LinearRegressionPerCrop.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "efccd121",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn import metrics\n",
    "\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "8eca6466",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>area</th>\n",
       "      <th>year</th>\n",
       "      <th>crop</th>\n",
       "      <th>yield</th>\n",
       "      <th>yield_1</th>\n",
       "      <th>yield_2</th>\n",
       "      <th>yield_3</th>\n",
       "      <th>yield_4</th>\n",
       "      <th>yield_5</th>\n",
       "      <th>avg_temp</th>\n",
       "      <th>...</th>\n",
       "      <th>tonnes_pesticide</th>\n",
       "      <th>tonnes_pesticide_1</th>\n",
       "      <th>tonnes_pesticide_2</th>\n",
       "      <th>tonnes_pesticide_3</th>\n",
       "      <th>tonnes_nitrogen</th>\n",
       "      <th>tonnes_nitrogen_1</th>\n",
       "      <th>tonnes_nitrogen_2</th>\n",
       "      <th>tonnes_nitrogen_3</th>\n",
       "      <th>arable_land</th>\n",
       "      <th>ag_land</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Albania</td>\n",
       "      <td>2008</td>\n",
       "      <td>Maize</td>\n",
       "      <td>50000</td>\n",
       "      <td>46732</td>\n",
       "      <td>50111</td>\n",
       "      <td>45442</td>\n",
       "      <td>46197</td>\n",
       "      <td>42805</td>\n",
       "      <td>16.71</td>\n",
       "      <td>...</td>\n",
       "      <td>1069.54</td>\n",
       "      <td>1006.57</td>\n",
       "      <td>943.61</td>\n",
       "      <td>880.64</td>\n",
       "      <td>28744.80</td>\n",
       "      <td>31483.82</td>\n",
       "      <td>30333.88</td>\n",
       "      <td>37210.64</td>\n",
       "      <td>610.0</td>\n",
       "      <td>1181.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Albania</td>\n",
       "      <td>2008</td>\n",
       "      <td>Potatoes</td>\n",
       "      <td>193878</td>\n",
       "      <td>188902</td>\n",
       "      <td>170745</td>\n",
       "      <td>167061</td>\n",
       "      <td>149346</td>\n",
       "      <td>151301</td>\n",
       "      <td>16.71</td>\n",
       "      <td>...</td>\n",
       "      <td>1069.54</td>\n",
       "      <td>1006.57</td>\n",
       "      <td>943.61</td>\n",
       "      <td>880.64</td>\n",
       "      <td>28744.80</td>\n",
       "      <td>31483.82</td>\n",
       "      <td>30333.88</td>\n",
       "      <td>37210.64</td>\n",
       "      <td>610.0</td>\n",
       "      <td>1181.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Albania</td>\n",
       "      <td>2008</td>\n",
       "      <td>Wheat</td>\n",
       "      <td>40168</td>\n",
       "      <td>35496</td>\n",
       "      <td>30787</td>\n",
       "      <td>31582</td>\n",
       "      <td>31130</td>\n",
       "      <td>28629</td>\n",
       "      <td>16.71</td>\n",
       "      <td>...</td>\n",
       "      <td>1069.54</td>\n",
       "      <td>1006.57</td>\n",
       "      <td>943.61</td>\n",
       "      <td>880.64</td>\n",
       "      <td>28744.80</td>\n",
       "      <td>31483.82</td>\n",
       "      <td>30333.88</td>\n",
       "      <td>37210.64</td>\n",
       "      <td>610.0</td>\n",
       "      <td>1181.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Albania</td>\n",
       "      <td>2008</td>\n",
       "      <td>Soybeans</td>\n",
       "      <td>20000</td>\n",
       "      <td>20000</td>\n",
       "      <td>20000</td>\n",
       "      <td>17500</td>\n",
       "      <td>13333</td>\n",
       "      <td>12107</td>\n",
       "      <td>16.71</td>\n",
       "      <td>...</td>\n",
       "      <td>1069.54</td>\n",
       "      <td>1006.57</td>\n",
       "      <td>943.61</td>\n",
       "      <td>880.64</td>\n",
       "      <td>28744.80</td>\n",
       "      <td>31483.82</td>\n",
       "      <td>30333.88</td>\n",
       "      <td>37210.64</td>\n",
       "      <td>610.0</td>\n",
       "      <td>1181.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Albania</td>\n",
       "      <td>2009</td>\n",
       "      <td>Maize</td>\n",
       "      <td>55693</td>\n",
       "      <td>50000</td>\n",
       "      <td>46732</td>\n",
       "      <td>50111</td>\n",
       "      <td>45442</td>\n",
       "      <td>46197</td>\n",
       "      <td>16.73</td>\n",
       "      <td>...</td>\n",
       "      <td>1132.50</td>\n",
       "      <td>1069.54</td>\n",
       "      <td>1006.57</td>\n",
       "      <td>943.61</td>\n",
       "      <td>33940.18</td>\n",
       "      <td>28744.80</td>\n",
       "      <td>31483.82</td>\n",
       "      <td>30333.88</td>\n",
       "      <td>609.0</td>\n",
       "      <td>1201.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3823</th>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>2013</td>\n",
       "      <td>Wheat</td>\n",
       "      <td>22888</td>\n",
       "      <td>24420</td>\n",
       "      <td>26274</td>\n",
       "      <td>27681</td>\n",
       "      <td>30000</td>\n",
       "      <td>30097</td>\n",
       "      <td>19.76</td>\n",
       "      <td>...</td>\n",
       "      <td>2550.07</td>\n",
       "      <td>3375.53</td>\n",
       "      <td>3340.35</td>\n",
       "      <td>3305.17</td>\n",
       "      <td>51800.00</td>\n",
       "      <td>41000.00</td>\n",
       "      <td>59563.00</td>\n",
       "      <td>70502.00</td>\n",
       "      <td>4000.0</td>\n",
       "      <td>16200.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3824</th>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>2013</td>\n",
       "      <td>Sorghum</td>\n",
       "      <td>3066</td>\n",
       "      <td>2046</td>\n",
       "      <td>2267</td>\n",
       "      <td>2702</td>\n",
       "      <td>3230</td>\n",
       "      <td>2577</td>\n",
       "      <td>19.76</td>\n",
       "      <td>...</td>\n",
       "      <td>2550.07</td>\n",
