fixed-time-horizon-prediction/Exp_1.2_GoodwinCreek_2hour.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import datetime\n",
    "import glob\n",
    "import os.path\n",
    "from pandas.compat import StringIO"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### NREL Bird Model implementation: for obtaining clear sky GHI"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import itertools\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "sns.set_color_codes()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pvlib\n",
    "from pvlib import clearsky, atmosphere\n",
    "from pvlib.location import Location"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "gwc = Location(34.2487,-89.8925, 'US/Central', 98, 'Goodwin Creek')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "times2009 = pd.DatetimeIndex(start='2009-01-01', end='2010-01-01', freq='1min',\n",
    "                        tz=gwc.tz)   # 12 months of 2009 - For testing\n",
    "times2010and2011 = pd.DatetimeIndex(start='2010-01-01', end='2012-01-01', freq='1min',\n",
    "                        tz=gwc.tz)   # 24 months of 2010 and 2011 - For training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "cs_2009 = gwc.get_clearsky(times2009) \n",
    "cs_2010and2011 = gwc.get_clearsky(times2010and2011) # ineichen with climatology table by default\n",
    "#cs_2011 = bvl.get_clearsky(times2011) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "cs_2009.drop(['dni','dhi'],axis=1, inplace=True) #updating the same dataframe by dropping two columns\n",
    "cs_2010and2011.drop(['dni','dhi'],axis=1, inplace=True) #updating the same dataframe by dropping two columns\n",
    "#cs_2011.drop(['dni','dhi'],axis=1, inplace=True) #updating the same dataframe by dropping two columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "cs_2009.reset_index(inplace=True)\n",
    "cs_2010and2011.reset_index(inplace=True)\n",
    "#cs_2011.reset_index(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "cs_2009['index']=cs_2009['index'].apply(lambda x:x.to_datetime())\n",
    "cs_2009['year'] = cs_2009['index'].apply(lambda x:x.year)\n",
    "cs_2009['month'] = cs_2009['index'].apply(lambda x:x.month)\n",
    "cs_2009['day'] = cs_2009['index'].apply(lambda x:x.day)\n",
    "cs_2009['hour'] = cs_2009['index'].apply(lambda x:x.hour)\n",
    "cs_2009['min'] = cs_2009['index'].apply(lambda x:x.minute)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "cs_2010and2011['index']=cs_2010and2011['index'].apply(lambda x:x.to_datetime())\n",
    "cs_2010and2011['year'] = cs_2010and2011['index'].apply(lambda x:x.year)\n",
    "cs_2010and2011['month'] = cs_2010and2011['index'].apply(lambda x:x.month)\n",
    "cs_2010and2011['day'] = cs_2010and2011['index'].apply(lambda x:x.day)\n",
    "cs_2010and2011['hour'] = cs_2010and2011['index'].apply(lambda x:x.hour)\n",
    "cs_2010and2011['min'] = cs_2010and2011['index'].apply(lambda x:x.minute)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(525601, 7)\n",
      "(1051201, 7)\n"
     ]
    }
   ],
   "source": [
    "print(cs_2009.shape)\n",
    "print(cs_2010and2011.shape)\n",
    "#print(cs_2011.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "cs_2009.drop(cs_2009.index[-1], inplace=True)\n",
    "cs_2010and2011.drop(cs_2010and2011.index[-1], inplace=True)\n",
    "#cs_2011.drop(cs_2011.index[-1], inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(525600, 7)\n",
      "(1051200, 7)\n"
     ]
    }
   ],
   "source": [
    "print(cs_2009.shape)\n",
    "print(cs_2010and2011.shape)\n",
    "#print(cs_2011.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>index</th>\n",
       "      <th>ghi</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>day</th>\n",
       "      <th>hour</th>\n",
       "      <th>min</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2010-01-01 00:00:00-06:00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2010</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2010-01-01 00:01:00-06:00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2010</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2010-01-01 00:02:00-06:00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2010</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2010-01-01 00:03:00-06:00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2010</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2010-01-01 00:04:00-06:00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2010</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      index  ghi  year  month  day  hour  min\n",
       "0 2010-01-01 00:00:00-06:00  0.0  2010      1    1     0    0\n",
       "1 2010-01-01 00:01:00-06:00  0.0  2010      1    1     0    1\n",
       "2 2010-01-01 00:02:00-06:00  0.0  2010      1    1     0    2\n",
       "3 2010-01-01 00:03:00-06:00  0.0  2010      1    1     0    3\n",
       "4 2010-01-01 00:04:00-06:00  0.0  2010      1    1     0    4"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cs_2010and2011.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import files from each year in a separate dataframe"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "\n",
    "- year            integer\t year, i.e., 1995\n",
    "- jday            integer\t Julian day (1 through 365 [or 366])\n",
    "- month           integer\t number of the month (1-12)\n",
    "- day             integer\t day of the month(1-31)\n",
    "- hour            integer\t hour of the day (0-23)\n",
    "- min             integer\t minute of the hour (0-59)\n",
    "- dt              real\t decimal time (hour.decimalminutes, e.g., 23.5 = 2330)\n",
    "- zen             real\t solar zenith angle (degrees)\n",
    "- dw_solar        real\t downwelling global solar (Watts m^-2)\n",
    "- uw_solar        real\t upwelling global solar (Watts m^-2)\n",
    "- direct_n        real\t direct-normal solar (Watts m^-2)\n",
    "- diffuse         real\t downwelling diffuse solar (Watts m^-2)\n",
    "- dw_ir           real\t downwelling thermal infrared (Watts m^-2)\n",
    "- dw_casetemp     real\t downwelling IR case temp. (K)\n",
    "- dw_dometemp     real\t downwelling IR dome temp. (K)\n",
    "- uw_ir           real\t upwelling thermal infrared (Watts m^-2)\n",
    "- uw_casetemp     real\t upwelling IR case temp. (K)\n",
    "- uw_dometemp     real\t upwelling IR dome temp. (K)\n",
    "- uvb             real\t global UVB (milliWatts m^-2)\n",
    "- par             real\t photosynthetically active radiation (Watts m^-2)\n",
    "- netsolar        real\t net solar (dw_solar - uw_solar) (Watts m^-2)\n",
    "- netir           real\t net infrared (dw_ir - uw_ir) (Watts m^-2)\n",
    "- totalnet        real\t net radiation (netsolar+netir) (Watts m^-2)\n",
    "- temp            real\t 10-meter air temperature (?C)\n",
    "- rh              real\t relative humidity (%)\n",
    "- windspd         real\t wind speed (ms^-1)\n",
    "- winddir         real\t wind direction (degrees, clockwise from north)\n",
    "- pressure        real\t station pressure (mb)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "cols = ['year', 'jday', 'month', 'day','hour','min','dt','zen','dw_solar','dw_solar_QC','uw_solar',\n",
    "       'uw_solar_QC', 'direct_n','direct_n_QC','diffuse', 'diffuse_QC', 'dw_ir', 'dw_ir_QC', 'dw_casetemp',\n",
    "       'dw_casetemp_QC', 'dw_dometemp','dw_dometemp_QC','uw_ir', 'uw_ir_QC', 'uw_casetemp','uw_casetemp_QC',\n",
    "       'uw_dometemp','uw_dometemp_QC','uvb','uvb_QC','par','par_QC','netsolar','netsolar_QC','netir','netir_QC',\n",
    "       'totalnet','totalnet_QC','temp','temp_QC','rh','rh_QC','windspd','windspd_QC','winddir','winddir_QC',\n",
    "       'pressure','pressure_QC']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1029728, 48)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path = r'.\\\\data\\\\Goodwin_Creek\\\\Exp_1_train'\n",
    "all_files = glob.glob(path + \"/*.dat\")\n",
    "all_files.sort()\n",
    "\n",
    "df_big_train = pd.concat([pd.read_csv(f, skipinitialspace = True, quotechar = '\"',skiprows=(2),delimiter=' ', \n",
    "                 index_col=False,header=None, names=cols) for f in all_files],ignore_index=True)\n",
    "df_big_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(513149, 48)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path = r'.\\\\data\\\\Goodwin_Creek\\\\Exp_1_test'\n",
    "all_files = glob.glob(path + \"/*.dat\")\n",
    "all_files.sort()\n",
    "\n",
    "df_big_test = pd.concat((pd.read_csv(f, skipinitialspace = True, quotechar = '\"',skiprows=(2),delimiter=' ', \n",
    "                 index_col=False,header=None, names=cols) for f in all_files),ignore_index=True)\n",
    "df_big_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1155, 48)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_big_test[df_big_test['dw_solar']==-9999.9].shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Merging Clear Sky GHI And the big dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1029728, 50)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train = pd.merge(df_big_train, cs_2010and2011, on=['year','month','day','hour','min'])\n",
    "df_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(513149, 50)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test = pd.merge(df_big_test, cs_2009, on=['year','month','day','hour','min'])\n",
    "df_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train.drop(['index'],axis=1, inplace=True) #updating the same dataframe by dropping the index columns from clear sky model\n",
    "df_test.drop(['index'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1029728, 49)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Managing missing values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Resetting index\n",
    "df_train.reset_index(drop=True, inplace=True)\n",
    "df_test.reset_index(drop=True, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dropping rows with two or more -9999.9 values in columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(987521, 49)"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Step1: Get indices of all rows with 2 or more -999\n",
    "missing_data_indices = np.where((df_train <=-9999.9).apply(sum, axis=1)>=2)[0]\n",
    "# Step2: Drop those indices\n",
    "df_train.drop(missing_data_indices, axis=0, inplace=True)\n",
    "# Checking that the rows are dropped\n",
    "df_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(496347, 49)"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "missing_data_indices_test = np.where((df_test <= -9999.9).apply(sum, axis=1)>=2)[0]\n",
    "df_test.drop(missing_data_indices_test, axis=0, inplace=True)\n",
    "df_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# For the rows with only one cell as -9999.9, replacing this cell with the mean of the column"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### First resetting index after dropping rows in the previous part of the code"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 2nd time - Reseting Index\n",
    "df_train.reset_index(drop=True, inplace=True)\n",
    "df_test.reset_index(drop=True, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "one_miss_train_idx = np.where((df_train <=-9999.9).apply(sum, axis=1)==1)[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2007"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(one_miss_train_idx)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(987521, 49)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\s269596\\AppData\\Local\\Continuum\\Anaconda3\\lib\\site-packages\\pandas\\core\\indexing.py:194: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  self._setitem_with_indexer(indexer, value)\n"
     ]
    }
   ],
   "source": [
    "col_names = df_train.columns\n",
    "from collections import defaultdict\n",
    "stats = defaultdict(int)\n",
    "total_single_missing_values = 0\n",
    "for name in col_names:\n",
    "    col_mean = df_train[~(df_train[name] == -9999.9)][name].mean()\n",
    "    missing_indices = np.where((df_train[name] == -9999.9))\n",
    "    stats[name] = len(missing_indices[0])\n",
    "    df_train[name].loc[missing_indices] = col_mean\n",
    "    total_single_missing_values += sum(df_train[name] == -9999.9)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_col_min = df_train.apply(min, axis=0)\n",
    "df_col_max = df_train.apply(max, axis =0)\n",
    "#print(df_col_min, df_col_max)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = np.where((df_train <=-9999.9).apply(sum, axis=1)==1)[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "# doing the same thing on test dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2206"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "one_miss_test_idx = np.where((df_test <=-9999.9).apply(sum, axis=1)==1)[0]\n",
    "len(one_miss_test_idx)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\s269596\\AppData\\Local\\Continuum\\Anaconda3\\lib\\site-packages\\pandas\\core\\indexing.py:194: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  self._setitem_with_indexer(indexer, value)\n"
     ]
    }
   ],
   "source": [
    "col_names_test = df_test.columns\n",
    "from collections import defaultdict\n",
    "stats_test = defaultdict(int)\n",
    "total_single_missing_values_test = 0\n",
    "for name in col_names_test:\n",
    "    col_mean = df_test[~(df_test[name] == -9999.9)][name].mean()\n",
    "    missing_indices = np.where((df_test[name] == -9999.9))\n",
    "    stats_test[name] = len(missing_indices[0])\n",
    "    df_test[name].loc[missing_indices] = col_mean\n",
    "    total_single_missing_values_test += sum(df_test[name] == -9999.9)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "test = np.where((df_test <=-9999.9).apply(sum, axis=1)==1)[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(987521, 49)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(496347, 49)"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Exploratory Data Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "dw_solar_everyday = df_test.groupby(['jday'])['dw_solar'].mean()\n",
    "ghi_everyday = df_test.groupby(['jday'])['ghi'].mean()\n",
    "j_day = df_test['jday'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY4AAAEgCAYAAACjEpTiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure()\n",
    "\n",
    "axes1 = fig.add_axes([0.1,0.1,0.8,0.8])\n",
    "#axes2 = fig.add_axes([0.1,0.1,0.8,0.8])\n",
    "\n",
    "axes1.scatter(j_day,dw_solar_everyday,label='Observed dw_solar',color='red')\n",
    "axes1.scatter(j_day, ghi_everyday, label='Clear Sky GHI',color='green')\n",
    "\n",
    "axes1.set_xlabel('Days')\n",
    "axes1.set_ylabel('Solar Irradiance (Watts /m^2)')\n",
    "axes1.set_title('Solar Irradiance - Test Year 2009')\n",
    "axes1.legend(loc='best')\n",
    "\n",
    "fig.savefig('Figure2a.png', bbox_inches = 'tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\s269596\\AppData\\Local\\Continuum\\Anaconda3\\lib\\site-packages\\matplotlib\\axes\\_axes.py:6462: UserWarning: The 'normed' kwarg is deprecated, and has been replaced by the 'density' kwarg.\n",
      "  warnings.warn(\"The 'normed' kwarg is deprecated, and has been \"\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbEAAAGpCAYAAAADVQhHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nJGlaUoklSZIkTcvfJTB7tEMwpJIAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x432 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.jointplot(x=dw_solar_everyday,y=ghi_everyday,kind='reg')\n",
    "#plt.title('observed dw_solar vs clear sky ghi')\n",
    "plt.xlabel('Observed global downwelling solar (Watts/m^2)')\n",
    "plt.ylabel('Clear Sky GHI (Watts/m^2)')\n",
    "plt.savefig('Figure3a.png', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### making the Kt (clear sky index at time t) column by first removing rows with ghi==0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train = df_train[df_train['ghi']!=0]\n",
    "df_test = df_test[df_test['ghi']!=0]\n",
    "df_train['Kt'] = df_train['dw_solar']/df_train['ghi']\n",
    "df_test['Kt'] = df_test['dw_solar']/df_test['ghi']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train.reset_index(inplace=True)\n",
    "df_test.reset_index(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "test Kt max: 16565.62428392545\n",
      "test Kt min: -858.197482511317\n",
      "test Kt mean: 2.3067127385087867\n",
      "\n",
      "\n",
      "train Kt max: 154389.83695910865\n",
      "train Kt min: -244.67261793490562\n",
      "train Kt mean: 3.284695720674705\n"
     ]
    }
   ],
   "source": [
    "print(\"test Kt max: \"+str(df_test['Kt'].max()))\n",
    "print(\"test Kt min: \"+str(df_test['Kt'].min()))\n",
    "print(\"test Kt mean: \"+str(df_test['Kt'].mean()))\n",
    "print(\"\\n\")\n",
    "print(\"train Kt max: \"+str(df_train['Kt'].max()))\n",
    "print(\"train Kt min: \"+str(df_train['Kt'].min()))\n",
    "print(\"train Kt mean: \"+str(df_train['Kt'].mean()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0xb8c0080>]"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAD4CAYAAAD2FnFTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nREQkkAJCREQC/X+Taznq3XBGqgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(df_train['Kt'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0xebc8d30>]"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX8AAAD3CAYAAAD10FRmAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nv4hIAP1/Utf10Ubxek4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(df_test['Kt'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train= df_train[df_train['Kt']< 5000]\n",
    "df_train= df_train[df_train['Kt']> -1000]\n",
    "df_test= df_test[df_test['Kt']< 5000]\n",
    "df_test= df_test[df_test['Kt']> -1000]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Group the data (train dataframe)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "zen = df_train.groupby(['year','month','day','hour'])['zen'].mean()\n",
    "dw_solar = df_train.groupby(['year','month','day','hour'])['dw_solar'].mean()\n",
    "uw_solar = df_train.groupby(['year','month','day','hour'])['uw_solar'].mean()\n",
    "direct_n = df_train.groupby(['year','month','day','hour'])['direct_n'].mean()\n",
    "diffuse = df_train.groupby(['year','month','day','hour'])['diffuse'].mean()\n",
    "dw_ir = df_train.groupby(['year','month','day','hour'])['dw_ir'].mean()\n",
    "dw_casetemp = df_train.groupby(['year','month','day','hour'])['dw_casetemp'].mean()\n",
    "dw_dometemp = df_train.groupby(['year','month','day','hour'])['dw_dometemp'].mean()\n",
    "uw_ir = df_train.groupby(['year','month','day','hour'])['uw_ir'].mean()\n",
    "uw_casetemp = df_train.groupby(['year','month','day','hour'])['uw_casetemp'].mean()\n",
    "uw_dometemp = df_train.groupby(['year','month','day','hour'])['uw_dometemp'].mean()\n",
    "uvb = df_train.groupby(['year','month','day','hour'])['uvb'].mean()\n",
    "par = df_train.groupby(['year','month','day','hour'])['par'].mean()\n",
    "netsolar = df_train.groupby(['year','month','day','hour'])['netsolar'].mean()\n",
    "netir = df_train.groupby(['year','month','day','hour'])['netir'].mean()\n",
    "totalnet = df_train.groupby(['year','month','day','hour'])['totalnet'].mean()\n",
