ML_Models_2_Phase.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Fire Size Predictors\n",
    "Instead of straight out predicting fire size, this approach will be to first use a classifier to predict weather the fire will be small, medium or large. Then based on that prediction, use the corresponding regression model to predict the exact size."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from helper_functions import evaluate_model\n",
    "import pickle\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler, OneHotEncoder, LabelEncoder, FunctionTransformer, PowerTransformer, MinMaxScaler\n",
    "from sklearn.pipeline import Pipeline, make_pipeline\n",
    "from sklearn.compose import ColumnTransformer, make_column_selector\n",
    "from sklearn.metrics import mean_squared_error, r2_score, f1_score, confusion_matrix\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.ensemble import AdaBoostRegressor, RandomForestClassifier, RandomForestRegressor\n",
    "from sklearn.linear_model import SGDClassifier, HuberRegressor, ElasticNetCV\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from imblearn.combine import SMOTEENN\n",
    "from imblearn.pipeline import Pipeline as imbPipeline\n",
    "import xgboost as xgb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(419872, 23)\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>FID</th>\n",
       "      <th>Src_Agency</th>\n",
       "      <th>Latitude</th>\n",
       "      <th>Longitude</th>\n",
       "      <th>Rep_Date</th>\n",
       "      <th>Size_Ha</th>\n",
       "      <th>Cause</th>\n",
       "      <th>Protzone</th>\n",
       "      <th>Ecoz_Name</th>\n",
       "      <th>Year</th>\n",
       "      <th>...</th>\n",
       "      <th>Province</th>\n",
       "      <th>Size_Bin</th>\n",
       "      <th>Decade</th>\n",
       "      <th>Elevation</th>\n",
       "      <th>2_Week_Rainfall</th>\n",
       "      <th>Avg_High</th>\n",
       "      <th>Start_Date_High</th>\n",
       "      <th>Avg_Temp</th>\n",
       "      <th>Start_Date_Wind_Speed</th>\n",
       "      <th>Wind_Direction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>BC</td>\n",
       "      <td>59.963</td>\n",
       "      <td>-128.172</td>\n",
       "      <td>1953-05-26</td>\n",
       "      <td>8.0</td>\n",
       "      <td>H</td>\n",
       "      <td></td>\n",
       "      <td>Boreal Cordillera</td>\n",
       "      <td>1953</td>\n",
       "      <td>...</td>\n",
       "      <td>BC</td>\n",
       "      <td>small</td>\n",
       "      <td>50s</td>\n",
       "      <td>835.0</td>\n",
       "      <td>21.9</td>\n",
       "      <td>12.400000</td>\n",
       "      <td>15.1</td>\n",
       "      <td>7.126667</td>\n",
       "      <td>11.9</td>\n",
       "      <td>200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>BC</td>\n",
       "      <td>59.318</td>\n",
       "      <td>-132.172</td>\n",
       "      <td>1950-06-22</td>\n",
       "      <td>8.0</td>\n",
       "      <td>L</td>\n",
       "      <td></td>\n",
       "      <td>Boreal Cordillera</td>\n",
       "      <td>1950</td>\n",
       "      <td>...</td>\n",
       "      <td>BC</td>\n",
       "      <td>small</td>\n",
       "      <td>40s</td>\n",
       "      <td>835.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>19.766667</td>\n",
       "      <td>17.0</td>\n",
       "      <td>13.660000</td>\n",
       "      <td>10.5</td>\n",
       "      <td>342</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>BC</td>\n",
       "      <td>59.876</td>\n",
       "      <td>-131.922</td>\n",
       "      <td>1950-06-04</td>\n",
       "      <td>12949.9</td>\n",
       "      <td>H</td>\n",
       "      <td></td>\n",
       "      <td>Boreal Cordillera</td>\n",
       "      <td>1950</td>\n",
       "      <td>...</td>\n",
       "      <td>BC</td>\n",
       "      <td>large</td>\n",
       "      <td>40s</td>\n",
       "      <td>853.0</td>\n",
       "      <td>11.2</td>\n",
       "      <td>9.146667</td>\n",
       "      <td>14.3</td>\n",
       "      <td>4.433333</td>\n",
       "      <td>12.9</td>\n",
       "      <td>310</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>BC</td>\n",
       "      <td>59.760</td>\n",
       "      <td>-132.808</td>\n",
       "      <td>1951-07-15</td>\n",
       "      <td>241.1</td>\n",
       "      <td>H</td>\n",
       "      <td></td>\n",
       "      <td>Boreal Cordillera</td>\n",
       "      <td>1951</td>\n",
       "      <td>...</td>\n",
       "      <td>BC</td>\n",
       "      <td>medium</td>\n",
       "      <td>50s</td>\n",
       "      <td>952.0</td>\n",
       "      <td>32.0</td>\n",
       "      <td>21.940000</td>\n",
       "      <td>21.9</td>\n",
       "      <td>17.086667</td>\n",
       "      <td>5.2</td>\n",
       "      <td>343</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>BC</td>\n",
       "      <td>59.434</td>\n",
       "      <td>-126.172</td>\n",
       "      <td>1952-06-12</td>\n",
       "      <td>1.2</td>\n",
       "      <td>H</td>\n",
       "      <td></td>\n",
       "      <td>Boreal Cordillera</td>\n",
       "      <td>1952</td>\n",
       "      <td>...</td>\n",
       "      <td>BC</td>\n",
       "      <td>small</td>\n",
       "      <td>50s</td>\n",
       "      <td>438.0</td>\n",
       "      <td>24.8</td>\n",
       "      <td>14.193333</td>\n",
       "      <td>21.2</td>\n",
       "      <td>9.986667</td>\n",
       "      <td>10.1</td>\n",
       "      <td>105</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   FID Src_Agency  Latitude  Longitude    Rep_Date  Size_Ha Cause Protzone  \\\n",
       "0    0         BC    59.963   -128.172  1953-05-26      8.0     H            \n",
       "1    1         BC    59.318   -132.172  1950-06-22      8.0     L            \n",
       "2    2         BC    59.876   -131.922  1950-06-04  12949.9     H            \n",
       "3    3         BC    59.760   -132.808  1951-07-15    241.1     H            \n",
       "4    4         BC    59.434   -126.172  1952-06-12      1.2     H            \n",
       "\n",
       "           Ecoz_Name  Year  ...  Province  Size_Bin  Decade Elevation  \\\n",
       "0  Boreal Cordillera  1953  ...        BC     small     50s     835.0   \n",
       "1  Boreal Cordillera  1950  ...        BC     small     40s     835.0   \n",
       "2  Boreal Cordillera  1950  ...        BC     large     40s     853.0   \n",
       "3  Boreal Cordillera  1951  ...        BC    medium     50s     952.0   \n",
       "4  Boreal Cordillera  1952  ...        BC     small     50s     438.0   \n",
       "\n",
       "  2_Week_Rainfall   Avg_High  Start_Date_High   Avg_Temp  \\\n",
       "0            21.9  12.400000             15.1   7.126667   \n",
       "1            13.0  19.766667             17.0  13.660000   \n",
       "2            11.2   9.146667             14.3   4.433333   \n",
       "3            32.0  21.940000             21.9  17.086667   \n",
       "4            24.8  14.193333             21.2   9.986667   \n",
       "\n",
       "   Start_Date_Wind_Speed  Wind_Direction  \n",
       "0                   11.9             200  \n",
       "1                   10.5             342  \n",
       "2                   12.9             310  \n",
       "3                    5.2             343  \n",
       "4                   10.1             105  \n",
       "\n",
       "[5 rows x 23 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# import fire and weather data and combine into single dataframe\n",
    "fire_df = pd.read_csv('data/wildfires_clean.csv')\n",
    "weather_df = pd.read_csv('data/clean_weather.csv')\n",
    "df = pd.merge(fire_df, weather_df, on='FID')\n",