       "      <td>3375.53</td>\n",
       "      <td>3340.35</td>\n",
       "      <td>3305.17</td>\n",
       "      <td>51800.00</td>\n",
       "      <td>41000.00</td>\n",
       "      <td>59563.00</td>\n",
       "      <td>70502.00</td>\n",
       "      <td>4000.0</td>\n",
       "      <td>16200.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3825</th>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>2013</td>\n",
       "      <td>Soybeans</td>\n",
       "      <td>13142</td>\n",
       "      <td>15300</td>\n",
       "      <td>12054</td>\n",
       "      <td>13557</td>\n",
       "      <td>15000</td>\n",
       "      <td>16129</td>\n",
       "      <td>19.76</td>\n",
       "      <td>...</td>\n",
       "      <td>2550.07</td>\n",
       "      <td>3375.53</td>\n",
       "      <td>3340.35</td>\n",
       "      <td>3305.17</td>\n",
       "      <td>51800.00</td>\n",
       "      <td>41000.00</td>\n",
       "      <td>59563.00</td>\n",
       "      <td>70502.00</td>\n",
       "      <td>4000.0</td>\n",
       "      <td>16200.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3826</th>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>2013</td>\n",
       "      <td>Cassava</td>\n",
       "      <td>46000</td>\n",
       "      <td>47010</td>\n",
       "      <td>45589</td>\n",
       "      <td>45152</td>\n",
       "      <td>44708</td>\n",
       "      <td>44259</td>\n",
       "      <td>19.76</td>\n",
       "      <td>...</td>\n",
       "      <td>2550.07</td>\n",
       "      <td>3375.53</td>\n",
       "      <td>3340.35</td>\n",
       "      <td>3305.17</td>\n",
       "      <td>51800.00</td>\n",
       "      <td>41000.00</td>\n",
       "      <td>59563.00</td>\n",
       "      <td>70502.00</td>\n",
       "      <td>4000.0</td>\n",
       "      <td>16200.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3827</th>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>2013</td>\n",
       "      <td>Sweet potatoes</td>\n",
       "      <td>22222</td>\n",
       "      <td>22222</td>\n",
       "      <td>21953</td>\n",
       "      <td>21905</td>\n",
       "      <td>21854</td>\n",
       "      <td>21798</td>\n",
       "      <td>19.76</td>\n",
       "      <td>...</td>\n",
       "      <td>2550.07</td>\n",
       "      <td>3375.53</td>\n",
       "      <td>3340.35</td>\n",
       "      <td>3305.17</td>\n",
       "      <td>51800.00</td>\n",
       "      <td>41000.00</td>\n",
       "      <td>59563.00</td>\n",
       "      <td>70502.00</td>\n",
       "      <td>4000.0</td>\n",
       "      <td>16200.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3828 rows × 35 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          area  year            crop   yield  yield_1  yield_2  yield_3  \\\n",
       "0      Albania  2008           Maize   50000    46732    50111    45442   \n",
       "1      Albania  2008        Potatoes  193878   188902   170745   167061   \n",
       "2      Albania  2008           Wheat   40168    35496    30787    31582   \n",
       "3      Albania  2008        Soybeans   20000    20000    20000    17500   \n",
       "4      Albania  2009           Maize   55693    50000    46732    50111   \n",
       "...        ...   ...             ...     ...      ...      ...      ...   \n",
       "3823  Zimbabwe  2013           Wheat   22888    24420    26274    27681   \n",
       "3824  Zimbabwe  2013         Sorghum    3066     2046     2267     2702   \n",
       "3825  Zimbabwe  2013        Soybeans   13142    15300    12054    13557   \n",
       "3826  Zimbabwe  2013         Cassava   46000    47010    45589    45152   \n",
       "3827  Zimbabwe  2013  Sweet potatoes   22222    22222    21953    21905   \n",
       "\n",
       "      yield_4  yield_5  avg_temp  ...  tonnes_pesticide  tonnes_pesticide_1  \\\n",
       "0       46197    42805     16.71  ...           1069.54             1006.57   \n",
       "1      149346   151301     16.71  ...           1069.54             1006.57   \n",
       "2       31130    28629     16.71  ...           1069.54             1006.57   \n",
       "3       13333    12107     16.71  ...           1069.54             1006.57   \n",
       "4       45442    46197     16.73  ...           1132.50             1069.54   \n",
       "...       ...      ...       ...  ...               ...                 ...   \n",
       "3823    30000    30097     19.76  ...           2550.07             3375.53   \n",
       "3824     3230     2577     19.76  ...           2550.07             3375.53   \n",
       "3825    15000    16129     19.76  ...           2550.07             3375.53   \n",
       "3826    44708    44259     19.76  ...           2550.07             3375.53   \n",
       "3827    21854    21798     19.76  ...           2550.07             3375.53   \n",
       "\n",
       "      tonnes_pesticide_2  tonnes_pesticide_3  tonnes_nitrogen  \\\n",
       "0                 943.61              880.64         28744.80   \n",
       "1                 943.61              880.64         28744.80   \n",
       "2                 943.61              880.64         28744.80   \n",
       "3                 943.61              880.64         28744.80   \n",
       "4                1006.57              943.61         33940.18   \n",
       "...                  ...                 ...              ...   \n",
       "3823             3340.35             3305.17         51800.00   \n",
       "3824             3340.35             3305.17         51800.00   \n",
       "3825             3340.35             3305.17         51800.00   \n",
       "3826             3340.35             3305.17         51800.00   \n",
       "3827             3340.35             3305.17         51800.00   \n",
       "\n",
       "      tonnes_nitrogen_1  tonnes_nitrogen_2  tonnes_nitrogen_3  arable_land  \\\n",
       "0              31483.82           30333.88           37210.64        610.0   \n",
