Crop-Yield-Prediction.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "b7cf3bce",
   "metadata": {
    "papermill": {
     "duration": 0.070885,
     "end_time": "2021-12-10T09:16:19.323546",
     "exception": false,
     "start_time": "2021-12-10T09:16:19.252661",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "<img src='https://i.imgur.com/ibzHl8U.png' width=\"1200\" height=\"1000\">"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d60fb0e1",
   "metadata": {
    "papermill": {
     "duration": 0.066794,
     "end_time": "2021-12-10T09:16:19.458275",
     "exception": false,
     "start_time": "2021-12-10T09:16:19.391481",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Crop Yield Analysis\n",
    " \n",
    "The science of training machines to learn and produce models for future predictions is widely used, and not for nothing.\n",
    "Agriculture plays a critical role in the global economy. With the continuing expansion of the human population understanding worldwide crop yield is central to addressing food security challenges and reducing the impacts of climate change. \n",
    "\n",
    "Crop yield prediction is an important agricultural problem. The Agricultural yield primarily depends on weather conditions (rain, temperature, etc), pesticides and accurate information about history of crop yield is an important thing for making decisions related to agricultural risk management and future predictions.  The basic ingredients that sustain humans are similar. We eat a lot of corn, wheat, rice and other simple crops. In this project the prediction of top 10 most consumed yields all over the world is established by applying machine learning techniques. It consist of 10 most consumed crops. It is a regression problem\n",
    " \n",
    " These corps include :\n",
    "\n",
    "- Cassava                \n",
    "- Maize                  \n",
    "- Plantains and others   \n",
    "- Potatoes                \n",
    "- Rice, paddy             \n",
    "- Sorghum                \n",
    "- Soybeans               \n",
    "- Sweet potatoes       \n",
    "- Wheat                  \n",
    "- Yams             "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "361a92a2",
   "metadata": {
    "papermill": {
     "duration": 0.067389,
     "end_time": "2021-12-10T09:16:19.593203",
     "exception": false,
     "start_time": "2021-12-10T09:16:19.525814",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Gathering & Cleaning Data\n",
    "\n",
    "### Crops Yield Data:\n",
    "\n",
    "\n",
    " \n",
    "After importing required libraries, crops yield of ten most consumed crops around the world was downloaded from FAO webiste.The collected data include country, item, year starting from 1961 to 2016 and yield value. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "55694a3f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:19.732122Z",
     "iopub.status.busy": "2021-12-10T09:16:19.730966Z",
     "iopub.status.idle": "2021-12-10T09:16:19.740504Z",
     "shell.execute_reply": "2021-12-10T09:16:19.741030Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.302223Z"
    },
    "papermill": {
     "duration": 0.081106,
     "end_time": "2021-12-10T09:16:19.741336",
     "exception": false,
     "start_time": "2021-12-10T09:16:19.660230",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import numpy as np \n",
    "import pandas as pd "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "91206406",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:19.882966Z",
     "iopub.status.busy": "2021-12-10T09:16:19.881902Z",
     "iopub.status.idle": "2021-12-10T09:16:20.025489Z",
     "shell.execute_reply": "2021-12-10T09:16:20.026014Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.330133Z"
    },
    "papermill": {
     "duration": 0.215779,
     "end_time": "2021-12-10T09:16:20.026207",
     "exception": false,
     "start_time": "2021-12-10T09:16:19.810428",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(56717, 12)"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_yield = pd.read_csv('../input/crop-yield-prediction-dataset/yield.csv')\n",
    "df_yield.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "b606902d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:20.173451Z",
     "iopub.status.busy": "2021-12-10T09:16:20.172470Z",
     "iopub.status.idle": "2021-12-10T09:16:20.190359Z",
     "shell.execute_reply": "2021-12-10T09:16:20.190845Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.456645Z"
    },
    "papermill": {
     "duration": 0.09382,
     "end_time": "2021-12-10T09:16:20.191032",
     "exception": false,
     "start_time": "2021-12-10T09:16:20.097212",
     "status": "completed"
    },
    "scrolled": true,
    "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>Domain Code</th>\n",
       "      <th>Domain</th>\n",
       "      <th>Area Code</th>\n",
       "      <th>Area</th>\n",
       "      <th>Element Code</th>\n",
       "      <th>Element</th>\n",
       "      <th>Item Code</th>\n",
       "      <th>Item</th>\n",
       "      <th>Year Code</th>\n",
       "      <th>Year</th>\n",
       "      <th>Unit</th>\n",
       "      <th>Value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>QC</td>\n",
       "      <td>Crops</td>\n",
       "      <td>2</td>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>5419</td>\n",
       "      <td>Yield</td>\n",
       "      <td>56</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1961</td>\n",
       "      <td>1961</td>\n",
       "      <td>hg/ha</td>\n",
       "      <td>14000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>QC</td>\n",
       "      <td>Crops</td>\n",
       "      <td>2</td>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>5419</td>\n",
       "      <td>Yield</td>\n",
       "      <td>56</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1962</td>\n",
       "      <td>1962</td>\n",
       "      <td>hg/ha</td>\n",
       "      <td>14000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>QC</td>\n",
       "      <td>Crops</td>\n",
       "      <td>2</td>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>5419</td>\n",
       "      <td>Yield</td>\n",
       "      <td>56</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1963</td>\n",
       "      <td>1963</td>\n",
       "      <td>hg/ha</td>\n",
       "      <td>14260</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>QC</td>\n",
       "      <td>Crops</td>\n",
       "      <td>2</td>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>5419</td>\n",
       "      <td>Yield</td>\n",
       "      <td>56</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1964</td>\n",
       "      <td>1964</td>\n",
       "      <td>hg/ha</td>\n",
       "      <td>14257</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>QC</td>\n",
       "      <td>Crops</td>\n",
       "      <td>2</td>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>5419</td>\n",
       "      <td>Yield</td>\n",
       "      <td>56</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1965</td>\n",
       "      <td>1965</td>\n",
       "      <td>hg/ha</td>\n",
       "      <td>14400</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Domain Code Domain  Area Code         Area  Element Code Element  Item Code  \\\n",
       "0          QC  Crops          2  Afghanistan          5419   Yield         56   \n",
       "1          QC  Crops          2  Afghanistan          5419   Yield         56   \n",
       "2          QC  Crops          2  Afghanistan          5419   Yield         56   \n",
       "3          QC  Crops          2  Afghanistan          5419   Yield         56   \n",
       "4          QC  Crops          2  Afghanistan          5419   Yield         56   \n",
       "\n",
       "    Item  Year Code  Year   Unit  Value  \n",
       "0  Maize       1961  1961  hg/ha  14000  \n",
       "1  Maize       1962  1962  hg/ha  14000  \n",
       "2  Maize       1963  1963  hg/ha  14260  \n",
       "3  Maize       1964  1964  hg/ha  14257  \n",
       "4  Maize       1965  1965  hg/ha  14400  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_yield.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b637c6eb",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:20.343094Z",
     "iopub.status.busy": "2021-12-10T09:16:20.332064Z",
     "iopub.status.idle": "2021-12-10T09:16:20.346800Z",
     "shell.execute_reply": "2021-12-10T09:16:20.346273Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.481427Z"
    },
    "papermill": {
     "duration": 0.08719,
     "end_time": "2021-12-10T09:16:20.346953",
     "exception": false,
     "start_time": "2021-12-10T09:16:20.259763",
     "status": "completed"
    },
    "scrolled": true,
    "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>Domain Code</th>\n",
       "      <th>Domain</th>\n",
       "      <th>Area Code</th>\n",
       "      <th>Area</th>\n",
       "      <th>Element Code</th>\n",
       "      <th>Element</th>\n",
       "      <th>Item Code</th>\n",
       "      <th>Item</th>\n",
       "      <th>Year Code</th>\n",
       "      <th>Year</th>\n",
       "      <th>Unit</th>\n",
       "      <th>Value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>56712</th>\n",
       "      <td>QC</td>\n",
       "      <td>Crops</td>\n",
       "      <td>181</td>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>5419</td>\n",
       "      <td>Yield</td>\n",
       "      <td>15</td>\n",
       "      <td>Wheat</td>\n",
       "      <td>2012</td>\n",
       "      <td>2012</td>\n",
       "      <td>hg/ha</td>\n",
       "      <td>24420</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56713</th>\n",
       "      <td>QC</td>\n",
       "      <td>Crops</td>\n",
       "      <td>181</td>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>5419</td>\n",
       "      <td>Yield</td>\n",
       "      <td>15</td>\n",
       "      <td>Wheat</td>\n",
       "      <td>2013</td>\n",
       "      <td>2013</td>\n",
       "      <td>hg/ha</td>\n",
       "      <td>22888</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56714</th>\n",
       "      <td>QC</td>\n",
       "      <td>Crops</td>\n",
       "      <td>181</td>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>5419</td>\n",
       "      <td>Yield</td>\n",
       "      <td>15</td>\n",
       "      <td>Wheat</td>\n",
       "      <td>2014</td>\n",
       "      <td>2014</td>\n",
       "      <td>hg/ha</td>\n",
       "      <td>21357</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56715</th>\n",
       "      <td>QC</td>\n",
       "      <td>Crops</td>\n",
       "      <td>181</td>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>5419</td>\n",
       "      <td>Yield</td>\n",
       "      <td>15</td>\n",
       "      <td>Wheat</td>\n",
       "      <td>2015</td>\n",
       "      <td>2015</td>\n",
       "      <td>hg/ha</td>\n",
       "      <td>19826</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56716</th>\n",
       "      <td>QC</td>\n",
       "      <td>Crops</td>\n",
       "      <td>181</td>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>5419</td>\n",
       "      <td>Yield</td>\n",
       "      <td>15</td>\n",
       "      <td>Wheat</td>\n",
       "      <td>2016</td>\n",
       "      <td>2016</td>\n",
       "      <td>hg/ha</td>\n",
       "      <td>18294</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Domain Code Domain  Area Code      Area  Element Code Element  \\\n",
       "56712          QC  Crops        181  Zimbabwe          5419   Yield   \n",
       "56713          QC  Crops        181  Zimbabwe          5419   Yield   \n",
       "56714          QC  Crops        181  Zimbabwe          5419   Yield   \n",
       "56715          QC  Crops        181  Zimbabwe          5419   Yield   \n",
       "56716          QC  Crops        181  Zimbabwe          5419   Yield   \n",
       "\n",
       "       Item Code   Item  Year Code  Year   Unit  Value  \n",
       "56712         15  Wheat       2012  2012  hg/ha  24420  \n",
       "56713         15  Wheat       2013  2013  hg/ha  22888  \n",
       "56714         15  Wheat       2014  2014  hg/ha  21357  \n",
       "56715         15  Wheat       2015  2015  hg/ha  19826  \n",
       "56716         15  Wheat       2016  2016  hg/ha  18294  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_yield.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0d730969",
   "metadata": {
    "papermill": {
     "duration": 0.069111,
     "end_time": "2021-12-10T09:16:20.485765",
     "exception": false,
     "start_time": "2021-12-10T09:16:20.416654",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Looking at the columns in the csv, we can rename **Value** to **hg/ha_yield** to make it easier to recognise that this is our crops yields production value. In addition to removal of unnecessary coloumns like Area Code, Domain, Item Code, etc."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "3e3a3c1c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:20.630991Z",
     "iopub.status.busy": "2021-12-10T09:16:20.630294Z",
     "iopub.status.idle": "2021-12-10T09:16:20.671482Z",
     "shell.execute_reply": "2021-12-10T09:16:20.672107Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.497225Z"
    },
    "papermill": {
     "duration": 0.115626,
     "end_time": "2021-12-10T09:16:20.672291",
     "exception": false,
     "start_time": "2021-12-10T09:16:20.556665",
     "status": "completed"
    },
    "scrolled": true,
    "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>Domain Code</th>\n",
       "      <th>Domain</th>\n",
       "      <th>Area Code</th>\n",
       "      <th>Area</th>\n",
       "      <th>Element Code</th>\n",
       "      <th>Element</th>\n",
       "      <th>Item Code</th>\n",
       "      <th>Item</th>\n",
       "      <th>Year Code</th>\n",
       "      <th>Year</th>\n",
       "      <th>Unit</th>\n",
       "      <th>hg/ha_yield</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>QC</td>\n",
       "      <td>Crops</td>\n",
       "      <td>2</td>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>5419</td>\n",
       "      <td>Yield</td>\n",
       "      <td>56</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1961</td>\n",
       "      <td>1961</td>\n",
       "      <td>hg/ha</td>\n",
       "      <td>14000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>QC</td>\n",
       "      <td>Crops</td>\n",
       "      <td>2</td>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>5419</td>\n",
       "      <td>Yield</td>\n",
       "      <td>56</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1962</td>\n",
       "      <td>1962</td>\n",
       "      <td>hg/ha</td>\n",
       "      <td>14000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>QC</td>\n",
       "      <td>Crops</td>\n",
       "      <td>2</td>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>5419</td>\n",
       "      <td>Yield</td>\n",
       "      <td>56</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1963</td>\n",
       "      <td>1963</td>\n",
       "      <td>hg/ha</td>\n",
       "      <td>14260</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>QC</td>\n",
       "      <td>Crops</td>\n",
       "      <td>2</td>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>5419</td>\n",
       "      <td>Yield</td>\n",
       "      <td>56</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1964</td>\n",
       "      <td>1964</td>\n",
       "      <td>hg/ha</td>\n",
       "      <td>14257</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>QC</td>\n",
       "      <td>Crops</td>\n",
       "      <td>2</td>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>5419</td>\n",
       "      <td>Yield</td>\n",
       "      <td>56</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1965</td>\n",
       "      <td>1965</td>\n",
       "      <td>hg/ha</td>\n",
       "      <td>14400</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Domain Code Domain  Area Code         Area  Element Code Element  Item Code  \\\n",
       "0          QC  Crops          2  Afghanistan          5419   Yield         56   \n",
       "1          QC  Crops          2  Afghanistan          5419   Yield         56   \n",
       "2          QC  Crops          2  Afghanistan          5419   Yield         56   \n",
       "3          QC  Crops          2  Afghanistan          5419   Yield         56   \n",
       "4          QC  Crops          2  Afghanistan          5419   Yield         56   \n",
       "\n",
       "    Item  Year Code  Year   Unit  hg/ha_yield  \n",
       "0  Maize       1961  1961  hg/ha        14000  \n",
       "1  Maize       1962  1962  hg/ha        14000  \n",
       "2  Maize       1963  1963  hg/ha        14260  \n",
       "3  Maize       1964  1964  hg/ha        14257  \n",
       "4  Maize       1965  1965  hg/ha        14400  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# rename columns.\n",
    "df_yield = df_yield.rename(index=str, columns={\"Value\": \"hg/ha_yield\"})\n",
    "df_yield.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "31309c4a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:20.815236Z",
     "iopub.status.busy": "2021-12-10T09:16:20.814543Z",
     "iopub.status.idle": "2021-12-10T09:16:20.830927Z",
     "shell.execute_reply": "2021-12-10T09:16:20.831481Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.537669Z"
    },
    "papermill": {
     "duration": 0.089989,
     "end_time": "2021-12-10T09:16:20.831655",
     "exception": false,
     "start_time": "2021-12-10T09:16:20.741666",
     "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>Area</th>\n",
       "      <th>Item</th>\n",
       "      <th>Year</th>\n",
       "      <th>hg/ha_yield</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1961</td>\n",
       "      <td>14000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1962</td>\n",
       "      <td>14000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1963</td>\n",
       "      <td>14260</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1964</td>\n",
       "      <td>14257</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1965</td>\n",
       "      <td>14400</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Area   Item  Year  hg/ha_yield\n",
       "0  Afghanistan  Maize  1961        14000\n",
       "1  Afghanistan  Maize  1962        14000\n",
       "2  Afghanistan  Maize  1963        14260\n",
       "3  Afghanistan  Maize  1964        14257\n",
       "4  Afghanistan  Maize  1965        14400"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# drop unwanted columns.\n",
    "df_yield = df_yield.drop(['Year Code','Element Code','Element','Year Code','Area Code','Domain Code','Domain','Unit','Item Code'], axis=1)\n",
    "df_yield.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "0d5d4008",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:20.974471Z",
     "iopub.status.busy": "2021-12-10T09:16:20.973831Z",
     "iopub.status.idle": "2021-12-10T09:16:21.000972Z",
     "shell.execute_reply": "2021-12-10T09:16:21.001486Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.555896Z"
    },
    "papermill": {
     "duration": 0.099815,
     "end_time": "2021-12-10T09:16:21.001786",
     "exception": false,
     "start_time": "2021-12-10T09:16:20.901971",
     "status": "completed"
    },
    "scrolled": true,
    "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>Year</th>\n",
       "      <th>hg/ha_yield</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>56717.000000</td>\n",
       "      <td>56717.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1989.669570</td>\n",
       "      <td>62094.660084</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>16.133198</td>\n",
       "      <td>67835.932856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1961.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1976.000000</td>\n",
       "      <td>15680.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1991.000000</td>\n",
       "      <td>36744.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2004.000000</td>\n",
       "      <td>86213.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2016.000000</td>\n",
       "      <td>1000000.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Year     hg/ha_yield\n",
       "count  56717.000000    56717.000000\n",
       "mean    1989.669570    62094.660084\n",
       "std       16.133198    67835.932856\n",
       "min     1961.000000        0.000000\n",
       "25%     1976.000000    15680.000000\n",
       "50%     1991.000000    36744.000000\n",
       "75%     2004.000000    86213.000000\n",
       "max     2016.000000  1000000.000000"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_yield.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "a8e19469",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:21.163892Z",
     "iopub.status.busy": "2021-12-10T09:16:21.163160Z",
     "iopub.status.idle": "2021-12-10T09:16:21.175384Z",
     "shell.execute_reply": "2021-12-10T09:16:21.176419Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.581597Z"
    },
    "papermill": {
     "duration": 0.102953,
     "end_time": "2021-12-10T09:16:21.176677",
     "exception": false,
     "start_time": "2021-12-10T09:16:21.073724",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Index: 56717 entries, 0 to 56716\n",
      "Data columns (total 4 columns):\n",
      " #   Column       Non-Null Count  Dtype \n",
      "---  ------       --------------  ----- \n",
      " 0   Area         56717 non-null  object\n",
      " 1   Item         56717 non-null  object\n",
      " 2   Year         56717 non-null  int64 \n",
      " 3   hg/ha_yield  56717 non-null  int64 \n",
      "dtypes: int64(2), object(2)\n",
      "memory usage: 2.2+ MB\n"
     ]
    }
   ],
   "source": [
    "df_yield.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "39086055",
   "metadata": {
    "papermill": {
     "duration": 0.070321,
     "end_time": "2021-12-10T09:16:21.319056",
     "exception": false,
     "start_time": "2021-12-10T09:16:21.248735",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Climate Data : Rainfall \n",
    "The climatic factors include rainfall and temperature. They are abiotic components, including pesticides and soil, of the environmental factors that influence plant growth and development.\n",
    "\n",
    "\n",
    "Rainfall has a dramatic effect on agriculture. For this project rain fall per year information was gathered from World Data Bank. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "03520e28",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:21.464412Z",
     "iopub.status.busy": "2021-12-10T09:16:21.463397Z",
     "iopub.status.idle": "2021-12-10T09:16:21.488444Z",
     "shell.execute_reply": "2021-12-10T09:16:21.488943Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.603541Z"
    },
    "papermill": {
     "duration": 0.099608,
     "end_time": "2021-12-10T09:16:21.489109",
     "exception": false,
     "start_time": "2021-12-10T09:16:21.389501",
     "status": "completed"
    },
    "scrolled": true,
    "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>Area</th>\n",
       "      <th>Year</th>\n",
       "      <th>average_rain_fall_mm_per_year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>1985</td>\n",
       "      <td>327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>1986</td>\n",
       "      <td>327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>1987</td>\n",
       "      <td>327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>1989</td>\n",
       "      <td>327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>1990</td>\n",
       "      <td>327</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Area  Year average_rain_fall_mm_per_year\n",
       "0  Afghanistan  1985                           327\n",
       "1  Afghanistan  1986                           327\n",
       "2  Afghanistan  1987                           327\n",
       "3  Afghanistan  1989                           327\n",
       "4  Afghanistan  1990                           327"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_rain = pd.read_csv('../input/crop-yield-prediction-dataset/rainfall.csv')\n",
    "df_rain.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "22cbf278",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:21.636377Z",
     "iopub.status.busy": "2021-12-10T09:16:21.635299Z",
     "iopub.status.idle": "2021-12-10T09:16:21.642507Z",
     "shell.execute_reply": "2021-12-10T09:16:21.642964Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.631691Z"
    },
    "papermill": {
     "duration": 0.082355,
     "end_time": "2021-12-10T09:16:21.643137",
     "exception": false,
     "start_time": "2021-12-10T09:16:21.560782",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "df_rain = df_rain.rename(index=str, columns={\" Area\": 'Area'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "e78652aa",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:21.799574Z",
     "iopub.status.busy": "2021-12-10T09:16:21.795301Z",
     "iopub.status.idle": "2021-12-10T09:16:21.802908Z",
     "shell.execute_reply": "2021-12-10T09:16:21.803422Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.640812Z"
    },
    "papermill": {
     "duration": 0.089364,
     "end_time": "2021-12-10T09:16:21.803599",
     "exception": false,
     "start_time": "2021-12-10T09:16:21.714235",
     "status": "completed"
    },
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Index: 6727 entries, 0 to 6726\n",
      "Data columns (total 3 columns):\n",
      " #   Column                         Non-Null Count  Dtype \n",
      "---  ------                         --------------  ----- \n",
