{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "nbpresent": {
     "id": "a20290f5-2a26-4622-9e8b-ce94cc1dcb1a"
    },
    "slideshow": {
     "slide_type": "slide"
    },
    "tags": []
   },
   "source": [
    "# Estimation of tree height using GEDI dataset - Random Forest prediction\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In order to run this notebook, please do what follows:\n",
    "\n",
    "    cd  /media/sf_LVM_shared/my_SE_data/exercise\n",
    "    pip3 install pyspatialml\n",
    "    jupyter-notebook Tree_Height_03RF_pred"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Find a relationship between tree height and enviromental predictors to be able to predict out side the GEDI observation.  \n",
    "The model, in its simplest form, looks like the following:  \n",
    "\n",
    "```\n",
    "y ~ f ( a*x1 + b*x2 + c*x3 + …)  + ε,  \n",
    "\n",
    "```\n",
    "    \n",
    "where y is a response variable, the x's are predictor variables, and ε is the associated error.\n",
    "\n",
    "**There are many different types of models**\n",
    "\n",
    "* Parametric models – make assumptions about the underlying distribution of the data.\n",
    "    * Maximum likelihood classification\n",
    "    * Discriminant analysis\n",
    "    * General linear models (ex. linear regression)\n",
    "\n",
    "\n",
    "* Nonparametric models – make no assumptions about the underlying data distributions.\n",
    "    * Generalized additive models\n",
    "    * Support Vector Machine\n",
    "    * Artificial neural networks\n",
    "    * Random Forest"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Random forest\n",
    "\n",
    "Random forest is a Supervised Machine Learning Algorithm that is used widely in **Classification** (containing categorical variables) and **Regression** (response continuous variables) problems. It builds decision trees on different samples and takes their majority vote for classification and average in case of regression.  \n",
    "\n",
    "One of the most important features of the Random Forest Algorithm is that it can handle continuous and categorical variables predictors without any assamption on their data distribution, dimension, and even if the predictors are autocorrelated."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Bagging**  \n",
    "Bagging, also known as Bootstrap Aggregation is the ensemble technique used by random forest. Choosesing a random sample from the data set it is creating a tree. Each tree is generated from the samples (Bootstrap Samples) provided by the Original Data with replacement known as row sampling. This step of row sampling with replacement is called bootstrap.\n",
    "Each is trained independently which generates results. The final output is based on majority voting after combining the results of all models. This step which involves combining all the results and generating output based on majority voting is known as tree aggregation.\n",
    "\n",
    "\n",
    "**Steps involved in random forest algorithm:**  \n",
    "\n",
    "* Step 1: In Random forest n number of random records are taken from the data set having k number of records.  \n",
    "\n",
    "* Step 2: Individual decision trees are constructed for each sample.  \n",
    "\n",
    "* Step 3: Each decision tree will generate an output.  \n",
    "* Step 4: Final output is considered based on Majority Voting or Averaging for Classification and regression respectively.  \n",
    " \n",
    "(source https://www.analyticsvidhya.com/blog/2021/06/understanding-random-forest/)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/jpeg": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 2,
     "metadata": {
      "image/jpeg": {
       "height": 700,
       "width": 700
      }
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.display import Image\n",
    "Image(\"../images/Random_Forest.jpg\" , width = 700, height = 700)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Random Forest exercise**   \n",
    "In this exercise we will use Random Forest Regression algorithm with the [sklearn](https://scikit-learn.org/stable/) python package to estimage the tree height. In last step, we are going to use the [Pyspatialml](https://pyspatialml.readthedocs.io/en/latest/index.html) for the spatial prediction on the raster files."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "nbpresent": {
     "id": "8f67df50-3050-47d2-a5d9-4329a61325fa"
    },
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import rasterio\n",
    "from rasterio import *\n",
    "from rasterio.plot import show\n",
    "from pyspatialml import Raster\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.model_selection import train_test_split,GridSearchCV\n",
    "from sklearn.pipeline import Pipeline\n",
    "from scipy.stats import pearsonr\n",
    "import matplotlib.pyplot as plt\n",
    "plt.rcParams[\"figure.figsize\"] = (10,6.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "nbpresent": {
     "id": "cf3cb8c6-45f0-48de-b171-598ebf92904f"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Import raw data, extracted predictors and show the data distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>X</th>\n",
       "      <th>Y</th>\n",
       "      <th>h</th>\n",
       "      <th>BLDFIE_WeigAver</th>\n",
       "      <th>CECSOL_WeigAver</th>\n",
       "      <th>CHELSA_bio18</th>\n",
       "      <th>CHELSA_bio4</th>\n",
       "      <th>convergence</th>\n",
       "      <th>cti</th>\n",
       "      <th>devmagnitude</th>\n",
       "      <th>eastness</th>\n",
       "      <th>elev</th>\n",
       "      <th>forestheight</th>\n",
       "      <th>glad_ard_SVVI_max</th>\n",
       "      <th>glad_ard_SVVI_med</th>\n",
       "      <th>glad_ard_SVVI_min</th>\n",
       "      <th>northness</th>\n",
       "      <th>ORCDRC_WeigAver</th>\n",
       "      <th>outlet_dist_dw_basin</th>\n",
       "      <th>SBIO3_Isothermality_5_15cm</th>\n",
       "      <th>SBIO4_Temperature_Seasonality_5_15cm</th>\n",
       "      <th>treecover</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>6.050001</td>\n",
       "      <td>49.727499</td>\n",
       "      <td>3139.00</td>\n",
       "      <td>1540</td>\n",
       "      <td>13</td>\n",
       "      <td>2113</td>\n",
       "      <td>5893</td>\n",
       "      <td>-10.486560</td>\n",
       "      <td>-238043120</td>\n",
       "      <td>1.158417</td>\n",
       "      <td>0.069094</td>\n",
       "      <td>353.983124</td>\n",
       "      <td>23</td>\n",
       "      <td>276.871094</td>\n",
       "      <td>46.444092</td>\n",
       "      <td>347.665405</td>\n",
       "      <td>0.042500</td>\n",
       "      <td>9</td>\n",
       "      <td>780403</td>\n",
       "      <td>19.798992</td>\n",
       "      <td>440.672211</td>\n",
       "      <td>85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>6.050002</td>\n",
       "      <td>49.922155</td>\n",
       "      <td>1454.75</td>\n",
       "      <td>1491</td>\n",
       "      <td>12</td>\n",
       "      <td>1993</td>\n",
       "      <td>5912</td>\n",
       "      <td>33.274361</td>\n",
       "      <td>-208915344</td>\n",
       "      <td>-1.755341</td>\n",
       "      <td>0.269112</td>\n",
       "      <td>267.511688</td>\n",
       "      <td>19</td>\n",
       "      <td>-49.526367</td>\n",
       "      <td>19.552734</td>\n",
       "      <td>-130.541748</td>\n",
       "      <td>0.182780</td>\n",
       "      <td>16</td>\n",
       "      <td>772777</td>\n",
       "      <td>20.889412</td>\n",
       "      <td>457.756195</td>\n",
       "      <td>85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>6.050002</td>\n",
       "      <td>48.602377</td>\n",
       "      <td>853.50</td>\n",
       "      <td>1521</td>\n",
       "      <td>17</td>\n",
       "      <td>2124</td>\n",
       "      <td>5983</td>\n",
       "      <td>0.045293</td>\n",
       "      <td>-137479792</td>\n",
       "      <td>1.908780</td>\n",
       "      <td>-0.016055</td>\n",
       "      <td>389.751160</td>\n",
       "      <td>21</td>\n",
       "      <td>93.257324</td>\n",
       "      <td>50.743652</td>\n",
       "      <td>384.522461</td>\n",
       "      <td>0.036253</td>\n",
       "      <td>14</td>\n",
       "      <td>898820</td>\n",
       "      <td>20.695877</td>\n",
       "      <td>481.879700</td>\n",
       "      <td>62</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>6.050009</td>\n",
       "      <td>48.151979</td>\n",
       "      <td>3141.00</td>\n",
       "      <td>1526</td>\n",
       "      <td>16</td>\n",
       "      <td>2569</td>\n",
       "      <td>6130</td>\n",
       "      <td>-33.654274</td>\n",
       "      <td>-267223072</td>\n",
       "      <td>0.965787</td>\n",
       "      <td>0.067767</td>\n",
       "      <td>380.207703</td>\n",
       "      <td>27</td>\n",
       "      <td>542.401367</td>\n",
       "      <td>202.264160</td>\n",
       "      <td>386.156738</td>\n",
       "      <td>0.005139</td>\n",
       "      <td>15</td>\n",
       "      <td>831824</td>\n",
       "      <td>19.375000</td>\n",
       "      <td>479.410278</td>\n",
       "      <td>85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>6.050010</td>\n",
       "      <td>49.588410</td>\n",
       "      <td>2065.25</td>\n",
       "      <td>1547</td>\n",
       "      <td>14</td>\n",
       "      <td>2108</td>\n",
       "      <td>5923</td>\n",
       "      <td>27.493824</td>\n",
       "      <td>-107809368</td>\n",
       "      <td>-0.162624</td>\n",
       "      <td>0.014065</td>\n",
       "      <td>308.042786</td>\n",
       "      <td>25</td>\n",
       "      <td>136.048340</td>\n",
       "      <td>146.835205</td>\n",
       "      <td>198.127441</td>\n",
       "      <td>0.028847</td>\n",
       "      <td>17</td>\n",
       "      <td>796962</td>\n",
       "      <td>18.777500</td>\n",
       "      <td>457.880066</td>\n",
       "      <td>85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>6.050014</td>\n",
       "      <td>48.608456</td>\n",
       "      <td>1246.50</td>\n",
       "      <td>1515</td>\n",
       "      <td>19</td>\n",
       "      <td>2124</td>\n",
       "      <td>6010</td>\n",
       "      <td>-1.602039</td>\n",
       "      <td>17384282</td>\n",
       "      <td>1.447979</td>\n",
       "      <td>-0.018912</td>\n",
       "      <td>364.527100</td>\n",
       "      <td>18</td>\n",
       "      <td>221.339844</td>\n",
       "      <td>247.387207</td>\n",
       "      <td>480.387939</td>\n",
       "      <td>0.042747</td>\n",
       "      <td>14</td>\n",
       "      <td>897945</td>\n",
       "      <td>19.398880</td>\n",
       "      <td>474.331329</td>\n",
       "      <td>62</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7</td>\n",
       "      <td>6.050016</td>\n",
       "      <td>48.571401</td>\n",
       "      <td>2938.75</td>\n",
       "      <td>1520</td>\n",
       "      <td>19</td>\n",
       "      <td>2169</td>\n",
       "      <td>6147</td>\n",
       "      <td>27.856503</td>\n",
       "      <td>-66516432</td>\n",
       "      <td>-1.073956</td>\n",
       "      <td>0.002280</td>\n",
       "      <td>254.679596</td>\n",
       "      <td>19</td>\n",
       "      <td>125.250488</td>\n",
       "      <td>87.865234</td>\n",
       "      <td>160.696777</td>\n",
       "      <td>0.037254</td>\n",
       "      <td>11</td>\n",
       "      <td>908426</td>\n",
       "      <td>20.170450</td>\n",
       "      <td>476.414520</td>\n",
       "      <td>96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8</td>\n",
       "      <td>6.050019</td>\n",
       "      <td>49.921613</td>\n",
       "      <td>3294.75</td>\n",
       "      <td>1490</td>\n",
       "      <td>12</td>\n",
       "      <td>1995</td>\n",
       "      <td>5912</td>\n",
       "      <td>22.102139</td>\n",
       "      <td>-297770784</td>\n",
       "      <td>-1.402633</td>\n",
       "      <td>0.309765</td>\n",
       "      <td>294.927765</td>\n",
       "      <td>26</td>\n",
       "      <td>-86.729492</td>\n",
       "      <td>-145.584229</td>\n",
       "      <td>-190.062988</td>\n",
       "      <td>0.222435</td>\n",
       "      <td>15</td>\n",
       "      <td>772784</td>\n",
       "      <td>20.855963</td>\n",
       "      <td>457.195404</td>\n",
       "      <td>86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>9</td>\n",
       "      <td>6.050020</td>\n",
       "      <td>48.822645</td>\n",
       "      <td>1623.50</td>\n",
       "      <td>1554</td>\n",
       "      <td>18</td>\n",
       "      <td>1973</td>\n",
       "      <td>6138</td>\n",
       "      <td>18.496584</td>\n",
       "      <td>-25336536</td>\n",
       "      <td>-0.800016</td>\n",
       "      <td>0.010370</td>\n",
       "      <td>240.493759</td>\n",
       "      <td>22</td>\n",
       "      <td>-51.470703</td>\n",
       "      <td>-245.886719</td>\n",
       "      <td>172.074707</td>\n",
       "      <td>0.004428</td>\n",
       "      <td>8</td>\n",
       "      <td>839132</td>\n",
       "      <td>21.812290</td>\n",
       "      <td>496.231110</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10</td>\n",
       "      <td>6.050024</td>\n",
       "      <td>49.847522</td>\n",
       "      <td>1400.00</td>\n",
       "      <td>1521</td>\n",
       "      <td>15</td>\n",
       "      <td>2187</td>\n",
       "      <td>5886</td>\n",
       "      <td>-5.660453</td>\n",
       "      <td>-278652608</td>\n",
       "      <td>1.477951</td>\n",
       "      <td>-0.068720</td>\n",
       "      <td>376.671143</td>\n",
       "      <td>12</td>\n",
       "      <td>277.297363</td>\n",
       "      <td>273.141846</td>\n",
       "      <td>-138.895996</td>\n",
       "      <td>0.098817</td>\n",
       "      <td>13</td>\n",
       "      <td>768873</td>\n",
       "      <td>21.137711</td>\n",
       "      <td>466.976685</td>\n",
       "      <td>70</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ID         X          Y        h  BLDFIE_WeigAver  CECSOL_WeigAver  \\\n",
       "0   1  6.050001  49.727499  3139.00             1540               13   \n",
       "1   2  6.050002  49.922155  1454.75             1491               12   \n",
       "2   3  6.050002  48.602377   853.50             1521               17   \n",
       "3   4  6.050009  48.151979  3141.00             1526               16   \n",
       "4   5  6.050010  49.588410  2065.25             1547               14   \n",
       "5   6  6.050014  48.608456  1246.50             1515               19   \n",
       "6   7  6.050016  48.571401  2938.75             1520               19   \n",
       "7   8  6.050019  49.921613  3294.75             1490               12   \n",
       "8   9  6.050020  48.822645  1623.50             1554               18   \n",
       "9  10  6.050024  49.847522  1400.00             1521               15   \n",
       "\n",
       "   CHELSA_bio18  CHELSA_bio4  convergence        cti  devmagnitude  eastness  \\\n",
       "0          2113         5893   -10.486560 -238043120      1.158417  0.069094   \n",
       "1          1993         5912    33.274361 -208915344     -1.755341  0.269112   \n",
       "2          2124         5983     0.045293 -137479792      1.908780 -0.016055   \n",
       "3          2569         6130   -33.654274 -267223072      0.965787  0.067767   \n",
       "4          2108         5923    27.493824 -107809368     -0.162624  0.014065   \n",
       "5          2124         6010    -1.602039   17384282      1.447979 -0.018912   \n",
       "6          2169         6147    27.856503  -66516432     -1.073956  0.002280   \n",
       "7          1995         5912    22.102139 -297770784     -1.402633  0.309765   \n",
       "8          1973         6138    18.496584  -25336536     -0.800016  0.010370   \n",
       "9          2187         5886    -5.660453 -278652608      1.477951 -0.068720   \n",
       "\n",
       "         elev  forestheight  glad_ard_SVVI_max  glad_ard_SVVI_med  \\\n",
       "0  353.983124            23         276.871094          46.444092   \n",
       "1  267.511688            19         -49.526367          19.552734   \n",
       "2  389.751160            21          93.257324          50.743652   \n",
       "3  380.207703            27         542.401367         202.264160   \n",
       "4  308.042786            25         136.048340         146.835205   \n",
       "5  364.527100            18         221.339844         247.387207   \n",
       "6  254.679596            19         125.250488          87.865234   \n",
       "7  294.927765            26         -86.729492        -145.584229   \n",
       "8  240.493759            22         -51.470703        -245.886719   \n",
       "9  376.671143            12         277.297363         273.141846   \n",
       "\n",
       "   glad_ard_SVVI_min  northness  ORCDRC_WeigAver  outlet_dist_dw_basin  \\\n",
