Neural_Net_FINAL.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Neural Network"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## I. Import and Load Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Training Data was pulled from two sources. \n",
    "\n",
    "Solar Irradience data was pulled from the [National Solar Radiation Database](https://nsrdb.nrel.gov/). Pointwise GHI, DHI, and DNI data is returned hourly for an entire calendar year. All irradience values are recorded in w/m^2. Location is for a latitude longitude coordinate pair. Coordinates were chosen based on the locations of [U.S. Climate Reference Network (USCRN)](https://www.ncdc.noaa.gov/crn/) monitoring stations. This is because the USCRN has real time measurements of GHI, which can be used to make real-time predictions of DHI. The NRSDB is not updated so frequently. \n",
    "\n",
    "The production prediction module in the [omf](https://github.com/dpinney/omf/wiki) uses GHI measuremeants read from USCRN. \n",
    "\n",
    "In addition to solar irradience data, this project uses other weather data such as cloud cover, air pressure, and season data to try and model the relationship between GHI and DHI. For that, this project pulls data from [darksky](https://darksky.net/dev). And that data is combined with the NRSDB data for testing."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Please NOTE! In this notebook, the directory \"Testing_Data\" does not mean it is solely Testing Data. Rather for this Neural network, it contains both testing and training data. This is because the neural network requires much more data to train than regression. So much of the same data used to test the regression model is now being used to train the neural network. The True training data for the neural network is in \"Proudction_Test\", where the data is from USCRN and combined with data combed from darksky."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import os\n",
    "import numpy as np\n",
    "from glob import glob\n",
    "from matplotlib import pyplot as plt\n",
    "import seaborn as sns\n",
    "from random import randint\n",
    "from glob import glob\n",
    "from math import sin, cos, asin, acos, sqrt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def featureHistograms(df, features=['GHI','DHI','Cloud Cover'],):\n",
    "    sns.set_style('darkgrid')\n",
    "    fig, axs = plt.subplots(ncols=len(features))\n",
    "    for i in range(len(features)):\n",
    "        sns.distplot(df[features[i]], ax=axs[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Lets try and make a neural network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data\n"
     ]
    }
   ],
   "source": [
    "#Get 2018 Data\n",
    "_Curr_Dir = os.getcwd()\n",
    "_Testing_Data_Dir = os.path.join(_Curr_Dir, 'Testing_Data')\n",
    "print(_Testing_Data_Dir)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Day</th>\n",
       "      <th>Hour</th>\n",
       "      <th>Minute</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Clearsky DHI</th>\n",
       "      <th>Clearsky DNI</th>\n",
       "      <th>Clearsky GHI</th>\n",
       "      <th>Cloud Type</th>\n",
       "      <th>...</th>\n",
       "      <th>GHI</th>\n",
       "      <th>Relative Humidity</th>\n",
       "      <th>Solar Zenith Angle</th>\n",
       "      <th>Surface Albedo</th>\n",
       "      <th>Pressure</th>\n",
       "      <th>Precipitable Water</th>\n",
       "      <th>Wind Direction</th>\n",
       "      <th>Wind Speed</th>\n",
       "      <th>timestamps</th>\n",
       "      <th>Cloud Cover</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>30</td>\n",
       "      <td>-22.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>62.75</td>\n",
       "      <td>103.76</td>\n",
       "      <td>0.69</td>\n",
       "      <td>978</td>\n",
       "      <td>0.1</td>\n",
       "      <td>296</td>\n",
       "      <td>2.1</td>\n",
       "      <td>1514739600</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>-23.1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>66.39</td>\n",
       "      <td>114.24</td>\n",
       "      <td>0.69</td>\n",
       "      <td>978</td>\n",
       "      <td>0.1</td>\n",
       "      <td>287</td>\n",
       "      <td>2.2</td>\n",
       "      <td>1514743200</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>30</td>\n",
       "      <td>-23.6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>69.34</td>\n",
       "      <td>125.01</td>\n",
       "      <td>0.69</td>\n",
       "      <td>978</td>\n",
       "      <td>0.1</td>\n",
       "      <td>283</td>\n",
       "      <td>2.3</td>\n",
       "      <td>1514746800</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>30</td>\n",
       "      <td>-23.9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>71.25</td>\n",
       "      <td>135.74</td>\n",
       "      <td>0.69</td>\n",
       "      <td>978</td>\n",
       "      <td>0.1</td>\n",
       "      <td>283</td>\n",
       "      <td>2.3</td>\n",
       "      <td>1514750400</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>30</td>\n",
       "      <td>-24.1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>72.87</td>\n",
       "      <td>145.93</td>\n",
       "      <td>0.69</td>\n",
       "      <td>979</td>\n",
       "      <td>0.1</td>\n",
       "      <td>284</td>\n",
       "      <td>2.3</td>\n",
       "      <td>1514754000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Year  Month  Day  Hour  Minute  Temperature  Clearsky DHI  Clearsky DNI  \\\n",
       "0  2018      1    1     0      30        -22.5             0             0   \n",
       "1  2018      1    1     1      30        -23.1             0             0   \n",
       "2  2018      1    1     2      30        -23.6             0             0   \n",
       "3  2018      1    1     3      30        -23.9             0             0   \n",
       "4  2018      1    1     4      30        -24.1             0             0   \n",
       "\n",
       "   Clearsky GHI  Cloud Type  ...  GHI  Relative Humidity  Solar Zenith Angle  \\\n",
       "0             0           8  ...    0              62.75              103.76   \n",
       "1             0           0  ...    0              66.39              114.24   \n",
       "2             0           8  ...    0              69.34              125.01   \n",
       "3             0           8  ...    0              71.25              135.74   \n",
       "4             0           8  ...    0              72.87              145.93   \n",
       "\n",
       "   Surface Albedo  Pressure  Precipitable Water  Wind Direction  Wind Speed  \\\n",
       "0            0.69       978                 0.1             296         2.1   \n",
       "1            0.69       978                 0.1             287         2.2   \n",
       "2            0.69       978                 0.1             283         2.3   \n",
       "3            0.69       978                 0.1             283         2.3   \n",
       "4            0.69       979                 0.1             284         2.3   \n",
       "\n",
       "   timestamps  Cloud Cover  \n",
       "0  1514739600          0.0  \n",
       "1  1514743200          0.0  \n",
       "2  1514746800          0.0  \n",
       "3  1514750400          0.0  \n",
       "4  1514754000          0.0  \n",
       "\n",
       "[5 rows x 24 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.read_csv(os.path.join(_Testing_Data_Dir, 'psm_south_dakota', 'psm_testing_data2018.csv')).head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df = pd.read_csv(os.path.join(_Testing_Data_Dir, 'psm_south_dakota', 'psm_testing_data2018.csv'))\n",
    "featureHistograms(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## II. Features and Data Engineering"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So we have all the above variables....Which ones do we think would work best in a NN?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So we want to predict DHI, at the very least. Maybe DNI as well. What can we use to predict? \n",
    "[More Information on GHI, DHI, and DNI here!](https://firstgreenconsulting.wordpress.com/2012/04/26/differentiate-between-the-dni-dhi-and-ghi/#:~:text=Global%20Horizontal%20Irradiance%20(GHI)%20is,Diffuse%20Horizontal%20Irradiance%20(DHI).)\n",
    "\n",
    "But a useful relationship to keep in mind is:\n",
    "\n",
    "*Global Horizontal (GHI) = Direct Normal (DNI) X cos(θ) + Diffuse Horizontal (DHI)*\n",
    "\n",
    "**GHI** is a must, as it is the primary metric our sensors will feed into such a model. \n",
    "\n",
    "**Cloud cover** is also essential, as cloud cover is one of the primary factors in how much direct normal solar irradience is diffused.\n",
    "\n",
    "**Solar Zenith Angle** is important as it dictates how much irradience is direct normal (DNI)\n",
    "\n",
    "Another useful relatinship to capture is **time**. We are given timestamps in addition to month, day, year, second, hour and minute variables. Timestamps and/or a combination between day + month would not be super useful, as it would overfit to the day. \n",
    "A more useful relationship would be to capture time of day. This is useful because in a non - featurized dataset, many GHI readings will read '0', simply because it is dark out. We are primarily concerned about the relationship between GHI and DHI, therefore factoring in readings when there is no sunlight to begin with could be confounding. Therefore adding in **minute** and **hour** variables in the input vector could adequately capture this relationship. \n",
    "\n",
    "Another solution would be to featurize this dataset, whereby we create a seperate feature which is a combination of day + time. OR, we could completly remove the nighttime entries entirely to ensure we only are concerned with readings that occur when the sun is out. However this occurs on a gradient, because sunlight is defracted even after sunset. So for now, let's just fit the model on the unedited dataset. \n",
    "\n",
    "Another good idea is to create a **season** variable, which is a dummy variable for each season (Fall, Winter, Spring, Summer), as different amounts of sunlight occur during each. The problem is that, for different locations on earth, seasonality could be inversely correlated with sunlight and thus GHI, DHI, and DNI. \n",
    "\n",
    "To keep the model as general as possible, we will exclude seasonality. \n",
    "\n",
    "Another useful metric is **Pressure**. According to the renowned [DISC model](https://www.nrel.gov/docs/legosti/old/3087.pdf), air pressure is included as a proxy for *air mass*, which is the amount of atmosphere solar rays have to travel through. The larger the solar mass, the more solar rays are scattered. \n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Loading the Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Get training arrays and combine them into a useable vector\n",
    "\n",
    "-Useful params:  \n",
    "-Hours  \n",
    "-Cloud Cover  \n",
    "-GHI  \n",
    "-Solar Zenith  \n",
    "-Pressure  \n",
    "\n",
    "\n",
    "Training Locations with Lat/Lon:  \n",
    "-**Reliance, South Dakota** (43.85, -99.50) from 2010-2018  \n",
    "-**Everglades, FL** (26.004157 ,-81.119239) from 2010-2018  \n",
    "-**Spokane, WA** (47.41,-117.52) from 2010-2018  \n",
    "-**Austin, TX** (30.581736,-98.024098) from 2010-2018  \n",
    "-**Charlottesville, VA** (38.0086,-78.4532) from 2010-2018  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Please NOTE! In this notebook, the directory \"Testing_Data\" does not mean it is solely Testing Data. Rather for this Neural network, it contains both testing and training data. \n",
    "This is because the neural network requires much more data to train than regression. So much of the same data used to test the regression model is now being used to train the neural network. The True training data for the neural network is in \"Proudction_Test\", where the data is from USCRN and combined with data combed from darksky. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "def combine_training_arrays(df, input_arry, output_arr, dni_arr):\n",
    "    ghi  = df['GHI'].values\n",
    "    cloud_cover = df['Cloud Cover'].values\n",
    "    hours = df['Hour'].values\n",
    "    minutes = df['Minute'].values\n",
    "    solar_zenith = df['Solar Zenith Angle'].values\n",
    "    pressure = df['Pressure'].values\n",
    "    dhi = df['DHI'].values\n",
    "    dni = df['DNI'].values\n",
    "    #Sanity check, should all be 8760\n",
    "    assert len(pressure)==len(solar_zenith)==len(minutes)==len(hours)==len(cloud_cover)==len(ghi)==len(dni)==8760, \"len of input array not 8760\"\n",
    "    ar = np.array([ghi, cloud_cover, hours, minutes, solar_zenith, pressure]).T\n",
    "    input_arry = np.concatenate((input_arry, ar))\n",
    "    output_arr = np.concatenate((output_arr, np.array(dhi).T))\n",
    "    dni_arr = np.concatenate((dni_arr, np.array(dni).T))\n",
    "    return input_arry, output_arr, dni_arr\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[]\n",
      "[]\n",
      "[]\n",
      "394200 (394200, 6)\n",
      "394200 (394200,)\n",
      "394200 (394200,)\n"
     ]
    }
   ],
   "source": [
    "# test_files = ['Murphey_ID', 'OldTown_ME', 'Quinault_WA', 'Lincoln_NE', 'SanDiego_CA', 'Boulder_CO', 'Miami_FL', 'outer_banks']\n",
    "training_files = ['psm_south_dakota', 'Everglades_FL', 'Spokane_WA', 'Austin_TX', 'Charlottesville']\n",
    "input_size = 6\n",
    "input_arry = np.array([], dtype=np.float64).reshape(0,input_size)\n",
    "output_arry = np.array([], dtype=np.float64)\n",
    "dni_arr = np.array([], dtype=np.float64)\n",
    "print(input_arry)\n",
    "print(output_arry)\n",
    "print(dni_arr)\n",
    "for directory in training_files:\n",
    "    for file in glob(_Testing_Data_Dir+'/'+directory+'/psm_*'):\n",
    "        df = pd.read_csv(file)\n",
    "        input_arry, output_arry, dni_arr = combine_training_arrays(df, input_arry, output_arry, dni_arr)\n",
    "print(len(input_arry), input_arry.shape)\n",
    "print(len(output_arry), output_arry.shape)\n",
    "print(len(dni_arr), dni_arr.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "630720\n"
     ]
    }
   ],
   "source": [
    "#sanity check, make sure 8760*number of files we look at == 429240\n",
    "count = 0\n",
    "for path in glob(_Testing_Data_Dir+'/*'):\n",
    "    for file in glob(path+'/psm_'+'*'):\n",
    "        count+=1\n",
    "print(count*8760)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# III. Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.optimizers import Adam"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "keras.backend.clear_session()\n",
    "tf.keras.backend.clear_session()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Epoch size of 50 seems reasonable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<function tensorflow.python.framework.ops.reset_default_graph()>"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "keras.backend.clear_session()\n",
    "tf.keras.backend.clear_session()\n",
    "tf.compat.v1.reset_default_graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "keras.backend.clear_session()\n",
    "tf.keras.backend.clear_session()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [],
   "source": [
    "#make RMSE a custom metric"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [],
   "source": [
    "# def rmse(y_true, y_pred):\n",
    "#     return keras.backend.sqrt(keras.backend.mean(keras.backend.square(y_pred - y_true), axis=-1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = keras.Sequential()\n",
    "model.add(layers.Dense(300, activation=\"relu\"))\n",
    "model.add(layers.Dense(200, activation=\"relu\"))\n",
    "model.add(layers.Dense(100, activation=\"relu\"))\n",
    "model.add(layers.Dense(50, activation=\"relu\"))\n",
    "model.add(layers.Dense(25, activation=\"relu\"))\n",
    "model.add(layers.Dense(1, activation='relu'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 1012.1399 - mae: 14.4956 - root_mean_squared_error: 31.8141\n",
      "Epoch 2/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 445.2907 - mae: 9.3514 - root_mean_squared_error: 21.1019\n",
      "Epoch 3/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 396.3455 - mae: 8.7124 - root_mean_squared_error: 19.9084\n",
      "Epoch 4/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 370.4831 - mae: 8.3535 - root_mean_squared_error: 19.2479\n",
      "Epoch 5/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 340.4404 - mae: 7.9697 - root_mean_squared_error: 18.4510\n",
      "Epoch 6/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 312.7972 - mae: 7.5872 - root_mean_squared_error: 17.6861\n",
      "Epoch 7/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 292.4026 - mae: 7.3077 - root_mean_squared_error: 17.0998\n",
      "Epoch 8/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 263.2337 - mae: 6.8830 - root_mean_squared_error: 16.2245\n",
      "Epoch 9/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 244.8074 - mae: 6.6382 - root_mean_squared_error: 15.6463\n",
      "Epoch 10/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 222.2033 - mae: 6.2699 - root_mean_squared_error: 14.9065: 0s - loss: 223.1092 - mae: 6.2826 - root_m\n",
      "Epoch 11/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 203.0965 - mae: 5.9993 - root_mean_squared_error: 14.2512\n",
      "Epoch 12/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 191.9607 - mae: 5.8103 - root_mean_squared_error: 13.8550: 0s - loss: 191.5089 - mae: 5.8010 - root_mean_squared_error: 1 - ETA: 0s - loss: 192.2996 - mae: 5.8132 - root_mean_squared_er\n",
      "Epoch 13/100\n",
      "6160/6160 [==============================] - 9s 2ms/step - loss: 179.8606 - mae: 5.6234 - root_mean_squared_error: 13.4112: 2s - loss: 179.0194 - mae: 5 - E\n",
      "Epoch 14/100\n",
      "6160/6160 [==============================] - 14s 2ms/step - loss: 174.7091 - mae: 5.5609 - root_mean_squared_error: 13.2178\n",
      "Epoch 15/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 170.5586 - mae: 5.4800 - root_mean_squared_error: 13.0598\n",
      "Epoch 16/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 168.9539 - mae: 5.4464 - root_mean_squared_error: 12.9982\n",
      "Epoch 17/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 165.4859 - mae: 5.4057 - root_mean_squared_error: 12.8641\n",
      "Epoch 18/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 160.8212 - mae: 5.3327 - root_mean_squared_error: 12.6815\n",
      "Epoch 19/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 163.3155 - mae: 5.3722 - root_mean_squared_error: 12.7795\n",
