22_선형회귀.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 딥러닝에서의 회귀\n",
    "- 딥러닝은 회귀를 수행하는 작업이다.\n",
    "- 따라서 회귀를 수행할 때는 분류와 크게 다르지 않다.\n",
    "- 분류가 회귀를 통해 얻은 결과를 활성화 함수를 통해 0 ~ 1사이로 맞추고 그 결과를 통해 최종 결과를 도출하는 작업을 하게 된다.\n",
    "- 분류에서는 마지막에 활성화 함수를 배치를 하고 회귀를 배치를 하지 않으면 된다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "pygame 2.0.1 (SDL 2.0.14, Python 3.8.5)\n",
      "Hello from the pygame community. https://www.pygame.org/contribute.html\n"
     ]
    }
   ],
   "source": [
    "# 기본\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# 경고 뜨지 않게 설정\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "# 그래프 설정\n",
    "plt.rcParams['font.family'] = 'Malgun Gothic'\n",
    "# plt.rcParams['font.family'] = 'AppleGothic'\n",
    "plt.rcParams['font.size'] = 16\n",
    "plt.rcParams['figure.figsize'] = 20, 10\n",
    "plt.rcParams['axes.unicode_minus'] = False\n",
    "\n",
    "# 데이터 전처리 알고리즘\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "# 학습용과 검증용으로 나누는 함수\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# 교차 검증\n",
    "# 지표를 하나만 설정할 경우\n",
    "from sklearn.model_selection import cross_val_score\n",
    "# 지표를 하나 이상 설정할 경우\n",
    "from sklearn.model_selection import cross_validate\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn.model_selection import StratifiedKFold\n",
    "\n",
    "# 모델의 최적의 하이퍼파라미터를 찾기 위한 도구\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "# 평가함수\n",
    "# 분류용\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import precision_score\n",
    "from sklearn.metrics import recall_score\n",
    "from sklearn.metrics import f1_score\n",
    "from sklearn.metrics import roc_auc_score\n",
    "\n",
    "# 회귀용\n",
    "from sklearn.metrics import r2_score\n",
    "from sklearn.metrics import mean_squared_error\n",
    "\n",
    "# 머신러닝 알고리즘 - 분류\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "from lightgbm import LGBMClassifier\n",
    "from xgboost import XGBClassifier\n",
    "from sklearn.ensemble import VotingClassifier\n",
    "\n",
    "# 머신러닝 알고리즘 - 회귀\n",
    "from sklearn.neighbors import KNeighborsRegressor\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.linear_model import Ridge\n",
    "from sklearn.linear_model import Lasso\n",
    "from sklearn.linear_model import ElasticNet\n",
    "from sklearn.svm import SVR\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.ensemble import AdaBoostRegressor\n",
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "from lightgbm import LGBMRegressor\n",
    "from xgboost import XGBRegressor\n",
    "from sklearn.ensemble import VotingRegressor\n",
    "\n",
    "# 차원축소\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
    "\n",
    "# 군집화\n",
    "from sklearn.cluster import KMeans\n",
    "from sklearn.cluster import MeanShift\n",
    "from sklearn.cluster import estimate_bandwidth\n",
    "\n",
    "# ARIMA (시계열 예측)\n",
    "from statsmodels.tsa.arima_model import ARIMA\n",
    "import statsmodels.api as sm\n",
    "\n",
    "# 시간 측정을 위한 시간 모듈\n",
    "import datetime\n",
    "# 주식 정보를 읽어오기 위한 라이브러리\n",
    "from pandas_datareader import data\n",
    "\n",
    "# 형태소 백터를 생성하기 위한 라이브러리\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "# 형태소 백터를 학습 백터로 변환한다.\n",
    "from sklearn.feature_extraction.text import TfidfTransformer\n",
    "\n",
    "# 데이터 수집\n",
    "import requests\n",
    "from bs4 import BeautifulSoup\n",
    "import re\n",
    "import time\n",
    "import os\n",
    "import json\n",
    "\n",
    "# 한국어 형태소 분석\n",
    "from konlpy.tag import Okt, Hannanum, Kkma, Mecab, Komoran\n",
    "\n",
    "# 워드 클라우드를 위한 라이브러리\n",
    "from collections import Counter\n",
    "import pytagcloud\n",
    "from IPython.display import Image\n",
    "\n",
    "# 출력 창 청소를 위한 함수\n",
    "from IPython.display import clear_output\n",
    "\n",
    "# 저장\n",
    "import pickle\n",
    "\n",
    "# 딥러닝\n",
    "import tensorflow as tf\n",
    "\n",
    "# 딥러닝 모델 구조를 정의하는 것\n",
    "from tensorflow.keras.models import Sequential\n",
    "# 층구조를 정의하는 것\n",
    "from tensorflow.keras.layers import Dense\n",
    "# 활성화 함수를 정의하는 것\n",
    "from tensorflow.keras.layers import Activation\n",
    "\n",
    "# 다중 분류를 위한 원핫 인코딩\n",
    "# 결과데이터의 종류 수 만큼 결과데이터의 컬럼을 늘리는 작업\n",
    "from tensorflow.keras.utils import to_categorical\n",
    "\n",
    "# 저장된 학습 모델을 읽어온다.\n",
    "from tensorflow.keras.models import load_model\n",
    "\n",
    "# epoch마다 모델을 저장하는 함수\n",
    "from tensorflow.keras.callbacks import ModelCheckpoint\n",
    "# 더이상 성능 향상이 이루어지지 않는다면 조기 중단 시킬 수 있다.\n",
    "from tensorflow.keras.callbacks import EarlyStopping\n",
    "\n",
    "# 현재 프로젝트를 gpu에 할당한다.\n",
    "# 컴퓨터의 GPU는 메모리를 가지고 있다.\n",
    "gpus = tf.config.experimental.list_physical_devices('GPU')\n",
    "# gpu가 있다면..\n",
    "if len(gpus) > 0 :\n",
    "    try :\n",
    "        for gpu in gpus :\n",
    "            tf.config.experimental.set_memory_growth(gpu, True)\n",
    "    except RuntimeError as e :\n",
    "        print(e)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# seed 설정\n",
    "np.random.seed(1)\n",
    "tf.random.set_seed(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.00632</td>\n",
       "      <td>18.0</td>\n",
       "      <td>2.31</td>\n",
       "      <td>0</td>\n",
       "      <td>0.538</td>\n",
       "      <td>6.575</td>\n",
       "      <td>65.2</td>\n",
       "      <td>4.0900</td>\n",
       "      <td>1</td>\n",
       "      <td>296.0</td>\n",
       "      <td>15.3</td>\n",
       "      <td>396.90</td>\n",
       "      <td>4.98</td>\n",
       "      <td>24.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.02731</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.07</td>\n",
       "      <td>0</td>\n",
       "      <td>0.469</td>\n",
       "      <td>6.421</td>\n",
       "      <td>78.9</td>\n",
       "      <td>4.9671</td>\n",
       "      <td>2</td>\n",
       "      <td>242.0</td>\n",
       "      <td>17.8</td>\n",
       "      <td>396.90</td>\n",
       "      <td>9.14</td>\n",
       "      <td>21.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.02729</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.07</td>\n",
       "      <td>0</td>\n",
       "      <td>0.469</td>\n",
       "      <td>7.185</td>\n",
       "      <td>61.1</td>\n",
       "      <td>4.9671</td>\n",
       "      <td>2</td>\n",
       "      <td>242.0</td>\n",
       "      <td>17.8</td>\n",
       "      <td>392.83</td>\n",
       "      <td>4.03</td>\n",
       "      <td>34.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.03237</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.18</td>\n",
       "      <td>0</td>\n",
       "      <td>0.458</td>\n",
       "      <td>6.998</td>\n",
       "      <td>45.8</td>\n",
       "      <td>6.0622</td>\n",
       "      <td>3</td>\n",
       "      <td>222.0</td>\n",
       "      <td>18.7</td>\n",
       "      <td>394.63</td>\n",
       "      <td>2.94</td>\n",
       "      <td>33.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.06905</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.18</td>\n",
       "      <td>0</td>\n",
       "      <td>0.458</td>\n",
       "      <td>7.147</td>\n",
       "      <td>54.2</td>\n",
       "      <td>6.0622</td>\n",
       "      <td>3</td>\n",
       "      <td>222.0</td>\n",
       "      <td>18.7</td>\n",
       "      <td>396.90</td>\n",
       "      <td>5.33</td>\n",
       "      <td>36.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>501</th>\n",
       "      <td>0.06263</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>0</td>\n",
       "      <td>0.573</td>\n",
       "      <td>6.593</td>\n",
       "      <td>69.1</td>\n",
       "      <td>2.4786</td>\n",
       "      <td>1</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>391.99</td>\n",
       "      <td>9.67</td>\n",
       "      <td>22.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>502</th>\n",
       "      <td>0.04527</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>0</td>\n",
       "      <td>0.573</td>\n",
       "      <td>6.120</td>\n",
       "      <td>76.7</td>\n",
       "      <td>2.2875</td>\n",
       "      <td>1</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>396.90</td>\n",
       "      <td>9.08</td>\n",
       "      <td>20.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>503</th>\n",
       "      <td>0.06076</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>0</td>\n",
       "      <td>0.573</td>\n",
       "      <td>6.976</td>\n",
       "      <td>91.0</td>\n",
       "      <td>2.1675</td>\n",
       "      <td>1</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>396.90</td>\n",
       "      <td>5.64</td>\n",
       "      <td>23.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>504</th>\n",
       "      <td>0.10959</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>0</td>\n",
       "      <td>0.573</td>\n",
       "      <td>6.794</td>\n",
       "      <td>89.3</td>\n",
       "      <td>2.3889</td>\n",
       "      <td>1</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>393.45</td>\n",
       "      <td>6.48</td>\n",
       "      <td>22.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>505</th>\n",
       "      <td>0.04741</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>0</td>\n",
       "      <td>0.573</td>\n",
       "      <td>6.030</td>\n",
       "      <td>80.8</td>\n",
       "      <td>2.5050</td>\n",
       "      <td>1</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>396.90</td>\n",
       "      <td>7.88</td>\n",
       "      <td>11.9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>506 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          0     1      2   3      4      5     6       7   8      9     10  \\\n",
       "0    0.00632  18.0   2.31   0  0.538  6.575  65.2  4.0900   1  296.0  15.3   \n",
       "1    0.02731   0.0   7.07   0  0.469  6.421  78.9  4.9671   2  242.0  17.8   \n",
       "2    0.02729   0.0   7.07   0  0.469  7.185  61.1  4.9671   2  242.0  17.8   \n",
       "3    0.03237   0.0   2.18   0  0.458  6.998  45.8  6.0622   3  222.0  18.7   \n",
       "4    0.06905   0.0   2.18   0  0.458  7.147  54.2  6.0622   3  222.0  18.7   \n",
       "..       ...   ...    ...  ..    ...    ...   ...     ...  ..    ...   ...   \n",
       "501  0.06263   0.0  11.93   0  0.573  6.593  69.1  2.4786   1  273.0  21.0   \n",
       "502  0.04527   0.0  11.93   0  0.573  6.120  76.7  2.2875   1  273.0  21.0   \n",
       "503  0.06076   0.0  11.93   0  0.573  6.976  91.0  2.1675   1  273.0  21.0   \n",
       "504  0.10959   0.0  11.93   0  0.573  6.794  89.3  2.3889   1  273.0  21.0   \n",
       "505  0.04741   0.0  11.93   0  0.573  6.030  80.8  2.5050   1  273.0  21.0   \n",
       "\n",
       "         11    12    13  \n",
       "0    396.90  4.98  24.0  \n",
       "1    396.90  9.14  21.6  \n",
       "2    392.83  4.03  34.7  \n",
       "3    394.63  2.94  33.4  \n",
       "4    396.90  5.33  36.2  \n",
       "..      ...   ...   ...  \n",
       "501  391.99  9.67  22.4  \n",
       "502  396.90  9.08  20.6  \n",
       "503  396.90  5.64  23.9  \n",
       "504  393.45  6.48  22.0  \n",
       "505  396.90  7.88  11.9  \n",
       "\n",
       "[506 rows x 14 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 데이터를 읽어온다.\n",
    "# 이 파일은 데이터가 띄어쓰기로 구분되어 있으므로\n",
    "# delim_whitespace를 True로 설정한다.\n",
    "df1 = pd.read_csv('data/housing.csv', delim_whitespace=True, header=None)\n",
    "df1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 506 entries, 0 to 505\n",
      "Data columns (total 14 columns):\n",
      " #   Column  Non-Null Count  Dtype  \n",
      "---  ------  --------------  -----  \n",
      " 0   0       506 non-null    float64\n",
      " 1   1       506 non-null    float64\n",
      " 2   2       506 non-null    float64\n",
      " 3   3       506 non-null    int64  \n",
      " 4   4       506 non-null    float64\n",
      " 5   5       506 non-null    float64\n",
      " 6   6       506 non-null    float64\n",
      " 7   7       506 non-null    float64\n",
      " 8   8       506 non-null    int64  \n",
      " 9   9       506 non-null    float64\n",
      " 10  10      506 non-null    float64\n",
      " 11  11      506 non-null    float64\n",
      " 12  12      506 non-null    float64\n",
      " 13  13      506 non-null    float64\n",
      "dtypes: float64(12), int64(2)\n",
      "memory usage: 55.5 KB\n"
     ]
    }
   ],
   "source": [
    "df1.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     0\n",
       "1     0\n",
       "2     0\n",
       "3     0\n",
       "4     0\n",
       "5     0\n",
       "6     0\n",
       "7     0\n",
       "8     0\n",
       "9     0\n",
       "10    0\n",
       "11    0\n",
       "12    0\n",
       "13    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 결측치 확인\n",
    "df1.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.00632</td>\n",
       "      <td>18.0</td>\n",
       "      <td>2.31</td>\n",
       "      <td>0</td>\n",
       "      <td>0.538</td>\n",
       "      <td>6.575</td>\n",
       "      <td>65.2</td>\n",
       "      <td>4.0900</td>\n",
       "      <td>1</td>\n",
       "      <td>296.0</td>\n",
       "      <td>15.3</td>\n",
       "      <td>396.90</td>\n",
       "      <td>4.98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.02731</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.07</td>\n",
       "      <td>0</td>\n",
       "      <td>0.469</td>\n",
       "      <td>6.421</td>\n",
       "      <td>78.9</td>\n",
       "      <td>4.9671</td>\n",
       "      <td>2</td>\n",
       "      <td>242.0</td>\n",
       "      <td>17.8</td>\n",
       "      <td>396.90</td>\n",
       "      <td>9.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.02729</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.07</td>\n",
       "      <td>0</td>\n",
       "      <td>0.469</td>\n",
       "      <td>7.185</td>\n",
       "      <td>61.1</td>\n",
       "      <td>4.9671</td>\n",
       "      <td>2</td>\n",
       "      <td>242.0</td>\n",
       "      <td>17.8</td>\n",
       "      <td>392.83</td>\n",
       "      <td>4.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.03237</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.18</td>\n",
       "      <td>0</td>\n",
       "      <td>0.458</td>\n",
       "      <td>6.998</td>\n",
       "      <td>45.8</td>\n",
       "      <td>6.0622</td>\n",
       "      <td>3</td>\n",
       "      <td>222.0</td>\n",
       "      <td>18.7</td>\n",
       "      <td>394.63</td>\n",
       "      <td>2.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.06905</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.18</td>\n",
       "      <td>0</td>\n",
       "      <td>0.458</td>\n",
       "      <td>7.147</td>\n",
       "      <td>54.2</td>\n",
       "      <td>6.0622</td>\n",
       "      <td>3</td>\n",
       "      <td>222.0</td>\n",
       "      <td>18.7</td>\n",
       "      <td>396.90</td>\n",
       "      <td>5.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>501</th>\n",
       "      <td>0.06263</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>0</td>\n",
       "      <td>0.573</td>\n",
       "      <td>6.593</td>\n",
       "      <td>69.1</td>\n",
       "      <td>2.4786</td>\n",
       "      <td>1</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>391.99</td>\n",
       "      <td>9.67</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>502</th>\n",
       "      <td>0.04527</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>0</td>\n",
       "      <td>0.573</td>\n",
       "      <td>6.120</td>\n",
       "      <td>76.7</td>\n",
       "      <td>2.2875</td>\n",
       "      <td>1</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>396.90</td>\n",
       "      <td>9.08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>503</th>\n",
       "      <td>0.06076</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>0</td>\n",
       "      <td>0.573</td>\n",
       "      <td>6.976</td>\n",
       "      <td>91.0</td>\n",
       "      <td>2.1675</td>\n",
       "      <td>1</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>396.90</td>\n",
       "      <td>5.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>504</th>\n",
       "      <td>0.10959</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>0</td>\n",
       "      <td>0.573</td>\n",
       "      <td>6.794</td>\n",
       "      <td>89.3</td>\n",
       "      <td>2.3889</td>\n",
       "      <td>1</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>393.45</td>\n",
       "      <td>6.48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>505</th>\n",
       "      <td>0.04741</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>0</td>\n",
       "      <td>0.573</td>\n",
       "      <td>6.030</td>\n",
       "      <td>80.8</td>\n",
       "      <td>2.5050</td>\n",
       "      <td>1</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>396.90</td>\n",
       "      <td>7.88</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>506 rows × 13 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          0     1      2   3      4      5     6       7   8      9     10  \\\n",
       "0    0.00632  18.0   2.31   0  0.538  6.575  65.2  4.0900   1  296.0  15.3   \n",
       "1    0.02731   0.0   7.07   0  0.469  6.421  78.9  4.9671   2  242.0  17.8   \n",
       "2    0.02729   0.0   7.07   0  0.469  7.185  61.1  4.9671   2  242.0  17.8   \n",
       "3    0.03237   0.0   2.18   0  0.458  6.998  45.8  6.0622   3  222.0  18.7   \n",
       "4    0.06905   0.0   2.18   0  0.458  7.147  54.2  6.0622   3  222.0  18.7   \n",
       "..       ...   ...    ...  ..    ...    ...   ...     ...  ..    ...   ...   \n",
       "501  0.06263   0.0  11.93   0  0.573  6.593  69.1  2.4786   1  273.0  21.0   \n",
       "502  0.04527   0.0  11.93   0  0.573  6.120  76.7  2.2875   1  273.0  21.0   \n",
       "503  0.06076   0.0  11.93   0  0.573  6.976  91.0  2.1675   1  273.0  21.0   \n",
       "504  0.10959   0.0  11.93   0  0.573  6.794  89.3  2.3889   1  273.0  21.0   \n",
       "505  0.04741   0.0  11.93   0  0.573  6.030  80.8  2.5050   1  273.0  21.0   \n",
       "\n",
       "         11    12  \n",
       "0    396.90  4.98  \n",
       "1    396.90  9.14  \n",
       "2    392.83  4.03  \n",
       "3    394.63  2.94  \n",
       "4    396.90  5.33  \n",
       "..      ...   ...  \n",
       "501  391.99  9.67  \n",
       "502  396.90  9.08  \n",
       "503  396.90  5.64  \n",
       "504  393.45  6.48  \n",
       "505  396.90  7.88  \n",
       "\n",
       "[506 rows x 13 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0      24.0\n",
       "1      21.6\n",
       "2      34.7\n",
       "3      33.4\n",
       "4      36.2\n",
       "       ... \n",
       "501    22.4\n",
       "502    20.6\n",
       "503    23.9\n",
       "504    22.0\n",
       "505    11.9\n",
       "Name: 13, Length: 506, dtype: float64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 입력과 결과로 나눈다.\n",
    "X = df1.drop(13, axis=1)\n",
    "y = df1[13]\n",
    "\n",
    "display(X)\n",
    "display(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "\n",
    "model.add(Dense(30, input_dim=13))\n",
    "model.add(Activation('relu'))\n",
    "\n",
    "model.add(Dense(20))\n",
    "model.add(Activation('relu'))\n",
    "\n",
    "# 회귀는 활성화 함수를 두지 않는다.\n",
    "model.add(Dense(1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 모델 컴파일\n",
    "model.compile(loss='mean_squared_error', optimizer='adam')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 모델들이 저장될 위치\n",
    "# epoch : 학습 횟수\n",
    "# val_loss : 검증 데이터의 손실률\n",
    "path1 = 'models/22_선형회귀/{epoch}-{val_loss}.hdf5'\n",
    "path2 = 'models/22_선형회귀/최종.hdf5'\n",
    "\n",
    "# 저장 콜백 설정\n",
    "# save_best_only : True를 넣어주면 모니터링하는 값이 개선 되었을 경우에만\n",
    "# 저장을 한다. 기본은 False 이며 False시 매번 저장을 한다.\n",
    "call1 = ModelCheckpoint(filepath=path1, monitor='val_loss', \n",
    "                       save_best_only=True)\n",
    "\n",
    "call2 = ModelCheckpoint(filepath=path2, monitor='val_loss', \n",
    "                       save_best_only=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 조기 중단\n",
    "# patience : 현재 평가 값이 이전 평가 값 보다 좋다고 판단될 경우 앞으로\n",
    "# 몇번을 더 볼 것인가를 설정하는 부분\n",
    "call3 = EarlyStopping(monitor='val_loss', patience=50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/2000\n",
      "1/1 [==============================] - 1s 1s/step - loss: 930.2162 - val_loss: 877.3423\n",
      "Epoch 2/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 826.0595 - val_loss: 791.7953\n",
      "Epoch 3/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 738.4684 - val_loss: 718.2267\n",
      "Epoch 4/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 664.1970 - val_loss: 652.2855\n",
      "Epoch 5/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 599.4406 - val_loss: 593.2211\n",
      "Epoch 6/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 542.0021 - val_loss: 538.0031\n",
      "Epoch 7/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 488.8558 - val_loss: 487.5325\n",
      "Epoch 8/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 440.2903 - val_loss: 441.2656\n",
      "Epoch 9/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 395.9294 - val_loss: 397.8363\n",
      "Epoch 10/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 354.6013 - val_loss: 357.2337\n",
      "Epoch 11/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 316.8027 - val_loss: 320.4661\n",
      "Epoch 12/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 282.8561 - val_loss: 286.0987\n",
      "Epoch 13/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 251.4147 - val_loss: 255.4344\n",
      "Epoch 14/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 224.0156 - val_loss: 227.9373\n",
      "Epoch 15/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 199.6208 - val_loss: 203.9082\n",
      "Epoch 16/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 178.5656 - val_loss: 183.0361\n",
      "Epoch 17/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 160.2626 - val_loss: 164.7182\n",
      "Epoch 18/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 144.2380 - val_loss: 148.6811\n",
      "Epoch 19/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 130.4500 - val_loss: 134.7306\n",
      "Epoch 20/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 118.5624 - val_loss: 122.9957\n",
      "Epoch 21/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 108.6438 - val_loss: 113.9659\n",
      "Epoch 22/2000\n",
      "1/1 [==============================] - 0s 22ms/step - loss: 100.8619 - val_loss: 106.7596\n",
      "Epoch 23/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 95.1044 - val_loss: 100.9460\n",
      "Epoch 24/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 90.5064 - val_loss: 96.3497\n",
      "Epoch 25/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 86.8918 - val_loss: 92.8195\n",
      "Epoch 26/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 84.2796 - val_loss: 89.9734\n",
      "Epoch 27/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 82.3040 - val_loss: 87.9207\n",
      "Epoch 28/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 81.0898 - val_loss: 86.4890\n",
      "Epoch 29/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 80.4599 - val_loss: 85.6003\n",
      "Epoch 30/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 80.2846 - val_loss: 84.8062\n",
      "Epoch 31/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 80.2811 - val_loss: 84.3020\n",
      "Epoch 32/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 80.3765 - val_loss: 83.7004\n",
      "Epoch 33/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 80.3995 - val_loss: 83.3434\n",
      "Epoch 34/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 80.5714 - val_loss: 82.8674\n",
      "Epoch 35/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 80.5776 - val_loss: 82.3982\n",
      "Epoch 36/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 80.6472 - val_loss: 82.0437\n",
      "Epoch 37/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 80.6353 - val_loss: 81.7016\n",
      "Epoch 38/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 80.6848 - val_loss: 81.3717\n",
      "Epoch 39/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 80.5901 - val_loss: 81.0411\n",
      "Epoch 40/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 80.3396 - val_loss: 80.7159\n",
      "Epoch 41/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 79.9265 - val_loss: 80.4117\n",
      "Epoch 42/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 79.5216 - val_loss: 80.0849\n",
      "Epoch 43/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 78.9583 - val_loss: 79.6975\n",
      "Epoch 44/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 78.2401 - val_loss: 79.3916\n",
      "Epoch 45/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 77.5342 - val_loss: 78.9912\n",
      "Epoch 46/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 76.7035 - val_loss: 78.6333\n",
      "Epoch 47/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 75.9284 - val_loss: 78.3321\n",
      "Epoch 48/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 75.1584 - val_loss: 78.0230\n",
      "Epoch 49/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 74.3461 - val_loss: 77.6197\n",
      "Epoch 50/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 73.5282 - val_loss: 77.3174\n",
      "Epoch 51/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 72.7538 - val_loss: 76.9670\n",
      "Epoch 52/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 72.0142 - val_loss: 76.6269\n",
      "Epoch 53/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 71.2210 - val_loss: 76.3470\n",
      "Epoch 54/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 70.5565 - val_loss: 76.0712\n",
      "Epoch 55/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 69.9269 - val_loss: 75.9138\n",
      "Epoch 56/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 69.3420 - val_loss: 75.8146\n",
      "Epoch 57/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 68.8466 - val_loss: 75.7154\n",
      "Epoch 58/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 68.3613 - val_loss: 75.6339\n",
      "Epoch 59/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 67.9248 - val_loss: 75.6163\n",
      "Epoch 60/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 67.5303 - val_loss: 75.7000\n",
      "Epoch 61/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 67.1628 - val_loss: 75.6518\n",
      "Epoch 62/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 66.8233 - val_loss: 75.7052\n",
      "Epoch 63/2000\n",
      "1/1 [==============================] - 0s 22ms/step - loss: 66.5066 - val_loss: 75.7587\n",
      "Epoch 64/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 66.2243 - val_loss: 75.8654\n",
      "Epoch 65/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 65.9718 - val_loss: 75.8652\n",
      "Epoch 66/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 65.6996 - val_loss: 75.8446\n",
      "Epoch 67/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 65.4369 - val_loss: 75.7401\n",
      "Epoch 68/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 65.1980 - val_loss: 75.6710\n",
      "Epoch 69/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 64.9802 - val_loss: 75.5277\n",
      "Epoch 70/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 64.7416 - val_loss: 75.3845\n",
      "Epoch 71/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 64.4903 - val_loss: 75.2092\n",
      "Epoch 72/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 64.2496 - val_loss: 74.9086\n",
      "Epoch 73/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 63.9633 - val_loss: 74.6525\n",
      "Epoch 74/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 63.6786 - val_loss: 74.3799\n",
      "Epoch 75/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 63.4039 - val_loss: 74.0540\n",
      "Epoch 76/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 63.1203 - val_loss: 73.8104\n",
      "Epoch 77/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 62.8450 - val_loss: 73.5188\n",
      "Epoch 78/2000\n",
      "1/1 [==============================] - 0s 25ms/step - loss: 62.5948 - val_loss: 73.1840\n",
      "Epoch 79/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 62.3493 - val_loss: 72.9031\n",
      "Epoch 80/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 62.1492 - val_loss: 72.6932\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 81/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 62.0309 - val_loss: 72.5421\n",
      "Epoch 82/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 61.8588 - val_loss: 72.3909\n",
      "Epoch 83/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 61.7022 - val_loss: 72.2878\n",
      "Epoch 84/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 61.5723 - val_loss: 72.2040\n",
      "Epoch 85/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 61.3915 - val_loss: 72.0620\n",
      "Epoch 86/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 61.2527 - val_loss: 71.9907\n",
      "Epoch 87/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 61.1007 - val_loss: 71.8391\n",
      "Epoch 88/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 60.9473 - val_loss: 71.8054\n",
      "Epoch 89/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 60.7904 - val_loss: 71.6744\n",
      "Epoch 90/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 60.6252 - val_loss: 71.5733\n",
      "Epoch 91/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 60.4596 - val_loss: 71.4502\n",
      "Epoch 92/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 60.2818 - val_loss: 71.4084\n",
      "Epoch 93/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 60.1102 - val_loss: 71.4168\n",
      "Epoch 94/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 59.9277 - val_loss: 71.4619\n",
      "Epoch 95/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 59.7872 - val_loss: 71.4541\n",
      "Epoch 96/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 59.6329 - val_loss: 71.4743\n",
      "Epoch 97/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 59.4992 - val_loss: 71.4186\n",
      "Epoch 98/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 59.3720 - val_loss: 71.3791\n",
      "Epoch 99/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 59.2298 - val_loss: 71.3501\n",
      "Epoch 100/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 59.1150 - val_loss: 71.2874\n",
      "Epoch 101/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 58.9968 - val_loss: 71.1948\n",
      "Epoch 102/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 58.8734 - val_loss: 71.0715\n",
      "Epoch 103/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 58.7431 - val_loss: 70.9468\n",
      "Epoch 104/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 58.5868 - val_loss: 70.8092\n",
      "Epoch 105/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 58.4631 - val_loss: 70.6782\n",
      "Epoch 106/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 58.3058 - val_loss: 70.5826\n",
      "Epoch 107/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 58.2111 - val_loss: 70.4626\n",
      "Epoch 108/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 58.0922 - val_loss: 70.3850\n",
      "Epoch 109/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 57.9977 - val_loss: 70.3720\n",
      "Epoch 110/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 57.8790 - val_loss: 70.3101\n",
      "Epoch 111/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 57.7914 - val_loss: 70.2094\n",
      "Epoch 112/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 57.6725 - val_loss: 70.1908\n",
      "Epoch 113/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 57.5623 - val_loss: 70.0961\n",
      "Epoch 114/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 57.4475 - val_loss: 70.0323\n",
