Neural Network with Normalization.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "f35f7977",
   "metadata": {},
   "source": [
    "# Neural Network"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a805ff85",
   "metadata": {},
   "source": [
    "At first, we have to import datasets to Dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "f2fe9cd6",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 93203 entries, 0 to 1044\n",
      "Data columns (total 9 columns):\n",
      " #   Column       Non-Null Count  Dtype \n",
      "---  ------       --------------  ----- \n",
      " 0   Date         93203 non-null  object\n",
      " 1   Time         93203 non-null  object\n",
      " 2   震央地名         93203 non-null  object\n",
      " 3   Lat          93203 non-null  object\n",
      " 4   Long         93203 non-null  object\n",
      " 5   Depth        93203 non-null  int64 \n",
      " 6   M            93203 non-null  object\n",
      " 7   最大震度         93203 non-null  object\n",
      " 8   Source.Name  89371 non-null  object\n",
      "dtypes: int64(1), object(8)\n",
      "memory usage: 7.1+ MB\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import glob\n",
    "\n",
    "# for multiple files\n",
    "PATH = './All/*.csv'\n",
    "data = pd.concat(map(pd.read_csv, glob.glob(PATH)))\n",
    "print(data.info())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c641e996",
   "metadata": {},
   "source": [
    "For analyzation, we use only date, time, position, depth and magnitude."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "a529c453",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 85334 entries, 0 to 1044\n",
      "Data columns (total 6 columns):\n",
      " #   Column  Non-Null Count  Dtype  \n",
      "---  ------  --------------  -----  \n",
      " 0   Date    85334 non-null  object \n",
      " 1   Time    85334 non-null  object \n",
      " 2   Lat     85334 non-null  object \n",
      " 3   Long    85334 non-null  object \n",
      " 4   Depth   85334 non-null  int64  \n",
      " 5   M       85334 non-null  float64\n",
      "dtypes: float64(1), int64(1), object(4)\n",
      "memory usage: 4.6+ MB\n"
     ]
    }
   ],
   "source": [
    "data = data[['Date', 'Time', 'Lat', 'Long', 'Depth', 'M']]\n",
    "data['M'] = data['M'].apply(pd.to_numeric, errors='coerce')\n",
    "data = data.dropna()\n",
    "data['M'] = data['M'].astype(np.float64)\n",
    "data.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f82dfac8",
   "metadata": {},
   "source": [
    "As you can see, latitude and longitude of the positions are not in decimal. For analyzation, we have to convert from dms to decimal."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "2b0e9f5b",
   "metadata": {},
   "outputs": [],
   "source": [
    "import re\n",
    "def dms2dd(s):\n",
    "    degrees, minutes, direction = re.split('[°\\']+', s)\n",
    "    dd = float(degrees) + float(minutes)/60\n",
    "    return dd"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0c4e05c5",
   "metadata": {},
   "source": [
    "Apply dms2dd function to the dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "61b0d05b",
   "metadata": {},
   "outputs": [],
   "source": [
    "data['Lat'] = data['Lat'].apply(dms2dd)\n",
    "data['Long'] = data['Long'].apply(dms2dd)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "08e1b561",
   "metadata": {},
   "source": [
    "In some places, the data is incorrect or unusable, so we have to remove all of these. After removing, we have to check the amount of data for analyzation. After conversion, we have to cast the data to be float."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "892da8d0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 85334 entries, 0 to 1044\n",
      "Data columns (total 6 columns):\n",
      " #   Column  Non-Null Count  Dtype  \n",
      "---  ------  --------------  -----  \n",
      " 0   Date    85334 non-null  object \n",
      " 1   Time    85334 non-null  object \n",
      " 2   Lat     85334 non-null  float64\n",
      " 3   Long    85334 non-null  float64\n",
      " 4   Depth   85334 non-null  float64\n",
      " 5   M       85334 non-null  float64\n",
      "dtypes: float64(4), object(2)\n",
      "memory usage: 4.6+ MB\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "data = data.astype({'Lat': np.float64, 'Long': np.float64, 'M': np.float64, 'Depth': np.float64})\n",
    "data.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a8d44b97",
   "metadata": {},
   "source": [
    "# Convert 'Date' and 'Time' to Timestamp"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a108432a",
   "metadata": {},
   "source": [
    "At first, we have to use array to store the Timestamp after convert from 'Date' and 'Time'."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "436929a6",
   "metadata": {},
   "outputs": [],
   "source": [
    "import datetime\n",
    "import time\n",
    "\n",
    "timestamp = []\n",
    "for d, t in zip(data['Date'], data['Time']):\n",
    "    try:\n",
    "        ts = datetime.datetime.strptime(d+' '+t, '%m/%d/%Y %H:%M:%S')\n",
    "        timestamp.append(time.mktime(ts.timetuple()))\n",
    "    except ValueError:\n",
    "        # if the date and time cannot be converted\n",
    "        timestamp.append('ValueError')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "60a066cd",
   "metadata": {},
   "source": [
    "Insert Timestamp to dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "e88eb2fc",
   "metadata": {},
   "outputs": [],
   "source": [
    "timeStamp = pd.Series(timestamp)\n",
    "data['Timestamp'] = timeStamp.values"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c45dc137",
   "metadata": {},
   "source": [
    "As you can see above, the data in dataframe is not in order by time, so we have to reorder it. We also have to remove the Timestamp Error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "c35db9e1",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date</th>\n",
       "      <th>Time</th>\n",
       "      <th>Lat</th>\n",
       "      <th>Long</th>\n",
       "      <th>Depth</th>\n",
       "      <th>M</th>\n",
       "      <th>Timestamp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>680</th>\n",
       "      <td>1/1/1985</td>\n",
       "      <td>0:58:14</td>\n",
       "      <td>36.103333</td>\n",
       "      <td>141.156667</td>\n",
       "      <td>38.0</td>\n",
       "      <td>3.9</td>\n",
       "      <td>473356694.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>679</th>\n",
       "      <td>1/1/1985</td>\n",
       "      <td>13:19:52</td>\n",
       "      <td>35.825000</td>\n",
       "      <td>137.565000</td>\n",
       "      <td>11.0</td>\n",
       "      <td>3.9</td>\n",
       "      <td>473401192.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>678</th>\n",
       "      <td>1/1/1985</td>\n",
       "      <td>19:41:54</td>\n",
       "      <td>42.181667</td>\n",
       "      <td>142.558333</td>\n",
       "      <td>29.0</td>\n",
       "      <td>3.2</td>\n",
       "      <td>473424114.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>677</th>\n",
       "      <td>1/2/1985</td>\n",
       "      <td>3:31:04</td>\n",
       "      <td>37.018333</td>\n",
       "      <td>141.791667</td>\n",
       "      <td>46.0</td>\n",
       "      <td>4.6</td>\n",
       "      <td>473452264.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>676</th>\n",
       "      <td>1/3/1985</td>\n",
       "      <td>3:54:45</td>\n",
       "      <td>43.361667</td>\n",
       "      <td>145.801667</td>\n",
       "      <td>48.0</td>\n",
       "      <td>3.1</td>\n",
       "      <td>473540085.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Date      Time        Lat        Long  Depth    M    Timestamp\n",
       "680  1/1/1985   0:58:14  36.103333  141.156667   38.0  3.9  473356694.0\n",
       "679  1/1/1985  13:19:52  35.825000  137.565000   11.0  3.9  473401192.0\n",
       "678  1/1/1985  19:41:54  42.181667  142.558333   29.0  3.2  473424114.0\n",
       "677  1/2/1985   3:31:04  37.018333  141.791667   46.0  4.6  473452264.0\n",
       "676  1/3/1985   3:54:45  43.361667  145.801667   48.0  3.1  473540085.0"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = data[data.Timestamp != 'ValueError']\n",
    "data.sort_values(by='Timestamp', inplace=True)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "6f12bf21",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 85334 entries, 680 to 1537\n",
      "Data columns (total 7 columns):\n",
      " #   Column     Non-Null Count  Dtype  \n",
      "---  ------     --------------  -----  \n",
      " 0   Date       85334 non-null  object \n",
      " 1   Time       85334 non-null  object \n",
      " 2   Lat        85334 non-null  float64\n",
      " 3   Long       85334 non-null  float64\n",
      " 4   Depth      85334 non-null  float64\n",
      " 5   M          85334 non-null  float64\n",
      " 6   Timestamp  85334 non-null  float64\n",
      "dtypes: float64(5), object(2)\n",
      "memory usage: 5.2+ MB\n"
     ]
    }
   ],
   "source": [
    "data.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c5717a93",
   "metadata": {},
   "source": [
    "# Split the dataframe to train and test set"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7aafc82a",
   "metadata": {},
   "source": [
    "Check the columns of dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "67265370",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date</th>\n",
       "      <th>Time</th>\n",
       "      <th>Lat</th>\n",
       "      <th>Long</th>\n",
       "      <th>Depth</th>\n",
       "      <th>M</th>\n",
       "      <th>Timestamp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>680</th>\n",
       "      <td>1/1/1985</td>\n",
       "      <td>0:58:14</td>\n",
       "      <td>36.103333</td>\n",
       "      <td>141.156667</td>\n",
       "      <td>38.0</td>\n",
       "      <td>3.9</td>\n",
       "      <td>473356694.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>679</th>\n",
       "      <td>1/1/1985</td>\n",
       "      <td>13:19:52</td>\n",
       "      <td>35.825000</td>\n",
       "      <td>137.565000</td>\n",
       "      <td>11.0</td>\n",
       "      <td>3.9</td>\n",
       "      <td>473401192.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>678</th>\n",
       "      <td>1/1/1985</td>\n",
       "      <td>19:41:54</td>\n",
       "      <td>42.181667</td>\n",
       "      <td>142.558333</td>\n",
       "      <td>29.0</td>\n",
       "      <td>3.2</td>\n",
       "      <td>473424114.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>677</th>\n",
       "      <td>1/2/1985</td>\n",
       "      <td>3:31:04</td>\n",
       "      <td>37.018333</td>\n",
       "      <td>141.791667</td>\n",
       "      <td>46.0</td>\n",
       "      <td>4.6</td>\n",
       "      <td>473452264.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>676</th>\n",
       "      <td>1/3/1985</td>\n",
       "      <td>3:54:45</td>\n",
       "      <td>43.361667</td>\n",
       "      <td>145.801667</td>\n",
       "      <td>48.0</td>\n",
       "      <td>3.1</td>\n",
       "      <td>473540085.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Date      Time        Lat        Long  Depth    M    Timestamp\n",
       "680  1/1/1985   0:58:14  36.103333  141.156667   38.0  3.9  473356694.0\n",
       "679  1/1/1985  13:19:52  35.825000  137.565000   11.0  3.9  473401192.0\n",
       "678  1/1/1985  19:41:54  42.181667  142.558333   29.0  3.2  473424114.0\n",
       "677  1/2/1985   3:31:04  37.018333  141.791667   46.0  4.6  473452264.0\n",
       "676  1/3/1985   3:54:45  43.361667  145.801667   48.0  3.1  473540085.0"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0374d0ca",
   "metadata": {},
   "source": [
    "For Machine Learning, 'Date' and 'Time' are unnecessary. So, we remove both of them from the dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "d9542f7f",
   "metadata": {},
   "outputs": [],
   "source": [
    "data = data.drop(['Date', 'Time'], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e8e457ee",
   "metadata": {},
   "source": [
    "Before that, we normalize the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "35b4ad40",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "          Lat      Long     Depth         M  Timestamp\n",
      "680  0.136779  0.703352  0.333917  0.649842  -2.465423\n",
      "679  0.063646 -0.038494 -0.417347  0.649842  -2.465267\n",
      "678  1.733879  0.992862  0.083496 -0.105507  -2.465187\n",
      "677  0.377198  0.834509  0.556514  1.405192  -2.465088\n",
      "676  2.043927  1.662761  0.612163 -0.213414  -2.464780\n"
     ]
    }
   ],
   "source": [
    "norm_Lat = (data['Lat']-data['Lat'].mean())/data['Lat'].std()\n",
    "norm_Long = (data['Long']-data['Long'].mean())/data['Long'].std()\n",
    "norm_M = (data['M']-data['M'].mean())/data['M'].std()\n",
    "norm_Time = (data['Timestamp']-data['Timestamp'].mean())/data['Timestamp'].std()\n",
    "norm_depth = (data['Depth']-data['Depth'].mean())/data['Depth'].std()\n",
    "data_copy = data.copy()\n",
    "data_copy['Lat'] = norm_Lat\n",
    "data_copy['Long'] = norm_Long\n",
    "data_copy['M'] = norm_M\n",
    "data_copy['Timestamp'] = norm_Time\n",
    "data_copy['Depth'] = norm_depth\n",
    "print(data_copy.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d4a22324",
   "metadata": {},
   "source": [
    "# Neural Network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "681b13e0",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = data_copy[['Timestamp', 'Lat', 'Long']]\n",
    "y = data_copy[['M', 'Depth']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "9809ca7f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(76800, 3) (8534, 3)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "# define the test size would be 20% of all data\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=42)\n",
    "print(X_train.shape, X_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "d7fdd6f4",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = X_train.astype(np.float64)\n",
    "X_test = X_test.astype(np.float64)\n",
    "y_train = y_train.astype(np.float64)\n",
    "y_test = y_test.astype(np.float64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "7b0de182",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras.metrics import RootMeanSquaredError\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "\n",
    "def create_model(activation, optimizer, loss):\n",
    "    model = Sequential()\n",
    "    model.add(Dense(256, activation=activation, input_shape=(3,)))\n",
    "    model.add(Dense(480, activation=activation))\n",
    "    model.add(Dense(2, activation='linear'))\n",
    "    \n",
    "    model.compile(optimizer=optimizer, loss=loss, metrics=['MeanSquaredError'])\n",
    "    \n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "d59ece15",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Metal device set to: Apple M1\n",
      "\n",
      "systemMemory: 16.00 GB\n",
      "maxCacheSize: 5.33 GB\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-11-03 09:25:40.504447: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:306] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.\n",
      "2022-11-03 09:25:40.504896: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:272] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/500\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-11-03 09:25:40.787696: W tensorflow/core/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz\n",
      "2022-11-03 09:25:41.104138: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7/7 [==============================] - 1s 65ms/step - loss: 11.9724 - mean_squared_error: 11.9724 - val_loss: 1.1374 - val_mean_squared_error: 1.1374\n",
      "Epoch 2/500\n",
      "4/7 [================>.............] - ETA: 0s - loss: 1.1246 - mean_squared_error: 1.1246"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-11-03 09:25:41.671547: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7/7 [==============================] - 0s 24ms/step - loss: 1.0698 - mean_squared_error: 1.0698 - val_loss: 1.0048 - val_mean_squared_error: 1.0048\n",
      "Epoch 3/500\n",
      "7/7 [==============================] - 0s 25ms/step - loss: 0.9575 - mean_squared_error: 0.9575 - val_loss: 0.8911 - val_mean_squared_error: 0.8911\n",
      "Epoch 4/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.8813 - mean_squared_error: 0.8813 - val_loss: 0.8143 - val_mean_squared_error: 0.8143\n",
      "Epoch 5/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.8265 - mean_squared_error: 0.8265 - val_loss: 0.7807 - val_mean_squared_error: 0.7807\n",
      "Epoch 6/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.7927 - mean_squared_error: 0.7927 - val_loss: 0.7510 - val_mean_squared_error: 0.7510\n",
      "Epoch 7/500\n",
      "7/7 [==============================] - 0s 39ms/step - loss: 0.7746 - mean_squared_error: 0.7746 - val_loss: 0.7455 - val_mean_squared_error: 0.7455\n",
      "Epoch 8/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.7570 - mean_squared_error: 0.7570 - val_loss: 0.7216 - val_mean_squared_error: 0.7216\n",
      "Epoch 9/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.7402 - mean_squared_error: 0.7402 - val_loss: 0.7063 - val_mean_squared_error: 0.7063\n",
      "Epoch 10/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.7211 - mean_squared_error: 0.7211 - val_loss: 0.7025 - val_mean_squared_error: 0.7025\n",
      "Epoch 11/500\n",
      "7/7 [==============================] - 0s 38ms/step - loss: 0.7058 - mean_squared_error: 0.7058 - val_loss: 0.6809 - val_mean_squared_error: 0.6809\n",
      "Epoch 12/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.6864 - mean_squared_error: 0.6864 - val_loss: 0.6649 - val_mean_squared_error: 0.6649\n",
      "Epoch 13/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.6678 - mean_squared_error: 0.6678 - val_loss: 0.6510 - val_mean_squared_error: 0.6510\n",
      "Epoch 14/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.6491 - mean_squared_error: 0.6491 - val_loss: 0.6521 - val_mean_squared_error: 0.6521\n",
      "Epoch 15/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.6289 - mean_squared_error: 0.6289 - val_loss: 0.6214 - val_mean_squared_error: 0.6214\n",
      "Epoch 16/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.6100 - mean_squared_error: 0.6100 - val_loss: 0.6091 - val_mean_squared_error: 0.6091\n",
      "Epoch 17/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.5949 - mean_squared_error: 0.5949 - val_loss: 0.5966 - val_mean_squared_error: 0.5966\n",
      "Epoch 18/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.5844 - mean_squared_error: 0.5844 - val_loss: 0.5883 - val_mean_squared_error: 0.5883\n",
      "Epoch 19/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.5724 - mean_squared_error: 0.5724 - val_loss: 0.5856 - val_mean_squared_error: 0.5856\n",
      "Epoch 20/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.5567 - mean_squared_error: 0.5567 - val_loss: 0.5746 - val_mean_squared_error: 0.5746\n",
