g2-forestcovertype-finalmodelevaluation-notebook.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "f3bdfca5",
   "metadata": {
    "papermill": {
     "duration": 0.025594,
     "end_time": "2021-12-12T04:16:18.524864",
     "exception": false,
     "start_time": "2021-12-12T04:16:18.499270",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Final Model Evaluation Notebook\n",
    "\n",
    "This notebook consists of an evaluation of our best model for the Forest Cover Type Prediction competition. Our best model is an Extra Trees model, which outperformed our ensemble. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "180afb42",
   "metadata": {
    "papermill": {
     "duration": 0.020935,
     "end_time": "2021-12-12T04:16:18.567488",
     "exception": false,
     "start_time": "2021-12-12T04:16:18.546553",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Import Statements"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b73abddb",
   "metadata": {
    "papermill": {
     "duration": 0.020237,
     "end_time": "2021-12-12T04:16:18.608290",
     "exception": false,
     "start_time": "2021-12-12T04:16:18.588053",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "First, we will import the necessary packages for data manipulation and analysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "997d9a4f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:18.661196Z",
     "iopub.status.busy": "2021-12-12T04:16:18.660378Z",
     "iopub.status.idle": "2021-12-12T04:16:25.454150Z",
     "shell.execute_reply": "2021-12-12T04:16:25.453451Z",
     "shell.execute_reply.started": "2021-12-12T04:14:34.489538Z"
    },
    "papermill": {
     "duration": 6.825411,
     "end_time": "2021-12-12T04:16:25.454333",
     "exception": false,
     "start_time": "2021-12-12T04:16:18.628922",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import numpy as np, pandas as pd, matplotlib.pyplot as plt\n",
    "import joblib\n",
    "import sklearn \n",
    "\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.ensemble import VotingClassifier\n",
    "from sklearn.metrics import confusion_matrix, classification_report\n",
    "from keras.metrics import top_k_categorical_accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "3a0fccb7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:25.502181Z",
     "iopub.status.busy": "2021-12-12T04:16:25.501258Z",
     "iopub.status.idle": "2021-12-12T04:16:25.620593Z",
     "shell.execute_reply": "2021-12-12T04:16:25.621083Z",
     "shell.execute_reply.started": "2021-12-12T04:14:34.542212Z"
    },
    "papermill": {
     "duration": 0.144357,
     "end_time": "2021-12-12T04:16:25.621266",
     "exception": false,
     "start_time": "2021-12-12T04:16:25.476909",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Id</th>\n",
       "      <th>Elevation</th>\n",
       "      <th>Aspect</th>\n",
       "      <th>Slope</th>\n",
       "      <th>Horizontal_Distance_To_Hydrology</th>\n",
       "      <th>Vertical_Distance_To_Hydrology</th>\n",
       "      <th>Horizontal_Distance_To_Roadways</th>\n",
       "      <th>Hillshade_9am</th>\n",
       "      <th>Hillshade_Noon</th>\n",
       "      <th>Hillshade_3pm</th>\n",
       "      <th>...</th>\n",
       "      <th>Soil_Type32</th>\n",
       "      <th>Soil_Type33</th>\n",
       "      <th>Soil_Type34</th>\n",
       "      <th>Soil_Type35</th>\n",
       "      <th>Soil_Type36</th>\n",
       "      <th>Soil_Type37</th>\n",
       "      <th>Soil_Type38</th>\n",
       "      <th>Soil_Type39</th>\n",
       "      <th>Soil_Type40</th>\n",
       "      <th>Cover_Type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>2596</td>\n",
       "      <td>51</td>\n",
       "      <td>3</td>\n",
       "      <td>258</td>\n",
       "      <td>0</td>\n",
       "      <td>510</td>\n",
       "      <td>221</td>\n",
       "      <td>232</td>\n",
       "      <td>148</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>2590</td>\n",
       "      <td>56</td>\n",
       "      <td>2</td>\n",
       "      <td>212</td>\n",
       "      <td>-6</td>\n",
       "      <td>390</td>\n",
       "      <td>220</td>\n",
       "      <td>235</td>\n",
       "      <td>151</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>2804</td>\n",
       "      <td>139</td>\n",
       "      <td>9</td>\n",
       "      <td>268</td>\n",
       "      <td>65</td>\n",
       "      <td>3180</td>\n",
       "      <td>234</td>\n",
       "      <td>238</td>\n",
       "      <td>135</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>2785</td>\n",
       "      <td>155</td>\n",
       "      <td>18</td>\n",
       "      <td>242</td>\n",
       "      <td>118</td>\n",
       "      <td>3090</td>\n",
       "      <td>238</td>\n",
       "      <td>238</td>\n",
       "      <td>122</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>2595</td>\n",
       "      <td>45</td>\n",
       "      <td>2</td>\n",
       "      <td>153</td>\n",
       "      <td>-1</td>\n",
       "      <td>391</td>\n",
       "      <td>220</td>\n",
       "      <td>234</td>\n",
       "      <td>150</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 56 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Id  Elevation  Aspect  Slope  Horizontal_Distance_To_Hydrology  \\\n",
       "0   1       2596      51      3                               258   \n",
       "1   2       2590      56      2                               212   \n",
       "2   3       2804     139      9                               268   \n",
       "3   4       2785     155     18                               242   \n",
       "4   5       2595      45      2                               153   \n",
       "\n",
       "   Vertical_Distance_To_Hydrology  Horizontal_Distance_To_Roadways  \\\n",
       "0                               0                              510   \n",
       "1                              -6                              390   \n",
       "2                              65                             3180   \n",
       "3                             118                             3090   \n",
       "4                              -1                              391   \n",
       "\n",
       "   Hillshade_9am  Hillshade_Noon  Hillshade_3pm  ...  Soil_Type32  \\\n",
       "0            221             232            148  ...            0   \n",
       "1            220             235            151  ...            0   \n",
       "2            234             238            135  ...            0   \n",
       "3            238             238            122  ...            0   \n",
       "4            220             234            150  ...            0   \n",
       "\n",
       "   Soil_Type33  Soil_Type34  Soil_Type35  Soil_Type36  Soil_Type37  \\\n",
       "0            0            0            0            0            0   \n",
       "1            0            0            0            0            0   \n",
       "2            0            0            0            0            0   \n",
       "3            0            0            0            0            0   \n",
       "4            0            0            0            0            0   \n",
       "\n",
       "   Soil_Type38  Soil_Type39  Soil_Type40  Cover_Type  \n",
       "0            0            0            0           5  \n",
       "1            0            0            0           5  \n",
       "2            0            0            0           2  \n",
       "3            0            0            0           2  \n",
       "4            0            0            0           5  \n",
       "\n",
       "[5 rows x 56 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load data\n",
    "train = pd.read_csv('../input/forest-cover-type-prediction/train.csv')\n",
    "# view data\n",
    "train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f190e464",
   "metadata": {
    "papermill": {
     "duration": 0.021109,
     "end_time": "2021-12-12T04:16:25.664037",
     "exception": false,
     "start_time": "2021-12-12T04:16:25.642928",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Preprocessing"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9b637f6c",
   "metadata": {
    "papermill": {
     "duration": 0.020952,
     "end_time": "2021-12-12T04:16:25.706282",
     "exception": false,
     "start_time": "2021-12-12T04:16:25.685330",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Next, we will preprocess the data to remove the ID column from the set, add new features from our feature engineering and PCA analysis, split the data into training and validation sets, and scale the data. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "de1889c5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:25.755111Z",
     "iopub.status.busy": "2021-12-12T04:16:25.754486Z",