      "Epoch 115/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 57.3489 - val_loss: 69.9748\n",
      "Epoch 116/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 57.2494 - val_loss: 69.9645\n",
      "Epoch 117/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 57.1467 - val_loss: 69.8671\n",
      "Epoch 118/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 57.0427 - val_loss: 69.7894\n",
      "Epoch 119/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 56.9631 - val_loss: 69.7235\n",
      "Epoch 120/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 56.8614 - val_loss: 69.6494\n",
      "Epoch 121/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 56.7751 - val_loss: 69.5895\n",
      "Epoch 122/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 56.6900 - val_loss: 69.5072\n",
      "Epoch 123/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 56.5970 - val_loss: 69.4651\n",
      "Epoch 124/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 56.5108 - val_loss: 69.4269\n",
      "Epoch 125/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 56.4275 - val_loss: 69.3799\n",
      "Epoch 126/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 56.3389 - val_loss: 69.2816\n",
      "Epoch 127/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 56.2517 - val_loss: 69.2185\n",
      "Epoch 128/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 56.1595 - val_loss: 69.1191\n",
      "Epoch 129/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 56.0862 - val_loss: 69.0407\n",
      "Epoch 130/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 55.9737 - val_loss: 69.0224\n",
      "Epoch 131/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 55.9145 - val_loss: 68.9875\n",
      "Epoch 132/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 55.8238 - val_loss: 68.8792\n",
      "Epoch 133/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 55.7441 - val_loss: 68.8508\n",
      "Epoch 134/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 55.6509 - val_loss: 68.7933\n",
      "Epoch 135/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 55.5788 - val_loss: 68.7419\n",
      "Epoch 136/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 55.5076 - val_loss: 68.6600\n",
      "Epoch 137/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 55.4383 - val_loss: 68.5773\n",
      "Epoch 138/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 55.3546 - val_loss: 68.5356\n",
      "Epoch 139/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 55.2614 - val_loss: 68.4627\n",
      "Epoch 140/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 55.1850 - val_loss: 68.3722\n",
      "Epoch 141/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 55.1288 - val_loss: 68.2620\n",
      "Epoch 142/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 55.0636 - val_loss: 68.2386\n",
      "Epoch 143/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 54.9875 - val_loss: 68.2062\n",
      "Epoch 144/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 54.8963 - val_loss: 68.1691\n",
      "Epoch 145/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 54.8396 - val_loss: 68.1217\n",
      "Epoch 146/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 54.7743 - val_loss: 68.0246\n",
      "Epoch 147/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 54.6992 - val_loss: 67.9512\n",
      "Epoch 148/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 54.6144 - val_loss: 67.8950\n",
      "Epoch 149/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 54.5456 - val_loss: 67.8776\n",
      "Epoch 150/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 54.4834 - val_loss: 67.8284\n",
      "Epoch 151/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 54.4151 - val_loss: 67.7337\n",
      "Epoch 152/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 54.3432 - val_loss: 67.6624\n",
      "Epoch 153/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 54.2692 - val_loss: 67.6505\n",
      "Epoch 154/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 54.2040 - val_loss: 67.6008\n",
      "Epoch 155/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 54.1343 - val_loss: 67.5128\n",
      "Epoch 156/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 54.0759 - val_loss: 67.4875\n",
      "Epoch 157/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 54.0080 - val_loss: 67.4159\n",
      "Epoch 158/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 53.9351 - val_loss: 67.3593\n",
      "Epoch 159/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 53.8561 - val_loss: 67.3045\n",
      "Epoch 160/2000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 16ms/step - loss: 53.7969 - val_loss: 67.2215\n",
      "Epoch 161/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 53.7361 - val_loss: 67.1738\n",
      "Epoch 162/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 53.6756 - val_loss: 67.1246\n",
      "Epoch 163/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 53.6144 - val_loss: 67.1177\n",
      "Epoch 164/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 53.5491 - val_loss: 67.0602\n",
      "Epoch 165/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 53.4795 - val_loss: 66.9825\n",
      "Epoch 166/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 53.4190 - val_loss: 66.8988\n",
      "Epoch 167/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 53.3576 - val_loss: 66.7863\n",
      "Epoch 168/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 53.2993 - val_loss: 66.7371\n",
      "Epoch 169/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 53.2443 - val_loss: 66.7351\n",
      "Epoch 170/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 53.1608 - val_loss: 66.6854\n",
      "Epoch 171/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 53.1001 - val_loss: 66.6591\n",
      "Epoch 172/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 53.0389 - val_loss: 66.5526\n",
      "Epoch 173/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 52.9875 - val_loss: 66.5724\n",
      "Epoch 174/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 52.9098 - val_loss: 66.4692\n",
      "Epoch 175/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 52.8421 - val_loss: 66.3785\n",
      "Epoch 176/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 52.7971 - val_loss: 66.2963\n",
      "Epoch 177/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 52.7417 - val_loss: 66.2096\n",
      "Epoch 178/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 52.6835 - val_loss: 66.1702\n",
      "Epoch 179/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 52.6276 - val_loss: 66.0301\n",
      "Epoch 180/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 52.5700 - val_loss: 66.0167\n",
      "Epoch 181/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 52.4815 - val_loss: 65.9666\n",
      "Epoch 182/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 52.4287 - val_loss: 65.9624\n",
      "Epoch 183/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 52.3707 - val_loss: 65.9115\n",
      "Epoch 184/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 52.3045 - val_loss: 65.8608\n",
      "Epoch 185/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 52.2538 - val_loss: 65.8040\n",
      "Epoch 186/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 52.1997 - val_loss: 65.7217\n",
      "Epoch 187/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 52.1404 - val_loss: 65.6358\n",
      "Epoch 188/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 52.0677 - val_loss: 65.4968\n",
      "Epoch 189/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 52.0143 - val_loss: 65.4493\n",
      "Epoch 190/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 51.9374 - val_loss: 65.3815\n",
      "Epoch 191/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 51.8829 - val_loss: 65.3675\n",
      "Epoch 192/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 51.8132 - val_loss: 65.2651\n",
      "Epoch 193/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 51.7616 - val_loss: 65.2139\n",
      "Epoch 194/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 51.7081 - val_loss: 65.1836\n",
      "Epoch 195/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 51.6479 - val_loss: 65.1676\n",
      "Epoch 196/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 51.5855 - val_loss: 65.1119\n",
      "Epoch 197/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 51.5113 - val_loss: 64.9631\n",
      "Epoch 198/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 51.4508 - val_loss: 64.8703\n",
      "Epoch 199/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 51.4062 - val_loss: 64.8105\n",
      "Epoch 200/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 51.3671 - val_loss: 64.7236\n",
      "Epoch 201/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 51.3097 - val_loss: 64.7112\n",
      "Epoch 202/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 51.2380 - val_loss: 64.6686\n",
      "Epoch 203/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 51.1717 - val_loss: 64.6627\n",
      "Epoch 204/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 51.1212 - val_loss: 64.5894\n",
      "Epoch 205/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 51.0595 - val_loss: 64.5081\n",
      "Epoch 206/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 51.0281 - val_loss: 64.4352\n",
      "Epoch 207/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 50.9649 - val_loss: 64.3143\n",
      "Epoch 208/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 50.9190 - val_loss: 64.2647\n",
      "Epoch 209/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 50.8622 - val_loss: 64.1985\n",
      "Epoch 210/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 50.8155 - val_loss: 64.1412\n",
      "Epoch 211/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 50.7511 - val_loss: 64.0594\n",
      "Epoch 212/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 50.6969 - val_loss: 63.9667\n",
      "Epoch 213/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 50.6520 - val_loss: 63.9229\n",
      "Epoch 214/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 50.5999 - val_loss: 63.8485\n",
      "Epoch 215/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 50.5361 - val_loss: 63.8219\n",
      "Epoch 216/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 50.4755 - val_loss: 63.6972\n",
      "Epoch 217/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 50.4207 - val_loss: 63.6776\n",
      "Epoch 218/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 50.3685 - val_loss: 63.6366\n",
      "Epoch 219/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 50.3104 - val_loss: 63.5984\n",
      "Epoch 220/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 50.2767 - val_loss: 63.5399\n",
      "Epoch 221/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 50.2157 - val_loss: 63.5024\n",
      "Epoch 222/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 50.1569 - val_loss: 63.4198\n",
      "Epoch 223/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 50.0877 - val_loss: 63.3367\n",
      "Epoch 224/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 50.0265 - val_loss: 63.2870\n",
      "Epoch 225/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 49.9990 - val_loss: 63.1736\n",
      "Epoch 226/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 49.9534 - val_loss: 63.1038\n",
      "Epoch 227/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 49.9009 - val_loss: 63.0413\n",
      "Epoch 228/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 49.8280 - val_loss: 63.0527\n",
      "Epoch 229/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 49.7907 - val_loss: 62.9394\n",
      "Epoch 230/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 49.7360 - val_loss: 62.9054\n",
      "Epoch 231/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 49.6920 - val_loss: 62.7924\n",
      "Epoch 232/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 49.6227 - val_loss: 62.6818\n",
      "Epoch 233/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 49.5588 - val_loss: 62.6389\n",
      "Epoch 234/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 49.5107 - val_loss: 62.5686\n",
      "Epoch 235/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 49.4600 - val_loss: 62.5754\n",
      "Epoch 236/2000\n",
      "1/1 [==============================] - 0s 21ms/step - loss: 49.4130 - val_loss: 62.5049\n",
      "Epoch 237/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 49.3411 - val_loss: 62.4368\n",
      "Epoch 238/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 49.2982 - val_loss: 62.3832\n",
      "Epoch 239/2000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 18ms/step - loss: 49.2456 - val_loss: 62.3150\n",
      "Epoch 240/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 49.1856 - val_loss: 62.2054\n",
      "Epoch 241/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 49.1433 - val_loss: 62.1326\n",
      "Epoch 242/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 49.0802 - val_loss: 62.0890\n",
      "Epoch 243/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 49.0324 - val_loss: 62.0289\n",
      "Epoch 244/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 48.9772 - val_loss: 61.9293\n",
      "Epoch 245/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 48.9265 - val_loss: 61.9022\n",
      "Epoch 246/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 48.8779 - val_loss: 61.8393\n",
      "Epoch 247/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 48.8265 - val_loss: 61.7836\n",
      "Epoch 248/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 48.7787 - val_loss: 61.7113\n",
      "Epoch 249/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 48.7179 - val_loss: 61.6387\n",
      "Epoch 250/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 48.6661 - val_loss: 61.5477\n",
      "Epoch 251/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 48.6172 - val_loss: 61.5227\n",
      "Epoch 252/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 48.5607 - val_loss: 61.4266\n",
      "Epoch 253/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 48.5182 - val_loss: 61.3920\n",
      "Epoch 254/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 48.4630 - val_loss: 61.3513\n",
      "Epoch 255/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 48.4037 - val_loss: 61.2657\n",
      "Epoch 256/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 48.3517 - val_loss: 61.2068\n",
      "Epoch 257/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 48.2964 - val_loss: 61.0939\n",
      "Epoch 258/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 48.2585 - val_loss: 61.0458\n",
      "Epoch 259/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 48.1996 - val_loss: 60.9496\n",
      "Epoch 260/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 48.1282 - val_loss: 60.8539\n",
      "Epoch 261/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 48.0852 - val_loss: 60.7564\n",
      "Epoch 262/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 48.0228 - val_loss: 60.7639\n",
      "Epoch 263/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 47.9833 - val_loss: 60.7130\n",
      "Epoch 264/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 47.9164 - val_loss: 60.6546\n",
      "Epoch 265/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 47.8668 - val_loss: 60.5849\n",
      "Epoch 266/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 47.8052 - val_loss: 60.4487\n",
      "Epoch 267/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 47.7616 - val_loss: 60.4229\n",
      "Epoch 268/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 47.7186 - val_loss: 60.3150\n",
      "Epoch 269/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 47.6611 - val_loss: 60.2521\n",
      "Epoch 270/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 47.6270 - val_loss: 60.1685\n",
      "Epoch 271/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 47.5642 - val_loss: 60.0894\n",
      "Epoch 272/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 47.5014 - val_loss: 60.0407\n",
      "Epoch 273/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 47.4458 - val_loss: 59.9608\n",
      "Epoch 274/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 47.4034 - val_loss: 59.9132\n",
      "Epoch 275/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 47.3434 - val_loss: 59.8669\n",
      "Epoch 276/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 47.3007 - val_loss: 59.7884\n",
      "Epoch 277/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 47.2494 - val_loss: 59.7330\n",
      "Epoch 278/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 47.1863 - val_loss: 59.6185\n",
      "Epoch 279/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 47.1280 - val_loss: 59.5336\n",
      "Epoch 280/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 47.0746 - val_loss: 59.4960\n",
      "Epoch 281/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 47.0272 - val_loss: 59.4566\n",
      "Epoch 282/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 46.9794 - val_loss: 59.4088\n",
      "Epoch 283/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 46.9316 - val_loss: 59.3213\n",
      "Epoch 284/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 46.8816 - val_loss: 59.2284\n",
      "Epoch 285/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 46.8353 - val_loss: 59.1722\n",
      "Epoch 286/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 46.7651 - val_loss: 59.1327\n",
      "Epoch 287/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 46.7118 - val_loss: 59.0915\n",
      "Epoch 288/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 46.6657 - val_loss: 59.0244\n",
      "Epoch 289/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 46.6144 - val_loss: 58.9169\n",
      "Epoch 290/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 46.5709 - val_loss: 58.8766\n",
      "Epoch 291/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 46.5166 - val_loss: 58.7837\n",
      "Epoch 292/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 46.4534 - val_loss: 58.6913\n",
      "Epoch 293/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 46.3979 - val_loss: 58.6336\n",
      "Epoch 294/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 46.3460 - val_loss: 58.5249\n",
      "Epoch 295/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 46.2971 - val_loss: 58.4686\n",
      "Epoch 296/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 46.2453 - val_loss: 58.4831\n",
      "Epoch 297/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 46.1969 - val_loss: 58.4135\n",
      "Epoch 298/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 46.1442 - val_loss: 58.3483\n",
      "Epoch 299/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 46.0930 - val_loss: 58.3042\n",
      "Epoch 300/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 46.0486 - val_loss: 58.2306\n",
      "Epoch 301/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 45.9980 - val_loss: 58.1375\n",
      "Epoch 302/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 45.9469 - val_loss: 58.1117\n",
      "Epoch 303/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 45.8717 - val_loss: 58.0325\n",
      "Epoch 304/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 45.8269 - val_loss: 57.9711\n",
      "Epoch 305/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 45.7693 - val_loss: 57.9014\n",
      "Epoch 306/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 45.7153 - val_loss: 57.8075\n",
      "Epoch 307/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 45.6733 - val_loss: 57.6904\n",
      "Epoch 308/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 45.6236 - val_loss: 57.6262\n",
      "Epoch 309/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 45.5782 - val_loss: 57.6094\n",
      "Epoch 310/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 45.5170 - val_loss: 57.5773\n",
      "Epoch 311/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 45.4690 - val_loss: 57.5571\n",
      "Epoch 312/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 45.3971 - val_loss: 57.4599\n",
      "Epoch 313/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 45.3650 - val_loss: 57.4038\n",
      "Epoch 314/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 45.3135 - val_loss: 57.2969\n",
      "Epoch 315/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 45.2571 - val_loss: 57.2324\n",
      "Epoch 316/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 45.1974 - val_loss: 57.1923\n",
      "Epoch 317/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 45.1449 - val_loss: 57.1265\n",
      "Epoch 318/2000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 18ms/step - loss: 45.0919 - val_loss: 57.0725\n",
      "Epoch 319/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 45.0311 - val_loss: 57.0102\n",
      "Epoch 320/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 44.9702 - val_loss: 56.8968\n",
      "Epoch 321/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 44.9167 - val_loss: 56.7912\n",
      "Epoch 322/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 44.8771 - val_loss: 56.7491\n",
      "Epoch 323/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 44.8189 - val_loss: 56.7278\n",
      "Epoch 324/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 44.7650 - val_loss: 56.6572\n",
      "Epoch 325/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 44.7262 - val_loss: 56.6156\n",
      "Epoch 326/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 44.6740 - val_loss: 56.5218\n",
      "Epoch 327/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 44.6151 - val_loss: 56.4169\n",
      "Epoch 328/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 44.5681 - val_loss: 56.2927\n",
      "Epoch 329/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 44.5047 - val_loss: 56.2802\n",
      "Epoch 330/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 44.4451 - val_loss: 56.2023\n",
      "Epoch 331/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 44.4010 - val_loss: 56.1519\n",
      "Epoch 332/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 44.3494 - val_loss: 56.1062\n",
      "Epoch 333/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 44.3021 - val_loss: 56.0168\n",
      "Epoch 334/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 44.2468 - val_loss: 55.9909\n",
      "Epoch 335/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 44.1865 - val_loss: 55.9240\n",
      "Epoch 336/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 44.1267 - val_loss: 55.9165\n",
      "Epoch 337/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 44.0782 - val_loss: 55.8732\n",
      "Epoch 338/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 44.0261 - val_loss: 55.8601\n",
      "Epoch 339/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 43.9728 - val_loss: 55.7597\n",
      "Epoch 340/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 43.9325 - val_loss: 55.6293\n",
      "Epoch 341/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 43.8657 - val_loss: 55.5401\n",
      "Epoch 342/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 43.8241 - val_loss: 55.5073\n",
      "Epoch 343/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 43.7672 - val_loss: 55.4687\n",
      "Epoch 344/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 43.7070 - val_loss: 55.4412\n",
      "Epoch 345/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 43.6593 - val_loss: 55.4008\n",
      "Epoch 346/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 43.6103 - val_loss: 55.2990\n",
      "Epoch 347/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 43.5491 - val_loss: 55.1263\n",
      "Epoch 348/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 43.5045 - val_loss: 55.0499\n",
      "Epoch 349/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 43.4321 - val_loss: 54.9490\n",
      "Epoch 350/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 43.3797 - val_loss: 54.9191\n",
      "Epoch 351/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 43.3250 - val_loss: 54.9168\n",
      "Epoch 352/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 43.2771 - val_loss: 54.8865\n",
      "Epoch 353/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 43.2283 - val_loss: 54.8179\n",
      "Epoch 354/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 43.1657 - val_loss: 54.7086\n",
      "Epoch 355/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 43.1331 - val_loss: 54.6765\n",
      "Epoch 356/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 43.0646 - val_loss: 54.5469\n",
      "Epoch 357/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 43.0108 - val_loss: 54.5646\n",
      "Epoch 358/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 42.9543 - val_loss: 54.5243\n",
      "Epoch 359/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 42.9111 - val_loss: 54.4264\n",
      "Epoch 360/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 42.8520 - val_loss: 54.3116\n",
      "Epoch 361/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 42.8083 - val_loss: 54.2146\n",
      "Epoch 362/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 42.7576 - val_loss: 54.1163\n",
      "Epoch 363/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 42.7086 - val_loss: 54.0765\n",
      "Epoch 364/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 42.6488 - val_loss: 53.9932\n",
      "Epoch 365/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 42.5894 - val_loss: 53.9431\n",
      "Epoch 366/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 42.5395 - val_loss: 53.9708\n",
      "Epoch 367/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 42.4901 - val_loss: 53.8890\n",
      "Epoch 368/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 42.4162 - val_loss: 53.7841\n",
      "Epoch 369/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 42.3700 - val_loss: 53.7697\n",
      "Epoch 370/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 42.3055 - val_loss: 53.6501\n",
      "Epoch 371/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 42.2718 - val_loss: 53.5989\n",
      "Epoch 372/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 42.2178 - val_loss: 53.5349\n",
      "Epoch 373/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 42.1536 - val_loss: 53.4803\n",
      "Epoch 374/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 42.1129 - val_loss: 53.3903\n",
      "Epoch 375/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 42.0478 - val_loss: 53.2848\n",
      "Epoch 376/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 41.9889 - val_loss: 53.2191\n",
      "Epoch 377/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 41.9275 - val_loss: 53.1592\n",
      "Epoch 378/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 41.8671 - val_loss: 53.0646\n",
      "Epoch 379/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 41.8328 - val_loss: 53.0023\n",
      "Epoch 380/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 41.7755 - val_loss: 52.9006\n",
      "Epoch 381/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 41.7104 - val_loss: 52.8632\n",
      "Epoch 382/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 41.6495 - val_loss: 52.8199\n",
      "Epoch 383/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 41.6173 - val_loss: 52.7022\n",
      "Epoch 384/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 41.5688 - val_loss: 52.6633\n",
      "Epoch 385/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 41.5022 - val_loss: 52.5815\n",
      "Epoch 386/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 41.4410 - val_loss: 52.5093\n",
      "Epoch 387/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 41.3969 - val_loss: 52.4192\n",
      "Epoch 388/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 41.3365 - val_loss: 52.3255\n",
      "Epoch 389/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 41.2887 - val_loss: 52.2821\n",
      "Epoch 390/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 41.2375 - val_loss: 52.2258\n",
      "Epoch 391/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 41.1742 - val_loss: 52.1943\n",
      "Epoch 392/2000\n",
      "1/1 [==============================] - 0s 22ms/step - loss: 41.1103 - val_loss: 52.1029\n",
      "Epoch 393/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 41.0554 - val_loss: 51.9966\n",
      "Epoch 394/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 41.0080 - val_loss: 51.8891\n",
      "Epoch 395/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 40.9386 - val_loss: 51.8647\n",
      "Epoch 396/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 40.8762 - val_loss: 51.8271\n",
      "Epoch 397/2000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 17ms/step - loss: 40.8158 - val_loss: 51.7344\n",
      "Epoch 398/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 40.7642 - val_loss: 51.6687\n",
      "Epoch 399/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 40.7281 - val_loss: 51.6247\n",
      "Epoch 400/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 40.6665 - val_loss: 51.5162\n",
      "Epoch 401/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 40.6100 - val_loss: 51.4486\n",
      "Epoch 402/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 40.5440 - val_loss: 51.4169\n",
      "Epoch 403/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 40.4803 - val_loss: 51.2645\n",
      "Epoch 404/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 40.4438 - val_loss: 51.1922\n",
      "Epoch 405/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 40.3958 - val_loss: 51.1158\n",
      "Epoch 406/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 40.3340 - val_loss: 51.0706\n",
      "Epoch 407/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 40.2751 - val_loss: 51.0883\n",
      "Epoch 408/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 40.2166 - val_loss: 50.9205\n",
      "Epoch 409/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 40.1690 - val_loss: 50.8770\n",
      "Epoch 410/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 40.0976 - val_loss: 50.7856\n",
      "Epoch 411/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 40.0361 - val_loss: 50.6827\n",
      "Epoch 412/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 40.0005 - val_loss: 50.5854\n",
      "Epoch 413/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 39.9381 - val_loss: 50.5623\n",
      "Epoch 414/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 39.8799 - val_loss: 50.5016\n",
      "Epoch 415/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 39.8354 - val_loss: 50.4289\n",
      "Epoch 416/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 39.7626 - val_loss: 50.3115\n",
      "Epoch 417/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 39.7171 - val_loss: 50.1964\n",
      "Epoch 418/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 39.6484 - val_loss: 50.0844\n",
      "Epoch 419/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 39.5942 - val_loss: 50.0120\n",
      "Epoch 420/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 39.5445 - val_loss: 50.0729\n",
      "Epoch 421/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 39.4943 - val_loss: 49.9816\n",
      "Epoch 422/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 39.4381 - val_loss: 49.8535\n",
      "Epoch 423/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 39.3773 - val_loss: 49.6964\n",
      "Epoch 424/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 39.3212 - val_loss: 49.6492\n",
      "Epoch 425/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 39.2590 - val_loss: 49.6048\n",
      "Epoch 426/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 39.1968 - val_loss: 49.6169\n",
      "Epoch 427/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 39.1492 - val_loss: 49.5958\n",
      "Epoch 428/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 39.0966 - val_loss: 49.4867\n",