       "1              31483.82           30333.88           37210.64        610.0   \n",
       "2              31483.82           30333.88           37210.64        610.0   \n",
       "3              31483.82           30333.88           37210.64        610.0   \n",
       "4              28744.80           31483.82           30333.88        609.0   \n",
       "...                 ...                ...                ...          ...   \n",
       "3823           41000.00           59563.00           70502.00       4000.0   \n",
       "3824           41000.00           59563.00           70502.00       4000.0   \n",
       "3825           41000.00           59563.00           70502.00       4000.0   \n",
       "3826           41000.00           59563.00           70502.00       4000.0   \n",
       "3827           41000.00           59563.00           70502.00       4000.0   \n",
       "\n",
       "      ag_land  \n",
       "0      1181.0  \n",
       "1      1181.0  \n",
       "2      1181.0  \n",
       "3      1181.0  \n",
       "4      1201.3  \n",
       "...       ...  \n",
       "3823  16200.0  \n",
       "3824  16200.0  \n",
       "3825  16200.0  \n",
       "3826  16200.0  \n",
       "3827  16200.0  \n",
       "\n",
       "[3828 rows x 35 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"processed_data/total_yields.csv\")\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "b88e31df",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['area', 'year', 'crop', 'yield', 'yield_1', 'yield_2', 'yield_3',\n",
       "       'yield_4', 'yield_5', 'avg_temp', 'avg_temp_1', 'avg_temp_2',\n",
       "       'avg_temp_3', 'avg_temp_4', 'avg_temp_5', 'latitude', 'longitude',\n",
       "       'tonnes_potash', 'tonnes_potash_1', 'tonnes_potash_2',\n",
       "       'tonnes_potash_3', 'tonnes_phosph', 'tonnes_phosph_1',\n",
       "       'tonnes_phosph_2', 'tonnes_phosph_3', 'tonnes_pesticide',\n",
       "       'tonnes_pesticide_1', 'tonnes_pesticide_2', 'tonnes_pesticide_3',\n",
       "       'tonnes_nitrogen', 'tonnes_nitrogen_1', 'tonnes_nitrogen_2',\n",
       "       'tonnes_nitrogen_3', 'arable_land', 'ag_land'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "7f515de5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Albania', 'Algeria', 'Angola', 'Argentina', 'Armenia',\n",
       "       'Australia', 'Austria', 'Azerbaijan', 'Bahrain', 'Bangladesh',\n",
       "       'Belarus', 'Belgium', 'Bolivia', 'Brazil', 'Bulgaria',\n",
       "       'Burkina Faso', 'Burundi', 'Cameroon', 'Canada',\n",
       "       'Central African Republic', 'Chile', 'China', 'Taiwan', 'Colombia',\n",
       "       \"Côte d'Ivoire\", 'Croatia', 'Czechia',\n",
       "       'Democratic Republic of the Congo', 'Denmark',\n",
       "       'Dominican Republic', 'Ecuador', 'Egypt', 'El Salvador', 'Estonia',\n",
       "       'Finland', 'France', 'Germany', 'Ghana', 'Greece', 'Guatemala',\n",
       "       'Guinea', 'Honduras', 'Hungary', 'India', 'Indonesia', 'Iran',\n",
       "       'Iraq', 'Ireland', 'Italy', 'Jamaica', 'Japan', 'Kazakhstan',\n",
       "       'Kenya', 'Latvia', 'Lebanon', 'Libya', 'Lithuania', 'Madagascar',\n",
       "       'Malawi', 'Malaysia', 'Mali', 'Mauritius', 'Mexico', 'Morocco',\n",
       "       'Mozambique', 'Namibia', 'Netherlands', 'New Zealand', 'Nicaragua',\n",
       "       'Niger', 'North Macedonia', 'Norway', 'Pakistan',\n",
       "       'Papua New Guinea', 'Peru', 'Poland', 'Portugal', 'South Korea',\n",
       "       'Moldova', 'Romania', 'Russia', 'Saudi Arabia', 'Senegal',\n",
       "       'Slovakia', 'Slovenia', 'South Africa', 'Spain', 'Sri Lanka',\n",
       "       'Suriname', 'Sweden', 'Switzerland', 'Syria', 'Tajikistan',\n",
       "       'Thailand', 'Tunisia', 'Turkey', 'Uganda', 'Ukraine',\n",
       "       'United Kingdom', 'Tanzania', 'United States', 'Uruguay',\n",
       "       'Venezuela', 'Vietnam', 'Zambia', 'Zimbabwe'], dtype=object)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "countries = df[\"area\"].unique()\n",
    "countries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "60fe82fd",
   "metadata": {},
   "outputs": [],
   "source": [
    "x = df[df['area'] == 'Albania'][['yield_1', 'yield_2', 'yield_3', 'yield_4', 'yield_5', \\\n",
    "                   'avg_temp', 'avg_temp_1','avg_temp_2','avg_temp_3', 'avg_temp_4', 'avg_temp_5', \\\n",
    "                   'tonnes_potash', 'tonnes_potash_1', 'tonnes_potash_2', 'tonnes_potash_3', \\\n",
    "                   'tonnes_phosph', 'tonnes_phosph_1','tonnes_phosph_2', 'tonnes_phosph_3', \\\n",
    "                   'tonnes_pesticide', 'tonnes_pesticide_1', 'tonnes_pesticide_2', 'tonnes_pesticide_3', \\\n",
    "                  'tonnes_nitrogen', 'tonnes_nitrogen_1', 'tonnes_nitrogen_2', 'tonnes_nitrogen_3', \\\n",
    "                  'arable_land', 'ag_land']]\n",
    "y = df[df['area'] == 'Albania']['yield']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "047c47f0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>area</th>\n",
       "      <th>crop</th>\n",
       "      <th>lat</th>\n",
       "      <th>long</th>\n",
       "      <th>yield_2013</th>\n",
       "      <th>yield_2013_pred</th>\n",
       "      <th>yield_2013_diff</th>\n",
       "      <th>perc_err</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Maize</td>\n",
       "      <td>41.153332</td>\n",
       "      <td>20.168331</td>\n",
       "      <td>69533</td>\n",
       "      <td>66770.792339</td>\n",
       "      <td>2762.207661</td>\n",
       "      <td>3.972513</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Potatoes</td>\n",
       "      <td>41.153332</td>\n",
       "      <td>20.168331</td>\n",
       "      <td>260110</td>\n",
       "      <td>249846.62375</td>\n",
       "      <td>10263.37625</td>\n",
       "      <td>3.945783</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Wheat</td>\n",