     "iopub.status.idle": "2021-12-12T04:16:25.766417Z",
     "shell.execute_reply": "2021-12-12T04:16:25.766943Z",
     "shell.execute_reply.started": "2021-12-12T04:14:34.620522Z"
    },
    "papermill": {
     "duration": 0.03942,
     "end_time": "2021-12-12T04:16:25.767113",
     "exception": false,
     "start_time": "2021-12-12T04:16:25.727693",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Elevation</th>\n",
       "      <th>Aspect</th>\n",
       "      <th>Slope</th>\n",
       "      <th>Horizontal_Distance_To_Hydrology</th>\n",
       "      <th>Vertical_Distance_To_Hydrology</th>\n",
       "      <th>Horizontal_Distance_To_Roadways</th>\n",
       "      <th>Hillshade_9am</th>\n",
       "      <th>Hillshade_Noon</th>\n",
       "      <th>Hillshade_3pm</th>\n",
       "      <th>Horizontal_Distance_To_Fire_Points</th>\n",
       "      <th>...</th>\n",
       "      <th>Soil_Type32</th>\n",
       "      <th>Soil_Type33</th>\n",
       "      <th>Soil_Type34</th>\n",
       "      <th>Soil_Type35</th>\n",
       "      <th>Soil_Type36</th>\n",
       "      <th>Soil_Type37</th>\n",
       "      <th>Soil_Type38</th>\n",
       "      <th>Soil_Type39</th>\n",
       "      <th>Soil_Type40</th>\n",
       "      <th>Cover_Type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2596</td>\n",
       "      <td>51</td>\n",
       "      <td>3</td>\n",
       "      <td>258</td>\n",
       "      <td>0</td>\n",
       "      <td>510</td>\n",
       "      <td>221</td>\n",
       "      <td>232</td>\n",
       "      <td>148</td>\n",
       "      <td>6279</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2590</td>\n",
       "      <td>56</td>\n",
       "      <td>2</td>\n",
       "      <td>212</td>\n",
       "      <td>-6</td>\n",
       "      <td>390</td>\n",
       "      <td>220</td>\n",
       "      <td>235</td>\n",
       "      <td>151</td>\n",
       "      <td>6225</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2804</td>\n",
       "      <td>139</td>\n",
       "      <td>9</td>\n",
       "      <td>268</td>\n",
       "      <td>65</td>\n",
       "      <td>3180</td>\n",
       "      <td>234</td>\n",
       "      <td>238</td>\n",
       "      <td>135</td>\n",
       "      <td>6121</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2785</td>\n",
       "      <td>155</td>\n",
       "      <td>18</td>\n",
       "      <td>242</td>\n",
       "      <td>118</td>\n",
       "      <td>3090</td>\n",
       "      <td>238</td>\n",
       "      <td>238</td>\n",
       "      <td>122</td>\n",
       "      <td>6211</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2595</td>\n",
       "      <td>45</td>\n",
       "      <td>2</td>\n",
       "      <td>153</td>\n",
       "      <td>-1</td>\n",
       "      <td>391</td>\n",
       "      <td>220</td>\n",
       "      <td>234</td>\n",
       "      <td>150</td>\n",
       "      <td>6172</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 55 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Elevation  Aspect  Slope  Horizontal_Distance_To_Hydrology  \\\n",
       "0       2596      51      3                               258   \n",
       "1       2590      56      2                               212   \n",
       "2       2804     139      9                               268   \n",
       "3       2785     155     18                               242   \n",
       "4       2595      45      2                               153   \n",
       "\n",
       "   Vertical_Distance_To_Hydrology  Horizontal_Distance_To_Roadways  \\\n",
       "0                               0                              510   \n",
       "1                              -6                              390   \n",
       "2                              65                             3180   \n",
       "3                             118                             3090   \n",
       "4                              -1                              391   \n",
       "\n",
       "   Hillshade_9am  Hillshade_Noon  Hillshade_3pm  \\\n",
       "0            221             232            148   \n",
       "1            220             235            151   \n",
       "2            234             238            135   \n",
       "3            238             238            122   \n",
       "4            220             234            150   \n",
       "\n",
       "   Horizontal_Distance_To_Fire_Points  ...  Soil_Type32  Soil_Type33  \\\n",
       "0                                6279  ...            0            0   \n",
       "1                                6225  ...            0            0   \n",
       "2                                6121  ...            0            0   \n",
       "3                                6211  ...            0            0   \n",
       "4                                6172  ...            0            0   \n",
       "\n",
       "   Soil_Type34  Soil_Type35  Soil_Type36  Soil_Type37  Soil_Type38  \\\n",
       "0            0            0            0            0            0   \n",
       "1            0            0            0            0            0   \n",
       "2            0            0            0            0            0   \n",
       "3            0            0            0            0            0   \n",
       "4            0            0            0            0            0   \n",
       "\n",
       "   Soil_Type39  Soil_Type40  Cover_Type  \n",
       "0            0            0           5  \n",
       "1            0            0           5  \n",
       "2            0            0           2  \n",
       "3            0            0           2  \n",
       "4            0            0           5  \n",
       "\n",
       "[5 rows x 55 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# remove ID column from set\n",
    "train = train.iloc[:, 1:]\n",
    "train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "6f7053cd",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:25.816510Z",
     "iopub.status.busy": "2021-12-12T04:16:25.815753Z",
     "iopub.status.idle": "2021-12-12T04:16:25.831114Z",
     "shell.execute_reply": "2021-12-12T04:16:25.831646Z",
     "shell.execute_reply.started": "2021-12-12T04:14:34.640338Z"
    },
    "papermill": {
     "duration": 0.041076,
     "end_time": "2021-12-12T04:16:25.831821",
     "exception": false,
     "start_time": "2021-12-12T04:16:25.790745",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# add new features from feature engineering\n",
    "train['Elev_to_Horizontal_Hyd'] = train.Elevation - 0.2 * train.Horizontal_Distance_To_Hydrology \n",
    "train['Elev_to_Horizontal_Road'] = train.Elevation - 0.05 * train.Horizontal_Distance_To_Roadways  \n",
    "train['Elev_to_Verticle_Hyd'] = train.Elevation - train.Vertical_Distance_To_Hydrology \n",
    "train['Mean_Horizontal_Dist'] = (train.Horizontal_Distance_To_Fire_Points + train.Horizontal_Distance_To_Hydrology + \n",
    "                                 train.Horizontal_Distance_To_Roadways)/3 \n",
    "train['Mean_Fire_Hydro'] = (train.Horizontal_Distance_To_Fire_Points + train.Horizontal_Distance_To_Hydrology)/2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "4dac93d8",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:25.878700Z",
     "iopub.status.busy": "2021-12-12T04:16:25.878094Z",
     "iopub.status.idle": "2021-12-12T04:16:25.900806Z",
     "shell.execute_reply": "2021-12-12T04:16:25.901316Z",
     "shell.execute_reply.started": "2021-12-12T04:14:34.658661Z"
    },
    "papermill": {
     "duration": 0.047599,
     "end_time": "2021-12-12T04:16:25.901510",
     "exception": false,
     "start_time": "2021-12-12T04:16:25.853911",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Cover_Type</th>\n",
       "      <th>Elevation</th>\n",
       "      <th>Aspect</th>\n",
       "      <th>Slope</th>\n",
       "      <th>Horizontal_Distance_To_Hydrology</th>\n",
       "      <th>Vertical_Distance_To_Hydrology</th>\n",
       "      <th>Horizontal_Distance_To_Roadways</th>\n",
       "      <th>Hillshade_9am</th>\n",
       "      <th>Hillshade_Noon</th>\n",
       "      <th>Hillshade_3pm</th>\n",
       "      <th>...</th>\n",
       "      <th>Soil_Type36</th>\n",
       "      <th>Soil_Type37</th>\n",
       "      <th>Soil_Type38</th>\n",
       "      <th>Soil_Type39</th>\n",
       "      <th>Soil_Type40</th>\n",
       "      <th>Elev_to_Horizontal_Hyd</th>\n",
       "      <th>Elev_to_Horizontal_Road</th>\n",
       "      <th>Elev_to_Verticle_Hyd</th>\n",
       "      <th>Mean_Horizontal_Dist</th>\n",
       "      <th>Mean_Fire_Hydro</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5</td>\n",
       "      <td>2596</td>\n",
       "      <td>51</td>\n",
       "      <td>3</td>\n",
       "      <td>258</td>\n",
       "      <td>0</td>\n",