      "Epoch 429/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 39.0378 - val_loss: 49.3199\n",
      "Epoch 430/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 38.9746 - val_loss: 49.1565\n",
      "Epoch 431/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 38.9244 - val_loss: 49.0852\n",
      "Epoch 432/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 38.8821 - val_loss: 49.1325\n",
      "Epoch 433/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 38.8057 - val_loss: 49.1207\n",
      "Epoch 434/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 38.7606 - val_loss: 49.0830\n",
      "Epoch 435/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 38.6951 - val_loss: 48.9511\n",
      "Epoch 436/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 38.6348 - val_loss: 48.7864\n",
      "Epoch 437/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 38.5907 - val_loss: 48.7582\n",
      "Epoch 438/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 38.5386 - val_loss: 48.6896\n",
      "Epoch 439/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 38.4703 - val_loss: 48.7016\n",
      "Epoch 440/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 38.4163 - val_loss: 48.6011\n",
      "Epoch 441/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 38.3534 - val_loss: 48.3997\n",
      "Epoch 442/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 38.2946 - val_loss: 48.2419\n",
      "Epoch 443/2000\n",
      "1/1 [==============================] - 0s 21ms/step - loss: 38.2386 - val_loss: 48.1574\n",
      "Epoch 444/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 38.1904 - val_loss: 48.1223\n",
      "Epoch 445/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 38.1301 - val_loss: 48.1426\n",
      "Epoch 446/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 38.0672 - val_loss: 48.0499\n",
      "Epoch 447/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 38.0071 - val_loss: 47.9400\n",
      "Epoch 448/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 37.9501 - val_loss: 47.9103\n",
      "Epoch 449/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 37.8907 - val_loss: 47.7656\n",
      "Epoch 450/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 37.8429 - val_loss: 47.6967\n",
      "Epoch 451/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 37.7883 - val_loss: 47.6138\n",
      "Epoch 452/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 37.7225 - val_loss: 47.5435\n",
      "Epoch 453/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 37.6558 - val_loss: 47.4495\n",
      "Epoch 454/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 37.6086 - val_loss: 47.3864\n",
      "Epoch 455/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 37.5445 - val_loss: 47.2890\n",
      "Epoch 456/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 37.5020 - val_loss: 47.2553\n",
      "Epoch 457/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 37.4290 - val_loss: 47.2512\n",
      "Epoch 458/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 37.3660 - val_loss: 47.2052\n",
      "Epoch 459/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 37.3219 - val_loss: 47.0615\n",
      "Epoch 460/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 37.2611 - val_loss: 46.9529\n",
      "Epoch 461/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 37.1971 - val_loss: 46.8511\n",
      "Epoch 462/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 37.1609 - val_loss: 46.8271\n",
      "Epoch 463/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 37.0879 - val_loss: 46.8325\n",
      "Epoch 464/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 37.0298 - val_loss: 46.7963\n",
      "Epoch 465/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 36.9753 - val_loss: 46.6259\n",
      "Epoch 466/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 36.9007 - val_loss: 46.4375\n",
      "Epoch 467/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 36.8413 - val_loss: 46.3745\n",
      "Epoch 468/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 36.7876 - val_loss: 46.3918\n",
      "Epoch 469/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 36.7253 - val_loss: 46.3929\n",
      "Epoch 470/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 36.6728 - val_loss: 46.2673\n",
      "Epoch 471/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 36.6118 - val_loss: 46.1013\n",
      "Epoch 472/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 36.5479 - val_loss: 46.0972\n",
      "Epoch 473/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 36.4935 - val_loss: 45.9450\n",
      "Epoch 474/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 36.4221 - val_loss: 45.8468\n",
      "Epoch 475/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 36.3699 - val_loss: 45.8207\n",
      "Epoch 476/2000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 17ms/step - loss: 36.3179 - val_loss: 45.7759\n",
      "Epoch 477/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 36.2589 - val_loss: 45.7234\n",
      "Epoch 478/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 36.1923 - val_loss: 45.5502\n",
      "Epoch 479/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 36.1383 - val_loss: 45.5133\n",
      "Epoch 480/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 36.0729 - val_loss: 45.4775\n",
      "Epoch 481/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 36.0122 - val_loss: 45.3998\n",
      "Epoch 482/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 35.9569 - val_loss: 45.2738\n",
      "Epoch 483/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 35.8874 - val_loss: 45.0823\n",
      "Epoch 484/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 35.8203 - val_loss: 44.9858\n",
      "Epoch 485/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 35.7804 - val_loss: 44.9734\n",
      "Epoch 486/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 35.7215 - val_loss: 44.8612\n",
      "Epoch 487/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 35.6450 - val_loss: 44.7307\n",
      "Epoch 488/2000\n",
      "1/1 [==============================] - 0s 25ms/step - loss: 35.5930 - val_loss: 44.7613\n",
      "Epoch 489/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 35.5423 - val_loss: 44.6512\n",
      "Epoch 490/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 35.4734 - val_loss: 44.5508\n",
      "Epoch 491/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 35.4124 - val_loss: 44.5520\n",
      "Epoch 492/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 35.3508 - val_loss: 44.4176\n",
      "Epoch 493/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 35.2891 - val_loss: 44.3477\n",
      "Epoch 494/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 35.2280 - val_loss: 44.2745\n",
      "Epoch 495/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 35.1753 - val_loss: 44.1534\n",
      "Epoch 496/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 35.1093 - val_loss: 44.0215\n",
      "Epoch 497/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 35.0532 - val_loss: 43.9877\n",
      "Epoch 498/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 34.9946 - val_loss: 43.9631\n",
      "Epoch 499/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 34.9225 - val_loss: 43.8620\n",
      "Epoch 500/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 34.8757 - val_loss: 43.7747\n",
      "Epoch 501/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 34.8192 - val_loss: 43.6805\n",
      "Epoch 502/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 34.7444 - val_loss: 43.5659\n",
      "Epoch 503/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 34.6751 - val_loss: 43.5046\n",
      "Epoch 504/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 34.6197 - val_loss: 43.4764\n",
      "Epoch 505/2000\n",
      "1/1 [==============================] - 0s 23ms/step - loss: 34.5707 - val_loss: 43.3807\n",
      "Epoch 506/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 34.5061 - val_loss: 43.1776\n",
      "Epoch 507/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 34.4438 - val_loss: 43.0346\n",
      "Epoch 508/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 34.3884 - val_loss: 42.7882\n",
      "Epoch 509/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 34.3004 - val_loss: 42.5459\n",
      "Epoch 510/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 34.1835 - val_loss: 42.2612\n",
      "Epoch 511/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 34.0835 - val_loss: 42.1429\n",
      "Epoch 512/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 33.9824 - val_loss: 42.0060\n",
      "Epoch 513/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 33.9236 - val_loss: 41.8643\n",
      "Epoch 514/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 33.8659 - val_loss: 41.8212\n",
      "Epoch 515/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 33.7710 - val_loss: 41.7054\n",
      "Epoch 516/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 33.6990 - val_loss: 41.6837\n",
      "Epoch 517/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 33.6012 - val_loss: 41.5511\n",
      "Epoch 518/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 33.4760 - val_loss: 41.4722\n",
      "Epoch 519/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 33.3992 - val_loss: 41.3975\n",
      "Epoch 520/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 33.3026 - val_loss: 41.2421\n",
      "Epoch 521/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 33.1938 - val_loss: 41.1331\n",
      "Epoch 522/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 33.1150 - val_loss: 41.0349\n",
      "Epoch 523/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 32.9901 - val_loss: 40.9352\n",
      "Epoch 524/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 32.8798 - val_loss: 40.8119\n",
      "Epoch 525/2000\n",
      "1/1 [==============================] - 0s 21ms/step - loss: 32.7749 - val_loss: 40.6823\n",
      "Epoch 526/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 32.6932 - val_loss: 40.5188\n",
      "Epoch 527/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 32.5889 - val_loss: 40.3526\n",
      "Epoch 528/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 32.4777 - val_loss: 40.2126\n",
      "Epoch 529/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 32.3947 - val_loss: 40.1454\n",
      "Epoch 530/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 32.3261 - val_loss: 40.0128\n",
      "Epoch 531/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 32.2354 - val_loss: 39.8617\n",
      "Epoch 532/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 32.1387 - val_loss: 39.7882\n",
      "Epoch 533/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 32.0417 - val_loss: 39.7089\n",
      "Epoch 534/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 31.9729 - val_loss: 39.6367\n",
      "Epoch 535/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 31.8817 - val_loss: 39.5023\n",
      "Epoch 536/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 31.7917 - val_loss: 39.4360\n",
      "Epoch 537/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 31.7124 - val_loss: 39.3919\n",
      "Epoch 538/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 31.6176 - val_loss: 39.3256\n",
      "Epoch 539/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 31.5215 - val_loss: 39.2532\n",
      "Epoch 540/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 31.4380 - val_loss: 39.1574\n",
      "Epoch 541/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 31.3625 - val_loss: 39.0411\n",
      "Epoch 542/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 31.2955 - val_loss: 38.9710\n",
      "Epoch 543/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 31.2166 - val_loss: 38.8501\n",
      "Epoch 544/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 31.1420 - val_loss: 38.7024\n",
      "Epoch 545/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 31.0637 - val_loss: 38.5665\n",
      "Epoch 546/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 30.9944 - val_loss: 38.4371\n",
      "Epoch 547/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 30.9158 - val_loss: 38.3437\n",
      "Epoch 548/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 30.8532 - val_loss: 38.1903\n",
      "Epoch 549/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 30.7681 - val_loss: 38.1107\n",
      "Epoch 550/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 30.6890 - val_loss: 38.0132\n",
      "Epoch 551/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 30.6097 - val_loss: 37.9571\n",
      "Epoch 552/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 30.5220 - val_loss: 37.8917\n",
      "Epoch 553/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 30.4812 - val_loss: 37.7399\n",
      "Epoch 554/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 30.3857 - val_loss: 37.6531\n",
      "Epoch 555/2000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 17ms/step - loss: 30.3119 - val_loss: 37.6051\n",
      "Epoch 556/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 30.2337 - val_loss: 37.5217\n",
      "Epoch 557/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 30.1605 - val_loss: 37.3551\n",
      "Epoch 558/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 30.0931 - val_loss: 37.2109\n",
      "Epoch 559/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 30.0159 - val_loss: 37.1936\n",
      "Epoch 560/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 29.9503 - val_loss: 37.0859\n",
      "Epoch 561/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 29.8901 - val_loss: 36.9287\n",
      "Epoch 562/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 29.7953 - val_loss: 36.8073\n",
      "Epoch 563/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 29.7436 - val_loss: 36.7760\n",
      "Epoch 564/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 29.6581 - val_loss: 36.6838\n",
      "Epoch 565/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 29.6122 - val_loss: 36.5315\n",
      "Epoch 566/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 29.5080 - val_loss: 36.3702\n",
      "Epoch 567/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 29.4429 - val_loss: 36.2534\n",
      "Epoch 568/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 29.3744 - val_loss: 36.3192\n",
      "Epoch 569/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 29.3280 - val_loss: 36.1705\n",
      "Epoch 570/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 29.2417 - val_loss: 36.0267\n",
      "Epoch 571/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 29.1702 - val_loss: 35.9346\n",
      "Epoch 572/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 29.0904 - val_loss: 35.8555\n",
      "Epoch 573/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 28.9978 - val_loss: 35.7639\n",
      "Epoch 574/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 28.9308 - val_loss: 35.6362\n",
      "Epoch 575/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 28.8545 - val_loss: 35.5263\n",
      "Epoch 576/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 28.8041 - val_loss: 35.4706\n",
      "Epoch 577/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 28.7311 - val_loss: 35.3621\n",
      "Epoch 578/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 28.6782 - val_loss: 35.2413\n",
      "Epoch 579/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 28.5964 - val_loss: 35.1386\n",
      "Epoch 580/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 28.5285 - val_loss: 35.0931\n",
      "Epoch 581/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 28.4441 - val_loss: 34.9735\n",
      "Epoch 582/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 28.3475 - val_loss: 34.8570\n",
      "Epoch 583/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 28.2843 - val_loss: 34.8501\n",
      "Epoch 584/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 28.2209 - val_loss: 34.7023\n",
      "Epoch 585/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 28.1583 - val_loss: 34.5927\n",
      "Epoch 586/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 28.0884 - val_loss: 34.5170\n",
      "Epoch 587/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 28.0006 - val_loss: 34.4469\n",
      "Epoch 588/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 27.9496 - val_loss: 34.3160\n",
      "Epoch 589/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 27.8594 - val_loss: 34.2041\n",
      "Epoch 590/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 27.8010 - val_loss: 34.1147\n",
      "Epoch 591/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 27.7042 - val_loss: 34.0988\n",
      "Epoch 592/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 27.6826 - val_loss: 33.9184\n",
      "Epoch 593/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 27.5932 - val_loss: 33.8068\n",
      "Epoch 594/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 27.5434 - val_loss: 33.7383\n",
      "Epoch 595/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 27.4724 - val_loss: 33.7041\n",
      "Epoch 596/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 27.4567 - val_loss: 33.5870\n",
      "Epoch 597/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 27.3399 - val_loss: 33.4774\n",
      "Epoch 598/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 27.2892 - val_loss: 33.4090\n",
      "Epoch 599/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 27.2191 - val_loss: 33.3656\n",
      "Epoch 600/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 27.1703 - val_loss: 33.2563\n",
      "Epoch 601/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 27.0873 - val_loss: 33.1344\n",
      "Epoch 602/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 27.0339 - val_loss: 33.0482\n",
      "Epoch 603/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 26.9599 - val_loss: 33.0349\n",
      "Epoch 604/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 26.9006 - val_loss: 33.0058\n",
      "Epoch 605/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 26.8465 - val_loss: 32.9047\n",
      "Epoch 606/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 26.7783 - val_loss: 32.8290\n",
      "Epoch 607/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 26.7215 - val_loss: 32.7557\n",
      "Epoch 608/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 26.6196 - val_loss: 32.6657\n",
      "Epoch 609/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 26.5897 - val_loss: 32.5323\n",
      "Epoch 610/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 26.5347 - val_loss: 32.4611\n",
      "Epoch 611/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 26.4593 - val_loss: 32.3668\n",
      "Epoch 612/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 26.3867 - val_loss: 32.2141\n",
      "Epoch 613/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 26.3388 - val_loss: 32.1871\n",
      "Epoch 614/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 26.2737 - val_loss: 32.1149\n",
      "Epoch 615/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 26.2211 - val_loss: 32.0209\n",
      "Epoch 616/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 26.1343 - val_loss: 31.9745\n",
      "Epoch 617/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 26.0575 - val_loss: 31.8867\n",
      "Epoch 618/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 26.0384 - val_loss: 31.8273\n",
      "Epoch 619/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 25.9605 - val_loss: 31.7841\n",
      "Epoch 620/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 25.9074 - val_loss: 31.7243\n",
      "Epoch 621/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 25.8206 - val_loss: 31.6550\n",
      "Epoch 622/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 25.7742 - val_loss: 31.5900\n",
      "Epoch 623/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 25.7033 - val_loss: 31.5359\n",
      "Epoch 624/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 25.6391 - val_loss: 31.4314\n",
      "Epoch 625/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 25.5864 - val_loss: 31.3222\n",
      "Epoch 626/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 25.5385 - val_loss: 31.2371\n",
      "Epoch 627/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 25.4814 - val_loss: 31.1635\n",
      "Epoch 628/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 25.4027 - val_loss: 31.0500\n",
      "Epoch 629/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 25.3849 - val_loss: 30.9896\n",
      "Epoch 630/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 25.3188 - val_loss: 30.9231\n",
      "Epoch 631/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 25.2679 - val_loss: 30.8147\n",
      "Epoch 632/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 25.2287 - val_loss: 30.7021\n",
      "Epoch 633/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 25.1751 - val_loss: 30.7569\n",
      "Epoch 634/2000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 16ms/step - loss: 25.0962 - val_loss: 30.7177\n",
      "Epoch 635/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 25.0453 - val_loss: 30.5519\n",
      "Epoch 636/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 24.9992 - val_loss: 30.5013\n",
      "Epoch 637/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 24.9467 - val_loss: 30.4801\n",
      "Epoch 638/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 24.9216 - val_loss: 30.4138\n",
      "Epoch 639/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 24.8676 - val_loss: 30.3510\n",
      "Epoch 640/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 24.8091 - val_loss: 30.2528\n",
      "Epoch 641/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 24.7598 - val_loss: 30.1962\n",
      "Epoch 642/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 24.6835 - val_loss: 30.0657\n",
      "Epoch 643/2000\n",
      "1/1 [==============================] - 0s 21ms/step - loss: 24.6185 - val_loss: 30.0088\n",
      "Epoch 644/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 24.6113 - val_loss: 29.9247\n",
      "Epoch 645/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 24.5478 - val_loss: 29.9174\n",
      "Epoch 646/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 24.5130 - val_loss: 29.7994\n",
      "Epoch 647/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 24.4666 - val_loss: 29.7202\n",
      "Epoch 648/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 24.4348 - val_loss: 29.6927\n",
      "Epoch 649/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 24.3543 - val_loss: 29.6095\n",
      "Epoch 650/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 24.3021 - val_loss: 29.5491\n",
      "Epoch 651/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 24.2728 - val_loss: 29.4474\n",
      "Epoch 652/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 24.2242 - val_loss: 29.4047\n",
      "Epoch 653/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 24.1655 - val_loss: 29.3570\n",
      "Epoch 654/2000\n",
      "1/1 [==============================] - 0s 21ms/step - loss: 24.1336 - val_loss: 29.2993\n",
      "Epoch 655/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 24.1042 - val_loss: 29.2288\n",
      "Epoch 656/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 24.0200 - val_loss: 29.1500\n",
      "Epoch 657/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 23.9743 - val_loss: 29.0729\n",
      "Epoch 658/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 23.9431 - val_loss: 28.9666\n",
      "Epoch 659/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 23.8941 - val_loss: 28.9538\n",
      "Epoch 660/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 23.8571 - val_loss: 28.8866\n",
      "Epoch 661/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 23.7906 - val_loss: 28.8207\n",
      "Epoch 662/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 23.7881 - val_loss: 28.7489\n",
      "Epoch 663/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 23.7249 - val_loss: 28.6953\n",
      "Epoch 664/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 23.6707 - val_loss: 28.6458\n",
      "Epoch 665/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 23.6217 - val_loss: 28.6015\n",
      "Epoch 666/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 23.5909 - val_loss: 28.5281\n",
      "Epoch 667/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 23.5624 - val_loss: 28.4684\n",
      "Epoch 668/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 23.5162 - val_loss: 28.3753\n",
      "Epoch 669/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 23.4738 - val_loss: 28.3121\n",
      "Epoch 670/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 23.4212 - val_loss: 28.3393\n",
      "Epoch 671/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 23.3744 - val_loss: 28.2094\n",
      "Epoch 672/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 23.3169 - val_loss: 28.1608\n",
      "Epoch 673/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 23.2847 - val_loss: 28.0748\n",
      "Epoch 674/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 23.2522 - val_loss: 28.0334\n",
      "Epoch 675/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 23.2081 - val_loss: 27.9725\n",
      "Epoch 676/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 23.1950 - val_loss: 27.8863\n",
      "Epoch 677/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 23.1616 - val_loss: 27.8277\n",
      "Epoch 678/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 23.1175 - val_loss: 27.7899\n",
      "Epoch 679/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 23.1230 - val_loss: 27.7475\n",
      "Epoch 680/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 23.0681 - val_loss: 27.6643\n",
      "Epoch 681/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 23.0103 - val_loss: 27.6525\n",
      "Epoch 682/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 22.9886 - val_loss: 27.5982\n",
      "Epoch 683/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 22.9327 - val_loss: 27.5675\n",
      "Epoch 684/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 22.9156 - val_loss: 27.4882\n",
      "Epoch 685/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 22.8695 - val_loss: 27.4856\n",
      "Epoch 686/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 22.8455 - val_loss: 27.3721\n",
      "Epoch 687/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 22.8058 - val_loss: 27.3151\n",
      "Epoch 688/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 22.7920 - val_loss: 27.2853\n",
      "Epoch 689/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 22.7032 - val_loss: 27.3035\n",
      "Epoch 690/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 22.7189 - val_loss: 27.2309\n",
      "Epoch 691/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 22.6653 - val_loss: 27.1900\n",
      "Epoch 692/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 22.6457 - val_loss: 27.1668\n",
      "Epoch 693/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 22.6168 - val_loss: 27.0678\n",
      "Epoch 694/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 22.5571 - val_loss: 27.0352\n",
      "Epoch 695/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 22.5300 - val_loss: 26.9224\n",
      "Epoch 696/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 22.4876 - val_loss: 26.9986\n",
      "Epoch 697/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 22.5213 - val_loss: 26.9154\n",
      "Epoch 698/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 22.4881 - val_loss: 26.8310\n",
      "Epoch 699/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 22.4023 - val_loss: 26.8179\n",
      "Epoch 700/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 22.4633 - val_loss: 26.6797\n",
      "Epoch 701/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 22.2959 - val_loss: 26.6729\n",
      "Epoch 702/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 22.3028 - val_loss: 26.5916\n",
      "Epoch 703/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 22.2144 - val_loss: 26.5257\n",
      "Epoch 704/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 22.1948 - val_loss: 26.4433\n",
      "Epoch 705/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 22.1170 - val_loss: 26.4736\n",
      "Epoch 706/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 22.1240 - val_loss: 26.3955\n",
      "Epoch 707/2000\n",
      "1/1 [==============================] - 0s 22ms/step - loss: 22.0989 - val_loss: 26.3301\n",
      "Epoch 708/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 22.0599 - val_loss: 26.2626\n",
      "Epoch 709/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 22.0222 - val_loss: 26.2451\n",
      "Epoch 710/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 21.9984 - val_loss: 26.1774\n",
      "Epoch 711/2000\n",
      "1/1 [==============================] - ETA: 0s - loss: 21.95 - 0s 17ms/step - loss: 21.9587 - val_loss: 26.1444\n",
      "Epoch 712/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 21.9632 - val_loss: 26.0753\n",
      "Epoch 713/2000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 16ms/step - loss: 21.9053 - val_loss: 26.0978\n",
      "Epoch 714/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 21.9007 - val_loss: 25.9826\n",
      "Epoch 715/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 21.8415 - val_loss: 25.9425\n",
      "Epoch 716/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 21.8274 - val_loss: 25.8758\n",
      "Epoch 717/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 21.7870 - val_loss: 25.8316\n",
      "Epoch 718/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 21.7670 - val_loss: 25.8139\n",
      "Epoch 719/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 21.7337 - val_loss: 25.7636\n",
      "Epoch 720/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 21.7236 - val_loss: 25.7089\n",
      "Epoch 721/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 21.6818 - val_loss: 25.7020\n",
      "Epoch 722/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 21.6411 - val_loss: 25.6821\n",
      "Epoch 723/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 21.6084 - val_loss: 25.6373\n",
      "Epoch 724/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 21.5772 - val_loss: 25.5538\n",
      "Epoch 725/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 21.5520 - val_loss: 25.5193\n",
      "Epoch 726/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 21.5277 - val_loss: 25.5004\n",
      "Epoch 727/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 21.5421 - val_loss: 25.3954\n",
      "Epoch 728/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 21.5240 - val_loss: 25.3851\n",
      "Epoch 729/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 21.4840 - val_loss: 25.3343\n",
      "Epoch 730/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 21.4428 - val_loss: 25.3100\n",
      "Epoch 731/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 21.4188 - val_loss: 25.2338\n",
      "Epoch 732/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 21.3673 - val_loss: 25.2067\n",
      "Epoch 733/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 21.3120 - val_loss: 25.1529\n",
      "Epoch 734/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 21.2822 - val_loss: 25.1414\n",
      "Epoch 735/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 21.2803 - val_loss: 25.0762\n",
      "Epoch 736/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 21.2263 - val_loss: 25.0917\n",
      "Epoch 737/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 21.2288 - val_loss: 25.0523\n",
      "Epoch 738/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 21.2380 - val_loss: 25.0095\n",
      "Epoch 739/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 21.1654 - val_loss: 24.9544\n",
      "Epoch 740/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 21.1641 - val_loss: 24.9203\n",
      "Epoch 741/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 21.1364 - val_loss: 24.8897\n",