      "Epoch 20/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 157.6197 - mae: 5.2856 - root_mean_squared_error: 12.5547\n",
      "Epoch 21/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 156.9302 - mae: 5.2849 - root_mean_squared_error: 12.5272\n",
      "Epoch 22/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 159.0625 - mae: 5.3004 - root_mean_squared_error: 12.6120\n",
      "Epoch 23/100\n",
      "6160/6160 [==============================] - 13s 2ms/step - loss: 154.3304 - mae: 5.2290 - root_mean_squared_error: 12.4230\n",
      "Epoch 24/100\n",
      "6160/6160 [==============================] - 19s 3ms/step - loss: 154.7006 - mae: 5.2152 - root_mean_squared_error: 12.4379\n",
      "Epoch 25/100\n",
      "6160/6160 [==============================] - 13s 2ms/step - loss: 154.9596 - mae: 5.2282 - root_mean_squared_error: 12.4483\n",
      "Epoch 26/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 147.2494 - mae: 5.0815 - root_mean_squared_error: 12.1346\n",
      "Epoch 27/100\n",
      "6160/6160 [==============================] - 13s 2ms/step - loss: 152.0258 - mae: 5.1752 - root_mean_squared_error: 12.3299\n",
      "Epoch 28/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 151.9592 - mae: 5.1578 - root_mean_squared_error: 12.3272\n",
      "Epoch 29/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 147.6575 - mae: 5.0804 - root_mean_squared_error: 12.1514\n",
      "Epoch 30/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 146.6293 - mae: 5.0639 - root_mean_squared_error: 12.1091\n",
      "Epoch 31/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 149.9240 - mae: 5.1259 - root_mean_squared_error: 12.2443\n",
      "Epoch 32/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 146.4019 - mae: 5.0593 - root_mean_squared_error: 12.0997\n",
      "Epoch 33/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 144.2696 - mae: 5.0312 - root_mean_squared_error: 12.0112\n",
      "Epoch 34/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 143.5508 - mae: 4.9922 - root_mean_squared_error: 11.9813\n",
      "Epoch 35/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 144.2494 - mae: 5.0164 - root_mean_squared_error: 12.0104 1s - loss: 144.1275 \n",
      "Epoch 36/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 139.9419 - mae: 4.9414 - root_mean_squared_error: 11.8297\n",
      "Epoch 37/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 143.5862 - mae: 5.0025 - root_mean_squared_error: 11.9827\n",
      "Epoch 38/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 142.4152 - mae: 4.9942 - root_mean_squared_error: 11.9338 0s - loss: 142.3435 - mae: 4.9920 - root_mean_squared_error: 1\n",
      "Epoch 39/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 139.7550 - mae: 4.9345 - root_mean_squared_error: 11.8218\n",
      "Epoch 40/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 140.8963 - mae: 4.9416 - root_mean_squared_error: 11.8700 1s - loss: 142\n",
      "Epoch 41/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 140.9589 - mae: 4.9564 - root_mean_squared_error: 11.8726 1s - loss: 139.0666 - ma\n",
      "Epoch 42/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 137.3897 - mae: 4.8710 - root_mean_squared_error: 11.7213\n",
      "Epoch 43/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 139.6133 - mae: 4.9100 - root_mean_squared_error: 11.8158\n",
      "Epoch 44/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 139.2838 - mae: 4.8928 - root_mean_squared_error: 11.8019 0s - loss: 1\n",
      "Epoch 45/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 134.4771 - mae: 4.8148 - root_mean_squared_error: 11.5964\n",
      "Epoch 46/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 134.8975 - mae: 4.8266 - root_mean_squared_error: 11.6145\n",
      "Epoch 47/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 135.6002 - mae: 4.8570 - root_mean_squared_error: 11.6448\n",
      "Epoch 48/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 135.8402 - mae: 4.8262 - root_mean_squared_error: 11.6551\n",
      "Epoch 49/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 133.6612 - mae: 4.8119 - root_mean_squared_error: 11.5612\n",
      "Epoch 50/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 132.9666 - mae: 4.7829 - root_mean_squared_error: 11.5311 0s - loss: 133.0261 - mae: 4.7843 - root_mea\n",
      "Epoch 51/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 134.7263 - mae: 4.8140 - root_mean_squared_error: 11.6072 0s - loss: 135.5518 - mae: 4.8334 - root\n",
      "Epoch 52/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 133.1790 - mae: 4.7881 - root_mean_squared_error: 11.5403\n",
      "Epoch 53/100\n",
      "6160/6160 [==============================] - 13s 2ms/step - loss: 132.4406 - mae: 4.7833 - root_mean_squared_error: 11.5083\n",
      "Epoch 54/100\n",
      "6160/6160 [==============================] - 13s 2ms/step - loss: 132.7711 - mae: 4.7806 - root_mean_squared_error: 11.5226 0s - loss: 132.3691 - mae: 4.7686 - root_mean_squared_\n",
      "Epoch 55/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 132.6906 - mae: 4.7661 - root_mean_squared_error: 11.5191\n",
      "Epoch 56/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6160/6160 [==============================] - 12s 2ms/step - loss: 131.9631 - mae: 4.7476 - root_mean_squared_error: 11.4875 1s - loss: 131.6468 - mae: 4.7\n",
      "Epoch 57/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 130.6452 - mae: 4.7363 - root_mean_squared_error: 11.4300\n",
      "Epoch 58/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 131.4985 - mae: 4.7599 - root_mean_squared_error: 11.4673 0s - loss: 131.7068 - mae: 4.7653 - root_mean_squared_error: 11.47 - ETA: 0s - loss: 131.6584 - mae: 4.7638 - ro\n",
      "Epoch 59/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 130.1663 - mae: 4.7236 - root_mean_squared_error: 11.4090 0s - loss: 129.4740 - mae: 4.7154 - root_mean_\n",
      "Epoch 60/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 127.1773 - mae: 4.6620 - root_mean_squared_error: 11.2773 0s - loss: 126.9708 - mae: 4.6596 - root_mean_squared_\n",
      "Epoch 61/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 130.0475 - mae: 4.7096 - root_mean_squared_error: 11.4038 0s - loss: 130.0600 - mae: 4.7076 - root\n",
      "Epoch 62/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 127.4631 - mae: 4.6658 - root_mean_squared_error: 11.2900 0s - loss: 127.5345 - mae: 4.6627 - root_m\n",
      "Epoch 63/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 127.6020 - mae: 4.6568 - root_mean_squared_error: 11.2961\n",
      "Epoch 64/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 125.1774 - mae: 4.6242 - root_mean_squared_error: 11.1883\n",
      "Epoch 65/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 124.8820 - mae: 4.5948 - root_mean_squared_error: 11.1751\n",
      "Epoch 66/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 125.3257 - mae: 4.6167 - root_mean_squared_error: 11.1949\n",
      "Epoch 67/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 123.9765 - mae: 4.5821 - root_mean_squared_error: 11.1345\n",
      "Epoch 68/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 123.9359 - mae: 4.5763 - root_mean_squared_error: 11.1326\n",
      "Epoch 69/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 123.2154 - mae: 4.5644 - root_mean_squared_error: 11.1002\n",
      "Epoch 70/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 121.8663 - mae: 4.5329 - root_mean_squared_error: 11.0393\n",
      "Epoch 71/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 121.8820 - mae: 4.5393 - root_mean_squared_error: 11.0400 1s - los - ETA: 0s - loss: 121.5485 - mae: 4.5353 - root_mean_squa\n",
      "Epoch 72/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 120.3734 - mae: 4.5006 - root_mean_squared_error: 10.9715\n",
      "Epoch 73/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 121.5694 - mae: 4.5179 - root_mean_squared_error: 11.0259\n",
      "Epoch 74/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 118.7868 - mae: 4.4758 - root_mean_squared_error: 10.8989 1s - loss: 118\n",
      "Epoch 75/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 118.4425 - mae: 4.4662 - root_mean_squared_error: 10.8831 2s - loss: 119.7441  - ETA: 1s - loss: 118.7\n",
      "Epoch 76/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 119.6849 - mae: 4.4723 - root_mean_squared_error: 10.9401\n",
      "Epoch 77/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 118.2670 - mae: 4.4564 - root_mean_squared_error: 10.8751\n",
      "Epoch 78/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 117.9102 - mae: 4.4599 - root_mean_squared_error: 10.8586\n",
      "Epoch 79/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 119.4891 - mae: 4.4798 - root_mean_squared_error: 10.9311\n",
      "Epoch 80/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 116.5976 - mae: 4.4137 - root_mean_squared_error: 10.7980 1s - loss: 116.3013 - mae: 4 - ETA: 0s - loss: 116.5436 - mae: 4.4112 - root_mean_squa\n",
      "Epoch 81/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 117.8149 - mae: 4.4350 - root_mean_squared_error: 10.8543\n",
      "Epoch 82/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 117.8128 - mae: 4.4354 - root_mean_squared_error: 10.8542\n",
      "Epoch 83/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 116.1864 - mae: 4.4102 - root_mean_squared_error: 10.7790\n",
      "Epoch 84/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 116.0883 - mae: 4.3929 - root_mean_squared_error: 10.7744\n",
      "Epoch 85/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 115.3354 - mae: 4.3785 - root_mean_squared_error: 10.7394\n",
      "Epoch 86/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 115.1078 - mae: 4.3804 - root_mean_squared_error: 10.7288\n",
      "Epoch 87/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 115.6509 - mae: 4.3730 - root_mean_squared_error: 10.7541 0s - loss: 115.7025 - mae: 4.3731 - root_mean_squared_error: 10.75\n",
      "Epoch 88/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 114.5234 - mae: 4.3740 - root_mean_squared_error: 10.7016\n",
      "Epoch 89/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 114.8072 - mae: 4.3713 - root_mean_squared_error: 10.7148\n",
      "Epoch 90/100\n",
      "6160/6160 [==============================] - 9s 2ms/step - loss: 113.6757 - mae: 4.3526 - root_mean_squared_error: 10.6619\n",
      "Epoch 91/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 113.5022 - mae: 4.3259 - root_mean_squared_error: 10.6537\n",
      "Epoch 92/100\n",
      "6160/6160 [==============================] - 13s 2ms/step - loss: 113.2030 - mae: 4.3412 - root_mean_squared_error: 10.6397\n",
      "Epoch 93/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 114.5600 - mae: 4.3563 - root_mean_squared_error: 10.7033A: 13s - loss: 119.6298 - mae: 4.4161 - root_mean_ - ETA: 11s - loss: 118.6957 - mae: 4.4280 - ro\n",
      "Epoch 94/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 113.4968 - mae: 4.3402 - root_mean_squared_error: 10.6535 2s - loss: 111.9676 - mae: 4.311 - ETA: 1s - loss: 112.3\n",
      "Epoch 95/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 113.6744 - mae: 4.3487 - root_mean_squared_error: 10.6618 \n",
      "Epoch 96/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 114.2506 - mae: 4.3451 - root_mean_squared_error: 10.6888\n",
      "Epoch 97/100\n",
      "6160/6160 [==============================] - 9s 2ms/step - loss: 112.0852 - mae: 4.3095 - root_mean_squared_error: 10.5870\n",
      "Epoch 98/100\n",
      "6160/6160 [==============================] - 9s 2ms/step - loss: 113.0469 - mae: 4.3278 - root_mean_squared_error: 10.6323\n",
      "Epoch 99/100\n",
      "6160/6160 [==============================] - 9s 2ms/step - loss: 113.8454 - mae: 4.3310 - root_mean_squared_error: 10.6698\n",
      "Epoch 100/100\n",
      "6160/6160 [==============================] - 9s 2ms/step - loss: 111.3855 - mae: 4.2959 - root_mean_squared_error: 10.5539\n"
     ]
    }
   ],
   "source": [
    "#Compile and train\n",
    "model.compile(optimizer=Adam(lr=0.0001), \n",
    "              loss='mse',\n",
    "              metrics=['mae',keras.metrics.RootMeanSquaredError()])\n",
    "\n",
    "history = model.fit(input_arry, output_arry, epochs=100, batch_size=64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense (Dense)                (None, 300)               2100      \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 200)               60200     \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 100)               20100     \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 50)                5050      \n",
      "_________________________________________________________________\n",
      "dense_4 (Dense)              (None, 25)                1275      \n",
      "_________________________________________________________________\n",
      "dense_5 (Dense)              (None, 1)                 26        \n",
      "=================================================================\n",
      "Total params: 88,751\n",
      "Trainable params: 88,751\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['mae'])\n",
    "plt.plot(history.history['root_mean_squared_error'])\n",
    "plt.title('MAE and RMSE per epoch')\n",
    "plt.ylabel('Error')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['MAE', 'RMSE'], loc='middle')\n",
    "plt.savefig('error-v-epoch.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "#mae: 4.2719\n",
    "#rmse 10.51"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "#MAE of around 4 w/m2. Considering DHI values go from 0 - 400, 4 represents a 1% error value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Prediction with DNI included\n",
    "\n",
    "\n",
    "keras.backend.clear_session()\n",
    "tf.keras.backend.clear_session()\n",
    "\n",
    "keras.backend.clear_session()\n",
    "tf.keras.backend.clear_session()\n",
    "tf.compat.v1.reset_default_graph\n",
    "\n",
    "keras.backend.clear_session()\n",
    "tf.keras.backend.clear_session()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_dni = keras.Sequential()\n",
    "model_dni.add(layers.Dense(300, activation=\"relu\"))\n",
    "model_dni.add(layers.Dense(200, activation=\"relu\"))\n",
    "model_dni.add(layers.Dense(100, activation=\"relu\"))\n",
    "model_dni.add(layers.Dense(50, activation=\"relu\"))\n",
    "model_dni.add(layers.Dense(25, activation=\"relu\"))\n",
    "model_dni.add(layers.Dense(1, activation='relu'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "6160/6160 [==============================] - 9s 2ms/step - loss: 9529.4062 - mae: 49.0188 - root_mean_squared_error: 97.6187\n",
      "Epoch 2/100\n",
      "6160/6160 [==============================] - 9s 2ms/step - loss: 2861.3137 - mae: 25.7829 - root_mean_squared_error: 53.4912\n",
      "Epoch 3/100\n",
      "6160/6160 [==============================] - 9s 2ms/step - loss: 2433.6155 - mae: 23.4008 - root_mean_squared_error: 49.3317\n",
      "Epoch 4/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 2124.7478 - mae: 21.7659 - root_mean_squared_error: 46.0950\n",
      "Epoch 5/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 1865.5197 - mae: 20.3368 - root_mean_squared_error: 43.1917: 1s -\n",
      "Epoch 6/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 1592.6884 - mae: 18.7421 - root_mean_squared_error: 39.9085\n",
      "Epoch 7/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 1296.1600 - mae: 16.8822 - root_mean_squared_error: 36.0022\n",
      "Epoch 8/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 996.2180 - mae: 14.7837 - root_mean_squared_error: 31.5629\n",
      "Epoch 9/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 778.9993 - mae: 13.0857 - root_mean_squared_error: 27.9106\n",
      "Epoch 10/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 713.4011 - mae: 12.4539 - root_mean_squared_error: 26.7096\n",
      "Epoch 11/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 673.1339 - mae: 12.0620 - root_mean_squared_error: 25.9448A: 5s - loss: 664.4157 - mae: 12.0199 - ro\n",
      "Epoch 12/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 640.1256 - mae: 11.7559 - root_mean_squared_error: 25.3007\n",
      "Epoch 13/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 639.5390 - mae: 11.6984 - root_mean_squared_error: 25.2891\n",
      "Epoch 14/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 611.5479 - mae: 11.4388 - root_mean_squared_error: 24.7295\n",
      "Epoch 15/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 605.9249 - mae: 11.3442 - root_mean_squared_error: 24.6155A\n",
      "Epoch 16/100\n",
      "6160/6160 [==============================] - 9s 2ms/step - loss: 594.8214 - mae: 11.2070 - root_mean_squared_error: 24.3890\n",
      "Epoch 17/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 590.8195 - mae: 11.2011 - root_mean_squared_error: 24.3068A: 1s - loss: 584.5526 - mae: 11.1455 - root_mean\n",
      "Epoch 18/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 572.9796 - mae: 10.9984 - root_mean_squared_error: 23.9370\n",
      "Epoch 19/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 558.0630 - mae: 10.8696 - root_mean_squared_error: 23.6234A: 2s - loss: 557.2150 - mae\n",
      "Epoch 20/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 559.0411 - mae: 10.8675 - root_mean_squared_error: 23.6441\n",
      "Epoch 21/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 561.0959 - mae: 10.8462 - root_mean_squared_error: 23.6875\n",
      "Epoch 22/100\n",
      "6160/6160 [==============================] - 9s 2ms/step - loss: 552.9545 - mae: 10.7541 - root_mean_squared_error: 23.5150\n",
      "Epoch 23/100\n",
      "6160/6160 [==============================] - 15s 2ms/step - loss: 542.6432 - mae: 10.6888 - root_mean_squared_error: 23.2947: 0s - loss: 543.4939 - mae: 10.7063 - root_mean\n",
      "Epoch 24/100\n",
      "6160/6160 [==============================] - 15s 2ms/step - loss: 540.5709 - mae: 10.6554 - root_mean_squared_error: 23.2502\n",
      "Epoch 25/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 557.5766 - mae: 10.8270 - root_mean_squared_error: 23.6131: 8s - loss: 562.1246 - mae: 10.8351 - root_mean_squared_error: 23.709 - ETA: 8s - lo - ETA\n",
      "Epoch 26/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 530.8529 - mae: 10.5160 - root_mean_squared_error: 23.0402\n",
      "Epoch 27/100\n",
      "6160/6160 [==============================] - 18s 3ms/step - loss: 526.1266 - mae: 10.4706 - root_mean_squared_error: 22.9375: 9s - loss: 528.0287 - mae: 10.5000 - ro\n",
      "Epoch 28/100\n",