      "Epoch 21/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.5542 - mean_squared_error: 0.5542 - val_loss: 0.5823 - val_mean_squared_error: 0.5823\n",
      "Epoch 22/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.5544 - mean_squared_error: 0.5544 - val_loss: 0.5592 - val_mean_squared_error: 0.5592\n",
      "Epoch 23/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.5380 - mean_squared_error: 0.5380 - val_loss: 0.5485 - val_mean_squared_error: 0.5485\n",
      "Epoch 24/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.5294 - mean_squared_error: 0.5294 - val_loss: 0.5575 - val_mean_squared_error: 0.5575\n",
      "Epoch 25/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.5271 - mean_squared_error: 0.5271 - val_loss: 0.5434 - val_mean_squared_error: 0.5434\n",
      "Epoch 26/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.5208 - mean_squared_error: 0.5208 - val_loss: 0.5441 - val_mean_squared_error: 0.5441\n",
      "Epoch 27/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.5226 - mean_squared_error: 0.5226 - val_loss: 0.5363 - val_mean_squared_error: 0.5363\n",
      "Epoch 28/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.5234 - mean_squared_error: 0.5234 - val_loss: 0.5260 - val_mean_squared_error: 0.5260\n",
      "Epoch 29/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.5105 - mean_squared_error: 0.5105 - val_loss: 0.5287 - val_mean_squared_error: 0.5287\n",
      "Epoch 30/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.5117 - mean_squared_error: 0.5117 - val_loss: 0.5136 - val_mean_squared_error: 0.5136\n",
      "Epoch 31/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.5085 - mean_squared_error: 0.5085 - val_loss: 0.5068 - val_mean_squared_error: 0.5068\n",
      "Epoch 32/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4962 - mean_squared_error: 0.4962 - val_loss: 0.5116 - val_mean_squared_error: 0.5116\n",
      "Epoch 33/500\n",
      "7/7 [==============================] - 0s 34ms/step - loss: 0.4930 - mean_squared_error: 0.4930 - val_loss: 0.5070 - val_mean_squared_error: 0.5070\n",
      "Epoch 34/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.5023 - mean_squared_error: 0.5023 - val_loss: 0.5444 - val_mean_squared_error: 0.5444\n",
      "Epoch 35/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.5034 - mean_squared_error: 0.5034 - val_loss: 0.5108 - val_mean_squared_error: 0.5108\n",
      "Epoch 36/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4964 - mean_squared_error: 0.4964 - val_loss: 0.5084 - val_mean_squared_error: 0.5084\n",
      "Epoch 37/500\n",
      "7/7 [==============================] - 0s 52ms/step - loss: 0.4916 - mean_squared_error: 0.4916 - val_loss: 0.5160 - val_mean_squared_error: 0.5160\n",
      "Epoch 38/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4963 - mean_squared_error: 0.4963 - val_loss: 0.4888 - val_mean_squared_error: 0.4888\n",
      "Epoch 39/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4860 - mean_squared_error: 0.4860 - val_loss: 0.4957 - val_mean_squared_error: 0.4957\n",
      "Epoch 40/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.5015 - mean_squared_error: 0.5015 - val_loss: 0.5054 - val_mean_squared_error: 0.5054\n",
      "Epoch 41/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.4852 - mean_squared_error: 0.4852 - val_loss: 0.5001 - val_mean_squared_error: 0.5001\n",
      "Epoch 42/500\n",
      "7/7 [==============================] - 0s 39ms/step - loss: 0.4814 - mean_squared_error: 0.4814 - val_loss: 0.4916 - val_mean_squared_error: 0.4916\n",
      "Epoch 43/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.4806 - mean_squared_error: 0.4806 - val_loss: 0.4987 - val_mean_squared_error: 0.4987\n",
      "Epoch 44/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.4770 - mean_squared_error: 0.4770 - val_loss: 0.4801 - val_mean_squared_error: 0.4801\n",
      "Epoch 45/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.4831 - mean_squared_error: 0.4831 - val_loss: 0.4882 - val_mean_squared_error: 0.4882\n",
      "Epoch 46/500\n",
      "7/7 [==============================] - 0s 34ms/step - loss: 0.4805 - mean_squared_error: 0.4805 - val_loss: 0.4751 - val_mean_squared_error: 0.4751\n",
      "Epoch 47/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4722 - mean_squared_error: 0.4722 - val_loss: 0.4853 - val_mean_squared_error: 0.4853\n",
      "Epoch 48/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.4727 - mean_squared_error: 0.4727 - val_loss: 0.4770 - val_mean_squared_error: 0.4770\n",
      "Epoch 49/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.4684 - mean_squared_error: 0.4684 - val_loss: 0.4879 - val_mean_squared_error: 0.4879\n",
      "Epoch 50/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.4727 - mean_squared_error: 0.4727 - val_loss: 0.4716 - val_mean_squared_error: 0.4716\n",
      "Epoch 51/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4678 - mean_squared_error: 0.4678 - val_loss: 0.4859 - val_mean_squared_error: 0.4859\n",
      "Epoch 52/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4621 - mean_squared_error: 0.4621 - val_loss: 0.4618 - val_mean_squared_error: 0.4618\n",
      "Epoch 53/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4589 - mean_squared_error: 0.4589 - val_loss: 0.4652 - val_mean_squared_error: 0.4652\n",
      "Epoch 54/500\n",
      "7/7 [==============================] - 0s 45ms/step - loss: 0.4602 - mean_squared_error: 0.4602 - val_loss: 0.4702 - val_mean_squared_error: 0.4702\n",
      "Epoch 55/500\n",
      "7/7 [==============================] - 0s 37ms/step - loss: 0.4599 - mean_squared_error: 0.4599 - val_loss: 0.4798 - val_mean_squared_error: 0.4798\n",
      "Epoch 56/500\n",
      "7/7 [==============================] - 0s 56ms/step - loss: 0.4612 - mean_squared_error: 0.4612 - val_loss: 0.4697 - val_mean_squared_error: 0.4697\n",
      "Epoch 57/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.4658 - mean_squared_error: 0.4658 - val_loss: 0.4625 - val_mean_squared_error: 0.4625\n",
      "Epoch 58/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4547 - mean_squared_error: 0.4547 - val_loss: 0.4566 - val_mean_squared_error: 0.4566\n",
      "Epoch 59/500\n",
      "7/7 [==============================] - 0s 56ms/step - loss: 0.4600 - mean_squared_error: 0.4600 - val_loss: 0.4555 - val_mean_squared_error: 0.4555\n",
      "Epoch 60/500\n",
      "7/7 [==============================] - 0s 36ms/step - loss: 0.4707 - mean_squared_error: 0.4707 - val_loss: 0.5000 - val_mean_squared_error: 0.5000\n",
      "Epoch 61/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.4647 - mean_squared_error: 0.4647 - val_loss: 0.4680 - val_mean_squared_error: 0.4680\n",
      "Epoch 62/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4639 - mean_squared_error: 0.4639 - val_loss: 0.4774 - val_mean_squared_error: 0.4774\n",
      "Epoch 63/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.4696 - mean_squared_error: 0.4696 - val_loss: 0.4723 - val_mean_squared_error: 0.4723\n",
      "Epoch 64/500\n",
      "7/7 [==============================] - 0s 38ms/step - loss: 0.4680 - mean_squared_error: 0.4680 - val_loss: 0.4702 - val_mean_squared_error: 0.4702\n",
      "Epoch 65/500\n",
      "7/7 [==============================] - 0s 48ms/step - loss: 0.4629 - mean_squared_error: 0.4629 - val_loss: 0.4695 - val_mean_squared_error: 0.4695\n",
      "Epoch 66/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.4594 - mean_squared_error: 0.4594 - val_loss: 0.4658 - val_mean_squared_error: 0.4658\n",
      "Epoch 67/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.4496 - mean_squared_error: 0.4496 - val_loss: 0.4510 - val_mean_squared_error: 0.4510\n",
      "Epoch 68/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.4502 - mean_squared_error: 0.4502 - val_loss: 0.4695 - val_mean_squared_error: 0.4695\n",
      "Epoch 69/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4585 - mean_squared_error: 0.4585 - val_loss: 0.4610 - val_mean_squared_error: 0.4610\n",
      "Epoch 70/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4482 - mean_squared_error: 0.4482 - val_loss: 0.4681 - val_mean_squared_error: 0.4681\n",
      "Epoch 71/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4500 - mean_squared_error: 0.4500 - val_loss: 0.4604 - val_mean_squared_error: 0.4604\n",
      "Epoch 72/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.4548 - mean_squared_error: 0.4548 - val_loss: 0.4529 - val_mean_squared_error: 0.4529\n",
      "Epoch 73/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4431 - mean_squared_error: 0.4431 - val_loss: 0.4402 - val_mean_squared_error: 0.4402\n",
      "Epoch 74/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4384 - mean_squared_error: 0.4384 - val_loss: 0.4419 - val_mean_squared_error: 0.4419\n",
      "Epoch 75/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.4391 - mean_squared_error: 0.4391 - val_loss: 0.4596 - val_mean_squared_error: 0.4596\n",
      "Epoch 76/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4532 - mean_squared_error: 0.4532 - val_loss: 0.4704 - val_mean_squared_error: 0.4704\n",
      "Epoch 77/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4618 - mean_squared_error: 0.4618 - val_loss: 0.4775 - val_mean_squared_error: 0.4775\n",
      "Epoch 78/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.4690 - mean_squared_error: 0.4690 - val_loss: 0.4679 - val_mean_squared_error: 0.4679\n",
      "Epoch 79/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.4577 - mean_squared_error: 0.4577 - val_loss: 0.4846 - val_mean_squared_error: 0.4846\n",
      "Epoch 80/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.4643 - mean_squared_error: 0.4643 - val_loss: 0.4492 - val_mean_squared_error: 0.4492\n",
      "Epoch 81/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.4426 - mean_squared_error: 0.4426 - val_loss: 0.4535 - val_mean_squared_error: 0.4535\n",
      "Epoch 82/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4385 - mean_squared_error: 0.4385 - val_loss: 0.4418 - val_mean_squared_error: 0.4418\n",
      "Epoch 83/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.4440 - mean_squared_error: 0.4440 - val_loss: 0.4362 - val_mean_squared_error: 0.4362\n",
      "Epoch 84/500\n",
      "7/7 [==============================] - 0s 38ms/step - loss: 0.4438 - mean_squared_error: 0.4438 - val_loss: 0.4630 - val_mean_squared_error: 0.4630\n",
      "Epoch 85/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.4416 - mean_squared_error: 0.4416 - val_loss: 0.4547 - val_mean_squared_error: 0.4547\n",
      "Epoch 86/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4470 - mean_squared_error: 0.4470 - val_loss: 0.4448 - val_mean_squared_error: 0.4448\n",
      "Epoch 87/500\n",
      "7/7 [==============================] - 0s 53ms/step - loss: 0.4630 - mean_squared_error: 0.4630 - val_loss: 0.4997 - val_mean_squared_error: 0.4997\n",
      "Epoch 88/500\n",
      "7/7 [==============================] - 0s 34ms/step - loss: 0.4653 - mean_squared_error: 0.4653 - val_loss: 0.4717 - val_mean_squared_error: 0.4717\n",
      "Epoch 89/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.4438 - mean_squared_error: 0.4438 - val_loss: 0.4563 - val_mean_squared_error: 0.4563\n",
      "Epoch 90/500\n",
      "7/7 [==============================] - 0s 37ms/step - loss: 0.4389 - mean_squared_error: 0.4389 - val_loss: 0.4512 - val_mean_squared_error: 0.4512\n",
      "Epoch 91/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.4407 - mean_squared_error: 0.4407 - val_loss: 0.4465 - val_mean_squared_error: 0.4465\n",
      "Epoch 92/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.4296 - mean_squared_error: 0.4296 - val_loss: 0.4471 - val_mean_squared_error: 0.4471\n",
      "Epoch 93/500\n",
      "7/7 [==============================] - 0s 34ms/step - loss: 0.4317 - mean_squared_error: 0.4317 - val_loss: 0.4458 - val_mean_squared_error: 0.4458\n",
      "Epoch 94/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4279 - mean_squared_error: 0.4279 - val_loss: 0.4434 - val_mean_squared_error: 0.4434\n",
      "Epoch 95/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.4268 - mean_squared_error: 0.4268 - val_loss: 0.4345 - val_mean_squared_error: 0.4345\n",
      "Epoch 96/500\n",
      "7/7 [==============================] - 0s 38ms/step - loss: 0.4304 - mean_squared_error: 0.4304 - val_loss: 0.4387 - val_mean_squared_error: 0.4387\n",
      "Epoch 97/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4239 - mean_squared_error: 0.4239 - val_loss: 0.4274 - val_mean_squared_error: 0.4274\n",
      "Epoch 98/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.4204 - mean_squared_error: 0.4204 - val_loss: 0.4226 - val_mean_squared_error: 0.4226\n",
      "Epoch 99/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.4255 - mean_squared_error: 0.4255 - val_loss: 0.4285 - val_mean_squared_error: 0.4285\n",
      "Epoch 100/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4269 - mean_squared_error: 0.4269 - val_loss: 0.4414 - val_mean_squared_error: 0.4414\n",
      "Epoch 101/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.4312 - mean_squared_error: 0.4312 - val_loss: 0.4523 - val_mean_squared_error: 0.4523\n",
      "Epoch 102/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.4447 - mean_squared_error: 0.4447 - val_loss: 0.4486 - val_mean_squared_error: 0.4486\n",
      "Epoch 103/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4449 - mean_squared_error: 0.4449 - val_loss: 0.4524 - val_mean_squared_error: 0.4524\n",
      "Epoch 104/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.4393 - mean_squared_error: 0.4393 - val_loss: 0.4385 - val_mean_squared_error: 0.4385\n",
      "Epoch 105/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.4426 - mean_squared_error: 0.4426 - val_loss: 0.4661 - val_mean_squared_error: 0.4661\n",
      "Epoch 106/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4504 - mean_squared_error: 0.4504 - val_loss: 0.4449 - val_mean_squared_error: 0.4449\n",
      "Epoch 107/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.4373 - mean_squared_error: 0.4373 - val_loss: 0.4589 - val_mean_squared_error: 0.4589\n",
      "Epoch 108/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.4340 - mean_squared_error: 0.4340 - val_loss: 0.4327 - val_mean_squared_error: 0.4327\n",
      "Epoch 109/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4317 - mean_squared_error: 0.4317 - val_loss: 0.4350 - val_mean_squared_error: 0.4350\n",
      "Epoch 110/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4259 - mean_squared_error: 0.4259 - val_loss: 0.4302 - val_mean_squared_error: 0.4302\n",
      "Epoch 111/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4212 - mean_squared_error: 0.4212 - val_loss: 0.4330 - val_mean_squared_error: 0.4330\n",
      "Epoch 112/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4309 - mean_squared_error: 0.4309 - val_loss: 0.4894 - val_mean_squared_error: 0.4894\n",
      "Epoch 113/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.4500 - mean_squared_error: 0.4500 - val_loss: 0.5234 - val_mean_squared_error: 0.5234\n",
      "Epoch 114/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.4468 - mean_squared_error: 0.4468 - val_loss: 0.4436 - val_mean_squared_error: 0.4436\n",
      "Epoch 115/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.4222 - mean_squared_error: 0.4222 - val_loss: 0.4397 - val_mean_squared_error: 0.4397\n",
      "Epoch 116/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4181 - mean_squared_error: 0.4181 - val_loss: 0.4261 - val_mean_squared_error: 0.4261\n",
      "Epoch 117/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.4156 - mean_squared_error: 0.4156 - val_loss: 0.4205 - val_mean_squared_error: 0.4205\n",
      "Epoch 118/500\n",
      "7/7 [==============================] - 0s 42ms/step - loss: 0.4192 - mean_squared_error: 0.4192 - val_loss: 0.4432 - val_mean_squared_error: 0.4432\n",
      "Epoch 119/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.4178 - mean_squared_error: 0.4178 - val_loss: 0.4336 - val_mean_squared_error: 0.4336\n",
      "Epoch 120/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.4182 - mean_squared_error: 0.4182 - val_loss: 0.4354 - val_mean_squared_error: 0.4354\n",
      "Epoch 121/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4131 - mean_squared_error: 0.4131 - val_loss: 0.4217 - val_mean_squared_error: 0.4217\n",
      "Epoch 122/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4084 - mean_squared_error: 0.4084 - val_loss: 0.4224 - val_mean_squared_error: 0.4224\n",
      "Epoch 123/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4231 - mean_squared_error: 0.4231 - val_loss: 0.4284 - val_mean_squared_error: 0.4284\n",
      "Epoch 124/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4177 - mean_squared_error: 0.4177 - val_loss: 0.4299 - val_mean_squared_error: 0.4299\n",
      "Epoch 125/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.4199 - mean_squared_error: 0.4199 - val_loss: 0.4344 - val_mean_squared_error: 0.4344\n",
      "Epoch 126/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4317 - mean_squared_error: 0.4317 - val_loss: 0.4303 - val_mean_squared_error: 0.4303\n",
      "Epoch 127/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4451 - mean_squared_error: 0.4451 - val_loss: 0.4345 - val_mean_squared_error: 0.4345\n",
      "Epoch 128/500\n",
      "7/7 [==============================] - 0s 38ms/step - loss: 0.4341 - mean_squared_error: 0.4341 - val_loss: 0.4435 - val_mean_squared_error: 0.4435\n",
      "Epoch 129/500\n",
      "7/7 [==============================] - 0s 40ms/step - loss: 0.4306 - mean_squared_error: 0.4306 - val_loss: 0.4211 - val_mean_squared_error: 0.4211\n",
      "Epoch 130/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.4223 - mean_squared_error: 0.4223 - val_loss: 0.4180 - val_mean_squared_error: 0.4180\n",
      "Epoch 131/500\n",
      "7/7 [==============================] - 0s 36ms/step - loss: 0.4245 - mean_squared_error: 0.4245 - val_loss: 0.4233 - val_mean_squared_error: 0.4233\n",
      "Epoch 132/500\n",
      "7/7 [==============================] - 0s 37ms/step - loss: 0.4193 - mean_squared_error: 0.4193 - val_loss: 0.4231 - val_mean_squared_error: 0.4231\n",
      "Epoch 133/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.4282 - mean_squared_error: 0.4282 - val_loss: 0.4701 - val_mean_squared_error: 0.4701\n",
      "Epoch 134/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.4380 - mean_squared_error: 0.4380 - val_loss: 0.4379 - val_mean_squared_error: 0.4379\n",
      "Epoch 135/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.4306 - mean_squared_error: 0.4306 - val_loss: 0.4219 - val_mean_squared_error: 0.4219\n",
      "Epoch 136/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.4363 - mean_squared_error: 0.4363 - val_loss: 0.4462 - val_mean_squared_error: 0.4462\n",
      "Epoch 137/500\n",