       "0         347.665405   0.042500                9                780403   \n",
       "1        -130.541748   0.182780               16                772777   \n",
       "2         384.522461   0.036253               14                898820   \n",
       "3         386.156738   0.005139               15                831824   \n",
       "4         198.127441   0.028847               17                796962   \n",
       "5         480.387939   0.042747               14                897945   \n",
       "6         160.696777   0.037254               11                908426   \n",
       "7        -190.062988   0.222435               15                772784   \n",
       "8         172.074707   0.004428                8                839132   \n",
       "9        -138.895996   0.098817               13                768873   \n",
       "\n",
       "   SBIO3_Isothermality_5_15cm  SBIO4_Temperature_Seasonality_5_15cm  treecover  \n",
       "0                   19.798992                            440.672211         85  \n",
       "1                   20.889412                            457.756195         85  \n",
       "2                   20.695877                            481.879700         62  \n",
       "3                   19.375000                            479.410278         85  \n",
       "4                   18.777500                            457.880066         85  \n",
       "5                   19.398880                            474.331329         62  \n",
       "6                   20.170450                            476.414520         96  \n",
       "7                   20.855963                            457.195404         86  \n",
       "8                   21.812290                            496.231110         64  \n",
       "9                   21.137711                            466.976685         70  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictors = pd.read_csv(\"tree_height/txt/eu_x_y_height_predictors_select.txt\", sep=\" \",  index_col=False)\n",
    "pd.set_option('display.max_columns',None)\n",
    "# change column name\n",
    "predictors = predictors.rename({'dev-magnitude':'devmagnitude'} , axis='columns')\n",
    "predictors.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1267239"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(predictors)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plotting the data "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x468 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "bins = np.linspace(min(predictors['h']),max(predictors['h']),100)\n",
    "plt.hist((predictors['h']),bins,alpha=0.8);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Select only tree with height less then 70m (highest tree in germany 67m https://visit.freiburg.de/en/attractions/waldtraut-germany-s-tallest-tree)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>ID</th>\n",
       "      <th>X</th>\n",
       "      <th>Y</th>\n",
       "      <th>h</th>\n",
       "      <th>hm</th>\n",
       "      <th>BLDFIE_WeigAver</th>\n",
       "      <th>CECSOL_WeigAver</th>\n",
       "      <th>CHELSA_bio18</th>\n",
       "      <th>CHELSA_bio4</th>\n",
       "      <th>convergence</th>\n",
       "      <th>cti</th>\n",
       "      <th>devmagnitude</th>\n",
       "      <th>eastness</th>\n",
       "      <th>elev</th>\n",
       "      <th>forestheight</th>\n",
       "      <th>glad_ard_SVVI_max</th>\n",
       "      <th>glad_ard_SVVI_med</th>\n",
       "      <th>glad_ard_SVVI_min</th>\n",
       "      <th>northness</th>\n",
       "      <th>ORCDRC_WeigAver</th>\n",
       "      <th>outlet_dist_dw_basin</th>\n",
       "      <th>SBIO3_Isothermality_5_15cm</th>\n",
       "      <th>SBIO4_Temperature_Seasonality_5_15cm</th>\n",
       "      <th>treecover</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>211869</th>\n",
       "      <td>211870</td>\n",
       "      <td>6.763981</td>\n",
       "      <td>49.923656</td>\n",
       "      <td>2504.00</td>\n",
       "      <td>25.0400</td>\n",
       "      <td>1533</td>\n",
       "      <td>13</td>\n",
       "      <td>2114</td>\n",
       "      <td>6069</td>\n",
       "      <td>-51.807003</td>\n",
       "      <td>-204555312</td>\n",
       "      <td>1.713994</td>\n",
       "      <td>-0.030101</td>\n",
       "      <td>356.479462</td>\n",
       "      <td>23</td>\n",
       "      <td>168.107666</td>\n",
       "      <td>0.701416</td>\n",
       "      <td>-27.481689</td>\n",
       "      <td>0.001850</td>\n",
       "      <td>15</td>\n",
       "      <td>667874</td>\n",
       "      <td>19.075720</td>\n",
       "      <td>444.208923</td>\n",
       "      <td>95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1243664</th>\n",
       "      <td>1243665</td>\n",
       "      <td>9.833275</td>\n",
       "      <td>49.526917</td>\n",
       "      <td>649.25</td>\n",
       "      <td>6.4925</td>\n",
       "      <td>1534</td>\n",
       "      <td>15</td>\n",
       "      <td>2102</td>\n",
       "      <td>6516</td>\n",
       "      <td>0.582322</td>\n",
       "      <td>-200281328</td>\n",
       "      <td>-0.082161</td>\n",
       "      <td>0.021150</td>\n",
       "      <td>338.562622</td>\n",
       "      <td>23</td>\n",
       "      <td>663.329102</td>\n",
       "      <td>66.410156</td>\n",
       "      <td>208.560791</td>\n",
       "      <td>-0.054791</td>\n",
       "      <td>8</td>\n",
       "      <td>801897</td>\n",
       "      <td>18.012062</td>\n",
       "      <td>487.378571</td>\n",
       "      <td>89</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>862800</th>\n",
       "      <td>862801</td>\n",
       "      <td>8.747207</td>\n",
       "      <td>49.396667</td>\n",
       "      <td>4145.25</td>\n",
       "      <td>41.4525</td>\n",
       "      <td>1464</td>\n",
       "      <td>13</td>\n",
       "      <td>2770</td>\n",
       "      <td>6187</td>\n",
       "      <td>-5.705454</td>\n",
       "      <td>-193031872</td>\n",
       "      <td>2.207440</td>\n",
       "      <td>0.082736</td>\n",
       "      <td>471.601471</td>\n",
       "      <td>29</td>\n",
       "      <td>682.764160</td>\n",
       "      <td>196.173340</td>\n",
       "      <td>410.016113</td>\n",
       "      <td>-0.105913</td>\n",
       "      <td>9</td>\n",
       "      <td>687733</td>\n",
       "      <td>19.400467</td>\n",
       "      <td>449.129852</td>\n",
       "      <td>85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>184466</th>\n",
       "      <td>184467</td>\n",
       "      <td>6.692364</td>\n",
       "      <td>48.311297</td>\n",
       "      <td>902.75</td>\n",
       "      <td>9.0275</td>\n",
       "      <td>1503</td>\n",
       "      <td>16</td>\n",
       "      <td>2458</td>\n",
       "      <td>6333</td>\n",
       "      <td>2.532082</td>\n",
       "      <td>-153840256</td>\n",
       "      <td>-0.687262</td>\n",
       "      <td>-0.040297</td>\n",
       "      <td>326.341187</td>\n",
       "      <td>12</td>\n",
       "      <td>-155.665771</td>\n",
       "      <td>-179.105957</td>\n",
       "      <td>-273.086426</td>\n",
       "      <td>0.005218</td>\n",
       "      <td>11</td>\n",
       "      <td>938821</td>\n",
       "      <td>19.768513</td>\n",
       "      <td>504.342621</td>\n",
       "      <td>86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>210193</th>\n",
       "      <td>210194</td>\n",
       "      <td>6.759972</td>\n",
       "      <td>48.427197</td>\n",
       "      <td>1215.75</td>\n",
       "      <td>12.1575</td>\n",
       "      <td>1499</td>\n",
       "      <td>11</td>\n",
       "      <td>2473</td>\n",
       "      <td>6283</td>\n",
       "      <td>-34.008671</td>\n",
       "      <td>-184876336</td>\n",
       "      <td>0.869934</td>\n",
       "      <td>0.029229</td>\n",
       "      <td>331.943359</td>\n",
       "      <td>26</td>\n",
       "      <td>214.386475</td>\n",
       "      <td>39.022705</td>\n",
       "      <td>-19.717529</td>\n",
       "      <td>-0.010323</td>\n",
       "      <td>20</td>\n",
       "      <td>937918</td>\n",
       "      <td>18.850109</td>\n",
       "      <td>450.874207</td>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>485841</th>\n",
       "      <td>485842</td>\n",
       "      <td>7.421320</td>\n",
       "      <td>48.972783</td>\n",
       "      <td>2629.50</td>\n",
       "      <td>26.2950</td>\n",
       "      <td>1443</td>\n",
       "      <td>13</td>\n",
       "      <td>2366</td>\n",
       "      <td>6249</td>\n",
       "      <td>-12.187319</td>\n",
       "      <td>-237638256</td>\n",
       "      <td>0.527734</td>\n",
       "      <td>0.184359</td>\n",
       "      <td>317.114014</td>\n",
       "      <td>25</td>\n",
       "      <td>55.184082</td>\n",
       "      <td>58.465332</td>\n",
       "      <td>91.532715</td>\n",
       "      <td>-0.045335</td>\n",
       "      <td>10</td>\n",
       "      <td>828316</td>\n",
       "      <td>19.876814</td>\n",
       "      <td>445.643494</td>\n",
       "      <td>73</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>120379</th>\n",
       "      <td>120380</td>\n",
       "      <td>6.482271</td>\n",
       "      <td>49.914016</td>\n",
       "      <td>2173.00</td>\n",
       "      <td>21.7300</td>\n",
       "      <td>1538</td>\n",
       "      <td>16</td>\n",
       "      <td>1847</td>\n",
       "      <td>6105</td>\n",
       "      <td>-17.632570</td>\n",
       "      <td>-197189520</td>\n",
       "      <td>-1.088760</td>\n",
       "      <td>-0.035505</td>\n",
       "      <td>272.008911</td>\n",
       "      <td>23</td>\n",
       "      <td>-126.860352</td>\n",
       "      <td>-195.746094</td>\n",
       "      <td>75.838867</td>\n",
       "      <td>-0.008322</td>\n",
       "      <td>18</td>\n",
       "      <td>728450</td>\n",
       "      <td>19.971024</td>\n",
       "      <td>459.339722</td>\n",
       "      <td>97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>388316</th>\n",
       "      <td>388317</td>\n",
       "      <td>7.193769</td>\n",
       "      <td>48.376309</td>\n",
       "      <td>527.00</td>\n",
       "      <td>5.2700</td>\n",
       "      <td>1460</td>\n",
       "      <td>11</td>\n",
       "      <td>2972</td>\n",
       "      <td>6227</td>\n",
       "      <td>24.105930</td>\n",
       "      <td>-287456672</td>\n",
       "      <td>-1.292012</td>\n",
       "      <td>0.108061</td>\n",
       "      <td>523.706970</td>\n",
       "      <td>11</td>\n",
       "      <td>309.591309</td>\n",
       "      <td>440.787842</td>\n",
       "      <td>600.310059</td>\n",
       "      <td>0.088250</td>\n",
       "      <td>17</td>\n",
       "      <td>870673</td>\n",
       "      <td>20.666460</td>\n",
       "      <td>459.003754</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>825800</th>\n",
       "      <td>825801</td>\n",
       "      <td>8.663616</td>\n",
       "      <td>49.674021</td>\n",
       "      <td>2836.25</td>\n",
       "      <td>28.3625</td>\n",
       "      <td>1533</td>\n",
       "      <td>14</td>\n",
       "      <td>2789</td>\n",
       "      <td>6461</td>\n",
       "      <td>19.265608</td>\n",
       "      <td>-253300688</td>\n",
       "      <td>-0.924922</td>\n",
       "      <td>-0.056384</td>\n",
       "      <td>208.336411</td>\n",
       "      <td>26</td>\n",
       "      <td>155.725586</td>\n",
       "      <td>211.511230</td>\n",
       "      <td>500.423096</td>\n",
       "      <td>0.082024</td>\n",
       "      <td>10</td>\n",
       "      <td>649353</td>\n",
       "      <td>20.394165</td>\n",
       "      <td>502.623260</td>\n",
       "      <td>85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1116257</th>\n",
       "      <td>1116258</td>\n",
       "      <td>9.384341</td>\n",
       "      <td>49.143288</td>\n",
       "      <td>1670.00</td>\n",
       "      <td>16.7000</td>\n",
       "      <td>1527</td>\n",
       "      <td>14</td>\n",
       "      <td>2207</td>\n",
       "      <td>6644</td>\n",
       "      <td>8.773687</td>\n",
       "      <td>-330216832</td>\n",
       "      <td>-0.236958</td>\n",
       "      <td>-0.159152</td>\n",
       "      <td>255.992538</td>\n",
       "      <td>26</td>\n",
       "      <td>154.274658</td>\n",
       "      <td>110.248535</td>\n",
       "      <td>181.868408</td>\n",
       "      <td>0.074600</td>\n",
       "      <td>5</td>\n",
       "      <td>780885</td>\n",
       "      <td>19.781750</td>\n",
       "      <td>518.130005</td>\n",
       "      <td>72</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              ID         X          Y        h       hm  BLDFIE_WeigAver  \\\n",
       "211869    211870  6.763981  49.923656  2504.00  25.0400             1533   \n",
       "1243664  1243665  9.833275  49.526917   649.25   6.4925             1534   \n",
       "862800    862801  8.747207  49.396667  4145.25  41.4525             1464   \n",
       "184466    184467  6.692364  48.311297   902.75   9.0275             1503   \n",
       "210193    210194  6.759972  48.427197  1215.75  12.1575             1499   \n",
       "485841    485842  7.421320  48.972783  2629.50  26.2950             1443   \n",
       "120379    120380  6.482271  49.914016  2173.00  21.7300             1538   \n",
       "388316    388317  7.193769  48.376309   527.00   5.2700             1460   \n",
       "825800    825801  8.663616  49.674021  2836.25  28.3625             1533   \n",
       "1116257  1116258  9.384341  49.143288  1670.00  16.7000             1527   \n",
       "\n",
       "         CECSOL_WeigAver  CHELSA_bio18  CHELSA_bio4  convergence        cti  \\\n",
       "211869                13          2114         6069   -51.807003 -204555312   \n",
       "1243664               15          2102         6516     0.582322 -200281328   \n",
       "862800                13          2770         6187    -5.705454 -193031872   \n",
       "184466                16          2458         6333     2.532082 -153840256   \n",
       "210193                11          2473         6283   -34.008671 -184876336   \n",
       "485841                13          2366         6249   -12.187319 -237638256   \n",
       "120379                16          1847         6105   -17.632570 -197189520   \n",
       "388316                11          2972         6227    24.105930 -287456672   \n",
       "825800                14          2789         6461    19.265608 -253300688   \n",
       "1116257               14          2207         6644     8.773687 -330216832   \n",
       "\n",
       "         devmagnitude  eastness        elev  forestheight  glad_ard_SVVI_max  \\\n",
       "211869       1.713994 -0.030101  356.479462            23         168.107666   \n",
       "1243664     -0.082161  0.021150  338.562622            23         663.329102   \n",
       "862800       2.207440  0.082736  471.601471            29         682.764160   \n",
       "184466      -0.687262 -0.040297  326.341187            12        -155.665771   \n",
       "210193       0.869934  0.029229  331.943359            26         214.386475   \n",
       "485841       0.527734  0.184359  317.114014            25          55.184082   \n",
       "120379      -1.088760 -0.035505  272.008911            23        -126.860352   \n",
       "388316      -1.292012  0.108061  523.706970            11         309.591309   \n",
       "825800      -0.924922 -0.056384  208.336411            26         155.725586   \n",
       "1116257     -0.236958 -0.159152  255.992538            26         154.274658   \n",
       "\n",
       "         glad_ard_SVVI_med  glad_ard_SVVI_min  northness  ORCDRC_WeigAver  \\\n",
       "211869            0.701416         -27.481689   0.001850               15   \n",
       "1243664          66.410156         208.560791  -0.054791                8   \n",
       "862800          196.173340         410.016113  -0.105913                9   \n",
       "184466         -179.105957        -273.086426   0.005218               11   \n",
       "210193           39.022705         -19.717529  -0.010323               20   \n",
       "485841           58.465332          91.532715  -0.045335               10   \n",
       "120379         -195.746094          75.838867  -0.008322               18   \n",
       "388316          440.787842         600.310059   0.088250               17   \n",
       "825800          211.511230         500.423096   0.082024               10   \n",
       "1116257         110.248535         181.868408   0.074600                5   \n",
       "\n",
       "         outlet_dist_dw_basin  SBIO3_Isothermality_5_15cm  \\\n",
       "211869                 667874                   19.075720   \n",
       "1243664                801897                   18.012062   \n",
       "862800                 687733                   19.400467   \n",