      "6160/6160 [==============================] - 14s 2ms/step - loss: 528.0776 - mae: 10.4892 - root_mean_squared_error: 22.9799\n",
      "Epoch 29/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 513.3091 - mae: 10.3485 - root_mean_squared_error: 22.6563\n",
      "Epoch 30/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 527.3453 - mae: 10.4627 - root_mean_squared_error: 22.9640\n",
      "Epoch 31/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 511.9346 - mae: 10.3178 - root_mean_squared_error: 22.6260\n",
      "Epoch 32/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 508.4727 - mae: 10.2883 - root_mean_squared_error: 22.5493: \n",
      "Epoch 33/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 502.9880 - mae: 10.2020 - root_mean_squared_error: 22.4274: 7s - loss: 511.5997 -  - ETA: 0s - loss: 505.5038 - mae: 10.2274 - root_mean_sq\n",
      "Epoch 34/100\n",
      "6160/6160 [==============================] - 9s 2ms/step - loss: 502.4905 - mae: 10.2157 - root_mean_squared_error: 22.4163A: 3s - loss: 501.6923 - mae:  - ETA: 0s - loss: 500.9402 - mae: 10.2106 - root_mean_sq\n",
      "Epoch 35/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 498.1489 - mae: 10.1590 - root_mean_squared_error: 22.3192\n",
      "Epoch 36/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 498.9307 - mae: 10.1824 - root_mean_squared_error: 22.3368\n",
      "Epoch 37/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 499.8658 - mae: 10.1729 - root_mean_squared_error: 22.3577\n",
      "Epoch 38/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 503.9407 - mae: 10.2247 - root_mean_squared_error: 22.4486\n",
      "Epoch 39/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 490.7081 - mae: 10.0802 - root_mean_squared_error: 22.1519\n",
      "Epoch 40/100\n",
      "6160/6160 [==============================] - 9s 2ms/step - loss: 486.0206 - mae: 10.0292 - root_mean_squared_error: 22.0459A: 1s - loss: 486.6059 - mae: 10.0547 - root_\n",
      "Epoch 41/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 499.1147 - mae: 10.1765 - root_mean_squared_error: 22.3409: 2s - loss:\n",
      "Epoch 42/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 494.8697 - mae: 10.1117 - root_mean_squared_error: 22.2457\n",
      "Epoch 43/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 484.3696 - mae: 10.0107 - root_mean_squared_error: 22.0084\n",
      "Epoch 44/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 466.2708 - mae: 9.7976 - root_mean_squared_error: 21.5933\n",
      "Epoch 45/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 476.3433 - mae: 9.9167 - root_mean_squared_error: 21.8253 1s - loss: 471.0804 - mae: 9.875 - ETA: 0s - loss: 475.9321 - mae: 9.9133 - root_mean_squared_error: 21. - ETA: 0s - loss: 476.7399 - mae: 9.9189 \n",
      "Epoch 46/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 470.3488 - mae: 9.8668 - root_mean_squared_error: 21.6875\n",
      "Epoch 47/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 476.2301 - mae: 9.9020 - root_mean_squared_error: 21.8227\n",
      "Epoch 48/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 475.2742 - mae: 9.8908 - root_mean_squared_error: 21.8008\n",
      "Epoch 49/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 465.4186 - mae: 9.7983 - root_mean_squared_error: 21.5736\n",
      "Epoch 50/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 479.1547 - mae: 9.9243 - root_mean_squared_error: 21.8896\n",
      "Epoch 51/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 462.8831 - mae: 9.7459 - root_mean_squared_error: 21.5147\n",
      "Epoch 52/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 462.8835 - mae: 9.7464 - root_mean_squared_error: 21.5147 5s - loss:\n",
      "Epoch 53/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 471.1370 - mae: 9.8334 - root_mean_squared_error: 21.7057 0s - loss: 472.8032 - mae:\n",
      "Epoch 54/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 461.9884 - mae: 9.7371 - root_mean_squared_error: 21.4939\n",
      "Epoch 55/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6160/6160 [==============================] - 9s 1ms/step - loss: 460.0359 - mae: 9.7115 - root_mean_squared_error: 21.4484\n",
      "Epoch 56/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 455.6392 - mae: 9.6479 - root_mean_squared_error: 21.3457\n",
      "Epoch 57/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 451.5148 - mae: 9.6181 - root_mean_squared_error: 21.2489: 1s - loss: 449.9117 - ma\n",
      "Epoch 58/100\n",
      "6160/6160 [==============================] - 9s 2ms/step - loss: 455.8871 - mae: 9.6610 - root_mean_squared_error: 21.3515\n",
      "Epoch 59/100\n",
      "6160/6160 [==============================] - 9s 2ms/step - loss: 451.0406 - mae: 9.6131 - root_mean_squared_error: 21.2377\n",
      "Epoch 60/100\n",
      "6160/6160 [==============================] - 9s 2ms/step - loss: 447.3058 - mae: 9.5531 - root_mean_squared_error: 21.1496: 0s - loss: 448.4530 - mae: 9.5630 - root_mean_square\n",
      "Epoch 61/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 447.4164 - mae: 9.5676 - root_mean_squared_error: 21.1522 3s - loss: 442.184\n",
      "Epoch 62/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 449.2356 - mae: 9.5992 - root_mean_squared_error: 21.1952\n",
      "Epoch 63/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 447.4100 - mae: 9.5428 - root_mean_squared_error: 21.1521\n",
      "Epoch 64/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 441.3284 - mae: 9.4965 - root_mean_squared_error: 21.0078\n",
      "Epoch 65/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 438.3840 - mae: 9.4484 - root_mean_squared_error: 20.9376 5s - loss: 444.3674 - mae: 9.5201 - root_mean_sq - ETA: 5s - loss: 442.3891 - mae: 9.5024 - root_mean_squared_erro - ETA: \n",
      "Epoch 66/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 444.1095 - mae: 9.5113 - root_mean_squared_error: 21.0739\n",
      "Epoch 67/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 441.5146 - mae: 9.4872 - root_mean_squared_error: 21.0122\n",
      "Epoch 68/100\n",
      "6160/6160 [==============================] - 9s 2ms/step - loss: 444.9576 - mae: 9.5202 - root_mean_squared_error: 21.0940\n",
      "Epoch 69/100\n",
      "6160/6160 [==============================] - 9s 2ms/step - loss: 453.7552 - mae: 9.6227 - root_mean_squared_error: 21.3015\n",
      "Epoch 70/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 442.9788 - mae: 9.4911 - root_mean_squared_error: 21.0471\n",
      "Epoch 71/100\n",
      "6160/6160 [==============================] - 9s 2ms/step - loss: 439.0236 - mae: 9.4409 - root_mean_squared_error: 20.9529: 1s - loss: 442.3371 -  - ETA: 0s - loss: 439.1139 - mae: 9.4403 - root_mean_squared_error:\n",
      "Epoch 72/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 439.0179 - mae: 9.4473 - root_mean_squared_error: 20.9528\n",
      "Epoch 73/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 439.3368 - mae: 9.4516 - root_mean_squared_error: 20.9604\n",
      "Epoch 74/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 439.5672 - mae: 9.4460 - root_mean_squared_error: 20.9659\n",
      "Epoch 75/100\n",
      "6160/6160 [==============================] - 13s 2ms/step - loss: 438.7420 - mae: 9.4410 - root_mean_squared_error: 20.9462\n",
      "Epoch 76/100\n",
      "6160/6160 [==============================] - 10s 2ms/step - loss: 431.2198 - mae: 9.3534 - root_mean_squared_error: 20.7658\n",
      "Epoch 77/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 431.8368 - mae: 9.3646 - root_mean_squared_error: 20.7807\n",
      "Epoch 78/100\n",
      "6160/6160 [==============================] - 13s 2ms/step - loss: 429.8522 - mae: 9.3370 - root_mean_squared_error: 20.7329\n",
      "Epoch 79/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 435.0251 - mae: 9.4047 - root_mean_squared_error: 20.8573: 0s - loss: 434.4819 - mae: 9.401\n",
      "Epoch 80/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 425.6935 - mae: 9.2994 - root_mean_squared_error: 20.6323: 1s - loss: 428.7341 - mae: 9.3469 - root_m - ETA: 1s - loss: 428.5\n",
      "Epoch 81/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 430.3108 - mae: 9.3377 - root_mean_squared_error: 20.7439\n",
      "Epoch 82/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 425.1190 - mae: 9.2623 - root_mean_squared_error: 20.6184: 1s - loss: 428.4\n",
      "Epoch 83/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 425.4610 - mae: 9.2782 - root_mean_squared_error: 20.6267\n",
      "Epoch 84/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 421.2116 - mae: 9.2147 - root_mean_squared_error: 20.5234\n",
      "Epoch 85/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 422.7060 - mae: 9.2306 - root_mean_squared_error: 20.5598\n",
      "Epoch 86/100\n",
      "6160/6160 [==============================] - 9s 1ms/step - loss: 419.2908 - mae: 9.1904 - root_mean_squared_error: 20.4766\n",
      "Epoch 87/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 422.2617 - mae: 9.2274 - root_mean_squared_error: 20.5490\n",
      "Epoch 88/100\n",
      "6160/6160 [==============================] - 13s 2ms/step - loss: 416.7957 - mae: 9.1609 - root_mean_squared_error: 20.4156 0s - loss: 416.4063 - mae: 9.1557 - root_mean_sq\n",
      "Epoch 89/100\n",
      "6160/6160 [==============================] - 13s 2ms/step - loss: 419.5353 - mae: 9.1916 - root_mean_squared_error: 20.4826\n",
      "Epoch 90/100\n",
      "6160/6160 [==============================] - 17s 3ms/step - loss: 417.4928 - mae: 9.1714 - root_mean_squared_error: 20.4326\n",
      "Epoch 91/100\n",
      "6160/6160 [==============================] - 18s 3ms/step - loss: 416.5030 - mae: 9.1632 - root_mean_squared_error: 20.4084 1s - loss: 417.5564 - mae: 9.1755 - ro\n",
      "Epoch 92/100\n",
      "6160/6160 [==============================] - 15s 2ms/step - loss: 423.8202 - mae: 9.2543 - root_mean_squared_error: 20.5869\n",
      "Epoch 93/100\n",
      "6160/6160 [==============================] - 17s 3ms/step - loss: 413.8411 - mae: 9.1095 - root_mean_squared_error: 20.3431\n",
      "Epoch 94/100\n",
      "6160/6160 [==============================] - 12s 2ms/step - loss: 418.8125 - mae: 9.1797 - root_mean_squared_error: 20.4649\n",
      "Epoch 95/100\n",
      "6160/6160 [==============================] - 13s 2ms/step - loss: 413.8310 - mae: 9.1298 - root_mean_squared_error: 20.3428\n",
      "Epoch 96/100\n",
      "6160/6160 [==============================] - 14s 2ms/step - loss: 410.1364 - mae: 9.0920 - root_mean_squared_error: 20.2518\n",
      "Epoch 97/100\n",
      "6160/6160 [==============================] - 17s 3ms/step - loss: 410.5172 - mae: 9.0912 - root_mean_squared_error: 20.2612 1s - l\n",
      "Epoch 98/100\n",
      "6160/6160 [==============================] - 17s 3ms/step - loss: 411.3690 - mae: 9.0836 - root_mean_squared_error: 20.2822\n",
      "Epoch 99/100\n",
      "6160/6160 [==============================] - 13s 2ms/step - loss: 406.0085 - mae: 9.0240 - root_mean_squared_error: 20.1497 1s - loss: 405.5868 - ma\n",
      "Epoch 100/100\n",
      "6160/6160 [==============================] - 11s 2ms/step - loss: 415.2852 - mae: 9.1309 - root_mean_squared_error: 20.3785\n"
     ]
    }
   ],
   "source": [
    "#Compile and train\n",
    "model_dni.compile(optimizer=Adam(lr=0.0001), \n",
    "              loss='mse',\n",
    "              metrics=['mae',keras.metrics.RootMeanSquaredError()])\n",
    "\n",
    "history_dni = model_dni.fit(input_arry, dni_arr, epochs=100, batch_size=64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['mae'])\n",
    "plt.plot(history.history['root_mean_squared_error'])\n",
    "plt.plot(history_dni.history['mae'])\n",
    "plt.plot(history_dni.history['root_mean_squared_error'])\n",
    "plt.title('MAE and RMSE per epoch')\n",
    "plt.ylabel('Error')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['MAE', 'RMSE','MAE-DNI', 'RMSE-DNI' ], loc='middle')\n",
    "plt.savefig('error-v-epoch.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## IV. Model Validation - Outer Banks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Day</th>\n",
       "      <th>Hour</th>\n",
       "      <th>Minute</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Clearsky DHI</th>\n",
       "      <th>Clearsky DNI</th>\n",
       "      <th>Clearsky GHI</th>\n",
       "      <th>Cloud Type</th>\n",
       "      <th>...</th>\n",
       "      <th>GHI</th>\n",
       "      <th>Relative Humidity</th>\n",
       "      <th>Solar Zenith Angle</th>\n",
       "      <th>Surface Albedo</th>\n",
       "      <th>Pressure</th>\n",
       "      <th>Precipitable Water</th>\n",
       "      <th>Wind Direction</th>\n",
       "      <th>Wind Speed</th>\n",
       "      <th>timestamps</th>\n",
       "      <th>Cloud Cover</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>30</td>\n",
       "      <td>1.8</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>85.84</td>\n",
       "      <td>119.89</td>\n",
       "      <td>0.03</td>\n",
       "      <td>1026</td>\n",
       "      <td>1.2</td>\n",
       "      <td>350</td>\n",
       "      <td>8.6</td>\n",
       "      <td>1514782800</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>1.6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>87.76</td>\n",
       "      <td>132.05</td>\n",
       "      <td>0.03</td>\n",
       "      <td>1026</td>\n",
       "      <td>1.2</td>\n",
       "      <td>348</td>\n",
       "      <td>8.7</td>\n",
       "      <td>1514786400</td>\n",
       "      <td>0.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>30</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>88.35</td>\n",
       "      <td>144.19</td>\n",
       "      <td>0.03</td>\n",
       "      <td>1026</td>\n",
       "      <td>1.1</td>\n",
       "      <td>346</td>\n",
       "      <td>8.7</td>\n",
       "      <td>1514790000</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>30</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>89.16</td>\n",
       "      <td>155.85</td>\n",
       "      <td>0.03</td>\n",
       "      <td>1026</td>\n",
       "      <td>1.1</td>\n",
       "      <td>346</td>\n",
       "      <td>8.7</td>\n",
       "      <td>1514793600</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>30</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>89.52</td>\n",
       "      <td>165.32</td>\n",
       "      <td>0.03</td>\n",
       "      <td>1026</td>\n",
       "      <td>1.1</td>\n",
       "      <td>347</td>\n",
       "      <td>8.9</td>\n",
       "      <td>1514797200</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Year  Month  Day  Hour  Minute  Temperature  Clearsky DHI  Clearsky DNI  \\\n",
       "0  2018      1    1     0      30          1.8             0             0   \n",
       "1  2018      1    1     1      30          1.6             0             0   \n",
       "2  2018      1    1     2      30          1.5             0             0   \n",
       "3  2018      1    1     3      30          1.4             0             0   \n",
       "4  2018      1    1     4      30          1.4             0             0   \n",
       "\n",
       "   Clearsky GHI  Cloud Type  ...  GHI  Relative Humidity  Solar Zenith Angle  \\\n",
       "0             0           4  ...    0              85.84              119.89   \n",
       "1             0           4  ...    0              87.76              132.05   \n",
       "2             0           4  ...    0              88.35              144.19   \n",
       "3             0           4  ...    0              89.16              155.85   \n",
       "4             0           4  ...    0              89.52              165.32   \n",
       "\n",
       "   Surface Albedo  Pressure  Precipitable Water  Wind Direction  Wind Speed  \\\n",
       "0            0.03      1026                 1.2             350         8.6   \n",
       "1            0.03      1026                 1.2             348         8.7   \n",
       "2            0.03      1026                 1.1             346         8.7   \n",
       "3            0.03      1026                 1.1             346         8.7   \n",
       "4            0.03      1026                 1.1             347         8.9   \n",
       "\n",
       "   timestamps  Cloud Cover  \n",
       "0  1514782800         1.00  \n",
       "1  1514786400         0.94  \n",
       "2  1514790000         1.00  \n",
       "3  1514793600         1.00  \n",
       "4  1514797200         1.00  \n",
       "\n",
       "[5 rows x 24 columns]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "validation_df = pd.read_csv(os.path.join(_Testing_Data_Dir, 'outer_banks', 'r_banks.csv'))\n",
    "validation_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "input_size = 6\n",
    "val_input_arry = np.array([], dtype=np.float64).reshape(0,input_size)\n",
    "val_output_arry = np.array([], dtype=np.float64)\n",
    "val_dni_arr = np.array([], dtype=np.float64)\n",
    "val_input_arry, val_output_arry, val_dni_arr = combine_training_arrays(validation_df, val_input_arry, val_output_arry, val_dni_arr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "#keras model.evaluate reveals a mae of 6.9433 w/m2 in the outer banks test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "274/274 [==============================] - 0s 783us/step - loss: 269.0166 - mean_absolute_error: 7.1228 - root_mean_squared_error: 16.4017\n"
     ]
    }
   ],
   "source": [
    "score = model.evaluate(val_input_arry, val_output_arry, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "preds = model.predict(val_input_arry)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Mape too high because of 0 values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Compare random induvidual resutls!!!!!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "index is  4521\n",
      "Actual  0.0   Predicted  [0.]\n"
     ]
    }
   ],
   "source": [
    "i = randint(0,8760)\n",
    "print(\"index is \",i)\n",
    "print(\"Actual \",val_output_arry[i],\" \",\"Predicted \", preds[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "274/274 [==============================] - 0s 770us/step - loss: 650.3802 - mae: 12.5543 - root_mean_squared_error: 25.5026\n"
     ]
    }
   ],
   "source": [
    "# Now lets see how it fares for DNI\n",
    "score_dni = model_dni.evaluate(val_input_arry, val_dni_arr, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "preds_dni = model_dni.predict(val_input_arry)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "index is  3520\n",
      "Actual  359.0   Predicted  [373.17288]\n"
     ]
    }
   ],
   "source": [
    "i = randint(0,8760)\n",
    "print(\"index is \",i)\n",
