8. 预测体积模量.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "88621b94",
   "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>material_id</th>\n",
       "      <th>formula</th>\n",
       "      <th>nsites</th>\n",
       "      <th>space_group</th>\n",
       "      <th>volume</th>\n",
       "      <th>structure</th>\n",
       "      <th>elastic_anisotropy</th>\n",
       "      <th>G_Reuss</th>\n",
       "      <th>G_VRH</th>\n",
       "      <th>G_Voigt</th>\n",
       "      <th>K_Reuss</th>\n",
       "      <th>K_VRH</th>\n",
       "      <th>K_Voigt</th>\n",
       "      <th>poisson_ratio</th>\n",
       "      <th>compliance_tensor</th>\n",
       "      <th>elastic_tensor</th>\n",
       "      <th>elastic_tensor_original</th>\n",
       "      <th>cif</th>\n",
       "      <th>kpoint_density</th>\n",
       "      <th>poscar</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mp-10003</td>\n",
       "      <td>Nb4CoSi</td>\n",
       "      <td>12</td>\n",
       "      <td>124</td>\n",
       "      <td>194.419802</td>\n",
       "      <td>[[0.94814328 2.07280467 2.5112    ] Nb, [5.273...</td>\n",
       "      <td>0.030688</td>\n",
       "      <td>96.844535</td>\n",
       "      <td>97.141604</td>\n",
       "      <td>97.438674</td>\n",
       "      <td>194.267623</td>\n",
       "      <td>194.268884</td>\n",
       "      <td>194.270146</td>\n",
       "      <td>0.285701</td>\n",
       "      <td>[[0.004385293093993, -0.0016070693558990002, -...</td>\n",
       "      <td>[[311.33514638650246, 144.45092552856926, 126....</td>\n",
       "      <td>[[311.33514638650246, 144.45092552856926, 126....</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>7000</td>\n",
       "      <td>Nb8 Co2 Si2\\n1.0\\n6.221780 0.000000 0.000000\\n...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mp-10010</td>\n",
       "      <td>Al(CoSi)2</td>\n",
       "      <td>5</td>\n",
       "      <td>164</td>\n",
       "      <td>61.987320</td>\n",
       "      <td>[[0. 0. 0.] Al, [1.96639263 1.13529553 0.75278...</td>\n",
       "      <td>0.266910</td>\n",
       "      <td>93.939650</td>\n",
       "      <td>96.252006</td>\n",
       "      <td>98.564362</td>\n",
       "      <td>173.647763</td>\n",
       "      <td>175.449907</td>\n",
       "      <td>177.252050</td>\n",
       "      <td>0.268105</td>\n",
       "      <td>[[0.0037715428949660003, -0.000844229828709, -...</td>\n",
       "      <td>[[306.93357350984974, 88.02634955100905, 105.6...</td>\n",
       "      <td>[[306.93357350984974, 88.02634955100905, 105.6...</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>7000</td>\n",
       "      <td>Al1 Co2 Si2\\n1.0\\n3.932782 0.000000 0.000000\\n...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>mp-10015</td>\n",
       "      <td>SiOs</td>\n",
       "      <td>2</td>\n",
       "      <td>221</td>\n",
       "      <td>25.952539</td>\n",
       "      <td>[[1.480346 1.480346 1.480346] Si, [0. 0. 0.] Os]</td>\n",
       "      <td>0.756489</td>\n",
       "      <td>120.962289</td>\n",
       "      <td>130.112955</td>\n",
       "      <td>139.263621</td>\n",
       "      <td>295.077545</td>\n",
       "      <td>295.077545</td>\n",
       "      <td>295.077545</td>\n",
       "      <td>0.307780</td>\n",
       "      <td>[[0.0019959391925840004, -0.000433146670736000...</td>\n",
       "      <td>[[569.5291276937579, 157.8517489654999, 157.85...</td>\n",
       "      <td>[[569.5291276937579, 157.8517489654999, 157.85...</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>7000</td>\n",
       "      <td>Si1 Os1\\n1.0\\n2.960692 0.000000 0.000000\\n0.00...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>mp-10021</td>\n",
       "      <td>Ga</td>\n",
       "      <td>4</td>\n",
       "      <td>63</td>\n",
       "      <td>76.721433</td>\n",
       "      <td>[[0.         1.09045794 0.84078375] Ga, [0.   ...</td>\n",
       "      <td>2.376805</td>\n",
       "      <td>12.205989</td>\n",
       "      <td>15.101901</td>\n",
       "      <td>17.997812</td>\n",
       "      <td>49.025963</td>\n",
       "      <td>49.130670</td>\n",
       "      <td>49.235377</td>\n",
       "      <td>0.360593</td>\n",
       "      <td>[[0.021647143908635, -0.005207263618160001, -0...</td>\n",
       "      <td>[[69.28798774976904, 34.7875015216915, 37.3877...</td>\n",
       "      <td>[[70.13259066665267, 40.60474945058445, 37.387...</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>7000</td>\n",
       "      <td>Ga4\\n1.0\\n2.803229 0.000000 0.000000\\n0.000000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>mp-10025</td>\n",
       "      <td>SiRu2</td>\n",
       "      <td>12</td>\n",
       "      <td>62</td>\n",
       "      <td>160.300999</td>\n",
       "      <td>[[1.0094265  4.24771709 2.9955487 ] Si, [3.028...</td>\n",
       "      <td>0.196930</td>\n",
       "      <td>100.110773</td>\n",
       "      <td>101.947798</td>\n",
       "      <td>103.784823</td>\n",
       "      <td>255.055257</td>\n",
       "      <td>256.768081</td>\n",
       "      <td>258.480904</td>\n",
       "      <td>0.324682</td>\n",
       "      <td>[[0.00410214297725, -0.001272204332729, -0.001...</td>\n",
       "      <td>[[349.3767766177825, 186.67131003104407, 176.4...</td>\n",
       "      <td>[[407.4791016459293, 176.4759188081947, 213.83...</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>7000</td>\n",
       "      <td>Si4 Ru8\\n1.0\\n4.037706 0.000000 0.000000\\n0.00...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  material_id    formula  nsites  space_group      volume  \\\n",
       "0    mp-10003    Nb4CoSi      12          124  194.419802   \n",
       "1    mp-10010  Al(CoSi)2       5          164   61.987320   \n",
       "2    mp-10015       SiOs       2          221   25.952539   \n",
       "3    mp-10021         Ga       4           63   76.721433   \n",
       "4    mp-10025      SiRu2      12           62  160.300999   \n",
       "\n",
       "                                           structure  elastic_anisotropy  \\\n",
       "0  [[0.94814328 2.07280467 2.5112    ] Nb, [5.273...            0.030688   \n",
       "1  [[0. 0. 0.] Al, [1.96639263 1.13529553 0.75278...            0.266910   \n",
       "2   [[1.480346 1.480346 1.480346] Si, [0. 0. 0.] Os]            0.756489   \n",
       "3  [[0.         1.09045794 0.84078375] Ga, [0.   ...            2.376805   \n",
       "4  [[1.0094265  4.24771709 2.9955487 ] Si, [3.028...            0.196930   \n",
       "\n",
       "      G_Reuss       G_VRH     G_Voigt     K_Reuss       K_VRH     K_Voigt  \\\n",
       "0   96.844535   97.141604   97.438674  194.267623  194.268884  194.270146   \n",
       "1   93.939650   96.252006   98.564362  173.647763  175.449907  177.252050   \n",
       "2  120.962289  130.112955  139.263621  295.077545  295.077545  295.077545   \n",
       "3   12.205989   15.101901   17.997812   49.025963   49.130670   49.235377   \n",
       "4  100.110773  101.947798  103.784823  255.055257  256.768081  258.480904   \n",
       "\n",
       "   poisson_ratio                                  compliance_tensor  \\\n",
       "0       0.285701  [[0.004385293093993, -0.0016070693558990002, -...   \n",
       "1       0.268105  [[0.0037715428949660003, -0.000844229828709, -...   \n",
       "2       0.307780  [[0.0019959391925840004, -0.000433146670736000...   \n",
       "3       0.360593  [[0.021647143908635, -0.005207263618160001, -0...   \n",
       "4       0.324682  [[0.00410214297725, -0.001272204332729, -0.001...   \n",
       "\n",
       "                                      elastic_tensor  \\\n",
       "0  [[311.33514638650246, 144.45092552856926, 126....   \n",
       "1  [[306.93357350984974, 88.02634955100905, 105.6...   \n",
       "2  [[569.5291276937579, 157.8517489654999, 157.85...   \n",
       "3  [[69.28798774976904, 34.7875015216915, 37.3877...   \n",
       "4  [[349.3767766177825, 186.67131003104407, 176.4...   \n",
       "\n",
       "                             elastic_tensor_original  \\\n",
       "0  [[311.33514638650246, 144.45092552856926, 126....   \n",
       "1  [[306.93357350984974, 88.02634955100905, 105.6...   \n",
       "2  [[569.5291276937579, 157.8517489654999, 157.85...   \n",
       "3  [[70.13259066665267, 40.60474945058445, 37.387...   \n",
       "4  [[407.4791016459293, 176.4759188081947, 213.83...   \n",
       "\n",
       "                                                 cif  kpoint_density  \\\n",
       "0  #\\#CIF1.1\\n###################################...            7000   \n",
       "1  #\\#CIF1.1\\n###################################...            7000   \n",
       "2  #\\#CIF1.1\\n###################################...            7000   \n",
       "3  #\\#CIF1.1\\n###################################...            7000   \n",
       "4  #\\#CIF1.1\\n###################################...            7000   \n",
       "\n",
       "                                              poscar  \n",
       "0  Nb8 Co2 Si2\\n1.0\\n6.221780 0.000000 0.000000\\n...  \n",
       "1  Al1 Co2 Si2\\n1.0\\n3.932782 0.000000 0.000000\\n...  \n",
       "2  Si1 Os1\\n1.0\\n2.960692 0.000000 0.000000\\n0.00...  \n",
       "3  Ga4\\n1.0\\n2.803229 0.000000 0.000000\\n0.000000...  \n",
