRJktQ4FquSJEmSpMaxWJUkSZIkNY7FqiRJkiSpcSxWJUmSJEmNY7EqSZIkSWoci1VJkiRJUuNYrEqSJEmSGsdiVZIkSZLUOBarkiRJkqTGsViVJEmSJDWOxaokSZIkqXEsViVJkiRJjWOxKkmSJElqHItVSZIkSVLjWKxKkiRJkhrHYlWSJEmS1DgWq5IkSZKkxrFYlSRJkiQ1jsWqJEmSJKlxLFYlSZIkSY1jsSpJkiRJahyLVUmSJElS41isSpIkSZIax2JVkiRJktQ4FquSJEmSpMaxWJUkSZIkNY7FqiRJkiSpcSxWJUmSJEmNY7EqSZIkSWoci1VJkiRJUuNYrEqSJEmSGsdiVZIkSZLUOBarkiRJkqTGGdzaPyAi5gHvA34H/Cozz9naP1OSJEmS1N+60bO6K/DuzHwLcGQXfp4kSZIkqc91vGc1Ik4FTh721EmZeW9EvBf4fKd/niRJkiSpPB0vVjPzk8AnW48jYvuIuARYmplf6/TPkyRJkiSVpxvDgM8Anga8MiI+1YWfJ0mSJEnqc+PuWY2InYAbgRdl5p31cycDHwCmARdk5sVbvi4zT2ujfVMBpkwZaONbqBvMqFxmWzbzLZfZlstsy2W25TLbkQ37vUwd6fjAxo0bx/wmEfFM4FJgDnBQZt4ZEXsD36bqNV1NVcj+SWbe2oF2tzwLuL6D30+SJEmS1CzzqWrLzYy3Z/WNwJuBRcOeOwa4NjPvA4iIJcAJwF+3187N3ELV8HuA9R38vpIkSZKk3poK7ElV9/2ecRWrmfkGgIgY/vReVEVkyz3AMybVxNGtZoQKW5IkSZJUhDtGO9DOAksjDbze0Mb3kyRJkiQJaK9YXQ7sMezxnsCK9pojSZIkSVJ7+6x+EzgzInYDVgIvB07tSKskSZIkSdu0SfesZuZy4HRgGfAD4MrM/G6H2iVJkiRJ2oaNa+saSZIkSZK6qZ05q5IkSZIkbRUWq5IkSZKkxrFYlaSCRcTUXrdB0vhFxEBE7NrrdkhSE1isakwRcWJE/HFEHFA/HmmPXfWhiHhORJzZ63ao8yLilIi4Fnh2r9uizoqIV0TEmyJiQf3Y9+RCRMRLgfuBs3vcFHVY/Z58akQ8rX7sNXghIuKkiDg5Ip5SP/Y9uYNcYEmjiojHAouBacAtwFOAN2bm3T1tmDomIt4GfAQ4NDNv7XV71L6I2Bf4J+DHwDmZmcOODWSmb/p9qL74eQxwIbAfcAPwdOAVmXlfL9um9kXEEcBpVJ+3G4FFmXllREzJzA29bZ0mqz5vdwQuBXYFrgVeCRyfmT/pZdvUvojYBbgC2B74IvAe4NTM/FpPG1YY7+ro0RwI3J2ZRwPnAf8H+KFZll2Be4H3OFy0GPsCPwfOBF4aEWdFxKsALFT7V53dDOCxwAsz84PAqy1U+19EzAD+FvjnzDwW+E+qz18sVPtbfd5OB9YBx2XmucCNVDcl1P8eDzyQmc/NzIuAvwc+FhH79bhdRbFY1WYi4qiIeHz9cB7ww/rvHwBeDry9HqbkMIc+U2e7d/337eridCVwEnB4/Ud9aIvz9sH6z6XALsDPgDMj4k/rr/W87SPDz1uq0S2zM/ORiDgduCYiLoqIF9Vfa7Z9pM5238xcBSzMzMvrQ9sD/1N/jddpfWiL83Z/YCbw5xFxDnAK8PqIOKlnDdSkbZHtU4EYdvhaYAB4cf21vid3gG+CGhIRc4BzgIUAmXl5Zl5SH/4pML/+70URMcNemv4xQrZrM3M98EyqId7nAp+OiC8NexNWHxgh2+8Da4H/yMz3ZeYVwGuBMxwG3F+GZXs0QGZ+A9gnIi4AdgLeQDXc++KI2NFs+8ewbBcAZOb6iNiuPjyN6vMWqiHB6iMjnLc3AX8HHEy1hsBc4CbgXRExtzet1GSMkO0VwOyI+HBEvBl4B3A5VcfOoO/JnWGxquF3fl4IzAEOi4hD62NTATLzksz8bmZ+HLgLOL4XbdXEjJDtM4ZlO4tqIY8Dqe4C7gL8MjOXd7+lmqhRztun18+dTzUciYiYlpnXUfXUPLfrDdWEjZLtYfVzZwNvAq7NyseAFcCJ3W+pJurR3pPZVJguBR4TEbt4sds/RjlvnwqQmcuoRrmclZk/y8wrqaZWPa8njdWEjPGe/DLgJ8AzgM9l5tnAcuobyGqfxaqGz2MbBP4G+C1wTN0Lsz4i/mDY6nUzgTuB63vRVk3Mo2Q7JTN/S/XmupRqYYCXAydExF49aKomaJRsn11nexuwY0Qcm5lrImJ34DdUvehquFGyParO9lNUc5KPhmpIP9WF0Xd60FRN0Bift+vqY1Oopmjs1P0WarJGyXbhsKHchwPPgaHi5zdUC6Wp4cb4vP1ePYT/tMz8Sn0N9Wvg5t60tjwWq9uw4WPpI2I6cENmXgDcTjXH4jn14SnAJyLiH4FvU93Ft/etwcaR7bPrw28FDs7MxXXv2/uoPkDVUBPIdj3woYj4DPAt4A7gt86haa4xsj2AukAFTqbqkVtEdQPiV9RzHNVME/i8BbiOanjwYajxxpHtMfXh9wLPr9+Tf0hV8Hy/u63VRIyR7X7Un7cRMRv4h4j4LPBNqp7WB/287Qy3rtmGRMRzgScCN2fmd4Y9P7Wev9h6vAfwF8Aa4MLMvK/eDuMAqmGiiRplktmen5kP1M9Py8w1XW62xqHN83Yf4CCq8/a2LjddY2gz252psr0vM+/octM1hklm+9HM/F39/AuA/3aruOZp87zdg2oY6YrMvL3LTdcY2sx2T6r5yCv8vO0si9VtQH036BNUd/i+DLyGat7ElRExPTNXj/Ca5wMnANdn5qe72FxNQJvZ/nu9OIAayPO2XJ635Woz2+sy8zPdbK/Gz/fkcpltsw32ugHqit2osj4mM1dHxL3A64Er68fbAR8Gbs3MT9avWQY8mWp+qpqrnWzv6kmLNV6et+XyvC1XO9nai9psvieXy2wbzGK1UBFxAnAv1fLos4GZw+4MTafa7oCICOBfga8DV9XPDWS1j9956YbkjWO25TLbcpltucy2XGZbLrPtHw4DLkg9kXs34AvABqoVI3cE/oxqVcH7MvOBiLgcuD8z31W/bk5rfH24D2MjmW25zLZcZlsusy2X2ZbLbPuTqwEXpD55dq/+mkcDb6FatXdJZt5Zn4A7UA1buByqvTYz87aIGIxqCW5PwAYy23KZbbnMtlxmWy6zLZfZ9ieL1fIcRrWcNpn5SGa+E3hyRBxXH38WcEdm/jgizgBujojdM3OdQxkaz2zLZbblMttymW25zLZcZttnLFb7XERM3eKpK6hOuvnDnnsncGb99wOAP4qIG4F9gIWZee9Wb6gmzGzLZbblMttymW25zLZcZtv/LFb7UETsGREfB8jN933arn58PnBu/dwU4CvAPRGxF9WY/JXAWzPz9Zm5vOv/AxqV2ZbLbMtltuUy23KZbbnMtiwWq/1pb+DUiHghQD2OfjAz19bHLwQeExF/WQ9ZmA2syswVwHmZeVBmfq83TdcYzLZcZlsusy2X2ZbLbMtltgVx65o+EptWINsLuA24FNgrM9cN+5oPAg8ArwUujIiDgacBi+svcbx9A5ltucy2XGZbLrMtl9mWy2zL5NY1DRcRRwHrgJtaQxki4q+ALwMXAP+WmR+MiF2oNij+MXBWZt4eEbOApwI/z8yf9+R/QKMy23KZbbnMtlxmWy6zLZfZls9itaEiYibweeBxwD3Az4DFmfn9iDgX+AbwX8BvgGuBVwG7ZuaP6tdPcdWyZjLbcpltucy2XGZbLrMtl9luO5yz2lz7ACsy8wjgbVQn23n1sRlUd5FeUz+/T2be4wnYN8y2XGZbLrMtl9mWy2zLZbbbCIvVBomIoyJin/rh/sCR9d9XAJ8AdoiIk4HVwFJgD+BJwIER0fpaPAGbx2zLZbblMttymW25zLZcZrttchhwA0REANcAy6kmdn8e+GL930sz86r6644D3g18FLglM39ZP38K8F3gp/XEcjWE2ZbLbMtltuUy23KZbbnMdtvmasA9FBHTM3M1MB+4LDM/GhEvAZ4HPAH4OPAG4Kr6JbcDdwG3ZeYvo9roeENmLv79765eMttymW25zLZcZlsusy2X2QosVnsiIgaBs4G9I+JzwAuAX9SHvwE8THVn6CaqTYo/lJnvBR4CZgF3wuYbHasZzLZcZlsusy2X2ZbLbMtlthrOOatdFhG7AUuAmcC/AK8EfgjsExFPyMyHqVYvuxE4nOpkfVlEXAh8rT62NiIGetF+jc5sy2W25TLbcpltucy2XGarLdmz2n27Abtl5kthaBz+EcAPgFcDZwH/SzVZfG5m3hkRC4GDgM9m5i29aLTGxWzLZbblMttymW25zLZcZqvN2LPafb8DroqIPevHd1GdcDcBh0TE8+thC48A0wEy8xeZ+S1PwMYz23KZbbnMtlxmWy6zLZfZajP2rHbfCuCKzLy/fvwK4AvAl4GdgIsi4ovAS4DTetNETZLZlstsy2W25TLbcpltucxWm3Hrmh6KiD8ErgPmZeaDEXEAcACwM3BzZt7Zy/Zp8sy2XGZbLrMtl9mWy2zLZbYCe1Z7bX+qTYv3jIirgPuBd2Tmr3rbLHWA2ZbLbMtltuUy23KZbbnMVharPfZk4M3AYVSbGl/W4/aoc8y2XGZbLrMtl9mWy2zLZbayWO2x+4AzgA9l5ppeN0YdZbblMttymW25zLZcZlsus5XFao99OjOdNFwmsy2X2ZbLbMtltuUy23KZrVxgSZIkSZLUPO6zKkmSJElqHItVSZIkSVLjWKxKkiRJkhrHYlWSJEmS1DgWq5IkSZKkxrFYlSRJkiQ1jsWqJEmSJKlx/h/P08FFWtaq/gAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 1152x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [],
   "outputs": [],
   "metadata": {}
  }
 ],
 "metadata": {
  "kernelspec": {
   "name": "python3",
   "display_name": "Python 3.8.11 64-bit ('py38tf2': conda)"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.11"
  },
  "interpreter": {
   "hash": "f786b47412788e78e454191c2f87ff8c55e16bdd567362c780933370cac632e2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}