    "print(df.shape)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "features = ['Latitude', 'Longitude', 'Year', 'Month', 'Date', 'Cause', 'Elevation', '2_Week_Rainfall', 'Avg_High', 'Start_Date_High', 'Avg_Temp', 'Start_Date_Wind_Speed', 'Wind_Direction', 'Size_Ha', 'Size_Bin']\n",
    "X = df[features]"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Stage 1: Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "small     0.902980\n",
       "medium    0.087034\n",
       "large     0.009986\n",
       "Name: Size_Bin, dtype: float64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check distribution of size bins\n",
    "df['Size_Bin'].value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create training and test data for classifier\n",
    "X_train, X_test, y_train, y_test = train_test_split(X.drop(columns=['Size_Ha', 'Size_Bin']), X['Size_Bin'], test_size=.3, random_state=58, stratify=X['Size_Bin'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "sgd_pipe = imbPipeline([\n",
    "    # Scaling, Normalizng, Encoding\n",
    "    ('preprocessing', ColumnTransformer([\n",
    "        ('numeric', make_pipeline(\n",
    "            PowerTransformer(),\n",
    "            StandardScaler()),\n",
    "            make_column_selector(dtype_include=np.number)\n",
    "        ),\n",
    "         ('categorical', make_pipeline(\n",
    "             OneHotEncoder()),\n",
    "             make_column_selector(dtype_include='object')\n",
    "         )])\n",
    "    ),\n",
    "    # Sampling to address class imbalance\n",
    "    ('smote_enn', SMOTEENN()),\n",
    "\n",
    "    # classifier\n",
    "    ('model', SGDClassifier())\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;background-color: white;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Pipeline(steps=[(&#x27;preprocessing&#x27;,\n",
       "                 ColumnTransformer(transformers=[(&#x27;numeric&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;powertransformer&#x27;,\n",
       "                                                                   PowerTransformer()),\n",
       "                                                                  (&#x27;standardscaler&#x27;,\n",
       "                                                                   StandardScaler())]),\n",
       "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8FA8D730&gt;),\n",
       "                                                 (&#x27;categorical&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;onehotencoder&#x27;,\n",
       "                                                                   OneHotEncoder())]),\n",
       "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8FA8D0A0&gt;)])),\n",
       "                (&#x27;smote_enn&#x27;, SMOTEENN()), (&#x27;model&#x27;, SGDClassifier())])</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" ><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;preprocessing&#x27;,\n",
       "                 ColumnTransformer(transformers=[(&#x27;numeric&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;powertransformer&#x27;,\n",
       "                                                                   PowerTransformer()),\n",
       "                                                                  (&#x27;standardscaler&#x27;,\n",
       "                                                                   StandardScaler())]),\n",
       "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8FA8D730&gt;),\n",
       "                                                 (&#x27;categorical&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;onehotencoder&#x27;,\n",
       "                                                                   OneHotEncoder())]),\n",
       "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8FA8D0A0&gt;)])),\n",
       "                (&#x27;smote_enn&#x27;, SMOTEENN()), (&#x27;model&#x27;, SGDClassifier())])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" ><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">preprocessing: ColumnTransformer</label><div class=\"sk-toggleable__content\"><pre>ColumnTransformer(transformers=[(&#x27;numeric&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;powertransformer&#x27;,\n",
       "                                                  PowerTransformer()),\n",
       "                                                 (&#x27;standardscaler&#x27;,\n",
       "                                                  StandardScaler())]),\n",
       "                                 &lt;sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8FA8D730&gt;),\n",
       "                                (&#x27;categorical&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;onehotencoder&#x27;,\n",
       "                                                  OneHotEncoder())]),\n",
       "                                 &lt;sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8FA8D0A0&gt;)])</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-3\" type=\"checkbox\" ><label for=\"sk-estimator-id-3\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">numeric</label><div class=\"sk-toggleable__content\"><pre>&lt;sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8FA8D730&gt;</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-4\" type=\"checkbox\" ><label for=\"sk-estimator-id-4\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">PowerTransformer</label><div class=\"sk-toggleable__content\"><pre>PowerTransformer()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-5\" type=\"checkbox\" ><label for=\"sk-estimator-id-5\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">StandardScaler</label><div class=\"sk-toggleable__content\"><pre>StandardScaler()</pre></div></div></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-6\" type=\"checkbox\" ><label for=\"sk-estimator-id-6\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">categorical</label><div class=\"sk-toggleable__content\"><pre>&lt;sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8FA8D0A0&gt;</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-7\" type=\"checkbox\" ><label for=\"sk-estimator-id-7\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">OneHotEncoder</label><div class=\"sk-toggleable__content\"><pre>OneHotEncoder()</pre></div></div></div></div></div></div></div></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-8\" type=\"checkbox\" ><label for=\"sk-estimator-id-8\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SMOTEENN</label><div class=\"sk-toggleable__content\"><pre>SMOTEENN()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-9\" type=\"checkbox\" ><label for=\"sk-estimator-id-9\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SGDClassifier</label><div class=\"sk-toggleable__content\"><pre>SGDClassifier()</pre></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "Pipeline(steps=[('preprocessing',\n",
       "                 ColumnTransformer(transformers=[('numeric',\n",
       "                                                  Pipeline(steps=[('powertransformer',\n",
       "                                                                   PowerTransformer()),\n",
       "                                                                  ('standardscaler',\n",
       "                                                                   StandardScaler())]),\n",
       "                                                  <sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8FA8D730>),\n",
       "                                                 ('categorical',\n",
       "                                                  Pipeline(steps=[('onehotencoder',\n",
       "                                                                   OneHotEncoder())]),\n",
       "                                                  <sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8FA8D0A0>)])),\n",