    "temp = df_train.groupby(['year','month','day','hour'])['temp'].mean()\n",
    "rh = df_train.groupby(['year','month','day','hour'])['rh'].mean()\n",
    "windspd = df_train.groupby(['year','month','day','hour'])['windspd'].mean()\n",
    "winddir = df_train.groupby(['year','month','day','hour'])['winddir'].mean()\n",
    "pressure = df_train.groupby(['year','month','day','hour'])['pressure'].mean()\n",
    "ghi = df_train.groupby(['year','month','day','hour'])['ghi'].mean()\n",
    "Kt = df_train.groupby(['year','month','day','hour'])['Kt'].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_new_train = pd.concat([zen,dw_solar,uw_solar,direct_n,diffuse,dw_ir,dw_casetemp,dw_dometemp,uw_ir,uw_casetemp,uw_dometemp,\n",
    "                    uvb,par,netsolar,netir,totalnet,temp,rh,windspd,winddir,pressure,ghi,Kt], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>zen</th>\n",
       "      <th>dw_solar</th>\n",
       "      <th>uw_solar</th>\n",
       "      <th>direct_n</th>\n",
       "      <th>diffuse</th>\n",
       "      <th>dw_ir</th>\n",
       "      <th>dw_casetemp</th>\n",
       "      <th>dw_dometemp</th>\n",
       "      <th>uw_ir</th>\n",
       "      <th>uw_casetemp</th>\n",
       "      <th>...</th>\n",
       "      <th>netsolar</th>\n",
       "      <th>netir</th>\n",
       "      <th>totalnet</th>\n",
       "      <th>temp</th>\n",
       "      <th>rh</th>\n",
       "      <th>windspd</th>\n",
       "      <th>winddir</th>\n",
       "      <th>pressure</th>\n",
       "      <th>ghi</th>\n",
       "      <th>Kt</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>day</th>\n",
       "      <th>hour</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">2010.0</th>\n",
       "      <th rowspan=\"5\" valign=\"top\">1.0</th>\n",
       "      <th rowspan=\"5\" valign=\"top\">1.0</th>\n",
       "      <th>7.0</th>\n",
       "      <td>157.281154</td>\n",
       "      <td>-4.348077</td>\n",
       "      <td>0.090385</td>\n",
       "      <td>0.150000</td>\n",
       "      <td>-1.123077</td>\n",
       "      <td>320.623077</td>\n",
       "      <td>280.961538</td>\n",
       "      <td>280.546154</td>\n",
       "      <td>335.715385</td>\n",
       "      <td>277.776923</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-15.086538</td>\n",
       "      <td>-15.086538</td>\n",
       "      <td>5.005769</td>\n",
       "      <td>79.425000</td>\n",
       "      <td>2.300000</td>\n",
       "      <td>69.582692</td>\n",
       "      <td>1013.025000</td>\n",
       "      <td>58.673464</td>\n",
       "      <td>-0.092544</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8.0</th>\n",
       "      <td>146.229333</td>\n",
       "      <td>-4.288333</td>\n",
       "      <td>0.113333</td>\n",
       "      <td>0.156667</td>\n",
       "      <td>-1.106667</td>\n",
       "      <td>322.270000</td>\n",
       "      <td>280.761667</td>\n",
       "      <td>280.345000</td>\n",
       "      <td>334.833333</td>\n",
       "      <td>277.581667</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-12.571667</td>\n",
       "      <td>-12.571667</td>\n",
       "      <td>4.783333</td>\n",
       "      <td>75.178333</td>\n",
       "      <td>2.541667</td>\n",
       "      <td>84.746667</td>\n",
       "      <td>1013.470000</td>\n",
       "      <td>199.226885</td>\n",
       "      <td>-0.022905</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9.0</th>\n",
       "      <td>133.906167</td>\n",
       "      <td>-4.328333</td>\n",
       "      <td>0.178333</td>\n",
       "      <td>0.701667</td>\n",
       "      <td>-1.108333</td>\n",
       "      <td>318.151667</td>\n",
       "      <td>280.491667</td>\n",
       "      <td>280.046667</td>\n",
       "      <td>333.480000</td>\n",
       "      <td>277.301667</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-15.333333</td>\n",
       "      <td>-15.333333</td>\n",
       "      <td>4.445000</td>\n",
       "      <td>76.843333</td>\n",
       "      <td>2.831667</td>\n",
       "      <td>159.561667</td>\n",
       "      <td>1013.861667</td>\n",
       "      <td>357.002941</td>\n",
       "      <td>-0.012248</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10.0</th>\n",
       "      <td>121.558667</td>\n",
       "      <td>-4.560000</td>\n",
       "      <td>0.316667</td>\n",
       "      <td>-0.558333</td>\n",
       "      <td>-1.170000</td>\n",
       "      <td>300.935000</td>\n",
       "      <td>279.631667</td>\n",
       "      <td>279.101667</td>\n",
       "      <td>328.243333</td>\n",
       "      <td>276.451667</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-27.313333</td>\n",
       "      <td>-27.313333</td>\n",
       "      <td>3.465000</td>\n",
       "      <td>78.943333</td>\n",
       "      <td>3.700000</td>\n",
       "      <td>133.698333</td>\n",
       "      <td>1014.648333</td>\n",
       "      <td>475.457844</td>\n",
       "      <td>-0.009625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11.0</th>\n",
       "      <td>109.427167</td>\n",
       "      <td>-4.868333</td>\n",
       "      <td>0.553333</td>\n",
       "      <td>-0.383333</td>\n",
       "      <td>-1.276667</td>\n",
       "      <td>241.326667</td>\n",
       "      <td>277.370000</td>\n",
       "      <td>276.700000</td>\n",
       "      <td>314.453333</td>\n",
       "      <td>274.423333</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-73.125000</td>\n",
       "      <td>-73.125000</td>\n",
       "      <td>1.351667</td>\n",
       "      <td>75.265000</td>\n",
       "      <td>3.646667</td>\n",
       "      <td>135.245000</td>\n",
       "      <td>1015.618333</td>\n",
       "      <td>539.708973</td>\n",
       "      <td>-0.009017</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                              zen  dw_solar  uw_solar  direct_n   diffuse  \\\n",
       "year   month day hour                                                       \n",
       "2010.0 1.0   1.0 7.0   157.281154 -4.348077  0.090385  0.150000 -1.123077   \n",
       "                 8.0   146.229333 -4.288333  0.113333  0.156667 -1.106667   \n",
       "                 9.0   133.906167 -4.328333  0.178333  0.701667 -1.108333   \n",
       "                 10.0  121.558667 -4.560000  0.316667 -0.558333 -1.170000   \n",
       "                 11.0  109.427167 -4.868333  0.553333 -0.383333 -1.276667   \n",
       "\n",
       "                            dw_ir  dw_casetemp  dw_dometemp       uw_ir  \\\n",
       "year   month day hour                                                     \n",
       "2010.0 1.0   1.0 7.0   320.623077   280.961538   280.546154  335.715385   \n",
       "                 8.0   322.270000   280.761667   280.345000  334.833333   \n",
       "                 9.0   318.151667   280.491667   280.046667  333.480000   \n",
       "                 10.0  300.935000   279.631667   279.101667  328.243333   \n",
       "                 11.0  241.326667   277.370000   276.700000  314.453333   \n",
       "\n",
       "                       uw_casetemp    ...     netsolar      netir   totalnet  \\\n",
       "year   month day hour                 ...                                      \n",
       "2010.0 1.0   1.0 7.0    277.776923    ...          0.0 -15.086538 -15.086538   \n",
       "                 8.0    277.581667    ...          0.0 -12.571667 -12.571667   \n",
       "                 9.0    277.301667    ...          0.0 -15.333333 -15.333333   \n",
       "                 10.0   276.451667    ...          0.0 -27.313333 -27.313333   \n",
       "                 11.0   274.423333    ...          0.0 -73.125000 -73.125000   \n",
       "\n",
       "                           temp         rh   windspd     winddir     pressure  \\\n",
       "year   month day hour                                                           \n",
       "2010.0 1.0   1.0 7.0   5.005769  79.425000  2.300000   69.582692  1013.025000   \n",
       "                 8.0   4.783333  75.178333  2.541667   84.746667  1013.470000   \n",
       "                 9.0   4.445000  76.843333  2.831667  159.561667  1013.861667   \n",
       "                 10.0  3.465000  78.943333  3.700000  133.698333  1014.648333   \n",
       "                 11.0  1.351667  75.265000  3.646667  135.245000  1015.618333   \n",
       "\n",
       "                              ghi        Kt  \n",
       "year   month day hour                        \n",
       "2010.0 1.0   1.0 7.0    58.673464 -0.092544  \n",
       "                 8.0   199.226885 -0.022905  \n",
       "                 9.0   357.002941 -0.012248  \n",
       "                 10.0  475.457844 -0.009625  \n",
       "                 11.0  539.708973 -0.009017  \n",
       "\n",
       "[5 rows x 23 columns]"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_new_train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Groupdata - test dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_zen = df_test.groupby(['month','day','hour'])['zen'].mean()\n",
    "test_dw_solar = df_test.groupby(['month','day','hour'])['dw_solar'].mean()\n",
    "test_uw_solar = df_test.groupby(['month','day','hour'])['uw_solar'].mean()\n",
    "test_direct_n = df_test.groupby(['month','day','hour'])['direct_n'].mean()\n",
    "test_diffuse = df_test.groupby(['month','day','hour'])['diffuse'].mean()\n",
    "test_dw_ir = df_test.groupby(['month','day','hour'])['dw_ir'].mean()\n",
    "test_dw_casetemp = df_test.groupby(['month','day','hour'])['dw_casetemp'].mean()\n",
    "test_dw_dometemp = df_test.groupby(['month','day','hour'])['dw_dometemp'].mean()\n",
    "test_uw_ir = df_test.groupby(['month','day','hour'])['uw_ir'].mean()\n",
    "test_uw_casetemp = df_test.groupby(['month','day','hour'])['uw_casetemp'].mean()\n",
    "test_uw_dometemp = df_test.groupby(['month','day','hour'])['uw_dometemp'].mean()\n",
    "test_uvb = df_test.groupby(['month','day','hour'])['uvb'].mean()\n",
    "test_par = df_test.groupby(['month','day','hour'])['par'].mean()\n",
    "test_netsolar = df_test.groupby(['month','day','hour'])['netsolar'].mean()\n",
    "test_netir = df_test.groupby(['month','day','hour'])['netir'].mean()\n",
    "test_totalnet = df_test.groupby(['month','day','hour'])['totalnet'].mean()\n",
    "test_temp = df_test.groupby(['month','day','hour'])['temp'].mean()\n",
    "test_rh = df_test.groupby(['month','day','hour'])['rh'].mean()\n",
    "test_windspd = df_test.groupby(['month','day','hour'])['windspd'].mean()\n",
    "test_winddir = df_test.groupby(['month','day','hour'])['winddir'].mean()\n",
    "test_pressure = df_test.groupby(['month','day','hour'])['pressure'].mean()\n",
    "test_ghi = df_test.groupby(['month','day','hour'])['ghi'].mean()\n",
    "test_Kt = df_test.groupby(['month','day','hour'])['Kt'].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_new_test = pd.concat([test_zen,test_dw_solar,test_uw_solar,test_direct_n,test_diffuse,test_dw_ir,\n",
    "                         test_dw_casetemp,test_dw_dometemp,test_uw_ir,test_uw_casetemp,test_uw_dometemp,\n",
    "                    test_uvb,test_par,test_netsolar,test_netir,test_totalnet,test_temp,test_rh,\n",
    "                         test_windspd,test_winddir,test_pressure,test_ghi,test_Kt], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "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>zen</th>\n",
       "      <th>dw_solar</th>\n",
       "      <th>uw_solar</th>\n",
       "      <th>direct_n</th>\n",
       "      <th>diffuse</th>\n",
       "      <th>dw_ir</th>\n",
       "      <th>dw_casetemp</th>\n",
       "      <th>dw_dometemp</th>\n",
       "      <th>uw_ir</th>\n",
       "      <th>uw_casetemp</th>\n",
       "      <th>...</th>\n",
       "      <th>netsolar</th>\n",
       "      <th>netir</th>\n",
       "      <th>totalnet</th>\n",
       "      <th>temp</th>\n",
       "      <th>rh</th>\n",
       "      <th>windspd</th>\n",
       "      <th>winddir</th>\n",
       "      <th>pressure</th>\n",
       "      <th>ghi</th>\n",
       "      <th>Kt</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hour</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>6.0</th>\n",
       "      <td>156.063125</td>\n",
       "      <td>-5.293750</td>\n",
       "      <td>-0.418750</td>\n",
       "      <td>1.537500</td>\n",
       "      <td>-0.381250</td>\n",
       "      <td>234.900000</td>\n",
       "      <td>272.150000</td>\n",
       "      <td>271.862500</td>\n",
       "      <td>291.925000</td>\n",
       "      <td>272.281250</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-57.043750</td>\n",
       "      <td>-57.043750</td>\n",
       "      <td>0.593750</td>\n",
       "      <td>61.012500</td>\n",
       "      <td>1.137500</td>\n",
       "      <td>116.625000</td>\n",
       "      <td>1013.318750</td>\n",
       "      <td>30.663084</td>\n",
       "      <td>-0.182511</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7.0</th>\n",
       "      <td>151.636833</td>\n",
       "      <td>-5.338333</td>\n",
       "      <td>-0.555000</td>\n",
       "      <td>1.555000</td>\n",
       "      <td>-0.473333</td>\n",
       "      <td>235.656667</td>\n",
       "      <td>272.565000</td>\n",
       "      <td>272.235000</td>\n",
       "      <td>291.751667</td>\n",
       "      <td>271.600000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-56.091667</td>\n",
       "      <td>-56.091667</td>\n",
       "      <td>0.088333</td>\n",
       "      <td>63.146667</td>\n",
       "      <td>0.618333</td>\n",
       "      <td>112.858333</td>\n",
       "      <td>1013.015000</td>\n",
       "      <td>115.766684</td>\n",
       "      <td>-0.058923</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8.0</th>\n",
       "      <td>141.881000</td>\n",
       "      <td>-5.150000</td>\n",
       "      <td>-0.876667</td>\n",
       "      <td>0.881667</td>\n",
       "      <td>-0.500000</td>\n",
       "      <td>236.418333</td>\n",
       "      <td>272.275000</td>\n",
       "      <td>271.936667</td>\n",
       "      <td>291.446667</td>\n",
       "      <td>271.781667</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-55.028333</td>\n",
       "      <td>-55.028333</td>\n",
       "      <td>0.830000</td>\n",
       "      <td>57.490000</td>\n",
       "      <td>1.448333</td>\n",
       "      <td>132.271667</td>\n",
       "      <td>1011.976667</td>\n",
       "      <td>310.197139</td>\n",
       "      <td>-0.017190</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9.0</th>\n",
       "      <td>130.346167</td>\n",
       "      <td>-5.693333</td>\n",
       "      <td>-0.785000</td>\n",
       "      <td>2.293333</td>\n",
       "      <td>-0.325000</td>\n",
       "      <td>242.050000</td>\n",
       "      <td>272.941667</td>\n",
       "      <td>272.681667</td>\n",
       "      <td>295.341667</td>\n",
       "      <td>273.095000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-53.295000</td>\n",
       "      <td>-53.295000</td>\n",
       "      <td>1.676667</td>\n",
       "      <td>53.623333</td>\n",
       "      <td>2.033333</td>\n",
       "      <td>137.526667</td>\n",
       "      <td>1011.566667</td>\n",
       "      <td>491.373002</td>\n",
       "      <td>-0.011644</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10.0</th>\n",
       "      <td>118.142333</td>\n",
       "      <td>-5.468333</td>\n",
       "      <td>-0.343333</td>\n",
       "      <td>2.310000</td>\n",
       "      <td>-0.261667</td>\n",
       "      <td>293.071667</td>\n",
       "      <td>275.710000</td>\n",
       "      <td>275.516667</td>\n",
       "      <td>308.071667</td>\n",
       "      <td>273.635000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-16.146667</td>\n",
       "      <td>-16.146667</td>\n",
       "      <td>1.591667</td>\n",
       "      <td>56.341667</td>\n",
       "      <td>1.473333</td>\n",
       "      <td>151.643333</td>\n",
       "      <td>1011.851667</td>\n",
       "      <td>628.392880</td>\n",
       "      <td>-0.008778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11.0</th>\n",
       "      <td>105.758667</td>\n",
       "      <td>-4.468333</td>\n",
       "      <td>-0.058333</td>\n",
       "      <td>1.616667</td>\n",
       "      <td>-0.338333</td>\n",
       "      <td>321.446667</td>\n",
       "      <td>277.086667</td>\n",
       "      <td>276.876667</td>\n",
       "      <td>315.970000</td>\n",
       "      <td>273.856667</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.773333</td>\n",
       "      <td>59.400000</td>\n",
       "      <td>0.791667</td>\n",
       "      <td>162.733333</td>\n",
       "      <td>1012.278333</td>\n",
       "      <td>707.999345</td>\n",
       "      <td>-0.006309</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12.0</th>\n",
       "      <td>93.360000</td>\n",
       "      <td>-5.085000</td>\n",
       "      <td>0.435000</td>\n",
       "      <td>0.523333</td>\n",
       "      <td>0.573333</td>\n",
       "      <td>324.706667</td>\n",
       "      <td>277.511667</td>\n",
       "      <td>277.048333</td>\n",
       "      <td>318.663333</td>\n",
       "      <td>274.298333</td>\n",
       "      <td>...</td>\n",
       "      <td>0.543333</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.543333</td>\n",
       "      <td>2.150000</td>\n",
       "      <td>62.286667</td>\n",
       "      <td>1.993333</td>\n",
       "      <td>152.851667</td>\n",
       "      <td>1012.153333</td>\n",
       "      <td>723.565963</td>\n",
       "      <td>-0.007017</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13.0</th>\n",
       "      <td>81.477500</td>\n",
       "      <td>10.581667</td>\n",
       "      <td>3.041667</td>\n",
       "      <td>0.323333</td>\n",
       "      <td>15.745000</td>\n",
       "      <td>326.311667</td>\n",
       "      <td>277.568333</td>\n",
       "      <td>277.051667</td>\n",
       "      <td>320.978333</td>\n",
       "      <td>274.485000</td>\n",
       "      <td>...</td>\n",
       "      <td>12.793333</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>12.793333</td>\n",
       "      <td>2.198333</td>\n",
       "      <td>71.641667</td>\n",
       "      <td>2.133333</td>\n",
       "      <td>160.955000</td>\n",
       "      <td>1012.345000</td>\n",
       "      <td>673.851495</td>\n",
       "      <td>0.015885</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14.0</th>\n",
       "      <td>70.368167</td>\n",
       "      <td>44.298333</td>\n",
       "      <td>7.703333</td>\n",
       "      <td>1.603333</td>\n",
       "      <td>47.981667</td>\n",
       "      <td>327.441667</td>\n",
       "      <td>277.428333</td>\n",
       "      <td>276.891667</td>\n",
       "      <td>323.908333</td>\n",
       "      <td>274.638333</td>\n",
       "      <td>...</td>\n",
       "      <td>40.848333</td>\n",
       "      <td>-0.015000</td>\n",
       "      <td>40.835000</td>\n",
       "      <td>2.188333</td>\n",
       "      <td>82.856667</td>\n",
       "      <td>2.678333</td>\n",
       "      <td>166.960000</td>\n",
       "      <td>1012.361667</td>\n",
       "      <td>562.891372</td>\n",