      " 0   Area                           6727 non-null   object\n",
      " 1   Year                           6727 non-null   int64 \n",
      " 2   average_rain_fall_mm_per_year  5953 non-null   object\n",
      "dtypes: int64(1), object(2)\n",
      "memory usage: 210.2+ KB\n"
     ]
    }
   ],
   "source": [
    "# check data types \n",
    "df_rain.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "f29a5b0e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:21.951923Z",
     "iopub.status.busy": "2021-12-10T09:16:21.951208Z",
     "iopub.status.idle": "2021-12-10T09:16:21.969558Z",
     "shell.execute_reply": "2021-12-10T09:16:21.970115Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.659797Z"
    },
    "papermill": {
     "duration": 0.092841,
     "end_time": "2021-12-10T09:16:21.970308",
     "exception": false,
     "start_time": "2021-12-10T09:16:21.877467",
     "status": "completed"
    },
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Index: 6727 entries, 0 to 6726\n",
      "Data columns (total 3 columns):\n",
      " #   Column                         Non-Null Count  Dtype  \n",
      "---  ------                         --------------  -----  \n",
      " 0   Area                           6727 non-null   object \n",
      " 1   Year                           6727 non-null   int64  \n",
      " 2   average_rain_fall_mm_per_year  5947 non-null   float64\n",
      "dtypes: float64(1), int64(1), object(1)\n",
      "memory usage: 210.2+ KB\n"
     ]
    }
   ],
   "source": [
    "# convert average_rain_fall_mm_per_year from object to float\n",
    "df_rain['average_rain_fall_mm_per_year'] = pd.to_numeric(df_rain['average_rain_fall_mm_per_year'],errors = 'coerce')\n",
    "df_rain.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a384c482",
   "metadata": {
    "papermill": {
     "duration": 0.071249,
     "end_time": "2021-12-10T09:16:22.113946",
     "exception": false,
     "start_time": "2021-12-10T09:16:22.042697",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Next, droping any empty rows from dataset and merge yield dataframe with rain dataframe by year and area columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "89009a87",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:22.260531Z",
     "iopub.status.busy": "2021-12-10T09:16:22.259909Z",
     "iopub.status.idle": "2021-12-10T09:16:22.266315Z",
     "shell.execute_reply": "2021-12-10T09:16:22.266847Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.685433Z"
    },
    "papermill": {
     "duration": 0.081475,
     "end_time": "2021-12-10T09:16:22.267012",
     "exception": false,
     "start_time": "2021-12-10T09:16:22.185537",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "df_rain = df_rain.dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "0352abd2",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:22.414028Z",
     "iopub.status.busy": "2021-12-10T09:16:22.413358Z",
     "iopub.status.idle": "2021-12-10T09:16:22.427886Z",
     "shell.execute_reply": "2021-12-10T09:16:22.428407Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.700070Z"
    },
    "papermill": {
     "duration": 0.089865,
     "end_time": "2021-12-10T09:16:22.428575",
     "exception": false,
     "start_time": "2021-12-10T09:16:22.338710",
     "status": "completed"
    },
    "scrolled": true,
    "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>Year</th>\n",
       "      <th>average_rain_fall_mm_per_year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>5947.000000</td>\n",
       "      <td>5947.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2001.365899</td>\n",
       "      <td>1124.743232</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>9.526335</td>\n",
       "      <td>786.257365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1985.000000</td>\n",
       "      <td>51.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1993.000000</td>\n",
       "      <td>534.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2001.000000</td>\n",
       "      <td>1010.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2010.000000</td>\n",
       "      <td>1651.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2017.000000</td>\n",
       "      <td>3240.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Year  average_rain_fall_mm_per_year\n",
       "count  5947.000000                    5947.000000\n",
       "mean   2001.365899                    1124.743232\n",
       "std       9.526335                     786.257365\n",
       "min    1985.000000                      51.000000\n",
       "25%    1993.000000                     534.000000\n",
       "50%    2001.000000                    1010.000000\n",
       "75%    2010.000000                    1651.000000\n",
       "max    2017.000000                    3240.000000"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_rain.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c18cc20d",
   "metadata": {
    "papermill": {
     "duration": 0.071358,
     "end_time": "2021-12-10T09:16:22.571715",
     "exception": false,
     "start_time": "2021-12-10T09:16:22.500357",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "The rainfall dataframe begins at 1985 and ends at 2016. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "97b319cd",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:22.718320Z",
     "iopub.status.busy": "2021-12-10T09:16:22.717686Z",
     "iopub.status.idle": "2021-12-10T09:16:22.744369Z",
     "shell.execute_reply": "2021-12-10T09:16:22.743682Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.723947Z"
    },
    "papermill": {
     "duration": 0.101174,
     "end_time": "2021-12-10T09:16:22.744516",
     "exception": false,
     "start_time": "2021-12-10T09:16:22.643342",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# merge yield dataframe with rain dataframe by year and area columns \n",
    "yield_df = pd.merge(df_yield, df_rain, on=['Year','Area'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cea1c597",
   "metadata": {
    "papermill": {
     "duration": 0.072425,
     "end_time": "2021-12-10T09:16:22.889516",
     "exception": false,
     "start_time": "2021-12-10T09:16:22.817091",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Now, we view the final shape of the dataframe and info of values:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "16ed46ec",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:23.038829Z",
     "iopub.status.busy": "2021-12-10T09:16:23.038077Z",
     "iopub.status.idle": "2021-12-10T09:16:23.040916Z",
     "shell.execute_reply": "2021-12-10T09:16:23.041436Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.750914Z"
    },
    "papermill": {
     "duration": 0.080068,
     "end_time": "2021-12-10T09:16:23.041605",
     "exception": false,
     "start_time": "2021-12-10T09:16:22.961537",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(25385, 5)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yield_df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "4e11c6fe",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:23.198782Z",
     "iopub.status.busy": "2021-12-10T09:16:23.198107Z",
     "iopub.status.idle": "2021-12-10T09:16:23.200793Z",
     "shell.execute_reply": "2021-12-10T09:16:23.201301Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.758624Z"
    },
    "papermill": {
     "duration": 0.085975,
     "end_time": "2021-12-10T09:16:23.201472",
     "exception": false,
     "start_time": "2021-12-10T09:16:23.115497",
     "status": "completed"
    },
    "scrolled": true,
    "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>Area</th>\n",
       "      <th>Item</th>\n",
       "      <th>Year</th>\n",
       "      <th>hg/ha_yield</th>\n",
       "      <th>average_rain_fall_mm_per_year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1985</td>\n",
       "      <td>16652</td>\n",
       "      <td>327.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>Potatoes</td>\n",
       "      <td>1985</td>\n",
       "      <td>140909</td>\n",
       "      <td>327.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>Rice, paddy</td>\n",
       "      <td>1985</td>\n",
       "      <td>22482</td>\n",
       "      <td>327.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>Wheat</td>\n",
       "      <td>1985</td>\n",
       "      <td>12277</td>\n",
       "      <td>327.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1986</td>\n",
       "      <td>16875</td>\n",
       "      <td>327.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Area         Item  Year  hg/ha_yield  average_rain_fall_mm_per_year\n",
       "0  Afghanistan        Maize  1985        16652                          327.0\n",
       "1  Afghanistan     Potatoes  1985       140909                          327.0\n",
       "2  Afghanistan  Rice, paddy  1985        22482                          327.0\n",
       "3  Afghanistan        Wheat  1985        12277                          327.0\n",
       "4  Afghanistan        Maize  1986        16875                          327.0"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yield_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a270fa4d",
   "metadata": {
    "papermill": {
     "duration": 0.073587,
     "end_time": "2021-12-10T09:16:23.348242",
     "exception": false,
     "start_time": "2021-12-10T09:16:23.274655",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "We can see that now the years start from the first yield dataframe the starting year was 1961, now it's 1985 because that's when the rainfall data begins. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "e5f30053",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:23.498747Z",
     "iopub.status.busy": "2021-12-10T09:16:23.498082Z",
     "iopub.status.idle": "2021-12-10T09:16:23.520085Z",
     "shell.execute_reply": "2021-12-10T09:16:23.519442Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.779164Z"
    },
    "papermill": {
     "duration": 0.099113,
     "end_time": "2021-12-10T09:16:23.520264",
     "exception": false,
     "start_time": "2021-12-10T09:16:23.421151",
     "status": "completed"
    },
    "scrolled": true,
    "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>Year</th>\n",
       "      <th>hg/ha_yield</th>\n",
       "      <th>average_rain_fall_mm_per_year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>25385.000000</td>\n",
       "      <td>25385.000000</td>\n",
       "      <td>25385.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2001.278787</td>\n",
       "      <td>68312.278353</td>\n",
       "      <td>1254.849754</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>9.143915</td>\n",
       "      <td>75213.292733</td>\n",
       "      <td>804.449430</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1985.000000</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>51.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1994.000000</td>\n",
       "      <td>17432.000000</td>\n",
       "      <td>630.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2001.000000</td>\n",
       "      <td>38750.000000</td>\n",
       "      <td>1150.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2009.000000</td>\n",
       "      <td>94286.000000</td>\n",
       "      <td>1761.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2016.000000</td>\n",
       "      <td>554855.000000</td>\n",
       "      <td>3240.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Year    hg/ha_yield  average_rain_fall_mm_per_year\n",
       "count  25385.000000   25385.000000                   25385.000000\n",
       "mean    2001.278787   68312.278353                    1254.849754\n",
       "std        9.143915   75213.292733                     804.449430\n",
       "min     1985.000000      50.000000                      51.000000\n",
       "25%     1994.000000   17432.000000                     630.000000\n",
       "50%     2001.000000   38750.000000                    1150.000000\n",
       "75%     2009.000000   94286.000000                    1761.000000\n",
       "max     2016.000000  554855.000000                    3240.000000"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yield_df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cff06fec",
   "metadata": {
    "papermill": {
     "duration": 0.073554,
     "end_time": "2021-12-10T09:16:23.668529",
     "exception": false,
     "start_time": "2021-12-10T09:16:23.594975",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Pesticides Data:\n",
    "Pesticides used for each item and country was also collected from FAO database.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "8a04a160",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:23.820506Z",
     "iopub.status.busy": "2021-12-10T09:16:23.819881Z",
     "iopub.status.idle": "2021-12-10T09:16:23.854233Z",
     "shell.execute_reply": "2021-12-10T09:16:23.854767Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.800582Z"
    },
    "papermill": {
     "duration": 0.112287,
     "end_time": "2021-12-10T09:16:23.854933",
     "exception": false,
     "start_time": "2021-12-10T09:16:23.742646",
     "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>Domain</th>\n",
       "      <th>Area</th>\n",
       "      <th>Element</th>\n",
       "      <th>Item</th>\n",
       "      <th>Year</th>\n",
       "      <th>Unit</th>\n",
       "      <th>Value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Pesticides Use</td>\n",
       "      <td>Albania</td>\n",
       "      <td>Use</td>\n",
       "      <td>Pesticides (total)</td>\n",
       "      <td>1990</td>\n",
       "      <td>tonnes of active ingredients</td>\n",
       "      <td>121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Pesticides Use</td>\n",
       "      <td>Albania</td>\n",
       "      <td>Use</td>\n",
       "      <td>Pesticides (total)</td>\n",
       "      <td>1991</td>\n",
       "      <td>tonnes of active ingredients</td>\n",
       "      <td>121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Pesticides Use</td>\n",
       "      <td>Albania</td>\n",
       "      <td>Use</td>\n",
       "      <td>Pesticides (total)</td>\n",
       "      <td>1992</td>\n",
       "      <td>tonnes of active ingredients</td>\n",
       "      <td>121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Pesticides Use</td>\n",
       "      <td>Albania</td>\n",
       "      <td>Use</td>\n",
       "      <td>Pesticides (total)</td>\n",
       "      <td>1993</td>\n",
       "      <td>tonnes of active ingredients</td>\n",
       "      <td>121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Pesticides Use</td>\n",
       "      <td>Albania</td>\n",
       "      <td>Use</td>\n",
       "      <td>Pesticides (total)</td>\n",
       "      <td>1994</td>\n",
       "      <td>tonnes of active ingredients</td>\n",
       "      <td>201.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           Domain     Area Element                Item  Year  \\\n",
       "0  Pesticides Use  Albania     Use  Pesticides (total)  1990   \n",
       "1  Pesticides Use  Albania     Use  Pesticides (total)  1991   \n",
       "2  Pesticides Use  Albania     Use  Pesticides (total)  1992   \n",
       "3  Pesticides Use  Albania     Use  Pesticides (total)  1993   \n",
       "4  Pesticides Use  Albania     Use  Pesticides (total)  1994   \n",
       "\n",
       "                           Unit  Value  \n",
       "0  tonnes of active ingredients  121.0  \n",
       "1  tonnes of active ingredients  121.0  \n",
       "2  tonnes of active ingredients  121.0  \n",
       "3  tonnes of active ingredients  121.0  \n",
       "4  tonnes of active ingredients  201.0  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_pes = pd.read_csv('../input/crop-yield-prediction-dataset/pesticides.csv')\n",
    "df_pes.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "e9b3764e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:24.006068Z",
     "iopub.status.busy": "2021-12-10T09:16:24.005466Z",
     "iopub.status.idle": "2021-12-10T09:16:24.019763Z",
     "shell.execute_reply": "2021-12-10T09:16:24.020225Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.837300Z"
    },
    "papermill": {
     "duration": 0.091576,
     "end_time": "2021-12-10T09:16:24.020393",
     "exception": false,
     "start_time": "2021-12-10T09:16:23.928817",
     "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>Area</th>\n",
       "      <th>Year</th>\n",
       "      <th>pesticides_tonnes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Albania</td>\n",
       "      <td>1990</td>\n",
       "      <td>121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Albania</td>\n",
       "      <td>1991</td>\n",
       "      <td>121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Albania</td>\n",
       "      <td>1992</td>\n",
       "      <td>121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Albania</td>\n",
       "      <td>1993</td>\n",
       "      <td>121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Albania</td>\n",
       "      <td>1994</td>\n",
       "      <td>201.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Area  Year  pesticides_tonnes\n",
       "0  Albania  1990              121.0\n",
       "1  Albania  1991              121.0\n",
       "2  Albania  1992              121.0\n",
       "3  Albania  1993              121.0\n",
       "4  Albania  1994              201.0"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_pes = df_pes.rename(index=str, columns={\"Value\": \"pesticides_tonnes\"})\n",
    "df_pes = df_pes.drop(['Element','Domain','Unit','Item'], axis=1)\n",
    "df_pes.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "bcac204f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:24.174006Z",
     "iopub.status.busy": "2021-12-10T09:16:24.173411Z",
     "iopub.status.idle": "2021-12-10T09:16:24.188268Z",
     "shell.execute_reply": "2021-12-10T09:16:24.187765Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.853788Z"
    },
    "papermill": {
     "duration": 0.093598,
     "end_time": "2021-12-10T09:16:24.188411",
     "exception": false,
     "start_time": "2021-12-10T09:16:24.094813",
     "status": "completed"
    },
    "scrolled": true,
    "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>Year</th>\n",
       "      <th>pesticides_tonnes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>4349.000000</td>\n",
       "      <td>4.349000e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2003.138883</td>\n",
       "      <td>2.030334e+04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>7.728044</td>\n",
       "      <td>1.177362e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1990.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1996.000000</td>\n",
       "      <td>9.300000e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2003.000000</td>\n",
       "      <td>1.137560e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2010.000000</td>\n",
       "      <td>7.869000e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2016.000000</td>\n",
       "      <td>1.807000e+06</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Year  pesticides_tonnes\n",
       "count  4349.000000       4.349000e+03\n",
       "mean   2003.138883       2.030334e+04\n",
       "std       7.728044       1.177362e+05\n",
       "min    1990.000000       0.000000e+00\n",
       "25%    1996.000000       9.300000e+01\n",
       "50%    2003.000000       1.137560e+03\n",
       "75%    2010.000000       7.869000e+03\n",
       "max    2016.000000       1.807000e+06"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_pes.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "9a6aaf33",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:24.341466Z",
     "iopub.status.busy": "2021-12-10T09:16:24.340546Z",
     "iopub.status.idle": "2021-12-10T09:16:24.354006Z",
     "shell.execute_reply": "2021-12-10T09:16:24.354757Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.878922Z"
    },
    "papermill": {
     "duration": 0.091743,
     "end_time": "2021-12-10T09:16:24.354989",
     "exception": false,
     "start_time": "2021-12-10T09:16:24.263246",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Index: 4349 entries, 0 to 4348\n",
      "Data columns (total 3 columns):\n",
      " #   Column             Non-Null Count  Dtype  \n",
      "---  ------             --------------  -----  \n",
      " 0   Area               4349 non-null   object \n",
      " 1   Year               4349 non-null   int64  \n",
      " 2   pesticides_tonnes  4349 non-null   float64\n",
      "dtypes: float64(1), int64(1), object(1)\n",
      "memory usage: 135.9+ KB\n"
     ]
    }
   ],
   "source": [
    "df_pes.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "b2927cdf",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:24.512970Z",
     "iopub.status.busy": "2021-12-10T09:16:24.511922Z",
     "iopub.status.idle": "2021-12-10T09:16:24.523658Z",
     "shell.execute_reply": "2021-12-10T09:16:24.523050Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.894462Z"
    },
    "papermill": {
     "duration": 0.092298,
     "end_time": "2021-12-10T09:16:24.523831",
     "exception": false,
     "start_time": "2021-12-10T09:16:24.431533",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(18949, 6)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# merge Pesticides dataframe with yield dataframe \n",
    "yield_df = pd.merge(yield_df, df_pes, on=['Year','Area'])\n",
    "yield_df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "9b4c5047",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:24.687941Z",
     "iopub.status.busy": "2021-12-10T09:16:24.687292Z",
     "iopub.status.idle": "2021-12-10T09:16:24.690703Z",
     "shell.execute_reply": "2021-12-10T09:16:24.690157Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.910205Z"
    },
    "papermill": {
     "duration": 0.09091,
     "end_time": "2021-12-10T09:16:24.690840",
     "exception": false,
     "start_time": "2021-12-10T09:16:24.599930",
     "status": "completed"
    },
    "scrolled": true,
    "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>Area</th>\n",
       "      <th>Item</th>\n",
       "      <th>Year</th>\n",
       "      <th>hg/ha_yield</th>\n",
       "      <th>average_rain_fall_mm_per_year</th>\n",
       "      <th>pesticides_tonnes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1990</td>\n",
       "      <td>36613</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Potatoes</td>\n",
       "      <td>1990</td>\n",
       "      <td>66667</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Rice, paddy</td>\n",
       "      <td>1990</td>\n",
       "      <td>23333</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Sorghum</td>\n",
       "      <td>1990</td>\n",
       "      <td>12500</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Soybeans</td>\n",
       "      <td>1990</td>\n",
       "      <td>7000</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Area         Item  Year  hg/ha_yield  average_rain_fall_mm_per_year  \\\n",
       "0  Albania        Maize  1990        36613                         1485.0   \n",
       "1  Albania     Potatoes  1990        66667                         1485.0   \n",
       "2  Albania  Rice, paddy  1990        23333                         1485.0   \n",
       "3  Albania      Sorghum  1990        12500                         1485.0   \n",
       "4  Albania     Soybeans  1990         7000                         1485.0   \n",
       "\n",
       "   pesticides_tonnes  \n",
       "0              121.0  \n",
       "1              121.0  \n",
       "2              121.0  \n",
       "3              121.0  \n",
       "4              121.0  "
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yield_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1393b1f5",
   "metadata": {
    "papermill": {
     "duration": 0.080573,
     "end_time": "2021-12-10T09:16:24.849470",
     "exception": false,
     "start_time": "2021-12-10T09:16:24.768897",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Average Temprature: \n",
    "\n",
    "Average Temprature for each country was colleced from World Bank Data. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "09555b71",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:25.008889Z",
     "iopub.status.busy": "2021-12-10T09:16:25.008143Z",
     "iopub.status.idle": "2021-12-10T09:16:25.062208Z",
     "shell.execute_reply": "2021-12-10T09:16:25.062764Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.925199Z"
    },
    "papermill": {
     "duration": 0.136184,
     "end_time": "2021-12-10T09:16:25.062936",
     "exception": false,
     "start_time": "2021-12-10T09:16:24.926752",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "avg_temp=  pd.read_csv('../input/crop-yield-prediction-dataset/temp.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "686e6cb8",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:25.218945Z",
     "iopub.status.busy": "2021-12-10T09:16:25.218301Z",
     "iopub.status.idle": "2021-12-10T09:16:25.227064Z",
     "shell.execute_reply": "2021-12-10T09:16:25.227687Z",
     "shell.execute_reply.started": "2021-12-10T08:05:20.987233Z"
    },
    "papermill": {
     "duration": 0.088762,
     "end_time": "2021-12-10T09:16:25.227860",
     "exception": false,
     "start_time": "2021-12-10T09:16:25.139098",
     "status": "completed"
    },
    "scrolled": true,
    "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>year</th>\n",
       "      <th>country</th>\n",
       "      <th>avg_temp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1849</td>\n",
       "      <td>Côte D'Ivoire</td>\n",
       "      <td>25.58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1850</td>\n",
       "      <td>Côte D'Ivoire</td>\n",