       "4  Si4 Ru8\\n1.0\\n4.037706 0.000000 0.000000\\n0.00...  "
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from matminer.datasets import load_dataset\n",
    "df = load_dataset('elastic_tensor_2015',data_home = '.')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "4f32a242",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['material_id', 'formula', 'nsites', 'space_group', 'volume',\n",
       "       'structure', 'elastic_anisotropy', 'G_Reuss', 'G_VRH', 'G_Voigt',\n",
       "       'K_Reuss', 'K_VRH', 'K_Voigt', 'poisson_ratio', 'compliance_tensor',\n",
       "       'elastic_tensor', 'elastic_tensor_original', 'cif', 'kpoint_density',\n",
       "       'poscar'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b8d8a75d",
   "metadata": {},
   "source": [
    "df = df['cif','kpoint_density']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e44f02ad",
   "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>material_id</th>\n",
       "      <th>formula</th>\n",
       "      <th>space_group</th>\n",
       "      <th>structure</th>\n",
       "      <th>elastic_anisotropy</th>\n",
       "      <th>G_VRH</th>\n",
       "      <th>K_VRH</th>\n",
       "      <th>poisson_ratio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mp-10003</td>\n",
       "      <td>Nb4CoSi</td>\n",
       "      <td>124</td>\n",
       "      <td>[[0.94814328 2.07280467 2.5112    ] Nb, [5.273...</td>\n",
       "      <td>0.030688</td>\n",
       "      <td>97.141604</td>\n",
       "      <td>194.268884</td>\n",
       "      <td>0.285701</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mp-10010</td>\n",
       "      <td>Al(CoSi)2</td>\n",
       "      <td>164</td>\n",
       "      <td>[[0. 0. 0.] Al, [1.96639263 1.13529553 0.75278...</td>\n",
       "      <td>0.266910</td>\n",
       "      <td>96.252006</td>\n",
       "      <td>175.449907</td>\n",
       "      <td>0.268105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>mp-10015</td>\n",
       "      <td>SiOs</td>\n",
       "      <td>221</td>\n",
       "      <td>[[1.480346 1.480346 1.480346] Si, [0. 0. 0.] Os]</td>\n",
       "      <td>0.756489</td>\n",
       "      <td>130.112955</td>\n",
       "      <td>295.077545</td>\n",
       "      <td>0.307780</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>mp-10021</td>\n",
       "      <td>Ga</td>\n",
       "      <td>63</td>\n",
       "      <td>[[0.         1.09045794 0.84078375] Ga, [0.   ...</td>\n",
       "      <td>2.376805</td>\n",
       "      <td>15.101901</td>\n",
       "      <td>49.130670</td>\n",
       "      <td>0.360593</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>mp-10025</td>\n",
       "      <td>SiRu2</td>\n",
       "      <td>62</td>\n",
       "      <td>[[1.0094265  4.24771709 2.9955487 ] Si, [3.028...</td>\n",
       "      <td>0.196930</td>\n",
       "      <td>101.947798</td>\n",
       "      <td>256.768081</td>\n",
       "      <td>0.324682</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  material_id    formula  space_group  \\\n",
       "0    mp-10003    Nb4CoSi          124   \n",
       "1    mp-10010  Al(CoSi)2          164   \n",
       "2    mp-10015       SiOs          221   \n",
       "3    mp-10021         Ga           63   \n",
       "4    mp-10025      SiRu2           62   \n",
       "\n",
       "                                           structure  elastic_anisotropy  \\\n",
       "0  [[0.94814328 2.07280467 2.5112    ] Nb, [5.273...            0.030688   \n",
       "1  [[0. 0. 0.] Al, [1.96639263 1.13529553 0.75278...            0.266910   \n",
       "2   [[1.480346 1.480346 1.480346] Si, [0. 0. 0.] Os]            0.756489   \n",
       "3  [[0.         1.09045794 0.84078375] Ga, [0.   ...            2.376805   \n",
       "4  [[1.0094265  4.24771709 2.9955487 ] Si, [3.028...            0.196930   \n",
       "\n",
       "        G_VRH       K_VRH  poisson_ratio  \n",
       "0   97.141604  194.268884       0.285701  \n",
       "1   96.252006  175.449907       0.268105  \n",
       "2  130.112955  295.077545       0.307780  \n",
       "3   15.101901   49.130670       0.360593  \n",
       "4  101.947798  256.768081       0.324682  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unwanted_columns = ['volume','nsites','compliance_tensor','elastic_tensor',\n",
    "                   'elastic_tensor_original','G_Reuss','G_Voigt','K_Reuss',\n",
    "                    'K_Voigt','cif','kpoint_density','poscar']\n",
    "df = df.drop(unwanted_columns, axis = 1)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "30170480",
   "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>space_group</th>\n",
       "      <th>elastic_anisotropy</th>\n",
       "      <th>G_VRH</th>\n",
       "      <th>K_VRH</th>\n",
       "      <th>poisson_ratio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1181.000000</td>\n",
       "      <td>1181.000000</td>\n",
       "      <td>1181.000000</td>\n",
       "      <td>1181.000000</td>\n",
       "      <td>1181.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>163.403895</td>\n",
       "      <td>2.145013</td>\n",
       "      <td>67.543145</td>\n",
       "      <td>136.259661</td>\n",
       "      <td>0.287401</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>65.040733</td>\n",
       "      <td>19.140097</td>\n",
       "      <td>44.579408</td>\n",
       "      <td>72.886978</td>\n",
       "      <td>0.062177</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>4.000000</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>2.722175</td>\n",
       "      <td>6.476135</td>\n",
       "      <td>0.042582</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>124.000000</td>\n",
       "      <td>0.145030</td>\n",
       "      <td>34.117959</td>\n",
       "      <td>76.435350</td>\n",
       "      <td>0.249159</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>193.000000</td>\n",
       "      <td>0.355287</td>\n",
       "      <td>59.735163</td>\n",
       "      <td>130.382766</td>\n",
       "      <td>0.290198</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>221.000000</td>\n",
       "      <td>0.923117</td>\n",
       "      <td>91.332142</td>\n",
       "      <td>189.574194</td>\n",
       "      <td>0.328808</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>229.000000</td>\n",
       "      <td>397.297866</td>\n",
       "      <td>522.921225</td>\n",
       "      <td>435.661487</td>\n",
       "      <td>0.467523</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       space_group  elastic_anisotropy        G_VRH        K_VRH  \\\n",
       "count  1181.000000         1181.000000  1181.000000  1181.000000   \n",
       "mean    163.403895            2.145013    67.543145   136.259661   \n",
       "std      65.040733           19.140097    44.579408    72.886978   \n",
       "min       4.000000            0.000005     2.722175     6.476135   \n",
       "25%     124.000000            0.145030    34.117959    76.435350   \n",
       "50%     193.000000            0.355287    59.735163   130.382766   \n",
       "75%     221.000000            0.923117    91.332142   189.574194   \n",
       "max     229.000000          397.297866   522.921225   435.661487   \n",
       "\n",
       "       poisson_ratio  \n",
       "count    1181.000000  \n",
       "mean        0.287401  \n",
       "std         0.062177  \n",
       "min         0.042582  \n",
       "25%         0.249159  \n",
       "50%         0.290198  \n",
       "75%         0.328808  \n",
       "max         0.467523  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "52ee1f04",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6d70ada1f547499f9f79a926e5aeedad",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "StrToComposition:   0%|          | 0/1181 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>material_id</th>\n",
       "      <th>formula</th>\n",
       "      <th>space_group</th>\n",
       "      <th>structure</th>\n",
       "      <th>elastic_anisotropy</th>\n",
       "      <th>G_VRH</th>\n",
       "      <th>K_VRH</th>\n",
       "      <th>poisson_ratio</th>\n",
       "      <th>composition</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mp-10003</td>\n",
       "      <td>Nb4CoSi</td>\n",
       "      <td>124</td>\n",
       "      <td>[[0.94814328 2.07280467 2.5112    ] Nb, [5.273...</td>\n",
       "      <td>0.030688</td>\n",
       "      <td>97.141604</td>\n",
       "      <td>194.268884</td>\n",
       "      <td>0.285701</td>\n",
       "      <td>(Nb, Co, Si)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mp-10010</td>\n",
       "      <td>Al(CoSi)2</td>\n",
       "      <td>164</td>\n",
       "      <td>[[0. 0. 0.] Al, [1.96639263 1.13529553 0.75278...</td>\n",
       "      <td>0.266910</td>\n",
       "      <td>96.252006</td>\n",
       "      <td>175.449907</td>\n",
       "      <td>0.268105</td>\n",
       "      <td>(Al, Co, Si)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>mp-10015</td>\n",
       "      <td>SiOs</td>\n",
       "      <td>221</td>\n",
       "      <td>[[1.480346 1.480346 1.480346] Si, [0. 0. 0.] Os]</td>\n",
       "      <td>0.756489</td>\n",
       "      <td>130.112955</td>\n",
       "      <td>295.077545</td>\n",
       "      <td>0.307780</td>\n",