       "184466                 938821                   19.768513   \n",
       "210193                 937918                   18.850109   \n",
       "485841                 828316                   19.876814   \n",
       "120379                 728450                   19.971024   \n",
       "388316                 870673                   20.666460   \n",
       "825800                 649353                   20.394165   \n",
       "1116257                780885                   19.781750   \n",
       "\n",
       "         SBIO4_Temperature_Seasonality_5_15cm  treecover  \n",
       "211869                             444.208923         95  \n",
       "1243664                            487.378571         89  \n",
       "862800                             449.129852         85  \n",
       "184466                             504.342621         86  \n",
       "210193                             450.874207         45  \n",
       "485841                             445.643494         73  \n",
       "120379                             459.339722         97  \n",
       "388316                             459.003754         11  \n",
       "825800                             502.623260         85  \n",
       "1116257                            518.130005         72  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictors_sel = predictors.loc[(predictors['h'] < 7000)  ].sample(100000)\n",
    "predictors_sel.insert ( 4, 'hm' ,  predictors_sel['h']/100 ) # add a culumn of heigh in meter\n",
    "len(predictors_sel)\n",
    "predictors_sel.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plotting response variables distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x468 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "bins = np.linspace(min(predictors_sel['hm']),max(predictors_sel['hm']),100)\n",
    "plt.hist((predictors_sel['hm']),bins,alpha=0.8);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The Global Forest Canopy Height, 2019 map has been release in 2020 (scientific publication https://doi.org/10.1016/j.rse.2020.112165). The authors use a regression tree model that was calibrated and applied to each individual Landsat GLAD ARD tile (1 × 1◦) in a “moving window” mode. Such tree height estimation is storede in forestheight.tiff and in the table as forestheight column. \n",
    "A correlation plot and its pearson coefficient is show below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fa02257ddb0>]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe4AAAFzCAYAAAD47+rLAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAABcVUlEQVR4nO2de5gU1Z2/39M9NdCD6DAKBkYuigYSFVSIoLiJxhiSaAzBW0zYNVeN2f3lYpYEEp9Vs2ZlQ2Ium9WsJibuxngFJxrcqOsl2RBBQVTEy3pDYCRKgEFlBqan+/z+6K6mpqfuXdXVl+/7PDwzU111zvdUNXOmut76lNJaIwiCIAhCfZBKugBBEARBEPwjE7cgCIIg1BEycQuCIAhCHSETtyAIgiDUETJxC4IgCEIdIRO3IAiCINQRLUkX4IeDDjpIT5o0KekyBEEQBKEqrF279q9a69F2r9XFxD1p0iTWrFmTdBmCIAiCUBWUUq86vSYflQuCIAhCHSETtyAIgiDUETJxC4IgCEIdIRO3IAiCINQRMnELgiAIQh0hE7cgCIIg1BEycQuCIAhCHSETtyAIgiDUETJxC4IgCEIdEWtymlJqI/AWkAMGtNYzlVIdwK3AJGAjcK7WemecdQiCIAhCo1CNM+5TtNbHaK1nFn9eBDygtT4CeKD4syAIgiAIPkjio/KPATcWv78RmJdADYIgCIJQl8Q9cWvgPqXUWqXUhcVlB2uttwIUv46x21ApdaFSao1Sas22bdtiLlMQHPjpT2HJEtA66UoEQRCA+J8ONkdr/ZpSagxwv1LqOb8baq2vA64DmDlzpvzWFKpPby9cfjmccAIolXQ1giAIQMwTt9b6teLXN5RSdwLHA68rpcZqrbcqpcYCb8RZgyCEJpOBm2+GMbYfCgmCICRCbB+VK6VGKKVGmt8DHwSeBu4CLiiudgHw27hqEISKUApOOw2mT0+6EkEQhBJxXuM+GPiTUupJ4FFghdb698AS4DSl1AvAacWfBaG2uOUW+PrXYc+epCsRBEEYRGwflWutXwaGnKporbcDp8bVryBUTD4PV1wBw4fDsGFJVyMIgjCIuOU0Qag/urrguecKZ90ipQmCUGNI5KkgWNEarroKDj8czj476WoEQRCGIGfcgmDlgQdgzRq47jpIp5OuRhAEYQhyxi0IVsaPh4svhr/7u6QrEQRBsEXOuAXBypQpcM01SVchCILgiJxxC4LJD38I69cnXYUgCIIrMnELAsCGDXDJJXDHHUlXIgiC4IpM3IIA8K//Cm1t8OUvJ12JIAiCKzJxC8LGjfCb38CFF8KBByZdjSAIgisycQvC978PqVQh4lQQBKHGkYlbEEaNKtwCdsghSVciCILgidwOJgj//M9JVyAIguAbOeMWmpc334T77ivEnAqCINQJMnELzcu118LcuYVbwQRBEOoEmbiF5qSvrxC48sEPwlFHJV2NIAiCb2TiFpqTX/0KXn8dFi9OuhJBEIRAyMQtNB8DA7B0KcyeDe97X9LVCIIgBEImbqH5ePFF2LOncLatVNLVCIIgBEJuBxOaj6lT4ZVXwDCSrkQQBCEwcsYtNBdbtkA2C8OGFdLSBEEQ6gz5zSU0D1rDuefCaaclXYkgCEJoZOIWmoc//hEeeaQweQuCINQpMnELzcNVV8GYMfCZzyRdiSAIQmhk4haag8cfh3vvha99DTKZpKsRBEEIjUzcQnNw002w//6Fp4AJgiDUMTJxC83B0qXw6KNwwAFJVyIIglARMnELjU9/f+HWrylTkq5EEAShYmTiFhqbLVvgkEPgrruSrkQQBCESZOIWGpurr4YdO2DatKQrEQRBiASZuIXGZft2uO46OP98mDQp6WoEQRAiQSZuoXH5t3+D3bth0aKkKxEEQYgMmbiFxqSvrzBxn3kmHHlk0tUIgiBEhjwdTGhMMhn4/e+hrS3pSgRBECJFJm6hcXnPe5KuQBAEIXLko3Kh8bjxxkIeeW9v0pUIgiBEjpxxC41FLgff/S6MHCmZ5IIgNCQycQuNxbJl8MILcPvtoFTS1QiCIESOfFQuNA5aFx7d+c53wsc/nnQ1giAIsSBn3ELjcO+98MQT8ItfQDqddDWCIAixIGfcQuPw7nfDN78JCxYkXYkgCEJsyBm30DhMmABLliRdhSAIQqzIGbfQGHznO7ByZdJVCIIgxI6ccQv1z/r1cNllhe/nzEm2FkEQhJiRM26h/lmyBPbbD/7hH5KuRBAEIXZk4hbqm5dfhltugYsugo6OpKsRBEGIHZm4hfpm6VJoaYFLLkm6EkEQhKogE7dQ30yeXJi0x41LuhJBEISqIHKaUN/84z8mXYEgCEJVkTNuoT7p6YFbby08VEQQBKGJkIlbqE+uuQY+8QnYsCHpSgRBEKpK7BO3UiqtlFqnlPpd8ecOpdT9SqkXil9HxV2D0GD09sKPfgQf/jBMm5Z0NYIgCFWlGmfcXwGetfy8CHhAa30E8EDxZ0Hwzw03wLZtsHhx0pUIgiBUnVgnbqXUIcDpwM8tiz8G3Fj8/kZgXpw1CA1GNlu4BWzOHPibv0m6GkEQhKoT9xn3j4BvAHnLsoO11lsBil/HxFyD0Ehs2gTDh8vZtiAITUtst4Mppc4A3tBar1VKnRxi+wuBCwEmTJgQbXFC/TJ5MjzzDKTEqxQEoTmJ87ffHOBMpdRG4Bbg/UqpXwOvK6XGAhS/vmG3sdb6Oq31TK31zNGjR8dYplA3vPAC7N4N6TQolXQ1giAIiRDbxK21Xqy1PkRrPQn4BPCg1noBcBdwQXG1C4DfxlWD0EBoDZ/6FLz//UlXIgiCkChJfN64BDhNKfUCcFrxZ0Fw58EH4bHH4LOfTboSQRCERKlK5KnW+mHg4eL324FTq9Gv0EBcdRW84x1wwQXe6wqCIDQwYvgItc9jj8EDDxQeJjJ8eNLVCIIgJIpM3ELt89vfQns7fPGLSVciCIKQODJxC7XPlVfC00/DyJFJVyIIgpA4MnELtc3bbxe+dnYmW4cgCEKNIBO3ULts2gRjx8JttyVdiSAIQs0gE7dQu/zgB7BnD8yenXQlgiAINYNM3EJtsm0bXH89LFgAEnkrCIJQQiZuoTb5yU8KZ9vf+EbSlQiCINQUMnELtcfevXDttTBvHrzrXUlXIwiCUFNUJTlNEAIxbBg88kjSVQiCINQkMnELtckRRyRdgSAIQk0iH5ULtcX11xc+Ijfv3xYEQRAGIWfcQu0wMABLlsBBB8GIEUlXIwiCUJPIxC3UDrffDi+/XLh/W6mkqxEEQahJ5KNyoTbQunC2/e53w5lnJl2NIAhCzSJn3EJtcM898NRTcOONkJK/JwVBEJyQ35BCbTBrFlx1FZx/ftKVCIIg1DRyxi3UBgcdBIsWJV2FIAhCzSNn3ELyfPOb8PvfJ12FIAhCXSATt5AsTzwB3/serF2bdCWCIAh1gUzcQrIsWQIjR8Lf/33SlQiCINQFMnELyfHii4V7ty++GNrbk65GEAShLpCJW0iO730PDAO+9rWkKxEEQagbxCoXkuP442H8eHjHO5KuRBAEoW6QiVtIjs9/PukKBEEQ6g75qFyoPjt2wM9+Bnv2JF2JIAhC3SETt1B9fvrTgpD2wgtJVyIIglB3yEflgidd67pZeu/zvNbTxwEZA6WgpzfLuPYMC+dOYd6xnf4b270bfvITOOMMOPro+IoWBEFoUGTiFlzpWtfN4uXr6cvmAOjpy5Ze6+7pY/Hy9QD+J+/rr4ft22Hx4shrFQRBaAbko3LBlaX3Pl+atO3oy+ZYeu/z/hrr7y88a/u974UTT4yoQkEQhOZCJm7Bldd6+iJZB4CtW2HcODnbFgRBqAD5qFxwZVx7hm6PiXlce8ZfYxMnwqpVEVQlCILQvMgZt+DKwrlTyBhpx9czRpqFc6d4N/T00/DXv4JShX+CIAhCKOSMO0asNrZfA/vSrvXcvHozOa1JK8X5s8Zz5Tx3+9raT3ubwd5sjt5sHoA2I0VrS5pdffss8DWv7ij1YdJZfA3g8rs2lCS0YS0p9g7kbfsd0ZrGSKf42q1PsPTe5zll6mgeem5bqY492Rx92Txozd03fpW0znP6p3+CLk7co9oMLvvokQBccfcGdvYW+mzPGJwxfWypLbt9F2bfCoIgNAJKW3551yozZ87Ua9asSbqMQJTb2FA4O71q/tGOE8ylXev59apNQ5YvmD3BcfK268eNlIK8wyE30opcTmM/TYfnb155nP+67Z9YNPcfuOWYDw2pRylFzqmoItZ9F2bfCoIg1BNKqbVa65l2r8lH5TFhZ2N7Gdg3r94caLlTP264zY/ZGCZtgC+tup2/7NfB8qNOta3Ha9KGwfsuzL4VBEFoFGTijgkn09rNwM45fPrhtNyrvVrg2O7nOGHTeq5/z8fpbzEqassca5h9KwiC0CjIxB0TTqa1m4GddpC2nJZ7tVcLzN68nh2Z/bm57CPyMJhjDbNvBUEQGgWZuGPCzsb2MrDPnzU+0HKnftxIuQjdRlpF/oa4dvY5nHzhdfS22k+qKQVpt6KKWPddmH0rCILQKIicVsSPpWyu093TR1qpkvlt/Wq1sxcvf6pgVdugAHPPZ4wU/QN5cg6Hos1I8S/zpwHw7TvXs7vf/zXtJOno3cWOtgNia18psL5924wUOc0gC77NSNGXzbtmrHet6x5k0pu2u5/j3ylGuyAIMeAmp8nEjT8DPIi9HZed7WaE1xqH7HqdB66/iMUf+n+2UlrSmMcXYOHtT5It27FGWrH07Om+jr8Y7YIgRI1Y5R74sZSD2Ntx2dn1MmkDfOHR5SgNf54wPelSbDGP79J7nx8yaUPhGPo9/mK0C4JQTSSABX+WshjL/jlo907Oe+p+lh/1fv6y/0FJl+OI1zENcvzl/SEIQrWQM278WcpiLPvnM2vuonUgy3/MOivpUlwZ155xPa5Bjr+8PwRBqBYycePPUg5ib8dhZ4O7EV4rtOQGOHf9/dwzZQ6vdNTuNV/z+C6cOwXDZscaaeX7+IvRLghCNRE5rUi5WSyEZ/TbOxiWy7LlgIOTLsWTzvYMp0wdzbK1WxzvAIDCvfSzDxvFM1vfKmWqQyFX/fIzj4wlRz1Mbr0gCI2Bm5wm17gtOD1MQ/CH0nm0SrFtv46kS/FNd08ftz66uXB/ngs5rVn50o4h95yXv2fK7fPunj4WL18PEGjyLs+tz2ld+lkmb0FobuSj8iJBM7+FoSxYdw+3/mYRI/fuTrqUQGTzmqzTTfRllOeq+7n7IIx1Hia3XhCE5kAm7iJiBVdGS26Ai1YvpyU3wFutbUmXU1X82OdB319hcusFQWgOZOIuIlZwZXz02T9yyJtvcM0J5xQizZoIP/Z50PdXmNx6QRCaA5m4iwTN/Bb2oXSei1fdwXMHTeTBye9JupzAGCmFkfY3IZZf4/Zz90EY6zxMbr0gCM1BbHKaUmo48EdgWLGfO7TWlymlOoBbgUnARuBcrfXOuOpwo1wAEsJx6ouP8c7tm/jKGV9Hq/r7W9AuOc2JXH5oLr1VOjO/d7PK/VjnpoDmZpVHZa+X49Ru+fJTpo7moee2Rd6/IAjuxHY7mFJKASO01m8rpQzgT8BXgPnADq31EqXUImCU1vqbbm3FcTuYTNrR0dbfx8c3PMQt0+eSSzXHpxZh88n95OJXsx2/7Z41o5Nla7tdBU7JbBeE6Egkq1wXeLv4o1H8p4GPATcWl98IzIurBjfEzo2O3tYMNx37kaaZtCF8PnlU1nlU7fht9+bVmz3vupDMdkGoDrF+rqmUSiulngDeAO7XWq8GDtZabwUofh3jsO2FSqk1Sqk127Zti7w2sXOj4Z/vu4aPPPenpMtIhDB3IkRlnUfVjt/t/f5/kbszBCF+Yp24tdY5rfUxwCHA8UqpowJse53WeqbWeubo0aMjr03s3Mo56i8v8rfr7mFiz9akS0mEMHciRGWdR9WO3+39/n+RuzMEIX48J26l1Ff8LHNDa90DPAx8CHhdKTW22M5YCmfjVUfs3Mq5eNXtvNnaxq+P/UjSpVSdsPnkUVnnUbXjt93zZ433vOtCMtsFoTp4ymlKqce11seVLVuntT7WY7vRQFZr3aOUygD3Af8KvA/YbpHTOrTW33BrKw457VPXP8LKl3ZE2mYzcdj2LfzPzy/m2tlns/R9FyRdTk1h2uZQuGbc3dM3yET3Y2ObBrd12/aMgVLQ05tlXHuGSQdmWPXyTtuPsf1Y6E51RGmV27Vl7pdq2uhhDfyozf2g7fk9FkmNR4gPNznNceJWSp0PfBI4Cfhfy0sjgZzW+gMenU6jIJ+lKZzZ36a1/o5S6kDgNmACsAk4R2vtOoNGPXHLpF05/3rPj/nYs39gzhdvYPuI9qTLqTmMtAJtf6uZl31tZ3aHZcHsCVw572hfbUZthdv1aaQUKAZFzMZto4c18KM294O2F8TwT2I8QryEtcr/DPwAeK741fz3dQofebuitX5Ka32s1nqa1voorfV3isu3a61P1VofUfxa9RlUJu3KefiwGXz/bxbIpO1ANqcd7w/3sq+jzM03757w02bUVrhdn3a58HHb6GEN/KjN/aDtBTH8kxiPkByOASxa61eBV4ETqleOUC/899STki6hrnGzr6M0s82P0f22GWXfQdqK00YPa+BHbe4HbS+o4V/t8QjJ4UdOm6+UekEptUsp9aZS6i2l1JvVKE6oPUb17uJLj9xWd08AqzXc7OsozWzTBvfbZpR9B2krThs9rIEftbkftL2ghn+1xyMkh5/bwb4HnKm1PkBrvb/WeqTWev+4C4uTOZPr53nRtcan1/6Ob/zxPzn4re1Jl1LTGGlVuJ5rg5d9HWVuvnn3hJ82o7bC7fq0y4WP20YPa+BHbe4HbS+I4Z/EeITk8GOVr9Raz6lSPbbEYZUfvngFA5LBEogRe3tZ+bPPsnr8UVw0/9Kky6lbUgoCxKM7twPkK28GAAWcOLmDjdv7hljw5ebxpV3rSxnqJp3tGdpaU7zwxr5PYuZM7uCcmRO44u4N7OzNlpaPaE1jpFP09GVRFOIUzRo+NXsCMyd2sHj5U/RlB4+uNa3oL14fN7dTCswyzP1q1u5EWilmHzaKZ7a+Nagu6+t+7gDwsuytPx9QvCNgZ2/Wcd8GMfC97PBKbPSwbfuhWlZ7I9jzYa3y+cVv3we8A+gC9pqva62XR1umM1FP3LO+ez+vv9UfWXvNwucfXc6lD93AvL/9AU+Mk7/SmwWreRw04/+IMSPY+NfeIaJeOqXI5zW1/rezk3Vdqflf3m6UxnclbXltG2fbUdEo9nxYq/yjxX/7A73ABy3Lzoi6yGoik3ZwWgeyfP6xLv48YZpM2k2G1TwOmvH/whu7be36XB1M2uBsXVdq/pe3G6XxXUlbXtvG2XZUNIM972aVf6aahQi1TXvfmzw/ehLXv+fjSZciJIBpHjdjxr+ddR2FiW1tI0rju5K2vLaNs+2oaAZ73vN53Eqpn9gs3gWs0Vr/NvqShFrkjZEHcsG530m6DCEhTPPY6/pxI2JnXY9rz9Bd4URgbdepvbB5+GHb8to2zrajolr9JIkfq3w4cAzwQvHfNKAD+JxS6kexVRYjB49sTbqEuuLorS/QuSuRSHmhBrCax0Ez/o8YM8LWrk+nFPXwmB8n67pS87+83SiN70ra8to2zrajohnseT9W+YPAB7XWA8WfWyjkjp8GrNdavzvuIuOwyictWhFpew2L1tzzqy+Tyuf50Gd/WtB4BaECvKx6p9fTCnIRnOxnjBQDuTzZAEp+uf1uV0Z53UaKQX2Yr49oTdPbn0MzOFPeLp/eNNtXPLW1ZMC3ZwzOmD7WNSfeav7b5dY70bWue9BdAHZ9+cmod2v/8rs20NO3z+Yf1WZw2UePDJXZ7vR6JTXWCm5ymudH5UAnMILCx+MUvx+ntc4ppfY6b1a7HL5YJm2/nPzyWt79xiss/PBXZNIWIsHrVjin18NM2gpoSatBMavlt5n5QTt8b6W87vJu8rrwScPu/n3iVE7rkqVvTqxWI7q7p2+Ixd/Tlx20rLunj8XL1wOUzO9la7tLlzRyWrNsbTczJ3YEzjLf3T/ArY9uLgmG3T19LFvbXZGhvXvvwKCfd/ZmWXjHk6X6veqyG2/565XWWOv4DWB5Qin1S6XUr4B1wPeVUiOA/4mzuLiQ+7f9c/Gq23lt5EF0HXly0qUIQmA0DMlGT5Kcw18lQTLl7YjC/LbNlrfJ3K80r93uLoNsTgfObI/CdK9XPM+4tda/UErdAxxP4Q/Yb2mtXyu+vDDO4oRkmbHlGWZt2cAVp36BbNpIuhxBaFiCZsrbUan5XY1s+TAZ/XGa7vWK4xm3Umpq8etxwFhgM4XHcL6juExocI7+y4u8vl8Ht0ybm3QpgtDQBM2Ut8Nqfru97rV9kL6CEiaj32t5M2awu31Ufknx6w9s/n0/5rpipUUu1friVzPP5L0XXk9f6/CkSxGEUCgYko2eJGmH/PogmfJ2RGF+22bL22TuV5rXbneXgZFWgTPbozDd6xVPq7wWEKu8+hyy63W2HHBw0mUIgi+iyn+PG/MBRytf2jHktYyRYk82X8o17+nN0t5m8Pae7BDRzUjBQH6oKGfmn4N9xrmJ1cJubzPQGnb1ZRlupNg7kC9lvp8/azwzJ3b4zv32m4VuNcvjsMrr0SIvJ1RWuWXjNgpn3xO01hcqpY4Apmitfxd9qfZEPXHLpO3OhJ1befD6i/inD17Mb475cNLlCELsGGlFLq+rMvk73U5mJWOkOWtG5yCj2y9eudxBctaDZHw3SkZ4rRA2q9zkl0A/cGLx5y3AlRHVJtQgFz26jFwqxf9MPj7pUgShKmRz1Zm0wXvShoIVffPq4JO2ua2bUR3EXA9iZzej3Z0UfibuyVrr7wFZAK11H9RF6JEQgtFv7+Ds9f/DsqM+wBsjD0y6HEFoWiqJlg1jb1e6fjPa3UnhZ+LuV0plKP6hqJSajOXxnkJj8bnHumjJ5/nZrLOSLkUQmpp0BYFHYeztStdvRrs7KfxM3JcBvwfGK6VuAh4AvhFrVUIipPM5znz2j6yYehKbRo1NuhxBqBpGWuEgfEeOn24yRprzZ423NbD9bOtmVAcx14PY2c1odyeFL6tcKXUgMJvCe26V1vqvcRdmRazy6jFy727a+vt4feRBSZciCA2HkYLzjp/A7Wu2sHfAPnq1PWNw+ZlHAgzKDffLiNY0RjrFrr6sY243MMjsVgq0LvSdzeVLsaxmLU5RpOV2+OnT3DPUze38mOfVssj9tmVdz2r+x2WxV2SVFxvoBCZiSVrTWv8xsgo9EKs8ftL5HDmVkjxyQYiZFIDL7WumUb5sbXeo+FMvnNp3W15uhnet62bh7U8OkeeMtGLp2dMDGe3l7XutE6W97rctLxM/Dnu+IqtcKfWvwErg2xQiThcC/xhZdUJN8LePr+C/f/n/2H/P20mXIggNTR73e85NozyOSdutfbfl5WZ4mMxxczuv9quZTe63LS8Tv9r2vJ+ng82jcN+2CGkNipHL8oVH76T7gNG8OXy/pMsRhKanEqO8kvadlpeb4WGtdT/meTWzyf225aftatrzfuS0lwF5wkQD87Fn/kDnW9u4ZvY5SZciCAKVGeWVtO+0vNwMD2ut+zHPq5lN7rctP21X0553e8jIvymlfgL0Unis538opX5i/qtahUKsKJ3ni6vu4NnRk3j4MNvLKYIgREgKXA120ygPk1nuB6f23ZaXm+FhMsfN7bzar2Y2ud+2vEz8atvzbmfca4C1wF3APwN/Lv68tvha3bJxyelJl1AznPriYxy+YwvXzj5HxDRBiAhzTmszhv6KdbvGnVaKs2Z0MnNiB8Na/HwgOhRFIfdcUcguXzB7Ap3tGRQFSzyl4NerNg25ZjvcSDFzYgdXzT+azuLZY1qp0vXbrnXdpXXnHdvJ0nOm057Z92HsqDbDVUwztztrRmfpzN4cr3UbP+sMzZ8Ld2lh3rGdpfGa+8tOMitfrz1jMKrNcN0mTvxklX9Fa/1jr2VxIlZ5fLTkBvjw8yu5Z+pJ5FLx/IUvCM1GJVnjRlqBJlTcqbWN8km0a103C+94kmzOuV3TjgZiyR2v1CoHbG12ACOlWHqO+x8O9USlDxl5XGt9XNmydVrrYyOs0RWZuAVBqDfSSsUumbnR2Z5h5aL3l36es+RBun0IVObZtt265W0GxakGa7tu6zjVFVV9tYTbxO1olSulzgc+CRyqlLrL8tL+wPZoSxSS4Orf/YBV44/mtukfTLoUQWg4kpy0IZwZ7bVepeZ0FFZ5mPYbDbfbwf4MbAUOAn5gWf4W8FScRQnxM23r/zF/w0M8O/rQpEsRhIYk6TNuOzPazxn3OJcz20rNaacayq1yt3XcxtAsueiO9oPW+lWt9cPAB4D/1Vr/gcJEfgjydLC650urbqdn+H785pgPJV2KIDQclWSNG2kVarvyNmxN8LR7u6YdHVfueKVWuZPNDoVr3M2Si+4ngOWPwN8opUZReMDIGuA84FNxFhYnG5ec3tTXuSf/dTMf+r9H+PGJn2D3sLakyxGEmqU9Y/DmnmygZ3WnleK4CQewfO0WX4LZsJYUGSNdyha3yxI388fNn91oM1LMn3EIS+99nq/e+kTpzL89Y9iecSn2OdmmyW4KXlHkgZdngZ81o5MVT20tZbAPa0mx5tUdQ8bbnjEG7RNr39Z1zTG3tqT56q1P8PXbniSnNZ0Ba44y/zxufMtpSqn/B2S01t8TOa2++f6KH/KR5//EnC/ewM62A5IuRxBCkU4pUlRmX5sYaTXItvZjMYfuK6VAYdufm1ltt53TWCqx0qPM3a50HF63lzn1YeJ3LFHmn0dFRVnlhe3VCRTOsM0Zz8+ZulCj3HnkKXz3lM/JpC3UNbm8jmxCLZ9EzHuXnTK5K+orrx37M7HLxrbbzrb9XGX7Jcrc7UrH4acOtxxxv2OJMv+8GviZgL8KLAbu1FpvUEodBjwUa1VCrKycdAwrJx2TdBmCUNNU21D2Y1ZXi6j6j8tCD7JOJW0kfRyc8Dzj1lr/QWt9JvDT4s8va62/HHtlQuQcuLuHRQ/dwEG7dyZdiiDUPOPaM1W1lP3kdVeLqPqPwkKvdJ1K2kj6ODjh57GeJyilngGeLf48XSl1TeyVCZHzmbV3ceGjd7L/nt1JlyIIFZNOVW5fm5Tb1n4s5tB9pZRjfyZ2ZrXddrbtV2ilR5m7Xek4/NThliPudyxxWfRx4UdOWw2cDdxlCmlKqae11kdVoT4gejkNmk9Q229vL3++9jP8aeJ0vvTxbyVdjiBURErBCYd1sG5TD73ZfCTtpRXYNTWsJcXeAfc+0qqQP+515XaQ4a2gp3eoNW3azdb7ldNKcf6s8cyc2FEynzNGynXs5XW3GSkG8pr+suvL5bWcMnU0Dz23zdaudjKv3YzsrnXdXHH3hpJF3p4xOGP62FIf7W0Ge7I5+ixjGdVmcNlHjwxshHf39JX2cbWs8rhs9FDJaVa01pvV4AdQxPOE9ypx+OLmmrQBFqy7h/337uaaE85NuhRBqJi8hpUv7Yi0PSefy2vSNlKKPGB3EmSa0TA4+7unL0vGSPPD844ZkiduZ0jntGbZ2m5mTuxg5aL307Wum0tufcK1rvK6szk9JIEjY6S5/MwjHW327p4+Fi9fX1rf7rU1r+5g2dpu223MdvdYJuWevizL1naXZaLvez1jpANN2mY/lU6WYdpw219x2uh+rPLNSqkTAa2UalVK/SPFj83rlYFkkwirzrCBfj63pos/TjqWp99xeNLlCEJDkc1rcg6zvmlG+7WW/RrSS+99nqCfM4S12a2Gvd1rN6/e7Dq2MG3Wqs1dTlL1+znj/iLwY6AT2ALcB/x9nEUJ0bLf3l7+PGE6Nx8zN+lSBKHpCJL97deQjtJ2rjQn3CnW1avWODPRq0VSNrrrxK2USgM/0lrXbUqaANtHtPOVMxcmXYYgNCVBsr+98sTN9f3mjgepz61dtzE4ZbJ71RpnJnq18JO9HgeuH5VrrXPAaKVUa6xVVJmWJkpan7HlGaZs25h0GYLQsBgp5fjwBtOMtrOWFXDK1NGDlrlZ7FbLeeHcKb6uc5bX6WSzd63rdn3s5ylTRzua1+fPGu9qZHtlj0dhc5v1H7poBXOWPEjXuu5Ar4clKRvdz0flG4GVxUd7lu4j0lpfHVdRceMjtKcx0Jor77uGlM4z97P/DqqJ/mIRhACklWL2YaN4ZutbJfvZDTPfu7NoYd/62OYh14/bjBT/Mn9aSVJa8+oOblq1qWSeaygJZ1aRyemj57Nm7JOnzK+Llz81SOyycsSYEfT25wfZzjA0f7zQjn1kqMmtj21m5sQOrpp/tK1BbbXdy81qP7nnlVjZXoJYnAJZlJnuQfBzO9hldsu11lfEUpENklUejve/+Cg3LPsOl5z+NZYfdWrS5QhCTdHZnmHlove7ruN0Fmrd1s86lbblVq/f/p1w6zNMe9XGa/yV7p+kqOh2MHOCVkqNLPyo3464PiEOtOZLq25ny/5juOtd70u6GkGoOaKKwvQrKFXSVpB+/LQV53rVxmv89RZn6gc/yWlHKaXWAU8DG5RSa5VSR8ZfmlAJx2/ZwMzuZ7nu+I8zkJZnwghCOVFFYfqNy6ykrSD9+GkrzvWqjdf46y3O1A9+/IbrgEu01hO11hOBrwPXx1tWvDTDld6JO19j0wEHc9u005IuRRBqjiijMP0KSn7bspPT3OI/KxWk3CJD/fSfNF7jr7c4Uz/4ucb9pNZ6utcym+3GA/8JvAPIA9dprX+slOoAbgUmURDfztVauz71QiJPw5HO58il3P9DCoKV9oxBT1/W8RafaqCAEyd3+BbFwmCkYCCPr0jP29dsGpLSZkaQXjnv6CGRnm1GitaWNLv6nONM3USmrnXdXH7XBnr6Cu35if+sNHbTun15BKlb/3ZxptYkNqc+DnCJfA3DpV3ruXn1ZnJa28bDtrcZaI3tMXGrs1qymR1u17j9TNx3Ao8D/1VctACYqbWe57HdWGCs1vrx4vXxtcA84NPADq31EqXUImCU1vqbbm1FPXFP/fY97GlgtXzKto08f9BEscibnAWzJ3DlvKNLP1/atZ5fr9rkus0RY0Zw/yUnO0ZvVosUcHUxDtSuFiOlQA19jnYYMka6LH5zXz8p5RyFCjBncgePb9rlup/M9pP45R8nXeu6WXjHk0OOgZFSLD1nuq8oV5NK9pHf94efPuzaSur4uU3cjh+VK6XMifp/gdHAcuBO4CDgM16daq23aq0fL37/FoWY1E7gY8CNxdVupDCZV5VGnrQn7ejmnl9+mc+svSvpUoSEuXn1Ztef7XjhjcIdn27Rm9UgX6zBqRa7+M6wuMVvuk3aUMhL99pP9RThGYSl9z5vewyyeR0oyhUq20d+3x9++qiXCFY3a2mGUmoicAFwCvtuXYSAl4mVUpOAY4HVwMFa661QmNyVUmMctrkQuBBgwoQJQbprai5avYyBVJrfTX1v0qUICVP+UXeQj75rwbiNI97Tq696bT8Jooxy9btOpdv5jZStpI9q4Can/Qz4PTAVWAM8Vvy6tvjVF0qp/YBlwFe11m/63U5rfZ3WeqbWeubo0aO9NxA4+K2/ctbTD3LbtNPYtt+opMsREiZddqmk/Gc3asG49bKCo+4rzn5qYX9GTRADvhKDP8rtvNatFwPdceLWWv9Ea/0u4Aat9WGWf4dqrQ/z07hSyqAwad+ktV5eXPx68fq3eR38jQrHEJjhPh7iXo98/rEuUjrPfxw/P+lShBrg/FnjXX+244gxIwB/pnGcpIo1ONViF98ZFrf4TYf00RJzJnd47qd6N5idWDh3iu0xMFJDDXSv91Ml+8jv+8NPH/VioHvKaaEbLjzA+0YKItpXLcuXAtstclqH1vobbm2JVe5NKp/jf35+MU+Mm8IlZ3w96XKEmLBer4py3Urx01d7xuDIcSNZ9fLOQR/bZ4yUY2ynW1ujR7aWrsmHrc80poFBJrfZx+VnHsmaV3eUjOVyzMjTh57bVrKQJx2YKY3Rap/XI16GdblVrhRoXdgv5RGrVpM8SqvcyWy39h3E7vd6za9FXykVWeUVdHoSBbFtPZQeHfstCte5bwMmAJuAc7TWO2wbKRL1xD3ru/fz+lv9kbVXKwwb6GdEfx872g5IuhQhIsoN3U9d/8iQW5NqHS8rN8kxGSnFeceP59ZHN5MtM9GMtGLp2d52tHV8tWQlV0qQsdia3WkFmkH7Nep9EWZ/hz1GQSz6KAhllVeK1vpPWmultZ6mtT6m+O8erfV2rfWpWusjil+r/j+20Sbt1oEsLbkB9ra0yqTdYJQbuvU2aYO3lZvkmLJ5zc2rh07aULiVyI8dbR1fvVjJfggyFluzO6eH7Neo90WY/R32GAWx6OMmtolbqB4L1t3Dw9ddyKjeXUmXIsRArRmtYajlMbjZ9kHzxuvFSvZDkLFEaXYHIcz+DnuMwrQZFzJx1zlGLsvnH7uTLQeMYaecbTcktWa0hqGWx+Bm2wfNG68XK9kPQcYSpdkdhDD7O+wxCtNmXDTlxH3wyNakS4iMeRseYtxbf+Wa2eckXYoQA+WG7pzJHQlWEw4vKzfJMRmpgjzmNx+8mXKxg4zF1uxOqyH7Nep9EWZ/hz1GQSz6uIlNTosSscrtSeVz3P+LL9FnDOeMC34kEacNzIjWNLv7c54RnEkxrCXF3oGhZvioNoPTp43ld09uHZS9/e6xQ+3yIGSMFLm8pt/mmuOcyR08sXkXu/vdE82MFCw9Z1+sql0+OBSubVqf52w11Mszy+3WbWtN09ufSzT32o2whrWfdspz3s19Ds7Gt9/a3NYBXO3vuNqNkkSs8iiJeuJuhEkb4JSXHuOXd1zBlz62iHumnpR0OU1JrU6kTqRTihTYylhBsLNwnTKjc1qH3kdhti+Z4o9t9oxFLc9zt+I3rz1jpDlrRifL1nb7WreWDPM4LXi3bHwjrTxzxJO0v2vh7oBErHIhfh4+bAYXnHMFv3/nCUmX0rTU06QNkMsPNX3DYGfhOmVGV9JdmO1LpriPLHO3/Ha/ee192Rw3r97se91aMszjtODd9q2fHPEk7e9avzvALatcqHG0SvGHw2YkXYbQpITJo64Wfj+CD2KUR9Ff0HbjJk4LPuhlkKAGv992gmxbad/VQs6465T/WH4ln33st0mXITQxYfKoq4XfXPYgRnkU/QVtN27itOCD7BO7PpO0v2v97gCZuOuQ47qfZe4Lq1BVC7QUnPDKsq410qmhpm8Y7Cxcp8zoSroLs33JFPeRZe6W3+43rz1jpDl/1njf69aSYR6nBe+2b/3kiCdpf9f63QFNKadBfQtq1y/7DjO3PMuci2+gt7U2/gIUaoOWlGIgxEVl89ec05ZWkzpjpBhupEs5022tKcfM8BSQKRrxQWkzUrS2pOnpyw7qf0RrmrzWtvnmCjhxcgcbt/fR3dNXys4GaC0KUZrCH1zDWlLsyeZdTer2NgOtGZRhbt0fnZZtrVZ5WhWkunafmdx+7W0/2eF2r0dljgeha103375z/aBjnzFSXDV/GhCdVe7Ud6X2d1z7xS9ilZdx2tUP+344Qa3xzm0bue+Gf+CHcz7Jj0/6ZNLlCA1CCkiXmb61RhB7u3w7qw3sZgwDtq/Z9RuVZezXYA6bkx5n7ZWOSXBGrPIy6nXSBrh41R3sNobzqxkfTboUoYHIM9T0rTWC2Nvl21ltYDdj2Ok1u36jsoz9Gsxhc9LjrN2JWrey6x2xyuuMX848k/899Fh2ZUYmXYogVJ2wgS1WGziMMezUbxSWsd96wuakx1m7E7VuZdc7TXnGXc88NfadLD/q1KTLEIRECGoqm1htYDdj2Ok1p36jsIz9Gsxhc9LjrN2JWrey652mnLiPGDMi6RICM/rtnVz13z9h3JtvJF2K0ICkGGr61hpB7O3y7aw2sJsx7PSaXb9RWcZ+DeawOelx1u5ErVvZ9U5TflR+/yUn151V/tk1v+W8p+7nullnJV2KECGmfdzpYWdb6WzPcMrU0ax4amvJmoWChT3MYnvbreNEHshbrnGPajNoTashz65PK8Vho9t4eVtv6I+traQUTB49Ykh75hgfem7bkKzoFU9tLV0/Ne1u6360267c+rbeYqbYd/114dwpXDX/aEeb+ObVm8lpTVopjptwAEvvfZ6v3fpEaX9b+510YKaUx27NNIehxvJZMzoH5bkPN4aeU5k1ONXm9vrMiR2hDekwdrVXrUJlNKVVPvXb97CnxkUcK/vveZuV136GPxw6g3+YtyjpcoQIMNKKpWfvy012y3W2kk4pflBB3rKRUuQpRJ9GjfWWrSD4tY29ssPd2gmSO27Xht/tvVgwewIzJ3bYHhcUnvnd1Ubs8OQQq7yMepq0ARasu4eR/X1ce4I8urNRyOYG5ya75TpbyVWYt5zN61gmbQg3aYN/29grO9ytnSC543Zt+N3ei5tXb3Y8Ln7yu6uN2OG1SVNO3PXE8OwePrvmt/zh0OPYcPDkpMsRIsRq2MaRdV1PBq+fWitZJ8i+sFs3qn2Z07riWqqJ2OG1iUzcNc6wgSwrpp7ET084N+lShIixGrZxZF3Xk8Hrp9ZK1gmyL+zWjWpfppWquJZqInZ4bdKUE/fwGrdnrezKjOSy0y7msfFHJV2KECFGenBusluus5V0hXnLRkqRjilgPWyrfm1jr+xwt3aC5I7bteF3ey/OnzXe8bj4ye+uNmKH1yZNKadBfWSVn/Dqkwyk0jJpJ4iTcBVWxLJub2Zqm9atl1U+ojXNx4/rtDWmnbAzgoEhGdKDalNw4mEdPPrKDsrjwK3WtjWT2/xqpBiyTTnlWed+8rTtxmPmiO/qc88BN7dzGnPGsM8s99qX5Ra5db+U42aVZ4wUfQN5rL+K3ez4OHF6v1j3+Z5sblBO/Kg2g8s+GiwHXPBGssrLOHTRipp/rpbSee79xT+QS6X48Gf+rfDbVKgZwuZmB+3Db8Z20IcnBDWko7CtjZRiqYMRH6e93LWum4V3POkY6ZpScPW5x0Qy8TjdHbBg9oTSpO1n/TmTO3h8066ayhfvWtfNwtufJGsjN5bfJSFUjljlZdT6pA3wgRcf5Z3bN3Ht7LNl0q5BwuZmB+3Db8Z2EMIY0lHY1lkXIz5Oe3npvc+75rDnNZFZ0k53BwRdvvKlHTWXL7703udtJ20YepeEEC9NOXHXPFrzpUduZ9MBB7Ni6t8kXY3gQBQBJF5UmrEdxfpu20XVd5z2clTGuh+c3hNBlzuRZL64V99imlcPmbhrkBM2refYrc9z3ayzyKUqF2KEeAibmx0EvxnbYdusdLuo+o7TXo7KWPeD03si6HInkswX9+pbTPPq0ZQTd61/8HzQ7p08d9BEbj/6A0mXIjgQNjc7aB9+M7aDEMaQjsK2NlyM+Djt5YVzp7jmsKcUkVnSTncHBF0+Z3JHzeWLL5w7pZDwZkP5XRJCvDSlnAZ1YJVrLde2a4wUhUxvE6UKh8k0qk072Y+FroDWlhR7BwottmcMzpg+tmQRm8Z0T1+21H57xqB/IEdv0ei12rxuxrPVvj4gY6AU7OzNklKF67sw1PQOYjRf2rV+UIb37MNG8czWtwZlpLdnDC4/0908DpOJ7daGOdae3sK4s7n8EKu8zUjxL/OnDRIA7bYPsk/K94fVJi+v9Yq7NwzaT0rBp2YVRLYo9kdQvPrsWtfN5XdtKGWqQ2Nb5UkcAxOxysuYdtnveXNvfFJRJczY8gzrxk0hLx+RJ0rQ272MtALNIHkniIldbu8GydUGIsnRjspQTyrLupIscz/bB23Pq1Y7093NvBeqS9LvbbHKy6jVSXvy9s3cftM3+cJjdyZdStMT9M/ZbE4PMW6DmNjl9m6QXO2ocrSjMtSTyrKuJMvcz/ZB23PDyXR3M++F6lJL7+1ymvKxnrXKRauX0d9icMdRcm27UQhiYvu1d/20FZa4LfE4qdQeD1NzlIZ+pW0K0VJL7+1ymvKMuxYZ++Y2Pr7hIW6Z9kG2j2hPuhwhIoKY2H7t3fJtorR547bE46RSezxMzVEa+pW2KURLLb23y2nKiXv/YbV3/fgLjxY+Hr/++PkJVyJA8DsPjLQaYtwGMbHL7d0gudpR5WhHZagnlWVdSZa5n+2DtueGk+nuZt4L1aWW3tvlNKWcBnD44hUM1MjQlc7zu199lWfHHMo/nv61pMtpelpSioGQz6wu33ZYS4r+AQfTvGilQyGH3EinSrnbbrnXpe0pXIsf1WawN7vPNjeXmzb6KIuhbu2zzUjR2pJ2zPouN4hHtRm8e+xIVr28c5Ax/cq2t1n50o5BtVlNeKuZHSSX3Irf9cNY4VYLXCnItKToy+Y9t7fLbO8sy/e2e81qsFut8vI7C+yywqttNrtRTeM6qb6C5OFHjVjlZTjlAydJKp9jRHYPbw0bkXQpTc3+w9I1IS+a9uqaV3dU9F51M8/dDFm3XOpKcMp496qlUrvXqY3jJhww5I8OcM4Wd2vPxEgpUNjKZ0HHGeRuhWpSTeO6UfvyQqzyMpzygZNgeHYPmf495FNpmbRrgFqYtGGfvVrpe9XNPHczZN1yqSutxy7j3auWSu1epzbsJm3w/h3hZqBn89oxGz3oOIPcrVBNqmlcN2pfldCUE3c1Mqb98reP38PKn32Wg3bvTLoUocZ4racvkvfqaz19iWSEO+E0pjhrDDoer/1eyf5JYjxRU833TaP2VQlNOXFXI2PaD60DWT6/potnx0ziryNGJV2OUGOMa89E8l51M8+TMGedxhRnjUHH47XfK9k/SYwnaqr5vmnUviqhKSdup3zganPW0w9w8Ns7uGb2uUmXIhSplTsOTHu10veqm3nuZsi65VJXWo9dxrtXLZXavU5tzJncYbu+1353M9CNlHLMRg86ziB3K1STahrXjdpXJTSlnAbJx56m8jkevP6L7Bq+Hx/7u6sll7xGOGLMCF54Y3fo7Q8e2crrb/X7WneUxVi1yxM3beZJB2Z45OUdBL3kbKQgawlXb00rRgxroafXvT+r+e1mlZsGdq+1kyJBrHIjrei3XBM+YswIevvzg+ztFU9tHWRfm7nndjnpG7f32drHTmay32zxcsz2rOa/uf3MiR2eVrlbm9W0ysMa20G2q9QKb1SD3Q2xyss47eqHK/rlHAXvfXkt/3n7ZVw071vcO+XERGsRosP64A47wuRlh8kxdyKdUpx//PghVneQGv3U6ZdPXf+IoyDmRBDjPm4juJYs5DBUo/5630dJIVZ5GUlP2gB/PGwGH1/wfe575+ykSxEixOusOExedpgccydyeW1rdQep0U+dfgk6aVv78WPcx20E14uF7EQ16q/3fVSLSFZ5EhQf2bmuc2rSlQgJECanOkyOuRN+THWvfpK2b1/r6fP9IJg4a0p6P1RKNeqv931UizTlGXfS/Odt/8SXV96cdBlCQoTJqQ6TY+6EH1Pdq5+k7dsgxn2cNSW9HyqlGvXX+z6qRZpy4j5iTHJBJzO3bOC9G9exa/h+idUgxIeXiB0mLztMjrkT6ZSytbqD1OinTr84Wd1uBDHu4zaC68VCdqIa9df7PqpFmvKj8vsvOZlJi1Yk0veXHrmd7Zn9uXXaaYn0L0TDiNY0/QO5Qdb2iNY0u/tzroJaXzbHFXdvABgk5gzKyy621dufK1mta17dwddufWLIx8PmeppCRnlrS4q9A4Wiyq1yKFzjthO6zAzztFKDrj+WZ5dbbduzZnQOyuo2t7t9zaYheeZOlvY5MyfwyEs7GOql76vrxMM6bE1xszarEX7Y6DZe3tZb+vmsGZ2BDOmgFrj5WpQWsh+rOSrzudL6/dQRxz5qdprSKp/67XvY4xBJGCfveuNl/vuXX+b7f7OAn574iar338wYacXxk0b5lqFSqjCJ2WVEw9Dc7zD1LD17eul2JLvJ1MzLdsvWt2ZqB83j9sLLZnfKHfeq08RvHrp1X7kR1l52M/WrbT/7GUOtWNq1UkejIlZ5GUlM2gBfXLWMt1sz/OdxZyTSfzOTzelAk1e+7MEO4J77HaYe86zWyY42l7vZ09bXguZxe+FlszvljnvVaa3XTx66dV+5EdZedjue1baf/YyhViztWqmjGWnKj8qT4kcnfZLfv/ME3pTr23VLHLatk+VtLnezwK2vxZnd7NS23yx1u/WizuYOay9X+nqU+BlDrVjatVJHM9KUZ9xJ8UpHJ/899aSkyxAqwC33O0xb4Gx5m8vd7Gnra3FmNzu17dfstlsv6mzusPZypa9HiZ8x1IqlXSt1NCOxTdxKqRuUUm8opZ62LOtQSt2vlHqh+DWRJ2sMd8gRjovRb+/gmjv/hUN3dFe1X2EfRloFMphTCseM6ErNbrMe06p1sqPN5W72tPW1oHncXnjZ7E654151Wuv1k4du3VduhLWX3Y5nte1nP2OoFUu7VupoRmKT05RS7wXeBv5Ta31Ucdn3gB1a6yVKqUXAKK31N73aiiOrvJpW+aKHbuALj3Vxyhf+g02jxlatX6EyFJQsbqP4J65NLHckWO1wOxP70q713LRq0yCr3BS+rGZvee54ec63E0rB4aNHDDKyy2uwiyftLPbx0HPbSsawU7a6XZvleejWfQ6FfPTTp40d1L5bdrmdHe5kiVvXNa162Bdba12/a103V9y9odRnxkgx3EgPyWCPgqBWuVPWfNA2w9Zq3S/WYxFmXMI+EssqV0pNAn5nmbifB07WWm9VSo0FHtZae/55FvXEfeiiFb5Tlypl/z1v8+drP8MDk4/nK2curFKvQq0QxEQPY+S6mb0AC+94kmxIGdNaj5vZ7sd6tmJnmDvhN5vdSCmWnjPdM8/dz/GwG4/XfkzKpk7aQvfbthjowaklq/xgrfVWgOLXMVXuH6BqkzbA3z3+O/br7+Pa2WdXsVehVghioocxct3M3qX3Ph960i6vx81s92M9W/GTMe63LZNsXrsa8OW1BjHJ/ezHpGzqpC10v22LgR4tNWuVK6UuBC4EmDBhQsLVhGN4dg+fWXMXD0x+D8+NOTTpcoSEiNqe9rN+VGavl/lu159X335NdD9t2a0bhSUeZDxB14uSpC10v22LgR4t1T7jfr34ETnFr284rai1vk5rPVNrPXP06NFVKzBK0vk8Nx3zYf79hHOTLkVIkCAmelAj183sjcLu9TLf7erw6tevie6nLbt1/VjiQUzyuI5dFCRtofttWwz0aKn2xH0XcEHx+wuA31a5f6AgwFSD3cPauPq9f8vjne+qUo9CrRHERA9j5LqZvQvnTsGo4A4Kaz1uZrsf69mKn4xxv22ZGCnlasCX1xrEJPezH5OyqZO20P22LQZ6tMRpld8MnAwcBLwOXAZ0AbcBE4BNwDlaa89Yp3q0yk9+6TFacwPcd8TsgrIr1DUHj2zl9bf6By2zywK3ooA2S+a4aV93O3w8qIBPlUWYWi1c6/ZWCxr2ZY3bmdNW6xcKBntea/rKijdN6Z292ZKJ3W4xltss5ruJU563nQVvjrF8WbmJ7GWGl1vlbUaK1pY0PX376i7fPyZGClpbCpny1nrK8+XbjBTDLNa4kylvbcNPtnkQ/FrYcWabR9l2NazyRjLXE7PKoyLqifvwxSsYiHHYqXyO+3/x9+xpaeX0T/9YJu46Ys7kDh7duNNTRgqTW15u0bpllM+c2BE4D91PpreXpW2kFdjEvTqNwakPPxnkg/otWuHg3/b2Mx4/pIozbxR3+lXb1o6TWqghCPVWrxe1ZJXXBHFO2gAffGEVk3dsKZjkMmnXFStf2uHLxA6TW15u0bpllIfJQ/eT6e3VbjanXSdcv9nfQSZt2GeFB80NjyI3Ph/RpA3Vt7XjpBZqCEK91VsJNWuV1y1ac/GqO3hl1FjumTIn6WqEGAljxFq3ccsoD2vbViN3O64+wuSR16KVXE1bO05qoYYg1Fu9ldCUZ9xxMufVJ5n+lxf4j+PPIp+qLBZTqG3CmNvW9d0yysPattXI3Y6rj6C2dyV9xUk1be04qYUaglBv9VZCU07cLTF+ej18YC9rOt/F8qNOja8TITbmTO7wZWKHyS0vt2jdMsrD5KH7yfT2atdIK9f8cL/Z334yyAf1W7TCg+aGR5Ebn1LR/SKstq0dJ7VQQxDqrd5KaEo5DaqbVS7UF3bms90yq83st53y7O2MkaJvII/W+7K8Z07sKFnVVtvZSEFO42o2m31c9lH7vGgoy+i22Q7wfN2aL27XX3kGuRttRor5Mw5xNObBPufcbjzlFno5RgpSqRR7BwpXtU2Tf+bEjkH1GikYyA/Oqt9vuMHO3qztcYWCHX/G9MG56uakEYXRXZ4JX4kx7de+vrRrPTev3uyYXx+mzUpriruNWkGs8jJmfff+Ibf2RMHJL63hkYnT2NvSGnnbQuNiNV+DWNIZI81ZMzq59dHNQ2QwL8Pcy8B1ej1If34yvqH4FDaFb5Pf76RnZ7anFKRTalBffrPJU0A6rRzrTKcUKQYb+XZjizqTPszkFnW+eKX1NZoRHgVilZcRx6R9+F838as7Lufzj94ZedtCY2M1X4Na6jevHjqJgrdh7mXgOr0epD+/WenZvA5k8vvByWzP66F/IPjNJs/j/sdFLj/UyLcbW9SZ9EGJI1+80vqayQiPgqacuOPg4tV30GsM4zfHfCjpUoQ6xG/Odjluud9ubXkZuE6vB+kvDps3rtzwMNnklZBkJn0c+eKV1tdMRngUyMQdAZ273uDMZ/7ALdPmsrPtgKTLEeoQvznb5bjlfru15WXgOr0epL84bN5qZL5Xw0KOMpM+qr4ryRevtL5mMsKjoCkn7oNHRnsN+guPLkejuP74j0fartAcKKC7p485Sx7klKmjfVvSCvczYLPNrnXdg5Z3reumt3/AdRu7OjJGmvNnjcdOGLcz2k+Z6v/hQH4c9CCGcJC+AXbvHeDSrvUc+537HCNpU2A79tLrdvslpYbcpWCOo2tdN3OWPMihi1YMOU7lr9kdD/N9c+x37uOYK+6zbceOOPLFKzW6/Wzvtr+ajaacuLe/Hd01bqXzvGvbRrqOPJmt+9fnU8yE6DlizAjfD7Mxp97unj6Wre3mrBmddLZnUBSM7faMgaJgLY9qMwB7Y92O7p4+Fi9fX/olZ0pA1uxyu23K6+hsz3DV/KOZObHD9qz7vPeMHyI2LVtr/4t1WMvQXzteYxnVZgQSnZz6dqKnL8uvV21y3C/tGYNPznZ+vPCwltSQ/aKA844fz9Kzpw/Zj1CIdu3u6UMz+DiZx8j6mvV4mG2b+2xnb5aevuyQdpyYd2wnV80/ekhN5fvW73pB1w1Tk90+8RpnI9OUVnnkt4JpzfCBvewxhkfbrlC3ON2O5IfO9gwrF73f8fU5Sx50PCv0ajPItnZ1OG1fvq7bekDo+v0QZv/46dutXafj7VR3mP0T5BgG2V/1gN/3XSPhZpVL5GkFtPX3MWygn51tB8ikLQwi7KQN8USKhpHfKpGQopaNKq27EvzsO6fjHXQ/+BEKw0TD1jsirw2mKT8qj4q/e3wFf/rZ5xj9tueTSYUmw03i8iKOSNEw8lslEpLbelEKVZWuG6Q9t3adjndQ6cpt/wQ5ho0mdYm8NpimnLijiDwdlt3L59Z0sbbzXWzbr6PyBoWqkVL4iuRMp9zjP50wUoWEqXLZxitOFPxHigaJ+bS26XfbSiUkt/Uqqd8PUcSgmphRrGa7dsfPSNsfb7e6g+6fIMewEWM+mynO1A9NeY0bKr/OvWDdPVx53zWc/4l/4ZGJ0yKqSqg1WtOKttYWX9GdJmZM6ag2A61hV1+WAzIG2Vye3f37QiaUghMP62Dj9j7HWEvzZ2ucZ2fZegdkDJSCnt4s49ozTDoww6qXd5LTGgW0tabp7c+VIiBvX7OJlS+5f0o0ojWNkU6xqy87KLrzirs3DBG4yqNS3cay4qmtrmIcQLq4/8zfTKaIlVaK2YeNYuP2vkH7o71s/AvnTmHNqztKUZ0pVWgz6/Lszk6f9SlV+MPf2lZ7xuDIcSP580s7SjWPaE3z3Y97P7fcKZ7TK7rT+nq75X3mtW75cSl/70QZERp1/GgjxZn6QSJPyzjt6od54Y3dobdP53M8dN2FbG9r5+N/+3155nYTYM0MD0IpJvSxzbapW9ao0KBxp36jJ6MYh5FS5Cmkg9m+nlagB8d9lkeo+ok/jQKvCFWvKM1PXf+I5x826ZRy3BcmC2ZPcMz1rhZB3lMQXcyoRJhWjkSellHJpA1w/OYNTNj1OteccI5M2k1CmMkOLDGhDpOINSo0aNyp3+hJK2HHkc1r14kqmxsa91keoVqNSRu8I1S9ojS9Jm1w/gPGys2rN3uuEzdB3lMQXcyoRJjGi1jlIXhk4jRO/dy1vHyg/OUoeONlmIeNOw1ifSdF2LHFTTXqqeTOgqiIMhI1ijZq7X1QrzTlGXcltOQKiVMvHTQerWT3Cd54GeZh406DWN9JEXZscVONeiq5syAq4jb4g7ZRa++DeqUpZ54jxowIt6HW/OaWb/GtB38RbUFCzRNCLgf2xYSWx16aWKNCg9jQQaxvK2HHYaQUaZeN7Yz5chPaaR9EjV3MqFNddsyZ7H2XiNu+MDl/1njPdeImboM/SL/NbIFHTVN+VH7/JSeHsspnbX6a47c8w93vem8MVQm1TJBrw9Y4yr5sjmVrt9CaTpHNDb3WmM1pLr9rA0BJ2rHawnuyOfqKCrMplnVaLO9jv3NfyYJuzxhcfuaRnDWjs2RUhx2HdQxmu242ejanyRgpWltSJXPejDY1beBsTvuS444YM4Le/jyv9fQx3EiVxm/Smlb0u1zDVgrX1/uyOb566xN89dYnSuMsN9OPGDPC0YVpM1LMn3HIIGPeavKnVeH2sCBimpMxbS4vv6vAyai2a+eq+UeHssov7Vpfeh+5jcnN9g5igVdijTuZ8373W73RlFb5tMt+z5t7/QsbJjfe9k+8+/WXOemLv2CvMSyyegTBSCmWnjN9SDazk5XrZGkXbn1Sts+iDkvGSHPchAN8SVvl2BneRlqRy2mc7s6ymvaXdq3n16s2haw8PObdAMvWdtvKXVEa0k7HOmj/UZrcTvu93JSPqs9K2oniboxaRKzyMsJM2kf+5UXe98rj/HLmmTJpC5GTzetBxq2XletkaefLbsmKgr5sLtSkDfaGd9Zl0jZfN8eZlJlt3g3gNBlEaUg7Heug/Udpcjvt9/LlUfVZSTtR3I1RbzTlxB2Gix5dzputbfzXcacnXYrQoFiNWy8rt9HtXHN8SZrZfu8GqBSndoL2H6XJ7dR3+fKo+qyknaj6qidk4vbJP7//83z5zG/w1rCQYpsgeGA1bqPIq65nzPElaWb7vRugUpzaCdp/lCa3U9/ly6Pqs5J2ouqrnmjKiXv/YcFzjLft18HDk20vNwhCxVgzscHbynWytP3msAchY6R9mdZ22BneRlq5/uKxmvZJmdnm3QBORnaUhrTTsQ7af5Qmt9N+L18eVZ+VtBPF3Rj1RlNO3E9d8SHf6x781l/5r1suZcq2jfEVJDQ0KQp53m60tuz7r9i1rpvL79ow6LqdAs6a0VmSauYd28nSs6cPaTevoSWthiwvTOhD+zWn1PaMwag2A1X2fWd7hqvmH81NXzjBdvIeZqm7PWOwYPYE2jP7+ra7TWwgpzlhcseg9ax1mte4u9Z1c+W8o1kwe0LFAYWKgo3uRJuRKvWRVoqzZnRy5byjuWr+0aVnZJtnm+Y+cRKcutZ1M2fJgxy6aAVzljxI17pu19rmHdvJWTM6S+1X0r/1eKTUvmu6djW41Wnud2tNpphmbjdp0Qq+ftuT9GVzg47PcLs3mgfzju0sjdX6vvMjkdltu2D2hMDHrZ5oSqs8yK1glz5wPZ9eezcnX3Q9Ww44OLIahPrEKwfbiTmTO3jkpR2uUpaRVpz3nvHc+uhmW8Gs3DwHHO1yp3UrMXe/fvuTtlGfVtM4iOHrtZ1ZG8Altz1hexvZnMkdPPrKTl/7K4i9HdY+DrOPozCzvfZ7eXth+/R7fOvJ3q5VxCoPSXvfm5z/5L3c9e73yaQtAN452E6s9Ji0ofDHwM2r7Sdts287k9iuHqd1KzF3nfK5raZxEMPXazuztqX3Pu947/fKl3b43l9B7O2w9nGYfRyFme2138vbC9un3+PbKPZ2rdKUASx++fTauxmR3cO1s85OuhShSYjKJA6ybqXmrrXmIMaun+0qNYD9mPpO+zzKnO8gxylM/0GPYdg+o65JCIeccTvQ1t/Hp9fezf2Hz+KF0ROTLkdoEqIyiYOsW6m5a605iLHrZ7tx7ZmKLGA/pr7TPo8y5zvIcQrTf9BjGLbPqGsSwiETtwM5leLfTvwEPznxE0mXItQQXjnYTsyZ3OH5n81IF2IlnazwcvMcnO1yp3UrMXed8rmtpnEQw9drO7O2hXOnOGasz5nc4Xt/BbG3w9rHYfZxFGa2134vby9sn36Pb6PY27VKU35UvnHJ6Z6C2l5jGL94z7zqFCTUBe0Zg9EjW0M9z33Da28x3EjRm3W+0j2Q0/x61SZH+3m/4UP/u655dQcDNtd47dY1RaEr7t5Qyjfvy+a44u7BWeld67oHrWNmlf/gnOl8444nh+SA/3rVJn69ahPDWlLsHRg8voyRIqVUKb8cCrbzJ2cNFtOuuHuwRT+qzeCyjx4JwLeWP2V7jVvh/OxsReEa99dve5Kv3vpEKa+6UE/Bvk8rxXETDmDFU1vLJK0Uw40UX7v1CZbe+7zvjG1rLrZdPrZTFrfZ9uV3baCnr7DPy81sr9zw8mxwtwxy6/pOfXrVWj7eUW0GWsOuPvv+vNqtJlHVkORYmnLi9pq05/7fnxm5t5dlR71fHt0plHhzT7b0S86NdEoNEbnKt7Ozmc0tnB6QsbM3y+Ll6wE8c7zL17Xy9p6BIesuvOPJ0s/llnpPX5aFtz/JecePx+0KfPmkDdCXzQ85Ix7WkmbmxMKtZU5W/Nt7Bljz6g5+s3qTo5jmVov5mnn92vxqfWBJTmvbib8vmy+t193T57gfzfqtlrW1P/Os05y0revZtWvdf9bjt+bVHYOOc07r0s/lk3fQicOuzzWv7hj0viyvNUw/MHRfee3bOIiqhqTHIrNSGal8jm8+/CsuWHs3muAfiQqNi98IcCf72opXFrXbdn5zvJ3yrO0sbPPeaTdL/ebVm0MZ9eX9+clcN/uLOHY9FG6GtJtlXT5Or+x5p9f95oYHpRqWvZ/+qmmf10K2ehTIxF3Gh5//M4ftfI1rZp9DxakPguBC2BzuIDneQS10v/Z4pfjJXE8yp7ycSi3ssNnzr/X0+c4ND0o1LHs/21fTPq+FbPUokInbitZ8adXtvNTRyb3vPCHpaoQGJ2wOd5Ac76AWul97vFL8ZK4nmVNeTqUWdtjs+XHtGd+54UGphmXvZ/tq2ue1kK0eBTJxW3jfK49z5Bsv87NZZ5FPBc8zFxobvxHgTva1Fa8sarft/OZ4O+VZ21nYZj64m6V+/qzxoYz68v78ZK6b/UUcux4KN0PazbIuH6dX9rzT635zw4NSDcveT3/VtM9rIVs9Cppy4t64xP7RnP3pFh46bAZdR55S5YqEpFGqkFcNgydo8+Sjsz3D1eceMySv++CRrUPynH9wzvQhOcl2+d9OWdTW3GyzLrv85vI8abd1TeYd28nSc6YPygkf1Waw9OzpJemoPAO9PWOw9JzpXDnvaNt8dBNrTjYU7OwfnXcMS4v7w64ur/6uPveY0nExx7hg9oTB4wZGtKZdc6qdjsOC2ROG9G1u7ycz25qTbe3HbpxuWdxur7vlhleCU5/W92XQ3PAw/VXTKo+qhqTH0pRZ5YIgCIJQy0hWuR9uuw16epKuQhAEQRBckYkb4Omn4bzz4Kc/TboSQRAEQXBFJm6AJUtgv/3gS19KuhJBEARBcEUm7ldegVtugYsugo4O7/UFQRAEIUFk4l66FNJpuOSSpCsRBEEQBE+ae+LWGl57DS64AMaNS7oaQRAEQfCkKR8yUkIp6OqCgQHPVQVBEAShFmjeM+4334RNxSfutDT33y+CIAhC/dC8E/dPfwqHHw5btiRdiSAIgiD4pjkn7t5e+NGP4AMfgEMOSboaQRAEQfBNIp8RK6U+BPwYSAM/11ovqWoBv/gFbNvGOaPex2OLVlS1a6Ex6GzPcMrU0fzuya309GUramtYS4p8Pk82P3h5xkhx1fxpzDu2k0u71nPTqk1EHVCcUoXnjI9qM9AadvVlGVcc24qntrKzN9jY2jMGZ0wf67hfzP6gkDH+qdkTWP3ydl54Y3cEo9mHkQKlFP02z/puM1L0ZfO241Sq4Kxa67Rivg6FsV5+5pGseXUHN6/eHOgxmwrQxTayuTy7+3ODXmtJUXo/lB+jnr4saaXIaV362tmeYeHcKUOysrvWdbP03ufp7ukbtO4pU0fz0HPb6O7pK9ViJaUK78s9xf1kbfvSrvX8etWmQfV+qpidbvb3Wk/foO2clltrdNqmvHY/29rtA7d1grblVlM1qHpWuVIqDfwfcBqwBXgMOF9r/YzTNpFmlWezbDmok60jR3POgu9F06YgxEQKOGFyBytf2pF0KUKNkzHSgx500bWum8XL19OXzXls6b/tNa/uGDRpW5kzuYPHN+0a1F/GSHPWjE6Wre0esvyq+YWHpJTX6LSN32299kH5On7Xc9ufdm1WiltWeRIT9wnA5VrrucWfFwNora9y2ibSiftPfyL7vpO5cP6lPDT5PdG0KQiCUAN0tmdYuej9AMxZ8iDdPX2Rtv2XXXsCfbIAlM5M7doDbGt02sbPtn72gXUdv+t57c/yNivFbeJO4qPyTmCz5ectwKzylZRSFwIXAkyYMCG63k86ifde9HO2jjwoujYFQRBqgNcsE8trEU7aZnthTvOcJmC3+rz+OHDb1s8+KF/uZz2v/Rn1/nYjCTlN2SwbcpS01tdprWdqrWeOHj060gK27j9634OWBUEQGoRxxTPR8u+jajsd4vem0zbj2jOONXr147atn31QvryStvy+HiVJTNxbgPGWnw8BXkugDkGoeVIUrh0KghcZI83CuVNKPy+cO4WMkY607fNnjXdcZ87kjiH9ZYw0588ab7t84dwptjU6beN3W699UL6O3/Xc9qddm3GSxMT9GHCEUupQpVQr8AngrmoWsHHJ6dXsTmhAOtszLJg9gfaMUXFbw1pSGDb/EzNGiqvPO4abvnACC2ZPsP2oqlJSxUZHtRm0ZwwU+8Y2qi342Nozhut+SVkGoYAFsydwxJgRwQv3wEhBa9p+j7UZKcdxmid6KYedbT0RbM8Y/Oi8Y1gwe0LgM1Fz7faMwYjW9JDXrO+H8mME+85Iza+d7ZkhctS8Yzu5av7RpevB1nUXzJ5QWm5XeUoV3n/mfjLbvnLe0SyYPfjSpXkcb/rCCaX+rNtdOe9o2+Xzju0cVKPTNk7jdNrWaR84reN3Pbf9GbWY5kXV5TQApdRHgB9RuB3sBq31d93Wj1ROEwRBEIQap9bkNLTW9wD3JNG3IAiCINQzzZmcJgiCIAh1ikzcgiAIglBHyMQtCIIgCHWETNyCIAiCUEfIxC0IgiAIdYRM3IIgCIJQR8jELQiCIAh1hEzcgiAIglBHyMQtCIIgCHVEIpGnQVFKbQNejbDJg4C/RtherSHjq29kfPVNo48PGn+MtTC+iVpr20dj1sXEHTVKqTVOGbCNgIyvvpHx1TeNPj5o/DHW+vjko3JBEARBqCNk4hYEQRCEOqJZJ+7rki4gZmR89Y2Mr75p9PFB44+xpsfXlNe4BUEQBKFeadYzbkEQBEGoS5pu4lZKfUgp9bxS6kWl1KKk66kUpdR4pdRDSqlnlVIblFJfKS7vUErdr5R6ofh1VNK1hkUplVZKrVNK/a74c8OMDUAp1a6UukMp9VzxOJ7QSGNUSn2t+N58Wil1s1JqeD2PTyl1g1LqDaXU05ZljuNRSi0u/r55Xik1N5mq/eMwvqXF9+dTSqk7lVLtltfqfnyW1/5RKaWVUgdZltXc+Jpq4lZKpYF/Bz4MvBs4Xyn17mSrqpgB4Ota63cBs4G/L45pEfCA1voI4IHiz/XKV4BnLT830tgAfgz8Xms9FZhOYawNMUalVCfwZWCm1vooIA18gvoe36+AD5Utsx1P8f/iJ4Aji9tcU/w9VMv8iqHjux84Sms9Dfg/YDE01PhQSo0HTgM2WZbV5PiaauIGjgde1Fq/rLXuB24BPpZwTRWhtd6qtX68+P1bFH7pd1IY143F1W4E5iVSYIUopQ4BTgd+blncEGMDUErtD7wX+AWA1rpfa91DA40RaAEySqkWoA14jToen9b6j8COssVO4/kYcIvWeq/W+hXgRQq/h2oWu/Fpre/TWg8Uf1wFHFL8viHGV+SHwDcAq/hVk+Nrtom7E9hs+XlLcVlDoJSaBBwLrAYO1lpvhcLkDoxJsLRK+BGF/0x5y7JGGRvAYcA24JfFywE/V0qNoEHGqLXuBr5P4SxmK7BLa30fDTI+C07jacTfOZ8F/rv4fUOMTyl1JtCttX6y7KWaHF+zTdzKZllDaPVKqf2AZcBXtdZvJl1PFCilzgDe0FqvTbqWGGkBjgOu1VofC+ymvj42dqV4rfdjwKHAOGCEUmpBslVVlYb6naOU+jaFy3M3mYtsVqur8Sml2oBvA/9k97LNssTH12wT9xZgvOXnQyh8bFfXKKUMCpP2TVrr5cXFryulxhZfHwu8kVR9FTAHOFMptZHCZY33K6V+TWOMzWQLsEVrvbr48x0UJvJGGeMHgFe01tu01llgOXAijTM+E6fxNMzvHKXUBcAZwKf0vvuIG2F8kyn8Yflk8XfNIcDjSql3UKPja7aJ+zHgCKXUoUqpVgrSwV0J11QRSilF4fros1rrqy0v3QVcUPz+AuC31a6tUrTWi7XWh2itJ1E4Vg9qrRfQAGMz0Vr/BdislJpSXHQq8AyNM8ZNwGylVFvxvXoqBQ+jUcZn4jSeu4BPKKWGKaUOBY4AHk2gvopQSn0I+CZwpta61/JS3Y9Pa71eaz1Gaz2p+LtmC3Bc8f9mbY5Pa91U/4CPULAiXwK+nXQ9EYznJAof3TwFPFH89xHgQAp26wvFrx1J11rhOE8Gflf8vtHGdgywpngMu4BRjTRG4ArgOeBp4L+AYfU8PuBmCtfrsxR+yX/ObTwUPoZ9CXge+HDS9Ycc34sUrvWav2N+1kjjK3t9I3BQLY9PktMEQRAEoY5oto/KBUEQBKGukYlbEARBEOoImbgFQRAEoY6QiVsQBEEQ6giZuAVBEAShjpCJWxCaHKXUJLsnJQmCUJvIxC0IgiAIdYRM3IIgAKSVUtcXn5t9n1Iqo5R6WCn1Q6XUH4vPCX+PUmp58ZnTVyZdsCA0KzJxC4IAhSjHf9daHwn0AGcVl/drrd8L/IxCjOffA0cBn1ZKHZhEoYLQ7MjELQgCFB4E8kTx+7XApOL3Zpb/emCDLjz/fS/wMoMfviAIQpWQiVsQBIC9lu9zFB43al2eL1snb1lHEIQqIhO3IAiCINQRMnELgiAIQh0hTwcTBEEQhDpCzrgFQRAEoY6QiVsQBEEQ6giZuAVBEAShjpCJWxAEQRDqCJm4BUEQBKGOkIlbEARBEOoImbgFQRAEoY6QiVsQBEEQ6oj/Dz2Zm8pnxuYBAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams[\"figure.figsize\"] = (8,6)\n",
    "plt.scatter(predictors['h']/100,predictors['forestheight'])\n",
    "plt.xlabel('hm')\n",