       "      <td>25.52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1851</td>\n",
       "      <td>Côte D'Ivoire</td>\n",
       "      <td>25.67</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1852</td>\n",
       "      <td>Côte D'Ivoire</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1853</td>\n",
       "      <td>Côte D'Ivoire</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   year        country  avg_temp\n",
       "0  1849  Côte D'Ivoire     25.58\n",
       "1  1850  Côte D'Ivoire     25.52\n",
       "2  1851  Côte D'Ivoire     25.67\n",
       "3  1852  Côte D'Ivoire       NaN\n",
       "4  1853  Côte D'Ivoire       NaN"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "avg_temp.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "ede46efd",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:25.386061Z",
     "iopub.status.busy": "2021-12-10T09:16:25.385431Z",
     "iopub.status.idle": "2021-12-10T09:16:25.405574Z",
     "shell.execute_reply": "2021-12-10T09:16:25.406130Z",
     "shell.execute_reply.started": "2021-12-10T08:05:21.001871Z"
    },
    "papermill": {
     "duration": 0.099996,
     "end_time": "2021-12-10T09:16:25.406328",
     "exception": false,
     "start_time": "2021-12-10T09:16:25.306332",
     "status": "completed"
    },
    "scrolled": true,
    "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>year</th>\n",
       "      <th>avg_temp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>71311.000000</td>\n",
       "      <td>68764.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1905.799007</td>\n",
       "      <td>16.183876</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>67.102099</td>\n",
       "      <td>7.592960</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1743.000000</td>\n",
       "      <td>-14.350000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1858.000000</td>\n",
       "      <td>9.750000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1910.000000</td>\n",
       "      <td>16.140000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1962.000000</td>\n",
       "      <td>23.762500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2013.000000</td>\n",
       "      <td>30.730000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               year      avg_temp\n",
       "count  71311.000000  68764.000000\n",
       "mean    1905.799007     16.183876\n",
       "std       67.102099      7.592960\n",
       "min     1743.000000    -14.350000\n",
       "25%     1858.000000      9.750000\n",
       "50%     1910.000000     16.140000\n",
       "75%     1962.000000     23.762500\n",
       "max     2013.000000     30.730000"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "avg_temp.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a65c3f56",
   "metadata": {
    "papermill": {
     "duration": 0.078072,
     "end_time": "2021-12-10T09:16:25.561365",
     "exception": false,
     "start_time": "2021-12-10T09:16:25.483293",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "So average temprature starts from 1743 and ends at 2013, with some empty rows that we have to drop."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "c246f8a1",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:25.720461Z",
     "iopub.status.busy": "2021-12-10T09:16:25.719847Z",
     "iopub.status.idle": "2021-12-10T09:16:25.754531Z",
     "shell.execute_reply": "2021-12-10T09:16:25.755184Z",
     "shell.execute_reply.started": "2021-12-10T08:05:21.024873Z"
    },
    "papermill": {
     "duration": 0.11466,
     "end_time": "2021-12-10T09:16:25.755353",
     "exception": false,
     "start_time": "2021-12-10T09:16:25.640693",
     "status": "completed"
    },
    "scrolled": true,
    "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>Year</th>\n",
       "      <th>Area</th>\n",
       "      <th>avg_temp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1849</td>\n",
       "      <td>Côte D'Ivoire</td>\n",
       "      <td>25.58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1850</td>\n",
       "      <td>Côte D'Ivoire</td>\n",
       "      <td>25.52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1851</td>\n",
       "      <td>Côte D'Ivoire</td>\n",
       "      <td>25.67</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1852</td>\n",
       "      <td>Côte D'Ivoire</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1853</td>\n",
       "      <td>Côte D'Ivoire</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Year           Area  avg_temp\n",
       "0  1849  Côte D'Ivoire     25.58\n",
       "1  1850  Côte D'Ivoire     25.52\n",
       "2  1851  Côte D'Ivoire     25.67\n",
       "3  1852  Côte D'Ivoire       NaN\n",
       "4  1853  Côte D'Ivoire       NaN"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "avg_temp = avg_temp.rename(index=str, columns={\"year\": \"Year\", \"country\":'Area'})\n",
    "avg_temp.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "62a5346d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:25.919520Z",
     "iopub.status.busy": "2021-12-10T09:16:25.918877Z",
     "iopub.status.idle": "2021-12-10T09:16:25.951780Z",
     "shell.execute_reply": "2021-12-10T09:16:25.951232Z",
     "shell.execute_reply.started": "2021-12-10T08:05:21.062437Z"
    },
    "papermill": {
     "duration": 0.116166,
     "end_time": "2021-12-10T09:16:25.951925",
     "exception": false,
     "start_time": "2021-12-10T09:16:25.835759",
     "status": "completed"
    },
    "scrolled": true,
    "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>Area</th>\n",
       "      <th>Item</th>\n",
       "      <th>Year</th>\n",
       "      <th>hg/ha_yield</th>\n",
       "      <th>average_rain_fall_mm_per_year</th>\n",
       "      <th>pesticides_tonnes</th>\n",
       "      <th>avg_temp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1990</td>\n",
       "      <td>36613</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Potatoes</td>\n",
       "      <td>1990</td>\n",
       "      <td>66667</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Rice, paddy</td>\n",
       "      <td>1990</td>\n",
       "      <td>23333</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Sorghum</td>\n",
       "      <td>1990</td>\n",
       "      <td>12500</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Soybeans</td>\n",
       "      <td>1990</td>\n",
       "      <td>7000</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Area         Item  Year  hg/ha_yield  average_rain_fall_mm_per_year  \\\n",
       "0  Albania        Maize  1990        36613                         1485.0   \n",
       "1  Albania     Potatoes  1990        66667                         1485.0   \n",
       "2  Albania  Rice, paddy  1990        23333                         1485.0   \n",
       "3  Albania      Sorghum  1990        12500                         1485.0   \n",
       "4  Albania     Soybeans  1990         7000                         1485.0   \n",
       "\n",
       "   pesticides_tonnes  avg_temp  \n",
       "0              121.0     16.37  \n",
       "1              121.0     16.37  \n",
       "2              121.0     16.37  \n",
       "3              121.0     16.37  \n",
       "4              121.0     16.37  "
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yield_df = pd.merge(yield_df,avg_temp, on=['Area','Year'])\n",
    "yield_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "9937f396",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:26.112491Z",
     "iopub.status.busy": "2021-12-10T09:16:26.111838Z",
     "iopub.status.idle": "2021-12-10T09:16:26.116161Z",
     "shell.execute_reply": "2021-12-10T09:16:26.116704Z",
     "shell.execute_reply.started": "2021-12-10T08:05:21.098180Z"
    },
    "papermill": {
     "duration": 0.086058,
     "end_time": "2021-12-10T09:16:26.116876",
     "exception": false,
     "start_time": "2021-12-10T09:16:26.030818",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(28242, 7)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yield_df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "d9f4e311",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:26.277897Z",
     "iopub.status.busy": "2021-12-10T09:16:26.277243Z",
     "iopub.status.idle": "2021-12-10T09:16:26.303475Z",
     "shell.execute_reply": "2021-12-10T09:16:26.304047Z",
     "shell.execute_reply.started": "2021-12-10T08:05:21.105975Z"
    },
    "papermill": {
     "duration": 0.109268,
     "end_time": "2021-12-10T09:16:26.304240",
     "exception": false,
     "start_time": "2021-12-10T09:16:26.194972",
     "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>Year</th>\n",
       "      <th>hg/ha_yield</th>\n",
       "      <th>average_rain_fall_mm_per_year</th>\n",
       "      <th>pesticides_tonnes</th>\n",
       "      <th>avg_temp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>28242.000000</td>\n",
       "      <td>28242.000000</td>\n",
       "      <td>28242.00000</td>\n",
       "      <td>28242.000000</td>\n",
       "      <td>28242.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2001.544296</td>\n",
       "      <td>77053.332094</td>\n",
       "      <td>1149.05598</td>\n",
       "      <td>37076.909344</td>\n",
       "      <td>20.542627</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>7.051905</td>\n",
       "      <td>84956.612897</td>\n",
       "      <td>709.81215</td>\n",
       "      <td>59958.784665</td>\n",
       "      <td>6.312051</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1990.000000</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>51.00000</td>\n",
       "      <td>0.040000</td>\n",
       "      <td>1.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1995.000000</td>\n",
       "      <td>19919.250000</td>\n",
       "      <td>593.00000</td>\n",
       "      <td>1702.000000</td>\n",
       "      <td>16.702500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2001.000000</td>\n",
       "      <td>38295.000000</td>\n",
       "      <td>1083.00000</td>\n",
       "      <td>17529.440000</td>\n",
       "      <td>21.510000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2008.000000</td>\n",
       "      <td>104676.750000</td>\n",
       "      <td>1668.00000</td>\n",
       "      <td>48687.880000</td>\n",
       "      <td>26.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2013.000000</td>\n",
       "      <td>501412.000000</td>\n",
       "      <td>3240.00000</td>\n",
       "      <td>367778.000000</td>\n",
       "      <td>30.650000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Year    hg/ha_yield  average_rain_fall_mm_per_year  \\\n",
       "count  28242.000000   28242.000000                    28242.00000   \n",
       "mean    2001.544296   77053.332094                     1149.05598   \n",
       "std        7.051905   84956.612897                      709.81215   \n",
       "min     1990.000000      50.000000                       51.00000   \n",
       "25%     1995.000000   19919.250000                      593.00000   \n",
       "50%     2001.000000   38295.000000                     1083.00000   \n",
       "75%     2008.000000  104676.750000                     1668.00000   \n",
       "max     2013.000000  501412.000000                     3240.00000   \n",
       "\n",
       "       pesticides_tonnes      avg_temp  \n",
       "count       28242.000000  28242.000000  \n",
       "mean        37076.909344     20.542627  \n",
       "std         59958.784665      6.312051  \n",
       "min             0.040000      1.300000  \n",
       "25%          1702.000000     16.702500  \n",
       "50%         17529.440000     21.510000  \n",
       "75%         48687.880000     26.000000  \n",
       "max        367778.000000     30.650000  "
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yield_df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "3749b75b",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:26.474011Z",
     "iopub.status.busy": "2021-12-10T09:16:26.473212Z",
     "iopub.status.idle": "2021-12-10T09:16:26.480241Z",
     "shell.execute_reply": "2021-12-10T09:16:26.479581Z",
     "shell.execute_reply.started": "2021-12-10T08:05:21.141697Z"
    },
    "papermill": {
     "duration": 0.096698,
     "end_time": "2021-12-10T09:16:26.480383",
     "exception": false,
     "start_time": "2021-12-10T09:16:26.383685",
     "status": "completed"
    },
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Area                             0\n",
       "Item                             0\n",
       "Year                             0\n",
       "hg/ha_yield                      0\n",
       "average_rain_fall_mm_per_year    0\n",
       "pesticides_tonnes                0\n",
       "avg_temp                         0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yield_df.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2501c083",
   "metadata": {
    "papermill": {
     "duration": 0.079373,
     "end_time": "2021-12-10T09:16:26.640105",
     "exception": false,
     "start_time": "2021-12-10T09:16:26.560732",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Great, no empty values!"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a69a77f1",
   "metadata": {
    "papermill": {
     "duration": 0.079003,
     "end_time": "2021-12-10T09:16:26.798430",
     "exception": false,
     "start_time": "2021-12-10T09:16:26.719427",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Data Exploration\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c56fd60a",
   "metadata": {
    "papermill": {
     "duration": 0.078734,
     "end_time": "2021-12-10T09:16:26.957312",
     "exception": false,
     "start_time": "2021-12-10T09:16:26.878578",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "**yield_df** is the final obtained dataframe; "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "331ebbac",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:27.124234Z",
     "iopub.status.busy": "2021-12-10T09:16:27.123541Z",
     "iopub.status.idle": "2021-12-10T09:16:27.138928Z",
     "shell.execute_reply": "2021-12-10T09:16:27.138381Z",
     "shell.execute_reply.started": "2021-12-10T08:05:21.154456Z"
    },
    "papermill": {
     "duration": 0.102335,
     "end_time": "2021-12-10T09:16:27.139069",
     "exception": false,
     "start_time": "2021-12-10T09:16:27.036734",
     "status": "completed"
    },
    "scrolled": true,
    "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>Area</th>\n",
       "      <th>Year</th>\n",
       "      <th>hg/ha_yield</th>\n",
       "      <th>average_rain_fall_mm_per_year</th>\n",
       "      <th>pesticides_tonnes</th>\n",
       "      <th>avg_temp</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Item</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Cassava</th>\n",
       "      <td>2045</td>\n",
       "      <td>2045</td>\n",
       "      <td>2045</td>\n",
       "      <td>2045</td>\n",
       "      <td>2045</td>\n",
       "      <td>2045</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Maize</th>\n",
       "      <td>4121</td>\n",
       "      <td>4121</td>\n",
       "      <td>4121</td>\n",
       "      <td>4121</td>\n",
       "      <td>4121</td>\n",
       "      <td>4121</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Plantains and others</th>\n",
       "      <td>556</td>\n",
       "      <td>556</td>\n",
       "      <td>556</td>\n",
       "      <td>556</td>\n",
       "      <td>556</td>\n",
       "      <td>556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Potatoes</th>\n",
       "      <td>4276</td>\n",
       "      <td>4276</td>\n",
       "      <td>4276</td>\n",
       "      <td>4276</td>\n",
       "      <td>4276</td>\n",
       "      <td>4276</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Rice, paddy</th>\n",
       "      <td>3388</td>\n",
       "      <td>3388</td>\n",
       "      <td>3388</td>\n",
       "      <td>3388</td>\n",
       "      <td>3388</td>\n",
       "      <td>3388</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sorghum</th>\n",
       "      <td>3039</td>\n",
       "      <td>3039</td>\n",
       "      <td>3039</td>\n",
       "      <td>3039</td>\n",
       "      <td>3039</td>\n",
       "      <td>3039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Soybeans</th>\n",
       "      <td>3223</td>\n",
       "      <td>3223</td>\n",
       "      <td>3223</td>\n",
       "      <td>3223</td>\n",
       "      <td>3223</td>\n",
       "      <td>3223</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sweet potatoes</th>\n",
       "      <td>2890</td>\n",
       "      <td>2890</td>\n",
       "      <td>2890</td>\n",
       "      <td>2890</td>\n",
       "      <td>2890</td>\n",
       "      <td>2890</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Wheat</th>\n",
       "      <td>3857</td>\n",
       "      <td>3857</td>\n",
       "      <td>3857</td>\n",
       "      <td>3857</td>\n",
       "      <td>3857</td>\n",
       "      <td>3857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Yams</th>\n",
       "      <td>847</td>\n",
       "      <td>847</td>\n",
       "      <td>847</td>\n",
       "      <td>847</td>\n",
       "      <td>847</td>\n",
       "      <td>847</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      Area  Year  hg/ha_yield  average_rain_fall_mm_per_year  \\\n",
       "Item                                                                           \n",
       "Cassava               2045  2045         2045                           2045   \n",
       "Maize                 4121  4121         4121                           4121   \n",
       "Plantains and others   556   556          556                            556   \n",
       "Potatoes              4276  4276         4276                           4276   \n",
       "Rice, paddy           3388  3388         3388                           3388   \n",
       "Sorghum               3039  3039         3039                           3039   \n",
       "Soybeans              3223  3223         3223                           3223   \n",
       "Sweet potatoes        2890  2890         2890                           2890   \n",
       "Wheat                 3857  3857         3857                           3857   \n",
       "Yams                   847   847          847                            847   \n",
       "\n",
       "                      pesticides_tonnes  avg_temp  \n",
       "Item                                               \n",
       "Cassava                            2045      2045  \n",
       "Maize                              4121      4121  \n",
       "Plantains and others                556       556  \n",
       "Potatoes                           4276      4276  \n",
       "Rice, paddy                        3388      3388  \n",
       "Sorghum                            3039      3039  \n",
       "Soybeans                           3223      3223  \n",
       "Sweet potatoes                     2890      2890  \n",
       "Wheat                              3857      3857  \n",
       "Yams                                847       847  "
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yield_df.groupby('Item').count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "088b5691",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:27.306979Z",
     "iopub.status.busy": "2021-12-10T09:16:27.306334Z",
     "iopub.status.idle": "2021-12-10T09:16:27.331029Z",
     "shell.execute_reply": "2021-12-10T09:16:27.330439Z",
     "shell.execute_reply.started": "2021-12-10T08:05:21.173201Z"
    },
    "papermill": {
     "duration": 0.110255,
     "end_time": "2021-12-10T09:16:27.331191",
     "exception": false,
     "start_time": "2021-12-10T09:16:27.220936",
     "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>Year</th>\n",
       "      <th>hg/ha_yield</th>\n",
       "      <th>average_rain_fall_mm_per_year</th>\n",
       "      <th>pesticides_tonnes</th>\n",
       "      <th>avg_temp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>28242.000000</td>\n",
       "      <td>28242.000000</td>\n",
       "      <td>28242.00000</td>\n",
       "      <td>28242.000000</td>\n",
       "      <td>28242.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2001.544296</td>\n",
       "      <td>77053.332094</td>\n",
       "      <td>1149.05598</td>\n",
       "      <td>37076.909344</td>\n",
       "      <td>20.542627</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>7.051905</td>\n",
       "      <td>84956.612897</td>\n",
       "      <td>709.81215</td>\n",
       "      <td>59958.784665</td>\n",
       "      <td>6.312051</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1990.000000</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>51.00000</td>\n",
       "      <td>0.040000</td>\n",
       "      <td>1.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1995.000000</td>\n",
       "      <td>19919.250000</td>\n",
       "      <td>593.00000</td>\n",
       "      <td>1702.000000</td>\n",
       "      <td>16.702500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2001.000000</td>\n",
       "      <td>38295.000000</td>\n",
       "      <td>1083.00000</td>\n",
       "      <td>17529.440000</td>\n",
       "      <td>21.510000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2008.000000</td>\n",
       "      <td>104676.750000</td>\n",
       "      <td>1668.00000</td>\n",
       "      <td>48687.880000</td>\n",
       "      <td>26.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2013.000000</td>\n",
       "      <td>501412.000000</td>\n",
       "      <td>3240.00000</td>\n",
       "      <td>367778.000000</td>\n",
       "      <td>30.650000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Year    hg/ha_yield  average_rain_fall_mm_per_year  \\\n",
       "count  28242.000000   28242.000000                    28242.00000   \n",
       "mean    2001.544296   77053.332094                     1149.05598   \n",
       "std        7.051905   84956.612897                      709.81215   \n",
       "min     1990.000000      50.000000                       51.00000   \n",
       "25%     1995.000000   19919.250000                      593.00000   \n",
       "50%     2001.000000   38295.000000                     1083.00000   \n",
       "75%     2008.000000  104676.750000                     1668.00000   \n",
       "max     2013.000000  501412.000000                     3240.00000   \n",
       "\n",
       "       pesticides_tonnes      avg_temp  \n",
       "count       28242.000000  28242.000000  \n",
       "mean        37076.909344     20.542627  \n",
       "std         59958.784665      6.312051  \n",
       "min             0.040000      1.300000  \n",
       "25%          1702.000000     16.702500  \n",
       "50%         17529.440000     21.510000  \n",
       "75%         48687.880000     26.000000  \n",
       "max        367778.000000     30.650000  "
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yield_df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "91156c9b",
   "metadata": {
    "papermill": {
     "duration": 0.079856,
     "end_time": "2021-12-10T09:16:27.490909",
     "exception": false,
     "start_time": "2021-12-10T09:16:27.411053",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "It can be noticed the high variance in the values for each columns, later on I'll account for that will scaling. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "416b8b22",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:27.657608Z",
     "iopub.status.busy": "2021-12-10T09:16:27.656759Z",
     "iopub.status.idle": "2021-12-10T09:16:27.661670Z",
     "shell.execute_reply": "2021-12-10T09:16:27.661163Z",
     "shell.execute_reply.started": "2021-12-10T08:05:21.200826Z"
    },
    "papermill": {
     "duration": 0.090769,
     "end_time": "2021-12-10T09:16:27.661814",
     "exception": false,
     "start_time": "2021-12-10T09:16:27.571045",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "101"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yield_df['Area'].nunique()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "85ffcaf4",
   "metadata": {
    "papermill": {
     "duration": 0.080098,
     "end_time": "2021-12-10T09:16:27.822630",
     "exception": false,
     "start_time": "2021-12-10T09:16:27.742532",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "The dataframe has 101 Countries, ordering these by 10 the highest yield production: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "07df458d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:27.994752Z",
     "iopub.status.busy": "2021-12-10T09:16:27.993730Z",
     "iopub.status.idle": "2021-12-10T09:16:28.003284Z",
     "shell.execute_reply": "2021-12-10T09:16:28.003750Z",
     "shell.execute_reply.started": "2021-12-10T08:05:21.210771Z"
    },
    "papermill": {
     "duration": 0.097959,
     "end_time": "2021-12-10T09:16:28.003922",
     "exception": false,
     "start_time": "2021-12-10T09:16:27.905963",
     "status": "completed"
    },
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Area\n",
       "India             327420324\n",
       "Brazil            167550306\n",
       "Mexico            130788528\n",
       "Japan             124470912\n",
       "Australia         109111062\n",
       "Pakistan           73897434\n",
       "Indonesia          69193506\n",
       "United Kingdom     55419990\n",
       "Turkey             52263950\n",
       "Spain              46773540\n",
       "Name: hg/ha_yield, dtype: int64"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yield_df.groupby(['Area'],sort=True)['hg/ha_yield'].sum().nlargest(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "11604777",
   "metadata": {
    "papermill": {
     "duration": 0.08055,
     "end_time": "2021-12-10T09:16:28.166479",