       "                ('smote_enn', SMOTEENN()), ('model', SGDClassifier())])"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sgd_pipe.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F1 Score: [0.06717705 0.09628487 0.84073795]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[ 1110,    36,   112],\n",
       "       [ 6217,   749,  3997],\n",
       "       [24462,  3810, 85469]], dtype=int64)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_pred = sgd_pipe.predict(X_test)\n",
    "f1 = f1_score(y_test, clf_pred, average=None)\n",
    "print('F1 Score:', f1)\n",
    "confusion_matrix(y_test, clf_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# attempt with a naive bayes classifier\n",
    "nbc_pipe = imbPipeline([\n",
    "    # Scaling, Normalizng, Encoding\n",
    "    ('preprocessing', ColumnTransformer([\n",
    "        ('numeric', make_pipeline(\n",
    "            PowerTransformer(),\n",
    "            MinMaxScaler()),\n",
    "            make_column_selector(dtype_include=np.number)\n",
    "        ),\n",
    "         ('categorical', make_pipeline(\n",
    "             OneHotEncoder()),\n",
    "             make_column_selector(dtype_include='object')\n",
    "         )])\n",
    "    ),\n",
    "    # Sampling to address class imbalance\n",
    "    ('smote_enn', SMOTEENN()),\n",
    "\n",
    "    # classifier\n",
    "    ('model', MultinomialNB())\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {color: black;background-color: white;}#sk-container-id-2 pre{padding: 0;}#sk-container-id-2 div.sk-toggleable {background-color: white;}#sk-container-id-2 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-2 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-2 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-2 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-2 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-2 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-2 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-2 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-2 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-2 div.sk-item {position: relative;z-index: 1;}#sk-container-id-2 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-2 div.sk-item::before, #sk-container-id-2 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-2 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-2 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-2 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-2 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-2 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-2 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-2 div.sk-label-container {text-align: center;}#sk-container-id-2 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-2 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Pipeline(steps=[(&#x27;preprocessing&#x27;,\n",
       "                 ColumnTransformer(transformers=[(&#x27;numeric&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;powertransformer&#x27;,\n",
       "                                                                   PowerTransformer()),\n",
       "                                                                  (&#x27;minmaxscaler&#x27;,\n",
       "                                                                   MinMaxScaler())]),\n",
       "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8DD2EF40&gt;),\n",
       "                                                 (&#x27;categorical&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;onehotencoder&#x27;,\n",
       "                                                                   OneHotEncoder())]),\n",
       "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8FAA1310&gt;)])),\n",
       "                (&#x27;smote_enn&#x27;, SMOTEENN()), (&#x27;model&#x27;, MultinomialNB())])</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-10\" type=\"checkbox\" ><label for=\"sk-estimator-id-10\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;preprocessing&#x27;,\n",
       "                 ColumnTransformer(transformers=[(&#x27;numeric&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;powertransformer&#x27;,\n",
       "                                                                   PowerTransformer()),\n",
       "                                                                  (&#x27;minmaxscaler&#x27;,\n",
       "                                                                   MinMaxScaler())]),\n",
       "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8DD2EF40&gt;),\n",
       "                                                 (&#x27;categorical&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;onehotencoder&#x27;,\n",
       "                                                                   OneHotEncoder())]),\n",
       "                                                  &lt;sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8FAA1310&gt;)])),\n",
       "                (&#x27;smote_enn&#x27;, SMOTEENN()), (&#x27;model&#x27;, MultinomialNB())])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-11\" type=\"checkbox\" ><label for=\"sk-estimator-id-11\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">preprocessing: ColumnTransformer</label><div class=\"sk-toggleable__content\"><pre>ColumnTransformer(transformers=[(&#x27;numeric&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;powertransformer&#x27;,\n",
       "                                                  PowerTransformer()),\n",
       "                                                 (&#x27;minmaxscaler&#x27;,\n",
       "                                                  MinMaxScaler())]),\n",
       "                                 &lt;sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8DD2EF40&gt;),\n",
       "                                (&#x27;categorical&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;onehotencoder&#x27;,\n",
       "                                                  OneHotEncoder())]),\n",
       "                                 &lt;sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8FAA1310&gt;)])</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-12\" type=\"checkbox\" ><label for=\"sk-estimator-id-12\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">numeric</label><div class=\"sk-toggleable__content\"><pre>&lt;sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8DD2EF40&gt;</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-13\" type=\"checkbox\" ><label for=\"sk-estimator-id-13\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">PowerTransformer</label><div class=\"sk-toggleable__content\"><pre>PowerTransformer()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-14\" type=\"checkbox\" ><label for=\"sk-estimator-id-14\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">MinMaxScaler</label><div class=\"sk-toggleable__content\"><pre>MinMaxScaler()</pre></div></div></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-15\" type=\"checkbox\" ><label for=\"sk-estimator-id-15\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">categorical</label><div class=\"sk-toggleable__content\"><pre>&lt;sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8FAA1310&gt;</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-16\" type=\"checkbox\" ><label for=\"sk-estimator-id-16\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">OneHotEncoder</label><div class=\"sk-toggleable__content\"><pre>OneHotEncoder()</pre></div></div></div></div></div></div></div></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-17\" type=\"checkbox\" ><label for=\"sk-estimator-id-17\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SMOTEENN</label><div class=\"sk-toggleable__content\"><pre>SMOTEENN()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-18\" type=\"checkbox\" ><label for=\"sk-estimator-id-18\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">MultinomialNB</label><div class=\"sk-toggleable__content\"><pre>MultinomialNB()</pre></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "Pipeline(steps=[('preprocessing',\n",