      "Epoch 742/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 21.1227 - val_loss: 24.8537\n",
      "Epoch 743/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 21.0954 - val_loss: 24.8482\n",
      "Epoch 744/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 21.0608 - val_loss: 24.7666\n",
      "Epoch 745/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 21.0240 - val_loss: 24.7492\n",
      "Epoch 746/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 21.0081 - val_loss: 24.7400\n",
      "Epoch 747/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 20.9749 - val_loss: 24.7066\n",
      "Epoch 748/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 20.9462 - val_loss: 24.6774\n",
      "Epoch 749/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 20.9240 - val_loss: 24.6575\n",
      "Epoch 750/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 20.9006 - val_loss: 24.5878\n",
      "Epoch 751/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 20.8751 - val_loss: 24.5486\n",
      "Epoch 752/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 20.8505 - val_loss: 24.4705\n",
      "Epoch 753/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 20.8321 - val_loss: 24.4478\n",
      "Epoch 754/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 20.8074 - val_loss: 24.4557\n",
      "Epoch 755/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 20.7975 - val_loss: 24.4153\n",
      "Epoch 756/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 20.7589 - val_loss: 24.3779\n",
      "Epoch 757/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 20.7439 - val_loss: 24.3600\n",
      "Epoch 758/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 20.7182 - val_loss: 24.2641\n",
      "Epoch 759/2000\n",
      "1/1 [==============================] - 0s 22ms/step - loss: 20.6829 - val_loss: 24.2055\n",
      "Epoch 760/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 20.6591 - val_loss: 24.1658\n",
      "Epoch 761/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 20.6365 - val_loss: 24.1403\n",
      "Epoch 762/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 20.6351 - val_loss: 24.1068\n",
      "Epoch 763/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 20.5892 - val_loss: 24.1478\n",
      "Epoch 764/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 20.5930 - val_loss: 24.1070\n",
      "Epoch 765/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 20.5673 - val_loss: 24.0902\n",
      "Epoch 766/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 20.5159 - val_loss: 24.0454\n",
      "Epoch 767/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 20.5185 - val_loss: 24.0174\n",
      "Epoch 768/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 20.4785 - val_loss: 24.0239\n",
      "Epoch 769/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 20.4514 - val_loss: 24.0320\n",
      "Epoch 770/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 20.4263 - val_loss: 23.9546\n",
      "Epoch 771/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 20.4098 - val_loss: 23.9049\n",
      "Epoch 772/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 20.3975 - val_loss: 23.8704\n",
      "Epoch 773/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 20.3734 - val_loss: 23.8473\n",
      "Epoch 774/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 20.3694 - val_loss: 23.7889\n",
      "Epoch 775/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 20.3519 - val_loss: 23.7600\n",
      "Epoch 776/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 20.3158 - val_loss: 23.7404\n",
      "Epoch 777/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 20.2961 - val_loss: 23.7196\n",
      "Epoch 778/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 20.2579 - val_loss: 23.6970\n",
      "Epoch 779/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 20.2518 - val_loss: 23.6315\n",
      "Epoch 780/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 20.2286 - val_loss: 23.6765\n",
      "Epoch 781/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 20.2210 - val_loss: 23.6022\n",
      "Epoch 782/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 20.1939 - val_loss: 23.5457\n",
      "Epoch 783/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 20.1632 - val_loss: 23.5314\n",
      "Epoch 784/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 20.1712 - val_loss: 23.4969\n",
      "Epoch 785/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 20.1516 - val_loss: 23.5120\n",
      "Epoch 786/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 20.1312 - val_loss: 23.5190\n",
      "Epoch 787/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 20.1205 - val_loss: 23.4437\n",
      "Epoch 788/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 20.0795 - val_loss: 23.4332\n",
      "Epoch 789/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 20.0620 - val_loss: 23.4284\n",
      "Epoch 790/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 20.0475 - val_loss: 23.3813\n",
      "Epoch 791/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 20.0360 - val_loss: 23.3509\n",
      "Epoch 792/2000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 17ms/step - loss: 20.0137 - val_loss: 23.3025\n",
      "Epoch 793/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.9948 - val_loss: 23.2876\n",
      "Epoch 794/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 19.9786 - val_loss: 23.2686\n",
      "Epoch 795/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 19.9595 - val_loss: 23.2498\n",
      "Epoch 796/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.9527 - val_loss: 23.2345\n",
      "Epoch 797/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 19.9479 - val_loss: 23.1917\n",
      "Epoch 798/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 19.9005 - val_loss: 23.2401\n",
      "Epoch 799/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.9178 - val_loss: 23.1528\n",
      "Epoch 800/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 19.8862 - val_loss: 23.1327\n",
      "Epoch 801/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 19.8941 - val_loss: 23.1377\n",
      "Epoch 802/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.8572 - val_loss: 23.0925\n",
      "Epoch 803/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 19.8154 - val_loss: 23.1317\n",
      "Epoch 804/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.9221 - val_loss: 23.0038\n",
      "Epoch 805/2000\n",
      "1/1 [==============================] - 0s 22ms/step - loss: 19.7997 - val_loss: 23.1486\n",
      "Epoch 806/2000\n",
      "1/1 [==============================] - 0s 22ms/step - loss: 19.9314 - val_loss: 22.9811\n",
      "Epoch 807/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 19.7796 - val_loss: 23.0219\n",
      "Epoch 808/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.8644 - val_loss: 22.9168\n",
      "Epoch 809/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 19.7140 - val_loss: 22.9348\n",
      "Epoch 810/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.7679 - val_loss: 22.8371\n",
      "Epoch 811/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 19.6849 - val_loss: 22.8585\n",
      "Epoch 812/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 19.7123 - val_loss: 22.7844\n",
      "Epoch 813/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 19.6574 - val_loss: 22.8406\n",
      "Epoch 814/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.7155 - val_loss: 22.7848\n",
      "Epoch 815/2000\n",
      "1/1 [==============================] - 0s 21ms/step - loss: 19.6043 - val_loss: 22.8117\n",
      "Epoch 816/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.6645 - val_loss: 22.7741\n",
      "Epoch 817/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 19.5982 - val_loss: 22.7877\n",
      "Epoch 818/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.6059 - val_loss: 22.7736\n",
      "Epoch 819/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 19.5956 - val_loss: 22.7765\n",
      "Epoch 820/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.5978 - val_loss: 22.7754\n",
      "Epoch 821/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.6081 - val_loss: 22.6908\n",
      "Epoch 822/2000\n",
      "1/1 [==============================] - 0s 22ms/step - loss: 19.5515 - val_loss: 22.7259\n",
      "Epoch 823/2000\n",
      "1/1 [==============================] - 0s 21ms/step - loss: 19.5753 - val_loss: 22.6574\n",
      "Epoch 824/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 19.5030 - val_loss: 22.6453\n",
      "Epoch 825/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.5266 - val_loss: 22.5887\n",
      "Epoch 826/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 19.4884 - val_loss: 22.4977\n",
      "Epoch 827/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 19.5161 - val_loss: 22.6085\n",
      "Epoch 828/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 19.5137 - val_loss: 22.5925\n",
      "Epoch 829/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.4983 - val_loss: 22.4999\n",
      "Epoch 830/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 19.5522 - val_loss: 22.4571\n",
      "Epoch 831/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 19.4348 - val_loss: 22.6553\n",
      "Epoch 832/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 19.5166 - val_loss: 22.4459\n",
      "Epoch 833/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.4149 - val_loss: 22.4359\n",
      "Epoch 834/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 19.4080 - val_loss: 22.5042\n",
      "Epoch 835/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.3950 - val_loss: 22.4519\n",
      "Epoch 836/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.3690 - val_loss: 22.4562\n",
      "Epoch 837/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.3505 - val_loss: 22.4034\n",
      "Epoch 838/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 19.3481 - val_loss: 22.4109\n",
      "Epoch 839/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.3208 - val_loss: 22.3591\n",
      "Epoch 840/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.3656 - val_loss: 22.3543\n",
      "Epoch 841/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.3248 - val_loss: 22.2951\n",
      "Epoch 842/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.3074 - val_loss: 22.2889\n",
      "Epoch 843/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 19.3200 - val_loss: 22.2561\n",
      "Epoch 844/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 19.2485 - val_loss: 22.2906\n",
      "Epoch 845/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.2570 - val_loss: 22.2973\n",
      "Epoch 846/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.2524 - val_loss: 22.3428\n",
      "Epoch 847/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.2607 - val_loss: 22.2437\n",
      "Epoch 848/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 19.2327 - val_loss: 22.2891\n",
      "Epoch 849/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 19.2356 - val_loss: 22.2735\n",
      "Epoch 850/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 19.2082 - val_loss: 22.2985\n",
      "Epoch 851/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.2234 - val_loss: 22.2400\n",
      "Epoch 852/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.1698 - val_loss: 22.2148\n",
      "Epoch 853/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.2155 - val_loss: 22.2474\n",
      "Epoch 854/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 19.1728 - val_loss: 22.1794\n",
      "Epoch 855/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.1400 - val_loss: 22.1421\n",
      "Epoch 856/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.1913 - val_loss: 22.1454\n",
      "Epoch 857/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.1289 - val_loss: 22.1067\n",
      "Epoch 858/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 19.1418 - val_loss: 22.0527\n",
      "Epoch 859/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 19.1225 - val_loss: 22.0558\n",
      "Epoch 860/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.1293 - val_loss: 22.0606\n",
      "Epoch 861/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 19.1041 - val_loss: 22.1247\n",
      "Epoch 862/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 19.0671 - val_loss: 22.0538\n",
      "Epoch 863/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.0507 - val_loss: 22.0329\n",
      "Epoch 864/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 19.1324 - val_loss: 22.0818\n",
      "Epoch 865/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.0536 - val_loss: 22.0251\n",
      "Epoch 866/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.0224 - val_loss: 22.0220\n",
      "Epoch 867/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 19.0903 - val_loss: 21.9661\n",
      "Epoch 868/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 19.0349 - val_loss: 22.1198\n",
      "Epoch 869/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 19.0908 - val_loss: 21.9232\n",
      "Epoch 870/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.9968 - val_loss: 21.9778\n",
      "Epoch 871/2000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 18ms/step - loss: 19.0944 - val_loss: 21.9779\n",
      "Epoch 872/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 18.9936 - val_loss: 22.0071\n",
      "Epoch 873/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.9870 - val_loss: 21.9095\n",
      "Epoch 874/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 18.9954 - val_loss: 21.9385\n",
      "Epoch 875/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.9845 - val_loss: 21.9296\n",
      "Epoch 876/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.9178 - val_loss: 21.9943\n",
      "Epoch 877/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.9735 - val_loss: 21.8652\n",
      "Epoch 878/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.8965 - val_loss: 21.8307\n",
      "Epoch 879/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 18.9366 - val_loss: 21.8052\n",
      "Epoch 880/2000\n",
      "1/1 [==============================] - 0s 22ms/step - loss: 18.8625 - val_loss: 21.9612\n",
      "Epoch 881/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.9595 - val_loss: 21.7815\n",
      "Epoch 882/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 18.8481 - val_loss: 21.8100\n",
      "Epoch 883/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.9319 - val_loss: 21.7923\n",
      "Epoch 884/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.8306 - val_loss: 21.9139\n",
      "Epoch 885/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.8512 - val_loss: 21.7812\n",
      "Epoch 886/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.8454 - val_loss: 21.7935\n",
      "Epoch 887/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.8285 - val_loss: 21.7810\n",
      "Epoch 888/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 18.7647 - val_loss: 21.8170\n",
      "Epoch 889/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.8184 - val_loss: 21.7464\n",
      "Epoch 890/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.7629 - val_loss: 21.6962\n",
      "Epoch 891/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.8116 - val_loss: 21.6773\n",
      "Epoch 892/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.7784 - val_loss: 21.6807\n",
      "Epoch 893/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.7828 - val_loss: 21.6198\n",
      "Epoch 894/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 18.7468 - val_loss: 21.6565\n",
      "Epoch 895/2000\n",
      "1/1 [==============================] - 0s 21ms/step - loss: 18.7261 - val_loss: 21.6580\n",
      "Epoch 896/2000\n",
      "1/1 [==============================] - 0s 21ms/step - loss: 18.7257 - val_loss: 21.6318\n",
      "Epoch 897/2000\n",
      "1/1 [==============================] - 0s 22ms/step - loss: 18.7015 - val_loss: 21.6729\n",
      "Epoch 898/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.6849 - val_loss: 21.6860\n",
      "Epoch 899/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.6864 - val_loss: 21.6704\n",
      "Epoch 900/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.6719 - val_loss: 21.6455\n",
      "Epoch 901/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.6606 - val_loss: 21.6185\n",
      "Epoch 902/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 18.6399 - val_loss: 21.6872\n",
      "Epoch 903/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.6260 - val_loss: 21.5780\n",
      "Epoch 904/2000\n",
      "1/1 [==============================] - 0s 21ms/step - loss: 18.5826 - val_loss: 21.4848\n",
      "Epoch 905/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.6194 - val_loss: 21.5304\n",
      "Epoch 906/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.6054 - val_loss: 21.5316\n",
      "Epoch 907/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.6001 - val_loss: 21.5135\n",
      "Epoch 908/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 18.5871 - val_loss: 21.4666\n",
      "Epoch 909/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.5459 - val_loss: 21.4739\n",
      "Epoch 910/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.5607 - val_loss: 21.4705\n",
      "Epoch 911/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 18.5571 - val_loss: 21.4979\n",
      "Epoch 912/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.5441 - val_loss: 21.4806\n",
      "Epoch 913/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.5166 - val_loss: 21.4596\n",
      "Epoch 914/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 18.5491 - val_loss: 21.4431\n",
      "Epoch 915/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.5021 - val_loss: 21.4313\n",
      "Epoch 916/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 18.4977 - val_loss: 21.4185\n",
      "Epoch 917/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.4728 - val_loss: 21.4365\n",
      "Epoch 918/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.4552 - val_loss: 21.3936\n",
      "Epoch 919/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.4359 - val_loss: 21.3704\n",
      "Epoch 920/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.4328 - val_loss: 21.3322\n",
      "Epoch 921/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.4119 - val_loss: 21.3279\n",
      "Epoch 922/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.3726 - val_loss: 21.2666\n",
      "Epoch 923/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.3994 - val_loss: 21.2991\n",
      "Epoch 924/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.3506 - val_loss: 21.2710\n",
      "Epoch 925/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 18.3879 - val_loss: 21.2336\n",
      "Epoch 926/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 18.3559 - val_loss: 21.2354\n",
      "Epoch 927/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.3757 - val_loss: 21.2255\n",
      "Epoch 928/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.3319 - val_loss: 21.3049\n",
      "Epoch 929/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.3885 - val_loss: 21.1404\n",
      "Epoch 930/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.2892 - val_loss: 21.1062\n",
      "Epoch 931/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 18.2841 - val_loss: 21.2338\n",
      "Epoch 932/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 18.2476 - val_loss: 21.2391\n",
      "Epoch 933/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.2374 - val_loss: 21.1531\n",
      "Epoch 934/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.2493 - val_loss: 21.1332\n",
      "Epoch 935/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.2575 - val_loss: 21.1745\n",
      "Epoch 936/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.2207 - val_loss: 21.1457\n",
      "Epoch 937/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.2069 - val_loss: 21.1507\n",
      "Epoch 938/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.1971 - val_loss: 21.1064\n",
      "Epoch 939/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 18.1629 - val_loss: 21.1917\n",
      "Epoch 940/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 18.1971 - val_loss: 21.0030\n",
      "Epoch 941/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.1627 - val_loss: 21.0146\n",
      "Epoch 942/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.1794 - val_loss: 21.0992\n",
      "Epoch 943/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 18.1143 - val_loss: 21.0854\n",
      "Epoch 944/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.1031 - val_loss: 21.0721\n",
      "Epoch 945/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.1787 - val_loss: 20.9614\n",
      "Epoch 946/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.1027 - val_loss: 21.0877\n",
      "Epoch 947/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.1342 - val_loss: 20.9622\n",
      "Epoch 948/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 18.0943 - val_loss: 20.9712\n",
      "Epoch 949/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.0953 - val_loss: 20.9546\n",
      "Epoch 950/2000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 19ms/step - loss: 18.0401 - val_loss: 21.1160\n",
      "Epoch 951/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.1383 - val_loss: 20.9684\n",
      "Epoch 952/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.0364 - val_loss: 20.9200\n",
      "Epoch 953/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 18.0453 - val_loss: 20.9727\n",
      "Epoch 954/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 18.0139 - val_loss: 21.0361\n",
      "Epoch 955/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 18.0140 - val_loss: 20.8801\n",
      "Epoch 956/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.9828 - val_loss: 20.8607\n",
      "Epoch 957/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 17.9860 - val_loss: 20.9281\n",
      "Epoch 958/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.9552 - val_loss: 20.9134\n",
      "Epoch 959/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 17.9444 - val_loss: 20.8690\n",
      "Epoch 960/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.9436 - val_loss: 20.8531\n",
      "Epoch 961/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 17.9473 - val_loss: 20.9105\n",
      "Epoch 962/2000\n",
      "1/1 [==============================] - 0s 22ms/step - loss: 17.9289 - val_loss: 20.8346\n",
      "Epoch 963/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.8950 - val_loss: 20.8047\n",
      "Epoch 964/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.8999 - val_loss: 20.8148\n",
      "Epoch 965/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.8820 - val_loss: 20.8162\n",
      "Epoch 966/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 17.8440 - val_loss: 20.8141\n",
      "Epoch 967/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.8509 - val_loss: 20.7971\n",
      "Epoch 968/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.8302 - val_loss: 20.8341\n",
      "Epoch 969/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 17.8440 - val_loss: 20.7602\n",
      "Epoch 970/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 17.8210 - val_loss: 20.7532\n",
      "Epoch 971/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.8138 - val_loss: 20.7093\n",
      "Epoch 972/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.8277 - val_loss: 20.7323\n",
      "Epoch 973/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.7879 - val_loss: 20.7489\n",
      "Epoch 974/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 17.7851 - val_loss: 20.6552\n",
      "Epoch 975/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.7750 - val_loss: 20.6781\n",
      "Epoch 976/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.7847 - val_loss: 20.6381\n",
      "Epoch 977/2000\n",
      "1/1 [==============================] - 0s 25ms/step - loss: 17.7751 - val_loss: 20.6329\n",
      "Epoch 978/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 17.7706 - val_loss: 20.6623\n",
      "Epoch 979/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.7521 - val_loss: 20.6627\n",
      "Epoch 980/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 17.7243 - val_loss: 20.6668\n",
      "Epoch 981/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.7223 - val_loss: 20.6278\n",
      "Epoch 982/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 17.6834 - val_loss: 20.6548\n",
      "Epoch 983/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.6791 - val_loss: 20.6075\n",
      "Epoch 984/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 17.6784 - val_loss: 20.6187\n",
      "Epoch 985/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 17.6715 - val_loss: 20.6192\n",
      "Epoch 986/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.6697 - val_loss: 20.6447\n",
      "Epoch 987/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.6515 - val_loss: 20.6047\n",
      "Epoch 988/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.6255 - val_loss: 20.6004\n",
      "Epoch 989/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.6003 - val_loss: 20.5476\n",
      "Epoch 990/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.6051 - val_loss: 20.5581\n",
      "Epoch 991/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.5862 - val_loss: 20.5336\n",
      "Epoch 992/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 17.6062 - val_loss: 20.5704\n",
      "Epoch 993/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.5990 - val_loss: 20.5512\n",
      "Epoch 994/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.5750 - val_loss: 20.4825\n",
      "Epoch 995/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.5833 - val_loss: 20.4950\n",
      "Epoch 996/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.5491 - val_loss: 20.5538\n",
      "Epoch 997/2000\n",
      "1/1 [==============================] - 0s 24ms/step - loss: 17.5699 - val_loss: 20.5115\n",
      "Epoch 998/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 17.5475 - val_loss: 20.4928\n",
      "Epoch 999/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.5610 - val_loss: 20.5075\n",
      "Epoch 1000/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.4901 - val_loss: 20.5354\n",
      "Epoch 1001/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.4966 - val_loss: 20.5045\n",
      "Epoch 1002/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.4838 - val_loss: 20.4955\n",
      "Epoch 1003/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 17.4827 - val_loss: 20.4687\n",
      "Epoch 1004/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.4537 - val_loss: 20.5285\n",
      "Epoch 1005/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.4710 - val_loss: 20.4191\n",
      "Epoch 1006/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.4548 - val_loss: 20.4441\n",
      "Epoch 1007/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.4506 - val_loss: 20.4743\n",
      "Epoch 1008/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.4228 - val_loss: 20.4685\n",
      "Epoch 1009/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.4190 - val_loss: 20.4251\n",
      "Epoch 1010/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.3829 - val_loss: 20.4302\n",
      "Epoch 1011/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.3623 - val_loss: 20.4603\n",
      "Epoch 1012/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.3624 - val_loss: 20.4400\n",
      "Epoch 1013/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.3491 - val_loss: 20.4392\n",
      "Epoch 1014/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.3122 - val_loss: 20.4295\n",
      "Epoch 1015/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.3331 - val_loss: 20.4314\n",
      "Epoch 1016/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 17.3008 - val_loss: 20.3895\n",
      "Epoch 1017/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.2790 - val_loss: 20.3910\n",
      "Epoch 1018/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.3134 - val_loss: 20.4030\n",
      "Epoch 1019/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 17.3070 - val_loss: 20.3754\n",
      "Epoch 1020/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 17.2704 - val_loss: 20.3580\n",
      "Epoch 1021/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.2739 - val_loss: 20.3559\n",
      "Epoch 1022/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 17.2601 - val_loss: 20.3934\n",
      "Epoch 1023/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.2998 - val_loss: 20.3156\n",
      "Epoch 1024/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 17.2251 - val_loss: 20.3419\n",
      "Epoch 1025/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.2090 - val_loss: 20.3632\n",
      "Epoch 1026/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.1842 - val_loss: 20.2903\n",
      "Epoch 1027/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.1844 - val_loss: 20.2510\n",
      "Epoch 1028/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 17.1782 - val_loss: 20.2980\n",
      "Epoch 1029/2000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 16ms/step - loss: 17.1656 - val_loss: 20.2745\n",
      "Epoch 1030/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 17.1508 - val_loss: 20.2632\n",
      "Epoch 1031/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.1331 - val_loss: 20.2575\n",
      "Epoch 1032/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 17.1082 - val_loss: 20.2149\n",
      "Epoch 1033/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.0898 - val_loss: 20.3045\n",
      "Epoch 1034/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 17.0994 - val_loss: 20.3082\n",
      "Epoch 1035/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.0960 - val_loss: 20.2436\n",
      "Epoch 1036/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.0605 - val_loss: 20.2231\n",
      "Epoch 1037/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.0436 - val_loss: 20.1651\n",
      "Epoch 1038/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 17.0172 - val_loss: 20.1235\n",
      "Epoch 1039/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 17.0017 - val_loss: 20.0852\n",
      "Epoch 1040/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.9979 - val_loss: 20.0582\n",
      "Epoch 1041/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 16.9774 - val_loss: 20.0559\n",
      "Epoch 1042/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.9599 - val_loss: 20.0331\n",
      "Epoch 1043/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 16.9458 - val_loss: 20.0414\n",
      "Epoch 1044/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.9254 - val_loss: 19.9915\n",
      "Epoch 1045/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.9186 - val_loss: 20.0454\n",
      "Epoch 1046/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.8994 - val_loss: 20.0397\n",
      "Epoch 1047/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.9091 - val_loss: 20.1008\n",
      "Epoch 1048/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.8345 - val_loss: 20.0842\n",
      "Epoch 1049/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.8469 - val_loss: 20.0163\n",
      "Epoch 1050/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 16.8611 - val_loss: 19.9760\n",
      "Epoch 1051/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 16.8343 - val_loss: 20.0311\n",
      "Epoch 1052/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 16.8235 - val_loss: 19.8962\n",
      "Epoch 1053/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 16.7827 - val_loss: 19.9186\n",
      "Epoch 1054/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 16.8012 - val_loss: 19.9074\n",
      "Epoch 1055/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 16.7630 - val_loss: 19.9514\n",
      "Epoch 1056/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.7658 - val_loss: 19.9922\n",
