2. ML - LazyPredict.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Finding the Best Machine Learning Regressor for the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\K.N.Dheeraj\\anaconda3\\lib\\site-packages\\sklearn\\utils\\deprecation.py:143: FutureWarning: The sklearn.utils.testing module is  deprecated in version 0.22 and will be removed in version 0.24. The corresponding classes / functions should instead be imported from sklearn.utils. Anything that cannot be imported from sklearn.utils is now part of the private API.\n",
      "  warnings.warn(message, FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "# Basic Data Science Packages\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "#LazyPredict package and Regressor Import\n",
    "import lazypredict\n",
    "from lazypredict.Supervised import LazyRegressor\n",
    "\n",
    "#Plotting Graph\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "#Getting Regression Metrics\n",
    "from utility_functions import metrics\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "thesis_dataset = pd.read_csv('data/final_thesis_dataset.csv',parse_dates=[0], index_col=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>DE_wind_generation_actual</th>\n",
       "      <th>DE_solar_generation_actual</th>\n",
       "      <th>v1</th>\n",
       "      <th>v2</th>\n",
       "      <th>v_50m</th>\n",
       "      <th>h1</th>\n",
       "      <th>h2</th>\n",
       "      <th>z0</th>\n",
       "      <th>SWTDN</th>\n",
       "      <th>SWGDN</th>\n",
       "      <th>...</th>\n",
       "      <th>humidity</th>\n",
       "      <th>dew</th>\n",
       "      <th>windspeed</th>\n",
       "      <th>winddir</th>\n",
       "      <th>pressure</th>\n",
       "      <th>visibility</th>\n",
       "      <th>cloudcover</th>\n",
       "      <th>conditions</th>\n",
       "      <th>skyType</th>\n",
       "      <th>total_renewable_generation</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>utc_timestamp</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-01-01 00:00:00+00:00</th>\n",
       "      <td>8579.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>2.62</td>\n",
       "      <td>3.71</td>\n",
       "      <td>5.13</td>\n",
       "      <td>2.55</td>\n",
       "      <td>10.55</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>100.00</td>\n",
       "      <td>1.90</td>\n",
       "      <td>4.40</td>\n",
       "      <td>180.00</td>\n",
       "      <td>nan</td>\n",
       "      <td>0.40</td>\n",
       "      <td>98.40</td>\n",
       "      <td>Overcast</td>\n",
       "      <td>1.00</td>\n",
       "      <td>8579.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-01-01 01:00:00+00:00</th>\n",
       "      <td>8542.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>2.56</td>\n",
       "      <td>3.63</td>\n",
       "      <td>4.98</td>\n",
       "      <td>2.55</td>\n",
       "      <td>10.55</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>93.62</td>\n",
       "      <td>1.90</td>\n",
       "      <td>5.60</td>\n",
       "      <td>196.00</td>\n",
       "      <td>nan</td>\n",
       "      <td>0.30</td>\n",
       "      <td>59.20</td>\n",
       "      <td>Partially cloudy</td>\n",
       "      <td>2.00</td>\n",
       "      <td>8542.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-01-01 02:00:00+00:00</th>\n",
       "      <td>8443.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>2.48</td>\n",
       "      <td>3.52</td>\n",
       "      <td>4.79</td>\n",
       "      <td>2.55</td>\n",
       "      <td>10.55</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>94.10</td>\n",
       "      <td>2.00</td>\n",
       "      <td>6.70</td>\n",
       "      <td>230.00</td>\n",
       "      <td>nan</td>\n",
       "      <td>1.10</td>\n",
       "      <td>98.40</td>\n",
       "      <td>Overcast</td>\n",
       "      <td>1.00</td>\n",
       "      <td>8443.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-01-01 03:00:00+00:00</th>\n",
       "      <td>8295.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>2.40</td>\n",
       "      <td>3.40</td>\n",
       "      <td>4.66</td>\n",
       "      <td>2.55</td>\n",
       "      <td>10.55</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>94.06</td>\n",
       "      <td>1.10</td>\n",
       "      <td>5.10</td>\n",
       "      <td>218.00</td>\n",
       "      <td>nan</td>\n",
       "      <td>1.30</td>\n",
       "      <td>98.40</td>\n",
       "      <td>Overcast</td>\n",
       "      <td>1.00</td>\n",
       "      <td>8295.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-01-01 04:00:00+00:00</th>\n",
       "      <td>7319.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>2.31</td>\n",
       "      <td>3.29</td>\n",
       "      <td>4.57</td>\n",
       "      <td>2.55</td>\n",
       "      <td>10.55</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>93.61</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.90</td>\n",
       "      <td>202.00</td>\n",
       "      <td>nan</td>\n",
       "      <td>2.10</td>\n",
       "      <td>13.80</td>\n",
       "      <td>Clear</td>\n",
       "      <td>0.00</td>\n",
       "      <td>7319.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-12-31 19:00:00+00:00</th>\n",
       "      <td>15303.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>3.68</td>\n",
       "      <td>5.23</td>\n",
       "      <td>7.10</td>\n",
       "      <td>2.55</td>\n",
       "      <td>10.55</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>98.45</td>\n",
       "      <td>1.60</td>\n",
       "      <td>15.00</td>\n",
       "      <td>240.00</td>\n",
       "      <td>nan</td>\n",
       "      <td>3.90</td>\n",
       "      <td>98.40</td>\n",
       "      <td>Overcast</td>\n",
       "      <td>1.00</td>\n",
       "      <td>15303.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-12-31 20:00:00+00:00</th>\n",
       "      <td>15354.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>3.76</td>\n",
       "      <td>5.31</td>\n",
       "      <td>7.08</td>\n",
       "      <td>2.55</td>\n",
       "      <td>10.55</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>98.46</td>\n",
       "      <td>1.70</td>\n",
       "      <td>14.20</td>\n",
       "      <td>239.00</td>\n",
       "      <td>nan</td>\n",
       "      <td>4.30</td>\n",
       "      <td>98.40</td>\n",
       "      <td>Overcast</td>\n",
       "      <td>1.00</td>\n",
       "      <td>15354.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-12-31 21:00:00+00:00</th>\n",
       "      <td>14847.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>3.79</td>\n",
       "      <td>5.34</td>\n",
       "      <td>7.12</td>\n",
       "      <td>2.55</td>\n",
       "      <td>10.55</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>92.66</td>\n",
       "      <td>1.90</td>\n",