       "      <td>0.082049</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15.0</th>\n",
       "      <td>60.402833</td>\n",
       "      <td>56.543333</td>\n",
       "      <td>8.775000</td>\n",
       "      <td>0.658333</td>\n",
       "      <td>59.681667</td>\n",
       "      <td>331.398333</td>\n",
       "      <td>278.208333</td>\n",
       "      <td>277.673333</td>\n",
       "      <td>327.438333</td>\n",
       "      <td>275.178333</td>\n",
       "      <td>...</td>\n",
       "      <td>51.280000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>51.280000</td>\n",
       "      <td>2.543333</td>\n",
       "      <td>89.940000</td>\n",
       "      <td>2.470000</td>\n",
       "      <td>150.240000</td>\n",
       "      <td>1012.115000</td>\n",
       "      <td>400.469352</td>\n",
       "      <td>0.149594</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16.0</th>\n",
       "      <td>52.384000</td>\n",
       "      <td>187.235000</td>\n",
       "      <td>29.745000</td>\n",
       "      <td>2.215000</td>\n",
       "      <td>185.133333</td>\n",
       "      <td>335.448333</td>\n",
       "      <td>279.595000</td>\n",
       "      <td>279.348333</td>\n",
       "      <td>337.640000</td>\n",
       "      <td>276.196667</td>\n",
       "      <td>...</td>\n",
       "      <td>156.750000</td>\n",
       "      <td>-2.490000</td>\n",
       "      <td>154.268333</td>\n",
       "      <td>3.485000</td>\n",
       "      <td>91.268333</td>\n",
       "      <td>3.808333</td>\n",
       "      <td>153.068333</td>\n",
       "      <td>1011.073333</td>\n",
       "      <td>205.793937</td>\n",
       "      <td>1.063090</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17.0</th>\n",
       "      <td>47.737500</td>\n",
       "      <td>163.261364</td>\n",
       "      <td>27.654545</td>\n",
       "      <td>0.834091</td>\n",
       "      <td>162.845455</td>\n",
       "      <td>339.390909</td>\n",
       "      <td>281.243182</td>\n",
       "      <td>280.965909</td>\n",
       "      <td>342.279545</td>\n",
       "      <td>277.679545</td>\n",
       "      <td>...</td>\n",
       "      <td>135.852273</td>\n",
       "      <td>-3.006818</td>\n",
       "      <td>132.838636</td>\n",
       "      <td>4.775000</td>\n",
       "      <td>86.770455</td>\n",
       "      <td>4.115909</td>\n",
       "      <td>165.429545</td>\n",
       "      <td>1010.206818</td>\n",
       "      <td>55.699158</td>\n",
       "      <td>3.182868</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>12 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             zen    dw_solar   uw_solar  direct_n     diffuse       dw_ir  \\\n",
       "hour                                                                        \n",
       "6.0   156.063125   -5.293750  -0.418750  1.537500   -0.381250  234.900000   \n",
       "7.0   151.636833   -5.338333  -0.555000  1.555000   -0.473333  235.656667   \n",
       "8.0   141.881000   -5.150000  -0.876667  0.881667   -0.500000  236.418333   \n",
       "9.0   130.346167   -5.693333  -0.785000  2.293333   -0.325000  242.050000   \n",
       "10.0  118.142333   -5.468333  -0.343333  2.310000   -0.261667  293.071667   \n",
       "11.0  105.758667   -4.468333  -0.058333  1.616667   -0.338333  321.446667   \n",
       "12.0   93.360000   -5.085000   0.435000  0.523333    0.573333  324.706667   \n",
       "13.0   81.477500   10.581667   3.041667  0.323333   15.745000  326.311667   \n",
       "14.0   70.368167   44.298333   7.703333  1.603333   47.981667  327.441667   \n",
       "15.0   60.402833   56.543333   8.775000  0.658333   59.681667  331.398333   \n",
       "16.0   52.384000  187.235000  29.745000  2.215000  185.133333  335.448333   \n",
       "17.0   47.737500  163.261364  27.654545  0.834091  162.845455  339.390909   \n",
       "\n",
       "      dw_casetemp  dw_dometemp       uw_ir  uw_casetemp    ...       netsolar  \\\n",
       "hour                                                       ...                  \n",
       "6.0    272.150000   271.862500  291.925000   272.281250    ...       0.000000   \n",
       "7.0    272.565000   272.235000  291.751667   271.600000    ...       0.000000   \n",
       "8.0    272.275000   271.936667  291.446667   271.781667    ...       0.000000   \n",
       "9.0    272.941667   272.681667  295.341667   273.095000    ...       0.000000   \n",
       "10.0   275.710000   275.516667  308.071667   273.635000    ...       0.000000   \n",
       "11.0   277.086667   276.876667  315.970000   273.856667    ...       0.000000   \n",
       "12.0   277.511667   277.048333  318.663333   274.298333    ...       0.543333   \n",
       "13.0   277.568333   277.051667  320.978333   274.485000    ...      12.793333   \n",
       "14.0   277.428333   276.891667  323.908333   274.638333    ...      40.848333   \n",
       "15.0   278.208333   277.673333  327.438333   275.178333    ...      51.280000   \n",
       "16.0   279.595000   279.348333  337.640000   276.196667    ...     156.750000   \n",
       "17.0   281.243182   280.965909  342.279545   277.679545    ...     135.852273   \n",
       "\n",
       "          netir    totalnet      temp         rh   windspd     winddir  \\\n",
       "hour                                                                     \n",
       "6.0  -57.043750  -57.043750  0.593750  61.012500  1.137500  116.625000   \n",
       "7.0  -56.091667  -56.091667  0.088333  63.146667  0.618333  112.858333   \n",
       "8.0  -55.028333  -55.028333  0.830000  57.490000  1.448333  132.271667   \n",
       "9.0  -53.295000  -53.295000  1.676667  53.623333  2.033333  137.526667   \n",
       "10.0 -16.146667  -16.146667  1.591667  56.341667  1.473333  151.643333   \n",
       "11.0   0.000000    0.000000  1.773333  59.400000  0.791667  162.733333   \n",
       "12.0   0.000000    0.543333  2.150000  62.286667  1.993333  152.851667   \n",
       "13.0   0.000000   12.793333  2.198333  71.641667  2.133333  160.955000   \n",
       "14.0  -0.015000   40.835000  2.188333  82.856667  2.678333  166.960000   \n",
       "15.0   0.000000   51.280000  2.543333  89.940000  2.470000  150.240000   \n",
       "16.0  -2.490000  154.268333  3.485000  91.268333  3.808333  153.068333   \n",
       "17.0  -3.006818  132.838636  4.775000  86.770455  4.115909  165.429545   \n",
       "\n",
       "         pressure         ghi        Kt  \n",
       "hour                                     \n",
       "6.0   1013.318750   30.663084 -0.182511  \n",
       "7.0   1013.015000  115.766684 -0.058923  \n",
       "8.0   1011.976667  310.197139 -0.017190  \n",
       "9.0   1011.566667  491.373002 -0.011644  \n",
       "10.0  1011.851667  628.392880 -0.008778  \n",
       "11.0  1012.278333  707.999345 -0.006309  \n",
       "12.0  1012.153333  723.565963 -0.007017  \n",
       "13.0  1012.345000  673.851495  0.015885  \n",
       "14.0  1012.361667  562.891372  0.082049  \n",
       "15.0  1012.115000  400.469352  0.149594  \n",
       "16.0  1011.073333  205.793937  1.063090  \n",
       "17.0  1010.206818   55.699158  3.182868  \n",
       "\n",
       "[12 rows x 23 columns]"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_new_test.loc[2].xs(17,level='day')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Shifting Kt values to make 1 hour ahead forecast"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Train dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "levels_index= []\n",
    "for m in df_new_train.index.levels:\n",
    "    levels_index.append(m)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "for i in levels_index[0]:\n",
    "    for j in levels_index[1]:\n",
    "        df_new_train.loc[i].loc[j]['Kt'] = df_new_train.loc[i].loc[j]['Kt'].shift(-2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_new_train = df_new_train[~(df_new_train['Kt'].isnull())]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Test dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "levels_index2= []\n",
    "for m in df_new_test.index.levels:\n",
    "    levels_index2.append(m)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in levels_index2[0]:\n",
    "    for j in levels_index2[1]:\n",
    "        df_new_test.loc[i].loc[j]['Kt'] = df_new_test.loc[i].loc[j]['Kt'].shift(-2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_new_test = df_new_test[~(df_new_test['Kt'].isnull())]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>zen</th>\n",
       "      <th>dw_solar</th>\n",
       "      <th>uw_solar</th>\n",
       "      <th>direct_n</th>\n",
       "      <th>diffuse</th>\n",
       "      <th>dw_ir</th>\n",
       "      <th>dw_casetemp</th>\n",
       "      <th>dw_dometemp</th>\n",
       "      <th>uw_ir</th>\n",
       "      <th>uw_casetemp</th>\n",
       "      <th>...</th>\n",
       "      <th>netsolar</th>\n",
       "      <th>netir</th>\n",
       "      <th>totalnet</th>\n",
       "      <th>temp</th>\n",
       "      <th>rh</th>\n",
       "      <th>windspd</th>\n",
       "      <th>winddir</th>\n",
       "      <th>pressure</th>\n",
       "      <th>ghi</th>\n",
       "      <th>Kt</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>month</th>\n",
       "      <th>day</th>\n",
       "      <th>hour</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>7.0</th>\n",
       "      <th>31.0</th>\n",
       "      <th>18.0</th>\n",
       "      <td>17.352667</td>\n",
       "      <td>804.265</td>\n",
       "      <td>143.933333</td>\n",
       "      <td>474.62</td>\n",
       "      <td>338.386667</td>\n",
       "      <td>422.245</td>\n",
       "      <td>308.041667</td>\n",
       "      <td>307.96</td>\n",
       "      <td>482.356667</td>\n",
       "      <td>303.85</td>\n",
       "      <td>...</td>\n",
       "      <td>646.77</td>\n",
       "      <td>-60.106667</td>\n",
       "      <td>586.668333</td>\n",
       "      <td>28.04</td>\n",
       "      <td>64.46</td>\n",
       "      <td>1.555</td>\n",
       "      <td>260.845</td>\n",
       "      <td>1005.741667</td>\n",
       "      <td>213.182009</td>\n",
       "      <td>1801.872849</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                       zen  dw_solar    uw_solar  direct_n     diffuse  \\\n",
       "month day  hour                                                          \n",
       "7.0   31.0 18.0  17.352667   804.265  143.933333    474.62  338.386667   \n",
       "\n",
       "                   dw_ir  dw_casetemp  dw_dometemp       uw_ir  uw_casetemp  \\\n",
       "month day  hour                                                               \n",
       "7.0   31.0 18.0  422.245   308.041667       307.96  482.356667       303.85   \n",
       "\n",
       "                    ...       netsolar      netir    totalnet   temp     rh  \\\n",
       "month day  hour     ...                                                       \n",
       "7.0   31.0 18.0     ...         646.77 -60.106667  586.668333  28.04  64.46   \n",
       "\n",
       "                 windspd  winddir     pressure         ghi           Kt  \n",
       "month day  hour                                                          \n",
       "7.0   31.0 18.0    1.555  260.845  1005.741667  213.182009  1801.872849  \n",
       "\n",
       "[1 rows x 23 columns]"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_new_test[df_new_test['Kt']==df_new_test['Kt'].max()]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Normalize train and test dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_norm = (df_new_train - df_new_train.mean()) / (df_new_train.max() - df_new_train.min())\n",
    "test_norm =  (df_new_test - df_new_test.mean()) / (df_new_test.max() - df_new_test.min())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_norm.reset_index(inplace=True,drop=True)\n",
    "test_norm.reset_index(inplace=True,drop=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "### Making train and test sets with train_norm and test_norm"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "#### finding the gcf (greatest common factor) of train and test dataset's length and chop off the extra rows to make it divisible with the batchsize"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\s269596\\AppData\\Local\\Continuum\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:2: DeprecationWarning: fractions.gcd() is deprecated. Use math.gcd() instead.\n",
      "  \n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from fractions import gcd\n",
    "gcd(train_norm.shape[0],test_norm.shape[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "import math\n",
    "def roundup(x):\n",
    "    return int(math.ceil(x / 100.0)) * 100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_lim = roundup(train_norm.shape[0])\n",
    "test_lim = roundup(test_norm.shape[0])\n",
    "\n",
    "train_random = train_norm.sample(train_lim-train_norm.shape[0])\n",
    "test_random = test_norm.sample(test_lim-test_norm.shape[0])\n",
    "\n",
    "train_norm = train_norm.append(train_random)\n",
    "test_norm = test_norm.append(test_random)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "X1 = train_norm.drop('Kt',axis=1)\n",
    "y1 = train_norm['Kt']\n",
    "\n",
    "X2 = test_norm.drop('Kt',axis=1)\n",
    "y2 = test_norm['Kt']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X1_train shape is (9000, 22)\n",
      "y1_train shape is (9000,)\n",
      "X2_test shape is (3900, 22)\n",
      "y2_test shape is (3900,)\n"
     ]
    }
   ],
   "source": [
    "print(\"X1_train shape is {}\".format(X1.shape))\n",
    "print(\"y1_train shape is {}\".format(y1.shape))\n",
    "print(\"X2_test shape is {}\".format(X2.shape))\n",
    "print(\"y2_test shape is {}\".format(y2.shape))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = np.array(X1)\n",
    "y_train  = np.array(y1)\n",
    "X_test = np.array(X2)\n",
    "y_test = np.array(y2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### start of RNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torchvision.transforms as transforms\n",
    "import torchvision.datasets as dsets\n",
    "from torch.autograd import Variable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "class RNNModel(nn.Module):\n",
    "    def __init__(self, input_dim, hidden_dim, layer_dim, output_dim):\n",
    "        super(RNNModel, self).__init__()\n",
    "        #Hidden Dimension\n",
    "        self.hidden_dim = hidden_dim\n",
    "        \n",
    "        # Number of hidden layers\n",
    "        self.layer_dim = layer_dim\n",
    "        \n",
    "        #Building the RNN\n",
    "        self.rnn = nn.RNN(input_dim, hidden_dim, layer_dim, batch_first=True, nonlinearity='relu')\n",
    "        \n",
    "        # Readout layer\n",
    "        self.fc = nn.Linear(hidden_dim, output_dim)\n",
    "        \n",
    "    def forward(self, x):\n",
    "        # Initializing the hidden state with zeros\n",
    "        # (layer_dim, batch_size, hidden_dim)\n",
    "        h0 = Variable(torch.zeros(self.layer_dim, x.size(0), self.hidden_dim))\n",
    "        \n",
    "        #One time step (the last one perhaps?)\n",
    "        out, hn = self.rnn(x, h0)\n",
    "        \n",
    "        # Indexing hidden state of the last time step\n",
    "        # out.size() --> ??\n",
    "        #out[:,-1,:] --> is it going to be 100,100\n",
    "        out = self.fc(out[:,-1,:])\n",
    "        # out.size() --> 100,1\n",
    "        return out\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Instantiating Model Class\n",
    "input_dim = 22\n",
    "hidden_dim = 15\n",
    "layer_dim = 1\n",
    "output_dim = 1\n",
    "batch_size = 100\n",
    "\n",
    "model = RNNModel(input_dim, hidden_dim, layer_dim, output_dim)\n",
    "\n",
    "# Instantiating Loss Class\n",
    "criterion = nn.MSELoss()\n",
    "\n",
    "# Instantiate Optimizer Class\n",
    "learning_rate = 0.001\n",
    "optimizer = torch.optim.SGD(model.parameters(), lr=learning_rate)\n",
    "\n",
    "# converting numpy array to torch tensor\n",
    "X_train = torch.from_numpy(X_train)\n",
    "y_train = torch.from_numpy(y_train)\n",
    "X_test = torch.from_numpy(X_test)\n",
    "y_test = torch.from_numpy(y_test)\n",
    "\n",
    "# initializing lists to store losses over epochs:\n",
    "train_loss = []\n",
    "test_loss = []\n",
    "train_iter = []\n",
    "test_iter = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 0 Iteration: 0. Train_MSE: 0.07102621346712112. Test_MSE: 0.002616565671594597\n",
      "Epoch: 1 Iteration: 100. Train_MSE: 0.04818740859627724. Test_MSE: 0.0017127833451811853\n",
      "Epoch: 2 Iteration: 200. Train_MSE: 0.032102301716804504. Test_MSE: 0.0010895137739285131\n",
      "Epoch: 3 Iteration: 300. Train_MSE: 0.015069985762238503. Test_MSE: 0.0006701087734201043\n",
      "Epoch: 4 Iteration: 400. Train_MSE: 0.001872098888270557. Test_MSE: 0.00042209753473087755\n",
      "Epoch: 5 Iteration: 500. Train_MSE: 0.003314315341413021. Test_MSE: 0.0002986154289320017\n",
      "Epoch: 6 Iteration: 600. Train_MSE: 0.007565425243228674. Test_MSE: 0.00027366245757225345\n",
      "Epoch: 7 Iteration: 700. Train_MSE: 0.0017244565533474088. Test_MSE: 0.00030086124995269887\n",
      "Epoch: 8 Iteration: 800. Train_MSE: 0.0004533612809609622. Test_MSE: 0.00033080792318494696\n",
      "Epoch: 10 Iteration: 900. Train_MSE: 0.0023514230269938707. Test_MSE: 0.00035154160463967196\n",
      "Epoch: 11 Iteration: 1000. Train_MSE: 0.0044293953105807304. Test_MSE: 0.00037339692407983225\n",
      "Epoch: 12 Iteration: 1100. Train_MSE: 0.003608967876061797. Test_MSE: 0.0003912210387469892\n",
      "Epoch: 13 Iteration: 1200. Train_MSE: 0.0005701781483367085. Test_MSE: 0.00039324139187611073\n",
      "Epoch: 14 Iteration: 1300. Train_MSE: 0.0010083867236971855. Test_MSE: 0.00039287478881694664\n",
      "Epoch: 15 Iteration: 1400. Train_MSE: 0.0028091459535062313. Test_MSE: 0.00039635923103968553\n",
      "Epoch: 16 Iteration: 1500. Train_MSE: 0.001920151524245739. Test_MSE: 0.00039971476463845554\n",
      "Epoch: 17 Iteration: 1600. Train_MSE: 0.0006401970167644322. Test_MSE: 0.0003956040479848542\n",
      "Epoch: 19 Iteration: 1700. Train_MSE: 0.0014413599856197834. Test_MSE: 0.0003875558203438552\n",
      "Epoch: 20 Iteration: 1800. Train_MSE: 0.0022299436386674643. Test_MSE: 0.00038694096518842203\n",
      "Epoch: 21 Iteration: 1900. Train_MSE: 0.0016140778316184878. Test_MSE: 0.00038876492977413636\n",
      "Epoch: 22 Iteration: 2000. Train_MSE: 0.0012325349962338805. Test_MSE: 0.0003816465010603325\n",
      "Epoch: 23 Iteration: 2100. Train_MSE: 0.0010695158271118999. Test_MSE: 0.000375128914829625\n",
      "Epoch: 24 Iteration: 2200. Train_MSE: 0.0018752493197098374. Test_MSE: 0.0003730981427060384\n",
      "Epoch: 25 Iteration: 2300. Train_MSE: 0.0014407822163775563. Test_MSE: 0.00037293585528344603\n",
      "Epoch: 26 Iteration: 2400. Train_MSE: 0.00043363674194552004. Test_MSE: 0.0003679264647131213\n",
      "Epoch: 28 Iteration: 2500. Train_MSE: 0.0006601743516512215. Test_MSE: 0.0003609885554708832\n",
      "Epoch: 29 Iteration: 2600. Train_MSE: 0.002052027964964509. Test_MSE: 0.000358776481432988\n",
      "Epoch: 30 Iteration: 2700. Train_MSE: 0.0014215355040505528. Test_MSE: 0.0003598309118032179\n",