      "7/7 [==============================] - 0s 38ms/step - loss: 0.4285 - mean_squared_error: 0.4285 - val_loss: 0.4358 - val_mean_squared_error: 0.4358\n",
      "Epoch 138/500\n",
      "7/7 [==============================] - 0s 36ms/step - loss: 0.4285 - mean_squared_error: 0.4285 - val_loss: 0.4294 - val_mean_squared_error: 0.4294\n",
      "Epoch 139/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.4146 - mean_squared_error: 0.4146 - val_loss: 0.4138 - val_mean_squared_error: 0.4138\n",
      "Epoch 140/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.4090 - mean_squared_error: 0.4090 - val_loss: 0.4157 - val_mean_squared_error: 0.4157\n",
      "Epoch 141/500\n",
      "7/7 [==============================] - 0s 37ms/step - loss: 0.4043 - mean_squared_error: 0.4043 - val_loss: 0.4148 - val_mean_squared_error: 0.4148\n",
      "Epoch 142/500\n",
      "7/7 [==============================] - 0s 37ms/step - loss: 0.4050 - mean_squared_error: 0.4050 - val_loss: 0.4090 - val_mean_squared_error: 0.4090\n",
      "Epoch 143/500\n",
      "7/7 [==============================] - 0s 40ms/step - loss: 0.4020 - mean_squared_error: 0.4020 - val_loss: 0.4146 - val_mean_squared_error: 0.4146\n",
      "Epoch 144/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4002 - mean_squared_error: 0.4002 - val_loss: 0.4109 - val_mean_squared_error: 0.4109\n",
      "Epoch 145/500\n",
      "7/7 [==============================] - 0s 42ms/step - loss: 0.4005 - mean_squared_error: 0.4005 - val_loss: 0.4112 - val_mean_squared_error: 0.4112\n",
      "Epoch 146/500\n",
      "7/7 [==============================] - 0s 47ms/step - loss: 0.4016 - mean_squared_error: 0.4016 - val_loss: 0.4188 - val_mean_squared_error: 0.4188\n",
      "Epoch 147/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.4120 - mean_squared_error: 0.4120 - val_loss: 0.4305 - val_mean_squared_error: 0.4305\n",
      "Epoch 148/500\n",
      "7/7 [==============================] - 0s 24ms/step - loss: 0.4123 - mean_squared_error: 0.4123 - val_loss: 0.4201 - val_mean_squared_error: 0.4201\n",
      "Epoch 149/500\n",
      "7/7 [==============================] - 0s 24ms/step - loss: 0.4124 - mean_squared_error: 0.4124 - val_loss: 0.4211 - val_mean_squared_error: 0.4211\n",
      "Epoch 150/500\n",
      "7/7 [==============================] - 0s 40ms/step - loss: 0.4070 - mean_squared_error: 0.4070 - val_loss: 0.4128 - val_mean_squared_error: 0.4128\n",
      "Epoch 151/500\n",
      "7/7 [==============================] - 0s 44ms/step - loss: 0.4062 - mean_squared_error: 0.4062 - val_loss: 0.4129 - val_mean_squared_error: 0.4129\n",
      "Epoch 152/500\n",
      "7/7 [==============================] - 0s 38ms/step - loss: 0.4087 - mean_squared_error: 0.4087 - val_loss: 0.4314 - val_mean_squared_error: 0.4314\n",
      "Epoch 153/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.4152 - mean_squared_error: 0.4152 - val_loss: 0.4140 - val_mean_squared_error: 0.4140\n",
      "Epoch 154/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.4104 - mean_squared_error: 0.4104 - val_loss: 0.4365 - val_mean_squared_error: 0.4365\n",
      "Epoch 155/500\n",
      "7/7 [==============================] - 0s 41ms/step - loss: 0.4286 - mean_squared_error: 0.4286 - val_loss: 0.4176 - val_mean_squared_error: 0.4176\n",
      "Epoch 156/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.4121 - mean_squared_error: 0.4121 - val_loss: 0.4167 - val_mean_squared_error: 0.4167\n",
      "Epoch 157/500\n",
      "7/7 [==============================] - 0s 34ms/step - loss: 0.4052 - mean_squared_error: 0.4052 - val_loss: 0.4040 - val_mean_squared_error: 0.4040\n",
      "Epoch 158/500\n",
      "7/7 [==============================] - 0s 46ms/step - loss: 0.4085 - mean_squared_error: 0.4085 - val_loss: 0.4176 - val_mean_squared_error: 0.4176\n",
      "Epoch 159/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.4075 - mean_squared_error: 0.4075 - val_loss: 0.4308 - val_mean_squared_error: 0.4308\n",
      "Epoch 160/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4098 - mean_squared_error: 0.4098 - val_loss: 0.4242 - val_mean_squared_error: 0.4242\n",
      "Epoch 161/500\n",
      "7/7 [==============================] - 0s 44ms/step - loss: 0.4035 - mean_squared_error: 0.4035 - val_loss: 0.4137 - val_mean_squared_error: 0.4137\n",
      "Epoch 162/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.4008 - mean_squared_error: 0.4008 - val_loss: 0.4193 - val_mean_squared_error: 0.4193\n",
      "Epoch 163/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.4006 - mean_squared_error: 0.4006 - val_loss: 0.4199 - val_mean_squared_error: 0.4199\n",
      "Epoch 164/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.4015 - mean_squared_error: 0.4015 - val_loss: 0.4162 - val_mean_squared_error: 0.4162\n",
      "Epoch 165/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.4034 - mean_squared_error: 0.4034 - val_loss: 0.4421 - val_mean_squared_error: 0.4421\n",
      "Epoch 166/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.4011 - mean_squared_error: 0.4011 - val_loss: 0.4124 - val_mean_squared_error: 0.4124\n",
      "Epoch 167/500\n",
      "7/7 [==============================] - 0s 34ms/step - loss: 0.4092 - mean_squared_error: 0.4092 - val_loss: 0.4076 - val_mean_squared_error: 0.4076\n",
      "Epoch 168/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.4168 - mean_squared_error: 0.4168 - val_loss: 0.4174 - val_mean_squared_error: 0.4174\n",
      "Epoch 169/500\n",
      "7/7 [==============================] - 0s 41ms/step - loss: 0.4048 - mean_squared_error: 0.4048 - val_loss: 0.4092 - val_mean_squared_error: 0.4092\n",
      "Epoch 170/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.4027 - mean_squared_error: 0.4027 - val_loss: 0.4092 - val_mean_squared_error: 0.4092\n",
      "Epoch 171/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3987 - mean_squared_error: 0.3987 - val_loss: 0.4102 - val_mean_squared_error: 0.4102\n",
      "Epoch 172/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.4015 - mean_squared_error: 0.4015 - val_loss: 0.4157 - val_mean_squared_error: 0.4157\n",
      "Epoch 173/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.4028 - mean_squared_error: 0.4028 - val_loss: 0.4227 - val_mean_squared_error: 0.4227\n",
      "Epoch 174/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.4048 - mean_squared_error: 0.4048 - val_loss: 0.4263 - val_mean_squared_error: 0.4263\n",
      "Epoch 175/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.4052 - mean_squared_error: 0.4052 - val_loss: 0.4223 - val_mean_squared_error: 0.4223\n",
      "Epoch 176/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.4018 - mean_squared_error: 0.4018 - val_loss: 0.4128 - val_mean_squared_error: 0.4128\n",
      "Epoch 177/500\n",
      "7/7 [==============================] - 0s 34ms/step - loss: 0.4088 - mean_squared_error: 0.4088 - val_loss: 0.4227 - val_mean_squared_error: 0.4227\n",
      "Epoch 178/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.4128 - mean_squared_error: 0.4128 - val_loss: 0.4481 - val_mean_squared_error: 0.4481\n",
      "Epoch 179/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.4194 - mean_squared_error: 0.4194 - val_loss: 0.4542 - val_mean_squared_error: 0.4542\n",
      "Epoch 180/500\n",
      "7/7 [==============================] - 0s 42ms/step - loss: 0.4156 - mean_squared_error: 0.4156 - val_loss: 0.4209 - val_mean_squared_error: 0.4209\n",
      "Epoch 181/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4073 - mean_squared_error: 0.4073 - val_loss: 0.4273 - val_mean_squared_error: 0.4273\n",
      "Epoch 182/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.4077 - mean_squared_error: 0.4077 - val_loss: 0.4327 - val_mean_squared_error: 0.4327\n",
      "Epoch 183/500\n",
      "7/7 [==============================] - 0s 34ms/step - loss: 0.4055 - mean_squared_error: 0.4055 - val_loss: 0.4274 - val_mean_squared_error: 0.4274\n",
      "Epoch 184/500\n",
      "7/7 [==============================] - 0s 34ms/step - loss: 0.4004 - mean_squared_error: 0.4004 - val_loss: 0.4149 - val_mean_squared_error: 0.4149\n",
      "Epoch 185/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.4030 - mean_squared_error: 0.4030 - val_loss: 0.4318 - val_mean_squared_error: 0.4318\n",
      "Epoch 186/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.3992 - mean_squared_error: 0.3992 - val_loss: 0.4287 - val_mean_squared_error: 0.4287\n",
      "Epoch 187/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.4035 - mean_squared_error: 0.4035 - val_loss: 0.4426 - val_mean_squared_error: 0.4426\n",
      "Epoch 188/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4069 - mean_squared_error: 0.4069 - val_loss: 0.4107 - val_mean_squared_error: 0.4107\n",
      "Epoch 189/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4097 - mean_squared_error: 0.4097 - val_loss: 0.4060 - val_mean_squared_error: 0.4060\n",
      "Epoch 190/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.4125 - mean_squared_error: 0.4125 - val_loss: 0.4607 - val_mean_squared_error: 0.4607\n",
      "Epoch 191/500\n",
      "7/7 [==============================] - 0s 43ms/step - loss: 0.4223 - mean_squared_error: 0.4223 - val_loss: 0.4202 - val_mean_squared_error: 0.4202\n",
      "Epoch 192/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.4123 - mean_squared_error: 0.4123 - val_loss: 0.4249 - val_mean_squared_error: 0.4249\n",
      "Epoch 193/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4101 - mean_squared_error: 0.4101 - val_loss: 0.4157 - val_mean_squared_error: 0.4157\n",
      "Epoch 194/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4026 - mean_squared_error: 0.4026 - val_loss: 0.4204 - val_mean_squared_error: 0.4204\n",
      "Epoch 195/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4036 - mean_squared_error: 0.4036 - val_loss: 0.4061 - val_mean_squared_error: 0.4061\n",
      "Epoch 196/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3930 - mean_squared_error: 0.3930 - val_loss: 0.3994 - val_mean_squared_error: 0.3994\n",
      "Epoch 197/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.3907 - mean_squared_error: 0.3907 - val_loss: 0.3985 - val_mean_squared_error: 0.3985\n",
      "Epoch 198/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3873 - mean_squared_error: 0.3873 - val_loss: 0.3981 - val_mean_squared_error: 0.3981\n",
      "Epoch 199/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3890 - mean_squared_error: 0.3890 - val_loss: 0.4061 - val_mean_squared_error: 0.4061\n",
      "Epoch 200/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3970 - mean_squared_error: 0.3970 - val_loss: 0.4023 - val_mean_squared_error: 0.4023\n",
      "Epoch 201/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3950 - mean_squared_error: 0.3950 - val_loss: 0.4283 - val_mean_squared_error: 0.4283\n",
      "Epoch 202/500\n",
      "7/7 [==============================] - 0s 44ms/step - loss: 0.4021 - mean_squared_error: 0.4021 - val_loss: 0.4061 - val_mean_squared_error: 0.4061\n",
      "Epoch 203/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3998 - mean_squared_error: 0.3998 - val_loss: 0.4263 - val_mean_squared_error: 0.4263\n",
      "Epoch 204/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4065 - mean_squared_error: 0.4065 - val_loss: 0.4259 - val_mean_squared_error: 0.4259\n",
      "Epoch 205/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4043 - mean_squared_error: 0.4043 - val_loss: 0.4068 - val_mean_squared_error: 0.4068\n",
      "Epoch 206/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4046 - mean_squared_error: 0.4046 - val_loss: 0.4044 - val_mean_squared_error: 0.4044\n",
      "Epoch 207/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4064 - mean_squared_error: 0.4064 - val_loss: 0.4073 - val_mean_squared_error: 0.4073\n",
      "Epoch 208/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3903 - mean_squared_error: 0.3903 - val_loss: 0.3990 - val_mean_squared_error: 0.3990\n",
      "Epoch 209/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3911 - mean_squared_error: 0.3911 - val_loss: 0.3969 - val_mean_squared_error: 0.3969\n",
      "Epoch 210/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3871 - mean_squared_error: 0.3871 - val_loss: 0.3989 - val_mean_squared_error: 0.3989\n",
      "Epoch 211/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.3878 - mean_squared_error: 0.3878 - val_loss: 0.4130 - val_mean_squared_error: 0.4130\n",
      "Epoch 212/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3913 - mean_squared_error: 0.3913 - val_loss: 0.4148 - val_mean_squared_error: 0.4148\n",
      "Epoch 213/500\n",
      "7/7 [==============================] - 0s 36ms/step - loss: 0.3925 - mean_squared_error: 0.3925 - val_loss: 0.4001 - val_mean_squared_error: 0.4001\n",
      "Epoch 214/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3912 - mean_squared_error: 0.3912 - val_loss: 0.3991 - val_mean_squared_error: 0.3991\n",
      "Epoch 215/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3828 - mean_squared_error: 0.3828 - val_loss: 0.3995 - val_mean_squared_error: 0.3995\n",
      "Epoch 216/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3866 - mean_squared_error: 0.3866 - val_loss: 0.4031 - val_mean_squared_error: 0.4031\n",
      "Epoch 217/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3882 - mean_squared_error: 0.3882 - val_loss: 0.4152 - val_mean_squared_error: 0.4152\n",
      "Epoch 218/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3968 - mean_squared_error: 0.3968 - val_loss: 0.3984 - val_mean_squared_error: 0.3984\n",
      "Epoch 219/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3895 - mean_squared_error: 0.3895 - val_loss: 0.4089 - val_mean_squared_error: 0.4089\n",
      "Epoch 220/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.3866 - mean_squared_error: 0.3866 - val_loss: 0.4050 - val_mean_squared_error: 0.4050\n",
      "Epoch 221/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.3940 - mean_squared_error: 0.3940 - val_loss: 0.4134 - val_mean_squared_error: 0.4134\n",
      "Epoch 222/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3945 - mean_squared_error: 0.3945 - val_loss: 0.4081 - val_mean_squared_error: 0.4081\n",
      "Epoch 223/500\n",
      "7/7 [==============================] - 0s 42ms/step - loss: 0.3951 - mean_squared_error: 0.3951 - val_loss: 0.4133 - val_mean_squared_error: 0.4133\n",
      "Epoch 224/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.4061 - mean_squared_error: 0.4061 - val_loss: 0.4291 - val_mean_squared_error: 0.4291\n",
      "Epoch 225/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.4056 - mean_squared_error: 0.4056 - val_loss: 0.4060 - val_mean_squared_error: 0.4060\n",
      "Epoch 226/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.3990 - mean_squared_error: 0.3990 - val_loss: 0.4193 - val_mean_squared_error: 0.4193\n",
      "Epoch 227/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3947 - mean_squared_error: 0.3947 - val_loss: 0.4097 - val_mean_squared_error: 0.4097\n",
      "Epoch 228/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3851 - mean_squared_error: 0.3851 - val_loss: 0.4001 - val_mean_squared_error: 0.4001\n",
      "Epoch 229/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3855 - mean_squared_error: 0.3855 - val_loss: 0.4212 - val_mean_squared_error: 0.4212\n",
      "Epoch 230/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3900 - mean_squared_error: 0.3900 - val_loss: 0.4140 - val_mean_squared_error: 0.4140\n",
      "Epoch 231/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4018 - mean_squared_error: 0.4018 - val_loss: 0.4222 - val_mean_squared_error: 0.4222\n",
      "Epoch 232/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4072 - mean_squared_error: 0.4072 - val_loss: 0.4296 - val_mean_squared_error: 0.4296\n",
      "Epoch 233/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4019 - mean_squared_error: 0.4019 - val_loss: 0.4228 - val_mean_squared_error: 0.4228\n",
      "Epoch 234/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3985 - mean_squared_error: 0.3985 - val_loss: 0.4011 - val_mean_squared_error: 0.4011\n",
      "Epoch 235/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4084 - mean_squared_error: 0.4084 - val_loss: 0.4115 - val_mean_squared_error: 0.4115\n",
      "Epoch 236/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.3984 - mean_squared_error: 0.3984 - val_loss: 0.4200 - val_mean_squared_error: 0.4200\n",
      "Epoch 237/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3941 - mean_squared_error: 0.3941 - val_loss: 0.4200 - val_mean_squared_error: 0.4200\n",
      "Epoch 238/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3922 - mean_squared_error: 0.3922 - val_loss: 0.4045 - val_mean_squared_error: 0.4045\n",
      "Epoch 239/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3891 - mean_squared_error: 0.3891 - val_loss: 0.4140 - val_mean_squared_error: 0.4140\n",
      "Epoch 240/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.4011 - mean_squared_error: 0.4011 - val_loss: 0.4161 - val_mean_squared_error: 0.4161\n",
      "Epoch 241/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.3984 - mean_squared_error: 0.3984 - val_loss: 0.4017 - val_mean_squared_error: 0.4017\n",
      "Epoch 242/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.3925 - mean_squared_error: 0.3925 - val_loss: 0.4037 - val_mean_squared_error: 0.4037\n",
      "Epoch 243/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3873 - mean_squared_error: 0.3873 - val_loss: 0.4095 - val_mean_squared_error: 0.4095\n",
      "Epoch 244/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3955 - mean_squared_error: 0.3955 - val_loss: 0.3953 - val_mean_squared_error: 0.3953\n",
      "Epoch 245/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3865 - mean_squared_error: 0.3865 - val_loss: 0.4014 - val_mean_squared_error: 0.4014\n",
      "Epoch 246/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3820 - mean_squared_error: 0.3820 - val_loss: 0.4051 - val_mean_squared_error: 0.4051\n",
      "Epoch 247/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3862 - mean_squared_error: 0.3862 - val_loss: 0.3969 - val_mean_squared_error: 0.3969\n",
      "Epoch 248/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3881 - mean_squared_error: 0.3881 - val_loss: 0.4097 - val_mean_squared_error: 0.4097\n",
      "Epoch 249/500\n",
      "7/7 [==============================] - 0s 34ms/step - loss: 0.3893 - mean_squared_error: 0.3893 - val_loss: 0.4134 - val_mean_squared_error: 0.4134\n",
      "Epoch 250/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3876 - mean_squared_error: 0.3876 - val_loss: 0.4017 - val_mean_squared_error: 0.4017\n",
      "Epoch 251/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.3850 - mean_squared_error: 0.3850 - val_loss: 0.3960 - val_mean_squared_error: 0.3960\n",
      "Epoch 252/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.3836 - mean_squared_error: 0.3836 - val_loss: 0.4078 - val_mean_squared_error: 0.4078\n",
      "Epoch 253/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.3871 - mean_squared_error: 0.3871 - val_loss: 0.3979 - val_mean_squared_error: 0.3979\n",