       "      <td>16.60</td>\n",
       "      <td>244.00</td>\n",
       "      <td>nan</td>\n",
       "      <td>8.00</td>\n",
       "      <td>98.40</td>\n",
       "      <td>Overcast</td>\n",
       "      <td>1.00</td>\n",
       "      <td>14847.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-12-31 22:00:00+00:00</th>\n",
       "      <td>15056.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>3.85</td>\n",
       "      <td>5.42</td>\n",
       "      <td>7.22</td>\n",
       "      <td>2.55</td>\n",
       "      <td>10.55</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>87.69</td>\n",
       "      <td>2.00</td>\n",
       "      <td>13.20</td>\n",
       "      <td>226.00</td>\n",
       "      <td>nan</td>\n",
       "      <td>10.00</td>\n",
       "      <td>98.40</td>\n",
       "      <td>Overcast</td>\n",
       "      <td>1.00</td>\n",
       "      <td>15056.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-12-31 23:00:00+00:00</th>\n",
       "      <td>15422.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>3.88</td>\n",
       "      <td>5.46</td>\n",
       "      <td>7.28</td>\n",
       "      <td>2.55</td>\n",
       "      <td>10.55</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>87.23</td>\n",
       "      <td>1.90</td>\n",
       "      <td>14.80</td>\n",
       "      <td>226.00</td>\n",
       "      <td>nan</td>\n",
       "      <td>10.00</td>\n",
       "      <td>98.40</td>\n",
       "      <td>Overcast</td>\n",
       "      <td>1.00</td>\n",
       "      <td>15422.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8711 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                           DE_wind_generation_actual  \\\n",
       "utc_timestamp                                          \n",
       "2016-01-01 00:00:00+00:00                    8579.00   \n",
       "2016-01-01 01:00:00+00:00                    8542.00   \n",
       "2016-01-01 02:00:00+00:00                    8443.00   \n",
       "2016-01-01 03:00:00+00:00                    8295.00   \n",
       "2016-01-01 04:00:00+00:00                    7319.00   \n",
       "...                                              ...   \n",
       "2016-12-31 19:00:00+00:00                   15303.00   \n",
       "2016-12-31 20:00:00+00:00                   15354.00   \n",
       "2016-12-31 21:00:00+00:00                   14847.00   \n",
       "2016-12-31 22:00:00+00:00                   15056.00   \n",
       "2016-12-31 23:00:00+00:00                   15422.00   \n",
       "\n",
       "                           DE_solar_generation_actual   v1   v2  v_50m   h1  \\\n",
       "utc_timestamp                                                                 \n",
       "2016-01-01 00:00:00+00:00                        0.00 2.62 3.71   5.13 2.55   \n",
       "2016-01-01 01:00:00+00:00                        0.00 2.56 3.63   4.98 2.55   \n",
       "2016-01-01 02:00:00+00:00                        0.00 2.48 3.52   4.79 2.55   \n",
       "2016-01-01 03:00:00+00:00                        0.00 2.40 3.40   4.66 2.55   \n",
       "2016-01-01 04:00:00+00:00                        0.00 2.31 3.29   4.57 2.55   \n",
       "...                                               ...  ...  ...    ...  ...   \n",
       "2016-12-31 19:00:00+00:00                        0.00 3.68 5.23   7.10 2.55   \n",
       "2016-12-31 20:00:00+00:00                        0.00 3.76 5.31   7.08 2.55   \n",
       "2016-12-31 21:00:00+00:00                        0.00 3.79 5.34   7.12 2.55   \n",
       "2016-12-31 22:00:00+00:00                        0.00 3.85 5.42   7.22 2.55   \n",
       "2016-12-31 23:00:00+00:00                        0.00 3.88 5.46   7.28 2.55   \n",
       "\n",
       "                             h2   z0  SWTDN  SWGDN  ...  humidity  dew  \\\n",
       "utc_timestamp                                       ...                  \n",
       "2016-01-01 00:00:00+00:00 10.55 0.16   0.00   0.00  ...    100.00 1.90   \n",
       "2016-01-01 01:00:00+00:00 10.55 0.16   0.00   0.00  ...     93.62 1.90   \n",
       "2016-01-01 02:00:00+00:00 10.55 0.16   0.00   0.00  ...     94.10 2.00   \n",
       "2016-01-01 03:00:00+00:00 10.55 0.16   0.00   0.00  ...     94.06 1.10   \n",
       "2016-01-01 04:00:00+00:00 10.55 0.16   0.00   0.00  ...     93.61 1.00   \n",
       "...                         ...  ...    ...    ...  ...       ...  ...   \n",
       "2016-12-31 19:00:00+00:00 10.55 0.16   0.00   0.00  ...     98.45 1.60   \n",
       "2016-12-31 20:00:00+00:00 10.55 0.16   0.00   0.00  ...     98.46 1.70   \n",
       "2016-12-31 21:00:00+00:00 10.55 0.16   0.00   0.00  ...     92.66 1.90   \n",
       "2016-12-31 22:00:00+00:00 10.55 0.16   0.00   0.00  ...     87.69 2.00   \n",
       "2016-12-31 23:00:00+00:00 10.55 0.16   0.00   0.00  ...     87.23 1.90   \n",
       "\n",
       "                           windspeed  winddir  pressure  visibility  \\\n",
       "utc_timestamp                                                         \n",
       "2016-01-01 00:00:00+00:00       4.40   180.00       nan        0.40   \n",
       "2016-01-01 01:00:00+00:00       5.60   196.00       nan        0.30   \n",
       "2016-01-01 02:00:00+00:00       6.70   230.00       nan        1.10   \n",
       "2016-01-01 03:00:00+00:00       5.10   218.00       nan        1.30   \n",
       "2016-01-01 04:00:00+00:00       1.90   202.00       nan        2.10   \n",
       "...                              ...      ...       ...         ...   \n",
       "2016-12-31 19:00:00+00:00      15.00   240.00       nan        3.90   \n",
       "2016-12-31 20:00:00+00:00      14.20   239.00       nan        4.30   \n",
       "2016-12-31 21:00:00+00:00      16.60   244.00       nan        8.00   \n",
       "2016-12-31 22:00:00+00:00      13.20   226.00       nan       10.00   \n",
       "2016-12-31 23:00:00+00:00      14.80   226.00       nan       10.00   \n",
       "\n",
       "                           cloudcover        conditions  skyType  \\\n",
       "utc_timestamp                                                      \n",
       "2016-01-01 00:00:00+00:00       98.40          Overcast     1.00   \n",
       "2016-01-01 01:00:00+00:00       59.20  Partially cloudy     2.00   \n",
       "2016-01-01 02:00:00+00:00       98.40          Overcast     1.00   \n",
       "2016-01-01 03:00:00+00:00       98.40          Overcast     1.00   \n",
       "2016-01-01 04:00:00+00:00       13.80             Clear     0.00   \n",
       "...                               ...               ...      ...   \n",
       "2016-12-31 19:00:00+00:00       98.40          Overcast     1.00   \n",
       "2016-12-31 20:00:00+00:00       98.40          Overcast     1.00   \n",
       "2016-12-31 21:00:00+00:00       98.40          Overcast     1.00   \n",
       "2016-12-31 22:00:00+00:00       98.40          Overcast     1.00   \n",
       "2016-12-31 23:00:00+00:00       98.40          Overcast     1.00   \n",