       "      <td>(Si, Os)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>mp-10021</td>\n",
       "      <td>Ga</td>\n",
       "      <td>63</td>\n",
       "      <td>[[0.         1.09045794 0.84078375] Ga, [0.   ...</td>\n",
       "      <td>2.376805</td>\n",
       "      <td>15.101901</td>\n",
       "      <td>49.130670</td>\n",
       "      <td>0.360593</td>\n",
       "      <td>(Ga)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>mp-10025</td>\n",
       "      <td>SiRu2</td>\n",
       "      <td>62</td>\n",
       "      <td>[[1.0094265  4.24771709 2.9955487 ] Si, [3.028...</td>\n",
       "      <td>0.196930</td>\n",
       "      <td>101.947798</td>\n",
       "      <td>256.768081</td>\n",
       "      <td>0.324682</td>\n",
       "      <td>(Si, Ru)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  material_id    formula  space_group  \\\n",
       "0    mp-10003    Nb4CoSi          124   \n",
       "1    mp-10010  Al(CoSi)2          164   \n",
       "2    mp-10015       SiOs          221   \n",
       "3    mp-10021         Ga           63   \n",
       "4    mp-10025      SiRu2           62   \n",
       "\n",
       "                                           structure  elastic_anisotropy  \\\n",
       "0  [[0.94814328 2.07280467 2.5112    ] Nb, [5.273...            0.030688   \n",
       "1  [[0. 0. 0.] Al, [1.96639263 1.13529553 0.75278...            0.266910   \n",
       "2   [[1.480346 1.480346 1.480346] Si, [0. 0. 0.] Os]            0.756489   \n",
       "3  [[0.         1.09045794 0.84078375] Ga, [0.   ...            2.376805   \n",
       "4  [[1.0094265  4.24771709 2.9955487 ] Si, [3.028...            0.196930   \n",
       "\n",
       "        G_VRH       K_VRH  poisson_ratio   composition  \n",
       "0   97.141604  194.268884       0.285701  (Nb, Co, Si)  \n",
       "1   96.252006  175.449907       0.268105  (Al, Co, Si)  \n",
       "2  130.112955  295.077545       0.307780      (Si, Os)  \n",
       "3   15.101901   49.130670       0.360593          (Ga)  \n",
       "4  101.947798  256.768081       0.324682      (Si, Ru)  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from matminer.featurizers.conversions import StrToComposition\n",
    "df = StrToComposition().featurize_dataframe(df, 'formula')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "a974600a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b041b831537b4bf9ae64342e87169240",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "ElementProperty:   0%|          | 0/1181 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>material_id</th>\n",
       "      <th>formula</th>\n",
       "      <th>space_group</th>\n",
       "      <th>structure</th>\n",
       "      <th>elastic_anisotropy</th>\n",
       "      <th>G_VRH</th>\n",
       "      <th>K_VRH</th>\n",
       "      <th>poisson_ratio</th>\n",
       "      <th>composition</th>\n",
       "      <th>MagpieData minimum Number</th>\n",
       "      <th>...</th>\n",
       "      <th>MagpieData range GSmagmom</th>\n",
       "      <th>MagpieData mean GSmagmom</th>\n",
       "      <th>MagpieData avg_dev GSmagmom</th>\n",
       "      <th>MagpieData mode GSmagmom</th>\n",
       "      <th>MagpieData minimum SpaceGroupNumber</th>\n",
       "      <th>MagpieData maximum SpaceGroupNumber</th>\n",
       "      <th>MagpieData range SpaceGroupNumber</th>\n",
       "      <th>MagpieData mean SpaceGroupNumber</th>\n",
       "      <th>MagpieData avg_dev SpaceGroupNumber</th>\n",
       "      <th>MagpieData mode SpaceGroupNumber</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mp-10003</td>\n",
       "      <td>Nb4CoSi</td>\n",
       "      <td>124</td>\n",
       "      <td>[[0.94814328 2.07280467 2.5112    ] Nb, [5.273...</td>\n",
       "      <td>0.030688</td>\n",
       "      <td>97.141604</td>\n",
       "      <td>194.268884</td>\n",
       "      <td>0.285701</td>\n",
       "      <td>(Nb, Co, Si)</td>\n",
       "      <td>14.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.548471</td>\n",
       "      <td>0.258079</td>\n",
       "      <td>0.430131</td>\n",
       "      <td>0.0</td>\n",
       "      <td>194.0</td>\n",
       "      <td>229.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>222.833333</td>\n",
       "      <td>9.611111</td>\n",
       "      <td>229.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mp-10010</td>\n",
       "      <td>Al(CoSi)2</td>\n",
       "      <td>164</td>\n",
       "      <td>[[0. 0. 0.] Al, [1.96639263 1.13529553 0.75278...</td>\n",
       "      <td>0.266910</td>\n",
       "      <td>96.252006</td>\n",
       "      <td>175.449907</td>\n",
       "      <td>0.268105</td>\n",
       "      <td>(Al, Co, Si)</td>\n",
       "      <td>13.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.548471</td>\n",
       "      <td>0.619388</td>\n",
       "      <td>0.743266</td>\n",
       "      <td>0.0</td>\n",
       "      <td>194.0</td>\n",
       "      <td>227.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>213.400000</td>\n",
       "      <td>15.520000</td>\n",
       "      <td>194.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>mp-10015</td>\n",
       "      <td>SiOs</td>\n",
       "      <td>221</td>\n",
       "      <td>[[1.480346 1.480346 1.480346] Si, [0. 0. 0.] Os]</td>\n",
       "      <td>0.756489</td>\n",
       "      <td>130.112955</td>\n",
       "      <td>295.077545</td>\n",
       "      <td>0.307780</td>\n",
       "      <td>(Si, Os)</td>\n",
       "      <td>14.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>194.0</td>\n",
       "      <td>227.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>210.500000</td>\n",
       "      <td>16.500000</td>\n",
       "      <td>194.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>mp-10021</td>\n",
       "      <td>Ga</td>\n",
       "      <td>63</td>\n",
       "      <td>[[0.         1.09045794 0.84078375] Ga, [0.   ...</td>\n",
       "      <td>2.376805</td>\n",
       "      <td>15.101901</td>\n",
       "      <td>49.130670</td>\n",
       "      <td>0.360593</td>\n",
       "      <td>(Ga)</td>\n",
       "      <td>31.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>64.0</td>\n",
       "      <td>64.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>64.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>64.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>mp-10025</td>\n",
       "      <td>SiRu2</td>\n",
       "      <td>62</td>\n",
       "      <td>[[1.0094265  4.24771709 2.9955487 ] Si, [3.028...</td>\n",
       "      <td>0.196930</td>\n",
       "      <td>101.947798</td>\n",
       "      <td>256.768081</td>\n",
       "      <td>0.324682</td>\n",
       "      <td>(Si, Ru)</td>\n",
       "      <td>14.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>194.0</td>\n",
       "      <td>227.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>205.000000</td>\n",
       "      <td>14.666667</td>\n",
       "      <td>194.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 141 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  material_id    formula  space_group  \\\n",
       "0    mp-10003    Nb4CoSi          124   \n",
       "1    mp-10010  Al(CoSi)2          164   \n",
       "2    mp-10015       SiOs          221   \n",
       "3    mp-10021         Ga           63   \n",
       "4    mp-10025      SiRu2           62   \n",
       "\n",
       "                                           structure  elastic_anisotropy  \\\n",
       "0  [[0.94814328 2.07280467 2.5112    ] Nb, [5.273...            0.030688   \n",
       "1  [[0. 0. 0.] Al, [1.96639263 1.13529553 0.75278...            0.266910   \n",
       "2   [[1.480346 1.480346 1.480346] Si, [0. 0. 0.] Os]            0.756489   \n",
       "3  [[0.         1.09045794 0.84078375] Ga, [0.   ...            2.376805   \n",
       "4  [[1.0094265  4.24771709 2.9955487 ] Si, [3.028...            0.196930   \n",
       "\n",
       "        G_VRH       K_VRH  poisson_ratio   composition  \\\n",
       "0   97.141604  194.268884       0.285701  (Nb, Co, Si)   \n",
       "1   96.252006  175.449907       0.268105  (Al, Co, Si)   \n",
       "2  130.112955  295.077545       0.307780      (Si, Os)   \n",
       "3   15.101901   49.130670       0.360593          (Ga)   \n",
       "4  101.947798  256.768081       0.324682      (Si, Ru)   \n",
       "\n",
       "   MagpieData minimum Number  ...  MagpieData range GSmagmom  \\\n",
       "0                       14.0  ...                   1.548471   \n",
       "1                       13.0  ...                   1.548471   \n",
       "2                       14.0  ...                   0.000000   \n",
       "3                       31.0  ...                   0.000000   \n",
       "4                       14.0  ...                   0.000000   \n",
       "\n",
       "   MagpieData mean GSmagmom  MagpieData avg_dev GSmagmom  \\\n",
       "0                  0.258079                     0.430131   \n",
       "1                  0.619388                     0.743266   \n",
       "2                  0.000000                     0.000000   \n",
       "3                  0.000000                     0.000000   \n",
       "4                  0.000000                     0.000000   \n",
       "\n",
       "   MagpieData mode GSmagmom  MagpieData minimum SpaceGroupNumber  \\\n",