       "      <td>510</td>\n",
       "      <td>221</td>\n",
       "      <td>232</td>\n",
       "      <td>148</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2544.4</td>\n",
       "      <td>2570.50</td>\n",
       "      <td>2596</td>\n",
       "      <td>2349.000000</td>\n",
       "      <td>3268.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5</td>\n",
       "      <td>2590</td>\n",
       "      <td>56</td>\n",
       "      <td>2</td>\n",
       "      <td>212</td>\n",
       "      <td>-6</td>\n",
       "      <td>390</td>\n",
       "      <td>220</td>\n",
       "      <td>235</td>\n",
       "      <td>151</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2547.6</td>\n",
       "      <td>2570.50</td>\n",
       "      <td>2596</td>\n",
       "      <td>2275.666667</td>\n",
       "      <td>3218.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>2804</td>\n",
       "      <td>139</td>\n",
       "      <td>9</td>\n",
       "      <td>268</td>\n",
       "      <td>65</td>\n",
       "      <td>3180</td>\n",
       "      <td>234</td>\n",
       "      <td>238</td>\n",
       "      <td>135</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2750.4</td>\n",
       "      <td>2645.00</td>\n",
       "      <td>2739</td>\n",
       "      <td>3189.666667</td>\n",
       "      <td>3194.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2</td>\n",
       "      <td>2785</td>\n",
       "      <td>155</td>\n",
       "      <td>18</td>\n",
       "      <td>242</td>\n",
       "      <td>118</td>\n",
       "      <td>3090</td>\n",
       "      <td>238</td>\n",
       "      <td>238</td>\n",
       "      <td>122</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2736.6</td>\n",
       "      <td>2630.50</td>\n",
       "      <td>2667</td>\n",
       "      <td>3181.000000</td>\n",
       "      <td>3226.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>2595</td>\n",
       "      <td>45</td>\n",
       "      <td>2</td>\n",
       "      <td>153</td>\n",
       "      <td>-1</td>\n",
       "      <td>391</td>\n",
       "      <td>220</td>\n",
       "      <td>234</td>\n",
       "      <td>150</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2564.4</td>\n",
       "      <td>2575.45</td>\n",
       "      <td>2596</td>\n",
       "      <td>2238.666667</td>\n",
       "      <td>3162.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 60 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Cover_Type  Elevation  Aspect  Slope  Horizontal_Distance_To_Hydrology  \\\n",
       "0           5       2596      51      3                               258   \n",
       "1           5       2590      56      2                               212   \n",
       "2           2       2804     139      9                               268   \n",
       "3           2       2785     155     18                               242   \n",
       "4           5       2595      45      2                               153   \n",
       "\n",
       "   Vertical_Distance_To_Hydrology  Horizontal_Distance_To_Roadways  \\\n",
       "0                               0                              510   \n",
       "1                              -6                              390   \n",
       "2                              65                             3180   \n",
       "3                             118                             3090   \n",
       "4                              -1                              391   \n",
       "\n",
       "   Hillshade_9am  Hillshade_Noon  Hillshade_3pm  ...  Soil_Type36  \\\n",
       "0            221             232            148  ...            0   \n",
       "1            220             235            151  ...            0   \n",
       "2            234             238            135  ...            0   \n",
       "3            238             238            122  ...            0   \n",
       "4            220             234            150  ...            0   \n",
       "\n",
       "   Soil_Type37  Soil_Type38  Soil_Type39  Soil_Type40  Elev_to_Horizontal_Hyd  \\\n",
       "0            0            0            0            0                  2544.4   \n",
       "1            0            0            0            0                  2547.6   \n",
       "2            0            0            0            0                  2750.4   \n",
       "3            0            0            0            0                  2736.6   \n",
       "4            0            0            0            0                  2564.4   \n",
       "\n",
       "   Elev_to_Horizontal_Road  Elev_to_Verticle_Hyd  Mean_Horizontal_Dist  \\\n",
       "0                  2570.50                  2596           2349.000000   \n",
       "1                  2570.50                  2596           2275.666667   \n",
       "2                  2645.00                  2739           3189.666667   \n",
       "3                  2630.50                  2667           3181.000000   \n",
       "4                  2575.45                  2596           2238.666667   \n",
       "\n",
       "   Mean_Fire_Hydro  \n",
       "0           3268.5  \n",
       "1           3218.5  \n",
       "2           3194.5  \n",
       "3           3226.5  \n",
       "4           3162.5  \n",
       "\n",
       "[5 rows x 60 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# move target to first column\n",
    "first_column = train.pop('Cover_Type')\n",
    "  \n",
    "# insert column using insert(position,column_name,first_column) function\n",
    "train.insert(0, 'Cover_Type', first_column)\n",
    "  \n",
    "# view\n",
    "train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "3edf041c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:25.952222Z",
     "iopub.status.busy": "2021-12-12T04:16:25.951630Z",
     "iopub.status.idle": "2021-12-12T04:16:25.967574Z",
     "shell.execute_reply": "2021-12-12T04:16:25.968064Z",
     "shell.execute_reply.started": "2021-12-12T04:14:34.926113Z"
    },
    "papermill": {
     "duration": 0.043516,
     "end_time": "2021-12-12T04:16:25.968231",
     "exception": false,
     "start_time": "2021-12-12T04:16:25.924715",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# create cat, num, and y\n",
    "X_cat = train.iloc[:, 11:55].values\n",
    "B = train.iloc[:, 55:60]\n",
    "A = train.iloc[:, 1:11]\n",
    "X_num = pd.concat([A, B], axis = 1).values\n",
    "y = train.iloc[:, 0].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "9fa1826c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:26.017965Z",
     "iopub.status.busy": "2021-12-12T04:16:26.016987Z",
     "iopub.status.idle": "2021-12-12T04:16:26.026840Z",
     "shell.execute_reply": "2021-12-12T04:16:26.027398Z",
     "shell.execute_reply.started": "2021-12-12T04:14:34.940859Z"
    },
    "papermill": {
     "duration": 0.036772,
     "end_time": "2021-12-12T04:16:26.027574",
     "exception": false,
     "start_time": "2021-12-12T04:16:25.990802",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Categorical Shape: (15120, 44)\n",
      "Numerical Shape: (15120, 15)\n",
      "Label Shape: (15120,)\n"
     ]
    }
   ],
   "source": [
    "# scale/standardizing numerical columns\n",
    "# scaler object\n",
    "scaler = StandardScaler()\n",
    "# fit to training data\n",
    "scaler.fit(X_num)\n",
    "# scale num columns\n",
    "X_num = scaler.transform(X_num)\n",
    "\n",
    "# shape\n",
    "print(f'Categorical Shape: {X_cat.shape}')\n",
    "print(f'Numerical Shape: {X_num.shape}')\n",
    "print(f'Label Shape: {y.shape}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "891bab35",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:26.076583Z",
     "iopub.status.busy": "2021-12-12T04:16:26.075981Z",
     "iopub.status.idle": "2021-12-12T04:16:26.085252Z",
     "shell.execute_reply": "2021-12-12T04:16:26.085720Z",
     "shell.execute_reply.started": "2021-12-12T04:14:34.956586Z"
    },
    "papermill": {
     "duration": 0.035594,
     "end_time": "2021-12-12T04:16:26.085995",
     "exception": false,
     "start_time": "2021-12-12T04:16:26.050401",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(15120, 59)\n"
     ]
    }
   ],
   "source": [
    "# combine num and cat\n",
    "X = np.hstack((X_num, X_cat))\n",
    "print(X.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "c8ae64f5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:26.140741Z",
     "iopub.status.busy": "2021-12-12T04:16:26.139962Z",
     "iopub.status.idle": "2021-12-12T04:16:26.152160Z",
     "shell.execute_reply": "2021-12-12T04:16:26.152921Z",
     "shell.execute_reply.started": "2021-12-12T04:14:34.967012Z"
    },
    "papermill": {
     "duration": 0.044063,
     "end_time": "2021-12-12T04:16:26.153147",
     "exception": false,