      "Epoch 1057/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.7314 - val_loss: 19.9321\n",
      "Epoch 1058/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 16.6926 - val_loss: 19.9957\n",
      "Epoch 1059/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 16.6988 - val_loss: 19.9119\n",
      "Epoch 1060/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 16.6476 - val_loss: 19.8997\n",
      "Epoch 1061/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 16.6589 - val_loss: 19.8599\n",
      "Epoch 1062/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 16.6544 - val_loss: 19.8731\n",
      "Epoch 1063/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 16.6363 - val_loss: 19.8285\n",
      "Epoch 1064/2000\n",
      "1/1 [==============================] - 0s 21ms/step - loss: 16.6189 - val_loss: 19.8203\n",
      "Epoch 1065/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.6271 - val_loss: 19.8966\n",
      "Epoch 1066/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.6092 - val_loss: 19.8935\n",
      "Epoch 1067/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.5880 - val_loss: 19.8772\n",
      "Epoch 1068/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 16.5996 - val_loss: 19.8907\n",
      "Epoch 1069/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 16.5899 - val_loss: 19.9061\n",
      "Epoch 1070/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 16.5757 - val_loss: 19.7726\n",
      "Epoch 1071/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 16.5549 - val_loss: 19.7494\n",
      "Epoch 1072/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 16.5161 - val_loss: 19.7387\n",
      "Epoch 1073/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.5169 - val_loss: 19.7176\n",
      "Epoch 1074/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.5193 - val_loss: 19.7206\n",
      "Epoch 1075/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 16.4684 - val_loss: 19.7036\n",
      "Epoch 1076/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.4558 - val_loss: 19.7543\n",
      "Epoch 1077/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 16.4889 - val_loss: 19.7872\n",
      "Epoch 1078/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 16.4609 - val_loss: 19.7914\n",
      "Epoch 1079/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 16.4443 - val_loss: 19.7962\n",
      "Epoch 1080/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 16.4481 - val_loss: 19.7969\n",
      "Epoch 1081/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.4049 - val_loss: 19.7699\n",
      "Epoch 1082/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 16.4041 - val_loss: 19.6583\n",
      "Epoch 1083/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.4099 - val_loss: 19.6764\n",
      "Epoch 1084/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 16.3883 - val_loss: 19.6909\n",
      "Epoch 1085/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 16.3901 - val_loss: 19.7183\n",
      "Epoch 1086/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 16.3763 - val_loss: 19.6665\n",
      "Epoch 1087/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.3499 - val_loss: 19.6867\n",
      "Epoch 1088/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 16.3435 - val_loss: 19.7056\n",
      "Epoch 1089/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 16.3178 - val_loss: 19.7387\n",
      "Epoch 1090/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 16.3075 - val_loss: 19.7517\n",
      "Epoch 1091/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.3053 - val_loss: 19.7395\n",
      "Epoch 1092/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 16.2942 - val_loss: 19.7260\n",
      "Epoch 1093/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 16.2566 - val_loss: 19.6674\n",
      "Epoch 1094/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.2568 - val_loss: 19.6840\n",
      "Epoch 1095/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 16.2273 - val_loss: 19.7639\n",
      "Epoch 1096/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 16.2355 - val_loss: 19.7523\n",
      "Epoch 1097/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.2357 - val_loss: 19.7881\n",
      "Epoch 1098/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.1899 - val_loss: 19.7756\n",
      "Epoch 1099/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 16.1836 - val_loss: 19.7577\n",
      "Epoch 1100/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.1868 - val_loss: 19.8533\n",
      "Epoch 1101/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 16.2078 - val_loss: 19.7438\n",
      "Epoch 1102/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 16.1878 - val_loss: 19.7565\n",
      "Epoch 1103/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.1535 - val_loss: 19.7967\n",
      "Epoch 1104/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 16.1553 - val_loss: 19.7264\n",
      "Epoch 1105/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 16.1165 - val_loss: 19.7555\n",
      "Epoch 1106/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 16.1312 - val_loss: 19.7644\n",
      "Epoch 1107/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.1200 - val_loss: 19.8173\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1108/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.1555 - val_loss: 19.7208\n",
      "Epoch 1109/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 16.0466 - val_loss: 19.7217\n",
      "Epoch 1110/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.0897 - val_loss: 19.7095\n",
      "Epoch 1111/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.0523 - val_loss: 19.7459\n",
      "Epoch 1112/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 16.0691 - val_loss: 19.6988\n",
      "Epoch 1113/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 16.0072 - val_loss: 19.6715\n",
      "Epoch 1114/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.9935 - val_loss: 19.6775\n",
      "Epoch 1115/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 16.0340 - val_loss: 19.6585\n",
      "Epoch 1116/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.9957 - val_loss: 19.6569\n",
      "Epoch 1117/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 16.0174 - val_loss: 19.6699\n",
      "Epoch 1118/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.9598 - val_loss: 19.6295\n",
      "Epoch 1119/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.9601 - val_loss: 19.5577\n",
      "Epoch 1120/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.9390 - val_loss: 19.6105\n",
      "Epoch 1121/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.9380 - val_loss: 19.5743\n",
      "Epoch 1122/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.8871 - val_loss: 19.6104\n",
      "Epoch 1123/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.9006 - val_loss: 19.5509\n",
      "Epoch 1124/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.9066 - val_loss: 19.5635\n",
      "Epoch 1125/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.8407 - val_loss: 19.6649\n",
      "Epoch 1126/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.9542 - val_loss: 19.6000\n",
      "Epoch 1127/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.8404 - val_loss: 19.6070\n",
      "Epoch 1128/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.8771 - val_loss: 19.5316\n",
      "Epoch 1129/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.8501 - val_loss: 19.5830\n",
      "Epoch 1130/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.8724 - val_loss: 19.5383\n",
      "Epoch 1131/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.8687 - val_loss: 19.5610\n",
      "Epoch 1132/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.8117 - val_loss: 19.5602\n",
      "Epoch 1133/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.8450 - val_loss: 19.5816\n",
      "Epoch 1134/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.7731 - val_loss: 19.5770\n",
      "Epoch 1135/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.7550 - val_loss: 19.5367\n",
      "Epoch 1136/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.7501 - val_loss: 19.4789\n",
      "Epoch 1137/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.7283 - val_loss: 19.4813\n",
      "Epoch 1138/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.7240 - val_loss: 19.4438\n",
      "Epoch 1139/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.7119 - val_loss: 19.4563\n",
      "Epoch 1140/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.7021 - val_loss: 19.4994\n",
      "Epoch 1141/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.7127 - val_loss: 19.4956\n",
      "Epoch 1142/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.6852 - val_loss: 19.4288\n",
      "Epoch 1143/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.6745 - val_loss: 19.4522\n",
      "Epoch 1144/2000\n",
      "1/1 [==============================] - 0s 22ms/step - loss: 15.6994 - val_loss: 19.4445\n",
      "Epoch 1145/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 15.6713 - val_loss: 19.4806\n",
      "Epoch 1146/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 15.6577 - val_loss: 19.4621\n",
      "Epoch 1147/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 15.6598 - val_loss: 19.4701\n",
      "Epoch 1148/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.6211 - val_loss: 19.4130\n",
      "Epoch 1149/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 15.6222 - val_loss: 19.3837\n",
      "Epoch 1150/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.6107 - val_loss: 19.4007\n",
      "Epoch 1151/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.6302 - val_loss: 19.4081\n",
      "Epoch 1152/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.6247 - val_loss: 19.3291\n",
      "Epoch 1153/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.5835 - val_loss: 19.3522\n",
      "Epoch 1154/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.5793 - val_loss: 19.3820\n",
      "Epoch 1155/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.5970 - val_loss: 19.4499\n",
      "Epoch 1156/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.6534 - val_loss: 19.3968\n",
      "Epoch 1157/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.5605 - val_loss: 19.3832\n",
      "Epoch 1158/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.5923 - val_loss: 19.3656\n",
      "Epoch 1159/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.5316 - val_loss: 19.3437\n",
      "Epoch 1160/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.5254 - val_loss: 19.3325\n",
      "Epoch 1161/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.5337 - val_loss: 19.3389\n",
      "Epoch 1162/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.5382 - val_loss: 19.3398\n",
      "Epoch 1163/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.5226 - val_loss: 19.3251\n",
      "Epoch 1164/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.5109 - val_loss: 19.3736\n",
      "Epoch 1165/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.4867 - val_loss: 19.3215\n",
      "Epoch 1166/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.4683 - val_loss: 19.3267\n",
      "Epoch 1167/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.4511 - val_loss: 19.3164\n",
      "Epoch 1168/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.4828 - val_loss: 19.2356\n",
      "Epoch 1169/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.4551 - val_loss: 19.2412\n",
      "Epoch 1170/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.4284 - val_loss: 19.2726\n",
      "Epoch 1171/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.4690 - val_loss: 19.2095\n",
      "Epoch 1172/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.3750 - val_loss: 19.2001\n",
      "Epoch 1173/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 15.4484 - val_loss: 19.2181\n",
      "Epoch 1174/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 15.3980 - val_loss: 19.2180\n",
      "Epoch 1175/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.3662 - val_loss: 19.1974\n",
      "Epoch 1176/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.3685 - val_loss: 19.1940\n",
      "Epoch 1177/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.3369 - val_loss: 19.2344\n",
      "Epoch 1178/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.3235 - val_loss: 19.1987\n",
      "Epoch 1179/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.3447 - val_loss: 19.1924\n",
      "Epoch 1180/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 15.3393 - val_loss: 19.1606\n",
      "Epoch 1181/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 15.3108 - val_loss: 19.1197\n",
      "Epoch 1182/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.3183 - val_loss: 19.1023\n",
      "Epoch 1183/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.3152 - val_loss: 19.0939\n",
      "Epoch 1184/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.3078 - val_loss: 19.0973\n",
      "Epoch 1185/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.3179 - val_loss: 19.0525\n",
      "Epoch 1186/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.2823 - val_loss: 19.0886\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1187/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.2578 - val_loss: 19.0998\n",
      "Epoch 1188/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.2751 - val_loss: 19.0879\n",
      "Epoch 1189/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.2703 - val_loss: 19.1071\n",
      "Epoch 1190/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 15.2737 - val_loss: 19.1133\n",
      "Epoch 1191/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.2640 - val_loss: 19.1008\n",
      "Epoch 1192/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.2414 - val_loss: 19.0581\n",
      "Epoch 1193/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.2573 - val_loss: 19.0763\n",
      "Epoch 1194/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.2609 - val_loss: 19.0380\n",
      "Epoch 1195/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.2779 - val_loss: 18.9895\n",
      "Epoch 1196/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 15.2459 - val_loss: 19.0719\n",
      "Epoch 1197/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.3109 - val_loss: 19.0259\n",
      "Epoch 1198/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.2213 - val_loss: 19.0422\n",
      "Epoch 1199/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.2415 - val_loss: 18.9757\n",
      "Epoch 1200/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.2021 - val_loss: 19.0382\n",
      "Epoch 1201/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.2066 - val_loss: 19.0083\n",
      "Epoch 1202/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.1673 - val_loss: 19.0585\n",
      "Epoch 1203/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.2600 - val_loss: 18.9734\n",
      "Epoch 1204/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.2080 - val_loss: 18.9356\n",
      "Epoch 1205/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.1687 - val_loss: 18.9562\n",
      "Epoch 1206/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.1658 - val_loss: 18.9372\n",
      "Epoch 1207/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.1389 - val_loss: 19.0128\n",
      "Epoch 1208/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.1893 - val_loss: 18.9239\n",
      "Epoch 1209/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.1274 - val_loss: 18.9256\n",
      "Epoch 1210/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.1775 - val_loss: 18.9192\n",
      "Epoch 1211/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.1233 - val_loss: 18.9198\n",
      "Epoch 1212/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.1627 - val_loss: 18.8255\n",
      "Epoch 1213/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.1035 - val_loss: 18.8015\n",
      "Epoch 1214/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.1029 - val_loss: 18.7900\n",
      "Epoch 1215/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.0963 - val_loss: 18.7826\n",
      "Epoch 1216/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.0710 - val_loss: 18.8014\n",
      "Epoch 1217/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.0740 - val_loss: 18.7846\n",
      "Epoch 1218/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.0486 - val_loss: 18.7743\n",
      "Epoch 1219/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.0494 - val_loss: 18.7930\n",
      "Epoch 1220/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.0617 - val_loss: 18.7726\n",
      "Epoch 1221/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.0674 - val_loss: 18.7419\n",
      "Epoch 1222/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 15.0707 - val_loss: 18.7919\n",
      "Epoch 1223/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.0746 - val_loss: 18.7804\n",
      "Epoch 1224/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.0632 - val_loss: 18.7584\n",
      "Epoch 1225/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.0169 - val_loss: 18.7640\n",
      "Epoch 1226/2000\n",
      "1/1 [==============================] - 0s 24ms/step - loss: 15.0480 - val_loss: 18.7503\n",
      "Epoch 1227/2000\n",
      "1/1 [==============================] - 0s 21ms/step - loss: 14.9845 - val_loss: 18.7396\n",
      "Epoch 1228/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.0485 - val_loss: 18.6893\n",
      "Epoch 1229/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.9663 - val_loss: 18.6789\n",
      "Epoch 1230/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.9732 - val_loss: 18.7653\n",
      "Epoch 1231/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.0576 - val_loss: 18.7247\n",
      "Epoch 1232/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.0103 - val_loss: 18.6891\n",
      "Epoch 1233/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.9636 - val_loss: 18.8593\n",
      "Epoch 1234/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.1216 - val_loss: 18.7549\n",
      "Epoch 1235/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.0432 - val_loss: 18.7611\n",
      "Epoch 1236/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.0725 - val_loss: 18.8358\n",
      "Epoch 1237/2000\n",
      "1/1 [==============================] - ETA: 0s - loss: 15.09 - 0s 17ms/step - loss: 15.0996 - val_loss: 18.7044\n",
      "Epoch 1238/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.9782 - val_loss: 18.7845\n",
      "Epoch 1239/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.1809 - val_loss: 18.6740\n",
      "Epoch 1240/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.9648 - val_loss: 19.0860\n",
      "Epoch 1241/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.3431 - val_loss: 18.7248\n",
      "Epoch 1242/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.9879 - val_loss: 18.8043\n",
      "Epoch 1243/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.1460 - val_loss: 18.7562\n",
      "Epoch 1244/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.0083 - val_loss: 18.7680\n",
      "Epoch 1245/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.9687 - val_loss: 18.7647\n",
      "Epoch 1246/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.0619 - val_loss: 18.6893\n",
      "Epoch 1247/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.9542 - val_loss: 18.8261\n",
      "Epoch 1248/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 15.0115 - val_loss: 18.6895\n",
      "Epoch 1249/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.8783 - val_loss: 18.6680\n",
      "Epoch 1250/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 15.0054 - val_loss: 18.6084\n",
      "Epoch 1251/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 14.8694 - val_loss: 18.7282\n",
      "Epoch 1252/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.9801 - val_loss: 18.6228\n",
      "Epoch 1253/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.9237 - val_loss: 18.6429\n",
      "Epoch 1254/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 14.9130 - val_loss: 18.6299\n",
      "Epoch 1255/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.8778 - val_loss: 18.7893\n",
      "Epoch 1256/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 15.0038 - val_loss: 18.6242\n",
      "Epoch 1257/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.9424 - val_loss: 18.5837\n",
      "Epoch 1258/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.9142 - val_loss: 18.7052\n",
      "Epoch 1259/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.9227 - val_loss: 18.5854\n",
      "Epoch 1260/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.8482 - val_loss: 18.4992\n",
      "Epoch 1261/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 14.8647 - val_loss: 18.4796\n",
      "Epoch 1262/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.8523 - val_loss: 18.5800\n",
      "Epoch 1263/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.8602 - val_loss: 18.5416\n",
      "Epoch 1264/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.8596 - val_loss: 18.5215\n",
      "Epoch 1265/2000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 17ms/step - loss: 14.8271 - val_loss: 18.5173\n",
      "Epoch 1266/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.8241 - val_loss: 18.5444\n",
      "Epoch 1267/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.8313 - val_loss: 18.4938\n",
      "Epoch 1268/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.7959 - val_loss: 18.4786\n",
      "Epoch 1269/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.8021 - val_loss: 18.4815\n",
      "Epoch 1270/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.7821 - val_loss: 18.4960\n",
      "Epoch 1271/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.7920 - val_loss: 18.4472\n",
      "Epoch 1272/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.7726 - val_loss: 18.4910\n",
      "Epoch 1273/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 14.8228 - val_loss: 18.4530\n",
      "Epoch 1274/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.7911 - val_loss: 18.4678\n",
      "Epoch 1275/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.8045 - val_loss: 18.4507\n",
      "Epoch 1276/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.7771 - val_loss: 18.4503\n",
      "Epoch 1277/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.7710 - val_loss: 18.4232\n",
      "Epoch 1278/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.7515 - val_loss: 18.4725\n",
      "Epoch 1279/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.7891 - val_loss: 18.3904\n",
      "Epoch 1280/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.7695 - val_loss: 18.4048\n",
      "Epoch 1281/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.7909 - val_loss: 18.4985\n",
      "Epoch 1282/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.7475 - val_loss: 18.4542\n",
      "Epoch 1283/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.7207 - val_loss: 18.4045\n",
      "Epoch 1284/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.7635 - val_loss: 18.3913\n",
      "Epoch 1285/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.7141 - val_loss: 18.5051\n",
      "Epoch 1286/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 14.7713 - val_loss: 18.3699\n",
      "Epoch 1287/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.6876 - val_loss: 18.3939\n",
      "Epoch 1288/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 14.7505 - val_loss: 18.3888\n",
      "Epoch 1289/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.6980 - val_loss: 18.4070\n",
      "Epoch 1290/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.7110 - val_loss: 18.3603\n",
      "Epoch 1291/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.6925 - val_loss: 18.3602\n",
      "Epoch 1292/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.7052 - val_loss: 18.3798\n",
      "Epoch 1293/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.6973 - val_loss: 18.3942\n",
      "Epoch 1294/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.6754 - val_loss: 18.3089\n",
      "Epoch 1295/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.6667 - val_loss: 18.3001\n",
      "Epoch 1296/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.6850 - val_loss: 18.3300\n",
      "Epoch 1297/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.6676 - val_loss: 18.3157\n",
      "Epoch 1298/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.6557 - val_loss: 18.3265\n",
      "Epoch 1299/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.6573 - val_loss: 18.3051\n",
      "Epoch 1300/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.6681 - val_loss: 18.2987\n",
      "Epoch 1301/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.6592 - val_loss: 18.3334\n",
      "Epoch 1302/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.6556 - val_loss: 18.3535\n",
      "Epoch 1303/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.6878 - val_loss: 18.3265\n",
      "Epoch 1304/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.6589 - val_loss: 18.2525\n",
      "Epoch 1305/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.6437 - val_loss: 18.2508\n",
      "Epoch 1306/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.6579 - val_loss: 18.2362\n",
      "Epoch 1307/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.6321 - val_loss: 18.2434\n",
      "Epoch 1308/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.6292 - val_loss: 18.2375\n",
      "Epoch 1309/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.6238 - val_loss: 18.2412\n",
      "Epoch 1310/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.6354 - val_loss: 18.2384\n",
      "Epoch 1311/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.6211 - val_loss: 18.2270\n",
      "Epoch 1312/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.6295 - val_loss: 18.2610\n",
      "Epoch 1313/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.6208 - val_loss: 18.2860\n",
      "Epoch 1314/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.6328 - val_loss: 18.2401\n",
      "Epoch 1315/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.6030 - val_loss: 18.1916\n",
      "Epoch 1316/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.6068 - val_loss: 18.2596\n",
      "Epoch 1317/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.6132 - val_loss: 18.2616\n",
      "Epoch 1318/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 14.6116 - val_loss: 18.2245\n",
      "Epoch 1319/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.5986 - val_loss: 18.1831\n",
      "Epoch 1320/2000\n",
      "1/1 [==============================] - 0s 23ms/step - loss: 14.5736 - val_loss: 18.2205\n",
      "Epoch 1321/2000\n",
      "1/1 [==============================] - 0s 22ms/step - loss: 14.5699 - val_loss: 18.2051\n",
      "Epoch 1322/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 14.5724 - val_loss: 18.1833\n",
      "Epoch 1323/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 14.6185 - val_loss: 18.1768\n",
      "Epoch 1324/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.5489 - val_loss: 18.2653\n",
      "Epoch 1325/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.5932 - val_loss: 18.1655\n",
      "Epoch 1326/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.5606 - val_loss: 18.1685\n",
      "Epoch 1327/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.6120 - val_loss: 18.1824\n",
      "Epoch 1328/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 14.5596 - val_loss: 18.1632\n",
      "Epoch 1329/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.5564 - val_loss: 18.1364\n",
      "Epoch 1330/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.5641 - val_loss: 18.1385\n",
      "Epoch 1331/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.5628 - val_loss: 18.3486\n",
      "Epoch 1332/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.6795 - val_loss: 18.1817\n",
      "Epoch 1333/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.6065 - val_loss: 18.1753\n",
      "Epoch 1334/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.5785 - val_loss: 18.3146\n",
      "Epoch 1335/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.6350 - val_loss: 18.1456\n",
      "Epoch 1336/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.5413 - val_loss: 18.1724\n",
      "Epoch 1337/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.5617 - val_loss: 18.1211\n",
      "Epoch 1338/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.5266 - val_loss: 18.2912\n",
      "Epoch 1339/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.6225 - val_loss: 18.1154\n",
      "Epoch 1340/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.5285 - val_loss: 18.1331\n",
      "Epoch 1341/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 14.5391 - val_loss: 18.0781\n",
      "Epoch 1342/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.5238 - val_loss: 18.1507\n",
      "Epoch 1343/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.5193 - val_loss: 18.1161\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1344/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.4958 - val_loss: 18.1430\n",
      "Epoch 1345/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.5077 - val_loss: 18.0563\n",
      "Epoch 1346/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.5034 - val_loss: 18.0965\n",
      "Epoch 1347/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.4997 - val_loss: 18.0765\n",
      "Epoch 1348/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.4779 - val_loss: 18.0928\n",
      "Epoch 1349/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.4804 - val_loss: 18.0946\n",
      "Epoch 1350/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.4796 - val_loss: 18.0523\n",
      "Epoch 1351/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.4624 - val_loss: 18.0092\n",
      "Epoch 1352/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.4466 - val_loss: 18.0204\n",
      "Epoch 1353/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 14.4431 - val_loss: 18.0632\n",
      "Epoch 1354/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.4825 - val_loss: 18.0301\n",
      "Epoch 1355/2000\n",
      "1/1 [==============================] - 0s 23ms/step - loss: 14.4444 - val_loss: 18.0515\n",
      "Epoch 1356/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 14.4348 - val_loss: 18.0375\n",
      "Epoch 1357/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.4186 - val_loss: 18.0130\n",
      "Epoch 1358/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.4183 - val_loss: 17.9878\n",
      "Epoch 1359/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 14.3965 - val_loss: 18.0196\n",
      "Epoch 1360/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3972 - val_loss: 18.0468\n",
      "Epoch 1361/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3906 - val_loss: 18.0265\n",
      "Epoch 1362/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3840 - val_loss: 18.0285\n",
      "Epoch 1363/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.3811 - val_loss: 18.0311\n",
      "Epoch 1364/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3964 - val_loss: 18.0783\n",
      "Epoch 1365/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.3928 - val_loss: 18.0687\n",
      "Epoch 1366/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3958 - val_loss: 18.0934\n",
      "Epoch 1367/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3928 - val_loss: 18.0227\n",
      "Epoch 1368/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.3952 - val_loss: 18.0334\n",
      "Epoch 1369/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3897 - val_loss: 18.0187\n",
      "Epoch 1370/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.3863 - val_loss: 18.0201\n",
      "Epoch 1371/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3777 - val_loss: 18.0227\n",
      "Epoch 1372/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.3763 - val_loss: 18.0232\n",