    "print(\"Actual \",val_dni_arr[i],\" \",\"Predicted \", preds_dni[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot([float(i) for i in preds])\n",
    "plt.plot([float(i) for i in val_output_arry])\n",
    "\n",
    "plt.title('Hourly Fitted vs actual')\n",
    "plt.ylabel('Prediction')\n",
    "plt.xlabel('Hour')\n",
    "plt.legend(['Estimated-DHI','Actual-DHI'], loc='middle')\n",
    "plt.savefig('OBX-actual-pred.png')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/site-packages/matplotlib-2.2.4-py3.7-macosx-10.15-x86_64.egg/matplotlib/legend.py:640: UserWarning: Unrecognized location \"middle\". Falling back on \"best\"; valid locations are\n",
      "\tbest\n",
      "\tupper right\n",
      "\tupper left\n",
      "\tlower left\n",
      "\tlower right\n",
      "\tright\n",
      "\tcenter left\n",
      "\tcenter right\n",
      "\tlower center\n",
      "\tupper center\n",
      "\tcenter\n",
      "\n",
      "  % (loc, '\\n\\t'.join(self.codes)))\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot([float(i) for i in preds_dni])\n",
    "plt.plot([float(i) for i in val_dni_arr])\n",
    "\n",
    "plt.title('Hourly Fitted vs actual')\n",
    "plt.ylabel('Prediction')\n",
    "plt.xlabel('Hour')\n",
    "plt.legend(['Estimated-DNI', 'Actual-DNI'], loc='middle')\n",
    "plt.savefig('OBX-actual-pred-dni.png')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# model.save('Neural_Net_National')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /usr/local/lib/python3.7/site-packages/tensorflow/python/training/tracking/tracking.py:111: Model.state_updates (from tensorflow.python.keras.engine.training) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "This property should not be used in TensorFlow 2.0, as updates are applied automatically.\n",
      "WARNING:tensorflow:From /usr/local/lib/python3.7/site-packages/tensorflow/python/training/tracking/tracking.py:111: Layer.updates (from tensorflow.python.keras.engine.base_layer) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "This property should not be used in TensorFlow 2.0, as updates are applied automatically.\n",
      "INFO:tensorflow:Assets written to: Neural_Nets/Neural_Net_National_dni/assets\n"
     ]
    }
   ],
   "source": [
    "model_dni.save('Neural_Nets/Neural_Net_National_dni')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## V. Model Testing - Other Locations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lets take the global model and test in new locations. Lets test on the following locations for the year 2018.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1. Outer Banks (35.5585, -75.4665)\n",
    "2. Miami Florida (25.7617, -80.1918)\n",
    "3. Boulder CO (40.035,-105.54)\n",
    "4. San Diego CA (32.7157, -117.1611)\n",
    "5. Lincoln Nebraska (40.848,-96.56)\n",
    "6. Quinault Washington (47.513,-123.81)\n",
    "7. Old Town Maine (44.928,-68.70)\n",
    "8. Murphey Idaho (43.204,-116.75)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Test General model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "National_model = keras.models.load_model('Neural_Nets/Neural_Net_National')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_dni = keras.models.load_model('Neural_Nets/Neural_Net_National_dni')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense (Dense)                (None, 300)               2100      \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 200)               60200     \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 100)               20100     \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 50)                5050      \n",
      "_________________________________________________________________\n",
      "dense_4 (Dense)              (None, 25)                1275      \n",
      "_________________________________________________________________\n",
      "dense_5 (Dense)              (None, 1)                 26        \n",
      "=================================================================\n",
      "Total params: 88,751\n",
      "Trainable params: 88,751\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "National_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Murphey_ID/psm_ID_Murphey2018.csv\n",
      "274/274 [==============================] - 0s 798us/step - loss: 256.5927 - mean_absolute_error: 6.7838 - root_mean_squared_error: 16.0185\n",
      "274/274 [==============================] - 0s 773us/step - loss: 939.1105 - mean_absolute_error: 15.0760 - root_mean_squared_error: 30.6449\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/OldTown_ME/psm_ME_OldTown2018.csv\n",
      "274/274 [==============================] - 0s 1ms/step - loss: 229.4214 - mean_absolute_error: 6.8104 - root_mean_squared_error: 15.1467\n",
      "274/274 [==============================] - 0s 809us/step - loss: 557.3621 - mean_absolute_error: 10.7451 - root_mean_squared_error: 23.6085\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Quinault_WA/psm_WA_Quinault2018.csv\n",
      "274/274 [==============================] - 0s 787us/step - loss: 104.5538 - mean_absolute_error: 4.2643 - root_mean_squared_error: 10.2252\n",
      "274/274 [==============================] - 0s 876us/step - loss: 256.6161 - mean_absolute_error: 7.1516 - root_mean_squared_error: 16.0192\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Lincoln_NE/psm_NE_Lincoln2018.csv\n",
      "274/274 [==============================] - 0s 938us/step - loss: 161.9369 - mean_absolute_error: 5.2631 - root_mean_squared_error: 12.7254\n",
      "274/274 [==============================] - 0s 769us/step - loss: 461.1183 - mean_absolute_error: 9.5104 - root_mean_squared_error: 21.4737\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/SanDiego_CA/psm_CA_SanDiego2018.csv\n",
      "274/274 [==============================] - 0s 828us/step - loss: 289.1909 - mean_absolute_error: 7.7851 - root_mean_squared_error: 17.0056\n",
      "274/274 [==============================] - 0s 846us/step - loss: 624.3943 - mean_absolute_error: 11.8969 - root_mean_squared_error: 24.9879\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Boulder_CO/psm_CO_Boulder2018.csv\n",
      "274/274 [==============================] - 0s 907us/step - loss: 1862.3188 - mean_absolute_error: 16.5438 - root_mean_squared_error: 43.1546\n",
      "274/274 [==============================] - 0s 968us/step - loss: 2644.2048 - mean_absolute_error: 28.0222 - root_mean_squared_error: 51.4218\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Miami_FL/psm_FL_Miami2018.csv\n",
      "274/274 [==============================] - 0s 883us/step - loss: 675.2378 - mean_absolute_error: 10.5119 - root_mean_squared_error: 25.9853\n",
      "274/274 [==============================] - 0s 847us/step - loss: 1217.5833 - mean_absolute_error: 16.9184 - root_mean_squared_error: 34.8939\n"
     ]
    }
   ],
   "source": [
    "test_files = ['Murphey_ID', 'OldTown_ME', 'Quinault_WA', 'Lincoln_NE', 'SanDiego_CA', 'Boulder_CO', 'Miami_FL', 'outer_banks']\n",
    "scores = {}\n",
    "scores_dni = {}\n",
    "pred_actuals = {}\n",
    "pred_actuals_dni = {}\n",
    "for directory in test_files:\n",
    "    for file in glob(_Testing_Data_Dir+'/'+directory+'/psm_*2018.csv'):\n",
    "        test_df = pd.read_csv(str(file))\n",
    "        input_size = 6\n",
    "        test_input_arry = np.array([], dtype=np.float64).reshape(0,input_size)\n",
    "        test_output_arry = np.array([], dtype=np.float64)\n",
    "        test_dni_arry = np.array([], dtype=np.float64)\n",
    "        test_input_arry, test_output_arry, test_dni_arry = combine_training_arrays(test_df, test_input_arry, test_output_arry, test_dni_arry)\n",
    "        print(file)\n",
    "        score = National_model.evaluate(test_input_arry, test_output_arry, verbose=1)\n",
    "        scores[directory] = score\n",
    "        score_dni = model_dni.evaluate(test_input_arry, test_dni_arry, verbose=1)\n",
    "        scores_dni[directory] = score_dni\n",
    "        preds = National_model.predict(test_input_arry)\n",
    "        pred_actuals[directory] = {'preds':list(preds), 'actuals': list(test_output_arry)}\n",
    "        preds_dni = model_dni.predict(test_input_arry)\n",
    "        pred_actuals_dni[directory] = {'preds':list(preds_dni), 'actuals': list(test_dni_arry)}\n",
    "        \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_items([('Murphey_ID', [256.5926818847656, 6.783843994140625, 16.018510818481445]), ('OldTown_ME', [229.42138671875, 6.810364246368408, 15.146662712097168]), ('Quinault_WA', [104.5538101196289, 4.2642903327941895, 10.2251558303833]), ('Lincoln_NE', [161.93692016601562, 5.263095855712891, 12.725443840026855]), ('SanDiego_CA', [289.1909484863281, 7.785069942474365, 17.005615234375]), ('Boulder_CO', [1862.31884765625, 16.543825149536133, 43.15459060668945]), ('Miami_FL', [675.23779296875, 10.51191234588623, 25.98533821105957])])"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scores.items()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_items([('Murphey_ID', [939.1104736328125, 15.076038360595703, 30.644908905029297]), ('OldTown_ME', [557.3621215820312, 10.745118141174316, 23.608518600463867]), ('Quinault_WA', [256.6160888671875, 7.151605129241943, 16.019241333007812]), ('Lincoln_NE', [461.11834716796875, 9.51038646697998, 21.47366714477539]), ('SanDiego_CA', [624.3943481445312, 11.896857261657715, 24.987884521484375]), ('Boulder_CO', [2644.204833984375, 28.022197723388672, 51.42183303833008]), ('Miami_FL', [1217.583251953125, 16.91837501525879, 34.89388656616211])])"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scores_dni.items()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/site-packages/matplotlib-2.2.4-py3.7-macosx-10.15-x86_64.egg/matplotlib/legend.py:640: UserWarning: Unrecognized location \"middle\". Falling back on \"best\"; valid locations are\n",
      "\tbest\n",
      "\tupper right\n",
      "\tupper left\n",
      "\tlower left\n",
      "\tlower right\n",
      "\tright\n",
      "\tcenter left\n",
      "\tcenter right\n",
      "\tlower center\n",
      "\tupper center\n",
      "\tcenter\n",
      "\n",
      "  % (loc, '\\n\\t'.join(self.codes)))\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Graphs\n",
    "legend_names = []\n",
    "for directory in pred_actuals.keys():\n",
    "    preds = ([float(i) for i in pred_actuals[directory]['preds']])\n",
    "    actuals =  ([float(i) for i in pred_actuals[directory]['actuals']])\n",
    "    plt.plot(preds)\n",
    "    legend_names.append(str(directory) +'-pred')\n",
    "    plt.plot(actuals)\n",
    "    legend_names.append(str(directory) +'actual')\n",
    "\n",
    "plt.title('Hourly Fitted vs actual')\n",
    "plt.ylabel('DHI Prediction')\n",
    "plt.xlabel('Hour')\n",
    "plt.legend(legend_names, loc='middle')\n",
    "plt.savefig('DHI-vs-Hour.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Graphs\n",
    "legend_names = []\n",
    "for directory in pred_actuals_dni.keys():\n",
    "    preds = ([float(i) for i in pred_actuals[directory]['preds']])\n",
    "    actuals =  ([float(i) for i in pred_actuals[directory]['actuals']])\n",
    "    plt.plot(preds)\n",
    "    legend_names.append(str(directory) +'-pred')\n",
    "    plt.plot(actuals)\n",
    "    legend_names.append(str(directory) +'actual')\n",
    "\n",
    "plt.title('Hourly Fitted vs actual')\n",
    "plt.ylabel('DNI Prediction')\n",
    "plt.xlabel('Hour')\n",
    "plt.legend(legend_names, loc='middle')\n",
    "plt.savefig('DNI-vs-Hour.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Only Miami and Boulder_CO seem to be outliers"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## V.I Model Testing, Boulder Specific"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lets see if we can train a model specific for Boulder, CO and see if we can get the mae down.\n",
    "Let's fetch more data from NRSDB and combine it with our readings from Darksky. Lets get data from 2008-2018."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[]\n",
      "[]\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Boulder_CO/psm_CO_Boulder2018.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Boulder_CO/psm_CO_Boulder2008.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Boulder_CO/psm_CO_Boulder2009.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Boulder_CO/psm_CO_Boulder2010.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Boulder_CO/psm_CO_Boulder2011.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Boulder_CO/psm_CO_Boulder2013.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Boulder_CO/psm_CO_Boulder2012.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Boulder_CO/psm_CO_Boulder2016.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Boulder_CO/psm_CO_Boulder2017.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Boulder_CO/psm_CO_Boulder2015.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Boulder_CO/psm_CO_Boulder2014.csv\n"
     ]
    }
   ],
   "source": [
    "input_size = 6\n",
    "input_arry = np.array([], dtype=np.float64).reshape(0,input_size)\n",
    "output_arry = np.array([], dtype=np.float64)\n",
    "dni_arry = np.array([], dtype=np.float64)\n",
    "\n",
    "print(input_arry)\n",
    "print(output_arry)\n",
    "from glob import glob\n",
    "for path in glob(_Testing_Data_Dir+'/Boulder_CO'):\n",
    "    for file in glob(path+'/psm_'+'*'):\n",
    "        print(file)\n",
    "        df = pd.read_csv(file)\n",
    "        input_arry, output_arry, dni_arry = combine_training_arrays(df, input_arry, output_arry ,dni_arry)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "keras.backend.clear_session()\n",
    "tf.keras.backend.clear_session()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<function tensorflow.python.framework.ops.reset_default_graph()>"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "keras.backend.clear_session()\n",
    "tf.keras.backend.clear_session()\n",
    "tf.compat.v1.reset_default_graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_Boulder = keras.Sequential()\n",
    "model_Boulder.add(keras.layers.Dense(300, activation=\"relu\"))\n",
    "model_Boulder.add(layers.Dense(200, activation=\"relu\"))\n",
    "model_Boulder.add(layers.Dense(100, activation=\"relu\"))\n",
    "model_Boulder.add(layers.Dense(50, activation=\"relu\"))\n",
    "model_Boulder.add(layers.Dense(25, activation=\"relu\"))\n",
    "model_Boulder.add(layers.Dense(1, activation='relu'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 2877.0916 - mae: 25.5355 - root_mean_squared_error: 53.6385\n",
      "Epoch 2/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 702.4658 - mae: 11.3388 - root_mean_squared_error: 26.5041\n",
      "Epoch 3/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 609.1577 - mae: 10.3836 - root_mean_squared_error: 24.6811A: 0s - loss: 609.8953 - mae: 10.3996 - root_mean_squared_error: 24.696\n",
      "Epoch 4/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 601.2148 - mae: 10.1433 - root_mean_squared_error: 24.5197\n",
      "Epoch 5/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 579.6346 - mae: 9.7413 - root_mean_squared_error: 24.0756\n",
      "Epoch 6/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 546.2700 - mae: 9.3879 - root_mean_squared_error: 23.3724\n",
      "Epoch 7/100\n",
      "3012/3012 [==============================] - 6s 2ms/step - loss: 543.9135 - mae: 9.3714 - root_mean_squared_error: 23.3220\n",
      "Epoch 8/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 536.4670 - mae: 9.2550 - root_mean_squared_error: 23.1618\n",
      "Epoch 9/100\n",
      "3012/3012 [==============================] - 6s 2ms/step - loss: 546.4626 - mae: 9.3199 - root_mean_squared_error: 23.3765\n",
      "Epoch 10/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 519.8393 - mae: 9.0285 - root_mean_squared_error: 22.8000: 0s - loss: 521.4241 - mae: 9.0491 - root_mean_squared_error:\n",
      "Epoch 11/100\n",
      "3012/3012 [==============================] - 6s 2ms/step - loss: 516.1115 - mae: 8.9876 - root_mean_squared_error: 22.7181\n",
      "Epoch 12/100\n",
      "3012/3012 [==============================] - 9s 3ms/step - loss: 505.0688 - mae: 8.8529 - root_mean_squared_error: 22.4737\n",
      "Epoch 13/100\n",
      "3012/3012 [==============================] - 8s 3ms/step - loss: 498.1188 - mae: 8.7605 - root_mean_squared_error: 22.3186\n",
      "Epoch 14/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 496.4182 - mae: 8.7621 - root_mean_squared_error: 22.2804\n",
      "Epoch 15/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 500.7352 - mae: 8.7955 - root_mean_squared_error: 22.3771\n",
      "Epoch 16/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 499.4591 - mae: 8.7359 - root_mean_squared_error: 22.3486\n",
      "Epoch 17/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 486.5204 - mae: 8.6440 - root_mean_squared_error: 22.0572\n",
      "Epoch 18/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 491.2159 - mae: 8.6442 - root_mean_squared_error: 22.1634\n",
      "Epoch 19/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 483.5406 - mae: 8.6728 - root_mean_squared_error: 21.9896\n",
      "Epoch 20/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 485.7330 - mae: 8.5713 - root_mean_squared_error: 22.0394\n",
      "Epoch 21/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 475.2501 - mae: 8.4550 - root_mean_squared_error: 21.8002\n",
      "Epoch 22/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 473.5088 - mae: 8.4377 - root_mean_squared_error: 21.7603: 3s - loss: 484\n",
      "Epoch 23/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 473.8627 - mae: 8.4650 - root_mean_squared_error: 21.7684\n",
      "Epoch 24/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 471.5439 - mae: 8.3815 - root_mean_squared_error: 21.7151\n",
      "Epoch 25/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 468.9697 - mae: 8.3188 - root_mean_squared_error: 21.6557\n",
      "Epoch 26/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 468.9990 - mae: 8.4258 - root_mean_squared_error: 21.6564: 3s - loss: 475.4723 - ma\n",
      "Epoch 27/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 467.1933 - mae: 8.3236 - root_mean_squared_error: 21.6147: 0s - loss: 465.0727 - mae: 8.2891 - root_mean_squared_\n",
      "Epoch 28/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 467.1798 - mae: 8.3114 - root_mean_squared_error: 21.6143\n",
      "Epoch 29/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 463.9851 - mae: 8.3178 - root_mean_squared_error: 21.5403\n",
      "Epoch 30/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 471.3863 - mae: 8.3525 - root_mean_squared_error: 21.7114: 1s - los\n",