      "Epoch 254/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3920 - mean_squared_error: 0.3920 - val_loss: 0.3997 - val_mean_squared_error: 0.3997\n",
      "Epoch 255/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3984 - mean_squared_error: 0.3984 - val_loss: 0.4104 - val_mean_squared_error: 0.4104\n",
      "Epoch 256/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3909 - mean_squared_error: 0.3909 - val_loss: 0.4030 - val_mean_squared_error: 0.4030\n",
      "Epoch 257/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.3904 - mean_squared_error: 0.3904 - val_loss: 0.4114 - val_mean_squared_error: 0.4114\n",
      "Epoch 258/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.3878 - mean_squared_error: 0.3878 - val_loss: 0.3999 - val_mean_squared_error: 0.3999\n",
      "Epoch 259/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.3847 - mean_squared_error: 0.3847 - val_loss: 0.3949 - val_mean_squared_error: 0.3949\n",
      "Epoch 260/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.3803 - mean_squared_error: 0.3803 - val_loss: 0.3947 - val_mean_squared_error: 0.3947\n",
      "Epoch 261/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3803 - mean_squared_error: 0.3803 - val_loss: 0.3917 - val_mean_squared_error: 0.3917\n",
      "Epoch 262/500\n",
      "7/7 [==============================] - 0s 25ms/step - loss: 0.3827 - mean_squared_error: 0.3827 - val_loss: 0.4028 - val_mean_squared_error: 0.4028\n",
      "Epoch 263/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.3913 - mean_squared_error: 0.3913 - val_loss: 0.4295 - val_mean_squared_error: 0.4295\n",
      "Epoch 264/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3947 - mean_squared_error: 0.3947 - val_loss: 0.4011 - val_mean_squared_error: 0.4011\n",
      "Epoch 265/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3971 - mean_squared_error: 0.3971 - val_loss: 0.4023 - val_mean_squared_error: 0.4023\n",
      "Epoch 266/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.3944 - mean_squared_error: 0.3944 - val_loss: 0.4059 - val_mean_squared_error: 0.4059\n",
      "Epoch 267/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.4011 - mean_squared_error: 0.4011 - val_loss: 0.4117 - val_mean_squared_error: 0.4117\n",
      "Epoch 268/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3983 - mean_squared_error: 0.3983 - val_loss: 0.4010 - val_mean_squared_error: 0.4010\n",
      "Epoch 269/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3971 - mean_squared_error: 0.3971 - val_loss: 0.4037 - val_mean_squared_error: 0.4037\n",
      "Epoch 270/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3905 - mean_squared_error: 0.3905 - val_loss: 0.4212 - val_mean_squared_error: 0.4212\n",
      "Epoch 271/500\n",
      "7/7 [==============================] - 0s 36ms/step - loss: 0.3948 - mean_squared_error: 0.3948 - val_loss: 0.4099 - val_mean_squared_error: 0.4099\n",
      "Epoch 272/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3932 - mean_squared_error: 0.3932 - val_loss: 0.4102 - val_mean_squared_error: 0.4102\n",
      "Epoch 273/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3887 - mean_squared_error: 0.3887 - val_loss: 0.3932 - val_mean_squared_error: 0.3932\n",
      "Epoch 274/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3820 - mean_squared_error: 0.3820 - val_loss: 0.4031 - val_mean_squared_error: 0.4031\n",
      "Epoch 275/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3889 - mean_squared_error: 0.3889 - val_loss: 0.4208 - val_mean_squared_error: 0.4208\n",
      "Epoch 276/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.4164 - mean_squared_error: 0.4164 - val_loss: 0.4291 - val_mean_squared_error: 0.4291\n",
      "Epoch 277/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.4084 - mean_squared_error: 0.4084 - val_loss: 0.4432 - val_mean_squared_error: 0.4432\n",
      "Epoch 278/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3993 - mean_squared_error: 0.3993 - val_loss: 0.4155 - val_mean_squared_error: 0.4155\n",
      "Epoch 279/500\n",
      "7/7 [==============================] - 0s 42ms/step - loss: 0.3947 - mean_squared_error: 0.3947 - val_loss: 0.4194 - val_mean_squared_error: 0.4194\n",
      "Epoch 280/500\n",
      "7/7 [==============================] - 0s 38ms/step - loss: 0.3961 - mean_squared_error: 0.3961 - val_loss: 0.4016 - val_mean_squared_error: 0.4016\n",
      "Epoch 281/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3974 - mean_squared_error: 0.3974 - val_loss: 0.4210 - val_mean_squared_error: 0.4210\n",
      "Epoch 282/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3963 - mean_squared_error: 0.3963 - val_loss: 0.3967 - val_mean_squared_error: 0.3967\n",
      "Epoch 283/500\n",
      "7/7 [==============================] - 0s 47ms/step - loss: 0.3884 - mean_squared_error: 0.3884 - val_loss: 0.3971 - val_mean_squared_error: 0.3971\n",
      "Epoch 284/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3832 - mean_squared_error: 0.3832 - val_loss: 0.3988 - val_mean_squared_error: 0.3988\n",
      "Epoch 285/500\n",
      "7/7 [==============================] - 0s 40ms/step - loss: 0.3829 - mean_squared_error: 0.3829 - val_loss: 0.3973 - val_mean_squared_error: 0.3973\n",
      "Epoch 286/500\n",
      "7/7 [==============================] - 0s 38ms/step - loss: 0.3850 - mean_squared_error: 0.3850 - val_loss: 0.4094 - val_mean_squared_error: 0.4094\n",
      "Epoch 287/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3910 - mean_squared_error: 0.3910 - val_loss: 0.3998 - val_mean_squared_error: 0.3998\n",
      "Epoch 288/500\n",
      "7/7 [==============================] - 0s 38ms/step - loss: 0.3895 - mean_squared_error: 0.3895 - val_loss: 0.3924 - val_mean_squared_error: 0.3924\n",
      "Epoch 289/500\n",
      "7/7 [==============================] - 0s 36ms/step - loss: 0.3873 - mean_squared_error: 0.3873 - val_loss: 0.4044 - val_mean_squared_error: 0.4044\n",
      "Epoch 290/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3882 - mean_squared_error: 0.3882 - val_loss: 0.3993 - val_mean_squared_error: 0.3993\n",
      "Epoch 291/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3904 - mean_squared_error: 0.3904 - val_loss: 0.4074 - val_mean_squared_error: 0.4074\n",
      "Epoch 292/500\n",
      "7/7 [==============================] - 0s 25ms/step - loss: 0.3975 - mean_squared_error: 0.3975 - val_loss: 0.4060 - val_mean_squared_error: 0.4060\n",
      "Epoch 293/500\n",
      "7/7 [==============================] - 0s 36ms/step - loss: 0.3931 - mean_squared_error: 0.3931 - val_loss: 0.4281 - val_mean_squared_error: 0.4281\n",
      "Epoch 294/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.3989 - mean_squared_error: 0.3989 - val_loss: 0.4060 - val_mean_squared_error: 0.4060\n",
      "Epoch 295/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3928 - mean_squared_error: 0.3928 - val_loss: 0.4005 - val_mean_squared_error: 0.4005\n",
      "Epoch 296/500\n",
      "7/7 [==============================] - 0s 36ms/step - loss: 0.3879 - mean_squared_error: 0.3879 - val_loss: 0.4191 - val_mean_squared_error: 0.4191\n",
      "Epoch 297/500\n",
      "7/7 [==============================] - 0s 44ms/step - loss: 0.3983 - mean_squared_error: 0.3983 - val_loss: 0.4089 - val_mean_squared_error: 0.4089\n",
      "Epoch 298/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3848 - mean_squared_error: 0.3848 - val_loss: 0.3928 - val_mean_squared_error: 0.3928\n",
      "Epoch 299/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3789 - mean_squared_error: 0.3789 - val_loss: 0.3983 - val_mean_squared_error: 0.3983\n",
      "Epoch 300/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3809 - mean_squared_error: 0.3809 - val_loss: 0.3918 - val_mean_squared_error: 0.3918\n",
      "Epoch 301/500\n",
      "7/7 [==============================] - 0s 37ms/step - loss: 0.3780 - mean_squared_error: 0.3780 - val_loss: 0.3893 - val_mean_squared_error: 0.3893\n",
      "Epoch 302/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3812 - mean_squared_error: 0.3812 - val_loss: 0.3998 - val_mean_squared_error: 0.3998\n",
      "Epoch 303/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3960 - mean_squared_error: 0.3960 - val_loss: 0.4560 - val_mean_squared_error: 0.4560\n",
      "Epoch 304/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.4237 - mean_squared_error: 0.4237 - val_loss: 0.4277 - val_mean_squared_error: 0.4277\n",
      "Epoch 305/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.4140 - mean_squared_error: 0.4140 - val_loss: 0.4314 - val_mean_squared_error: 0.4314\n",
      "Epoch 306/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.4019 - mean_squared_error: 0.4019 - val_loss: 0.4064 - val_mean_squared_error: 0.4064\n",
      "Epoch 307/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3913 - mean_squared_error: 0.3913 - val_loss: 0.3950 - val_mean_squared_error: 0.3950\n",
      "Epoch 308/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.3850 - mean_squared_error: 0.3850 - val_loss: 0.4013 - val_mean_squared_error: 0.4013\n",
      "Epoch 309/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3837 - mean_squared_error: 0.3837 - val_loss: 0.4011 - val_mean_squared_error: 0.4011\n",
      "Epoch 310/500\n",
      "7/7 [==============================] - 0s 26ms/step - loss: 0.3863 - mean_squared_error: 0.3863 - val_loss: 0.4060 - val_mean_squared_error: 0.4060\n",
      "Epoch 311/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3877 - mean_squared_error: 0.3877 - val_loss: 0.3917 - val_mean_squared_error: 0.3917\n",
      "Epoch 312/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3892 - mean_squared_error: 0.3892 - val_loss: 0.4123 - val_mean_squared_error: 0.4123\n",
      "Epoch 313/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3941 - mean_squared_error: 0.3941 - val_loss: 0.3986 - val_mean_squared_error: 0.3986\n",
      "Epoch 314/500\n",
      "7/7 [==============================] - 1s 81ms/step - loss: 0.3884 - mean_squared_error: 0.3884 - val_loss: 0.4069 - val_mean_squared_error: 0.4069\n",
      "Epoch 315/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3978 - mean_squared_error: 0.3978 - val_loss: 0.4055 - val_mean_squared_error: 0.4055\n",
      "Epoch 316/500\n",
      "7/7 [==============================] - 0s 47ms/step - loss: 0.3938 - mean_squared_error: 0.3938 - val_loss: 0.4124 - val_mean_squared_error: 0.4124\n",
      "Epoch 317/500\n",
      "7/7 [==============================] - 0s 38ms/step - loss: 0.3943 - mean_squared_error: 0.3943 - val_loss: 0.4211 - val_mean_squared_error: 0.4211\n",
      "Epoch 318/500\n",
      "7/7 [==============================] - 0s 47ms/step - loss: 0.3912 - mean_squared_error: 0.3912 - val_loss: 0.4259 - val_mean_squared_error: 0.4259\n",
      "Epoch 319/500\n",
      "7/7 [==============================] - 0s 44ms/step - loss: 0.3954 - mean_squared_error: 0.3954 - val_loss: 0.4051 - val_mean_squared_error: 0.4051\n",
      "Epoch 320/500\n",
      "7/7 [==============================] - 0s 38ms/step - loss: 0.3857 - mean_squared_error: 0.3857 - val_loss: 0.3948 - val_mean_squared_error: 0.3948\n",
      "Epoch 321/500\n",
      "7/7 [==============================] - 0s 44ms/step - loss: 0.3824 - mean_squared_error: 0.3824 - val_loss: 0.4044 - val_mean_squared_error: 0.4044\n",
      "Epoch 322/500\n",
      "7/7 [==============================] - 0s 38ms/step - loss: 0.3798 - mean_squared_error: 0.3798 - val_loss: 0.3928 - val_mean_squared_error: 0.3928\n",
      "Epoch 323/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3767 - mean_squared_error: 0.3767 - val_loss: 0.3891 - val_mean_squared_error: 0.3891\n",
      "Epoch 324/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3776 - mean_squared_error: 0.3776 - val_loss: 0.3942 - val_mean_squared_error: 0.3942\n",
      "Epoch 325/500\n",
      "7/7 [==============================] - 0s 46ms/step - loss: 0.3841 - mean_squared_error: 0.3841 - val_loss: 0.3993 - val_mean_squared_error: 0.3993\n",
      "Epoch 326/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3869 - mean_squared_error: 0.3869 - val_loss: 0.4114 - val_mean_squared_error: 0.4114\n",
      "Epoch 327/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3832 - mean_squared_error: 0.3832 - val_loss: 0.3902 - val_mean_squared_error: 0.3902\n",
      "Epoch 328/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3836 - mean_squared_error: 0.3836 - val_loss: 0.3919 - val_mean_squared_error: 0.3919\n",
      "Epoch 329/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3784 - mean_squared_error: 0.3784 - val_loss: 0.3971 - val_mean_squared_error: 0.3971\n",
      "Epoch 330/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3788 - mean_squared_error: 0.3788 - val_loss: 0.3951 - val_mean_squared_error: 0.3951\n",
      "Epoch 331/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3772 - mean_squared_error: 0.3772 - val_loss: 0.3925 - val_mean_squared_error: 0.3925\n",
      "Epoch 332/500\n",
      "7/7 [==============================] - 0s 45ms/step - loss: 0.3744 - mean_squared_error: 0.3744 - val_loss: 0.3828 - val_mean_squared_error: 0.3828\n",
      "Epoch 333/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3826 - mean_squared_error: 0.3826 - val_loss: 0.3903 - val_mean_squared_error: 0.3903\n",
      "Epoch 334/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3845 - mean_squared_error: 0.3845 - val_loss: 0.4104 - val_mean_squared_error: 0.4104\n",
      "Epoch 335/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3827 - mean_squared_error: 0.3827 - val_loss: 0.4094 - val_mean_squared_error: 0.4094\n",
      "Epoch 336/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3798 - mean_squared_error: 0.3798 - val_loss: 0.3862 - val_mean_squared_error: 0.3862\n",
      "Epoch 337/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3731 - mean_squared_error: 0.3731 - val_loss: 0.3910 - val_mean_squared_error: 0.3910\n",
      "Epoch 338/500\n",
      "7/7 [==============================] - 0s 45ms/step - loss: 0.3717 - mean_squared_error: 0.3717 - val_loss: 0.3856 - val_mean_squared_error: 0.3856\n",
      "Epoch 339/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3696 - mean_squared_error: 0.3696 - val_loss: 0.3860 - val_mean_squared_error: 0.3860\n",
      "Epoch 340/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3753 - mean_squared_error: 0.3753 - val_loss: 0.3956 - val_mean_squared_error: 0.3956\n",
      "Epoch 341/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3733 - mean_squared_error: 0.3733 - val_loss: 0.3958 - val_mean_squared_error: 0.3958\n",
      "Epoch 342/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3909 - mean_squared_error: 0.3909 - val_loss: 0.3984 - val_mean_squared_error: 0.3984\n",
      "Epoch 343/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.3865 - mean_squared_error: 0.3865 - val_loss: 0.3939 - val_mean_squared_error: 0.3939\n",
      "Epoch 344/500\n",
      "7/7 [==============================] - 0s 57ms/step - loss: 0.3795 - mean_squared_error: 0.3795 - val_loss: 0.3964 - val_mean_squared_error: 0.3964\n",
      "Epoch 345/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3772 - mean_squared_error: 0.3772 - val_loss: 0.3942 - val_mean_squared_error: 0.3942\n",
      "Epoch 346/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3754 - mean_squared_error: 0.3754 - val_loss: 0.3973 - val_mean_squared_error: 0.3973\n",
      "Epoch 347/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3746 - mean_squared_error: 0.3746 - val_loss: 0.3855 - val_mean_squared_error: 0.3855\n",
      "Epoch 348/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3735 - mean_squared_error: 0.3735 - val_loss: 0.3913 - val_mean_squared_error: 0.3913\n",
      "Epoch 349/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3746 - mean_squared_error: 0.3746 - val_loss: 0.4077 - val_mean_squared_error: 0.4077\n",
      "Epoch 350/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3779 - mean_squared_error: 0.3779 - val_loss: 0.3929 - val_mean_squared_error: 0.3929\n",
      "Epoch 351/500\n",
      "7/7 [==============================] - 0s 43ms/step - loss: 0.3741 - mean_squared_error: 0.3741 - val_loss: 0.3962 - val_mean_squared_error: 0.3962\n",
      "Epoch 352/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3787 - mean_squared_error: 0.3787 - val_loss: 0.3978 - val_mean_squared_error: 0.3978\n",
      "Epoch 353/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3819 - mean_squared_error: 0.3819 - val_loss: 0.4000 - val_mean_squared_error: 0.4000\n",
      "Epoch 354/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3806 - mean_squared_error: 0.3806 - val_loss: 0.3906 - val_mean_squared_error: 0.3906\n",
      "Epoch 355/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3776 - mean_squared_error: 0.3776 - val_loss: 0.3978 - val_mean_squared_error: 0.3978\n",
      "Epoch 356/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3805 - mean_squared_error: 0.3805 - val_loss: 0.3978 - val_mean_squared_error: 0.3978\n",
      "Epoch 357/500\n",
      "7/7 [==============================] - 0s 46ms/step - loss: 0.3765 - mean_squared_error: 0.3765 - val_loss: 0.3881 - val_mean_squared_error: 0.3881\n",
      "Epoch 358/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3709 - mean_squared_error: 0.3709 - val_loss: 0.3857 - val_mean_squared_error: 0.3857\n",
      "Epoch 359/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3693 - mean_squared_error: 0.3693 - val_loss: 0.3846 - val_mean_squared_error: 0.3846\n",
      "Epoch 360/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3678 - mean_squared_error: 0.3678 - val_loss: 0.3830 - val_mean_squared_error: 0.3830\n",
      "Epoch 361/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.3675 - mean_squared_error: 0.3675 - val_loss: 0.4049 - val_mean_squared_error: 0.4049\n",
      "Epoch 362/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.3701 - mean_squared_error: 0.3701 - val_loss: 0.3990 - val_mean_squared_error: 0.3990\n",
      "Epoch 363/500\n",
      "7/7 [==============================] - 0s 44ms/step - loss: 0.3760 - mean_squared_error: 0.3760 - val_loss: 0.4022 - val_mean_squared_error: 0.4022\n",
      "Epoch 364/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3770 - mean_squared_error: 0.3770 - val_loss: 0.4002 - val_mean_squared_error: 0.4002\n",
      "Epoch 365/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.3753 - mean_squared_error: 0.3753 - val_loss: 0.3978 - val_mean_squared_error: 0.3978\n",
      "Epoch 366/500\n",
      "7/7 [==============================] - 0s 40ms/step - loss: 0.3859 - mean_squared_error: 0.3859 - val_loss: 0.3940 - val_mean_squared_error: 0.3940\n",
      "Epoch 367/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3876 - mean_squared_error: 0.3876 - val_loss: 0.4028 - val_mean_squared_error: 0.4028\n",
      "Epoch 368/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3818 - mean_squared_error: 0.3818 - val_loss: 0.3901 - val_mean_squared_error: 0.3901\n",