      "Epoch 1373/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3690 - val_loss: 17.9959\n",
      "Epoch 1374/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 14.3590 - val_loss: 17.9164\n",
      "Epoch 1375/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.3657 - val_loss: 17.9738\n",
      "Epoch 1376/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3488 - val_loss: 18.0307\n",
      "Epoch 1377/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.3702 - val_loss: 17.9031\n",
      "Epoch 1378/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3729 - val_loss: 17.9135\n",
      "Epoch 1379/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.3769 - val_loss: 18.0603\n",
      "Epoch 1380/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3894 - val_loss: 17.8950\n",
      "Epoch 1381/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3891 - val_loss: 17.9325\n",
      "Epoch 1382/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.3880 - val_loss: 17.9262\n",
      "Epoch 1383/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.3880 - val_loss: 18.0020\n",
      "Epoch 1384/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3635 - val_loss: 17.8934\n",
      "Epoch 1385/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3660 - val_loss: 17.8645\n",
      "Epoch 1386/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.3515 - val_loss: 17.9118\n",
      "Epoch 1387/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.3445 - val_loss: 17.9673\n",
      "Epoch 1388/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 14.3819 - val_loss: 17.9322\n",
      "Epoch 1389/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3186 - val_loss: 17.9278\n",
      "Epoch 1390/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3093 - val_loss: 17.9228\n",
      "Epoch 1391/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3603 - val_loss: 17.8688\n",
      "Epoch 1392/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.3413 - val_loss: 17.8529\n",
      "Epoch 1393/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3122 - val_loss: 17.8531\n",
      "Epoch 1394/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3198 - val_loss: 18.0438\n",
      "Epoch 1395/2000\n",
      "1/1 [==============================] - 0s 21ms/step - loss: 14.3704 - val_loss: 17.8174\n",
      "Epoch 1396/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 14.2986 - val_loss: 17.8265\n",
      "Epoch 1397/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3013 - val_loss: 17.8849\n",
      "Epoch 1398/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.3623 - val_loss: 17.8789\n",
      "Epoch 1399/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.2889 - val_loss: 17.8853\n",
      "Epoch 1400/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.2766 - val_loss: 17.8597\n",
      "Epoch 1401/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.3264 - val_loss: 17.9261\n",
      "Epoch 1402/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.2719 - val_loss: 17.8881\n",
      "Epoch 1403/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.2535 - val_loss: 17.8710\n",
      "Epoch 1404/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.3161 - val_loss: 17.8358\n",
      "Epoch 1405/2000\n",
      "1/1 [==============================] - ETA: 0s - loss: 14.26 - 0s 17ms/step - loss: 14.2630 - val_loss: 17.8159\n",
      "Epoch 1406/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.2648 - val_loss: 17.8074\n",
      "Epoch 1407/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.2533 - val_loss: 17.7788\n",
      "Epoch 1408/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.2556 - val_loss: 17.8699\n",
      "Epoch 1409/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.2425 - val_loss: 17.8342\n",
      "Epoch 1410/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.2379 - val_loss: 17.8067\n",
      "Epoch 1411/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.2982 - val_loss: 17.8280\n",
      "Epoch 1412/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.2365 - val_loss: 17.8635\n",
      "Epoch 1413/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.3052 - val_loss: 17.7724\n",
      "Epoch 1414/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.2264 - val_loss: 17.7874\n",
      "Epoch 1415/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.2241 - val_loss: 17.7999\n",
      "Epoch 1416/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.2266 - val_loss: 17.7911\n",
      "Epoch 1417/2000\n",
      "1/1 [==============================] - 0s 26ms/step - loss: 14.2338 - val_loss: 17.7728\n",
      "Epoch 1418/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 14.2260 - val_loss: 17.7700\n",
      "Epoch 1419/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.2005 - val_loss: 17.8199\n",
      "Epoch 1420/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 14.2257 - val_loss: 17.8067\n",
      "Epoch 1421/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.2114 - val_loss: 17.8240\n",
      "Epoch 1422/2000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 16ms/step - loss: 14.2060 - val_loss: 17.8570\n",
      "Epoch 1423/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.2096 - val_loss: 17.8402\n",
      "Epoch 1424/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.1602 - val_loss: 17.7769\n",
      "Epoch 1425/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.1743 - val_loss: 17.7512\n",
      "Epoch 1426/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.1711 - val_loss: 17.7343\n",
      "Epoch 1427/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.2218 - val_loss: 17.7545\n",
      "Epoch 1428/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.1680 - val_loss: 17.7385\n",
      "Epoch 1429/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.1998 - val_loss: 17.7572\n",
      "Epoch 1430/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 14.1532 - val_loss: 17.7259\n",
      "Epoch 1431/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.1628 - val_loss: 17.6694\n",
      "Epoch 1432/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.2014 - val_loss: 17.6691\n",
      "Epoch 1433/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.2156 - val_loss: 17.7096\n",
      "Epoch 1434/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.1959 - val_loss: 17.7183\n",
      "Epoch 1435/2000\n",
      "1/1 [==============================] - 0s 24ms/step - loss: 14.2004 - val_loss: 17.6968\n",
      "Epoch 1436/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 14.1884 - val_loss: 17.6755\n",
      "Epoch 1437/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 14.1751 - val_loss: 17.6888\n",
      "Epoch 1438/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.1666 - val_loss: 17.6815\n",
      "Epoch 1439/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.1720 - val_loss: 17.6971\n",
      "Epoch 1440/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 14.1688 - val_loss: 17.6542\n",
      "Epoch 1441/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.1685 - val_loss: 17.6641\n",
      "Epoch 1442/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.1789 - val_loss: 17.6511\n",
      "Epoch 1443/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.1658 - val_loss: 17.6670\n",
      "Epoch 1444/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 14.1379 - val_loss: 17.6850\n",
      "Epoch 1445/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.1300 - val_loss: 17.6797\n",
      "Epoch 1446/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.1434 - val_loss: 17.6896\n",
      "Epoch 1447/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 14.1283 - val_loss: 17.6460\n",
      "Epoch 1448/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.1289 - val_loss: 17.6980\n",
      "Epoch 1449/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.1621 - val_loss: 17.7697\n",
      "Epoch 1450/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.2006 - val_loss: 17.6086\n",
      "Epoch 1451/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.0861 - val_loss: 17.8130\n",
      "Epoch 1452/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.2683 - val_loss: 17.7270\n",
      "Epoch 1453/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 14.2118 - val_loss: 17.7369\n",
      "Epoch 1454/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.2030 - val_loss: 17.8606\n",
      "Epoch 1455/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.2564 - val_loss: 17.6691\n",
      "Epoch 1456/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.1428 - val_loss: 17.6540\n",
      "Epoch 1457/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.1363 - val_loss: 17.6674\n",
      "Epoch 1458/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.1441 - val_loss: 17.5714\n",
      "Epoch 1459/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 14.0702 - val_loss: 17.7192\n",
      "Epoch 1460/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.1665 - val_loss: 17.6310\n",
      "Epoch 1461/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.1366 - val_loss: 17.5990\n",
      "Epoch 1462/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.0656 - val_loss: 17.7408\n",
      "Epoch 1463/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.1722 - val_loss: 17.5460\n",
      "Epoch 1464/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.0501 - val_loss: 17.5435\n",
      "Epoch 1465/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.0718 - val_loss: 17.5432\n",
      "Epoch 1466/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.0814 - val_loss: 17.5487\n",
      "Epoch 1467/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.0688 - val_loss: 17.5403\n",
      "Epoch 1468/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.0632 - val_loss: 17.5356\n",
      "Epoch 1469/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.0464 - val_loss: 17.5213\n",
      "Epoch 1470/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.0379 - val_loss: 17.5075\n",
      "Epoch 1471/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.0431 - val_loss: 17.4914\n",
      "Epoch 1472/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.0400 - val_loss: 17.5170\n",
      "Epoch 1473/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 14.0299 - val_loss: 17.5135\n",
      "Epoch 1474/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.0111 - val_loss: 17.4913\n",
      "Epoch 1475/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.0343 - val_loss: 17.4888\n",
      "Epoch 1476/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 14.0184 - val_loss: 17.5351\n",
      "Epoch 1477/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.0278 - val_loss: 17.5227\n",
      "Epoch 1478/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.0010 - val_loss: 17.4908\n",
      "Epoch 1479/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.0115 - val_loss: 17.5025\n",
      "Epoch 1480/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 14.0040 - val_loss: 17.4752\n",
      "Epoch 1481/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.9969 - val_loss: 17.4521\n",
      "Epoch 1482/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.9774 - val_loss: 17.4455\n",
      "Epoch 1483/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.9580 - val_loss: 17.5057\n",
      "Epoch 1484/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.0307 - val_loss: 17.3871\n",
      "Epoch 1485/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.9836 - val_loss: 17.4001\n",
      "Epoch 1486/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.9769 - val_loss: 17.4444\n",
      "Epoch 1487/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.0071 - val_loss: 17.4299\n",
      "Epoch 1488/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.9714 - val_loss: 17.4937\n",
      "Epoch 1489/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.9507 - val_loss: 17.4766\n",
      "Epoch 1490/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 14.0140 - val_loss: 17.4776\n",
      "Epoch 1491/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.9390 - val_loss: 17.4144\n",
      "Epoch 1492/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.9406 - val_loss: 17.4567\n",
      "Epoch 1493/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.9813 - val_loss: 17.4239\n",
      "Epoch 1494/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.9572 - val_loss: 17.4000\n",
      "Epoch 1495/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.9668 - val_loss: 17.4112\n",
      "Epoch 1496/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.9447 - val_loss: 17.4269\n",
      "Epoch 1497/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.9552 - val_loss: 17.4357\n",
      "Epoch 1498/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.9313 - val_loss: 17.4365\n",
      "Epoch 1499/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.9297 - val_loss: 17.4107\n",
      "Epoch 1500/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.9142 - val_loss: 17.4146\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1501/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.9047 - val_loss: 17.4112\n",
      "Epoch 1502/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.9277 - val_loss: 17.3769\n",
      "Epoch 1503/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.9376 - val_loss: 17.3885\n",
      "Epoch 1504/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.8967 - val_loss: 17.4233\n",
      "Epoch 1505/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.9204 - val_loss: 17.4314\n",
      "Epoch 1506/2000\n",
      "1/1 [==============================] - 0s 25ms/step - loss: 13.9396 - val_loss: 17.4226\n",
      "Epoch 1507/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.9164 - val_loss: 17.4498\n",
      "Epoch 1508/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.8897 - val_loss: 17.4159\n",
      "Epoch 1509/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.9153 - val_loss: 17.4469\n",
      "Epoch 1510/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.9090 - val_loss: 17.4327\n",
      "Epoch 1511/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.8912 - val_loss: 17.4134\n",
      "Epoch 1512/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.8872 - val_loss: 17.3740\n",
      "Epoch 1513/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.8804 - val_loss: 17.3754\n",
      "Epoch 1514/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.8773 - val_loss: 17.3557\n",
      "Epoch 1515/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.8985 - val_loss: 17.3652\n",
      "Epoch 1516/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.8743 - val_loss: 17.3544\n",
      "Epoch 1517/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.8584 - val_loss: 17.3939\n",
      "Epoch 1518/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.8916 - val_loss: 17.3536\n",
      "Epoch 1519/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.8652 - val_loss: 17.3326\n",
      "Epoch 1520/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.8441 - val_loss: 17.3406\n",
      "Epoch 1521/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.8402 - val_loss: 17.3410\n",
      "Epoch 1522/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.8419 - val_loss: 17.3361\n",
      "Epoch 1523/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.8450 - val_loss: 17.3603\n",
      "Epoch 1524/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.8083 - val_loss: 17.4192\n",
      "Epoch 1525/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.8585 - val_loss: 17.3925\n",
      "Epoch 1526/2000\n",
      "1/1 [==============================] - 0s 21ms/step - loss: 13.8651 - val_loss: 17.4015\n",
      "Epoch 1527/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.8213 - val_loss: 17.3625\n",
      "Epoch 1528/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.8553 - val_loss: 17.3827\n",
      "Epoch 1529/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.8128 - val_loss: 17.3218\n",
      "Epoch 1530/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.8340 - val_loss: 17.3126\n",
      "Epoch 1531/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.8270 - val_loss: 17.3840\n",
      "Epoch 1532/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.8265 - val_loss: 17.3184\n",
      "Epoch 1533/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.8698 - val_loss: 17.3265\n",
      "Epoch 1534/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.7986 - val_loss: 17.4091\n",
      "Epoch 1535/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.8365 - val_loss: 17.3585\n",
      "Epoch 1536/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.8791 - val_loss: 17.3192\n",
      "Epoch 1537/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.8456 - val_loss: 17.4876\n",
      "Epoch 1538/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.9132 - val_loss: 17.2981\n",
      "Epoch 1539/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 13.8067 - val_loss: 17.3551\n",
      "Epoch 1540/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.9521 - val_loss: 17.4007\n",
      "Epoch 1541/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.8241 - val_loss: 17.3641\n",
      "Epoch 1542/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.8118 - val_loss: 17.3268\n",
      "Epoch 1543/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.8944 - val_loss: 17.2522\n",
      "Epoch 1544/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.7689 - val_loss: 17.3746\n",
      "Epoch 1545/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.8135 - val_loss: 17.2394\n",
      "Epoch 1546/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.7699 - val_loss: 17.3203\n",
      "Epoch 1547/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.8121 - val_loss: 17.3874\n",
      "Epoch 1548/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.8015 - val_loss: 17.2901\n",
      "Epoch 1549/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.7603 - val_loss: 17.2613\n",
      "Epoch 1550/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.7609 - val_loss: 17.2745\n",
      "Epoch 1551/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.7539 - val_loss: 17.3268\n",
      "Epoch 1552/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.7595 - val_loss: 17.2913\n",
      "Epoch 1553/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.7442 - val_loss: 17.2996\n",
      "Epoch 1554/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.7400 - val_loss: 17.2930\n",
      "Epoch 1555/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.7375 - val_loss: 17.2923\n",
      "Epoch 1556/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.7331 - val_loss: 17.2776\n",
      "Epoch 1557/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.7390 - val_loss: 17.2574\n",
      "Epoch 1558/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.7301 - val_loss: 17.2944\n",
      "Epoch 1559/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.7228 - val_loss: 17.2981\n",
      "Epoch 1560/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.7181 - val_loss: 17.2908\n",
      "Epoch 1561/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.7002 - val_loss: 17.2949\n",
      "Epoch 1562/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.7067 - val_loss: 17.3167\n",
      "Epoch 1563/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.7125 - val_loss: 17.2691\n",
      "Epoch 1564/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.7149 - val_loss: 17.2731\n",
      "Epoch 1565/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.6989 - val_loss: 17.2738\n",
      "Epoch 1566/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.6959 - val_loss: 17.2414\n",
      "Epoch 1567/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.7046 - val_loss: 17.2504\n",
      "Epoch 1568/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.6984 - val_loss: 17.2512\n",
      "Epoch 1569/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.6858 - val_loss: 17.2336\n",
      "Epoch 1570/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.7083 - val_loss: 17.2604\n",
      "Epoch 1571/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.7107 - val_loss: 17.2490\n",
      "Epoch 1572/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.6664 - val_loss: 17.3417\n",
      "Epoch 1573/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.7447 - val_loss: 17.3433\n",
      "Epoch 1574/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.8181 - val_loss: 17.2594\n",
      "Epoch 1575/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.6813 - val_loss: 17.3136\n",
      "Epoch 1576/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.7726 - val_loss: 17.3871\n",
      "Epoch 1577/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.8728 - val_loss: 17.2094\n",
      "Epoch 1578/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.7042 - val_loss: 17.6012\n",
      "Epoch 1579/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.9980 - val_loss: 17.3045\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1580/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.7421 - val_loss: 17.2684\n",
      "Epoch 1581/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.7693 - val_loss: 17.4067\n",
      "Epoch 1582/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.8257 - val_loss: 17.2796\n",
      "Epoch 1583/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.6697 - val_loss: 17.3912\n",
      "Epoch 1584/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.8601 - val_loss: 17.2349\n",
      "Epoch 1585/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.6417 - val_loss: 17.4561\n",
      "Epoch 1586/2000\n",
      "1/1 [==============================] - 0s 23ms/step - loss: 13.7960 - val_loss: 17.2891\n",
      "Epoch 1587/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.6962 - val_loss: 17.2442\n",
      "Epoch 1588/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.6668 - val_loss: 17.4106\n",
      "Epoch 1589/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.7355 - val_loss: 17.2453\n",
      "Epoch 1590/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.6480 - val_loss: 17.2623\n",
      "Epoch 1591/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.6672 - val_loss: 17.2583\n",
      "Epoch 1592/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.6374 - val_loss: 17.2990\n",
      "Epoch 1593/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.6530 - val_loss: 17.2104\n",
      "Epoch 1594/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.6396 - val_loss: 17.1822\n",
      "Epoch 1595/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.6764 - val_loss: 17.1838\n",
      "Epoch 1596/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.6343 - val_loss: 17.1785\n",
      "Epoch 1597/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.6376 - val_loss: 17.1998\n",
      "Epoch 1598/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.6558 - val_loss: 17.2165\n",
      "Epoch 1599/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.6368 - val_loss: 17.2860\n",
      "Epoch 1600/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.6510 - val_loss: 17.2208\n",
      "Epoch 1601/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.6213 - val_loss: 17.2593\n",
      "Epoch 1602/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.6645 - val_loss: 17.2325\n",
      "Epoch 1603/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.6101 - val_loss: 17.2803\n",
      "Epoch 1604/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.6451 - val_loss: 17.2510\n",
      "Epoch 1605/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.6580 - val_loss: 17.2050\n",
      "Epoch 1606/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.6202 - val_loss: 17.2862\n",
      "Epoch 1607/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.6395 - val_loss: 17.1685\n",
      "Epoch 1608/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.5851 - val_loss: 17.2479\n",
      "Epoch 1609/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.6623 - val_loss: 17.2549\n",
      "Epoch 1610/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.6414 - val_loss: 17.1949\n",
      "Epoch 1611/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.5716 - val_loss: 17.2135\n",
      "Epoch 1612/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.6055 - val_loss: 17.2000\n",
      "Epoch 1613/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.5804 - val_loss: 17.2786\n",
      "Epoch 1614/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.6373 - val_loss: 17.2371\n",
      "Epoch 1615/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.6368 - val_loss: 17.1694\n",
      "Epoch 1616/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 13.5780 - val_loss: 17.3610\n",
      "Epoch 1617/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.7129 - val_loss: 17.2363\n",
      "Epoch 1618/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.6270 - val_loss: 17.1314\n",
      "Epoch 1619/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.5877 - val_loss: 17.2729\n",
      "Epoch 1620/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.6652 - val_loss: 17.1410\n",
      "Epoch 1621/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.5517 - val_loss: 17.1961\n",
      "Epoch 1622/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.6307 - val_loss: 17.1857\n",
      "Epoch 1623/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.5458 - val_loss: 17.2781\n",
      "Epoch 1624/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.6653 - val_loss: 17.1597\n",
      "Epoch 1625/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.5885 - val_loss: 17.1814\n",
      "Epoch 1626/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.5732 - val_loss: 17.3376\n",
      "Epoch 1627/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.7023 - val_loss: 17.1232\n",
      "Epoch 1628/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.5553 - val_loss: 17.0999\n",
      "Epoch 1629/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.5711 - val_loss: 17.2333\n",
      "Epoch 1630/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.6253 - val_loss: 17.1708\n",
      "Epoch 1631/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.5391 - val_loss: 17.1481\n",
      "Epoch 1632/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.5265 - val_loss: 17.1627\n",
      "Epoch 1633/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.5324 - val_loss: 17.1809\n",
      "Epoch 1634/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4970 - val_loss: 17.1357\n",
      "Epoch 1635/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.5019 - val_loss: 17.1315\n",
      "Epoch 1636/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.5128 - val_loss: 17.1570\n",
      "Epoch 1637/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.5123 - val_loss: 17.1087\n",
      "Epoch 1638/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.5161 - val_loss: 17.1111\n",
      "Epoch 1639/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4921 - val_loss: 17.1462\n",
      "Epoch 1640/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4983 - val_loss: 17.1288\n",
      "Epoch 1641/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4946 - val_loss: 17.1431\n",
      "Epoch 1642/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.4974 - val_loss: 17.1319\n",
      "Epoch 1643/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.4873 - val_loss: 17.0930\n",
      "Epoch 1644/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4830 - val_loss: 17.1319\n",
      "Epoch 1645/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.5308 - val_loss: 17.1418\n",
      "Epoch 1646/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4934 - val_loss: 17.1334\n",
      "Epoch 1647/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4660 - val_loss: 17.1231\n",
      "Epoch 1648/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.4885 - val_loss: 17.1244\n",
      "Epoch 1649/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4635 - val_loss: 17.1233\n",
      "Epoch 1650/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4737 - val_loss: 17.1244\n",
      "Epoch 1651/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.4647 - val_loss: 17.1181\n",
      "Epoch 1652/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.4765 - val_loss: 17.0997\n",
      "Epoch 1653/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.4870 - val_loss: 17.0452\n",
      "Epoch 1654/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.4483 - val_loss: 17.0615\n",
      "Epoch 1655/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.4736 - val_loss: 17.0785\n",
      "Epoch 1656/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.4785 - val_loss: 17.0978\n",
      "Epoch 1657/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.4649 - val_loss: 17.1064\n",
      "Epoch 1658/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.4651 - val_loss: 17.1040\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1659/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.4626 - val_loss: 17.1014\n",
      "Epoch 1660/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.4516 - val_loss: 17.0823\n",
      "Epoch 1661/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4418 - val_loss: 17.0907\n",
      "Epoch 1662/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.4282 - val_loss: 17.0895\n",
      "Epoch 1663/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4557 - val_loss: 17.0985\n",
      "Epoch 1664/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4294 - val_loss: 17.0377\n",
      "Epoch 1665/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.4111 - val_loss: 17.0594\n",
      "Epoch 1666/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.4044 - val_loss: 17.0284\n",
      "Epoch 1667/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4183 - val_loss: 17.0791\n",
      "Epoch 1668/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4081 - val_loss: 17.0817\n",
      "Epoch 1669/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4147 - val_loss: 17.1120\n",
      "Epoch 1670/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4152 - val_loss: 17.0970\n",
      "Epoch 1671/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.4240 - val_loss: 17.0760\n",
      "Epoch 1672/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4025 - val_loss: 17.0363\n",
      "Epoch 1673/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4011 - val_loss: 17.0906\n",
      "Epoch 1674/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 13.4309 - val_loss: 17.0264\n",
      "Epoch 1675/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.3960 - val_loss: 17.0599\n",
      "Epoch 1676/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3922 - val_loss: 17.1231\n",
      "Epoch 1677/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.4509 - val_loss: 17.0822\n",
      "Epoch 1678/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4001 - val_loss: 17.0769\n",
      "Epoch 1679/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3848 - val_loss: 17.0719\n",
      "Epoch 1680/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3881 - val_loss: 17.0333\n",
      "Epoch 1681/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4217 - val_loss: 16.9990\n",
      "Epoch 1682/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.3851 - val_loss: 17.0544\n",
      "Epoch 1683/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.4242 - val_loss: 17.0294\n",
      "Epoch 1684/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4420 - val_loss: 17.0306\n",
      "Epoch 1685/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.3957 - val_loss: 17.0707\n",
      "Epoch 1686/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4439 - val_loss: 17.0205\n",
      "Epoch 1687/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4210 - val_loss: 16.9974\n",
      "Epoch 1688/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.3897 - val_loss: 17.0216\n",
      "Epoch 1689/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.4245 - val_loss: 17.0152\n",
      "Epoch 1690/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3666 - val_loss: 17.0412\n",