      "Epoch 31/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 464.0964 - mae: 8.3537 - root_mean_squared_error: 21.5429\n",
      "Epoch 32/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 458.7533 - mae: 8.2715 - root_mean_squared_error: 21.4185\n",
      "Epoch 33/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 454.6998 - mae: 8.1552 - root_mean_squared_error: 21.3237\n",
      "Epoch 34/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 456.1515 - mae: 8.2247 - root_mean_squared_error: 21.3577\n",
      "Epoch 35/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 451.9662 - mae: 8.1337 - root_mean_squared_error: 21.2595\n",
      "Epoch 36/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 456.8525 - mae: 8.1742 - root_mean_squared_error: 21.3741\n",
      "Epoch 37/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 457.0673 - mae: 8.2077 - root_mean_squared_error: 21.3791\n",
      "Epoch 38/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 452.8080 - mae: 8.1821 - root_mean_squared_error: 21.2793: 0s - loss: 452.0220 - mae: 8.1718 - root_mean_squared_error: 21.\n",
      "Epoch 39/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 450.1650 - mae: 8.1370 - root_mean_squared_error: 21.2171\n",
      "Epoch 40/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 448.3389 - mae: 8.0499 - root_mean_squared_error: 21.1740\n",
      "Epoch 41/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 450.3363 - mae: 8.1058 - root_mean_squared_error: 21.2211\n",
      "Epoch 42/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 449.1070 - mae: 8.0745 - root_mean_squared_error: 21.1921\n",
      "Epoch 43/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 445.5112 - mae: 8.0402 - root_mean_squared_error: 21.1071\n",
      "Epoch 44/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 443.2002 - mae: 7.9831 - root_mean_squared_error: 21.0523\n",
      "Epoch 45/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 446.1323 - mae: 8.0329 - root_mean_squared_error: 21.1218\n",
      "Epoch 46/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 448.5466 - mae: 8.0815 - root_mean_squared_error: 21.1789\n",
      "Epoch 47/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 443.1236 - mae: 7.9949 - root_mean_squared_error: 21.0505\n",
      "Epoch 48/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 442.8403 - mae: 8.0040 - root_mean_squared_error: 21.0438\n",
      "Epoch 49/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 438.0711 - mae: 7.9794 - root_mean_squared_error: 20.9301\n",
      "Epoch 50/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 440.8987 - mae: 8.0173 - root_mean_squared_error: 20.9976\n",
      "Epoch 51/100\n",
      "3012/3012 [==============================] - 6s 2ms/step - loss: 438.4687 - mae: 7.9183 - root_mean_squared_error: 20.9396\n",
      "Epoch 52/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 437.4593 - mae: 7.9422 - root_mean_squared_error: 20.9155\n",
      "Epoch 53/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 434.9712 - mae: 7.8904 - root_mean_squared_error: 20.8560\n",
      "Epoch 54/100\n",
      "3012/3012 [==============================] - 6s 2ms/step - loss: 434.3372 - mae: 7.8961 - root_mean_squared_error: 20.8408\n",
      "Epoch 55/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 434.1020 - mae: 7.8546 - root_mean_squared_error: 20.8351\n",
      "Epoch 56/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 433.6989 - mae: 7.8907 - root_mean_squared_error: 20.8254\n",
      "Epoch 57/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 431.3864 - mae: 7.8492 - root_mean_squared_error: 20.7698\n",
      "Epoch 58/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3012/3012 [==============================] - 7s 2ms/step - loss: 431.9071 - mae: 7.8272 - root_mean_squared_error: 20.7824\n",
      "Epoch 59/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 430.3178 - mae: 7.8230 - root_mean_squared_error: 20.7441\n",
      "Epoch 60/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 424.1770 - mae: 7.7587 - root_mean_squared_error: 20.5956\n",
      "Epoch 61/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 424.7537 - mae: 7.7656 - root_mean_squared_error: 20.6096\n",
      "Epoch 62/100\n",
      "3012/3012 [==============================] - 7s 2ms/step - loss: 425.8828 - mae: 7.7637 - root_mean_squared_error: 20.6369\n",
      "Epoch 63/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 421.8170 - mae: 7.7149 - root_mean_squared_error: 20.5382\n",
      "Epoch 64/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 421.1922 - mae: 7.6932 - root_mean_squared_error: 20.5230\n",
      "Epoch 65/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 423.4917 - mae: 7.7649 - root_mean_squared_error: 20.5789\n",
      "Epoch 66/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 430.0306 - mae: 7.7958 - root_mean_squared_error: 20.7372\n",
      "Epoch 67/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 424.8602 - mae: 7.6974 - root_mean_squared_error: 20.6121\n",
      "Epoch 68/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 419.5780 - mae: 7.6533 - root_mean_squared_error: 20.4836\n",
      "Epoch 69/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 423.6875 - mae: 7.7178 - root_mean_squared_error: 20.5837\n",
      "Epoch 70/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 411.9170 - mae: 7.5797 - root_mean_squared_error: 20.2957\n",
      "Epoch 71/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 416.9352 - mae: 7.6000 - root_mean_squared_error: 20.4190\n",
      "Epoch 72/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 414.2979 - mae: 7.5533 - root_mean_squared_error: 20.3543: 1s - loss: 398.7578 \n",
      "Epoch 73/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 416.1005 - mae: 7.6079 - root_mean_squared_error: 20.3985\n",
      "Epoch 74/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 411.9357 - mae: 7.5369 - root_mean_squared_error: 20.2962\n",
      "Epoch 75/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 413.6200 - mae: 7.6275 - root_mean_squared_error: 20.3377\n",
      "Epoch 76/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 412.7784 - mae: 7.5907 - root_mean_squared_error: 20.3169\n",
      "Epoch 77/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 412.9873 - mae: 7.5424 - root_mean_squared_error: 20.3221: 1s - los\n",
      "Epoch 78/100\n",
      "3012/3012 [==============================] - 5s 1ms/step - loss: 409.9065 - mae: 7.5176 - root_mean_squared_error: 20.2461\n",
      "Epoch 79/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 409.4922 - mae: 7.5075 - root_mean_squared_error: 20.2359\n",
      "Epoch 80/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 412.0279 - mae: 7.5348 - root_mean_squared_error: 20.2985\n",
      "Epoch 81/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 409.0555 - mae: 7.5307 - root_mean_squared_error: 20.2251\n",
      "Epoch 82/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 405.4362 - mae: 7.4476 - root_mean_squared_error: 20.1354: 0s - loss: 405.7073 - mae: 7.4478 - root_mean_squared_error: 20.14\n",
      "Epoch 83/100\n",
      "3012/3012 [==============================] - 5s 1ms/step - loss: 406.5412 - mae: 7.4895 - root_mean_squared_error: 20.1629\n",
      "Epoch 84/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 405.4978 - mae: 7.4757 - root_mean_squared_error: 20.1370\n",
      "Epoch 85/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 410.7215 - mae: 7.4867 - root_mean_squared_error: 20.2663: 0s - loss: 412.7445 - mae: 7\n",
      "Epoch 86/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 395.9564 - mae: 7.2883 - root_mean_squared_error: 19.8987\n",
      "Epoch 87/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 403.3719 - mae: 7.4353 - root_mean_squared_error: 20.0841: 4s - loss: 394.3188 - \n",
      "Epoch 88/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 399.6424 - mae: 7.3556 - root_mean_squared_error: 19.9911\n",
      "Epoch 89/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 395.3971 - mae: 7.3239 - root_mean_squared_error: 19.8846\n",
      "Epoch 90/100\n",
      "3012/3012 [==============================] - 6s 2ms/step - loss: 400.9569 - mae: 7.3347 - root_mean_squared_error: 20.0239: 0s - loss: 403.5268 - mae: 7.3475 - root_mea\n",
      "Epoch 91/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 391.0229 - mae: 7.2593 - root_mean_squared_error: 19.7743\n",
      "Epoch 92/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 396.0735 - mae: 7.3170 - root_mean_squared_error: 19.9016\n",
      "Epoch 93/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 396.8293 - mae: 7.3026 - root_mean_squared_error: 19.9206: 0s - loss: 397.7376 - mae: 7.3170 - root\n",
      "Epoch 94/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 394.7733 - mae: 7.2771 - root_mean_squared_error: 19.8689\n",
      "Epoch 95/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 392.5875 - mae: 7.2084 - root_mean_squared_error: 19.8138\n",
      "Epoch 96/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 388.7122 - mae: 7.1757 - root_mean_squared_error: 19.7158\n",
      "Epoch 97/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 383.7001 - mae: 7.1293 - root_mean_squared_error: 19.5883\n",
      "Epoch 98/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 382.0702 - mae: 7.1194 - root_mean_squared_error: 19.5466: 0s - loss: 386.4940 - mae: 7.1539 - \n",
      "Epoch 99/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 384.7579 - mae: 7.1876 - root_mean_squared_error: 19.6152\n",
      "Epoch 100/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 382.7167 - mae: 7.0977 - root_mean_squared_error: 19.5631\n"
     ]
    }
   ],
   "source": [
    "#Compile and train\n",
    "model_Boulder.compile(optimizer=Adam(lr=0.0001), \n",
    "              loss='mse',\n",
    "              metrics=['mae',keras.metrics.RootMeanSquaredError()])\n",
    "\n",
    "history_Boulder = model_Boulder.fit(input_arry, output_arry, epochs=100, batch_size=32)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "#MAE falls to around 6.99"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history_Boulder.history['mae'])\n",
    "plt.plot(history_Boulder.history['root_mean_squared_error'])\n",
    "plt.title('mae and rmse per epoch')\n",
    "plt.ylabel('Error')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['mae', 'rmse'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "# And for DNI\n",
    "model_Boulder_dni = keras.Sequential()\n",
    "model_Boulder_dni.add(keras.layers.Dense(300, activation=\"relu\"))\n",
    "model_Boulder_dni.add(layers.Dense(200, activation=\"relu\"))\n",
    "model_Boulder_dni.add(layers.Dense(100, activation=\"relu\"))\n",
    "model_Boulder_dni.add(layers.Dense(50, activation=\"relu\"))\n",
    "model_Boulder_dni.add(layers.Dense(25, activation=\"relu\"))\n",
    "model_Boulder_dni.add(layers.Dense(1, activation='relu'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 25947.3262 - mae: 89.8100 - root_mean_squared_error: 161.0817\n",
      "Epoch 2/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 4233.6587 - mae: 33.2195 - root_mean_squared_error: 65.0666\n",
      "Epoch 3/100\n",
      "3012/3012 [==============================] - 6s 2ms/step - loss: 3369.9451 - mae: 29.2935 - root_mean_squared_error: 58.0512\n",
      "Epoch 4/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 3020.7935 - mae: 27.5213 - root_mean_squared_error: 54.9617\n",
      "Epoch 5/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 2698.4561 - mae: 25.8562 - root_mean_squared_error: 51.9467\n",
      "Epoch 6/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 2462.8945 - mae: 24.4448 - root_mean_squared_error: 49.6276\n",
      "Epoch 7/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 2358.8652 - mae: 23.6506 - root_mean_squared_error: 48.5681\n",
      "Epoch 8/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 2220.2063 - mae: 22.6926 - root_mean_squared_error: 47.1191\n",
      "Epoch 9/100\n",
      "3012/3012 [==============================] - 6s 2ms/step - loss: 2132.6843 - mae: 22.0771 - root_mean_squared_error: 46.1810\n",
      "Epoch 10/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 2110.5571 - mae: 21.8106 - root_mean_squared_error: 45.9408: 0s - loss: 2136.6016 - mae\n",
      "Epoch 11/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 2075.9771 - mae: 21.5989 - root_mean_squared_error: 45.5629: 0s - loss: 2066.4460 - mae: 21.5605 - root_mean_squared_er - ETA: 0s - loss: 2069.6772 - mae: 21.5594 - root_mean_squared_\n",
      "Epoch 12/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 2013.6254 - mae: 21.2056 - root_mean_squared_error: 44.8734: 1s - loss: 2051.3887 - mae: 21.4322 - root_mea - ETA: 1s - loss: 2040.7695 -\n",
      "Epoch 13/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 2028.7344 - mae: 21.1915 - root_mean_squared_error: 45.0415\n",
      "Epoch 14/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1990.7424 - mae: 20.9493 - root_mean_squared_error: 44.6177\n",
      "Epoch 15/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1933.1182 - mae: 20.5742 - root_mean_squared_error: 43.9672\n",
      "Epoch 16/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1921.8600 - mae: 20.4304 - root_mean_squared_error: 43.8390: 1s - los\n",
      "Epoch 17/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 1913.6267 - mae: 20.3658 - root_mean_squared_error: 43.7450\n",
      "Epoch 18/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1913.6514 - mae: 20.3745 - root_mean_squared_error: 43.7453\n",
      "Epoch 19/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 1906.8167 - mae: 20.2649 - root_mean_squared_error: 43.6671\n",
      "Epoch 20/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 1843.4546 - mae: 19.8166 - root_mean_squared_error: 42.9355\n",
      "Epoch 21/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1837.7496 - mae: 19.6685 - root_mean_squared_error: 42.8690\n",
      "Epoch 22/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1807.5874 - mae: 19.4034 - root_mean_squared_error: 42.5157\n",
      "Epoch 23/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1773.0984 - mae: 19.2113 - root_mean_squared_error: 42.1082\n",
      "Epoch 24/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1783.8252 - mae: 19.1931 - root_mean_squared_error: 42.2354\n",
      "Epoch 25/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1730.9371 - mae: 18.8747 - root_mean_squared_error: 41.6045: 0s - loss: 1734.7469 - mae: 18.8978 - root_mean_squared_\n",
      "Epoch 26/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1787.0518 - mae: 19.2076 - root_mean_squared_error: 42.2735\n",
      "Epoch 27/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 1740.4387 - mae: 18.9354 - root_mean_squared_error: 41.7186\n",
      "Epoch 28/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1723.0708 - mae: 18.8128 - root_mean_squared_error: 41.5099\n",
      "Epoch 29/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1701.1428 - mae: 18.6469 - root_mean_squared_error: 41.2449\n",
      "Epoch 30/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 1656.4185 - mae: 18.4038 - root_mean_squared_error: 40.6991\n",
      "Epoch 31/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 1673.9032 - mae: 18.5148 - root_mean_squared_error: 40.9134\n",
      "Epoch 32/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1672.4615 - mae: 18.5031 - root_mean_squared_error: 40.8957\n",
      "Epoch 33/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1628.4130 - mae: 18.2503 - root_mean_squared_error: 40.3536\n",
      "Epoch 34/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1629.5129 - mae: 18.2551 - root_mean_squared_error: 40.3672\n",
      "Epoch 35/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1624.5217 - mae: 18.1600 - root_mean_squared_error: 40.3054\n",
      "Epoch 36/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1625.2983 - mae: 18.1141 - root_mean_squared_error: 40.3150\n",
      "Epoch 37/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 1591.1639 - mae: 17.8954 - root_mean_squared_error: 39.8894: 3s - loss: 1627.0101 - mae: 17.9126 -  - ETA: 2s - loss: 1585.6514 - mae: 17.737\n",
      "Epoch 38/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 1587.1304 - mae: 17.9193 - root_mean_squared_error: 39.8388\n",
      "Epoch 39/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 1568.7975 - mae: 17.8355 - root_mean_squared_error: 39.6080\n",
      "Epoch 40/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1547.3710 - mae: 17.6072 - root_mean_squared_error: 39.3366: 1s - loss: 1552.9675 - mae: 17\n",
      "Epoch 41/100\n",
      "3012/3012 [==============================] - 6s 2ms/step - loss: 1546.1097 - mae: 17.6266 - root_mean_squared_error: 39.3206\n",
      "Epoch 42/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 1508.6191 - mae: 17.4177 - root_mean_squared_error: 38.8409: 0s - loss: 1505.2142 - mae: 17.3889 - root_mean_squa\n",
      "Epoch 43/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 1540.0850 - mae: 17.5898 - root_mean_squared_error: 39.2439\n",
      "Epoch 44/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1529.7539 - mae: 17.4803 - root_mean_squared_error: 39.1121\n",
      "Epoch 45/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1477.2386 - mae: 17.2076 - root_mean_squared_error: 38.4349\n",
      "Epoch 46/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1488.4642 - mae: 17.2450 - root_mean_squared_error: 38.5806: 0s - loss: 1492.3796 - mae: 17.2683 \n",
      "Epoch 47/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1468.5955 - mae: 17.1484 - root_mean_squared_error: 38.3223\n",
      "Epoch 48/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 1448.6167 - mae: 17.0106 - root_mean_squared_error: 38.0607\n",
      "Epoch 49/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 1431.0316 - mae: 16.8643 - root_mean_squared_error: 37.8290\n",
      "Epoch 50/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1435.4811 - mae: 16.9034 - root_mean_squared_error: 37.8877\n",
      "Epoch 51/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1429.1732 - mae: 16.8157 - root_mean_squared_error: 37.8044\n",
      "Epoch 52/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1434.0421 - mae: 16.8605 - root_mean_squared_error: 37.8687: 1s - loss: 1422.89\n",
      "Epoch 53/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1384.9900 - mae: 16.5131 - root_mean_squared_error: 37.2155\n",
      "Epoch 54/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1364.8818 - mae: 16.3748 - root_mean_squared_error: 36.9443: 2s - loss: 1321.3507 - mae - ETA: 1s\n",