       "0                       0.0                                194.0   \n",
       "1                       0.0                                194.0   \n",
       "2                       0.0                                194.0   \n",
       "3                       0.0                                 64.0   \n",
       "4                       0.0                                194.0   \n",
       "\n",
       "   MagpieData maximum SpaceGroupNumber  MagpieData range SpaceGroupNumber  \\\n",
       "0                                229.0                               35.0   \n",
       "1                                227.0                               33.0   \n",
       "2                                227.0                               33.0   \n",
       "3                                 64.0                                0.0   \n",
       "4                                227.0                               33.0   \n",
       "\n",
       "   MagpieData mean SpaceGroupNumber  MagpieData avg_dev SpaceGroupNumber  \\\n",
       "0                        222.833333                             9.611111   \n",
       "1                        213.400000                            15.520000   \n",
       "2                        210.500000                            16.500000   \n",
       "3                         64.000000                             0.000000   \n",
       "4                        205.000000                            14.666667   \n",
       "\n",
       "   MagpieData mode SpaceGroupNumber  \n",
       "0                             229.0  \n",
       "1                             194.0  \n",
       "2                             194.0  \n",
       "3                              64.0  \n",
       "4                             194.0  \n",
       "\n",
       "[5 rows x 141 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from matminer.featurizers.composition import ElementProperty\n",
    "ep_feat = ElementProperty.from_preset(preset_name = 'magpie')\n",
    "df = ep_feat.featurize_dataframe(df, col_id = 'composition')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "e39d9d7c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Help on method from_preset in module matminer.featurizers.composition.composite:\n",
      "\n",
      "from_preset(preset_name) method of abc.ABCMeta instance\n",
      "    Return ElementProperty from a preset string\n",
      "    Args:\n",
      "        preset_name: (str) can be one of \"magpie\", \"deml\", \"matminer\",\n",
      "            \"matscholar_el\", or \"megnet_el\".\n",
      "    \n",
      "    Returns:\n",
      "        ElementProperty based on the preset name.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "help(ElementProperty.from_preset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "e2d6f85d",
   "metadata": {},
   "outputs": [],
   "source": [
    "from matminer.featurizers.conversions import CompositionToOxidComposition\n",
    "from matminer.featurizers.composition import OxidationStates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "531675d5",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2de30313a34d42139db9a42b9378aaec",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "CompositionToOxidComposition:   0%|          | 0/1181 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "659201aee1f04363bc570164fe0d6bef",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "OxidationStates:   0%|          | 0/1181 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>material_id</th>\n",
       "      <th>formula</th>\n",
       "      <th>space_group</th>\n",
       "      <th>structure</th>\n",
       "      <th>elastic_anisotropy</th>\n",
       "      <th>G_VRH</th>\n",
       "      <th>K_VRH</th>\n",
       "      <th>poisson_ratio</th>\n",
       "      <th>composition</th>\n",
       "      <th>MagpieData minimum Number</th>\n",
       "      <th>...</th>\n",
       "      <th>MagpieData maximum SpaceGroupNumber</th>\n",
       "      <th>MagpieData range SpaceGroupNumber</th>\n",
       "      <th>MagpieData mean SpaceGroupNumber</th>\n",
       "      <th>MagpieData avg_dev SpaceGroupNumber</th>\n",
       "      <th>MagpieData mode SpaceGroupNumber</th>\n",
       "      <th>composition_oxid</th>\n",
       "      <th>minimum oxidation state</th>\n",
       "      <th>maximum oxidation state</th>\n",
       "      <th>range oxidation state</th>\n",
       "      <th>std_dev oxidation state</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mp-10003</td>\n",
       "      <td>Nb4CoSi</td>\n",
       "      <td>124</td>\n",
       "      <td>[[0.94814328 2.07280467 2.5112    ] Nb, [5.273...</td>\n",
       "      <td>0.030688</td>\n",
       "      <td>97.141604</td>\n",
       "      <td>194.268884</td>\n",
       "      <td>0.285701</td>\n",
       "      <td>(Nb, Co, Si)</td>\n",
       "      <td>14.0</td>\n",
       "      <td>...</td>\n",
       "      <td>229.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>222.833333</td>\n",
       "      <td>9.611111</td>\n",
       "      <td>229.0</td>\n",
       "      <td>(Nb0+, Co0+, Si0+)</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mp-10010</td>\n",
       "      <td>Al(CoSi)2</td>\n",
       "      <td>164</td>\n",
       "      <td>[[0. 0. 0.] Al, [1.96639263 1.13529553 0.75278...</td>\n",
       "      <td>0.266910</td>\n",
       "      <td>96.252006</td>\n",
       "      <td>175.449907</td>\n",
       "      <td>0.268105</td>\n",
       "      <td>(Al, Co, Si)</td>\n",
       "      <td>13.0</td>\n",
       "      <td>...</td>\n",
       "      <td>227.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>213.400000</td>\n",
       "      <td>15.520000</td>\n",
       "      <td>194.0</td>\n",
       "      <td>(Al3+, Co2+, Co3+, Si4-)</td>\n",
       "      <td>-4</td>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>3.872983</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>mp-10015</td>\n",
       "      <td>SiOs</td>\n",
       "      <td>221</td>\n",
       "      <td>[[1.480346 1.480346 1.480346] Si, [0. 0. 0.] Os]</td>\n",
       "      <td>0.756489</td>\n",
       "      <td>130.112955</td>\n",
       "      <td>295.077545</td>\n",
       "      <td>0.307780</td>\n",
       "      <td>(Si, Os)</td>\n",
       "      <td>14.0</td>\n",
       "      <td>...</td>\n",
       "      <td>227.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>210.500000</td>\n",
       "      <td>16.500000</td>\n",
       "      <td>194.0</td>\n",
       "      <td>(Si4-, Os4+)</td>\n",
       "      <td>-4</td>\n",
       "      <td>4</td>\n",
       "      <td>8</td>\n",
       "      <td>5.656854</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>mp-10021</td>\n",
       "      <td>Ga</td>\n",
       "      <td>63</td>\n",
       "      <td>[[0.         1.09045794 0.84078375] Ga, [0.   ...</td>\n",
       "      <td>2.376805</td>\n",
       "      <td>15.101901</td>\n",
       "      <td>49.130670</td>\n",
       "      <td>0.360593</td>\n",
       "      <td>(Ga)</td>\n",
       "      <td>31.0</td>\n",
       "      <td>...</td>\n",
       "      <td>64.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>64.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>64.0</td>\n",
       "      <td>(Ga0+)</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>mp-10025</td>\n",
       "      <td>SiRu2</td>\n",
       "      <td>62</td>\n",
       "      <td>[[1.0094265  4.24771709 2.9955487 ] Si, [3.028...</td>\n",
       "      <td>0.196930</td>\n",
       "      <td>101.947798</td>\n",
       "      <td>256.768081</td>\n",
       "      <td>0.324682</td>\n",
       "      <td>(Si, Ru)</td>\n",
       "      <td>14.0</td>\n",
       "      <td>...</td>\n",
       "      <td>227.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>205.000000</td>\n",
       "      <td>14.666667</td>\n",
       "      <td>194.0</td>\n",
       "      <td>(Si4-, Ru2+)</td>\n",
       "      <td>-4</td>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>4.242641</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 146 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  material_id    formula  space_group  \\\n",
       "0    mp-10003    Nb4CoSi          124   \n",
       "1    mp-10010  Al(CoSi)2          164   \n",
       "2    mp-10015       SiOs          221   \n",
       "3    mp-10021         Ga           63   \n",
       "4    mp-10025      SiRu2           62   \n",
       "\n",
       "                                           structure  elastic_anisotropy  \\\n",
       "0  [[0.94814328 2.07280467 2.5112    ] Nb, [5.273...            0.030688   \n",
       "1  [[0. 0. 0.] Al, [1.96639263 1.13529553 0.75278...            0.266910   \n",
       "2   [[1.480346 1.480346 1.480346] Si, [0. 0. 0.] Os]            0.756489   \n",
       "3  [[0.         1.09045794 0.84078375] Ga, [0.   ...            2.376805   \n",
       "4  [[1.0094265  4.24771709 2.9955487 ] Si, [3.028...            0.196930   \n",
       "\n",
       "        G_VRH       K_VRH  poisson_ratio   composition  \\\n",
       "0   97.141604  194.268884       0.285701  (Nb, Co, Si)   \n",
       "1   96.252006  175.449907       0.268105  (Al, Co, Si)   \n",
       "2  130.112955  295.077545       0.307780      (Si, Os)   \n",
       "3   15.101901   49.130670       0.360593          (Ga)   \n",