      "Epoch 1691/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3799 - val_loss: 17.0347\n",
      "Epoch 1692/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.4142 - val_loss: 17.0108\n",
      "Epoch 1693/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.3688 - val_loss: 17.1091\n",
      "Epoch 1694/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.3997 - val_loss: 17.0669\n",
      "Epoch 1695/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.4002 - val_loss: 17.0164\n",
      "Epoch 1696/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.3531 - val_loss: 17.0694\n",
      "Epoch 1697/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3974 - val_loss: 17.0094\n",
      "Epoch 1698/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3693 - val_loss: 17.0359\n",
      "Epoch 1699/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.3508 - val_loss: 17.0584\n",
      "Epoch 1700/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.4023 - val_loss: 17.0184\n",
      "Epoch 1701/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.3657 - val_loss: 17.0298\n",
      "Epoch 1702/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.3140 - val_loss: 16.9985\n",
      "Epoch 1703/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.3654 - val_loss: 16.9881\n",
      "Epoch 1704/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.3370 - val_loss: 17.0043\n",
      "Epoch 1705/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.3457 - val_loss: 16.9821\n",
      "Epoch 1706/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3300 - val_loss: 17.0042\n",
      "Epoch 1707/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3319 - val_loss: 16.9653\n",
      "Epoch 1708/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3472 - val_loss: 16.9397\n",
      "Epoch 1709/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3386 - val_loss: 17.0110\n",
      "Epoch 1710/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.3308 - val_loss: 17.0309\n",
      "Epoch 1711/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3309 - val_loss: 16.9996\n",
      "Epoch 1712/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3439 - val_loss: 16.9596\n",
      "Epoch 1713/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3384 - val_loss: 16.9358\n",
      "Epoch 1714/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.3421 - val_loss: 16.9419\n",
      "Epoch 1715/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.3412 - val_loss: 16.9266\n",
      "Epoch 1716/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3391 - val_loss: 16.9343\n",
      "Epoch 1717/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3197 - val_loss: 16.9573\n",
      "Epoch 1718/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.3242 - val_loss: 16.9349\n",
      "Epoch 1719/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.3472 - val_loss: 16.8980\n",
      "Epoch 1720/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3556 - val_loss: 16.8953\n",
      "Epoch 1721/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 13.3132 - val_loss: 16.9133\n",
      "Epoch 1722/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.3610 - val_loss: 17.1053\n",
      "Epoch 1723/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3780 - val_loss: 16.9570\n",
      "Epoch 1724/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3015 - val_loss: 17.0370\n",
      "Epoch 1725/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.4337 - val_loss: 17.1313\n",
      "Epoch 1726/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.4432 - val_loss: 16.8988\n",
      "Epoch 1727/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.2989 - val_loss: 17.0506\n",
      "Epoch 1728/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.5235 - val_loss: 17.0032\n",
      "Epoch 1729/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.3576 - val_loss: 16.9857\n",
      "Epoch 1730/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3158 - val_loss: 17.1339\n",
      "Epoch 1731/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.5291 - val_loss: 16.8803\n",
      "Epoch 1732/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.3017 - val_loss: 17.0611\n",
      "Epoch 1733/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.3655 - val_loss: 17.0044\n",
      "Epoch 1734/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3895 - val_loss: 16.8785\n",
      "Epoch 1735/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.2863 - val_loss: 16.9177\n",
      "Epoch 1736/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3084 - val_loss: 16.8495\n",
      "Epoch 1737/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.2681 - val_loss: 16.9045\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1738/2000\n",
      "1/1 [==============================] - 0s 23ms/step - loss: 13.2819 - val_loss: 16.9322\n",
      "Epoch 1739/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.2933 - val_loss: 16.8894\n",
      "Epoch 1740/2000\n",
      "1/1 [==============================] - 0s 21ms/step - loss: 13.2713 - val_loss: 16.8731\n",
      "Epoch 1741/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.3066 - val_loss: 16.9157\n",
      "Epoch 1742/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.2833 - val_loss: 16.8829\n",
      "Epoch 1743/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.2628 - val_loss: 16.8552\n",
      "Epoch 1744/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.2513 - val_loss: 16.8690\n",
      "Epoch 1745/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.2302 - val_loss: 16.9465\n",
      "Epoch 1746/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.3030 - val_loss: 16.8478\n",
      "Epoch 1747/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.2827 - val_loss: 16.8193\n",
      "Epoch 1748/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.2448 - val_loss: 17.0076\n",
      "Epoch 1749/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.3210 - val_loss: 16.7989\n",
      "Epoch 1750/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 13.2567 - val_loss: 16.8070\n",
      "Epoch 1751/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.2582 - val_loss: 16.9177\n",
      "Epoch 1752/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.2211 - val_loss: 16.8520\n",
      "Epoch 1753/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.2192 - val_loss: 16.7955\n",
      "Epoch 1754/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.2413 - val_loss: 16.8416\n",
      "Epoch 1755/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.2230 - val_loss: 16.8305\n",
      "Epoch 1756/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 13.1924 - val_loss: 16.8663\n",
      "Epoch 1757/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.2287 - val_loss: 16.8738\n",
      "Epoch 1758/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.2488 - val_loss: 16.8255\n",
      "Epoch 1759/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.2025 - val_loss: 16.8514\n",
      "Epoch 1760/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.1956 - val_loss: 16.8586\n",
      "Epoch 1761/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.1968 - val_loss: 16.8377\n",
      "Epoch 1762/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.2048 - val_loss: 16.8539\n",
      "Epoch 1763/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.2205 - val_loss: 16.8156\n",
      "Epoch 1764/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.1907 - val_loss: 16.8638\n",
      "Epoch 1765/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.1975 - val_loss: 16.8739\n",
      "Epoch 1766/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.1705 - val_loss: 16.8458\n",
      "Epoch 1767/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1682 - val_loss: 16.8563\n",
      "Epoch 1768/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.1777 - val_loss: 16.8526\n",
      "Epoch 1769/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.2002 - val_loss: 16.8397\n",
      "Epoch 1770/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1815 - val_loss: 16.8190\n",
      "Epoch 1771/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1776 - val_loss: 16.7875\n",
      "Epoch 1772/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.1986 - val_loss: 16.7449\n",
      "Epoch 1773/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.1854 - val_loss: 16.7414\n",
      "Epoch 1774/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.1628 - val_loss: 16.8291\n",
      "Epoch 1775/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1719 - val_loss: 16.8425\n",
      "Epoch 1776/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1677 - val_loss: 16.7953\n",
      "Epoch 1777/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.1545 - val_loss: 16.8101\n",
      "Epoch 1778/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1520 - val_loss: 16.8198\n",
      "Epoch 1779/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1586 - val_loss: 16.8028\n",
      "Epoch 1780/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 13.1372 - val_loss: 16.7973\n",
      "Epoch 1781/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1553 - val_loss: 16.7959\n",
      "Epoch 1782/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.1550 - val_loss: 16.7960\n",
      "Epoch 1783/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1590 - val_loss: 16.8013\n",
      "Epoch 1784/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1356 - val_loss: 16.8087\n",
      "Epoch 1785/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1490 - val_loss: 16.8053\n",
      "Epoch 1786/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.1370 - val_loss: 16.7793\n",
      "Epoch 1787/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1276 - val_loss: 16.7562\n",
      "Epoch 1788/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1843 - val_loss: 16.7531\n",
      "Epoch 1789/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1677 - val_loss: 16.7999\n",
      "Epoch 1790/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.1379 - val_loss: 16.7798\n",
      "Epoch 1791/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.1082 - val_loss: 16.7786\n",
      "Epoch 1792/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1316 - val_loss: 16.8232\n",
      "Epoch 1793/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1342 - val_loss: 16.7766\n",
      "Epoch 1794/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.1089 - val_loss: 16.7615\n",
      "Epoch 1795/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1017 - val_loss: 16.7330\n",
      "Epoch 1796/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.1103 - val_loss: 16.7309\n",
      "Epoch 1797/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.1077 - val_loss: 16.7173\n",
      "Epoch 1798/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.1142 - val_loss: 16.7691\n",
      "Epoch 1799/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1210 - val_loss: 16.7227\n",
      "Epoch 1800/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.1236 - val_loss: 16.7500\n",
      "Epoch 1801/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1038 - val_loss: 16.7653\n",
      "Epoch 1802/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.0936 - val_loss: 16.8028\n",
      "Epoch 1803/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1233 - val_loss: 16.7850\n",
      "Epoch 1804/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.0892 - val_loss: 16.7589\n",
      "Epoch 1805/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1042 - val_loss: 16.7384\n",
      "Epoch 1806/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.1068 - val_loss: 16.7135\n",
      "Epoch 1807/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.1034 - val_loss: 16.7088\n",
      "Epoch 1808/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 13.0962 - val_loss: 16.7249\n",
      "Epoch 1809/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 13.0955 - val_loss: 16.7118\n",
      "Epoch 1810/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.1065 - val_loss: 16.7463\n",
      "Epoch 1811/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.1152 - val_loss: 16.7163\n",
      "Epoch 1812/2000\n",
      "1/1 [==============================] - 0s 21ms/step - loss: 13.1235 - val_loss: 16.6799\n",
      "Epoch 1813/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.1064 - val_loss: 16.7743\n",
      "Epoch 1814/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.1242 - val_loss: 16.6631\n",
      "Epoch 1815/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.1032 - val_loss: 16.6778\n",
      "Epoch 1816/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.0919 - val_loss: 16.7449\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1817/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.0960 - val_loss: 16.6789\n",
      "Epoch 1818/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.0869 - val_loss: 16.6514\n",
      "Epoch 1819/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.0977 - val_loss: 16.6815\n",
      "Epoch 1820/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.0606 - val_loss: 16.6727\n",
      "Epoch 1821/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.0703 - val_loss: 16.6812\n",
      "Epoch 1822/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.0812 - val_loss: 16.7029\n",
      "Epoch 1823/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.0770 - val_loss: 16.7220\n",
      "Epoch 1824/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.0946 - val_loss: 16.7052\n",
      "Epoch 1825/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.0719 - val_loss: 16.6677\n",
      "Epoch 1826/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.1103 - val_loss: 16.7565\n",
      "Epoch 1827/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1434 - val_loss: 16.6612\n",
      "Epoch 1828/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.0673 - val_loss: 16.6872\n",
      "Epoch 1829/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.1031 - val_loss: 16.8163\n",
      "Epoch 1830/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1352 - val_loss: 16.6860\n",
      "Epoch 1831/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.0815 - val_loss: 16.7177\n",
      "Epoch 1832/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1569 - val_loss: 16.8469\n",
      "Epoch 1833/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.1488 - val_loss: 16.6668\n",
      "Epoch 1834/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.0514 - val_loss: 16.7378\n",
      "Epoch 1835/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.1864 - val_loss: 16.7196\n",
      "Epoch 1836/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.0732 - val_loss: 16.7533\n",
      "Epoch 1837/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.0718 - val_loss: 16.6868\n",
      "Epoch 1838/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.1472 - val_loss: 16.6749\n",
      "Epoch 1839/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.0224 - val_loss: 16.8432\n",
      "Epoch 1840/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.0907 - val_loss: 16.6756\n",
      "Epoch 1841/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.0873 - val_loss: 16.6355\n",
      "Epoch 1842/2000\n",
      "1/1 [==============================] - 0s 24ms/step - loss: 13.0436 - val_loss: 16.8252\n",
      "Epoch 1843/2000\n",
      "1/1 [==============================] - 0s 25ms/step - loss: 13.1976 - val_loss: 16.7010\n",
      "Epoch 1844/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.1663 - val_loss: 16.6072\n",
      "Epoch 1845/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.0324 - val_loss: 16.7755\n",
      "Epoch 1846/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.1256 - val_loss: 16.6268\n",
      "Epoch 1847/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.0323 - val_loss: 16.6911\n",
      "Epoch 1848/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 13.0927 - val_loss: 16.7365\n",
      "Epoch 1849/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.0595 - val_loss: 16.6685\n",
      "Epoch 1850/2000\n",
      "1/1 [==============================] - 0s 21ms/step - loss: 13.0004 - val_loss: 16.6186\n",
      "Epoch 1851/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.0839 - val_loss: 16.5718\n",
      "Epoch 1852/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 13.0075 - val_loss: 16.7166\n",
      "Epoch 1853/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 13.0713 - val_loss: 16.6126\n",
      "Epoch 1854/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 13.0313 - val_loss: 16.5900\n",
      "Epoch 1855/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.9990 - val_loss: 16.7127\n",
      "Epoch 1856/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.0540 - val_loss: 16.6327\n",
      "Epoch 1857/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.9869 - val_loss: 16.6067\n",
      "Epoch 1858/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.9959 - val_loss: 16.5926\n",
      "Epoch 1859/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.9897 - val_loss: 16.6844\n",
      "Epoch 1860/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 13.0477 - val_loss: 16.5621\n",
      "Epoch 1861/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.9760 - val_loss: 16.5448\n",
      "Epoch 1862/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.9945 - val_loss: 16.6325\n",
      "Epoch 1863/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.9819 - val_loss: 16.6350\n",
      "Epoch 1864/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.9838 - val_loss: 16.5580\n",
      "Epoch 1865/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.9654 - val_loss: 16.5584\n",
      "Epoch 1866/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.9604 - val_loss: 16.5852\n",
      "Epoch 1867/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 12.9603 - val_loss: 16.5978\n",
      "Epoch 1868/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.9398 - val_loss: 16.5466\n",
      "Epoch 1869/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.9595 - val_loss: 16.5397\n",
      "Epoch 1870/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.9684 - val_loss: 16.5259\n",
      "Epoch 1871/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 12.9612 - val_loss: 16.5395\n",
      "Epoch 1872/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 12.9197 - val_loss: 16.5765\n",
      "Epoch 1873/2000\n",
      "1/1 [==============================] - 0s 22ms/step - loss: 12.9353 - val_loss: 16.5224\n",
      "Epoch 1874/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 12.9532 - val_loss: 16.5657\n",
      "Epoch 1875/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.9609 - val_loss: 16.6009\n",
      "Epoch 1876/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.9505 - val_loss: 16.5058\n",
      "Epoch 1877/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.9654 - val_loss: 16.5373\n",
      "Epoch 1878/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 12.9612 - val_loss: 16.5052\n",
      "Epoch 1879/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 12.9264 - val_loss: 16.5311\n",
      "Epoch 1880/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 12.9344 - val_loss: 16.5921\n",
      "Epoch 1881/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.9662 - val_loss: 16.5990\n",
      "Epoch 1882/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.9435 - val_loss: 16.6069\n",
      "Epoch 1883/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.9303 - val_loss: 16.5213\n",
      "Epoch 1884/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.9671 - val_loss: 16.5537\n",
      "Epoch 1885/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.9396 - val_loss: 16.5439\n",
      "Epoch 1886/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.9281 - val_loss: 16.5026\n",
      "Epoch 1887/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 12.9693 - val_loss: 16.5293\n",
      "Epoch 1888/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.9299 - val_loss: 16.5693\n",
      "Epoch 1889/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.9311 - val_loss: 16.5742\n",
      "Epoch 1890/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.9182 - val_loss: 16.5667\n",
      "Epoch 1891/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.9301 - val_loss: 16.6302\n",
      "Epoch 1892/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.9389 - val_loss: 16.5701\n",
      "Epoch 1893/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.9102 - val_loss: 16.4928\n",
      "Epoch 1894/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 12.9320 - val_loss: 16.4583\n",
      "Epoch 1895/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 12.9107 - val_loss: 16.5222\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1896/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 12.9613 - val_loss: 16.4611\n",
      "Epoch 1897/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.8783 - val_loss: 16.4684\n",
      "Epoch 1898/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.9375 - val_loss: 16.5893\n",
      "Epoch 1899/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.9424 - val_loss: 16.5715\n",
      "Epoch 1900/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.9008 - val_loss: 16.5386\n",
      "Epoch 1901/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.9114 - val_loss: 16.4491\n",
      "Epoch 1902/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 12.9020 - val_loss: 16.5506\n",
      "Epoch 1903/2000\n",
      "1/1 [==============================] - ETA: 0s - loss: 12.97 - 0s 18ms/step - loss: 12.9784 - val_loss: 16.4306\n",
      "Epoch 1904/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 12.9159 - val_loss: 16.4061\n",
      "Epoch 1905/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.9000 - val_loss: 16.5884\n",
      "Epoch 1906/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.9653 - val_loss: 16.4782\n",
      "Epoch 1907/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.8788 - val_loss: 16.4825\n",
      "Epoch 1908/2000\n",
      "1/1 [==============================] - 0s 25ms/step - loss: 12.8643 - val_loss: 16.6535\n",
      "Epoch 1909/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.9142 - val_loss: 16.4820\n",
      "Epoch 1910/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 12.8802 - val_loss: 16.4128\n",
      "Epoch 1911/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.8638 - val_loss: 16.5379\n",
      "Epoch 1912/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.9174 - val_loss: 16.3716\n",
      "Epoch 1913/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 12.8555 - val_loss: 16.4051\n",
      "Epoch 1914/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.8517 - val_loss: 16.5779\n",
      "Epoch 1915/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.8707 - val_loss: 16.4499\n",
      "Epoch 1916/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.8297 - val_loss: 16.4533\n",
      "Epoch 1917/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.8471 - val_loss: 16.5381\n",
      "Epoch 1918/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.8408 - val_loss: 16.5299\n",
      "Epoch 1919/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.8550 - val_loss: 16.4205\n",
      "Epoch 1920/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.8434 - val_loss: 16.3954\n",
      "Epoch 1921/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.8280 - val_loss: 16.4713\n",
      "Epoch 1922/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 12.8778 - val_loss: 16.4217\n",
      "Epoch 1923/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.8213 - val_loss: 16.4291\n",
      "Epoch 1924/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.8393 - val_loss: 16.5614\n",
      "Epoch 1925/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.8485 - val_loss: 16.4987\n",
      "Epoch 1926/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.8235 - val_loss: 16.4944\n",
      "Epoch 1927/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.8342 - val_loss: 16.4728\n",
      "Epoch 1928/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.8233 - val_loss: 16.4557\n",
      "Epoch 1929/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.8260 - val_loss: 16.4347\n",
      "Epoch 1930/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.7949 - val_loss: 16.4967\n",
      "Epoch 1931/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.8232 - val_loss: 16.5347\n",
      "Epoch 1932/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.8265 - val_loss: 16.4651\n",
      "Epoch 1933/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7889 - val_loss: 16.4831\n",
      "Epoch 1934/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.8172 - val_loss: 16.4002\n",
      "Epoch 1935/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.7920 - val_loss: 16.3763\n",
      "Epoch 1936/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.8014 - val_loss: 16.4557\n",
      "Epoch 1937/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.8057 - val_loss: 16.3794\n",
      "Epoch 1938/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7833 - val_loss: 16.4340\n",
      "Epoch 1939/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.7825 - val_loss: 16.4461\n",
      "Epoch 1940/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.8511 - val_loss: 16.3792\n",
      "Epoch 1941/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7903 - val_loss: 16.3996\n",
      "Epoch 1942/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.8015 - val_loss: 16.4595\n",
      "Epoch 1943/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 12.8270 - val_loss: 16.3643\n",
      "Epoch 1944/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7715 - val_loss: 16.4131\n",
      "Epoch 1945/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.7885 - val_loss: 16.4188\n",
      "Epoch 1946/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.7837 - val_loss: 16.3850\n",
      "Epoch 1947/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7758 - val_loss: 16.3764\n",
      "Epoch 1948/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.7706 - val_loss: 16.3995\n",
      "Epoch 1949/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.7616 - val_loss: 16.4066\n",
      "Epoch 1950/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.7938 - val_loss: 16.3638\n",
      "Epoch 1951/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.8357 - val_loss: 16.3902\n",
      "Epoch 1952/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7646 - val_loss: 16.3632\n",
      "Epoch 1953/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 12.7610 - val_loss: 16.3589\n",
      "Epoch 1954/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.7746 - val_loss: 16.3701\n",
      "Epoch 1955/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 12.7656 - val_loss: 16.3041\n",
      "Epoch 1956/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7682 - val_loss: 16.3168\n",
      "Epoch 1957/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 12.7439 - val_loss: 16.3553\n",
      "Epoch 1958/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7588 - val_loss: 16.3126\n",
      "Epoch 1959/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.7511 - val_loss: 16.3311\n",
      "Epoch 1960/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7555 - val_loss: 16.3666\n",
      "Epoch 1961/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.7642 - val_loss: 16.3010\n",
      "Epoch 1962/2000\n",
      "1/1 [==============================] - 0s 21ms/step - loss: 12.7506 - val_loss: 16.3366\n",
      "Epoch 1963/2000\n",
      "1/1 [==============================] - 0s 20ms/step - loss: 12.7609 - val_loss: 16.3770\n",
      "Epoch 1964/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7682 - val_loss: 16.3419\n",
      "Epoch 1965/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.7419 - val_loss: 16.3383\n",
      "Epoch 1966/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7506 - val_loss: 16.3630\n",
      "Epoch 1967/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7852 - val_loss: 16.3606\n",
      "Epoch 1968/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.7374 - val_loss: 16.3645\n",
      "Epoch 1969/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7714 - val_loss: 16.4895\n",
      "Epoch 1970/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.8116 - val_loss: 16.4095\n",
      "Epoch 1971/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.8139 - val_loss: 16.3608\n",
      "Epoch 1972/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7224 - val_loss: 16.3626\n",
      "Epoch 1973/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7554 - val_loss: 16.3224\n",
      "Epoch 1974/2000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 16ms/step - loss: 12.7471 - val_loss: 16.2959\n",
      "Epoch 1975/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 12.7353 - val_loss: 16.3582\n",
      "Epoch 1976/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7573 - val_loss: 16.2994\n",
      "Epoch 1977/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.7152 - val_loss: 16.3395\n",
      "Epoch 1978/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.7083 - val_loss: 16.4983\n",
      "Epoch 1979/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.7959 - val_loss: 16.3775\n",
      "Epoch 1980/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7528 - val_loss: 16.3293\n",
      "Epoch 1981/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.6967 - val_loss: 16.3800\n",
      "Epoch 1982/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7207 - val_loss: 16.2831\n",
      "Epoch 1983/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 12.6879 - val_loss: 16.2781\n",
      "Epoch 1984/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.7278 - val_loss: 16.2959\n",
      "Epoch 1985/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.6972 - val_loss: 16.3508\n",
      "Epoch 1986/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7014 - val_loss: 16.3448\n",
      "Epoch 1987/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7356 - val_loss: 16.3311\n",
      "Epoch 1988/2000\n",
      "1/1 [==============================] - 0s 15ms/step - loss: 12.6978 - val_loss: 16.3845\n",
      "Epoch 1989/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7229 - val_loss: 16.3350\n",
      "Epoch 1990/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.6890 - val_loss: 16.2478\n",
      "Epoch 1991/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 12.7510 - val_loss: 16.4010\n",
      "Epoch 1992/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.7214 - val_loss: 16.2285\n",
      "Epoch 1993/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.6842 - val_loss: 16.2943\n",
      "Epoch 1994/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.7049 - val_loss: 16.3796\n",
      "Epoch 1995/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.6851 - val_loss: 16.3776\n",
      "Epoch 1996/2000\n",
      "1/1 [==============================] - 0s 17ms/step - loss: 12.6746 - val_loss: 16.3638\n",
      "Epoch 1997/2000\n",
      "1/1 [==============================] - 0s 18ms/step - loss: 12.7092 - val_loss: 16.3923\n",
      "Epoch 1998/2000\n",
      "1/1 [==============================] - 0s 19ms/step - loss: 12.6609 - val_loss: 16.4278\n",
      "Epoch 1999/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.6736 - val_loss: 16.3089\n",
      "Epoch 2000/2000\n",
      "1/1 [==============================] - 0s 16ms/step - loss: 12.6908 - val_loss: 16.2880\n"
     ]
    }
   ],
   "source": [
    "# 테스트를 위해 학습과 검증으로 나눠서 한다.\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, \n",
    "                                        test_size=0.3, random_state=1)\n",
    "\n",
    "history = model.fit(X_train, y_train, validation_data=(X_test, y_test), \n",
    "                    epochs=2000, batch_size=500, \n",