      "Epoch 55/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1377.0914 - mae: 16.4977 - root_mean_squared_error: 37.1092: 2s - loss: 1396.9694 - mae: 16\n",
      "Epoch 56/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1342.8401 - mae: 16.1875 - root_mean_squared_error: 36.6448\n",
      "Epoch 57/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1346.8761 - mae: 16.2503 - root_mean_squared_error: 36.6998\n",
      "Epoch 58/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1347.7358 - mae: 16.2661 - root_mean_squared_error: 36.7115\n",
      "Epoch 59/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1338.3065 - mae: 16.1892 - root_mean_squared_error: 36.5829\n",
      "Epoch 60/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1326.2108 - mae: 16.0783 - root_mean_squared_error: 36.4172\n",
      "Epoch 61/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1281.8542 - mae: 15.7775 - root_mean_squared_error: 35.8030\n",
      "Epoch 62/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1284.9010 - mae: 15.8259 - root_mean_squared_error: 35.8455\n",
      "Epoch 63/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1235.3823 - mae: 15.4519 - root_mean_squared_error: 35.1480\n",
      "Epoch 64/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1260.1962 - mae: 15.6052 - root_mean_squared_error: 35.4992\n",
      "Epoch 65/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1237.9576 - mae: 15.3912 - root_mean_squared_error: 35.1846\n",
      "Epoch 66/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1213.6095 - mae: 15.2397 - root_mean_squared_error: 34.8369\n",
      "Epoch 67/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1214.4796 - mae: 15.2992 - root_mean_squared_error: 34.8494\n",
      "Epoch 68/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1194.5222 - mae: 15.1166 - root_mean_squared_error: 34.5619\n",
      "Epoch 69/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1173.4229 - mae: 14.9589 - root_mean_squared_error: 34.2553\n",
      "Epoch 70/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1167.8854 - mae: 14.8850 - root_mean_squared_error: 34.1743\n",
      "Epoch 71/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1154.4100 - mae: 14.7542 - root_mean_squared_error: 33.9766\n",
      "Epoch 72/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 1153.4061 - mae: 14.7412 - root_mean_squared_error: 33.9618\n",
      "Epoch 73/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 1156.8718 - mae: 14.8256 - root_mean_squared_error: 34.0128: 2s - loss: 1170.9377 - mae: 14.8739  - ETA: 1s -\n",
      "Epoch 74/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1142.0243 - mae: 14.6968 - root_mean_squared_error: 33.7938: 0s - loss: 1131.8583 - mae: 14.6186 - ro\n",
      "Epoch 75/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1118.7277 - mae: 14.4798 - root_mean_squared_error: 33.4474\n",
      "Epoch 76/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1125.0350 - mae: 14.4509 - root_mean_squared_error: 33.5415\n",
      "Epoch 77/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1096.1298 - mae: 14.2988 - root_mean_squared_error: 33.1078\n",
      "Epoch 78/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1113.1403 - mae: 14.3986 - root_mean_squared_error: 33.3638\n",
      "Epoch 79/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1105.5052 - mae: 14.3439 - root_mean_squared_error: 33.2491\n",
      "Epoch 80/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1072.9741 - mae: 14.0271 - root_mean_squared_error: 32.7563\n",
      "Epoch 81/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1098.4060 - mae: 14.2443 - root_mean_squared_error: 33.1422\n",
      "Epoch 82/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1069.0121 - mae: 14.0206 - root_mean_squared_error: 32.6958: 1s - loss: 106\n",
      "Epoch 83/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1061.9070 - mae: 13.9587 - root_mean_squared_error: 32.5869: 1s - loss: 107\n",
      "Epoch 84/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1039.7192 - mae: 13.8026 - root_mean_squared_error: 32.2447\n",
      "Epoch 85/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 1037.6296 - mae: 13.7604 - root_mean_squared_error: 32.2123: \n",
      "Epoch 86/100\n",
      "3012/3012 [==============================] - 5s 2ms/step - loss: 1049.4423 - mae: 13.8902 - root_mean_squared_error: 32.3951\n",
      "Epoch 87/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1017.1714 - mae: 13.5974 - root_mean_squared_error: 31.8931\n",
      "Epoch 88/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1016.3913 - mae: 13.5532 - root_mean_squared_error: 31.8809\n",
      "Epoch 89/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 995.9191 - mae: 13.3989 - root_mean_squared_error: 31.5582TA: 1s - loss: 997.0596 - mae: 13.4345 - root_mean_sq - ETA: 0s - loss: 989.6720 - mae: 13.3361 - root_mean_squared_error\n",
      "Epoch 90/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 1008.0516 - mae: 13.4947 - root_mean_squared_error: 31.7498\n",
      "Epoch 91/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 995.5656 - mae: 13.4091 - root_mean_squared_error: 31.5526A: 1s - loss: 972.8979 - mae: 13.\n",
      "Epoch 92/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 981.0503 - mae: 13.2688 - root_mean_squared_error: 31.3217\n",
      "Epoch 93/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 989.4658 - mae: 13.3640 - root_mean_squared_error: 31.4558\n",
      "Epoch 94/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 974.7287 - mae: 13.1975 - root_mean_squared_error: 31.2206\n",
      "Epoch 95/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 955.0452 - mae: 13.0517 - root_mean_squared_error: 30.9038\n",
      "Epoch 96/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 958.1776 - mae: 13.0037 - root_mean_squared_error: 30.9544\n",
      "Epoch 97/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 962.2823 - mae: 13.0990 - root_mean_squared_error: 31.0207A: 2s - loss\n",
      "Epoch 98/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 951.6418 - mae: 12.9891 - root_mean_squared_error: 30.8487\n",
      "Epoch 99/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 955.4750 - mae: 12.9962 - root_mean_squared_error: 30.9108\n",
      "Epoch 100/100\n",
      "3012/3012 [==============================] - 4s 1ms/step - loss: 930.5453 - mae: 12.7675 - root_mean_squared_error: 30.5048\n"
     ]
    }
   ],
   "source": [
    "#Compile and train\n",
    "model_Boulder_dni.compile(optimizer=Adam(lr=0.0001), \n",
    "              loss='mse',\n",
    "              metrics=['mae',keras.metrics.RootMeanSquaredError()])\n",
    "\n",
    "history_Boulder_dni = model_Boulder_dni.fit(input_arry, dni_arry, epochs=100, batch_size=32)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history_Boulder.history['mae'])\n",
    "plt.plot(history_Boulder.history['root_mean_squared_error'])\n",
    "plt.plot(history_Boulder_dni.history['mae'])\n",
    "plt.plot(history_Boulder_dni.history['root_mean_squared_error'])\n",
    "plt.title('MAE and RMSE per epoch')\n",
    "plt.ylabel('Error')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['MAE', 'RMSE','MAE-DNI', 'RMSE-DNI' ], loc='middle')\n",
    "plt.savefig('error-v-epoch-Boulder.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## V.II Model Testing,  Boulder Specific Model on Boulder 2018 data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Test on 2018 data only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_Boulder = keras.models.load_model('Neural_Nets/Neural_Net_Boulder_CO')\n",
    "model_Boulder_dni = keras.models.load_model('Neural_Nets/Neural_Net_Boulder_CO-dni')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "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>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Day</th>\n",
       "      <th>Hour</th>\n",
       "      <th>Minute</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Clearsky DHI</th>\n",
       "      <th>Clearsky DNI</th>\n",
       "      <th>Clearsky GHI</th>\n",
       "      <th>Cloud Type</th>\n",
       "      <th>...</th>\n",
       "      <th>Solar Zenith Angle</th>\n",
       "      <th>Surface Albedo</th>\n",
       "      <th>Pressure</th>\n",
       "      <th>Precipitable Water</th>\n",
       "      <th>Wind Direction</th>\n",
       "      <th>Wind Speed</th>\n",
       "      <th>Global Horizontal UV Irradiance (280-400nm)</th>\n",
       "      <th>Global Horizontal UV Irradiance (285-385nm)</th>\n",
       "      <th>timestamps</th>\n",
       "      <th>Cloud Cover</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>30</td>\n",
       "      <td>-14.3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>98.21</td>\n",
       "      <td>0.67</td>\n",
       "      <td>711</td>\n",
       "      <td>0.2</td>\n",
       "      <td>261</td>\n",
       "      <td>0.9</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1514782800</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>-15.6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>109.06</td>\n",
       "      <td>0.67</td>\n",
       "      <td>711</td>\n",
       "      <td>0.2</td>\n",
       "      <td>254</td>\n",
       "      <td>1.2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1514786400</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>30</td>\n",
       "      <td>-16.3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>120.35</td>\n",
       "      <td>0.67</td>\n",
       "      <td>711</td>\n",
       "      <td>0.1</td>\n",
       "      <td>252</td>\n",
       "      <td>1.1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1514790000</td>\n",
       "      <td>0.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>30</td>\n",
       "      <td>-16.7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>131.81</td>\n",
       "      <td>0.67</td>\n",
       "      <td>711</td>\n",
       "      <td>0.1</td>\n",
       "      <td>256</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1514793600</td>\n",
       "      <td>0.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>30</td>\n",
       "      <td>-17.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>143.11</td>\n",
       "      <td>0.67</td>\n",
       "      <td>711</td>\n",
       "      <td>0.1</td>\n",
       "      <td>261</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1514797200</td>\n",
       "      <td>0.64</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 26 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Year  Month  Day  Hour  Minute  Temperature  Clearsky DHI  Clearsky DNI  \\\n",
       "0  2018      1    1     0      30        -14.3             0             0   \n",
       "1  2018      1    1     1      30        -15.6             0             0   \n",
       "2  2018      1    1     2      30        -16.3             0             0   \n",
       "3  2018      1    1     3      30        -16.7             0             0   \n",
       "4  2018      1    1     4      30        -17.0             0             0   \n",
       "\n",
       "   Clearsky GHI  Cloud Type  ...  Solar Zenith Angle  Surface Albedo  \\\n",
       "0             0           0  ...               98.21            0.67   \n",
       "1             0           0  ...              109.06            0.67   \n",
       "2             0           0  ...              120.35            0.67   \n",
       "3             0           0  ...              131.81            0.67   \n",
       "4             0           0  ...              143.11            0.67   \n",
       "\n",
       "   Pressure  Precipitable Water  Wind Direction  Wind Speed  \\\n",
       "0       711                 0.2             261         0.9   \n",
       "1       711                 0.2             254         1.2   \n",
       "2       711                 0.1             252         1.1   \n",
       "3       711                 0.1             256         1.0   \n",
       "4       711                 0.1             261         1.0   \n",
       "\n",
       "   Global Horizontal UV Irradiance (280-400nm)  \\\n",
       "0                                          NaN   \n",
       "1                                          NaN   \n",
       "2                                          NaN   \n",
       "3                                          NaN   \n",
       "4                                          NaN   \n",
       "\n",
       "   Global Horizontal UV Irradiance (285-385nm)  timestamps  Cloud Cover  \n",
       "0                                          NaN  1514782800         0.00  \n",
       "1                                          NaN  1514786400         0.00  \n",
       "2                                          NaN  1514790000         0.64  \n",
       "3                                          NaN  1514793600         0.64  \n",
       "4                                          NaN  1514797200         0.64  \n",
       "\n",
       "[5 rows x 26 columns]"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df = pd.read_csv(os.path.join(_Testing_Data_Dir, 'Boulder_CO', 'psm_CO_Boulder2018.csv'))\n",
    "test_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "input_size = 6\n",
    "test_input_arry = np.array([], dtype=np.float64).reshape(0,input_size)\n",
    "test_output_arry = np.array([], dtype=np.float64)\n",
    "test_dni_arry = np.array([], dtype=np.float64)\n",
    "\n",
    "test_input_arry, test_output_arry, test_dni_arry = combine_training_arrays(test_df, test_input_arry, test_output_arry, test_dni_arry)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "274/274 [==============================] - 0s 932us/step - loss: 978.8859 - mean_absolute_error: 12.8149 - root_mean_squared_error: 31.2872\n"
     ]
    }
   ],
   "source": [
    "score = model_Boulder.evaluate(test_input_arry, test_output_arry, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "#12.8149 is better than 15"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "274/274 [==============================] - 0s 820us/step - loss: 1701.9760 - mae: 19.8515 - root_mean_squared_error: 41.2550\n"
     ]
    }
   ],
   "source": [
    "score_dni = model_Boulder_dni.evaluate(test_input_arry, test_dni_arry, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "preds = model_Boulder_dni.predict()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## V.III Model Testing Miami Specific\n",
    "Lets train on NRSDB + Darksky data from 08-18"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[]\n",
      "[]\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Miami_FL/psm_FL_Miami2009.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Miami_FL/psm_FL_Miami2008.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Miami_FL/psm_FL_Miami2018.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Miami_FL/psm_FL_Miami2017.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Miami_FL/psm_FL_Miami2016.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Miami_FL/psm_FL_Miami2014.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Miami_FL/psm_FL_Miami2015.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Miami_FL/psm_FL_Miami2011.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Miami_FL/psm_FL_Miami2010.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Miami_FL/psm_FL_Miami2012.csv\n",
      "/Users/tuomastalvitie/Documents/Diffuse:Direct/solarIrradiencePredictor/Testing_Data/Miami_FL/psm_FL_Miami2013.csv\n"
     ]
    }
   ],
   "source": [
    "input_size = 6\n",
    "input_arry = np.array([], dtype=np.float64).reshape(0,input_size)\n",
    "output_arry = np.array([], dtype=np.float64)\n",
    "dni_arry = np.array([], dtype=np.float64)\n",
    "\n",
    "print(input_arry)\n",
    "print(output_arry)\n",
    "from glob import glob\n",
    "for path in glob(_Testing_Data_Dir+'/Miami_FL'):\n",
    "    for file in glob(path+'/psm_'+'*'):\n",
    "        print(file)\n",
    "        df = pd.read_csv(file)\n",
    "        input_arry, output_arry, dni_arry = combine_training_arrays(df, input_arry, output_arry, dni_arry)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<function tensorflow.python.framework.ops.reset_default_graph()>"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "keras.backend.clear_session()\n",
    "tf.keras.backend.clear_session()\n",
    "tf.compat.v1.reset_default_graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_Miami = keras.Sequential()\n",
    "model_Miami.add(keras.layers.Dense(300, activation=\"relu\"))\n",
    "model_Miami.add(layers.Dense(200, activation=\"relu\"))\n",
    "model_Miami.add(layers.Dense(100, activation=\"relu\"))\n",
    "model_Miami.add(layers.Dense(50, activation=\"relu\"))\n",
    "model_Miami.add(layers.Dense(25, activation=\"relu\"))\n",
    "model_Miami.add(layers.Dense(1, activation='relu'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 2821.8845 - mae: 28.0470 - root_mean_squared_error: 53.1214: 0s - loss: 3199.4131 - mae: 30.5490 - root_mean_squared_\n",
      "Epoch 2/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 347.6064 - mae: 9.5146 - root_mean_squared_error: 18.6442\n",
      "Epoch 3/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 238.4990 - mae: 7.1761 - root_mean_squared_error: 15.4434\n",
      "Epoch 4/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 220.8985 - mae: 6.7411 - root_mean_squared_error: 14.8627\n",
      "Epoch 5/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 210.2945 - mae: 6.4900 - root_mean_squared_error: 14.5015\n",
      "Epoch 6/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 213.8370 - mae: 6.4973 - root_mean_squared_error: 14.6232\n",
      "Epoch 7/100\n",
      "1506/1506 [==============================] - 4s 3ms/step - loss: 206.8211 - mae: 6.3599 - root_mean_squared_error: 14.3813\n",
      "Epoch 8/100\n",
      "1506/1506 [==============================] - 4s 2ms/step - loss: 214.3339 - mae: 6.4584 - root_mean_squared_error: 14.6401\n",
      "Epoch 9/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 203.5709 - mae: 6.2699 - root_mean_squared_error: 14.2678\n",
      "Epoch 10/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 207.8142 - mae: 6.3201 - root_mean_squared_error: 14.4158\n",
      "Epoch 11/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 200.8087 - mae: 6.1515 - root_mean_squared_error: 14.1707\n",
      "Epoch 12/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 196.8871 - mae: 6.0559 - root_mean_squared_error: 14.0316\n",
      "Epoch 13/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 197.8395 - mae: 6.1094 - root_mean_squared_error: 14.0655\n",
      "Epoch 14/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 198.9715 - mae: 6.1421 - root_mean_squared_error: 14.1057\n",