       "\n",
       "                           total_renewable_generation  \n",
       "utc_timestamp                                          \n",
       "2016-01-01 00:00:00+00:00                     8579.00  \n",
       "2016-01-01 01:00:00+00:00                     8542.00  \n",
       "2016-01-01 02:00:00+00:00                     8443.00  \n",
       "2016-01-01 03:00:00+00:00                     8295.00  \n",
       "2016-01-01 04:00:00+00:00                     7319.00  \n",
       "...                                               ...  \n",
       "2016-12-31 19:00:00+00:00                    15303.00  \n",
       "2016-12-31 20:00:00+00:00                    15354.00  \n",
       "2016-12-31 21:00:00+00:00                    14847.00  \n",
       "2016-12-31 22:00:00+00:00                    15056.00  \n",
       "2016-12-31 23:00:00+00:00                    15422.00  \n",
       "\n",
       "[8711 rows x 24 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "thesis_dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Split data into train and test datasets\n",
    "- Fetch the first 70% records as train and rest as test\n",
    "    - 8711 * 70% = 6098 records from the start (Train)\n",
    "    - 8711 - 6098 = 2613 remaining records as (Test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = thesis_dataset[:6098] \n",
    "test = thesis_dataset[6098:] \n",
    "\n",
    "X_solar_train = train[['SWTDN', 'SWGDN', 'T']]\n",
    "y_solar_train = train['DE_solar_generation_actual']\n",
    "X_solar_test = test[['SWTDN', 'SWGDN', 'T']]\n",
    "y_solar_test = test['DE_solar_generation_actual']\n",
    "\n",
    "X_wind_train = train[['v1', 'v2', 'v_50m', 'z0']]\n",
    "y_wind_train = train['DE_wind_generation_actual']\n",
    "X_wind_test = test[['v1', 'v2', 'v_50m', 'z0']]\n",
    "y_wind_test = test['DE_wind_generation_actual']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Applying LazyRegressor on Solar Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████████████████████████████████████████████| 42/42 [00:27<00:00,  1.54it/s]\n"
     ]
    },
    {
     "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>Adjusted R-Squared</th>\n",
       "      <th>R-Squared</th>\n",
       "      <th>RMSE</th>\n",
       "      <th>Time Taken</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Model</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>HistGradientBoostingRegressor</th>\n",
       "      <td>0.93</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1075.28</td>\n",
       "      <td>0.86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LGBMRegressor</th>\n",
       "      <td>0.93</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1075.40</td>\n",
       "      <td>0.15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>XGBRegressor</th>\n",
       "      <td>0.93</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1090.94</td>\n",
       "      <td>1.76</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PassiveAggressiveRegressor</th>\n",
       "      <td>0.93</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1091.96</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RANSACRegressor</th>\n",
       "      <td>0.93</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1093.64</td>\n",
       "      <td>0.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HuberRegressor</th>\n",
       "      <td>0.93</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1098.57</td>\n",
       "      <td>0.06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RandomForestRegressor</th>\n",
       "      <td>0.93</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1108.30</td>\n",
       "      <td>1.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OrthogonalMatchingPursuitCV</th>\n",
       "      <td>0.93</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1108.96</td>\n",
       "      <td>0.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LinearRegression</th>\n",
       "      <td>0.93</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1113.42</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TransformedTargetRegressor</th>\n",
       "      <td>0.93</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1113.42</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Lars</th>\n",
       "      <td>0.93</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1113.42</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LarsCV</th>\n",
       "      <td>0.93</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1113.42</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LassoLarsCV</th>\n",
       "      <td>0.93</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1113.42</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LassoLarsIC</th>\n",
       "      <td>0.93</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1113.42</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BayesianRidge</th>\n",
       "      <td>0.93</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1113.48</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RidgeCV</th>\n",
       "      <td>0.93</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1113.49</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ridge</th>\n",
       "      <td>0.93</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1114.12</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Lasso</th>\n",
       "      <td>0.93</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1114.18</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SGDRegressor</th>\n",
       "      <td>0.92</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1117.27</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LassoCV</th>\n",
       "      <td>0.92</td>\n",
       "      <td>0.92</td>\n",
       "      <td>1118.70</td>\n",
       "      <td>0.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BaggingRegressor</th>\n",
       "      <td>0.92</td>\n",
       "      <td>0.92</td>\n",
       "      <td>1121.58</td>\n",
       "      <td>0.11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ExtraTreesRegressor</th>\n",
       "      <td>0.92</td>\n",
       "      <td>0.92</td>\n",
       "      <td>1127.87</td>\n",
       "      <td>0.60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GradientBoostingRegressor</th>\n",
       "      <td>0.92</td>\n",
       "      <td>0.92</td>\n",
       "      <td>1144.04</td>\n",
       "      <td>0.44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OrthogonalMatchingPursuit</th>\n",
       "      <td>0.92</td>\n",
       "      <td>0.92</td>\n",
       "      <td>1159.02</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LassoLars</th>\n",
       "      <td>0.92</td>\n",
       "      <td>0.92</td>\n",
       "      <td>1166.32</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GaussianProcessRegressor</th>\n",
       "      <td>0.92</td>\n",