     "exception": false,
     "start_time": "2021-12-10T09:16:28.085929",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "India has the highest yield production in the dataset. Inclusing items in the groupby:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "e058a038",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:28.335530Z",
     "iopub.status.busy": "2021-12-10T09:16:28.334515Z",
     "iopub.status.idle": "2021-12-10T09:16:28.348747Z",
     "shell.execute_reply": "2021-12-10T09:16:28.348098Z",
     "shell.execute_reply.started": "2021-12-10T08:05:21.227146Z"
    },
    "papermill": {
     "duration": 0.101751,
     "end_time": "2021-12-10T09:16:28.348885",
     "exception": false,
     "start_time": "2021-12-10T09:16:28.247134",
     "status": "completed"
    },
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Item            Area          \n",
       "Cassava         India             142810624\n",
       "Potatoes        India              92122514\n",
       "                Brazil             49602168\n",
       "                United Kingdom     46705145\n",
       "                Australia          45670386\n",
       "Sweet potatoes  India              44439538\n",
       "Potatoes        Japan              42918726\n",
       "                Mexico             42053880\n",
       "Sweet potatoes  Mexico             35808592\n",
       "                Australia          35550294\n",
       "Name: hg/ha_yield, dtype: int64"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yield_df.groupby(['Item','Area'],sort=True)['hg/ha_yield'].sum().nlargest(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d2504a48",
   "metadata": {
    "papermill": {
     "duration": 0.080225,
     "end_time": "2021-12-10T09:16:28.510362",
     "exception": false,
     "start_time": "2021-12-10T09:16:28.430137",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "India is the highest for production of cassava and potatoes. Potatoes seems to be the dominated crop in the dataset, being the highest in 4 countries. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "936b422b",
   "metadata": {
    "papermill": {
     "duration": 0.080556,
     "end_time": "2021-12-10T09:16:28.671867",
     "exception": false,
     "start_time": "2021-12-10T09:16:28.591311",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "The final dataframe starts from 1990 and ends in 2013, that's 23 years worth of data for 101 countries. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3c1f4ef3",
   "metadata": {
    "papermill": {
     "duration": 0.080547,
     "end_time": "2021-12-10T09:16:28.833568",
     "exception": false,
     "start_time": "2021-12-10T09:16:28.753021",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Now, exploring the relationships between the colunms of the dataframe, a good way to quickly check correlations among columns is by visualizing the correlation matrix as a heatmap."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "b5fd473f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:29.028298Z",
     "iopub.status.busy": "2021-12-10T09:16:29.027519Z",
     "iopub.status.idle": "2021-12-10T09:16:30.127945Z",
     "shell.execute_reply": "2021-12-10T09:16:30.127353Z",
     "shell.execute_reply.started": "2021-12-10T08:05:21.244973Z"
    },
    "papermill": {
     "duration": 1.204724,
     "end_time": "2021-12-10T09:16:30.128087",
     "exception": false,
     "start_time": "2021-12-10T09:16:28.923363",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import sklearn\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "450f78a3",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:30.305275Z",
     "iopub.status.busy": "2021-12-10T09:16:30.304632Z",
     "iopub.status.idle": "2021-12-10T09:16:30.682288Z",
     "shell.execute_reply": "2021-12-10T09:16:30.681358Z",
     "shell.execute_reply.started": "2021-12-10T08:05:22.320659Z"
    },
    "papermill": {
     "duration": 0.471802,
     "end_time": "2021-12-10T09:16:30.682432",
     "exception": false,
     "start_time": "2021-12-10T09:16:30.210630",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 792x648 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "correlation_data=yield_df.select_dtypes(include=[np.number]).corr()\n",
    "\n",
    "mask = np.zeros_like(correlation_data, dtype=np.bool)\n",
    "mask[np.triu_indices_from(mask)] = True\n",
    "\n",
    "f, ax = plt.subplots(figsize=(11, 9))\n",
    "\n",
    "# Generate a custom diverging colormap\n",
    "cmap = sns.palette=\"vlag\"\n",
    "\n",
    "# Draw the heatmap with the mask and correct aspect ratio\n",
    "sns.heatmap(correlation_data, mask=mask, cmap=cmap, vmax=.3, center=0,\n",
    "            square=True, linewidths=.5, cbar_kws={\"shrink\": .5});"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9bd6e510",
   "metadata": {
    "papermill": {
     "duration": 0.082816,
     "end_time": "2021-12-10T09:16:30.847770",
     "exception": false,
     "start_time": "2021-12-10T09:16:30.764954",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "It can be seen from the above correlation map that there is no correlation between any of the colmuns in the dataframe. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8cf42ca8",
   "metadata": {
    "papermill": {
     "duration": 0.08305,
     "end_time": "2021-12-10T09:16:31.015310",
     "exception": false,
     "start_time": "2021-12-10T09:16:30.932260",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Data Preprocessing"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0049116b",
   "metadata": {
    "papermill": {
     "duration": 0.082282,
     "end_time": "2021-12-10T09:16:31.180108",
     "exception": false,
     "start_time": "2021-12-10T09:16:31.097826",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Data Preprocessing is a technique that is used to convert the raw data into a clean data set. In other words, whenever the data is gathered from different sources it is collected in raw format which is not feasible for the analysis.  \n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "8204ee29",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:31.347902Z",
     "iopub.status.busy": "2021-12-10T09:16:31.347289Z",
     "iopub.status.idle": "2021-12-10T09:16:31.360156Z",
     "shell.execute_reply": "2021-12-10T09:16:31.360726Z",
     "shell.execute_reply.started": "2021-12-10T08:05:22.637117Z"
    },
    "papermill": {
     "duration": 0.098123,
     "end_time": "2021-12-10T09:16:31.360896",
     "exception": false,
     "start_time": "2021-12-10T09:16:31.262773",
     "status": "completed"
    },
    "scrolled": true,
    "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>Area</th>\n",
       "      <th>Item</th>\n",
       "      <th>Year</th>\n",
       "      <th>hg/ha_yield</th>\n",
       "      <th>average_rain_fall_mm_per_year</th>\n",
       "      <th>pesticides_tonnes</th>\n",
       "      <th>avg_temp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Maize</td>\n",
       "      <td>1990</td>\n",
       "      <td>36613</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Potatoes</td>\n",
       "      <td>1990</td>\n",
       "      <td>66667</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Rice, paddy</td>\n",
       "      <td>1990</td>\n",
       "      <td>23333</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Sorghum</td>\n",
       "      <td>1990</td>\n",
       "      <td>12500</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Soybeans</td>\n",
       "      <td>1990</td>\n",
       "      <td>7000</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Area         Item  Year  hg/ha_yield  average_rain_fall_mm_per_year  \\\n",
       "0  Albania        Maize  1990        36613                         1485.0   \n",
       "1  Albania     Potatoes  1990        66667                         1485.0   \n",
       "2  Albania  Rice, paddy  1990        23333                         1485.0   \n",
       "3  Albania      Sorghum  1990        12500                         1485.0   \n",
       "4  Albania     Soybeans  1990         7000                         1485.0   \n",
       "\n",
       "   pesticides_tonnes  avg_temp  \n",
       "0              121.0     16.37  \n",
       "1              121.0     16.37  \n",
       "2              121.0     16.37  \n",
       "3              121.0     16.37  \n",
       "4              121.0     16.37  "
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yield_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b6badc54",
   "metadata": {
    "papermill": {
     "duration": 0.08264,
     "end_time": "2021-12-10T09:16:31.526507",
     "exception": false,
     "start_time": "2021-12-10T09:16:31.443867",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Encoding Categorical Variables:\n",
    "There are two categorical columns in the dataframe, categorical data are variables that contain label values rather than numeric values. The number of possible values is often limited to a fixed set, like in this case, items and countries values.\n",
    "Many machine learning algorithms cannot operate on label data directly. They require all input variables and output variables to be numeric.\n",
    "\n",
    "This means that categorical data must be converted to a numerical form. One hot encoding is a process by which categorical variables are converted into a form that could be provided to ML algorithms to do a better job in prediction. For that purpose, One-Hot Encoding will be used to convert these two columns to one-hot numeric array.\n",
    "\n",
    "The categorical value represents the numerical value of the entry in the dataset. This encoding will create a binary column for each category and returns a matrix with the results. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "8dd9e719",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:31.695882Z",
     "iopub.status.busy": "2021-12-10T09:16:31.695285Z",
     "iopub.status.idle": "2021-12-10T09:16:31.723899Z",
     "shell.execute_reply": "2021-12-10T09:16:31.724435Z",
     "shell.execute_reply.started": "2021-12-10T08:05:22.652633Z"
    },
    "papermill": {
     "duration": 0.114627,
     "end_time": "2021-12-10T09:16:31.724603",
     "exception": false,
     "start_time": "2021-12-10T09:16:31.609976",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import OneHotEncoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "69cbf842",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:31.895529Z",
     "iopub.status.busy": "2021-12-10T09:16:31.894870Z",
     "iopub.status.idle": "2021-12-10T09:16:31.944275Z",
     "shell.execute_reply": "2021-12-10T09:16:31.944764Z",
     "shell.execute_reply.started": "2021-12-10T08:05:22.688240Z"
    },
    "papermill": {
     "duration": 0.13712,
     "end_time": "2021-12-10T09:16:31.944927",
     "exception": false,
     "start_time": "2021-12-10T09:16:31.807807",
     "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>Year</th>\n",
       "      <th>average_rain_fall_mm_per_year</th>\n",
       "      <th>pesticides_tonnes</th>\n",
       "      <th>avg_temp</th>\n",
       "      <th>Country_Albania</th>\n",
       "      <th>Country_Algeria</th>\n",
       "      <th>Country_Angola</th>\n",
       "      <th>Country_Argentina</th>\n",
       "      <th>Country_Armenia</th>\n",
       "      <th>Country_Australia</th>\n",
       "      <th>...</th>\n",
       "      <th>Item_Cassava</th>\n",
       "      <th>Item_Maize</th>\n",
       "      <th>Item_Plantains and others</th>\n",
       "      <th>Item_Potatoes</th>\n",
       "      <th>Item_Rice, paddy</th>\n",
       "      <th>Item_Sorghum</th>\n",
       "      <th>Item_Soybeans</th>\n",
       "      <th>Item_Sweet potatoes</th>\n",
       "      <th>Item_Wheat</th>\n",
       "      <th>Item_Yams</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1990</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1990</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1990</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1990</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1990</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 115 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Year  average_rain_fall_mm_per_year  pesticides_tonnes  avg_temp  \\\n",
       "0  1990                         1485.0              121.0     16.37   \n",
       "1  1990                         1485.0              121.0     16.37   \n",
       "2  1990                         1485.0              121.0     16.37   \n",
       "3  1990                         1485.0              121.0     16.37   \n",
       "4  1990                         1485.0              121.0     16.37   \n",
       "\n",
       "   Country_Albania  Country_Algeria  Country_Angola  Country_Argentina  \\\n",
       "0                1                0               0                  0   \n",
       "1                1                0               0                  0   \n",
       "2                1                0               0                  0   \n",
       "3                1                0               0                  0   \n",
       "4                1                0               0                  0   \n",
       "\n",
       "   Country_Armenia  Country_Australia  ...  Item_Cassava  Item_Maize  \\\n",
       "0                0                  0  ...             0           1   \n",
       "1                0                  0  ...             0           0   \n",
       "2                0                  0  ...             0           0   \n",
       "3                0                  0  ...             0           0   \n",
       "4                0                  0  ...             0           0   \n",
       "\n",
       "   Item_Plantains and others  Item_Potatoes  Item_Rice, paddy  Item_Sorghum  \\\n",
       "0                          0              0                 0             0   \n",
       "1                          0              1                 0             0   \n",
       "2                          0              0                 1             0   \n",
       "3                          0              0                 0             1   \n",
       "4                          0              0                 0             0   \n",
       "\n",
       "   Item_Soybeans  Item_Sweet potatoes  Item_Wheat  Item_Yams  \n",
       "0              0                    0           0          0  \n",
       "1              0                    0           0          0  \n",
       "2              0                    0           0          0  \n",
       "3              0                    0           0          0  \n",
       "4              1                    0           0          0  \n",
       "\n",
       "[5 rows x 115 columns]"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yield_df_onehot = pd.get_dummies(yield_df, columns=['Area',\"Item\"], prefix = ['Country',\"Item\"])\n",
    "features=yield_df_onehot.loc[:, yield_df_onehot.columns != 'hg/ha_yield']\n",
    "label=yield_df['hg/ha_yield']\n",
    "features.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "9006729c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:32.116522Z",
     "iopub.status.busy": "2021-12-10T09:16:32.115869Z",
     "iopub.status.idle": "2021-12-10T09:16:32.123815Z",
     "shell.execute_reply": "2021-12-10T09:16:32.124357Z",
     "shell.execute_reply.started": "2021-12-10T08:05:22.738995Z"
    },
    "papermill": {
     "duration": 0.095436,
     "end_time": "2021-12-10T09:16:32.124521",
     "exception": false,
     "start_time": "2021-12-10T09:16:32.029085",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "features = features.drop(['Year'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "51de10be",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:32.294270Z",
     "iopub.status.busy": "2021-12-10T09:16:32.293645Z",
     "iopub.status.idle": "2021-12-10T09:16:32.310060Z",
     "shell.execute_reply": "2021-12-10T09:16:32.309051Z",
     "shell.execute_reply.started": "2021-12-10T08:05:22.748347Z"
    },
    "papermill": {
     "duration": 0.102112,
     "end_time": "2021-12-10T09:16:32.310326",
     "exception": false,
     "start_time": "2021-12-10T09:16:32.208214",
     "status": "completed"
    },
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 28242 entries, 0 to 28241\n",
      "Columns: 114 entries, average_rain_fall_mm_per_year to Item_Yams\n",
      "dtypes: float64(3), uint8(111)\n",
      "memory usage: 3.9 MB\n"
     ]
    }
   ],
   "source": [
    "features.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "8b7eec54",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:32.482388Z",
     "iopub.status.busy": "2021-12-10T09:16:32.481780Z",
     "iopub.status.idle": "2021-12-10T09:16:32.500420Z",
     "shell.execute_reply": "2021-12-10T09:16:32.500939Z",
     "shell.execute_reply.started": "2021-12-10T08:05:22.769349Z"
    },
    "papermill": {
     "duration": 0.105341,
     "end_time": "2021-12-10T09:16:32.501106",
     "exception": false,
     "start_time": "2021-12-10T09:16:32.395765",
     "status": "completed"
    },
    "scrolled": true,
    "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>average_rain_fall_mm_per_year</th>\n",
       "      <th>pesticides_tonnes</th>\n",
       "      <th>avg_temp</th>\n",
       "      <th>Country_Albania</th>\n",
       "      <th>Country_Algeria</th>\n",
       "      <th>Country_Angola</th>\n",
       "      <th>Country_Argentina</th>\n",
       "      <th>Country_Armenia</th>\n",
       "      <th>Country_Australia</th>\n",
       "      <th>Country_Austria</th>\n",
       "      <th>...</th>\n",
       "      <th>Item_Cassava</th>\n",
       "      <th>Item_Maize</th>\n",
       "      <th>Item_Plantains and others</th>\n",
       "      <th>Item_Potatoes</th>\n",
       "      <th>Item_Rice, paddy</th>\n",
       "      <th>Item_Sorghum</th>\n",
       "      <th>Item_Soybeans</th>\n",
       "      <th>Item_Sweet potatoes</th>\n",
       "      <th>Item_Wheat</th>\n",
       "      <th>Item_Yams</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "      <td>1</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>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "      <td>1</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>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "      <td>1</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>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "      <td>1</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>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "      <td>1</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>...</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>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 114 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   average_rain_fall_mm_per_year  pesticides_tonnes  avg_temp  \\\n",
       "0                         1485.0              121.0     16.37   \n",
       "1                         1485.0              121.0     16.37   \n",
       "2                         1485.0              121.0     16.37   \n",
       "3                         1485.0              121.0     16.37   \n",
       "4                         1485.0              121.0     16.37   \n",
       "\n",
       "   Country_Albania  Country_Algeria  Country_Angola  Country_Argentina  \\\n",
       "0                1                0               0                  0   \n",
       "1                1                0               0                  0   \n",
       "2                1                0               0                  0   \n",
       "3                1                0               0                  0   \n",
       "4                1                0               0                  0   \n",
       "\n",
       "   Country_Armenia  Country_Australia  Country_Austria  ...  Item_Cassava  \\\n",
       "0                0                  0                0  ...             0   \n",
       "1                0                  0                0  ...             0   \n",
       "2                0                  0                0  ...             0   \n",
       "3                0                  0                0  ...             0   \n",
       "4                0                  0                0  ...             0   \n",
       "\n",
       "   Item_Maize  Item_Plantains and others  Item_Potatoes  Item_Rice, paddy  \\\n",
       "0           1                          0              0                 0   \n",
       "1           0                          0              1                 0   \n",
       "2           0                          0              0                 1   \n",
       "3           0                          0              0                 0   \n",
       "4           0                          0              0                 0   \n",
       "\n",
       "   Item_Sorghum  Item_Soybeans  Item_Sweet potatoes  Item_Wheat  Item_Yams  \n",
       "0             0              0                    0           0          0  \n",
       "1             0              0                    0           0          0  \n",
       "2             0              0                    0           0          0  \n",
       "3             1              0                    0           0          0  \n",
       "4             0              1                    0           0          0  \n",
       "\n",
       "[5 rows x 114 columns]"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7afc0054",
   "metadata": {
    "papermill": {
     "duration": 0.08413,
     "end_time": "2021-12-10T09:16:32.670776",
     "exception": false,
     "start_time": "2021-12-10T09:16:32.586646",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "\n",
    "### Scaling Features: \n",
    "\n",
    "Taking a look at the dataset above, it contains features highly varying in magnitudes, units and range. The features with high magnitudes will weigh in a lot more in the distance calculations than features with low magnitudes.\n",
    "\n",
    "To supress this effect, we need to bring all features to the same level of magnitudes. This can be acheived by scaling."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "32622a54",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:32.843308Z",
     "iopub.status.busy": "2021-12-10T09:16:32.842597Z",
     "iopub.status.idle": "2021-12-10T09:16:32.898202Z",
     "shell.execute_reply": "2021-12-10T09:16:32.898705Z",
     "shell.execute_reply.started": "2021-12-10T08:05:22.794468Z"
    },
    "papermill": {
     "duration": 0.143264,
     "end_time": "2021-12-10T09:16:32.898888",
     "exception": false,
     "start_time": "2021-12-10T09:16:32.755624",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler\n",
    "scaler=MinMaxScaler()\n",
    "features=scaler.fit_transform(features) "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "117cd2e4",
   "metadata": {
    "papermill": {
     "duration": 0.085935,
     "end_time": "2021-12-10T09:16:33.069575",
     "exception": false,
     "start_time": "2021-12-10T09:16:32.983640",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "After dropping year column in addition to scaling all values in features, the resulting array will look something like this : "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "c8c1e738",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:33.244271Z",
     "iopub.status.busy": "2021-12-10T09:16:33.243271Z",
     "iopub.status.idle": "2021-12-10T09:16:33.249029Z",
     "shell.execute_reply": "2021-12-10T09:16:33.249673Z",
     "shell.execute_reply.started": "2021-12-10T08:05:22.856360Z"
    },
    "papermill": {
     "duration": 0.094914,
     "end_time": "2021-12-10T09:16:33.249845",
     "exception": false,
     "start_time": "2021-12-10T09:16:33.154931",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[4.49670743e-01, 3.28894097e-04, 5.13458262e-01, ...,\n",
       "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00],\n",
       "       [4.49670743e-01, 3.28894097e-04, 5.13458262e-01, ...,\n",
       "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00],\n",
       "       [4.49670743e-01, 3.28894097e-04, 5.13458262e-01, ...,\n",
       "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00],\n",
       "       ...,\n",
       "       [1.90028222e-01, 6.93361288e-03, 6.28960818e-01, ...,\n",
       "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00],\n",
       "       [1.90028222e-01, 6.93361288e-03, 6.28960818e-01, ...,\n",
       "        1.00000000e+00, 0.00000000e+00, 0.00000000e+00],\n",
       "       [1.90028222e-01, 6.93361288e-03, 6.28960818e-01, ...,\n",
       "        0.00000000e+00, 1.00000000e+00, 0.00000000e+00]])"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a07ae43e",
   "metadata": {
    "papermill": {
     "duration": 0.084624,
     "end_time": "2021-12-10T09:16:33.421613",
     "exception": false,
     "start_time": "2021-12-10T09:16:33.336989",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Training Data: \n",
    "\n",
    "The dataset will be split to two datasets, the training dataset and test dataset. The data is usually tend to be split inequality because training the model usually requires as much data-points as possible.The common splits are 70/30 or 80/20 for train/test.\n",
    "\n",
    "The training dataset is the intial dataset used to train ML algorithm to learn and produce right predictions. (70% of dataset is training dataset)\n",
    "\n",
    "The test dataset, however, is used to assess how well ML algorithm is trained with the training dataset. You can’t simply reuse the training dataset in the testing stage because ML algorithm will already “know” the expected output, which defeats the purpose of testing the algorithm. (30% of dataset is testing dataset) \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "c33c9058",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:33.598325Z",