       "      <td>41.153332</td>\n",
       "      <td>20.168331</td>\n",
       "      <td>41292</td>\n",
       "      <td>40802.600664</td>\n",
       "      <td>489.399336</td>\n",
       "      <td>1.185216</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Soybeans</td>\n",
       "      <td>41.153332</td>\n",
       "      <td>20.168331</td>\n",
       "      <td>10000</td>\n",
       "      <td>17743.403174</td>\n",
       "      <td>-7743.403174</td>\n",
       "      <td>77.434032</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Algeria</td>\n",
       "      <td>Maize</td>\n",
       "      <td>28.033886</td>\n",
       "      <td>1.659626</td>\n",
       "      <td>33649</td>\n",
       "      <td>25145.994589</td>\n",
       "      <td>8503.005411</td>\n",
       "      <td>25.269712</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>633</th>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>Wheat</td>\n",
       "      <td>-19.015438</td>\n",
       "      <td>29.154857</td>\n",
       "      <td>22888</td>\n",
       "      <td>28548.460675</td>\n",
       "      <td>-5660.460675</td>\n",
       "      <td>24.731128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>634</th>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>Sorghum</td>\n",
       "      <td>-19.015438</td>\n",
       "      <td>29.154857</td>\n",
       "      <td>3066</td>\n",
       "      <td>2945.082863</td>\n",
       "      <td>120.917137</td>\n",
       "      <td>3.943807</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>635</th>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>Soybeans</td>\n",
       "      <td>-19.015438</td>\n",
       "      <td>29.154857</td>\n",
       "      <td>13142</td>\n",
       "      <td>17551.723221</td>\n",
       "      <td>-4409.723221</td>\n",
       "      <td>33.55443</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>636</th>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>Cassava</td>\n",
       "      <td>-19.015438</td>\n",
       "      <td>29.154857</td>\n",
       "      <td>46000</td>\n",
       "      <td>45570.213407</td>\n",
       "      <td>429.786593</td>\n",
       "      <td>0.934319</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>637</th>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>Sweet potatoes</td>\n",
       "      <td>-19.015438</td>\n",
       "      <td>29.154857</td>\n",
       "      <td>22222</td>\n",
       "      <td>22003.171039</td>\n",
       "      <td>218.828961</td>\n",
       "      <td>0.98474</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>638 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         area            crop        lat       long yield_2013  \\\n",
       "0     Albania           Maize  41.153332  20.168331      69533   \n",
       "1     Albania        Potatoes  41.153332  20.168331     260110   \n",
       "2     Albania           Wheat  41.153332  20.168331      41292   \n",
       "3     Albania        Soybeans  41.153332  20.168331      10000   \n",
       "4     Algeria           Maize  28.033886   1.659626      33649   \n",
       "..        ...             ...        ...        ...        ...   \n",
       "633  Zimbabwe           Wheat -19.015438  29.154857      22888   \n",
       "634  Zimbabwe         Sorghum -19.015438  29.154857       3066   \n",
       "635  Zimbabwe        Soybeans -19.015438  29.154857      13142   \n",
       "636  Zimbabwe         Cassava -19.015438  29.154857      46000   \n",
       "637  Zimbabwe  Sweet potatoes -19.015438  29.154857      22222   \n",
       "\n",
       "    yield_2013_pred yield_2013_diff   perc_err  \n",
       "0      66770.792339     2762.207661   3.972513  \n",
       "1      249846.62375     10263.37625   3.945783  \n",
       "2      40802.600664      489.399336   1.185216  \n",
       "3      17743.403174    -7743.403174  77.434032  \n",
       "4      25145.994589     8503.005411  25.269712  \n",
       "..              ...             ...        ...  \n",
       "633    28548.460675    -5660.460675  24.731128  \n",
       "634     2945.082863      120.917137   3.943807  \n",
       "635    17551.723221    -4409.723221   33.55443  \n",
       "636    45570.213407      429.786593   0.934319  \n",
       "637    22003.171039      218.828961    0.98474  \n",
       "\n",
       "[638 rows x 8 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions_df = pd.DataFrame(columns = ['area', 'crop', 'lat', 'long', 'yield_2013', \\\n",
    "                                         'yield_2013_pred', 'yield_2013_diff', 'perc_err'])\n",
    "\n",
    "for country in countries:\n",
    "    country_rows = df[df['area'] == country]\n",
    "    crops = country_rows['crop'].unique()\n",
    "    for crop in crops:\n",
    "        crop_rows = country_rows[country_rows['crop'] == crop] #rows = 6\n",
    "        \n",
    "        crops_08_12 = crop_rows[crop_rows['year'] != 2013] #rows = 5\n",
    "        x = crops_08_12[['yield_1', 'yield_2', 'yield_3', 'yield_4', 'yield_5', \\\n",
    "                   'avg_temp', 'avg_temp_1','avg_temp_2','avg_temp_3', 'avg_temp_4', 'avg_temp_5', \\\n",
    "                   'tonnes_potash', 'tonnes_potash_1', 'tonnes_potash_2', 'tonnes_potash_3', \\\n",
    "                   'tonnes_phosph', 'tonnes_phosph_1','tonnes_phosph_2', 'tonnes_phosph_3', \\\n",
    "                   'tonnes_pesticide', 'tonnes_pesticide_1', 'tonnes_pesticide_2', 'tonnes_pesticide_3', \\\n",
    "                  'tonnes_nitrogen', 'tonnes_nitrogen_1', 'tonnes_nitrogen_2', 'tonnes_nitrogen_3', \\\n",
    "                  'arable_land', 'ag_land']]\n",
    "        y = crops_08_12['yield']\n",
    "        \n",
    "        #x_train, x_test, y_train, y_test = train_test_split(x, y, test_size = 0.3, random_state = 100)\n",
    "        \n",
    "        scaler = StandardScaler()\n",
    "        x_scaler = scaler.fit(x)\n",
    "        #x_scaler = scaler.fit(x_train)\n",