      "Epoch: 31 Iteration: 2800. Train_MSE: 0.0008599397260695696. Test_MSE: 0.0003547212122437017\n",
      "Epoch: 32 Iteration: 2900. Train_MSE: 0.00045770886936224997. Test_MSE: 0.00034907151324406184\n",
      "Epoch: 33 Iteration: 3000. Train_MSE: 0.0015887366607785225. Test_MSE: 0.0003471599753202232\n",
      "Epoch: 34 Iteration: 3100. Train_MSE: 0.0016259914264082909. Test_MSE: 0.0003470239325583537\n",
      "Epoch: 35 Iteration: 3200. Train_MSE: 0.0002332744188606739. Test_MSE: 0.0003434724834452564\n",
      "Epoch: 37 Iteration: 3300. Train_MSE: 0.0011399490758776665. Test_MSE: 0.0003384915300987705\n",
      "Epoch: 38 Iteration: 3400. Train_MSE: 0.0018335336353629827. Test_MSE: 0.00033600919488899233\n",
      "Epoch: 39 Iteration: 3500. Train_MSE: 0.0018240265781059861. Test_MSE: 0.0003372393755991921\n",
      "Epoch: 40 Iteration: 3600. Train_MSE: 0.00025576248299330473. Test_MSE: 0.0003338247367555408\n",
      "Epoch: 41 Iteration: 3700. Train_MSE: 0.0005192547105252743. Test_MSE: 0.00032945581151947264\n",
      "Epoch: 42 Iteration: 3800. Train_MSE: 0.0021124589256942272. Test_MSE: 0.00032775768995742054\n",
      "Epoch: 43 Iteration: 3900. Train_MSE: 0.0009881570003926754. Test_MSE: 0.0003278437424821244\n",
      "Epoch: 44 Iteration: 4000. Train_MSE: 0.0005815416807308793. Test_MSE: 0.00032520799704383\n",
      "Epoch: 46 Iteration: 4100. Train_MSE: 0.0006175626185722649. Test_MSE: 0.00032210211828452806\n",
      "Epoch: 47 Iteration: 4200. Train_MSE: 0.0008044940768741071. Test_MSE: 0.0003194928737944514\n",
      "Epoch: 48 Iteration: 4300. Train_MSE: 0.0014393961755558848. Test_MSE: 0.00032044283538331704\n",
      "Epoch: 49 Iteration: 4400. Train_MSE: 0.0007242759456858039. Test_MSE: 0.0003180044659904182\n",
      "Epoch: 50 Iteration: 4500. Train_MSE: 0.00034258130472153425. Test_MSE: 0.00031489355872170086\n",
      "Epoch: 51 Iteration: 4600. Train_MSE: 0.0012142352061346173. Test_MSE: 0.0003135139787797188\n",
      "Epoch: 52 Iteration: 4700. Train_MSE: 0.0006440460565499961. Test_MSE: 0.0003134720988663905\n",
      "Epoch: 53 Iteration: 4800. Train_MSE: 0.00028478604508563876. Test_MSE: 0.00031189854844134\n",
      "Epoch: 55 Iteration: 4900. Train_MSE: 0.0007957136840559542. Test_MSE: 0.0003096498797610492\n",
      "Epoch: 56 Iteration: 5000. Train_MSE: 0.00349333044141531. Test_MSE: 0.0003072294961847591\n",
      "Epoch: 57 Iteration: 5100. Train_MSE: 0.0006226516561582685. Test_MSE: 0.00030780630210678733\n",
      "Epoch: 58 Iteration: 5200. Train_MSE: 0.001488480600528419. Test_MSE: 0.000306453530656783\n",
      "Epoch: 59 Iteration: 5300. Train_MSE: 0.0006176016759127378. Test_MSE: 0.0003038629373478297\n",
      "Epoch: 60 Iteration: 5400. Train_MSE: 0.0012090046657249331. Test_MSE: 0.0003024636018194962\n",
      "Epoch: 61 Iteration: 5500. Train_MSE: 0.0009699916117824614. Test_MSE: 0.00030262942272357094\n",
      "Epoch: 62 Iteration: 5600. Train_MSE: 0.0006060784799046814. Test_MSE: 0.00030150716596989794\n",
      "Epoch: 64 Iteration: 5700. Train_MSE: 0.0007580694509670138. Test_MSE: 0.0003001753689653357\n",
      "Epoch: 65 Iteration: 5800. Train_MSE: 0.002006145194172859. Test_MSE: 0.0002979624412435585\n",
      "Epoch: 66 Iteration: 5900. Train_MSE: 0.0007198739913292229. Test_MSE: 0.000298117055066631\n",
      "Epoch: 67 Iteration: 6000. Train_MSE: 0.0006208755075931549. Test_MSE: 0.0002973840384710902\n",
      "Epoch: 68 Iteration: 6100. Train_MSE: 0.0005322492215782404. Test_MSE: 0.00029519335880687204\n",
      "Epoch: 69 Iteration: 6200. Train_MSE: 0.0007017983007244766. Test_MSE: 0.00029404085300998743\n",
      "Epoch: 70 Iteration: 6300. Train_MSE: 0.008544504642486572. Test_MSE: 0.0002939333863125462\n",
      "Epoch: 71 Iteration: 6400. Train_MSE: 0.000846377806738019. Test_MSE: 0.00029322837875502016\n",
      "Epoch: 73 Iteration: 6500. Train_MSE: 0.00029486120911315084. Test_MSE: 0.0002924974942864172\n",
      "Epoch: 74 Iteration: 6600. Train_MSE: 0.0005307464743964374. Test_MSE: 0.00029030930594510035\n",
      "Epoch: 75 Iteration: 6700. Train_MSE: 0.01316571794450283. Test_MSE: 0.0002902219629493129\n",
      "Epoch: 76 Iteration: 6800. Train_MSE: 0.0007554935873486102. Test_MSE: 0.0002899766381015293\n",
      "Epoch: 77 Iteration: 6900. Train_MSE: 0.0008643235778436065. Test_MSE: 0.0002878516734496557\n",
      "Epoch: 78 Iteration: 7000. Train_MSE: 0.0007870603585615754. Test_MSE: 0.0002870312676938508\n",
      "Epoch: 79 Iteration: 7100. Train_MSE: 0.001295192982070148. Test_MSE: 0.00028696359476692527\n",
      "Epoch: 80 Iteration: 7200. Train_MSE: 0.0005922724376432598. Test_MSE: 0.000286562954246255\n",
      "Epoch: 82 Iteration: 7300. Train_MSE: 0.0009477842831984162. Test_MSE: 0.00028585000676606436\n",
      "Epoch: 83 Iteration: 7400. Train_MSE: 0.0007081324001774192. Test_MSE: 0.0002838594702457034\n",
      "Epoch: 84 Iteration: 7500. Train_MSE: 0.0014495992800220847. Test_MSE: 0.00028388195738451757\n",
      "Epoch: 85 Iteration: 7600. Train_MSE: 0.0004179048410151154. Test_MSE: 0.0002838581031889452\n",
      "Epoch: 86 Iteration: 7700. Train_MSE: 0.0003096265427302569. Test_MSE: 0.0002819436644377193\n",
      "Epoch: 87 Iteration: 7800. Train_MSE: 0.0004005471128039062. Test_MSE: 0.0002812207197177176\n",
      "Epoch: 88 Iteration: 7900. Train_MSE: 0.0013357080752030015. Test_MSE: 0.0002810317505457886\n",
      "Epoch: 89 Iteration: 8000. Train_MSE: 0.0002966810716316104. Test_MSE: 0.00028075032740901924\n",
      "Epoch: 91 Iteration: 8100. Train_MSE: 0.0005872498732060194. Test_MSE: 0.00028039538630238083\n",
      "Epoch: 92 Iteration: 8200. Train_MSE: 0.0008076991070993245. Test_MSE: 0.00027853510145984514\n",
      "Epoch: 93 Iteration: 8300. Train_MSE: 0.0010569870937615633. Test_MSE: 0.0002783146703345253\n",
      "Epoch: 94 Iteration: 8400. Train_MSE: 0.0006355616496875882. Test_MSE: 0.0002785433675713286\n",
      "Epoch: 95 Iteration: 8500. Train_MSE: 0.001533408765681088. Test_MSE: 0.0002767362630447465\n",
      "Epoch: 96 Iteration: 8600. Train_MSE: 0.0006976796430535614. Test_MSE: 0.0002761602654910329\n",
      "Epoch: 97 Iteration: 8700. Train_MSE: 0.0009095440036617219. Test_MSE: 0.0002758205918577466\n",
      "Epoch: 98 Iteration: 8800. Train_MSE: 0.0005327793769538403. Test_MSE: 0.00027560250700113455\n",
      "Epoch: 100 Iteration: 8900. Train_MSE: 0.0007809080998413265. Test_MSE: 0.0002754630904865886\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 101 Iteration: 9000. Train_MSE: 0.00027039804263040423. Test_MSE: 0.0002738220768785274\n",
      "Epoch: 102 Iteration: 9100. Train_MSE: 0.0007372400723397732. Test_MSE: 0.000273459988002357\n",
      "Epoch: 103 Iteration: 9200. Train_MSE: 0.0005497653619386256. Test_MSE: 0.00027373653760794976\n",
      "Epoch: 104 Iteration: 9300. Train_MSE: 0.000511855585500598. Test_MSE: 0.0002721888599591102\n",
      "Epoch: 105 Iteration: 9400. Train_MSE: 0.0005754136363975704. Test_MSE: 0.00027162687531975344\n",
      "Epoch: 106 Iteration: 9500. Train_MSE: 0.0010389707749709487. Test_MSE: 0.0002712090164594662\n",
      "Epoch: 107 Iteration: 9600. Train_MSE: 0.0005902088014408946. Test_MSE: 0.00027107776665497593\n",
      "Epoch: 108 Iteration: 9700. Train_MSE: 0.000554964761249721. Test_MSE: 0.0002707857240461164\n",
      "Epoch: 110 Iteration: 9800. Train_MSE: 0.0005419874214567244. Test_MSE: 0.00026942382198095\n",
      "Epoch: 111 Iteration: 9900. Train_MSE: 0.0008617540588602424. Test_MSE: 0.00026901384077881064\n",
      "Epoch: 112 Iteration: 10000. Train_MSE: 0.0010139583609998226. Test_MSE: 0.0002692944760168529\n",
      "Epoch: 113 Iteration: 10100. Train_MSE: 0.00033736974000930786. Test_MSE: 0.0002679357959387593\n",
      "Epoch: 114 Iteration: 10200. Train_MSE: 0.0005416656495071948. Test_MSE: 0.0002674989104485493\n",
      "Epoch: 115 Iteration: 10300. Train_MSE: 0.0008909296011552215. Test_MSE: 0.0002669668216511023\n",
      "Epoch: 116 Iteration: 10400. Train_MSE: 0.00039618604932911694. Test_MSE: 0.0002667734287314579\n",
      "Epoch: 117 Iteration: 10500. Train_MSE: 0.0010714492527768016. Test_MSE: 0.00026650605880161854\n",
      "Epoch: 119 Iteration: 10600. Train_MSE: 0.0006193164153955877. Test_MSE: 0.0002653993328940377\n",
      "Epoch: 120 Iteration: 10700. Train_MSE: 0.0004583821864798665. Test_MSE: 0.0002648564698375886\n",
      "Epoch: 121 Iteration: 10800. Train_MSE: 0.0005190403899177909. Test_MSE: 0.00026517412865470236\n",
      "Epoch: 122 Iteration: 10900. Train_MSE: 0.00042784298420883715. Test_MSE: 0.0002639383619667742\n",
      "Epoch: 123 Iteration: 11000. Train_MSE: 0.00034639108343981206. Test_MSE: 0.0002635926183450335\n",
      "Epoch: 124 Iteration: 11100. Train_MSE: 0.0006778863025829196. Test_MSE: 0.0002630513866110862\n",
      "Epoch: 125 Iteration: 11200. Train_MSE: 0.0005331965512596071. Test_MSE: 0.00026290655673153726\n",
      "Epoch: 126 Iteration: 11300. Train_MSE: 0.0004485942190513015. Test_MSE: 0.0002624060101551282\n",
      "Epoch: 128 Iteration: 11400. Train_MSE: 0.00033739354694262147. Test_MSE: 0.0002617530878317075\n",
      "Epoch: 129 Iteration: 11500. Train_MSE: 0.0006231768638826907. Test_MSE: 0.0002610358313293963\n",
      "Epoch: 130 Iteration: 11600. Train_MSE: 0.000529537268448621. Test_MSE: 0.00026131423110375084\n",
      "Epoch: 131 Iteration: 11700. Train_MSE: 0.0004714626702480018. Test_MSE: 0.00026032104574472126\n",
      "Epoch: 132 Iteration: 11800. Train_MSE: 0.00028400844894349575. Test_MSE: 0.00025987288952129833\n",
      "Epoch: 133 Iteration: 11900. Train_MSE: 0.000671841436997056. Test_MSE: 0.0002593646680711133\n",
      "Epoch: 134 Iteration: 12000. Train_MSE: 0.0005988713819533587. Test_MSE: 0.00025922179530049903\n",
      "Epoch: 135 Iteration: 12100. Train_MSE: 0.00035622130963020027. Test_MSE: 0.00025867840662361627\n",
      "Epoch: 137 Iteration: 12200. Train_MSE: 0.000676696770824492. Test_MSE: 0.0002582476859195859\n",
      "Epoch: 138 Iteration: 12300. Train_MSE: 0.001184051507152617. Test_MSE: 0.00025738751207892417\n",
      "Epoch: 139 Iteration: 12400. Train_MSE: 0.0007940036011859775. Test_MSE: 0.00025769155176704394\n",
      "Epoch: 140 Iteration: 12500. Train_MSE: 0.00035506609128788114. Test_MSE: 0.00025684394429099157\n",
      "Epoch: 141 Iteration: 12600. Train_MSE: 0.00047071967856027186. Test_MSE: 0.00025637188801956086\n",
      "Epoch: 142 Iteration: 12700. Train_MSE: 0.0015030945651233196. Test_MSE: 0.00025589442427602024\n",
      "Epoch: 143 Iteration: 12800. Train_MSE: 0.0004463530203793198. Test_MSE: 0.0002557667958268407\n",
      "Epoch: 144 Iteration: 12900. Train_MSE: 0.0003251680755056441. Test_MSE: 0.00025514130853626193\n",
      "Epoch: 146 Iteration: 13000. Train_MSE: 0.00047068591811694205. Test_MSE: 0.00025499666548822037\n",
      "Epoch: 147 Iteration: 13100. Train_MSE: 0.0005241403123363853. Test_MSE: 0.0002540074863920548\n",
      "Epoch: 148 Iteration: 13200. Train_MSE: 0.0006728190346620977. Test_MSE: 0.00025419510962900924\n",
      "Epoch: 149 Iteration: 13300. Train_MSE: 0.0005010644672438502. Test_MSE: 0.00025341028799260375\n",
      "Epoch: 150 Iteration: 13400. Train_MSE: 0.00026448024436831474. Test_MSE: 0.00025305423672996355\n",
      "Epoch: 151 Iteration: 13500. Train_MSE: 0.0006231562583707273. Test_MSE: 0.0002526436109036494\n",
      "Epoch: 152 Iteration: 13600. Train_MSE: 0.00036010946496389806. Test_MSE: 0.00025242222012695283\n",
      "Epoch: 153 Iteration: 13700. Train_MSE: 0.00024104351177811623. Test_MSE: 0.0002519690757453507\n",
      "Epoch: 155 Iteration: 13800. Train_MSE: 0.0005099886911921203. Test_MSE: 0.0002518355415960703\n",
      "Epoch: 156 Iteration: 13900. Train_MSE: 0.0034859669394791126. Test_MSE: 0.0002508512616884116\n",
      "Epoch: 157 Iteration: 14000. Train_MSE: 0.0002554072707425803. Test_MSE: 0.0002508980263882798\n",
      "Epoch: 158 Iteration: 14100. Train_MSE: 0.0013662789715453982. Test_MSE: 0.0002503658272769063\n",
      "Epoch: 159 Iteration: 14200. Train_MSE: 0.0005821635131724179. Test_MSE: 0.0002499140912209517\n",
      "Epoch: 160 Iteration: 14300. Train_MSE: 0.000851030636113137. Test_MSE: 0.0002494418067432291\n",
      "Epoch: 161 Iteration: 14400. Train_MSE: 0.0005065402947366238. Test_MSE: 0.00024931817750790947\n",
      "Epoch: 162 Iteration: 14500. Train_MSE: 0.0005214883130975068. Test_MSE: 0.0002488826788138729\n",
      "Epoch: 164 Iteration: 14600. Train_MSE: 0.0004803059855476022. Test_MSE: 0.0002488966979800278\n",
      "Epoch: 165 Iteration: 14700. Train_MSE: 0.001875470858067274. Test_MSE: 0.000247953822465583\n",
      "Epoch: 166 Iteration: 14800. Train_MSE: 0.0003781262494158. Test_MSE: 0.00024782206549198393\n",
      "Epoch: 167 Iteration: 14900. Train_MSE: 0.0005137142725288868. Test_MSE: 0.0002474569414611172\n",
      "Epoch: 168 Iteration: 15000. Train_MSE: 0.0005379862850531936. Test_MSE: 0.00024693166591566233\n",
      "Epoch: 169 Iteration: 15100. Train_MSE: 0.0004083254316356033. Test_MSE: 0.00024653230192935147\n",
      "Epoch: 170 Iteration: 15200. Train_MSE: 0.008584882132709026. Test_MSE: 0.00024629989068804514\n",
      "Epoch: 171 Iteration: 15300. Train_MSE: 0.0005954820662736893. Test_MSE: 0.00024596608694327\n",
      "Epoch: 173 Iteration: 15400. Train_MSE: 0.00023299730673898011. Test_MSE: 0.0002460584285069019\n",
      "Epoch: 174 Iteration: 15500. Train_MSE: 0.0003495560958981514. Test_MSE: 0.0002450945681427414\n",
      "Epoch: 175 Iteration: 15600. Train_MSE: 0.013281464576721191. Test_MSE: 0.0002448599470519642\n",
      "Epoch: 176 Iteration: 15700. Train_MSE: 0.000590726500377059. Test_MSE: 0.00024468409130918114\n",
      "Epoch: 177 Iteration: 15800. Train_MSE: 0.0006889399373903871. Test_MSE: 0.00024397729337479913\n",
      "Epoch: 178 Iteration: 15900. Train_MSE: 0.0004717077245004475. Test_MSE: 0.0002436966037239296\n",
      "Epoch: 179 Iteration: 16000. Train_MSE: 0.0011989549966529012. Test_MSE: 0.00024352475475834668\n",
      "Epoch: 180 Iteration: 16100. Train_MSE: 0.00042781635420396924. Test_MSE: 0.0002432602798839302\n",
      "Epoch: 182 Iteration: 16200. Train_MSE: 0.0007579992525279522. Test_MSE: 0.00024321533070770484\n",
      "Epoch: 183 Iteration: 16300. Train_MSE: 0.0005341307260096073. Test_MSE: 0.000242301183812199\n",
      "Epoch: 184 Iteration: 16400. Train_MSE: 0.001310031278990209. Test_MSE: 0.00024215653639860955\n",
      "Epoch: 185 Iteration: 16500. Train_MSE: 0.0003543188504409045. Test_MSE: 0.00024205493118724822\n",
      "Epoch: 186 Iteration: 16600. Train_MSE: 0.0002572369121480733. Test_MSE: 0.00024129536794586132\n",
      "Epoch: 187 Iteration: 16700. Train_MSE: 0.00027454219525679946. Test_MSE: 0.00024102420717756528\n",
      "Epoch: 188 Iteration: 16800. Train_MSE: 0.0011360832722857594. Test_MSE: 0.00024081430480969734\n",
      "Epoch: 189 Iteration: 16900. Train_MSE: 0.00020876486087217927. Test_MSE: 0.00024057815744413252\n",
      "Epoch: 191 Iteration: 17000. Train_MSE: 0.0005087875761091709. Test_MSE: 0.00024061648987591335\n",
      "Epoch: 192 Iteration: 17100. Train_MSE: 0.0005835791816934943. Test_MSE: 0.00023970339980437525\n",
      "Epoch: 193 Iteration: 17200. Train_MSE: 0.0009877291740849614. Test_MSE: 0.00023944958691042735\n",
      "Epoch: 194 Iteration: 17300. Train_MSE: 0.000471111765364185. Test_MSE: 0.0002394765143019782\n",
      "Epoch: 195 Iteration: 17400. Train_MSE: 0.0013012283015996218. Test_MSE: 0.0002386403387621595\n",
      "Epoch: 196 Iteration: 17500. Train_MSE: 0.00047359836753457785. Test_MSE: 0.0002384244587491155\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 197 Iteration: 17600. Train_MSE: 0.0007237786194309592. Test_MSE: 0.0002381522288813406\n",
      "Epoch: 198 Iteration: 17700. Train_MSE: 0.0004133535549044609. Test_MSE: 0.0002379265009284767\n",
      "Epoch: 200 Iteration: 17800. Train_MSE: 0.0007750818622298539. Test_MSE: 0.00023800919034853215\n",
      "Epoch: 201 Iteration: 17900. Train_MSE: 0.00017503215349279344. Test_MSE: 0.00023717441349410833\n",
      "Epoch: 202 Iteration: 18000. Train_MSE: 0.0006250643637031317. Test_MSE: 0.00023685871425823825\n",
      "Epoch: 203 Iteration: 18100. Train_MSE: 0.00039197865407913923. Test_MSE: 0.00023691862834323074\n",
      "Epoch: 204 Iteration: 18200. Train_MSE: 0.0004452329012565315. Test_MSE: 0.0002361434483365631\n",
      "Epoch: 205 Iteration: 18300. Train_MSE: 0.00041164804133586586. Test_MSE: 0.00023590976708761795\n",
      "Epoch: 206 Iteration: 18400. Train_MSE: 0.0008791125728748739. Test_MSE: 0.00023560830863826398\n",
      "Epoch: 207 Iteration: 18500. Train_MSE: 0.00043885334162041545. Test_MSE: 0.00023544330202728288\n",
      "Epoch: 208 Iteration: 18600. Train_MSE: 0.000495817163027823. Test_MSE: 0.00023537351906848745\n",
      "Epoch: 210 Iteration: 18700. Train_MSE: 0.0003930033999495208. Test_MSE: 0.0002346747273315747\n",
      "Epoch: 211 Iteration: 18800. Train_MSE: 0.0006640934152528644. Test_MSE: 0.00023434568427115038\n",
      "Epoch: 212 Iteration: 18900. Train_MSE: 0.0009312587208114564. Test_MSE: 0.00023441675264602963\n",
      "Epoch: 213 Iteration: 19000. Train_MSE: 0.00029345080838538706. Test_MSE: 0.0002336806175873814\n",
      "Epoch: 214 Iteration: 19100. Train_MSE: 0.00037721951957792044. Test_MSE: 0.00023350062177337702\n",
      "Epoch: 215 Iteration: 19200. Train_MSE: 0.0007587569416500628. Test_MSE: 0.00023314510946496905\n",
      "Epoch: 216 Iteration: 19300. Train_MSE: 0.0003113826969638467. Test_MSE: 0.0002329506103647694\n",
      "Epoch: 217 Iteration: 19400. Train_MSE: 0.0009880973957479. Test_MSE: 0.0002328532998912872\n",
      "Epoch: 219 Iteration: 19500. Train_MSE: 0.00045290670823305845. Test_MSE: 0.00023227662100929497\n",
      "Epoch: 220 Iteration: 19600. Train_MSE: 0.00033532732049934566. Test_MSE: 0.00023188075222145224\n",
      "Epoch: 221 Iteration: 19700. Train_MSE: 0.00040002475725486875. Test_MSE: 0.00023198576808815927\n",
      "Epoch: 222 Iteration: 19800. Train_MSE: 0.0002989170025102794. Test_MSE: 0.00023127507675988542\n",
      "Epoch: 223 Iteration: 19900. Train_MSE: 0.000271965516731143. Test_MSE: 0.00023113485464073183\n",
      "Epoch: 224 Iteration: 20000. Train_MSE: 0.0005515358061529696. Test_MSE: 0.00023076747935841128\n",
      "Epoch: 225 Iteration: 20100. Train_MSE: 0.0003780866099987179. Test_MSE: 0.00023060592226973732\n",
      "Epoch: 226 Iteration: 20200. Train_MSE: 0.0004311750817578286. Test_MSE: 0.00023033587280971388\n",
      "Epoch: 228 Iteration: 20300. Train_MSE: 0.00026261209859512746. Test_MSE: 0.00023000735712677022\n",
      "Epoch: 229 Iteration: 20400. Train_MSE: 0.0004813634732272476. Test_MSE: 0.00022950838304642727\n",
      "Epoch: 230 Iteration: 20500. Train_MSE: 0.00041411840356886387. Test_MSE: 0.00022960632948679832\n",
      "Epoch: 231 Iteration: 20600. Train_MSE: 0.0003880726872012019. Test_MSE: 0.00022900078119993833\n",
      "Epoch: 232 Iteration: 20700. Train_MSE: 0.00023137536481954157. Test_MSE: 0.00022878878786042944\n",
      "Epoch: 233 Iteration: 20800. Train_MSE: 0.0005755365127697587. Test_MSE: 0.0002284456122704408\n",
      "Epoch: 234 Iteration: 20900. Train_MSE: 0.0004842574126087129. Test_MSE: 0.00022830219339896047\n",