     "start_time": "2021-12-12T04:16:26.109084",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(12096, 59)\n",
      "(3024, 59)\n",
      "(12096,)\n",
      "(3024,)\n"
     ]
    }
   ],
   "source": [
    "# train/validate split\n",
    "X_train, X_valid, y_train, y_valid = train_test_split(X, y, test_size = .20, random_state = 1)\n",
    "print(X_train.shape)\n",
    "print(X_valid.shape)\n",
    "print(y_train.shape)\n",
    "print(y_valid.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "02332231",
   "metadata": {
    "papermill": {
     "duration": 0.023766,
     "end_time": "2021-12-12T04:16:26.201775",
     "exception": false,
     "start_time": "2021-12-12T04:16:26.178009",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Load Model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d9b41f94",
   "metadata": {
    "papermill": {
     "duration": 0.023773,
     "end_time": "2021-12-12T04:16:26.250080",
     "exception": false,
     "start_time": "2021-12-12T04:16:26.226307",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Below, we will load in the best model from our Training Model Notebook and fit it to the training set. Normally, we would not refit the model, but we are interested in exploring the strength of the model by splitting the training set into a training and validation set, so we need to refit it to the training portion of the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "f7ee6621",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:26.303087Z",
     "iopub.status.busy": "2021-12-12T04:16:26.302408Z",
     "iopub.status.idle": "2021-12-12T04:16:36.130934Z",
     "shell.execute_reply": "2021-12-12T04:16:36.130086Z",
     "shell.execute_reply.started": "2021-12-12T04:14:34.983909Z"
    },
    "papermill": {
     "duration": 9.856322,
     "end_time": "2021-12-12T04:16:36.131163",
     "exception": false,
     "start_time": "2021-12-12T04:16:26.274841",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# load final model\n",
    "final_model = joblib.load('../input/g2-forestcovertype-finalmodels/tree_model_final.joblib')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "609053c4",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:36.184930Z",
     "iopub.status.busy": "2021-12-12T04:16:36.183631Z",
     "iopub.status.idle": "2021-12-12T04:16:51.516316Z",
     "shell.execute_reply": "2021-12-12T04:16:51.515808Z",
     "shell.execute_reply.started": "2021-12-12T04:14:37.211801Z"
    },
    "papermill": {
     "duration": 15.360748,
     "end_time": "2021-12-12T04:16:51.516507",
     "exception": false,
     "start_time": "2021-12-12T04:16:36.155759",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ExtraTreesClassifier(max_depth=33, n_estimators=1000, random_state=1)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# fit model to training set \n",
    "final_model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a2a00dab",
   "metadata": {
    "papermill": {
     "duration": 0.024139,
     "end_time": "2021-12-12T04:16:51.564847",
     "exception": false,
     "start_time": "2021-12-12T04:16:51.540708",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Accuracy\n",
    "As we do not have access to the labels for the test set for the Forest Cover Type competition, we are going to examine the accuracy of the model with a validation set. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "5518d11e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:51.617171Z",
     "iopub.status.busy": "2021-12-12T04:16:51.616553Z",
     "iopub.status.idle": "2021-12-12T04:16:56.076836Z",
     "shell.execute_reply": "2021-12-12T04:16:56.077272Z",
     "shell.execute_reply.started": "2021-12-12T04:14:52.151264Z"
    },
    "papermill": {
     "duration": 4.489087,
     "end_time": "2021-12-12T04:16:56.077533",
     "exception": false,
     "start_time": "2021-12-12T04:16:51.588446",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Accuracy for Final Model {1.0}\n",
      "Validation Accuracy for Final Model {0.8932}\n"
     ]
    }
   ],
   "source": [
    "# score for training set\n",
    "train_acc = final_model.score(X_train, y_train)\n",
    "# score for validation set\n",
    "valid_acc = final_model.score(X_valid, y_valid)\n",
    "\n",
    "print('Training Accuracy for Final Model', {round(train_acc, 4)})\n",
    "print('Validation Accuracy for Final Model', {round(valid_acc, 4)})"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b9cab248",
   "metadata": {
    "papermill": {
     "duration": 0.024625,
     "end_time": "2021-12-12T04:16:56.127201",
     "exception": false,
     "start_time": "2021-12-12T04:16:56.102576",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "The training accuracy for the model is 1.0, with the validation accuracy at .8932. This means that the model is overfitting the training set. \n",
    "The validation score is .8932 and the test score accuracy is .7962. This means that the model is fitting the validation set better than the test set. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "380faf73",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:56.182052Z",
     "iopub.status.busy": "2021-12-12T04:16:56.181398Z",
     "iopub.status.idle": "2021-12-12T04:16:58.247230Z",
     "shell.execute_reply": "2021-12-12T04:16:58.246677Z",
     "shell.execute_reply.started": "2021-12-12T04:14:56.643268Z"
    },
    "papermill": {
     "duration": 2.095272,
     "end_time": "2021-12-12T04:16:58.247406",
     "exception": false,
     "start_time": "2021-12-12T04:16:56.152134",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# predictions\n",
    "valid_pred = final_model.predict(X_valid)\n",
    "# prob predictions\n",
    "valid_proba = final_model.predict_proba(X_valid)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f7e1a44a",
   "metadata": {
    "papermill": {
     "duration": 0.024115,
     "end_time": "2021-12-12T04:16:58.296155",
     "exception": false,
     "start_time": "2021-12-12T04:16:58.272040",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Top K-Accuracy"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "63910745",
   "metadata": {
    "papermill": {
     "duration": 0.025836,
     "end_time": "2021-12-12T04:16:58.346102",
     "exception": false,
     "start_time": "2021-12-12T04:16:58.320266",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "We will now explore the accuracy of the final model by calculating the top number of successes of finding the actual label in the top 2 and 3 predicted labels."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "21f3b94e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:58.401162Z",
     "iopub.status.busy": "2021-12-12T04:16:58.400534Z",
     "iopub.status.idle": "2021-12-12T04:16:58.403305Z",
     "shell.execute_reply": "2021-12-12T04:16:58.402797Z",
     "shell.execute_reply.started": "2021-12-12T04:14:58.872627Z"
    },
    "papermill": {
     "duration": 0.033206,
     "end_time": "2021-12-12T04:16:58.403468",
     "exception": false,
     "start_time": "2021-12-12T04:16:58.370262",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def top_k_accuracy(y_true, pred_prob, K):\n",
    "    count = 0\n",
    "    for i in range(len(y_true)):\n",
    "        p = pred_prob[i, :]          # Get predictions for current observation\n",
    "        rank = np.argsort(p) + 1     # Rank classes in increasing order; add 1 to get from 0-6 to 1-7 for label\n",
    "        correct = y_true[i]          # Get correct class.\n",
    "        if correct in rank[-K:]:     # See if correct class is in top k\n",
    "            count += 1               # Increment count if so.\n",
    "\n",
    "    return count / len(y_true)       # Return score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "169adaf9",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:58.503202Z",
     "iopub.status.busy": "2021-12-12T04:16:58.494154Z",
     "iopub.status.idle": "2021-12-12T04:16:58.506708Z",
     "shell.execute_reply": "2021-12-12T04:16:58.507329Z",
     "shell.execute_reply.started": "2021-12-12T04:14:58.883874Z"
    },
    "papermill": {
     "duration": 0.080302,
     "end_time": "2021-12-12T04:16:58.507593",
     "exception": false,
     "start_time": "2021-12-12T04:16:58.427291",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9798\n"
     ]
    }
   ],
   "source": [
    "# calculate accuracy for top 2 predictions\n",
    "top_2_accuracy = top_k_accuracy(y_valid, valid_proba, 2)\n",
    "print(round(top_2_accuracy, 4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "b4f81d64",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:58.561327Z",
     "iopub.status.busy": "2021-12-12T04:16:58.560701Z",
     "iopub.status.idle": "2021-12-12T04:16:58.589028Z",
     "shell.execute_reply": "2021-12-12T04:16:58.589561Z",
     "shell.execute_reply.started": "2021-12-12T04:14:58.949315Z"
    },
    "papermill": {
     "duration": 0.057068,