       "4  101.947798  256.768081       0.324682      (Si, Ru)   \n",
       "\n",
       "   MagpieData minimum Number  ...  MagpieData maximum SpaceGroupNumber  \\\n",
       "0                       14.0  ...                                229.0   \n",
       "1                       13.0  ...                                227.0   \n",
       "2                       14.0  ...                                227.0   \n",
       "3                       31.0  ...                                 64.0   \n",
       "4                       14.0  ...                                227.0   \n",
       "\n",
       "   MagpieData range SpaceGroupNumber  MagpieData mean SpaceGroupNumber  \\\n",
       "0                               35.0                        222.833333   \n",
       "1                               33.0                        213.400000   \n",
       "2                               33.0                        210.500000   \n",
       "3                                0.0                         64.000000   \n",
       "4                               33.0                        205.000000   \n",
       "\n",
       "   MagpieData avg_dev SpaceGroupNumber  MagpieData mode SpaceGroupNumber  \\\n",
       "0                             9.611111                             229.0   \n",
       "1                            15.520000                             194.0   \n",
       "2                            16.500000                             194.0   \n",
       "3                             0.000000                              64.0   \n",
       "4                            14.666667                             194.0   \n",
       "\n",
       "           composition_oxid  minimum oxidation state  maximum oxidation state  \\\n",
       "0        (Nb0+, Co0+, Si0+)                        0                        0   \n",
       "1  (Al3+, Co2+, Co3+, Si4-)                       -4                        3   \n",
       "2              (Si4-, Os4+)                       -4                        4   \n",
       "3                    (Ga0+)                        0                        0   \n",
       "4              (Si4-, Ru2+)                       -4                        2   \n",
       "\n",
       "   range oxidation state  std_dev oxidation state  \n",
       "0                      0                 0.000000  \n",
       "1                      7                 3.872983  \n",
       "2                      8                 5.656854  \n",
       "3                      0                 0.000000  \n",
       "4                      6                 4.242641  \n",
       "\n",
       "[5 rows x 146 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = CompositionToOxidComposition().featurize_dataframe(df, 'composition')\n",
    "\n",
    "os_feat = OxidationStates()\n",
    "df = os_feat.featurize_dataframe(df, 'composition_oxid')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "1c98ca78",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "10a964f9f991457e88604d353642041a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "DensityFeatures:   0%|          | 0/1181 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>material_id</th>\n",
       "      <th>formula</th>\n",
       "      <th>space_group</th>\n",
       "      <th>structure</th>\n",
       "      <th>elastic_anisotropy</th>\n",
       "      <th>G_VRH</th>\n",
       "      <th>K_VRH</th>\n",
       "      <th>poisson_ratio</th>\n",
       "      <th>composition</th>\n",
       "      <th>MagpieData minimum Number</th>\n",
       "      <th>...</th>\n",
       "      <th>MagpieData avg_dev SpaceGroupNumber</th>\n",
       "      <th>MagpieData mode SpaceGroupNumber</th>\n",
       "      <th>composition_oxid</th>\n",
       "      <th>minimum oxidation state</th>\n",
       "      <th>maximum oxidation state</th>\n",
       "      <th>range oxidation state</th>\n",
       "      <th>std_dev oxidation state</th>\n",
       "      <th>density</th>\n",
       "      <th>vpa</th>\n",
       "      <th>packing fraction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mp-10003</td>\n",
       "      <td>Nb4CoSi</td>\n",
       "      <td>124</td>\n",
       "      <td>[[0.94814328 2.07280467 2.5112    ] Nb, [5.273...</td>\n",
       "      <td>0.030688</td>\n",
       "      <td>97.141604</td>\n",
       "      <td>194.268884</td>\n",
       "      <td>0.285701</td>\n",
       "      <td>(Nb, Co, Si)</td>\n",
       "      <td>14.0</td>\n",
       "      <td>...</td>\n",
       "      <td>9.611111</td>\n",
       "      <td>229.0</td>\n",
       "      <td>(Nb0+, Co0+, Si0+)</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.834556</td>\n",
       "      <td>16.201654</td>\n",
       "      <td>0.688834</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mp-10010</td>\n",
       "      <td>Al(CoSi)2</td>\n",
       "      <td>164</td>\n",
       "      <td>[[0. 0. 0.] Al, [1.96639263 1.13529553 0.75278...</td>\n",
       "      <td>0.266910</td>\n",
       "      <td>96.252006</td>\n",
       "      <td>175.449907</td>\n",
       "      <td>0.268105</td>\n",
       "      <td>(Al, Co, Si)</td>\n",
       "      <td>13.0</td>\n",
       "      <td>...</td>\n",
       "      <td>15.520000</td>\n",
       "      <td>194.0</td>\n",
       "      <td>(Al3+, Co2+, Co3+, Si4-)</td>\n",
       "      <td>-4</td>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>3.872983</td>\n",
       "      <td>5.384968</td>\n",
       "      <td>12.397466</td>\n",
       "      <td>0.644386</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>mp-10015</td>\n",
       "      <td>SiOs</td>\n",
       "      <td>221</td>\n",
       "      <td>[[1.480346 1.480346 1.480346] Si, [0. 0. 0.] Os]</td>\n",
       "      <td>0.756489</td>\n",
       "      <td>130.112955</td>\n",
       "      <td>295.077545</td>\n",
       "      <td>0.307780</td>\n",
       "      <td>(Si, Os)</td>\n",
       "      <td>14.0</td>\n",
       "      <td>...</td>\n",
       "      <td>16.500000</td>\n",
       "      <td>194.0</td>\n",
       "      <td>(Si4-, Os4+)</td>\n",
       "      <td>-4</td>\n",
       "      <td>4</td>\n",
       "      <td>8</td>\n",
       "      <td>5.656854</td>\n",
       "      <td>13.968635</td>\n",
       "      <td>12.976265</td>\n",
       "      <td>0.569426</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>mp-10021</td>\n",
       "      <td>Ga</td>\n",
       "      <td>63</td>\n",
       "      <td>[[0.         1.09045794 0.84078375] Ga, [0.   ...</td>\n",
       "      <td>2.376805</td>\n",
       "      <td>15.101901</td>\n",
       "      <td>49.130670</td>\n",
       "      <td>0.360593</td>\n",
       "      <td>(Ga)</td>\n",
       "      <td>31.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>64.0</td>\n",
       "      <td>(Ga0+)</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6.036267</td>\n",
       "      <td>19.180359</td>\n",
       "      <td>0.479802</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>mp-10025</td>\n",
       "      <td>SiRu2</td>\n",
       "      <td>62</td>\n",
       "      <td>[[1.0094265  4.24771709 2.9955487 ] Si, [3.028...</td>\n",
       "      <td>0.196930</td>\n",
       "      <td>101.947798</td>\n",
       "      <td>256.768081</td>\n",
       "      <td>0.324682</td>\n",
       "      <td>(Si, Ru)</td>\n",
       "      <td>14.0</td>\n",
       "      <td>...</td>\n",
       "      <td>14.666667</td>\n",
       "      <td>194.0</td>\n",
       "      <td>(Si4-, Ru2+)</td>\n",
       "      <td>-4</td>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>4.242641</td>\n",
       "      <td>9.539514</td>\n",
       "      <td>13.358418</td>\n",
       "      <td>0.598395</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 149 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  material_id    formula  space_group  \\\n",
       "0    mp-10003    Nb4CoSi          124   \n",
       "1    mp-10010  Al(CoSi)2          164   \n",
       "2    mp-10015       SiOs          221   \n",
       "3    mp-10021         Ga           63   \n",
       "4    mp-10025      SiRu2           62   \n",
       "\n",
       "                                           structure  elastic_anisotropy  \\\n",
       "0  [[0.94814328 2.07280467 2.5112    ] Nb, [5.273...            0.030688   \n",
       "1  [[0. 0. 0.] Al, [1.96639263 1.13529553 0.75278...            0.266910   \n",
       "2   [[1.480346 1.480346 1.480346] Si, [0. 0. 0.] Os]            0.756489   \n",
       "3  [[0.         1.09045794 0.84078375] Ga, [0.   ...            2.376805   \n",
       "4  [[1.0094265  4.24771709 2.9955487 ] Si, [3.028...            0.196930   \n",
       "\n",
       "        G_VRH       K_VRH  poisson_ratio   composition  \\\n",
       "0   97.141604  194.268884       0.285701  (Nb, Co, Si)   \n",
       "1   96.252006  175.449907       0.268105  (Al, Co, Si)   \n",
       "2  130.112955  295.077545       0.307780      (Si, Os)   \n",
       "3   15.101901   49.130670       0.360593          (Ga)   \n",
       "4  101.947798  256.768081       0.324682      (Si, Ru)   \n",
       "\n",
       "   MagpieData minimum Number  ...  MagpieData avg_dev SpaceGroupNumber  \\\n",
       "0                       14.0  ...                             9.611111   \n",