      "Epoch: 235 Iteration: 21000. Train_MSE: 0.00035214555100537837. Test_MSE: 0.0002279724422130707\n",
      "Epoch: 237 Iteration: 21100. Train_MSE: 0.0005222692270763218. Test_MSE: 0.00022777190054895028\n",
      "Epoch: 238 Iteration: 21200. Train_MSE: 0.0010838733287528157. Test_MSE: 0.00022718514177968353\n",
      "Epoch: 239 Iteration: 21300. Train_MSE: 0.0006280384259298444. Test_MSE: 0.00022731240813384605\n",
      "Epoch: 240 Iteration: 21400. Train_MSE: 0.00032471827580593526. Test_MSE: 0.00022676449989764334\n",
      "Epoch: 241 Iteration: 21500. Train_MSE: 0.0003797941899392754. Test_MSE: 0.00022652417702406232\n",
      "Epoch: 242 Iteration: 21600. Train_MSE: 0.0014219399308785796. Test_MSE: 0.00022620572959503675\n",
      "Epoch: 243 Iteration: 21700. Train_MSE: 0.00033563305623829365. Test_MSE: 0.00022607422138193055\n",
      "Epoch: 244 Iteration: 21800. Train_MSE: 0.00025183457182720304. Test_MSE: 0.00022566778061648815\n",
      "Epoch: 246 Iteration: 21900. Train_MSE: 0.000368267297744751. Test_MSE: 0.00022563833826034187\n",
      "Epoch: 247 Iteration: 22000. Train_MSE: 0.00044567190343514085. Test_MSE: 0.0002249643338860532\n",
      "Epoch: 248 Iteration: 22100. Train_MSE: 0.000541536370292306. Test_MSE: 0.00022502801846107672\n",
      "Epoch: 249 Iteration: 22200. Train_MSE: 0.0004378034791443497. Test_MSE: 0.00022449872065839944\n",
      "Epoch: 250 Iteration: 22300. Train_MSE: 0.00022063589130993932. Test_MSE: 0.00022432025900948445\n",
      "Epoch: 251 Iteration: 22400. Train_MSE: 0.0005131697980687022. Test_MSE: 0.00022404667465558364\n",
      "Epoch: 252 Iteration: 22500. Train_MSE: 0.00030548765789717436. Test_MSE: 0.00022385508295154658\n",
      "Epoch: 253 Iteration: 22600. Train_MSE: 0.0002209044760093093. Test_MSE: 0.00022354137549399143\n",
      "Epoch: 255 Iteration: 22700. Train_MSE: 0.00038872554432600737. Test_MSE: 0.000223512571748544\n",
      "Epoch: 256 Iteration: 22800. Train_MSE: 0.003383349161595106. Test_MSE: 0.00022283534477054244\n",
      "Epoch: 257 Iteration: 22900. Train_MSE: 0.00020244518236722797. Test_MSE: 0.0002228139532382397\n",
      "Epoch: 258 Iteration: 23000. Train_MSE: 0.001219945726916194. Test_MSE: 0.00022244075125925168\n",
      "Epoch: 259 Iteration: 23100. Train_MSE: 0.0004907760885544121. Test_MSE: 0.00022218224575061966\n",
      "Epoch: 260 Iteration: 23200. Train_MSE: 0.000744212418794632. Test_MSE: 0.00022186939411000776\n",
      "Epoch: 261 Iteration: 23300. Train_MSE: 0.0004384499916341156. Test_MSE: 0.00022174415903053452\n",
      "Epoch: 262 Iteration: 23400. Train_MSE: 0.0004387332883197814. Test_MSE: 0.00022143047667632143\n",
      "Epoch: 264 Iteration: 23500. Train_MSE: 0.00035979304811917245. Test_MSE: 0.00022149038856564767\n",
      "Epoch: 265 Iteration: 23600. Train_MSE: 0.0018095186678692698. Test_MSE: 0.0002208472813661075\n",
      "Epoch: 266 Iteration: 23700. Train_MSE: 0.0003202637890353799. Test_MSE: 0.00022070386264287658\n",
      "Epoch: 267 Iteration: 23800. Train_MSE: 0.0004383689956739545. Test_MSE: 0.00022043836206624975\n",
      "Epoch: 268 Iteration: 23900. Train_MSE: 0.00047177658416330814. Test_MSE: 0.000220110816284179\n",
      "Epoch: 269 Iteration: 24000. Train_MSE: 0.0003240156511310488. Test_MSE: 0.00021984803409554906\n",
      "Epoch: 270 Iteration: 24100. Train_MSE: 0.008503091521561146. Test_MSE: 0.000219648569593756\n",
      "Epoch: 271 Iteration: 24200. Train_MSE: 0.0004736226692330092. Test_MSE: 0.00021940436552899412\n",
      "Epoch: 273 Iteration: 24300. Train_MSE: 0.0001987572613870725. Test_MSE: 0.0002195014433232026\n",
      "Epoch: 274 Iteration: 24400. Train_MSE: 0.0002920868282672018. Test_MSE: 0.00021884957332840983\n",
      "Epoch: 275 Iteration: 24500. Train_MSE: 0.01323779858648777. Test_MSE: 0.00021863204123410743\n",
      "Epoch: 276 Iteration: 24600. Train_MSE: 0.0004835741128772497. Test_MSE: 0.0002184937175974154\n",
      "Epoch: 277 Iteration: 24700. Train_MSE: 0.0005629307124763727. Test_MSE: 0.00021802659637264754\n",
      "Epoch: 278 Iteration: 24800. Train_MSE: 0.00034639344085007906. Test_MSE: 0.00021784187521172848\n",
      "Epoch: 279 Iteration: 24900. Train_MSE: 0.0011466305004432797. Test_MSE: 0.0002176985635706157\n",
      "Epoch: 280 Iteration: 25000. Train_MSE: 0.00036536221159622073. Test_MSE: 0.00021748544121518037\n",
      "Epoch: 282 Iteration: 25100. Train_MSE: 0.0006338685052469373. Test_MSE: 0.0002174808912987314\n",
      "Epoch: 283 Iteration: 25200. Train_MSE: 0.00044095568591728806. Test_MSE: 0.00021685826082371557\n",
      "Epoch: 284 Iteration: 25300. Train_MSE: 0.0012615942396223545. Test_MSE: 0.00021671924523357575\n",
      "Epoch: 285 Iteration: 25400. Train_MSE: 0.0003026063786819577. Test_MSE: 0.00021662034587587506\n",
      "Epoch: 286 Iteration: 25500. Train_MSE: 0.0002160574367735535. Test_MSE: 0.00021610412454960569\n",
      "Epoch: 287 Iteration: 25600. Train_MSE: 0.00022814450494479388. Test_MSE: 0.00021592257367924873\n",
      "Epoch: 288 Iteration: 25700. Train_MSE: 0.0010568612487986684. Test_MSE: 0.00021576129918977394\n",
      "Epoch: 289 Iteration: 25800. Train_MSE: 0.00018349070160184056. Test_MSE: 0.00021556667896441024\n",
      "Epoch: 291 Iteration: 25900. Train_MSE: 0.0004332654061727226. Test_MSE: 0.00021561709960273625\n",
      "Epoch: 292 Iteration: 26000. Train_MSE: 0.00046511783148162067. Test_MSE: 0.00021498771772332292\n",
      "Epoch: 293 Iteration: 26100. Train_MSE: 0.0009411562932655215. Test_MSE: 0.00021477641479766883\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 294 Iteration: 26200. Train_MSE: 0.00040216362685896456. Test_MSE: 0.00021476980493773164\n",
      "Epoch: 295 Iteration: 26300. Train_MSE: 0.0011077290400862694. Test_MSE: 0.00021418591927481722\n",
      "Epoch: 296 Iteration: 26400. Train_MSE: 0.00037953307037241757. Test_MSE: 0.00021404185944881685\n",
      "Epoch: 297 Iteration: 26500. Train_MSE: 0.0006625734968110919. Test_MSE: 0.00021384258947727334\n",
      "Epoch: 298 Iteration: 26600. Train_MSE: 0.0003597511094994843. Test_MSE: 0.0002136495210500817\n",
      "Epoch: 300 Iteration: 26700. Train_MSE: 0.0007128105498850346. Test_MSE: 0.00021372562549293\n",
      "Epoch: 301 Iteration: 26800. Train_MSE: 0.000139163137646392. Test_MSE: 0.00021314589632918238\n",
      "Epoch: 302 Iteration: 26900. Train_MSE: 0.0005733567522838712. Test_MSE: 0.0002128932460692584\n",
      "Epoch: 303 Iteration: 27000. Train_MSE: 0.00032864109380170703. Test_MSE: 0.00021291043997922023\n",
      "Epoch: 304 Iteration: 27100. Train_MSE: 0.0003847041225526482. Test_MSE: 0.0002123595672103405\n",
      "Epoch: 305 Iteration: 27200. Train_MSE: 0.00033134117256850004. Test_MSE: 0.00021219662865966184\n",
      "Epoch: 306 Iteration: 27300. Train_MSE: 0.0008276158478111029. Test_MSE: 0.00021197781455954524\n",
      "Epoch: 307 Iteration: 27400. Train_MSE: 0.0003435242979321629. Test_MSE: 0.00021183411548504356\n",
      "Epoch: 308 Iteration: 27500. Train_MSE: 0.00041425571544095874. Test_MSE: 0.0002117895965367393\n",
      "Epoch: 310 Iteration: 27600. Train_MSE: 0.0003185522509738803. Test_MSE: 0.00021130457952066092\n",
      "Epoch: 311 Iteration: 27700. Train_MSE: 0.0005849468871019781. Test_MSE: 0.0002110448831189439\n",
      "Epoch: 312 Iteration: 27800. Train_MSE: 0.0009053330286405981. Test_MSE: 0.0002110707461039282\n",
      "Epoch: 313 Iteration: 27900. Train_MSE: 0.00024917165865190327. Test_MSE: 0.00021053747006904205\n",
      "Epoch: 314 Iteration: 28000. Train_MSE: 0.0002825675474014133. Test_MSE: 0.00021041163013931188\n",
      "Epoch: 315 Iteration: 28100. Train_MSE: 0.0007094357861205935. Test_MSE: 0.00021015562351264208\n",
      "Epoch: 316 Iteration: 28200. Train_MSE: 0.0002751787251327187. Test_MSE: 0.00020999356966229537\n",
      "Epoch: 317 Iteration: 28300. Train_MSE: 0.0008519722614437342. Test_MSE: 0.00020992434599256205\n",
      "Epoch: 319 Iteration: 28400. Train_MSE: 0.00035759914317168295. Test_MSE: 0.00020952077867111572\n",
      "Epoch: 320 Iteration: 28500. Train_MSE: 0.00028791461954824626. Test_MSE: 0.0002092161664578902\n",
      "Epoch: 321 Iteration: 28600. Train_MSE: 0.00033996798447333276. Test_MSE: 0.00020926782357984968\n",
      "Epoch: 322 Iteration: 28700. Train_MSE: 0.00023227682686410844. Test_MSE: 0.000208747911962709\n",
      "Epoch: 323 Iteration: 28800. Train_MSE: 0.0002347471017856151. Test_MSE: 0.00020865061219720396\n",
      "Epoch: 324 Iteration: 28900. Train_MSE: 0.0004957691417075694. Test_MSE: 0.00020838168819856134\n",
      "Epoch: 325 Iteration: 29000. Train_MSE: 0.00029961601831018925. Test_MSE: 0.00020824425812179608\n",
      "Epoch: 326 Iteration: 29100. Train_MSE: 0.0003886246995534748. Test_MSE: 0.00020804749440153376\n",
      "Epoch: 328 Iteration: 29200. Train_MSE: 0.00021746530546806753. Test_MSE: 0.00020781874900987379\n",
      "Epoch: 329 Iteration: 29300. Train_MSE: 0.00042345680412836373. Test_MSE: 0.00020743900650334898\n",
      "Epoch: 330 Iteration: 29400. Train_MSE: 0.0003543123311828822. Test_MSE: 0.00020748899428387832\n",
      "Epoch: 331 Iteration: 29500. Train_MSE: 0.000332761206664145. Test_MSE: 0.00020703644403603923\n",
      "Epoch: 332 Iteration: 29600. Train_MSE: 0.0001983305555768311. Test_MSE: 0.00020688642122229807\n",
      "Epoch: 333 Iteration: 29700. Train_MSE: 0.0005308105610311031. Test_MSE: 0.00020663400498252107\n",
      "Epoch: 334 Iteration: 29800. Train_MSE: 0.0004247943579684943. Test_MSE: 0.0002065156594061313\n",
      "Epoch: 335 Iteration: 29900. Train_MSE: 0.00032286590430885553. Test_MSE: 0.0002062665912889951\n",
      "Epoch: 337 Iteration: 30000. Train_MSE: 0.00042963336454704404. Test_MSE: 0.00020613052781395767\n",
      "Epoch: 338 Iteration: 30100. Train_MSE: 0.0010297701228410006. Test_MSE: 0.0002056845555496159\n",
      "Epoch: 339 Iteration: 30200. Train_MSE: 0.0005380993825383484. Test_MSE: 0.00020575904044564837\n",
      "Epoch: 340 Iteration: 30300. Train_MSE: 0.00028295646188780665. Test_MSE: 0.00020534154710966974\n",
      "Epoch: 341 Iteration: 30400. Train_MSE: 0.0003111552505288273. Test_MSE: 0.00020516983572646958\n",
      "Epoch: 342 Iteration: 30500. Train_MSE: 0.0013815666316077113. Test_MSE: 0.00020493768116509398\n",
      "Epoch: 343 Iteration: 30600. Train_MSE: 0.0002773101150523871. Test_MSE: 0.00020482772565748943\n",
      "Epoch: 344 Iteration: 30700. Train_MSE: 0.00020748504903167486. Test_MSE: 0.00020451772848207916\n",
      "Epoch: 346 Iteration: 30800. Train_MSE: 0.00029899587389081717. Test_MSE: 0.0002045089844629869\n",
      "Epoch: 347 Iteration: 30900. Train_MSE: 0.00039717816980555654. Test_MSE: 0.00020399693387174178\n",
      "Epoch: 348 Iteration: 31000. Train_MSE: 0.0004690956266131252. Test_MSE: 0.00020402693157193323\n",
      "Epoch: 349 Iteration: 31100. Train_MSE: 0.00039185237255878747. Test_MSE: 0.0002036193782707715\n",
      "Epoch: 350 Iteration: 31200. Train_MSE: 0.00018963502952829003. Test_MSE: 0.0002034931376645116\n",
      "Epoch: 351 Iteration: 31300. Train_MSE: 0.0004597464867401868. Test_MSE: 0.00020329488523411566\n",
      "Epoch: 352 Iteration: 31400. Train_MSE: 0.0002735803718678653. Test_MSE: 0.00020313779319921753\n",
      "Epoch: 353 Iteration: 31500. Train_MSE: 0.0002021980326389894. Test_MSE: 0.000202892531699178\n",
      "Epoch: 355 Iteration: 31600. Train_MSE: 0.0003085499338340014. Test_MSE: 0.00020288574187627225\n",
      "Epoch: 356 Iteration: 31700. Train_MSE: 0.0033024856820702553. Test_MSE: 0.00020236752271515437\n",
      "Epoch: 357 Iteration: 31800. Train_MSE: 0.00017702339391689748. Test_MSE: 0.0002023353211906952\n",
      "Epoch: 358 Iteration: 31900. Train_MSE: 0.00111039774492383. Test_MSE: 0.00020204558061843585\n",
      "Epoch: 359 Iteration: 32000. Train_MSE: 0.0004138619697187096. Test_MSE: 0.0002018545575232327\n",
      "Epoch: 360 Iteration: 32100. Train_MSE: 0.0006828800542280078. Test_MSE: 0.00020162735874498585\n",
      "Epoch: 361 Iteration: 32200. Train_MSE: 0.0004009041585959494. Test_MSE: 0.0002015191514134166\n",
      "Epoch: 362 Iteration: 32300. Train_MSE: 0.0003759853425435722. Test_MSE: 0.0002012721344134075\n",
      "Epoch: 364 Iteration: 32400. Train_MSE: 0.0002832354512065649. Test_MSE: 0.0002013306163755733\n",
      "Epoch: 365 Iteration: 32500. Train_MSE: 0.0017715075518935919. Test_MSE: 0.0002008448314248369\n",
      "Epoch: 366 Iteration: 32600. Train_MSE: 0.0002898494421970099. Test_MSE: 0.00020071557300152632\n",
      "Epoch: 367 Iteration: 32700. Train_MSE: 0.00038311496609821916. Test_MSE: 0.00020050887327887785\n",
      "Epoch: 368 Iteration: 32800. Train_MSE: 0.00041395824518986046. Test_MSE: 0.00020025865623949182\n",
      "Epoch: 369 Iteration: 32900. Train_MSE: 0.0002789265417959541. Test_MSE: 0.00020006997590227003\n",
      "Epoch: 370 Iteration: 33000. Train_MSE: 0.008443611674010754. Test_MSE: 0.00019990330491763837\n",
      "Epoch: 371 Iteration: 33100. Train_MSE: 0.00039103065500967205. Test_MSE: 0.00019971323638412177\n",
      "Epoch: 373 Iteration: 33200. Train_MSE: 0.00017680201563052833. Test_MSE: 0.00019979494075115316\n",
      "Epoch: 374 Iteration: 33300. Train_MSE: 0.0002536402316763997. Test_MSE: 0.00019930684060320925\n",
      "Epoch: 375 Iteration: 33400. Train_MSE: 0.013207676820456982. Test_MSE: 0.0001991161133859099\n",
      "Epoch: 376 Iteration: 33500. Train_MSE: 0.00040653126779943705. Test_MSE: 0.00019900804367466821\n",
      "Epoch: 377 Iteration: 33600. Train_MSE: 0.00046975977602414787. Test_MSE: 0.00019864824946798135\n",
      "Epoch: 378 Iteration: 33700. Train_MSE: 0.00027345711714588106. Test_MSE: 0.0001985160866065649\n",
      "Epoch: 379 Iteration: 33800. Train_MSE: 0.0011095310328528285. Test_MSE: 0.00019840180620382184\n",
      "Epoch: 380 Iteration: 33900. Train_MSE: 0.0003209649003110826. Test_MSE: 0.00019822651408864727\n",
      "Epoch: 382 Iteration: 34000. Train_MSE: 0.0005424206610769033. Test_MSE: 0.00019822872454921693\n",
      "Epoch: 383 Iteration: 34100. Train_MSE: 0.0003766365407500416. Test_MSE: 0.0001977605844738422\n",
      "Epoch: 384 Iteration: 34200. Train_MSE: 0.0012328692246228456. Test_MSE: 0.000197642120442974\n",
      "Epoch: 385 Iteration: 34300. Train_MSE: 0.0002620440791361034. Test_MSE: 0.00019755630929736226\n",
      "Epoch: 386 Iteration: 34400. Train_MSE: 0.00018478529818821698. Test_MSE: 0.00019715502173815584\n",
      "Epoch: 387 Iteration: 34500. Train_MSE: 0.00019795693515334278. Test_MSE: 0.00019702156998638914\n",
      "Epoch: 388 Iteration: 34600. Train_MSE: 0.0010109766153618693. Test_MSE: 0.00019689658468354817\n",
      "Epoch: 389 Iteration: 34700. Train_MSE: 0.00016666828014422208. Test_MSE: 0.00019673432615389944\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 391 Iteration: 34800. Train_MSE: 0.00037406408227980137. Test_MSE: 0.0001967783071506456\n",
      "Epoch: 392 Iteration: 34900. Train_MSE: 0.00038693618262186646. Test_MSE: 0.00019630072845282568\n",
      "Epoch: 393 Iteration: 35000. Train_MSE: 0.0009137842571362853. Test_MSE: 0.00019612638617285565\n",
      "Epoch: 394 Iteration: 35100. Train_MSE: 0.0003540042380336672. Test_MSE: 0.0001961124284672935\n",
      "Epoch: 395 Iteration: 35200. Train_MSE: 0.000957963929977268. Test_MSE: 0.00019565362082629218\n",
      "Epoch: 396 Iteration: 35300. Train_MSE: 0.0003190386341884732. Test_MSE: 0.000195547432801098\n",
      "Epoch: 397 Iteration: 35400. Train_MSE: 0.0006254275795072317. Test_MSE: 0.00019539721956250076\n",
      "Epoch: 398 Iteration: 35500. Train_MSE: 0.00032055587507784367. Test_MSE: 0.0001952334747448042\n",
      "Epoch: 400 Iteration: 35600. Train_MSE: 0.0006533071864396334. Test_MSE: 0.0001952971098054992\n",
      "Epoch: 401 Iteration: 35700. Train_MSE: 0.00011803409142885357. Test_MSE: 0.00019485500351938868\n",
      "Epoch: 402 Iteration: 35800. Train_MSE: 0.0005423419061116874. Test_MSE: 0.0001946492241121529\n",
      "Epoch: 403 Iteration: 35900. Train_MSE: 0.0002860765380319208. Test_MSE: 0.00019465404189907825\n",
      "Epoch: 404 Iteration: 36000. Train_MSE: 0.00033440557308495045. Test_MSE: 0.00019421797275899182\n",
      "Epoch: 405 Iteration: 36100. Train_MSE: 0.0002786064869724214. Test_MSE: 0.00019409259826633332\n",
      "Epoch: 406 Iteration: 36200. Train_MSE: 0.0008000516681931913. Test_MSE: 0.00019392748508054283\n",
      "Epoch: 407 Iteration: 36300. Train_MSE: 0.00028126436518505216. Test_MSE: 0.00019380602735267607\n",
      "Epoch: 408 Iteration: 36400. Train_MSE: 0.00034668456646613777. Test_MSE: 0.0001937696411089479\n",
      "Epoch: 410 Iteration: 36500. Train_MSE: 0.00026919078663922846. Test_MSE: 0.00019340076228371335\n",
      "Epoch: 411 Iteration: 36600. Train_MSE: 0.0005336840404197574. Test_MSE: 0.00019319097333394048\n",
      "Epoch: 412 Iteration: 36700. Train_MSE: 0.0008890041499398649. Test_MSE: 0.0001932034013927826\n",
      "Epoch: 413 Iteration: 36800. Train_MSE: 0.00021296275372151285. Test_MSE: 0.00019277831789042398\n",
      "Epoch: 414 Iteration: 36900. Train_MSE: 0.00022373493993654847. Test_MSE: 0.00019268152991860628\n",
      "Epoch: 415 Iteration: 37000. Train_MSE: 0.0006790924235247076. Test_MSE: 0.0001924882699666223\n",
      "Epoch: 416 Iteration: 37100. Train_MSE: 0.0002518232213333249. Test_MSE: 0.00019235391730624413\n",
      "Epoch: 417 Iteration: 37200. Train_MSE: 0.000730337924323976. Test_MSE: 0.0001922973298609076\n",
      "Epoch: 419 Iteration: 37300. Train_MSE: 0.00029421132057905197. Test_MSE: 0.00019198832310745434\n",
      "Epoch: 420 Iteration: 37400. Train_MSE: 0.0002579071733634919. Test_MSE: 0.00019174509178943158\n",
      "Epoch: 421 Iteration: 37500. Train_MSE: 0.0003017550625372678. Test_MSE: 0.00019177771413455476\n",
      "Epoch: 422 Iteration: 37600. Train_MSE: 0.00019006650836672634. Test_MSE: 0.00019136196044947757\n",
      "Epoch: 423 Iteration: 37700. Train_MSE: 0.0002092345239361748. Test_MSE: 0.00019128677203032485\n",
      "Epoch: 424 Iteration: 37800. Train_MSE: 0.0004602277767844498. Test_MSE: 0.00019108089846950658\n",
      "Epoch: 425 Iteration: 37900. Train_MSE: 0.00025064105284400284. Test_MSE: 0.00019096616495465133\n",
      "Epoch: 426 Iteration: 38000. Train_MSE: 0.0003462102613411844. Test_MSE: 0.0001908093963803468\n",
      "Epoch: 428 Iteration: 38100. Train_MSE: 0.00018802825070451945. Test_MSE: 0.0001906357740927671\n",
      "Epoch: 429 Iteration: 38200. Train_MSE: 0.0003876116534229368. Test_MSE: 0.00019033232703777737\n",
      "Epoch: 430 Iteration: 38300. Train_MSE: 0.00031280601979233325. Test_MSE: 0.00019036498763884695\n",
      "Epoch: 431 Iteration: 38400. Train_MSE: 0.00029125018045306206. Test_MSE: 0.00018999884268722534\n",
      "Epoch: 432 Iteration: 38500. Train_MSE: 0.00017466496501583606. Test_MSE: 0.00018988138617736044\n",
      "Epoch: 433 Iteration: 38600. Train_MSE: 0.0005005747661925852. Test_MSE: 0.00018968786461532542\n",
      "Epoch: 434 Iteration: 38700. Train_MSE: 0.0003805097076110542. Test_MSE: 0.00018959193801555005\n",
      "Epoch: 435 Iteration: 38800. Train_MSE: 0.00029007659759372473. Test_MSE: 0.00018938988238235318\n",
      "Epoch: 437 Iteration: 38900. Train_MSE: 0.00036601375904865563. Test_MSE: 0.00018928915806804296\n",
      "Epoch: 438 Iteration: 39000. Train_MSE: 0.0009918877622112632. Test_MSE: 0.00018893168580426715\n",