    "        #x_train_scaled = x_scaler.transform(x_train)\n",
    "        #x_test_scaled = x_scaler.transform(x_test)\n",
    "        x_train_scaled = x_scaler.transform(x)\n",
    "        y_train = y\n",
    "        \n",
    "        mlr = LinearRegression()\n",
    "        mlr.fit(x_train_scaled, y_train)\n",
    "        \n",
    "        crops_13 = crop_rows[crop_rows['year'] == 2013] #rows = 1\n",
    "        x_test = crops_13[['yield_1', 'yield_2', 'yield_3', 'yield_4', 'yield_5', \\\n",
    "                   'avg_temp', 'avg_temp_1','avg_temp_2','avg_temp_3', 'avg_temp_4', 'avg_temp_5', \\\n",
    "                   'tonnes_potash', 'tonnes_potash_1', 'tonnes_potash_2', 'tonnes_potash_3', \\\n",
    "                   'tonnes_phosph', 'tonnes_phosph_1','tonnes_phosph_2', 'tonnes_phosph_3', \\\n",
    "                   'tonnes_pesticide', 'tonnes_pesticide_1', 'tonnes_pesticide_2', 'tonnes_pesticide_3', \\\n",
    "                  'tonnes_nitrogen', 'tonnes_nitrogen_1', 'tonnes_nitrogen_2', 'tonnes_nitrogen_3', \\\n",
    "                  'arable_land', 'ag_land']]\n",
    "        x_test_scaled = x_scaler.transform(x_test)\n",
    "        y_test = crops_13['yield']\n",
    "        pred_13 = mlr.predict(x_test_scaled)\n",
    "        perc_err = np.abs((y_test - pred_13) / y_test * 100)\n",
    "        \n",
    "        #pred_08_12 = mlr.predict(x_test_scaled)\n",
    "        #rmse = np.sqrt(metrics.mean_squared_error(y_test, pred_08_12))\n",
    "        #rsq = mlr.score(x_test_scaled, y_test)*100\n",
    "        #rsq = mlr.score(x_pred_scaled, y_actual)*100\n",
    "        \n",
    "        row = {'area': country, 'crop': crop, 'lat':crops_13['latitude'], 'long':crops_13['longitude'], 'yield_2013': y_test, 'yield_2013_pred': pred_13, \\\n",
    "               'yield_2013_diff': y_test - pred_13, 'perc_err': perc_err}\n",
    "        row_df = pd.DataFrame(row)\n",
    "        predictions_df = pd.concat([predictions_df, row_df], axis=0, ignore_index=True)\n",
    "predictions_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "6767dc74",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "area               object\n",
       "crop               object\n",
       "lat                object\n",
       "long               object\n",
       "yield_2013         object\n",
       "yield_2013_pred    object\n",
       "yield_2013_diff    object\n",
       "perc_err           object\n",
       "dtype: object"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions_df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "262c79aa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>area</th>\n",
       "      <th>crop</th>\n",
       "      <th>lat</th>\n",
       "      <th>long</th>\n",
       "      <th>yield_2013</th>\n",
       "      <th>yield_2013_pred</th>\n",
       "      <th>yield_2013_diff</th>\n",
       "      <th>perc_err</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Maize</td>\n",
       "      <td>41.153332</td>\n",
       "      <td>20.168331</td>\n",
       "      <td>69533</td>\n",
       "      <td>66771</td>\n",
       "      <td>2762</td>\n",
       "      <td>3.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Potatoes</td>\n",
       "      <td>41.153332</td>\n",
       "      <td>20.168331</td>\n",
       "      <td>260110</td>\n",
       "      <td>249847</td>\n",
       "      <td>10263</td>\n",
       "      <td>3.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Wheat</td>\n",
       "      <td>41.153332</td>\n",
       "      <td>20.168331</td>\n",
       "      <td>41292</td>\n",
       "      <td>40803</td>\n",
       "      <td>489</td>\n",
       "      <td>1.19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Soybeans</td>\n",
       "      <td>41.153332</td>\n",
       "      <td>20.168331</td>\n",
       "      <td>10000</td>\n",
       "      <td>17743</td>\n",
       "      <td>-7743</td>\n",
       "      <td>77.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Algeria</td>\n",
       "      <td>Maize</td>\n",
       "      <td>28.033886</td>\n",
       "      <td>1.659626</td>\n",
       "      <td>33649</td>\n",
       "      <td>25146</td>\n",
       "      <td>8503</td>\n",
       "      <td>25.27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>633</th>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>Wheat</td>\n",
       "      <td>-19.015438</td>\n",
       "      <td>29.154857</td>\n",
       "      <td>22888</td>\n",
       "      <td>28548</td>\n",
       "      <td>-5660</td>\n",
       "      <td>24.73</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>634</th>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>Sorghum</td>\n",
       "      <td>-19.015438</td>\n",
       "      <td>29.154857</td>\n",
       "      <td>3066</td>\n",
       "      <td>2945</td>\n",
       "      <td>121</td>\n",
       "      <td>3.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>635</th>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>Soybeans</td>\n",
       "      <td>-19.015438</td>\n",
       "      <td>29.154857</td>\n",
       "      <td>13142</td>\n",
       "      <td>17552</td>\n",
       "      <td>-4410</td>\n",
       "      <td>33.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>636</th>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>Cassava</td>\n",
       "      <td>-19.015438</td>\n",
       "      <td>29.154857</td>\n",
       "      <td>46000</td>\n",
       "      <td>45570</td>\n",
       "      <td>430</td>\n",
       "      <td>0.93</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>637</th>\n",
       "      <td>Zimbabwe</td>\n",
       "      <td>Sweet potatoes</td>\n",
       "      <td>-19.015438</td>\n",
       "      <td>29.154857</td>\n",
       "      <td>22222</td>\n",
       "      <td>22003</td>\n",
       "      <td>219</td>\n",