       "      <td>0.92</td>\n",
       "      <td>1172.76</td>\n",
       "      <td>7.15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KNeighborsRegressor</th>\n",
       "      <td>0.92</td>\n",
       "      <td>0.92</td>\n",
       "      <td>1174.03</td>\n",
       "      <td>0.04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AdaBoostRegressor</th>\n",
       "      <td>0.91</td>\n",
       "      <td>0.91</td>\n",
       "      <td>1204.41</td>\n",
       "      <td>0.23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DecisionTreeRegressor</th>\n",
       "      <td>0.90</td>\n",
       "      <td>0.90</td>\n",
       "      <td>1303.01</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ExtraTreeRegressor</th>\n",
       "      <td>0.89</td>\n",
       "      <td>0.89</td>\n",
       "      <td>1370.81</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ElasticNet</th>\n",
       "      <td>0.84</td>\n",
       "      <td>0.84</td>\n",
       "      <td>1633.54</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MLPRegressor</th>\n",
       "      <td>0.78</td>\n",
       "      <td>0.78</td>\n",
       "      <td>1935.15</td>\n",
       "      <td>6.73</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GeneralizedLinearRegressor</th>\n",
       "      <td>0.77</td>\n",
       "      <td>0.77</td>\n",
       "      <td>1961.87</td>\n",
       "      <td>0.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TweedieRegressor</th>\n",
       "      <td>0.77</td>\n",
       "      <td>0.77</td>\n",
       "      <td>1961.87</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PoissonRegressor</th>\n",
       "      <td>0.65</td>\n",
       "      <td>0.65</td>\n",
       "      <td>2419.96</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LinearSVR</th>\n",
       "      <td>0.65</td>\n",
       "      <td>0.65</td>\n",
       "      <td>2425.47</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ElasticNetCV</th>\n",
       "      <td>0.22</td>\n",
       "      <td>0.22</td>\n",
       "      <td>3613.48</td>\n",
       "      <td>0.08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SVR</th>\n",
       "      <td>0.21</td>\n",
       "      <td>0.21</td>\n",
       "      <td>3631.97</td>\n",
       "      <td>1.32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NuSVR</th>\n",
       "      <td>0.05</td>\n",
       "      <td>0.05</td>\n",
       "      <td>3974.40</td>\n",
       "      <td>1.04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DummyRegressor</th>\n",
       "      <td>-0.39</td>\n",
       "      <td>-0.38</td>\n",
       "      <td>4803.94</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KernelRidge</th>\n",
       "      <td>-0.58</td>\n",
       "      <td>-0.57</td>\n",
       "      <td>5122.03</td>\n",
       "      <td>5.14</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                               Adjusted R-Squared  R-Squared    RMSE  \\\n",
       "Model                                                                  \n",
       "HistGradientBoostingRegressor                0.93       0.93 1075.28   \n",
       "LGBMRegressor                                0.93       0.93 1075.40   \n",
       "XGBRegressor                                 0.93       0.93 1090.94   \n",
       "PassiveAggressiveRegressor                   0.93       0.93 1091.96   \n",
       "RANSACRegressor                              0.93       0.93 1093.64   \n",
       "HuberRegressor                               0.93       0.93 1098.57   \n",
       "RandomForestRegressor                        0.93       0.93 1108.30   \n",
       "OrthogonalMatchingPursuitCV                  0.93       0.93 1108.96   \n",
       "LinearRegression                             0.93       0.93 1113.42   \n",
       "TransformedTargetRegressor                   0.93       0.93 1113.42   \n",
       "Lars                                         0.93       0.93 1113.42   \n",
       "LarsCV                                       0.93       0.93 1113.42   \n",
       "LassoLarsCV                                  0.93       0.93 1113.42   \n",
       "LassoLarsIC                                  0.93       0.93 1113.42   \n",
       "BayesianRidge                                0.93       0.93 1113.48   \n",
       "RidgeCV                                      0.93       0.93 1113.49   \n",
       "Ridge                                        0.93       0.93 1114.12   \n",
       "Lasso                                        0.93       0.93 1114.18   \n",
       "SGDRegressor                                 0.92       0.93 1117.27   \n",
       "LassoCV                                      0.92       0.92 1118.70   \n",
       "BaggingRegressor                             0.92       0.92 1121.58   \n",
       "ExtraTreesRegressor                          0.92       0.92 1127.87   \n",
       "GradientBoostingRegressor                    0.92       0.92 1144.04   \n",
       "OrthogonalMatchingPursuit                    0.92       0.92 1159.02   \n",
       "LassoLars                                    0.92       0.92 1166.32   \n",
       "GaussianProcessRegressor                     0.92       0.92 1172.76   \n",
       "KNeighborsRegressor                          0.92       0.92 1174.03   \n",
       "AdaBoostRegressor                            0.91       0.91 1204.41   \n",
       "DecisionTreeRegressor                        0.90       0.90 1303.01   \n",
       "ExtraTreeRegressor                           0.89       0.89 1370.81   \n",
       "ElasticNet                                   0.84       0.84 1633.54   \n",
       "MLPRegressor                                 0.78       0.78 1935.15   \n",
       "GeneralizedLinearRegressor                   0.77       0.77 1961.87   \n",
       "TweedieRegressor                             0.77       0.77 1961.87   \n",
       "PoissonRegressor                             0.65       0.65 2419.96   \n",
       "LinearSVR                                    0.65       0.65 2425.47   \n",
       "ElasticNetCV                                 0.22       0.22 3613.48   \n",
       "SVR                                          0.21       0.21 3631.97   \n",
       "NuSVR                                        0.05       0.05 3974.40   \n",
       "DummyRegressor                              -0.39      -0.38 4803.94   \n",
       "KernelRidge                                 -0.58      -0.57 5122.03   \n",
       "\n",
       "                               Time Taken  \n",
       "Model                                      \n",
       "HistGradientBoostingRegressor        0.86  \n",
       "LGBMRegressor                        0.15  \n",
       "XGBRegressor                         1.76  \n",
       "PassiveAggressiveRegressor           0.02  \n",
       "RANSACRegressor                      0.03  \n",
       "HuberRegressor                       0.06  \n",