       "                 ColumnTransformer(transformers=[('numeric',\n",
       "                                                  Pipeline(steps=[('powertransformer',\n",
       "                                                                   PowerTransformer()),\n",
       "                                                                  ('minmaxscaler',\n",
       "                                                                   MinMaxScaler())]),\n",
       "                                                  <sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8DD2EF40>),\n",
       "                                                 ('categorical',\n",
       "                                                  Pipeline(steps=[('onehotencoder',\n",
       "                                                                   OneHotEncoder())]),\n",
       "                                                  <sklearn.compose._column_transformer.make_column_selector object at 0x000001FD8FAA1310>)])),\n",
       "                ('smote_enn', SMOTEENN()), ('model', MultinomialNB())])"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nbc_pipe.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F1 Score: [0.0381741  0.00795372 0.70679376]\n",
      "[[ 1111     2   145]\n",
      " [ 6701    44  4218]\n",
      " [49137    55 64549]]\n"
     ]
    }
   ],
   "source": [
    "clf_pred = nbc_pipe.predict(X_test)\n",
    "f1 = f1_score(y_test, clf_pred, average=None)\n",
    "print('F1 Score:', f1)\n",
    "print(confusion_matrix(y_test, clf_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "# attempt with a Random Forest classifier\n",
    "rfc_pipe = imbPipeline([\n",
    "    # Scaling, Normalizng, Encoding\n",
    "    ('preprocessing', ColumnTransformer([\n",
    "        ('numeric', make_pipeline(\n",
    "            PowerTransformer(),\n",
    "            StandardScaler()),\n",
    "            make_column_selector(dtype_include=np.number)\n",
    "        ),\n",
    "         ('categorical', make_pipeline(\n",
    "             OneHotEncoder()),\n",
    "             make_column_selector(dtype_include='object')\n",
    "         )])\n",
    "    ),\n",
    "    # Sampling to address class imbalance\n",
    "    ('smote_enn', SMOTEENN()),\n",
    "\n",
    "    # classifier\n",
    "    ('model', RandomForestClassifier())\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F1 Score: [0.18821351 0.40855973 0.90889492]\n",
      "[[  305   803   150]\n",
      " [  603  6835  3525]\n",
      " [ 1075 14858 97808]]\n"
     ]
    }
   ],
   "source": [
    "rfc_pipe.fit(X_train, y_train)\n",
    "clf_pred = rfc_pipe.predict(X_test)\n",
    "f1 = f1_score(y_test, clf_pred, average=None)\n",
    "print('F1 Score:', f1)\n",
    "print(confusion_matrix(y_test, clf_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F1 Score: [0.19504547 0.40582212 0.90818494]\n",
      "[[  311   787   160]\n",
      " [  601  6817  3545]\n",
      " [ 1019 15029 97693]]\n"
     ]
    }
   ],
   "source": [
    "rfc_params = {\n",
    "    'model__max_features': ['sqrt', 'log2', None]\n",
    "}\n",
    "\n",
    "rfc_grid = GridSearchCV(estimator=rfc_pipe, param_grid=rfc_params)\n",
    "rfc_grid.fit(X_train, y_train)\n",
    "clf_pred = rfc_grid.predict(X_test)\n",
    "f1 = f1_score(y_test, clf_pred, average=None)\n",
    "print('F1 Score:', f1)\n",
    "print(confusion_matrix(y_test, clf_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'model__max_features': 'log2'}"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfc_grid.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "filename = 'models/rfcgrid.sav'\n",
    "pickle.dump(rfc_grid, open(filename, 'wb'))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\HP USER\\AppData\\Local\\Temp\\ipykernel_17244\\2519040305.py:5: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  X['Size_Bin'] = le.fit_transform(X['Size_Bin'])\n"
     ]
    }
   ],
   "source": [
    "# XGBoost Classifier\n",
    "\n",
    "#XGBoost requires encoded y values.\n",
    "le = LabelEncoder()\n",
    "X['Size_Bin'] = le.fit_transform(X['Size_Bin'])\n",
    "X_train, X_test, y_train, y_test = train_test_split(X.drop(columns=['Size_Ha', 'Size_Bin']), X['Size_Bin'], test_size=.3, random_state=58, stratify=X['Size_Bin'])\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'large': 0, 'medium': 1, 'small': 2}\n"
     ]
    }
   ],
   "source": [
    "le_name_mapping = dict(zip(le.classes_, le.transform(le.classes_)))\n",
    "print(le_name_mapping)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F1 Score: [0.16333938 0.36147093 0.9091246 ]\n",
      "[[  675   450   133]\n",
      " [ 2072  5023  3868]\n",
      " [ 4260 11356 98125]]\n"
     ]
    }
   ],
   "source": [
    "\n",
    "xgc_pipe = imbPipeline([\n",
    "    # Scaling, Normalizng, Encoding\n",
    "    ('preprocessing', ColumnTransformer([\n",
    "        ('numeric', make_pipeline(\n",
    "            PowerTransformer(),\n",
    "            StandardScaler()),\n",
    "            make_column_selector(dtype_include=np.number)\n",
    "        ),\n",
    "         ('categorical', make_pipeline(\n",
    "             OneHotEncoder()),\n",
    "             make_column_selector(dtype_include='object')\n",
    "         )])\n",
    "    ),\n",
    "    # Sampling to address class imbalance\n",
    "    ('smote_enn', SMOTEENN()),\n",
    "\n",
    "    # classifier\n",
    "    ('model', xgb.XGBClassifier())\n",
    "])\n",
    "\n",
    "xgc_pipe.fit(X_train, y_train)\n",
    "clf_pred = xgc_pipe.predict(X_test)\n",
    "f1 = f1_score(y_test, clf_pred, average=None)\n",
    "print('F1 Score:', f1)\n",
    "print(confusion_matrix(y_test, clf_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x1fe76172e20>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import ConfusionMatrixDisplay\n",
    "cm = confusion_matrix(y_test, clf_pred)\n",
    "cm_plot = ConfusionMatrixDisplay(cm, display_labels=xgc_pipe.classes_)\n",
    "cm_plot.plot()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Stage 2: Regressors\n",
    "### Small Fire Regressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create training and test data for small regressor\n",
    "small_filter = X['Size_Bin'] == 'small'\n",
    "X_train, X_test, y_train, y_test = train_test_split(X[small_filter].drop(columns=['Size_Ha', 'Size_Bin']), X[small_filter]['Size_Ha'], test_size=.3, random_state=58)\n",
    "\n",
    "# Preprocessing Pipeline\n",
    "\n",
    "# isolate categorical features\n",
    "cat_feats = X_train.dtypes[X_train.dtypes == 'object'].index.tolist()\n",
    "\n",
    "# isolate numeric features\n",
    "num_feats = X_train.dtypes[X_train.dtypes == np.number].index.tolist()\n",
    "\n",
    "numeric_transform = Pipeline([\n",
    "    ('transform', PowerTransformer()),\n",
    "    ('scaler', StandardScaler()),\n",
    "])\n",
    "categorical_transform = Pipeline([\n",
    "    ('encoder', OneHotEncoder())\n",
    "])\n",
    "\n",
    "preprocessing = ColumnTransformer([\n",
    "    ('numeric', numeric_transform, num_feats),\n",
    "    ('categorical', categorical_transform, cat_feats)\n",
    "])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 2.406\n",
      "R2 of the model: -0.338\n"
     ]
    }
   ],
   "source": [
    "# AdaBoost model\n",
    "adb_small_pipe = Pipeline([\n",
    "    ('preprocessing', preprocessing),\n",
    "    ('model', AdaBoostRegressor())\n",
    "])\n",
    "\n",
    "adb_small_pipe.fit(X_train, y_train)\n",
    "\n",