       "1                       13.0  ...                            15.520000   \n",
       "2                       14.0  ...                            16.500000   \n",
       "3                       31.0  ...                             0.000000   \n",
       "4                       14.0  ...                            14.666667   \n",
       "\n",
       "   MagpieData mode SpaceGroupNumber          composition_oxid  \\\n",
       "0                             229.0        (Nb0+, Co0+, Si0+)   \n",
       "1                             194.0  (Al3+, Co2+, Co3+, Si4-)   \n",
       "2                             194.0              (Si4-, Os4+)   \n",
       "3                              64.0                    (Ga0+)   \n",
       "4                             194.0              (Si4-, Ru2+)   \n",
       "\n",
       "   minimum oxidation state  maximum oxidation state  range oxidation state  \\\n",
       "0                        0                        0                      0   \n",
       "1                       -4                        3                      7   \n",
       "2                       -4                        4                      8   \n",
       "3                        0                        0                      0   \n",
       "4                       -4                        2                      6   \n",
       "\n",
       "   std_dev oxidation state    density        vpa  packing fraction  \n",
       "0                 0.000000   7.834556  16.201654          0.688834  \n",
       "1                 3.872983   5.384968  12.397466          0.644386  \n",
       "2                 5.656854  13.968635  12.976265          0.569426  \n",
       "3                 0.000000   6.036267  19.180359          0.479802  \n",
       "4                 4.242641   9.539514  13.358418          0.598395  \n",
       "\n",
       "[5 rows x 149 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from matminer.featurizers.structure import DensityFeatures\n",
    "\n",
    "df_feat = DensityFeatures()\n",
    "df = df_feat.featurize_dataframe(df, 'structure')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "6e0e5846",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['density', 'vpa', 'packing fraction']"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_feat.feature_labels()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "006c9106",
   "metadata": {},
   "outputs": [],
   "source": [
    "y = df['K_VRH'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "520e6310",
   "metadata": {},
   "outputs": [],
   "source": [
    "excluded = ['G_VRH','K_VRH','elastic_anisotropy','poisson_ratio',\n",
    "           'formula','material_id','structure','composition',\n",
    "            'composition_oxid']\n",
    "X = df.drop(excluded, axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "c86d982d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 140 possible descriptor: \n",
      " ['space_group' 'MagpieData minimum Number' 'MagpieData maximum Number'\n",
      " 'MagpieData range Number' 'MagpieData mean Number'\n",
      " 'MagpieData avg_dev Number' 'MagpieData mode Number'\n",
      " 'MagpieData minimum MendeleevNumber' 'MagpieData maximum MendeleevNumber'\n",
      " 'MagpieData range MendeleevNumber' 'MagpieData mean MendeleevNumber'\n",
      " 'MagpieData avg_dev MendeleevNumber' 'MagpieData mode MendeleevNumber'\n",
      " 'MagpieData minimum AtomicWeight' 'MagpieData maximum AtomicWeight'\n",
      " 'MagpieData range AtomicWeight' 'MagpieData mean AtomicWeight'\n",
      " 'MagpieData avg_dev AtomicWeight' 'MagpieData mode AtomicWeight'\n",
      " 'MagpieData minimum MeltingT' 'MagpieData maximum MeltingT'\n",
      " 'MagpieData range MeltingT' 'MagpieData mean MeltingT'\n",
      " 'MagpieData avg_dev MeltingT' 'MagpieData mode MeltingT'\n",
      " 'MagpieData minimum Column' 'MagpieData maximum Column'\n",
      " 'MagpieData range Column' 'MagpieData mean Column'\n",
      " 'MagpieData avg_dev Column' 'MagpieData mode Column'\n",
      " 'MagpieData minimum Row' 'MagpieData maximum Row' 'MagpieData range Row'\n",
      " 'MagpieData mean Row' 'MagpieData avg_dev Row' 'MagpieData mode Row'\n",
      " 'MagpieData minimum CovalentRadius' 'MagpieData maximum CovalentRadius'\n",
      " 'MagpieData range CovalentRadius' 'MagpieData mean CovalentRadius'\n",
      " 'MagpieData avg_dev CovalentRadius' 'MagpieData mode CovalentRadius'\n",
      " 'MagpieData minimum Electronegativity'\n",
      " 'MagpieData maximum Electronegativity'\n",
      " 'MagpieData range Electronegativity' 'MagpieData mean Electronegativity'\n",
      " 'MagpieData avg_dev Electronegativity'\n",
      " 'MagpieData mode Electronegativity' 'MagpieData minimum NsValence'\n",
      " 'MagpieData maximum NsValence' 'MagpieData range NsValence'\n",
      " 'MagpieData mean NsValence' 'MagpieData avg_dev NsValence'\n",
      " 'MagpieData mode NsValence' 'MagpieData minimum NpValence'\n",
      " 'MagpieData maximum NpValence' 'MagpieData range NpValence'\n",
      " 'MagpieData mean NpValence' 'MagpieData avg_dev NpValence'\n",
      " 'MagpieData mode NpValence' 'MagpieData minimum NdValence'\n",
      " 'MagpieData maximum NdValence' 'MagpieData range NdValence'\n",
      " 'MagpieData mean NdValence' 'MagpieData avg_dev NdValence'\n",
      " 'MagpieData mode NdValence' 'MagpieData minimum NfValence'\n",
      " 'MagpieData maximum NfValence' 'MagpieData range NfValence'\n",
      " 'MagpieData mean NfValence' 'MagpieData avg_dev NfValence'\n",
      " 'MagpieData mode NfValence' 'MagpieData minimum NValence'\n",
      " 'MagpieData maximum NValence' 'MagpieData range NValence'\n",
      " 'MagpieData mean NValence' 'MagpieData avg_dev NValence'\n",
      " 'MagpieData mode NValence' 'MagpieData minimum NsUnfilled'\n",
      " 'MagpieData maximum NsUnfilled' 'MagpieData range NsUnfilled'\n",
      " 'MagpieData mean NsUnfilled' 'MagpieData avg_dev NsUnfilled'\n",
      " 'MagpieData mode NsUnfilled' 'MagpieData minimum NpUnfilled'\n",
      " 'MagpieData maximum NpUnfilled' 'MagpieData range NpUnfilled'\n",
      " 'MagpieData mean NpUnfilled' 'MagpieData avg_dev NpUnfilled'\n",
      " 'MagpieData mode NpUnfilled' 'MagpieData minimum NdUnfilled'\n",
      " 'MagpieData maximum NdUnfilled' 'MagpieData range NdUnfilled'\n",
      " 'MagpieData mean NdUnfilled' 'MagpieData avg_dev NdUnfilled'\n",
      " 'MagpieData mode NdUnfilled' 'MagpieData minimum NfUnfilled'\n",
      " 'MagpieData maximum NfUnfilled' 'MagpieData range NfUnfilled'\n",
      " 'MagpieData mean NfUnfilled' 'MagpieData avg_dev NfUnfilled'\n",
      " 'MagpieData mode NfUnfilled' 'MagpieData minimum NUnfilled'\n",
      " 'MagpieData maximum NUnfilled' 'MagpieData range NUnfilled'\n",
      " 'MagpieData mean NUnfilled' 'MagpieData avg_dev NUnfilled'\n",
      " 'MagpieData mode NUnfilled' 'MagpieData minimum GSvolume_pa'\n",
      " 'MagpieData maximum GSvolume_pa' 'MagpieData range GSvolume_pa'\n",
      " 'MagpieData mean GSvolume_pa' 'MagpieData avg_dev GSvolume_pa'\n",
      " 'MagpieData mode GSvolume_pa' 'MagpieData minimum GSbandgap'\n",
      " 'MagpieData maximum GSbandgap' 'MagpieData range GSbandgap'\n",
      " 'MagpieData mean GSbandgap' 'MagpieData avg_dev GSbandgap'\n",
      " 'MagpieData mode GSbandgap' 'MagpieData minimum GSmagmom'\n",
      " 'MagpieData maximum GSmagmom' 'MagpieData range GSmagmom'\n",
      " 'MagpieData mean GSmagmom' 'MagpieData avg_dev GSmagmom'\n",
      " 'MagpieData mode GSmagmom' 'MagpieData minimum SpaceGroupNumber'\n",
      " 'MagpieData maximum SpaceGroupNumber' 'MagpieData range SpaceGroupNumber'\n",
      " 'MagpieData mean SpaceGroupNumber' 'MagpieData avg_dev SpaceGroupNumber'\n",
      " 'MagpieData mode SpaceGroupNumber' 'minimum oxidation state'\n",
      " 'maximum oxidation state' 'range oxidation state'\n",
      " 'std_dev oxidation state' 'density' 'vpa' 'packing fraction']\n"
     ]
    }
   ],
   "source": [
    "print('There are %s possible descriptor: \\n %s'%(X.shape[1], X.columns.values))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "d3ca8a15",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.metrics import mean_squared_error\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "8081ca59",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "training R2 = 0.926\n",
      "training RMSE = 19.773\n"
     ]
    }
   ],
   "source": [
    "lr = LinearRegression()\n",
    "lr.fit(X,y)\n",
    "\n",
    "print('training R2 = '+ str(round(lr.score(X, y), 3)))\n",
    "print('training RMSE = %.3f' % np.sqrt(mean_squared_error(y_true = y,\n",
    "                                                          y_pred = lr.predict(X))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "9168dae7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x213f9bb6eb0>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.figure(figsize = (5, 5))\n",
    "plt.plot([0,400],[0,400],'r--')\n",
    "plt.scatter(y, lr.predict(X), s = 80, c = 'b', edgecolor = 'k',\n",
    "           alpha = 0.7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "fccf23cb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cross-validation results:\n",
      "Folds: 10, mean RMSE: 22.388\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import KFold,cross_val_score\n",
    "\n",