      "Epoch 369/500\n",
      "7/7 [==============================] - 0s 40ms/step - loss: 0.3755 - mean_squared_error: 0.3755 - val_loss: 0.3889 - val_mean_squared_error: 0.3889\n",
      "Epoch 370/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3727 - mean_squared_error: 0.3727 - val_loss: 0.3929 - val_mean_squared_error: 0.3929\n",
      "Epoch 371/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3731 - mean_squared_error: 0.3731 - val_loss: 0.3807 - val_mean_squared_error: 0.3807\n",
      "Epoch 372/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3683 - mean_squared_error: 0.3683 - val_loss: 0.3901 - val_mean_squared_error: 0.3901\n",
      "Epoch 373/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.3765 - mean_squared_error: 0.3765 - val_loss: 0.3934 - val_mean_squared_error: 0.3934\n",
      "Epoch 374/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3723 - mean_squared_error: 0.3723 - val_loss: 0.3866 - val_mean_squared_error: 0.3866\n",
      "Epoch 375/500\n",
      "7/7 [==============================] - 0s 46ms/step - loss: 0.3701 - mean_squared_error: 0.3701 - val_loss: 0.3888 - val_mean_squared_error: 0.3888\n",
      "Epoch 376/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.3675 - mean_squared_error: 0.3675 - val_loss: 0.3874 - val_mean_squared_error: 0.3874\n",
      "Epoch 377/500\n",
      "7/7 [==============================] - 0s 36ms/step - loss: 0.3667 - mean_squared_error: 0.3667 - val_loss: 0.3840 - val_mean_squared_error: 0.3840\n",
      "Epoch 378/500\n",
      "7/7 [==============================] - 0s 37ms/step - loss: 0.3678 - mean_squared_error: 0.3678 - val_loss: 0.3818 - val_mean_squared_error: 0.3818\n",
      "Epoch 379/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3693 - mean_squared_error: 0.3693 - val_loss: 0.3910 - val_mean_squared_error: 0.3910\n",
      "Epoch 380/500\n",
      "7/7 [==============================] - 0s 51ms/step - loss: 0.3691 - mean_squared_error: 0.3691 - val_loss: 0.3811 - val_mean_squared_error: 0.3811\n",
      "Epoch 381/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3642 - mean_squared_error: 0.3642 - val_loss: 0.3913 - val_mean_squared_error: 0.3913\n",
      "Epoch 382/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3676 - mean_squared_error: 0.3676 - val_loss: 0.4017 - val_mean_squared_error: 0.4017\n",
      "Epoch 383/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.3788 - mean_squared_error: 0.3788 - val_loss: 0.4036 - val_mean_squared_error: 0.4036\n",
      "Epoch 384/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3725 - mean_squared_error: 0.3725 - val_loss: 0.3959 - val_mean_squared_error: 0.3959\n",
      "Epoch 385/500\n",
      "7/7 [==============================] - 0s 39ms/step - loss: 0.3746 - mean_squared_error: 0.3746 - val_loss: 0.3955 - val_mean_squared_error: 0.3955\n",
      "Epoch 386/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.3737 - mean_squared_error: 0.3737 - val_loss: 0.3841 - val_mean_squared_error: 0.3841\n",
      "Epoch 387/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3751 - mean_squared_error: 0.3751 - val_loss: 0.3929 - val_mean_squared_error: 0.3929\n",
      "Epoch 388/500\n",
      "7/7 [==============================] - 0s 40ms/step - loss: 0.3772 - mean_squared_error: 0.3772 - val_loss: 0.3858 - val_mean_squared_error: 0.3858\n",
      "Epoch 389/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3801 - mean_squared_error: 0.3801 - val_loss: 0.3938 - val_mean_squared_error: 0.3938\n",
      "Epoch 390/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.3839 - mean_squared_error: 0.3839 - val_loss: 0.4035 - val_mean_squared_error: 0.4035\n",
      "Epoch 391/500\n",
      "7/7 [==============================] - 0s 37ms/step - loss: 0.3779 - mean_squared_error: 0.3779 - val_loss: 0.4022 - val_mean_squared_error: 0.4022\n",
      "Epoch 392/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3746 - mean_squared_error: 0.3746 - val_loss: 0.4011 - val_mean_squared_error: 0.4011\n",
      "Epoch 393/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.3721 - mean_squared_error: 0.3721 - val_loss: 0.3942 - val_mean_squared_error: 0.3942\n",
      "Epoch 394/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3690 - mean_squared_error: 0.3690 - val_loss: 0.3915 - val_mean_squared_error: 0.3915\n",
      "Epoch 395/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3651 - mean_squared_error: 0.3651 - val_loss: 0.3806 - val_mean_squared_error: 0.3806\n",
      "Epoch 396/500\n",
      "7/7 [==============================] - 0s 43ms/step - loss: 0.3678 - mean_squared_error: 0.3678 - val_loss: 0.3862 - val_mean_squared_error: 0.3862\n",
      "Epoch 397/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3646 - mean_squared_error: 0.3646 - val_loss: 0.3872 - val_mean_squared_error: 0.3872\n",
      "Epoch 398/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3671 - mean_squared_error: 0.3671 - val_loss: 0.4000 - val_mean_squared_error: 0.4000\n",
      "Epoch 399/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.3729 - mean_squared_error: 0.3729 - val_loss: 0.3875 - val_mean_squared_error: 0.3875\n",
      "Epoch 400/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3698 - mean_squared_error: 0.3698 - val_loss: 0.4115 - val_mean_squared_error: 0.4115\n",
      "Epoch 401/500\n",
      "7/7 [==============================] - 0s 45ms/step - loss: 0.3782 - mean_squared_error: 0.3782 - val_loss: 0.4006 - val_mean_squared_error: 0.4006\n",
      "Epoch 402/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3717 - mean_squared_error: 0.3717 - val_loss: 0.3983 - val_mean_squared_error: 0.3983\n",
      "Epoch 403/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3686 - mean_squared_error: 0.3686 - val_loss: 0.3869 - val_mean_squared_error: 0.3869\n",
      "Epoch 404/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3649 - mean_squared_error: 0.3649 - val_loss: 0.3848 - val_mean_squared_error: 0.3848\n",
      "Epoch 405/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3671 - mean_squared_error: 0.3671 - val_loss: 0.3868 - val_mean_squared_error: 0.3868\n",
      "Epoch 406/500\n",
      "7/7 [==============================] - 0s 45ms/step - loss: 0.3736 - mean_squared_error: 0.3736 - val_loss: 0.3908 - val_mean_squared_error: 0.3908\n",
      "Epoch 407/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.3800 - mean_squared_error: 0.3800 - val_loss: 0.4034 - val_mean_squared_error: 0.4034\n",
      "Epoch 408/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3848 - mean_squared_error: 0.3848 - val_loss: 0.3921 - val_mean_squared_error: 0.3921\n",
      "Epoch 409/500\n",
      "7/7 [==============================] - 0s 41ms/step - loss: 0.3757 - mean_squared_error: 0.3757 - val_loss: 0.3890 - val_mean_squared_error: 0.3890\n",
      "Epoch 410/500\n",
      "7/7 [==============================] - 0s 43ms/step - loss: 0.3776 - mean_squared_error: 0.3776 - val_loss: 0.3888 - val_mean_squared_error: 0.3888\n",
      "Epoch 411/500\n",
      "7/7 [==============================] - 0s 47ms/step - loss: 0.3788 - mean_squared_error: 0.3788 - val_loss: 0.3902 - val_mean_squared_error: 0.3902\n",
      "Epoch 412/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3732 - mean_squared_error: 0.3732 - val_loss: 0.3827 - val_mean_squared_error: 0.3827\n",
      "Epoch 413/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3723 - mean_squared_error: 0.3723 - val_loss: 0.3970 - val_mean_squared_error: 0.3970\n",
      "Epoch 414/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3801 - mean_squared_error: 0.3801 - val_loss: 0.3946 - val_mean_squared_error: 0.3946\n",
      "Epoch 415/500\n",
      "7/7 [==============================] - 0s 43ms/step - loss: 0.3808 - mean_squared_error: 0.3808 - val_loss: 0.4057 - val_mean_squared_error: 0.4057\n",
      "Epoch 416/500\n",
      "7/7 [==============================] - 0s 37ms/step - loss: 0.3778 - mean_squared_error: 0.3778 - val_loss: 0.3906 - val_mean_squared_error: 0.3906\n",
      "Epoch 417/500\n",
      "7/7 [==============================] - 0s 43ms/step - loss: 0.3729 - mean_squared_error: 0.3729 - val_loss: 0.3969 - val_mean_squared_error: 0.3969\n",
      "Epoch 418/500\n",
      "7/7 [==============================] - 0s 36ms/step - loss: 0.3714 - mean_squared_error: 0.3714 - val_loss: 0.3822 - val_mean_squared_error: 0.3822\n",
      "Epoch 419/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3634 - mean_squared_error: 0.3634 - val_loss: 0.3857 - val_mean_squared_error: 0.3857\n",
      "Epoch 420/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3659 - mean_squared_error: 0.3659 - val_loss: 0.3755 - val_mean_squared_error: 0.3755\n",
      "Epoch 421/500\n",
      "7/7 [==============================] - 0s 48ms/step - loss: 0.3644 - mean_squared_error: 0.3644 - val_loss: 0.3835 - val_mean_squared_error: 0.3835\n",
      "Epoch 422/500\n",
      "7/7 [==============================] - 0s 34ms/step - loss: 0.3656 - mean_squared_error: 0.3656 - val_loss: 0.3763 - val_mean_squared_error: 0.3763\n",
      "Epoch 423/500\n",
      "7/7 [==============================] - 0s 28ms/step - loss: 0.3611 - mean_squared_error: 0.3611 - val_loss: 0.3762 - val_mean_squared_error: 0.3762\n",
      "Epoch 424/500\n",
      "7/7 [==============================] - 0s 42ms/step - loss: 0.3640 - mean_squared_error: 0.3640 - val_loss: 0.4001 - val_mean_squared_error: 0.4001\n",
      "Epoch 425/500\n",
      "7/7 [==============================] - 0s 46ms/step - loss: 0.3655 - mean_squared_error: 0.3655 - val_loss: 0.3867 - val_mean_squared_error: 0.3867\n",
      "Epoch 426/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3633 - mean_squared_error: 0.3633 - val_loss: 0.3879 - val_mean_squared_error: 0.3879\n",
      "Epoch 427/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3642 - mean_squared_error: 0.3642 - val_loss: 0.3978 - val_mean_squared_error: 0.3978\n",
      "Epoch 428/500\n",
      "7/7 [==============================] - 0s 36ms/step - loss: 0.3711 - mean_squared_error: 0.3711 - val_loss: 0.3868 - val_mean_squared_error: 0.3868\n",
      "Epoch 429/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3711 - mean_squared_error: 0.3711 - val_loss: 0.3863 - val_mean_squared_error: 0.3863\n",
      "Epoch 430/500\n",
      "7/7 [==============================] - 0s 43ms/step - loss: 0.3716 - mean_squared_error: 0.3716 - val_loss: 0.3795 - val_mean_squared_error: 0.3795\n",
      "Epoch 431/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3659 - mean_squared_error: 0.3659 - val_loss: 0.3827 - val_mean_squared_error: 0.3827\n",
      "Epoch 432/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3642 - mean_squared_error: 0.3642 - val_loss: 0.3805 - val_mean_squared_error: 0.3805\n",
      "Epoch 433/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3684 - mean_squared_error: 0.3684 - val_loss: 0.3870 - val_mean_squared_error: 0.3870\n",
      "Epoch 434/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3663 - mean_squared_error: 0.3663 - val_loss: 0.3773 - val_mean_squared_error: 0.3773\n",
      "Epoch 435/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3630 - mean_squared_error: 0.3630 - val_loss: 0.3908 - val_mean_squared_error: 0.3908\n",
      "Epoch 436/500\n",
      "7/7 [==============================] - 0s 38ms/step - loss: 0.3618 - mean_squared_error: 0.3618 - val_loss: 0.4007 - val_mean_squared_error: 0.4007\n",
      "Epoch 437/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3672 - mean_squared_error: 0.3672 - val_loss: 0.3975 - val_mean_squared_error: 0.3975\n",
      "Epoch 438/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3721 - mean_squared_error: 0.3721 - val_loss: 0.3794 - val_mean_squared_error: 0.3794\n",
      "Epoch 439/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3701 - mean_squared_error: 0.3701 - val_loss: 0.3813 - val_mean_squared_error: 0.3813\n",
      "Epoch 440/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3664 - mean_squared_error: 0.3664 - val_loss: 0.3777 - val_mean_squared_error: 0.3777\n",
      "Epoch 441/500\n",
      "7/7 [==============================] - 0s 40ms/step - loss: 0.3647 - mean_squared_error: 0.3647 - val_loss: 0.3807 - val_mean_squared_error: 0.3807\n",
      "Epoch 442/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3618 - mean_squared_error: 0.3618 - val_loss: 0.3811 - val_mean_squared_error: 0.3811\n",
      "Epoch 443/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3626 - mean_squared_error: 0.3626 - val_loss: 0.3879 - val_mean_squared_error: 0.3879\n",
      "Epoch 444/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3704 - mean_squared_error: 0.3704 - val_loss: 0.3881 - val_mean_squared_error: 0.3881\n",
      "Epoch 445/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3682 - mean_squared_error: 0.3682 - val_loss: 0.3797 - val_mean_squared_error: 0.3797\n",
      "Epoch 446/500\n",
      "7/7 [==============================] - 0s 41ms/step - loss: 0.3640 - mean_squared_error: 0.3640 - val_loss: 0.3781 - val_mean_squared_error: 0.3781\n",
      "Epoch 447/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3623 - mean_squared_error: 0.3623 - val_loss: 0.3917 - val_mean_squared_error: 0.3917\n",
      "Epoch 448/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3680 - mean_squared_error: 0.3680 - val_loss: 0.3807 - val_mean_squared_error: 0.3807\n",
      "Epoch 449/500\n",
      "7/7 [==============================] - 0s 36ms/step - loss: 0.3642 - mean_squared_error: 0.3642 - val_loss: 0.3832 - val_mean_squared_error: 0.3832\n",
      "Epoch 450/500\n",
      "7/7 [==============================] - 0s 47ms/step - loss: 0.3655 - mean_squared_error: 0.3655 - val_loss: 0.3826 - val_mean_squared_error: 0.3826\n",
      "Epoch 451/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.3645 - mean_squared_error: 0.3645 - val_loss: 0.3803 - val_mean_squared_error: 0.3803\n",
      "Epoch 452/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3613 - mean_squared_error: 0.3613 - val_loss: 0.3860 - val_mean_squared_error: 0.3860\n",
      "Epoch 453/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3624 - mean_squared_error: 0.3624 - val_loss: 0.3856 - val_mean_squared_error: 0.3856\n",
      "Epoch 454/500\n",
      "7/7 [==============================] - 0s 36ms/step - loss: 0.3640 - mean_squared_error: 0.3640 - val_loss: 0.3921 - val_mean_squared_error: 0.3921\n",
      "Epoch 455/500\n",
      "7/7 [==============================] - 0s 34ms/step - loss: 0.3691 - mean_squared_error: 0.3691 - val_loss: 0.3907 - val_mean_squared_error: 0.3907\n",
      "Epoch 456/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3719 - mean_squared_error: 0.3719 - val_loss: 0.3796 - val_mean_squared_error: 0.3796\n",
      "Epoch 457/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3643 - mean_squared_error: 0.3643 - val_loss: 0.3807 - val_mean_squared_error: 0.3807\n",
      "Epoch 458/500\n",
      "7/7 [==============================] - 0s 37ms/step - loss: 0.3577 - mean_squared_error: 0.3577 - val_loss: 0.3773 - val_mean_squared_error: 0.3773\n",
      "Epoch 459/500\n",
      "7/7 [==============================] - 0s 45ms/step - loss: 0.3603 - mean_squared_error: 0.3603 - val_loss: 0.3774 - val_mean_squared_error: 0.3774\n",
      "Epoch 460/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3681 - mean_squared_error: 0.3681 - val_loss: 0.4159 - val_mean_squared_error: 0.4159\n",
      "Epoch 461/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3748 - mean_squared_error: 0.3748 - val_loss: 0.3923 - val_mean_squared_error: 0.3923\n",
      "Epoch 462/500\n",
      "7/7 [==============================] - 0s 36ms/step - loss: 0.3642 - mean_squared_error: 0.3642 - val_loss: 0.3743 - val_mean_squared_error: 0.3743\n",
      "Epoch 463/500\n",
      "7/7 [==============================] - 0s 42ms/step - loss: 0.3600 - mean_squared_error: 0.3600 - val_loss: 0.3833 - val_mean_squared_error: 0.3833\n",
      "Epoch 464/500\n",
      "7/7 [==============================] - 0s 27ms/step - loss: 0.3596 - mean_squared_error: 0.3596 - val_loss: 0.3804 - val_mean_squared_error: 0.3804\n",
      "Epoch 465/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3626 - mean_squared_error: 0.3626 - val_loss: 0.4184 - val_mean_squared_error: 0.4184\n",
      "Epoch 466/500\n",
      "7/7 [==============================] - 0s 37ms/step - loss: 0.3692 - mean_squared_error: 0.3692 - val_loss: 0.3912 - val_mean_squared_error: 0.3912\n",
      "Epoch 467/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.3682 - mean_squared_error: 0.3682 - val_loss: 0.3778 - val_mean_squared_error: 0.3778\n",
      "Epoch 468/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3622 - mean_squared_error: 0.3622 - val_loss: 0.3823 - val_mean_squared_error: 0.3823\n",
      "Epoch 469/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3622 - mean_squared_error: 0.3622 - val_loss: 0.3854 - val_mean_squared_error: 0.3854\n",
      "Epoch 470/500\n",
      "7/7 [==============================] - 0s 37ms/step - loss: 0.3605 - mean_squared_error: 0.3605 - val_loss: 0.3979 - val_mean_squared_error: 0.3979\n",
      "Epoch 471/500\n",
      "7/7 [==============================] - 0s 39ms/step - loss: 0.3615 - mean_squared_error: 0.3615 - val_loss: 0.3741 - val_mean_squared_error: 0.3741\n",
      "Epoch 472/500\n",
      "7/7 [==============================] - 0s 39ms/step - loss: 0.3603 - mean_squared_error: 0.3603 - val_loss: 0.3786 - val_mean_squared_error: 0.3786\n",
      "Epoch 473/500\n",
      "7/7 [==============================] - 0s 38ms/step - loss: 0.3639 - mean_squared_error: 0.3639 - val_loss: 0.3958 - val_mean_squared_error: 0.3958\n",
      "Epoch 474/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.3688 - mean_squared_error: 0.3688 - val_loss: 0.3769 - val_mean_squared_error: 0.3769\n",
      "Epoch 475/500\n",
      "7/7 [==============================] - 0s 39ms/step - loss: 0.3603 - mean_squared_error: 0.3603 - val_loss: 0.3799 - val_mean_squared_error: 0.3799\n",
      "Epoch 476/500\n",
      "7/7 [==============================] - 0s 42ms/step - loss: 0.3625 - mean_squared_error: 0.3625 - val_loss: 0.3907 - val_mean_squared_error: 0.3907\n",
      "Epoch 477/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3617 - mean_squared_error: 0.3617 - val_loss: 0.3910 - val_mean_squared_error: 0.3910\n",
      "Epoch 478/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3682 - mean_squared_error: 0.3682 - val_loss: 0.3765 - val_mean_squared_error: 0.3765\n",
      "Epoch 479/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.3636 - mean_squared_error: 0.3636 - val_loss: 0.3881 - val_mean_squared_error: 0.3881\n",
      "Epoch 480/500\n",
      "7/7 [==============================] - 0s 37ms/step - loss: 0.3680 - mean_squared_error: 0.3680 - val_loss: 0.3716 - val_mean_squared_error: 0.3716\n",
      "Epoch 481/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3613 - mean_squared_error: 0.3613 - val_loss: 0.3857 - val_mean_squared_error: 0.3857\n",