      "Epoch 15/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 197.6367 - mae: 6.0797 - root_mean_squared_error: 14.0583\n",
      "Epoch 16/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 197.0545 - mae: 6.0644 - root_mean_squared_error: 14.0376\n",
      "Epoch 17/100\n",
      "1506/1506 [==============================] - 4s 3ms/step - loss: 197.4313 - mae: 6.0725 - root_mean_squared_error: 14.0510\n",
      "Epoch 18/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 189.1157 - mae: 5.9246 - root_mean_squared_error: 13.7519\n",
      "Epoch 19/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 194.5186 - mae: 6.0056 - root_mean_squared_error: 13.9470\n",
      "Epoch 20/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 195.1071 - mae: 6.0083 - root_mean_squared_error: 13.9681\n",
      "Epoch 21/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 196.4428 - mae: 6.0260 - root_mean_squared_error: 14.0158\n",
      "Epoch 22/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 192.1096 - mae: 5.9641 - root_mean_squared_error: 13.8604\n",
      "Epoch 23/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 192.8724 - mae: 5.9798 - root_mean_squared_error: 13.8878\n",
      "Epoch 24/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 189.6416 - mae: 5.9224 - root_mean_squared_error: 13.7710\n",
      "Epoch 25/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 193.5787 - mae: 6.0033 - root_mean_squared_error: 13.9133\n",
      "Epoch 26/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 187.9339 - mae: 5.8696 - root_mean_squared_error: 13.7089\n",
      "Epoch 27/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 190.8081 - mae: 5.9508 - root_mean_squared_error: 13.8133\n",
      "Epoch 28/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 189.6211 - mae: 5.8983 - root_mean_squared_error: 13.7703\n",
      "Epoch 29/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 192.2378 - mae: 5.9568 - root_mean_squared_error: 13.8650\n",
      "Epoch 30/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 189.0360 - mae: 5.9033 - root_mean_squared_error: 13.7490\n",
      "Epoch 31/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 186.4806 - mae: 5.8329 - root_mean_squared_error: 13.6558\n",
      "Epoch 32/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 186.7839 - mae: 5.8258 - root_mean_squared_error: 13.6669\n",
      "Epoch 33/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 185.5470 - mae: 5.7943 - root_mean_squared_error: 13.6216: 0s - loss: 181.7961 - mae: 5.7235 - root_mean_square\n",
      "Epoch 34/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 187.7018 - mae: 5.8464 - root_mean_squared_error: 13.7004\n",
      "Epoch 35/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 188.8217 - mae: 5.8503 - root_mean_squared_error: 13.7412\n",
      "Epoch 36/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 185.6020 - mae: 5.8107 - root_mean_squared_error: 13.6236: 0s - loss: 179.1104 - mae: 5.7119 - ro\n",
      "Epoch 37/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 186.2074 - mae: 5.8349 - root_mean_squared_error: 13.6458\n",
      "Epoch 38/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 184.0135 - mae: 5.7483 - root_mean_squared_error: 13.5652: 2s - loss: 168.3643 - mae: 5.5389 - root_mean_squared_error: 12. - ETA: 2s - loss: 170.7218 - mae: 5.4918 - root - ETA\n",
      "Epoch 39/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 185.1743 - mae: 5.8020 - root_mean_squared_error: 13.6079\n",
      "Epoch 40/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 179.4367 - mae: 5.6880 - root_mean_squared_error: 13.3954\n",
      "Epoch 41/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 185.6613 - mae: 5.8068 - root_mean_squared_error: 13.6258: 1s\n",
      "Epoch 42/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 185.2198 - mae: 5.8003 - root_mean_squared_error: 13.6095\n",
      "Epoch 43/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 183.7556 - mae: 5.7565 - root_mean_squared_error: 13.5556\n",
      "Epoch 44/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 186.5438 - mae: 5.8210 - root_mean_squared_error: 13.6581\n",
      "Epoch 45/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 181.4390 - mae: 5.7052 - root_mean_squared_error: 13.4699\n",
      "Epoch 46/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 182.0834 - mae: 5.7238 - root_mean_squared_error: 13.4938\n",
      "Epoch 47/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 180.1544 - mae: 5.6753 - root_mean_squared_error: 13.4222\n",
      "Epoch 48/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 178.8092 - mae: 5.6463 - root_mean_squared_error: 13.3720\n",
      "Epoch 49/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 184.5269 - mae: 5.7801 - root_mean_squared_error: 13.5841\n",
      "Epoch 50/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 181.5386 - mae: 5.7335 - root_mean_squared_error: 13.4736\n",
      "Epoch 51/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 180.8496 - mae: 5.6785 - root_mean_squared_error: 13.4480\n",
      "Epoch 52/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 181.1335 - mae: 5.7015 - root_mean_squared_error: 13.4586\n",
      "Epoch 53/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 181.9534 - mae: 5.7066 - root_mean_squared_error: 13.4890\n",
      "Epoch 54/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 181.3833 - mae: 5.7194 - root_mean_squared_error: 13.4679\n",
      "Epoch 55/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 178.3869 - mae: 5.6553 - root_mean_squared_error: 13.3562\n",
      "Epoch 56/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 180.9613 - mae: 5.7047 - root_mean_squared_error: 13.4522: 1s -\n",
      "Epoch 57/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 184.8574 - mae: 5.7481 - root_mean_squared_error: 13.5962\n",
      "Epoch 58/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 179.1429 - mae: 5.6545 - root_mean_squared_error: 13.3844\n",
      "Epoch 59/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1506/1506 [==============================] - 3s 2ms/step - loss: 176.5036 - mae: 5.5849 - root_mean_squared_error: 13.2855: 1s -\n",
      "Epoch 60/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 180.8584 - mae: 5.6784 - root_mean_squared_error: 13.4484\n",
      "Epoch 61/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 180.4018 - mae: 5.6680 - root_mean_squared_error: 13.4314\n",
      "Epoch 62/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 177.2824 - mae: 5.6184 - root_mean_squared_error: 13.3147\n",
      "Epoch 63/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 176.3687 - mae: 5.5835 - root_mean_squared_error: 13.2804: 0s - loss: 177.0906 - mae: 5.6186 - root_m\n",
      "Epoch 64/100\n",
      "1506/1506 [==============================] - 4s 2ms/step - loss: 175.8892 - mae: 5.6005 - root_mean_squared_error: 13.2623\n",
      "Epoch 65/100\n",
      "1506/1506 [==============================] - 4s 2ms/step - loss: 178.9175 - mae: 5.6409 - root_mean_squared_error: 13.3760\n",
      "Epoch 66/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 178.9872 - mae: 5.6318 - root_mean_squared_error: 13.3786\n",
      "Epoch 67/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 176.0798 - mae: 5.5700 - root_mean_squared_error: 13.2695\n",
      "Epoch 68/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 175.4893 - mae: 5.5776 - root_mean_squared_error: 13.2472\n",
      "Epoch 69/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 174.2338 - mae: 5.5443 - root_mean_squared_error: 13.1998: 0s - loss: 174.5295 - mae: 5.5535 - ro\n",
      "Epoch 70/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 174.9166 - mae: 5.5774 - root_mean_squared_error: 13.2256\n",
      "Epoch 71/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 175.3379 - mae: 5.5592 - root_mean_squared_error: 13.2415: 0s - loss: 176.9247 - mae: 5.5971 - root_mea\n",
      "Epoch 72/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 177.2766 - mae: 5.5908 - root_mean_squared_error: 13.3145\n",
      "Epoch 73/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 176.7438 - mae: 5.5887 - root_mean_squared_error: 13.2945\n",
      "Epoch 74/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 176.8699 - mae: 5.5951 - root_mean_squared_error: 13.2992\n",
      "Epoch 75/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 174.3920 - mae: 5.5368 - root_mean_squared_error: 13.2058: 0s - loss: 168.4373 \n",
      "Epoch 76/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 173.6733 - mae: 5.5244 - root_mean_squared_error: 13.1785\n",
      "Epoch 77/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 173.6910 - mae: 5.5060 - root_mean_squared_error: 13.1792\n",
      "Epoch 78/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 176.5367 - mae: 5.5788 - root_mean_squared_error: 13.2867\n",
      "Epoch 79/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 174.2255 - mae: 5.5445 - root_mean_squared_error: 13.1995\n",
      "Epoch 80/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 175.1773 - mae: 5.5529 - root_mean_squared_error: 13.2355: 0s - loss: 174.0788 - mae: 5.5228 - root_mean_squared_error:\n",
      "Epoch 81/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 172.6248 - mae: 5.4995 - root_mean_squared_error: 13.1387\n",
      "Epoch 82/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 173.9524 - mae: 5.5386 - root_mean_squared_error: 13.1891\n",
      "Epoch 83/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 173.3291 - mae: 5.4896 - root_mean_squared_error: 13.1654\n",
      "Epoch 84/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 175.1248 - mae: 5.5663 - root_mean_squared_error: 13.2335\n",
      "Epoch 85/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 176.1440 - mae: 5.6041 - root_mean_squared_error: 13.2719\n",
      "Epoch 86/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 171.4935 - mae: 5.4505 - root_mean_squared_error: 13.0956\n",
      "Epoch 87/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 172.4312 - mae: 5.4770 - root_mean_squared_error: 13.1313\n",
      "Epoch 88/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 172.1769 - mae: 5.4837 - root_mean_squared_error: 13.1216\n",
      "Epoch 89/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 171.0313 - mae: 5.4436 - root_mean_squared_error: 13.0779\n",
      "Epoch 90/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 175.6559 - mae: 5.5796 - root_mean_squared_error: 13.2535\n",
      "Epoch 91/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 171.0574 - mae: 5.4500 - root_mean_squared_error: 13.0789\n",
      "Epoch 92/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 171.8045 - mae: 5.4684 - root_mean_squared_error: 13.1074\n",
      "Epoch 93/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 174.8482 - mae: 5.5479 - root_mean_squared_error: 13.2230\n",
      "Epoch 94/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 171.5979 - mae: 5.4683 - root_mean_squared_error: 13.0995\n",
      "Epoch 95/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 173.4941 - mae: 5.5173 - root_mean_squared_error: 13.1717\n",
      "Epoch 96/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 172.0383 - mae: 5.4657 - root_mean_squared_error: 13.1163: 0s - loss: 172.2038 - mae: 5.4640 - root_mean_squared_error:\n",
      "Epoch 97/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 170.1415 - mae: 5.4399 - root_mean_squared_error: 13.0438\n",
      "Epoch 98/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 169.2349 - mae: 5.4042 - root_mean_squared_error: 13.0090\n",
      "Epoch 99/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 172.1746 - mae: 5.5191 - root_mean_squared_error: 13.1215\n",
      "Epoch 100/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 169.4010 - mae: 5.3901 - root_mean_squared_error: 13.0154\n"
     ]
    }
   ],
   "source": [
    "#Compile and train\n",
    "model_Miami.compile(optimizer=Adam(lr=0.0001), \n",
    "              loss='mse',\n",
    "              metrics=['mae', keras.metrics.RootMeanSquaredError()])\n",
    "\n",
    "history_Miami = model_Miami.fit(input_arry, output_arry, epochs=100, batch_size=64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Mae down to 5.4439"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "#"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history_Miami.history['mae'])\n",
    "plt.plot(history_Miami.history['root_mean_squared_error'])\n",
    "plt.title('mae and rmse per epoch')\n",
    "plt.ylabel('Error')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['mae', 'rmse'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "# And DNI\n",
    "model_Miami_dni = keras.Sequential()\n",
    "model_Miami_dni.add(keras.layers.Dense(300, activation=\"relu\"))\n",
    "model_Miami_dni.add(layers.Dense(200, activation=\"relu\"))\n",
    "model_Miami_dni.add(layers.Dense(100, activation=\"relu\"))\n",
    "model_Miami_dni.add(layers.Dense(50, activation=\"relu\"))\n",
    "model_Miami_dni.add(layers.Dense(25, activation=\"relu\"))\n",
    "model_Miami_dni.add(layers.Dense(1, activation='relu'))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 24441.1426 - mae: 95.3410 - root_mean_squared_error: 156.3366TA: 0s - loss: 26539.7012 - mae: 99.8543 - root_mean_\n",
      "Epoch 2/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 3993.5186 - mae: 33.8525 - root_mean_squared_error: 63.1943\n",
      "Epoch 3/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 1245.0220 - mae: 18.3697 - root_mean_squared_error: 35.2849\n",
      "Epoch 4/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 879.0259 - mae: 15.0913 - root_mean_squared_error: 29.6484\n",
      "Epoch 5/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 718.7422 - mae: 13.1955 - root_mean_squared_error: 26.8094\n",
      "Epoch 6/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 691.8619 - mae: 12.8896 - root_mean_squared_error: 26.3033\n",
      "Epoch 7/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 650.6275 - mae: 12.3042 - root_mean_squared_error: 25.5074\n",
      "Epoch 8/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 633.1435 - mae: 12.1261 - root_mean_squared_error: 25.1623\n",
      "Epoch 9/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 633.6764 - mae: 12.0898 - root_mean_squared_error: 25.1729\n",
      "Epoch 10/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 615.7643 - mae: 11.9000 - root_mean_squared_error: 24.8146\n",
      "Epoch 11/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 659.0756 - mae: 12.3419 - root_mean_squared_error: 25.6725\n",
      "Epoch 12/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 623.2617 - mae: 11.9586 - root_mean_squared_error: 24.9652\n",
      "Epoch 13/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 592.5181 - mae: 11.6339 - root_mean_squared_error: 24.3417A: 1s - loss: 590.9510 - mae: 11.5824 - roo\n",
      "Epoch 14/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 609.8351 - mae: 11.8568 - root_mean_squared_error: 24.6948\n",
      "Epoch 15/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 596.3159 - mae: 11.6515 - root_mean_squared_error: 24.4196\n",
      "Epoch 16/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 598.0421 - mae: 11.6448 - root_mean_squared_error: 24.4549\n",
      "Epoch 17/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 602.2962 - mae: 11.7003 - root_mean_squared_error: 24.5417\n",
      "Epoch 18/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 601.7975 - mae: 11.6804 - root_mean_squared_error: 24.5316\n",
      "Epoch 19/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 588.3111 - mae: 11.5283 - root_mean_squared_error: 24.2551\n",
      "Epoch 20/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 592.3995 - mae: 11.5854 - root_mean_squared_error: 24.3393\n",
      "Epoch 21/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 588.8544 - mae: 11.5274 - root_mean_squared_error: 24.2663A: 0s - loss: 590.2387 - mae: 11.5334 - root_mean_squared_error: 24 - ETA: 0s - loss: 584.5585 - mae: 11.4932 - root_mean_squared_error: \n",
      "Epoch 22/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 593.1629 - mae: 11.5734 - root_mean_squared_error: 24.3549\n",
      "Epoch 23/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 580.7520 - mae: 11.4883 - root_mean_squared_error: 24.0988\n",
      "Epoch 24/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 590.0010 - mae: 11.5583 - root_mean_squared_error: 24.2899\n",
      "Epoch 25/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 575.6038 - mae: 11.3980 - root_mean_squared_error: 23.9917\n",
      "Epoch 26/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 582.2065 - mae: 11.4798 - root_mean_squared_error: 24.1290\n",
      "Epoch 27/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 583.8294 - mae: 11.4598 - root_mean_squared_error: 24.1626\n",
      "Epoch 28/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 548.9653 - mae: 11.0796 - root_mean_squared_error: 23.4300\n",
      "Epoch 29/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 570.8430 - mae: 11.3448 - root_mean_squared_error: 23.8923\n",
      "Epoch 30/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 563.2712 - mae: 11.2609 - root_mean_squared_error: 23.7333\n",
      "Epoch 31/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 568.4005 - mae: 11.3159 - root_mean_squared_error: 23.8412\n",
      "Epoch 32/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 562.7807 - mae: 11.2398 - root_mean_squared_error: 23.7230A: 0s - loss: 571.7056 - mae: 11.3758 - root_mean_sq\n",
      "Epoch 33/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 548.8454 - mae: 11.0805 - root_mean_squared_error: 23.4275\n",
      "Epoch 34/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 565.1383 - mae: 11.2518 - root_mean_squared_error: 23.7726\n",
      "Epoch 35/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 560.4287 - mae: 11.2163 - root_mean_squared_error: 23.6734A: 0s - loss: 557.1215 - mae: 11.1748 - root_mean_squared_error: 23.\n",
      "Epoch 36/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 555.7074 - mae: 11.1635 - root_mean_squared_error: 23.5734A: 1s - loss: 551.6968 - mae: 11.2226 -\n",
      "Epoch 37/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 552.5774 - mae: 11.1255 - root_mean_squared_error: 23.5070\n",
      "Epoch 38/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 558.8302 - mae: 11.1940 - root_mean_squared_error: 23.6396\n",
      "Epoch 39/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 556.3363 - mae: 11.1554 - root_mean_squared_error: 23.5868\n",