       "RandomForestRegressor                1.03  \n",
       "OrthogonalMatchingPursuitCV          0.03  \n",
       "LinearRegression                     0.01  \n",
       "TransformedTargetRegressor           0.01  \n",
       "Lars                                 0.02  \n",
       "LarsCV                               0.02  \n",
       "LassoLarsCV                          0.02  \n",
       "LassoLarsIC                          0.01  \n",
       "BayesianRidge                        0.02  \n",
       "RidgeCV                              0.01  \n",
       "Ridge                                0.01  \n",
       "Lasso                                0.01  \n",
       "SGDRegressor                         0.02  \n",
       "LassoCV                              0.07  \n",
       "BaggingRegressor                     0.11  \n",
       "ExtraTreesRegressor                  0.60  \n",
       "GradientBoostingRegressor            0.44  \n",
       "OrthogonalMatchingPursuit            0.01  \n",
       "LassoLars                            0.01  \n",
       "GaussianProcessRegressor             7.15  \n",
       "KNeighborsRegressor                  0.04  \n",
       "AdaBoostRegressor                    0.23  \n",
       "DecisionTreeRegressor                0.02  \n",
       "ExtraTreeRegressor                   0.02  \n",
       "ElasticNet                           0.01  \n",
       "MLPRegressor                         6.73  \n",
       "GeneralizedLinearRegressor           0.03  \n",
       "TweedieRegressor                     0.01  \n",
       "PoissonRegressor                     0.02  \n",
       "LinearSVR                            0.01  \n",
       "ElasticNetCV                         0.08  \n",
       "SVR                                  1.32  \n",
       "NuSVR                                1.04  \n",
       "DummyRegressor                       0.01  \n",
       "KernelRidge                          5.14  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "solar_reg = LazyRegressor(verbose=0,ignore_warnings=True)\n",
    "models, predictions = solar_reg.fit(X_solar_train, X_solar_test, y_solar_train, y_solar_test)\n",
    "models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Applying LazyRegressor for Wind Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████████████████████████████████████████████| 42/42 [00:21<00:00,  1.95it/s]\n"
     ]
    },
    {
     "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>Adjusted R-Squared</th>\n",
       "      <th>R-Squared</th>\n",
       "      <th>RMSE</th>\n",
       "      <th>Time Taken</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Model</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>GradientBoostingRegressor</th>\n",
       "      <td>0.90</td>\n",
       "      <td>0.90</td>\n",
       "      <td>2383.22</td>\n",
       "      <td>0.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ExtraTreesRegressor</th>\n",
       "      <td>0.90</td>\n",
       "      <td>0.90</td>\n",
       "      <td>2393.15</td>\n",
       "      <td>0.90</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HistGradientBoostingRegressor</th>\n",
       "      <td>0.90</td>\n",
       "      <td>0.90</td>\n",
       "      <td>2406.22</td>\n",
       "      <td>0.80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RandomForestRegressor</th>\n",
       "      <td>0.90</td>\n",
       "      <td>0.90</td>\n",
       "      <td>2429.91</td>\n",
       "      <td>1.61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BaggingRegressor</th>\n",
       "      <td>0.89</td>\n",
       "      <td>0.89</td>\n",
       "      <td>2449.92</td>\n",
       "      <td>0.17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LGBMRegressor</th>\n",
       "      <td>0.89</td>\n",
       "      <td>0.89</td>\n",
       "      <td>2460.63</td>\n",
       "      <td>0.12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KNeighborsRegressor</th>\n",
       "      <td>0.88</td>\n",
       "      <td>0.88</td>\n",
       "      <td>2647.33</td>\n",
       "      <td>0.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SGDRegressor</th>\n",
       "      <td>0.88</td>\n",
       "      <td>0.88</td>\n",
       "      <td>2650.90</td>\n",
       "      <td>0.06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LassoCV</th>\n",
       "      <td>0.88</td>\n",
       "      <td>0.88</td>\n",
       "      <td>2658.27</td>\n",
       "      <td>0.09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AdaBoostRegressor</th>\n",
       "      <td>0.87</td>\n",
       "      <td>0.87</td>\n",
       "      <td>2669.60</td>\n",
       "      <td>0.29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Lasso</th>\n",
       "      <td>0.87</td>\n",
       "      <td>0.87</td>\n",
       "      <td>2670.75</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RANSACRegressor</th>\n",
       "      <td>0.87</td>\n",
       "      <td>0.87</td>\n",
       "      <td>2675.82</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ridge</th>\n",
       "      <td>0.87</td>\n",
       "      <td>0.87</td>\n",
       "      <td>2676.73</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OrthogonalMatchingPursuit</th>\n",
       "      <td>0.87</td>\n",
       "      <td>0.87</td>\n",
       "      <td>2683.40</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RidgeCV</th>\n",
       "      <td>0.87</td>\n",
       "      <td>0.87</td>\n",
       "      <td>2684.49</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BayesianRidge</th>\n",
       "      <td>0.87</td>\n",
       "      <td>0.87</td>\n",
       "      <td>2685.00</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LinearRegression</th>\n",
       "      <td>0.87</td>\n",
       "      <td>0.87</td>\n",
       "      <td>2685.61</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TransformedTargetRegressor</th>\n",
       "      <td>0.87</td>\n",
       "      <td>0.87</td>\n",
       "      <td>2685.61</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LassoLarsIC</th>\n",
       "      <td>0.87</td>\n",
       "      <td>0.87</td>\n",
       "      <td>2685.61</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OrthogonalMatchingPursuitCV</th>\n",
       "      <td>0.87</td>\n",
       "      <td>0.87</td>\n",
       "      <td>2685.61</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LassoLarsCV</th>\n",
       "      <td>0.87</td>\n",
       "      <td>0.87</td>\n",
       "      <td>2685.61</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PassiveAggressiveRegressor</th>\n",
       "      <td>0.87</td>\n",
       "      <td>0.87</td>\n",
       "      <td>2695.39</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>XGBRegressor</th>\n",
       "      <td>0.87</td>\n",
       "      <td>0.87</td>\n",
       "      <td>2696.26</td>\n",
       "      <td>0.27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LassoLars</th>\n",