    "                    callbacks=[call1, call2, call3])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIwAAAJGCAYAAAA59R5HAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAABVC0lEQVR4nO39ebxsaV0f/n6eGvZ0xh5paIZuiDL4Y9IGwcSBcEkE1ABGgz+R4CXyM0ZvQElEuD9RiUBu0OSqIUBMvBrD78UPQRKBiEpU6GCURtEwtYA2U8/dp/uMe6p67h+rak9nn3N2d59TtfbZ7/frtV5VtZ61Vj1Va1ftsz/n+zyr1FoDAAAAAGOdaXcAAAAAgHYRGAEAAACwicAIAAAAgE0ERgAAAABsIjACAAAAYBOBEQAAAACb9KbdgZ24/PLL6zXXXDPtbgAAAABcND72sY/dWWu9Yru2XREYXXPNNbnhhhum3Q0AAACAi0Yp5QtnajMkDQAAAIBNBEYAAAAAbCIwAgAAAGATgREAAAAAmwiMAAAAANhEYAQAAADAJr1pdwAAAABor6NHj+b222/PysrKtLvCDvX7/Vx55ZU5ePDg/T6GwAgAAADY1tGjR3Pbbbfl6quvzvz8fEop0+4S51BrzalTp/KVr3wlSe53aGRIGgAAALCt22+/PVdffXUWFhaERbtEKSULCwu5+uqrc/vtt9/v4wiMAAAAgG2trKxkfn5+2t3gfpifn39AwwgFRgAAAMAZqSzanR7oeRMYAQAAALCJwAgAAACATQRGAAAAwJ7xkpe8JC960Yum3Y3WExgBAAAAsInACAAAAIBNBEYAAAAAbCIwAgAAAPas97///Xn605+ehYWFHD58ON/1Xd+VL37xi5u2edvb3pbHPvaxmZuby7XXXpsPfvCDO2rbzXrT7gAAAADANLzrXe/KC1/4wvzET/xE3vzmN+euu+7Ka1/72nzzN39z/vzP/zwHDx7MO9/5zrzqVa/Kr/3ar+WRj3xkPv7xj6fTaepvzta22wmMAAAAgPvkp3/rk/nUzUen2ofHPeRgXvvtX3O/96+15pWvfGVe+cpX5md+5mfW1j/taU/LNddck1/+5V/Oj/7oj+YP//AP803f9E35tm/7tuZ5H/e4tW3P1rbbXRyxFwAAAMB98LnPfS433XRTXvKSl2xav3///jz3uc/N9ddfnyR5+tOfng984AP5pV/6pSwuLm7a9mxtu50KIwAAAOA+eSCVPW1x++23J0muvvrq09oe/OAHr81j9L3f+71ZXFzMa1/72rzuda/Lj//4j+flL395Op3OWdt2u93/CnaJv7ztWP72z/1BPvL5O6fdFQAAANjzDh8+nCS55ZZbTmu77bbbcsUVV6w9fulLX5q/+qu/yutf//r81E/9VN70pjftqG03ExhNyPLqMH91x4mcWBpMuysAAACw5z3mMY/JVVddlV/91V/dtP7kyZN5//vfn7/7d//upvUzMzN56Utfmpe+9KX50Ic+tOO23cqQtAnplJIkGQzrlHsCAAAAdLvdvOENb8gP/MAPpNfr5XnPe16OHDmSn/7pn87DHvawfN/3fV+S5HWve10e+tCH5ilPeUruueeefOADH8iLX/zic7btdgKjCel2msCoVoERAAAAtMFLXvKSzM/P5/Wvf33e8IY35PDhw3n+85+fN77xjZmZmUmSPPzhD8/P/MzP5Ctf+Uoe/OAH5x/+w3+Yf/7P//k523a7shsCjOuuu67ecMMN0+7GA/LZ247lWf/6Q/ml//3J+bYnPGTa3QEAAIBz+vSnP53HPvax0+4G99O5zl8p5WO11uu2azOH0YR0RhVGRqQBAAAAbScwmpDxHEZDiREAAADQcgKjCemOA6NdMAQQAAAA2NsERhMyyotcJQ0AAABoPYHRhKxfJW3KHQEAAAA4B4HRhIznMBpIjAAAAICWExhNSGf0TpvDCAAAAGg7gdGEuEoaAAAAsFsIjCZk/SppU+4IAAAAwDkIjCZkbQ4jiREAAADQcgKjCTGHEQAAALBbCIwmZG0OI4ERAAAA7DrXXHNNfvmXf3lH25ZS8nu/93sXuEcXlsBoQrodcxgBAAAAu4PAaEJGBUbmMAIAAABaT2A0IeOrpFVD0gAAAICWExhNyPpV0qbcEQAAANiDvu3bvi0vetGLTlv/jd/4jXnVq16Vt771rfm6r/u67N+/P494xCPyhje84bw999LSUl7zmtfk2muvzczMTK699tq8/vWvz3C4HhLcddddeelLX5oHPehBWVhYyDd8wzestZ+t7ULpXdCjs6YzmsNooMIIAAAAJu6FL3xhfviHfzjLy8uZmZlJktx88835yEc+kre85S156Utfmh//8R/PE57whPzRH/1R/tE/+kd58pOfnG/91m99wM/93d/93fn4xz+eN73pTXnMYx6TP/7jP86P/diP5a677srP/dzPJUle/OIXZzAY5AMf+ECWl5fzvve9b23/s7VdKAKjCeoUQ9IAAAC4CPy3VyW3/q/p9uGqxyfPfuOON3/e856Xl73sZfngBz+YZz/72UmSd77znXnSk56Ur/mar8mHP/zh9Pv9JMljHvOY/Kf/9J/ywQ9+8AEHRr//+7+f9773vflf/+t/5XGPe1yS5PGPf3z27duXF7/4xfnxH//xXHnllfnDP/zD/Pqv/3qe9KQnJUme+tSnrh3jbG0XiiFpE9TtFJNeAwAAwBTs378/z33uc/Pud797bd073/nOvPjFL06S9Pv9fPazn83b3/72vPa1r80Xv/jF3HrrrQ/4eX/v934vT33qU9fCorHv+q7vSq01//N//s8kydOf/vS89rWvzUc+8pHTjnG2tgtFhdEElVIiLwIAAGDXuw+VPW3yPd/zPfnBH/zBvOUtb8mtt96aj370o3n3u9+du+66Ky94wQvyqU99Kk996lPzVV/1VTl8+PB5GSV0++235+qrrz5tfa/Xy+WXX54jR44kacKrV77ylfnmb/7mPOUpT8nP//zP52lPe9o52y4UFUYT1C0lQ0PSAAAAYCqe85znZGlpKR/+8Ifzzne+M8961rNy5ZVX5o1vfGOWl5fzpS99Ke973/vyb/7Nv8kjH/nI8/Kchw8fzi233HLa+sFgkLvuuitXXHHF2na//Mu/nL/+67/Oox/96DzjGc/IF7/4xXO2XSgCownqlGSoxAgAAACmYm5uLs973vPynve8J+9617vWhqN98pOfzNOe9rTMzc0laa5qNh4q9kA94xnPyB//8R/nM5/5zKb1v/mbv5nZ2dn8zb/5Nzetf+hDH5r/+B//Y+bn5/Mnf/InO2473wRGE9TpFFdJAwAAgCl64QtfmN/5nd/JZz/72XzHd3xHkuSJT3xi3vGOd+SDH/xgbrjhhrzwhS9cmwD7gXr2s5+db/qmb8q3fuu35jd+4zfyiU98Ir/yK7+Sl73sZfmZn/mZHDp0KEnynd/5nfnd3/3d3HjjjfkP/+E/ZHFxMV/3dV93zrYLxRxGk3Ly7nxHPpQDS89M8jXT7g0AAADsSc961rNyxx135PnPf/5aRdGrX/3qfOELX8gLXvCC7N+/P695zWty2WWXZXFx8QE/Xyklv/Vbv5VXv/rV+ZEf+ZHcdddd+eqv/ur83M/9XL7/+79/bbtOp5MXvvCFWVpayhOe8IT81m/9Vq699tpztl0oZTdc5v26666rN9xww7S78cDc/GfJ274l//naN+Z7/+E/nnZvAAAA4Jw+/elP57GPfey0u8H9dK7zV0r5WK31uu3aDEmblNJtbutguv0AAAAA7rcPfehDmZub23b5gR/4gWl377wxJG1SOqPAaCgwAgAAgN3quuuuy8c//vFt28bzEV0MBEaTslZhNJxuPwAAAID7bWFhIY95zGOm3Y0LzpC0SVFhBAAAAOwSAqNJKaO32hxGAAAAQMsJjCZlVGFUVBgBAAAALScwmhRzGAEAALAL1Vqn3QXuhwd63gRGkzKew8iQNAAAAHaJfr+fU6dOTbsb3A+nTp1Kv9+/3/sLjCZlVGFUBEYAAADsEldeeWW+8pWv5OTJkyqNdolaa06ePJmvfOUrufLKK+/3cXrnsU+cTceQNAAAAHaXgwcPJkluvvnmrKysTLk37FS/38+DHvSgtfN3fwiMJmV0lTQVRgAAAOwmBw8efEDBA7uTIWmTsnaVNBVGAAAAQLsJjCalmPQaAAAA2B0ERpMyrjAyhxEAAADQcgKjSRlfJS0qjAAAAIB2ExhNyvgqaUOBEQAAANBuAqNJWbtKmiFpAAAAQLsJjCallAzTSTHpNQAAANByAqMJagIjFUYAAABAuwmMJmhYOia9BgAAAFpPYDRBw3RShiqMAAAAgHYTGE3QMJ2UCIwAAACAdhMYTVAtXZNeAwAAAK0nMJogV0kDAAAAdgOB0QQNSycdQ9IAAACAlhMYTVBNN6UKjAAAAIB2ExhN0LB00jEkDQAAAGg5gdEE1eIqaQAAAED7CYwmqKarwggAAABoPYHRBA1LNx1zGAEAAAAtJzCaoGHppESFEQAAANBuAqMJqqWjwggAAABoPYHRBNV00zHpNQAAANByAqMJUmEEAAAA7AYCowmqpWsOIwAAAKD1BEYTVEvHkDQAAACg9XYcGJVS/p+llBtLKSdLKR8tpTxrS/uLSimfKaUsllL+rJTyjG2O8YpSyk2llFOllOtLKU84Hy9it6ila0gaAAAA0Ho7CoxKKc9P8pYkP5/kqUnem+R9pZSvHbV/e5K3JnlDkicn+YMk7y2lPHLDMX4kyauTvDzJU5J8KckHSimHztNraT0VRgAAAMBusNMKo+9L8uu11rfWWj9Ra/3pJH+U5HtG7a9L8sZa66/WWj9da31FkhuT/FCSlFJ6SX4yyY/WWt9Ta/1EkpdsOPaeUIurpAEAAADtt9PAaJjk1JZ1x5N0SylXJ3likndsaf/NJONhaV+f5HCSd48ba61LSd6/YZuLnqukAQAAALvBTgOjtyZ5USnlmaWUfinlhUm+Mcm/T/K4JCtJPrtlnxuTPGp0/3FJbqq1njjLNhc/FUYAAADALtDbyUa11t8tpfzbJL+XpCYpSV5Wa/10KeVJSY7UWuuW3Y4kOTC6f3mSu7c59JEkB7d7zlLKy5K8LEke/vCH76SbrdfMYTSYdjcAAAAAzmpHgVEp5R8n+f408w59IsnTk/zLUsotSbrJtilIHS3Z4TabG2p9W5K3Jcl111237Ta7TS3ddFNTa00pZdrdAQAAANjWOQOjUsrBJP8qyQtqrb8zWv2xUspSkjcn+ZFsXyV0KE0FUZIcO8s221UeXZTGk14PhjW9rsAIAAAAaKedzGH02CT7kvzxlvV/lORhSe5Msm80+fVGj07yqdH9zyV55OhqaWfa5uJXOulmmOFFUS8FAAAAXKx2EhjdOrr92i3rn5JkOcmfJrklyQu2tD8/yftG968fPdezx42llNkkz92wzUVvXGE0PG26JwAAAID2OOeQtFrrF0op70jyK6WUH0vy6SRPS/KmJG+utZ4qpbwxyetKKbemmePoB5NcleQto2PcW0p5c5I3l1KWk9yc5DVJbkvyrgvwutqp0x1VGAmMAAAAgPba0aTXSV6c5JVJXp/koUluSvLaNHMYJckvprki2i+kmZfo+iTPrLUe3XCMV6W5utrbk8wk+e0kz6m17p3Lho2GpA2MSQMAAABabEeBUa11OU1Y9PoztNckPztaznaMV4yWPamWbjrFHEYAAABAu+1kDiPOlzIakiYxAgAAAFpMYDRJnY45jAAAAIDWExhN0PgqaQOBEQAAANBiAqNJGl0lTV4EAAAAtJnAaJLGFUbmMAIAAABaTGA0SaMKI3MYAQAAAG0mMJqkMpr0ejjtjgAAAACcmcBokjrNkDQVRgAAAECbCYwmqTRD0lwlDQAAAGgzgdEkdbrplppqTBoAAADQYgKjSSrdJMlgIDACAAAA2ktgNEGl0wRGdbgy5Z4AAAAAnJnAaJI64wqjwZQ7AgAAAHBmAqNJKr3mVmAEAAAAtJjAaIJKt3m7h8PVKfcEAAAA4MwERpM0mvR6OFRhBAAAALSXwGiSxpNeD1QYAQAAAO0lMJqgTqeZw6gOh1PuCQAAAMCZCYwmqHZGcxipMAIAAABaTGA0QZ3RHEYxhxEAAADQYgKjSeo2Q9JcJQ0AAABoM4HRBJWOq6QBAAAA7ScwmqBSRm+3wAgAAABoMYHRBJXxkLSBwAgAAABoL4HRBI2HpFVzGAEAAAAtJjCaoPXASIURAAAA0F4CownqjIakCYwAAACANhMYTdDaVdIGhqQBAAAA7SUwmqBOZ1RhVFUYAQAAAO0lMJqgTndUYbQqMAIAAADaS2A0QePASIURAAAA0GYCowkaT3ptDiMAAACgzQRGEzSuMIqrpAEAAAAtJjCaoPGk10OBEQAAANBiAqMJ6vRGcxgJjAAAAIAWExhNUHdUYVQHAiMAAACgvQRGE9Tp9ps71aTXAAAAQHsJjCao01NhBAAAALSfwGiCuuMKo6EKIwAAAKC9BEYT1Ok2FUYCIwAAAKDNBEaT1GmukiYwAgAAANpMYDRJo6ukpZrDCAAAAGgvgdEkdUx6DQAAALSfwGiSRoFRqYakAQAAAO0lMJokcxgBAAAAu4DAaJLGcxgNDUkDAAAA2ktgNEnjIWkqjAAAAIAWExhNUmmGpBUVRgAAAECLCYwmaTyHkUmvAQAAgBYTGE1SKRmkk1QVRgAAAEB7CYwmbJCuIWkAAABAqwmMJmyQboohaQAAAECLCYwmTIURAAAA0HYCowkblG6KOYwAAACAFhMYTdgwHUPSAAAAgFYTGE3YsHTTUWEEAAAAtJjAaMKG6SbmMAIAAABaTGA0YYPSTScCIwAAAKC9BEYTNizddFQYAQAAAC0mMJqwYVwlDQAAAGg3gdGEDQ1JAwAAAFpOYDRhtXTTqavT7gYAAADAGQmMJmxYOukYkgYAAAC0mMBowmrpptThtLsBAAAAcEYCowkbll665jACAAAAWkxgNGHD0jUkDQAAAGg1gdGEVYERAAAA0HICowmrpZuOIWkAAABAiwmMJmxYeumqMAIAAABaTGA0aZ2OSa8BAACAVhMYTVgtPUPSAAAAgFYTGE1YLb106nDa3QAAAAA4I4HRhNVON90IjAAAAID2EhhNWC1dcxgBAAAArSYwmrDa6QmMAAAAgFYTGE3aKDCqtU67JwAAAADbEhhNWumkl2GG8iIAAACgpQRGkzaqMBpIjAAAAICWEhhNWC29dDMUGAEAAACtJTCatE43vQwzMIcRAAAA0FICo0kbD0kbCIwAAACAdhIYTVqnl14ZZjAcTrsnAAAAANsSGE1at5ckWR2sTLkjAAAAANsTGE1YKd0kyXB1dco9AQAAANiewGjSOk1gNFhVYQQAAAC0k8Bo0jrNkLThYDDljgAAAABsT2A0YWU0h9FgsDzlngAAAABsT2A0aaMKozowhxEAAADQTgKjCVurMDKHEQAAANBSAqNJW5vDSIURAAAA0E4CowkrAiMAAACg5QRGE1Y63STJcGBIGgAAANBOAqMJG89hNBwMptwTAAAAgO0JjCasdPtJVBgBAAAA7bXjwKg0Xl5K+UwpZamUcksp5fkb2l80alsspfxZKeUZ2xzjFaWUm0opp0op15dSnnC+Xshu0RlVGNVVcxgBAAAA7XRfKox+Kck/TvITSZ6Q5B8k+eskKaV8e5K3JnlDkicn+YMk7y2lPHK8cynlR5K8OsnLkzwlyZeSfKCUcuiBvojdZG1I2lBgBAAAALRTbycblVK+IcmLkjy61nrraPWNGzZ5XZI31lp/dfT4FaWUb07yQ0leWUrpJfnJJD9aa33P6JgvSXJTku9LE0btCePAqLpKGgAAANBSO60w+qEk/35DWLSmlHJ1kicmeceWpt9MMh6W9vVJDid597ix1rqU5P0bttkTOh2BEQAAANBuOw2MnpnkhlLKr5VS7iylfKGU8tpSSifJ45KsJPnsln1uTPKo0f3HJbmp1nriLNvsCWsVRkOTXgMAAADtdM4haaWUg0muSvLPk/yXJM9O8rVJ3pTkeJKbkxyptdYtux5JcmB0//Ikd29z+CNJDt6vnu9SHUPSAAAAgJbbyRxG40DnD2utPz26/9FSyv4kr0zyz5IMttmvjpYk6e5gm01KKS9L8rIkefjDH76Dbu4O48BoONju7QAAAACYvp0MSVse3X5gy/rfT1N51M32VUKH0lQQJcmxs2yzXeVRaq1vq7VeV2u97oorrthBN3eHbm8mSVJdJQ0AAABoqZ0ERnckOZHTA5+aZJjkhiT7RpNfb/ToJJ8a3f9ckkeOrpZ2pm32hO5ahZHACAAAAGincwZGo7mJ/iDJd25penaST6YJfG5J8oIt7c9P8r7R/etHz/XscWMpZTbJczdssyd0euM5jJbPsSUAAADAdOxkDqMkeWOSPyil/GyS/zvJ1yd5TZIX1lprKeWNSV5XSrk1ySeS/GCa4WpvSZJa672llDcneXMpZTnNRNmvSXJbknedzxfUdt1eP4lJrwEAAID22lFgVGu9vpTy95P8izQTXX8hyQ/WWn9rtMkvprki2i+kmZfo+iTPrLUe3XCYVyUpSd6eZCbJbyd5Tq11T83+vD6H0Z562QAAAMAustMKo9Ra35PkPWdoq0l+drScaf/lJK8YLXtWZ63CyJA0AAAAoJ12Muk151Gv3wRGMSQNAAAAaCmB0YR1u6PAaCgwAgAAANpJYDRhvX4zh1EGK9PtCAAAAMAZCIwmbDzptQojAAAAoK0ERpPWaeYZrwIjAAAAoKUERpPWaeYwKgIjAAAAoKUERpPW6WRYizmMAAAAgNYSGE3BaukmVYURAAAA0E4CoykYpGtIGgAAANBaAqMpWE03ZWhIGgAAANBOAqMpWE0vGQ6m3Q0AAACAbQmMpmBYuukYkgYAAAC0lMBoClbTTTHpNQAAANBSAqMpGJr0GgAAAGgxgdEUDEovpZrDCAAAAGgngdEUmMMIAAAAaDOB0RQMSi8dcxgBAAAALSUwmoJhMek1AAAA0F4CoykYppeuwAgAAABoKYHRFAw6Jr0GAAAA2ktgNAW1dFUYAQAAAK0lMJqCYemlo8IIAAAAaCmB0RQ0FUYCIwAAAKCdBEZTMOz00o0haQAAAEA7CYymoBqSBgAAALSYwGgKhkWFEQAAANBeAqMpqJ2eOYwAAACA1hIYTUHt9NKNwAgAAABoJ4HRFAiMAAAAgDYTGE1D6aZnSBoAAADQUgKjKajdfnomvQYAAABaSmA0DaWXbobT7gUAAADAtgRGU1A7vfSymuGwTrsrAAAAAKcRGE1Dt59ehhlUgREAAADQPgKjaej00ik1qyvmMQIAAADaR2A0BaXbS5Ksri5PuScAAAAApxMYTUOnCYwGAiMAAACghQRG09DpJ0lWV1am3BEAAACA0wmMpmA8JG2wKjACAAAA2kdgNA3dpsLIkDQAAACgjQRGU1DWAiMVRgAAAED7CIymwJA0AAAAoM0ERlMwrjAaDgRGAAAAQPsIjKagdEYVRq6SBgAAALSQwGgKOr2mwqia9BoAAABoIYHRFKxNem1IGgAAANBCAqMp6I7nMDLpNQAAANBCAqMpGFcY1cHqlHsCAAAAcDqB0RR0eqNJrw1JAwAAAFpIYDQFne5MkqQKjAAAAIAWEhhNwfgqaeYwAgAAANpIYDQF48DIHEYAAABAGwmMpmDtKmmD5Sn3BAAAAOB0AqMp6PZHFUaGpAEAAAAtJDCagm5/NklShwIjAAAAoH0ERlPQ642ukrZqSBoAAADQPgKjKVirMBqoMAIAAADaR2A0Bd2ZpsIoAiMAAACghQRGUzCzVmFkSBoAAADQPgKjKej1VRgBAAAA7SUwmoLuaNLrIjACAAAAWkhgNAWl08ly7SZDgREAAADQPgKjKVlNL8UcRgAAAEALCYymZLX0kuHqtLsBAAAAcBqB0ZSspptiSBoAAADQQgKjKVlNT2AEAAAAtJLAaEpWi8AIAAAAaCeB0ZSsppeOwAgAAABoIYHRlAxKLx2TXgMAAAAtJDCakkHpqzACAAAAWklgNCWD0kunqjACAAAA2kdgNCVNYKTCCAAAAGgfgdGUNEPSVBgBAAAA7SMwmpJh6aZrSBoAAADQQgKjKRl2+gIjAAAAoJUERlMyLP10zWEEAAAAtJDAaEqGnV56UWEEAAAAtI/AaEqqIWkAAABASwmMpmRY+ukJjAAAAIAWEhhNSe320zUkDQAAAGghgdGU1E4vfYERAAAA0EICoympnZn0Mph2NwAAAABOIzCalk7PHEYAAABAKwmMpqR2Z9Ivg6TWaXcFAAAAYBOB0bR0e0mSOlieckcAAAAANhMYTUt3JkkyWFmackcAAAAANhMYTUu3nyRZXVmZckcAAAAANhMYTUunqTBaXVmcckcAAAAANhMYTYsKIwAAAKClBEZT0umNAqNlFUYAAABAuwiMpqSMJ71eVWEEAAAAtIvAaErKaEiaq6QBAAAAbSMwmpLSU2EEAAAAtJPAaEo6a4GRCiMAAACgXQRGU7IWGC0vT7knAAAAAJvdp8ColNIppXyylHL9lvUvKqV8ppSyWEr5s1LKM7bZ9xWllJtKKadKKdeXUp7wQDu/m42HpA0HAiMAAACgXe5rhdF3J3ncxhWllG9P8tYkb0jy5CR/kOS9pZRHbtjmR5K8OsnLkzwlyZeSfKCUcuj+dny3644DoxWBEQAAANAuOw6MSinzSV6f5He3NL0uyRtrrb9aa/10rfUVSW5M8kOj/XpJfjLJj9Za31Nr/USSl4z2/b4H2P9da63CaFVgBAAAALTLfakwek2S/5HkI+MVpZSrkzwxyTu2bPubScbD0r4+yeEk7x431lqXkrx/wzZ7TrfXT2JIGgAAANA+OwqMSilfm+T/SPLjW5oel2QlyWe3rL8xyaM2bHNTrfXEWbbZc3r92STJcHVlyj0BAAAA2OycgdFoKNp/TvJjtdabtzRfnuRIrbVuWX8kyYEN29y9zaGPJDl4lud9WSnlhlLKDXfccce5urnrjK+SVleXptwTAAAAgM12UmH0b5N8stb6a9u0dZMMtllfR8tOtzm9sda31Vqvq7Ved8UVV+ygm7tLrz8OjAxJAwAAANqld7bGUso/STPP0NeeYZNj2b5K6FCaCqJzbbNd5dGe0BkNSasDQ9IAAACAdjlrYJTklUmuSXJ3KWVTQymlJvnhJPtKKVfXWr+yofnRST41uv+5JI8spfRqratn2GbPWaswEhgBAAAALXOuIWnPTfLkLctbk3x8dP/XktyS5AVb9nt+kveN7l8/ep5njxtLKbOjY78ve1R/dlxhZEgaAAAA0C5nrTCqtZ5WAVRKuTXJiVrrx0eP35jkdaP1n0jyg0muSvKW0THuLaW8OcmbSynLSW5O8poktyV51/l7KbvLzHhImjmMAAAAgJY515C0nfjFNFdE+4U08xJdn+SZtdajG7Z5VZKS5O1JZpL8dpLn1Fq3mwx7T+iPAqMYkgYAAAC0zH0OjGqtP5XkpzY8rkl+drScaZ/lJK8YLSTp97pZqV2BEQAAANA655rDiAuk3y1ZSc+k1wAAAEDrCIympJSS1XSToTmMAAAAgHYRGE3RSnopKowAAACAlhEYTdGgdJOhwAgAAABoF4HRFK2kr8IIAAAAaB2B0RStll7KcHXa3QAAAADYRGA0RcP00jEkDQAAAGgZgdEUrZZeShUYAQAAAO0iMJqiYemlY0gaAAAA0DICoylaLf0UQ9IAAACAlhEYTVHt9NI1JA0AAABoGYHRFA1KL91qSBoAAADQLgKjKRqWfjoCIwAAAKBlBEZTNOz00zWHEQAAANAyAqMpGnZ66WYw7W4AAAAAbCIwmqLa6adXl6fdDQAAAIBNBEZTVDsz6ZnDCAAAAGgZgdEU1W4//QiMAAAAgHYRGE1R7c6kH5NeAwAAAO0iMJqiYWcmM1VgBAAAALSLwGiaurPpl0EyHE67JwAAAABrBEZTVHszze3AldIAAACA9hAYTVHpNoHRyvLilHsCAAAAsE5gNE2jCqNlgREAAADQIgKjaerOJklWlwRGAAAAQHsIjKap1wRGAxVGAAAAQIsIjKao028Co5UVgREAAADQHgKjKep0xxVGS1PuCQAAAMA6gdEUlVGF0UCFEQAAANAiAqMp6ozmMFpVYQQAAAC0iMBoijozTWA0XBEYAQAAAO0hMJqi7vgqaasCIwAAAKA9BEZTNL5KmgojAAAAoE0ERlPUXQuMTHoNAAAAtIfAaIq6/bkkyXB1eco9AQAAAFgnMJqi3mwTGFVzGAEAAAAtIjCaov5oSJrACAAAAGgTgdEU9WbnkyTVpNcAAABAiwiMpqg/MxqSNjCHEQAAANAeAqMpmpkxJA0AAABoH4HRFM32u1mqfYERAAAA0CoCoyma6XaynF7KqiFpAAAAQHsIjKao0ylZSS8ZCowAAACA9hAYTdly+okKIwAAAKBFBEZTtlL66agwAgAAAFpEYDRlg9JLGQiMAAAAgPYQGE2ZCiMAAACgbQRGU7aamXSGK9PuBgAAAMAagdGUDToqjAAAAIB2ERhN2aD00xUYAQAAAC0iMJqyYWcmXUPSAAAAgBYRGE3ZoNNPtwqMAAAAgPYQGE3ZsDOTnsAIAAAAaBGB0ZQNuwIjAAAAoF0ERlM27M4KjAAAAIBWERhNWe3OZjZL0+4GAAAAwBqB0ZTV7mxmosIIAAAAaA+B0ZQNe3OZzUpS67S7AgAAAJBEYDR93dnmdnVxuv0AAAAAGBEYTVtvLkmyunRqyh0BAAAAaAiMpq3fBEbLSyen3BEAAACAhsBoysqowmhlUWAEAAAAtIPAaMrKqMJoZdmQNAAAAKAdBEZT1hkHRosCIwAAAKAdBEZTVmbmkyQDFUYAAABASwiMpqzTbwKjFZNeAwAAAC0hMJqy3kwzJG2wsjjlngAAAAA0BEZT1h0NSRsakgYAAAC0hMBoynqz5jACAAAA2kVgNGW9mYUkSTUkDQAAAGgJgdGU9eaawGgoMAIAAABaQmA0ZTOjIWl1xZA0AAAAoB0ERlPWn22ukmZIGgAAANAWAqMpm5uZzUrtpq4KjAAAAIB2EBhN2fxMN4uZSVQYAQAAAC0hMJqy2V4nS+knKowAAACAlhAYTVkpJcvpJ4OlaXcFAAAAIInAqBWWy0zKqsAIAAAAaAeBUQuslJl0DEkDAAAAWkJg1AIrZSadoQojAAAAoB0ERi2wWmbTNYcRAAAA0BICoxZY7cymq8IIAAAAaAmBUQsMujPpDZen3Q0AAACAJAKjVhh2ZtOrAiMAAACgHQRGLTDozqZvSBoAAADQEgKjFhh25zKjwggAAABoCYFRCwx785nL4rS7AQAAAJBEYNQKtb+Q+Swlw+G0uwIAAAAgMGqDYX+hubOqyggAAACYPoFRG/SawGiwdGLKHQEAAAAQGLVCmZlPkiydPDblngAAAAAIjFqhzO5LkiyfEhgBAAAA0ycwaoFOvwmMlk4ZkgYAAABMn8CoBTpz+5Mkq4vHp9wTAAAAAIFRK3Rnm0mvVwRGAAAAQAvsKDAqpTyklPKfSil3llLuLaV8sJTypC3bvKiU8plSymIp5c9KKc/Y5jivKKXcVEo5VUq5vpTyhPP0Ona13mgOo+GSwAgAAACYvp1WGP1ikqNJnp3kmUmOJPndUsqVSVJK+fYkb03yhiRPTvIHSd5bSnnk+ACllB9J8uokL0/ylCRfSvKBUsqh8/FCdrP+/IEkyeriySn3BAAAAGDngdGra63/pNb60VrrDUm+N8kwybeP2l+X5I211l+ttX661vqKJDcm+aEkKaX0kvxkkh+ttb6n1vqJJC8Z7ft95+m17Fr90RxGwyWTXgMAAADTt6PAqNZ645bHS0m+kOTKUsrVSZ6Y5B1bdvvNJONhaV+f5HCSd285xvs3bLNnzcw3Q9LqisAIAAAAmL77Nel1KWUhyVcn+VSSxyVZSfLZLZvdmORRo/uPS3JTrXVrIrJxmz1rdn48h5HACAAAAJi++3uVtH+Z5JYk70tyeZIjtda6ZZsjSQ6M7l+e5O5tjnMkycHtnqCU8rJSyg2llBvuuOOO+9nN3WFupp+TdTZZOTXtrgAAAADct8ColDJTSnlbkr+X5O/VWleTdJMMttm8jpbscJvNDbW+rdZ6Xa31uiuuuOK+dHPX2TfbzcnMJssmvQYAAACmr7fTDUspD0vyG0mWkzy11nrrqOlYtq8SOpSmguhc22xXebSnzPe7uTsziTmMAAAAgBbYUYVRKeXhST6S5ENJvmVDWJQkn0uybzT59UaPTjPH0XibR46ulnambfasUkoWM5fOqiFpAAAAwPTtdEja25L8l1rrP6u1bh1a9qk08xm9YMv656eZ4yhJrh8917PHjaWU2STP3bDNnrbUERgBAAAA7XDOIWmllH1JnpXkbaWUa7Y0L9VabymlvDHJ60optyb5RJIfTHJVkrckSa313lLKm5O8uZSynOTmJK9JcluSd52vF7ObLXfmMiMwAgAAAFpgJ3MYXZamOmi7YOdjSa5L8otproj2C2nmJbo+yTNrrUc3bPuqJCXJ25PMJPntJM/ZpmJpT1rtzKU/OHruDQEAAAAusHMGRrXWL6YJes62TU3ys6PlTNssJ3nFaGGL1e58+qu3TbsbAAAAADuew4gLbLU7n/5wcdrdAAAAABAYtcWwt5BZgREAAADQAgKjlhj092dfTia1TrsrAAAAwB4nMGqJ4ez+9DJIVlUZAQAAANMlMGqLmYNJkrroSmkAAADAdAmM2mKuCYyWT9475Y4AAAAAe53AqCU6cweSJKeO3TPdjgAAAAB7nsCoJTpzh5IkyydUGAEAAADTJTBqif5CMyRt6eQ90+0IAAAAsOcJjFqiv9BUGK2awwgAAACYMoFRS8zuHwVGp1wlDQAAAJgugVFLzO47nCSpiwIjAAAAYLoERi2xML8vS7WX4eKxaXcFAAAA2OMERi2xb7ab45lPlgRGAAAAwHQJjFriwFw/x+t8smRIGgAAADBdAqOWmOl1cqIspLN8fNpdAQAAAPY4gVGLLHYW0l0RGAEAAADTJTBqkaXuvvRXBUYAAADAdAmMWmSlty8zgxPT7gYAAACwxwmMWmS1tz9zAiMAAABgygRGLTKYPZT99XhS67S7AgAAAOxhAqMWGcxekl4GydKxaXcFAAAA2MMERm0yf0mSpJ68a8odAQAAAPYygVGLlIVLkySnjt455Z4AAAAAe5nAqEW6+y5Lkpy6V2AEAAAATI/AqEX6By5PkiwevWPKPQEAAAD2MoFRi8wfaiqMVo6ZwwgAAACYHoFRi8wfvCJJMjghMAIAAACmR2DUIgf3zeVoXUg9eWTaXQEAAAD2MIFRixyc6+dI3Z+cunvaXQEAAAD2MIFRixya7+fe7E93UYURAAAAMD0CoxbpdEqOdw+mt3TPtLsCAAAA7GECo5ZZ7B3K7Mq90+4GAAAAsIcJjFpmZfZw9g0ERgAAAMD0CIxaZmnuQdlXTyRLx6fdFQAAAGCPEhi1zMqBhzR3jn5luh0BAAAA9iyBUcuUQw9Nkizf9YUp9wQAAADYqwRGLdO/9OFJkhN3CIwAAACA6RAYtczCpQ/NoBYVRgAAAMDUCIxa5tKDC7k1l2Z4z5en3RUAAABgjxIYtczl+2dzS70s3WMmvQYAAACmQ2DUMpfvn83N9bLMnrh52l0BAAAA9iiBUcvMz3RzZ/dB2bd0azJYmXZ3AAAAgD1IYNRCd+9/VHp1Nbnrc9PuCgAAALAHCYxa6MThxzR3bv3EdDsCAAAA7EkCoxbqXfnoLNduqsAIAAAAmAKBUQv9jQdfks/Vh2bxK38x7a4AAAAAe5DAqIUec9XBfLo+LOU2FUYAAADA5AmMWuirH3Qgfz78qswt3p7caeJrAAAAYLIERi00P9PNZw9/Q/PgL397up0BAAAA9hyBUUtdevXfyF+VhwuMAAAAgIkTGLXU468+lP+28qTUL3wkuffL0+4OAAAAsIcIjFrq6Y+8LP/X4JmpSfL7r09qnXaXAAAAgD1CYNRS/9vVh7K47+r8zuF/kHz8Pyfv+7Fk8ei0uwUAAADsAQKjlup2Sr79iQ/JP73923Ly6/5xcsN/SP6/T0g+/POCIwAAAOCCEhi12D98+jVZGSb/7Nh357YX/nZOXvl1yQd/OoN/9Tdy56/87zn2F+9NBivT7iYAAABwkSl1F8yNc91119Ubbrhh2t2Yin/7+5/Lv/rAjWuPH1/+Kn+/+4f59u4f5dJyPMe7B1O/5jtz4Cnfmzz0uqSUKfYWAAAA2C1KKR+rtV63bZvAqP0+9oUj+cvbjmWm28ml+2ayb7aXleWlfOmjv5WDf/nu/O3ODZnLSnLo4cljv71ZHvbUpNOddtcBAACAlhIYXcS+dPfJ/NNf/VAeeefv54ev+mQecc8fpwyWk31XNhVHC5clnV6ydCxZOZn055PZg8mVj00e8uTkwU9KejPTfhkAAADAhAmMLnInl1fzz37jL/K+v7gl3/34Q/kX/9stmfnL9yV3/GVy8s6kDpPZA0l/IVk5lZy6Ozl1pNm5N588/OuTa74xufabmhCp25/uCwIAAAAuuLMFRr1Jd4bzb2Gml1/6nifncQ8+mDf9zo351N1X5W3f9+/ykMPz2+9Qa3L05uQrNyQ3/Y/kpuuT//66pm1mf3LV45vKpIVLk/lLkvlLm/tzh5LeXDKzLznw4GaZWZjcCwUAAAAmQoXRReb3PnVbXv6Oj2eu38lPfcfX5LmPf3DKTibCPnFnExzd9OHk9k8nJ+9uKpFO3p0Mz3IltoXLkkMPTRYuT2b3J3OHRyHTJcn86P7c4c33Zw+YnBsAAACmzJC0Peaztx3Ly9/x8Xzy5qN5yjWX5B9/y6PyzV99Zbqd+xHS1Josn2jCo8V7k9XlZPlYcuzW5N4vj5YvNUPclo4li0ebbQfLZz5m6TbVSuNQae7w6PGG+3MHm7mW5g6Nbkf35w438zAJnAAAAOABERjtQYNhzdv/+Av5d3/w+dx872KuPDCb5zz+wfmWR1+Rp1xzafbNXsDRiLU2E2yfOpKcuidZvOfs9xfvHT2+t1k3XD378bsz6+HR/OH1+1tDp+3uzx5MOp0L8rIBAABgNxEY7WErg2F+55O35b/++Vfy+zfekeXVYXqdkic+7HC+4VGX5emPuixf+/BLMtfvTrurjXHYtHg0WTo6ur13FCbduyFcuucM9+9N6uAsT1BG1UqHR0HTlmVtON14SN2l64+FTQAAAFxEBEYkSU4tD/KxLxzJRz5/Zz7y+bvyF1++J8OazPQ6ue4Rl6wFSI998MEszOzS+dBrHQ2NG1Urbaxc2vb+aFk61qxfPn7mY5fO+hxNc4eaOZsWLkv2Xbn+eG0Op8PrQdPc4WaicMPoAAAAaBGBEds6uriSj/713fnI5+/KH33+rnzqlqNrbVcfns8jr9iXv3Hl/jzqitFy5b5csX92Z5No71aDlSZMOnX3aOjcluXkaP3iPc3cTifvSo7f0VRD5SyfpU5/PURauGz9ynPjq9EtXDZav/GqdIeT7i4N7gAAAGg9gRE7cveJ5fzJX9+dz952LJ+/43g+f8eJfP6O4zm5vD7E68BcL4+6Yv+GIGlfrr5kPlccmM1l+2bv38TaF4Px5OAb52gah0yL95weOp28uwmbTt519qvQzR7aME/TofX5mPZd0SwLlzdXnZvdv2Eo3aUmBgcAAOCcBEbcb7XW3HLvYhMg3b4eIn3+juO57ejSpm07Jbls/2yuPDCbKw6s316xfzZXHpzbtG7XDnk732pthsGNw6NTR5KT43Dp7vWKpqWj68PnTt6dnLzz7JODd2c3XIXu0IZJwi9pqpfGwdL8JcnChqDJPE0AAAB7xtkCI3+1c1allDzk8Hwecng+3/hVV2xqO7a4ks/fcSK33nsqdxxbyu3HljbdfuaWY7nj+FIGw9NDyX0z3SZE2j+bKw42odI4XDq00M8lCzM5vNDP4fl+Di30M9tryaTc51spowqhA8kl1+x8v1qb8OjEnU2YtHz8DEPnRiHT8duTO25sKp+W7j1LfzqbK5VOC5gObxM4XZrM7FfRBAAAcBERGHG/HZjr50kPO5w87PAZtxkOa46cXF4LkTYGSnccX8rtRxfz6VuO5kNHl3Js6cwVMwsz3VF4NNPcjpaD870N90fL3Oa2izJsKmU0J9Lh+77vYHV9mNw4WBrP2bT18bFbkts/3Tw+24Tgnf560LSTgMnQOQAAgFYTGHFBdToll+2fzWX7Z/PYB59921PLg9x1Yin3nFzJvadWcuTk8vr9E8u559RK7jm5kntOLufzdxzP0cWmbXFleNbjzvY6a4HSofl+Ds71tjxeD5j2z/azf66X/bO9HBjdLsx0L66Jvru9ZN/lzXJfrC5vqGI6Q8A0fnzPF5Nb/rx5vHrqLH2ZPXvAtHBZ08+Fy5N9lzXzNqlmAgAAuOAERrTG/Ew3D51ZyEMvuW/7La0OcvTU6lqAdO+plRwdL4ura4/HbXccX8rn7ji+ts+5pvHqlIwCpH4OzPVycK4JlQ6sLf3sn+3l4FxvFDb118Km9dt+Znq7fG6g3kxy4EHNcl+snDp9qNxpgdNouevzyZdvaNoHy9sfrzs7CpE2hknbPR6FTHOHBUwAAAD3kcCIXW+2180VB7q54sDsfd53OKw5vryae0+u5OjiSk4sDXJiaTXHllZzbHElxxdXc2xxNceXmnDp2GKz/vZji/n8Hatrj1cG5548fqbXyYFxiDQKkvbP9jcETaPwaa6XA7Oj9nEotaHyadcFT/35Zjn4kJ3vM77q3Mk7kxN3jW7v3HB71/rjuz7fPD7TkLlOrwmTFi5vqpcWLh1VL21zO3eomfh74dKkd99/ngAAAC4WAiP2tE6n5OBcMyzt/qq1Zml1mONLo3BpcTXHllbW7h8fhU/HllY3BVDHF1fz5SMnm/ujfbebIHyrmV4nB+fWK57GgdKBLesOzp2+bnx/rt/yeZ1KSWb3N8tOJwNfObUhVDpDyHTy7mZOpnFlUx2c+Xj7H9QsM/ua24MPGQVLl4yql65ILnlEsu/KpvoKAADgIiIwggeolJK5fjdz/W4u33//q1JqrTm1MlgLk44trgdJzbqVDdVOoxBqXPF0dGnt/onls4QgIzPdzmkh0ub7/VEoNQ6edkHo1J9PDj+sWXZiOGyuMHfq7uTkaEjc4j3rV5+754vJiTuSlZPJbZ9IPvd7Z65imtnfVCXtu7IJkvZf0dzOX5ocuCo5eHVy6OrkwIOT7v0PJwEAACZFYAQtUUrJwkwvCzO9XHng/h9nMKw5vrh5CN2xDVVPxza1rbffeeeJtXXHz3LFurH7Ejod2Kba6eBcP7O9zvQmFO901q80d+kO91ldbgKmE3cmJ25PjtzU3D91ZLTujuTeLyc3/2nzeGsFU6efXPX45KHXJVc8pqlaWrh8NDeUMAkAAGgPgRFcZLqdkkML/RxauP/hw2BYm0qmUxc2dOp3y2lD6w7Otzh06s00FUMHrjr3trU2FUvHbk2OfiW59yvJ3Z9PvvKnyZ/9elO5tFHpJPuvSg49dMPysM2P5y8xgTcAADARAiPgNN1OyaH5fg7NP/DQ6dg2wdKxxZXRsLrN644truamO09uCKjuX+i0MVg6eIaw6cBcE05dsNCplCbgmb8kufKxm9uGg+T47cmxm5v5lo7d0lQm3fvl5OiXk1s+nnzmfclgafN+3Zn1uZUOPyw5/PDk8COaMGnfFetLf+78vhYAAGDPERgBF8T5CJ3GV7Fbr3TaWej0hbvWQ6fjy6up55hL/P6GTgfn+7l0YSYH5/vpdu5D4NTpJgcf3CxnfvHNhN3jIOneLyfHbx0FTbckt/zFKFRaPn3ffVc08yYdfMj6cuAhmx/P7Nt5fwEAgD1HYAS01vm4it04dNouWNo6efjGUOqLd59cG3p3fOnsoVMpyaFReHR4oZ9LFmZyyb6ZXLLQz+GFmVy65f54m363c7YXn+y/slmu/tozvbgmRLr3K838SSfuGA2B+3Jy9ObkyBeSL3ykGRq31dyhDSHSg0eTdV+1Pp/Sgauax6qVAABgTxIYARe1zaHT/P06xnBYc2ItdBrP37SSo6dWc+Tkco6cWM6RkyvN/ZPLueXexXz6lqO5++RyFleGZzzugdleDu8bBUwLM7l8/2wefGguVx2ay1UH53Llwdlcvr9ZZnrbhEudznrF0Nksn2yqko7ePFq+svnx7Z9qJukerpy+79zhZgjcvsubZeHy0dC30eN9VzRXh1u4tNm269cKAABcDPzLHuAcOp3xkLX7Xum0uDLIkZPLufvEcu5ZC5VWRiFTs65pW85nbzuW244tZTA8vZzp8EJ/FB7NrFUwjSuamgqmDev29XNgtrc+L9PMQnLZo5rlTGpNTt7dVCwdu6WpVBovJ25vAqXbPz26KtzdZz7O7MFm3qZ9VzTVUePb+UubfozndZq/tGlbuEzIBAAALeRf6QAX0Fy/mwcfms+DD+2sumkwrLnz+FJuuXcxdxxbyp3Hl3LHsfXlrhNL+eztx3PkxHLuObWybbiUJL1OGQ2BW69gGg+TWw+YNrcdXLg0Zd9lyYO+5hydXE1O3tXMsXR8FCadOrK+nLyrGR53zxeTL9/QbFfPXGmVucNNtdLc4WT+cDNcbvZgMncwmT2QzB4a3d+47uD6dgInAAA47/wrG6BFup2SBx2cy4MOnnvuoOGw5tjiau4eDYVbGxp3Yjl3n2yqlu4erfv8Hcdz5AtNhdPZQ6aNIVI/h+dncmihmbz84GgS80Pz/Ryen82h+Wty6KqvOvek38NBsnhvsnJyS6h054bbO5ttTt6d3P3XydLRZPHo6VeK205/YUOItDVYOlPYdHBD24GkN9dMRgUAACQRGAHsWp1OacKchX6uzc6uelZrzdHF1bUwaTwkbjz/0t0nVtba/vrOE7nn5D2599RKllbPUiGUZP/s+tXkDs73mnmj5sdXmVtfd2Du8hycvyr7DvRy4PJe9s/1sm+2l30zve1Dp9WlJjhaOroeIq3dHhvdv3dz29KxZm6m8bqVEzt4M/tbgqWDTRA1s5D09zW3M/vW7/fHjxeS/nyz9Gab4GnTMtu0dbo7Oj8AANAWAiOAPaSUslYl9IjLdhYyJc1cTEdPreTeUyu559RK7j3Z3B8v46vLHR1NBn7r0cV89vbjo8crOUNR0yZz/U72zTQB0sJMtwmSZnvZN9PNwkwvCzP9LMxcmfmZB2dhppv5mV72Hehm4bJu5vrdzPa6met3Mj/TzXy/WeZmupnr1PRXT6RsGzqdJYg6dktTFbV8sgmdlk9uPzH4TnR66wFSb2O4NHuGsGn0uL/l8Xjp9pPuzGi5L/f7SUpSOs2k6QAAcAYCIwDOaa7fhDJX7mCo3Fa11pxYHqxdWe7o4kqOL63m+OJqTiyt5vjSak4sDXJyubl/cnkwul3NvadWcss9p3JiaTUnVwY5uTzI8jmqnbbTKVkLlOb63cz357Mwuz8L/YdnptfJTK+TfrdkptdNv18yO99Jv9vJTLeTfq+5nel1MlsGWchi5sty5uqpzGcpc3Up/axkNsvp1+XMZCX94XL6dSm94XJ6dSm94VK6w+V0B81tGSymrC4lq4vrVVSrd6w/Xj01ul1MBsv3+fXuTGmCrLWlez8e73Cf1eVmrqnuTFK66/uWrbed7dffl21LZ/25N7V1znyMTm8H23aFbADAniIwAuCCKqVk/2wv+2d7efChB368wbDm5PJqTq0Mcmq5CZFOrQyyuLYMc2qbdYsrgyyujtrX9m2Oc++plawMhlleHWZ5dLuydluzPDhbSFWSzIyWnZsZBVFNUDUKqMbh1IbAaqZbs68zyEJ3JQtlNQudlcx2hpkrg8x2BulnNRksp5/V9LOahe4wc51BunUl3eFKemU1veFq87iupBn4V9Opg/TKMDNlmG4G6WaQTh0tG+6XutrcDpv7pQ5ShsOU1ZV06uKGdYOkrqYMV5PRthkOmknSu72mfbDcrKuDzbfZQQlaW5TulkDqLOFS6STZMNRyY8g1rvIqnW3Crt7myrBxW0oz19am+yVrVWOb2kbPNXdoFIiN+7Fhn2TD/qPbOmyCylrXjzl+/nFf1+5vaUtpzunW1zTuy8b+b3estfens2Xpbtmvs822ZZtjbtxu6/5btxu/BztQa7J8Ipndfx5+oACgvQRGAOwq3U7Jgbl+Dsz1J/actda14GhlQ6i0PGiCpZXVmuXBIMurm7dZGQyztCl8Gu9Xt4RS24RVo+PeuzLMHas1K4NOlge9rKx2Nj3/YFhTSslwWFOTM05qPk2lJN3RH+OlJCUlvW5Jt1PSGy2znZpeqel3amY6w/Q6Nf3U9DvDZilJrwzTL8O17XpptuulNm2dmm4Zpp9humXUXpp1vTJMN8Pmfpp13XF7Bk3baL/x/c4ZbwfpZJhObdaVOhgtm7ftpI5e7yhbGe1TUkeB2TCpTWDXSU1JE8SV4amU4UrKcCUZLDdhW+po+2y4P7rd+Hjj/eEgWT42tfO+65wWVG0MljYETnXQTN4/e7B5rzcFh1vCxN58s264uiG42hKk1doMd63D0f79LRV72wWGOT0s3Bj8nXFdthznTPtsuL/1Z6wORx/qmfW+bFU3fg9tuN/pJvuv2rx+/HO7cb/ebPM+rb3ODa93fP+0/m/oe+p6X+uwCatLaa7G2e2vB7Nr+43fl2yzrpyHdTl93aZzeH/XbfO+bN3uxB3N+zFzoLmQxOpyM9y5Dpt5+Lbbf+Pxt31NWd8utRmyvXBp83O+3Xtxtj6WsvlnoHSan62N32/j5xtfcXW7Y2y9vS8XshgOmucbLDfzFAJrBEYAcA6llMz0mkqgzE67N9uroz+0lgfDLC4PU1MzrMnqcJjhMGuPa63Nv8NrsjwY5NTyevA0GNYMa83qsGY4bG4Hw5pam32HtWkfDNe3GYweb1pqzWDQ7NPvlSwuDzIYPW+SDOr68VcH49vh2vpBTQbDcZ+a510Z1ixtOP54/+Ggbtv3Zpus93XUz7XXVde32/S3bQuV0lzFsAnYOmtB21rg1u00t72SbqepWut2SrqljEK1wdq+vZL0OkmnlOa20/xjsNNpQr1ut5Pam02ndNPtZC2U65akm0F6JemU5nETwtV0OzXd1HRS0+n2mvYMmuNuCOo6SbqdpJNm+26p6ZTh6HHSLcOUDNePVeoogGvCtG6pKalNUFdqurVZP96uCexG90tNqc22pQzTGf3xWdeCvZqyMQCp6wHe2jL+I3LT+kHqcJhBby6dJJ1Od7Tt6nrF3Fr13GqycmoUBI2DocGmY639AdzpN38oD1fXl9WlZHhidOxxaJOcHhBuDA63W5ezbHeW44zXbxck1ZoMVtb7P/5B3fyTe3rb6tLOrn4J58U5QqXx7WBlfY7CTr+5f65qxZTmszl7YOfdGX/2N/Yv2RKAbgyotwkhx9ufK9g7bbttnvM+H+Nc221ty4a2XfJc9+UYBx6UPPMnc7ETGAHARaCM/iEz22smAGfnNgZf68FSE7atBU91c4g2rE3YVbMeONXabLc6aIYyjgOswbB5vDFoG/+7s9ZkeXWYpcFww/GHp4V246U5znD0POvtK6PQb9y+HpKVLNduBqsbgrmNr7NmU+DW3D+26XWOb8fbXngb/rGeC/ezXEoTnHVK8/nplCY065TSZCIZ5yZ1tH1Zq+4b77/QbyboH78t8zOdDIfrx954WzJ+vtHjtecebzN6vGG7lKz1b7u/i0rWj9Xst36/+U5YD3u7o+cY1rrW3u00VZullFH7qC9Zf85y2nM152jbtqx/F52+Pil1kNnhqZSU1FF/1o9XUkbVOP3hchP6pTkJnVI3PFfWq/dGAVcpSacmZRQeNgftpo7+8K6lm06GmRmcSHc89HY0RHd8zKYXzXMlJaXWUR/GS/N+NtvWjGur1rdp9h8fp6zd1tH70ISY432a7da32fh47XlKmn6MnnP9uBv3G/V3w+P197umdvsZdmebYcWdTmqnn85wJel00xksberzuA/rx2gCx03PX9aecu31pDeX7tK9qd1+81pTU8qwOcYoeGz22/D9Maq2LGsHK6P3fZgybCrCSuk1Pyfj17VW0bYlHE1NqaevX3t/NgWlOX3fTjed/lxWazd1+XgGpZ+kZr5bUrIlSF4Lk0fDb1cXsyN1fFI2hK7bvZatIfXW6ruNr+NsVXpb97lfx7ifz1XrBXyuuuHmfL+uM7WdYbtLrsleIDACAPa0Tqekk5K+nG1Hat1Y6dWEEWsVW8P1gGItaBoFbuNqtbVwbsu+40Bqc0jVVJlt3beOg66N+27qw/q+w9H6pDnXGwO49cq79e3G/d8YeDSvu6mYm+11c2xxJd1SMqg1p5YH6XSaYaFLq8N0SlkLEod1823N+uuuo/dyXPk3PMN2TftwyznI2v41aQKrtfvj52wCpHFANX4/uqO/uQfDjF7v+vs4Osx6P+r6uvFrGj/Opscbthvdz5a2bHOcnHbcuuH4F5oP/Pl15bQ7cEGNvwfWwtSsh6LZGDpuE6COw79Nxxu3ldOPuTWQvU/9zMbAdj3ATZITy6tZ6HfT7ZasrNYsrQ6yMNNLd/S92Os2wfHG15MNx9j6Gje+Lxu33bjdmbavdXMfN4bqG9/XjYH9pvYNx9yuH9v1fVM4vOn9Hr1zZf07cWOwv+lcb/Cgg7P56e2f9qIiMAIAYMdKaeag8o9ILrTxH4vbBlNZb0u2CaO2bL81nNraXkcJ15nCrfFw3pytfUPb5kBtc3837rvpNZwxPNv6Ppwerm3s/2nH39i/jf3epn1j39fbTu/btu/DlvdqsOH8bXx9a69o3L7xWKe9rtOfb6v1dz7n2G6bdXU94O53Oyklmel2sjIY5uTy4LR+bheerj3/ltd/WvuWvu0kkL0vtvs5Gq+vqVmY6WZxZZjVwXDtCrEnlwZNQLIhTN/0tBt/vsbH2mbdxvdn445bP2/jlk4p64H5pvesrg2jrzXpjq5OOg73x/8hsLGPm34+tn0/mgfbfbY39rHWmm6nqe4c1s0/7+N9NoZGK2e9IMrFY+K/60spfzfJG5I8LslNSX6y1vp/T7ofAABAe20c3jZaM7W+AOxFZ7i0wYVRSvm6JO9J8utJnpzkV5K8vZTyDZPsBwAAAABnNukKo/8zydtrrT8/evzpUsrTkvxoko9MuC8AAAAAbGNiFUallF6Sv5PkHVua3p3kGZPqBwAAAABnN8khadcmmU/yyS3rb0xyaSnl8AT7AgAAAMAZTDIwunx0e/eW9UdGtwcn2BcAAAAAzmCSgVF3dDvYsr5uuU2SlFJeVkq5oZRywx133HHBOwcAAABAY5KB0bHR7dZKokOj2yMbV9Za31Zrva7Wet0VV1xxwTsHAAAAQGOSgdFfJRkmefSW9Y9O8sVa6/EJ9gUAAACAM5hYYFRrPZbkT5K8YEvTC5K8b1L9AAAAAODsehN+vjckeWcp5cYkH0ryvCTPSvLECfcDAAAAgDOY5JC01Fr/a5IfTvITST6e5O8l+Tu11psm2Q8AAAAAzmzSFUaptf77JP9+0s8LAAAAwM5MtMIIAAAAgPYTGAEAAACwicAIAAAAgE0ERgAAAABsIjACAAAAYBOBEQAAAACbCIwAAAAA2ERgBAAAAMAmAiMAAAAANhEYAQAAALCJwAgAAACATUqtddp9OKdSyh1JvjDtfpwnlye5c9qdYOKc973Lud+7nPu9y7nfu5z7vcu537uc+73rYjn3j6i1XrFdw64IjC4mpZQbaq3XTbsfTJbzvnc593uXc793Ofd7l3O/dzn3e5dzv3fthXNvSBoAAAAAmwiMAAAAANhEYDR5b5t2B5gK533vcu73Lud+73Lu9y7nfu9y7vcu537vuujPvTmMAAAAANhEhREAAAAAmwiMAAAAANhEYDQhpZS/W0r501LKYinlM6WU7552n3hgSikPKaX8p1LKnaWUe0spHyylPGlD+3eUUurWZZvjvKKUclMp5VQp5fpSyhMm+kK4T0opT9juvJZSrtmwzYtGn/PFUsqflVKesc1xnPddppTyLWc497WUsjLaxuf+IlJKOVhK+eVSyr/Ypu0Bf85LKb1Sys+WUm4upZwspfy3UsojLuRrYmfOdO5LKY8ppbxn9Hv/rtH9a7ds8//a5nvgpi3bOPctdZZzf16+35379tru3JdSXnKW3/2f3bCdz/0uVM7x99xomz39+15gNAGllK9L8p4kv57kyUl+JcnbSynfMM1+8YD9YpKjSZ6d5JlJjiT53VLKlaP2S5N8Msm1W5Y1pZQfSfLqJC9P8pQkX0rygVLKoQn0n/vn0iT35vTz+uUkKaV8e5K3JnlDms/7HyR5bynlkeMDOO+71v/M6ef92iTvTfP9nvjcXxRKKZeUUn4syV8meck27efrc/4vk/yDJN+X5G8lmUnyvlJK97y/KHbkXOc+yf8vySeSfFOSv5fkkiS/XUqZ27DNpUk+kM3fAX9ry3Gc+5bZwbk/X9/vzn3LnOPc/0a2/93/8az/7k987ners/495/d9klqr5QIvacKi/7Bl3W8m+Y1p983ygM7ro7c8nk1yW5KXjh7/WJLfOcv+vSR3JPm+Lce4JckPT/v1Wc543r4zyV+epf3jSf7PLev+NMmbnPeLb0nyVUkWkzxq9Njn/iJY0vzB8MUk/0eafxz+iy3tD/hznuTKJMtJvnHDNlcmWUrybdN+D/bqsoNzv/V3/+VJVpM8c8O6X0zytrM8h3PfwmUH5/4Bf7879+1cznXut9n+b6cJFg5tWOdzvwuXbb7Tt/49t+d/36swusBKKb0kfyfJO7Y0vTvJaeVs7B611hu3PF5K8oU0XwBJclmS289yiK9PcjjNz8LGY7w/fjba7IzntZRydZIn5vTP+29m/Zw67xeX1yR5e63186PHPvcXh/ckubbW+tatDefxc/6sJHfWWj+8YZvbk1wfPwvT9J6c4dwn2/7uvzPJ3Vn/3Z+c+3vAuW+n9+Qs5z7n5/vduW+n9+Ts536rn0zyr2ut925Y53O/C53t7zm/7xsCowvv2iTzaUpYN7oxyaWllMMT7xEXRCllIclXJ/nUaNVlSb6zlHKslPLFUspbSimXbtjlcUluqrWe2HKoG5M86sL3mPvpsiRPKaUcLaXcUkr5z6WUh4/aHpdkJclnt+yz8Zw67xeJ0T8kvifJmzas9rm/CNRa76m1Ds7QfL4+54/L6f822LoNE3aOc3+aUspD01QZfWrD6suSvGL0PfD5Usq/LKXMb2h37ltoB+f+fHy/O/ctdF8+96WUp6QJCX5pS5PP/UVgy99zft+nKaHiwrp8dHv3lvVHRrcHk9wzsd5wIf3LNOWH7xs9/qUk/zHNF80Tkrw2TdDw1NEvpctz+s9F0vxsHLzw3eV++o005cqn0vxCeU2S60spj09zTo/UUa3pBkeSHBjdd94vHv8kyR/VWjf+oehzf/E7X5/zs23zmPPQTy6wUkonzWf+v9da/3xD0/87SU1S0vxh+do05/Tvjdqd+93pfHy/O/e73yuSvKPWuvU8+txfHDb+Pfdd8fteYDQB44mstqbWdcstu1QpZSbNPyK+Ncn/o9a6miS11v+1YbM/LaV8JM1Emc9J8ltpfja2+9+MGj8XrVVr/WzW/6fh46WUD6aZJPHFab74z3VOnfeLwGi48fcn+acb1/vc7wk7OYfnaxtaqpRyMM2Et9emmc9kTa31oxsefrSU8skk/72U8vjRd4Rzvwudp+93534XK6VcluQFaSa938Tnfnfb7u+50YTUe/73vSFpF96x0e3W/zkez5p+JOxapZSHJflwkscmeWqt9S/PtO2o7TNp/lcqaX42tqsoOJTtU2haqNZ6V5L/kea8nu2cjj/rzvvF4e8k2Z/kv5xtI5/7i9L5+pz7WdilRhWlN6SZ7Pobaq13nG37Wuvvp7kKj++Bi8j9/H537ne3f5Dki7XWPznXhj73u8dZ/p7z+z4Co0n4qyTDJI/esv7Rab5wjk++S5wPo3lrPpLkQ0m+pdZ66w5266cpZU6SzyV55KhSYaNHZ/NcCLTf+Lx+Lsm+0dw2G208p877xeG7k3xgNLHhufjcX1zO1+f8czn93wZbt6FlSilfm2ZY8r+rtb6g1nrsHLuklFLSVPVv/B5w7i8O9/X73bnf3b47yX/dyYY+97vDOf6e8/s+AqMLbvQPiT9JU7640QuyPtcNu9PbkvyXWus/28lEeaWUJ6SZ8+aPRquuT/MZfPaGbWaTPDd+NnaNUspVSf5mmvP6qTTjnrd+3p+f9XPqvO9yo38EjocgnGtbn/uLz/n6nH8wyd8Y/YyMt7k8yTfGz0IrjT77v57kDbXWf30fdn1Okrkkfzx67NxfBO7n97tzv0uVUg4l+VvZwe/+EZ/73eFsf8/5fR9zGE3KG5K8s5RyY5r08nlpLq/3xGl2ivuvlLIvzTl8Wynlmi3NS7XWW0op70jyziSfTvI1Sf5VmoqEDydJrfXeUsqbk7y5lLKc5OY0EyjfluRdk3kl3FellLek+Z+IP03yiDSf78+nmQCxllLemOR1pZRb08xt8INJrkrylsR5v0g8PskVWf9H4Bqf+4vf+fqc11o/U0r5zSS/Xkr5oTQT6f9/kvzuToY7MBWPTjNk4UPb/O4/MR6aVkr5QJJfSHNp5q9PM4nqW2utX0ic+93qfHy/O/e72jenmdD6o9s1+tzvPjv8e27P/74XGE1ArfW/llJ+OMmr03yR/FmSv1NrvWmqHeOBuCxNmrzdH3gfS3JdktvTTJ52WZp0+v9Kc8WEjV6V5pfP25PMJPntJM+5L5f0ZeK+kORnkzwozbjj/5LkJ2qty6P2X0xz5YRfSDM2+fokz6y1Ht1wDOd9d/vaJCfTTHa+lc/93nC+PuffPzrG+9NMiPkbSX70gvee++vK0e1pYXGafw/8/dH9E0l+Nc2cFV9IEyz8qy3bO/e7z/n6fnfud6evTfLZWuvJM7T73O8+O/l7bs//vi+nXyUOAAAAgL3MHEYAAAAAbCIwAgAAAGATgREAAAAAmwiMAAAAANhEYAQAAADAJgIjAAAAADYRGAEAAACwicAIAAAAgE0ERgAAAABsIjACAAAAYJP/P+GgN0iLswbXAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 오차 정도와 정확도를 그래프로 그린다.\n",
    "plt.plot(history.history['loss'], label='loss')\n",
    "plt.plot(history.history['val_loss'], label='val_loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5/5 [==============================] - 0s 2ms/step - loss: 16.2639\n",
      "손실률 : 16.2639217376709\n"
     ]
    }
   ],
   "source": [
    "# 성능평가\n",
    "a1 = model.evaluate(X_test, y_test)\n",
    "print(f'손실률 : {a1}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.engine.sequential.Sequential at 0x2703c821280>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 모델을 복원한다.\n",
    "best_model = load_model('models/22_선형회귀/최종.hdf5')\n",
    "best_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([31.691868 , 29.1035   , 19.149124 , 23.474897 , 24.747076 ,\n",
       "       18.748978 , 29.975866 , 18.129461 , 21.424498 , 23.164734 ,\n",
       "       30.005682 , 30.685875 , 17.40433  , 23.117313 , 21.512554 ,\n",
       "       25.67001  , 12.804648 , 37.079247 , 28.040628 , 16.88409  ,\n",
       "       21.38309  , 13.78387  , 22.49964  , 23.577705 , 31.135145 ,\n",
       "       11.704777 , 11.796269 , 20.519335 , 36.202465 , 14.762757 ,\n",
       "       19.710518 , 17.37924  , 47.919865 , 15.920772 , 15.337752 ,\n",
       "       19.3997   , 16.640562 , 32.99471  , 11.11016  , 16.494686 ,\n",
       "       20.12177  , 22.238659 , 26.562418 , 14.615733 , 15.407184 ,\n",
       "       11.753483 , 52.456486 , 15.078021 , 15.614341 , 12.219623 ,\n",
       "       24.188372 , 20.301397 , 29.656452 , 17.422468 , 15.142171 ,\n",
       "       22.118887 , 11.159282 , 28.02412  , 18.670504 , 36.534935 ,\n",
       "       20.43214  , 25.94847  , 15.300746 , 13.393136 , 15.745534 ,\n",
       "       44.299564 , 39.25745  , 21.143742 , 23.247753 , 28.52164  ,\n",
       "       23.574497 ,  9.372737 , 19.351204 , 19.397837 , 18.482637 ,\n",
       "       21.410955 , 46.88257  , 23.364262 , 32.984947 , 34.23444  ,\n",
       "       18.672361 , 22.36752  , 35.21265  , 13.70208  , 22.007593 ,\n",
       "       27.353632 , 16.00078  , 23.152586 , 19.320833 , 18.07553  ,\n",
       "       23.203363 , 45.14459  , 18.27002  , 22.202072 , 12.274849 ,\n",
       "       21.596783 , 21.648615 , 23.058867 , 37.004124 , 19.232235 ,\n",
       "       12.638896 , 15.331176 , 22.125017 , 21.930128 , 10.788151 ,\n",
       "       18.25945  , 12.873426 , 33.26141  , 23.928164 , 28.062368 ,\n",
       "       45.088955 , 23.465763 , 12.012196 , 35.618473 , 33.487362 ,\n",
       "       36.00923  , 22.240534 , 11.5659685, 37.011684 , 38.4592   ,\n",
       "       16.216251 , 13.892727 , 25.074675 , 18.934338 , 21.077318 ,\n",
       "       19.66201  , 23.414486 , 21.755823 , 21.160273 , 25.44659  ,\n",
       "       21.106813 , 22.402395 , 23.197866 , 12.597093 , 22.63583  ,\n",
       "       32.89289  , 18.402372 , 13.333512 , 38.763508 , 25.850376 ,\n",
       "       20.603682 , 17.859753 , 18.447153 , 25.756405 , 33.657887 ,\n",
       "       21.265484 , 23.114624 , 15.63061  , 21.442772 , 18.313759 ,\n",
       "       31.99624  , 17.231115 ], dtype=float32)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 결과를 예측\n",
    "pred1 = best_model.predict(X_test).flatten()\n",
    "pred1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(y_test.values, label='original')\n",
    "plt.plot(pred1, label='prediction')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}