      "Epoch 40/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 546.4764 - mae: 11.0733 - root_mean_squared_error: 23.3768\n",
      "Epoch 41/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 541.8518 - mae: 10.9818 - root_mean_squared_error: 23.2777\n",
      "Epoch 42/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 545.5273 - mae: 11.0657 - root_mean_squared_error: 23.3565A: 1s - loss: 548.2637 - mae: 11.1054 - root - ETA: 0s - loss: 532.6329 - mae: 10.9295 - root_mean_squared_error: \n",
      "Epoch 43/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 560.4739 - mae: 11.2144 - root_mean_squared_error: 23.6743\n",
      "Epoch 44/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 561.5544 - mae: 11.2337 - root_mean_squared_error: 23.6971\n",
      "Epoch 45/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 547.4160 - mae: 11.0388 - root_mean_squared_error: 23.3969\n",
      "Epoch 46/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 524.5800 - mae: 10.8225 - root_mean_squared_error: 22.9037\n",
      "Epoch 47/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 542.1990 - mae: 10.9893 - root_mean_squared_error: 23.2852\n",
      "Epoch 48/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 536.6486 - mae: 10.9162 - root_mean_squared_error: 23.1657\n",
      "Epoch 49/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 532.4825 - mae: 10.8809 - root_mean_squared_error: 23.0756A: 1s - loss: 501.3364 - mae: 10.5894 - root_mean - ETA: 0s - loss: 527.5193 - mae: 10.8276 - root_mean_squared_err\n",
      "Epoch 50/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 544.5115 - mae: 11.0171 - root_mean_squared_error: 23.3348\n",
      "Epoch 51/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 524.5927 - mae: 10.7903 - root_mean_squared_error: 22.9040\n",
      "Epoch 52/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 538.9692 - mae: 10.9457 - root_mean_squared_error: 23.2157\n",
      "Epoch 53/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 533.2513 - mae: 10.9279 - root_mean_squared_error: 23.0922\n",
      "Epoch 54/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 534.6094 - mae: 10.9206 - root_mean_squared_error: 23.1216\n",
      "Epoch 55/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 540.4200 - mae: 10.9953 - root_mean_squared_error: 23.2469\n",
      "Epoch 56/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1506/1506 [==============================] - 2s 1ms/step - loss: 529.7480 - mae: 10.8679 - root_mean_squared_error: 23.0163A: 0s - loss: 534.2669 - mae: 10.8995 - root_mean_squared_err\n",
      "Epoch 57/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 528.3816 - mae: 10.8613 - root_mean_squared_error: 22.9866A: 0s - loss: 514.9348 - mae: 10.7114 - root_mean_sq\n",
      "Epoch 58/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 524.3574 - mae: 10.7890 - root_mean_squared_error: 22.8989A: 1s - loss: 528.5540 - mae: 10.8230 -\n",
      "Epoch 59/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 528.8050 - mae: 10.8465 - root_mean_squared_error: 22.9958\n",
      "Epoch 60/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 526.3825 - mae: 10.8099 - root_mean_squared_error: 22.9430\n",
      "Epoch 61/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 523.4461 - mae: 10.7752 - root_mean_squared_error: 22.8789\n",
      "Epoch 62/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 519.2808 - mae: 10.7368 - root_mean_squared_error: 22.7877A: 0s - loss: 517.4283 - mae: 10.7055 - root_mean_squared_error: 22.\n",
      "Epoch 63/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 526.5548 - mae: 10.8212 - root_mean_squared_error: 22.9468\n",
      "Epoch 64/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 518.3933 - mae: 10.7612 - root_mean_squared_error: 22.7683A: 0s - loss: 516.3441 - mae: 10.7232 - root_mean_square\n",
      "Epoch 65/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 523.6528 - mae: 10.8115 - root_mean_squared_error: 22.8835A: 2s - loss: 564.9816 - mae: 11.1951 - root_mean_squa - ETA: 1s - loss: 539.5856 - mae: 10.9459 - root_mean_\n",
      "Epoch 66/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 516.1694 - mae: 10.7386 - root_mean_squared_error: 22.7194\n",
      "Epoch 67/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 519.0080 - mae: 10.7357 - root_mean_squared_error: 22.7817\n",
      "Epoch 68/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 516.5555 - mae: 10.7081 - root_mean_squared_error: 22.7279\n",
      "Epoch 69/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 523.9865 - mae: 10.7861 - root_mean_squared_error: 22.8908\n",
      "Epoch 70/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 502.6943 - mae: 10.5534 - root_mean_squared_error: 22.4208\n",
      "Epoch 71/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 515.0601 - mae: 10.6887 - root_mean_squared_error: 22.6949\n",
      "Epoch 72/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 527.1155 - mae: 10.8299 - root_mean_squared_error: 22.9590\n",
      "Epoch 73/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 512.8336 - mae: 10.6624 - root_mean_squared_error: 22.6458\n",
      "Epoch 74/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 515.7070 - mae: 10.6879 - root_mean_squared_error: 22.7092\n",
      "Epoch 75/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 525.7398 - mae: 10.8142 - root_mean_squared_error: 22.9290A: 1s - loss: 518.8105 - mae: 10.7733 - root_mean_squared_er - ETA: 0s - loss: 528.9537 - mae: 10.9000 - root_mean_squared_er\n",
      "Epoch 76/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 514.6820 - mae: 10.6622 - root_mean_squared_error: 22.6866\n",
      "Epoch 77/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 508.7832 - mae: 10.6096 - root_mean_squared_error: 22.5562\n",
      "Epoch 78/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 513.6507 - mae: 10.6686 - root_mean_squared_error: 22.6639\n",
      "Epoch 79/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 512.8212 - mae: 10.6671 - root_mean_squared_error: 22.6456\n",
      "Epoch 80/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 513.6895 - mae: 10.6801 - root_mean_squared_error: 22.6647A: 2s - loss: 509.9710 - mae: 1\n",
      "Epoch 81/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 509.1351 - mae: 10.5958 - root_mean_squared_error: 22.5640A: 0s - loss: 506.8447 - mae: 10.5943 - root_mean_squared_er\n",
      "Epoch 82/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 506.5152 - mae: 10.5503 - root_mean_squared_error: 22.5059\n",
      "Epoch 83/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 504.4672 - mae: 10.5637 - root_mean_squared_error: 22.4603\n",
      "Epoch 84/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 508.9622 - mae: 10.6057 - root_mean_squared_error: 22.5602\n",
      "Epoch 85/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 516.4899 - mae: 10.7291 - root_mean_squared_error: 22.7264A: 1s - loss: 509.7965 - mae: \n",
      "Epoch 86/100\n",
      "1506/1506 [==============================] - 4s 3ms/step - loss: 502.4246 - mae: 10.5441 - root_mean_squared_error: 22.4148\n",
      "Epoch 87/100\n",
      "1506/1506 [==============================] - 4s 3ms/step - loss: 506.8261 - mae: 10.5947 - root_mean_squared_error: 22.5128\n",
      "Epoch 88/100\n",
      "1506/1506 [==============================] - 5s 3ms/step - loss: 517.8685 - mae: 10.7140 - root_mean_squared_error: 22.7567\n",
      "Epoch 89/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 509.4811 - mae: 10.6315 - root_mean_squared_error: 22.5717\n",
      "Epoch 90/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 500.9059 - mae: 10.4885 - root_mean_squared_error: 22.3809\n",
      "Epoch 91/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 493.9611 - mae: 10.4138 - root_mean_squared_error: 22.2252\n",
      "Epoch 92/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 497.2519 - mae: 10.4889 - root_mean_squared_error: 22.2991\n",
      "Epoch 93/100\n",
      "1506/1506 [==============================] - 3s 2ms/step - loss: 499.5031 - mae: 10.4862 - root_mean_squared_error: 22.3496\n",
      "Epoch 94/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 497.3579 - mae: 10.4820 - root_mean_squared_error: 22.3015\n",
      "Epoch 95/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 499.1127 - mae: 10.4983 - root_mean_squared_error: 22.3408\n",
      "Epoch 96/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 498.1273 - mae: 10.4923 - root_mean_squared_error: 22.3188\n",
      "Epoch 97/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 489.7868 - mae: 10.3912 - root_mean_squared_error: 22.1311\n",
      "Epoch 98/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 496.9417 - mae: 10.4883 - root_mean_squared_error: 22.2922A: 2s - loss: 488.2690 - mae: 10.\n",
      "Epoch 99/100\n",
      "1506/1506 [==============================] - 2s 1ms/step - loss: 498.0988 - mae: 10.4644 - root_mean_squared_error: 22.3181\n",
      "Epoch 100/100\n",
      "1506/1506 [==============================] - 2s 2ms/step - loss: 490.8223 - mae: 10.4317 - root_mean_squared_error: 22.1545\n"
     ]
    }
   ],
   "source": [
    "#Compile and train\n",
    "model_Miami_dni.compile(optimizer=Adam(lr=0.0001), \n",
    "              loss='mse',\n",
    "              metrics=['mae', keras.metrics.RootMeanSquaredError()])\n",
    "\n",
    "history_Miami_dni = model_Miami_dni.fit(input_arry, dni_arry, epochs=100, batch_size=64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/site-packages/matplotlib-2.2.4-py3.7-macosx-10.15-x86_64.egg/matplotlib/legend.py:640: UserWarning: Unrecognized location \"middle\". Falling back on \"best\"; valid locations are\n",
      "\tbest\n",
      "\tupper right\n",
      "\tupper left\n",
      "\tlower left\n",
      "\tlower right\n",
      "\tright\n",
      "\tcenter left\n",
      "\tcenter right\n",
      "\tlower center\n",
      "\tupper center\n",
      "\tcenter\n",
      "\n",
      "  % (loc, '\\n\\t'.join(self.codes)))\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history_Miami.history['mae'])\n",
    "plt.plot(history_Miami.history['root_mean_squared_error'])\n",
    "plt.plot(history_Miami_dni.history['mae'])\n",
    "plt.plot(history_Miami_dni.history['root_mean_squared_error'])\n",
    "plt.title('MAE and RMSE per epoch')\n",
    "plt.ylabel('Error')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['MAE', 'RMSE','MAE-DNI', 'RMSE-DNI' ], loc='middle')\n",
    "plt.savefig('error-v-epoch-Miami.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Test on Miami 2018 data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "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>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Day</th>\n",
       "      <th>Hour</th>\n",
       "      <th>Minute</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Clearsky DHI</th>\n",
       "      <th>Clearsky DNI</th>\n",
       "      <th>Clearsky GHI</th>\n",
       "      <th>Cloud Type</th>\n",
       "      <th>...</th>\n",
       "      <th>Solar Zenith Angle</th>\n",
       "      <th>Surface Albedo</th>\n",
       "      <th>Pressure</th>\n",
       "      <th>Precipitable Water</th>\n",
       "      <th>Wind Direction</th>\n",
       "      <th>Wind Speed</th>\n",
       "      <th>Global Horizontal UV Irradiance (280-400nm)</th>\n",
       "      <th>Global Horizontal UV Irradiance (285-385nm)</th>\n",
       "      <th>timestamps</th>\n",
       "      <th>Cloud Cover</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>30</td>\n",
       "      <td>21.1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>113.89</td>\n",
       "      <td>0.08</td>\n",
       "      <td>1019</td>\n",
       "      <td>2.7</td>\n",
       "      <td>58</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1514782800</td>\n",
       "      <td>0.30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>21.2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>127.06</td>\n",
       "      <td>0.08</td>\n",
       "      <td>1019</td>\n",
       "      <td>2.8</td>\n",
       "      <td>63</td>\n",
       "      <td>1.8</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1514786400</td>\n",
       "      <td>0.32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>30</td>\n",
       "      <td>21.2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>140.43</td>\n",
       "      <td>0.08</td>\n",
       "      <td>1019</td>\n",
       "      <td>2.8</td>\n",
       "      <td>65</td>\n",
       "      <td>1.6</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1514790000</td>\n",
       "      <td>0.23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>30</td>\n",
       "      <td>21.2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>153.91</td>\n",
       "      <td>0.08</td>\n",
       "      <td>1018</td>\n",
       "      <td>2.9</td>\n",
       "      <td>57</td>\n",
       "      <td>1.3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1514793600</td>\n",
       "      <td>0.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>30</td>\n",
       "      <td>21.1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>167.37</td>\n",
       "      <td>0.08</td>\n",
       "      <td>1018</td>\n",
       "      <td>2.9</td>\n",
       "      <td>41</td>\n",
       "      <td>1.1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1514797200</td>\n",
       "      <td>0.14</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 26 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Year  Month  Day  Hour  Minute  Temperature  Clearsky DHI  Clearsky DNI  \\\n",
       "0  2018      1    1     0      30         21.1             0             0   \n",
       "1  2018      1    1     1      30         21.2             0             0   \n",
       "2  2018      1    1     2      30         21.2             0             0   \n",
       "3  2018      1    1     3      30         21.2             0             0   \n",
       "4  2018      1    1     4      30         21.1             0             0   \n",
       "\n",
       "   Clearsky GHI  Cloud Type  ...  Solar Zenith Angle  Surface Albedo  \\\n",
       "0             0           0  ...              113.89            0.08   \n",
       "1             0           0  ...              127.06            0.08   \n",
       "2             0           0  ...              140.43            0.08   \n",
       "3             0           0  ...              153.91            0.08   \n",
       "4             0           0  ...              167.37            0.08   \n",
       "\n",
       "   Pressure  Precipitable Water  Wind Direction  Wind Speed  \\\n",
       "0      1019                 2.7              58         2.0   \n",
       "1      1019                 2.8              63         1.8   \n",
       "2      1019                 2.8              65         1.6   \n",
       "3      1018                 2.9              57         1.3   \n",
       "4      1018                 2.9              41         1.1   \n",
       "\n",
       "   Global Horizontal UV Irradiance (280-400nm)  \\\n",
       "0                                          NaN   \n",
       "1                                          NaN   \n",
       "2                                          NaN   \n",
       "3                                          NaN   \n",
       "4                                          NaN   \n",
       "\n",
       "   Global Horizontal UV Irradiance (285-385nm)  timestamps  Cloud Cover  \n",
       "0                                          NaN  1514782800         0.30  \n",
       "1                                          NaN  1514786400         0.32  \n",
       "2                                          NaN  1514790000         0.23  \n",
       "3                                          NaN  1514793600         0.14  \n",
       "4                                          NaN  1514797200         0.14  \n",
       "\n",
       "[5 rows x 26 columns]"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df = pd.read_csv(os.path.join(_Testing_Data_Dir, 'Miami_FL', 'psm_FL_Miami2018.csv'))\n",
    "test_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "input_size = 6\n",
    "test_input_arry = np.array([], dtype=np.float64).reshape(0,input_size)\n",
    "test_output_arry = np.array([], dtype=np.float64)\n",
    "test_dni_arry = np.array([], dtype=np.float64)\n",
    "test_input_arry, test_output_arry, test_dni_arry = combine_training_arrays(test_df, test_input_arry, test_output_arry, test_dni_arry)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "274/274 [==============================] - 0s 794us/step - loss: 672.0304 - mae: 10.6658 - root_mean_squared_error: 25.9235\n"
     ]
    }
   ],
   "source": [
    "score = model_Miami.evaluate(test_input_arry, test_output_arry, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Mae of 10.1002....not bad!~"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "274/274 [==============================] - 0s 817us/step - loss: 1450.7910 - mae: 18.8467 - root_mean_squared_error: 38.0893\n"
     ]
    }
   ],
   "source": [
    "score_dni = model_Miami_dni.evaluate(test_input_arry, test_dni_arry, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Save models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: Neural_Nets/Neural_Net_Boulder_CO/assets\n"
     ]
    }
   ],
   "source": [
    "model_Boulder.save('Neural_Nets/Neural_Net_Boulder_CO')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /usr/local/lib/python3.7/site-packages/tensorflow/python/training/tracking/tracking.py:111: Model.state_updates (from tensorflow.python.keras.engine.training) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "This property should not be used in TensorFlow 2.0, as updates are applied automatically.\n",
      "WARNING:tensorflow:From /usr/local/lib/python3.7/site-packages/tensorflow/python/training/tracking/tracking.py:111: Layer.updates (from tensorflow.python.keras.engine.base_layer) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "This property should not be used in TensorFlow 2.0, as updates are applied automatically.\n",
      "INFO:tensorflow:Assets written to: Neural_Nets/Neural_Net_Boulder_CO-dni/assets\n"
     ]
    }
   ],
   "source": [
    "model_Boulder_dni.save('Neural_Nets/Neural_Net_Boulder_CO-dni')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: Neural_Nets/Neural_Net_Miami_FL/assets\n"
     ]
    }
   ],
   "source": [
    "model_Miami.save('Neural_Nets/Neural_Net_Miami_FL')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: Neural_Nets/Neural_Net_Miami_FL-dni/assets\n"
     ]
    }
   ],
   "source": [
    "model_Miami_dni.save('Neural_Nets/Neural_Net_Miami_FL-dni')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}