    "evaluate_model(adb_small_pipe, X_test, y_test)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 2.040\n",
      "R2 of the model: 0.037\n",
      "Best Parameters: {'model__learning_rate': 0.001, 'model__loss': 'linear', 'model__n_estimators': 50}\n"
     ]
    }
   ],
   "source": [
    "adb_small_params = {\n",
    "    'model__n_estimators': [50, 100],\n",
    "    'model__loss': ['linear', 'square'],\n",
    "    'model__learning_rate': [.001, .01]\n",
    "}\n",
    "\n",
    "adb_small_grid = GridSearchCV(estimator=adb_small_pipe, param_grid=adb_small_params)\n",
    "adb_small_grid.fit(X_train, y_train)\n",
    "evaluate_model(adb_small_grid, X_test, y_test)\n",
    "print('Best Parameters:', adb_small_grid.best_params_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "filename = 'models/adb_small_grid.sav'\n",
    "pickle.dump(adb_small_grid, open(filename, 'wb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 2.040\n",
      "R2 of the model: 0.038\n"
     ]
    }
   ],
   "source": [
    "en_small_pipe =  Pipeline([\n",
    "    ('preprocessing', preprocessing),\n",
    "    ('model', ElasticNetCV())\n",
    "])\n",
    "\n",
    "en_small_pipe.fit(X_train, y_train)\n",
    "evaluate_model(en_small_pipe, X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 2.039\n",
      "R2 of the model: 0.038\n"
     ]
    }
   ],
   "source": [
    "en_params = {\n",
    "    'model__l1_ratio': [.1, .3, .5, .8,  1]\n",
    "}\n",
    "en_small_grid = GridSearchCV(estimator=en_small_pipe, param_grid=en_params)\n",
    "\n",
    "en_small_grid.fit(X_train, y_train)\n",
    "evaluate_model(en_small_grid, X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 1.977\n",
      "R2 of the model: 0.096\n"
     ]
    }
   ],
   "source": [
    "# XGBoost\n",
    "xgb_small_pipe = Pipeline([\n",
    "    ('preprocessing', preprocessing),\n",
    "    ('model', xgb.XGBRegressor())\n",
    "])\n",
    "\n",
    "xgb_small_pipe.fit(X_train, y_train)\n",
    "evaluate_model(xgb_small_pipe, X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.1837102380620621"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xgb_small_pipe.score(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 1.974\n",
      "R2 of the model: 0.100\n",
      "{'model__eta': 0.3, 'model__max_depth': 6, 'model__min_child_weight': 50}\n"
     ]
    }
   ],
   "source": [
    "xgb_params = {\n",
    "    'model__eta': [.01, .1, .3],\n",
    "    'model__max_depth': [2, 4, 6],\n",
    "    'model__min_child_weight': [1, 10, 50]\n",
    "}\n",
    "xgb_small_grid = GridSearchCV(estimator=xgb_small_pipe, param_grid=xgb_params)\n",
    "xgb_small_grid.fit(X_train, y_train)\n",
    "evaluate_model(xgb_small_grid,X_test, y_test)\n",
    "print(xgb_small_grid.best_params_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 1.970\n",
      "R2 of the model: 0.102\n",
      "{'model__eta': 0.3, 'model__max_depth': 10, 'model__min_child_weight': 500}\n"
     ]
    }
   ],
   "source": [
    "xgb_params = {\n",
    "    'model__eta': [.3, .6, .8],\n",
    "    'model__max_depth': [6, 10, 15],\n",
    "    'model__min_child_weight': [100, 50, 500]\n",
    "}\n",
    "xgb_small_grid2 = GridSearchCV(estimator=xgb_small_pipe, param_grid=xgb_params)\n",
    "xgb_small_grid2.fit(X_train, y_train)\n",
    "evaluate_model(xgb_small_grid2,X_test, y_test)\n",
    "print(xgb_small_grid2.best_params_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "filename = 'models/xgb_small.sav'\n",
    "pickle.dump(xgb_small_grid, open(filename, 'wb'))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Medium Fire Regressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create training and test data for medium regressor\n",
    "medium_filter = X['Size_Bin'] == 'medium'\n",
    "X_train, X_test, y_train, y_test = train_test_split(X[medium_filter].drop(columns=['Size_Ha', 'Size_Bin']), X[medium_filter]['Size_Ha'], test_size=.25, random_state=58)\n",
    "\n",
    "# Preprocessing Pipeline\n",
    "\n",
    "# isolate categorical features\n",
    "cat_feats = X_train.dtypes[X_train.dtypes == 'object'].index.tolist()\n",
    "\n",
    "# isolate numeric features\n",
    "num_feats = X_train.dtypes[X_train.dtypes == np.number].index.tolist()\n",
    "\n",
    "numeric_transform = Pipeline([\n",
    "    ('transform', PowerTransformer()),\n",
    "    ('scaler', StandardScaler()),\n",
    "])\n",
    "categorical_transform = Pipeline([\n",
    "    ('encoder', OneHotEncoder())\n",
    "])\n",
    "\n",
    "preprocessing = ColumnTransformer([\n",
    "    ('numeric', numeric_transform, num_feats),\n",
    "    ('categorical', categorical_transform, cat_feats)\n",
    "])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 904.671\n",
      "R2 of the model: 0.052\n"
     ]
    }
   ],
   "source": [
    "# AdaBoost model\n",
    "adb_med_pipe = Pipeline([\n",
    "    ('preprocessing', preprocessing),\n",
    "    ('model', AdaBoostRegressor(learning_rate=0.01, loss='linear', n_estimators=50)) # these parameters came from gridsearch below\n",
    "])\n",
    "\n",
    "adb_med_pipe.fit(X_train, y_train)\n",
    "\n",
    "evaluate_model(adb_med_pipe, X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 904.831\n",
      "R2 of the model: 0.052\n",
      "Best Parameters: {'model__learning_rate': 0.01, 'model__loss': 'linear', 'model__n_estimators': 50}\n"
     ]
    }
   ],
   "source": [
    "adb_med_params = {\n",
    "    'model__n_estimators': [50, 100, 500],\n",
    "    'model__loss': ['linear', 'square'],\n",
    "    'model__learning_rate': [.01, .1, 1]\n",
    "}\n",
    "\n",
    "med_grid = GridSearchCV(estimator=adb_med_pipe, param_grid=adb_med_params)\n",
    "med_grid.fit(X_train, y_train)\n",
    "evaluate_model(med_grid, X_test, y_test)\n",
    "print('Best Parameters:', med_grid.best_params_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 902.786\n",
      "R2 of the model: 0.056\n"
     ]
    }
   ],
   "source": [
    "# XGBoost\n",
    "xgb_med_pipe = Pipeline([\n",
    "    ('preprocessing', preprocessing),\n",
    "    ('model', xgb.XGBRegressor())\n",
    "])\n",
    "\n",
    "xgb_med_pipe.fit(X_train, y_train)\n",
    "evaluate_model(xgb_med_pipe, X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 889.295\n",
      "R2 of the model: 0.084\n",
      "{'model__eta': 0.1, 'model__max_depth': 6, 'model__min_child_weight': 50}\n"
     ]
    }
   ],
   "source": [
    "xgb_params = {\n",
    "    'model__eta': [.01, .1, .3],\n",
    "    'model__max_depth': [2, 4, 6],\n",
    "    'model__min_child_weight': [1, 10, 50]\n",
    "}\n",
    "xgb_med_grid = GridSearchCV(estimator=xgb_med_pipe, param_grid=xgb_params)\n",
    "xgb_med_grid.fit(X_train, y_train)\n",
    "evaluate_model(xgb_med_grid,X_test, y_test)\n",
    "print(xgb_med_grid.best_params_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 888.042\n",
      "R2 of the model: 0.087\n",
      "{'model__eta': 0.1, 'model__max_depth': 8, 'model__min_child_weight': 200}\n"
     ]
    }
   ],
   "source": [
    "xgb_params = {\n",
    "    'model__eta': [.01, .1, .3],\n",
    "    'model__max_depth': [4, 6, 8, 10],\n",
    "    'model__min_child_weight': [50, 100, 200]\n",
    "}\n",
    "xgb_med_grid = GridSearchCV(estimator=xgb_med_pipe, param_grid=xgb_params)\n",
    "xgb_med_grid.fit(X_train, y_train)\n",
    "evaluate_model(xgb_med_grid,X_test, y_test)\n",
    "print(xgb_med_grid.best_params_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "filename = 'models/xgb_med.sav'\n",
    "pickle.dump(xgb_med_grid, open(filename, 'wb'))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Large Fire Regressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3354, 13)\n"
     ]
    }
   ],
   "source": [
    "# Create training and test data for large regressor\n",
    "large_filter = X['Size_Bin'] == 'large'\n",