     "iopub.status.busy": "2021-12-10T09:16:33.597527Z",
     "iopub.status.idle": "2021-12-10T09:16:33.686083Z",
     "shell.execute_reply": "2021-12-10T09:16:33.685533Z",
     "shell.execute_reply.started": "2021-12-10T08:05:22.863769Z"
    },
    "papermill": {
     "duration": 0.179776,
     "end_time": "2021-12-10T09:16:33.686278",
     "exception": false,
     "start_time": "2021-12-10T09:16:33.506502",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "train_data, test_data, train_labels, test_labels = train_test_split(features, label, test_size=0.3, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "82d19ecf",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:33.860352Z",
     "iopub.status.busy": "2021-12-10T09:16:33.859076Z",
     "iopub.status.idle": "2021-12-10T09:16:33.863264Z",
     "shell.execute_reply": "2021-12-10T09:16:33.862738Z",
     "shell.execute_reply.started": "2021-12-10T08:05:22.949470Z"
    },
    "papermill": {
     "duration": 0.092165,
     "end_time": "2021-12-10T09:16:33.863413",
     "exception": false,
     "start_time": "2021-12-10T09:16:33.771248",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "#write final df to csv file \n",
    "#yield_df.to_csv('../input/crop-yield-prediction-dataset/yield_df.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "83eb6fa5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:34.043236Z",
     "iopub.status.busy": "2021-12-10T09:16:34.041018Z",
     "iopub.status.idle": "2021-12-10T09:16:34.083420Z",
     "shell.execute_reply": "2021-12-10T09:16:34.082836Z",
     "shell.execute_reply.started": "2021-12-10T08:05:22.955438Z"
    },
    "papermill": {
     "duration": 0.133821,
     "end_time": "2021-12-10T09:16:34.083575",
     "exception": false,
     "start_time": "2021-12-10T09:16:33.949754",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "train_data, test_data, train_labels, test_labels = train_test_split(features, label, test_size=0.3, random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cb2f9a15",
   "metadata": {
    "papermill": {
     "duration": 0.084597,
     "end_time": "2021-12-10T09:16:34.252882",
     "exception": false,
     "start_time": "2021-12-10T09:16:34.168285",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Model Comparison & Selection \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "3106ac40",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:34.430556Z",
     "iopub.status.busy": "2021-12-10T09:16:34.429875Z",
     "iopub.status.idle": "2021-12-10T09:16:34.432390Z",
     "shell.execute_reply": "2021-12-10T09:16:34.431869Z",
     "shell.execute_reply.started": "2021-12-10T08:05:23.008454Z"
    },
    "papermill": {
     "duration": 0.095033,
     "end_time": "2021-12-10T09:16:34.432534",
     "exception": false,
     "start_time": "2021-12-10T09:16:34.337501",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import r2_score\n",
    "def compare_models(model):\n",
    "    model_name = model.__class__.__name__\n",
    "    fit=model.fit(train_data,train_labels)\n",
    "    y_pred=fit.predict(test_data)\n",
    "    r2=r2_score(test_labels,y_pred)\n",
    "    return([model_name,r2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "5a419de4",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:34.610866Z",
     "iopub.status.busy": "2021-12-10T09:16:34.610121Z",
     "iopub.status.idle": "2021-12-10T09:16:34.814501Z",
     "shell.execute_reply": "2021-12-10T09:16:34.813939Z",
     "shell.execute_reply.started": "2021-12-10T08:05:23.015829Z"
    },
    "papermill": {
     "duration": 0.295595,
     "end_time": "2021-12-10T09:16:34.814651",
     "exception": false,
     "start_time": "2021-12-10T09:16:34.519056",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "from sklearn import svm\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "\n",
    "models = [\n",
    "    GradientBoostingRegressor(n_estimators=200, max_depth=3, random_state=0),\n",
    "     RandomForestRegressor(n_estimators=200, max_depth=3, random_state=0),\n",
    "    svm.SVR(),\n",
    "   DecisionTreeRegressor()\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "dc8554ec",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:16:34.988166Z",
     "iopub.status.busy": "2021-12-10T09:16:34.987529Z",
     "iopub.status.idle": "2021-12-10T09:18:55.705540Z",
     "shell.execute_reply": "2021-12-10T09:18:55.704793Z",
     "shell.execute_reply.started": "2021-12-10T08:05:23.207274Z"
    },
    "papermill": {
     "duration": 140.805221,
     "end_time": "2021-12-10T09:18:55.705699",
     "exception": false,
     "start_time": "2021-12-10T09:16:34.900478",
     "status": "completed"
    },
    "scrolled": true,
    "tags": []
   },
   "outputs": [],
   "source": [
    "model_train=list(map(compare_models,models)) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "b419ab07",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:18:55.882610Z",
     "iopub.status.busy": "2021-12-10T09:18:55.881607Z",
     "iopub.status.idle": "2021-12-10T09:18:55.887867Z",
     "shell.execute_reply": "2021-12-10T09:18:55.888547Z",
     "shell.execute_reply.started": "2021-12-10T08:07:43.079175Z"
    },
    "papermill": {
     "duration": 0.097112,
     "end_time": "2021-12-10T09:18:55.888792",
     "exception": false,
     "start_time": "2021-12-10T09:18:55.791680",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['GradientBoostingRegressor', 0.8965768919264416]\n",
      "['RandomForestRegressor', 0.6842532317855172]\n",
      "['SVR', -0.19543203867357395]\n",
      "['DecisionTreeRegressor', 0.9606383868613704]\n"
     ]
    }
   ],
   "source": [
    "print(*model_train, sep = \"\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2258bf9f",
   "metadata": {
    "papermill": {
     "duration": 0.085948,
     "end_time": "2021-12-10T09:18:56.068396",
     "exception": false,
     "start_time": "2021-12-10T09:18:55.982448",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "The evaluation metric is set based on **R^2 (coefficient of determination)** regression score function, that will represents the proportion of the variance for items (crops) in the regression model. **R^2** score shows how well terms (data points) fit a curve or line.\n",
    "\n",
    "**R^2** is a statistical measure between 0 and 1 which calculates how similar a regression line is to the data it’s fitted to. If it’s a 1, the model 100% predicts the data variance; if it’s a 0, the model predicts none of the variance. \n",
    "\n",
    "From results viewd above, **Decision Tree Regressor** has the highest R^2 score 0f **96%**, **GradientBoostingRegressor** comes second. \n",
    "\n",
    "\n",
    " I'll also calculate **Adjusted R^2** also indicates how well terms fit a curve or line, but adjusts for the number of terms in a model. If you add more and more useless variables to a model, adjusted r-squared will decrease. If you add more useful variables, adjusted r-squared will increase.\n",
    "Adjusted R2 will always be less than or equal to R2. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "93723d80",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:18:56.247410Z",
     "iopub.status.busy": "2021-12-10T09:18:56.246731Z",
     "iopub.status.idle": "2021-12-10T09:18:56.250556Z",
     "shell.execute_reply": "2021-12-10T09:18:56.251016Z",
     "shell.execute_reply.started": "2021-12-10T08:07:43.086582Z"
    },
    "papermill": {
     "duration": 0.095279,
     "end_time": "2021-12-10T09:18:56.251235",
     "exception": false,
     "start_time": "2021-12-10T09:18:56.155956",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "yield_df_onehot = yield_df_onehot.drop(['Year'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "ccf0644c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:18:56.426611Z",
     "iopub.status.busy": "2021-12-10T09:18:56.425906Z",
     "iopub.status.idle": "2021-12-10T09:18:56.444283Z",
     "shell.execute_reply": "2021-12-10T09:18:56.444841Z",
     "shell.execute_reply.started": "2021-12-10T08:07:43.101986Z"
    },
    "papermill": {
     "duration": 0.107229,
     "end_time": "2021-12-10T09:18:56.445013",
     "exception": false,
     "start_time": "2021-12-10T09:18:56.337784",
     "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>hg/ha_yield</th>\n",
       "      <th>average_rain_fall_mm_per_year</th>\n",
       "      <th>pesticides_tonnes</th>\n",
       "      <th>avg_temp</th>\n",
       "      <th>Country_Albania</th>\n",
       "      <th>Country_Algeria</th>\n",
       "      <th>Country_Angola</th>\n",
       "      <th>Country_Argentina</th>\n",
       "      <th>Country_Armenia</th>\n",
       "      <th>Country_Australia</th>\n",
       "      <th>...</th>\n",
       "      <th>Item_Cassava</th>\n",
       "      <th>Item_Maize</th>\n",
       "      <th>Item_Plantains and others</th>\n",
       "      <th>Item_Potatoes</th>\n",
       "      <th>Item_Rice, paddy</th>\n",
       "      <th>Item_Sorghum</th>\n",
       "      <th>Item_Soybeans</th>\n",
       "      <th>Item_Sweet potatoes</th>\n",
       "      <th>Item_Wheat</th>\n",
       "      <th>Item_Yams</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>36613</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>66667</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>23333</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>12500</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7000</td>\n",
       "      <td>1485.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>16.37</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 115 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   hg/ha_yield  average_rain_fall_mm_per_year  pesticides_tonnes  avg_temp  \\\n",
       "0        36613                         1485.0              121.0     16.37   \n",
       "1        66667                         1485.0              121.0     16.37   \n",
       "2        23333                         1485.0              121.0     16.37   \n",
       "3        12500                         1485.0              121.0     16.37   \n",
       "4         7000                         1485.0              121.0     16.37   \n",
       "\n",
       "   Country_Albania  Country_Algeria  Country_Angola  Country_Argentina  \\\n",
       "0                1                0               0                  0   \n",
       "1                1                0               0                  0   \n",
       "2                1                0               0                  0   \n",
       "3                1                0               0                  0   \n",
       "4                1                0               0                  0   \n",
       "\n",
       "   Country_Armenia  Country_Australia  ...  Item_Cassava  Item_Maize  \\\n",
       "0                0                  0  ...             0           1   \n",
       "1                0                  0  ...             0           0   \n",
       "2                0                  0  ...             0           0   \n",
       "3                0                  0  ...             0           0   \n",
       "4                0                  0  ...             0           0   \n",
       "\n",
       "   Item_Plantains and others  Item_Potatoes  Item_Rice, paddy  Item_Sorghum  \\\n",
       "0                          0              0                 0             0   \n",
       "1                          0              1                 0             0   \n",
       "2                          0              0                 1             0   \n",
       "3                          0              0                 0             1   \n",
       "4                          0              0                 0             0   \n",
       "\n",
       "   Item_Soybeans  Item_Sweet potatoes  Item_Wheat  Item_Yams  \n",
       "0              0                    0           0          0  \n",
       "1              0                    0           0          0  \n",
       "2              0                    0           0          0  \n",
       "3              0                    0           0          0  \n",
       "4              1                    0           0          0  \n",
       "\n",
       "[5 rows x 115 columns]"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yield_df_onehot.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "f122b875",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:18:56.620132Z",
     "iopub.status.busy": "2021-12-10T09:18:56.619450Z",
     "iopub.status.idle": "2021-12-10T09:18:56.819629Z",
     "shell.execute_reply": "2021-12-10T09:18:56.819010Z",
     "shell.execute_reply.started": "2021-12-10T08:07:43.128500Z"
    },
    "papermill": {
     "duration": 0.288652,
     "end_time": "2021-12-10T09:18:56.819800",
     "exception": false,
     "start_time": "2021-12-10T09:18:56.531148",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "#setting test data to columns from dataframe and excluding 'hg/ha_yield' values where ML model should be predicting \n",
    "\n",
    "test_df=pd.DataFrame(test_data,columns=yield_df_onehot.loc[:, yield_df_onehot.columns != 'hg/ha_yield'].columns) \n",
    "\n",
    "# using stack function to return a reshaped DataFrame by pivoting the columns of the current dataframe\n",
    "\n",
    "cntry=test_df[[col for col in test_df.columns if 'Country' in col]].stack()[test_df[[col for col in test_df.columns if 'Country' in col]].stack()>0]\n",
    "cntrylist=list(pd.DataFrame(cntry).index.get_level_values(1))\n",
    "countries=[i.split(\"_\")[1] for i in cntrylist]\n",
    "itm=test_df[[col for col in test_df.columns if 'Item' in col]].stack()[test_df[[col for col in test_df.columns if 'Item' in col]].stack()>0]\n",
    "itmlist=list(pd.DataFrame(itm).index.get_level_values(1))\n",
    "items=[i.split(\"_\")[1] for i in itmlist]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "c14be74d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:18:56.999994Z",
     "iopub.status.busy": "2021-12-10T09:18:56.998970Z",
     "iopub.status.idle": "2021-12-10T09:18:57.025230Z",
     "shell.execute_reply": "2021-12-10T09:18:57.025787Z",
     "shell.execute_reply.started": "2021-12-10T08:07:43.294753Z"
    },
    "papermill": {
     "duration": 0.119426,
     "end_time": "2021-12-10T09:18:57.025955",
     "exception": false,
     "start_time": "2021-12-10T09:18:56.906529",
     "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>average_rain_fall_mm_per_year</th>\n",
       "      <th>pesticides_tonnes</th>\n",
       "      <th>avg_temp</th>\n",
       "      <th>Country_Albania</th>\n",
       "      <th>Country_Algeria</th>\n",
       "      <th>Country_Angola</th>\n",
       "      <th>Country_Argentina</th>\n",
       "      <th>Country_Armenia</th>\n",
       "      <th>Country_Australia</th>\n",
       "      <th>Country_Austria</th>\n",
       "      <th>...</th>\n",
       "      <th>Item_Cassava</th>\n",
       "      <th>Item_Maize</th>\n",
       "      <th>Item_Plantains and others</th>\n",
       "      <th>Item_Potatoes</th>\n",
       "      <th>Item_Rice, paddy</th>\n",
       "      <th>Item_Sorghum</th>\n",
       "      <th>Item_Soybeans</th>\n",
       "      <th>Item_Sweet potatoes</th>\n",
       "      <th>Item_Wheat</th>\n",
       "      <th>Item_Yams</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.183443</td>\n",
       "      <td>0.110716</td>\n",
       "      <td>0.542078</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.458451</td>\n",
       "      <td>0.000413</td>\n",
       "      <td>0.627257</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.183443</td>\n",
       "      <td>0.106159</td>\n",
       "      <td>0.518228</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.224154</td>\n",
       "      <td>0.890971</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.458451</td>\n",
       "      <td>0.000355</td>\n",
       "      <td>0.625213</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 114 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   average_rain_fall_mm_per_year  pesticides_tonnes  avg_temp  \\\n",
       "0                       0.183443           0.110716  0.542078   \n",
       "1                       0.458451           0.000413  0.627257   \n",
       "2                       0.183443           0.106159  0.518228   \n",
       "3                       1.000000           0.224154  0.890971   \n",
       "4                       0.458451           0.000355  0.625213   \n",
       "\n",
       "   Country_Albania  Country_Algeria  Country_Angola  Country_Argentina  \\\n",
       "0              0.0              0.0             0.0                0.0   \n",
       "1              0.0              0.0             0.0                0.0   \n",
       "2              0.0              0.0             0.0                0.0   \n",
       "3              0.0              0.0             0.0                0.0   \n",
       "4              0.0              0.0             0.0                0.0   \n",
       "\n",
       "   Country_Armenia  Country_Australia  Country_Austria  ...  Item_Cassava  \\\n",
       "0              0.0                0.0              0.0  ...           0.0   \n",
       "1              0.0                0.0              0.0  ...           0.0   \n",
       "2              0.0                0.0              0.0  ...           0.0   \n",
       "3              0.0                0.0              0.0  ...           0.0   \n",
       "4              0.0                0.0              0.0  ...           0.0   \n",
       "\n",
       "   Item_Maize  Item_Plantains and others  Item_Potatoes  Item_Rice, paddy  \\\n",
       "0         0.0                        0.0            0.0               1.0   \n",
       "1         0.0                        0.0            0.0               0.0   \n",
       "2         0.0                        0.0            0.0               0.0   \n",
       "3         0.0                        0.0            1.0               0.0   \n",
       "4         0.0                        0.0            0.0               0.0   \n",
       "\n",
       "   Item_Sorghum  Item_Soybeans  Item_Sweet potatoes  Item_Wheat  Item_Yams  \n",
       "0           0.0            0.0                  0.0         0.0        0.0  \n",
       "1           0.0            0.0                  0.0         1.0        0.0  \n",
       "2           1.0            0.0                  0.0         0.0        0.0  \n",
       "3           0.0            0.0                  0.0         0.0        0.0  \n",
       "4           0.0            0.0                  1.0         0.0        0.0  \n",
       "\n",
       "[5 rows x 114 columns]"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "1555cc60",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:18:57.203137Z",
     "iopub.status.busy": "2021-12-10T09:18:57.202205Z",
     "iopub.status.idle": "2021-12-10T09:18:57.217983Z",
     "shell.execute_reply": "2021-12-10T09:18:57.217404Z",
     "shell.execute_reply.started": "2021-12-10T08:07:43.322790Z"
    },
    "papermill": {
     "duration": 0.105131,
     "end_time": "2021-12-10T09:18:57.218123",
     "exception": false,
     "start_time": "2021-12-10T09:18:57.112992",
     "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>average_rain_fall_mm_per_year</th>\n",
       "      <th>pesticides_tonnes</th>\n",
       "      <th>avg_temp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.183443</td>\n",
       "      <td>0.110716</td>\n",
       "      <td>0.542078</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.458451</td>\n",
       "      <td>0.000413</td>\n",
       "      <td>0.627257</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.183443</td>\n",
       "      <td>0.106159</td>\n",
       "      <td>0.518228</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.224154</td>\n",
       "      <td>0.890971</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.458451</td>\n",
       "      <td>0.000355</td>\n",
       "      <td>0.625213</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   average_rain_fall_mm_per_year  pesticides_tonnes  avg_temp\n",
       "0                       0.183443           0.110716  0.542078\n",
       "1                       0.458451           0.000413  0.627257\n",
       "2                       0.183443           0.106159  0.518228\n",
       "3                       1.000000           0.224154  0.890971\n",
       "4                       0.458451           0.000355  0.625213"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df.drop([col for col in test_df.columns if 'Item' in col],axis=1,inplace=True)\n",
    "test_df.drop([col for col in test_df.columns if 'Country' in col],axis=1,inplace=True)\n",
    "test_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "6372fe2a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:18:57.397042Z",
     "iopub.status.busy": "2021-12-10T09:18:57.396402Z",
     "iopub.status.idle": "2021-12-10T09:18:57.410066Z",
     "shell.execute_reply": "2021-12-10T09:18:57.410597Z",
     "shell.execute_reply.started": "2021-12-10T08:07:43.338057Z"
    },
    "papermill": {
     "duration": 0.105196,
     "end_time": "2021-12-10T09:18:57.410766",
     "exception": false,
     "start_time": "2021-12-10T09:18:57.305570",
     "status": "completed"
    },
    "scrolled": true,
    "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>average_rain_fall_mm_per_year</th>\n",
       "      <th>pesticides_tonnes</th>\n",
       "      <th>avg_temp</th>\n",
       "      <th>Country</th>\n",
       "      <th>Item</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.183443</td>\n",
       "      <td>0.110716</td>\n",
       "      <td>0.542078</td>\n",
       "      <td>Spain</td>\n",
       "      <td>Rice, paddy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.458451</td>\n",
       "      <td>0.000413</td>\n",
       "      <td>0.627257</td>\n",
       "      <td>Madagascar</td>\n",
       "      <td>Wheat</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.183443</td>\n",
       "      <td>0.106159</td>\n",
       "      <td>0.518228</td>\n",
       "      <td>Spain</td>\n",
       "      <td>Sorghum</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.224154</td>\n",
       "      <td>0.890971</td>\n",
       "      <td>Colombia</td>\n",
       "      <td>Potatoes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.458451</td>\n",
       "      <td>0.000355</td>\n",
       "      <td>0.625213</td>\n",
       "      <td>Madagascar</td>\n",
       "      <td>Sweet potatoes</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   average_rain_fall_mm_per_year  pesticides_tonnes  avg_temp     Country  \\\n",
       "0                       0.183443           0.110716  0.542078       Spain   \n",
       "1                       0.458451           0.000413  0.627257  Madagascar   \n",
       "2                       0.183443           0.106159  0.518228       Spain   \n",
       "3                       1.000000           0.224154  0.890971    Colombia   \n",
       "4                       0.458451           0.000355  0.625213  Madagascar   \n",
       "\n",
       "             Item  \n",
       "0     Rice, paddy  \n",
       "1           Wheat  \n",
       "2         Sorghum  \n",
       "3        Potatoes  \n",
       "4  Sweet potatoes  "
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df['Country']=countries\n",
    "test_df['Item']=items\n",
    "test_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "d10e3d86",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:18:57.590048Z",
     "iopub.status.busy": "2021-12-10T09:18:57.589022Z",
     "iopub.status.idle": "2021-12-10T09:18:57.817639Z",
     "shell.execute_reply": "2021-12-10T09:18:57.816940Z",
     "shell.execute_reply.started": "2021-12-10T08:07:43.358626Z"
    },
    "papermill": {
     "duration": 0.319458,
     "end_time": "2021-12-10T09:18:57.817783",
     "exception": false,
     "start_time": "2021-12-10T09:18:57.498325",
     "status": "completed"
    },
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Item\n",
       "Cassava                 0.929995\n",
       "Maize                   0.895655\n",
       "Plantains and others    0.807287\n",
       "Potatoes                0.911998\n",
       "Rice, paddy             0.894790\n",
       "Sorghum                 0.797737\n",
       "Soybeans                0.848586\n",
       "Sweet potatoes          0.841783\n",
       "Wheat                   0.921756\n",
       "Yams                    0.928672\n",
       "dtype: float64"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf=DecisionTreeRegressor()\n",
    "model=clf.fit(train_data,train_labels)\n",
    "\n",
    "test_df[\"yield_predicted\"]= model.predict(test_data)\n",
    "test_df[\"yield_actual\"]=pd.DataFrame(test_labels)[\"hg/ha_yield\"].tolist()\n",
    "test_group=test_df.groupby(\"Item\")\n",
    "test_group.apply(lambda x: r2_score(x.yield_actual,x.yield_predicted))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "a5c627eb",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:18:58.018462Z",
     "iopub.status.busy": "2021-12-10T09:18:58.003104Z",
     "iopub.status.idle": "2021-12-10T09:18:58.226149Z",
     "shell.execute_reply": "2021-12-10T09:18:58.226680Z",
     "shell.execute_reply.started": "2021-12-10T08:07:43.590323Z"
    },
    "papermill": {
     "duration": 0.318865,
     "end_time": "2021-12-10T09:18:58.226859",
     "exception": false,