    "crossvalidation = KFold(n_splits = 10, shuffle=True, random_state = 1)\n",
    "scores = cross_val_score(lr, X, y, scoring = 'neg_mean_squared_error',\n",
    "                        cv = crossvalidation, n_jobs = 1)\n",
    "rmse_scores = [np.sqrt(abs(s)) for s in scores]\n",
    "\n",
    "print('cross-validation results:')\n",
    "print('Folds: %i, mean RMSE: %.3f' %(len(scores), np.mean(rmse_scores)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "2aecbda4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "training R2 = 0.989\n",
      "training RMSE = 7.687\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "rf = RandomForestRegressor(n_estimators = 50, random_state = 1)\n",
    "rf.fit(X, y)\n",
    "\n",
    "print('training R2 = ' + str(round(rf.score(X, y),3)))\n",
    "print('training RMSE = %.3f' % np.sqrt(mean_squared_error(y_true = y,\n",
    "                                                         y_pred = rf.predict(X))) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "e5333df1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x213fb3281f0>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(5,5))\n",
    "plt.plot([0,400],[0,400],'r--')\n",
    "plt.scatter(y,rf.predict(X), s= 80, c = 'b', edgecolor = 'k',alpha = 0.7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "41898498",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cross-validation results:\n",
      "Folds: 10, mean R2: 0.924\n",
      "Folds: 10, mean RMSE: 19.277\n"
     ]
    }
   ],
   "source": [
    "crossvalidation = KFold(n_splits = 10, shuffle=True, random_state = 1)\n",
    "\n",
    "r2_scores = cross_val_score(rf,X, y, scoring = 'r2', cv = crossvalidation,\n",
    "                            n_jobs = 2)\n",
    "scores = cross_val_score(rf, X, y, scoring = 'neg_mean_squared_error',\n",
    "                        cv = crossvalidation, n_jobs = 1)\n",
    "rmse_scores = [np.sqrt(abs(s)) for s in scores]\n",
    "\n",
    "print('cross-validation results:')\n",
    "print('Folds: %i, mean R2: %.3f' % (len(r2_scores), np.mean(r2_scores)))\n",
    "print('Folds: %i, mean RMSE: %.3f' %(len(scores), np.mean(rmse_scores)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "d9d0597b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x213fb1c8460>"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_val_predict\n",
    "plt.figure(figsize=(5,5))\n",
    "plt.plot([0,400],[0,400],'r--')\n",
    "plt.scatter(y, cross_val_predict(rf, X, y, cv = crossvalidation), s = 80,\n",
    "           c = 'b', edgecolor = 'k', alpha = 0.7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "987e87e4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsUAAAH2CAYAAACYzfGGAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAACMlklEQVR4nO2dd7gdVdWH35UGIQQIEAw9gNKr0qVIBwEpUqRIF5H2SbGC0kSki6B0KQrSQZAqvQnSRKWj9BIRQ+9kfX+sPdx9J+cmN8m9d+bO+b3Ps5/cs2fmZO+zpqxZexVzd4QQQgghhGhnBlQ9ACGEEEIIIapGSrEQQgghhGh7pBQLIYQQQoi2R0qxEEIIIYRoe6QUCyGEEEKItkdKsRBCCCGEaHsGVT2Arph55pl99OjRVQ9DCCGEEGKiPPDAA/9195FVj0NMPrVVikePHs39999f9TCEEEIIISaKmT1X9RjElCH3CSGEEEII0fZIKRZCCCGEEG2PlGIhhBBCCNH2SCkWQgghhBBtj5RiIYQQQgjR9kgpFkIIIYQQbY+UYiGEEEII0fZIKRZCCCGEEG2PlGIhhBBCCNH2SCkWQgghhBBtj5RiIYQQQgjR9rS1UjxqjlGYWe3bqDlGVf1TCSGEEEI0mkFVD6BKxrw0Bg6uehQTZ8zBY6oeghBCCCFEo2lrS7EQQgghhBAgpVgIIYQQQggpxUIIIYQQQkgpFkIIIYQQbY+UYiGEEEII0fZIKRZCCCGEEG2PlGIhhBBCCNH2SCkWQgghhBBtj5RiIYQQQgjR9kgpFkIIIYQQbY+UYiGEEEII0fZIKRZCCCGEEG2PlGIhhBBCCNH2SCkWQgghhBBtj5RiIYQQQgjR9kgpFkIIIYQQbU+3lGIzW9jMbjKz98zsZTM71MwGTuSY0WbmLdoFPTN0IYQQQggheoZBE9vBzEYANwKPAhsB8wHHEgr1gd34P/YH7so+/3fShymEEEIIIUTvMVGlGNgNGAps6u5vAX82s+mAg83sqNQ3IZ5w93umdKBCCCGEEEL0Ft1xn1gPuL6k/F5AKMqr9sqohBBCCCGE6EO6oxQvCDyed7j788B7advEOMvMPjWzV8zsODMbOhnjFEIIIYQQotfojvvECOCNFv1j07au+BD4NXAD8BbwFeAHhE/yRpMySCGEEEIIIXqT7ijFAN6iz7rojwPcXwH2zLpuNbMxwG/MbEl3/9t4X2i2K7ArwFxzzdXNoQkhhBBCCDFldMd9YiwwQ4v+6WltQZ4Ql6R/v9hqo7uf5u5Lu/vSI0eOnMSvFkIIIYQQYvLojlL8OCXfYTObExhGyde4G3jpXyGEEEIIISqnO0rxtcA6ZjY869sSeB+4bRL/v83Svw9M4nFCCCGEEEL0Gt3xKT4F2Bu4zMyOBOYFDgaOy9O0mdnTwG3uvnP6fDAwnCjc8RawCvA94DJ3/3sPzkEIIYQQQogpYqJKsbuPNbM1gJOAqwg/4uMJxbj8XXnp58eJana7EDmNnweOBg6f0kELIYQQQgjRk3Qr+4S7PwqsPpF9Rpc+X0AU+RBCCCGEEKLWdMenWAghhBBCiEYjpVgIIYQQQrQ9UoqFEEIIIUTbI6VYCCGEEEK0PVKKhRBCCCFE2yOlWAghhBBCtD1SioUQQgghRNsjpVgIIYQQQrQ9UoqFEEIIIUTbI6VYCCGEEEK0PVKKhRBCCCFE2yOlWAghhBBCtD1SioUQQgghRNsjpVgIIYQQQrQ9UoqFEEIIIUTbI6VYCCGEEEK0PVKKhRBCCCFE2yOlWAghhBBCtD1SioUQQgghRNsjpVgIIYQQQrQ9UoqFEEIIIUTbI6VYCCGEEEK0PVKKhRBCCCFE2yOlWAghhBBCtD1SioUQQgghRNsjpVgIIYQQQrQ9UoqFEEIIIUTbI6VYCCGEEEK0PVKKhRBCCCFE2yOlWAghhBBCtD1SioUQQgghRNsjpVgIIYQQQrQ9UoqFEEIIIUTbI6VYCCGEEEK0Pd1Sis1sYTO7yczeM7OXzexQMxvY3f/EzAaY2QNm5ma2weQPVwghhBBCiJ5n0MR2MLMRwI3Ao8BGwHzAsYRCfWA3/59dgNknc4xCCCGEEEL0Kt2xFO8GDAU2dfc/u/spwCHAvmY23cQOTkr14cABUzRSIYQQQggheonuKMXrAde7+1tZ3wWEorxqN44/DLgLuGnShyeEEEIIIUTv0x2leEHg8bzD3Z8H3kvbusTMFgd2BPaf3AEKIYQQQgjR23RHKR4BvNGif2zaNiFOBH7t7k9P4riEEEIIIYToMyYaaJfwFn3WRX9sNPsGsACwYXcHY2a7ArsCzDXXXN09TAghhBBCiCmiO5biscAMLfqnp7UFGTMbDBwNHAkMMLMZgCIob5iZDW91nLuf5u5Lu/vSI0eO7MbQhBBCCCGEmHK6oxQ/Tsl32MzmBIZR8jXOGAbMARxHKNVjgYfTtguAhyZnsEIIIYQQQvQG3XGfuBb4npkNd/e3U9+WwPvAbV0c8w6wWqlvFPAH4MfAzZMxViGEEEIIIXqF7ijFpwB7A5eZ2ZHAvMDBwHF5mjYzexq4zd13dvdPgFvzLzGz0enPf7j7vVM+dCGEEEIIIXqGiSrF7j7WzNYATgKuIvyIjycU4/J3dbv0sxBCCCGEEHWhW9kn3P1RYPWJ7DN6ItufJTJWCCGEEEIIUSu6E2gnhBBCCCFEo5FSLIQQQggh2h4pxUIIIYQQou2RUiyEEEIIIdoeKcVCCCGEEKLtkVIshBBCCCHaHinFQgghhBCi7ZFSLIQQQggh2h4pxUIIIYQQou2RUiyEEEIIIdoeKcVCCCGEEKLtkVIshBBCCCHaHinFQgghhBCi7ZFSLIQQQggh2h4pxUIIIYQQou2RUiyEEEIIIdoeKcVCCCGEEKLtkVIshBBCCCHaHinFQgghhBCi7ZFSLIQQQggh2h4pxUIIIYQQou2RUiyEEEIIIdoeKcVCCCGEEKLtkVIshBBCCCHaHinFQgghhBCi7ZFSLIQQQggh2h4pxUIIIYQQou2RUiyEEEIIIdoeKcVCCCGEEKLtkVIshBBCCCHaHinFQgghhBCi7ZFSLIQQQggh2h4pxUIIIYQQou2RUiyEEEIIIdqebinFZrawmd1kZu+Z2ctmdqiZDZzIMYuY2XVp/w/N7HkzO8PMZu2ZoQshhBBCCNEzDJrYDmY2ArgReBTYCJgPOJZQqA+cwKHTA88A5wIvA/MABwFfMrNl3P2TKRu6EEIIIYQQPcNElWJgN2AosKm7vwX82cymAw42s6NS33i4+93A3VnXrWb2InADsDjw4JQNXQghhBBCiJ6hO+4T6wHXl5TfCwhFedVJ/P9eT/8OmcTjhBBCCCGE6DW6oxQvCDyed7j788B7adsEMbMBZjbEzBYAfgHcB/x1MsYqhBBCCCFEr9AdpXgE8EaL/rFp28S4BviQUKxnBDZw93HdHaAQQgghhBC9TXdTsnmLPuuiv8xewPLAN4FpgWvNbOpWO5rZrmZ2v5nd/9prr3VzaEIIIYQQQkwZ3VGKxwIztOifntYW5E64+1Pufq+7/x5YB1gK2LqLfU9z96XdfemRI0d2Y2hCCCGEEEJMOd1Rih+n5DtsZnMCwyj5Gk8Md38O+B8w76QcJ4QQQgghRG/SHaX4WmAdMxue9W0JvA/cNin/WQq2m4nIXyyEEEIIIUQt6E6e4lOAvYHLzOxIwsp7MHBcnqbNzJ4GbnP3ndPnY4BPgHsJN4uFgO8D/yJSugkhhBBCCFELJqoUu/tYM1sDOAm4ilBwjycU4/J35aWf7yeC7HYFpgaeBy4FjnD3d6d04EIIIYQQQvQU3bEU4+6PAqtPZJ/Rpc8XIIuwEEIIIYToB3Q3JZsQQgghhBCNRUqxEEIIIYRoe6QUCyGEEEKItkdKsRBCCCGEaHukFAshhBBCiLZHSrEQQgghhGh7pBQLIYQQQoi2R0qxEEIIIYRoe6QUCyGEEEKItkdKsRBCCCGEaHukFAshhBBCiLZHSrEQQgghhGh7pBQLIYQQQoi2R0qxEEIIIYRoe6QUCyGEEEKItkdKsRBCCCGEaHukFAshhBBCiLZHSrEQQgghhGh7pBQLIYQQQoi2R0qxEEIIIYRoe6QUCyGEEEKItkdKsRBCCCGEaHukFAshhBBCiLZHSrEQQgghhGh7pBQLIYQQQoi2R0qxEEIIIYRoe6QUCyGEEEKItkdKsRBCCCGEaHukFAshhBBCiLZHSrEQQgghhGh7pBQLIYQQQoi2R0qxEEIIIYRoe6QUCyGEEEKItkdKsRBCCCGEaHu6pRSb2cJmdpOZvWdmL5vZoWY2cCLHLGNmZ5nZ0+m4J8zsIDObumeGLoQQQgghRM8waGI7mNkI4EbgUWAjYD7gWEKhPnACh26Z9j0SeApYHDgs/fv1KRq1EEIIIYQQPchElWJgN2AosKm7vwX82cymAw42s6NSXyuOdPfXss+3mtkHwKlmNre7PzdlQxdCCCGEEKJn6I77xHrA9SXl9wJCUV61q4NKCnHBQ+nfWbo9QiGEEEIIIXqZ7ijFCwKP5x3u/jzwXto2KawIjAOemMTjhBBCCCGE6DW6oxSPAN5o0T82besWZjYKOAD43QRcLoQQQgghhOhzupuSzVv0WRf94+9oNgS4CHgH2GcC++1qZveb2f2vvdbK+0IIIYQQQoiepztK8Vhghhb909PagtwJMzPgXGAR4KvuPrarfd39NHdf2t2XHjlyZDeGJoQQQgghxJTTnewTj1PyHTazOYFhlHyNu+B4IpXbWu7enf2FEEIIIYToU7pjKb4WWMfMhmd9WwLvA7dN6EAz+xGwF7Ctu9852aMUQgghhBCiF+mOUnwK8CFwmZmtaWa7AgcDx+UBc6ly3ZnZ562BnxOuEy+Z2fJZk2+EEEIIIYSoDRN1n3D3sWa2BnAScBXhR3w8oRiXvysv/bx2+neH1HJ2BM6exLEKIYQQQgjRK3THpxh3fxRYfSL7jC593oHxlWEhhBBCCCFqR3dTsgkhhBBCCNFYpBQLIYQQQoi2R0qxEEIIIYRoe6QUCyGEEEKItkdKsRBCCCGEaHukFAshhBBCiLZHSrEQQgghhGh7pBQLIYQQQoi2R0qxEEIIIYRoe6QUCyGEEEKItkdKsRBCCCGEaHukFAshhBBCiLZHSrEQQgghhGh7pBQLIYQQQoi2R0qxEEIIIYRoe6QUCyGEEEKItkdKsRBCCCGEaHukFAshhBBCiLZHSrEQQgghhGh7pBQLIYQQQoi2R0qxEEIIIYRoe6QUCyGEEEKItkdKsRBCCCGEaHukFAshhBBCiLZHSrEQQgghhGh7pBQ3iFFzjMLMat1GzTGq6p9JCCGEEGI8BlU9ANFzjHlpDBxc9SgmzJiDx1Q9BCGEEEKI8ZClWAghhBBCtD1SioUQQgghRNsjpVgIIYQQQrQ9UoqFEEIIIUTbI6VYCCGEEEK0PVKKhRBCCCFE2yOlWAghhBBCtD3dUorNbGEzu8nM3jOzl83sUDMbOJFjhpjZ0WZ2h5m9b2beM0MWQgghhBCiZ5moUmxmI4AbAQc2Ag4F9gMOmcih0wC7AO8Bd0/ZMIUQQgghhOg9ulPRbjdgKLCpu78F/NnMpgMONrOjUt94uPsbZjaju7uZ7Qms3nPDFkIIIYQQoufojvvEesD1JeX3AkJRXnVCB7q7XCaEEEIIIUTt6Y5SvCDweN7h7s8TbhEL9saghBBCCCGE6Eu6oxSPAN5o0T82bRNCCCGEEKJf092UbK3cIKyL/snGzHY1s/vN7P7XXnutJ79aCCGEEEKILumOUjwWmKFF//S0tiBPNu5+mrsv7e5Ljxw5sie/WgghhBBCiC7pjlL8OCXfYTObExhGyddYCCGEEEKI/kh3lOJrgXXMbHjWtyXwPnBbr4xKCCGEEEKIPqQ7SvEpwIfAZWa2ppntChwMHJenaTOzp83szPxAM1vPzDYDlkyfN0tt7p6agBBCCCGEEFPKRIt3uPtYM1sDOAm4ivAjPp5QjMvfVS79fDKQK8AXp393BM6e5NEKIYQQQgjRC3Snoh3u/igTqUjn7qO70yeEEEIIIUTd6G5KNiGEEEIIIRqLlGIhhBBCCNH2SCkWQgghhBBtj5RiIYQQQgjR9kgpFkIIIYQQbY+UYiGEEEII0fZIKRZCCCGEEG2PlGIhhBBCCNH2SCkWQgghhBBtj5RiIYQQQgjR9kgpFkIIIYQQbY+UYiGEEEII0fZIKRZCCCGEEG2PlGJRW0bNMQozq3UbNceoqn8mIYQQQvQAg6oegBBdMealMXBw1aOYMGMOHlP1EIQQQgjRA8hSLIQQQggh2h4pxUIIIYQQou2RUiyEEEIIIdoeKcVCCCGEEKLtkVIshBBCCCHaHinFQgghhBCi7ZFSLIQQQggh2h4pxUIIIYQQou2RUiyEEEIIIdoeKcVCCCGEEKLtkVIshBBCCCHaHinFQvQRo+YYhZnVuo2aY1TVP5MQQghRCYOqHoAQ7cKYl8bAwVWPYsKMOXhM1UMQQgghKkGWYiHEJNMfrN6TYvlu2nyEEEJMOrIUCyEmmf5g9YbuW76bNp9Rc4yKOdWYz83+OV598dWqhyGEEJ8hpVgIIRpGf1Dy5aojhKgbcp8QQgghhBBtj5RiIYQQQgjR9kgpFkIIUWv6QyCkgiCF6P90y6fYzBYGTgRWAN4AzgAOcfdPJ3Lc9MAvgY0JBfxPwN7u/vpkj1gIIURbIR9pIURfMFFLsZmNAG4EHNgIOBTYDzikG99/IfAVYBdgB2AZ4IrJGqkQQgjRz+kPVm+lMxTtSncsxbsBQ4FN3f0t4M