    "X_train, X_test, y_train, y_test = train_test_split(X[large_filter].drop(columns=['Size_Ha', 'Size_Bin']), X[large_filter]['Size_Ha'], test_size=.2, random_state=58)\n",
    "print(X_train.shape)\n",
    "\n",
    "# isolate categorical features\n",
    "cat_feats = X_train.dtypes[X_train.dtypes == 'object'].index.tolist()\n",
    "\n",
    "# isolate numeric features\n",
    "num_feats = X_train.dtypes[X_train.dtypes == np.number].index.tolist()\n",
    "\n",
    "# Preprocessing Pipeline\n",
    "\n",
    "numeric_transform = Pipeline([\n",
    "    ('transform', PowerTransformer()),\n",
    "    ('scaler', StandardScaler()),\n",
    "])\n",
    "categorical_transform = Pipeline([\n",
    "    ('encoder', OneHotEncoder())\n",
    "])\n",
    "\n",
    "preprocessing = ColumnTransformer([\n",
    "    ('numeric', numeric_transform, num_feats),\n",
    "    ('categorical', categorical_transform, cat_feats)\n",
    "])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# AdaBoost model\n",
    "adb_large_pipe = Pipeline([\n",
    "    ('preprocessing', preprocessing),\n",
    "    ('model', AdaBoostRegressor())\n",
    "])\n",
    "\n",
    "adb_large_pipe.fit(X_train, y_train)\n",
    "\n",
    "evaluate_model(adb_large_pipe, X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 52590.670\n",
      "R2 of the model: -0.001\n"
     ]
    }
   ],
   "source": [
    "adb_large_params = {\n",
    "    'model__n_estimators': [50, 100, 500],\n",
    "    'model__loss': ['linear', 'square'],\n",
    "    'model__learning_rate': [.01, .1, 1]\n",
    "}\n",
    "\n",
    "large_grid = GridSearchCV(estimator=adb_large_pipe, param_grid=adb_large_params)\n",
    "large_grid.fit(X_train, y_train)\n",
    "evaluate_model(large_grid, X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'model__learning_rate': 0.01,\n",
       " 'model__loss': 'square',\n",
       " 'model__n_estimators': 50}"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "large_grid.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-3 {color: black;background-color: white;}#sk-container-id-3 pre{padding: 0;}#sk-container-id-3 div.sk-toggleable {background-color: white;}#sk-container-id-3 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-3 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-3 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-3 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-3 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-3 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-3 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-3 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-3 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-3 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-3 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-3 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-3 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-3 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-3 div.sk-item {position: relative;z-index: 1;}#sk-container-id-3 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-3 div.sk-item::before, #sk-container-id-3 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-3 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-3 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-3 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-3 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-3 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-3 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-3 div.sk-label-container {text-align: center;}#sk-container-id-3 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-3 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-3\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Pipeline(steps=[(&#x27;preprocessing&#x27;,\n",
       "                 ColumnTransformer(transformers=[(&#x27;numeric&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;transform&#x27;,\n",
       "                                                                   PowerTransformer()),\n",
       "                                                                  (&#x27;scaler&#x27;,\n",
       "                                                                   StandardScaler())]),\n",
       "                                                  [&#x27;LATITUDE&#x27;, &#x27;LONGITUDE&#x27;,\n",
       "                                                   &#x27;elevation&#x27;,\n",
       "                                                   &#x27;2_Week_Rainfall&#x27;,\n",
       "                                                   &#x27;Avg_High&#x27;,\n",
       "                                                   &#x27;Start_Date_High&#x27;,\n",
       "                                                   &#x27;Avg_Temp&#x27;,\n",
       "                                                   &#x27;Start_Date_Wind_Speed&#x27;]),\n",
       "                                                 (&#x27;categorical&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;encoder&#x27;,\n",
       "                                                                   OneHotEncoder())]),\n",
       "                                                  [&#x27;CAUSE&#x27;])])),\n",
       "                (&#x27;model&#x27;, ElasticNetCV())])</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-19\" type=\"checkbox\" ><label for=\"sk-estimator-id-19\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;preprocessing&#x27;,\n",
       "                 ColumnTransformer(transformers=[(&#x27;numeric&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;transform&#x27;,\n",
       "                                                                   PowerTransformer()),\n",
       "                                                                  (&#x27;scaler&#x27;,\n",
       "                                                                   StandardScaler())]),\n",
       "                                                  [&#x27;LATITUDE&#x27;, &#x27;LONGITUDE&#x27;,\n",
       "                                                   &#x27;elevation&#x27;,\n",
       "                                                   &#x27;2_Week_Rainfall&#x27;,\n",
       "                                                   &#x27;Avg_High&#x27;,\n",
       "                                                   &#x27;Start_Date_High&#x27;,\n",
       "                                                   &#x27;Avg_Temp&#x27;,\n",
       "                                                   &#x27;Start_Date_Wind_Speed&#x27;]),\n",
       "                                                 (&#x27;categorical&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;encoder&#x27;,\n",
       "                                                                   OneHotEncoder())]),\n",
       "                                                  [&#x27;CAUSE&#x27;])])),\n",
       "                (&#x27;model&#x27;, ElasticNetCV())])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-20\" type=\"checkbox\" ><label for=\"sk-estimator-id-20\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">preprocessing: ColumnTransformer</label><div class=\"sk-toggleable__content\"><pre>ColumnTransformer(transformers=[(&#x27;numeric&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;transform&#x27;,\n",
       "                                                  PowerTransformer()),\n",
       "                                                 (&#x27;scaler&#x27;, StandardScaler())]),\n",
       "                                 [&#x27;LATITUDE&#x27;, &#x27;LONGITUDE&#x27;, &#x27;elevation&#x27;,\n",
       "                                  &#x27;2_Week_Rainfall&#x27;, &#x27;Avg_High&#x27;,\n",
       "                                  &#x27;Start_Date_High&#x27;, &#x27;Avg_Temp&#x27;,\n",
       "                                  &#x27;Start_Date_Wind_Speed&#x27;]),\n",