       "      <td>0.87</td>\n",
       "      <td>0.87</td>\n",
       "      <td>2699.14</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HuberRegressor</th>\n",
       "      <td>0.87</td>\n",
       "      <td>0.87</td>\n",
       "      <td>2721.23</td>\n",
       "      <td>0.06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DecisionTreeRegressor</th>\n",
       "      <td>0.86</td>\n",
       "      <td>0.86</td>\n",
       "      <td>2813.82</td>\n",
       "      <td>0.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LarsCV</th>\n",
       "      <td>0.86</td>\n",
       "      <td>0.86</td>\n",
       "      <td>2815.02</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ExtraTreeRegressor</th>\n",
       "      <td>0.85</td>\n",
       "      <td>0.85</td>\n",
       "      <td>2874.03</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ElasticNet</th>\n",
       "      <td>0.79</td>\n",
       "      <td>0.79</td>\n",
       "      <td>3458.87</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PoissonRegressor</th>\n",
       "      <td>0.79</td>\n",
       "      <td>0.79</td>\n",
       "      <td>3480.81</td>\n",
       "      <td>0.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GeneralizedLinearRegressor</th>\n",
       "      <td>0.73</td>\n",
       "      <td>0.73</td>\n",
       "      <td>3936.41</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TweedieRegressor</th>\n",
       "      <td>0.73</td>\n",
       "      <td>0.73</td>\n",
       "      <td>3936.41</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GammaRegressor</th>\n",
       "      <td>0.68</td>\n",
       "      <td>0.68</td>\n",
       "      <td>4266.61</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MLPRegressor</th>\n",
       "      <td>0.49</td>\n",
       "      <td>0.49</td>\n",
       "      <td>5390.12</td>\n",
       "      <td>7.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ElasticNetCV</th>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>5954.86</td>\n",
       "      <td>0.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LinearSVR</th>\n",
       "      <td>0.10</td>\n",
       "      <td>0.10</td>\n",
       "      <td>7136.80</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NuSVR</th>\n",
       "      <td>0.03</td>\n",
       "      <td>0.03</td>\n",
       "      <td>7431.65</td>\n",
       "      <td>0.81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SVR</th>\n",
       "      <td>0.01</td>\n",
       "      <td>0.01</td>\n",
       "      <td>7511.57</td>\n",
       "      <td>1.23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DummyRegressor</th>\n",
       "      <td>-0.03</td>\n",
       "      <td>-0.03</td>\n",
       "      <td>7648.35</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KernelRidge</th>\n",
       "      <td>-0.72</td>\n",
       "      <td>-0.72</td>\n",
       "      <td>9881.43</td>\n",
       "      <td>2.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Lars</th>\n",
       "      <td>-158.73</td>\n",
       "      <td>-158.48</td>\n",
       "      <td>95226.44</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GaussianProcessRegressor</th>\n",
       "      <td>-21616.05</td>\n",
       "      <td>-21582.95</td>\n",
       "      <td>1107813.71</td>\n",
       "      <td>4.67</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                               Adjusted R-Squared  R-Squared       RMSE  \\\n",
       "Model                                                                     \n",
       "GradientBoostingRegressor                    0.90       0.90    2383.22   \n",
       "ExtraTreesRegressor                          0.90       0.90    2393.15   \n",
       "HistGradientBoostingRegressor                0.90       0.90    2406.22   \n",
       "RandomForestRegressor                        0.90       0.90    2429.91   \n",
       "BaggingRegressor                             0.89       0.89    2449.92   \n",
       "LGBMRegressor                                0.89       0.89    2460.63   \n",
       "KNeighborsRegressor                          0.88       0.88    2647.33   \n",
       "SGDRegressor                                 0.88       0.88    2650.90   \n",
       "LassoCV                                      0.88       0.88    2658.27   \n",
       "AdaBoostRegressor                            0.87       0.87    2669.60   \n",
       "Lasso                                        0.87       0.87    2670.75   \n",
       "RANSACRegressor                              0.87       0.87    2675.82   \n",
       "Ridge                                        0.87       0.87    2676.73   \n",
       "OrthogonalMatchingPursuit                    0.87       0.87    2683.40   \n",
       "RidgeCV                                      0.87       0.87    2684.49   \n",
       "BayesianRidge                                0.87       0.87    2685.00   \n",
       "LinearRegression                             0.87       0.87    2685.61   \n",
       "TransformedTargetRegressor                   0.87       0.87    2685.61   \n",
       "LassoLarsIC                                  0.87       0.87    2685.61   \n",
       "OrthogonalMatchingPursuitCV                  0.87       0.87    2685.61   \n",
       "LassoLarsCV                                  0.87       0.87    2685.61   \n",
       "PassiveAggressiveRegressor                   0.87       0.87    2695.39   \n",
       "XGBRegressor                                 0.87       0.87    2696.26   \n",
       "LassoLars                                    0.87       0.87    2699.14   \n",
       "HuberRegressor                               0.87       0.87    2721.23   \n",
       "DecisionTreeRegressor                        0.86       0.86    2813.82   \n",
       "LarsCV                                       0.86       0.86    2815.02   \n",
       "ExtraTreeRegressor                           0.85       0.85    2874.03   \n",
       "ElasticNet                                   0.79       0.79    3458.87   \n",
       "PoissonRegressor                             0.79       0.79    3480.81   \n",
       "GeneralizedLinearRegressor                   0.73       0.73    3936.41   \n",
       "TweedieRegressor                             0.73       0.73    3936.41   \n",
       "GammaRegressor                               0.68       0.68    4266.61   \n",
       "MLPRegressor                                 0.49       0.49    5390.12   \n",
       "ElasticNetCV                                 0.38       0.38    5954.86   \n",
       "LinearSVR                                    0.10       0.10    7136.80   \n",
       "NuSVR                                        0.03       0.03    7431.65   \n",
       "SVR                                          0.01       0.01    7511.57   \n",
       "DummyRegressor                              -0.03      -0.03    7648.35   \n",