      "Epoch: 439 Iteration: 39100. Train_MSE: 0.0004726083134301007. Test_MSE: 0.00018898555936136042\n",
      "Epoch: 440 Iteration: 39200. Train_MSE: 0.0002449513121973723. Test_MSE: 0.00018864490314542974\n",
      "Epoch: 441 Iteration: 39300. Train_MSE: 0.00025898555759340525. Test_MSE: 0.00018850968968270703\n",
      "Epoch: 442 Iteration: 39400. Train_MSE: 0.001354826963506639. Test_MSE: 0.00018833362968972368\n",
      "Epoch: 443 Iteration: 39500. Train_MSE: 0.00023937644436955452. Test_MSE: 0.00018824322313961122\n",
      "Epoch: 444 Iteration: 39600. Train_MSE: 0.00017626857152208686. Test_MSE: 0.00018799073422745783\n",
      "Epoch: 446 Iteration: 39700. Train_MSE: 0.0002508718171156943. Test_MSE: 0.0001879912846808352\n",
      "Epoch: 447 Iteration: 39800. Train_MSE: 0.0003663465322460979. Test_MSE: 0.00018757856846851454\n",
      "Epoch: 448 Iteration: 39900. Train_MSE: 0.0004165355348959565. Test_MSE: 0.00018759767747034177\n",
      "Epoch: 449 Iteration: 40000. Train_MSE: 0.0003541227779351175. Test_MSE: 0.00018726347624987072\n",
      "Epoch: 450 Iteration: 40100. Train_MSE: 0.00016618431254755706. Test_MSE: 0.00018716425971452815\n",
      "Epoch: 451 Iteration: 40200. Train_MSE: 0.000425001751864329. Test_MSE: 0.0001870153189258079\n",
      "Epoch: 452 Iteration: 40300. Train_MSE: 0.00025138602359220386. Test_MSE: 0.00018688448091717732\n",
      "Epoch: 453 Iteration: 40400. Train_MSE: 0.00018524577899370342. Test_MSE: 0.000186682595551092\n",
      "Epoch: 455 Iteration: 40500. Train_MSE: 0.00025079853367060423. Test_MSE: 0.000186685600933882\n",
      "Epoch: 456 Iteration: 40600. Train_MSE: 0.0032346430234611034. Test_MSE: 0.00018626581648477816\n",
      "Epoch: 457 Iteration: 40700. Train_MSE: 0.00016073169535957277. Test_MSE: 0.0001862351978627052\n",
      "Epoch: 458 Iteration: 40800. Train_MSE: 0.0010282156290486455. Test_MSE: 0.0001859976565773732\n",
      "Epoch: 459 Iteration: 40900. Train_MSE: 0.0003517404547892511. Test_MSE: 0.00018584347430037362\n",
      "Epoch: 460 Iteration: 41000. Train_MSE: 0.0006413646624423563. Test_MSE: 0.00018567043332828642\n",
      "Epoch: 461 Iteration: 41100. Train_MSE: 0.0003718351072166115. Test_MSE: 0.00018557740505470573\n",
      "Epoch: 462 Iteration: 41200. Train_MSE: 0.00032743532210588455. Test_MSE: 0.000185373494360659\n",
      "Epoch: 464 Iteration: 41300. Train_MSE: 0.0002296316233696416. Test_MSE: 0.0001854272332589453\n",
      "Epoch: 465 Iteration: 41400. Train_MSE: 0.0017450357554480433. Test_MSE: 0.0001850396275621866\n",
      "Epoch: 466 Iteration: 41500. Train_MSE: 0.00026749935932457447. Test_MSE: 0.00018492429348018129\n",
      "Epoch: 467 Iteration: 41600. Train_MSE: 0.00034161354415118694. Test_MSE: 0.00018475587881926624\n",
      "Epoch: 468 Iteration: 41700. Train_MSE: 0.00036657508462667465. Test_MSE: 0.00018454865109020422\n",
      "Epoch: 469 Iteration: 41800. Train_MSE: 0.0002485490695107728. Test_MSE: 0.00018440790027653996\n",
      "Epoch: 470 Iteration: 41900. Train_MSE: 0.008403634652495384. Test_MSE: 0.00018426478685181058\n",
      "Epoch: 471 Iteration: 42000. Train_MSE: 0.0003303496923763305. Test_MSE: 0.0001841112523168775\n",
      "Epoch: 473 Iteration: 42100. Train_MSE: 0.00016108494310174137. Test_MSE: 0.00018418055621861795\n",
      "Epoch: 474 Iteration: 42200. Train_MSE: 0.00022480259940493852. Test_MSE: 0.00018379520385335523\n",
      "Epoch: 475 Iteration: 42300. Train_MSE: 0.013185902498662472. Test_MSE: 0.00018362485549407447\n",
      "Epoch: 476 Iteration: 42400. Train_MSE: 0.0003480628947727382. Test_MSE: 0.00018353893048343893\n",
      "Epoch: 477 Iteration: 42500. Train_MSE: 0.00039993374957703054. Test_MSE: 0.00018324044339092618\n",
      "Epoch: 478 Iteration: 42600. Train_MSE: 0.0002251017722301185. Test_MSE: 0.00018314253467227603\n",
      "Epoch: 479 Iteration: 42700. Train_MSE: 0.0010832251282408834. Test_MSE: 0.00018304920215054968\n",
      "Epoch: 480 Iteration: 42800. Train_MSE: 0.0002849925949703902. Test_MSE: 0.00018290034036287542\n",
      "Epoch: 482 Iteration: 42900. Train_MSE: 0.0004730269720312208. Test_MSE: 0.00018290348745455398\n",
      "Epoch: 483 Iteration: 43000. Train_MSE: 0.0003283376863691956. Test_MSE: 0.0001825317737957416\n",
      "Epoch: 484 Iteration: 43100. Train_MSE: 0.0012106825597584248. Test_MSE: 0.00018242915962844275\n",
      "Epoch: 485 Iteration: 43200. Train_MSE: 0.0002302552165929228. Test_MSE: 0.0001823546327114678\n",
      "Epoch: 486 Iteration: 43300. Train_MSE: 0.0001612122869119048. Test_MSE: 0.0001820207715997933\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 487 Iteration: 43400. Train_MSE: 0.00017507215670775622. Test_MSE: 0.000181918331365943\n",
      "Epoch: 488 Iteration: 43500. Train_MSE: 0.0009807791793718934. Test_MSE: 0.00018181912377679535\n",
      "Epoch: 489 Iteration: 43600. Train_MSE: 0.00015361847181338817. Test_MSE: 0.00018168086834689207\n",
      "Epoch: 491 Iteration: 43700. Train_MSE: 0.0003273342445027083. Test_MSE: 0.00018171867863545356\n",
      "Epoch: 492 Iteration: 43800. Train_MSE: 0.0003314425121061504. Test_MSE: 0.00018133640478084896\n",
      "Epoch: 493 Iteration: 43900. Train_MSE: 0.0008989410707727075. Test_MSE: 0.0001811882060947146\n",
      "Epoch: 494 Iteration: 44000. Train_MSE: 0.0003163327928632498. Test_MSE: 0.0001811729993751487\n",
      "Epoch: 495 Iteration: 44100. Train_MSE: 0.0008406923734582961. Test_MSE: 0.00018078921396493403\n",
      "Epoch: 496 Iteration: 44200. Train_MSE: 0.0002758572227321565. Test_MSE: 0.00018070649495008164\n",
      "Epoch: 497 Iteration: 44300. Train_MSE: 0.0005983827286399901. Test_MSE: 0.00018058958430062998\n",
      "Epoch: 498 Iteration: 44400. Train_MSE: 0.0002881203545257449. Test_MSE: 0.0001804473224091621\n",
      "Epoch: 500 Iteration: 44500. Train_MSE: 0.0006008204072713852. Test_MSE: 0.00018049951812738562\n",
      "Epoch: 501 Iteration: 44600. Train_MSE: 0.00010400432802271098. Test_MSE: 0.00018014402692650554\n",
      "Epoch: 502 Iteration: 44700. Train_MSE: 0.0005210242816247046. Test_MSE: 0.00017997075269275928\n",
      "Epoch: 503 Iteration: 44800. Train_MSE: 0.00025349861243739724. Test_MSE: 0.00017997121088247488\n",
      "Epoch: 504 Iteration: 44900. Train_MSE: 0.000293188524665311. Test_MSE: 0.00017960486168903998\n",
      "Epoch: 505 Iteration: 45000. Train_MSE: 0.00023974390933290124. Test_MSE: 0.00017950303339010673\n",
      "Epoch: 506 Iteration: 45100. Train_MSE: 0.0007818948943167925. Test_MSE: 0.00017937472026166156\n",
      "Epoch: 507 Iteration: 45200. Train_MSE: 0.00023769040126353502. Test_MSE: 0.00017927033508560516\n",
      "Epoch: 508 Iteration: 45300. Train_MSE: 0.00029307001386769116. Test_MSE: 0.00017923709616040034\n",
      "Epoch: 510 Iteration: 45400. Train_MSE: 0.00023287738440558314. Test_MSE: 0.0001789409751302128\n",
      "Epoch: 511 Iteration: 45500. Train_MSE: 0.0004955615731887519. Test_MSE: 0.00017876518715334425\n",
      "Epoch: 512 Iteration: 45600. Train_MSE: 0.0008761920034885406. Test_MSE: 0.0001787720519572491\n",
      "Epoch: 513 Iteration: 45700. Train_MSE: 0.00018391115008853376. Test_MSE: 0.00017841368066137415\n",
      "Epoch: 514 Iteration: 45800. Train_MSE: 0.00018473068485036492. Test_MSE: 0.00017833415735624331\n",
      "Epoch: 515 Iteration: 45900. Train_MSE: 0.0006567458040080965. Test_MSE: 0.0001781829535012752\n",
      "Epoch: 516 Iteration: 46000. Train_MSE: 0.0002342910156585276. Test_MSE: 0.0001780686173665476\n",
      "Epoch: 517 Iteration: 46100. Train_MSE: 0.0006286637508310378. Test_MSE: 0.0001780190059115787\n",
      "Epoch: 519 Iteration: 46200. Train_MSE: 0.00024965425836853683. Test_MSE: 0.00017776923112936764\n",
      "Epoch: 520 Iteration: 46300. Train_MSE: 0.0002355823526158929. Test_MSE: 0.00017756723521782488\n",
      "Epoch: 521 Iteration: 46400. Train_MSE: 0.0002726050151977688. Test_MSE: 0.00017759069913621543\n",
      "Epoch: 522 Iteration: 46500. Train_MSE: 0.00016080739442259073. Test_MSE: 0.0001772391440077312\n",
      "Epoch: 523 Iteration: 46600. Train_MSE: 0.0001900868519442156. Test_MSE: 0.00017717657938775145\n",
      "Epoch: 524 Iteration: 46700. Train_MSE: 0.00043218093924224377. Test_MSE: 0.00017701242130938022\n",
      "Epoch: 525 Iteration: 46800. Train_MSE: 0.00021626042143907398. Test_MSE: 0.00017691385334780518\n",
      "Epoch: 526 Iteration: 46900. Train_MSE: 0.00030818258528597653. Test_MSE: 0.00017678108173401164\n",
      "Epoch: 528 Iteration: 47000. Train_MSE: 0.00016809170483611524. Test_MSE: 0.0001766409517662503\n",
      "Epoch: 529 Iteration: 47100. Train_MSE: 0.0003615347668528557. Test_MSE: 0.0001763872089594364\n",
      "Epoch: 530 Iteration: 47200. Train_MSE: 0.00028087786631658673. Test_MSE: 0.00017641106813071495\n",
      "Epoch: 531 Iteration: 47300. Train_MSE: 0.0002590295043773949. Test_MSE: 0.00017609895734676138\n",
      "Epoch: 532 Iteration: 47400. Train_MSE: 0.00015729079314041883. Test_MSE: 0.000176000263650516\n",
      "Epoch: 533 Iteration: 47500. Train_MSE: 0.0004787820507772267. Test_MSE: 0.0001758458668511267\n",
      "Epoch: 534 Iteration: 47600. Train_MSE: 0.00034365549799986184. Test_MSE: 0.0001757653105537947\n",
      "Epoch: 535 Iteration: 47700. Train_MSE: 0.00025963434018194675. Test_MSE: 0.0001755927338479436\n",
      "Epoch: 537 Iteration: 47800. Train_MSE: 0.00031915391446091235. Test_MSE: 0.00017551242986181938\n",
      "Epoch: 538 Iteration: 47900. Train_MSE: 0.0009630867862142622. Test_MSE: 0.00017521298538326853\n",
      "Epoch: 539 Iteration: 48000. Train_MSE: 0.00041970296297222376. Test_MSE: 0.00017525564160532868\n",
      "Epoch: 540 Iteration: 48100. Train_MSE: 0.00021281019144225866. Test_MSE: 0.00017496382539532887\n",
      "Epoch: 541 Iteration: 48200. Train_MSE: 0.00021841874695383012. Test_MSE: 0.00017484976078168063\n",
      "Epoch: 542 Iteration: 48300. Train_MSE: 0.0013353235553950071. Test_MSE: 0.00017471105249488377\n",
      "Epoch: 543 Iteration: 48400. Train_MSE: 0.00021175736037548631. Test_MSE: 0.00017463452802557702\n",
      "Epoch: 544 Iteration: 48500. Train_MSE: 0.0001528817811049521. Test_MSE: 0.00017441998448914026\n",
      "Epoch: 546 Iteration: 48600. Train_MSE: 0.00021643178479280323. Test_MSE: 0.00017442510147374604\n",
      "Epoch: 547 Iteration: 48700. Train_MSE: 0.0003446172922849655. Test_MSE: 0.00017407909807620926\n",
      "Epoch: 548 Iteration: 48800. Train_MSE: 0.000374192459275946. Test_MSE: 0.0001740930669511905\n",
      "Epoch: 549 Iteration: 48900. Train_MSE: 0.00032237693085335195. Test_MSE: 0.00017380686203070732\n",
      "Epoch: 550 Iteration: 49000. Train_MSE: 0.00014772186113987118. Test_MSE: 0.00017372294969532289\n",
      "Epoch: 551 Iteration: 49100. Train_MSE: 0.00040005959453992546. Test_MSE: 0.0001736076786360904\n",
      "Epoch: 552 Iteration: 49200. Train_MSE: 0.00023339735344052315. Test_MSE: 0.00017349506161084153\n",
      "Epoch: 553 Iteration: 49300. Train_MSE: 0.00017018472135532647. Test_MSE: 0.00017332250996455358\n",
      "Epoch: 555 Iteration: 49400. Train_MSE: 0.00020813895389437675. Test_MSE: 0.0001733299914471883\n",
      "Epoch: 556 Iteration: 49500. Train_MSE: 0.003169635310769081. Test_MSE: 0.00017297654443726806\n",
      "Epoch: 557 Iteration: 49600. Train_MSE: 0.00014837674098089337. Test_MSE: 0.00017294880850322232\n",
      "Epoch: 558 Iteration: 49700. Train_MSE: 0.0009650072315707803. Test_MSE: 0.00017274580909514188\n",
      "Epoch: 559 Iteration: 49800. Train_MSE: 0.00030265384702943265. Test_MSE: 0.0001726132317266231\n",
      "Epoch: 560 Iteration: 49900. Train_MSE: 0.0006108916713856161. Test_MSE: 0.00017247699019256306\n",
      "Epoch: 561 Iteration: 50000. Train_MSE: 0.00034776132088154554. Test_MSE: 0.00017239471901699634\n",
      "Epoch: 562 Iteration: 50100. Train_MSE: 0.0002889260358642787. Test_MSE: 0.0001722210555311204\n",
      "Epoch: 564 Iteration: 50200. Train_MSE: 0.0001905683020595461. Test_MSE: 0.00017226988704430396\n",
      "Epoch: 565 Iteration: 50300. Train_MSE: 0.0017275420250371099. Test_MSE: 0.00017194824738112597\n",
      "Epoch: 566 Iteration: 50400. Train_MSE: 0.0002499959373380989. Test_MSE: 0.00017184524108421775\n",
      "Epoch: 567 Iteration: 50500. Train_MSE: 0.0003084331110585481. Test_MSE: 0.00017170315457818212\n",
      "Epoch: 568 Iteration: 50600. Train_MSE: 0.00032765569631010294. Test_MSE: 0.00017152231494185156\n",
      "Epoch: 569 Iteration: 50700. Train_MSE: 0.00022582737437915057. Test_MSE: 0.00017141435534082845\n",
      "Epoch: 570 Iteration: 50800. Train_MSE: 0.00837279949337244. Test_MSE: 0.000171287298574032\n",
      "Epoch: 571 Iteration: 50900. Train_MSE: 0.00028379782452248037. Test_MSE: 0.00017115926895463766\n",
      "Epoch: 573 Iteration: 51000. Train_MSE: 0.0001488858397351578. Test_MSE: 0.00017121874409400652\n",
      "Epoch: 574 Iteration: 51100. Train_MSE: 0.00020185517496429384. Test_MSE: 0.0001709018434457618\n",
      "Epoch: 575 Iteration: 51200. Train_MSE: 0.013175193220376968. Test_MSE: 0.0001707478173621407\n",
      "Epoch: 576 Iteration: 51300. Train_MSE: 0.0003022942109964788. Test_MSE: 0.0001706782788634809\n",
      "Epoch: 577 Iteration: 51400. Train_MSE: 0.0003462779859546572. Test_MSE: 0.0001704187985335605\n",
      "Epoch: 578 Iteration: 51500. Train_MSE: 0.00019046863599214703. Test_MSE: 0.00017034491141950284\n",
      "Epoch: 579 Iteration: 51600. Train_MSE: 0.001063944073393941. Test_MSE: 0.0001702672698022433\n",
      "Epoch: 580 Iteration: 51700. Train_MSE: 0.00025461334735155106. Test_MSE: 0.0001701380829740649\n",
      "Epoch: 582 Iteration: 51800. Train_MSE: 0.0004182131960988045. Test_MSE: 0.0001701407914034362\n",
      "Epoch: 583 Iteration: 51900. Train_MSE: 0.000290616910206154. Test_MSE: 0.00016983396098399394\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 584 Iteration: 52000. Train_MSE: 0.0011918277014046907. Test_MSE: 0.0001697457319844509\n",
      "Epoch: 585 Iteration: 52100. Train_MSE: 0.00020420076907612383. Test_MSE: 0.00016968087133226298\n",
      "Epoch: 586 Iteration: 52200. Train_MSE: 0.0001427690003765747. Test_MSE: 0.00016939044739549557\n",
      "Epoch: 587 Iteration: 52300. Train_MSE: 0.00015577532758470625. Test_MSE: 0.0001693099615527244\n",
      "Epoch: 588 Iteration: 52400. Train_MSE: 0.0009606513194739819. Test_MSE: 0.00016922986737652476\n",
      "Epoch: 589 Iteration: 52500. Train_MSE: 0.0001423556386725977. Test_MSE: 0.00016911004221053795\n",
      "Epoch: 591 Iteration: 52600. Train_MSE: 0.00028939370531588793. Test_MSE: 0.00016914318732569057\n",
      "Epoch: 592 Iteration: 52700. Train_MSE: 0.0002892641059588641. Test_MSE: 0.00016882538834331922\n",
      "Epoch: 593 Iteration: 52800. Train_MSE: 0.0008895839564502239. Test_MSE: 0.0001686982314685773\n",
      "Epoch: 594 Iteration: 52900. Train_MSE: 0.00028481296612881124. Test_MSE: 0.00016868570412957815\n",
      "Epoch: 595 Iteration: 53000. Train_MSE: 0.0007460699998773634. Test_MSE: 0.00016835158921179387\n",
      "Epoch: 596 Iteration: 53100. Train_MSE: 0.0002429495070828125. Test_MSE: 0.00016828705594186413\n",
      "Epoch: 597 Iteration: 53200. Train_MSE: 0.0005787329864688218. Test_MSE: 0.0001681941798968902\n",
      "Epoch: 598 Iteration: 53300. Train_MSE: 0.0002596976701170206. Test_MSE: 0.00016806980676737013\n",
      "Epoch: 600 Iteration: 53400. Train_MSE: 0.0005543080624192953. Test_MSE: 0.00016811765125585978\n",
      "Epoch: 601 Iteration: 53500. Train_MSE: 9.395305824000388e-05. Test_MSE: 0.00016782030225852144\n",
      "Epoch: 602 Iteration: 53600. Train_MSE: 0.0005047028535045683. Test_MSE: 0.00016767189288532502\n",
      "Epoch: 603 Iteration: 53700. Train_MSE: 0.00022748396440874785. Test_MSE: 0.0001676731175565518\n",
      "Epoch: 604 Iteration: 53800. Train_MSE: 0.000258986052358523. Test_MSE: 0.00016735324191623438\n",
      "Epoch: 605 Iteration: 53900. Train_MSE: 0.00021022389410063624. Test_MSE: 0.00016726973626115475\n",
      "Epoch: 606 Iteration: 54000. Train_MSE: 0.0007696933462284505. Test_MSE: 0.0001671678320832433\n",
      "Epoch: 607 Iteration: 54100. Train_MSE: 0.00020546599989756942. Test_MSE: 0.00016707813712237173\n",
      "Epoch: 608 Iteration: 54200. Train_MSE: 0.0002500944829080254. Test_MSE: 0.00016704920891405465\n",
      "Epoch: 610 Iteration: 54300. Train_MSE: 0.00020398087508510798. Test_MSE: 0.0001668027918572921\n",
      "Epoch: 611 Iteration: 54400. Train_MSE: 0.000466005876660347. Test_MSE: 0.00016665263228633916\n",
      "Epoch: 612 Iteration: 54500. Train_MSE: 0.0008655400015413761. Test_MSE: 0.00016665952063057459\n",
      "Epoch: 613 Iteration: 54600. Train_MSE: 0.00016057034372352064. Test_MSE: 0.00016634589967323628\n",
      "Epoch: 614 Iteration: 54700. Train_MSE: 0.00015747727593407035. Test_MSE: 0.0001662807727518782\n",
      "Epoch: 615 Iteration: 54800. Train_MSE: 0.0006396358949132264. Test_MSE: 0.00016615942418108517\n",
      "Epoch: 616 Iteration: 54900. Train_MSE: 0.000220165413338691. Test_MSE: 0.00016606156097005316\n",
      "Epoch: 617 Iteration: 55000. Train_MSE: 0.0005440691020339727. Test_MSE: 0.0001660190454143145\n",
      "Epoch: 619 Iteration: 55100. Train_MSE: 0.00021717225899919868. Test_MSE: 0.00016581038023557123\n",
      "Epoch: 620 Iteration: 55200. Train_MSE: 0.00021868280600756407. Test_MSE: 0.00016563871590096865\n",
      "Epoch: 621 Iteration: 55300. Train_MSE: 0.00024933251552283764. Test_MSE: 0.00016565999456661287\n",
      "Epoch: 622 Iteration: 55400. Train_MSE: 0.00013949911226518452. Test_MSE: 0.00016535298512991282\n",
      "Epoch: 623 Iteration: 55500. Train_MSE: 0.00017513410421088338. Test_MSE: 0.00016530177775682037\n",
      "Epoch: 624 Iteration: 55600. Train_MSE: 0.0004073999880347401. Test_MSE: 0.00016516902046700882\n",
      "Epoch: 625 Iteration: 55700. Train_MSE: 0.0001908010453917086. Test_MSE: 0.00016508535020374153\n",
      "Epoch: 626 Iteration: 55800. Train_MSE: 0.0002748061087913811. Test_MSE: 0.0001649710122569005\n",
      "Epoch: 628 Iteration: 55900. Train_MSE: 0.00015392077330034226. Test_MSE: 0.0001648565263552774\n",
      "Epoch: 629 Iteration: 56000. Train_MSE: 0.00034189841244369745. Test_MSE: 0.00016464026477995563\n",
      "Epoch: 630 Iteration: 56100. Train_MSE: 0.0002551489742472768. Test_MSE: 0.00016466274769329645\n",
      "Epoch: 631 Iteration: 56200. Train_MSE: 0.00023269973462447524. Test_MSE: 0.0001643894942220749\n",
      "Epoch: 632 Iteration: 56300. Train_MSE: 0.0001438628532923758. Test_MSE: 0.00016430627821418958\n",
      "Epoch: 633 Iteration: 56400. Train_MSE: 0.0004598363593686372. Test_MSE: 0.00016418210633992232\n",
      "Epoch: 634 Iteration: 56500. Train_MSE: 0.00031298116664402187. Test_MSE: 0.00016411604686283692\n",
      "Epoch: 635 Iteration: 56600. Train_MSE: 0.00023270081146620214. Test_MSE: 0.0001639659561422174\n",
      "Epoch: 637 Iteration: 56700. Train_MSE: 0.000283511501038447. Test_MSE: 0.00016390297126362\n",
      "Epoch: 638 Iteration: 56800. Train_MSE: 0.0009408307960256934. Test_MSE: 0.00016364597273751003\n",
      "Epoch: 639 Iteration: 56900. Train_MSE: 0.0003762285050470382. Test_MSE: 0.00016368593387373488\n",
      "Epoch: 640 Iteration: 57000. Train_MSE: 0.00018607062520459294. Test_MSE: 0.00016342931132389522\n",