       "      <td>0.98</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>638 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         area            crop        lat       long  yield_2013  \\\n",
       "0     Albania           Maize  41.153332  20.168331       69533   \n",
       "1     Albania        Potatoes  41.153332  20.168331      260110   \n",
       "2     Albania           Wheat  41.153332  20.168331       41292   \n",
       "3     Albania        Soybeans  41.153332  20.168331       10000   \n",
       "4     Algeria           Maize  28.033886   1.659626       33649   \n",
       "..        ...             ...        ...        ...         ...   \n",
       "633  Zimbabwe           Wheat -19.015438  29.154857       22888   \n",
       "634  Zimbabwe         Sorghum -19.015438  29.154857        3066   \n",
       "635  Zimbabwe        Soybeans -19.015438  29.154857       13142   \n",
       "636  Zimbabwe         Cassava -19.015438  29.154857       46000   \n",
       "637  Zimbabwe  Sweet potatoes -19.015438  29.154857       22222   \n",
       "\n",
       "     yield_2013_pred  yield_2013_diff  perc_err  \n",
       "0              66771             2762      3.97  \n",
       "1             249847            10263      3.95  \n",
       "2              40803              489      1.19  \n",
       "3              17743            -7743     77.43  \n",
       "4              25146             8503     25.27  \n",
       "..               ...              ...       ...  \n",
       "633            28548            -5660     24.73  \n",
       "634             2945              121      3.94  \n",
       "635            17552            -4410     33.55  \n",
       "636            45570              430      0.93  \n",
       "637            22003              219      0.98  \n",
       "\n",
       "[638 rows x 8 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions_df['yield_2013'] = [int(x) for x in predictions_df['yield_2013']]\n",
    "predictions_df['yield_2013_pred'] = [round(float(x)) for x in predictions_df['yield_2013_pred']]\n",
    "predictions_df['yield_2013_diff'] = [round(float(x)) for x in predictions_df['yield_2013_diff']]\n",
    "predictions_df['perc_err'] = [round(float(x), 2) for x in predictions_df['perc_err']]\n",
    "predictions_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "11a65821",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "28.661645768025064"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions_df['perc_err'].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "36ed5f05",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6.805"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions_df['perc_err'].median()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "750c6be3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "507       0.00\n",
       "390       0.00\n",
       "434       0.01\n",
       "581       0.01\n",
       "416       0.02\n",
       "        ...   \n",
       "8       182.49\n",
       "208     260.68\n",
       "320     260.76\n",
       "143     476.79\n",
       "154    8484.30\n",
       "Name: perc_err, Length: 638, dtype: float64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions_df['perc_err'].sort_values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "623dfdd5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x1b7e6ddef70>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(predictions_df.index, predictions_df['perc_err'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "6435fbe2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x1b7e6ec9a90>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ruling out our biggest outlier\n",
    "no_ivory = predictions_df[predictions_df['perc_err'] < 6000]\n",
    "plt.scatter(no_ivory.index, no_ivory['perc_err'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "feff28b5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x1b7e6f44670>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "soybeans = predictions_df[predictions_df['crop'] == 'Soybeans']\n",
    "plt.scatter(soybeans.index, soybeans['perc_err'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "e42087db",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "no_ivory.boxplot(column=\"perc_err\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "9e45a8e7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Boxplot with the ~5 largest outliers removed with a cutoff of 150% error\n",
    "no_150 = predictions_df[predictions_df['perc_err'] < 150]\n",
    "no_150.boxplot(column=\"perc_err\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "730ae0f6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>area</th>\n",
       "      <th>year</th>\n",
       "      <th>crop</th>\n",
       "      <th>yield</th>\n",
       "      <th>yield_1</th>\n",
       "      <th>yield_2</th>\n",
       "      <th>yield_3</th>\n",
       "      <th>yield_4</th>\n",
       "      <th>yield_5</th>\n",
       "      <th>avg_temp</th>\n",
       "      <th>avg_temp_1</th>\n",
       "      <th>avg_temp_2</th>\n",
       "      <th>avg_temp_3</th>\n",
       "      <th>avg_temp_4</th>\n",
       "      <th>avg_temp_5</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>tonnes_potash</th>\n",
       "      <th>tonnes_potash_1</th>\n",
       "      <th>tonnes_potash_2</th>\n",
       "      <th>tonnes_potash_3</th>\n",
       "      <th>tonnes_phosph</th>\n",
       "      <th>tonnes_phosph_1</th>\n",
       "      <th>tonnes_phosph_2</th>\n",
       "      <th>tonnes_phosph_3</th>\n",
       "      <th>tonnes_pesticide</th>\n",
       "      <th>tonnes_pesticide_1</th>\n",
       "      <th>tonnes_pesticide_2</th>\n",
       "      <th>tonnes_pesticide_3</th>\n",
       "      <th>tonnes_nitrogen</th>\n",
       "      <th>tonnes_nitrogen_1</th>\n",
       "      <th>tonnes_nitrogen_2</th>\n",
       "      <th>tonnes_nitrogen_3</th>\n",
       "      <th>arable_land</th>\n",
       "      <th>ag_land</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>894</th>\n",