     "end_time": "2021-12-12T04:16:58.589738",
     "exception": false,
     "start_time": "2021-12-12T04:16:58.532670",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9954\n"
     ]
    }
   ],
   "source": [
    "# calculate accuracy for top 3 predictions\n",
    "top_3_accuracy = top_k_accuracy(y_valid, valid_proba, 3)\n",
    "print(round(top_3_accuracy, 4))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c1ebd4b3",
   "metadata": {
    "papermill": {
     "duration": 0.02479,
     "end_time": "2021-12-12T04:16:58.639517",
     "exception": false,
     "start_time": "2021-12-12T04:16:58.614727",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "While the model is 89.32% accurate on the validation set for predicting the correct label, the model is 97.98% accurate at predicting the label that is in the top 2 for each cover type and 99.54% accurate at predicting the label that is within the top 3 for each cover type. If selecting a model where predicting within the top 2 classifications were acceptable, then this model would be very strong. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8037044a",
   "metadata": {
    "papermill": {
     "duration": 0.024336,
     "end_time": "2021-12-12T04:16:58.688971",
     "exception": false,
     "start_time": "2021-12-12T04:16:58.664635",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Classification Report\n",
    "\n",
    "One of the ways to assess the accuracy of our model is through examining precision and recall. The classification report below includes the precision and recall for each cover type. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "78480fe9",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:58.742652Z",
     "iopub.status.busy": "2021-12-12T04:16:58.741963Z",
     "iopub.status.idle": "2021-12-12T04:16:58.754956Z",
     "shell.execute_reply": "2021-12-12T04:16:58.755462Z",
     "shell.execute_reply.started": "2021-12-12T04:14:59.014190Z"
    },
    "papermill": {
     "duration": 0.041792,
     "end_time": "2021-12-12T04:16:58.755632",
     "exception": false,
     "start_time": "2021-12-12T04:16:58.713840",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           1       0.81      0.78      0.80       434\n",
      "           2       0.83      0.74      0.78       458\n",
      "           3       0.91      0.90      0.90       437\n",
      "           4       0.96      0.98      0.97       447\n",
      "           5       0.90      0.96      0.93       418\n",
      "           6       0.90      0.94      0.92       411\n",
      "           7       0.92      0.97      0.95       419\n",
      "\n",
      "    accuracy                           0.89      3024\n",
      "   macro avg       0.89      0.90      0.89      3024\n",
      "weighted avg       0.89      0.89      0.89      3024\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# classification report\n",
    "c_report = classification_report(y_valid, valid_pred)\n",
    "print(c_report)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ca0c2093",
   "metadata": {
    "papermill": {
     "duration": 0.024934,
     "end_time": "2021-12-12T04:16:58.805916",
     "exception": false,
     "start_time": "2021-12-12T04:16:58.780982",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Precision = TP/TP + FP\n",
    "\n",
    "Recall = TP/TP + FN\n",
    "\n",
    "- The model precision was highest for Cover Type 4. \n",
    "- The model recall was highest for Cover Type 4 as well. \n",
    "- The model also yielded the highest f1-score for Cover Type 4. \n",
    "\n",
    "This makes sense as the support for Cover Type 4 was among the top 2 most common cover types in the data set. \n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0eac2e2e",
   "metadata": {
    "papermill": {
     "duration": 0.024191,
     "end_time": "2021-12-12T04:16:58.854603",
     "exception": false,
     "start_time": "2021-12-12T04:16:58.830412",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Confusion Matrix\n",
    "\n",
    "We will now examine the accuracy of the model via a confusion matrix. The confusion matrix displays the number of times the cover type was classified correctly and how many times it was misclassified as another cover type. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "010f2e46",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:58.969493Z",
     "iopub.status.busy": "2021-12-12T04:16:58.968844Z",
     "iopub.status.idle": "2021-12-12T04:16:58.985639Z",
     "shell.execute_reply": "2021-12-12T04:16:58.986072Z",
     "shell.execute_reply.started": "2021-12-12T04:14:59.032531Z"
    },
    "papermill": {
     "duration": 0.107052,
     "end_time": "2021-12-12T04:16:58.986249",
     "exception": false,
     "start_time": "2021-12-12T04:16:58.879197",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>340</td>\n",
       "      <td>57</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>68</td>\n",
       "      <td>338</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>33</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>392</td>\n",
       "      <td>13</td>\n",
       "      <td>4</td>\n",
       "      <td>27</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>440</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>401</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>16</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>385</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>12</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>405</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     1    2    3    4    5    6    7\n",
       "1  340   57    0    0    5    0   32\n",
       "2   68  338   10    0   33    8    1\n",
       "3    0    1  392   13    4   27    0\n",
       "4    0    0    5  440    0    2    0\n",
       "5    0    6    7    0  401    4    0\n",
       "6    0    3   16    5    2  385    0\n",
       "7   12    2    0    0    0    0  405"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# confusion matrix\n",
    "cm = confusion_matrix(y_valid, valid_pred)\n",
    "cm_df = pd.DataFrame(cm)\n",
    "# Change the column names\n",
    "cm_df.columns =[1, 2, 3, 4, 5, 6, 7]\n",
    "cm_df.index = [1, 2, 3, 4, 5, 6, 7]\n",
    "# display\n",
    "cm_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "501880b8",
   "metadata": {
    "papermill": {
     "duration": 0.025813,
     "end_time": "2021-12-12T04:16:59.038034",
     "exception": false,
     "start_time": "2021-12-12T04:16:59.012221",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Distribution of Class Probability Predictions"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ffbdaf1f",
   "metadata": {
    "papermill": {
     "duration": 0.024751,
     "end_time": "2021-12-12T04:16:59.089307",
     "exception": false,
     "start_time": "2021-12-12T04:16:59.064556",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "In this section, we provide some histograms displaying the distributions of probability estimates generated for each label."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "90274bc7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:59.146050Z",
     "iopub.status.busy": "2021-12-12T04:16:59.145333Z",
     "iopub.status.idle": "2021-12-12T04:16:59.157694Z",
     "shell.execute_reply": "2021-12-12T04:16:59.158145Z",
     "shell.execute_reply.started": "2021-12-12T04:14:59.055951Z"
    },
    "papermill": {
     "duration": 0.04388,
     "end_time": "2021-12-12T04:16:59.158317",
     "exception": false,
     "start_time": "2021-12-12T04:16:59.114437",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.016106</td>\n",
       "      <td>0.068118</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.915776</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.371000</td>\n",
       "      <td>0.425000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.202000</td>\n",
       "      <td>0.002000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.001083</td>\n",
       "      <td>0.002000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.996917</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.006087</td>\n",
       "      <td>0.276991</td>\n",
       "      <td>0.000054</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.716868</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.360165</td>\n",
       "      <td>0.559063</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.080688</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000083</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3019</th>\n",
       "      <td>0.011852</td>\n",
       "      <td>0.094562</td>\n",
       "      <td>0.001048</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.890405</td>\n",
       "      <td>0.002133</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3020</th>\n",
       "      <td>0.022944</td>\n",
       "      <td>0.071591</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.905412</td>\n",
       "      <td>0.000053</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3021</th>\n",
       "      <td>0.838245</td>\n",
       "      <td>0.150612</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.007143</td>\n",
       "      <td>0.001000</td>\n",
       "      <td>0.003000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3022</th>\n",
       "      <td>0.001000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.999000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3023</th>\n",
       "      <td>0.002000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.998000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3024 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             1         2         3    4         5         6         7\n",
       "0     0.000000  0.016106  0.068118  0.0  0.000000  0.915776  0.000000\n",
       "1     0.371000  0.425000  0.000000  0.0  0.202000  0.002000  0.000000\n",
       "2     0.001083  0.002000  0.000000  0.0  0.000000  0.000000  0.996917\n",
       "3     0.006087  0.276991  0.000054  0.0  0.716868  0.000000  0.000000\n",
       "4     0.360165  0.559063  0.000000  0.0  0.080688  0.000000  0.000083\n",
       "...        ...       ...       ...  ...       ...       ...       ...\n",
       "3019  0.011852  0.094562  0.001048  0.0  0.890405  0.002133  0.000000\n",
       "3020  0.022944  0.071591  0.000000  0.0  0.905412  0.000053  0.000000\n",
       "3021  0.838245  0.150612  0.000000  0.0  0.007143  0.001000  0.003000\n",
       "3022  0.001000  0.000000  0.000000  0.0  0.000000  0.000000  0.999000\n",
       "3023  0.002000  0.000000  0.000000  0.0  0.000000  0.000000  0.998000\n",
       "\n",
       "[3024 rows x 7 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# validation predictions data frame\n",
    "df_prob = pd.DataFrame(valid_proba)\n",
    "df_prob.columns = [1, 2, 3, 4, 5, 6, 7]\n",
    "df_prob"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "973a44d0",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:59.224869Z",
     "iopub.status.busy": "2021-12-12T04:16:59.213727Z",
     "iopub.status.idle": "2021-12-12T04:16:59.627119Z",
     "shell.execute_reply": "2021-12-12T04:16:59.627698Z",
     "shell.execute_reply.started": "2021-12-12T04:14:59.079666Z"
    },
    "papermill": {
     "duration": 0.443716,
     "end_time": "2021-12-12T04:16:59.627881",
     "exception": false,
     "start_time": "2021-12-12T04:16:59.184165",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# histogram chart for probabilities by cover type\n",
    "plt.figure(figsize = [12,9])\n",
    "plt.hist(df_prob)\n",
    "plt.title('Histogram for Probabilities for Each Cover Type')\n",
    "plt.ylabel('Count')\n",
    "plt.xlabel('Probability')\n",
    "plt.legend([1, 2, 3, 4, 5, 6, 7], title = 'Cover Type')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f5cede39",
   "metadata": {
    "papermill": {
     "duration": 0.027507,
     "end_time": "2021-12-12T04:16:59.683471",
     "exception": false,
     "start_time": "2021-12-12T04:16:59.655964",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "These visualizations provide us with information about the labels that our model is more likely to predict, or if there are labels that it tends to be more or less confident about."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "3a1d3389",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:16:59.743967Z",
     "iopub.status.busy": "2021-12-12T04:16:59.743282Z",
     "iopub.status.idle": "2021-12-12T04:17:00.534898Z",
     "shell.execute_reply": "2021-12-12T04:17:00.535385Z",
     "shell.execute_reply.started": "2021-12-12T04:14:59.520732Z"
    },
    "papermill": {
     "duration": 0.825482,
     "end_time": "2021-12-12T04:17:00.535557",
     "exception": false,
     "start_time": "2021-12-12T04:16:59.710075",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_prob.plot(kind = 'hist',\n",
    "        alpha = 0.7,\n",
    "        bins = 30,\n",
    "        title = 'Histogram of Probabilities by Cover Type',\n",
    "        rot = 45,\n",
    "        grid = False,\n",
    "        figsize = (12,8),\n",
    "        fontsize = 15, \n",
    "        color = ['purple', 'orange', 'gold', 'pink', 'forestgreen', 'lightblue', 'navy'])\n",
    "plt.xlabel('Probability')\n",
    "plt.ylabel(\"Count\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "69786fe2",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:17:00.598685Z",
     "iopub.status.busy": "2021-12-12T04:17:00.597704Z",
     "iopub.status.idle": "2021-12-12T04:17:00.610691Z",
     "shell.execute_reply": "2021-12-12T04:17:00.611202Z",
     "shell.execute_reply.started": "2021-12-12T04:15:00.463799Z"
    },
    "papermill": {
     "duration": 0.047841,
     "end_time": "2021-12-12T04:17:00.611392",
     "exception": false,
     "start_time": "2021-12-12T04:17:00.563551",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "maxValueIndex = df_prob.idxmax(axis=1)\n",
    "maxValues = df_prob.max(axis=1)\n",
    "pred_prob = pd.concat([maxValueIndex, maxValues], axis=1)\n",
    "df_1 = pred_prob.loc[lambda x: x[0] == 1]\n",
    "df_2 = pred_prob.loc[lambda x: x[0] == 2]\n",
    "df_3 = pred_prob.loc[lambda x: x[0] == 3]\n",
    "df_4 = pred_prob.loc[lambda x: x[0] == 4]\n",
    "df_5 = pred_prob.loc[lambda x: x[0] == 5]\n",
    "df_6 = pred_prob.loc[lambda x: x[0] == 6]\n",
    "df_7 = pred_prob.loc[lambda x: x[0] == 7]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "8ce0aedf",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:17:00.670709Z",
     "iopub.status.busy": "2021-12-12T04:17:00.669777Z",
     "iopub.status.idle": "2021-12-12T04:17:02.034390Z",
     "shell.execute_reply": "2021-12-12T04:17:02.033813Z",
     "shell.execute_reply.started": "2021-12-12T04:15:00.493881Z"
    },
    "papermill": {
     "duration": 1.395087,
     "end_time": "2021-12-12T04:17:02.034587",
     "exception": false,
     "start_time": "2021-12-12T04:17:00.639500",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 1296x648 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAARsAAAEbCAYAAADqLSAhAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAAa5UlEQVR4nO3deZRdZZ3u8e8TZgkQaMrcmIEgQ0PA7kCXCGpfo+AA62pAEaGZQdO0YGM3l16IjYJoNy5ttJ3gRoEEQQYFJAoOzNMV6AoyBmgDJGYiCVMI80343T/et8imqOFU5ez3VFWez1q16px3T799zj5P7f3uvU8pIjAzq9uIVhdgZusGh42ZFeGwMbMiHDZmVoTDxsyKcNiYWRHDPmwkhaTtBzjtPEn79DDsbyU92t24kk6V9JOBVTygOidIekHSek2a37mSTsuPp0ha2Iz55vm96XVrJkkHSFqQX4vd6liGDdygDJv8wX05bzRLJc2QNLLVdVVFxG0R8Zc9DPu3iPgsgKSJOfDWH8hyJB0laXV+LV6Q9ISkCyTtWFnenyNiZESsbmBet/e1zIg4LiLOHEi93SzzTWHf2+vWBN8GTsivxR+bMUNJH5V0q6SVkpZLukXSJ5ox77Wo6aHK9rBa0iuV56cWWP4kSR2Sns0/10ua1Nd0gzJsso9HxEhgd6Ad+NeuIwz0AzwE/SG/FlsA+wAvA7Ml7drsBTVr76hFtgEeGsiE3a23pAOBnwMXAuOA0cBXgI+vRY1rXVtE7JIDdSRwG2sCdmRE/FuBkhYDBwJbAVsDs4BL+5wqIgbdDzAP2Kfy/FvAr/PjAI4H/gQ8kds+B8wFnskr/o7KtAH8I/A48FSe14g8bDvgRuDpPOxiYFSXOr4EzAGeBS4ANs7DpgALu6sZOB24KD/+c67hhfzzgVznuyrTvh14CWjr5rU4Cri9m/ZfA7/IjyfmZaxfmeZxYCXwBHAosDPwCrA61/FcHncGcA5wLfAiKcxmAF+vridwan6N5gGHVuq4Gfhsd/UCt+a6XszL/Ew3r9vOeR7PkYLiE5VhM4AfAtfkdbkL2K6b12KjPP/OZT3W4LzftN5d5qn83p3cy3Y6gvRHcD6wjBRKW+RhvyGFQHX8+4BP5sc7AdflbeFR4KBGa+syzzdef+BB0h/pzmEb5PdsN9ZsI9NIYbEE+N9d1uUU4DHS5+FyYKsGPqvrkz6PL/U5bquDpa+wAcbnDeXMSnhcR0rVTYAP5Rd097zRfR+4tUvY3JTHnwD8d+XN2R74cJ6ujfTh+G6XOh7MNWwF3EGXD2EDYdP5Jq9fGfdHwDcrz08EftXDa3EU3YfNMcDSrssANgWeB/4yDxsD7NLTvPKGvQJ4X97gNuatYbMKODu/Th8gfQA6538zPYRN5fXfvvL8jdeN9GGYSwqyDfN7ubIy7xmkDX+PvG4XA5f2st28sawG5/2m9e4yr53y/LbtZXnH5GW8ExgJXAn8NA87ArijMu4kUuhtlN+jBcDReb12I23DkxqprZew+RfgssqwqcADXbaRS/Ly3wUsZ802eyJwJ2kPbiPg/wCX9PE5fS5vG68D/9rX53owH0b9UtJzwO3ALUB19/DfI+KZiHiZ9Ff7/Ii4JyJeJe2J7CVpYmX8b+bx/wx8FzgEICLmRsR1EfFqRCwnfaA+0KWOH0TEgoh4BvhG57RraSZwiCTl54cDP+3nPBaTArA7rwO7StokIpZERF+HFldHxB0R8XpEvNLDOKfl1+kW0p7GQf2stzt7kj6kZ0XEaxFxI2mPrfoaXxURd0fEKlLYTG7ivHtb77/Iv5f0soxDgbMj4vGIeIG07R2cD++vAiZL2qYy7pV5G/1fwLyIuCAiVkXqX7oC+HSDtfXkImA/SZvn591tV2dExIsR8QBpT73z9TgO+HJELMw1ng4c2FtXRUSMIh3anwD02Uc2mMNm/4gYFRHbRMTnc7B0WlB5/A7SbiwA+U1/Ghjbw/jz8zRIGi3pUkmLJD1PerO27lJHt9OujYi4i3TYNEXSTqQ9rFn9nM1Y0i5413m/SDpcOQ5YIumavIzeLOhj+LN5vp2a8jrkeSyIiNe7zLv63j1ZefwSKUCaNe/e1vvp/HtMH8uYX3k+n7SnMjoiVpJC+eA87BBSWELqW3qPpOc6f0hh9D8arK1bEbGYtPf9KUmjgH0ry+xuvtX3cRvgqko9D5MOuUf3scwXgXOBCyW9vbdxB3PY9KZ6q/pi0gsFgKRNSX+VFlXGGV95PCFPA2lvKUj9J5sDh5GO1Wlg2oHUWjUzL+9wUt9Lo3+9Oh1A6hx86wIjfhcRHyZ9UB4BftxHLX3d+r9lfl07VV+HF4G3VYZVPzB9WQyMl1TdDifw5vduoBqZd2/r/Sjpg/mpPpaxTeX5BNJhxdL8/BLSHuxepMPTm3L7AuCW/Me082dkRPxDg7X1pnO7+jTpxELX17Kn7XkBsG+XmjbuZvrujCBtA2P7GmmouwQ4WtJkSRuRAuSuiJhXGedkSVtKGk86Nr0st29G6lhcIWkscHI38z9e0jhJWwFfrkzbqOWkw5p3dmm/iBQYh5E6FvskaT1J20r6Pqnv44xuxhktaWoOh1dJ69f5130pME7Shv1cB4AzJG0o6W9JhwE/z+33Ap+U9LZ8ivvYLtMt5a3r3qlzD+9fJG0gaQrpTE/fZzb6tlbzjtQp8c/AaZKOlrS5pBGS3i9peh7tEuCf8nsykrTtXZYP+SB18G4DfC23d74PvwZ2lHR4rm0DSe+WtPParza/JPVfnkj329