      "Epoch: 641 Iteration: 57100. Train_MSE: 0.00018586110672913492. Test_MSE: 0.00016333183673468782\n",
      "Epoch: 642 Iteration: 57200. Train_MSE: 0.0013203526614233851. Test_MSE: 0.00016322194664987168\n",
      "Epoch: 643 Iteration: 57300. Train_MSE: 0.00019110986613668501. Test_MSE: 0.0001631587835666023\n",
      "Epoch: 644 Iteration: 57400. Train_MSE: 0.0001347177749266848. Test_MSE: 0.00016297197500340544\n",
      "Epoch: 646 Iteration: 57500. Train_MSE: 0.00019042026542592794. Test_MSE: 0.0001629841081936941\n",
      "Epoch: 647 Iteration: 57600. Train_MSE: 0.0003286823339294642. Test_MSE: 0.0001626856204865348\n",
      "Epoch: 648 Iteration: 57700. Train_MSE: 0.0003403696755412966. Test_MSE: 0.00016270157127756002\n",
      "Epoch: 649 Iteration: 57800. Train_MSE: 0.0002956017560791224. Test_MSE: 0.00016244990312776252\n",
      "Epoch: 650 Iteration: 57900. Train_MSE: 0.00013253795623313636. Test_MSE: 0.00016237909132299573\n",
      "Epoch: 651 Iteration: 58000. Train_MSE: 0.0003812535142060369. Test_MSE: 0.00016229021181508126\n",
      "Epoch: 652 Iteration: 58100. Train_MSE: 0.000219223671592772. Test_MSE: 0.00016219383707514345\n",
      "Epoch: 653 Iteration: 58200. Train_MSE: 0.00015731446910649538. Test_MSE: 0.00016204368215836886\n",
      "Epoch: 655 Iteration: 58300. Train_MSE: 0.00017582591681275517. Test_MSE: 0.00016205763585625812\n",
      "Epoch: 656 Iteration: 58400. Train_MSE: 0.0031165278051048517. Test_MSE: 0.00016175137105149442\n",
      "Epoch: 657 Iteration: 58500. Train_MSE: 0.00013865710934624076. Test_MSE: 0.00016173063785503792\n",
      "Epoch: 658 Iteration: 58600. Train_MSE: 0.0009146157535724342. Test_MSE: 0.0001615540030106399\n",
      "Epoch: 659 Iteration: 58700. Train_MSE: 0.00026331478147767484. Test_MSE: 0.00016143883659045365\n",
      "Epoch: 660 Iteration: 58800. Train_MSE: 0.0005872343899682164. Test_MSE: 0.00016133124001738462\n",
      "Epoch: 661 Iteration: 58900. Train_MSE: 0.000328907830407843. Test_MSE: 0.00016126078924437643\n",
      "Epoch: 662 Iteration: 59000. Train_MSE: 0.0002582253946457058. Test_MSE: 0.0001611106478814798\n",
      "Epoch: 664 Iteration: 59100. Train_MSE: 0.00016051992133725435. Test_MSE: 0.0001611597908760031\n",
      "Epoch: 665 Iteration: 59200. Train_MSE: 0.0017136301612481475. Test_MSE: 0.00016088588917528195\n",
      "Epoch: 666 Iteration: 59300. Train_MSE: 0.0002371928858337924. Test_MSE: 0.00016079527428577997\n",
      "Epoch: 667 Iteration: 59400. Train_MSE: 0.0002821047091856599. Test_MSE: 0.0001606742508270583\n",
      "Epoch: 668 Iteration: 59500. Train_MSE: 0.00029510821332223713. Test_MSE: 0.00016051337216799096\n",
      "Epoch: 669 Iteration: 59600. Train_MSE: 0.00020797003526240587. Test_MSE: 0.0001604316024792664\n",
      "Epoch: 670 Iteration: 59700. Train_MSE: 0.00834645889699459. Test_MSE: 0.00016031888737488235\n",
      "Epoch: 671 Iteration: 59800. Train_MSE: 0.0002473504573572427. Test_MSE: 0.00016021193137129847\n",
      "Epoch: 673 Iteration: 59900. Train_MSE: 0.00013875504373572767. Test_MSE: 0.00016026829608864774\n",
      "Epoch: 674 Iteration: 60000. Train_MSE: 0.00018370502220932394. Test_MSE: 0.0001600011902211021\n",
      "Epoch: 675 Iteration: 60100. Train_MSE: 0.013159551657736301. Test_MSE: 0.0001598614722928058\n",
      "Epoch: 676 Iteration: 60200. Train_MSE: 0.00026560804690234363. Test_MSE: 0.00015980705340318287\n",
      "Epoch: 677 Iteration: 60300. Train_MSE: 0.00030405877623707056. Test_MSE: 0.00015957580292796868\n",
      "Epoch: 678 Iteration: 60400. Train_MSE: 0.00016471062554046512. Test_MSE: 0.0001595223356396852\n",
      "Epoch: 679 Iteration: 60500. Train_MSE: 0.0010485830716788769. Test_MSE: 0.0001594589390922044\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 680 Iteration: 60600. Train_MSE: 0.0002291351556777954. Test_MSE: 0.00015934663183585554\n",
      "Epoch: 682 Iteration: 60700. Train_MSE: 0.00037424280890263617. Test_MSE: 0.0001593520552788226\n",
      "Epoch: 683 Iteration: 60800. Train_MSE: 0.0002602623717393726. Test_MSE: 0.00015909210076710018\n",
      "Epoch: 684 Iteration: 60900. Train_MSE: 0.0011749276891350746. Test_MSE: 0.00015901756188881863\n",
      "Epoch: 685 Iteration: 61000. Train_MSE: 0.00018273830937687308. Test_MSE: 0.0001589627371260972\n",
      "Epoch: 686 Iteration: 61100. Train_MSE: 0.00012805509322788566. Test_MSE: 0.00015870338276557336\n",
      "Epoch: 687 Iteration: 61200. Train_MSE: 0.0001399282191414386. Test_MSE: 0.00015864141198389744\n",
      "Epoch: 688 Iteration: 61300. Train_MSE: 0.0009448184864595532. Test_MSE: 0.0001585779538487158\n",
      "Epoch: 689 Iteration: 61400. Train_MSE: 0.0001333133695879951. Test_MSE: 0.00015847426111702056\n",
      "Epoch: 691 Iteration: 61500. Train_MSE: 0.00025812923558987677. Test_MSE: 0.0001585072943054425\n",
      "Epoch: 692 Iteration: 61600. Train_MSE: 0.0002562007284723222. Test_MSE: 0.00015823613275922602\n",
      "Epoch: 693 Iteration: 61700. Train_MSE: 0.0008829624857753515. Test_MSE: 0.00015812659722449682\n",
      "Epoch: 694 Iteration: 61800. Train_MSE: 0.0002584495523478836. Test_MSE: 0.00015811889233243836\n",
      "Epoch: 695 Iteration: 61900. Train_MSE: 0.0006675221957266331. Test_MSE: 0.00015781959368343283\n",
      "Epoch: 696 Iteration: 62000. Train_MSE: 0.00021770212333649397. Test_MSE: 0.000157770020043904\n",
      "Epoch: 697 Iteration: 62100. Train_MSE: 0.0005639006849378347. Test_MSE: 0.00015769730162160697\n",
      "Epoch: 698 Iteration: 62200. Train_MSE: 0.00023607017647009343. Test_MSE: 0.00015758705810050964\n",
      "Epoch: 700 Iteration: 62300. Train_MSE: 0.0005130691570229828. Test_MSE: 0.0001576326582255723\n",
      "Epoch: 701 Iteration: 62400. Train_MSE: 8.647044887766242e-05. Test_MSE: 0.00015737723522226302\n",
      "Epoch: 702 Iteration: 62500. Train_MSE: 0.0004912498407065868. Test_MSE: 0.00015724870235961846\n",
      "Epoch: 703 Iteration: 62600. Train_MSE: 0.0002065184962702915. Test_MSE: 0.00015725227027975122\n",
      "Epoch: 704 Iteration: 62700. Train_MSE: 0.00023036902712192386. Test_MSE: 0.00015696489620416716\n",
      "Epoch: 705 Iteration: 62800. Train_MSE: 0.0001870363048510626. Test_MSE: 0.00015689581862998828\n",
      "Epoch: 706 Iteration: 62900. Train_MSE: 0.0007614756468683481. Test_MSE: 0.00015681492072955273\n",
      "Epoch: 707 Iteration: 63000. Train_MSE: 0.00018081249436363578. Test_MSE: 0.00015673821576534024\n",
      "Epoch: 708 Iteration: 63100. Train_MSE: 0.00021571434626821429. Test_MSE: 0.00015671367256492\n",
      "Epoch: 710 Iteration: 63200. Train_MSE: 0.00018103748152498156. Test_MSE: 0.0001565035155793287\n",
      "Epoch: 711 Iteration: 63300. Train_MSE: 0.0004426214727573097. Test_MSE: 0.0001563741611605963\n",
      "Epoch: 712 Iteration: 63400. Train_MSE: 0.0008577824919484556. Test_MSE: 0.00015638316687775755\n",
      "Epoch: 713 Iteration: 63500. Train_MSE: 0.00014168974303174764. Test_MSE: 0.0001561013036155614\n",
      "Epoch: 714 Iteration: 63600. Train_MSE: 0.00013778269931208342. Test_MSE: 0.00015604822078341994\n",
      "Epoch: 715 Iteration: 63700. Train_MSE: 0.0006259672227315605. Test_MSE: 0.0001559504924539488\n",
      "Epoch: 716 Iteration: 63800. Train_MSE: 0.00020919731468893588. Test_MSE: 0.00015586684770630825\n",
      "Epoch: 717 Iteration: 63900. Train_MSE: 0.00047405422083102167. Test_MSE: 0.00015583080092993723\n",
      "Epoch: 719 Iteration: 64000. Train_MSE: 0.00019234584760852158. Test_MSE: 0.0001556527115652509\n",
      "Epoch: 720 Iteration: 64100. Train_MSE: 0.00020588908228091896. Test_MSE: 0.00015550533329147331\n",
      "Epoch: 721 Iteration: 64200. Train_MSE: 0.0002305889647686854. Test_MSE: 0.00015552698994696535\n",
      "Epoch: 722 Iteration: 64300. Train_MSE: 0.00012322395923547447. Test_MSE: 0.00015525130002308444\n",
      "Epoch: 723 Iteration: 64400. Train_MSE: 0.000162245036335662. Test_MSE: 0.00015520972980754037\n",
      "Epoch: 724 Iteration: 64500. Train_MSE: 0.000385704857762903. Test_MSE: 0.0001551008351456882\n",
      "Epoch: 725 Iteration: 64600. Train_MSE: 0.0001707470219116658. Test_MSE: 0.00015503010307396084\n",
      "Epoch: 726 Iteration: 64700. Train_MSE: 0.00024578699958510697. Test_MSE: 0.00015492993751868442\n",
      "Epoch: 728 Iteration: 64800. Train_MSE: 0.00014331881538964808. Test_MSE: 0.00015483515828819533\n",
      "Epoch: 729 Iteration: 64900. Train_MSE: 0.0003263502148911357. Test_MSE: 0.0001546470827338702\n",
      "Epoch: 730 Iteration: 65000. Train_MSE: 0.00023386241809930652. Test_MSE: 0.00015467050080128817\n",
      "Epoch: 731 Iteration: 65100. Train_MSE: 0.00021078057761769742. Test_MSE: 0.00015442484755161373\n",
      "Epoch: 732 Iteration: 65200. Train_MSE: 0.00013293459778651595. Test_MSE: 0.00015435405711476023\n",
      "Epoch: 733 Iteration: 65300. Train_MSE: 0.0004438958421815187. Test_MSE: 0.00015425283936871062\n",
      "Epoch: 734 Iteration: 65400. Train_MSE: 0.0002879776293411851. Test_MSE: 0.0001541992879771031\n",
      "Epoch: 735 Iteration: 65500. Train_MSE: 0.0002093712246278301. Test_MSE: 0.00015406573735774303\n",
      "Epoch: 737 Iteration: 65600. Train_MSE: 0.0002548474876675755. Test_MSE: 0.00015401674013189223\n",
      "Epoch: 738 Iteration: 65700. Train_MSE: 0.0009225734975188971. Test_MSE: 0.00015379119704020522\n",
      "Epoch: 739 Iteration: 65800. Train_MSE: 0.0003397763066459447. Test_MSE: 0.00015383099589616096\n",
      "Epoch: 740 Iteration: 65900. Train_MSE: 0.0001637718960409984. Test_MSE: 0.00015360002214429746\n",
      "Epoch: 741 Iteration: 66000. Train_MSE: 0.00015973769768606871. Test_MSE: 0.00015351582871563223\n",
      "Epoch: 742 Iteration: 66100. Train_MSE: 0.0013082544319331646. Test_MSE: 0.0001534282063153525\n",
      "Epoch: 743 Iteration: 66200. Train_MSE: 0.00017490211757831275. Test_MSE: 0.0001533766124385881\n",
      "Epoch: 744 Iteration: 66300. Train_MSE: 0.00012040557339787483. Test_MSE: 0.00015321000751845751\n",
      "Epoch: 746 Iteration: 66400. Train_MSE: 0.0001702223380561918. Test_MSE: 0.00015322894434219335\n",
      "Epoch: 747 Iteration: 66500. Train_MSE: 0.000316133868182078. Test_MSE: 0.0001529652432981401\n",
      "Epoch: 748 Iteration: 66600. Train_MSE: 0.00031275805667974055. Test_MSE: 0.00015298404131651555\n",
      "Epoch: 749 Iteration: 66700. Train_MSE: 0.0002724686055444181. Test_MSE: 0.0001527575556248509\n",
      "Epoch: 750 Iteration: 66800. Train_MSE: 0.00011991556675639004. Test_MSE: 0.00015269701022974756\n",
      "Epoch: 751 Iteration: 66900. Train_MSE: 0.0003667860582936555. Test_MSE: 0.00015262869713111938\n",
      "Epoch: 752 Iteration: 67000. Train_MSE: 0.00020800535276066512. Test_MSE: 0.0001525454091170148\n",
      "Epoch: 753 Iteration: 67100. Train_MSE: 0.00014612676750402898. Test_MSE: 0.00015241206953811422\n",
      "Epoch: 755 Iteration: 67200. Train_MSE: 0.0001506982371211052. Test_MSE: 0.00015243204423274752\n",
      "Epoch: 756 Iteration: 67300. Train_MSE: 0.0030699807684868574. Test_MSE: 0.00015216070884060222\n",
      "Epoch: 757 Iteration: 67400. Train_MSE: 0.00013075233437120914. Test_MSE: 0.00015214625130392516\n",
      "Epoch: 758 Iteration: 67500. Train_MSE: 0.0008741833153180778. Test_MSE: 0.0001519893456962195\n",
      "Epoch: 759 Iteration: 67600. Train_MSE: 0.00023142281861510128. Test_MSE: 0.00015188725469497932\n",
      "Epoch: 760 Iteration: 67700. Train_MSE: 0.0005682779010385275. Test_MSE: 0.00015180153083033769\n",
      "Epoch: 761 Iteration: 67800. Train_MSE: 0.0003127200179733336. Test_MSE: 0.00015174060397618754\n",
      "Epoch: 762 Iteration: 67900. Train_MSE: 0.00023304465867113322. Test_MSE: 0.00015160834826873345\n",
      "Epoch: 764 Iteration: 68000. Train_MSE: 0.00013717461843043566. Test_MSE: 0.00015165858735995689\n",
      "Epoch: 765 Iteration: 68100. Train_MSE: 0.0017047039000317454. Test_MSE: 0.00015142032244430282\n",
      "Epoch: 766 Iteration: 68200. Train_MSE: 0.0002258223103126511. Test_MSE: 0.0001513392326902278\n",
      "Epoch: 767 Iteration: 68300. Train_MSE: 0.0002609754737932235. Test_MSE: 0.00015123464479504565\n",
      "Epoch: 768 Iteration: 68400. Train_MSE: 0.0002678582677617669. Test_MSE: 0.00015108869297717907\n",
      "Epoch: 769 Iteration: 68500. Train_MSE: 0.0001931382284965366. Test_MSE: 0.00015102761237655596\n",
      "Epoch: 770 Iteration: 68600. Train_MSE: 0.008324628695845604. Test_MSE: 0.00015092609747546784\n",
      "Epoch: 771 Iteration: 68700. Train_MSE: 0.00021815398940816522. Test_MSE: 0.00015083565755683685\n",
      "Epoch: 773 Iteration: 68800. Train_MSE: 0.0001302216696785763. Test_MSE: 0.00015089056880577588\n",
      "Epoch: 774 Iteration: 68900. Train_MSE: 0.00016883434727787971. Test_MSE: 0.00015066087167849197\n",
      "Epoch: 775 Iteration: 69000. Train_MSE: 0.013136166147887707. Test_MSE: 0.0001505317167634737\n",
      "Epoch: 776 Iteration: 69100. Train_MSE: 0.00023615563986822963. Test_MSE: 0.00015048944988673013\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 777 Iteration: 69200. Train_MSE: 0.0002703395439311862. Test_MSE: 0.00015027857197966743\n",
      "Epoch: 778 Iteration: 69300. Train_MSE: 0.000144889549119398. Test_MSE: 0.00015024097996017087\n",
      "Epoch: 779 Iteration: 69400. Train_MSE: 0.0010355138219892979. Test_MSE: 0.00015018902863348676\n",
      "Epoch: 780 Iteration: 69500. Train_MSE: 0.0002074610092677176. Test_MSE: 0.00015008988213797937\n",
      "Epoch: 782 Iteration: 69600. Train_MSE: 0.00033813592744991183. Test_MSE: 0.00015009784836875044\n",
      "Epoch: 783 Iteration: 69700. Train_MSE: 0.00023551312915515155. Test_MSE: 0.00014987354423870602\n",
      "Epoch: 784 Iteration: 69800. Train_MSE: 0.0011605990584939718. Test_MSE: 0.0001498102629440649\n",
      "Epoch: 785 Iteration: 69900. Train_MSE: 0.00016508260159753263. Test_MSE: 0.00014976307997315198\n",
      "Epoch: 786 Iteration: 70000. Train_MSE: 0.00011614998220466077. Test_MSE: 0.00014952649844526697\n",
      "Epoch: 787 Iteration: 70100. Train_MSE: 0.00012649722339119762. Test_MSE: 0.00014947915018712372\n",
      "Epoch: 788 Iteration: 70200. Train_MSE: 0.0009329357999376953. Test_MSE: 0.00014942893578067927\n",
      "Epoch: 789 Iteration: 70300. Train_MSE: 0.00012536979920696467. Test_MSE: 0.0001493380168391606\n",
      "Epoch: 791 Iteration: 70400. Train_MSE: 0.00023156471434049308. Test_MSE: 0.00014937133856088194\n",
      "Epoch: 792 Iteration: 70500. Train_MSE: 0.00022996401821728796. Test_MSE: 0.00014913569679662423\n",
      "Epoch: 793 Iteration: 70600. Train_MSE: 0.0008791531436145306. Test_MSE: 0.00014904041690041815\n",
      "Epoch: 794 Iteration: 70700. Train_MSE: 0.0002355387550778687. Test_MSE: 0.00014903679884929117\n",
      "Epoch: 795 Iteration: 70800. Train_MSE: 0.000601945910602808. Test_MSE: 0.0001487634932787354\n",
      "Epoch: 796 Iteration: 70900. Train_MSE: 0.00019686443556565791. Test_MSE: 0.00014872588793966127\n",
      "Epoch: 797 Iteration: 71000. Train_MSE: 0.000552540528587997. Test_MSE: 0.00014866944306341173\n",
      "Epoch: 798 Iteration: 71100. Train_MSE: 0.00021574411948677152. Test_MSE: 0.0001485722916073319\n",
      "Epoch: 800 Iteration: 71200. Train_MSE: 0.00047666707541793585. Test_MSE: 0.00014861820440073908\n",
      "Epoch: 801 Iteration: 71300. Train_MSE: 8.069354225881398e-05. Test_MSE: 0.00014839604384260575\n",
      "Epoch: 802 Iteration: 71400. Train_MSE: 0.00048080741544254124. Test_MSE: 0.00014828430445723809\n",
      "Epoch: 803 Iteration: 71500. Train_MSE: 0.00018857068789657205. Test_MSE: 0.00014829226410164765\n",
      "Epoch: 804 Iteration: 71600. Train_MSE: 0.00020627342746593058. Test_MSE: 0.00014802977945500425\n",
      "Epoch: 805 Iteration: 71700. Train_MSE: 0.0001679600536590442. Test_MSE: 0.0001479729433870048\n",
      "Epoch: 806 Iteration: 71800. Train_MSE: 0.0007557193166576326. Test_MSE: 0.00014790954856223573\n",
      "Epoch: 807 Iteration: 71900. Train_MSE: 0.00016144987603183836. Test_MSE: 0.0001478440650604375\n",
      "Epoch: 808 Iteration: 72000. Train_MSE: 0.00018785141583066434. Test_MSE: 0.0001478238423033155\n",
      "Epoch: 810 Iteration: 72100. Train_MSE: 0.0001618934766156599. Test_MSE: 0.00014764256704622242\n",
      "Epoch: 811 Iteration: 72200. Train_MSE: 0.0004237012180965394. Test_MSE: 0.00014753036109539799\n",
      "Epoch: 812 Iteration: 72300. Train_MSE: 0.0008514787186868489. Test_MSE: 0.0001475431942742763\n",
      "Epoch: 813 Iteration: 72400. Train_MSE: 0.00012618987238965929. Test_MSE: 0.00014728583522150793\n",
      "Epoch: 814 Iteration: 72500. Train_MSE: 0.00012318372318986803. Test_MSE: 0.00014724302642499422\n",
      "Epoch: 815 Iteration: 72600. Train_MSE: 0.0006148596876300871. Test_MSE: 0.0001471642933823019\n",
      "Epoch: 816 Iteration: 72700. Train_MSE: 0.00020050765306223184. Test_MSE: 0.0001470928392073751\n",
      "Epoch: 817 Iteration: 72800. Train_MSE: 0.00041575287468731403. Test_MSE: 0.00014706242603731314\n",
      "Epoch: 819 Iteration: 72900. Train_MSE: 0.00017253329860977829. Test_MSE: 0.00014690859283952333\n",
      "Epoch: 820 Iteration: 73000. Train_MSE: 0.00019547404372133315. Test_MSE: 0.00014678066915887848\n",
      "Epoch: 821 Iteration: 73100. Train_MSE: 0.0002149759093299508. Test_MSE: 0.00014680469225122027\n",
      "Epoch: 822 Iteration: 73200. Train_MSE: 0.00011039273522328585. Test_MSE: 0.00014655327942410474\n",
      "Epoch: 823 Iteration: 73300. Train_MSE: 0.00015077136049512774. Test_MSE: 0.00014651990168290216\n",
      "Epoch: 824 Iteration: 73400. Train_MSE: 0.0003680535010062158. Test_MSE: 0.00014643011920650675\n",
      "Epoch: 825 Iteration: 73500. Train_MSE: 0.00015481789887417108. Test_MSE: 0.00014637002190189677\n",
      "Epoch: 826 Iteration: 73600. Train_MSE: 0.0002206710196333006. Test_MSE: 0.00014628065035833673\n",
      "Epoch: 828 Iteration: 73700. Train_MSE: 0.00013496428437065333. Test_MSE: 0.00014620175294935664\n",
      "Epoch: 829 Iteration: 73800. Train_MSE: 0.0003135902516078204. Test_MSE: 0.00014603627397808362\n",
      "Epoch: 830 Iteration: 73900. Train_MSE: 0.0002164015604648739. Test_MSE: 0.00014606147667892445\n",
      "Epoch: 831 Iteration: 74000. Train_MSE: 0.00019240727124270052. Test_MSE: 0.00014583751677433446\n",
      "Epoch: 832 Iteration: 74100. Train_MSE: 0.00012380009866319597. Test_MSE: 0.00014577653740741416\n",
      "Epoch: 833 Iteration: 74200. Train_MSE: 0.00043053284753113985. Test_MSE: 0.00014569387125282926\n",
      "Epoch: 834 Iteration: 74300. Train_MSE: 0.000266832736087963. Test_MSE: 0.00014565030592758092\n",
      "Epoch: 835 Iteration: 74400. Train_MSE: 0.0001889664854388684. Test_MSE: 0.00014552990907759582\n",
      "Epoch: 837 Iteration: 74500. Train_MSE: 0.00023159034026321024. Test_MSE: 0.00014549230144238597\n",
      "Epoch: 838 Iteration: 74600. Train_MSE: 0.0009074379340745509. Test_MSE: 0.0001452922269838242\n",
      "Epoch: 839 Iteration: 74700. Train_MSE: 0.0003088846569880843. Test_MSE: 0.00014533283464231094\n",
      "Epoch: 840 Iteration: 74800. Train_MSE: 0.00014506859588436782. Test_MSE: 0.00014512251336384928\n",
      "Epoch: 841 Iteration: 74900. Train_MSE: 0.00013857544399797916. Test_MSE: 0.0001450487212981273\n",
      "Epoch: 842 Iteration: 75000. Train_MSE: 0.0012984737986698747. Test_MSE: 0.00014497856090503223\n",
      "Epoch: 843 Iteration: 75100. Train_MSE: 0.00016207416774705052. Test_MSE: 0.0001449364093216789\n",