       "      <td>Côte d'Ivoire</td>\n",
       "      <td>2008</td>\n",
       "      <td>Sweet potatoes</td>\n",
       "      <td>21500</td>\n",
       "      <td>21500</td>\n",
       "      <td>21500</td>\n",
       "      <td>21500</td>\n",
       "      <td>21499</td>\n",
       "      <td>21500</td>\n",
       "      <td>26.94</td>\n",
       "      <td>27.01</td>\n",
       "      <td>26.99</td>\n",
       "      <td>26.98</td>\n",
       "      <td>26.99</td>\n",
       "      <td>27.05</td>\n",
       "      <td>7.539989</td>\n",
       "      <td>-5.54708</td>\n",
       "      <td>30468.00</td>\n",
       "      <td>27853.00</td>\n",
       "      <td>30489.0</td>\n",
       "      <td>22859.0</td>\n",
       "      <td>4740.00</td>\n",
       "      <td>15809.00</td>\n",
       "      <td>22312.0</td>\n",
       "      <td>10746.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>17606.00</td>\n",
       "      <td>25909.00</td>\n",
       "      <td>10962.0</td>\n",
       "      <td>16180.0</td>\n",
       "      <td>2900.0</td>\n",
       "      <td>20600.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>902</th>\n",
       "      <td>Côte d'Ivoire</td>\n",
       "      <td>2009</td>\n",
       "      <td>Sweet potatoes</td>\n",
       "      <td>18654</td>\n",
       "      <td>21500</td>\n",
       "      <td>21500</td>\n",
       "      <td>21500</td>\n",
       "      <td>21500</td>\n",
       "      <td>21499</td>\n",
       "      <td>26.98</td>\n",
       "      <td>26.94</td>\n",
       "      <td>27.01</td>\n",
       "      <td>26.99</td>\n",
       "      <td>26.98</td>\n",
       "      <td>26.99</td>\n",
       "      <td>7.539989</td>\n",
       "      <td>-5.54708</td>\n",
       "      <td>16198.00</td>\n",
       "      <td>30468.00</td>\n",
       "      <td>27853.0</td>\n",
       "      <td>30489.0</td>\n",
       "      <td>6798.00</td>\n",
       "      <td>4740.00</td>\n",
       "      <td>15809.0</td>\n",
       "      <td>22312.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>21396.00</td>\n",
       "      <td>17606.00</td>\n",
       "      <td>25909.0</td>\n",
       "      <td>10962.0</td>\n",
       "      <td>2900.0</td>\n",
       "      <td>20600.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>910</th>\n",
       "      <td>Côte d'Ivoire</td>\n",
       "      <td>2010</td>\n",
       "      <td>Sweet potatoes</td>\n",
       "      <td>19031</td>\n",
       "      <td>18654</td>\n",
       "      <td>21500</td>\n",
       "      <td>21500</td>\n",
       "      <td>21500</td>\n",
       "      <td>21500</td>\n",
       "      <td>27.45</td>\n",
       "      <td>26.98</td>\n",
       "      <td>26.94</td>\n",
       "      <td>27.01</td>\n",
       "      <td>26.99</td>\n",
       "      <td>26.98</td>\n",
       "      <td>7.539989</td>\n",
       "      <td>-5.54708</td>\n",
       "      <td>43788.00</td>\n",
       "      <td>16198.00</td>\n",
       "      <td>30468.0</td>\n",
       "      <td>27853.0</td>\n",
       "      <td>22537.00</td>\n",
       "      <td>6798.00</td>\n",
       "      <td>4740.0</td>\n",
       "      <td>15809.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>26724.00</td>\n",
       "      <td>21396.00</td>\n",
       "      <td>17606.0</td>\n",
       "      <td>25909.0</td>\n",
       "      <td>2900.0</td>\n",
       "      <td>20600.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>918</th>\n",
       "      <td>Côte d'Ivoire</td>\n",
       "      <td>2011</td>\n",
       "      <td>Sweet potatoes</td>\n",
       "      <td>18690</td>\n",
       "      <td>19031</td>\n",
       "      <td>18654</td>\n",
       "      <td>21500</td>\n",
       "      <td>21500</td>\n",
       "      <td>21500</td>\n",
       "      <td>27.02</td>\n",
       "      <td>27.45</td>\n",
       "      <td>26.98</td>\n",
       "      <td>26.94</td>\n",
       "      <td>27.01</td>\n",
       "      <td>26.99</td>\n",
       "      <td>7.539989</td>\n",
       "      <td>-5.54708</td>\n",
       "      <td>21840.00</td>\n",
       "      <td>43788.00</td>\n",
       "      <td>16198.0</td>\n",
       "      <td>30468.0</td>\n",
       "      <td>9192.00</td>\n",
       "      <td>22537.00</td>\n",
       "      <td>6798.0</td>\n",
       "      <td>4740.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>25236.00</td>\n",
       "      <td>26724.00</td>\n",
       "      <td>21396.0</td>\n",
       "      <td>17606.0</td>\n",
       "      <td>2900.0</td>\n",
       "      <td>20600.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>926</th>\n",
       "      <td>Côte d'Ivoire</td>\n",
       "      <td>2012</td>\n",
       "      <td>Sweet potatoes</td>\n",
       "      <td>19169</td>\n",
       "      <td>18690</td>\n",
       "      <td>19031</td>\n",
       "      <td>18654</td>\n",
       "      <td>21500</td>\n",
       "      <td>21500</td>\n",
       "      <td>26.77</td>\n",
       "      <td>27.02</td>\n",
       "      <td>27.45</td>\n",
       "      <td>26.98</td>\n",
       "      <td>26.94</td>\n",
       "      <td>27.01</td>\n",
       "      <td>7.539989</td>\n",
       "      <td>-5.54708</td>\n",
       "      <td>26895.86</td>\n",
       "      <td>21840.00</td>\n",
       "      <td>43788.0</td>\n",
       "      <td>16198.0</td>\n",
       "      <td>18636.64</td>\n",
       "      <td>9192.00</td>\n",
       "      <td>22537.0</td>\n",
       "      <td>6798.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>32530.85</td>\n",
       "      <td>25236.00</td>\n",
       "      <td>26724.0</td>\n",
       "      <td>21396.0</td>\n",
       "      <td>3000.0</td>\n",
       "      <td>20700.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>934</th>\n",
       "      <td>Côte d'Ivoire</td>\n",
       "      <td>2013</td>\n",
       "      <td>Sweet potatoes</td>\n",
       "      <td>19306</td>\n",
       "      <td>19169</td>\n",
       "      <td>18690</td>\n",