       "                                (&#x27;categorical&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;encoder&#x27;, OneHotEncoder())]),\n",
       "                                 [&#x27;CAUSE&#x27;])])</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-21\" type=\"checkbox\" ><label for=\"sk-estimator-id-21\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">numeric</label><div class=\"sk-toggleable__content\"><pre>[&#x27;LATITUDE&#x27;, &#x27;LONGITUDE&#x27;, &#x27;elevation&#x27;, &#x27;2_Week_Rainfall&#x27;, &#x27;Avg_High&#x27;, &#x27;Start_Date_High&#x27;, &#x27;Avg_Temp&#x27;, &#x27;Start_Date_Wind_Speed&#x27;]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-22\" type=\"checkbox\" ><label for=\"sk-estimator-id-22\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">PowerTransformer</label><div class=\"sk-toggleable__content\"><pre>PowerTransformer()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-23\" type=\"checkbox\" ><label for=\"sk-estimator-id-23\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">StandardScaler</label><div class=\"sk-toggleable__content\"><pre>StandardScaler()</pre></div></div></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-24\" type=\"checkbox\" ><label for=\"sk-estimator-id-24\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">categorical</label><div class=\"sk-toggleable__content\"><pre>[&#x27;CAUSE&#x27;]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-25\" type=\"checkbox\" ><label for=\"sk-estimator-id-25\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">OneHotEncoder</label><div class=\"sk-toggleable__content\"><pre>OneHotEncoder()</pre></div></div></div></div></div></div></div></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-26\" type=\"checkbox\" ><label for=\"sk-estimator-id-26\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">ElasticNetCV</label><div class=\"sk-toggleable__content\"><pre>ElasticNetCV()</pre></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "Pipeline(steps=[('preprocessing',\n",
       "                 ColumnTransformer(transformers=[('numeric',\n",
       "                                                  Pipeline(steps=[('transform',\n",
       "                                                                   PowerTransformer()),\n",
       "                                                                  ('scaler',\n",
       "                                                                   StandardScaler())]),\n",
       "                                                  ['LATITUDE', 'LONGITUDE',\n",
       "                                                   'elevation',\n",
       "                                                   '2_Week_Rainfall',\n",
       "                                                   'Avg_High',\n",
       "                                                   'Start_Date_High',\n",
       "                                                   'Avg_Temp',\n",
       "                                                   'Start_Date_Wind_Speed']),\n",
       "                                                 ('categorical',\n",
       "                                                  Pipeline(steps=[('encoder',\n",
       "                                                                   OneHotEncoder())]),\n",
       "                                                  ['CAUSE'])])),\n",
       "                ('model', ElasticNetCV())])"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "en_large_pipe = Pipeline([\n",
    "    ('preprocessing', preprocessing),\n",
    "    ('model', ElasticNetCV())\n",
    "])\n",
    "\n",
    "en_large_pipe.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 55968.119\n",
      "R2 of the model: 0.001\n"
     ]
    }
   ],
   "source": [
    "evaluate_model(en_large_pipe, X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 55959.623\n",
      "R2 of the model: 0.002\n"
     ]
    }
   ],
   "source": [
    "en_params = {\n",
    "    'model__l1_ratio': [.1, .3, .5, .8,  1]\n",
    "}\n",
    "en_large_grid = GridSearchCV(estimator=en_large_pipe, param_grid=en_params)\n",
    "\n",
    "en_large_grid.fit(X_train, y_train)\n",
    "evaluate_model(en_large_grid, X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\HP USER\\anaconda3\\envs\\lhl_data\\lib\\site-packages\\sklearn\\linear_model\\_huber.py:342: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "  self.n_iter_ = _check_optimize_result(\"lbfgs\", opt_res, self.max_iter)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;background-color: white;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Pipeline(steps=[(&#x27;preprocessing&#x27;,\n",
       "                 ColumnTransformer(transformers=[(&#x27;numeric&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;transform&#x27;,\n",
       "                                                                   PowerTransformer()),\n",
       "                                                                  (&#x27;scaler&#x27;,\n",
       "                                                                   StandardScaler())]),\n",
       "                                                  [&#x27;Latitude&#x27;, &#x27;Longitude&#x27;,\n",
       "                                                   &#x27;Elevation&#x27;,\n",
       "                                                   &#x27;2_Week_Rainfall&#x27;,\n",
       "                                                   &#x27;Avg_High&#x27;,\n",
       "                                                   &#x27;Start_Date_High&#x27;,\n",
       "                                                   &#x27;Avg_Temp&#x27;,\n",
       "                                                   &#x27;Start_Date_Wind_Speed&#x27;]),\n",
       "                                                 (&#x27;categorical&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;encoder&#x27;,\n",
       "                                                                   OneHotEncoder())]),\n",
       "                                                  [&#x27;Cause&#x27;])])),\n",
       "                (&#x27;model&#x27;, HuberRegressor())])</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" ><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;preprocessing&#x27;,\n",
       "                 ColumnTransformer(transformers=[(&#x27;numeric&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;transform&#x27;,\n",
       "                                                                   PowerTransformer()),\n",
       "                                                                  (&#x27;scaler&#x27;,\n",
       "                                                                   StandardScaler())]),\n",
       "                                                  [&#x27;Latitude&#x27;, &#x27;Longitude&#x27;,\n",
       "                                                   &#x27;Elevation&#x27;,\n",
       "                                                   &#x27;2_Week_Rainfall&#x27;,\n",
       "                                                   &#x27;Avg_High&#x27;,\n",
       "                                                   &#x27;Start_Date_High&#x27;,\n",
       "                                                   &#x27;Avg_Temp&#x27;,\n",
       "                                                   &#x27;Start_Date_Wind_Speed&#x27;]),\n",
       "                                                 (&#x27;categorical&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;encoder&#x27;,\n",
       "                                                                   OneHotEncoder())]),\n",
       "                                                  [&#x27;Cause&#x27;])])),\n",