     "start_time": "2021-12-10T09:18:57.907994",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaUAAAEWCAYAAADGjIh1AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8/fFQqAAAACXBIWXMAAAsTAAALEwEAmpwYAACED0lEQVR4nO2deXxU1fn/32f2JetkY0uCVbSlSq36+1qkrgUUFUqw/aLiQhUtoBYR0FqoX6tgWxYp1SpurcqiaCWWICjghkW6qLVUowJaQwAl+zZJJrOc3x/3zmUmMxNCyM55v17zYubMvfeceyfc557zPM/nEVJKFAqFQqHoDZh6egAKhUKhUIRRRkmhUCgUvQZllBQKhULRa1BGSaFQKBS9BmWUFAqFQtFrUEZJoVAoFL0GZZQUik5CCHGvEGJ1T4+jowghnhZCLNTfnyuE+Kyb+pVCiJO6oy9F70cZJUW/QQjxlhCiWghhb+f2U4UQf+3qcXUmQogvhRBNQogGIcQh3ZAkdXY/Usp3pJSntGM8fe4aKno3yigp+gVCiKHAuYAEJvTsaLqc8VLKJOAM4CxgQesNhBCWbh+VQtEJKKOk6C9cB/wNeBq4PvILIUSuEGK9EKJcCFEphHhYCPEtYCUwUp911OjbviWEmBaxb9RMQAixQghRKoSoE0K8L4Q4tz2DE0J8IoS4POKzRR/PGUIIhxBitT62GiHEP4UQOUc6ppTyALAZOFU/phRC3CKE2APs0dsuF0J8qB/3XSHEiIgxfFcI8YEQol4IsQ5wRHx3gRBifwevoV0IsVQIsU+fza0UQjgjjjVPCPGVEOKgEOKG9lw/xfGDMkqK/sJ1wBr9dXH4pi6EMAMbgRJgKDAYeF5K+QkwHdgppUySUqa1s59/AqcDHmAt8KIQwtHmHhrPAVdFfL4YqJBSfoBmRFOBXCBDH1fTkQ4ohMgFLgX+FdE8ETgbGC6E+C7wR+Cn+nEfAzboRsMGvAys0s/lReCKBP0c7TX8DXAy2nU6Sd/+Hv1YlwBzgTHAMGD0kc5TcXyhjJKizyOE+D6QD7wgpXwf+By4Wv/6f4BBwDwppVdK2Syl7LAPREq5WkpZKaUMSCmXAXbgiL4XNAM2QQjh0j9fjWaoAPxoRuMkKWVQSvm+lLKujWO9rM9K/gq8DTwQ8d2vpZRVUsom4GbgMSnl3/XjPgP4gO/pLyvwOymlX0r5ZzSDG492X0MhhND7na2Po14f35X6Jv8L/ElK+ZGU0gvc28Z5Ko5DlFFS9AeuB7ZIKSv0z2s5vISXC5RIKQOd0ZEQYq6+FFerG4ZUIPNI+0kp9wKfAON1wzRBHydos5XXgOf1Ja3FQghrG4ebKKVMk1LmSyln6gYoTGnE+3xgjr50V6OPNxfNwAwCDshoReaSBP0dzTXMAlzA+xF9vqq3o/cbOcZEfSqOU5QzVNGn0X0V/wuYhRBf6812IE0I8R20G2CeEMIS56YaTyLfi3ZTDTMgoq9zgTuBHwAfSylDQohqQLRzuOElPBNQrBsqpJR+4FfAr/SAjU3AZ8BT7TxuJJHnVAosklIuar2REOJ8YLAQQkQYpjy0WWZrjuYaVqAtPX5b93m15is0IxcmL/GpKI5H1ExJ0deZCASB4Wg+jNOBbwHvoPmZ/oF2I/yNEMKtBxWM0vc9BAzR/SthPgQmCSFceu7MjRHfJQMBoBywCCHuAVKOYqzPA2OBGRyeJSGEuFAIcZruu6lDW84LHcVxE/EEMF0IcbbQcAshLhNCJAM79XP5mRDCKoSYhLZMF492X0MpZUjvd7kQIls/v8FCiIv17V8Apgohhuszxv/rhPNU9COUUVL0da5H81Hsk1J+HX4BDwNT0GYx49Ec7vuA/cBkfd83gI+Br4UQ4aW/5UAL2s32GbTAiTCvoS1F7UZbdmomeimqTaSUX6EZg3OAdRFfDQD+jGaQPkHzE61q73Hb6O894Ca0a1EN7AWm6t+1AJP0z1Vo12R9guMEObpreJfe19+EEHXANnS/m5RyM/A7fb+9+r8KhYFQRf4UCoVC0VtQMyWFQqFQ9BqUUVIoFApFr0EZJYVCoVD0GpRRUigUCkWvQeUp6WRmZsqhQ4f29DAUCoWiT/H+++9XSCmzjrxl+1BGSWfo0KG89957PT0MhUKh6FMIITpVlUMt3ykUCoWi16CMkkKhUCh6DcooKRQKhaLX0KVGSWilm/+jFxl7T2/zCCG2CiH26P+m6+1CCPF7IcReIcQuIcQZEce5Xt9+jxDi+oj2M/Xj79X3FW31oVAoFIreTXfMlC6UUp4upTxL//xz4HUp5TDgdf0zwDi0ol/D0OqxPAqagUETbTwbTTDy/yKMzKNo2l7h/S45Qh8KhUKh6MX0xPLdD9GELtH/nRjR/qzU+Bta6YGBaBU6t+oFw6qBrcAl+ncpUsq/6dL7z7Y6Vrw+FAqFot+wZs1aThj2TUxmMycM+yZr1qw98k69nK4OCZfAFiGERKuA+TiQo6slA3wN5OjvBxOtuLxfb2urfX+cdtroQ6FQKPoFa9asZcbsebh+cCu5E4fTtL+YGbPnATBlytVH2Lv30tUzpe9LKc9AW5q7RQhxXuSX+gynS2XK2+pDCHGzEOI9IcR75eXlXTkMhUKh6FQW3Hsfrh/ciiN/BMJswZE/AtcPbmXBvffF3b6vzKq61CiFK09KKcuAQjSf0CF96Q393zJ98wNEV6Qcore11T4kTjtt9NF6fI9LKc+SUp6VldVpCckKhULR5ZR8sQf7kOFRbfYhwyn5Yk/MtuFZVdNZ15N7x3qazrqeGbPn9UrD1GVGSa9QmRx+j1Zx8yNgA1phNvR//6K/3wBcp0fhfQ+o1ZfgXgPGCiHS9QCHscBr+nd1Qojv6VF317U6Vrw+FAqFol+Q/41h+PYXR7X59heT/41hMdse7ayqJ+nKmVIO8FchxL/Ryim/IqV8FfgNMEYIsQcYrX8G2AR8gVaN8glgJoCUsgq4H/in/rpPb0Pf5kl9n8+BzXp7oj4UCoWiX7Dw3ntofP1hmkt2IYMBmkt20fj6wyy8956YbY9mVtXTqMqzOmeddZZU2ncKhaIvsWbNWhbcex8lX+wh/xvDWHjvPXGDHE4Y9k2azroeR/4Io625ZBfO957hv3s+PaYxCCHej0j5OWaUooNCoVD0UaZMuZr/7vmUUDDIf/d8mjDq7mhmVT2NUglXKBSKfk7YWC249z5KXtBmVcuWL+mVoeNq+U5HLd8pFArF0aOW7xQKhULRb1FGSaFQKBS9BmWUFAqFQtFrUEZJoVAo+hh9RTKoI6joO4VCoehD9Fch1jAq+k5HRd8pFIq+QKJE2KYty6mtjCvz2aWo6DuFQqHoJ3RkGS6RZFBdVQUzb7m1q4babSijpFAoFF1IIsPTUeVuR1JqXCFWc5KHJ59+tsvOo7tQRkmhUCi6iLYMT0eUu2fecitNjY1UbFoeJRlUsWk5IPA3NnTfyXURyqeko3xKCoWis2lLCLXkiz3k3rEeYT4cbyaDAUofnEQoGIx7PJs7BX9TAxmXzqbu73/GX7kfa8YQUs7+EZWblmN1JtHirevy84pE+ZQUCoWiB+hM/09Y1bu99ZDC+BsbsHpysSRnMujGR8i/cwODbnwES3ImwuZk2tTrOnZyvQhllBQKRb9i9NixmB1uhDBhdrgZPXbsMR+zo/6fzJxBCQ1PW8rd8QzgmjVrMdmdOE8eSeXmFVH7lRct5qLzRvHIHx4+5nPtadTynY5avlMo+j6jx47lzXffI2v8POxDhuPbX0x50RIuPOcstm3Z0uHjdqQe0Zo1a5k241b8wkLmpbON8dS9toInHnqQKVOujqmHNG7saFavXos3IMmacGfUPjLQgu3UsXiL38I9/AKadu/EX1WKsDm56LxRx3R+x0JnL98po6SjjJJC0fcxO9xkFSyIMR7lhQsJNnsT7nekYnnCZMLqycVfpflwUkdOxnXKqDb9P2FDFvRWU7tzHf7K/VhSc0izC8q/PhDT/+1z5lFZXQsCsgvmx57DhsWYXSn4K0sx2d2Emr2Y7E5+euNPenSG1NlGSSk6KBSKfkPI1xTXhxPyNSXc50gKCWvWrMXiTsczZroxc6ncvIKWin1t+n9KvthD7sThCLMF9/DzgcOBDPH6b/JLzK4UArVfx5xDoL4CAM/ow2Oo2LyCYF0Zj/zh4XZXoO0LKJ+SQqHoN5jszrg+HJPdmXCfBffehzjpXKq2rWTfsklUbVuJOOlcIzR7wb33kXH53KjQ7Yxxs/D+a2OblVvbE8jw7dNO49obbqK+/CuC3moCtV9j9eTG7Fe74zmyJtwZNYbMcbOwOpM67O/qrSijpFAo+g0XnjeK8qIlrYIAlnDheaMS7vPl57vxFr+FZ/R08uasxzN6Ot7it/jy891A4gi6kK+pzdnIkUqQD87N49P/HiB70gLy5haSfcU9CJsrbiBDoCZ29mQfMpxAk7dD+U69GbV8p1Ao+g3btmzRgh0KFxLyNWGyO7nwCEEAVmcSnnGzDB9OeBZSteHXgDbjadpfHOXj8e0vJv/ExEt3kLgE+Y533+X6aTcTbG7EkjaAoLfaMCbJZ1xOw64tJI0YS9XWlfirSjHZXKR4MvElGEN4mTAS+5DhlLyw5+guXi9BzZQUCkW/YtuWLQSbvUgZItjsPWJUWqDJm3AWAkee8bTFlClX8989nxIKBvnvnk/Z8e67PP7MWjInzidvbiEZl9xGzfZn8Ra/DUDaqKsIeWuof+8v+CtLyRowmGefepxHfv+7hGPoSL5Tb0bNlBQKxXFF66CAzAGDEs5CIPGMJ9HSXbyggx3vvsuTTz9LwO8n+4p7omZlGeNmUbVtJe7h5+PbX4ywOVn1xydijq8d49daAq0rifNGfU/r5/M9mPYvwpZ7GoGqg/irSjHbXdx8w9QuuHpdjwoJ11Eh4QpF/ycy0i4cxVb7yhKExUbKxbOMtsbXH+bRNgzP0Ry/qui3hDCROX4eh9YtIG9OrLTQvmWTyJm8kPKiJQxIT+JA6b4jHreiaAnuEWNJG3UVNTueo2HXlqj8rI6ew9Gi8pS6CGWUFIr+T6Ik2OCbD5OUnHzMIdXxjn/gsZvIuOQ2HPkjOPjUTDyjp8f0X/bSfSBMWDyDcAQbY4xJonFXbVvJoBsfSXjctpJ7OwuVp6RQKBQdJFFQQGn51zEJre0lcrlOhiQZp1Rw8KmZhlhqZN5R6sjJVG5eQca4w7Oy8g2LMTmSSL/gJ7iHn09zyS4W3HtflFFKNG5/5X4A/JX742vs9cFgB2WUFArFcUPCSLoOBgW0Trwt/f3V1LyzKkpWqKxwkeGzCifRVr76EIGar7GkDUCYLAyZ+bRxzHjGJNG4rRlDALBmDInvF+uDwQ4q+k6hUHQbHVHa7kyOJZIuHq1zhECSeensqJyh5DMup7xosdGn2Z2OlCFSRv4vBHy4T/tB1DEjjUn4en35+W4qNy6NrqFUtATnsJHIYADnsJFUtMrPOtJ59fRvkRAppXpJyZlnnikVCkXXsXr1GpmcNUjmXPmAzJv7ssy58gGZnDVIrl69ptvHMfSkU6QwmeTQk045qv5nzLxFWl3JEoQ02V0SzFLYnBKEFHaXBCHz5r4s8+/aaLzy5r4csf3hf4eedIqcMfOWhNek9fVKGTlZmh1uKYTJ2DfyPFp/buu8OvO3AN6TnXgvVoEOOirQQaHoWjqitN2bmHnLrVqOkR7hFi/irbxoMUkjLib9vGuN/ZpLdlG2fiEEmuOKtybSrevK69WZx1bRd12EMkoKRddiMpuPutJqb8LmTsEz4W7jRp4wkq5wIdkFC6JKZ5iTMxicZD6qG35XXq/OPHafqzwrhDALIf4lhNiofz5BCPF3IcReIcQ6IYRNb7frn/fq3w+NOMbdevtnQoiLI9ov0dv2CiF+HtEetw+FQtFz9BXlgUS+Fn9jQ1SEW6KIN+lromz9QvYtLaBs/UJCzY34v/4cr9drHKt1HzNvuTWmz668Xr35t+iOQIdZwCcRn38LLJdSngRUAzfq7TcC1Xr7cn07hBDDgSuBbwOXAI/ohs4M/AEYBwwHrtK3basPhULRQ3R2kEFX0FpxuyrrdK6degMmm6YyfvDJGYYkUDjiLRLf/mJMDjfZkxZgdqchQ0EItWDNyMU3dBQzZs9j5i23xqh6P/7MWqqyTo9S+R43dnSXXa9e/Vt0poOq9QsYArwOXARsBARQAVj070cCr+nvXwNG6u8t+nYCuBu4O+KYr+n7Gfvq7Xfrr4R9tPVSgQ4KRddzLEEG3cHQk06ROVc+IPPv2igzx8+TJmeqNKdkRQUEmFOyZMZld8iUkZOlyZ0e9Z3JnS6Tz7hc5t+1UeZc+YC0pA00vrOk5siUkZOl1ZVs9BF+5Vz5gLRm5kV9Dl+frrpenXVsOjnQoavzlH4H3Akk658zgBopZUD/vB8YrL8fDJQCSCkDQohaffvBwN8ijhm5T2mr9rOP0EcUQoibgZsB8vLyjv7sFArFUTFlytW9uvhcZJJq9fZnMdldhhoD6Aril86m7KX7saRmYU7OoGz9/ciWZkwON+7hF+AZMx3QRV1rD0XVYKraujJmGTC8bTgRNvy55IU9XXq9eutv0WXLd0KIy4EyKeX7XdXHsSKlfFxKeZaU8qysrKyeHo5C0atpb15Lr81/aQf53xhG+V9+w77fTSZYW5awjpEM+PCMnk6osRYZDDH0pJPJmvgLwyBBdHJreD9/VSlWV1LcZb/IbXuLf6cn6Eqf0ihgghDiS+B5tCW8FUCaECI8QxsChLU9DgC5APr3qUBlZHurfRK1V7bRh0Kh6ADtrW7a16ugnnhCHr6Dn5FdoJWWsKQNiGtAhMVO5asPEfI18oOLzovro6nYtJzUkZOj9jPbXUybel3sthGJsL3Kv9MTdOZaYKIXcAGwUX//InCl/n4lMFN/fwuwUn9/JfCC/v7bwL8BO3AC8AVgRvM7faG32fRtvt1WH229lE9J0R30dp9KIiJ9La39Hh3Zrjcx/NRTteTXcBKszSWtmUMNv1Jrn5LJnWZs94MxY4zjRP62WQOHSEdytL/JkuSRM2beErPt0Sa+tofu/DujLybPCiEuAOZKKS8XQnwDbebkAf4FXCOl9AkhHMAq4LtAlW5UvtD3nw/cAASA26WUm/X2S9H8Vmbgj1LKRXp73D7aGqPKU1J0NfHKD3RXeYFjpb15LX0tF2lwbh4H9x/A5E5tlQS7BLMzlUE3PkzDR29QteURpN+HsNqRIckPLjovbvHAyETYjKwBCJOJikMHj0l5/Gjp7r8zlTzbRSijpOhquipDP5EiQGfS3rH3JdWGb592Gp/s/gLZ0oQlbQBp515jCKaGVRjyZr9Ac8kuKl99CKdVtHlj7y0PHd39G/S55FmFQqFR8sWe+OUFvuh4eYHu8uG0N6+lV+e/RDB67Fg+/e8BsictiFuW3D5kOLKlieaSXZRvWIwT3xGNS2txVkf+CFw/uJUF997XXacFdM3fWXeijJJC0U10RRZ9d90Ip0y5mkeXL8H53jOUPjgJ53vPxL1Jt3e7Y6Ezovve3L6DrPHzoq5bxrhZ1O5cB+jBDDYH5YULCTXWkpF5ODo3Uf+9xRj0ZrWG9qCMkkLRTXTFLKI7b4RTplzNf/d8SigY5L97Pk1oaFpvB3RaiPixzAxn3nIrNncKQpgI+ZoS5go1l+yivGgJ0t9CVoE2k2o663qmTr8NIUxce+NNfPn5bizpQ6jKOt3oP54xqNnxHBaHu1vD4/vKbDURyqeko3xKiu6gs/0/vd2H09l+lo6eb2uF74NPzohKig0fRytLLpAtPnKuXBj1ffX2VTTseo2s8Xca51K5eQWu4RfgKf+QhffeE3WuNTuew7tri9Fnd/qYusPPGEYFOnQRyigpOoPuvBmE++sNzvVEdLbR7Gh0X2uFb2/x21S//XRUhdiwmrcj2EhDxdfkzonu5+CTM/GMiVUFr9q6kkD1fkLBYNTvb3G4o/o81nPvrahAB4Wil9ITiaPd4cM5Fjp7efFI/pLRY8didrgRwoTZ4Wb02LFArMK3e/j5pJ17LWUv3XdYzdtbw+AkM48uX0L+ibH9+KtK4y/5VZUa/UcuXQaavb3Cx9TXUEZJoWgH7XGu91T0VXt9PT1BZzvd2/KXDM7N4423dxDyNWFyuLHnjeDNd99j9NixcaV9LMmZWKxWpAwR8nmRMmRcv3j9CJsr7rmY7a64/pq+HnDQUyijpFAcgfbOgHpL9FVvorOd7olmhjvefZevqxuMEO+sib/Ad/Az7INO4c3tO5g29ToqipbESPtMm3rdkftZNomKlxfhyDuNis0roo5RuXEpN98wNe6DQF8POOgplE9JR/mUFInoj4mj3Ul3+NnMDjdZBQtirn35yw8QatZmQTNvuZUnn34Wf2MDVlcS06ZexyN/ePiozuHLz3djdSYRaPKSf+KRz6W7fYw9gQp06CKUUVIkor3O9d4edNDfiLzhy5Akb25hzG+0b2kBJruTYLO3B0fav1GBDgpFN9Ne30BvDzroT3z7tNOi8oWEzRFfzdvm5MLzRvXQKA/Tl8t5dDfKKCkUR+BofAO9OeigpzmaG3PkttmDcskaMBiT2YzFasdkd1H80ccgwZ73HTxjpiP9Acpb+YzKi5YgW3xxhVO7k75ezqPb6UzJ8b78UqUrFG3RV0tO9BZWr14jk7MGRZVySM4aFPc6xtsWq1OCWQq7SyKEtGbkGeXIHUO/K60DTpJY7FLY3VpZCbtbYrFHlZY4lrEfzW/fevvMnEF9rpzH0UBfLF3RF1A+JYWi6zhh2Depyjqdpj078Vfux5oxBOewkXjKP4wJAmkdMFK1dSUNH7+JsFjjqinUv19E7s/Wsm9pAcLmRLY0YbJry3bHOks6Wj9hvO3LNywm/cIbSDr1ImO73lzO42hRgQ5dhDJKCkXXIUwmzCnZZI6bZdysKzavIFhXhgyForYNB5ZUv/Ek3uK3CDV7ETYn2ZNio+uqtq7EX1lKzpWLKCtcyKqnnujUJdOjjahMtH3lqw8x+KdPtOsYfQ0V6KBQKHolbfmMrM4kMsfNikoszhw3C6szKeY4mTmDKP/Lb/B+toOsib8gb24h0p9AQLWqFGFzUl60mG8N+0an+/CONvcs0faBmq9VvlI7UUZJoVAcM62d+VVZp3P9tJsxmTQD5W9qiH+zbvLGHKeiooLmkl1RpSWsGbkJo+tki48Lz/l/fPyf/3T6eR2tKkOi7bMGDlZRme1EGSWFQnHMREosNX62g8bit8icOB/Ppbezv6IOYbHHv7mfqN3chTBjsru45pprQJiQLdEzI0feiDjRdYuRLT6kDHRZhN3RqjIk2n75kt+qqMx2YjnyJgqFQtE2JV/sIXeiZkRqd64jY9wsgt5qav+6hoxxswjUV1CxaXmUKnfj6w+zbPkShDAh7FqAgjUjF+fJI2nYtYWvVs9l0PW/A6B53y6SRoylattKI1AiacTFeMo/7NLzChuPBffeR8kLmirDsjZmOUe7vSIWFeig058CHfqDtEl/OIfjiUgHf8niCWRcejvVrz9OyOfFmpFL6sjJANS8s5pAzdcMPelkXA4rxR99gsmdnDCqLnvSAuxDhrNvWQF5c2IVG/pLBFtfRgU6KNqkPyTq9YdzON6IXLYSjiRq3lmlBSnMKcQzejo1258FYNC0RxEmQdlXB/n0vwcQNjtZ4++MKUvetHsnsqWJssJFWqi3Nb5iw9FUdVWqCn0DZZT6GT1VPqEz6Q/ncLwRKbFE0E/mpbNjDE3tznWaIbHYaPIHyRo/L8Z3BNFRdWZHEikj/5cklzPGV1NRtATndy5t14PL0TzoKOPVsyij1M/oD+UT+sM59Fdm3nIrNncKQpiwOJNxupIw2VwIYWL2vLtYeO89yJbmmN+vcttj+CsPUrb+fvwtPkBSVvhrhM3ZRlRdE2nnXUfaqKtoqKuJ0hWs2vBr3CPGkn7ete16cGnvg46apfc8yij1M/pDYbH+cA79kZm33Mrjz6zFM+Fu8uYWkjnxF7SYbDiGfoe8uYWYzp/JTbfdgdkRXQyvZPFEgnVlIIOYXWlkXDab7Em/RFgsyBZf/Ki6kMCSmg0c/u1bV3VNG3VV1Pg6kj/Uens1S+95lFHqZ/SHwmL94Rz6I08+/SyZEblDjvwRZI2fh6/0I+NzysWzcDnsVBQtoXr7KkoWT8TkSiZ70i/Jm1tIxiW3UfvXNQS91WSNn4ewOwj5milbv9AoS27LOoH8O7QIvpp3Vsf97Tsrf6j19mqW3vOokPB+Rn8ISe0P59Af8TfGT4ANRdQqsg8ZTlldDQMHDebrXa8hrDYjCRYw/EtV21YycOrvkb5GACTE1EMKKyE8unp1zG+/8N57mDF7HrTSpFu2fEncsbd3+/xvDKNpf3GUTJCapXcvyij1Q6ZMubrP38D7wzn0N6yuJO1mvvvdw5p0dhdEGBLf/mIw2/iqrILsSb/k0PPz4wcyVO7XfEd2F0iJ1WrBF8cYDD3p5Lh/Bx3JH9rx7rs8+fSvoyrPHquxU3Q+avlO0emo6KX+yXmjvsehP/8qSpMuu2A+JkcSla89okXEbVqOZ+xMI9ghUSCDJTWH8qIliFCIVX98gj8+9uhRL9keTe2qNWvWsvrFQsMf5plwN6tfLIz521SFGnselTyr05+SZ3sSVRK8/3LCsG9SUlpKdkG0Wnf19lXUv78B2dJsRM2FVb0rXlmODAXIGj8vqpRDyNeIMFtJctnJyMyi5Is9ZGQNoMXvp66qAmF1kOFJ53fLOufv5mjVvhXtRyXPKno1KnqpffSG2WSi6q6JxlPyxR6k73Bekbf4bfY/fB11O1+IiKpbgDk1G0f+CMqLFpN52WxkMGAEMpS//AD2wd/E5HDjHn4+9TXVRvh1yzfOxesLkHPlIjxjZ1Djk1xzzTVkD8o95uujAhj6DsqnpOhUIjXQwtiHDKfkBfWfP0zkbDJ34nCa9hdrfgzottlk6zH49hdT9pffYErK5MvPd3P9tJvZ8e67PPKHh4198r8xjJLS/Yf9Sp/tiJoBVW5eQdp515GpBzIkjbiYsvX3g5RkX3FPzCylrHARJleq0d60ZyeZ4+dFaeaFj32s10cFMPQdumymJIRwCCH+IYT4txDiYyHEr/T2E4QQfxdC7BVCrBNC2PR2u/55r/790Ihj3a23fyaEuDii/RK9ba8Q4ucR7XH7UHQ9KsfoyHT2bLIjs67WYwh6qzFZ7WReejt5cwrJnDifJ559zjjWmjVr8Xq9SF8jZYWLaPj4zajSEpGqDeFAhrRRVyFbmpF+X9xZimxpRPqbKV1xJSW/HY+/ohT7kOGGoGtnzrbbSjPoDbNWxWG6cvnOB1wkpfwOcDpwiRDie8BvgeVSypOAauBGffsbgWq9fbm+HUKI4cCVwLeBS4BHhBBmIYQZ+AMwDhgOXKVvSxt9KLqY/ppj1Jk3rvYuJbWnz44qELQeQ+3OdbHSQJfPZcG99xl9mM6fSd7cQpAS2dLYZlSdNWOIEV0nrPHLVlg9uchAC1kTf4E5NRuTKxXf/mL8lfs7faktUQADEPf6zbzlVmWoeoguM0pSo0H/aNVfErgI+LPe/gwwUX//Q/0z+vc/EEIIvf15KaVPSvlfYC/wP/prr5TyCyllC/A88EN9n0R9KLqY/hi91NnSM+2ZTba3z47OulqPIZEh+HLvbq69YVpUH7KlCasnftE9S2oO5RsW4xw2kvKiJTjyTsMkRIxqQ+XmFThPHok1Y4hRhVZY7VRsWo4lNadLZtvxovUSXb/H//h0n5Ya6suzvzaNkhDC09brSAfXZzQfAmXAVuBzoEZKGdA32Q8M1t8PBkoB9O9rgYzI9lb7JGrPaKOP1uO7WQjxnhDivfLy8iOdjqKdHE2obl+gs5fb2jObbKvPyBvOl3t3d2hW0XoM8QxBzY7nMLlTkf5oLTuTw43z5JFUbl4RLQ+0YTGBhipCTfXUv19EyFtDS+l/uOD872MOtmiK38sKqNq6EtfwC2gsfssoaWEfMpxgXTlp515LoL6S8g2Lu2W2nWjWGvQ19tlgnb6u33ekQIf30WY3AshDWwoTQBqwDzihrZ2llEHgdCFEGlAIfPPYhtu5SCkfBx4HLSS8h4ej6KUkCt748vndnDDsm0dd66k9iZ8JA0bW7YkKUDj45Iy4SadHmlW0HoPd4aKscCGypQmzOwMZ8hNqrMOSNgCRFJ3Y6h5+AQ27tmhF97auNBS9ZVAiTBYwW8gumE+gvoLaHc/x+tZtpHgykYEADaEQ/spS5Cd+0s67Dvfw8wHdANpdVG5ajtWZxLSpU9i85ZkuV/RIFABh9eRGbdeXgnUiH2hAU9FAN6p94QGxzZmSlPIEKeU3gG3AeCllppQyA7gcaHf9YSllDfAmMBJIE0KEjeEQ4ID+/gCQC6B/nwpURra32idRe2UbfSgUR02i5TZrRm6Hn0KPNJtM1KfF6Y6aQaWdew0Vm5Z3aFYRHsOqZ1dhdaeQXbCAjEtnI2WQrAl3GVp1MhSkrHCR0Yfr5HOQLT7qP9iIv7IUYXUgfU2YTWASmr+p4pUHqXlnFRmX3Ebe3EKcY2eD3cmqPz3J6tWrcVoFZnc6MhigevsqGnZtOVx/SU9sXXjvPV0+2443a63cuBTnySNjrn1fCdbp6+Hv7fUpfU9KuSn8QUq5GTinrR2EEFn6DAkhhBMYA3yCZpx+pG92PfAX/f0G/TP6929ILbN3A3ClHp13AjAM+AfwT2CYHmlnQwuG2KDvk6gPheKoiXvj2ryC1HMmd9nSTqIlPn9TtP6ce/j5pJ17LWUv3d9hH174yTroraZq66MxUXVZ4+eBlFHLb8lnjUdYrPxgzGhCvkZWr16NKzWdzIIFeoVYa0zghCviad3wOy6bRMMHRTF9dtdyWTwf6E3XXYXc+06fDdbp6xGw7TVKB4UQC4QQQ/XXfODgEfYZCLwphNiFZkC2Sik3AncBdwgh9qL5f57St38KyNDb7wB+DiCl/Bh4ASgGXgVukVIGdZ/RrcBraMbuBX1b2uhDoThqIm9c+5ZqN+XIpaeueApNFDAy9MSTY244luRM8vPzOjyr+HLvbio2/Y6KoqVxayFp4dtNJJ9xOcLmwl9ZivdfG/np9VPYtkVbMGntAwvUHmrzad2YKYaCMT6r1tseDR1x8LeetT7yh4f7dLBOX4+AbZfMkB7U8H/AeWg+pu3AfVLKqq4dXvehZIYU7aGn5WriyTjVvrKE5CQ3FYcOkv+NYUfl4/r2aafx6X/3kzX+TuxDNB9VxiW3xSa6vnQfmMwAXHTeOYYxCmMym8m9Y72h8n3wqZl4Rk9v13XqrGuqJK4Os2bNWs1f+MWeo/6bOFp6RGZISlklpZwFfF9KeYaU8vb+ZJAUivbSmU+hHX2qj3yKD775MMJiw3T+zHZHWo0eOxZhsWOyOSn+6GOEMBP0Vif0UVVsWo4MSZKcdlY99XiMQYLYJaPUkZPb7evqrGuqJK4O06cjYKWUR3yh+Y+KgX365+8Aj7Rn377yOvPMM6Wi65gx8xZpdSVLENLqSpYzZt7S00PqMKtXr5FDTzpFCpNJDj3pFLl69ZoOHSM5a5DMufIBmTf3ZZlz5QMyOWvQEY+1evUamTVwiAQhhdUphcUuU0ZOlvl3bTReOVc+IIeedErc/X8wZozEbJfC7pYIIa0ZeTJl5GRpTsmSmePnyfy7NsqMy+6QwuqQCJO0pA2Uwu6WNrvzqM/HkZwuswYOOeJ1Wr16jczMGSSF1SlByKyBQzp0TYXJJPPmvhx1LfLmviyFyXTUx1K0H+A92Yn34vZq3y0HLkYLOkBK+W8hxHmdbiEV/ZJwGe3MCXcbyyqPP6Nl00dqq/UVOqPWU0fCdtesWctNt91BysWzyNOvY8Wm5dR/+Cq2zLxoH1ec8GWL1U7IZIZgC+ZkD6mjf4olOZPKzStwf/sianeuwz38fCzJmSBMICXBxhoI+PEF/Ue8JuHzMsK4H334iNcpSoMvYsmtIyh9u/5Be31Kf5dSni2E+JeU8rt627+lJiHUL1A+pa7D5k7BM+HuGJ9B1YZf0+Kt68GR9RytfTAAMhig9MFJhILBuPsk8r1UvvoQwmJl0I2PGG2R/hhhsoDJjMnuImvCnTECqmZ3up5vtJ+cyQsp37BYG4+UyKY6hEkkHNOx0pk+OuVT6hl6qnRFqRDiHEAKIaxCiLloEW8KxRFJVEbb39iQYI/OpydkV9rqsyNhu4nyTwK1h7SE1Ah/zLixo7HYHAhhQtjsEPRjsjkN31GMgGpVKcJio2z9QkKNtZhdKSSffgmWtAEdmmm093p3Zk5Nf5S4Oh5p7/LddGAFmlzPAbTE2ZldNShF/yJcRjsma96V1C3990SpiCP12ZGy24mWpyypOYSa69m3rEBXQ7iOlU/8ERkIYHKnGpF14dkRaPlNUWXJbS48+nJexeYVOIeNxPvxG1hlgIX3Lu3Uc2/POXV0ya0zllYVPUt7Z0qnSCmnSClzpJTZUsprgG915cAU/YdpU6+jopUgZ0XREqZNva5b+u/KqKxEM4L29OmwmDj0/HxKV1xF8M0jLzONGzuassJFHHjsJkoWT+DAYzdR9vKvNb25Zi/C6sTf2KAZpFAQZAiTNf7sCDRpH2Fzcuj5+ZgsdoTJbIij1r+/AZob+Mm1R3+TP5rr3ddzahSdT3tnSg8BZ7SjTaGIIRzM8OTTv8bf2IDVlcTNU6/rtiCHrio8mGhGsOPdd/ly727yEvQZuV84YKFy41J2vPsuU6ZcHTfHZMe77/LYU39C+prAmULGpbdjSc6kvGgxjtzv0vL152ReNtuYEVVsWk7audcagQwQPTsKS/tkF8yPmUW5ThmF9PvInLyQ1S8+zKhzzjkqw3Q017s9OoCK44s2Ax2EECPRwsFvR4vAC5MCFKhAB0VfoKsSXhMdt+LlRQhHStwkVOd7WnWWRPvdfMNUVr9YGJMc6w9BZpwqr2Z3OmWFi8gumB8bSLJtJYNufCTmfdlL94EQZE/6Zdx9PKOnR21/tNeppxOMFd1Ldwc62IAktBlVcsSrjsPacgpFr2bhvfdQ99qKmOXDE0/I61DwQ3jJ7su9u6nauhJv8dvGd+GyB2nnXhNT2qHutRU01NfHLTcRqK8gFAzy6COP0FBXy6Hn5/PV0z8j6K0mYHaQmaDKa6C+AulLXHAv8n1zyS7KixYj/T4IxK8G668s1XT9IkpKHG3QgVqSUxwLbS7fSSnfBt4WQjwtpSzppjEpFJ1OoKWZylcfIlB7CEtqDsGWZt569z0yx887quCHqKW3ibHBA+GyB+GcoaptK/FX7kdYbNgdTswXz8O6dWVU4Ie3+G1q3llF9hX3HF5+04MNarY/S6CuLMaAeD/9K4G6cipfWY6wOeMHkmQMMd6HI+tkSzNShrTZTJx9hM0VpevXkaADtSSnOBbaG+jwZFjxG0AIkS6EeK1rhqRoTWeHMx/peB3t71jG2ZUh2wvuvQ/P+LsY/NMnyL9zA4N/+gQWd1rM7KM9wQ/xnPgZ42ZRtfVRShaPp6xwIRbPIEAzUqkjJ2NJzQEg9TKtv9RzJkfNomreWR2jqJ05bhbe4rfIGDcLYXNFhY8fWvdLGvfsJHvSL8mbW0jymeNjKrtWbFpOytk/0mdHS7BkDMFsc7B69Sog/mym7rUV2G02o6TEscxw+rTMTQL6cjXXvkR7Ax0ypVYTCQApZbUQIrtrhqSIpLPDmY90vI72dyzj7OqQ7XiO94Qq1kcIfkjkxA81e7Fm5OI8eSTeXVuo3r4Kq2cwNe+sIvPS2RxaN9/oz5hFbV2Jv7LUOEbrYwZry4zluYqNy3CfNhrvx28Saq6P8geln3ctgDYT8jcjLDaklPosyoFsaSZvYDYLlz5kXM+4s5mHHoxtUzMcoGfSCo5X2qvo8D5aYMM+/XM+UCil7DfRd7010KGzncZHOl5H+zuWcXa1Yzze8Q88dlPCQISOnGfZS/djdqcR8jWSdPolNP17EwF/gOwrNONRuuJKsib+Ima/8pcfwJzkiauoXVa4CGG2EvI1AhiKDIeen0/e3MIYNYh9SwvIuXIRZYWLADAnZxKo2IeUoWO4egpQwRtt0VOKDvOBvwohVgkhVqOVrri7swahSExnV5E80vE62l+i/b7cuxubO4WZt9za4TEdKwvvvYfKv/w6Kr8n4K2hvGjxUTvj4y17VWxajmesVvJBWCx4//M6gWYvBA8HE4SavVS0Cnyo2LyCULOX1JGTKd+wOOY7R95pyEALBP0Ii5Wk71xM1baVCKs9rhqEsLso37AYGQhgSc4iY/RPuy1Bub/T16u59iXaW7riVbScpHXA88CZUkrlU+oGOruK5JGO19H+2ioZ7plwN48/szahYeqOSpnSZNFKc89ZT8Ylt2GyOTAnZ1K14ddtStK09iMAUQX/KjYuAymp3PQ7qratxH3qaEK+BvK/MSzqvKyZubiHX0DVtpXsWzZJ23b4BVgzcwEI+byHK7tuWwmhEM37diH9TVgzcpG+RrzFb+EZPR3P2JkxZSHKi5YgW1pI+s7F5M1eh2f0zZQXLeG8Ud/rtGt4PNPXq7n2Jdo0SkKIb+r/ngHkoVWbPQjk6W2KLqazw2vjHa/2lSV4vV5MZjMN9fUx4dMdnUFElgzPHD+PJ59+tlvOsTUL7r0vJqgha/w88LcQaPYmdMaH/QhNZ11v1Cq66bY7uH3OPEq+2IOw2gkFA2Redgd5c9bjGT0db/FbSL+PhffeE3VeKWf/CO9H2/CMnn5424/fIOXsH1HzzmpMdjfS10jenELAhAz6ydZLi3vGTMfkSsU9/AIc+SNIOvUi0s+fSuWrD7FvaYGmV+etAYuF+veL2Le0gPKXH8A+6BT+8eHHHXbIr1mzlqwBgzHZXAhhIntQbqcFxfQ1VJh7N9JWXQvgCf3fN+O83ujMGho9/erN9ZQ6o35PouNl5gySzvScDtXBSXRc0Or0hOvzhOvaaC7M7jnHSBLV2UGIhHWHpJRy6EmnyJwrH4ipVWRJG2hcq8g6ROHvhe1w7aHI8zLbnNKSNlAiTNLs9kgsdokQEoRWw8jmlCkjJ0thc8Xt15ySHXsOCGkdcJI0udKksDqOqrZSW6xevUY6ktOlOSUr6m/DmZ5j/DYdrQnVV+nKv9G+DJ1cT6ldgQ7HA7010KGr6QoHbm8rVRHvHKu3r6L+/SLwN5N/Yvxy0YnKS+xbNon8OzcA0coJxvdLC4zggjVr1nL7nHlUVlUj/c1YM3JJPWcytTvXGcENpb+/mqTTx1H/wUZAIn1NCQMZ8u8qMtpaKzVUvvoQUoZIj8gzOlI5jLau2f6KujaDQZTzXwHdHOgghJjU1quzBqHoObrCgdvTAqytab30Yui+TVpA7pzEJcQT+8mGGJ8jlRO8xW9z8MkZgGaYR48dy7QZt1Ld5Cf7Ci2nyDNmOjXbn8VfWYp9yHD2rbiaUIuP+g+KyC6Yj/Q1IayO+IEMVnu0H2nDYkK+RrzFbxslLDIjBFfD+3XE71HyxZ7EYfPHGBSjULTFkfKUxuv/ZqNp4L2hf74QeBdY30XjUnQTXVGts6cFWFvTOifH4nCTFTGTS1T1NV55iYpNy0k/f6qxTbh0RMNHbxg5SeFt3yxagvS3kD1pXlRfGeNmaergv78K2dKMsDmRvibsQ4ZjcrjBZKFi0/KoY1VsWg4Iytbfj2xpxpI2gPQLbzAEV1sq9mHNGGJIBclgoM1yGPFEX41r9IV2jUImW1yliMigGFXpVdHptGeND61+0sCIzwOB1zpzHbGnX73Zp9SVHItfoD1r7L1xHT6Rj0mYTFLK6DE73SlS2JwShBQ2pxQ2d9S1MqdkGf6ceH4gYXNKc0q2tr/VKYUjWYKQJnd61HFM7nQpnGky+YzLpbC7pXCmGP4nS9pAbT+LXZpcqRKENKfmGL6snCsfkMLulpnj58mcKx+QVlfyEX+T1r+5Mz1HOpKjxyTsbmlOzlA+JUWb0Mk+pfYqOuRKKb+K+HwILRpP0cdpj05ZoqfqI2W499Ys+ERP+JjtZA0YjLclSMrFs8jVte18GxZjsrlxn/YD6j/YaMxWhNUBwgRIpD++wKlsaSZz0i+xDxlO+V9+Q9MXHyBsTrLGR8+essbPo2z9Qlwnn4MMhWj4z+sEAy0gJVKGSBp+Po17/x41e4oqNdHSiNmdTuPrD/Onx1e2eX0jpZLC/XPxLCpffSiqLbtgPhXr76fspfuR/mayBg7miYcebFsVQilAKI6R9io6PAwMA57TmyYDe6WUt3Xh2LqV4zXQ4UhEGpbICql2E5gvvLVNJ3ekI9xb/Da1O9fhryzFZHMSamli6Iknxw0w6IlzCtcfqt3xXFznfuWrD4EM4Rp+ge7/WUCgvsJYsqvauhLPmPiqDPaBJ9Nc8hEmpxuTzUmg5uuEgQzWzFz8FZr0UOQ2B5+aGVf1IVxqouyl+0hOTsZmtVJR9hVWZxKBJm/cII72BHCE2zoSJKE4vugRRQcp5a3ASuA7+uvx/mSQFIlJVEW0sqo6bvmFkpJ9Rs7Kl59rJRq8xW9Ts/1ZPUenkKyCBZhTsqnKOp2p029DmEzdmuMyZcrVPLp8CaG3H2Hf0gIqX32I9POnknTqRQmd+4HaQ2SMm0XT7p3IliaC3mqqX3+cYH0FVdtW4sgfEVOqorxoMdLXSHNpMcJiJtRYq1WBtbviBzLYnLqAazaWtAFR2/gr9ycsNVFetJhvnXIS0myl5RvnYk7JxjPh7oRBHIkCOMLCsZFtyj+k6G7aKzME8AHwipRyNvCaECK5i8ak6EUkirCS/uaoG9vh8gu/NBJNLe50anY8R+3OdWSMmxWjgt20ZycZl8/F6smNe/PsysTMKVOuxu12g4BB0x41QqitGUMSRtzZhwzHX1WK2Z1BzfZnyZr4Cy25dfR0mj7/J84T/x9VW1dGJLTWImxOhM1O9hX3kDe3kOwr7kGGQjGq3uVFS8Bso/Zd7Vq1rsdkSc2JOy6T3U3IW8vuLw8iTjqXpj07yWx1rVurnydSCLcEm1VyqKLHaZdREkLcBPwZeExvGgy83EVjUvQiEj1VZw0cHHVji1d+IePyuXj/tdEIf44kHEodvtFXvPIgDTVVXHPNNWQPymXmLbfGqCnEC9uOJJ4Rm3nLrdjcKQhhitHgK/liD1ZPbtT5pY6cHCPhEy56p4VlOwn5GmKMbMa4WTTv24Xz5JGY3KlkT1qgGaFJCzBZ7doMSd8254pfIv0tlL/8QJQig8lqM66Ve/j5pJ13HWUv3c++ZZOQQX/ccSV99zLMqdlkXD6Xpt07E86oIsO0wzNF53vPGBJLTzz0IE8++nBUWzzZJYWiq2mvT+lD4H+Av0spv6u3/UdKeVrXDq/7UD6l+CTyKT2qhxmHAyBkSMb1k5Qum4TF6Y6fTKv7Qw79+VeYXSnRIdBFS3CPGGuUZQjvkygxM944q15eRMhsiyojXl60hJC3hqEnnYzX68U3dBSNet2i8DZlhYsgFET6fVjSBpA66iosyZmUFy0hacRY6v72Inlz4vlkChA2J9kFCxKWJje2XVoAYJSWEDYn0t+Eye4m6buXGecd6UuK8svp23k/foP086fiOmUU+5YVaFqDcXxPKqFV0VX0lEq4T0rZEjEIC6CkII4D4j1Vh5+gIwu5DT3p5PiClScO40+Pr4xV1tYrq5YXLcHsSIotcjd+Hk27d0Ydr63EzEjfV+NnO6jatpJAKBRX887kcNN01vXUN3jxf/oWruEXaMtuywooK1xI8hmXkzvrecypOQSbaqnctJzylx8gSTeSiZb4hNVh5Bu1Hnc4wdbY1uYETGC2Y3anaTOrOYVkTfwFDXo9JhkMGNeouWQXrlNG4Rmt6eCFmr007dlJ+vlTjYq3ZrsL57CRMWrkfWEZ7njR0FMcmfaGhL8thPgF4BRCjAFmAkVH2EfRTwgboLaIl2gaTtwM73vtT6YhA82aH0S/qYa8NSBE/Bt5VWlUW2vHe2SoOmY77qS/UrF5BcHaMixpA5AtzXGPG2r2ajOJy+YRfPNhkso/pKR6P1ZnEpbB36bhX69Qt/MFhM1J8pnjST/vWkoWTyBt1FUAOPJGUF60hCx9Blaz4znqPyjSZjx2FzU7noua4YWDCMIJreUbFiNbmhAmgUUE8YyPTuTNGj+P8pcfoO5vL2LNGIL7lFFGQUCzw8WF557Djr+9RyDgp6JoKZVbHkH6GknxZGL74h3q6sqo2vBrI/qut4dp99bUAUXP0F6jdBcwDfgP8FNgE/BkVw1K0fdIlLMCWmh4WCXAMvRsAtUHDxeuc6Visjmp2fEcTXs0n4g1YwjOYSMx2100l+yKMXIQeyMLL80ljRhL2qirtGW49QvjKhKYHG5AM1Cl5V9T/vUBAIQwETz4GVkTf3F4GXHzCmyZecbsyJE/guZ9u0gaMZaKoqUEvdWY3GlkFyyIWiIEjHFockBe9i0twORMJtRUb2jjmczm+IbT5yX/zmidu8pXH0KGArz11nawuXCfeh7e4rfIjFh6bHz9YVavWt2nbuZx86biKGwojg+OuHwnhDADn0gpn5BS/lhK+SP9fZvLd0KIXCHEm0KIYiHEx0KIWXq7RwixVQixR/83XW8XQojfCyH2CiF2RZbGEEJcr2+/RwhxfUT7mUKI/+j7/F4IIdrqQ9G1RC7nhf0XkcEKzu9ciu/gZ1HlG4QQBBuqaNi1Jbqsw64tXHDuOQkd7/FC1bPGz6PhX68Y9YosnkFxo9xsA7TZlm9/MRaH21gyikxqjYwSrNr6KI48LeS7evsqArVl1O18gVBLI8Ieu48xjqUFVG1dSaixluTvXEz+XUUknT4OYXMghAmT3QVmW4LwcFdMUEPaudeQeelsQsKsLW+2I9KuLyyLKQ09RSRHnClJKYNCiM+EEHlSL4feTgLAHCnlB3r4+PtCiK3AVOB1KeVvhBA/B36ONhMbh5agOww4G3gUOFsI4QH+DzgLzY/1vhBig5SyWt/mJuDvaLO3S4DN+jHj9aHoRlo/ATft2RmjZJA5fh4VLz9g+H4i2z9vwzlf8sUecifGn2HkzSk0Zjkhbw1lhYuQvkZMDjf23FNpKfsv1dtX4d21Bfd3LiVt1FU07S9G7p2fcNbS+Pk/AUHDv18j+4p7jJnJoecT72PN1IIOyjf8lvSLpulCsK+RrSs8+PYXU/aX31LxyjIyL5sTNdOSgQBlL92HDLRgzRhCmq78LYMBZEtTVPRi675LXtBu5n1lWUxp6CkiaW+gQzrwsRDidSHEhvCrrR2klF9JKT/Q39cDn6CFkv8QeEbf7Blgov7+h8CzupzS34A0IcRA4GJgq5SySjdEW4FL9O9SpJR/02dtz7Y6Vrw+FN1I6yfgRDfRYLP3qJ+U26p0GznLEXYXJruLzPFzyZ31PNmTFpA5bhYNHxQZ0X3h7YUtvjq3NUOrGhtqqiXUVEvVtpU0frYDR/6ImCTX8D7C5tSCDjYuAwT7lhZQ/8FGssbfGTWzyf7hXYAwCvaFAyosyR5MrlRyJi9k0I2PGHlU4eXHcO5UW9VQEyU+R86kegOqgJ4ikvYapV8ClwP3AcsiXu1CCDEU+C7ajCYnQkfvayCcRj4YiPRs79fb2mrfH6edNvpoPa6bhRDvCSHeKy8vb+/pKNpJa8OR6CZqsjvjKww43AmXm+LdyCr0fKIw9iHDkb5GnN/4f8ZN3Vv8NhUbH0SGQtTtfIH9K2/EW/w2gJbUumFx1DHL1i/CX/mVNkOa9EsjWbZm+7N4i98mddRVMfuUb1iM9DVR/34RwYYqZFALXJW+xvhGuaGKQdMeBSEI+RpJG3UVgdqvkS3NccqeL0ZKwaHn5xNsrKPildjvv9y7G5s7hS/37u4Ty2JtRXgqjj/aXL4TQjiA6cBJaEEOT0kpA0fTgRAiCXgJuF1KWae7fQCQUkohRJeGlrfVh5TyceBx0PKUunIcxyOtI/Kcw0ZSUbQkKm+oYvMK7Hkj4ra7v3Np1HJTa2HYa35cwOYtz/Dl87sxObS8nbDxAd2wpQ2g6fO/s/+RfxCsrzRmTtlX3EOgvoLaHc8ZEWwEWpBmG2XrFyJbmsBiQ5gsCJuNrAl3xpSfCOdZAYdFWm0OZCAAFpsWYWd1sOqpJ5gy5WpNCzDOMlXYWFtScxAWq+FTyvrhXQS91VRtW4m/cj+W1BxCviZyfvR/h5cPX/wVZYXaeK2eXJJGXEzDri04R4yF3TvbLD3Rm2hPhKfi+OBIPqVnAD/wDprPZzgwq70HF0JY0QzSGilluPbSISHEQCnlV/oSXJnefgDIjdh9iN52ALigVftbevuQONu31YeiG4kXkTfl+qu1OktNDVgzco0qqdXbV1FeuJBQS1NUe3P+d5j5s9uBWFXy1S9qSbzX/mQaSd+9jMbit3Dmf+ewb2bDYqS/GcfQ0/Ed+BSTQwtFF84UGne/S9Pn/4xJmhUWM0kjLqVh1xZkSxNmVwqB2q8T6s5VbFqOlCE8Y2YYCbZIH56x06nctJz8/DzjOiy89x5uuu0OuHhWtPEdfgEVrywjFAiQ/K1LqHtthTGrEmZLVBXZfcsKooyjJdkTIyDrzP8OVdtWknrOZCo3r4g6x0T1lRSK3kKbig6Rqg16wuw/pJRnJNwhel+BZtSqpJS3R7QvASojghA8Uso7hRCXAbcCl6IFOvxeSvk/eqDD+0C43w+AM6WUVUKIfwA/43Cgw0NSyk2J+mhrvP1F0SFemYne9gQqTCasnlz8VfsxJ3nAZCZYV46w2PCMnUnSqRcZ24bVD1I8mTjHzo6rVNBQX091kx/3ty/C+/GbBOvKo5QYygoXIcxWsibcacyOgt5qsq+4J+p4Bx67CWtWPs37/oNsadTCaoTA6hkSXwF8/f3IYACzK5X0C36iGdGSXVQULSVz/FzKChcy/cafRBU3XLNmLbPn3UX5VwcOqzlY7VqdskALWQMHs3zJb5k97y5M58+MVYbYutIop+6v3A8yvpJGWHHckTeC5pJd+CtLGXpSz6iyK/o3na3ocKSZkj/8RkoZiFx6awejgGuB/+gyRQC/AH4DvCCEuBEoAf5X/24TmkHaCzQCP9H7rRJC3A/8U9/uPilllf5+JvA04ESLutustyfqo1/TF6Kt1qxZi8WdjmfM9MOlHyKe5Cs2LUeYzFGOfUvaAOqqviYtQaTZqmdXMW3GrTR+sp1QUx0pI/+Xpj07qdz0O6wZQwyDFPRWU/vXNVrl13ULDAXz8A1eWGwEGqog2ILVk0vI50VYbDhPHhkz4ygvWkzymROMXKTI2kZBbzVlhYtw5I1g85ZtUWMOL1NFlvUIEzay4d+qdTJyedFi7IO+Sc32Z42xHHxyRtwlurDcUOXmFbiGX8DgcmenyQz1hQcfRd/lSDOlIOANf0S7+Tfq76WUMqXLR9hN9IeZUls3us7WPWvvjan1dl6v15gBhHXdgt5qwzBYUnMItTQxZObT2o34L79FWO2EmupiZjaR5xbu58u9uzGnZkcZuvINi0m/8Abq/v5nnMNG0rRnJ4HacpLPmhCjexeuq2RJzqT8L79FyhAmmwP3ty/SBE+rShE2J5b0QQy6/ndRYwn7mCpffQgpQwy+6bGE9YgS1TSK3L71tTvxhDze2L4jSlvPW/w21W8/HVP8LxxC3lyyi4qXF/HMk493iuFoSwtRGabjk86eKbVLkPV4oD8Ypfbc6DqD9t6Y4m136Pn5WDO0pTtMFs3P462JFj7dsJhQYy3m5AwQgsxLZ0cV1Asfq+61FVGVUAGExYEl2UOg9hDWjCGkjpyM2Z1O5asPEaj52jBY3k//irf4LbInJRZPbS7ZRcXGBwk21SJMZqTfpwmntjTFXzJbVoAlJZvU70+hctPvyJm8MOEDQUcfIITJRN6c6L4bPnqDqi2PIv3NWDNyST1ncpQfqnTZJEKhzvn9u/PBR9E36O7lO0UforuSENsrC9N6u7AkT9TSXZwn/KwJd2pK3RIyLzvsRxIms2FcLGkDsARb2PHuu8ZsIiNrACa7k4xLbos6Zur3pxCo+Rphd5I5bha1f3sR38HPjCTUSCLFU+1DhhP0ViJsLgCSz7gMz5jpHHxqZnz5IrubtPOuw+xOR1hsVLy8iJtvmBr3Gi689x6mzbiVSrODQO0hLKk5WILNLHv04bjbhxl64skxv7ElOZP8/DyAGIMRFsXtLBIlLYcTdhWKY+VoivwpejmdnYSYqBZRe2VhWm9Xu3OdoehQ9/c/x9ZfGjeL2p3rtPyilkZCvoao/d3Dz9fzeWDwT59ADDiFlU/9iS8/340lfQjVXp8Ruh11zB3PIZwpSF8T3k//SkvZf8metABrRnQtJTgcoh1+r82MmvGM/imeMVr4d+rIyTFK3BWblpP+g5sxu9Op2LQcz9iZZE6czxPPPhdVwykSYbGRcclt5M1ZT8YltyEstiP+Jm39xt2RhJooaTn84NMXZI0UvRs1U+pHJBJF7cha/8xbbuXxZ9aSOeFuY9bx+DNaKHF7Z2Stt4tUdEhc3nu/Zhg8ufgrS+POSITNRdXWlfgOfhYlhJpI8idQ8zWYrQi7C+8nb5NdMB9H/oi4IdNhn5KRqBoIaOXJkzONY7qHn09LxT7KX36AkK8RYbFhsidR+cpyLGkDjHISABmXz+WxpxYy6pxzYmaRKRfPip5tXjzriCKk7fmNO+P3T0RbavB9IdBG0ftRM6V+RmtR1I7eDJ58+tmYWkSZ4+fx5NPPtuuJfM2atXi9Xg49P58Dj91Ew0dvRJX0TqTuYHanU7l5Bc6TR5LiyaS8aHGsooGvkYaP3ogRQk0s+ePAZHeRfMblUaoK4equRgnzl+4DKanc9DtNSNVbS+als+KqNjQWv4VnzAxyJi/EkppN5uV3gNmCMFup2LiMg0/NxFv8tq6D18Ttc+ZFjaut2WZbs43WwQ/jxo5mwb33GdsCnfL7J6It9YXwcm3QW81XT/+MQ+sW0OSXMeeuULSFCnTQ6Q+BDkdLWxF0QpgS5r9IGToc7fb5bqzOJKN2T9gwtQ5wKN+wGLdFEDDbSLl4luZT2v5MlBBpxablhHyNSJ+mvi0DfggGjOACYXNqKtw//Dn7lhXEdfhXv/lHsibcGV3nyNeEsDmwD/omvq92GzOlMEZZCBliyPSnotqQIVzDL6D+w1cx2RxGraZwUEY4+bX+w1cRJpNRYynsz3INv4DGT7YT9FYjAz7MdhfB5kaEzUH2pF/GjCP09iM0B0IJK/22vq7hCr3h0PSejIQzmc14xt1uhN1H/vbPPv4HNVvqp6jouy7ieDBKkUYoI2sA3pYgKRfPirn5TZlyNTZ3SvwS5ht+TYu3zjhe65tk5calOC3gumRu3AithffeYySPYrZiSc6IiZSr2roSz5jpUeHZ4Zt/Y/FbOE/8fzR8/KZmvKwOLfFUhhAmkx4hpyWlmtxpUUaivGgxMhhCmE1kjb8z2hg2N5J85uWHayAVLcF9yihcJ59DWeEiMsbONJQn6j/YiGxpRNhcSF8j5tRsAKOMROQ5lxUuwjP6p1S+shxzciaZl802DGbDri1R46t7bQVumxnzhbfGNVZutztu5FtkqfWejIQ7Ydg32V9RF6MwoaLz+jc9VQ5d0YN0hvM4bEDCtY1qfNLwacRTkJ429ToqWtUiqihawrSp1xnH+8nN02NUqDMun0uDN77waDgQojkQIufKRRAKMGjao+TfucFQwtYqzu7XlgsvnU3d3/9sKH437dmJa/gFeD/bQXbBfPLmFpJ9xS8x2R3azOOKe8i4bDYIE8Jqj1Pn6E4IBUgacTEVm1foS3b3IwN+ZLBFr8VUQNW2lSSNGEvzvl1G0IXrlFE0l+yiYdcWhNmqqT2EgmRcNpsh058iWFduJOMefGomJYsnULV1JdLXSO2O58BiMyIJhdlC+nnXkjRiLOUvP8C+ZZOofPUh3DYzleXxJY3KvzpAyefxl/wiS633pODqwnvvIVATf/y9TQRW0XtRgQ69nM5yHrcOzw7UHmqzFk9YGufJp3+Nv7EBqyuJm6dex6hzziF7UK4209H3aX0M6fclFAKN9DuY7C72LSvQcmtGark1kdFvrcOzNVmd2JpMWePvpPLVh2jc/a5hsMKKDTFja2miafdOQ1sPwjlGk8id9byxrQwGqPvbi0ao975lk7B6hhDy1pA3t5B9SwsAYQRAWDOGULPjuZhk3PKiJbi+dR4N/36NQH1F1HjSRl1F3d9eJPPyOdS+u46KQ6VYXUnxy6mnDUC0NCRQbxgS9TkzZ1CCv4KuZcqUq5k9764+IwKr6J2omVIvp7Nq4rR2rB+pFg9ohqnFW4eUIVq8dYw65xxmzJ6H6fyZ5M0tTBhYYE7Jig0M0AMhSr7Yg/fTv1L1xpNkTfyFUQqictvj7P/D9Rx6fj4hvw9v8dsx4dkmuwt/ZWn8CLvar2n4+E3DYCU6P2tGLp4xh0tPhNstqTkx21pSc6jcvALPmBnk37kBz5jpWDO1MPKhJ53Mqj89aZxnytk/ov6DjWS0qgSbpVeIzZpwpzZjan2tkjzUbH8Wz5jp5M0txDPhbhp2baF6+6qoqrOpo67C39QQW66jaAnOYSOjQtPrG7w9Foq9fMlvVW0kxTGhfEo6vdWn1FkqDa0z8ePJ0xzJSd6eY5QXLSbkrUXYnCQ57DQ01BpRYpu3bOPL//4XhMCSdFh1wZE3gsa9f486TsWm5YT8PjwXTYsOKPigyJDZMXTrKkqNYAiTw417+AXYB38rSiMunvyO4bt6ZTmhgI/sH/48yjkf8jWR8j8TDT9TeAxy7zvGdcoaMJganyRQeyixOOqySeTNWc++pQXkXLko6hxBkHnp7XF9UbKlKdrXtkFXV48ILKksK6PJ5Ijxy/WkD0dp4x1fqECHLqK3GqXOknWJF5RQ+8oSkpPcVBw6GHPziHdjufa6a2MMpCFxE2hG2LSwa+MmXrSEm6+/mlHnnMNNt91BysWzKPvLbxDCZETIhUtGREbEeYvfpuad1ZoKQ4SCtqGqbXfiyBtBS9l/o8VcdaPRsGsL7lNGaYZJP048+Z19SwuMchZa7SSTXhPJid0s+MlPpvLk08/ib2w4HAFodWC3mPBLQbDZa7Rb0gYgQ8G4BiasiVf2kha6HWxuRFjtmIQg2NKcMMoxyoAliLKL95t0hbSUQpEIZZS6iN5qlDpTAPNoRFRb91n32gqa62swJ3kI1H6Nya7dzC1pA5BBTUxemK0Eag9hTsmCUJBgQyXC6sRhtZAy/uc48kew78Efk31FdCh0yeLxRni3t/jtmBlOuNZQ9g/vijBkC6OESSHaAJS//ABZE39BedFikJKsCXcljFoLz2aQkozLZmNJzozS1Yu8Ht5P/0rjnp2tovqWYB90Cs0l/8FktcUWKxx+Ad5dW7j5+qujylhA4oeOcLRdyRd7sDjcOL9zaZSfKfxgArHSQiraTdGdqOi744zOLBXdnsTaNWvWcv20n1Jf8RVV21bS+NkOzUfzzQsQdpcui1NI1sRfYE7NxvWt8wgFWpDBgCGZkzluFghBxqWzyZ60AJ80GU5+6W+O8gl5i9/WAwkKOPjUTKp1gxSVtHvZHMx2V1RbW7p19iHDCTV7KStchH3QNwm1NMeWLC9aYpROD/uPrJm5JJ16EY78EaTo6grhKMP6iq+oeOVBvMVvEWqsibo2WePn4Sv9iOxJ80lPcuJ87xn2LSugvHAhwdoyGj4oIuitYfOWbcy85daoSMpxY0fH9cEsX/Jb47cKNHtJG3VVzLmWfLGnW6SFFIruREXf9QG6q1T0mjVruem2O8ic+IsoPwxA/YebMdvdHFq3wPBdZOolwc12d1RuSjicOzwTyZpwJxUbl5F06kUIu8uIzgrPirIi+mtLKqhk8QSjb6snN2EkWlizLmPsTFynjGLfskl4fvAzo6x4OIouHOYdTtpN+vaFUX2WrNvDjNnz8Ey42xCQzZ60IObauE4ZRajZi33IcErLvqb86wPG9YyccVbteI6VT/3JKF1elXU6q18sNMq6J5IGakvWqTOlpRSK3oCaKR0HtDfPKVKPLVLQtHLb4whhMmZCntFa9FqgvgJ/5f42yoVHqG03VGkRZb5GY9ZS++66mFlRoog+S9qAqL4tnkGUt86j2rwC57CRlBctIenUi4wQc0tqjibmeuMj5N+5AefJIxE2p5antP5+Qi3NOPJOw1v8FiWLJ3DwqZnU7HgOi9NtRD4mEpCt3v6sFhnocGt9Od1R1zO8f+NnO2gsfovsggVaxOGY6TQWv4U46Vw2b9kWNYMF2jWbCs+GOktaSqHoDaiZUj/naPKcEpUlIOAjK6LAnmGsXn0Ic0oWoea282dqdjyHsLuo2/mCFqwQ8BslKFobs7DOXGQgRMWm5aSfPzXKGJQVLsL9rfMpe+m+qDpH9e8X4cgfQfpF0/R6SEsJ+X00l+yKUlKInPFUbFpO8/5PyI6YsVUULSHYeFilPJGAbLC2TK8yexrlRUsINTXEvZ61Ow8b4MhrWLV1JSXVh5Nf4/1eq198+IizKYWiv6BmSr2M1rOa1j6I1uKcbW0785ZbDX9IpA8kUZ5TorIE0u9LuKQm/T6Sz7icylalHMqLlpBy9o+o3r5KMwJhBYZJv0RYbQSb6uLOivxVBwj5Gil76T5DJDXt3GuNqLlw37KlEfc3vw8mE5itRrAFQuAvLzFUEkJ+H9LXROWrD7Fv2SQa/vVKjNJD5qWzY3xWmePnYXa4jigga0kbAKEgzfv+g33QKQw98eS41zOhKnpVaVRuWKK8tNazqd5qkFTpCsWxooxSL6K1FFDTWdfz+DNrqco63fg8Y/Y81qxZe8Rtq7JO5/Fn1uKZcLeRoBpOGA3UV1BSsg+T2UzWgMFkD8rFZDbTUF9P1SvLYuoECZsjvvq21U6osZa0UVdpatvbVrJv2SRdYbuGyk2/i2sEssbPg1CIQH1VVADCwWdup2HXa+T8+F5yZz2vSRGZzFFlI8J9W1IH6IrhTVjTBxnbmJ0ppJ17DZmXz0FYrLpGnYOgz0vO5IWEfPElkAK1h2Lags2NxrJZytk/igmWCCe1hg2zpWZfVIBBZBCC1ZNAFd3uitqnvbWqeuPNP97fZPjvVaFoLyokXKc3hIQnCg+OJ7gJ8UOBw9sefGomntHTY76v2LwCZChhifGyl36F2R1dTrzq9ScQFmtscquvGSEkWRN/ETf5Myz3kzcnNo9m39ICsDoxWW1If7O2BGd3xoR5h2daUSHYemIrwRYwWzDZ3THLfdLfQlZECHl50RJCzV6EyUR2xFJkeLyVrz7E4J8+EX0OL93Pqj89aYTRmyx2hCstJlG17KX7yc/PixtiP/OWW41cJ5M7NUoItnLjUm667qqoMPH25KV1ZppAZ6JKpR+fqHLo/ZhEPp0YwU1dn66tbdvygVjSBhD0Vkc570Hzc8hAC4OmPRplRCo2LiPj0tsPR69lDCHt3Gup3PQ7pJSUFy2OvtluXoEj7zTKChchLLYEhfocyJYmQoEWhNWqfREKxdeH2/nCYd+RnkwbVlo4+OSMuJF/la8+1Eofbx5VW1fiPHkkFUVLonKJtNLrkpLfjteSaX1ehM1FktMRFfloGIPJC6OMwao/PZkwvH71i5p0UNifVfHyIkK+JvJPHMbTKx+K2a+tInph2luOvrtRpdIVnYEySr2IRKG/rQU3wz6ItrYN+0Biv9e03yo3ryBQVxZjuOKFWltSc7AkZxqzNdCegC2pOQiLFUfeCF0WpxGrJ1dT8i5+C2G2knTG5TGBC2FjJVuaERYLwmLHM2aGJie0aTnCZDZ8SGG/TbCxFjAhLDayf/jzIwrLxluO81eW4t9ZSnJaOqG3H2HfVwfAYsNkc5L0nYvxFr8VpRBR99oK1qxZa9zojzb8urXxSD/vWpz532lz5tCePnrrzb+9FYkVirZQPqVeRLxEyNaCm+FQ4CNt6xw2Mqb0RPlffkso4OPQugVaeQeLPcbP4Tx5ZEy115CvUYtQaxXIEGiowjlsJOkXTcORdxqYbATqyqjb+QKhxlpCjbU07d5JqLGWsvUL2be0gEMv/gphtpI14S699MQ9CIuV6jeeIuitJvPS2dS8szqq0mywqQ7Z4iP/rr8Qam6IMkIJAxDiCKwKmwNzUgbiW2Oorm8EQJjMZE24k6Y9O416SGHfVziBNpIpU65m4b33kJE1gJKSfVxzzTWYbE6yBgyO8Z0k8g99uXd3m76gI4V4JwpI6embv0rkVXQGaqbUi4j3lHzz9Vezecs2Sh58Me5Tc1vbTrn+ajZveYYvn9+NOSkDicQkIp5DhJmyv/wmSoi0YddrCIuDsvULkf4mbWY1+mYAI4zbmpFL0oixeP+zDe9Hr2uh3nYXZndKK3HWJThPHsnAUb/XQ7Ffw2R1Jlxuq925joFTf0+g5mv2LSvQwrx9zSDMCJsjKmAgvH/qyMlUbFoe7e/SBVbDYeBhuZ/kMyfg/fgNGj58laTTL6Fp9078VaVRShCRxJt9rFmzlmkzbsUvLGRf8csoKaRpM26N+h0TzRwsaQMYNO3RDpchac8SX0+gEnkVnYEKdNDpDYEOXcUJw77JvoOH4gYrBL11EGzRBELtSdgGnoTv4Gea0Wm1nBXOFwovrYUDK0K+RkItTXHLjJe9dD+W1CxCfh+Z42YlDnxYNgmAnMkLKXvpPhAmTQBVF1/1HfgUsysFf0VpTEXZsr/8BgItyIDP0OQzJ2cQamnSlhQj6jWFgxqE2Yq/SvOPeUZPNzTzjuSkb6u6asXGZZhCLQSavYYy+uoXC6PLl8e5hh0JBFBK3Ireggp0UBw1C++9h2t/ciNZE+6MmaGUvXQfubMLjQCFloO7jRBuW2aeHtxQirA68IyZEZMv5K8oxeRKRbbED7WWAR+e0dMN+aBEvq7wclvl5hVkXHKbFtW2fiHCbMN34FPSL7wBYTJrpSoqSzUflk9T25aBgBEsYU7yGOOMFHqNHJPmb5JYM3Jx5I3QFCBGjKVi84ooIxyOjouk5Is9SBm/uGGwoYpMXdk7XtIrZjuesbHXsCO+oO6SnlIouhvlUzoOmDLl6oQJsNLvi1JKCHqrje3C0jx5cwqRLc3U7njOkOEJF+ETNidZE+7EmpGboKjekCj5oPByW+tcqGBjHTLoN+odhavECrMFYbbgrzqgFcMbPZ28OYVkF8zH7E7DYjZjTkole9IvjXys6u3PUr19FcLmSuhvEjYXjrwRNH3+T5JGjKVp905NnUH3fVVtXYnrtDGsfrEwyu+T/41hWFJzEibStpX0mp+fFzfnqqd9QQpFb0IZpX7CkZIpswYOjp+8mZoNaGrdVVtXIqyxwQ81O57D5E6N0r6rfvtpytYvMtS6U0dOjqvq4MjTfT+6fJDZnU7auddSsXGZodgQ9NaSfOblDJn5dFTUnbA5yLzsDtLOuy5uVdfM8fMIIci8bE50+7hZ1H+wERkKxQRtVGxaTrCpDkfeaTSX7CJj3CzSz7uWQdMeIf+uIrInLcCamcugaY+Qft61MeoXC++9B0uwOTbwY8NiUhMoeUfuqwIBFIq2Uct3fYTWPoRwJdeSL/aQmTOI+gYvqZfNS6hvt3zJb7nptjvg4ug6Ra5hI9m/8kYjf8n97YtiAgciq72CtvTn/vZF1L+/AYCDT84g7dxrCDRUawESLU0ImxOTM4Wmz/+Jd/C3sCRnIqWkYvMKgrVlCKtdkwcKhSAUomHXFpz534kKkpCBAHVFv6HJWw8kWDLT1blbt0tfIxmXzabyleVUbFxGsKFK6xPJRRecx/Ydf8Pf2BB330R5YZHX8/Y58yh76X6kv9ko2HekWZAKBFAojowySn2AyAx+zykV7Hvzjzz29BqyJtxJ7kTtJu7ftJygt9qYLbROppwy5Wp2vPsujz21kJCvCXOSh5CvKaYMeeXmFbhOOtuItBP2JKQvunaRt/htvMVvkT3pcPRZ2fpFmByuqCTa8qIlmJIztBBvGSJj9M0xlV8taQMI+ryEvHWH/UR2F4QkKWdPMgyfsDnj5125kuK3Z+ZiSc7E7HDhSXZS7gWL1Uqgycvn/92nlRXPSFz+IvJz6+W1eP6c8G90pIg45QtSKNpGLd/1AcJJmEFvNbV/XaPlxehBC5GiorU71xn7xNNL27xlGyZnKggYPP0pTK7kuOUYmvftYtC0RwEQFnOMcGrtznUxOT1mVwpZ46PHlDV+HoGqAwRqvsY9/IIoB7+mnefQPgSDQAhkCARYkjPIuOQW0kZdhWxpxpqRi2xpilmKq3ttBdOmXqdVxY1XwmLDYtJTU/lRwQ9JzhyIZ8Ld5M7RNNks7nSt/MVffsuBx26iZPEEDjx2E2WFi+LmhR2JzizGqFAcz3TZTEkI8UfgcqBMSnmq3uYB1gFDgS+B/5VSVgshBLACuBRoBKZKKT/Q97keWKAfdqGU8hm9/UzgacAJbAJmSSlloj666jy7g3AG/1dP/4wMPaz6SMtO8Z7ww5FjYdWGYF15wuMYQQzj7yTordai4sbN0mooVZTG7JdIWUG2NGNypVL/wcbo5bkNi5GhgD4bc5Ey8n9p/GR7lP5c9fZVmFypeMZMN2R6ygq15UFL6gBCTV4AnnjoQWbPu4t9Xx3Q5Yua8f5nG+kX3oA5OZPH/7iIzInzo5YfXaeNoe4fL2Oya3lT4XFVvbIM2xfvUPr3+HlhbaFmQQrFsdOVM6WngUtatf0ceF1KOQx4Xf8MMA4Ypr9uBh4Fw4j9H3A28D/A/wkh0vV9HgVuitjvkiP00WcJZ/CHEzzbUjFo6wk/HDkWLpBndqfHPY7JmWLc/O1DhuMefj7OE/8fZS/dR/Wbf4xbciJRRJqwOciacCeEQpStv599yyZRsXkFUobI+fGvNFWHgvk07NpCwFsTNeOp/2Bj1Iww/bxryS5YgDklm8E/fYKsgl/w+B+fBqDsYClShsjPyyPnykUMueUZo7R5sJUyeHj50ZLsiZlxei6bQ1Jy8jGViOiNCt4KRV+hy4ySlHI7UNWq+YfAM/r7Z4CJEe3PSo2/AWlCiIHAxcBWKWWVPtvZClyif5cipfyb1LJ/n211rHh99EnWrFmL1+vl0PPzETYnNTueSxjpFqiv0qupLqS+/CsW3Hsfa9asZeYtt2Jzp/Dl3t0EG2tp2vs+IW8NoZbYKLKywkUAZBcsMHwu3uK3afr8n5jd6WRNuJO0c6+J6T/YVB+zvFZetARMFj30vBnMNvLv3IDJajf06yKX+oQQRt2jylcfQiYoMxGsLcNb/Lb23tfI7Hl3GUbgy727Y/YxJ2fFXX5MNLtrvex5tL9XbyvfoIykoi/R3YEOOVLKr/T3XwNhgbLBQGnEdvv1trba98dpb6uPGIQQN6PNzMjLyzvac+lyIgMc8iKCB5JGjCX1+1OMYASTIwkpQ6T8z8QoFYam/cX85KczCWEiU1eqPqyKbUH6G8GZRMUrDxJsqDJmWmZnCofWLcDsSuPQn38FMoQlOdOoFBtORg2rhguLDZM9iWBD5eFgBb0SrDk1W6s8a3OSocsVJZL0wd/MkMxcSuoOEaivQNjjBzdY0gZQtfVRKjYuxWR3U/7VAUxXLiJ34nAOPjkjZh8ZaImKKPRXlraZyHsseUPdoeB9NGoOR1N5WKHoDfRYoIM+w+lSjaMj9SGlfFxKeZaU8qysrKyuHEq7iXyq/cnN02OqkGaNn0fTvzdRtfl3pDvNRtXV7IL5cUVFpdVFZkSRvaC3GpPdRfKZE7B6cgnWlRFqaSL5u5eSOuoqTDankY+UOX4uZmcSwmwj45LbohJkw4m1OZMXYknJJuitRNicEAphcqWSPWkBeXMLyRw3i4ZdW5B+Py0V+zj45EyExUbNjuc4+NRMIxm3ZsdzYLEDsOrZVZitNpLPGB8/92nod5HBACAwOZIRdpcReej8xpmUFS6iZPF4Dj45k+rtq7RChOdeaxQhNNndRiJv6+NXFC1h3NjRHf792lukr6Mc7UwsUSXbeJWHFYreQHfPlA4JIQZKKb/Sl+DK9PYDQG7EdkP0tgPABa3a39Lbh8TZvq0+ej2tn2r3LSuIH4jQ2MCMmTNZ/WIhOT++1wh8iDcDCdRqs46DT83EX7kfk92FPfdUGovfIiNCUqe8aAmNu3dGSRFpN3obsqWKis0rcJ34/4yAh6gQ8uEXEPigApMjhVBzfYycUdb4eZS9dJ9RrM/76V9jC/cVLcH97Qtp+ub3mTF7HsHmRtJGXRUhdbQfq2cIIW8NTV/8k+wr7onSk6t+608ANH3+T7IL5kccdzFOd3JU6Q1v8dvGzClyxmlOzcY9Yiybt2zr8G/Y1eUbjnYm1lvLXCgUiejumdIG4Hr9/fXAXyLarxMa3wNq9SW414CxQoh0PcBhLPCa/l2dEOJ7euTeda2OFa+PXk/rp9pE0j2WtAE8/senjW3Dy1DxAiBMzhRq3lmFZ/R0Mi69HZMjmeYvP4xRR8gaP49gQ5Vh1LzFb1Oz/Vlt1qTPeBr3/h3nif+Pqq0r2be0gPKiJbiGX6DNhHxNDL5pZWINPL/P0NTz7f8obol03/6PjCf5cAn28Iws/84NeMZM11QeWoWxZ146m1Czl9qd6+Kc1534AoGosHGzOx2rDFD20v1UbvqdJlQ7fi5Dpj9F2qirjFlNIl9M2EcnhAmbO4WZt9xqnGtXqzYc7Uyst5a5UCgS0ZUh4c+hzXIyhRD70aLofgO8IIS4ESgB/lfffBNaOPhetJDwnwBIKauEEPcD/9S3u09KGQ6emMnhkPDN+os2+uj1tH6qTR05OUYktGLTcpCSYHMjVVtXknrOZGMZyjX8gpjtZaCFrAl3GjlObYWUhyWGHPkjom7wcFjAtWrbSjxjplOxcRkyFKTh368hZQhzarZRRDBupVmr3egzkU/JX1FKyeIJWD1DkC3NUYUDnSePxPvxG8iW5gRGrznquN7it3XxVs3nZQm24HzvmcNKCo8+zIJ774sp3x2+YSfyxfzpmad56933onx0jz+jJcg+8oeHu1y14WhnYr21zIVCkQhVukKnN5SuOGHYN2NukmXrF9Jc8m+tSqsePGBJG0DqqKuwJGdS/pffImwOgnVlCJtLDzJw6Ntr/+bNLeTAY9MQZiuB2kOY7C6SvnsZ6eddi7f4barf+hOh5gak34fJnUrW+DvbKDFRgCUlm7TzrtOVvO/HM2YGwmSmZvuzRtXZGEMakmRefgeO/BEcfGpm3DIRla8+xKBpj+q1l2KX92SLD5MzicxLb48tG/HyIkLBENlX/JKgt1qb5UUuT25YzLOP/yHKOIQNjzjpXKO2ktnu4uYbprJ5y7aY36K5ZBflhQvJipBcCrdXbfg1Ld66Tv+baE2ksQyfW+0rS0hOclNx6GDcwAdV5kLRlajSFf2Y1k+1NTuew3fwM604XasbfeXmFThP/H9IGdIL9wnMzhSCQPIZl5M26ioj0q5mx3PIYIDMy+6Iusn7K0po3v8JJqvN8NHsW1pA+csPICy2uDMek91tKHnLYADZ0owlOdPYrnbnOkNtW/q1JFekxH3aDygvWkLW+HmknP2j2MJ8ep0hYbbQtGensbwHh/1SFZtX4B5+gXGcyLLlzzz5OABTp98GFnvMLC9rwp0xfpew9NITzz5HxuVzjeOtfvFh6su/Ii+OLybUSnIp3O5vbIhq6ypD0HomlpE1QIt+PH8muUPiR9eppF5FX0LNlHR6w0xpzZq1zJ53F+VfHdDqBIUkhFow2d1kTfxFzNP5oT//CrMrJebmHvK3YHGn4q/cj9mdTqilkexJv4wtwLd+IWZXalTBuoNPzcQ5bCQNH76KsNpaVXRdRvoFN0QVqCsrXITJ7orRz4s0XPuWFWBOziTYUIMwm5GBFsxJHjCZCdaVIyw2PGNnknTqRQCULJ7Q5izNmn0CgQMfG8X0Im/4M2+5lUcfeYS8uYUx+5c+OIlQMBh1zePNTtuaEZWtX0j2pLZnSvFmM42vP9wlskOJxt+RwoEKRUfo7JmS0r7rJYRvZKbzZ2oqB1fcg9mdQsalswn54ithCyHiOv3RC+vlzVlP5uVzEvthWpoIeqsJ1FcY7akjJ+Mtfouk0y+BkKTspfuMEhMhvx+zOz1KY076Gkk799rD2xUuIlBXRu3OdVRtXcnBJ2foQfkCQgEwmUn53o8x2V0E68qxZgxBWB1RCtuJFCtMdjeu4RdgqdnHnx5fGVd14ZE/PJywTEc8v0uiwIGQr4mK1uHim1cgW5qoKFoSE0Y+bep1xv7dGYbd1SHoCkV3o5bveojw8s6Xn+/GZHMS8jUh7C7q1803yneHAwtaBw94i9+m+s0/tlm4z5E/wnD2RwYwhAkHJXjGTKdi03KEyYx7+Pm4h59PS8U+rR6RrxFhdeAcdjaBqoM4Tx55ODw7Ywju4RfQtGcnluRMTPYkcLjJvGxO1BJh0oixxlJi+YbFWLOGxvEXLaZs/SKyJ2mh3M5hI2OW6Mo3LCbk8+Ip/5CFR5hxLF/y23Y79xMFDlhdSTiHXxBzvp7yDxk3djRPPv1r/I0NWF1J3Dz1Oh75w8PG/t0Zht3VIegKRXejlu90unP5bs2atdx02x2k6LWN4jn2Kzev0HJoNv2OzMvnUL39WU0ap76Cmu3PkHnZHCo2/S7K6e8tfpuad1YTqPkazDZMdk1NPFBfQc07qxIusUUGGRw2AI1a4IS/CUuSR1OOcKdFjTHs4/F+/AYhXyPZBfNjl7W2rTTyg8LLffG2KytchNmZQqD2kFatNm8ETV+8bwi2Jtlt1NW2Vq1q+xq3x6eTaKntmh8XsPrFwg4twXXnklp3LhUqFPFQgQ59gJm33MqTTz9rPElPa/UkPXveXaRcfNgRH8+xnzFuFpWvPqQ/oWuzl/KXH0AGA2RfcQ+O/BGkn3+9ETAQNjzub1+E9z/bEBZblK9ImMxGkqg1I9cwSKA9xQdqvmbf0gItYs/fQvJ3x+EZM53q7ato2PUaOVcuIlBfEVFnSYv08378JtLfkjA/qXXBvER6drKlERxJpHzvxzTt3kn9v17RFCIsDuw2G48+8jBHQ3ud+22FcI8655wOhXZ3Zxi2Khyo6G8oo9TJzLzlVh5/Zm3CPBaA8q8OkBdxY06UtxOo+ZqMy2ZrCZjFb+EZM4OKjcuitxUmXazVQfakX1K1bSWZl8+JyUVyDz8f1ymjKF1xFZ4x02OWe4TNgdmVRrDZS+bltxoGK23UVdTtfIGy9fdrYeZ2FwCWpAyCZivCZCbzh3dRtW1luwrmJSzWl5GLI29E7Ixx41J+cu1VXXqTTWTAOhq11t2GQkXXKfoTKtChk3ny6WejtOYc+SNwjxjLY0/9CZPJjM2dAmZLlCM+kWNf2BxUvrKcsvULCdRqwQNmV5qh3F2z/Vncwy/Qi+A1U7V1Jf6KUgL1FZjsrvjHtNrj6Mktxj7omwz+6RNkT/x5VLHAsMGQLc2YUzLJLphP3txCPGOmg8AQaU2kXB5ZMK9y8woc+SOo3Lg0ersNiwk1N2oyR62uXcblc49J9qenmDLlav6759NjKoGhUByPKJ+STmf5lIQwRYUje4vfNvxBUYmggRZDoy1RsmhkkEDYf1P3j5cRVhvIEI680/Ad/MzQk/N+8hbS14Sw2gGBlCEIthiKCA27tiBbfCSfNd5IFhU2J5bkbAbdqM3iwqHXeXMKo7Tt6t/fED+sPMJHdFhFoRSrM0lbvszIxV+lBQqkjpyM65RR7FtWQNaAwZR/dSAqEfjQ8/PbHcp9tKgEUoWia1A+pV6O2eGOWp6KLB0OEQKl6xcaMjomu5tQs1dLOG1pMpQbmvbsxJaZh3v4+WSOm0XFKw9CsAXhcBHy1tC87z9kF8yncfe7NO7ZSXbBglb5Sj48F9+iKT8ULcacnEnIW03dP1+GQAvC5iD5jPGkn3etMX5tNuVk37ICrJ5cTaHh4zeQ/kRh5Y2GSKvrlFGY3emUFz5AWmoq5Y0NyKCfzMvnROU2DT3xZABMVy6KMnLh4oGdHUmmyjcoFH0HtXzXiaxZsxYTkvKIPJZw7Z5ItLDtJkxWJ8JiJ9TsxeQ+XO4he9ICzKnZOIeNpGb7s0ZBu1CzV4tEGzFWM1x60IC3+K2YZa/MS2djtrup+/ufDWHSQPVXuE8djRDhn11Q//7GmCU36fdjsrvxV5bS+Ml20s+fmlAc1mx34Rp+gSbSuqyAspd+hdXhMPKtMi65jeq3n6bhozeixEnj5dekjrqK8g3RhQI7Q8xUlW9QKPoOyih1IgvuvQ/PxPm4TxlF+csPaNFsVmf8m3lKNjIUMNpMVqdRE8iRP4LMcbPwfvQ6CBMVRUspfWgK0t+M9DXS8O/XyJ60AJMzGd/+YkLN8ZNrA7WHjOi38KymafdOZEDTw0s+czwIOPT8fEpXXEXFKw+SNGIs5uR0sib+Aiw2gk21VGxcSqi5kYpXlsUYjJtvmIqn/EMC1fsZeuLJZGZmkTbujhgDWbXlUZzvPWOEKsdTr7YkZ+JJduF87xlKH5wUtf2xoBJMFYq+g1q+60TCSZOO/BFaIADQ8NEblG9YTNaEOw8vrW1cRrC5AbMziawJ90T5kXwHPsEzZjreT/8aq1e3YTGhpnpkMEjZS/ch/T7K1i9MXKE1NYegt8aQDrJ6cvFXlWLNyKXxsx00Fr8VVXuocvMKrJ7BBOvKKfvzr7Q8p/GHx11WuIi6TUtobqhNGFFmMpvJTaDiXfLFHmN2Ei9sOiws2tl+H5VgqlD0HVSgg05nBDokSpoMvvkwScnJlHy+B5PdSbClBWGxxtVQKytchH3gyfi++ozsVtpr1dtXUf/+RkyOaK25ssJFCLM12vCFfUoXTdN9SkuwDzqF5n27yBh7C7U71yVU6hYWKzLgj8pzCn9/JDXsRNcgMjk3nNwJGMEHGVkD8LYEjYTizkwCVQmmCkXX0dmBDsoo6XSGUWrPzS9rwGAqq2uRLY1xI832LS1A2J1IXxPmlCxNsNTqwORwgzAhTOa4xuLQn3+FMJm0XCKrHWF14PnBTa3EUxcifU3k31WUWPR0aQGZ4+dSsXEpeXPij0/K0FFdg7ACeORYWqsbdLUKgoq+Uyi6BhV914tpT9JkZVU1ZncaQUJxl9wwW5C+JkyuVDIvvT3qxh6sKwch4vpHCLaAyQmA9PvInfV8lEEJC7AKu4vmkl1YPUPiF+PTk2OFzZWgdIUTk9mc8Mbe+hpgtuMZO8MwSOGxtNaB62q9OJVgqlD0DVSgQyfTVtLk6LFjkf5mArWHkC3NUVF61dtXcejPv8JkdyNsmmZd62ABYXdhSc2JGzhh9eRqEX2uVDBbEyTjukj69oW6yGhpVP/hyDv3t86n5p3VJJ9xefwk27wR5N6xnqazrmfG7HlGifBE1yA/Py9KATw8ltb+HFW2W6FQgFq+M+hKQdY1a9Zy08xbaaqr0bTlAgGExRJVTRaLDSEE0u8DiJ90urQAYXMhrLYY8VaXrtjtGT2dspfu0wxba5Vtv4+cK34Z5YsiFEL6tSq1YEL6m0BK8uYW0vjZDqOkuNUzBH9lKfl3FRnn1Z7ltfb6c5TfR6Hom3T28h1SSvWSkjPPPFN2hNWr18ihJ50ihTBJqytZIoQcetIpcvXqNcb3zvQcmXPlAzJv7ssy58oHpMmdJkFIa0aedA77nsRilyZ3uhTONCnsbmlyp0dtb0nNkSkjJ0tzUoYUVofE4pDC7pII7RgpIydLkzvdOCYImXHZHdKSNlCCkOakDGlOypQpIydr+yGksDqlOVlrS/FkSRAyb+7LMv+ujdKamSdzrnxA5t+10XjlXPmAtGbkRbXlzX1ZCpOp/dfIZIq6Nh3dTqFQ9B6A92Qn3ovVTEmnIzOluE59XQ6o5aMt2KwW6uvrST7rhzTt2Ym/ohSTw03I58Vkd2NOG4D/UAnCatVmS2YbwmxKKOej5TJVkjN5EYfW/R/CatFnOa6oEujlRYsJeWuxZoTlhV4j5K3VAihCMmq2VLFpOaGmOsxujxFAEdbVy4gsv75xKa7TxkSpP6gKpwqFQkXfdREdMUqJIsYqNq8AGSLz0tkcen4+5tRsre5Q8VutNPAWE2ryQihgLOMJqz2qNDgc1qMDEBY7mCyYbHbcp46m/oON8esTRUgWOfJHkPXDn+PbX8yhl+7XlglbmjA53LiHX0D9v14h49LZ1P51jWGIanY8R/0HRdDSTP6Jwxg3dnSH6wspFIr+i4q+60UkihgL1pbhHHa2FqZtc5I5bpZWUmLcLILear56+mf4K0oRdqdmkKx2hMWGZ8x0LMmZUZVgQVeAcHsI+ZsQNhcms9WY1dT97YWEMkaaX6qU5pJd7FtaAGYrJrsrJpEXk43KTcsxp2RT8cqDBBuqsKTmkJGWRvnXB4zjdrS+kEKhULQXFX13DMSLGKvZ8RzC6qBpz9+1Bhni0PPzjZISVdseJ9hYh8mVSnaBrnV3xT0Is5nqN54i6K0m89LZ1Lyz2oh6q9i0nGBzA9LXTKi+0igXASTUpBM2F54x08mbo2npWZLSSU5Oio3qu3wOlmQPeXMKyRw3C0xmUr73Y6S/mR9fURB1XFWOQaFQdDXKKB0DC++9h8bXH6a5ZBcNH73BvuWTNV26K35JxmWzMTuTSD7rh5hTs7GkDaBq62OE/M3IgC/WOFw2B2G2ULtz3eFKsMsKKH/5AdLOvRYCLSAk+XcVYc08bIji1jHasBgZDNC4+92oukQN3vhVXwO1h6I09xr+9QruEWP7ZB0jhULRt1HLd8dAeKYwY+atNDS3IEMBzLYUDq1boC2TTfwFVdtWan6bD14BwOxKIVhfEdc4BBuqCHqrNd26tAEgQ6Sddx1mdzqWtAEIsxU4bIjC5SJaKvZpZTB8jVjSBpB+4Q2GtBCAZ8x0fUnPd8TqsPYhwwn5GkkbdRUlD77YpddPoVAoWqNmSp2A16dp2ZlsTjIuuY28OesJNXup2Pgg/uqvaPj3a5idyZhdKWReOjvxkpvVjiU1h/INiwk0VJH6/SlafaKiJQSb6nCePBLQSpunnXcdla8+xL6lBTTt2UnyGZdjSRvA4J8+QdKpFxl1m7zFbxnHN6dkxcyqKjYtJ3Xk5KhxhCvhqsRVhULR3aiZ0jEye95dhFoaERYHyWdN0AIFGmsRNhehgA9hsRFqqkX6fchAc1Tp8MiQ64pNy7U4/aCf9AtvoPKV5VRu+h1WzxBC3hoQZrwfv4Ez/zvYhwzH7E7XZmZJGXhGT6d8w2LSL7whamzhGkzNJbuo3LgU92ljsGXm6YoO+zG70wn5m7VjBQNRIe2Nrz/MMl00VaFQKLoLZZSOkfKvDmi5R81e6t4rwmS1gbAggwFMFidZE+6OUlDw7S82ourCcj/CpunNJY8Yg2fMdE2bLjOXQTc+ovuIfkuosY60c6/V9onIdxIWO2WFv0YGW+LK+QibA+d7z3DTdVex+sVCzPm3MnDq741SGZb0gZS//IBWQFAPS/eUf8hCFVmnUCh6ALV8dwysWbMWkyvVKIgnhNQUve12LMmemGCG5DMuN/TmXKeMwjN6OiZnKjIQIPmMy0m/aJqxpJZy9o+M9+kX3qgt7SVnkjpyMpbUbLIm/kKLrLviHswOF84Tvkt50eIYLbvp027gv3s+5ZE/aDlFkQX0LjznLKg7RKjZi9WVxPRpP0HKkIqsUygUPYZKntU51uTZksU/BCkRdgfZBQs4tG5BwtIQwmpH+n2YkzyEgn7cp3yf5i//RaDma+27UAiCfsyp2aTrgQ6HXvw/TA43Jqszfp2jrdqsKzxrC//bVpkJhUKhOFZU8mwv4su9u0lK+iuHXrwXYXWAkMiWJuxDhhvBAjGlIWwOZDAIgLDYyLjwhlY1jxaRN+t5w7/TUrEP79tPY7K5sA/6Jk17/h43cs9fVYo5NZsh058yjuV875luuhIKhULROfTb5TshxCVCiM+EEHuFED/v7OOvWbMWhInG3e9iSc7A7EpBtjRh9WiRdQnzh/x+kk4bDQIGTXs0ps6QbGmMyhmqf38DhCRJp1+CpWYfWQMHJ0yWlX6f0VfdaytYeO89nX3aCoVC0aX0y5mSEMIM/AEYA+wH/imE2CClLG57z/bzkxtu1EpITLiTQ+sWAGD1aAKo4ci61O9PofLVh7RlOZsDGQyByUTDh5sRNmeCnKFc47NmpJoJBXxG8AHAdTffEiUVFK6D1PDhZvYtLSBr4GCeeOhB5RdSKBR9jn5plID/AfZKKb8AEEI8D/wQ6DSjFEAgW5qNpToZ8OM8eSSNxW/hGn6B5uOpKkXYnPoeAkwmhDAh7C5CLU2UFy0ma/ydUWHh6edPNfrw7S8mxZNJbWVZVN/X3zBNM3a1h7BmDMFz0TTM7nR8u9+hxVvXWaeoUCgU3U5/NUqDgdKIz/uBs1tvJIS4GbgZIC8v76g6CBfoCy/VVW17HO/Hb+D+9kU07d5pGCTpa9J3CGkF9YLNgAMCLYSCQcoKNTVvc3IWIb8vOmeoaAnPPPZwTN83T7uBx59ZS87khYcNWtESbp563VGdg0KhUPQ6OrM4U295AT8Cnoz4fC3wcFv7HG2RP2FzSeuAk6TJlSZzrnxAZlx2hzS50rQifAgpbE4JJinsbr2onkNitskZM2+JOs7q1Wtk1sAhEoQ0udKkOSVbIoQ0O9wx20YyY+YtWlFBhLS6ktvcVqFQKLoKOrnIX48bkK54ASOB1yI+3w3c3dY+R2uUhp96qjS506R1wElSWJ0RhijOv1aHzMwZ1GYlVVV1VaFQ9EU62yj1yzwlIYQF2A38ADgA/BO4Wkr5caJ9OpKn9O3TTuOT3Z9rS3kR+UVZAwezfMlvVaCBQqHo96g8pXYgpQwIIW4FXgPMwB/bMkgd5eP//KezD6lQKBTHNf3SKAFIKTcBm3p6HAqFQqFoP/02eVahUCgUfQ9llBQKhULRa1BGSaFQKBS9BmWUFAqFQtFr6Jch4R1BCFEOlHRw90ygohOH05dQ5358os79+CPReedLKbM6qxNllDoBIcR7nRmn35dQ567O/XjjeD337jpvtXynUCgUil6DMkoKhUKh6DUoo9Q5PN7TA+hB1Lkfn6hzP/7olvNWPiWFQqFQ9BrUTEmhUCgUvQZllBQKhULRa1BG6RgQQlwihPhMCLFXCPHznh7P0SCE+KMQokwI8VFEm0cIsVUIsUf/N11vF0KI3+vnuUsIcUbEPtfr2+8RQlwf0X6mEOI/+j6/F0KItvroToQQuUKIN4UQxUKIj4UQs46X8xdCOIQQ/xBC/Fs/91/p7ScIIf6uj3edEMKmt9v1z3v174dGHOtuvf0zIcTFEe1x/18k6qM7EUKYhRD/EkJsbGtM/fC8v9T/Hj8UQrynt/XOv/fOLM50PL3QSmJ8DnwDsAH/Bob39LiOYvznAWcAH0W0LQZ+rr//OfBb/f2lwGZAAN8D/q63e4Av9H/T9ffp+nf/0LcV+r7j2uqjm899IHCG/j4ZrfbW8OPh/PXxJOnvrcDf9XG+AFypt68EZujvZwIr9fdXAuv098P1v3k7cIL+f8Hc1v+LRH108/nfAawFNrY1pn543l8Cma3aeuXfe7demP70ogPVbXvbCxhKtFH6DBiovx8IfKa/fwy4qvV2wFXAYxHtj+ltA4FPI9qN7RL10cPX4S/AmOPt/AEX8AFwNlqmvkVvN/620WqSjdTfW/TtROu/9/B2if5f6PvE7aMbz3cI8DpwEbCxrTH1p/PW+/2SWKPUK//e1fJdxxkMlEZ83q+39WVypJRf6e+/BnL094nOta32/XHa2+qjR9CXZb6LNmM4Ls5fX8L6ECgDtqI94ddIKQP6JpHjNc5R/74WyODor0lGG310F78D7gRC+ue2xtSfzhtAAluEEO8LIW7W23rl33u/LfKnODaklFII0aX5At3RR1sIIZKAl4DbpZR1+jJ4t42tp85fShkEThdCpAGFwDe7ewzdjRDicqBMSvm+EOKCHh5OT/B9KeUBIUQ2sFUI8Wnkl73p713NlDrOASA34vMQva0vc0gIMRBA/7dMb090rm21D4nT3lYf3YoQwopmkNZIKdcfYWz97vwBpJQ1wJtoS0ppQojwQ2rkeI1z1L9PBSo5+mtS2UYf3cEoYIIQ4kvgebQlvBVtjKm/nDcAUsoD+r9laA8i/0Mv/XtXRqnj/BMYpkfW2NCcoRt6eEzHygYgHFFzPZqvJdx+nR6V8z2gVp+SvwaMFUKk61E1Y9HWy78C6oQQ39OjcK5rdax4fXQb+pieAj6RUj4Y8VW/P38hRJY+Q0II4UTzpX2CZpx+FGdckeP9EfCG1BwEG4Ar9Si1E4BhaM7uuP8v9H0S9dHlSCnvllIOkVIO1cf0hpRyShtj6hfnDSCEcAshksPv0f5OP6K3/r13t8OtP73QolR2o63Jz+/p8Rzl2J8DvgL8aGvAN6Ktf78O7AG2AR59WwH8QT/P/wBnRRznBmCv/vpJRPtZ+h/+58DDHFYPidtHN5/799HW2HcBH+qvS4+H8wdGAP/Sz/0j4B69/RtoN9e9wIuAXW936J/36t9/I+JY8/Xz+ww92qqt/xeJ+uiB3/8CDkff9fvz1vv/t/76ODy23vr3rmSGFAqFQtFrUMt3CoVCoeg1KKOkUCgUil6DMkoKhUKh6DUoo6RQKBSKXoMySgqFQqHoNSijpFB0I0KIiUIIKYRoU0VBCHG7EMJ1DP1MFUI83NH9FYqeQhklhaJ7uQr4q/5vW9yOJpiqUBxXKKOkUHQTutbe99ESla/U28xCiKVCiI/02jW3CSF+BgwC3hRCvKlv1xBxnB8JIZ7W348XWq2efwkhtgkhelTgVqE4VpQgq0LRffwQeFVKuVsIUSmEOBNNg2wocLqUMiCE8Egpq4QQdwAXSikrjnDMvwLfk1JKIcQ0NBXsOV15EgpFV6KMkkLRfVyFJgIKmijoVWiF4lZKvbSBlLLqKI85BFini13agP920lgVih5BGSWFohsQQnjQlKlP0+X7zWj6e/9s5yEi9cAcEe8fAh6UUm7QSzLce8yDVSh6EOVTUii6hx8Bq6SU+VLKoVLKXLRZzb+Bn4ZLG+jGC6AerVR7mENCiG8JIUxAQUR7KofLBFyPQtHHUUZJoegerkKrYxPJS2glovcBu4QQ/wau1r97HHg1HOgA/BythPe7aOruYe4FXhRCvI9Wdluh6NMolXCFQqFQ9BrUTEmhUCgUvQZllBQKhULRa1BGSaFQKBS9BmWUFAqFQtFrUEZJoVAoFL0GZZQUCoVC0WtQRkmhUCgUvYb/D3JDLv14XdFCAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# So let's run the model actual values against the predicted ones \n",
    "\n",
    "fig, ax = plt.subplots() \n",
    "\n",
    "ax.scatter(test_df[\"yield_actual\"], test_df[\"yield_predicted\"],edgecolors=(0, 0, 0))\n",
    "\n",
    "ax.set_xlabel('Actual')\n",
    "ax.set_ylabel('Predicted')\n",
    "ax.set_title(\"Actual vs Predicted\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "ec67a892",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:18:58.414214Z",
     "iopub.status.busy": "2021-12-10T09:18:58.413209Z",
     "iopub.status.idle": "2021-12-10T09:18:58.427249Z",
     "shell.execute_reply": "2021-12-10T09:18:58.426634Z",
     "shell.execute_reply.started": "2021-12-10T08:07:43.760789Z"
    },
    "papermill": {
     "duration": 0.111056,
     "end_time": "2021-12-10T09:18:58.427396",
     "exception": false,
     "start_time": "2021-12-10T09:18:58.316340",
     "status": "completed"
    },
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Item\n",
       "Cassava                 0.929086\n",
       "Maize                   0.894981\n",
       "Plantains and others    0.797340\n",
       "Potatoes                0.911456\n",
       "Rice, paddy             0.893915\n",
       "Sorghum                 0.795925\n",
       "Soybeans                0.847358\n",
       "Sweet potatoes          0.840310\n",
       "Wheat                   0.921211\n",
       "Yams                    0.926169\n",
       "dtype: float64"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def adjusted_r_squared(y,yhat,x):\n",
    "    score=1- (((1-(r2_score(y,yhat)))*(len(y)-1))/(len(y)-x.shape[1]-2))\n",
    "    return score\n",
    "\n",
    "test_group.apply(lambda x: adjusted_r_squared(x.yield_actual,x.yield_predicted,x))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "70538bd1",
   "metadata": {
    "papermill": {
     "duration": 0.091501,
     "end_time": "2021-12-10T09:18:58.611339",
     "exception": false,
     "start_time": "2021-12-10T09:18:58.519838",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Model Results & Conclusions\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "9fe52c2f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:18:58.798889Z",
     "iopub.status.busy": "2021-12-10T09:18:58.798203Z",
     "iopub.status.idle": "2021-12-10T09:18:58.800784Z",
     "shell.execute_reply": "2021-12-10T09:18:58.801276Z",
     "shell.execute_reply.started": "2021-12-10T08:07:43.781807Z"
    },
    "papermill": {
     "duration": 0.098091,
     "end_time": "2021-12-10T09:18:58.801451",
     "exception": false,
     "start_time": "2021-12-10T09:18:58.703360",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "varimp= {'imp':model.feature_importances_,'names':yield_df_onehot.columns[yield_df_onehot.columns!=\"hg/ha_yield\"]}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "90d15570",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:18:58.987674Z",
     "iopub.status.busy": "2021-12-10T09:18:58.986969Z",
     "iopub.status.idle": "2021-12-10T09:19:01.503147Z",
     "shell.execute_reply": "2021-12-10T09:19:01.503717Z",
     "shell.execute_reply.started": "2021-12-10T08:07:43.787613Z"
    },
    "papermill": {
     "duration": 2.610936,
     "end_time": "2021-12-10T09:19:01.503887",
     "exception": false,
     "start_time": "2021-12-10T09:18:58.892951",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 595.44x1202.4 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "a4_dims = (8.27,16.7)\n",
    "\n",
    "fig, ax = plt.subplots(figsize=a4_dims)\n",
    "df=pd.DataFrame.from_dict(varimp)\n",
    "df.sort_values(ascending=False,by=[\"imp\"],inplace=True)\n",
    "df=df.dropna()\n",
    "sns.barplot(x=\"imp\",y=\"names\",palette=\"vlag\",data=df,orient=\"h\",ax=ax);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a4d3b2c1",
   "metadata": {
    "papermill": {
     "duration": 0.092942,
     "end_time": "2021-12-10T09:19:01.690600",
     "exception": false,
     "start_time": "2021-12-10T09:19:01.597658",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Getting only top 7 of features importance in the model:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "05589b9b",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:19:01.883615Z",
     "iopub.status.busy": "2021-12-10T09:19:01.882950Z",
     "iopub.status.idle": "2021-12-10T09:19:02.152304Z",
     "shell.execute_reply": "2021-12-10T09:19:02.151665Z",
     "shell.execute_reply.started": "2021-12-10T08:07:46.021212Z"
    },
    "papermill": {
     "duration": 0.368335,
     "end_time": "2021-12-10T09:19:02.152447",
     "exception": false,
     "start_time": "2021-12-10T09:19:01.784112",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1202.4x595.44 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#7 most important factors that affect crops \n",
    "a4_dims = (16.7, 8.27)\n",
    "\n",
    "fig, ax = plt.subplots(figsize=a4_dims)\n",
    "df=pd.DataFrame.from_dict(varimp)\n",
    "df.sort_values(ascending=False,by=[\"imp\"],inplace=True)\n",
    "df=df.dropna()\n",
    "df=df.nlargest(7, 'imp')\n",
    "sns.barplot(x=\"imp\",y=\"names\",palette=\"vlag\",data=df,orient=\"h\",ax=ax);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "983682a2",
   "metadata": {
    "papermill": {
     "duration": 0.095626,
     "end_time": "2021-12-10T09:19:02.344099",
     "exception": false,
     "start_time": "2021-12-10T09:19:02.248473",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "The crop being potatoes has the highest importance in the decision making for the model, where it's the highest crops in the dataset. Cassava too, then as expected we see the effect of pesticides, where its the third most important feature, and then if the crop is sweet potatoes, we see some of the highest crops in features importance in dataset. \n",
    "\n",
    "If the crop is grown in India, makes sense since Indis has the largest crops sum in the dataset. Then comes rainfall and temprature. Thr first assumption about these features were correct, where they all significanally impact the expected crops yield in the model. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "edd62195",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-10T09:19:02.536353Z",
     "iopub.status.busy": "2021-12-10T09:19:02.535603Z",
     "iopub.status.idle": "2021-12-10T09:19:02.943759Z",
     "shell.execute_reply": "2021-12-10T09:19:02.944469Z",
     "shell.execute_reply.started": "2021-12-10T08:07:46.228572Z"
    },
    "papermill": {
     "duration": 0.506641,
     "end_time": "2021-12-10T09:19:02.944706",
     "exception": false,