       "      <td>19031</td>\n",
       "      <td>18654</td>\n",
       "      <td>21500</td>\n",
       "      <td>27.23</td>\n",
       "      <td>26.77</td>\n",
       "      <td>27.02</td>\n",
       "      <td>27.45</td>\n",
       "      <td>26.98</td>\n",
       "      <td>26.94</td>\n",
       "      <td>7.539989</td>\n",
       "      <td>-5.54708</td>\n",
       "      <td>39042.04</td>\n",
       "      <td>26895.86</td>\n",
       "      <td>21840.0</td>\n",
       "      <td>43788.0</td>\n",
       "      <td>36486.51</td>\n",
       "      <td>18636.64</td>\n",
       "      <td>9192.0</td>\n",
       "      <td>22537.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>33455.43</td>\n",
       "      <td>32530.85</td>\n",
       "      <td>25236.0</td>\n",
       "      <td>26724.0</td>\n",
       "      <td>3500.0</td>\n",
       "      <td>21200.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              area  year            crop  yield  yield_1  yield_2  yield_3  \\\n",
       "894  Côte d'Ivoire  2008  Sweet potatoes  21500    21500    21500    21500   \n",
       "902  Côte d'Ivoire  2009  Sweet potatoes  18654    21500    21500    21500   \n",
       "910  Côte d'Ivoire  2010  Sweet potatoes  19031    18654    21500    21500   \n",
       "918  Côte d'Ivoire  2011  Sweet potatoes  18690    19031    18654    21500   \n",
       "926  Côte d'Ivoire  2012  Sweet potatoes  19169    18690    19031    18654   \n",
       "934  Côte d'Ivoire  2013  Sweet potatoes  19306    19169    18690    19031   \n",
       "\n",
       "     yield_4  yield_5  avg_temp  avg_temp_1  avg_temp_2  avg_temp_3  \\\n",
       "894    21499    21500     26.94       27.01       26.99       26.98   \n",
       "902    21500    21499     26.98       26.94       27.01       26.99   \n",
       "910    21500    21500     27.45       26.98       26.94       27.01   \n",
       "918    21500    21500     27.02       27.45       26.98       26.94   \n",
       "926    21500    21500     26.77       27.02       27.45       26.98   \n",
       "934    18654    21500     27.23       26.77       27.02       27.45   \n",
       "\n",
       "     avg_temp_4  avg_temp_5  latitude  longitude  tonnes_potash  \\\n",
       "894       26.99       27.05  7.539989   -5.54708       30468.00   \n",
       "902       26.98       26.99  7.539989   -5.54708       16198.00   \n",
       "910       26.99       26.98  7.539989   -5.54708       43788.00   \n",
       "918       27.01       26.99  7.539989   -5.54708       21840.00   \n",
       "926       26.94       27.01  7.539989   -5.54708       26895.86   \n",
       "934       26.98       26.94  7.539989   -5.54708       39042.04   \n",
       "\n",
       "     tonnes_potash_1  tonnes_potash_2  tonnes_potash_3  tonnes_phosph  \\\n",
       "894         27853.00          30489.0          22859.0        4740.00   \n",
       "902         30468.00          27853.0          30489.0        6798.00   \n",
       "910         16198.00          30468.0          27853.0       22537.00   \n",
       "918         43788.00          16198.0          30468.0        9192.00   \n",
       "926         21840.00          43788.0          16198.0       18636.64   \n",
       "934         26895.86          21840.0          43788.0       36486.51   \n",
       "\n",
       "     tonnes_phosph_1  tonnes_phosph_2  tonnes_phosph_3  tonnes_pesticide  \\\n",
       "894         15809.00          22312.0          10746.0              93.0   \n",
       "902          4740.00          15809.0          22312.0              93.0   \n",
       "910          6798.00           4740.0          15809.0              93.0   \n",
       "918         22537.00           6798.0           4740.0              93.0   \n",
       "926          9192.00          22537.0           6798.0              93.0   \n",
       "934         18636.64           9192.0          22537.0              93.0   \n",
       "\n",
       "     tonnes_pesticide_1  tonnes_pesticide_2  tonnes_pesticide_3  \\\n",
       "894                93.0                93.0                93.0   \n",
       "902                93.0                93.0                93.0   \n",
       "910                93.0                93.0                93.0   \n",
       "918                93.0                93.0                93.0   \n",
       "926                93.0                93.0                93.0   \n",
       "934                93.0                93.0                93.0   \n",
       "\n",
       "     tonnes_nitrogen  tonnes_nitrogen_1  tonnes_nitrogen_2  tonnes_nitrogen_3  \\\n",
       "894         17606.00           25909.00            10962.0            16180.0   \n",
       "902         21396.00           17606.00            25909.0            10962.0   \n",
       "910         26724.00           21396.00            17606.0            25909.0   \n",
       "918         25236.00           26724.00            21396.0            17606.0   \n",
       "926         32530.85           25236.00            26724.0            21396.0   \n",
       "934         33455.43           32530.85            25236.0            26724.0   \n",
       "\n",
       "     arable_land  ag_land  \n",
       "894       2900.0  20600.0  \n",
       "902       2900.0  20600.0  \n",
       "910       2900.0  20600.0  \n",
       "918       2900.0  20600.0  \n",
       "926       3000.0  20700.0  \n",
       "934       3500.0  21200.0  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with pd.option_context('display.max_columns', None,):\n",
    "    display(df[(df['area'] == \"Côte d'Ivoire\") & (df['crop'] == 'Sweet potatoes')])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "c8d4dc74",
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions_df.to_csv(\"processed_data/predictions.csv\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "mlenv",
   "language": "python",
   "name": "mlenv"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}