      "Epoch 482/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3646 - mean_squared_error: 0.3646 - val_loss: 0.3839 - val_mean_squared_error: 0.3839\n",
      "Epoch 483/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3682 - mean_squared_error: 0.3682 - val_loss: 0.3933 - val_mean_squared_error: 0.3933\n",
      "Epoch 484/500\n",
      "7/7 [==============================] - 0s 40ms/step - loss: 0.3672 - mean_squared_error: 0.3672 - val_loss: 0.3865 - val_mean_squared_error: 0.3865\n",
      "Epoch 485/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3631 - mean_squared_error: 0.3631 - val_loss: 0.3754 - val_mean_squared_error: 0.3754\n",
      "Epoch 486/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.3559 - mean_squared_error: 0.3559 - val_loss: 0.3826 - val_mean_squared_error: 0.3826\n",
      "Epoch 487/500\n",
      "7/7 [==============================] - 0s 32ms/step - loss: 0.3568 - mean_squared_error: 0.3568 - val_loss: 0.3770 - val_mean_squared_error: 0.3770\n",
      "Epoch 488/500\n",
      "7/7 [==============================] - 0s 33ms/step - loss: 0.3575 - mean_squared_error: 0.3575 - val_loss: 0.3778 - val_mean_squared_error: 0.3778\n",
      "Epoch 489/500\n",
      "7/7 [==============================] - 0s 36ms/step - loss: 0.3562 - mean_squared_error: 0.3562 - val_loss: 0.3779 - val_mean_squared_error: 0.3779\n",
      "Epoch 490/500\n",
      "7/7 [==============================] - 0s 31ms/step - loss: 0.3540 - mean_squared_error: 0.3540 - val_loss: 0.3787 - val_mean_squared_error: 0.3787\n",
      "Epoch 491/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3534 - mean_squared_error: 0.3534 - val_loss: 0.3743 - val_mean_squared_error: 0.3743\n",
      "Epoch 492/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.3553 - mean_squared_error: 0.3553 - val_loss: 0.3755 - val_mean_squared_error: 0.3755\n",
      "Epoch 493/500\n",
      "7/7 [==============================] - 0s 39ms/step - loss: 0.3596 - mean_squared_error: 0.3596 - val_loss: 0.3788 - val_mean_squared_error: 0.3788\n",
      "Epoch 494/500\n",
      "7/7 [==============================] - 0s 29ms/step - loss: 0.3682 - mean_squared_error: 0.3682 - val_loss: 0.3846 - val_mean_squared_error: 0.3846\n",
      "Epoch 495/500\n",
      "7/7 [==============================] - 0s 39ms/step - loss: 0.3659 - mean_squared_error: 0.3659 - val_loss: 0.3801 - val_mean_squared_error: 0.3801\n",
      "Epoch 496/500\n",
      "7/7 [==============================] - 0s 42ms/step - loss: 0.3628 - mean_squared_error: 0.3628 - val_loss: 0.3780 - val_mean_squared_error: 0.3780\n",
      "Epoch 497/500\n",
      "7/7 [==============================] - 0s 35ms/step - loss: 0.3565 - mean_squared_error: 0.3565 - val_loss: 0.3762 - val_mean_squared_error: 0.3762\n",
      "Epoch 498/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3615 - mean_squared_error: 0.3615 - val_loss: 0.3780 - val_mean_squared_error: 0.3780\n",
      "Epoch 499/500\n",
      "7/7 [==============================] - 0s 30ms/step - loss: 0.3575 - mean_squared_error: 0.3575 - val_loss: 0.3812 - val_mean_squared_error: 0.3812\n",
      "Epoch 500/500\n",
      "7/7 [==============================] - 0s 45ms/step - loss: 0.3602 - mean_squared_error: 0.3602 - val_loss: 0.3756 - val_mean_squared_error: 0.3756\n"
     ]
    }
   ],
   "source": [
    "model = create_model('relu', Adam(learning_rate=0.05), 'mse')\n",
    "history = model.fit(X_train, y_train, epochs=500, batch_size=10000, validation_split=0.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "0d255480",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnkAAAHUCAYAAABGYz4MAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/av/WaAAAACXBIWXMAAA9hAAAPYQGoP6dpAACS6UlEQVR4nOzdd3gU1dfA8e9ueq8kISGE3jsCUqUJAiJFFJD2goIiolQFfyqgKCqIDbGggIgC0kSRjoBSBAXpTXqAJBAgve/e948xS0IKKZtsNjmf55knuzN3Zs7s7G7O3pl7r04ppRBCCCGEEKWK3tIBCCGEEEII85MkTwghhBCiFJIkTwghhBCiFJIkTwghhBCiFJIkTwghhBCiFJIkTwghhBCiFJIkTwghhBCiFJIkTwghhBCiFLK1dADWzmg0cv36ddzc3NDpdJYORwghhBClnFKK2NhYAgMD0etzrq+TJK+Qrl+/TnBwsKXDEEIIIUQZExoaSoUKFXJcLkleIbm5uQHaC+3u7m7haIQQQghR2sXExBAcHGzKQXIiSV4hpV+idXd3lyRPCCGEEMXmfreJScMLIYQQQohSSJI8IYQQQohSSJI8IYQQQohSSJI8IYQQQohSSJI8IYQQQohSyGqSvISEBDZu3MjMmTPp27cvISEh6HQ6dDod06dPz3Xda9euMX/+fJ544gmqVauGk5MTTk5OVK5cmYEDB/Lbb78Vz0EIIYQQQhQTq+lC5cCBA3Tv3j3f64WGhhISEoJSyjTP2dkZpRSXLl3i0qVLLF++nBEjRvDVV19hY2NjzrCFEEIIISzCamryALy8vOjUqROTJ09m2bJlBAQE3Hcdg8GAUopOnTrx7bffcu3aNeLj44mLi+PEiRP06tULgIULF963RlAIIYQQwlroVMYqrhLMYDBkqWWrVKkSly9fZtq0aTkmaNHR0Zw/f54mTZpku1wpRffu3dm0aROurq7cvHkTR0fHPMcVExODh4cH0dHR0hmyEGWUUorU1FSMRqOlQxFCWAG9Xo+dnV2Bx7zPa+5hNZdrC3oZ1cPDI8cED7TeokeMGMGmTZuIi4vj1KlTNG7cuKBhCiHKEIPBQGRkJLGxsaSmplo6HCGEFbGzs8PNzQ1fX98iu1XMapK8opSx5s5gMFgwEiGEtTAYDISGhpKcnIyHhweurq7Y2NgU+Je5EKJsUEphMBiIi4sjKiqKxMREgoODiyTRkyQP2LlzJwD29vbUqFHDssEIIaxCZGQkycnJVKxYEScnJ0uHI4SwMq6urnh4eHDlyhUiIyPx9/c3+z7KfJJ38eJFvvjiCwD69+9/3/vqkpOTSU5ONj2PiYkp0viEECWPUorY2Fg8PDwkwRNCFJiTkxPu7u7Exsbi5+dn9isBVtW61twSExN54oknSEhIwNfXl3ffffe+68yaNQsPDw/TFBwcXAyRCiFKktTUVFJTU3F1dbV0KEIIK+fm5mb6TjG3MpvkpaWl8dRTT3Hw4EHs7Oz4/vvvCQwMvO96U6dOJTo62jSFhoYWbaBHj8KDD0K/fkW7HyFEnqW3opV+NYUQhZX+PVIUrfPL5OVag8HAoEGD+Omnn7C1teWHH36gS5cueVrXwcEBBweHIo4wg9hY2L8fIiOLb59CiDyRRhZCiMIqyu+RMleTZzAYGDx4MD/++CM2NjYsXbqUfiW5lkz+iQghhBCiAMpUTV56Dd6KFStMCV7//v0tHVbu0pM86+izWgghhBAlRJlJ8gwGA0899VSmGrwBAwZYOqy8kyRPCCGEEPlQJi7Xptfg/fjjj9ja2vL9999bT4Inl2uFEKJY7dy5E51OVyT3Si1evBidTkelSpXMvm0h7mVVSd6dO3eIjIw0TektURISEjLNj4uLM62Tfg/eihUrTI0sSvwl2ozkcq0QohRKT6IKMi1evNjS4ZdJ06dPL7LkVxQNq7pc27hxYy5fvpxl/uzZs5k9e7bp+bBhw0xfAnv27GH58uWA9qUyduxYxo4dm+M+Pv7445KVBNrZgZcXeHpaOhIhhDCbnHr3j4uLIz4+PtcyRd0BtbOzMzVr1iySbXt4eFCzZk2CgoKKZPtCZGRVSV5BZOx3JjU1lYiIiFzLJyYmFnVI+dOkCdy+bekohBDCrMLDw7OdP336dGbMmJFrmaLWvHlzTp8+XSTb7tOnD3369CmSbQtxL6tK8i5dupTvddq3b4+SS51CCCGEKGOs6p48IYQQZVv6PWE7d+7kxo0bTJgwgRo1auDs7JzpXrGEhASWLVvG0KFDadSoEeXKlcPBwYHAwEB69+7Nxo0bc9xHbg0v7m04cfDgQZ588knKly+Pg4MDVapUYcKECdy5cyfbbefW8CL9nrf27dsDsH37dnr06EG5cuVwdHSkdu3azJgxg6SkpFxfo3Xr1tGxY0c8PT1xdXWlYcOGvP/++6SmpmbZR3GJjo7mzTffpEmTJri7u+Pk5ET16tUZPXo0Fy5cyHG9xMRE5syZQ8uWLfHy8sLOzo5y5cpRp04dhg0bxurVq7Osk5aWxldffUX79u3x9fXFzs4OHx8fatasSf/+/fnmm2+K8lBLFiUKJTo6WgEqOjq6aHZw9qxSHTsq1b9/0WxfCJFviYmJ6uTJkyoxMdHSoZQ606ZNU4DK6d9T+rIFCxYof39/BShHR0fl5uaWaZ1FixaZyup0OuXh4aGcnZ1N8wA1ceLEbPexY8eOHGNI325ISIj6/vvvlZ2dnQKUh4eH0uv1pvXq1q2rYmNjc10/p2N/6KGH1Pvvv690Op3S6XTK09NT6XQ607Y7dOig0tLSso194sSJmY7R09NT2draKkC1a9dOvfrqq6Z95Nf9zk1Ojh8/ripUqGBaN+P5ApSDg4NatWpVlvViYmJUw4YNM53HjMeT3euYlpamHn744UyvgYeHh3JwcMg0ryQpyPdJXnMPqckr6WJj4bffYPduS0cihBAlxvjx4/H09GT79u3Ex8cTExPDmTNnTMu9vLyYNGkSu3fvJi4ujqioKOLj47l+/TozZszAzs6ODz74gJ9//rlA+7958yYjRoxg2LBhXLlyhaioKGJjY5k3bx52dnacOHGC999/v0DbPnLkCFOmTGHKlCncuHGDO3fuEBUVxRtvvAHAjh07+Pbbb7Ost3z5cj744AMAnnrqKa5evcqdO3eIjY3lq6++4sCBA3z++ecFiqmgYmNj6dmzJ1evXiUoKIhff/3VdL4OHz7Mgw8+SHJyMoMGDeLIkSOZ1v344485cuQI3t7erF69msTERO7cuUNycjLXrl1jyZIlWYYkXbZsGVu3bsXR0ZGvv/6a2NhYoqKiSExMJCIigjVr1pTsUa7MrbAZaFlX5DV5hw4pBUoFBhbN9oUQ+ZavX95xcTlP966fW9mEhIKXjY/PuWx8fN7KFpO81uS5u7ur0NDQAu9n9uzZClCdOnXKsiwvNXmAGjZsWLbbnjBhggJUtWrVclw/t5o8QE2bNi3bbfft21cBqnPnzpnmG41GVa1aNQWohx9+WBmNxlxjL66avHfffVcBys7OTh07dizL8piYGFWpUiUFqB49emRa1q1bNwWod955J8/7Gz16tALUqFGj8ryOpUlNXlkm/REJYd1cXXOeHn88c1k/v5zLduuWuWylSjmXbdcuc9k6dXIu26xZ5rLNmmVfroQZMmQIFSpUKPD6PXr0AGDfvn0YDIYCbeO1117Ldn6vXr0AOHfuHAkJCfneroODA5MmTcp120ePHs00//Dhw5w7dw6AV199Ndv7CYcNG0bFihXzHU9hrFixAoB+/fpRr169LMvd3Nx4+eWXAdi4cSPR0dGmZZ7/dR0WFhaW5/2lr2OpltkljSR51kJaCAshhEnr1q3vWyYiIoJp06bRsmVLfHx8sLW1NTWoqFOnDqA10MipkURuvL29qVatWrbLAgMDTY8Lsu26devimkNinb7t2/d0rXXo0CEA7OzsaNWqVbbr6nQ6HnrooXzHU1ApKSmmZLRz5845lnv44YcBrcuz9OMAePTRRwGYN28eAwcO5KeffiIyMjLXfXbv3h2dTsfPP/9Mt27dWLZsGdevXy/soVgtq+pCpUySES+EsG4ZRuDJwsYm8/MbN3Iuq7/nN3luXUrdW/bkyZy/Q+6t8fnrL6v4vvHz88t1+b59++jevTtRUVGmea6urqZWuAaDwZQwxMfH4+vrm6/9u7m55bjM1vbuv9bU1NR8bTev205LS8s0/+bNmwD4+Phgb2+f4/rF2Qnz7du3TbWkue03Y43sjQyfgaeeeooDBw7w6aefsnz5ctPABtWqVaNLly6MGDGCpk2bZtpWmzZteO+993jttdfYtGkTmzZtMu2jc+fODB06lA4dOpjtGEs6qckr6STJE8K6ubjkPDk65r3svaM85Kess3POZZ2d81a2hLG5N0HOIC0tjYEDBxIVFUWjRo3YsGEDMTExxMbGEhERQXh4OH/++aepvCpF36+lbcixjz76iDNnzvDOO+/QrVs3PD09OXfuHPPnz+eBBx5g3LhxWdaZPHkyFy9e5MMPP6R37974+flx9epVFi9eTMeOHXniiScKlHxbI0nySjq9HuzttUkIIcR97du3j8uXL2NjY8P69evp1q1bltqx0nbPVrly5QCIjIwkJSUlx3LXrl0rrpDw9vY2JeNXr17NsVzGZdnV0FarVo2pU6eyYcMGbt26xb59++jduzegtcDNroV0YGAg48aNY+3atURERHD06FGeeeYZAFatWlXsrYwtRZK8kq5ePUhOhitXLB2JEEJYhdDQUEBLfHK6TLht27biDKnINWnSBNAuD+/duzfbMkopfv/992KLyd7engYNGgBax845ST8Xer3edBw50ev1PPjgg6xatcrUiGTr1q33jaV+/fosWLDAdC9nXtYpDSTJE0IIUap4eHgAWsOL7MYrv3r1Kp988klxh1WkGjVqZGoI8u6772Z7CXrp0qVcvny5WOMaMGAAoNWeHT9+PMvyuLg4U3+C3bt3N507gOTk5By3a2NjY7r3UJ/hHtTc1gFw+u9WBv29962WUmXjKIUQQpQZbdq0wcXFBaUUTz75JGfPngXAYDCwefNm2rdvX+ruXdPpdMyYMQOAzZs3M2zYMFOr0qSkJL755hueffZZvLy8zLK/yMjIXKf0Bi+jR4+mcuXKpKam0q1bNzZu3IjRaATg2LFjdO3alYsXL+Lg4MDMmTMz7aNFixa8+OKL7Ny5k/j4eNP869evM3bsWFOXMd27dzct6927NyNGjGDjxo2ZGt3cvn2bmTNnmmoU07vQKe2kdW1Jd+UKjBkDbm7www+WjkYIIUo8Dw8P5syZw+jRo/n999+pWbMmrq6upKWlkZSUhK+vL4sWLeKxxx6zdKhm9dRTT/HXX3/x0Ucf8d1337F06VI8PT2Ji4sjNTWVjh070qJFC2bNmoXjvY1+8in9HsCcNGzYkMOHD+Pm5sbPP//MI488wtWrV+nevTuOjo7Y29sTExMDaP0CLl26lIYNG2baRlRUFJ9++imffvopOp0ODw8PUlNTMyV848ePp2vXrqbniYmJLFq0iEWLFgHg7u4OYNoXaH32pd+fV9pJTV5JFxcH69fD5s2WjkQIIazGc889x6+//kr79u1NCV5QUBBjx47lyJEj1K9f39IhFokPP/yQNWvW0L59e9zc3EhOTqZ27drMnj2bzZs3mxKk9E6Di0O9evU4ceIE06dPp1GjRtja2pKcnEzVqlV57rnnOHHiRLZDjS1fvpwZM2bQqVMnKleuTEpKCqmpqYSEhNC/f3+2b9/O3LlzM63z6aef8t5779G9e3eqV6+OUorExEQCAwN57LHHWL16NStXriwzl2t1qjS1HbeAmJgYPDw8iI6ONv1iMKtTp7Te6r294dYt829fCJFvSUlJXLx4kcqVKxe6RkSI4tS6dWv27t3Lm2++yeuvv27pcAQF+z7Ja+5RNlJZayb95AkhhDCDXbt2mVrePvLIIxaORhQHSfKshSR5Qggh7mPMmDEsXryY8PBwUwvbqKgovvzyS9O4tx07dqTZvWMWi1JJGl6UdKWsBZgQQoiis2fPHubPnw9oDRqcnZ2JiooyJXx16tRhyZIllgxRFCNJ8ko6uVwrhBAij958801++ukn9u/fT0REBNHR0Xh5eVG3bl369u3LqFGjcL53KDtRakmSV9JJTZ4QQog8euyxx0pd1zCi4CTJK+mqVYO0NEtHIYQQQggrI0leSafTwX8DPAshhBBC5JW0rhVCCCGEKIUkySvpbtyAAQNg2DBLRyKEEEIIKyJJXkkXHw8rVsCqVZaORAghhBBWRJK8kk66UBFCCCFEAUiSV9JJFypCCCGEKABJ8qyF1OQJIYQQIh8kySvp5HKtEEIIIQpAkrySTpI8IYQQQhSAJHklndyTJ4QQQogCkBEvSrrAQIiOlmRPCCGEEPkiNXklnV4P7u7g5mbpSIQQotTYuXMnOp0OXTY/oHNbVthtF4fFixej0+moVKmSRfYvSg5J8oQQQhS7kSNHotPp8PHxITk5Oc/rVa9eHZ1Ox2OPPVaE0ZVMly5dYvr06UyfPt3SoRSJS5cumZLjxYsXWzqcUkGSvJIuKgqefhqeecbSkQghhNk8/fTTANy+fZt169blaZ1du3Zx7ty5TOsXBWdnZ2rWrEnNmjWLbB8FcenSJWbMmMGMGTNyLefh4UHNmjWpWrVqMUUmSipJ8kq6xERYuFCbhBCilHjwwQepU6cOAIsWLcrTOunl/P396dGjR5HF1rx5c06fPs3p06eLbB9FqU+fPpw+fZrt27dbOhRhYZLklXTS4EIIUUql18Zt2bKFa9eu5Vo2NjaWVf+N4T106FBsbaXdoBD3I0leSSf95AkhSqkhQ4ZgZ2eH0Wi87z1YK1asID4+HoARI0YAkJCQwLJlyxg6dCiNGjWiXLlyODg4EBgYSO/evdm4cWOB4spLw4nTp08zaNAgAgICcHR0pEqVKowdO5aIiIhct52amsrPP//MqFGjeOCBByhfvjz29vb4+fnRtWtXli1bhsrm+75SpUp06NDB9Dw9vvTp//7v/0zL8tLw4vz584wePZrq1avj5OSEu7s7TZo04c033yQmJiZPr8u5c+cYMWIEwcHBODg4UKFCBUaOHHnfhL2orFmzhkcffRR/f3/s7e3x9/fn0UcfZe3atbmut3nzZvr27UuFChWwt7fH3d2dKlWq0KVLF+bMmcPt27ezrLN//34GDRpE5cqVcXR0xMXFhZCQEB566CHeeustrl69WlSHmT9KFEp0dLQCVHR0dNHsIDxcKS3FK5rtCyHyLTExUZ08eVIlJiZaOhSr9/jjjytAVatWLddyrVq1UoBq1aqVad6iRYsUoACl0+mUh4eHcnZ2Ns0D1MSJE7Pd3o4dO0xl8rNMKaU2btyoHBwcTGVcXV2Vo6OjAlT58uXVwoUL87RtQLm7uys3N7dM85544gllMBgyrffAAw8oLy8vUxl/f/9M04svvpjldQkJCck2/hUrVmSK383NLdPz4OBgdfLkyVxj/+2335Srq6tpfVtbW9OywMBAdfXq1Wz3nZuLFy+atrFo0aI8r5ecnKz69+9vWlev1ysvLy+l1+tN8wYOHKhSUlKyrDtjxoxMr72zs7PpuNKnHTt2ZFpn8eLFSqfTmZY7ODgod3f3TOvkJ/6CfJ/kNfeQzKGQijzJi4iQJE+IEkaSPPPZsGGD6R/jrl27si1z+vRpU5lvvvnGNP+nn35SkyZNUrt371bx8fGm+devX1czZsxQdnZ2ClDr1q3Lss2CJnmhoaGmf+gNGjRQ+/fvV0opZTAY1MaNG1WFChWUp6dnjuvv379fPfvss2rr1q2Z/m/cunVLffzxx6Ztf/zxx/mKK6PckryDBw+aXpfWrVuro0ePmuL/+eefVfny5RWgqlatqmJjY3Pcv5eXl3rsscfUqVOnlFJaorVixQpTwjpkyJBcY8xOQZO8iRMnmhL9119/Xd25c0cppdTt27fVq6++atrmK6+8kmm9S5cumRLBCRMmqGvXrpmWRUVFqT/++EM9//zz6u+//zbNj4+PNx3j4MGD1blz50zL4uLi1N9//60mT56sfv311zzHL0leCVbkSd6NG3eTPKOxaPYhhMiXvHwpG41KxcWVrqkovoIMBoOqUKGCAtSwYcOyLfPyyy+baszuTTxyM3v2bAWoTp06ZVlW0CRv9OjRClA+Pj4qIiIiy/Jjx46ZkqiCXCxbuXKlKcnKT1wZ5ZbkPfLII6aa04yJcbpDhw6ZauVmz56d4/47dOiQpbZRKaU++eQTBSgnJyeVmpp6n6PNrCBJ3tWrV03xTp06NdsyEyZMUICys7NT169fN81fsWKFAlSNGjXyHOP+/fsVoFxcXPJ9fDkpyiTPau7JS0hIYOPGjcycOZO+ffsSEhJiujcgr30GRUREMHHiRGrWrImTkxPe3t60bduWr7/+Ott7IEoEaXghhFVKSABX19I1JSSY/3XS6/Wm+8lWrVpFXFxcpuUGg4HvvvsOgCeffBJXV9c8bzu9Be6+ffswGAyFjlUpxYoVKwB47rnn8PPzy1KmXr169OvXr8D7SI/5/PnzhIeHF3g72YmKimLz5s0ATJ48GWdn5yxlGjduTN++fQFYtmxZjtt69dVX0euzphC9evUCIDExkX///dccYedq9erVpKWl4ejoyJQpU7It89prr+Hg4EBqaqqp8Q6Ap6cnoDXqSb/f837S10lJSeHWrVuFir04WE2Sd+DAAbp3787rr7/O2rVruXLlSr7WP3jwIHXr1mXu3LmcPXsWW1tbYmNj2b17NyNHjqRbt26kpKQUUfSF4O0N169DWJilIxFCiCIxfPhwdDod8fHxpiQq3caNGwn77/svu77xIiIimDZtGi1btsTHxwdbW1tTBUB6Fy0JCQncuXOn0HFevHjRdBN+x44dcyyX2zLQkorZs2fz0EMP4efnh729vSnmjImXuW/eP3TokKlCo3PnzjmWe/jhhwE4evQoqamp2ZZp0aJFtvMDAwNNj7NrsGBuf//9NwDNmjXD3d092zJeXl488MADmcqD1lWOr68vYWFhtGjRgnnz5nH69OlcK32qVq1KrVq1SE1NpUWLFrz33nscPnzYLD8iioJVtUH38vKiSZMmpmn8+PF5+qUTHR3No48+yq1bt6hVqxbfffcdDzzwACkpKSxYsIDx48ezefNmxo0bx/z584vhSPJBr4fy5S0dhRAin5yd4Z5KKauXTcWPWVSpUoX27duzY8cOFi5cmCmZW/hfH6G1atWiVatWmdbbt28f3bt3JyoqyjTP1dUVZ2dndDodBoOByMhIAOLj4/H19S1UnDdu3DA9DgoKyrFchQoVclx29uxZOnXqlCmBc3Z2xtPT01Qzlt5CN6+1S3mV3/jT0tK4ffs2/v7+Wcq45TDUZsaubXJKEM0p/ZhyOx64e0wZXwNPT0+WLVvGU089xYkTJxg7diygdSbdrl07nnzySfr374+dnZ1pHRsbG5YvX06fPn24ePEiU6ZMYcqUKTg7O9OqVSv69u3LsGHDsq0ltQSrqclr27Ytt2/fZtu2bbz//vsMGDAABweHPK07Z84cwsPDcXJyYsOGDaaM3t7enjFjxph6D//qq684e/ZskR2DEKLs0OnAxaV0TUV590h6Yrd3717T9/DNmzdZv349cLfblHRpaWkMHDiQqKgoGjVqxIYNG4iJiSE2NpaIiAjCw8P5888/TeVLyi05w4cP5+rVq1SqVImVK1dy69Yt4uPjuXHjBuHh4Zm6HykpMZdmnTt35uLFiyxZsoRhw4ZRvXp1oqOj+eWXXxgyZAiNGzfO0iVMw4YNOX36NKtXr2bUqFHUq1ePxMREtm3bxvPPP0+tWrU4duyYhY4oM6tJ8mxsbAq87pIlSwAYMGAAlStXzrJ87NixuLq6YjAY+P777wu8nyKRkABjx2pTCa0OFkKIwnr88cdN9zul194tXbqU1NRUbG1tGTp0aKby+/bt4/Lly9jY2LB+/Xq6deuWpXbJ3Pe0ZbwHL7e+4HJaFhoayt69ewHtfrd+/frh7e2dqYy5Y84oY/y5XQpOX2Zra5slvpIm/Zjud2k7fXl291G6uLgwZMgQFi9ezNmzZ7l69Srvvfcejo6OmWr4MrK3t6dv3758+eWXHDt2jJs3b/LFF1/g7e1NaGgow4YNM8PRFZ7VJHkFdebMGdP9e926dcu2jKurK23btgW0ntdLlJQUmDdPmyTJE0KUUo6Ojjz11FOA9sPcYDCYhjFL7+A2o9DQUADKlSuX46W6bdu2mTXGypUrm5KeHTt25Fjut99+y3Z+esygNXDITm4xZ2zoUJBaviZNmpi2kduQZ+kxNGzYMNOlypIo47120dHR2ZaJiorKdO/e/QQFBfHyyy8zceJEALZu3XrfdXx8fHj22Wd57733APjnn39KRMOMUp/kHT9+3PS4Xr16OZZLX3by5MkijylfMl4fkap7IUQpln7JNiwsjLfeest0yeveS7Wg3TcF2v1r2Y0ycfXqVT755BOzxqfT6XjyyScB+OKLL0z3+2V08uTJTC04M0qPGeDIkSNZlsfGxjJz5swc95+xYUHG+xDzytPTk65duwIwe/ZsErJpLn3kyBFWr14NwMCBA/O9j+L2+OOPY2trS1JSkinButc777xDcnIydnZ2PP7446b5ycnJuW7byckJyJxc53Wde9ezFMtHUMSuX79uepzbjZnpy2JiYrI04RdCCFH0mjRpQqNGjQB46623AChfvjzdu3fPUrZNmza4uLiglOLJJ5803cdnMBjYvHkz7du3z3VYsoKaOnUqbm5uREZG8vDDD5tqiJRSbNmyhW7duuV4033t2rWpWLEioCWuBw8eNC3bt28f7du3z7UVcI0aNbC3twcocNdfM2fOxM7OjnPnztG1a1dTIm00GtmwYQPdu3cnLS2NqlWr8uyzz+Z7++YSFxdHZGRkrpPBYCAoKIiXXnoJgHfffZdp06aZEuCoqChef/11Zs+eDcCECRMon6Eh43vvvUe3bt347rvvMl3uTU5O5scffzStl96tDcDy5ctp3bo1X375JRcuXDDNT3/fpXfj0rJlS7y8vIrmxcmPgnXdVzKEhIQoQE2bNi3HMm+//bapc8XcOi786quvTOUydpZ4r6SkJBUdHW2aQkNDi7Yz5Ojou50hS+/6QpQIMuJF0fn0008zDQ81ZcqUHMt+/vnnmcpmHF7M19dX/fzzz6ZlFy9ezLRuYYY1W79+fZZhwZycnBTcf1izX375JdMQYM7Ozqah2FxcXNS2bdtyHE5LKaWefvrpTOtWrFhRhYSEZBq+7X7Dmi1fvlzZ29ubtuPu7m563SBvw5rlJrf4c5OxM+S8TP/8849SShtt48knnzTNz+uwZtOmTcu0PScnJ+Xt7Z1pyLLatWursLCwLK9t+uTg4KB8fHwy7SswMNA0EkheSGfIJcisWbPw8PAwTcHBwUW7Q7lcK4QoQwYNGoSjo6PpeXaXatM999xz/Prrr7Rv3x5XV1fS0tIICgpi7NixHDlyhPr16xdJjD169ODQoUMMGDAAPz8/UlJS8Pf354UXXuCff/7JtoFfukcffZTff/+dHj164OnpSVpaGr6+vgwfPpyDBw/SqVOnXPf92WefMX36dNOxXblyhcuXL2d76Tgn/fv358SJEzz77LNUrVqV5ORkbG1tadSoETNmzOD48ePUrl07z9uzNHt7e1asWMGqVavo1q0bPj4+xMbG4uPjQ7du3VizZg0//PBDlvsLR40axVdffcXAgQOpV68ezs7OxMTE4OXlRdu2bfnoo484dOgQAQEBpnUee+wxlixZwvDhw2nYsCEeHh5ER0fj5uZG8+bNeeuttzhx4gS1atUq7pchWzqlrDdzqFSpEpcvX2batGk5jnrx6aef8uKLLwJaf3k5dZb48ccfM27cOEC7LyKnXtWTk5MzXZOPiYkhODg4120XSlwcpLcYi48vuo6qhBB5lpSUxMWLF6lcuXKmhEQIIfKrIN8nMTExpgQzt9zDqjpDLoiMvW9fu3Ytxxcjvcm7u7t7rsPmODg45Ll/PrOQYc2EEEIIUQCl/nJtxha1GVva3it9WfowOCWGkxOcO6dNUmMghBBCiDwq9UlejRo1TK2ZNm3alG2Z+Ph4/vjjDwC6dOlSbLHliV4PVatqUwloji2EEEII61DqswadTmfqKX358uVcunQpS5nPPvuMuLg4bGxsGDRoUDFHKIQQQghhflaV5N25cydTHzlGoxGAhISETPPv7edu0qRJBAQEkJCQQI8ePUx9E6WkpPD555/z+uuvA1pLmxo1ahTvQd1PaipMnQpTpkBSkqWjEUIIIYSVsKrWtemtae9n2LBhLF68ONO8gwcP0rVrV9MwI25ubiQlJZGamgpol2l//vnnfDeqyGsLlwJLStLuywOIjoai2IcQIl+kda0QwlyKsnWtVdXkFUbTpk05ceIE48ePp3r16qSmpuLi4kKbNm1YsGABGzduLN5Ws3kl/eQJIYQQogCsqguV7O6nyw9/f3/mzp3L3LlzzRNQcZMkTwghhBB5VGZq8qyW9JMnhBBCiAKQJK+kk8u1QpRYVnRLsxCihCrK7xFJ8ko6SfKEKHH0//VZaTAYLByJEMLapX+P6IugL1xJ8ko6uVwrRIljZ2eHnZ1dlu6ahBAiv2JjY03fKeZmVQ0vyiS9Ho4c0R5L9ylClAg6nQ43NzeioqLw8PDAKb2bIyGEyIfExERiYmLw9PREVwSVOpLklXQ6HTRoYOkohBD38PX1JTExkStXruDu7o6bmxs2NjZF8kUthCg9lFIYDAZiY2OJiYnBwcEBX1/fItmXJHlCCFEANjY2BAcHExkZSWxsLFFRUZYOSQhhRezs7PD09MTX1xcbG5si2Yckedbgrbe0vxMmgIuLZWMRQpjY2Njg7++Pn58fqamppqEWhRAiN3q9Hjs7uyKv+beqYc1KoiIf1gzuNr4IDwd//6LZhxBCCCGsggxrVprIPT5CCCGEyCdJ8qyJVLoKIYQQIo8kybMG6TV5kuQJIYQQIo8kybMGkuQJIYQQIp8kybMGkuQJIYQQIp8kybMG0vBCCCGEEPkk/eRZg127tL9F1CO2EEIIIUofSfKsQcuWlo5ACCGEEFZGLtcKIYQQQpRCUpNnDT75BFJTYeRIKKpRNYQQQghRqkiSZw1eeQWSkqBfP0nyhBBCCJEncrnWGkgXKkIIIYTIJ0nyrIkkeUIIIYTII0nyrIH0kyeEEEKIfJIkzxrI5VohhBBC5JMkedZAkjwhhBBC5JMkedZALtcKIYQQIp+kCxVr8PPPYDBA+fKWjkQIIYQQVkKSPGvQvr2lIxBWwGg0cv36ddzc3NBJ7a8QZqeUIjY2lsDAQPR6uRAmSj5J8oQoIp999hmzZ88mPDychg0b8umnn9K8efMcy3/00Ud8/vnnXLlyBV9fX/r168esWbNwdHTM0/6uX79OcHCwucIXQuQgNDSUChUqWDoMIe5LkjxrsHixNuLFgAHg6WnpaEQerFixggkTJvDFF1/QokULPvroI7p27cqZM2fw8/PLUv6HH35gypQpLFy4kFatWnH27Fn+7//+D51Ox9y5c/O0Tzc3N0D7B+QuI6MIYXYxMTEEBwebPmtClHQ6paTJZmHExMTg4eFBdHR00f1j9faGO3fg1CmoVato9iHMqkWLFjRr1ox58+YB2qXU4OBgxo4dy5QpU7KUf+GFFzh16hTbt283zZs4cSL79+9n9+7dedpnsbwXhSjD5DMmSoq8vhflpgJrIPdXWZWUlBQOHjxI586dTfP0ej2dO3dm37592a7TqlUrDh48yIEDBwC4cOECGzZsoHv37jnuJzk5mZiYmEyTEEIIkU4u11oTqXS1CpGRkRgMBvz9/TPN9/f35/Tp09mu89RTTxEZGUmbNm1QSpGWlsZzzz3Hq6++muN+Zs2axYwZM8wauxBCiNJDavKsgXSGXOrt3LmTd955h/nz53Po0CHWrFnDr7/+yltvvZXjOlOnTiU6Oto0hYaGFmPEQgghSjqpybMGkuRZFV9fX2xsbIiIiMg0PyIigoCAgGzXef311xkyZAjPPPMMAPXr1yc+Pp5Ro0bxv//9L9vuGhwcHHBwcChYkGlpYCsffyGEKM2kJs8ayD15VsXe3p6mTZtmakRhNBrZvn07LVu2zHadhISELImcjY0NoPXNZTY3bkDPnlCpkpboCSGEKLXkp7w1kZo8qzFhwgSGDRvGAw88QPPmzfnoo4+Ij49n+PDhAAwdOpSgoCBmzZoFQM+ePZk7dy6NGzemRYsWnDt3jtdff52ePXuakj2z8PaGffvg1i3YvVs62hZCiFJMkjxrsGQJpKRAxYqWjkTkUf/+/bl58yZvvPEG4eHhNGrUiE2bNpkaY1y5ciVTzd1rr72GTqfjtdde49q1a5QrV46ePXvy9ttvmzcwW1t49FH49ltYt06SPCGEKMWkn7xCkn6TREmR5/fi2rXQt692yfbCBbkdQIg8ku97UVJIP3lCiOx16QKOjnDpEhw7ZulohBBCFBFJ8qzBqlXw3Xdw+7alIxGlgYsLpHfUvG6dZWMRQghRZCTJswYvvghDh8Lly5aORJQWvXtrfyXJE0KIUqvMJXlbt27lySefJCQkBEdHR5ycnKhSpQqDBg1i165dlg4ve3LPlDC3nj2hXTsYPFhabQshRClVZlrXKqUYPXo0X375pWmek5MTABcvXuTixYv88MMPjB8/nrlz51oqzNzJP2NhLn5+UFJ/1AghhDCLMlOTt3jxYlOC169fP86ePUtCQgIJCQmcPn2aXr16AfDhhx+ydu1aS4aalYx4IYQQQoh8KjNJ3pIlSwCoVq0ay5Yto3r16qZlNWvWZOXKlVSpUgWAH3/80SIx5kiSPFFUIiNh4UKtc2QhhBClSplJ8sLCwgBo2LAhttmM2WlnZ0ejRo0AiIuLK87Q7k/uyRNFpWtXePpp+PlnS0cihBDCzMpMkpdeS3fkyBHSshmzMzU1lcOHDwPwwAMPFGdoeSc1ecLc/rtNgZJ2i4IQQohCKzNJ3ujRowE4d+4cAwcO5Ny5c6ZlZ86c4cknn+TChQtUrVqV8ePHWyrM7H36KfzwA/yXqAphNn36aH+3bIHYWMvGIoQQwqzKTJLXs2dPPvzwQ+zt7Vm1ahXVq1fH2dkZZ2dnatWqxc6dOxk9ejQHDhzIdYiQ5ORkYmJiMk1FrlcvGDgQfHyKfl+i1NuyBUxv23r1oHp1SE6G9estGpcQQgjzKjNJHsC4ceNYs2YNfn5+ACQmJpKYmAhASkoKcXFxREdH57qNWbNm4eHhYZqCg4OLPG4hzGXtWujeXRvZLCoK7X7PJ57QFq5cacnQhBBCmFmZSfISEhLo378/jz76KBUrVmTLli3cvHmTmzdvsmXLFurUqcN3331H8+bNOXr0aI7bmTp1KtHR0aYpNDS06IPfvBlWr5YWkKLQKlUCDw/Yvx8efvi/kfLSk7yNG6GkNToSQghRYDqlysbd/GPGjGH+/PnUrFmTw4cP4+jomGl5YmIijRo14uzZs7Rp04Y//vgjT9uNiYnBw8OD6OjoXC/zFkrVqnDhAuzdCy1bFs0+hNXL63vxyBFt6NrISO3v1i0KatTQ3mNbtkCnTsUYtRDWo1i+74XIg7y+F8tETV5sbCxfffUVoCV79yZ4oI1+8cILLwCwe/dubty4Uawx5kq6UBFm1LAhbNumPd62De5E6eD77yEsTBI8IYQoRcpEknf27FlTtylVq1bNsVzGDpIvXrxY5HHlW9modBXFoGFDrb0FwL59QPPm2lBnQgghSo0ykeTp9XcP8/LlyzmWi4iIMD12c3Mr0pjyRUa8EEWgTRvt7+7d9yyQ95kQQpQKZSLJq1WrFk5OTgB8/fXX2XaGbDAYTJd0vby8qFmzZrHGmCtJ8kQRaN1a+2tK8rZvh/btYexYS4UkhBDCjMpEkufk5MQzzzwDwKFDh+jZsyfHjh3DaDRiNBo5evQo3bt3Z+/evYDW1YqNjY0lQ85M7skTRSC9Ju+vv7Ru8khJgV27YN06+UEhhBClQNZBXEup9957j3///ZdNmzaZJgcHB0Dr4DjdwIED+d///mepMHMn/3iFGdWoAb6+WivbQ4egZfv24OQEV6/CsWPQoIGlQxRCCFEIZaImD7TavA0bNrBy5Up69epFhQoVSO89Jjg4mMcff5z169fzww8/lKxaPIC334avv757p7wQZqDT3XPJ1skJOnbUZmzYYLG4hBBCmEeZqckD0Ol09OvXj379+lk6lPyxtniF1WjTRrs6u2cPTJ4M9OgBv/6qTVOmWDo8IYQQhVBmavKEEFllbGFrMKCNeQZax9u3b1ssLiGEEIUnSZ412LMHNm2SYc2E2TVpAl5e2ltr504gJES7F89o1IbSE0IIYbUkybMGo0ZBt26Qy5i6QhSEvT08+aT2eMmS/2YOG6a936pUsVhcQgghCk+SPGsirWtFERg6VPu7ejXExwMTJmgNL2SIMyGEsGqS5FkD6QxZFKGWLaFqVS3BW7vW0tEIIYQwF0nyrIEkeVbps88+o1KlSjg6OtKiRQsOHDiQY9n27duj0+myTD169CjyOHU6GDJEe2y6ZAtw5Qp8+qm874QQwkpJkmcNJMmzOitWrGDChAlMmzaNQ4cO0bBhQ7p27cqNGzeyLb9mzRrCwsJM0/Hjx7GxseGJJ54olngHD9b+bt0KL70ECXeSoV49ePFF2L+/WGIQQghhXpLkWQMZ1szqzJ07l5EjRzJ8+HDq1KnDF198gbOzMwsXLsy2vLe3NwEBAaZp69atODs7F1uSV7UqvPKK9viTT6B5Wwfiu/TRZqxfXywxCCGEMC9J8qyJ1ORZhZSUFA4ePEjnzp1N8/R6PZ07d2bfvn152sY333zDgAEDcHFxybFMcnIyMTExmabCePddraceX184cQI2lh+hLZDRL4QQwipJkmcNXnkFPv4Yata0dCQiDyIjIzEYDPj7+2ea7+/vT3h4+H3XP3DgAMePH+eZZ57JtdysWbPw8PAwTcHBwYWKG6BrV3jqKe3xb3HNtAf//ANhYYXethBCiOIlSZ41eOop7d6okBBLRyKKwTfffEP9+vVp3rx5ruWmTp1KdHS0aQoNDTXL/tOHr/1tnzM0+y/R27TJLNsWQghRfCTJE8LMfH19sbGxISIiItP8iIgIAgICcl03Pj6e5cuX8/TTT993Pw4ODri7u2eazOGhh0CvhzNn4Fqb/tpMuWQrhBBWR5I8a/DPP/DHHzKWqJWwt7enadOmbN++3TTPaDSyfft2WrZsmeu6K1euJDk5mcHpzV0twNMTmjbVHv/m3lt7sGuXNtSZEEIIqyFJnjUYORLatZOuLKzIhAkTWLBgAd9++y2nTp1i9OjRxMfHM3z4cACGDh3K1KlTs6z3zTff0Lt3b3x8fIo75ExMl2wvVYZ16+D8ea16TwghhNWwtXQAIh+kda3V6N+/Pzdv3uSNN94gPDycRo0asWnTJlNjjCtXrqC/J2k6c+YMu3fvZsuWLZYIOZOOHeG99+C3HXrUosekFx8hhLBCkuRZA+kM2Sq98MILvPDCC9ku27lzZ5Z5NWvWRJWQc9ymDdjZaYNenD8P1apZOiIhhBD5JddfrIEkeaKYOTtD69ba459/BmbOhCZN4Nw5i8YlhBAi7yTJswaS5AkLSB9sY9kyYMcOrQGQjH4hhBBWQ5I8ayA3RAkLeOIJsLGBv/+Gf1v819r3l18sG5QQQog8kyTPmkhNnihG5cpB+shsyxMe0x78/jtER1suKCGEEHkmSZ41GD0a3nkHate2dCSijBk4UPu7bIsPqmYtSEuT0S+EEMJKSJJnDYYNg6lToXp1S0ciypjevcHBAU6dgmMPjtRmyn15QghhFYo8yTMYDMybN49evXrRp08fvvnmm6LepRDCTDw8oEcP7fGihCe1Bxs2gMFguaCEEELkiVmSvIULF2JjY0P//v2zLBs4cCAvvfQS69evZ926dYwaNYoBAwaYY7dlx9mzcPAg3Llj6UhEGfTMM9rfb7cFkVi5DjzyCERFWTQmIYQQ92eWJC+9h/6nnnoq0/ydO3eyatUqlFK0atWKzv/dxb1y5UrWrVtnjl2XDSNGwAMPaN1YCFHMunSBSpXgzh0dK6cdh++/BwsPuyaEEOL+zJLkHT58GIDW6b2n/mfJkiUAjBw5kj/++IMtW7YwY8YMlFIsXrzYHLsuW6R1rbAAGxtt+GSAL7+S7nyEEMJamCXJi4yMxMHBAV9f30zzt23bhk6n48UXXzTNGzNmDAB///23OXZdNkhnyMLCRowAW1vYuxdatlRUCkrhl6XSlYoQQpRkZknyYmJicHR0zDQvLCyMq1ev4ufnR926dU3zvby8cHd35+bNm+bYddkgSZ6wsIAAraUtwJ9/6rh83Z5v3o+0aExCCCFyZ5Ykz8PDg+joaBISEkzzdu3aBUCrVq2yXefepFDkQpI8UQJ88ok2hO2rHf8E4Mg5FwtHJIQQIjdmSfLq1asHwI8//miat2TJEnQ6HQ899FCmstHR0cTExBAQEGCOXZcNMqyZKAHKl4f//Q8mznAH4FJiANERSRaOSgghRE7MkuQNHDgQpRRjxoxh9OjR9OnTh02bNmFvb8+TTz6Zqey+ffsAqC4d++af1OSJEsC7dW0q6K8DcGzZMQtHI4QQIidmSfKefvppOnfuTGJiIl999RXr1q1Dp9Mxc+bMLDV2K1euzLaGT+Ri6FB47TWoU8fSkQgBOh0Nymv31B799aqFgxFCCJETW3NsxMbGhk2bNrFs2TL27t2Lp6cn3bt3z9KlSkpKCmFhYbRr145u3bqZY9dlw4gRlo5AiEwaNrVlwzU4ckhGvhBCiJLKLEkegF6vZ9CgQQwaNCjHMvb29mzYsMFcuxRCWEiDHsHwMxy9HQRhYdoNe0IIIUoUsyV5ogiFhkJ8vPaP1MPD0tEIQcO2WuOLY47NMHoain4QbCGEEPlWLN/N69ev56WXXmL8+PFs3bq1OHZZuowYAbVrw6+/WjoSIQCoXh0cHSE+yZbzVx0sHY4QQohsmCXJW7NmDVWqVOG5557LsmzChAn06tWLefPm8cknn/DII48wefJkc+y27JHWtaKEsLWF9D7Ojx61bCxCCCGyZ5Yk7+eff+by5cu0bds20/xDhw7x0UcfoZQiODiYqlWropRi7ty57Ny50xy7LhukM2RRAjVsqP09MnsLHDxo2WCEEEJkYZYk76+//gKgU6dOmeYvXLgQgD59+nDhwgXOnj3LmDFjUEqxYMECc+y6bJAkT5RA6Unerv0OsGqVZYMRQgiRhVmSvJs3b2Jra5ulT7wtW7ag0+l45ZVX0Ou1Xb366qvA3U6RRR5IkidKoN69wUZv5Hce4p+V5ywdjhBCiHuYJcmLiorC1dU107xbt25x7tw5PD09ad68uWl++fLlcXFxISwszBy7LhtkWDNRAlWsCP37pgLwwflecPmyhSMSQgiRkVmSPFdXV6Kjo0lNTTXN2717NwAtW7bMUt7Ozg5bW8v13hITE8N7771Hq1atKFeuHA4ODlSoUIEOHTowffp0oqKiLBZbrqQmT5QwE6dqLWuXM4DQpbssHI0QQoiMzJLk1apVC6VUpo6OV6xYgU6ny9IYIyEhgejo6CyXdovLjh07qFGjBlOmTGHfvn3ExMTg7OzMtWvX2LlzJzNmzODSpUsWiS1HffrA+PEyrJmV+eyzz6hUqRKOjo60aNGCAwcO5Fo+KiqKMWPGUL58eRwcHKhRo0aJ7zy8SRPoWPUyBmz5+GtnS4cjhBAiA7NUp/Xt25c///yTZ555htOnTxMWFsaKFSvQ6/U88cQTmcr+9ddfKKWoXLmyOXadL3v27KFHjx4kJibSt29fpk6dStOmTdHpdCQkJHDixAnWrVuHR0nrcHjUKEtHIPJpxYoVTJgwgS+++IIWLVrw0Ucf0bVrV86cOYOfn1+W8ikpKTz88MP4+fmxatUqgoKCuHz5Mp6ensUffD5NfMWW30bBgksP88a5G7hXy3p8QgghLECZQWJiomrYsKHS6XRKr9crnU6ndDqdGj9+fJayL774otLr9eqtt94yx67zLD4+XlWpUkUBauzYsWbbbnR0tAJUdHS02bYprF/z5s3VmDFjTM8NBoMKDAxUs2bNyrb8559/rqpUqaJSUlIKvE9LvRcNBqVqO11QoNQHY84X676FKE7yfS9Kiry+F81yudbR0ZHdu3fz5ptv8sgjjzBw4EC+//575s6dm6lcSkoKu3btomLFinTp0sUcu86z7777jgsXLhAQEMD7779frPsutMhIbWizuDhLRyLyICUlhYMHD9K5c2fTPL1eT+fOnXNsVf7zzz/TsmVLxowZg7+/P/Xq1eOdd97BYDAUV9gFptfDhDe9APj4lyqkpVk4ICGEEIAZx651dXXltddey7WMvb09hw8fNtcu82XJkiUAPPHEEzg6OlokhgIbPhzWr4evv4ann7Z0NOI+IiMjMRgM+Pv7Z5rv7+/P6dOns13nwoUL/PbbbwwaNIgNGzZw7tw5nn/+eVJTU5k2bVq26yQnJ5OcnGx6HhMTY76DyKfBL3jy6vtw5QqsXg39+1ssFCGEEP8pE+OKJycn8/fffwPQtGlTrly5wqhRowgODsbe3h5/f3969uzJryV9bFhpXVtqGY1G/Pz8+Oqrr2jatCn9+/fnf//7H1988UWO68yaNQsPDw/TFBwcXIwRZ+boCGPG/BfX9CSpzRNCiBKgSJK82NhYdu3axcqVK1m5ciW7du0iNja2KHaVJ5cuXSIlJQXQakzq1avHggULuHHjBi4uLty4cYP169fz6KOPMnLkSFQuyVRycjIxMTGZpiInnSFbFV9fX2xsbIiIiMg0PyIiIsdW5eXLl6dGjRrY2NiY5tWuXZvw8HDTe/deU6dOJTo62jSFhoaa7yAK4PlOZ/DURXHktCOz30qyaCxCCCHMnOQdO3aMxx57DG9vbzp27MiAAQMYMGAAHTt2xNvbm969e3Ps2DFz7jJP7ty5Y3o8c+ZM7OzsWLlyJXFxcdy5c4fLly+bWgF//fXXfPjhhzluyyK1J5LkWRV7e3uaNm3K9u3bTfOMRiPbt2/Ptt9IgNatW3Pu3DmMRqNp3tmzZylfvjz29vbZruPg4IC7u3umyZLKPViVj/1nATBtpi0W+KgLIYTIyFwtPVavXq2cnJwyta69d9Lr9crJyUmtWbPGXLvNkz179ijANK1duzZLGYPBoBo2bKgA5ePjo1JTU7PdVlJSkoqOjjZNoaGhRd/aqlcvpUCpL78sun0Is1q+fLlycHBQixcvVidPnlSjRo1Snp6eKjw8XCml1JAhQ9SUKVNM5a9cuaLc3NzUCy+8oM6cOaPWr1+v/Pz81MyZM/O8z5LQ8s+4cpXqyToFSlWtlKpWrdJa3wpRGpSEz5gQShVz69qLFy8yaNAgkpKSCAkJYf78+fz7778kJiaSmJjIv//+y/z586lUqRJJSUkMGjSIixcvmmPXeeLm5mZ6XL16dXr37p2ljF6vZ9KkSYA2JNvBgwez3ZZFa0+kJs9q9O/fnzlz5vDGG2/QqFEjDh8+zKZNm0yNMa5cuZJpaL/g4GA2b97MX3/9RYMGDXjxxRd56aWXmDJliqUOoUB0j/fly4afE0AY5y/Z0q8fPPQQZGgfIoQQoriYI6McPXq00ul0qlWrVio2NjbHcnFxcapVq1ZKr9dn6kOsqN26dctUi9ezZ88cyx08eNBUbsWKFXnadrH8suvdW6vJ+/zzotuHsHolppZh+3Z1Cy/1un6mcnUxKFBq/nxt0bp1Sr3wglJJSZYNUYiCKDGfMVHmFWtN3rZt29DpdHzxxRe4urrmWM7FxYUvvvgCpRRbtmwxx67zxNvbm6CgoPuWUxlqynTp98GVBF27wrPPQu3alo5EiPvr2BHvtvV40/ga7z64DoC334b9++GJJ2DePFi1ysIxCiFEGWCWJO/q1au4ublRv379+5atX78+7u7uXL161Ry7zrP0zpdPnTqVY5mTJ0+aHlti2LUcPfccfPGFdt1LCGvwyisAPOOxkgoV4No16NAB0hsK//67BWMTQogywixJnp2dHampqXkqq5QiJSUFOzs7c+w6z4YPHw7AuXPn+Omnn7IsNxqNzJkzB4CgoCCaNGlSnOEJUbp07w5Hj+Kw+gfS+0hPTATb/7pf/+MPy4UmhBBlhVmSvGrVqpGUlMTmzZvvW3bz5s0kJSVRrVo1c+w6z9q2bUu/fv0AeOaZZ1i9ejVp//XYeuXKFQYOHMjRo0cBePvtt9HrS1A/0XFxcOsWJEnfY8JK6HTwX83+8OFQrRrY2MCyZdriU6fgxg0LxieEEGWAWTKZXr16oZRi5MiR970cOmrUKHQ6XbYtXIva4sWLadeuHbdu3aJfv364urri7e1NSEgIP/74IwDTpk1j2LBhxR5brkaMAF9f+OYbS0ciRL7ZR99k7w+XOHkS+vWDevW0+bt3WzYuIYQo7cyS5I0bN46goCCuXr1K48aNGTJkCIsXL2br1q1s3bqVRYsWMXjwYJo0acLVq1cJCgpi3Lhx5th1vri4uLBjxw4WLFhAu3btcHFxIS4ujqCgIAYMGMCePXuYPn16sceVZ9KFirA2CxZA+fKUm/0yNWpos9q10/7KfXlCCFG0bM2xEXd3dzZt2kTPnj25dOkSP/zwAz/88EOWckopKleuzM8//5yp77ripNfreeaZZ3jmmWcssv8CkREvhLVq2hQMBvjlF4iJAXd32rWD+fMlyRNCiKJmthvP6taty9GjR5k1axaNGjVCr9ejlEIphV6vp1GjRrz33nscOXKEunXrmmu3ZYMkecJaNW4MtWpp95OuWQNA27baosOHITracqEJIURpZ9bWBa6urrzyyiscPHiQhIQEwsLCCAsLIyEhgYMHDzJ58mQMBgNNmjShadOm5tx16VaS+uwTIj90Ohg0SHv8X+1+YCBUrar9Ztm1S1uUlgYbNmgtcIUQQphHkTUhtbOzw9/fH39//0zdpaSlpXH48GEOHz5cVLsuvaQmT1ijgQO1v9u3w4QJEBbGo49qs2bP1t7W48ZBjx4webLFohRCiFKnBPUTInIkl2uFNataFZ56CoxG+PBD2LOHyZPB0VFrYfvWW9o9egDffqv1GCSEEKLwJMmzBm3awODB2r1NQlijpUu167EdO0JiIkFBMGaMtmjatLu/X+LiYPlyy4UphBCliSR51uD55+G77+CRRywdiRAFo9NBt27aJdshQ+DWLaZsfAhXYgHw9IRJk7SiX31luTCFEKI0kSRPCFH8vL3xTQ1jOtPR6RSffqrdj2dnB3/9pTXE3bhRG+iloGJjtd9GUVFmi1oIIayKJHnWIDVV64Liv2HYhLB6Oh30789E5hLfoz+DB4OfH6QPhPP449rwtw89VPAWt59+CkOHwqxZZotaCCGsiiR51uDpp8HJCT76yNKRCGE+Tz4JgNOWdaYO8155Bby9tYTP1RVOnCh4i9sjR7S/Bw+aI1ghhLA+kuRZE2ldK0qTevWgdm1ISYEVKwBtgIxbtyAiAlav1op99hmsX5//zZ85o/09edJM8QohhJUpUJJnY2NT4MnPz8/cx1D6SRcqojTS6bRaaoD//Q9u3sy0uEsXGD9eezxyJMTH533TRiP8+6/2OCwM7twxQ7xCCGFlCpTkpQ9XVtBJ5JMkeaK0GjsWGjSAyEhYtizL4lmzoHJlCA/P390K169DQsLd51KbJ4Qoi2wLstK0adPMHYfIjQxrJkore3tYuBAuXdJaW9zDwQHeflvrS/m992DYMPjtN63Llccey3mzZ89mfn7iBLRubdbIhRCixJMkz5pITZ4ojZo21aZ0X32lXV995RUA+veHOXPg0CFt8IyUFK3YmTNQo0b2m7w3yZOaPCFEWSQNL6yBXK4VZYXRCJ98AlOmaDV8SqE/fZL3HtkB3E3wAJYsyXkz6Y0ufHy0vydOFFG8QghRgkmSZw2aNoU+fXKuthCitNDr716HHTUKqleHunXp/E5Hvvd9iUWLtPFtQevo2GjMfjPpNXnpm5KaPCFEWSRJnjUYM0YbAiCbe5aEKHXeflsb+sxggPPntRvzWrfmqd4J/N8wxRNPgIcHXLkCv/+e/SbSk7xevbS/16/LyBdCiLJHkjwhRMmi08HXX8O8efDjj1rL2927YcEC0OlwcjL1o2yq1csoJQUuXtQeN2sGFSpoj9Nr8+LitErC3C73CiFEaSBJnhCi5LG312qwn3hCG/oi3dGj8H//x9CALQCsWqW4ffvu4oQEuHBBqwR0dYXy5aFOHW1Z+n15Y8Zo+eK4cXKbqxCidJMkzxqMHg02NtplLCHKso0b4dtvaT27N7U5SVycjv5N/uX25Vh69NAu46YPg1ajhlYpWLeu9nz1aq2vvfQavDt3tIRQCCFKK0nyrIHReHcSVuOzzz6jUqVKODo60qJFCw4cOJBj2cWLF6PT6TJNjo6OxRitlXjmGXB1RZeUyDIG4kIc2y5Xp3JlxYYNkJZ2dwi09HZKDz6o/d28+e4IGjY22t+//y7e8IUQojhJkmcNpAsVq7NixQomTJjAtGnTOHToEA0bNqRr167cuHEjx3Xc3d0JCwszTZcvXy7GiK2Ejw/s3w+//ELD8C189+ppAGKUO+Xd43jttbsJXM2a2t8nnoC1a6FdO+15585argjw11+ZN//FF/DiixATUwzHIoQQRUySPGsgSZ7VmTt3LiNHjmT48OHUqVOHL774AmdnZxYuXJjjOjqdjoCAANPk7+9fjBFbkTp14NFHwd+fPm8/wOLplxj0SCT7j7vy1lvaiBijRsHI3jfhhRfQKSO9e8OuXXD1KmzYcLd2L2OSt3+/dmfEp59Cp05w65ZFjk4IIcxGkjxrIMOaWZWUlBQOHjxI586dTfP0ej2dO3dm3759Oa4XFxdHSEgIwcHB9OrVixP36cE3OTmZmJiYTFNZNGxaJZZu9CU4WHverh18+YUiaPJT8NlnWjcs/wlyjcbODh54QHt+6JDWSMNohLEv3P0R9fff8NBD2n17oJU5dqzk/M6aPx8mTSo58Zhbaiq8+irs3GnpSISwbpLkWZPS+o1eykRGRmIwGLLUxPn7+xMeHp7tOjVr1mThwoWsW7eOpUuXYjQaadWqFVevXs1xP7NmzcLDw8M0BadnOWXZjRuQlASrVsG2bdq12/h47bPz5ZdQqRIcPkzt2uDsrHWncmZXOEuWwF9/63Allt/oQGC5FE6cgIkTtQTwiSegQQOtA2ZLMxhgwgT44AOtsXFptHUrzJqltYQWQhScJHnWQC7XlnotW7Zk6NChNGrUiIceeog1a9ZQrlw5vvzyyxzXmTp1KtHR0aYpNDS0GCMugaZP15K45s21/lEAXnsNGjXSPjs//aT1iNy5Mzb7dtOknPZ6/dxnEa+8on22XuctOjxXix/X2qPTwaJFWp98a9dqm/vhh+I9pOyEhUFysvb42DHLxlJUrlzR/p46pSXiQoiCkSTPGtSpA127aqOzixLP19cXGxsbIiIiMs2PiIggICAgT9uws7OjcePGnDt3LscyDg4OuLu7Z5rKtMaNwc5Oy3yuX4fKleGVV7Rler2WoTVrpt1s164dzS6vBODV2CncuKGjnv4EL/ExDB5M69Ywdqy26urVd3exYwfExhbzcd3j0qW7j0trknf9uvZXKTh82KKhCGHVJMmzBmPGwKZNMHSopSMReWBvb0/Tpk3Zvn27aZ7RaGT79u20bNkyT9swGAwcO3aM8uXLF1WYpU+vXvDvv/Dss1ClCixcCE5Od5d7eWk3efXpA0rxgO4QAErp8HRJYa2xFw6Vg6BVK7hxg3f6/k3lytqqI0Zov7FSUrRLiUrB3r1a58s5iYkpmoQwY5J3/Lj5t18SpCd5AAcPWi4OIaydJHlCFIEJEyawYMECvv32W06dOsXo0aOJj49n+PDhAAwdOpSpU6eayr/55pts2bKFCxcucOjQIQYPHszly5d5Jr2vD5E3fn5aPyjnz0P79lmXOztr9+stXcqDa14GtEq+5XVnUo3zMHiwVl0XGIjLMwPZvEnx+edaQ4eePbVNrF8Pb74JrVvD88/f3XR4OCQmao+jo6FePa0j5vRE8NIl2LKl8IeYPmQblP6aPJAkT4jCsLV0AEKURv379+fmzZu88cYbhIeH06hRIzZt2mRqjHHlyhX0+ru/se7cucPIkSMJDw/Hy8uLpk2bsnfvXuqkj8klzEevh0GDqAIsWwYerga6Tt+gLRs0CIKCtGHVzp2j+oKXqT5zJjg48Oij2ogZq1dDfLwCdHz/vWLWLB1hYVoFYL16Wg3fhx9C+i2SGzZA377QpYtW0fjnn9CiRcHDz1iTFxqqJZQeHgXfnrmlpWn3MXp4aJWpBekcQJI8IcxDp5TczV8YMTExeHh4EB0dXXT3RL3yyt0+E6ZNK5p9CKtXLO/F0mrZMq3Z6uDB2vMZM7SGHKANnfHQQ6RUr0u5t14kJlbLWnQYUeiZ1mgdu5y6snOfNkLJmDFaK9z0Hm0ef1zrfy+9R52ZM+F//8s+jNu3tavKuSVGnTppfQGm271bq1UsKbZtg4cf1h7fuAHlyuV/G35+cPOm9liv115LFxfzxVhQ8hkTJUVe34tyudYaJCVpTczSm9QJIcxr4MC7CR5oP6bWrtUylLNnYcEC7F8exyPltGqlAPd4PtO/CMB7h7uwc58jNjba7+XPPtOSkvTbKX/9VavZS7dnT/YhLF8Ovr5a/3C5Sa/J8/TU/h47puWnGWu/ituZM3eT2pUr784/dSr/20pJuZvgubpqXdhI4wshCkaSPGsgXagIUfx699aylCVL4PXX4cEHmfJVFTp1gpW/uvDMhVcp75tCElrjjolNdjJgwN3VP/9cqwRMStISvXR792Ydhjo6Wuv1RSmYM0e7rJsdg+Fu9yKPPKL9PXZMa3MSFKRVQBb318SyZVCrllZTmZwMa9bcXXb6dP63l96VpJ2d1iE1yCVbIQpKkjxrIEmeEJbh4wNDhmgtLfbto3Enb7ZtgzZtwC4kkOfG2gPgzS2m/tWHef/3N40baw14H3sM+ve/u6kGDRSurlpCd+9gJm+9Bek97qSlwZQp2Ydz/bq23NZW61UJYOnSu0nk9OlaY5D0BiBF7ehRePpp7fFff8GwYRAZeXd5QZK89BrJ8uXvjkwiSZ4QBSNJnjWQYc2EKJHGj9eSquWdv8GTaHxq+HDokFabpbt2lf7X5prKjnJbTovmWhVexku2p07Bxx9rj2fP1u5BW7Mm+8u66ZdqK3pE03Dz+8Ddy6Tt22tfFV98oSVIzz6rPd68Ofs7PUJD7w7bVhBRUVoym5gI6d0/rlih/U2/lFyQy7XpSV5gIDRpoj3OOMawECLvJMmzJlKTJ0SJ4uam3YP38MpRWtVdSoq2IC4Oqlal7sKJ9GENdTjBoD2jaXNVy4LSE7hjx7TLnGlp0KOH1rZqxAht2XPPaSOyZZSe5FWqqqf2ydXoMWjPK8GG5TGsfvs0ISFabeFXX2kNPh55BBo2zJw0Xr2qXWJt0kQrm36ZeNIkbdzYvHj7bbhwwTRSHLWr3s0kX3pR+64qTE1eYKDWYtnWVksWC5IwClHWSZJnDeRyrRAlm6cnrFsHNWtqz48d07phcXFhzbIUTqw5i6djMq3PLgS0FrHr10PbtlpSU7eulpSB1vrW31/r6Hj06Mwf+/Q+8ir7xuJ4+jBN0a5jfjzsIE79etDnzcZc+Hwz27fDi0/H09NtJ+U8UzhzRtvX559r669dq/Xfd+kSvPyyNn/yZG083Lw04A8Lg3nztMfz54O/Sxwfpb4AQCDXeK661hH45cu5dxidHVOSV17h++8+Hqmg9fi8dGn+tiOEAJQolOjoaAWo6OjootvJvHlKtWmj1Pz5RbcPYfWK5b0o8u7WLaUiI+8+37VLxfQarPR6o9JSN21q00ap27czrGc0ql1bkpSNjbZ8/vQIpS5fVkopNfz/DAqUeustpdSaNSpUF6z28uDdjXl4KPX339p2xoxRCtRtOz81pPlpBUqVK6dUSopSDz+sMsVga3v3sU6n1G+/5X5oY8dqZVu2VMpoVEp98olSoHbwkDpFTaWaN1c+Ptpx/rPqnFI5vCf37lWqc2el/vrr7rxhw7Rtv1Pxc6VArXhwrgKlKlZUymDI5zkwM/mMiZIir+9FSfIKST70oqSQ96J1aNxYS2JsbIxq4oO7VcLoCUq1a6dUly5K9eypVGCgUk2bqvdnJitQSk+aWtn2Y6XS0lQHr0MKlPru41vaxtavV6p3b6Xq1lWqenWl9u+/u6OkJKWeeEIpUKnYKH+7SAVKLVmilJ2dFkP37neTu379lHr6ae1xYKBSN29mH//ly0rZ22vltm//b6bRqNScOUqtWaOUk5NSoFrXua1AqWX0V2rZsmy39V94qkoVpeLitHnpCehihirl6KgSRr6o3N21hHHnTvOcg4KSz5goKSTJKybyoRclhbwXrcO2bUqNGKHUsRFzM1enZZzs7JRx6ffq6V43tackq6/qfqQqcEWBUn+sDMvbztLSlPr4Y6VcXdUk3leglLe3ljDVrKlUzIWbqon9UdXM8YiKvh6n4uK0+ekJYHY1Z1Onasvbt89hnxMmKOXrq55ufFCBUm8wXalRo7ItGhJy95BfeEGbV7eu9nwLnU1XL9KTz2eeydthFxX5jImSIq/vRbknTwghilGnTvDNN1CvkS107AgvvKDdcLZkiXaD286dEB2NbtBTfLnalwFBv5OKPaNOvMRVggGo9GBA3nZmYwMvvggnTzLMYx0At29r9/j27AlurzzPwZQGHEhqiPvCj3BxgR9/BEdHbTi2D9r+dLf57n/WrtX+jhqFNm5ZbGzmfU6eDKmp1C6n9aVymlrEbttv6t8v3c2b2j176ebN04YNvn5Va0wSqI+Afv0gJYUh9f4B4PvvtdbCQog8Kqaks8SaNWuWAkxTfhXLL7uZM5Xy91dq2rSi24ewelLLUDql/HlQTdXNUo11/yg7W4Nq0aKA96bNm6ce4ICp5uz3adsy1x66uSl144ZSBoP6sssq7ZIyqerIMx+bNnHqlKmiUUWt/e3utdbExMz7unpVrf9Ou1wbwHXlyw3l6GBQ//xzt8ivv2qr16ql1HPPKVPtYno4tzv10wqeOaMM6FR33QbT/YNLlxbg+M1APmOipJCavDw4c+YMM2bMsHQY9xcbq/WUes8vaiFE6WfXognv/NONQxe9SEjUs2+f1pdevj37LMMHat2ceHtDy/KXtPmvvw6NG2vfMx99BDodIytupjdrMWDLnG/Laf2sAD/9pK3SsSN4LPpIe9K9u1b1l1FQELVbeQEQTnkiKUdSsp533rlb5MAB7W+zZjBrltbv9Jkz2jxHEvEc0lN7Ur06+gpBrFW9GNghnLQ0bQQ6S4zuIYS1KbNJntFoZMSIESQlJdGyZUtLh5M76UJFiLKtYUMICcHWthB9o9va8n9ft2HUKO2qsO2DD2jZ2muvaX2nvPMO/O9/oNOhm/8Z//vYH4DlqY8TNmsxcPdSbe/2UXeH2Xj++Wx3FxICXlqeRx+0sc5Wrbrbd156B8fNm2s90EyffnfdQN8UdL17aU90OujYEXtSWdr8EyZO1GZPnw4PPqgliY0aaf0KLl9efKN9CGENymyS9+mnn7J3714GDRpEly5dLB1O7iTJE0KYgbMzfPnlf8Ot1asH27Zp/fl16ABTp2oFAOzseODFVrSucYNU7Pn8k1Su/Ztgqn3rFfWtNpBumzZQu3a2+7Kx0e6f2zjzIGt4nMcct6AUvPceqF2/89d2rXaw2W/vwcaNPPuMgVq1tHUDa3mAh8fdjXXqBIB+2xbmzFZ8/bXWSfKBA/D333DkiHZcAwdCxYrwxhvamMF5JV+torQqk0nexYsX+d///oePjw8ffvihpcO5PxnWTAhhbjY29/1uGfemDwCfJw5j5MA4AB58UFH+x//GYRs5Mtf1mzWDR8bVgsmTeXWWGwBLvzOyofNcbiZ7YEsqDddOg+7dsatcgXkfpuLsfHdcXpNOnbRr1AcPwrPP8vSwNA4d0jpxXrdOq2GcMEFL8CIjtbGAX3gh7y/Fo49qY/De2zhECGtXJpO8kSNHEh8fz9y5cylXrpylw8k7+bkphChGvR+3oaJfIpGUY+NBPwAGh+zWht7w8NBav96Piwu8/z4txrWka/tk0gx6eqWtAqBhhds4vjBSu64bHk6n/e8QFaVdQc4kKEhrfqvTaRndmTPUr69don3sMejdW7vifP48fPuttso338Dvv2vzxo+Hf/7JPrxjx7SWxN9+K7+nRelT5pK8BQsWsH37djp37szQoUMtHU7eyOVaIYQF2NrCpwucaN0aJk7Ukqbnn/lvfN5nnrl7eTePlq50oEHFOxiwBaDZo/7w6afaWGZ//AFTp2Jnl8PKo0drCV58vFarpxT8/LPWl8t/3422tjB06H/du6A9btJEa0/yxBPaGMH3+vRT7W/fvhAcnK/DEaLEs7V0AMXp2rVrTJ48GScnJ7788ssCbSM5OZnk5LsDcccUR4vXwECt9VuFCkW/LyGEyOCxx7TJJKqp1qFdmzb53pavL2w/6EXHjopjx3Tpt9pprXPzsr1eveDECa38zZvazYVJSVqNYpcu2n2CNja8+652GTdjP3znz8OyZTBkyN15t27Bd99pj198Md+HI0SJV6Zq8p599lmio6OZPn06VapUKdA2Zs2ahYeHh2kKLo6ffs8/D4cOaSOJCyGEJXl6Qvv2WrVZAfj6wr59Ov74Ax5/vAAbqFwZypcHPz+tdg+0DK1xY+3S8Jtv4uWaysKFWs3clBfieKvzLgBmztTywHRff63liI0bQ+vWBTocIUq0MpPkLV26lF9//ZVGjRoxYcKEAm9n6tSpREdHm6bQ0FAzRimEEKWfi4tWcVfoe+BefhmcnLQO9g4fhuRkmDYNNm+me3e48sNuZn1fkZe2PYq37jZnz2rdrIDW7ehnn2mPXxxyR+7HE6VSmbhcGxERwbhx47CxsWHBggXYFvAXKICDgwMODg5mjE4IIUSBBARoDTJ+/FEbp83VFfbsgR49tOV16oCDA2728UxI+YDXeJvnRytcXHR8+CGEhkJ552gGvF4dXggj5xsChbBOZaImb8qUKdy6dYtRo0ZRq1Yt4uLiMk0pKSmmstnNs7iPP4YqVbSe6YUQQtw1YgRs2gRjxsCwYfDVV3erCL29tfsH//2X8YE/0o5dxMTq6NNHa0Ti7g6/TNqFY2rs3eE2hChFykSSd/HiRQA+//xz3NzcskyzZs0ylU2f93JJuv/tzh2ty4LISEtHIoQQ1qVWLahYEeeNq9lSbjDDWAyAs7NiwwZo+kpniIrSOocWopQpE0me1ZMuVIQQonAaNMDh9BEWDf+D9ba9OTjvT62xhbOzdl+fEKVQmbgnb+fOnbkunz59OjNmzABAlcRESpI8IYQoPG9vdAu/ocfUf8GaOsIXooDKRJJn9aTZlxBCmE/16paOQIhiIZdrrYnU5AkhhBAijyTJswZyuVYIIYQQ+SRJHto9eUqpknk/HmhdxNesCf7+lo5ECCGEEFZC7smzBs89p01C5CL9R0qxjKcsRBmU/tkqsRUCQtxDkjwhSonY2FiA4hlPWYgyLDY2Fg8PD0uHIcR9SZInRCkRGBhIaGgobm5u6Ky4RXZMTAzBwcGEhobi7u5u6XAKrbQdD5S+Y8rr8SiliI2NJTAwsBijE6LgJMmzBgsXwocfQu/e8NZblo5GlFB6vZ4KFSpYOgyzcXd3LxUJRLrSdjxQ+o4pL8cjNXjCmkjDC2tw8yYcPw5Xr1o6EiGEEEJYCUnyrIF0oSKEEEKIfJIkzxpY8f1VQuSXg4MD06ZNw8HBwdKhmEVpOx4ofcdU2o5HiHQ6JW3BCyUmJgYPDw+io6OL7t6U2bPh5Zdh6FD49tui2YcQQgghrEJecw+pybMGcrlWCCGEEPkkSZ41kCRPCCGEEPkkSZ41cHeHChXA29vSkQghhBDCSsg9eYVULPfkCSGEEEL8R+7JE0KUWLNmzaJZs2a4ubnh5+dH7969OXPmTKYy7du3R6fTZZqeK6FjOE+fPj1LrLVq1TItT0pKYsyYMfj4+ODq6srjjz9ORESEBSO+v0qVKmU5Jp1Ox5gxY4CSf35+//13evbsSWBgIDqdjp9++inTcqUUb7zxBuXLl8fJyYnOnTvz77//Zipz+/ZtBg0ahLu7O56enjz99NPExcUV41EIUTiS5Akhit2uXbsYM2YMf/75J1u3biU1NZUuXboQHx+fqdzIkSMJCwszTe+//76FIr6/unXrZop19+7dpmXjx4/nl19+YeXKlezatYvr16/Tt29fC0Z7f3/99Vem49m6dSsATzzxhKlMST4/8fHxNGzYkM8++yzb5e+//z6ffPIJX3zxBfv378fFxYWuXbuSlJRkKjNo0CBOnDjB1q1bWb9+Pb///jujRo0qrkMQovCUKJTo6GgFqOjo6KLbyfLlSrVoodRrrxXdPoSwoBs3bihA7dq1yzTvoYceUi+99JLlgsqHadOmqYYNG2a7LCoqStnZ2amVK1ea5p06dUoBat++fcUUYeG99NJLqmrVqspoNCqlrOv8AGrt2rWm50ajUQUEBKjZs2eb5kVFRSkHBwe1bNkypZRSJ0+eVID666+/TGU2btyodDqdunbtWrHFLkR28pp7SE2eNQgPh/374Z5LCUKUFtHR0QB439O46Pvvv8fX15d69eoxdepUEhISLBFenvz7778EBgZSpUoVBg0axJUrVwA4ePAgqampdO7c2VS2Vq1aVKxYkX379lkq3HxJSUlh6dKljBgxAl2Gztmt6fxkdPHiRcLDwzOdEw8PD1q0aGE6J/v27cPT05MHHnjAVKZz587o9Xr2799f7DELURC2lg5A5IF0oSJKMaPRyLhx42jdujX16tUzzX/qqacICQkhMDCQo0eP8sorr3DmzBnWrFljwWiz16JFCxYvXkzNmjUJCwtjxowZtG3bluPHjxMeHo69vT2enp6Z1vH39yc8PNwyAefTTz/9RFRUFP/3f/9nmmdN5+de6a+7v79/pvkZz0l4eDh+fn6Zltva2uLt7W01500ISfKsgQxrJkqxMWPGcPz48Uz3sAGZ7n2qX78+5cuXp1OnTpw/f56qVasWd5i56tatm+lxgwYNaNGiBSEhIfz44484OTlZMDLz+Oabb+jWrRuBgYGmedZ0foQoq+RyrTWRmjxRyrzwwgusX7+eHTt2UKFChVzLtmjRAoBz584VR2iF4unpSY0aNTh37hwBAQGkpKQQFRWVqUxERAQBAQGWCTAfLl++zLZt23jmmWdyLWdN5yf9db+3hXPGcxIQEMCNGzcyLU9LS+P27dtWcd6EAEnyrINcrhWljFKKF154gbVr1/Lbb79RuXLl+65z+PBhAMqXL1/E0RVeXFwc58+fp3z58jRt2hQ7Ozu2b99uWn7mzBmuXLlCy5YtLRhl3ixatAg/Pz969OiRazlrOj+VK1cmICAg0zmJiYlh//79pnPSsmVLoqKiOHjwoKnMb7/9htFoNCW0QpR0crnWGkiSJ0qZMWPG8MMPP7Bu3Trc3NxM9zh5eHjg5OTE+fPn+eGHH+jevTs+Pj4cPXqU8ePH065dOxo0aGDh6LOaNGkSPXv2JCQkhOvXrzNt2jRsbGwYOHAgHh4ePP3000yYMAFvb2/c3d0ZO3YsLVu25MEHH7R06LkyGo0sWrSIYcOGYWt799+FNZyfuLi4TLWKFy9e5PDhw3h7e1OxYkXGjRvHzJkzqV69OpUrV+b1118nMDCQ3r17A1C7dm0eeeQRRo4cyRdffEFqaiovvPACAwYMyHTZWogSrXga+5ZexdKFytdfK+XlpdSwYUW3DyGKEZDttGjRIqWUUleuXFHt2rVT3t7eysHBQVWrVk1Nnjy5aD9nhdC/f39Vvnx5ZW9vr4KCglT//v3VuXPnTMsTExPV888/r7y8vJSzs7Pq06ePCgsLs2DEebN582YFqDNnzmSabw3nZ8eOHdm+x4b99z1qNBrV66+/rvz9/ZWDg4Pq1KlTluO8deuWGjhwoHJ1dVXu7u5q+PDhKjY21gJHI0Rmec09ZFizQpJhzYQQQghRnGRYMyGEEEKIMkySPCGEEEKIUkiSPGuwfj107AhvvGHpSIQQQghhJaR1rTW4fh127AC5508IIYQQeSQ1edZAulARQgghRD5JkmcNZFgzIYQQQuSTJHnWRGryhBBCCJFHkuRZA7lcK4QQQoh8kiTPGkiSJ4QQQoh8kiTPGtjagoMD2NlZOhIhhBBCWAkZ1qyQZFgzIYQQQhQnGdZMCCGEEKIMkyRPCCGEEKIUkiTPGvz2G/TsCdOmWToSIYQQQlgJGdbMGly9qo1fm5Ji6UiEEEIIYSWkJs8aSBcqQgghhMgnSfKsgQxrJoQQQoh8kiTPmkhNnhBCCCHyqMwkebdu3WLRokUMHjyYOnXq4OLigoODAxUqVKB3796sXbvW0iHmTC7XCiGEECKfykzDi4CAANLS0kzPHR0dsbOz49q1a1y7do1169bRrVs3Vq1ahbOzswUjzYYkeUIIIYTIpzJTk5eWlkbz5s2ZP38+58+fJzExkbi4OC5evMjTTz8NwMaNG3n22WctHGk25J48IYQQQuRTmRnWbMeOHXTo0CHH5c899xxffvklAFeuXCE4ODhP2y2WYc2MRq0WT6cDfZnJy4UQQgiRDRnW7B65JXiAqTYP4O+//y7qcPJHrwcbG0nwhBBCCJFnkjX8x9HR0fTYYDBYMBIhhBBCiMKTJO8/O3fuND2uX7++5QLJzp9/woAB8Oablo5ECCGEEFZCkjwgKiqKWbNmAdC2bVtq1qxp4YjuERoKK1bA9u2WjkQIIYQQVqLMdKGSE6PRyJAhQwgLC8PR0ZF58+blWj45OZnk5GTT85iYmKIOUbpQEUIIIUS+lfmavJdeeon169cD8Nlnn9GgQYNcy8+aNQsPDw/TlNdWuIUiXagIIYQQIp/KdJI3adIkU83dhx9+yIgRI+67ztSpU4mOjjZNoaGhRR3mXVKTJ4QQQog8KrOXa19++WU++OADAObMmcO4cePytJ6DgwMODg5FGFk25HKtEEIIIfKpTCZ5kydPZs6cOQC8//77TJw40cIR3YckeUIIIYTIpzKX5E2aNMlUg/f+++8zefJkC0eUB3JPnhBCCCHyqUwleRkTvDlz5pT8Grx0PXpAdLQ26oUQQgghRB6UmSQv4z14c+fOZfz48RaOKB/s7LRJCCGEECKPdEqV/hu9rly5QkhICAB6vZ5y5crlWn7SpElMmjQpT9vO6yDBQgghhBDmkNfco0zU5BmNxkyPIyIici0fFxdX1CHlz+HD8MknULkyvP66paMRQgghhBUoE0lepUqVsOoKy9BQWLQImjWTJE8IIYQQeVKmO0O2GtKFihBCCCHySZI8ayBdqAghhBAinyTJsyZSkyeEEEKIPJIkzxrI5VohhBBC5JMkedZAkjwhhBBC5JMkedZA7skTQgghRD6ViS5UrF6HDhAWBrZyuoQQQgiRN5I1WAMHBwgIsHQUQgghhLAicrlWCCGEEKIUkiTPGpw5A2PHwqxZlo5ECCGEEFZCkjxrEBoK8+bBsmWWjkQIIYQQVkKSPGsgXagIIYQQIp8kyRNCCCGEKIUkybMGUpMnhBBCiHySJK+ECw+HBb8G8jVPS5InhBBCiDyTJK+EO3QIRs2txbtMkSRPCCGEEHkmSV4J16YN6PWK81TjWqqfpcMRQgghhJWQJK+Ec3eHxg21GrxdL662cDRCCCGEsBaS5FmBhzpop2nXcR8LRyKEEEIIayFJnhVo3177+9tvFg1DCCGEEFZEkjwr8NBDYGenOHcO/n1soqXDEUIIIYQVkCTPCri7Q9sHEgHY+EsanD9v4YiEEEIIUdJJkmcluj/uDMAvPAqffmrhaIQQQghR0kmSZyV69dL+bqcTFz7fDBcuWDYgIYQQQpRokuRZiWrV4JFHFAo9H6Y8D8OHQ1KSpcMSQgghRAklSZ4VeeklbQzbeYzlg98f0HpKjoiwcFRCCCGEKIkkybMiXbvC1Kna40l8QKcz81m3t5xlgxJCCCFEiSRJnhXR6eCdd+Ctt7Tnv8U1p3dfPf36wZIFydwc/QaEhVk2SCGEEEKUCDqlZNT7woiJicHDw4Po6Gjc3d2Lbb9//w3ffac1tE0/g17cZrr9LJ6e6InLqy+Bq2uxxSOEEEKI4pHX3EOSvEKyVJKX7uhR+PJL2PZLImdDnQBwJ5qathcYPTCK4d+0ATu7Yo9LCCGEEEUjr7mHXK61cg0awGefwfHzTsz/TFHVP44YPPgrrTEjvuvA4947WPXKARISLB2pEEIIIYqT1OQVkqVr8u5lMMCenal8P+0MX+2pZ5rv5KR1w9Kokfa8f3/o0cMyMQohhBCi4ORybTEpaUleRof/iGX5K4dYEdqKS1ezXrKt5hlJnK0HrVpClZp2NGsG7dqBoyN4et4td+ECrFypdc3n51d88QshhBAiK0nyiklJTvLSKQWnT8OZM7B76nr+PW1gPY9ixCbHddq3h7YVLhIW786SX71JSdER7JtAn0eSSLB154VhsSQ5eXH+PNStC8HBkJAAFSoUPM74eG0b5YqhV5jwcPDyAgeHotvH99/DtGkwbx488kjR7UcIIUTZIkleMbGGJC+Ta9fgzh2u7bvC3nmHsD97nINJdbiFD7t4iBPUu/82clHZ7SaeumgqesYS5+iLrZ2OWHsfvAKdMBq1fv5SEtI4sDuFm2FptK13h4fa6yA4mGYNU7gcbk/v3jpq14Ygv1QC0q7iVcWLGs08SEjUcfMm2NuDXg/16oGtrbZfgwEOHYLftqSRfCmMM+EeVK9lw7S37dHZa7WYcXHaugcPQseOCj+vNNauUTR50L7Ax3vxopbgpseRbsOGu5fD69SBEycKvAuTuDgtWW/aVOtOpzRSqvQemxBCmIskecXE6pK87MTEwL//gq0tCSG1uRltz4oVcGbpX3id2kNimh1PVdnP5sR2RN5U7Ehrw2lq4+UFtWrB8eOK2Nji/8/s6mSg48M21KsH3y1MITQ8a7Lmrouhc5PbBLasxKJF2jxnYyw3E91MZer636SGfzQeAU7Y+vng7udI/fpgb6eoXz2Jes2cCAuDH75NZf2qROxuR5CYrOd8UgXC7zig14Ori5GBHcJ5/AkbPKr40K2nDbdv331Nnu9/i1c/8CEpCX79FSpXBjc3aNVKa/wcGQmxsVoiFxcHly7B7dswqP01vOoGkpSso3lzxbFjOsb3PEebB1PZe8qbTk/506IFeHvn/DolJUFKCmR8e8bFQVoa3LypJd1162LKVOPiIDVVq+lMp5R2yT4wUBtoJScZk7SICHj9da27n379YMIEOHcOfH0hIABCQ7Xj3rULHn1UW9a7N0yaBDNn5ryPdKmp2mtnNGr7NFdyaDRCcrJ2H2t2kpO1yRIf98RE7XaK/Bzr6dPwyy/Qty9UrVqw/aa/1rn57DNYskT7IdeuXe7vSSFE4UiSV0xKRZKXG6W0/2iOjtpzoxHj7Sgu3fGgUlUb9HowxiUQu3Al4RcT2XfBnwTXcly7lIp/3HniEm1J9A3mcrVOnDwJZ08bCIo/S03O4Gtzhz205rShBgD2JNOmQSy2Ab5UrgxhxyO5sf8St9Lc+RetjC2pOJGIQkccbplC9SCKdvxOiq0ze9JaZFl+r45sZwcdUPdpZO5IIknk8B8/F8FcoRl/sYbHAe0f872fNle7JPTKSEyac7bbcCeaEH0ox4w517Da2UG3hxJwjLiEs4c9zo5G4pQL3t6KK2H2bD7gSWKaHZUr6/D1hSred9iww4noJEfTNprbHqRWLUi0cePn09VJTtZqU8vZRxN1x8iNGEfCo5zQ6RT/1+wk8WExOLvbEtTrAdIMOi5fhpN/xXPmiiOOtgbqBMdy9pozt+Lvvm56vcJo1LITRwcjScl3X3dHuzSSUjNXh6bXWHaufJ6uA7zwqOxNSgqcOK7Ysi6R1RudsLVRpBn02Npq94w+kLKX5FQ9jQfXxTvEjUOHYMsWbTu9e2u3AqSkwP79cOVCGjVDEkm0cSM5SXH9ShoujmmsXQtXbzpQv4EeHx/o0AHOn4cjR0CfmsyhY7akpNnwQI1oBjS7wI6LIVyO8SY4GKpUgevnEgjyiKNdVyeux7ji42mgTh3F5Wt22NrClStazW5QEHTqpN3z6ugIQYGKPb9GoXNyJDbNicBAbbm/v3aOFyyAiRPhgSZGxg+7TWS0LReOJVC9ioEk70BWrLIhKEhL5Nq1034ozH03mTMXHf57/ymaN07Fs5w9d+5or1eXhxUKHWFh2mtSty5Ur651zeTgoP3227IFtm5VPPvEHTyD3Th4xIYH68Xz1LNu6PUQEgLHj99t1AWgw8jUPmfoNz6Y2ymuxMRo8+PjtW3WqAFRUXD7agJ/HbHn73/0uDor3njNQIXK9uj12g8gnU6rLd+8WftR9NRT2rz4eLCxARcXLQH99lutXLt22shA6RISIDpaS9yNRq3G32jUPofBwdq6GzZAxYracd+6pZWvVk37m5ys/SBLT3BTUmD7du0HQPoPp3LltNfq/HntNfDwyPxjx2jUrjzk5PZtOHBA+9FVr56WHKf/K0m/1eb0aW0f3t5Qv7523OmiorRYDx+GY8e0W22qVdN+SKU7cUJ7zR54IHMsCQnaMZ8+rW2zbl3tvZjdbSyhoeDsDD4+2jrOzjn/EFJKi0Wng9q1s17pKKx//tF+VIwcCS1aaPO2bNG6E3vlFWjeXJuXlqa9vn5+WkxJSTnHbG0kySsmpT7JM7e0NK3m0M3N9M0ZdfEO147dxrdRBfwrZvPtkpzMyeVHSbsWQYNmDpCUhOF2NEd9OrBybxDXr8MDrqd5xvkHHEc8BTVrQkICu3cZuP7DTo78Y+RIpV60aq2je3f446O/ebCtPc0GVOXaO99ybE8M/972IeFiBGmNHiDygUc4cAAMUbH8c9KeFLSYWrGH/g7r8GgQQrJPeXbbtKfrQG9u3IAbv59i80Yjick23KAc3txmnsMkGg6sw4eRQ9hzpw579miHE2AXiU9qOOEEcAtf02G6EIerPgHXyuVwd9dx6YKBO9F375vUY6C/zWq20YmbBh8aVIkjXufK+fNFd7oKqzYnqcdxVvIkoPXPnZCg/eMDcCCJZBxz2YIoLD2GXO+/LSx7fSopRvP3xelom4qysSM5OcO+dCmkqLs19kFu0SiDkesJd6udHR2hvHs87k4pnAh1J82Y/bFrfcUr4uJyrxZ1d9eSQDc3xY1LCZwPc8mxrJ2d1mgtOlpRq9xtrka7cifenio+0dg76NDZ2uDjkYbRzoEL150oVw5OnFAYDHdj0OkUAW7xRCXY42ybwq2krJ3a+3kkUbdyApFxjpy57EhKatYssl5dhW1qIuE3dIRHaZmNvT4Vd4cUbHQGnFxtiYh1JjEx83puTqnUr55MgJ8RezsjDs626J0dWfKDLXo9VAgycvGSHmeHNKr5xRLsHY+Dmz06JweuXdfh4WvP2SuOXLyoba9SuTgerH4LT08dSTgSGe/EzSg7oiLTqFVHj4OXM3/8AdUrp1HOJZ5Ugx57O7hy05H4eDh/yQZHOwOtWyq8A+xxdobVq4xE3tJjozfySKtY4nWu7PxDO892tkZ69Ugj2WjPX39p918/2imBq2E2HDtjz2Otb3PH4MahI7a4O6dSuYYd6PR4eUFkeCrXrxqJuGVLs/pJVK1pS5LRgaAg7V9XWpqWJLq7w53bikvH44iLSqN8kB6P8s7obPXExoKDo546dXUMHVp0Nf6S5BUTSfJKt+RrkVzZexX/ig64V/fXrmHmdq3MaNS+VW7c0H7CZvhJHBGh/QKuFXMA/cXzpN6K4cQJsE+MplqPmtjbKa3apkUL0OtJToYDuxKJu3CDYL9kfOxiKN+lPolGB27dutvI5c8/4c/vz2Pz5x7u3IG4VAe8Um8Qk+yAt10s7UIuE/DWGC471uTKFQjdc5lqZzbg5W+H0cGZCpVsORRWnj92GtAbUhnxQ2eqVtOxdSskr/qFgPN7iE20pYHdKf6OqsZuXVt8QlwxJCYT27IrNjbaZdw6exZQ8/oOEmzd2RdemcByqXTpmIZDg5qcTa2MS6cHCQyEO3fg9w5v0MB4hCqJJ4h2DeKTpFH8q6ox5qHj2Ldtgapbj7NntdqHX6fu5ujNABLQajtrcgZ7UmiiO0xknbYM+LgVCQmw7AdF1Pa/McYnciyhKnG44scN+rCWKDzZ5fEYqd4B2NhASFAqNrt+I4zy1OQMNhioyBUScKah41kSnvw/Tjk1ISQE9qwOx3jwEA+zFR9u07JhAs7ejrzzTzfOp1Wkw/jGNGjpwvbtcHvXMRr8vZC99u25kuKPD7c4TS1u401VzhMV0hBPXzs6doRjP57i78u+eHEHgAScqc0pHEjGgWQuN+zF8dO2piSnHDfoyS9cJ5BoPPEhEj/HWE4aamDTvCktW+qws4Nz3+5m3fVmhHCZ0XzOiJancahbjc3bbTnj2YIN7gO4fRvCrxlIuJ2IHiP2pNCQI5ynKpepRA3OoHNzpVbHILy8oHpVA3tmbOdOmist2M+fPMiftMSWVNLQErz69eHnnyHk0i5WzL7Cu9sf4N/kilTiEk4kcpUK3MKHytXtcHDQaptct66hafzvPMDfbKcTixhONB4Y0RODB6B9hEJC4PLZnH8Q+NhG0aqbJ+vXZ60t12PABgN6jNpkqyfNxtH0utqT/F8JW1yJxYlEbuKHPcnY2kCCIfMPT1tSqcBVXIkjFjduUo4EXHCzTyI2pWA/WGpyGlvSOENN0+uZzsEmlQZN7DAY4NrlVCJu5ZxQV/e5RYqrD1euZH4d0l+DVLK//9jdOZWEBLLsuzAcHUFvTCUhpex2xn/nTuaeKsxJkrxiIkmeEMUgJkab4uO1a2w1amjVJtkl3EYjXL6sXZuJjNRa2zg6atfVAgPvlrt1S8sgzpzRfqLXqaOVs7XNvN3QUO36kKurdj3L3z/nOG/c0NYtV067ge7cOW2bSUna9asuXbSbEkG7Tnv+vHa9zc1Nuy5YubJ2nDduQM2aGHU2JCRA/E9b8Y25gE2j+loM7u7acdpkU0t14wbq0mV0Z89o2VGrVnfLZbyOmJamXWc9dUq7hufsDFWqkJSix5Ek7XirV7+7XaW0Xyo3b0K5ciRv2I7t9StcjffCWLsulYa0zXo6zp3TrnMrhcHLlyS3crg0r5v5XCUmaufSxka7lnvsGGnXb3AlyQ/dk09QsaK2KPHHXwgLTcMtPhy3+HCSkxT7z3mj3Dxo3ccP1/49uHRJCy/x5WnERyZSPSCWav6x2nXKFi20cxkUhOHB1vz1Fzjqkqk3/3kMbp7o4uOw93KBChW4Ee+Cp7qDbf3aHKvSi5s3ITY8npTX36Kzwx/4eKRp12U9PMDLi7SWbbHp+BAX0ioSH//fVYCxC6lvOIx/DQ8u3vZA3b6DITqOWwZPEh/sQMDInoSHQ5uGsVQd3g4aN4aoKC4n+XM6qRIBfkZiHXxp+nglnB57WDsFp04TM34af18rz4XbnoSk/Esg1ykfpCfGtwqVn+4IAwdy7Roc+/4oxvfnEOBnpEoTT5wbVicsQk/M9TiMjs6EtXoc11oVtMvip06SuGQlyVdvcvCILeGhKZxPrYjS2WBMSeNOx8dp90wN6taFyPX7qPnO/xHuW49w9xqcjQ9Cd+c2Kj4BL9dUrj76HCGdq/PYY2A8dJjvR2wn3uhEdIItTsnRlEu5hm/yNVwCPTje63/E+FSmVi2I2vQnaSvXYGdMIcHoSMXkszjrkqjqHMadcjXY3eF1jHXqkZAADpfP8vTOIfzj0oYNpytTOfUsjiQx2GEVfwY9zu5H38Wvtg9Vq0Lq9z/y69LbhOivUsXxOv8YGlA+4RwP2f/JSbcW/DvgdeJcA7hzB9re+Rnf7SsIdItlR2R9rse740kUt/DBljTsXhxNDB7Ex4P3yT8IPLAOT289YTEuxCZryawzCcTjQtjgySz+zszXqTOQJK+YSJInhBCiVLNEs/e0NC27v99+ldJarTk6aj/o7ufe+8xzc+uWdhNtesuuOnXu3qAZE3P3ByRoP4DSWygZDNpNkbndjFlIec09ii7NFEIIIYT1s0S/RnltraHT5e/GN50ubwkeaK1MfHyyX3bvPnOr4begMjd2bWxsLNOnT6d+/fq4urri4eFBs2bN+OCDD0hJSbF0eEIIIYQQZlGmLtdevnyZ9u3bc+nSJQCcnZ0xGAwk/3cHbuPGjdm+fTteGTsIuw+5XCuEEEKI4pTX3KPM1OSlpaXRs2dPLl26RPny5dm6dSvx8fEkJCSwfPly3Nzc+Oeffxg8eLClQxVCCCGEKLQyk+R9++23HDt2DIDVq1fTuXNnAPR6Pf379+fLL78EYMOGDWzfvt1icQohhBBCmEOZSvIAOnToQMuWLbMsHzBgAJUrVwZgyZIlxRqbEEIIIYS5lYkkLyEhgT3/DTfQrVu3bMvodDoeeeQRALZs2VJssQkhhBBCFIUykeSdOnUK43/jKNWrl/MYoOnLwsPDuX37drHEJoQQQghRFMpEP3nXr183PQ4KCsqxXMZl169fx9vbO0uZ5ORkU2tc0Fq4CCGEEEKUNGWiJi82Ntb02NnZOcdyGZdlXCejWbNm4eHhYZqCg4PNF6gQQgghhJmUiSTPnKZOnUp0dLRpCg0NtXRIQgghhBBZlInLtW5ubqbHCQkJOZbLuCzjOhk5ODjg4OBgvuCEEEIIIYpAmajJCwwMND2+du1ajuUyLsu4jhBCCCGEtSkTSV7t2rXR67VDPX78eI7l0pcFBARk2+hCCCGEEMJalIkkz9nZmdatWwOwadOmbMsopdi8eTMAXbp0KbbYhBBCCCGKQpm4Jw9g2LBh/PHHH+zYsYP9+/fTokWLTMtXrlzJhQsXABg6dGiet6uUAqQrFSGEEEIUj/ScIz0HyZEqI1JTU1X9+vUVoIKCgtS2bduUUkoZDAb1448/Knd3dwWobt265Wu7oaGhCpBJJplkkkkmmWQq1ik0NDTXHEWn1P3SwNLj0qVLdOjQgUuXLgHaZVyj0UhSUhIAjRs3Zvv27Xh5eeV5m0ajkevXr+Pm5oZOpzN7zDExMQQHBxMaGoq7u7vZty/yTs5FySDnoeSQc1FyyLkoOYrjXCiliI2NJTAw0NTmIDtl5nItQKVKlTh69Chz5sxhzZo1XLx4ETs7O+rWrcvAgQMZO3Ys9vb2+dqmXq+nQoUKRRTxXe7u7vLBLSHkXJQMch5KDjkXJYeci5KjqM+Fh4fHfcuUqSQPtP7vZsyYwYwZMywdihBCCCFEkSkTrWuFEEIIIcoaSfJKOAcHB6ZNmyajbJQAci5KBjkPJYeci5JDzkXJUZLORZlqeCGEEEIIUVZITZ4QQgghRCkkSZ4QQgghRCkkSZ4QQgghRCkkSZ4QQgghRCkkSV4JFRsby/Tp06lfvz6urq54eHjQrFkzPvjgA1JSUiwdnlVISEhg48aNzJw5k759+xISEoJOp0On0zF9+vQ8bSMiIoKJEydSs2ZNnJyc8Pb2pm3btnz99df3HzMQOH/+PM8++yyVK1fG0dGRcuXK0bVrV1avXl3Io7Met27dYtGiRQwePJg6derg4uKCg4MDFSpUoHfv3qxdu/a+2yjs56Gw57G0OHToEDNmzOCxxx6jVq1a+Pj4YGdnh4+PD61bt+btt9/m9u3buW5DPhNF69133zV9T91vFCX5XJjH4sWLM73mOU3btm3LcRuFfV8fOnSIwYMHU6FCBRwcHChfvjx9+vTht99+K9zB5XcMWFH0Ll26pCpVqmQam87Z2Vk5ODiYnjdu3Fjdvn3b0mGWeDt27MhxvL9p06bdd/2///5b+fj4mNZxdXVVtra2puddu3ZVycnJOa7/66+/KmdnZ1N5d3d3pdfrTc+HDx+ujEajGY+4ZMr4mgHK0dFRubi4ZJrXrVs3FR8fn+36hf08FPY8liZjxozJci7c3NwyzfP19VV79+7Ndn35TBSt06dPK0dHx0znIyfyuTCfRYsWKUDp9Xrl7++f4/T7779nu35h39cLFizI9Np7eHgonU6Xr/9XOZEkr4RJTU1V9evXV4AqX7682rp1q1JKKYPBoJYvX276Qu7evbuFIy35duzYoby8vFSnTp3U5MmT1bJly1RAQECePjRRUVGmsrVq1VJ//fWXUkqp5ORkNW/ePGVnZ6cANXr06GzXv3DhgimRad26tTpz5oxSSqnY2Fj1xhtvmD687733nlmPuSQCVPPmzdX8+fPV+fPnTfMvXryonn76adNrMXjw4CzrFvbzUNjzWNp8++23avbs2Wrfvn3qzp07pvmxsbHq22+/VeXKlVOA8vPzU1FRUZnWlc9E0TIYDKpVq1YKUC1btsw1yZPPhXmlJ3khISH5Xrew7+u9e/cqGxsbBajevXur0NBQpZRSkZGR6tlnnzWtv2LFigIdmyR5JczXX39tOqnZ/Zr+4YcfTMu3bdtmgQitR1paWpZ5ISEheUryXnvtNQUoJycndeHChSzL33nnHQUoGxsb04c6o8GDBytABQQEZPpnmm7UqFGmX3ylvVb2t99+y3V5xi+yK1euZFpW2M9DYc9jWbN582bT67l06dJMy+QzUbQ++ugjBahBgwapadOm5ZrkyefCvAqT5BX2fd2mTRsFqPr166uUlJQsy7t27aoAValSpWz/p92PJHklTNu2bRWgOnTokO1yo9GoKleurAA1dOjQYo7O+uU1yatYsaKpmj07sbGxytXVVQHqjTfeyLQsLi5OOTk5KUDNmDEj2/UvXrxo+hJeuHBhgY6ltDhw4IDptVizZk2mZYX9PBTmPJZF0dHRpnPx7rvvZlomn4mik14b5OPjo27cuHHfJE8+F+ZV0CSvsO/r8+fPm5Z9++232a6/c+dOU5n7/WDOjjS8KEESEhLYs2cPAN26dcu2jE6n45FHHgFgy5YtxRZbWXLmzBmuXLkC5HweXF1dadu2LZD1POzevZvExMRc169UqRK1a9fOdv2yxtHR0fTYYDCYHhf281DY81gW/fHHH6bHVatWNT2Wz0TRGjlyJPHx8cydO5dy5crlWlY+FyVHYd/XW7duNT1OP1/3atOmDW5ubtmunxeS5JUgp06dwmg0AlCvXr0cy6UvCw8Pv29LOJF/x48fNz3Oy3k4efJkodY/ceJEgeIsLXbu3Gl6XL9+fdPjwn4eCnsey4rk5GQuXbrEvHnzGDJkCADVqlWjZ8+epjLymSg6CxYsYPv27XTu3JmhQ4fet7x8LorOzZs3adq0Ka6urjg5OVGlShUGDx6c6Tsqo8K+r9PX9/Pzw8/PL9t1bWxsqFWrVrbr54UkeSXI9evXTY+DgoJyLJdxWcZ1hHnk9zzExMQQFxeXZX0vLy+cnJzuu35ZPodRUVHMmjULgLZt21KzZk3TssJ+Hgp7Hks7R0dHdDodjo6OVK5cmbFjx3Lnzh1at27N9u3bMw2uLp+JonHt2jUmT56Mk5MTX375ZZ7Wkc9F0UlISODQoUPY29tjNBq5ePEi33//PR06dGDEiBGkpaVlKl/Y93X689zOQ27r54UkeSVIbGys6bGzs3OO5TIuy7iOMI/Cnof0x7mtm3F5WT2HRqORIUOGEBYWhqOjI/Pmzcu03FznoaDrl3YBAQH4+/vj4uJimtehQwc++ugjKlasmKmsfCaKxrPPPkt0dDTTp0+nSpUqeVpHPhfmFxgYyLRp0zhy5AhJSUncvn3bdFm8c+fOACxatIjx48dnWq+w7+vi+FxIkieEsIiXXnqJ9evXA/DZZ5/RoEEDC0dUtly6dInw8HDi4uKIiIhgzpw5HD58mObNm/PGG29YOrxSb+nSpfz66680atSICRMmWDqcMq1Lly5Mnz6dBg0amGqwbWxsaNWqFZs3b6ZXr14AzJ8/n3///deSoeabJHklSPrNlaBVG+ck47KM6wjzKOx5SH+c27oZl5fFczhp0iRTzd2HH37IiBEjspQx13ko6PpliZ+fHxMnTmTTpk3odDreeustUwIO8pkwt4iICMaNG4eNjQ0LFizA1tY2z+vK56J46fV65syZA2hXH3755RfTssK+r4vjcyFJXgkSGBhoenzt2rUcy2VclnEdYR75PQ/u7u64urpmWf/OnTumlle5rV/WzuHLL7/MBx98AMCcOXMYN25ctuUK+3ko7Hksi5o3b06bNm0A+Oqrr0zz5TNhXlOmTOHWrVuMGjWKWrVqERcXl2nKOCTZvfPkc1H8qlWrhq+vLwAXLlwwzS/s+zr9eW7nIbf180KSvBKkdu3a6PXaKcnYaude6csCAgLw9vYultjKkoytpPJyHurUqVOo9evWrVugOK3R5MmTmT17NgDvv/8+EydOzLFsYT8PhT2PZVX6Td7nzp0zzZPPhHldvHgRgM8//xw3N7csU3pjJMA07+WXXwbkc1GSFPZ9nb7+jRs3uHnzZrbrGgwGTp8+ne36eSFJXgni7OxM69atAdi0aVO2ZZRSbN68GdDuIxDmV6NGDdON5zmdh/j4eFOfYveehzZt2phaWuW0/uXLlzl16lS265dWkyZNMl32eP/995k8eXKu5Qv7eSjseSyr0msqMl4aks9EySGfi+J3/vx5IiMjAahcubJpfmHf1w8//LDpcU7r79mzx9TgokDnIt/dJ4silT5cjU6nU3/++WeW5StWrJBhzQohv8OaOTs7q4sXL2ZZ/t577+VpCKfy5ctnGQNUKaVGjx6tAOXm5lYmhnCaOHGi6X07Z86cPK9X2M9DYc9jaZKWlpbrIOlKKbVt2zbTwOj/396dxTTRxVEAPy112qIIShUiKriiREXjLq7gElzigkZDjDtooj6QgImJUdEHH9ziEpcoipqYuOAbiRuCsWo0qHGJIS5pEBdQXABZROH/PfB1QqXQYlVwOL9kEjp37u2dtrc9zLRz161b51DGMfH3uDutGceF51yNierqapk9e7YAEL1eLzk5OQ7lnr6u7dOahYeHO53WLDo6Wp2Ng9OaaUDtiaeDgoLUAVpVVSVnz56Vtm3bCgCJjo5u4p7+Gz59+iQfPnxQly5duggASUpKclhfUlLiUK/2BN5hYWGSnZ0tIjUTeB84cEAURXF7MvYxY8bIs2fPRKRmGpzk5GT1g7QlTMaelJSkfuDs2rWrUXU9HQ+ePo9aYrPZJDw8XA4dOiQvX750+HB79eqVbNu2TX3Ntm/fXt69e+dQn2Pi73EV8jgufh+bzSZDhw6tMy6qqqrk9u3b6tyx9T0enr6ub968KV5eXgJA5syZI69fvxYRkY8fP6oBEYCcOXPml/aPIa8ZstlsEhISoj653t7eYjKZ1NuDBg1q8f/pust+5M7Vsnjx4jp1s7Ozxd/fX93Gx8dHWrVqpd6ePHmyVFRU1Hvf6enp4u3trW7v6+urDmb8P2+kq/8i/3W5ubnq/ur1egkICGhw2b59e502PB0Pnj6PWlF7Dk0AoiiKWCwW9QPKvnTr1k3u37/vtA2Oib/DVcgT4bj4XX4eF0ajUSwWixiNRof1S5cule/fvzttw9PX9ZEjR8RgMKjb+/n5qeHQnTNPDWHIa6aKi4tl48aN0q9fP2ndurX4+PjI4MGDZceOHfLt27em7t4/w5OQJyKSn58vCQkJ0qtXLzGZTOLn5yejR4+WI0eOSFVVlcv7f/HihcTFxUlISIj65jFp0iQ5f/78b97T5unnN1BXS31vZp6OB0+fRy349u2bnDt3TlavXi1DhgyRTp06iaIoYjabpWvXrjJjxgw5evSolJWVNdgOx8Sf507IE+G4+B3Kyspk3759EhsbK2FhYdKhQwcxGAzSpk0b6dOnjyxbtkysVqvLdjx9Xd+7d09iY2MlKChIFEWRgIAAmTVrlmRkZHi0fzoRERARERGRpvDXtUREREQaxJBHREREpEEMeUREREQaxJBHREREpEEMeUREREQaxJBHREREpEEMeUREREQaxJBHREREpEEMeUREREQaxJBHRKRhOp0OOp0OWVlZTd0VIvrLGPKIqEXZvHmzGnzcWYiI/lWGpu4AEVFTCQgIaOouEBH9MQx5RNRi5efnN3UXiIj+GJ6uJSIiItIghjwiIjeFhIRAp9MhNTUVJSUlWL9+PUJDQ2E2m2GxWDBr1izcuXOnwTaqqqpw7NgxREZGwmKxwGg0IigoCPPmzXPrxxF5eXlYt24dBg4cCF9fX5jNZvTo0QMzZ87EyZMnUVFRUW/dkpISbNiwAX369IHZbIa/vz+mT5/uss9E9I8SIqIWZNOmTQJAfuXtLzg4WADIrl27JDQ0VACIoijStm1btU29Xi8pKSlO63/58kXGjx+vbuvl5SV+fn6i0+nUdYmJifXe/8mTJ8VkMqnbKooi/v7+YjAY1HUPHjxwqGNff/r0aenZs6cAEJPJJN7e3g7tXLp0qdGPBxE1bzySR0TUSMnJyXj//j3Onj2L0tJSFBUV4enTpxg3bhyqq6uxcuVK3L9/v0695cuXIysrC4qiYO/evSguLsbnz5/x9u1bLFu2DACwY8cOHDp0qE7d9PR0LF68GBUVFYiIiMCNGzdQXl6OwsJClJaW4saNG4iLi4OiKE77vHr1aiiKgmvXrqG0tBRfv37F3bt3ERoaisrKSsTHx6O6uvr3PlBE1KR0IiJN3Qkior9l8+bNSE5OBuD617Xz58/Hnj171NshISHIzc0FAFy9ehVRUVEO25eXlyM8PBzPnz/H1KlTkZ6erpbduXMHI0aMAAAcPnwY8fHxde5v7ty5SEtLg8ViQV5eHkwmEwDgx48f6N27N2w2G0aPHo2MjIx6w9zP7JeB6dChA548eYKOHTs6lD9+/BgDBgwAAFitVkRERLjVLhE1fzySR0QtVkFBQYNLUVGR03oRERF1Ah4AmM1mJCUlAQAuXrzoUP/MmTMAgM6dO2PFihVO2926dSsAoLCwEFeuXFHXZ2ZmwmazAQB2797tdsCrLT4+vk7AA4D+/fujW7duAIBHjx41ul0iar4Y8oioxRKRBpfU1FSn9SIjI+tt015WXV3tcMo2OzsbADBhwgTo9c7fevv27YugoCCH7QHg1q1bAIDAwEAMGTLE/R2sZfjw4fWWderUCQDw6dOnX2qbiJonhjwiokayBzFXZe/fv6/zd0N1gZojfT/XtV/PLzg4uPGd/Z+Pj0+9ZQZDzSVTv3///svtE1Hzw5BHRNTMcXo1IvoVDHlERI305s0bt8pqfwfO/vfr168bbNteXrtuYGAgAKg/+iAicgdDHhFRI2VmZros0+v1GDRokLre/l26zMzMei9VkpOTo4bEoUOHqutHjRoFoOa0be3v6hERNYQhj4iokaxWq9PZKSoqKrBz504AwJQpU+Dn56eWLViwAEDNkb6jR486bXfjxo0AAIvFgokTJ6rrJ0yYgO7duwMAEhISUFlZ+Tt2g4g0jiGPiKiRfH19ERMTg/Pnz+PHjx8Aao7CTZs2DTk5OfDy8sKWLVsc6gwbNgwxMTEAgLVr12L//v0oKysDUHOELi4uDufOnQNQcykV+zXyAMDLywv79++HTqeD1WpFVFQUrFarekSwsrISWVlZWLhwIZ4+ffrH95+I/g2Gpu4AEVFTsX/XrSEXLlxQT5fabdq0CYcPH8a8efNgNBphMpnUa+LpdDocPHjQ6aVOUlJSUFhYiOvXr2Pt2rVISEiAj48Pvnz5Avt16RMTE7Fq1ao6daOjo5Gamor4+HhYrVaMGTMGRqMRbdq0QVFRkRo2ExMTG/04EJE2MeQRUYtVUFDgchtnp0bbtWuHu3fvYtu2bUhLS0NeXh7at2+PiIgIrF+/HiNHjnTalq+vLzIyMnDixAmcOnUKDx8+xNevXxEYGIhRo0ZhzZo1GD9+fL19WbRoEcaOHYs9e/bg8uXLyM3NRXl5OYKDg9G/f3/ExMSgb9++bu8/EWkbpzUjInKTfVqz48ePY8mSJU3dHSKiBvE7eUREREQaxJBHREREpEEMeUREREQaxJBHREREpEH84QURERGRBvFIHhEREZEGMeQRERERaRBDHhEREZEGMeQRERERaRBDHhEREZEGMeQRERERaRBDHhEREZEGMeQRERERaRBDHhEREZEG/QfOzvVImSnw9wAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 700x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from mpl_toolkits.axes_grid1.inset_locator import inset_axes\n",
    "# Get training and test loss histories\n",
    "training_loss = history.history['loss']\n",
    "test_loss = history.history['val_loss']\n",
    "\n",
    "# Create count of the number of epochs\n",
    "epoch_count = range(1, len(training_loss) + 1)\n",
    "\n",
    "fig, ax = plt.subplots(1, figsize=[7, 5])\n",
    "# Visualize loss history\n",
    "plt.plot(epoch_count, training_loss, 'r--')\n",
    "plt.plot(epoch_count, test_loss, 'b-')\n",
    "plt.legend(['Training Loss', 'Validation Loss'], loc='upper right', fontsize=18)\n",
    "plt.xlabel('Epoch', fontsize=18)\n",
    "plt.ylabel('Loss', fontsize=18)\n",
    "plt.tick_params(labelsize=18)\n",
    "axins2 = inset_axes(ax, width='40%', height='40%', loc='center')\n",
    "axins2.tick_params(labelleft=True, labelbottom=True)\n",
    "plt.plot(epoch_count[5:100], training_loss[5:100], 'r--')\n",
    "plt.plot(epoch_count[5:100], test_loss[5:100], 'b-')\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "68241a80",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "267/267 [==============================] - 1s 5ms/step - loss: 0.3905 - mean_squared_error: 0.3905\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.3905055522918701, 0.3905055522918701]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "cdcc147f",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " dense (Dense)               (None, 256)               1024      \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 480)               123360    \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 2)                 962       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 125,346\n",
      "Trainable params: 125,346\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "f62d9ad2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "267/267 [==============================] - 1s 2ms/step\n",
      "0.40684981982975676\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import mean_squared_error\n",
    "Y_test = y_test['M']*data['M'].std()+data['M'].mean()\n",
    "y_pred = model.predict(X_test)\n",
    "y_pred_M = ([row[0] for row in y_pred])\n",
    "y_pred_M = np.array(y_pred_M)\n",
    "Y_pred_M = y_pred_M*data['M'].std()+data['M'].mean()\n",
    "print(mean_squared_error(Y_test, Y_pred_M))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "189aac1d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2.  2.2 3.1 ... 3.1 2.6 3.7]\n"
     ]
    }
   ],
   "source": [
    "Y_test = np.array(Y_test)\n",
    "print(Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "b6bb1cff",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2.9133308 2.4597435 3.4839587 ... 3.8725743 2.549524  3.6369133]\n"
     ]
    }
   ],
   "source": [
    "print(Y_pred_M)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "247ce797",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "267/267 [==============================] - 1s 2ms/step\n",
      "396.8934362285743\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import mean_squared_error\n",
    "Y_test = y_test['Depth']*data['Depth'].std()+data['Depth'].mean()\n",
    "y_pred = model.predict(X_test)\n",
    "y_pred_D = ([row[1] for row in y_pred])\n",
    "y_pred_D = np.array(y_pred_D)\n",
    "Y_pred_D = y_pred_D*data['Depth'].std()+data['Depth'].mean()\n",
    "print(mean_squared_error(Y_test, Y_pred_D))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "44f98662",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.10.6 ('earthquake')",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.6"
  },
  "vscode": {
   "interpreter": {
    "hash": "bcc1dba778c8d39367a1997aa22541d7d8a758f7365af005214ff65c00692215"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}