       "KernelRidge                                 -0.72      -0.72    9881.43   \n",
       "Lars                                      -158.73    -158.48   95226.44   \n",
       "GaussianProcessRegressor                -21616.05  -21582.95 1107813.71   \n",
       "\n",
       "                               Time Taken  \n",
       "Model                                      \n",
       "GradientBoostingRegressor            0.64  \n",
       "ExtraTreesRegressor                  0.90  \n",
       "HistGradientBoostingRegressor        0.80  \n",
       "RandomForestRegressor                1.61  \n",
       "BaggingRegressor                     0.17  \n",
       "LGBMRegressor                        0.12  \n",
       "KNeighborsRegressor                  0.05  \n",
       "SGDRegressor                         0.06  \n",
       "LassoCV                              0.09  \n",
       "AdaBoostRegressor                    0.29  \n",
       "Lasso                                0.02  \n",
       "RANSACRegressor                      0.02  \n",
       "Ridge                                0.01  \n",
       "OrthogonalMatchingPursuit            0.01  \n",
       "RidgeCV                              0.01  \n",
       "BayesianRidge                        0.01  \n",
       "LinearRegression                     0.01  \n",
       "TransformedTargetRegressor           0.01  \n",
       "LassoLarsIC                          0.01  \n",
       "OrthogonalMatchingPursuitCV          0.02  \n",
       "LassoLarsCV                          0.02  \n",
       "PassiveAggressiveRegressor           0.02  \n",
       "XGBRegressor                         0.27  \n",
       "LassoLars                            0.01  \n",
       "HuberRegressor                       0.06  \n",
       "DecisionTreeRegressor                0.03  \n",
       "LarsCV                               0.02  \n",
       "ExtraTreeRegressor                   0.02  \n",
       "ElasticNet                           0.01  \n",
       "PoissonRegressor                     0.03  \n",
       "GeneralizedLinearRegressor           0.02  \n",
       "TweedieRegressor                     0.01  \n",
       "GammaRegressor                       0.02  \n",
       "MLPRegressor                         7.14  \n",
       "ElasticNetCV                         0.07  \n",
       "LinearSVR                            0.01  \n",
       "NuSVR                                0.81  \n",
       "SVR                                  1.23  \n",
       "DummyRegressor                       0.01  \n",
       "KernelRidge                          2.20  \n",
       "Lars                                 0.02  \n",
       "GaussianProcessRegressor             4.67  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wind_reg = LazyRegressor(verbose=0,ignore_warnings=True)\n",
    "models, predictions = wind_reg.fit(X_wind_train, X_wind_test, y_wind_train, y_wind_test)\n",
    "models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### End Results\n",
    "- For solar dataset **HistGradientBoostingRegressor** and\n",
    "- For wind dataset **GradientBoostingRegressor** works the best"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Cross-Validating the findings with the scores of Linear Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.experimental import enable_hist_gradient_boosting\n",
    "from sklearn.ensemble import HistGradientBoostingRegressor\n",
    "hgbr = HistGradientBoostingRegressor()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.85807896 0.94661228 0.94447233 0.95746803 0.88008925] \n",
      "average = 0.9173441664489002\n"
     ]
    }
   ],
   "source": [
    "X_solar = thesis_dataset[['SWTDN', 'SWGDN', 'T']]\n",
    "y_solar = thesis_dataset['DE_solar_generation_actual']\n",
    "\n",
    "scores_solar = cross_val_score(hgbr, X_solar, y_solar, cv=5)\n",
    "print(scores_solar, \"\\naverage =\", np.mean(scores_solar))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plotting Actual vs Predictions on same plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "hgbr = HistGradientBoostingRegressor().fit(X_solar_train, y_solar_train)\n",
    "\n",
    "train_solar_pred = hgbr.predict(X_solar_train)\n",
    "test_solar_pred = hgbr.predict(X_solar_test)\n",
    "\n",
    "# thesis_dataset = thesis_dataset.reset_index()\n",
    "df = thesis_dataset[['DE_solar_generation_actual']]\n",
    "\n",
    "train_df = train[['DE_solar_generation_actual']]\n",
    "train_df.loc[:, 'DE_solar_generation_actual'] = train_solar_pred\n",
    "\n",
    "test_df = test[['DE_solar_generation_actual']]\n",
    "test_df.loc[:, 'DE_solar_generation_actual'] = test_solar_pred\n",
    "\n",
    "# # Plot all predictions\n",
    "inversetransform, =plt.plot(df, label = 'Actual')\n",
    "trainpredx, =plt.plot(train_df, color='orange', label = 'Train')\n",
    "testpredx, =plt.plot(test_df, color='green', label = 'Test')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Solar Output')\n",
    "plt.legend(loc=\"upper right\")\n",
    "plt.title(\"Predicted vs. Actual Solar Generation - HistGradBoosting\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Regression Metrics for Solar"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>MAE</th>\n",
       "      <th>MSE</th>\n",
       "      <th>RMSE</th>\n",
       "      <th>RMSLE</th>\n",
       "      <th>R2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>model</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>histgrad</th>\n",
       "      <td>465.13</td>\n",
       "      <td>1075.28</td>\n",
       "      <td>32.79</td>\n",
       "      <td>3.49</td>\n",
       "      <td>0.93</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            MAE     MSE  RMSE  RMSLE   R2\n",
       "model                                    \n",
       "histgrad 465.13 1075.28 32.79   3.49 0.93"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics('histgrad', test['DE_solar_generation_actual'], test_solar_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Cross-validating the findings over Wind Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "gbr = GradientBoostingRegressor()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.89866486 0.8504175  0.8787059  0.92984511 0.88913014] \n",
      "average = 0.8893527020571004\n"
     ]
    }
   ],
   "source": [
    "X_wind = thesis_dataset[['v1', 'v2', 'v_50m', 'z0']]\n",
    "y_wind = thesis_dataset['DE_wind_generation_actual']\n",
    "\n",
    "scores_wind = cross_val_score(gbr, X_wind, y_wind, cv=5)\n",
    "print(scores_wind, \"\\naverage =\", np.mean(scores_wind))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plotting Actual vs Predictions on same plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "gbr = GradientBoostingRegressor().fit(X_wind_train, y_wind_train)\n",
    "\n",
    "train_wind_pred = gbr.predict(X_wind_train)\n",
    "test_wind_pred = gbr.predict(X_wind_test)\n",
    "\n",