      "Epoch: 844 Iteration: 75200. Train_MSE: 0.000108825319330208. Test_MSE: 0.00014478609083244732\n",
      "Epoch: 846 Iteration: 75300. Train_MSE: 0.00015444324526470155. Test_MSE: 0.00014481027733004637\n",
      "Epoch: 847 Iteration: 75400. Train_MSE: 0.000306187488604337. Test_MSE: 0.00014457492944275381\n",
      "Epoch: 848 Iteration: 75500. Train_MSE: 0.0002895650395657867. Test_MSE: 0.00014459618805183453\n",
      "Epoch: 849 Iteration: 75600. Train_MSE: 0.0002521602436900139. Test_MSE: 0.00014439030323533196\n",
      "Epoch: 850 Iteration: 75700. Train_MSE: 0.00010934110468951985. Test_MSE: 0.00014433764073815583\n",
      "Epoch: 851 Iteration: 75800. Train_MSE: 0.00035561557160690427. Test_MSE: 0.00014428545208965493\n",
      "Epoch: 852 Iteration: 75900. Train_MSE: 0.00019919105398003012. Test_MSE: 0.00014421286863072413\n",
      "Epoch: 853 Iteration: 76000. Train_MSE: 0.0001363328192383051. Test_MSE: 0.00014409353738923788\n",
      "Epoch: 855 Iteration: 76100. Train_MSE: 0.00013094085443299264. Test_MSE: 0.00014411764377905123\n",
      "Epoch: 856 Iteration: 76200. Train_MSE: 0.003030716674402356. Test_MSE: 0.00014387504394319111\n",
      "Epoch: 857 Iteration: 76300. Train_MSE: 0.0001244164741365239. Test_MSE: 0.0001438657947048146\n",
      "Epoch: 858 Iteration: 76400. Train_MSE: 0.0008407182758674026. Test_MSE: 0.00014372502162463943\n",
      "Epoch: 859 Iteration: 76500. Train_MSE: 0.00020540348486974835. Test_MSE: 0.00014363314879588162\n",
      "Epoch: 860 Iteration: 76600. Train_MSE: 0.000552079698536545. Test_MSE: 0.00014356453220082502\n",
      "Epoch: 861 Iteration: 76700. Train_MSE: 0.00029867739067412913. Test_MSE: 0.0001435108568878827\n",
      "Epoch: 862 Iteration: 76800. Train_MSE: 0.00021176075097173452. Test_MSE: 0.00014339365892270133\n",
      "Epoch: 864 Iteration: 76900. Train_MSE: 0.0001183714994112961. Test_MSE: 0.00014344356260629582\n",
      "Epoch: 865 Iteration: 77000. Train_MSE: 0.0016986194532364607. Test_MSE: 0.00014323477368325435\n",
      "Epoch: 866 Iteration: 77100. Train_MSE: 0.00021588263916783035. Test_MSE: 0.00014316164058657984\n",
      "Epoch: 867 Iteration: 77200. Train_MSE: 0.00024367874721065164. Test_MSE: 0.00014307037789398154\n",
      "Epoch: 868 Iteration: 77300. Train_MSE: 0.0002449823368806392. Test_MSE: 0.00014293650801975463\n",
      "Epoch: 869 Iteration: 77400. Train_MSE: 0.00018059383728541434. Test_MSE: 0.00014289163809309057\n",
      "Epoch: 870 Iteration: 77500. Train_MSE: 0.008309886790812016. Test_MSE: 0.00014279865729101632\n",
      "Epoch: 871 Iteration: 77600. Train_MSE: 0.00019442314805928618. Test_MSE: 0.00014272178881374003\n",
      "Epoch: 873 Iteration: 77700. Train_MSE: 0.00012299444642849267. Test_MSE: 0.00014277407339380346\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 874 Iteration: 77800. Train_MSE: 0.00015666353283450007. Test_MSE: 0.00014257571865762664\n",
      "Epoch: 875 Iteration: 77900. Train_MSE: 0.0131159583106637. Test_MSE: 0.00014245547002392534\n",
      "Epoch: 876 Iteration: 78000. Train_MSE: 0.0002111602370860055. Test_MSE: 0.00014242301647882997\n",
      "Epoch: 877 Iteration: 78100. Train_MSE: 0.00024273259623441845. Test_MSE: 0.0001422292821990323\n",
      "Epoch: 878 Iteration: 78200. Train_MSE: 0.0001292571978410706. Test_MSE: 0.00014220403136425857\n",
      "Epoch: 879 Iteration: 78300. Train_MSE: 0.0010250372579321265. Test_MSE: 0.0001421605339336587\n",
      "Epoch: 880 Iteration: 78400. Train_MSE: 0.00018905433535110205. Test_MSE: 0.00014207232122536904\n",
      "Epoch: 882 Iteration: 78500. Train_MSE: 0.0003080404130741954. Test_MSE: 0.00014208107269791085\n",
      "Epoch: 883 Iteration: 78600. Train_MSE: 0.0002146053739124909. Test_MSE: 0.00014188656954165898\n",
      "Epoch: 884 Iteration: 78700. Train_MSE: 0.0011481449473649263. Test_MSE: 0.0001418316949200183\n",
      "Epoch: 885 Iteration: 78800. Train_MSE: 0.00015032495139166713. Test_MSE: 0.00014179099682754933\n",
      "Epoch: 886 Iteration: 78900. Train_MSE: 0.00010644680151017383. Test_MSE: 0.0001415739407035003\n",
      "Epoch: 887 Iteration: 79000. Train_MSE: 0.00011581159924389794. Test_MSE: 0.00014153789225187973\n",
      "Epoch: 888 Iteration: 79100. Train_MSE: 0.000923352490644902. Test_MSE: 0.00014149762957944362\n",
      "Epoch: 889 Iteration: 79200. Train_MSE: 0.00011884392006322742. Test_MSE: 0.00014141720704603862\n",
      "Epoch: 891 Iteration: 79300. Train_MSE: 0.00020927382865920663. Test_MSE: 0.00014144928552766453\n",
      "Epoch: 892 Iteration: 79400. Train_MSE: 0.00020887261780444533. Test_MSE: 0.00014124363298241083\n",
      "Epoch: 893 Iteration: 79500. Train_MSE: 0.000875507656019181. Test_MSE: 0.00014115932596848462\n",
      "Epoch: 894 Iteration: 79600. Train_MSE: 0.00021574391575995833. Test_MSE: 0.00014115874752579255\n",
      "Epoch: 895 Iteration: 79700. Train_MSE: 0.0005465527647174895. Test_MSE: 0.0001409080070871831\n",
      "Epoch: 896 Iteration: 79800. Train_MSE: 0.00017987278988584876. Test_MSE: 0.00014087942066804153\n",
      "Epoch: 897 Iteration: 79900. Train_MSE: 0.0005436715437099338. Test_MSE: 0.00014083546246561505\n",
      "Epoch: 898 Iteration: 80000. Train_MSE: 0.00019893224816769361. Test_MSE: 0.00014074818265243224\n",
      "Epoch: 900 Iteration: 80100. Train_MSE: 0.00044490929576568305. Test_MSE: 0.00014079198435876502\n",
      "Epoch: 901 Iteration: 80200. Train_MSE: 7.625994476256892e-05. Test_MSE: 0.00014059750913164403\n",
      "Epoch: 902 Iteration: 80300. Train_MSE: 0.0004726024635601789. Test_MSE: 0.00014049873494009475\n",
      "Epoch: 903 Iteration: 80400. Train_MSE: 0.0001734239049255848. Test_MSE: 0.00014050867762345089\n",
      "Epoch: 904 Iteration: 80500. Train_MSE: 0.00018598644237499684. Test_MSE: 0.00014026811236927364\n",
      "Epoch: 905 Iteration: 80600. Train_MSE: 0.00015183906361926347. Test_MSE: 0.00014022014581924674\n",
      "Epoch: 906 Iteration: 80700. Train_MSE: 0.0007514269091188908. Test_MSE: 0.00014016993659136435\n",
      "Epoch: 907 Iteration: 80800. Train_MSE: 0.00014611409278586507. Test_MSE: 0.0001401130510536946\n",
      "Epoch: 908 Iteration: 80900. Train_MSE: 0.00016534223686903715. Test_MSE: 0.00014009514587800813\n",
      "Epoch: 910 Iteration: 81000. Train_MSE: 0.00014586206816602498. Test_MSE: 0.00013993779582110362\n",
      "Epoch: 911 Iteration: 81100. Train_MSE: 0.0004080096259713173. Test_MSE: 0.00013983914533033688\n",
      "Epoch: 912 Iteration: 81200. Train_MSE: 0.0008466175058856606. Test_MSE: 0.00013985344731728071\n",
      "Epoch: 913 Iteration: 81300. Train_MSE: 0.00011316765449009836. Test_MSE: 0.00013961777516249265\n",
      "Epoch: 914 Iteration: 81400. Train_MSE: 0.00011190273653483018. Test_MSE: 0.00013958213083409933\n",
      "Epoch: 915 Iteration: 81500. Train_MSE: 0.0006055153789930046. Test_MSE: 0.00013951813351921154\n",
      "Epoch: 916 Iteration: 81600. Train_MSE: 0.00019352664821781218. Test_MSE: 0.00013945622858469752\n",
      "Epoch: 917 Iteration: 81700. Train_MSE: 0.000367422093404457. Test_MSE: 0.00013942968120158315\n",
      "Epoch: 919 Iteration: 81800. Train_MSE: 0.00015604030340909958. Test_MSE: 0.00013929626480652219\n",
      "Epoch: 920 Iteration: 81900. Train_MSE: 0.00018643909425009042. Test_MSE: 0.00013918440497837114\n",
      "Epoch: 921 Iteration: 82000. Train_MSE: 0.00020217010751366615. Test_MSE: 0.00013920875169299058\n",
      "Epoch: 922 Iteration: 82100. Train_MSE: 0.000100050623586867. Test_MSE: 0.00013897925185140208\n",
      "Epoch: 923 Iteration: 82200. Train_MSE: 0.00014093529898673296. Test_MSE: 0.00013895190353063836\n",
      "Epoch: 924 Iteration: 82300. Train_MSE: 0.0003525330394040793. Test_MSE: 0.00013887792963551468\n",
      "Epoch: 925 Iteration: 82400. Train_MSE: 0.00014190160436555743. Test_MSE: 0.00013882665263686321\n",
      "Epoch: 926 Iteration: 82500. Train_MSE: 0.00019921688362956047. Test_MSE: 0.00013874648819733288\n",
      "Epoch: 928 Iteration: 82600. Train_MSE: 0.00012780536781065166. Test_MSE: 0.0001386808810719808\n",
      "Epoch: 929 Iteration: 82700. Train_MSE: 0.00030332893948070705. Test_MSE: 0.0001385344309126551\n",
      "Epoch: 930 Iteration: 82800. Train_MSE: 0.0002016889920923859. Test_MSE: 0.000138560163780053\n",
      "Epoch: 931 Iteration: 82900. Train_MSE: 0.00017689004016574472. Test_MSE: 0.0001383555867349457\n",
      "Epoch: 932 Iteration: 83000. Train_MSE: 0.00011607592750806361. Test_MSE: 0.00013830247351421222\n",
      "Epoch: 933 Iteration: 83100. Train_MSE: 0.00041949551086872816. Test_MSE: 0.00013823506493195506\n",
      "Epoch: 934 Iteration: 83200. Train_MSE: 0.0002487622550688684. Test_MSE: 0.00013819932136700878\n",
      "Epoch: 935 Iteration: 83300. Train_MSE: 0.00017137372924480587. Test_MSE: 0.00013809006808413027\n",
      "Epoch: 937 Iteration: 83400. Train_MSE: 0.00021239154739305377. Test_MSE: 0.00013806140763614923\n",
      "Epoch: 938 Iteration: 83500. Train_MSE: 0.0008951124036684632. Test_MSE: 0.0001378827824923676\n",
      "Epoch: 939 Iteration: 83600. Train_MSE: 0.0002826204872690141. Test_MSE: 0.0001379229046639063\n",
      "Epoch: 940 Iteration: 83700. Train_MSE: 0.00012950778182130307. Test_MSE: 0.0001377310463198138\n",
      "Epoch: 941 Iteration: 83800. Train_MSE: 0.00012119000894017518. Test_MSE: 0.00013766574805261972\n",
      "Epoch: 942 Iteration: 83900. Train_MSE: 0.0012909045908600092. Test_MSE: 0.000137610012781041\n",
      "Epoch: 943 Iteration: 84000. Train_MSE: 0.00015189274563454092. Test_MSE: 0.00013757547601253163\n",
      "Epoch: 944 Iteration: 84100. Train_MSE: 9.928983490681276e-05. Test_MSE: 0.0001374395362076485\n",
      "Epoch: 946 Iteration: 84200. Train_MSE: 0.00014189067587722093. Test_MSE: 0.00013746767352493144\n",
      "Epoch: 947 Iteration: 84300. Train_MSE: 0.00029841886134818196. Test_MSE: 0.00013725645003811231\n",
      "Epoch: 948 Iteration: 84400. Train_MSE: 0.0002706355298869312. Test_MSE: 0.00013727933664861778\n",
      "Epoch: 949 Iteration: 84500. Train_MSE: 0.00023429523571394384. Test_MSE: 0.00013709194993617694\n",
      "Epoch: 950 Iteration: 84600. Train_MSE: 0.00010039989865617827. Test_MSE: 0.00013704585664749814\n",
      "Epoch: 951 Iteration: 84700. Train_MSE: 0.0003461386659182608. Test_MSE: 0.00013700702362863106\n",
      "Epoch: 952 Iteration: 84800. Train_MSE: 0.00019243134011048824. Test_MSE: 0.00013694321611393555\n",
      "Epoch: 953 Iteration: 84900. Train_MSE: 0.000127893261378631. Test_MSE: 0.00013683606168889717\n",
      "Epoch: 955 Iteration: 85000. Train_MSE: 0.0001153556804638356. Test_MSE: 0.00013686355517660722\n",
      "Epoch: 956 Iteration: 85100. Train_MSE: 0.0029973448254168034. Test_MSE: 0.00013664573610118713\n",
      "Epoch: 957 Iteration: 85200. Train_MSE: 0.0001193546995637007. Test_MSE: 0.00013664070713768252\n",
      "Epoch: 958 Iteration: 85300. Train_MSE: 0.0008123635780066252. Test_MSE: 0.00013651441947134138\n",
      "Epoch: 959 Iteration: 85400. Train_MSE: 0.0001839831966208294. Test_MSE: 0.00013643146243994458\n",
      "Epoch: 960 Iteration: 85500. Train_MSE: 0.0005375164328143001. Test_MSE: 0.0001363773390273605\n",
      "Epoch: 961 Iteration: 85600. Train_MSE: 0.00028647214639931917. Test_MSE: 0.00013632971138897995\n",
      "Epoch: 962 Iteration: 85700. Train_MSE: 0.00019419321324676275. Test_MSE: 0.00013622562922264154\n",
      "Epoch: 964 Iteration: 85800. Train_MSE: 0.00010333243699278682. Test_MSE: 0.0001362751560264716\n",
      "Epoch: 965 Iteration: 85900. Train_MSE: 0.0016917305765673518. Test_MSE: 0.00013609167365928758\n",
      "Epoch: 966 Iteration: 86000. Train_MSE: 0.00020764610962942243. Test_MSE: 0.00013602524965532293\n",
      "Epoch: 967 Iteration: 86100. Train_MSE: 0.00022933514264877886. Test_MSE: 0.0001359457168514172\n",
      "Epoch: 968 Iteration: 86200. Train_MSE: 0.00022551346046384424. Test_MSE: 0.00013582234837301093\n",
      "Epoch: 969 Iteration: 86300. Train_MSE: 0.000169850216479972. Test_MSE: 0.00013579099459607772\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 970 Iteration: 86400. Train_MSE: 0.008298207074403763. Test_MSE: 0.00013570554008543491\n",
      "Epoch: 971 Iteration: 86500. Train_MSE: 0.0001750295195961371. Test_MSE: 0.00013564063756372688\n",
      "Epoch: 973 Iteration: 86600. Train_MSE: 0.00011672219989122823. Test_MSE: 0.0001356910273064918\n",
      "Epoch: 974 Iteration: 86700. Train_MSE: 0.000146513368235901. Test_MSE: 0.0001355193059871635\n",
      "Epoch: 975 Iteration: 86800. Train_MSE: 0.013096796348690987. Test_MSE: 0.00013540686344920482\n",
      "Epoch: 976 Iteration: 86900. Train_MSE: 0.00018971574900206178. Test_MSE: 0.00013538334978053783\n",
      "Epoch: 977 Iteration: 87000. Train_MSE: 0.00021997712610755116. Test_MSE: 0.00013520466054886567\n",
      "Epoch: 978 Iteration: 87100. Train_MSE: 0.00011683462798828259. Test_MSE: 0.00013518989067916117\n",
      "Epoch: 979 Iteration: 87200. Train_MSE: 0.001015855697914958. Test_MSE: 0.00013515409545484253\n",
      "Epoch: 980 Iteration: 87300. Train_MSE: 0.00017377789481543005. Test_MSE: 0.0001350754861392037\n",
      "Epoch: 982 Iteration: 87400. Train_MSE: 0.0002826914715114981. Test_MSE: 0.00013508526543185934\n",
      "Epoch: 983 Iteration: 87500. Train_MSE: 0.0001970294106286019. Test_MSE: 0.00013491587706460868\n",
      "Epoch: 984 Iteration: 87600. Train_MSE: 0.001135471393354237. Test_MSE: 0.00013486819488319225\n",
      "Epoch: 985 Iteration: 87700. Train_MSE: 0.00013773899991065264. Test_MSE: 0.00013483314768138497\n",
      "Epoch: 986 Iteration: 87800. Train_MSE: 9.826833411352709e-05. Test_MSE: 0.00013463285134055565\n",
      "Epoch: 987 Iteration: 87900. Train_MSE: 0.00010706187458708882. Test_MSE: 0.00013460614546552423\n",
      "Epoch: 988 Iteration: 88000. Train_MSE: 0.000915048411116004. Test_MSE: 0.0001345742740445551\n",
      "Epoch: 989 Iteration: 88100. Train_MSE: 0.00011366663238732144. Test_MSE: 0.00013450277916343258\n",
      "Epoch: 991 Iteration: 88200. Train_MSE: 0.00019032973796129227. Test_MSE: 0.00013453388031995294\n",
      "Epoch: 992 Iteration: 88300. Train_MSE: 0.00019141427765134722. Test_MSE: 0.00013435353179163035\n",
      "Epoch: 993 Iteration: 88400. Train_MSE: 0.0008722788188606501. Test_MSE: 0.00013427867431174055\n",
      "Epoch: 994 Iteration: 88500. Train_MSE: 0.00019901266205124557. Test_MSE: 0.00013428055874992683\n",
      "Epoch: 995 Iteration: 88600. Train_MSE: 0.0004995433264411986. Test_MSE: 0.0001340491289356079\n",
      "Epoch: 996 Iteration: 88700. Train_MSE: 0.00016622502880636603. Test_MSE: 0.00013402788634796315\n",
      "Epoch: 997 Iteration: 88800. Train_MSE: 0.0005369008285924792. Test_MSE: 0.0001339943207808915\n",
      "Epoch: 998 Iteration: 88900. Train_MSE: 0.00018483407620806247. Test_MSE: 0.0001339151453711189\n"
     ]
    }
   ],
   "source": [
    "# Training the model\n",
    "seq_dim = 1\n",
    "\n",
    "n_iter =0\n",
    "num_samples = len(X_train)\n",
    "test_samples = len(X_test)\n",
    "batch_size = 100\n",
    "num_epochs = 1000\n",
    "feat_dim = X_train.shape[1]\n",
    "\n",
    "X_train = X_train.type(torch.FloatTensor)\n",
    "y_train = y_train.type(torch.FloatTensor)\n",
    "X_test = X_test.type(torch.FloatTensor)\n",
    "y_test = y_test.type(torch.FloatTensor)\n",
    "\n",
    "for epoch in range(num_epochs):\n",
    "    for i in range(0, int(num_samples/batch_size -1)):\n",
    "        \n",
    "        \n",
    "        features = Variable(X_train[i*batch_size:(i+1)*batch_size, :]).view(-1, seq_dim, feat_dim)\n",
    "        Kt_value = Variable(y_train[i*batch_size:(i+1)*batch_size])\n",
    "        \n",
    "        #print(\"Kt_value={}\".format(Kt_value))\n",
    "        \n",
    "        optimizer.zero_grad()\n",
    "        \n",
    "        outputs = model(features)\n",
    "        #print(\"outputs ={}\".format(outputs))\n",
    "        \n",
    "        loss = criterion(outputs, Kt_value)\n",
    "        \n",
    "        train_loss.append(loss.data[0])\n",
    "        train_iter.append(n_iter)\n",
    "\n",
    "        #print(\"loss = {}\".format(loss))\n",
    "        loss.backward()\n",
    "        \n",
    "        optimizer.step()\n",
    "        \n",
    "       \n",
    "            \n",
    "        if n_iter%100 == 0:\n",
    "            for i in range(0,int(test_samples/batch_size -1)):\n",
    "                features = Variable(X_test[i*batch_size:(i+1)*batch_size, :]).view(-1, seq_dim, feat_dim)\n",
    "                Kt_test = Variable(y_test[i*batch_size:(i+1)*batch_size])\n",
    "                \n",
    "                outputs = model(features)\n",
    "                \n",
    "                mse = np.sqrt(np.mean((Kt_test.data.numpy() - outputs.data.numpy().squeeze())**2)/num_samples)\n",
    "                \n",
    "                test_iter.append(n_iter)\n",
    "                test_loss.append(mse)\n",
    "                \n",
    "            print('Epoch: {} Iteration: {}. Train_MSE: {}. Test_MSE: {}'.format(epoch, n_iter, loss.data[0], mse))\n",
    "        n_iter += 1       "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "33820\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x12330b70>]"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD3CAYAAADmBxSSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(len(test_loss))\n",
    "#plt.plot(test_loss)\n",
    "plt.plot(train_loss,'-')\n",
    "#plt.ylim([0.000,0.99])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0xc2e3898>]"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAD3CAYAAADsd3iFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nlsJCRERiKSxERCSWwkJERGL9fzH+NxJBwV4LAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(test_loss,'r')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "#### Demornamization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "rmse = np.sqrt(mse)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "rmse_denorm = (rmse * (df_new_test['Kt'].max() - df_new_test['Kt'].min()))+ df_new_test['Kt'].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "26.801033625268513"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rmse_denorm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    3846.000000\n",
       "mean        5.949008\n",
       "std        38.282310\n",
       "min        -0.038391\n",
       "25%        -0.007695\n",
       "50%         0.195282\n",
       "75%         1.414407\n",
       "max      1801.872849\n",
       "Name: Kt, dtype: float64"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_new_test['Kt'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "### Saving train and test losses to a csv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_trainLoss = pd.DataFrame(data={'Train Loss':train_loss,'iteration':train_iter}, columns=['Train Loss','iteration'])\n",
    "df_trainLoss.to_csv('RNN Paper Results/Exp1_GoodwinCreek_TrainLoss.csv')\n",
    "df_testLoss = pd.DataFrame(data={'Test Loss':test_loss,'iteration':test_iter}, columns=['Test Loss','iteration'])\n",
    "df_testLoss.to_csv('RNN Paper Results/Exp1_GoodwinCreek_TestLoss.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "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": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}