Vp+b3ahdRn1Lk9nwt8o/OwT1KbpKndLUDShyXtlrfHzUndD8+S9oZ6NOTDJiKuB04jHfMuIZ1hOrjLaFcDs0kfjGuA83L7GaQ3ZkVuv7KbRfwM+D3p7M5jwNf7Wd9LpL6eO/Iu6p65fQFwD+kvWLd7KBV7SXqB1PF7M7A58O583N3VCNKHZDHpMOsDQOdfzBtJne1PSnqqH6vxJGljWkzaLT8uIh7Jw74DvEYKlZm8dbf9dGBmXvc39fNExGukANiX1EH6I+CIyrwHrBnzjohfkA5JjyGt+1LS+391HuV8Up/IraSzfq8AX6hM/yppm9qHtB11tq8EPkLaTheTXt9vkjpm10rubrgC2Jbut+dbSJ3aNwDfjojf5/b/JB3K/17SSlJn8Xt6WMwoUtCuIH0mtgM+1tfeuXKvsrWApPOBxRHxlmuIzAZK0leAHSPisErbRFIgblDZ8ypqXbkobtDJb/4nSac9zZoiH+4fS+oLHFRqO4yStLGkuyXdly+vPiO3byvpLklzJV3W2X8gaaP8fG4ePrGu2lpN0pmk63e+FRFPtLoeGx4kfY7U0fubiLi11fV0VdthVL6GZNOIeEHSBqTrZU4k9SdcGRGXSjoXuC8izpH0eeCvIuI4SQcDB0TEZ2opzsyKq23PJpIX8tMN8k+QruT8RW6fCeyfH0/Nz8nD965c9GZmQ1ytfTb5BrLZpIvWfkjquX6u0kG1kDXn5seSLziKiFWSVpCul3mqyzynke7vYNNNN/2bnXbq63o1M1tbs2fPfioi2tZmHrWGTaSvPJicr2a8inS/ydrOczowHaC9vT06OjrWdpZm1gdJ8/seq3dFrrOJiOdIV0/uBYyq3G8xjjVXdC4iX92Yh2/BmkvGzWyIq/NsVFveo0HSJqS7qx8mhc6BebQjWXOB1Kz8nDz8xvBFQGbDRp2HUWNIV46uRwq1yyPi15LmAJdK+jrpTtHOq3nPA34qqfN7abpeBWxmQ1htYRMR99PNBWsR8Tjp+0m6tr/Cm2+xN7NhZMjfG2VmQ4PDxsyKcNiYWREOGzMrwmFjZkU4bMysCIeNmRXhsDGzIhw2ZlaEw8bMinDYmFkRDhszK8JhY2ZFOGzMrAiHjZkV4bAxsyIcNmZWhMPGzIpw2JhZEQ4bMyvCYWNmRThszKwIh42ZFeGwMbMiHDZmVoTDxsyKcNiYWREOGzMrorawkTRe0k2S5kh6SNKJuf10SYsk3Zt/9qtM8yVJcyU9KumjddVmZuWtX+O8VwEnRcQ9kjYDZku6Lg/7TkR8uzqypEnAwcAuwDuA6yXtGBGra6zRzAqpbc8mIpZExD358UrgYWBsL5NMBS6NiFcj4glgLrBHXfWZWVlF+mwkTQR2A+7KTSdIul/S+ZK2zG1jgQWVyRbSTThJmiapQ1LH8uXL6yzbzJqo9rCRNBK4AvhiRDwPnANsB0wGlgD/0Z/5RcT0iGiPiPa2trZml2tmNak1bCRtQAqaiyPiSoCIWBoRqyPideDHrDlUWgSMr0w+LreZ2TBQ59koAecBD0fE2ZX2MZXRDgAezI9nAQdL2kjStsAOwN111WdmZdV5Nup9wOHAA5LuzW2nAodImgwEMA/4e4CIeEjS5cAc0pms430mymz4qC1sIuJ2QN0MuraXab4BfKOumsysdXwFsZkV4bAxsyIcNmZWhMPGzIpw2JhZEQ4bMyvCYWNmRThszKwIh42ZFeGwMbMiHDZmVoTDxsyKcNiYWREOGzMrwmFjZkU4bMysCIeNmRXhsDGzIhw2ZlaEw8bMinDYmFkRDhszK8JhY2ZFOGzMrAiHjZkV4bAxsyIcNmZWhMPGzIqoLWwkjZd0k6Q5kh6SdGJu30rSdZL+lH9vmdsl6XuS5kq6X9LuddVmZuXVuWezCjgpIiYBewLHS5oEnALcEBE7ADfk5wD7Ajvkn2nAOTXWZmaF1RY2EbEkIu7Jj1cCDwNjganAzDzaTGD//HgqcGEkdwKjJI2pqz4zK6tIn42kicBuwF3A6IhYkgc9CYzOj8cCCyqTLcxtZjYM1B42kkYCVwBfjIjnq8MiIoDo5/ymSeqQ1LF8+fImVmpmdao1bCRtQAqaiyPiyty8tPPwKP9eltsXAeMrk4/LbW8SEdMjoj0i2tva2uor3syaqs6zUQLOAx6OiLMrg2YBR+bHRwJXV9qPyGel9gRWVA63zGyIW7/Geb8POBx4QNK9ue1U4CzgcknHAvOBg/Kwa4H9gLnAS8DRNdZmZoXVFjYRcTugHgbv3c34ARxfVz1m1lq+gtjMinDYmFkRDhszK8JhY2ZFOGzMrAiHjZkV4bAxsyIcNmZWhMPGzIpw2JhZEQ4bMyvCYWNmRThszKwIh42ZFeGwMbMiHDZmVoTDxsyKcNiYWREOGzMroqGwkfS+RtrMzHrS6J7N9xtsMzPrVq//XUHSXsB7gTZJ/1wZtDmwXp2Fmdnw0te/ctkQGJnH26zS/jxwYF1Fmdnw02vYRMQtwC2SZkTE/EI1mdkw1Og/qdtI0nRgYnWaiPhQHUWZ2fDTaNj8HDgX+Amwur5yzGy4ajRsVkXEObVWYmbDWqOnvn8l6fOSxkjaqvOn1srMbFhpNGyOBE4G/i8wO/909DaBpPMlLZP0YKXtdEmLJN2bf/arDPuSpLmSHpX00f6vipkNZg0dRkXEtgOY9wzgB8CFXdq/ExHfrjZImgQcDOwCvAO4XtKOEeH+IbNhoqGwkXREd+0R0TVIqsNulTSxwTqmApdGxKvAE5LmAnsAf2hwejMb5BrtIH535fHGwN7APbx1r6URJ+Tw6gBOiohngbHAnZVxFuY2MxsmGj2M+kL1uaRRwKUDWN45wJlA5N//ARzTnxlImgZMA5gwYcIASjCzVhjoV0y8CPS7HycilkbE6oh4Hfgx6VAJYBEwvjLquNzW3TymR0R7RLS3tbX1twQza5FG+2x+RdobgXQD5s7A5f1dmKQxEbEkPz0A6DxTNQv4maSzSR3EOwB393f+ZjZ4NdpnUz17tAqYHxELe5tA0iXAFGBrSQuBrwJTJE0mBdc84O8BIuIhSZcDc/L8j/eZKLPhRRHR91iApNGs6Si+OyKW1VZVg9rb26Ojo9fLfcysCSTNjoj2tZlHo9/UdxDpsObTwEHAXZL8FRNm1rBGD6O+DLy7c29GUhtwPfCLugozs+Gl0bNRI7ocNj3dj2nNzBres/mtpN8Bl+TnnwGurackMxuO+voO4u2B0RFxsqRPAu/Pg/4AXFx3cWY2fPS1Z/Nd4EsAEXElcCWApHflYR+vsTYzG0b66ncZHREPdG3MbRNrqcjMhqW+wmZUL8M2aWIdZjbM9RU2HZI+17VR0mdJX6BlZtaQvvpsvghcJelQ1oRLO+n/SR1QY11mNsz09X+jlgLvlfRBYNfcfE1E3Fh7ZWY2rDT6fTY3ATfVXIuZDWO+CtjMinDYmFkRDhszK8JhY2ZFOGzMrAiHjZkV4bAxsyIcNmZWhMPGzIpw2JhZEQ4bMyvCYWNmRThszKwIh42ZFeGwMbMiHDZmVkRtYSPpfEnLJD1YadtK0nWS/pR/b5nbJel7kuZKul/S7nXVZWatUeeezQzgY13aTgFuiIgdgBvyc4B9gR3yzzTgnBrrMrMWqC1sIuJW4JkuzVOBmfnxTGD/SvuFkdwJjJI0pq7azKy80n02oyNiSX78JDA6Px4LLKiMtzC3mdkw0bIO4ogIIPo7naRpkjokdSxfvryGysysDqXDZmnn4VH+vSy3LwLGV8Ybl9veIiKmR0R7RLS3tbXVWqyZNU/psJkFHJkfHwlcXWk/Ip+V2hNYUTncMrNhoKH/GzUQki4BpgBbS1oIfBU4C7hc0rHAfOCgPPq1wH7AXOAl4Oi66jKz1qgtbCLikB4G7d3NuAEcX1ctZtZ6voLYzIpw2JhZEQ4bMyvCYWNmRThszKwIh42ZFeGwMbMiHDZmVoTDxsyKcNiYWREOGzMrorZ7o8yGpEfU3Pnt1O+vbBq2vGdjZkU4bMysCIeNmRXhsDGzItxBbFYndzi/wXs2ZlaEw8bMinDYmFkRDhszK8JhY2ZF+GyU2VAyhM9uOWxsaGv2h89q48MoMyvCYWNmRThszKwIh42ZFeGwMbMiWnI2StI8YCWwGlgVEe2StgIuAyYC84CDIuLZVtRnZs3Xyj2bD0bE5Ihoz89PAW6IiB2AG/JzMxsmBtNh1FRgZn48E9i/daWYWbO1KmwC+L2k2ZKm5bbREbEkP34SGN3dhJKmSeqQ1LF8+fIStZpZE7TqCuL3R8QiSW8HrpP0SHVgRISkbq+jjojpwHSA9vb2oftNQmbrmJbs2UTEovx7GXAVsAewVNIYgPx7WStqM7N6FA8bSZtK2qzzMfAR4EFgFnBkHu1I4OrStZlZfVpxGDUauEpS5/J/FhG/lfRfwOWSjgXmAwe1oDYzq0nxsImIx4G/7qb9aWDv0vVYA3xntTXBYDr1bWbDmMPGzIpw2JhZEQ4bMyvCYWNmRThszKwIh42ZFeGwMbMiHDZmVoT/b9Rw5Ct+bRDyno2ZFeGwMbMiHDZmVoT7bAYD97HYOsB7NmZWhMPGzIpw2JhZEQ4bMyvCYWNmRThszKwIh42ZFeGwMbMifFHfQPgiPLN+856NmRXhsDGzIhw2ZlaEw8bMinDYmFkRg+5slKSPAf8JrAf8JCLOWuuZ+uyRWcsNqj0bSesBPwT2BSYBh0ia1NqqzKwZBlXYAHsAcyPi8Yh4DbgUmNrimsysCQbbYdRYYEHl+ULgPdURJE0DpuWnL0h6tFBtjdoaeKrVRRSwLqznOrCOgsbWc5u1XdJgC5s+RcR0YHqr6+iJpI6IaG91HXVbF9ZzXVhHKLeeg+0wahEwvvJ8XG4zsyFusIXNfwE7SNpW0obAwcCsFtdkZk0wqA6jImKVpBOA35FOfZ8fEQ+1uKz+GrSHeE22LqznurCOUGg9FREllmNm67jBdhhlZsOUw8bMinDYDJCkj0l6VNJcSaf0Mt6nJIWkIXcKtZF1lHSQpDmSHpL0s9I1NkNf6ylpgqSbJP1R0v2S9mtFnWtD0vmSlkl6sIfhkvS9/BrcL2n3phcREf7p5w+p8/ox4J3AhsB9wKRuxtsMuBW4E2hvdd3NXkdgB+CPwJb5+dtbXXdN6zkd+If8eBIwr9V1D2A9/yewO/BgD8P3A35DuspvT+CuZtfgPZuBafS2ijOBbwKvlCyuSRpZx88BP4yIZwEiYlnhGpuhkfUMYPP8eAtgccH6miIibgWe6WWUqcCFkdwJjJI0ppk1OGwGprvbKsZWR8i7oeMj4pqShTVRn+sI7AjsKOkOSXfmO/aHmkbW83TgMEkLgWuBL5QprahGXoe14rCpgaQRwNnASa2upWbrkw6lpgCHAD+WNKqVBdXkEGBGRIwjHW78NL/H1g9+wQamr9sqNgN2BW6WNI90DDxriHUSN3LryEJgVkT8v4h4AvhvUvgMJY2s57HA5QAR8QdgY9LNi8NJ7bcKOWwGptfbKiJiRURsHRETI2IiqYP4ExHR0ZpyB6SRW0d+SdqrQdLWpMOqxwvW2AyNrOefgb0BJO1MCpvlRaus3yzgiHxWak9gRUQsaeYCBtXtCkNF9HBbhaSvAR0RMeTv52pwHX8HfETSHGA1cHJEPN26qvuvwfU8iXSI+E+kzuKjIp/CGSokXUL6w7B17nv6KrABQEScS+qL2g+YC7wEHN30GobYa2ZmQ5QPo8ysCIeNmRXhsDGzIhw2ZlaEw8bMinDY2BskrZZ0r6QHJf1c0tv6Me1Rkn7Qz+W90EP71yTtkx/f3HkxpKRrJY3KP5/vz7Ks9Rw2VvVyREyOiF2B14DjqgMlFbkuKyK+EhHXd9O+X0Q8B4wCHDZDjMPGenIbsL2kKZJukzQLmCNpY0kXSHogf7/LByvTjM97In+S9NXORkm/lDQ7f+fNtOpCJH0nt98gqS23zZB0YNeCJM3LVyqfBWyX98K+JelCSftXxrtYkv+54SDjsLG3yHsw+wIP5KbdgRMjYkfgeCAi4l2kGxRnSto4j7cH8Cngr4BPV+4FOyYi/gZoB/5R0l/k9k1JV+nuAtxCuqq1EacAj+W9sJOB84Cjcu1bAO8Fhurd9sOWw8aqNpF0L9BBuh/ovNx+d77REuD9wEUAEfEIMJ90TxTAdRHxdES8DFyZx4UUMPeR7hEbz5qbNV8HLsuPL6qM3y8RcQvp/qY2UgBeERGrBjIvq4/vjbKqlyNicrVBEsCLDU7f9d6XkDQF2AfYKyJeknQz6UbGRqbvjwuBw0g3Ujb9vh5be96zsf66DTgUQNKOwASg8/+tf1jSVpI2AfYH7iB9s92zOWh2In3dRqcRQGffzN8BtzdYw0rS13hUzQC+CBARcxpfHSvFYWP99SNghKQHSIdAR0XEq3nY3cAVwP2kQ5kO4LfA+pIeJnXs3lmZ14vAHvlLuD8EfK2RAvKd5XfkU/Tfym1LgYeBC9Z2Ba0evuvbhoV8TdADwO4RsaLV9dhbec/Ghrx8AeDDwPcdNIOX92zMrAjv2ZhZEQ4bMyvCYWNmRThszKwIh42ZFfH/AZTjRI6xF9RYAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1296x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAARsAAAEbCAYAAADqLSAhAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAAa1ElEQVR4nO3de5RdZZ3m8e/DHQkQaMpMSAJBASFgD9AlgthjFLyQNRpuIgx30DQt2Ggz9AIcFER7dGmD4w0mNJAgyEUJEgVF7gFGoCsIhIu0ARKTEJJwC+Fuwm/+eN8ym+JU1alK7fdUFc9nrVp1zrtvv33OPk/t/e69TykiMDOr21qtLsDM3hkcNmZWhMPGzIpw2JhZEQ4bMyvCYWNmRQz7sJEUkrbt57TzJO3TzbC/l/RYo3ElnS7p3/tXcb/q3ErSS5LWHqD5nS/pjPx4oqSFAzHfPL+3vG4DSdL+khbk12LXOpZh/TcowyZ/cF/NG80SSdMkjWh1XVURcUdEvK+bYf8aEZ8HkDQ+B946/VmOpKMlrcqvxUuSnpR0saTtK8v7c0SMiIhVTczrzt6WGRHHR8TZ/am3wTLfEvY9vW4D4HvAifm1+MNAzFDSJyXNkrRC0jJJt0v6zEDMew1qeriyPayS9Frl+emFaniXpJ9IekbSckmzeptmUIZN9umIGAHsBrQD/6vrCP39AA9Bv8+vxabAPsCrwGxJOw/0ggZq76hFtgYe7s+EjdZb0kHAz4FLgLHAKOBrwKfXoMY1ri0idsqBOgK4g9UBOyIi/rVQWVOBzYEd8++v9DpFRAy6H2AesE/l+XeBX+fHAZwA/Al4Mrd9AZgLPAfMBLasTBvAPwFPAM/kea2Vh70XuAV4Ng+7DBjZpY7TgEeA54GLgQ3ysInAwkY1A2cCl+bHf841vJR/PpLrfH9l2ncDrwBtDV6Lo4E7G7T/GvhFfjw+L2OdyjRPACuAJ4HD8kbxGrAq1/FCHncacB5wPfAyKcymAd+sridwen6N5gGHVeq4Dfh8o3qBWbmul/MyP9fgddsxz+MFUlB8pjJsGvBj4Lq8LvcA723wWqyf59+5rMebnPdb1rvLPJXfu1N62E7XIv0RnA8sJYXSpnnYb0ghUB3/AeCA/HgH4Ma8LTwGHNxsbV3m+dfXH3iI9Ee6c9i6+T3bldXbyBTgKWAx8D+7rMupwOOkz8NVwObdLHMH4EVgkz59rlsdLL2FDTAubyhnV8LjRlKabgh8LL+gu+WN7ofArC5hc2sefyvgPytvzrbAx/N0baQPx/e71PFQrmFz4C66fAibCJvON3mdyrg/Ab5TeX4S8KtuXoujaRw2xwJLui4D2ChvCO/Lw0YDO3U3r7xhLwf2yhvcBrw9bFYC5+TX6SOkD0Dn/G+jm7CpvP7bVp7/9XUjfRjmkoJsvfxerqjMexppw989r9tlwBU9bDd/XVaT837Lejf4QAWwTQ/LOzYv4z3ACGAG8NM87Ejgrsq4E0iht35+jxYAx+T12pW0DU9oprYewuZfgCsrwyYDc7psI5fn5b8fWMbqbfYk4G7SHtz6wP8FLu9mmUcCc4Bzc91zgAN7+1wP5sOoX0p6AbgTuB2o7h7+74h4LiJeJf3Vvigi7ouI10l7IntKGl8Z/zt5/D8D3wcOBYiIuRFxY0S8HhHLSB+oj3Sp40cRsSAingO+1TntGpoOHCpJ+fkRwE/7OI+nSAHYyJvAzpI2jIjFEdHbocW1EXFXRLwZEa91M84Z+XW6nbSncXAf621kD9KH9NsR8UZE3ELaY6u+xtdExL0RsZIUNrsM4Lx7Wu+/yb8X97CMw4BzIuKJiHiJtO0dkg/vrwF2kbR1ZdwZeRv978C8iLg4IlZG6l+6Gvhsk7V151JgkqRN8vNG29VZEfFyRMwh7al3vh7HA1+NiIW5xjOBg7rpqhgL7EwKxC2BE4HpknbsqbjBHDb7RcTIiNg6Ir6Yg6XTgsrjLUm7sQDkN/1ZYEw348/P0yBplKQrJC2S9CLpzdqiSx0Np10TEXEP6bBpoqQdSHtYM/s4mzGkXfCu836ZdLhyPLBY0nV5GT1Z0Mvw5/N8Ow3I65DnsSAi3uwy7+p793Tl8SukABmoefe03s/m36N7Wcb8yvP5pD2VURGxghTKh+Rhh5LCElLf0gclvdD5Qwqj/9JkbQ1FxFOkve8DJY0E9q0ss9F8q+/j1sA1lXoeJR1yj2qwqFeBv5D2ft/If4BuBT7RU32DOWx6Ur1V/SnSCwWApI1If5UWVcYZV3m8VZ4G0t5SkPpPNgEOJx2r08S0/am1anpe3hGkvpdm/3p12p/UOfj2BUbcEBEfJ31Q/ghc0Estvd36v1l+XTtVX4eXgXdVhlU/ML15ChgnqbodbsVb37v+ambePa33Y6QP5oG9LGPryvOtSIecS/Lzy0l7sHuSDk9vze0LgNvzH9POnxER8Y9N1taTzu3qs6QTC11fy+625wXAvl1q2qDB9AAPNmjrtd6hGjZVlwPHSNpF0vqkALknIuZVxjlF0maSxpGOTa/M7RuTOhaXSxoDnNJg/idIGitpc+CrlWmbtYx0WPOeLu2XkgLjcFLHYq8krS1pG0k/JPV9nNVgnFGSJudweJ20fp1/3ZcAYyWt18d1ADhL0nqS/p50GPDz3H4/cEA+FbotcFyX6Zbw9nXv1LmH9y+S1pU0kXSm54p+1Deg847UOfHPwBmSjpG0iaS1JH1Y0tQ82uXAV/J7MoK07V2ZD/kgdfBuDXwjt3e+D78Gtpd0RK5tXUkf6O0wpEm/JPVfnkTj7eqM/F7tROoz6tyezwe+1XnYJ6lN0uRuljGL1Hl+mqR1JO0FfBS4oafChnzYRMRNwBmkY97FpDNMh3QZ7VpgNumDcR1wYW4/i/TGLM/tMxos4mfA70hndx4HvtnH+l4h9fXclXdR98jtC4D7SH8RGu6hVOwp6SVSx+9twCbAB/Jxd1drkT4kT5EOsz4CdP7FvIXU2f60pGf6sBpPk87GPUXaLT8+Iv6Yh50LvEEKlem8fbf9TNLx/AuS3tLPExFvkAJgX1JH40+AIyvz7reBmHdE/IJ0SHosad2XkN7/a/MoF5H6RGaRzvq9BnypMv3rpG1qH9J21Nm+gnTIcUie79PAd0gds2skdzdcDWxD4+35dlKn9s3A9yLid7n9/5AO5X8naQWps/iD3SzjL6TO50mkz84FNPHaKvcuWwtIugh4KiLedg2RWX9J+hqwfUQcXmkbTwrEdSt7XkW9Uy6KG3Tym38A6bSn2YDIh/vHkfoCB5XaDqMkbSDpXkkP5Murz8rt20i6R9JcSVd29h9IWj8/n5uHj6+rtlaTdDbp+p3vRsSTra7HhgdJXyB19P4mInq9faC02g6j8jUkG0XES5LWJV0vcxKpP2FGRFwh6XzggYg4T9IXgb+NiOMlHQLsHxGfq6U4Myuutj2bSF7KT9fNP0G6kvMXuX06sF9+PDk/Jw/fu3LRm5kNcbX22eQbyGaTLlr7MelszguVDqqFrL7Iagz5gqOIWClpOel6mWe6zHMK6f4ONtpoo7/bYYferlczszU1e/bsZyKibU3mUWvYRPrKg13y1YzXkO43WdN5TiXdcUp7e3t0dHSs6SzNrBeS5vc+Vs+KXGcTES+Qrp7cExhZud9iLKuv6FxEvroxD9+U1ZeMm9kQV+fZqLa8R4OkDUl3Vz9KCp2D8mhHsfoCqZn5OXn4LeGLgMyGjToPo0aTrhxdmxRqV0XEryU9Alwh6ZvAH1h9Ne+FwE8ldX4vTdergM1sCKstbCLiQRpcsBYRT5C+n6Rr+2u89RZ7MxtGhvy9UWY2NDhszKwIh42ZFeGwMbMiHDZmVoTDxsyKcNiYWREOGzMrwmFjZkU4bMysCIeNmRXhsDGzIhw2ZlaEw8bMinDYmFkRDhszK8JhY2ZFOGzMrAiHjZkV4bAxsyIcNmZWhMPGzIpw2JhZEQ4bMyvCYWNmRThszKwIh42ZFeGwMbMiagsbSeMk3SrpEUkPSzopt58paZGk+/PPpMo0p0maK+kxSZ+sqzYzK2+dGue9Ejg5Iu6TtDEwW9KNedi5EfG96siSJgCHADsBWwI3Sdo+IlbVWKOZFVLbnk1ELI6I+/LjFcCjwJgeJpkMXBERr0fEk8BcYPe66jOzsor02UgaD+wK3JObTpT0oKSLJG2W28YACyqTLaRBOEmaIqlDUseyZcvqLNvMBlDtYSNpBHA18OWIeBE4D3gvsAuwGPi3vswvIqZGRHtEtLe1tQ10uWZWk1rDRtK6pKC5LCJmAETEkohYFRFvAhew+lBpETCuMvnY3GZmw0CdZ6MEXAg8GhHnVNpHV0bbH3goP54JHCJpfUnbANsB99ZVn5mVVefZqL2AI4A5ku7PbacDh0raBQhgHvAPABHxsKSrgEdIZ7JO8Jkos+GjtrCJiDsBNRh0fQ/TfAv4Vl01mVnr+ApiMyvCYWNmRThszKwIh42ZFeGwMbMiHDZmVoTDxsyKcNiYWREOGzMrwmFjZkU4bMysCIeNmRXhsDGzIhw2ZlaEw8bMinDYmFkRDhszK8JhY2ZFOGzMrAiHjZkV4bAxsyIcNmZWhMPGzIpw2JhZEQ4bMyvCYWNmRThszKwIh42ZFVFb2EgaJ+lWSY9IeljSSbl9c0k3SvpT/r1ZbpekH0iaK+lBSbvVVZuZlVfnns1K4OSImADsAZwgaQJwKnBzRGwH3JyfA+wLbJd/pgDn1VibmRVWW9hExOKIuC8/XgE8CowBJgPT82jTgf3y48nAJZHcDYyUNLqu+sysrCJ9NpLGA7sC9wCjImJxHvQ0MCo/HgMsqEy2MLeZ2TBQe9hIGgFcDXw5Il6sDouIAKKP85siqUNSx7JlywawUjOrU61hI2ldUtBcFhEzcvOSzsOj/Htpbl8EjKtMPja3vUVETI2I9ohob2trq694MxtQdZ6NEnAh8GhEnFMZNBM4Kj8+Cri20n5kPiu1B7C8crhlZkPcOjXOey/gCGCOpPtz2+nAt4GrJB0HzAcOzsOuByYBc4FXgGNqrM3MCqstbCLiTkDdDN67wfgBnFBXPWbWWr6C2MyKcNiYWREOGzMrwmFjZkU4bMysCIeNmRXhsDGzIhw2ZlaEw8bMinDYmFkRDhszK8JhY2ZFOGzMrAiHjZkV4bAxsyIcNmZWhMPGzIpw2JhZEQ4bMyuiqbCRtFczbWZm3Wl2z+aHTbaZmTXU439XkLQn8CGgTdI/VwZtAqxdZ2FmNrz09q9c1gNG5PE2rrS/CBxUV1FmNvz0GDYRcTtwu6RpETG/UE1mNgw1+0/q1pc0FRhfnSYiPlZHUWY2/DQbNj8Hzgf+HVhVXzlmNlw1GzYrI+K8Wisxs2Gt2VPfv5L0RUmjJW3e+VNrZWY2rDQbNkcBpwD/D5idfzp6mkDSRZKWSnqo0nampEWS7s8/kyrDTpM0V9Jjkj7Z91Uxs8GsqcOoiNimH/OeBvwIuKRL+7kR8b1qg6QJwCHATsCWwE2Sto8I9w+ZDRNNhY2kIxu1R0TXIKkOmyVpfJN1TAauiIjXgSclzQV2B37f5PRmNsg120H8gcrjDYC9gft4+15LM07M4dUBnBwRzwNjgLsr4yzMbWY2TDR7GPWl6nNJI4Er+rG884Czgci//w04ti8zkDQFmAKw1VZb9aMEM2uF/n7FxMtAn/txImJJRKyKiDeBC0iHSgCLgHGVUcfmtkbzmBoR7RHR3tbW1tcSzKxFmu2z+RVpbwTSDZg7Alf1dWGSRkfE4vx0f6DzTNVM4GeSziF1EG8H3NvX+ZvZ4NVsn0317NFKYH5ELOxpAkmXAxOBLSQtBL4OTJS0Cym45gH/ABARD0u6Cngkz/8En4kyG14UEb2PBUgaxeqO4nsjYmltVTWpvb09Ojp6vNzHzAaApNkR0b4m82j2m/oOJh3WfBY4GLhHkr9iwsya1uxh1FeBD3TuzUhqA24CflFXYWY2vDR7NmqtLodNz/ZhWjOzpvdsfivpBuDy/PxzwPX1lGRmw1Fv30G8LTAqIk6RdADw4Tzo98BldRdnZsNHb3s23wdOA4iIGcAMAEnvz8M+XWNtZjaM9NbvMioi5nRtzG3ja6nIzIal3sJmZA/DNhzAOsxsmOstbDokfaFro6TPk75Ay8ysKb312XwZuEbSYawOl3bS/5Pav8a6zGyY6e3/Ri0BPiTpo8DOufm6iLil9srMbFhp9vtsbgVurbkWMxvGfBWwmRXhsDGzIhw2ZlaEw8bMinDYmFkRDhszK8JhY2ZFOGzMrAiHjZkV4bAxsyIcNmZWhMPGzIpw2JhZEQ4bMyvCYWNmRThszKyI2sJG0kWSlkp6qNK2uaQbJf0p/94st0vSDyTNlfSgpN3qqsvMWqPOPZtpwKe6tJ0K3BwR2wE35+cA+wLb5Z8pwHk11mVmLVBb2ETELOC5Ls2Tgen58XRgv0r7JZHcDYyUNLqu2sysvNJ9NqMiYnF+/DQwKj8eAyyojLcwt5nZMNGyDuKICCD6Op2kKZI6JHUsW7ashsrMrA6lw2ZJ5+FR/r00ty8CxlXGG5vb3iYipkZEe0S0t7W11VqsmQ2c0mEzEzgqPz4KuLbSfmQ+K7UHsLxyuGVmw0BT/zeqPyRdDkwEtpC0EPg68G3gKknHAfOBg/Po1wOTgLnAK8AxddVlZq1RW9hExKHdDNq7wbgBnFBXLWbWer6C2MyKcNiYWREOGzMrwmFjZkU4bMysCIeNmRXhsDGzImq7zsbMYMZjA3sh/AHvG7pfhuA9GzMrwmFjZkU4bMysCPfZmFUMdB+LreY9GzMrwmFjZkU4bMysCIeNmRXhsDGzIhw2ZlaEw8bMinDYmFkRvqjPbAgZyjd2es/GzIpw2JhZEQ4bMyvCYWNmRThszKwIh42ZFeGwMbMiHDZmVkRLLuqTNA9YAawCVkZEu6TNgSuB8cA84OCIeL4V9dnQ4W/WGzpauWfz0YjYJSLa8/NTgZsjYjvg5vzczIaJwXQYNRmYnh9PB/ZrXSlmNtBaFTYB/E7SbElTctuoiOjcJ34aGNVoQklTJHVI6li2bFmJWs1sALTqRswPR8QiSe8GbpT0x+rAiAhJ0WjCiJgKTAVob29vOI6ZDT4t2bOJiEX591LgGmB3YImk0QD599JW1GZm9SgeNpI2krRx52PgE8BDwEzgqDzaUcC1pWszs/q04jBqFHCNpM7l/ywifivpP4CrJB0HzAcObkFtVjOfqn7nKh42EfEE8F8btD8L7F26HjMrYzCd+jazYcxhY2ZFOGzMrAiHjZkV4bAxsyL8r1ysVz5dbQPBezZmVoTDxsyKcNiYWREOGzMrwh3Eg8BQ/v/NZs3yno2ZFeE9m2HIp6ptMPKejZkV4bAxsyIcNmZWhMPGzIpw2JhZEQ4bMyvCYWNmRThszKwIh42ZFeGwMbMiHDZmVoTDxsyKcNiYWRG+67sffFe1Wd+9I8LG4WDWeoPuMErSpyQ9JmmupFNbXY+ZDYxBFTaS1gZ+DOwLTAAOlTShtVWZ2UAYVGED7A7MjYgnIuIN4ApgcotrMrMBMNj6bMYACyrPFwIfrI4gaQowJT99SdJjNde0BfBMzcsYjLze7yy9rffWa7qAwRY2vYqIqcDUUsuT1BER7aWWN1h4vd9ZSqz3YDuMWgSMqzwfm9vMbIgbbGHzH8B2kraRtB5wCDCzxTWZ2QAYVIdREbFS0onADcDawEUR8XCLyyp2yDbIeL3fWWpfb0VE3cswMxt0h1FmNkw5bMysCIcNzd8iIelASSFpWJwabWa9JR0s6RFJD0v6Weka69DbekvaStKtkv4g6UFJk1pR50CTdJGkpZIe6ma4JP0gvy4PStptQAuIiHf0D6kj+nHgPcB6wAPAhAbjbQzMAu4G2ltdd4n1BrYD/gBslp+/u9V1F1rvqcA/5scTgHmtrnuA1v2/AbsBD3UzfBLwG0DAHsA9A7l879k0f4vE2cB3gNdKFlejZtb7C8CPI+J5gIhYWrjGOjSz3gFskh9vCjxVsL7aRMQs4LkeRpkMXBLJ3cBISaMHavkOm8a3SIypjpB3J8dFxHUlC6tZr+sNbA9sL+kuSXdL+lSx6urTzHqfCRwuaSFwPfClMqW1XDOvTb85bHohaS3gHODkVtfSAuuQDqUmAocCF0ga2cqCCjkUmBYRY0mHFj/N24GtAb+Avd8isTGwM3CbpHmkY9mZw6CTuJlbQxYCMyPiLxHxJPCfpPAZyppZ7+OAqwAi4vfABqQbFYe7Wm8Xctj0cotERCyPiC0iYnxEjCd1EH8mIjpaU+6AaebWkF+S9mqQtAXpsOqJgjXWoZn1/jOwN4CkHUlhs6xola0xEzgyn5XaA1geEQP2NZeD6naFVohubpGQ9A2gIyKG5b1ZTa73DcAnJD0CrAJOiYhnW1f1mmtyvU8mHTJ+hdRZfHTk0zVDmaTLSX88tsj9UV8H1gWIiPNJ/VOTgLnAK8AxA7r8YfAamtkQ4MMoMyvCYWNmRThszKwIh42ZFeGwMbMiHDb2V5JWSbpf0kOSfi7pXX2Y9mhJP+rj8l7qpv0bkvbJj2/rvIBS0vWSRuafL/ZlWdZ6DhurejUidomInYE3gOOrAyUVuS4rIr4WETc1aJ8UES8AIwGHzRDjsLHu3AFsK2mipDskzQQekbSBpIslzcnf9/LRyjTj8p7InyR9vbNR0i8lzc7fiTOluhBJ5+b2myW15bZpkg7qWpCkeflK5m8D7817Yd+VdImk/SrjXSbJ/9xwkHHY2NvkPZh9gTm5aTfgpIjYHjgBiIh4P+mGxemSNsjj7Q4cCPwt8NnK/WPHRsTfAe3AP0n6m9y+Eemq3Z2A20lXtDbjVODxvBd2CnAhcHSufVPgQ8BwukN/WHDYWNWGku4HOkj3B12Y2+/NN2ICfBi4FCAi/gjMJ90zBXBjRDwbEa8CM/K4kALmAdJ9ZeNYfTPnm8CV+fGllfH7JCJuJ93v1EYKwKsjYmV/5mX1ecffG2Vv8WpE7FJtkATwcpPTd733JSRNBPYB9oyIVyTdRrqxsZnp++IS4HDSjZUDek+PDQzv2Vhf3QEcBiBpe2AroPP/rX9c0uaSNgT2A+4ifdPd8zlodiB9RUentYDOvpn/AdzZZA0rSF/9UTUN+DJARDzS/OpYKQ4b66ufAGtJmkM6BDo6Il7Pw+4FrgYeJB3KdAC/BdaR9CipY/fuyrxeBnbPX8D9MeAbzRSQ7zy/K5+i/25uWwI8Cly8pito9fBd3zYs5GuC5gC7RcTyVtdjb+c9Gxvy8gWAjwI/dNAMXt6zMbMivGdjZkU4bMysCIeNmRXhsDGzIhw2ZlbE/wcd2VL5mPXVEgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1296x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# histograms of probability estimates generated for each label\n",
    "c_label = [1, 2, 3, 4, 5, 6, 7]\n",
    "palette = ['purple', 'orange', 'gold', 'pink', 'forestgreen', 'lightblue', 'navy']\n",
    "dflist = [df_1, df_2, df_3, df_4, df_5, df_6, df_7]\n",
    "plt.figure(figsize = [18, 9])\n",
    "\n",
    "for c in c_label:\n",
    "    plt.figure(figsize = [18, 9])\n",
    "    plt.subplot(2, 4, c)\n",
    "    plt.hist(dflist[c-1][1], color = palette[c-1])\n",
    "    plt.title(f'Probability Distribution for Cover Type {c_label[c-1]}')\n",
    "    plt.ylabel('Count')\n",
    "    plt.ylim(0, 300)\n",
    "    plt.xlabel('Probability')\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f4e1abaa",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-12T04:14:11.111998Z",
     "iopub.status.busy": "2021-12-12T04:14:11.111635Z",
     "iopub.status.idle": "2021-12-12T04:14:11.118964Z",
     "shell.execute_reply": "2021-12-12T04:14:11.117996Z",
     "shell.execute_reply.started": "2021-12-12T04:14:11.111950Z"
    },
    "papermill": {
     "duration": 0.031532,
     "end_time": "2021-12-12T04:17:02.099802",
     "exception": false,
     "start_time": "2021-12-12T04:17:02.068270",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "From these visualizations, we can see that our model is most confident about Cover Types 7, 4, and 5. This corresponds with the classification report data explored earlier. "
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.10"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 55.659647,
   "end_time": "2021-12-12T04:17:05.340866",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2021-12-12T04:16:09.681219",
   "version": "2.3.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}