    "plt.ylabel('forestheight')\n",
    "ident = [0, 50]\n",
    "plt.plot(ident,ident,'r--')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.4527925129990046"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pearsonr_Publication_Estimation = pearsonr(predictors['h'],predictors['forestheight'])[0]\n",
    "pearsonr_Publication_Estimation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will try to beats such error estimation using a more advance ML tecnques and different enviromental predictors that better express the ecological condition."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data set splitting\n",
    "Split the dataset (predictors_sel) in order to create response variable vs predictors variables (we are excluding forestheight predictors)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "   X = predictors_sel.iloc[:,[5,6,7,8,9,10,11,12,13,15,16,17,18,19,20,21,22,23]].values\n",
    "Y = predictors_sel.iloc[:,4:5].values\n",
    "feat = predictors_sel.iloc[:,[5,6,7,8,9,10,11,12,13,15,16,17,18,19,20,21,22,23]].columns.values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Double ceck that we select the right columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['BLDFIE_WeigAver', 'CECSOL_WeigAver', 'CHELSA_bio18',\n",
       "       'CHELSA_bio4', 'convergence', 'cti', 'devmagnitude', 'eastness',\n",
       "       'elev', 'glad_ard_SVVI_max', 'glad_ard_SVVI_med',\n",
       "       'glad_ard_SVVI_min', 'northness', 'ORCDRC_WeigAver',\n",
       "       'outlet_dist_dw_basin', 'SBIO3_Isothermality_5_15cm',\n",
       "       'SBIO4_Temperature_Seasonality_5_15cm', 'treecover'], dtype=object)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "feat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100000, 1)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100000, 18)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create 4 dataset for training and testing the algorithm "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.5, random_state=24)\n",
    "y_train = np.ravel(Y_train)\n",
    "y_test = np.ravel(Y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Random Forest default parameters\n",
    "\n",
    "Random Forest can be implemented using the [RandomForestRegressor in scikit-learn](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestRegressor.html)\n",
    "\n",
    "Training Random Forest using default parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'bootstrap': True,\n",
       " 'ccp_alpha': 0.0,\n",
       " 'criterion': 'mse',\n",
       " 'max_depth': None,\n",
       " 'max_features': 'auto',\n",
       " 'max_leaf_nodes': None,\n",
       " 'max_samples': None,\n",
       " 'min_impurity_decrease': 0.0,\n",
       " 'min_impurity_split': None,\n",
       " 'min_samples_leaf': 1,\n",
       " 'min_samples_split': 2,\n",
       " 'min_weight_fraction_leaf': 0.0,\n",
       " 'n_estimators': 100,\n",
       " 'n_jobs': None,\n",
       " 'oob_score': False,\n",
       " 'random_state': 42,\n",
       " 'verbose': 0,\n",
       " 'warm_start': False}"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf = RandomForestRegressor(random_state = 42)\n",
    "rf.get_params()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.6043347073198263, 0.5189400044676861]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfReg = RandomForestRegressor(min_samples_leaf=50, oob_score=True)\n",
    "rfReg.fit(X_train, y_train);\n",
    "dic_pred = {}\n",
    "dic_pred['train'] = rfReg.predict(X_train)\n",
    "dic_pred['test'] = rfReg.predict(X_test)\n",
    "pearsonr_all = [pearsonr(dic_pred['train'],y_train)[0],pearsonr(dic_pred['test'],y_test)[0]]\n",
    "pearsonr_all"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.2651550005190103"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# checking the oob score\n",
    "rfReg.oob_score_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Additional resources how to reduce the oob error can found at  \n",
    "https://scikit-learn.org/stable/auto_examples/ensemble/plot_ensemble_oob.html"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Random Forest tuning\n",
    "\n",
    "\n",
    "* \"max_features\": number of features to consider when looking for the best split. \n",
    "* \"max_samples\": number of samples to draw from X to train each base estimator.\n",
    "* \"n_estimators\": identify the number of trees that must grow. It must be large enough so that the error is stabilized. Defoult 100.\n",
    "* \"max_depth\": max number of levels in each decision tree."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using this pseudo code tune the parameters in order the improve the algoirithm performance\n",
    "\n",
    "\n",
    "    pipeline = Pipeline([('rf',RandomForestRegressor())])\n",
    "\n",
    "    parameters = {\n",
    "            'rf__max_features':(3,4,5),\n",
    "            'rf__max_samples':(0.5,0.6,0.7),\n",
    "            'rf__n_estimators':(500,1000),\n",
    "            'rf__max_depth':(50,100,200,300)}\n",
    "\n",
    "    grid_search = GridSearchCV(pipeline,parameters,n_jobs=6,cv=5,scoring='r2',verbose=1)\n",
    "    grid_search.fit(X_train,y_train)\n",
    "    \n",
    "\n",
    "    rfReg = RandomForestRegressor(n_estimators=5000,max_features=0.33,max_depth=500,max_samples=0.7,n_jobs=-1,random_state=24 , oob_score = True) \n",
    "    rfReg.fit(X_train, y_train); \n",
    "    dic_pred = {} \n",
    "    dic_pred['train'] = rfReg.predict(X_train) \n",
    "    dic_pred['test'] = rfReg.predict(X_test) \n",
    "    pearsonr_all_tune = [pearsonr(dic_pred['train'],y_train)[0],pearsonr(dic_pred['test'],y_test)[0]] \n",
    "    pearsonr_all_tune\n",
    "\n",
    "    grid_search.best_score_\n",
    "\n",
    "    print ('Best Training score: %0.3f' % grid_search.best_score_) \n",
    "    print ('Optimal parameters:') best_par = grid_search.best_estimator_.get_params() \n",
    "    for par_name in sorted(parameters.keys()): \n",
    "        print ('\\t%s: %r' % (par_name, best_par[par_name]))\n",
    "        \n",
    "Tracking the error rate trend as in https://scikit-learn.org/stable/auto_examples/ensemble/plot_ensemble_oob.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fa02261eec0>]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams[\"figure.figsize\"] = (8,6)\n",
    "plt.scatter(y_train,dic_pred['train'])\n",
    "plt.xlabel('training GEDI Height (all rows)')\n",
    "plt.ylabel('training prediction')\n",
    "ident = [0, 50]\n",
    "plt.plot(ident,ident,'r--')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fa03153b3a0>]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams[\"figure.figsize\"] = (8,6)\n",
    "plt.scatter(y_test,dic_pred['test'])\n",
    "plt.xlabel('testing GEDI Height (all rows)')\n",
    "plt.ylabel('testing prediction')\n",
    "ident = [0, 50]\n",
    "plt.plot(ident,ident,'r--')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "impt = [rfReg.feature_importances_, np.std([tree.feature_importances_ for tree in rfReg.estimators_],axis=1)] \n",
    "ind = np.argsort(impt[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1, 13,  0,  4,  6,  8, 16,  2,  5, 15, 12,  3, 11,  7,  9, 14, 10,\n",
       "       17])"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ind"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x864 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams[\"figure.figsize\"] = (6,12)\n",
    "plt.barh(range(len(feat)),impt[0][ind],color=\"b\", xerr=impt[1][ind], align=\"center\")\n",
    "plt.yticks(range(len(feat)),feat[ind]);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Most important variable is treecover followed by Spectral Variability Vegetation Index, followed by outlet_dist_dw_basin that is a proxy of water accumulation. Besides eastness and northness describes microclimat condition."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Base on data quality flag select more reilable tree height."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**File storing tree hight (cm) obtained by 6 algorithms, with their associate quality flags.**\n",
    "The quality flags can be used to refine and select the best tree height estimation and use it as tree height observation.\n",
    "\n",
    "* a?_95: tree hight (cm) at 95 quintile, for each algorithm  \n",
    "* min_rh_95: minimum value of tree hight (cm) ammong the 6 algorithms  \n",
    "* max_rh_95: maximum value of tree hight (cm) ammong the 6 algorithms  \n",
    "* BEAM: 1-4 coverage beam = lower power (worse) ; 5-8 power beam = higher power (better)  \n",
    "* digital_elev: digital mdoel elevation  \n",
    "* elev_low: elevation of center of lowest mode  \n",
    "* qc_a?: quality_flag for six algorithms quality_flag = 1 (better); = 0 (worse)  \n",
    "* se_a?: sensitivity for six algorithms sensitivity < 0.95 (worse);  sensitivity > 0.95  (beter )  \n",
    "* deg_fg: (degrade_flag) not-degraded 0 (better) ;  degraded > 0 (worse)  \n",
    "* solar_ele: solar elevation. > 0 day (worse); < 0 night (better)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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>ID</th>\n",
       "      <th>X</th>\n",
       "      <th>Y</th>\n",
       "      <th>a1_95</th>\n",
       "      <th>a2_95</th>\n",
       "      <th>a3_95</th>\n",
       "      <th>a4_95</th>\n",
       "      <th>a5_95</th>\n",
       "      <th>a6_95</th>\n",
       "      <th>min_rh_95</th>\n",
       "      <th>max_rh_95</th>\n",
       "      <th>BEAM</th>\n",
       "      <th>digital_elev</th>\n",
       "      <th>elev_low</th>\n",
       "      <th>qc_a1</th>\n",
       "      <th>qc_a2</th>\n",
       "      <th>qc_a3</th>\n",
       "      <th>qc_a4</th>\n",
       "      <th>qc_a5</th>\n",
       "      <th>qc_a6</th>\n",
       "      <th>se_a1</th>\n",
       "      <th>se_a2</th>\n",
       "      <th>se_a3</th>\n",
       "      <th>se_a4</th>\n",
       "      <th>se_a5</th>\n",
       "      <th>se_a6</th>\n",
       "      <th>deg_fg</th>\n",
       "      <th>solar_ele</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>6.050001</td>\n",
       "      <td>49.727499</td>\n",
       "      <td>3139</td>\n",
       "      <td>3139</td>\n",
       "      <td>3139</td>\n",
       "      <td>3120</td>\n",
       "      <td>3139</td>\n",
       "      <td>3139</td>\n",
       "      <td>3120</td>\n",
       "      <td>3139</td>\n",
       "      <td>5</td>\n",
       "      <td>410.0</td>\n",
       "      <td>383.72153</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.962</td>\n",
       "      <td>0.984</td>\n",
       "      <td>0.968</td>\n",
       "      <td>0.962</td>\n",
       "      <td>0.989</td>\n",
       "      <td>0.979</td>\n",
       "      <td>0</td>\n",
       "      <td>17.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>6.050002</td>\n",
       "      <td>49.922155</td>\n",
       "      <td>1022</td>\n",
       "      <td>2303</td>\n",
       "      <td>970</td>\n",
       "      <td>872</td>\n",
       "      <td>5596</td>\n",
       "      <td>1524</td>\n",
       "      <td>872</td>\n",
       "      <td>5596</td>\n",
       "      <td>5</td>\n",
       "      <td>290.0</td>\n",
       "      <td>2374.14110</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>0.948</td>\n",
       "      <td>0.990</td>\n",
       "      <td>0.960</td>\n",
       "      <td>0.948</td>\n",
       "      <td>0.994</td>\n",
       "      <td>0.980</td>\n",
       "      <td>0</td>\n",
       "      <td>43.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>6.050002</td>\n",
       "      <td>48.602377</td>\n",
       "      <td>380</td>\n",
       "      <td>1336</td>\n",
       "      <td>332</td>\n",
       "      <td>362</td>\n",
       "      <td>1336</td>\n",
       "      <td>1340</td>\n",
       "      <td>332</td>\n",
       "      <td>1340</td>\n",
       "      <td>4</td>\n",
       "      <td>440.0</td>\n",
       "      <td>435.97781</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.947</td>\n",
       "      <td>0.975</td>\n",
       "      <td>0.956</td>\n",
       "      <td>0.947</td>\n",
       "      <td>0.981</td>\n",
       "      <td>0.968</td>\n",
       "      <td>0</td>\n",
       "      <td>0.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>6.050009</td>\n",
       "      <td>48.151979</td>\n",
       "      <td>3153</td>\n",
       "      <td>3142</td>\n",
       "      <td>3142</td>\n",
       "      <td>3127</td>\n",
       "      <td>3138</td>\n",
       "      <td>3142</td>\n",
       "      <td>3127</td>\n",
       "      <td>3153</td>\n",
       "      <td>2</td>\n",
       "      <td>450.0</td>\n",
       "      <td>422.00537</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.930</td>\n",
       "      <td>0.970</td>\n",
       "      <td>0.943</td>\n",
       "      <td>0.930</td>\n",
       "      <td>0.978</td>\n",
       "      <td>0.962</td>\n",
       "      <td>0</td>\n",
       "      <td>-14.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>6.050010</td>\n",
       "      <td>49.588410</td>\n",
       "      <td>666</td>\n",
       "      <td>4221</td>\n",
       "      <td>651</td>\n",
       "      <td>33</td>\n",
       "      <td>5611</td>\n",
       "      <td>2723</td>\n",
       "      <td>33</td>\n",
       "      <td>5611</td>\n",
       "      <td>8</td>\n",
       "      <td>370.0</td>\n",
       "      <td>2413.74830</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>0.941</td>\n",
       "      <td>0.983</td>\n",
       "      <td>0.946</td>\n",
       "      <td>0.941</td>\n",
       "      <td>0.992</td>\n",
       "      <td>0.969</td>\n",
       "      <td>0</td>\n",
       "      <td>22.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>6.050014</td>\n",
       "      <td>48.608456</td>\n",
       "      <td>787</td>\n",
       "      <td>1179</td>\n",
       "      <td>1187</td>\n",
       "      <td>761</td>\n",
       "      <td>1833</td>\n",
       "      <td>1833</td>\n",
       "      <td>761</td>\n",
       "      <td>1833</td>\n",
       "      <td>3</td>\n",
       "      <td>420.0</td>\n",
       "      <td>415.51581</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.952</td>\n",
       "      <td>0.979</td>\n",
       "      <td>0.961</td>\n",
       "      <td>0.952</td>\n",
       "      <td>0.986</td>\n",
       "      <td>0.975</td>\n",
       "      <td>0</td>\n",
       "      <td>0.2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ID         X          Y  a1_95  a2_95  a3_95  a4_95  a5_95  a6_95  \\\n",
       "0   1  6.050001  49.727499   3139   3139   3139   3120   3139   3139   \n",
       "1   2  6.050002  49.922155   1022   2303    970    872   5596   1524   \n",
       "2   3  6.050002  48.602377    380   1336    332    362   1336   1340   \n",
       "3   4  6.050009  48.151979   3153   3142   3142   3127   3138   3142   \n",
       "4   5  6.050010  49.588410    666   4221    651     33   5611   2723   \n",
       "5   6  6.050014  48.608456    787   1179   1187    761   1833   1833   \n",
       "\n",
       "   min_rh_95  max_rh_95  BEAM  digital_elev    elev_low  qc_a1  qc_a2  qc_a3  \\\n",
       "0       3120       3139     5         410.0   383.72153      1      1      1   \n",
       "1        872       5596     5         290.0  2374.14110      0      0      0   \n",
       "2        332       1340     4         440.0   435.97781      1      1      1   \n",
       "3       3127       3153     2         450.0   422.00537      1      1      1   \n",
       "4         33       5611     8         370.0  2413.74830      0      0      0   \n",
       "5        761       1833     3         420.0   415.51581      1      1      1   \n",
       "\n",
       "   qc_a4  qc_a5  qc_a6  se_a1  se_a2  se_a3  se_a4  se_a5  se_a6  deg_fg  \\\n",
       "0      1      1      1  0.962  0.984  0.968  0.962  0.989  0.979       0   \n",
       "1      0      0      0  0.948  0.990  0.960  0.948  0.994  0.980       0   \n",
       "2      1      1      1  0.947  0.975  0.956  0.947  0.981  0.968       0   \n",
       "3      1      1      1  0.930  0.970  0.943  0.930  0.978  0.962       0   \n",
       "4      0      0      0  0.941  0.983  0.946  0.941  0.992  0.969       0   \n",
       "5      1      1      1  0.952  0.979  0.961  0.952  0.986  0.975       0   \n",
       "\n",
       "   solar_ele  \n",
       "0       17.7  \n",
       "1       43.7  \n",
       "2        0.2  \n",
       "3      -14.2  \n",
       "4       22.1  \n",
       "5        0.2  "
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "height_6algorithms = pd.read_csv(\"tree_height/txt/eu_y_x_select_6algorithms_fullTable.txt\", sep=\" \",  index_col=False)\n",
    "pd.set_option('display.max_columns',None)\n",
    "height_6algorithms.head(6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "height_6algorithms_sel = height_6algorithms.loc[(height_6algorithms['BEAM'] > 4) \n",
    "                                            &   (height_6algorithms['qc_a1'] == 1)\n",