    "df = thesis_dataset[['DE_wind_generation_actual']]\n",
    "\n",
    "train_df = train[['DE_wind_generation_actual']]\n",
    "train_df.loc[:, 'DE_wind_generation_actual'] = train_wind_pred\n",
    "\n",
    "test_df = test[['DE_wind_generation_actual']]\n",
    "test_df.loc[:, 'DE_wind_generation_actual'] = test_wind_pred\n",
    "\n",
    "# Plot all predictions\n",
    "inversetransform, =plt.plot(df, label = 'Actual')\n",
    "trainpredx, =plt.plot(train_df, color='orange', label = 'Train')\n",
    "testpredx, =plt.plot(test_df, color='green', label = 'Test')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('wind Output')\n",
    "plt.legend(loc=\"upper right\")\n",
    "plt.title(\"Predicted vs. Actual wind Generation - GradBoosting\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Regression Metrics for Wind"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>MAE</th>\n",
       "      <th>MSE</th>\n",
       "      <th>RMSE</th>\n",
       "      <th>RMSLE</th>\n",
       "      <th>R2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>model</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>grad</th>\n",
       "      <td>1684.61</td>\n",
       "      <td>2373.77</td>\n",
       "      <td>48.72</td>\n",
       "      <td>3.89</td>\n",
       "      <td>0.90</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          MAE     MSE  RMSE  RMSLE   R2\n",
       "model                                  \n",
       "grad  1684.61 2373.77 48.72   3.89 0.90"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics('grad', test['DE_wind_generation_actual'], test_wind_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the above results we can see that for solar dataset there hasn't be much difference **(92.4% -> 91.7%)** accuracy.\n",
    "\n",
    "But for Wind dataset there has being the accuracy jump from **(85.8% -> 88.9%)** accuracy."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conclusion\n",
    "- We have seen that due to variating nature of the way Wind and Solar produce their energy. These cannot be captured in a single model.\n",
    "- Different models for solar and wind are required.\n",
    "- So a combined model that captures the output their predictions seperately and combines them, could result in better accuracy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.9475020609796467, 0.9405725476558187, 0.9394510510620339, 0.9452853795411004, 0.9366545872702404] \n",
      "average = 0.941893125301768\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import KFold \n",
    "from sklearn.metrics import r2_score\n",
    "kf = KFold(5, True, 1)\n",
    "\n",
    "scores = []\n",
    "for train_index, test_index in kf.split(thesis_dataset):\n",
    "    X_solar = thesis_dataset.iloc[train_index, :][['SWTDN', 'SWGDN', 'T']]\n",
    "    y_solar = thesis_dataset.iloc[train_index, :]['DE_solar_generation_actual']\n",
    "    hgbr = HistGradientBoostingRegressor().fit(X_solar, y_solar)\n",
    "\n",
    "    X_wind = thesis_dataset.iloc[train_index, :][['v1', 'v2', 'v_50m', 'z0']]\n",
    "    y_wind = thesis_dataset.iloc[train_index, :]['DE_wind_generation_actual']\n",
    "    gbr = GradientBoostingRegressor().fit(X_wind, y_wind)\n",
    "    \n",
    "    scores.append(r2_score(\n",
    "        hgbr.predict(thesis_dataset.iloc[test_index, :][['SWTDN', 'SWGDN', 'T']]) + gbr.predict(thesis_dataset.iloc[test_index, :][['v1', 'v2', 'v_50m', 'z0']]),\n",
    "        thesis_dataset.iloc[test_index, :]['DE_solar_generation_actual'] + thesis_dataset.iloc[test_index, :]['DE_wind_generation_actual']))                \n",
    "\n",
    "print(scores, \"\\naverage =\", np.mean(scores))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Creating a Custom Regressor from the findings and plotting their predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "class CustomRegressor:\n",
    "    def __init__(self):\n",
    "        self.solarRegressor = None\n",
    "        self.windRegressor = None\n",
    "        \n",
    "    def fit(self, train_dataset):\n",
    "        X_solar = train_dataset[['SWTDN', 'SWGDN', 'T']]\n",
    "        y_solar = train_dataset['DE_solar_generation_actual']\n",
    "        self.solarRegressor = HistGradientBoostingRegressor().fit(X_solar, y_solar)\n",
    "\n",
    "        X_wind = train_dataset[['v1', 'v2', 'v_50m', 'z0']]\n",
    "        y_wind = train_dataset['DE_wind_generation_actual']\n",
    "        self.windRegressor = GradientBoostingRegressor().fit(X_wind, y_wind)\n",
    "        return self\n",
    "        \n",
    "    def predict(self, prediction_dataset):\n",
    "        return self.solarRegressor.predict(prediction_dataset[['SWTDN', 'SWGDN', 'T']]) + self.windRegressor.predict(prediction_dataset[['v1', 'v2', 'v_50m', 'z0']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cr = CustomRegressor().fit(train)\n",
    "\n",
    "train_total_pred = cr.predict(train)\n",
    "test_total_pred = cr.predict(test)\n",
    "\n",
    "df = thesis_dataset[['DE_solar_generation_actual']]\n",
    "df['Total Generation'] = thesis_dataset['DE_solar_generation_actual'] + thesis_dataset['DE_wind_generation_actual']\n",
    "df.drop(['DE_solar_generation_actual'], axis = 1, inplace = True)\n",
    "\n",
    "train_df = train[['DE_solar_generation_actual']]\n",
    "train_df['Total Generation'] = train_total_pred\n",
    "train_df.drop(['DE_solar_generation_actual'], axis = 1, inplace = True)\n",
    "\n",
    "test_df = test[['DE_solar_generation_actual']]\n",
    "test_df['Total Generation'] = test_total_pred\n",
    "test_df.drop(['DE_solar_generation_actual'], axis = 1, inplace = True)\n",
    "\n",
    "# Plot all predictions\n",
    "inversetransform, =plt.plot(df, label = 'Actual')\n",
    "trainpredx, =plt.plot(train_df, color='orange', label = 'Train')\n",
    "testpredx, =plt.plot(test_df, color='green', label = 'Test')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Combined Output')\n",
    "plt.legend(loc=\"upper right\")\n",
    "plt.title(\"Predicted vs. Actual wind + solar Generation - histgrad+grad\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>MAE</th>\n",
       "      <th>MSE</th>\n",
       "      <th>RMSE</th>\n",
       "      <th>RMSLE</th>\n",
       "      <th>R2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>model</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>histgrad+grad</th>\n",
       "      <td>1980.47</td>\n",
       "      <td>2721.37</td>\n",
       "      <td>52.17</td>\n",
       "      <td>3.95</td>\n",
       "      <td>0.89</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  MAE     MSE  RMSE  RMSLE   R2\n",
       "model                                          \n",
       "histgrad+grad 1980.47 2721.37 52.17   3.95 0.89"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics('histgrad+grad', test['DE_solar_generation_actual'] + test['DE_wind_generation_actual'], test_total_pred)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}