       "                (&#x27;model&#x27;, HuberRegressor())])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" ><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">preprocessing: ColumnTransformer</label><div class=\"sk-toggleable__content\"><pre>ColumnTransformer(transformers=[(&#x27;numeric&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;transform&#x27;,\n",
       "                                                  PowerTransformer()),\n",
       "                                                 (&#x27;scaler&#x27;, StandardScaler())]),\n",
       "                                 [&#x27;Latitude&#x27;, &#x27;Longitude&#x27;, &#x27;Elevation&#x27;,\n",
       "                                  &#x27;2_Week_Rainfall&#x27;, &#x27;Avg_High&#x27;,\n",
       "                                  &#x27;Start_Date_High&#x27;, &#x27;Avg_Temp&#x27;,\n",
       "                                  &#x27;Start_Date_Wind_Speed&#x27;]),\n",
       "                                (&#x27;categorical&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;encoder&#x27;, OneHotEncoder())]),\n",
       "                                 [&#x27;Cause&#x27;])])</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-3\" type=\"checkbox\" ><label for=\"sk-estimator-id-3\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">numeric</label><div class=\"sk-toggleable__content\"><pre>[&#x27;Latitude&#x27;, &#x27;Longitude&#x27;, &#x27;Elevation&#x27;, &#x27;2_Week_Rainfall&#x27;, &#x27;Avg_High&#x27;, &#x27;Start_Date_High&#x27;, &#x27;Avg_Temp&#x27;, &#x27;Start_Date_Wind_Speed&#x27;]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-4\" type=\"checkbox\" ><label for=\"sk-estimator-id-4\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">PowerTransformer</label><div class=\"sk-toggleable__content\"><pre>PowerTransformer()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-5\" type=\"checkbox\" ><label for=\"sk-estimator-id-5\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">StandardScaler</label><div class=\"sk-toggleable__content\"><pre>StandardScaler()</pre></div></div></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-6\" type=\"checkbox\" ><label for=\"sk-estimator-id-6\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">categorical</label><div class=\"sk-toggleable__content\"><pre>[&#x27;Cause&#x27;]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-7\" type=\"checkbox\" ><label for=\"sk-estimator-id-7\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">OneHotEncoder</label><div class=\"sk-toggleable__content\"><pre>OneHotEncoder()</pre></div></div></div></div></div></div></div></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-8\" type=\"checkbox\" ><label for=\"sk-estimator-id-8\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">HuberRegressor</label><div class=\"sk-toggleable__content\"><pre>HuberRegressor()</pre></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "Pipeline(steps=[('preprocessing',\n",
       "                 ColumnTransformer(transformers=[('numeric',\n",
       "                                                  Pipeline(steps=[('transform',\n",
       "                                                                   PowerTransformer()),\n",
       "                                                                  ('scaler',\n",
       "                                                                   StandardScaler())]),\n",
       "                                                  ['Latitude', 'Longitude',\n",
       "                                                   'Elevation',\n",
       "                                                   '2_Week_Rainfall',\n",
       "                                                   'Avg_High',\n",
       "                                                   'Start_Date_High',\n",
       "                                                   'Avg_Temp',\n",
       "                                                   'Start_Date_Wind_Speed']),\n",
       "                                                 ('categorical',\n",
       "                                                  Pipeline(steps=[('encoder',\n",
       "                                                                   OneHotEncoder())]),\n",
       "                                                  ['Cause'])])),\n",
       "                ('model', HuberRegressor())])"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hr_large_pipe = Pipeline([\n",
    "    ('preprocessing', preprocessing),\n",
    "    ('model', HuberRegressor())\n",
    "])\n",
    "\n",
    "hr_large_pipe.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 50889.811\n",
      "R2 of the model: -0.058\n"
     ]
    }
   ],
   "source": [
    "evaluate_model(hr_large_pipe, X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\HP USER\\anaconda3\\envs\\lhl_data\\lib\\site-packages\\sklearn\\linear_model\\_huber.py:342: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "  self.n_iter_ = _check_optimize_result(\"lbfgs\", opt_res, self.max_iter)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 49351.625\n",
      "R2 of the model: 0.005\n"
     ]
    }
   ],
   "source": [
    "hr_params = {\n",
    "    'model__alpha': [.0001, .01, .1],\n",
    "    'model__epsilon': [1, 10, 100]\n",
    "}\n",
    "hr_large_grid = GridSearchCV(estimator=hr_large_pipe, param_grid=hr_params)\n",
    "\n",
    "hr_large_grid.fit(X_train, y_train)\n",
    "evaluate_model(hr_large_grid, X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'model__alpha': 0.0001, 'model__epsilon': 10}"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hr_large_grid.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "filename = 'models/hr_large_grid.sav'\n",
    "pickle.dump(hr_large_grid, open(filename, 'wb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 50429.038\n",
      "R2 of the model: -0.039\n"
     ]
    }
   ],
   "source": [
    "rfr_large_pipe = Pipeline([\n",
    "    ('preprocessing', preprocessing),\n",
    "    ('model', RandomForestRegressor())\n",
    "])\n",
    "\n",
    "rfr_large_pipe.fit(X_train, y_train)\n",
    "evaluate_model(rfr_large_pipe, X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 50119.233\n",
      "R2 of the model: -0.026\n"
     ]
    }
   ],
   "source": [
    "rfr_params = {\n",
    "    'model__n_estimators': [50, 100, 500],\n",
    "    'model__criterion': ['squared_error', 'absolute_error']\n",
    "}\n",
    "rfr_large_grid = GridSearchCV(estimator=rfr_large_pipe, param_grid=rfr_params)\n",
    "\n",
    "rfr_large_grid.fit(X_train, y_train)\n",
    "evaluate_model(rfr_large_grid, X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 51971.722\n",
      "R2 of the model: -0.103\n"
     ]
    }
   ],
   "source": [
    "xgb_large_pipe = Pipeline([\n",
    "    ('preprocessing', preprocessing),\n",
    "    ('model', xgb.XGBRegressor())\n",
    "])\n",
    "\n",
    "xgb_large_pipe.fit(X_train, y_train)\n",
    "evaluate_model(xgb_large_pipe, X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of the model: 49176.651\n",
      "R2 of the model: 0.012\n",
      "{'model__eta': 0.1, 'model__max_depth': 2, 'model__min_child_weight': 10}\n"
     ]
    }
   ],
   "source": [
    "xgb_params = {\n",
    "    'model__eta': [.01, .1, .3],\n",
    "    'model__max_depth': [2, 4, 6],\n",
    "    'model__min_child_weight': [1, 10, 50]\n",
    "}\n",
    "xgb_large_grid = GridSearchCV(estimator=xgb_large_pipe, param_grid=xgb_params)\n",
    "xgb_large_grid.fit(X_train, y_train)\n",
    "evaluate_model(xgb_large_grid,X_test, y_test)\n",
    "print(xgb_large_grid.best_params_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "filename = 'models/xgb_large.sav'\n",
    "pickle.dump(xgb_large_grid, open(filename, 'wb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "lhl_data",
   "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": "3.8.16"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}