    "                                            &   (height_6algorithms['qc_a2'] == 1)\n",
    "                                            &   (height_6algorithms['qc_a3'] == 1)    \n",
    "                                            &   (height_6algorithms['qc_a4'] == 1)    \n",
    "                                            &   (height_6algorithms['qc_a5'] == 1)    \n",
    "                                            &   (height_6algorithms['qc_a6'] == 1)\n",
    "                                            &   (height_6algorithms['se_a1'] > 0.95)    \n",
    "                                            &   (height_6algorithms['se_a2'] > 0.95)\n",
    "                                            &   (height_6algorithms['se_a3'] > 0.95)\n",
    "                                            &   (height_6algorithms['se_a4'] > 0.95)\n",
    "                                            &   (height_6algorithms['se_a5'] > 0.95)   \n",
    "                                            &   (height_6algorithms['se_a6'] > 0.95)\n",
    "                                            &   (height_6algorithms['deg_fg'] == 0)  \n",
    "                                            &   (height_6algorithms['solar_ele'] < 0)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "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>ID</th>\n",
       "      <th>X</th>\n",
       "      <th>Y</th>\n",
       "      <th>a1_95</th>\n",
       "      <th>a2_95</th>\n",
       "      <th>a3_95</th>\n",
       "      <th>a4_95</th>\n",
       "      <th>a5_95</th>\n",
       "      <th>a6_95</th>\n",
       "      <th>min_rh_95</th>\n",
       "      <th>max_rh_95</th>\n",
       "      <th>BEAM</th>\n",
       "      <th>digital_elev</th>\n",
       "      <th>elev_low</th>\n",
       "      <th>qc_a1</th>\n",
       "      <th>qc_a2</th>\n",
       "      <th>qc_a3</th>\n",
       "      <th>qc_a4</th>\n",
       "      <th>qc_a5</th>\n",
       "      <th>qc_a6</th>\n",
       "      <th>se_a1</th>\n",
       "      <th>se_a2</th>\n",
       "      <th>se_a3</th>\n",
       "      <th>se_a4</th>\n",
       "      <th>se_a5</th>\n",
       "      <th>se_a6</th>\n",
       "      <th>deg_fg</th>\n",
       "      <th>solar_ele</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8</td>\n",
       "      <td>6.050019</td>\n",
       "      <td>49.921613</td>\n",
       "      <td>3303</td>\n",
       "      <td>3288</td>\n",
       "      <td>3296</td>\n",
       "      <td>3236</td>\n",
       "      <td>3857</td>\n",
       "      <td>3292</td>\n",
       "      <td>3236</td>\n",
       "      <td>3857</td>\n",
       "      <td>7</td>\n",
       "      <td>320.0</td>\n",
       "      <td>297.68533</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.971</td>\n",
       "      <td>0.988</td>\n",
       "      <td>0.976</td>\n",
       "      <td>0.971</td>\n",
       "      <td>0.992</td>\n",
       "      <td>0.984</td>\n",
       "      <td>0</td>\n",
       "      <td>-33.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>12</td>\n",
       "      <td>6.050039</td>\n",
       "      <td>47.995344</td>\n",
       "      <td>2762</td>\n",
       "      <td>2736</td>\n",
       "      <td>2740</td>\n",
       "      <td>2747</td>\n",
       "      <td>3893</td>\n",
       "      <td>2736</td>\n",
       "      <td>2736</td>\n",
       "      <td>3893</td>\n",
       "      <td>5</td>\n",
       "      <td>390.0</td>\n",
       "      <td>368.55121</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.975</td>\n",
       "      <td>0.990</td>\n",
       "      <td>0.979</td>\n",
       "      <td>0.975</td>\n",
       "      <td>0.994</td>\n",
       "      <td>0.987</td>\n",
       "      <td>0</td>\n",
       "      <td>-37.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>15</td>\n",
       "      <td>6.050046</td>\n",
       "      <td>49.865317</td>\n",
       "      <td>1398</td>\n",
       "      <td>2505</td>\n",
       "      <td>2509</td>\n",
       "      <td>1316</td>\n",
       "      <td>2848</td>\n",
       "      <td>2505</td>\n",
       "      <td>1316</td>\n",
       "      <td>2848</td>\n",
       "      <td>6</td>\n",
       "      <td>340.0</td>\n",
       "      <td>330.40564</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.973</td>\n",
       "      <td>0.990</td>\n",
       "      <td>0.979</td>\n",
       "      <td>0.973</td>\n",
       "      <td>0.994</td>\n",
       "      <td>0.986</td>\n",
       "      <td>0</td>\n",
       "      <td>-18.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>16</td>\n",
       "      <td>6.050048</td>\n",
       "      <td>49.050020</td>\n",
       "      <td>984</td>\n",
       "      <td>943</td>\n",
       "      <td>947</td>\n",
       "      <td>958</td>\n",
       "      <td>2617</td>\n",
       "      <td>947</td>\n",
       "      <td>943</td>\n",
       "      <td>2617</td>\n",
       "      <td>6</td>\n",
       "      <td>300.0</td>\n",
       "      <td>291.22598</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.978</td>\n",
       "      <td>0.991</td>\n",
       "      <td>0.982</td>\n",
       "      <td>0.978</td>\n",
       "      <td>0.995</td>\n",
       "      <td>0.988</td>\n",
       "      <td>0</td>\n",
       "      <td>-35.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>17</td>\n",
       "      <td>6.050049</td>\n",
       "      <td>48.391359</td>\n",
       "      <td>3362</td>\n",
       "      <td>3332</td>\n",
       "      <td>3336</td>\n",
       "      <td>3351</td>\n",
       "      <td>4467</td>\n",
       "      <td>3336</td>\n",
       "      <td>3332</td>\n",
       "      <td>4467</td>\n",
       "      <td>5</td>\n",
       "      <td>530.0</td>\n",
       "      <td>504.78122</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.973</td>\n",
       "      <td>0.988</td>\n",
       "      <td>0.977</td>\n",
       "      <td>0.973</td>\n",
       "      <td>0.992</td>\n",
       "      <td>0.984</td>\n",
       "      <td>0</td>\n",
       "      <td>-5.1</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",
       "      <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>1267207</th>\n",
       "      <td>1267208</td>\n",
       "      <td>9.949829</td>\n",
       "      <td>49.216272</td>\n",
       "      <td>2160</td>\n",
       "      <td>2816</td>\n",
       "      <td>2816</td>\n",
       "      <td>2104</td>\n",
       "      <td>3299</td>\n",
       "      <td>2816</td>\n",
       "      <td>2104</td>\n",
       "      <td>3299</td>\n",
       "      <td>8</td>\n",
       "      <td>420.0</td>\n",
       "      <td>386.44556</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.980</td>\n",
       "      <td>0.993</td>\n",
       "      <td>0.984</td>\n",
       "      <td>0.980</td>\n",
       "      <td>0.995</td>\n",
       "      <td>0.989</td>\n",
       "      <td>0</td>\n",
       "      <td>-16.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1267211</th>\n",
       "      <td>1267212</td>\n",
       "      <td>9.949856</td>\n",
       "      <td>49.881190</td>\n",
       "      <td>3190</td>\n",
       "      <td>3179</td>\n",
       "      <td>3179</td>\n",
       "      <td>3171</td>\n",
       "      <td>3822</td>\n",
       "      <td>3179</td>\n",
       "      <td>3171</td>\n",
       "      <td>3822</td>\n",
       "      <td>6</td>\n",
       "      <td>380.0</td>\n",
       "      <td>363.69348</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.968</td>\n",
       "      <td>0.986</td>\n",
       "      <td>0.974</td>\n",
       "      <td>0.968</td>\n",
       "      <td>0.990</td>\n",
       "      <td>0.982</td>\n",
       "      <td>0</td>\n",
       "      <td>-35.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1267216</th>\n",
       "      <td>1267217</td>\n",
       "      <td>9.949880</td>\n",
       "      <td>49.873435</td>\n",
       "      <td>2061</td>\n",
       "      <td>2828</td>\n",
       "      <td>2046</td>\n",
       "      <td>2024</td>\n",
       "      <td>2828</td>\n",
       "      <td>2828</td>\n",
       "      <td>2024</td>\n",
       "      <td>2828</td>\n",
       "      <td>7</td>\n",
       "      <td>380.0</td>\n",
       "      <td>361.06812</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.967</td>\n",
       "      <td>0.988</td>\n",
       "      <td>0.974</td>\n",
       "      <td>0.967</td>\n",
       "      <td>0.993</td>\n",
       "      <td>0.983</td>\n",
       "      <td>0</td>\n",
       "      <td>-35.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1267227</th>\n",
       "      <td>1267228</td>\n",
       "      <td>9.949958</td>\n",
       "      <td>49.127182</td>\n",
       "      <td>366</td>\n",
       "      <td>2307</td>\n",
       "      <td>1260</td>\n",
       "      <td>355</td>\n",
       "      <td>3531</td>\n",
       "      <td>2719</td>\n",
       "      <td>355</td>\n",
       "      <td>3531</td>\n",
       "      <td>6</td>\n",
       "      <td>500.0</td>\n",
       "      <td>493.52792</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.973</td>\n",
       "      <td>0.989</td>\n",
       "      <td>0.978</td>\n",
       "      <td>0.973</td>\n",
       "      <td>0.993</td>\n",
       "      <td>0.985</td>\n",
       "      <td>0</td>\n",
       "      <td>-36.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1267237</th>\n",
       "      <td>1267238</td>\n",
       "      <td>9.949999</td>\n",
       "      <td>49.936763</td>\n",
       "      <td>2513</td>\n",
       "      <td>2490</td>\n",
       "      <td>2490</td>\n",
       "      <td>2494</td>\n",
       "      <td>2490</td>\n",
       "      <td>2490</td>\n",
       "      <td>2490</td>\n",
       "      <td>2513</td>\n",
       "      <td>5</td>\n",
       "      <td>360.0</td>\n",
       "      <td>346.54227</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.968</td>\n",
       "      <td>0.988</td>\n",
       "      <td>0.974</td>\n",
       "      <td>0.968</td>\n",
       "      <td>0.993</td>\n",
       "      <td>0.983</td>\n",
       "      <td>0</td>\n",
       "      <td>-32.2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>226892 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              ID         X          Y  a1_95  a2_95  a3_95  a4_95  a5_95  \\\n",
       "7              8  6.050019  49.921613   3303   3288   3296   3236   3857   \n",
       "11            12  6.050039  47.995344   2762   2736   2740   2747   3893   \n",
       "14            15  6.050046  49.865317   1398   2505   2509   1316   2848   \n",
       "15            16  6.050048  49.050020    984    943    947    958   2617   \n",
       "16            17  6.050049  48.391359   3362   3332   3336   3351   4467   \n",
       "...          ...       ...        ...    ...    ...    ...    ...    ...   \n",
       "1267207  1267208  9.949829  49.216272   2160   2816   2816   2104   3299   \n",
       "1267211  1267212  9.949856  49.881190   3190   3179   3179   3171   3822   \n",
       "1267216  1267217  9.949880  49.873435   2061   2828   2046   2024   2828   \n",
       "1267227  1267228  9.949958  49.127182    366   2307   1260    355   3531   \n",
       "1267237  1267238  9.949999  49.936763   2513   2490   2490   2494   2490   \n",
       "\n",
       "         a6_95  min_rh_95  max_rh_95  BEAM  digital_elev   elev_low  qc_a1  \\\n",
       "7         3292       3236       3857     7         320.0  297.68533      1   \n",
       "11        2736       2736       3893     5         390.0  368.55121      1   \n",
       "14        2505       1316       2848     6         340.0  330.40564      1   \n",
       "15         947        943       2617     6         300.0  291.22598      1   \n",
       "16        3336       3332       4467     5         530.0  504.78122      1   \n",
       "...        ...        ...        ...   ...           ...        ...    ...   \n",
       "1267207   2816       2104       3299     8         420.0  386.44556      1   \n",
       "1267211   3179       3171       3822     6         380.0  363.69348      1   \n",
       "1267216   2828       2024       2828     7         380.0  361.06812      1   \n",
       "1267227   2719        355       3531     6         500.0  493.52792      1   \n",
       "1267237   2490       2490       2513     5         360.0  346.54227      1   \n",
       "\n",
       "         qc_a2  qc_a3  qc_a4  qc_a5  qc_a6  se_a1  se_a2  se_a3  se_a4  se_a5  \\\n",
       "7            1      1      1      1      1  0.971  0.988  0.976  0.971  0.992   \n",
       "11           1      1      1      1      1  0.975  0.990  0.979  0.975  0.994   \n",
       "14           1      1      1      1      1  0.973  0.990  0.979  0.973  0.994   \n",
       "15           1      1      1      1      1  0.978  0.991  0.982  0.978  0.995   \n",
       "16           1      1      1      1      1  0.973  0.988  0.977  0.973  0.992   \n",
       "...        ...    ...    ...    ...    ...    ...    ...    ...    ...    ...   \n",
       "1267207      1      1      1      1      1  0.980  0.993  0.984  0.980  0.995   \n",
       "1267211      1      1      1      1      1  0.968  0.986  0.974  0.968  0.990   \n",
       "1267216      1      1      1      1      1  0.967  0.988  0.974  0.967  0.993   \n",
       "1267227      1      1      1      1      1  0.973  0.989  0.978  0.973  0.993   \n",
       "1267237      1      1      1      1      1  0.968  0.988  0.974  0.968  0.993   \n",
       "\n",
       "         se_a6  deg_fg  solar_ele  \n",
       "7        0.984       0      -33.9  \n",
       "11       0.987       0      -37.3  \n",
       "14       0.986       0      -18.2  \n",
       "15       0.988       0      -35.4  \n",
       "16       0.984       0       -5.1  \n",
       "...        ...     ...        ...  \n",
       "1267207  0.989       0      -16.9  \n",
       "1267211  0.982       0      -35.1  \n",
       "1267216  0.983       0      -35.1  \n",
       "1267227  0.985       0      -36.0  \n",
       "1267237  0.983       0      -32.2  \n",
       "\n",
       "[226892 rows x 28 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "height_6algorithms_sel"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Calculate the mean height excluidng the maximum and minimum values "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "height_sel =  pd.DataFrame({'ID' : height_6algorithms_sel['ID'] , \n",
    "                            'hm_sel': (height_6algorithms_sel['a1_95'] + height_6algorithms_sel['a2_95'] + height_6algorithms_sel['a3_95'] + height_6algorithms_sel['a4_95'] \n",
    "                             + height_6algorithms_sel['a5_95'] + height_6algorithms_sel['a6_95'] - height_6algorithms_sel['min_rh_95'] - height_6algorithms_sel['max_rh_95']) / 400 } )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>ID</th>\n",
       "      <th>hm_sel</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8</td>\n",
       "      <td>32.9475</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>12</td>\n",
       "      <td>27.4625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>15</td>\n",
       "      <td>22.2925</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>16</td>\n",
       "      <td>9.5900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>17</td>\n",
       "      <td>33.4625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1267207</th>\n",
       "      <td>1267208</td>\n",
       "      <td>26.5200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1267211</th>\n",
       "      <td>1267212</td>\n",
       "      <td>31.8175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1267216</th>\n",
       "      <td>1267217</td>\n",
       "      <td>24.4075</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1267227</th>\n",
       "      <td>1267228</td>\n",
       "      <td>16.6300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1267237</th>\n",
       "      <td>1267238</td>\n",
       "      <td>24.9100</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>226892 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              ID   hm_sel\n",
       "7              8  32.9475\n",
       "11            12  27.4625\n",
       "14            15  22.2925\n",
       "15            16   9.5900\n",
       "16            17  33.4625\n",
       "...          ...      ...\n",
       "1267207  1267208  26.5200\n",
       "1267211  1267212  31.8175\n",
       "1267216  1267217  24.4075\n",
       "1267227  1267228  16.6300\n",
       "1267237  1267238  24.9100\n",
       "\n",
       "[226892 rows x 2 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "height_sel"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Merge the new height with the predictors table, using the ID as Primary Key"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "predictors_hm_sel = pd.merge( predictors ,  height_sel , left_on='ID' ,  right_on='ID' ,  how='right')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "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>ID</th>\n",