     "start_time": "2021-12-10T09:19:02.438065",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA+0AAAHuCAYAAADqTO2UAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8/fFQqAAAACXBIWXMAAAsTAAALEwEAmpwYAABKWElEQVR4nO3de3xcdZ3/8fcnCWC5tUMbCNBisy6KrqsoFVBJJGgRlF1wt3FR5FKxsKtoBS+L628jW3ddLyuI63oBJICgaLKwoAvWLAQSFAoFpQXKpZBAAw1tYdpCWyhpPr8/znfamXRymWYm58zk9Xw88sj5fudcPnPmzJnzOd9zvsfcXQAAAAAAIHmq4g4AAAAAAADkR9IOAAAAAEBCkbQDAAAAAJBQJO0AAAAAACQUSTsAAAAAAAlVE3cASTJjxgyfPXt23GEAAAAAACaZ+++/f5271w6tJ2nPMnv2bC1dujTuMAAAAAAAk4yZPZ2vnsvjAQAAAABIKJJ2AAAAAAASiqQdAAAAAICEImkHAAAAACChSNoBAAAAAEgoknYAAAAAABKKpB0AAAAAgIQiaQcAAAAAIKFI2gEAAAAASCiSdgAAAAAAEoqkHQAAAACAhCJpBwAAAAAgoUjaAQAAAABIKJJ2AAAAAAASiqQdAAAAAICEImkHdkE6nVZLS4vS6XTcoQAAAACoYCTtwC5oa2vTihUr1N7eHncoAAAAACpYyZN2M+s1s+Vm9iczWxrq9jOzDjN7IvxPhXozs++b2UozW2Zm78yaz5lh/CfM7Mys+iPC/FeGaW2kZQDjlU6n1dnZKXdXZ2cnre0AAAAASmaiWtqb3P1wd58TyhdKus3dD5V0WyhL0omSDg1/50j6kRQl4JK+JukoSUdK+lpWEv4jSQuypjthlGUA49LW1iZ3lyQNDg7S2g4AAACgZOK6PP5kSVeH4aslnZJVf41H7pE0zcwOlPRBSR3u/qK7pyV1SDohvLavu9/jURZ1zZB55VsGMC7d3d0aGBiQJA0MDKirqyvmiAAAAABUqolI2l3S78zsfjM7J9Qd4O6rw3C/pAPC8MGSVmVN2xfqRqrvy1M/0jJymNk5ZrbUzJauXbu24DeHyaehoUHhLgyZmRobG2OOCAAAAEClmoik/Rh3f6eiS98/Y2Y5GU5oIfdSBjDSMtz9Mnef4+5zamtrSxkGKsTcuXO3Xx7v7po7d27MEQEAAACoVCVP2t392fB/jaQbFd2T/ny4tF3h/5ow+rOSZmVNPjPUjVQ/M0+9RlgGMC4dHR05Le0dHR0xRwQAAACgUpU0aTezvcxsn8ywpOMlPSTpZkmZHuDPlHRTGL5Z0hmhF/mjJW0Il7gvlnS8maVCB3THS1ocXttoZkeHXuPPGDKvfMsAxqW7uzunpZ172gEAAACUSk2J53+ApBtDq2SNpJ+7+2/N7D5JvzKzsyU9LemjYfxbJH1I0kpJmyXNlyR3f9HMvi7pvjDeInd/MQx/WtJVkqZIujX8SdI3h1kGMC4NDQ267bbbtG3bNlVXV3NPOwAAAICSsUyLIaQ5c+b40qVL4w4DCZdOp3XOOefI3VVVVaWf/OQnSqVSo08IAAAAAMMws/uzHpO+XVyPfAMqAie9AAAAAJQSSTtQoLa2tpx72tvb22OOCAAAAEClImkHCjS047k777wzpkgAAAAAVDqSdqBAM2bMyCnX1tbGFAkAAACASkfSDhRo7dq1OeU1a9bEFAkAAACASkfSDhRov/32yylPnz49pkgAAAAAVDqSdqBA/f39OeXVq1fHFAkAAACASkfSDhRo6GPeeOwbAAAAgFIhaQcKVF1dPWIZAAAAAIqFpB0o0DHHHJNTbmhoiCkSAAAAAJWOpB0oUGNj44hlAAAAACgWknagQK2trSOWAQAAAKBYSNqBAvX19eWUV61aFVMkAAAAACodSTtQoJkzZ+aUZ82aFVMkAAAAACodSTtQoIULF45YBgAAAIBiIWkHCjRt2rSc8tSpU+MJBAAAAEDFI2kHCtTW1pZTbm9vjykSAAAAAJWOpB0o0J133plTvuOOO+IJBAAAAEDFI2kHClRTUzNiGQAAAACKhaQdKNCmTZtGLAMAAABAsZC0AwU68MADc8oHHXRQTJEAAAAAqHQk7UCBZs+ePWIZAAAAAIqFpB0o0AMPPJBTvv/++2OKBAAAAEClI2kHCkRHdAAAAAAmCkk7UCA6ogMAAAAwUUjaAQAAAABIKJJ2AAAAAAASiqQdAAAAAICEImkHAAAAACChSNoBAAAAAEgoknYAAAAAABKKpB0AAAAAgIQiaQcAAAAAIKFI2gEAAAAASCiSdgAAAAAAEoqkHQAAAACAhCJpBwAAAAAgoUjaAQAAAABIKJJ2AAAAAAASiqQdAAAAAICEImkHAAAAACChSNoBAAAAAEgoknYAAAAAABKKpB0AAAAAgIQiaQcAAAAAIKFI2gEAAAAASCiSdgAAAAAAEoqkHQAAAACAhCJpBwAAAAAgoUjaAQAAAABIKJJ2AAAAAAASiqQdAAAAAICEImkHAAAAACChSNoBAAAAAEgoknYAAAAAABKqJu4AgInS2tqqnp6eksy7paVll6etr6/X/PnzixgNAAAAgEpBSzsAAAAAFNmDDz6oj370o1q2bFncoaDMmbvHHUNizJkzx5cuXRp3GEi4n//857rhhhu2l+fNm6dTTz01xogAAACQNGeeeaY2bdqkvfbaS1dffXXc4aAMmNn97j5naD0t7UCBPv7xj+eUSdgBAACQ7cEHH9SmTZskSZs2baK1HeNC0g7sgv32209S1MoOAAAAZLv44otzyt/97ndjigSVgI7ogF1QV1enuro6WtkBAACwk0wr+3BloBC0tAMAAABAEe21114jloFCkLQDAAAAQBEtWLAgp3zuuefGFAkqAUk7AAAAABTRI488klN++OGHY4oElYCkHQAAAACKqKurK6d85513xhQJKgFJOwAAAAAU0YwZM3LKtbW1MUWCSkDSDgAAAABFtG7dupzy2rVrY4oElYCkHQAAAACKqLGxUWYmSTIzve9974s5IpQzknYAAAAAKKLm5mZVV1dLkmpqajRv3ryYI0I5I2kHAAAAgCJKpVI67rjjZGY67rjjlEql4g4JZawm7gAAAAAAoNI0Nzerr6+PVnaMGy3tAAAAQIVLp9NqaWlROp2OO5RJI5VKadGiRbSyY9wmJGk3s2oz+6OZ/SaU681siZmtNLNfmtnuoX6PUF4ZXp+dNY+vhPrHzOyDWfUnhLqVZnZhVn3eZQAAAACTTVtbm1asWKH29va4QwFQoIlqaV8oaUVW+VuSLnH3P5eUlnR2qD9bUjrUXxLGk5m9RdKpkv5C0gmSfhhOBFRL+i9JJ0p6i6SPhXFHWgYAAAAwaaTTaXV2dsrd1dnZSWs7UGZKnrSb2UxJH5Z0RSibpOMkZU7zXS3plDB8cigrvP7+MP7Jkq5391fdvUfSSklHhr+V7v6Uu2+VdL2kk0dZBgAAADBptLW1yd0lSYODg7S2A2VmIlravyfpy5IGQ3m6pPXuPhDKfZIODsMHS1olSeH1DWH87fVDphmufqRl5DCzc8xsqZktXbt27S6+RQAAACCZuru7NTAQHRYPDAyoq6sr5ogAFKKkSbuZnSRpjbvfX8rljIe7X+buc9x9Tm1tbdzhAAAAAEXV0NCgmprooVE1NTVqbGyMOaLJoaenR2eccYZ6e3vjDgVlrtQt7e+V9Ndm1qvo0vXjJF0qaZqZZR43N1PSs2H4WUmzJCm8PlXSC9n1Q6YZrv6FEZYBAAAATBrNzc2K7h6VqqqqeATZBLn00ku1efNmXXrppXGHgjJX0qTd3b/i7jPdfbaijuRud/fTJHVKyuwtzpR0Uxi+OZQVXr/doxtwbpZ0auhdvl7SoZLulXSfpENDT/G7h2XcHKYZbhkAAADApJFKpdTU1CQzU1NTE48gmwA9PT3q6+uTJK1atYrWdoxLXM9p/0dJF5jZSkX3n/801P9U0vRQf4GkCyXJ3R+W9CtJj0j6raTPuPu2cM/6eZIWK+qd/ldh3JGWAQAAAEwqzc3NevOb30wr+wQZ2rpOazvGo2b0UYrD3e+QdEcYfkpRz+9Dx3lFUvMw0/+bpH/LU3+LpFvy1OddBgAAADDZpFIpLVq0KO4wJo1MK3vGqlWrhhkTGF1cLe0AAAAAUJFmzpyZU541a9YwYwKjI2kHAAAAgCKaP3/+iGWgECTtAAAAAFBES5YsGbEMFIKkHQAAAACKqLu7O6fc1dUVUySoBCTtAAAAAFBERx6Z2x/2UUcdFVMkqAQT1ns8AAAAyldra6t6enrGPZ/+/n5JUl1d3bjnJUn19fXcLwygotHSDgAAgAnzyiuv6JVXXok7DKCk7r333pwy97RjPGhpBwAAwKiK1Zrd0tIiSTwzHBWtoaFBv/vd77aXGxsbY4wG5Y6WdgAAAAAooqH3sHNPO8aDpB0AAAAAiqi1tXXEMlAIknYAAAAAKKK+vr6c8qpVq2KKBJWApB0AAAAAimjmzJk55VmzZsUUCSoBSTsAAAAAFNHChQtHLAOFIGkHAAAAgCKqr6/f3to+a9YszZ49O96AUNZI2gEAAACgyBYuXKg999yTVnaMG89pBwAAAIAiq6+v1zXXXBN3GKgAtLQDAAAAAJBQJO0AAAAAACQUSTsAAAAAAAlF0g4AAAAAQEKRtAMAAAAAkFAk7QAAAAAAJBRJOwAAAAAACUXSDgAAAABAQpG0AwAAAACQUCTtAAAAAAAkFEk7AAAAAAAJRdIOAAAAAEBCkbQDAAAAAJBQJO0AAABAhUun02ppaVE6nY47FAAFImkHAAAAKlxbW5tWrFih9vb2uEMBUCCSdgAAAKCCpdNpdXZ2yt3V2dlJaztQZkjaAQAAgArW1tYmd5ckDQ4O0toOlBmSdgAAAKCCdXd3a2BgQJI0MDCgrq6umCMCUAiSdgAAAKCCNTQ05JQbGxtjigTAriBpBwAAACrY3nvvnVPeZ599YooEwK4gaQcAAAAq2A033JBT5p52oLyQtAMAAAAAkFAk7QAAAABQZOl0Wi0tLTxiD+NG0g4AAAAARdbW1qYVK1ZwOwLGjaQdAAAAAIoonU6rs7NT7q7Ozk5a2zEuJO0AAAAAUERtbW1yd0nS4OAgre0YF5J2AAAAoIKddtppOeXTTz89pkgmj+7ubg0MDEiSBgYG1NXVFXNEKGck7QAAAEAF+8hHPpJTPvnkk2OKZPJoaGhQTU2NJKmmpkaNjY0xR4RyRtIOAAAAVLCenp6ccm9vbzyBTCLNzc0yM0lSVVWV5s2bF3NEKGck7QAAAEAF+/rXvz5iGcWXSqXU1NQkM1NTU5NSqVTcIaGM1cQdAAAAAIDS2bhxY055w4YNMUUyuTQ3N6uvr49WdowbSTsAAAAAFFkqldKiRYviDgMVgMvjAQAAAABIKJJ2AAAAAAASiqQdAAAAAICEImkHAAAAACChSNoBAACACrbbbruNWAaQbCTtAAAAQAVramrKKR933HExRQJgV5C0AwAAABWsubk5p8xzw4HyQtIOAAAAVLD169fnlDds2BBPIAB2CUk7AAAAUMEuvfTSEcsAko2kHQAAAKhgfX19OeVVq1bFFAmAXUHSDgAAAFSwob3F77777jFFAmBXkLQDAAAAFey1117LKW/dujWmSADsCpJ2AAAAAAASiqQdAAAAAIosnU6rpaVF6XQ67lBQ5kjaAQAAAKDIWlpa9Mgjj+hrX/ta3KGgzJG0AwAAABVs1qxZOeXXv/71MUUyeaTTaa1evVqS9Nxzz9HajnEhaQcAAAAq2Lp163LKa9asiSmSyaOlpSWnTGs7xoOkHQAAAKhgDQ0NMjNJkpmpsbEx5ogqX6aVPeO5556LKRJUApJ2AAAAoII1NzerpqZGUvTM9nnz5sUcEYBCkLQDAAAAFSyVSqmurk6SdOCBByqVSsUcEYBCkLQDAAAAFW7VqlWSpKeffjrmSCaHqqqqEctAIdh6AAAAgAr205/+NKd81VVXxRPIJDJlypQRy0AhSNoBAACACnbrrbfmlH/zm9/EFMnksWnTphHLQCFI2gEAAACgiGbOnJlTnjVrVkyRoBKQtAMAAABAES1cuHDEMlCIkibtZvY6M7vXzB40s4fN7F9Cfb2ZLTGzlWb2SzPbPdTvEcorw+uzs+b1lVD/mJl9MKv+hFC30swuzKrPuwwAAABgMjnxxBNzyieddFJMkUwe9fX121vbZ82apdmzZ8cbEMpaqVvaX5V0nLu/XdLhkk4ws6MlfUvSJe7+55LSks4O458tKR3qLwnjyczeIulUSX8h6QRJPzSzajOrlvRfkk6U9BZJHwvjaoRlAAAAAJPG2WfnHgafddZZ8QQyySxcuFB77rknrewYt5Im7R55ORR3C38u6ThJ7aH+akmnhOGTQ1nh9febmYX66939VXfvkbRS0pHhb6W7P+XuWyVdL+nkMM1wywAAAAAmjZ6enpxyb29vPIFMMvX19brmmmtoZce4lfye9tAi/idJayR1SHpS0np3Hwij9Ek6OAwfLGmVJIXXN0ianl0/ZJrh6qePsIyh8Z1jZkvNbOnatWvH8U4BAACA5Ln00ktHLANItpIn7e6+zd0PlzRTUcv4YaVeZiHc/TJ3n+Puc2pra+MOBwAAACiqvr6+nPKqVauGGRNAEk1Y7/Huvl5Sp6R3S5pmZjXhpZmSng3Dz0qaJUnh9amSXsiuHzLNcPUvjLAMAAAAYNLg8WNAeSt17/G1ZjYtDE+RNFfSCkXJ+7ww2pmSbgrDN4eywuu3u7uH+lND7/L1kg6VdK+k+yQdGnqK311RZ3U3h2mGWwYAAAAwacyfP3/EMoBkK3VL+4GSOs1smaIEu8PdfyPpHyVdYGYrFd1//tMw/k8lTQ/1F0i6UJLc/WFJv5L0iKTfSvpMuOx+QNJ5khYrOhnwqzCuRlgGAAAAMGksWbJkxDKAZKsZfZRd5+7LJL0jT/1Tiu5vH1r/iqTmYeb1b5L+LU/9LZJuGesyAAAAgMmku7s7p9zV1aUFCxbEFA2AQk3YPe0AAAAAJl5DQ4OiJyJLZqbGxsaYIwJQCJJ2AAAAoILNnTtXUZdPkrtr7ty5MUcEoBAk7QAAAEAF6+joyGlp7+joiDmiySGdTqulpUXpdDruUFDmSNoBAACACtbd3Z3T0t7V1RVzRJNDW1ubVqxYofb29rhDQZkjaQcAAAAq2OGHH55Tfsc7duonGkWWTqd1++23y911++2309qOcSFpBwAAACpYb2/viGUUX1tbm7Zt2yZJGhgYoLUd40LSDgAAAFSw1atX55Sfe+65mCKZPLq6unJuSbjzzjtjjgjljKQdAAAAqGAHHHDAiGUU34wZM3LKtbW1MUWCSkDSDgAAAFSwTIsvJs7atWtzymvWrIkpElSCmtFGMLPlkob9prv724oaEQAAAICiGZowPv/88zFFMnnU1taqr69ve3n//fePMRqUu1GTdkknhf+fCf9/Fv6fVvxwAAAAAKC8rVu3Lqc8tOUdKMSol8e7+9Pu/rSkue7+ZXdfHv4ulHR86UMEAAAAgPLR2NgoM5MkmZne9773xRwRylkh97Sbmb03q/CeAqcHAAAAgIrX3NysmproouaamhrNmzcv5ohQzgpJus+W9EMz6zWzpyX9UNInSxMWAAAAAJSnVCqlpqYmmZmOO+44pVKpuENCGRvLPe2SJHe/X9LbzWxqKG8oWVQAAAAAUMaam5vV19dHKzvGbSy9x18wTL0kyd0vLnJMAAAAAFDWUqmUFi1aFHcYqABjaWnfp+RRAAAAAACAnYyatLv7v0xEIAAAAAB2aG1tVU9Pz7jns/vuu2vr1q055ZaWll2eX319vebPnz/uuACMzZg7ojOzN5rZbWb2UCi/zcz+X+lCAwAAADBeBx98cE555syZMUUCYFeMuSM6SZdL+pKkn0iSuy8zs59L+tdSBAYAAABMZsVszf74xz+urVu3atasWfr2t79dtPkCKL1CHvm2p7vfO6RuoJjBAAAAACi+gw8+WFVVVVq4cGHcoQAoUCFJ+zoze4MklyQzmydpdUmiAgAAAFA0U6ZM0WGHHabZs2fHHQqAAhVyefxnJF0m6TAze1ZSj6RPlCQqAAAAAAAw9pZ2d3/K3T8gqVbSYe5+jLv3liwyAAAAAChT6XRaLS0tSqfTcYeCMjdqS7uZfcLdrzWzC4bUS5Lc/eISxQYAAAAAZenaa6/VI488ouuuu07nnXde3OGgjI2lpX2v8H+fYf4AAAAAAEE6nVZXV5ckqauri9Z2jMuoLe3u/pMw+AN3f6HE8QAAAABAWbv22mvl7pKkwcFBWtsxLoX0Hn+PmbWZ2Ycsc208AAAAACDHXXfdlVPu7u6OKRJUgkKS9jcq6j3+dElPmNk3zOyNpQkLAAAAAAAU0nu8u3uHu39M0gJJZ0q618zuNLN3lyxCAAAAACgj73rXu3LKRx55ZEyRoBKM+TntZjZd0XPZT5f0vKTPSrpZ0uGS2iTVlyA+AAAAACgrQ+8m5u5ijMeYk3ZJd0v6maRT3L0vq36pmf24uGEBkdbWVvX09MQdxk56e3slSS0tLfEGMkR9fb3mz58fdxgAAACT2r333ptTXrJkSUyRoBIUkrS/yTNdIA7h7t8ys/90988WKS5AktTT06PHHl+pKftOjzuUHK9ti/4/05+cx3ds2cjDHQAAAIBKM+akfbiEPct7xxkLkNeUfafrsKP+Ou4wEu/RJTfHHQIAAAAkHXPMMbrzzju3lxsaGmKMBuWukJZ2AAAAAMAoPvGJT6irq0vuLjPTaaedFndIiVXM22H7+/slSXV1dUWZX1JuPS3kkW8AAAAAgFGkUqntieOBBx6oVCoVc0STwyuvvKJXXnkl7jCKrpgt7XSJCAAAAGDSS6fTWrdunSRp7dq1SqfTJO7DKGZLdqaT6EWLFhVtnklQzJb2S4s4LwAAAAAoS21tbcp0Cebuam9vjzkilLMxJ+1mVmtm/2Fmt5jZ7Zm/zOvuflVJIgQAAACAMtLd3a2BgQFJ0sDAgLq6umKOCOWskJb26yStkFQv6V8k9Uq6rwQxAQAAAEDZamhoUE1NdCdyTU2NGhsbY44I5ayQpH26u/9U0mvufqe7f1LScSWKCwAAAADKUnNz8/ZhM9O8efNijAblrpCk/bXwf7WZfdjM3iFpvxLEBAAAAABlK5VKqba2VpJUW1tLJ3QYl0J6j/9XM5sq6QuS/lPSvpLOL0lUAAAAAFCm0un09meGr169mt7jMS5jbml399+4+wZ3f8jdm9z9CHe/uZTBAQAAAEC5ufbaa3N6j7/uuutijgjlrNDe4//JzC4zsyszf6UMDgAAAADKzV133ZVT7u7ujikSVIJCLo+/SVK3pP+TtK004QAAAABAeRscHByxDBSikKR9T3f/x5JFAgAAAAAVoKqqStu2bcspA7uqkK3nN2b2oZJFAgAAAAAV4F3veldO+cgjj4wpElSCUVvazewlSS7JJP2Tmb2q6PFvJsndfd/ShggAAAAA5cPMRiwDhRjL5fHT3X1rySMBAAAAgApw77335pSXLFkSUySoBGNJ2v9gZn2Sfivpt+7eW9qQAAAAAACANIZ72t19jqTPh+L3zOw+M7vEzI43sz1KGh0AAAAAlBnuaUcxjakjOnfvdfcfu/spkt4j6deSPiCp28z+t4TxAQAAAEBZ2WOPPUYsA4Uo+NkD7v6au9/u7l929yMlnVOCuAAAAACgLHFPO4ppzEm7mS03s2VD/rolfdHMppcwRgAAAAAoG0Mvhz/qqKNiigSVYCwd0WXcKmmbpJ+H8qmS9pTUL+kqSX9V1MgAAAAAAJjkCrk8/gPu/hV3Xx7+virpfe7+LUmzSxMeAAAAAJSXoZfD33PPPTFFgkpQSNJebWbbr/MIw9WhOFDUqAAAAACgTM2YMSOnXFtbG1MkqASFXB7/KUlXmtneofySpLPNbC9J/170yAAAAACgDK1bty6nvHbt2pgiQSUopKV9maRvSrpC0v9I+m9JJ7r7Jnf/VQliAwAAAICyM7TjuaOPPjqmSFAJCmlpv0nSekkPSOorSTQAAAAomtbWVvX09MQdRo7e3l5JUktLS7yB5FFfX6/58+fHHQYA5CgkaZ/p7ieULBIAAAAUVU9Pj5568kkdvH9y7qfdrSq60PPVlzbGHEmuZ9dw+TKKJ99z2s8777yYokG5KyRp/4OZ/aW7Ly9ZNAAAACiqg/ev1XkfOzXuMBLvB7+4Pu4QUEEaGhp02223adu2baqurlZjY2PcIaGMjXpPu5ktN7Nlko6R9ICZPWZmy7LqAQAAAABBc3OzqsJVJdXV1Zo3b17MEaGcjaWl/aSSRwEAAAAAMStmPxCZpH2vvfbSJZdcMq550d/C5DZqS7u7Pz3S30QEieGl02m1tLQonU7HHQoAAACAoKqqSlVVVTyjHeNWyD3tSKC2tjatWLFC7e3tWrBgQdzhAKgg6XRal1xyic4//3ylUqm4wwEAoOSK2ZqdeULCokWLijZPTE6FPKcdCZNOp9XZ2Sl3V2dnJ63tAIoq+6QgAAAA4kHSXsba2trk7pKkwcFBDqwBFA0nBQEAAJKBpL2MdXd3a2BgQJI0MDCgrq6umCMCUCk4KQgAAJAMJO1lrKGhQTU1UbcENTU1PP8RQNFwUhAAACAZSNrLWHNzs8xMUtQ7Jc9/BFAsnBQEAABIhpIm7WY2y8w6zewRM3vYzBaG+v3MrMPMngj/U6HezOz7ZrbSzJaZ2Tuz5nVmGP8JMzszq/4IM1sepvm+hSx2uGVUklQqpaamJpmZmpqa6N0ZQNFwUhAAACAZSt3SPiDpC+7+FklHS/qMmb1F0oWSbnP3QyXdFsqSdKKkQ8PfOZJ+JEUJuKSvSTpK0pGSvpaVhP9I0oKs6U4I9cMto6I0NzfrzW9+MwfUAIqKk4IAAADJUNKk3d1Xu/sDYfglSSskHSzpZElXh9GulnRKGD5Z0jUeuUfSNDM7UNIHJXW4+4vunpbUIemE8Nq+7n6PRz0mXTNkXvmWUVFSqZQWLVrEATWAouOkIAAAQPxqJmpBZjZb0jskLZF0gLuvDi/1SzogDB8saVXWZH2hbqT6vjz1GmEZQ+M6R1Grvg455JBC3xYAVKzMSUEAAADEZ0I6ojOzvSX9t6TPu/vG7NdCC7mXcvkjLcPdL3P3Oe4+p7a2tpRhAAAAAABQkJIn7Wa2m6KE/Tp3vyFUPx8ubVf4vybUPytpVtbkM0PdSPUz89SPtIyKkk6n1dLSonQ6HXcoAAAAAIAiK3Xv8Sbpp5JWuPvFWS/dLCnTA/yZkm7Kqj8j9CJ/tKQN4RL3xZKON7NU6IDueEmLw2sbzezosKwzhswr3zIqSltbm1asWKH29va4QwEAAAAAFFmpW9rfK+l0SceZ2Z/C34ckfVPSXDN7QtIHQlmSbpH0lKSVki6X9GlJcvcXJX1d0n3hb1GoUxjnijDNk5JuDfXDLaNipNNpdXZ2yt3V2dlJazsAAAAAVJiSdkTn7ndJsmFefn+e8V3SZ4aZ15WSrsxTv1TSW/PUv5BvGZWkra1N0SqTBgcH1d7ergULFsQcFQAAAACgWCakIzqURnd3twYGBiRJAwMD6urqijkiAAAAAEAxkbSXsYaGBtXURBdL1NTUqLGxMeaIAAAAAADFRNJexpqbmxX1vydVVVVp3rx5MUcEAAAAACgmkvYylkql1NTUJDNTU1OTUqlU3CEBAAAAAIqopB3RofSam5vV19dHKzsAAAAAVCCS9jKXSqW0aNGiuMMAAAAAAJQAl8eXuZ6eHp1xxhnq7e2NOxQAAAAAQJGRtJe5Sy+9VJs3b9all14adygAAAAAgCLj8vgy1tPTo76+PknSqlWr1Nvbq9mzZ8cbVJH19/dr88aX9eiSm+MOJfE2b3xB/Xo17jAAAAAAFBEt7WVsaOs6re0AAAAAUFloaS9jmVb2jFWrVsUUSenU1dVpq9I67Ki/jjuUxHt0yc2qq+OxfwAAAEAloaW9jM2cOTOnPGvWrJgiAQAAAACUAkl7GVu4cOGIZQAAAABAeSNpL2P19fXbW9tnzZpVcZ3QAQAAAMBkR9Je5hYuXKg999yTVnYAAAAAqEB0RFfm6uvrdc0118QdBgAAAACgBGhpBwAAAAAgoUjaAQAAAABIKJJ2AAAAAAASiqQdAAAAAICEImkHAAAAACChSNoBAAAAAEgoknYAAAAAABKKpB0AAAAAgIQiaQcAAAAAIKFI2gEAAAAASCiSdgAAAAAAEoqkHQAAAACAhKqJOwAAAACgUrS2tqqnpyfuMHbS29srSWppaYk3kCHq6+s1f/78uMMAEo2kHQAAACiSnp4ePfnEE6qbOjXuUHJUDw5KkjatWRNzJDv0b9gQdwhAWSBpBwAAAIqobupUnXFsY9xhJN41d3TFHQJQFrinHQAAAACAhCJpBwAAAAAgoUjaAQAAAABIKO5pBwAAAAAUJIlPSqjUpySQtAMAAAAACtLT06MnVqxQqro67lC2823bJEnrHn885kh2SIeYxoOkHQAAAABQsFR1teZOS9bjDZOmY/34H23IPe0AAAAAACQUSTsAAAAAAAlF0g4AAAAAQEKRtAMAAAAAkFAk7QAAAAAAJBS9xyPxtmx8QY8uuTnuMHK8uinqBXKPvZLTW+aWjS9Idam4wwAAAABQRCTtZS6dTuuSSy7R+eefr1Sq8hK2+vr6uEPIq7c3StoPSVKSXJdK7PoCAAAAsGtI2stcW1ubVqxYofb2di1YsCDucIpu/vz5cYeQV0tLiyRp0aJFMUcCAAAAoJJxT3sZS6fT6uzslLurs7NT6XQ67pAAAAAAAEVE0l7G2traNDg4KEnatm2b2tvbY44IAAAAAFBMJO1lrLu7W9u2bZMUJe1dXV0xRwQAAAAAKCaS9jJ25JFH5pSPOuqomCIBAAAAAJQCSXsZe/XVV0csAwAAAADKG0l7Gbvvvvtyyvfee29MkQAAAAAASoGkHQAAAACAhOI57WXs7W9/ux544IHt5cMPPzy+YAAAQOL09/dr86aX9YNfXB93KIn37Jo12nPT5rjDAICd0NJexh599NERywAAAACA8kZLexnbvDn3bPCmTZtiigQAACRRXV2dXn1po8772Klxh5J4P/jF9dpjn33jDgMAdkJLOwAAAAAACUXSDgDIK51Oq6WlRel0Ou5QAAAAJi0ujwcA5NXW1qYVK1aovb1dCxYsiDscAACG1draqp6enrjDyNHb2ytJamlpiTeQIerr6zV//vy4w0ABSNoBADtJp9Pq7OyUu6uzs1Pz5s1TKpWKOywAAPLq6enRysce04wpU+IOZTt77TVJ0vpnnok5kh3WbdkSdwjYBSTtAICdtLW1yd0lSYODg7S2AwASb8aUKfqbw94YdxiJdsOjj8cdAnYB97QDAHbS3d2tgYEBSdLAwIC6urpijggAAGByImkHAOykoaFBNTXRxVg1NTVqbGyMOSIAAIDJiaQdALCT5uZmmZkkqaqqSvPmzYs5IgAAgMmJpB0AsJNUKqWmpiaZmZqamuiEDgAAICZ0RAcAyKu5uVl9fX20sgMAAMSIpB0AkFcqldKiRYviDgMAAGBS4/J4AAAAAAASiqQdAAAAAICEImkHAAAAACChSNoBlIUbb7xR8+bN00033RR3KAAAAMCEIWkHUBauu+46SdLPfvazmCMBAAAAJg5JO4DEu/HGG3PKtLYDAABgsiBpB5B4mVb2DFrbJ0Y6nVZLS4vS6XTcoQAlw3YOAEi6kibtZnalma0xs4ey6vYzsw4zeyL8T4V6M7Pvm9lKM1tmZu/MmubMMP4TZnZmVv0RZrY8TPN9M7ORlgEAGLu2tjatWLFC7e3tcYcClMyVV16pRx55RK2trXGHAgBAXqVuab9K0glD6i6UdJu7HyrptlCWpBMlHRr+zpH0IylKwCV9TdJRko6U9LWsJPxHkhZkTXfCKMsAAIxBOp1WZ2en3F2dnZ20QqIipdNp3X333ZKku+++m+0cAJBINaWcubt3mdnsIdUnSzo2DF8t6Q5J/xjqr3F3l3SPmU0zswPDuB3u/qIkmVmHpBPM7A5J+7r7PaH+GkmnSLp1hGUAAMagra1N0e5YGhwcVHt7uxYsWBBzVEBxXXnllduH3V2tra264IILYowIlaC/v1+bNm7UNXd0xR1K4vWvX6+9BgfjDgNIvDjuaT/A3VeH4X5JB4ThgyWtyhqvL9SNVN+Xp36kZezEzM4xs6VmtnTt2rW78HYAoPJ0d3drYGBAkjQwMKCuLg4+UXnuueeenHKm1R0AgCQpaUv7aNzdzczjXIa7XybpMkmaM2dOSWMBgHLR0NCg22+/XQMDA6qpqVFjY2PcIQFFl7maZLgysCvq6uq0qapKZxzLfnM019zRpb323z/uMIDEi6Ol/flw2bvC/zWh/llJs7LGmxnqRqqfmad+pGUAAMagublZoW9PVVVVad68eTFHBBTfjBkzRiwDAJAEcbS03yzpTEnfDP9vyqo/z8yuV9Tp3AZ3X21miyV9I6vzueMlfcXdXzSzjWZ2tKQlks6Q9J+jLCMxWltb1dPTU/T5trS0jGv6+vp6zZ8/v0jRAChXqVRKTU1N6ujoUFNTk1IpHsKBylNVldt2UV1dHVMkAAAMr9SPfPuFpLslvcnM+szsbEWJ9Fwze0LSB0JZkm6R9JSklZIul/RpSQod0H1d0n3hb1GmU7owzhVhmicVdUKnEZYBABijuXPnasqUKZo7d27coQAlsWZN7oV4zz//fEyRAAAwvFL3Hv+xYV56f55xXdJnhpnPlZKuzFO/VNJb89S/kG8ZSVKM1ux8l6suWrRo3PMFAEnq6OjQli1b1NHRQc/xqEhTpkzRli1bcsoAgLHp7+/XSwMD6li/Ie5QEi09MKCB/v5xzSOOe9pRJCeeeGJO+aSTToopEgCVhue0YzLITtjzlQEASIJYe4/H+Jx99tm69dZbt5fPOuus+IIBUFF4TjsAoJz09/fr5c2bdcOjj8cdSqKt27xZr4yz1Tejrq5ONRs3au60qUWZX6XqWL9BM+rqxjUPkvYyN23aNK1fv55WdgBFle857STtQHl6ds1a/eAX18cdxnbr0uslSTNS02KNY6hn16zVn+2zb9xhAMBOSNrL3EEHHaSDDjqIVnYARcVz2oHKUF9fH3cIO3nthag/4T0SliD/2T77JnJ9YWzq6uq0futW/c1hb4w7lES74dHHNW2crb6YeCTtAICdNDc3q7OzUxLPaQfKWRIf45p5PC2d5wLA2NARHQBgJ5nntJsZz2kHAACIES3tAEqmtbVVPT09JZl3pqVmV9TX1yey9Slpmpub1dfXRys7AABAjEjaAQB5pVIpLl8FAACIGUk7gJIpVmv2Jz/5SW3cuHF7eerUqSSTAAAAmBRI2gEk3j//8z/rS1/6Uk4Zpbd48WJdfvnlOvfcczV37ty4wwGAstG/YYOuuaMr7jByvPjyy5Kk/fbeO+ZIdujfsEFv2H//uMMAEo+kHUDi1dfXq6qqSoODg5o6dapmz54dd0iTwhVXXCFJuuyyy0jaAWCMkvrYuLWbN0uS9kpQkvyG/fdP7PoCkoSkHUBZeP3rX6+nn36aVvYJsnjxYrm7JMnd1dHRQeIOAGOQ1I5OedQeSiG9bZs61m+IO4ztXtq2TZK0T3V1zJHskN62TTPGOQ+SdgBlYcqUKTrssMNoZZ8gmVb2DFrbAQBAtiReJfFyb68kaUaCjhdnaPzriqQdALCTTCv7cGUAADC5JfGqkkq9oqQq7gAAAMljZiOWAQAAMDFI2gEAO/nUpz6VUz7nnHNiigQAAGByI2kHAOzkhRdeGLEMAACAiUHSDgDYyQ033JBTbm9vjykSAACAyY2kHQAAAACAhCJpBwAAAAAgoUjaAQAAAABIKJJ2AAAAAAASiqQdAAAAAICEqok7AAAAgEK1traqp6en6PNtaWkZ1/T19fWaP39+kaIBAICWdgAAAAAAEouWdgAAUHaK0Zo9b968neoWLVo07vkCAFBMJO0AAGBSOvHEE3XrrbduL5900kkxRgNgvNZt2aIbHn087jC22/Dqq5KkqXvsEXMkO6zbskXT4g4CBSNpBwAAk9LZZ5+dk7SfddZZ8QUDYFzq6+vjDmEn63t7JUnTDjkk3kCyTFMy1xVGRtIOAAAmrWnTpmn9+vW0sgNlLokdQGY6tuS2G4wXSTsAAJi0DjroIB100EG0sgMAEoukHQAqSKkegyXxKCwAAIA48Mg3AAAAAAASipZ2AKggxWrJ5lFYAAAAyUBLOwBgJwsWLMgpn3vuuTFFAgAAMLnR0g4gRynviR6P3vDYlPHeV11slXqf9gc/+EFdfvnl28tz586NMRoAAIDJi6S9AElMZpKayEiVm8xUup6eHj2x8klN22//uEPJ4eHCoLUvvhRzJDusf3FN3CGU1AEHHKDnn3+eVnYAAIAYkbQXoKenRysee0LVU/aLO5Tttr0W/X/8mRfiDWSIbVtejDsEjMO0/fZX04dOjTuMxOu85fq4Qyip6dOna/r06bSyAwAAxIikvUDVU/bTPoceH3cYiffSE7+LOwQAKDuLFy/W5ZdfrnPPPZeTJQAAQBId0QEAkBhXXHGFJOmyyy6LORIAAJAUJO0AACTA4sWL5e6SJHdXR0dHzBEBAIAkIGkHACABMq3sGbS2AwAAiaQdAIBEyLSyD1cGAACTE0k7AAAAAAAJRe/xAACMQ2trq3p6ekoy75aWlnFNX19fr/nz5xcpGgAAEAda2gEAAAAASCha2gHk6O/v10svbVLnLdfHHUrirX9hjbZt3RR3GIhZsVqyb7zxRl133XXby6effrpOPvnkosw7KUp5VcKu6u3tlTT+qxpKgSslAAASSTsAAInwkY98JCdpr7SEXZJ6enrU29OjQ2bOjDuU7fbYfXdJ0uBrr8UcSa5n+vriDgEAkBAk7QBy1NXVqXr3l9T0oVPjDiXxOm+5XrX77RN3GKggM2bM0Lp163T66afHHUrJHDJzpv7pCxfEHUbifeO7F8cdAgAgIUjaAQBIiP3331/7779/RbayAwCAXUNHdAAAAAAAJBRJOwAAAAAACUXSDgAAAABAQnFPO4CdrH9xTeIe+fbyxrQkae99UzFHssP6F9fQER0AAABKiqQdQI76+vq4Q8hr08YXJClRSXLtfvsUbX3x/Oqx49nVAABgMiFpL0B/f78GNr+kl574XdyhJN7A5hfV35+sZ95ibJKaDGUSx0WLFsUcSWn09PToySef0gF1B8YdynbV1dFPxMubtsQcyQ7P96+OOwSMQ39/v7Zs2cLjzMbg6b4+TZkyJe4wAAAJQNIOAAlxQN2BOu3Mc+MOI9Guu/oncYcAAAAwoUjaC1BXV6eNW3fTPoceH3coiffSE79TXd30uMMAgGFxS8LYFeuWhLq6Og2+9pr+6QsXFCGqyvaN716sqt12izsMAEACkLQDACalnp4e9fT0aNasQ+IOZbvdd99DkjQwsC3mSHZYteqZuEMAgAlTzBO6xTwRS38ukxtJOwBg0po16xB9+UtfjjuMRPv2d74ddwgAUJZe97rXxR0CKgRJOwAAmDDP9PUlqiO659eulSQdUFsbcyS5nunr0+yEPs0DqGS0ZiOJSNoLtG3Li4nqPX7bqy9Jkqr3SM5jsKRoPUnc0w4A2CGJj5R8detWSUrc/eOz6+sTub4AABOPpL0ASfzx7O2NkvbZhyQtQZ6eyPUFAIhPEluwKv1xkgCA8kfSXgAONgCgcmSeGc492yNbteoZnhcOAECMquIOAAAAAAAA5EdLOwAkQH9/vzZt2qzrrv5J3KEk2vP9q/XyXnsWZV51dXUaGNhG7/Gj+PZ3vq2amuq4wwAAYNIiaQcAAMCoivX86mI+u1ri+dVAuSvWvkWq3P0LSTsAJEBdXZ1e3rRFp515btyhJNp1V/9Ee+9VvPurV616JlH3tK9Zs0aStP/++8ccyQ6rVj1Dx6IoKp5dDaBUKnX/QtIOAJiUkpiIbt36qiQl6nL0eh49hiAJrU2TTVJbIJPS+ojKwLY0OpJ2TBr88CHpnu9fnah72tMvviBJSu2XnEdKPt+/Wnu/4c+KMq8kfu94IgiAUqnUFkhgMiBpB3YBP3wotiS2ZK5bOyBJRb0cfbz2fsOfJXJdAUApJPHkIoCJR9KOSYMfPiRZErdPWn0BAADix3PaAQAAAABIqIpO2s3sBDN7zMxWmtmFcccDAAAAAEAhKvbyeDOrlvRfkuZK6pN0n5nd7O6PxBsZMHnQ+R+AUuGZ4QCAyaJik3ZJR0pa6e5PSZKZXS/pZEmJSNo52AAKQ+d/Y5PUEyVS5e5fWOfljX0LACDpKjlpP1jSqqxyn6Sjho5kZudIOkeSDjnkkImJrIg42ECSkSyUN/YvE491PnbsXwAAk4W5e9wxlISZzZN0grt/KpRPl3SUu5833DRz5szxpUuXTlSIAAAAAABIkszsfnefM7S+kjuie1bSrKzyzFAHAAAAAEBZqOSk/T5Jh5pZvZntLulUSTfHHBMAAAAAAGNWsfe0u/uAmZ0nabGkaklXuvvDMYcFAAAAAMCYVWzSLknufoukW+KOAwAAAACAXVHJl8cDAAAAAFDWSNoBAAAAAEgoknYAAAAAABKKpB0AAAAAgIQiaQcAAAAAIKFI2gEAAAAASCiSdgAAAAAAEoqkHQAAAACAhCJpBwAAAAAgoUjaAQAAAABIKJJ2AAAAAAASiqQdAAAAAICEImkHAAAAACChSNoBAAAAAEgoknYAAAAAABLK3D3uGBLDzNZKejruOHbBDEnr4g5iEmK9TzzW+cRjnU881vnEY51PPNZ5PFjvE491PvHKeZ2/3t1rh1aStFcAM1vq7nPijmOyYb1PPNb5xGOdTzzW+cRjnU881nk8WO8Tj3U+8SpxnXN5PAAAAAAACUXSDgAAAABAQpG0V4bL4g5gkmK9TzzW+cRjnU881vnEY51PPNZ5PFjvE491PvEqbp1zTzsAAAAAAAlFSzsAAAAAAAlF0g4AAAAAQEKRtCeMmbmZXZtVrjGztWb2m1Gmm2Nm3y99hOXJzLaZ2Z/M7CEzazOzPUcY91gze88Y5jmm8SrVkHX6azObFuoPMrP2mMPLYWYXmdkX89TPNrOH4oipFMzsq2b2sJktC5/NUUWYZ951h+KsbzN7uRSxTTZmdomZfT6rvNjMrsgqf9fMLhjtt7SA5Z1iZm8pxrzKiZnVmdn1Zvakmd1vZreY2RvjjiuJSrE/HmY5RTsWCb+JHy/WeOOMJe9x23j2mWZ2lpkdNIbxFpnZB3Z1OcVSyHs1s2lm9ums8rHF2t/FzSJ3mdmJWXXNZvbbOOOaaCTtybNJ0lvNbEooz5X07GgTuftSd/9cSSMrb1vc/XB3f6ukrZL+foRxj5U0lh/AsY5XqbLX6YuSPiNJ7v6cu8+LN7TJx8zeLekkSe9097dJ+oCkVWOctrqUsVWi8axvlMTvFfbHZlYlaYakv8h6/T2Sdi/i8k6RNKmSdjMzSTdKusPd3+DuR0j6iqQD4o0seSZ4/3CsincsMlvSWJLxsY43HoUct43VWZJGTdrdvcXd/68Iy5tI0yR9erSRxsrMaoo1r/HyqAO2v5d0sZm9zsz2lvQNhePOyYKkPZlukfThMPwxSb/IvGBmR5rZ3Wb2RzP7g5m9KdRvP6MWznz/KfxtMLMzzazazL5jZveFs77nTvi7So5uSX9uZvuZ2f+E9XGPmb3NzGYr2jGcH9Zfg5n9lZktCev8/8zsgGHGm21mt4f53WZmh0iSmdWa2X+HdX+fmb031L8v63P6o5ntE9P6KIa7JR0s5bZeh+3uP8KZ8mVm9tlQf4SZ3Rlaahab2YEjzdyi1t6fhW3/CTNbEOr3Duv6ATNbbmYnZ03zVTN73MzukvSmrPojzOxBM3tQWTt8M+sys8OzyneZ2duLsXImyIGS1rn7q5Lk7uvc/Tkze3/Yvpab2ZVmtockmVmvmX3LzB6Q1GxmHzKzR8Nn8v0hZ+jfYmZ3mNlTZva5MH3OVQpm9kUzuygM32FRy+dSM1thZu8ysxvCZ/evE7VCSmzM69vMjjOz/8lMaGZzzezGrPIlFrXI3WZmtaHuDWb22/B5dJvZYaF+p/1RqL8oLG/o57SXmf1v2OYfMrO/m7hVNKH+IOndYfgvJD0k6SUzS4Vt/s2SHpC0t5m1h239OjMzafh9kpktCPvtB8N+fE+LWjX/WtJ3wv77DRP+buPRJOk1d/9xpsLdH5T0x3z74eG2PTP7ppk9En4T/iPU5fudrQr7qWmZ5YV9yAHDfQ8SZLj9w7vM7AZJMrOTzWyLme1uUSLyVKgf7ru/07GE5TkWyQ7Chv/tNIuOCR8Kn1lmv/BNSQ1hXudbtJ/vDp/tA7ajRX/oeK8zs9Ywrz+aWVNYTt5jTzM70KLf3ExLek7ceXRL+vMh7y3v73+IeYWZXW7RfvV3ZjbFzOZJmiPpurDcKWbWEmJ7yMwuy9ofXBXGz/xW/kvWcjKfx6jHcBYdY94f4jgnq/5lM/u38N24x3bsx+vDZ7XcRvittOiqoYfC3+ezPpM3hHi+E+oK3d/dYWbfM7OlkhZa1Jr9UIiza5TPqKTc/SFJv5b0j5JaJF0r6VrbOR86K6z3jvDZnRfW1x/Dut4vjPc527Efuj62N1YId+cvQX+SXpb0Nkntkl4n6U+KzqL+Jry+r6SaMPwBSf8dhrePkzWvIyQtkzRV0jmS/l+o30PSUkn1cb/fiVyv4X+NpJsk/YOk/5T0tVB/nKQ/heGLJH0xa9qUdjxp4VOSvjvMeL+WdGYY/qSk/wnDP5d0TBg+RNKKrPHfG4b3znyu5fKXtU6rJbVJOiGUZ0t6KAz/Q9iWM9vsfpJ2U3SAXRvq/k7SlaMs6yJJD0qaoqgFbZWis+U1kvYN48yQtFKShW1/uaQ9w3dmZeazCt+JxjD8naxYz5T0vTD8RklL417HBX4ee4f9xeOSfijpfYr2IaskvTGMc42kz4fhXklfDsOZ8epD+Rfasc+5KHxee4R1/EL4DLd/zmG8L0q6KAzfIelbYXihpOcUHcTuIalP0vS419dEru+wTT6atc3/XNJfhWGXdFoYbpH0gzB8m6RDw/BRkm4PwyPtj/J9Tn8r6fKsuKfGve5K+Jn0KNrHnqsokfm6pA9Jeq+ig/5jJW2QNFNRo8Xdko7RCPuk7G1V0r9K+mwYvkrSvLjf8wSv389JuiRP/XD74Z22PUnTJT2WtQ1PC/+H264vlTQ/DB8l6f9GGj8pf/n2D1nr6qkw/B+S7gvb5/sk/SLUD/fdH+5Y4iJlHYsMieMi5f/t/FtJHYp+vw+Q9IyiffSxyjqWVPQb+rowfKjC72Ke8b6Q9Z05LMzvdRrm2DOM/9VQXy1pnzyx73Tclqc+33Y3W9KApMPDa7+S9IkwfIekOVnL2C9r+GfasV++SuH7rei3MvO9/7SkK8LwqMdwmfmH9f+Qwv5E0X4/s6xvZ62jmyWdEYY/k3mvQ+aZOb7ZKyz3YUnv0M6/yceq8P3dHZJ+mDWP5ZIOzv6uxvy92kvR/mO5ov1JvnzorLAt7COpNqyDvw+vXaIdx0DPSdojKe9tLH+JufQBO7j7MovOnn5MUat7tqmSrjazQxV96XfLNw8zm6FoB/RRd99gZsdLelvmzGGYz6GKDnImgylm9qcw3C3pp5KWKPrhkrvfbmbTzWzfPNPOlPTLcCZydw2/zt4t6W/C8M8U7YilaGfylnCCU5L2tejSnt8rutTnOkk3uHvfrr65mGTW6cGSVig6ABjqA5J+7O4DkuTuL5rZWyW9VVJHWCfVklaPYXk3ufsWSVvMrFPSkZL+V9I3zKxR0mCI5QBJDZJudPfNkmRmN4f/0xTtnDNnjH8mKXOPVJukfzazLyk66XLV2FZDMrj7y2Z2hKL33iTpl5L+XVKPuz8eRrta0YHA90L5l+H/YYoOJDPb9i8UHWxl/K9HLUavmtkaje1y2JvD/+WSHnb31ZJkUWvSLEVJZdkqZH27+/fM7GeSPmFmrYr2FWeEcQa143O4VtINYf/wHkltWfuNPcL/kfZH+T6n5ZK+a2bfUnSQ3V28tZA4f1C03t4j6WJF+4P3KDpo+30Y597Mvjbsv2ZLWq/h90lvDS1e0xQdIC8u+bsoP6b8++Gdtj2LLrl9RdJPLbqaJ3NFz3Db9S8VncxqlXSqdnxXxvq7HIt8+wczu9Ddr7KoP4A3K/oNu1hSo6JtrnuU7/5wxxKjyffbeYyikwTbJD1vZndKepekjUOm3U3SDyy6Cm2bohPa+RyjqCFE7v6omT0dxh3u2PM+SVea2W6KGjj+lGee+Y7bsg233UnRfjgz7f2Kvuf5NJnZlxWdnNhPUQL86zzj3ZA1r8xx3liO4T5nZh8Jw7MUvfcXFF3un9n271d0K6wUncD52zD8M0nfyjPPYxQd32ySJIuu3GjQjt/cbIXu76Qd37HMe7zKzH6VtQ5i4+6bzOyXiho491UUW758qNPdX1J0tdUG7fhMlytqGJWiBpzrLLoK7n8mIPxxI2lPrpsVnYU9VtGZ6YyvK9oYPxIS+zuGTmjR/anXS1rk0eUkUrRz+6y7T9YDji3ufnh2RdYP32j+U9LF7n6zmR2r6Mx1IaokHe3urwyp/6aZ/a+ilqDfm9kH3f3RAucdpy3ufrhFncMsVpQMjqUzRFOUxL171DFzeZ7yaYrOpB7h7q+ZWa+is/sFc/fNZtYh6WRJH1V0NrushAOwOyTdYWbLNfr9XpvGOOtXs4a3KfrtGFDuLVZD13tmmsEh0w+qQn57ClzfrYoOHF6R1JY5kZVvtorW6/qh+6xgpP3RTp+Tuz9uZu9UtJ/5VzO7zd0XjfLWylXmvva/VNSqtUpRi95GRetfyr8tj7RPukrSKe7+oJmdpeg3ebJ6WFK+/kry7oeH2/bM7EhJ7w/zOk/RlW7Dbdd3K7qdrVZRPwKZS4bH+7tccnn2D2cq2p66FJ0sfk3S/4W6aklf0sjf/bzHEmM4lsn32zlW50t6XtLbw/KHHseMZthjz5Bsf1hR4nWxu18zZJSdjtuGGOn3f+j3fMqQaWVmr1N0FcQcd19l0e1dwx0/ZOaX2WfI3Uc8hgvb5QckvTscX9yRNf/XPDTxZs8zKOTzGU2h+zsp67jA3f/eog4UPyzpfjM7wt3jPuE+GP5GyoeGHnNkH49k1vWHFZ0w+ytJXzWzvxzhdzkRuKc9ua6U9C/uvnxI/VTt6JjurGGm/aakZe6efY/GYkn/EM5qyszeaGZ7FTHectStaKef2bmuc/eNkl5SdFlNRvY6PzOrfuh4f1DUEqAw30yL1u8kfTYzUjhjLTN7g7svd/dvKTrrfNh431AcQmv25yR9wXbuuKRD0rmZ+nAv0WOSai3qqEdmtpuZ/UUYPs/MzhtmUSdbdN/cdEUHzvcp+mzWhB/sJkmvD+N2STrFonvW9lG0U5a7r5e03syOCeOdNmQZVyg68XCfu6cLXRdxMrM3hTPOGYdLelLSbDPL3At4uqQ780z+mKQ/Cz98UnS53Giel7R/uEJlD0WdLk0aha5vd39O0eV4/087Ekgp+h3OJEIfl3RX2A/1mFlzWJbZjv4VhtsfDRfnQZI2u/u1im4HeWch77PM/EHRdviiu29z9xcVtZC/O7w2nGH3SYr28avDb2f2/mLo/n8yuF3SHpZ7b+7bFO13d9oP59v2QsvwVHe/RVFCOOJ2HRKbGxW1SK/IShgK+h5MtGH2D0+H4W5Ft83c7e5rFTXMvEnRpc0jfffzHkto9G0x329nt6S/s+ie81pFycu9eeY1VdJqdx9UtD/LdFo6dLzs46k3Krp8/zENc+xpZq+X9Ly7X67od3dX9kvD/f6PJDvuTAK9LmyXBXWgO4ZjuKmS0iFhP0zS0WOY7e+VexyZT7ei45s9w3H8R0LdWPdJI+3vcoT3uMTdWyStVXS1QFKMJR/Ky6LOSme5e6eie+SnKrqSKtFI2hPK3fvcPV+r5bcl/buZ/VHDt1Z9UdLxtqODjL9WtFN8RNIDFnUe9ZMRpp8sLpJ0hJktU3SiI/PD/2tJH7EdnbpcpOhStfslrcuafuh4n5U0P8zvdEX38kpRQjvHos4uHtGOHlA/b6GDNkVn3G8t1RstNXf/o6JLjT425KUrFN3btsyijt8+7u5bFf04fivU/Uk7er49TMNfNr1MUqekeyR9PSRB1ylat8sVXW78aIjnAUWXeD2oaL3elzWf+ZL+y6JLxXKaKNz9fuW2ypWTvRXdOvNI2KbeIulCRe+3LayjQUk/HjphuHTy05J+G7bzlxRdUjwsd39N0iJFB3odCut+EtmV9X2dpFXuviKrbpOkI8N++ThF61SKDtjODt+RhxVdASINvz8azl9Kujds71/TjpbKSrRc0b2t9wyp2+Duw66rUfZJ/6zoVqrfK3cbv17Slyzq3GhSdEQXEuiPSPqARZd4P6zolpBblGc/rPzb3j6SfhO+M3dJuiCMe5GG365/KekTyr1sd6TxkyDf/uGi8NoSRZdxZ27TWiZpeVbL63Df/eGOJYYeiwyV77fzxlD/oKKTMV929/5Qt82ijsfOV9QSfWaI5TDtaIXNN15V2AZ+KemscKvOcMeex0p6MBzL/p2ivgsKlff3fxRXSfpx2CZflXS5oqtyFiv3OGEsRjuG+62kGjNboegY856hM8hjoaTPhPd0cL4RwvHNVYp+e5cousf+j+GE1u9DTN/JN22YfqT93VDfsahTvIcUnfh8cAzvYaKMJR8aTrWiTuyWS/qjpO+HRp1Esx37CACIl0X3OP5N+FHJrr9IUYcs/1Hi5R+k6BKrw0LLwqRhZnuH+zBN0n9JesLdL4k7rkpiZj+Q9Ed3H3pvJgAU3UT9dgIoPVraASSGu580NGGfKGZ2hqKz1l+dbAl7sCC0Pjys6FKxn8QbTmUJLYJvU9TZHAAAwJjR0g4AAAAAQELR0g4AAAAAQEKRtAMAAAAAkFAk7QAAAAAAJBRJOwAAGJaZvRz+zzazj8cdDwAAkw1JOwAAGIvZkkjaAQCYYCTtAABgLL4pqcHM/mRm55tZtZl9x8zuM7NlZnauJJnZsWZ2p5ndZGZPmdk3zew0M7vXzJab2Rtifh8AAJSVmrgDAAAAZeFCSV9095MkyczOkbTB3d9lZntI+r2Z/S6M+3ZJb5b0oqSnJF3h7kea2UJJn5X0+QmPHgCAMkXSDgAAdsXxkt5mZvNCeaqkQyVtlXSfu6+WJDN7UlImmV8uqWmiAwUAoJyRtAMAgF1hkj7r7otzKs2OlfRqVtVgVnlQHHsAAFAQ7mkHAABj8ZKkfbLKiyX9g5ntJklm9kYz2yuWyAAAqGCc7QYAAGOxTNI2M3tQ0lWSLlXUo/wDZmaS1ko6Ja7gAACoVObucccAAAAAAADy4PJ4AAAAAAASiqQdAAAAAICEImkHAAAAACChSNoBAAAAAEgoknYAAAAAABKKpB0AAAAAgIQiaQcAAAAAIKH+P7xbrK/422OCAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1202.4x595.44 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Boxplot that shows yield for each item \n",
    "a4_dims = (16.7, 8.27)\n",
    "\n",
    "fig, ax = plt.subplots(figsize=a4_dims)\n",
    "sns.boxplot(x=\"Item\",y=\"hg/ha_yield\",palette=\"vlag\",data=yield_df,ax=ax);"
   ]
  }
 ],
 "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": "3.7.12"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 174.772109,
   "end_time": "2021-12-10T09:19:03.767946",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2021-12-10T09:16:08.995837",
   "version": "2.3.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}