       "      <th>X</th>\n",
       "      <th>Y</th>\n",
       "      <th>h</th>\n",
       "      <th>BLDFIE_WeigAver</th>\n",
       "      <th>CECSOL_WeigAver</th>\n",
       "      <th>CHELSA_bio18</th>\n",
       "      <th>CHELSA_bio4</th>\n",
       "      <th>convergence</th>\n",
       "      <th>cti</th>\n",
       "      <th>devmagnitude</th>\n",
       "      <th>eastness</th>\n",
       "      <th>elev</th>\n",
       "      <th>forestheight</th>\n",
       "      <th>glad_ard_SVVI_max</th>\n",
       "      <th>glad_ard_SVVI_med</th>\n",
       "      <th>glad_ard_SVVI_min</th>\n",
       "      <th>northness</th>\n",
       "      <th>ORCDRC_WeigAver</th>\n",
       "      <th>outlet_dist_dw_basin</th>\n",
       "      <th>SBIO3_Isothermality_5_15cm</th>\n",
       "      <th>SBIO4_Temperature_Seasonality_5_15cm</th>\n",
       "      <th>treecover</th>\n",
       "      <th>hm_sel</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8</td>\n",
       "      <td>6.050019</td>\n",
       "      <td>49.921613</td>\n",
       "      <td>3294.75</td>\n",
       "      <td>1490</td>\n",
       "      <td>12</td>\n",
       "      <td>1995</td>\n",
       "      <td>5912</td>\n",
       "      <td>22.102139</td>\n",
       "      <td>-297770784</td>\n",
       "      <td>-1.402633</td>\n",
       "      <td>0.309765</td>\n",
       "      <td>294.927765</td>\n",
       "      <td>26</td>\n",
       "      <td>-86.729492</td>\n",
       "      <td>-145.584229</td>\n",
       "      <td>-190.062988</td>\n",
       "      <td>0.222435</td>\n",
       "      <td>15</td>\n",
       "      <td>772784</td>\n",
       "      <td>20.855963</td>\n",
       "      <td>457.195404</td>\n",
       "      <td>86</td>\n",
       "      <td>32.9475</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>12</td>\n",
       "      <td>6.050039</td>\n",
       "      <td>47.995344</td>\n",
       "      <td>2746.25</td>\n",
       "      <td>1523</td>\n",
       "      <td>12</td>\n",
       "      <td>2612</td>\n",
       "      <td>6181</td>\n",
       "      <td>3.549103</td>\n",
       "      <td>-71279992</td>\n",
       "      <td>0.507727</td>\n",
       "      <td>-0.021408</td>\n",
       "      <td>322.920227</td>\n",
       "      <td>26</td>\n",
       "      <td>660.006104</td>\n",
       "      <td>92.722168</td>\n",
       "      <td>190.979736</td>\n",
       "      <td>-0.034787</td>\n",
       "      <td>16</td>\n",
       "      <td>784807</td>\n",
       "      <td>20.798000</td>\n",
       "      <td>460.501221</td>\n",
       "      <td>97</td>\n",
       "      <td>27.4625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15</td>\n",
       "      <td>6.050046</td>\n",
       "      <td>49.865317</td>\n",
       "      <td>2229.25</td>\n",
       "      <td>1517</td>\n",
       "      <td>13</td>\n",
       "      <td>2191</td>\n",
       "      <td>5901</td>\n",
       "      <td>31.054762</td>\n",
       "      <td>-186807440</td>\n",
       "      <td>-1.375050</td>\n",
       "      <td>-0.126880</td>\n",
       "      <td>291.412537</td>\n",
       "      <td>7</td>\n",
       "      <td>1028.385498</td>\n",
       "      <td>915.806396</td>\n",
       "      <td>841.586182</td>\n",
       "      <td>0.024677</td>\n",
       "      <td>16</td>\n",
       "      <td>766444</td>\n",
       "      <td>19.941267</td>\n",
       "      <td>454.185089</td>\n",
       "      <td>54</td>\n",
       "      <td>22.2925</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>16</td>\n",
       "      <td>6.050048</td>\n",
       "      <td>49.050020</td>\n",
       "      <td>959.00</td>\n",
       "      <td>1526</td>\n",
       "      <td>14</td>\n",
       "      <td>2081</td>\n",
       "      <td>6100</td>\n",
       "      <td>9.933455</td>\n",
       "      <td>-183562672</td>\n",
       "      <td>-0.382834</td>\n",
       "      <td>0.086874</td>\n",
       "      <td>246.288010</td>\n",
       "      <td>24</td>\n",
       "      <td>-12.283691</td>\n",
       "      <td>-58.179199</td>\n",
       "      <td>174.205566</td>\n",
       "      <td>0.094175</td>\n",
       "      <td>10</td>\n",
       "      <td>805730</td>\n",
       "      <td>19.849365</td>\n",
       "      <td>470.946533</td>\n",
       "      <td>78</td>\n",
       "      <td>9.5900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>17</td>\n",
       "      <td>6.050049</td>\n",
       "      <td>48.391359</td>\n",
       "      <td>3346.25</td>\n",
       "      <td>1489</td>\n",
       "      <td>19</td>\n",
       "      <td>2486</td>\n",
       "      <td>5966</td>\n",
       "      <td>-6.957157</td>\n",
       "      <td>-273522688</td>\n",
       "      <td>2.989759</td>\n",
       "      <td>0.214769</td>\n",
       "      <td>474.409088</td>\n",
       "      <td>24</td>\n",
       "      <td>125.583008</td>\n",
       "      <td>6.154297</td>\n",
       "      <td>128.129150</td>\n",
       "      <td>0.017164</td>\n",
       "      <td>15</td>\n",
       "      <td>950190</td>\n",
       "      <td>21.179420</td>\n",
       "      <td>491.398376</td>\n",
       "      <td>85</td>\n",
       "      <td>33.4625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>19</td>\n",
       "      <td>6.050053</td>\n",
       "      <td>49.877876</td>\n",
       "      <td>529.00</td>\n",
       "      <td>1531</td>\n",
       "      <td>12</td>\n",
       "      <td>2184</td>\n",
       "      <td>5915</td>\n",
       "      <td>-24.278454</td>\n",
       "      <td>-377335296</td>\n",
       "      <td>0.265329</td>\n",
       "      <td>-0.248356</td>\n",
       "      <td>335.534760</td>\n",
       "      <td>25</td>\n",
       "      <td>593.601074</td>\n",
       "      <td>228.712402</td>\n",
       "      <td>315.298340</td>\n",
       "      <td>-0.127365</td>\n",
       "      <td>17</td>\n",
       "      <td>764713</td>\n",
       "      <td>19.760756</td>\n",
       "      <td>448.580811</td>\n",
       "      <td>96</td>\n",
       "      <td>5.2900</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ID         X          Y        h  BLDFIE_WeigAver  CECSOL_WeigAver  \\\n",
       "0   8  6.050019  49.921613  3294.75             1490               12   \n",
       "1  12  6.050039  47.995344  2746.25             1523               12   \n",
       "2  15  6.050046  49.865317  2229.25             1517               13   \n",
       "3  16  6.050048  49.050020   959.00             1526               14   \n",
       "4  17  6.050049  48.391359  3346.25             1489               19   \n",
       "5  19  6.050053  49.877876   529.00             1531               12   \n",
       "\n",
       "   CHELSA_bio18  CHELSA_bio4  convergence        cti  devmagnitude  eastness  \\\n",
       "0          1995         5912    22.102139 -297770784     -1.402633  0.309765   \n",
       "1          2612         6181     3.549103  -71279992      0.507727 -0.021408   \n",
       "2          2191         5901    31.054762 -186807440     -1.375050 -0.126880   \n",
       "3          2081         6100     9.933455 -183562672     -0.382834  0.086874   \n",
       "4          2486         5966    -6.957157 -273522688      2.989759  0.214769   \n",
       "5          2184         5915   -24.278454 -377335296      0.265329 -0.248356   \n",
       "\n",
       "         elev  forestheight  glad_ard_SVVI_max  glad_ard_SVVI_med  \\\n",
       "0  294.927765            26         -86.729492        -145.584229   \n",
       "1  322.920227            26         660.006104          92.722168   \n",
       "2  291.412537             7        1028.385498         915.806396   \n",
       "3  246.288010            24         -12.283691         -58.179199   \n",
       "4  474.409088            24         125.583008           6.154297   \n",
       "5  335.534760            25         593.601074         228.712402   \n",
       "\n",
       "   glad_ard_SVVI_min  northness  ORCDRC_WeigAver  outlet_dist_dw_basin  \\\n",
       "0        -190.062988   0.222435               15                772784   \n",
       "1         190.979736  -0.034787               16                784807   \n",
       "2         841.586182   0.024677               16                766444   \n",
       "3         174.205566   0.094175               10                805730   \n",
       "4         128.129150   0.017164               15                950190   \n",
       "5         315.298340  -0.127365               17                764713   \n",
       "\n",
       "   SBIO3_Isothermality_5_15cm  SBIO4_Temperature_Seasonality_5_15cm  \\\n",
       "0                   20.855963                            457.195404   \n",
       "1                   20.798000                            460.501221   \n",
       "2                   19.941267                            454.185089   \n",
       "3                   19.849365                            470.946533   \n",
       "4                   21.179420                            491.398376   \n",
       "5                   19.760756                            448.580811   \n",
       "\n",
       "   treecover   hm_sel  \n",
       "0         86  32.9475  \n",
       "1         97  27.4625  \n",
       "2         54  22.2925  \n",
       "3         78   9.5900  \n",
       "4         85  33.4625  \n",
       "5         96   5.2900  "
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictors_hm_sel.head(6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "predictors_hm_sel = predictors_hm_sel.loc[(predictors['h'] < 7000) ].sample(100000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "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>ID</th>\n",
       "      <th>X</th>\n",
       "      <th>Y</th>\n",
       "      <th>h</th>\n",
       "      <th>BLDFIE_WeigAver</th>\n",
       "      <th>CECSOL_WeigAver</th>\n",
       "      <th>CHELSA_bio18</th>\n",
       "      <th>CHELSA_bio4</th>\n",
       "      <th>convergence</th>\n",
       "      <th>cti</th>\n",
       "      <th>devmagnitude</th>\n",
       "      <th>eastness</th>\n",
       "      <th>elev</th>\n",
       "      <th>forestheight</th>\n",
       "      <th>glad_ard_SVVI_max</th>\n",
       "      <th>glad_ard_SVVI_med</th>\n",
       "      <th>glad_ard_SVVI_min</th>\n",
       "      <th>northness</th>\n",
       "      <th>ORCDRC_WeigAver</th>\n",
       "      <th>outlet_dist_dw_basin</th>\n",
       "      <th>SBIO3_Isothermality_5_15cm</th>\n",
       "      <th>SBIO4_Temperature_Seasonality_5_15cm</th>\n",
       "      <th>treecover</th>\n",
       "      <th>hm_sel</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>83357</th>\n",
       "      <td>447901</td>\n",
       "      <td>7.325923</td>\n",
       "      <td>49.296261</td>\n",
       "      <td>1988.00</td>\n",
       "      <td>1504</td>\n",
       "      <td>11</td>\n",
       "      <td>1973</td>\n",
       "      <td>6329</td>\n",
       "      <td>-27.310654</td>\n",
       "      <td>-311542144</td>\n",
       "      <td>1.611667</td>\n",
       "      <td>-0.059867</td>\n",
       "      <td>315.589905</td>\n",
       "      <td>22</td>\n",
       "      <td>99.902344</td>\n",
       "      <td>182.999268</td>\n",
       "      <td>358.212158</td>\n",
       "      <td>-0.027319</td>\n",
       "      <td>13</td>\n",
       "      <td>852606</td>\n",
       "      <td>18.866175</td>\n",
       "      <td>470.286011</td>\n",
       "      <td>75</td>\n",
       "      <td>19.8800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>213105</th>\n",
       "      <td>1196478</td>\n",
       "      <td>9.622623</td>\n",
       "      <td>49.833063</td>\n",
       "      <td>2220.25</td>\n",
       "      <td>1520</td>\n",
       "      <td>13</td>\n",
       "      <td>1875</td>\n",
       "      <td>6606</td>\n",
       "      <td>-20.839262</td>\n",
       "      <td>-279513600</td>\n",
       "      <td>-0.550544</td>\n",
       "      <td>0.038211</td>\n",
       "      <td>244.721085</td>\n",
       "      <td>15</td>\n",
       "      <td>112.770508</td>\n",
       "      <td>-38.957520</td>\n",
       "      <td>-107.884766</td>\n",
       "      <td>-0.153021</td>\n",
       "      <td>11</td>\n",
       "      <td>770768</td>\n",
       "      <td>19.954826</td>\n",
       "      <td>507.755493</td>\n",
       "      <td>54</td>\n",
       "      <td>22.2025</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58293</th>\n",
       "      <td>313301</td>\n",
       "      <td>7.017546</td>\n",
       "      <td>49.874260</td>\n",
       "      <td>1389.75</td>\n",
       "      <td>1525</td>\n",
       "      <td>14</td>\n",
       "      <td>2127</td>\n",
       "      <td>6080</td>\n",
       "      <td>-16.012159</td>\n",
       "      <td>-287215200</td>\n",
       "      <td>1.066931</td>\n",
       "      <td>-0.218266</td>\n",
       "      <td>424.159332</td>\n",
       "      <td>22</td>\n",
       "      <td>-59.042236</td>\n",
       "      <td>-32.115967</td>\n",
       "      <td>126.038086</td>\n",
       "      <td>0.068847</td>\n",
       "      <td>13</td>\n",
       "      <td>619919</td>\n",
       "      <td>19.761299</td>\n",
       "      <td>465.827271</td>\n",
       "      <td>85</td>\n",
       "      <td>13.8975</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54985</th>\n",
       "      <td>297876</td>\n",
       "      <td>6.981116</td>\n",
       "      <td>48.889372</td>\n",
       "      <td>1883.25</td>\n",
       "      <td>1525</td>\n",
       "      <td>13</td>\n",
       "      <td>2161</td>\n",
       "      <td>6330</td>\n",
       "      <td>8.863924</td>\n",
       "      <td>431331168</td>\n",
       "      <td>-1.713272</td>\n",
       "      <td>0.013969</td>\n",
       "      <td>221.787323</td>\n",
       "      <td>23</td>\n",
       "      <td>560.937012</td>\n",
       "      <td>101.161133</td>\n",
       "      <td>27.694336</td>\n",
       "      <td>-0.025228</td>\n",
       "      <td>7</td>\n",
       "      <td>847969</td>\n",
       "      <td>18.343138</td>\n",
       "      <td>464.072418</td>\n",
       "      <td>67</td>\n",
       "      <td>18.8325</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18698</th>\n",
       "      <td>98303</td>\n",
       "      <td>6.397619</td>\n",
       "      <td>49.380758</td>\n",
       "      <td>2965.00</td>\n",
       "      <td>1538</td>\n",
       "      <td>11</td>\n",
       "      <td>2367</td>\n",
       "      <td>6042</td>\n",
       "      <td>23.690971</td>\n",
       "      <td>-80957808</td>\n",
       "      <td>-0.130503</td>\n",
       "      <td>-0.035726</td>\n",
       "      <td>276.400085</td>\n",
       "      <td>25</td>\n",
       "      <td>230.774902</td>\n",
       "      <td>120.549805</td>\n",
       "      <td>309.234375</td>\n",
       "      <td>0.012809</td>\n",
       "      <td>13</td>\n",
       "      <td>740733</td>\n",
       "      <td>18.482462</td>\n",
       "      <td>444.627533</td>\n",
       "      <td>78</td>\n",
       "      <td>29.6500</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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60228</th>\n",
       "      <td>322853</td>\n",
       "      <td>7.040689</td>\n",
       "      <td>49.728151</td>\n",
       "      <td>3375.00</td>\n",
       "      <td>1402</td>\n",
       "      <td>12</td>\n",
       "      <td>2362</td>\n",
       "      <td>5810</td>\n",
       "      <td>17.660843</td>\n",
       "      <td>15303386</td>\n",
       "      <td>1.776377</td>\n",
       "      <td>0.010083</td>\n",
       "      <td>517.456360</td>\n",
       "      <td>25</td>\n",
       "      <td>172.217773</td>\n",
       "      <td>256.016113</td>\n",
       "      <td>213.990967</td>\n",
       "      <td>-0.079195</td>\n",
       "      <td>25</td>\n",
       "      <td>663180</td>\n",
       "      <td>21.064062</td>\n",
       "      <td>430.348450</td>\n",
       "      <td>85</td>\n",
       "      <td>33.7500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42226</th>\n",
       "      <td>231457</td>\n",
       "      <td>6.810772</td>\n",
       "      <td>48.268025</td>\n",
       "      <td>2707.00</td>\n",
       "      <td>1409</td>\n",
       "      <td>15</td>\n",
       "      <td>3120</td>\n",
       "      <td>6229</td>\n",
       "      <td>0.276158</td>\n",
       "      <td>-217058432</td>\n",
       "      <td>-0.143515</td>\n",
       "      <td>-0.099901</td>\n",
       "      <td>471.342743</td>\n",
       "      <td>25</td>\n",
       "      <td>-153.505615</td>\n",
       "      <td>-142.315430</td>\n",
       "      <td>-232.547607</td>\n",
       "      <td>-0.198644</td>\n",
       "      <td>29</td>\n",
       "      <td>955008</td>\n",
       "      <td>20.322662</td>\n",
       "      <td>441.391357</td>\n",
       "      <td>85</td>\n",
       "      <td>27.0700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49495</th>\n",
       "      <td>271346</td>\n",
       "      <td>6.916373</td>\n",
       "      <td>49.514878</td>\n",
       "      <td>3097.25</td>\n",
       "      <td>1536</td>\n",
       "      <td>13</td>\n",
       "      <td>2213</td>\n",
       "      <td>6140</td>\n",
       "      <td>-0.606761</td>\n",
       "      <td>-299815488</td>\n",
       "      <td>0.547743</td>\n",
       "      <td>-0.104366</td>\n",
       "      <td>365.085297</td>\n",
       "      <td>24</td>\n",
       "      <td>29.947266</td>\n",
       "      <td>-2.760986</td>\n",
       "      <td>204.209473</td>\n",
       "      <td>0.163875</td>\n",
       "      <td>14</td>\n",
       "      <td>789703</td>\n",
       "      <td>20.452227</td>\n",
       "      <td>469.161865</td>\n",
       "      <td>85</td>\n",
       "      <td>30.9725</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>40</td>\n",
       "      <td>6.050120</td>\n",
       "      <td>48.058094</td>\n",
       "      <td>3012.25</td>\n",
       "      <td>1521</td>\n",
       "      <td>14</td>\n",
       "      <td>2474</td>\n",
       "      <td>6191</td>\n",
       "      <td>-15.363133</td>\n",
       "      <td>-195105152</td>\n",
       "      <td>1.290480</td>\n",
       "      <td>-0.043498</td>\n",
       "      <td>331.926483</td>\n",
       "      <td>25</td>\n",
       "      <td>750.277832</td>\n",
       "      <td>12.481934</td>\n",
       "      <td>206.658447</td>\n",
       "      <td>-0.008241</td>\n",
       "      <td>15</td>\n",
       "      <td>820336</td>\n",
       "      <td>19.240412</td>\n",
       "      <td>474.946625</td>\n",
       "      <td>85</td>\n",
       "      <td>30.1225</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>185044</th>\n",
       "      <td>1039892</td>\n",
       "      <td>9.187669</td>\n",
       "      <td>49.500969</td>\n",
       "      <td>1117.00</td>\n",
       "      <td>1472</td>\n",
       "      <td>15</td>\n",
       "      <td>2561</td>\n",
       "      <td>6294</td>\n",
       "      <td>10.249912</td>\n",
       "      <td>-141432176</td>\n",
       "      <td>1.770781</td>\n",
       "      <td>-0.046596</td>\n",
       "      <td>483.281250</td>\n",
       "      <td>27</td>\n",
       "      <td>-15.297119</td>\n",
       "      <td>-161.964844</td>\n",
       "      <td>-210.821777</td>\n",
       "      <td>-0.015172</td>\n",
       "      <td>23</td>\n",
       "      <td>761513</td>\n",
       "      <td>19.522852</td>\n",
       "      <td>437.534729</td>\n",
       "      <td>38</td>\n",
       "      <td>11.1700</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100000 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             ID         X          Y        h  BLDFIE_WeigAver  \\\n",
       "83357    447901  7.325923  49.296261  1988.00             1504   \n",
       "213105  1196478  9.622623  49.833063  2220.25             1520   \n",
       "58293    313301  7.017546  49.874260  1389.75             1525   \n",
       "54985    297876  6.981116  48.889372  1883.25             1525   \n",
       "18698     98303  6.397619  49.380758  2965.00             1538   \n",
       "...         ...       ...        ...      ...              ...   \n",
       "60228    322853  7.040689  49.728151  3375.00             1402   \n",
       "42226    231457  6.810772  48.268025  2707.00             1409   \n",
       "49495    271346  6.916373  49.514878  3097.25             1536   \n",
       "11           40  6.050120  48.058094  3012.25             1521   \n",
       "185044  1039892  9.187669  49.500969  1117.00             1472   \n",
       "\n",
       "        CECSOL_WeigAver  CHELSA_bio18  CHELSA_bio4  convergence        cti  \\\n",
       "83357                11          1973         6329   -27.310654 -311542144   \n",
       "213105               13          1875         6606   -20.839262 -279513600   \n",
       "58293                14          2127         6080   -16.012159 -287215200   \n",
       "54985                13          2161         6330     8.863924  431331168   \n",
       "18698                11          2367         6042    23.690971  -80957808   \n",
       "...                 ...           ...          ...          ...        ...   \n",
       "60228                12          2362         5810    17.660843   15303386   \n",
       "42226                15          3120         6229     0.276158 -217058432   \n",
       "49495                13          2213         6140    -0.606761 -299815488   \n",
       "11                   14          2474         6191   -15.363133 -195105152   \n",
       "185044               15          2561         6294    10.249912 -141432176   \n",
       "\n",
       "        devmagnitude  eastness        elev  forestheight  glad_ard_SVVI_max  \\\n",
       "83357       1.611667 -0.059867  315.589905            22          99.902344   \n",
       "213105     -0.550544  0.038211  244.721085            15         112.770508   \n",
       "58293       1.066931 -0.218266  424.159332            22         -59.042236   \n",
       "54985      -1.713272  0.013969  221.787323            23         560.937012   \n",
       "18698      -0.130503 -0.035726  276.400085            25         230.774902   \n",
       "...              ...       ...         ...           ...                ...   \n",
       "60228       1.776377  0.010083  517.456360            25         172.217773   \n",
       "42226      -0.143515 -0.099901  471.342743            25        -153.505615   \n",
       "49495       0.547743 -0.104366  365.085297            24          29.947266   \n",
       "11          1.290480 -0.043498  331.926483            25         750.277832   \n",
       "185044      1.770781 -0.046596  483.281250            27         -15.297119   \n",
       "\n",
       "        glad_ard_SVVI_med  glad_ard_SVVI_min  northness  ORCDRC_WeigAver  \\\n",
       "83357          182.999268         358.212158  -0.027319               13   \n",
       "213105         -38.957520        -107.884766  -0.153021               11   \n",
       "58293          -32.115967         126.038086   0.068847               13   \n",
       "54985          101.161133          27.694336  -0.025228                7   \n",
       "18698          120.549805         309.234375   0.012809               13   \n",
       "...                   ...                ...        ...              ...   \n",
       "60228          256.016113         213.990967  -0.079195               25   \n",
       "42226         -142.315430        -232.547607  -0.198644               29   \n",
       "49495           -2.760986         204.209473   0.163875               14   \n",
       "11              12.481934         206.658447  -0.008241               15   \n",
       "185044        -161.964844        -210.821777  -0.015172               23   \n",
       "\n",
       "        outlet_dist_dw_basin  SBIO3_Isothermality_5_15cm  \\\n",
       "83357                 852606                   18.866175   \n",
       "213105                770768                   19.954826   \n",
       "58293                 619919                   19.761299   \n",
       "54985                 847969                   18.343138   \n",
       "18698                 740733                   18.482462   \n",
       "...                      ...                         ...   \n",
       "60228                 663180                   21.064062   \n",
       "42226                 955008                   20.322662   \n",
       "49495                 789703                   20.452227   \n",
       "11                    820336                   19.240412   \n",
       "185044                761513                   19.522852   \n",
       "\n",
       "        SBIO4_Temperature_Seasonality_5_15cm  treecover   hm_sel  \n",
       "83357                             470.286011         75  19.8800  \n",
       "213105                            507.755493         54  22.2025  \n",
       "58293                             465.827271         85  13.8975  \n",
       "54985                             464.072418         67  18.8325  \n",
       "18698                             444.627533         78  29.6500  \n",
       "...                                      ...        ...      ...  \n",
       "60228                             430.348450         85  33.7500  \n",
       "42226                             441.391357         85  27.0700  \n",
       "49495                             469.161865         85  30.9725  \n",
       "11                                474.946625         85  30.1225  \n",
       "185044                            437.534729         38  11.1700  \n",
       "\n",
       "[100000 rows x 24 columns]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictors_hm_sel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "   X = predictors_hm_sel.iloc[:,[4,5,6,7,8,9,10,11,12,14,15,16,17,18,19,20,21,22]].values\n",
    "Y = predictors_hm_sel.iloc[:,23:24].values\n",
    "feat = predictors_hm_sel.iloc[:,[4,5,6,7,8,9,10,11,12,14,15,16,17,18,19,20,21,22]].columns.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['BLDFIE_WeigAver', 'CECSOL_WeigAver', 'CHELSA_bio18',\n",
       "       'CHELSA_bio4', 'convergence', 'cti', 'devmagnitude', 'eastness',\n",
       "       'elev', 'glad_ard_SVVI_max', 'glad_ard_SVVI_med',\n",
       "       'glad_ard_SVVI_min', 'northness', 'ORCDRC_WeigAver',\n",
       "       'outlet_dist_dw_basin', 'SBIO3_Isothermality_5_15cm',\n",
       "       'SBIO4_Temperature_Seasonality_5_15cm', 'treecover'], dtype=object)"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "feat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100000, 1)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100000, 18)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.5, random_state=24)\n",
    "y_train = np.ravel(Y_train)\n",
    "y_test = np.ravel(Y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Random Forest default parameters with select row base on quality flag\n",
    "Training Random Forest using default parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "rfReg = RandomForestRegressor(min_samples_leaf=20, oob_score=True)\n",
    "rfReg.fit(X_train, y_train);\n",
    "dic_pred = {}\n",
    "dic_pred['train'] = rfReg.predict(X_train)\n",
    "dic_pred['test'] = rfReg.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.37582774369799943"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# checking the oob score\n",
    "rfReg.oob_score_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Final assesment"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Selected data**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.747400101639897, 0.615726616845201]"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pearsonr_all_sel = [pearsonr(dic_pred['train'],y_train)[0],pearsonr(dic_pred['test'],y_test)[0]]\n",
    "pearsonr_all_sel"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**All data**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.6043347073198263, 0.5189400044676861]"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pearsonr_all"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Publication results** "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.4527925129990046"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pearsonr_Publication_Estimation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Tone the RF\n",
    "\n",
    "    pipeline = Pipeline([('rf',RandomForestRegressor())])\n",
    "\n",
    "    parameters = {\n",
    "            'rf__max_features':(\"log2\",\"sqrt\",0.33),\n",
    "            'rf__max_samples':(0.5,0.6,0.7),\n",
    "            'rf__n_estimators':(500,1000),\n",
    "            'rf__max_depth':(50,100,200)}\n",
    "\n",
    "    grid_search = GridSearchCV(pipeline,parameters,n_jobs=-1,cv=3,scoring='r2',verbose=1)\n",
    "    grid_search.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "    grid_search.best_score_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "    print ('Best Training score: %0.3f' % grid_search.best_score_)\n",
    "    print ('Optimal parameters:')\n",
    "    best_par = grid_search.best_estimator_.get_params()\n",
    "    for par_name in sorted(parameters.keys()):\n",
    "        print ('\\t%s: %r' % (par_name, best_par[par_name]))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "    rfReg = RandomForestRegressor(n_estimators=500,max_features='sqrt',max_depth=50,max_samples=0.6,n_jobs=-1,random_state=24)\n",
    "    rfReg.fit(X_train, y_train);\n",
    "    dic_pred = {}\n",
    "    dic_pred['train'] = rfReg.predict(X_train)\n",
    "    dic_pred['test'] = rfReg.predict(X_test)\n",
    "    [pearsonr(dic_pred['train'],y_train)[0],pearsonr(dic_pred['test'],y_test)[0]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fa0315145e0>]"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams[\"figure.figsize\"] = (8,6)\n",
    "plt.scatter(y_train,dic_pred['train'])\n",
    "plt.xlabel('trening GEDI Height (selected rows)')\n",
    "plt.ylabel('trening prediction')\n",
    "ident = [0, 50]\n",
    "plt.plot(ident,ident,'r--')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fa0226b4d30>]"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams[\"figure.figsize\"] = (8,6)\n",
    "plt.scatter(y_test,dic_pred['test'])\n",
    "plt.xlabel('testing GEDI Height (selected rows)')\n",
    "plt.ylabel('testing prediction')\n",
    "ident = [0, 50]\n",
    "plt.plot(ident,ident,'r--')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "impt = [rfReg.feature_importances_, np.std([tree.feature_importances_ for tree in rfReg.estimators_],axis=1)] \n",
    "ind = np.argsort(impt[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1, 13,  0,  6, 16,  4,  8,  5,  3,  2, 15, 11, 12,  9, 14,  7, 10,\n",
       "       17])"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ind"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x864 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams[\"figure.figsize\"] = (6,12)\n",
    "plt.barh(range(len(feat)),impt[0][ind],color=\"b\", xerr=impt[1][ind], align=\"center\")\n",
    "plt.yticks(range(len(feat)),feat[ind]);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Predict on the raster using pyspatialml"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import satalite indeces\n",
    "glad_ard_SVVI_min = \"tree_height/geodata_raster/glad_ard_SVVI_min.tif\"\n",
    "glad_ard_SVVI_med = \"tree_height/geodata_raster/glad_ard_SVVI_med.tif\"\n",
    "glad_ard_SVVI_max = \"tree_height/geodata_raster/glad_ard_SVVI_max.tif\"\n",
    "\n",
    "# import climate\n",
    "CHELSA_bio4 = \"tree_height/geodata_raster/CHELSA_bio4.tif\"\n",
    "CHELSA_bio18 = \"tree_height/geodata_raster/CHELSA_bio18.tif\"\n",
    "\n",
    "# soil\n",
    "BLDFIE_WeigAver = \"tree_height/geodata_raster/BLDFIE_WeigAver.tif\"\n",
    "CECSOL_WeigAver = \"tree_height/geodata_raster/CECSOL_WeigAver.tif\"\n",
    "ORCDRC_WeigAver = \"tree_height/geodata_raster/ORCDRC_WeigAver.tif\"\n",
    "\n",
    "# Geomorphological\n",
    "elev = \"tree_height/geodata_raster/elev.tif\"\n",
    "convergence = \"tree_height/geodata_raster/convergence.tif\"\n",
    "northness = \"tree_height/geodata_raster/northness.tif\"\n",
    "eastness = \"tree_height/geodata_raster/eastness.tif\"\n",
    "devmagnitude = \"tree_height/geodata_raster/devmagnitude.tif\"\n",
    "\n",
    "# Hydrography\n",
    "cti = \"tree_height/geodata_raster/cti.tif\"\n",
    "outlet_dist_dw_basin = \"tree_height/geodata_raster/outlet_dist_dw_basin.tif\"\n",
    "\n",
    "# Soil climate\n",
    "\n",
    "SBIO3_Isothermality_5_15cm = \"tree_height/geodata_raster/SBIO3_Isothermality_5_15cm.tif\"\n",
    "SBIO4_Temperature_Seasonality_5_15cm = \"tree_height/geodata_raster/SBIO4_Temperature_Seasonality_5_15cm.tif\"\n",
    "\n",
    "# forest\n",
    "\n",
    "treecover = \"tree_height/geodata_raster/treecover.tif\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'tree_height/geodata_raster/treecover.tif'"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "treecover"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "predictors_rasters = [glad_ard_SVVI_min, glad_ard_SVVI_med, glad_ard_SVVI_max,\n",
    "                      CHELSA_bio4,CHELSA_bio18,\n",
    "                      BLDFIE_WeigAver,CECSOL_WeigAver,ORCDRC_WeigAver,\n",
    "                      elev,convergence,northness,eastness,devmagnitude,cti,outlet_dist_dw_basin,\n",
    "                      SBIO3_Isothermality_5_15cm,SBIO4_Temperature_Seasonality_5_15cm,treecover]\n",
    "\n",
    "stack = Raster(predictors_rasters)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "result = stack.predict(estimator=rfReg, dtype='int16', nodata=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x864 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot regression result\n",
    "plt.rcParams[\"figure.figsize\"] = (12,12)\n",
    "result.iloc[0].cmap = \"plasma\"\n",
    "result.plot()\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "celltoolbar": "Slideshow",
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}