9mNh1wsJkdlfrGw8xWANYBVnX321PfS8C9Zramu9/YM1MQQggh+gf9weoN7ZvOULQ33fEpXg+4vqT8XkAoyqtO5LgxhUIM4O5/BZ5J24QQQgghhKgF3VGKFwQezzvc/XngvbSt28clHpvIcUIIIYQQQvQp3VGKRxDBdWXGpm09fZwQQgghhBB9irn7hHcw+xjY391PKPW/BJzt7gd0cdyfgXfcfZNS/3nAaHf/cotjdgV2TR8XAJ7o7kRqxMzAf6seRA+i+dSXJs0FNJ+606T5NGkuoPnUhbndfWTVgxCTT3cC7cYCM7Ton57WluD8uFYnxwxdHefupwGndWNMtcXM7nf3paseR0+h+dSXJs0FNJ+606T5NGkuoPkI0VN0x33icUo+wGY2JzCM1j7DXR6X6MrXWAghhBBCiErojlJ8LbCOmQ3P+rYE3gdum8hxo8xspaLDzJYG5k3bhBBCCCGEqAXdUYpPAT4ELjOzNZPf78HAcXmaNjN72szOLD67+1+A64FzzWxTM9sYOA+4s+E5ivu1+0cLNJ/60qS5gOZTd5o0nybNBTQfIXqEiQbawWdlnk+ic5nng/Myz2b2LHCru++Q9c0AHA9sQucyz/3RgV4IIYQQQjSUbinFQgghhBBCNJnuuE8IIYToZcxsqqrHIFoj2QjRHkgprhAzm97MZql6HKI1kk99aZpskqvZ98zs+1WPpSdoknwkm3rTtPmIapFSXBHpRrsHcKOZLVvxcEQJyae+NE02aT47EFl9fmFme1Y6oCmkSfKRbOpN0+YjqkdKcQWY2fTATkQA4qLASWY2xMys2pFNOmY2yMx+ZGbDqh5LT9Ek+ZQxs359zTdNNmY2HbAj8G1gbiLTzwOVDmoKaJJ8JJt607T5iHrQnYp2ogdJN9qdgN2BIcDLwOHu/lGlA5t8FgUOB54ELq14LFNM0+RjZkOBqYHpgJfd/WMzG+zuH1c8tEmmgbIp5vND4GEi7/ubwH1mNsDdx1U5vkmlSfKRbOpN0+Yj6oOU4j4ku5B/DNwC3A8sRj+1PqT5HAjcBFxZ8XCmmAbKp3hhWRxw4Fkz27jIL25mI4ER7v5khcPsFg2UTTGfA4DLgZ8BVwPHu/snaZ+BedrLOtMk+Ug29aZp8xH1ol8vpfYnShfylcB2xNLck+7+YtqnXyz7ZOMcDSwDXNcfLY85TZIPgJmtC9wLjAL+APwSGApcbWYDk0J8DXCdmU1b2UC7QQNlk1sh/+zuuwJHAa+5+2+L/dz90+SetGFFQ+0WTZKPZFNvmjYfUUPcXa2XG7F0/V3gVeCi1Pcz4L7SfsOAWYETgKWrHnc35nUZ8Gipb0Cax9RVj69d5QMsArwDXAAsmPUvQzxIFgRuBJ4D1iblK69ja6Bsivm8AlyW+hYD/gqsl+23MPAd4C5gHLE0XPn4mywfyaa+smnifNTq2SofQNNbupD/r3SjnQb4C1Hdr9hvuXQR35putK8Dc1c9/gnMax3gaWC7rG8U4U7xMrHcOLDqcbabfAj/4cvSg3yh0rZRwGPAC+nBshr1V4ibJJvpgf3S739J1v9b4E7AgM8D+wMvAWcB/wXGAhcBo6qeQ1PlI9nUVzZNnI9afZvcJ3qRlJFhb+B7wJ3uvmna9ENgVnf/lZmtbGbHEYEcn5ACooDrgOEVDLu7rA98DFxoZguY2R7A34BtgauA7wP7pkCvWtJQ+UwFLAHc5u6PFZ1pSXEp4sH+JrAVcIe7uyUqGW0XNFQ28xA+3g+4+2YAZrYm8DXiQf4j4BxgN8IS+RIxrwuAn7v7q1UMuhUNlI9kU1PZNG0+ouZUrZU3uRHWhSuBm7K+RQlr3UnAGcBTwPXAskTqn/8CpwNLVD3+CcxreeAj4qa0O5F54rU0p/nTPvukuRwADKh6zO0iH2CONObvleY5J2FV+RewWuofUmxP/9bGst9E2aQ5fDX7eyhhefyYsFA+Q/hLDgfWBd4iLJJLZMfUwrLfRPlINvWUTdPmo1bvVvkAmtoyRWNA1jcIWIuw1L0K3AOslLZtQPiBnlPHG21pbgcQS1MvpbkcDHwl2/5T4FPgP8Sb+oHA8KrH3Q7ySQ+Qa4CHgBUIJXn9NJdngK8A8yUZ3p4eJMcC0xa/QR3m0DTZUHoxLH5n4BvA2cTS/YDUtxXwNvC7Oj7UmyYfyabWsmnUfNTq3yofQJNb6UIemP6dgViC+1q2bbvsRrt41j/eDaEOLd2UzgN+A8xe2nYU8DxwH5FUfV/gBuBRYGjVY2+yfLIH92AiTd5VwOPEcuLThEK8Ydr2fpLfeUTA0OMkxbgOrWmymchcp8/+/hZhhfwdsGjWPxcwV3l+ko9k03TZNG0+avVulQ+gnVqmtAzM+oob7bmlG+2SxFvv1Olz5Ra88jhIy+/Z58MIhfhsYLas/0tEwMOWVY+/XeRDZAGZJz0k/kaHhfgGwsr//eyY+Ylcn8dV/UBvsmxazMny35tYni/mkz/U5yEs+zcCm+XHVz2HpspHsqm1bBo1H7V6tcoH0G6tdKPdMbuQl8z6hwBbA9cCN5NcD6iJzyetl7R+RvgZP0BSiOnwWZ2dWOraqeqxt4l8iofGAMJavG76fCDwAeEu8QmwbXbMJWnf2jzMmyibCcxtZ+LFsdN8su2fT9fYm2QZX+rUmiofyaZ+smnafNTq0yofQLs2IkDtf4Tv05JZ/+yZMrkEYYH4BzBN1WMujT9/Sz8KeIQIttsy9Q3Jtm+XlLB+kzOyAfIp++ANItwmDk19BxIW4y3S52uBEyYk57q0/i6b0lwMWCjJ4nw6+0EeRywFHw7MmfoOIIIlF6h67E2Xj2RTX9k0dT5q1bfKB9COLd1sbyAsc0tm/ZcQwVD3AD9LfbMROWe/U/W4u5jL4cATwMlESdQflLavTbhUnJHfkIgMFlNVPf6my6c0r0uBc7PPPySi688gXmi+nW3bjpr5gDdcNmsDS2Wf70oyuTP9/QqwAJGb9QVSBpG6tSbKR7KpfuztMB+1erRBiD7FzMzdPZXhHe3u/079NxKVks4ABgLbmdkC7r65mX1IVCGrFWZ2ELArcRM6GPgmsLGZ3QaMAVYllhUfA8529/fScVsCewFvmNlW7v52BcNvSZPk04K/AxuY2ULu/pi7/8LMPgWOBH7v7qcCmNnewE+ABczswPR7DHT3TysceyNlY2YD3H2cu9+Q9e1D+OGvSzzYRxHX133ADwg/8U/6frQTpmnykWxqLZtGzUfUiKq18nZsjJ8CaAXC13OX9HkgkW/xBaIyz83AnmlbbXw+icCt4+nwIZ6a8CkeS6TF+QQ4k8zKkvYbTDxYiiWtYVXPpYnyaTGvQUSqtruAlekIPlkj26fIL/0KYfn/Fh3p2ip3pWiqbEpz2gt4kZTzO/UNAf5ILOXfQ5btoHRsbTIfNFE+kk19WtPmo1aPpop2FeDu40pdHwLvElYG3P1Td/8rcBChvMwO3JG2eX6gmQ3s9QF3gbv/C9jP3V82s6nc/QPiJrQLUfVpWXffGXjCzLY1sxvN7CLgu+7+AJEH9C3CBaM2NEU+OcnS+wmwDDGPQ4AjzWxad78p7bM/IbcXiFR69xJp9e41s2m8YksxNFM2LXiBSDm1lJnNlvo2TH3/AfZ19+eL8ZvZbGY2E8Qcq6xO2AbykWyoh2yaNh9RE6rWytXijZe4WO8gCi2MJCx0dxE32pXTfkOBmYnSo6tSszQztLAkEmWH9ydKbv6VyIv7AnBx2n4kqZZ9XVuD5FOkaxsIbAZskG3bl8hVfCGwcNa/AOEKs3vV42+ybFrM6wDgn8DdhIVrXJrPGmn7TESRnPuIQKNHyLKJ1KU1UT6STT1l07T5qFV0HlU9gHZvdCQjHwxcQFSAeyLdaMcAq6ftsxDBbM+lbf8lfERrVSmuxfyGpwfIVenzAGAlQjE+D7gMuKXqcbaLfGiRwJ6o2PUocBEpcp6OtG5fJNxhvt1XY2xX2ZTlAyxGBEP+h6geWVTt+hwdBXEuJ8rankwsHa+ZHV/1cn2j5CPZ1Fo2jZqPWnWt8gGodbLgGbAKYbF7GVgh9c8C3AI8DPySKN27JpFi67fl76lTA0YTden3L/XPk25KLwOrFvOverxtKJ+tCGvWH8j8JNO2aQnXlnHAKi2OrbxCVBNlk18HROaDj4vfH5ieSA/2d7JIesKf/5i0/3xVz6Gp8pFsai2bRs1HraLzqOoBqCVBZAUxgD8Ba6XPQ4Ffpwt5V2BwdszGRL7MnalZsFo+rzSf+4ngupmBuYmqd+8D/0c/eEtvsHwGAocCi5T6pyEsYZ8QfpIQgXojCN+84gFU+QOkqbLJxjpH9vcqRKDXgWUZACcAbxC+lUtWPe52kI9kUy/ZNG0+ahWcQ1UPQC0TRuul7RmBZ4FfFRcyKb8vsBZRRW4cJUtsHRodS/CDiZyeVxJ+xS8S2Sn2A2asepxtLJ+W2SQIhfjHwKfAgZkM9yf8KP9J5AatTYWopskmnxOdl+2PAMZmn6fK/r6UWCo+kVIBCapfrm+UfCSb+sumKfNR69um7BM1wlM0rZnlclkBmAs43t0/TlkePkzbvk4oLscA1+TfVYdoWncfZ2aD3P1jwoqyBWFp/ITIg3uWu/+vyjFOCg2Uz3jZJMxsGLA3YT0+iI7MIPMS2UL+SmSuGEZkpZjW3T8t/SZ9TtNkA59dP+ado+w/Al4ysxFpnw8BzOwQIlPIH4BfufsTZjbQzL6Y9vM+Hn4nmiYfyabeskljacR8RB9TtVau1rlRshoASxM+n+uV+g8m3mxPAD6f+gYCPyJVIit/V4VzKqwqRkQG/xSYuepxST4t5/fNNO6flPoXI3IXr5s+jyb88y4hc6EglOWWeVolmyma24aE69FGhB/rdsSLykzEw/yjNJ/50v4DgFMJf/7Fu/qNJB/Jpomyadp81Prw3Kl6AGpdCAa2ISyrRiSEvxfYHNgg3VDHEcEC+Y32GCKa9vt0KKKV+3wW40v/Gv3Ah7jd5JPNaxCwY/p7KjpcJGYn/MJPSp8HEoU+fkZn/7z1khIwR1+Ou8myISp03QZ8PX3+QZrTLYQrUquH+lFE1pB/E377y2bfV/VyfWPkI9nUVzZNnI9aH5wzVQ9ArYVQwqfzXCJFlqUL9TfpYv2ECFDLLQ8DgWPpyJd5DXAYNfL5TONoxBt3g+WTW3yNUHqPB6ZJfV9Lc/hq+jw1MCL7TQpfvbWqknWDZXMakUaqiKQflub5Uhp/Pp9fpYf6E8DpRFDru8B2NZhH4+Qj2dRTNk2bj1ofnTdVD0CtC8HAQsDrwO+JpetRwNGp72g6Wx5+BTwP3A6sC+wJ/JkIiBpa9Vya2NpBPsB3Cevwl+lQeK8k8oBOm+03TXqYPEAE5A3MtvV5NHeTZEPnFGCXEy4s9xDluj8kLPXzp+0DiJeYfxMK10ypf2rgF8AzwOw1mFMj5CPZ1Fc2TZ2PWh+cM1UPQG0CwoEl0030P8B7wJuEr1N5qecF4JTiRpu2LUf4UG3S4nu1FCT5dGd+A9MD4g5gUcKd4gHgimyfYYSP+IfAEaXjfwF8jywKX7KZrLnkGQ72S+O9lKjGVShXA9N8xhEFcWbMx0sUkfgvMKr03TMCP5J8JJumyaap81Hr5fOl6gGoTURAkWB8Q2A3IripKEk5kLA8jCPSnRU34CLNzNxE2rNtS983DaHkLNtXc2hya6p86Fwh6h4iv+f/0sNlz7RtOKEQjwO+n/qmJSp6nZX6dwaGSDZTPJcuC6Wk+ZyUZPM/kj83MISOvK2HEynBZigduzRRfW1ByUeyaZpsmjoftd5rgxC1xt1fJAI2PiOliTkBWI1YBjrK3V8vpZlZn7jxPlT6vvfM7FbiohZTSFPl45FmbZBH+qIvE8EpsxLVuu42MyMsYz8lrC7HpkM/BVYmfPh+4O5n9v3ogybJxjun/sLMBnpHKrxjifncTVi2ZgJedPeP0r6bEkvBx7r7G6lvZmAxd7/FzFYE3u6zySSaIh/Jpr6yyf7/Rs1H9CJVa+Vqk9aIN9vjiECNEwhftn1K+2xMBHmcSEdamZGE9U5pZiSfSZpPi77BRK7icUS0fe5DfAqRfeJx4Gk6LDKVz6eBsjHgnGw+CwB/JNLqFQGQ3wT+lfoXSX0zEQ/5/wAjq55HE+Uj2dRXNk2cj1oPnhtVD0BtEgUWfpqvEUs+I4ADiGWflYEvEFaHl4kSl8umY2YkfEM/BJapeg5Nbk2XD2EZKXKvfp/OCvGpqX+vpATcSyz/Fq4YVUejN042RHqvc4Dp0+cDiKj6h4C/EVH25wArZ/P5C/HCslbVMmmyfCSbWsumUfNR68Fzo+oBqE2iwGC+dEHn0cv3AW8QQQQfElG0y6TtI9KN9llimUjBAZLPlMyvKMByGK0txHsD06W+UcAixBLkNDUYe2NkQ+fMB0NL27YlrF+/I/Kxfi6bzz1EhP3qdOHrTUXWr6bIR7Kpr2yaOh+1nmuFo7/oR6Tyol74PiXfqK8SVryHgWdS/wjgWmA2YHvgLnf/qDg++76B3qLkr5g8miqfbF7l8Z0I7EpkmjjH3d9M/QOI7BR3EvP8kVd8w2mSbPKxJB/vctnhfN8ZgauJIizbu/stqX9WYEFCCRjj7v8qf3df0hT5SDb1lU32/zdqPqJnkFLcz+nqQjSzmYiln/KNdlHCV2oaYpnuAo+ggUHu/knfjbw9aJp8WjwIVgFuAg4Fjnf3d1L/AHcfZ2bLExW/tnb3SysZdBc0TTZlMhnMTKQCmxvYwSOAa0ai0teRRGliJ7IjHO7u51Y26Iwmy0eyqZdsmjYfMQVMyIys1j8bEQxwB2mpJ+v7HpEd4FHiQn6ISExeVOxR0IDkM6lzWZWI6v5qi23TEhawq7K+PK9rHefTGNmkcU2VxvkKadk39f2QyGpwDOH/vRJwIPABsEHV424H+Ug29ZVNE+ej1k25Vz0AtV4Qavh3fpButENS3/eIpOV/Sp+HEgUZbgeuQj5Sks/kzWVgenDcB8yW+gakfxcAHgP2KPVPTUfhgi5zvEo2PTanHYG16ahKuDbh//0h8Mdsv+mIwhOnEAUN8heYEVXPo4nykWxqLZtGzUetm3KvegBqvSDUUFS+mCkeixFvtrcRy3FbZ/vulRSXWUrfMTc1KDvaxNYU+dC5wMffgFVK288mUh4V+40gqktdmx4gU1cti6bKJo1jPIsVYYm8hijVvTBRxeuabPvNwO9bzOdk4Kc1mFMj5CPZ1Fc2TZ2PWvfaAESjKHyj3P1B7/BtWp9Ij7UmcBDwezP7Rto2NZEa6OPsO+Yi0m3dY1G4QfQQTZKPR4GCwe7+MbCUu9+egoowsyWICO+j0n47Ab8l5jkEeJJ4wNSGJskGwNNTucQnwAxEAYlHiQwIi5rZ1WY2CpgZeC6T41zAHsA2wMFmdkafDL4FTZKPZFNf2aSxNGo+ovuool3D8NbRr4MIK90gd/+5mY0DzksBHesRFokiY8BcRFqtLYk8jjeb2ZrufkefTKDhNE0+SSFuxWLA/MAaZrYJsC5wCbCbu59V7FRVJH0rmiabMkmZcuAfwHRmNrW7P2xmGxLFC14m5nOUu7uZjSbyte5KVGa7ErjRzD509z36evxNlo9kUy/ZNG0+YhKo2lSt1vsN2Ax4EPhS1vd9oiLZf+nwBZ2HCO54Ezgk9R0NvIWSlUs+XY9/KLAMHX53SxBBKOMIf+PLgaWpQa7idpNNF3PaNo1z+0ImxDLxJXQEFM1bnk/q/yrwDKXqX5KPZNMGsmnUfNS6kHPVA1DrAyGHb9Q9RPLxJenwkdoWWCP9nV/Ih2bHzkT4rN2BfKMkn9bjn5GIoP8rEQz0DpHn82hgNDBt2q8ItOs30dn9XTYTmNePCQvWjzNlq4ienzPN561c6UrbhhElvm+h5D8p+Ug2TZZN0+aj1oWcqx6AWi8LuHMw1B3ArURpyxkK5QSYo9WNNtu+BuEDum7V82laa4p8gKWA84E/ArsAC5a29xtFuGmyKc0pz1qwM7Hs/hs6Xljm6mI+efXCFQl/8OUkH8mmTWTTqPmoTUDWVQ9ArQ+E3PFGO5DIgfkzOlIAdXWjHQCdypX+Gzix6rk0sTVFPoTP3ZBSX61SrrWrbLqSCVGUoHjgz53N59BW+6fPixNpw9aowVwaJR/JptayadR81Fo3Bdq1Ae7+SVax5xdZect5gN2BbwG/dPeDoKPaUvp7IPE2DPB+BcNvPE2Rj7eo5ORdlLbtLzRFNjkeldTMg5cAzOwLxHy2A45z94NTfz6folrX4sQLUOVzapp8JJtay6ZR8xGtkVLcJngWTZsu5NmIt93NiPK8B8P4F7JHOq35CUvFU30/8vZA8qkvTZSNe5itMmYhouWvncB8PjGz2QmL2FXufndfjrkrmiYfyabWsmnUfMT4KE9x+/IK4QM1jAiIytMC5RfyrMB5RBDVZdl+oneRfOpL42Tj7ncBPwVWN7Ovm9lULR7qswJXAO8Cp5a/o0Zza5R8JJv6yobmzaftkVLchhTLPsDGwL+A681sAcI/zdP2T5Pl4UriAj/B3V+HDktGWhISPYzkU1+aKBszGwDg7ocDvyAqEe6U5gUwtZktSZSynZF4+F+bjl3SzBZLx3vVD/qmyUeyqbVsGjUfERRRkaLNsOSDZmaDiSo97wBnESm1RgCLAKcTtd+PAU5JF/pqRKaBk939fevwZRM9iORTX5oom9Jy73eBnQADXiR8IZcEHicKRZyX5vN54JfArMCx7n5+Or7SgixNk49kU2vZNGo+QkpxW5Nd0IOAM4mb6wLEBTwdkXbml4SPmpvZVMBPiLQ/w4A13f1t6wg+ED2I5FNfmiibkvK1EvAl4GtEztX7gLPc/dW0fZi7v5uWhbclyt7u4u4XVDP6zjRNPpJNrWXTqPm0O1KK25zM58mAhYDlCSvE/9z98rTP1MQb77Pu/npaDjqXqGS2hrsrmraXkHzqSxNlU7Yklh/UFmWHtwa+APwdONzd/2VmBwEbAV9z9xdbfVdf0zT5SDa1lk2j5tPWeA3ywqlV25hALlngm8Sb7jjgn8BJqf8LRCLyTbrzPWqSTxNbO8iGDuPJmkSlwseIYK6bgTHAYsDngeeBOdK+Q7Lj55Z8JJumy6Zp82nXppRsAu8il6yZLQTsRVzImxJRtvub2ezuvomZfQjMlvbNl/fWBB5w97F9MoGGI/nUl3aQjacnNbAOUaRgHnd/I1m6DiOCvH5FpA4bnPYt5vMrYKiZHewp725f0nT5SDb1kU3T5tO2VK2Vq9W3AZsTF/LS6fNUwDJEVZ6H07ZV0rZZ0r/Hp21frnr8TW+ST31b02RDVPE6nVCypiptu52oonZw+jwU+AthrXyH8GkdUfUcmiofyaa+smnifJrelJJNTAgj8jAOAXD3D939PuAMwm/qaCLvIsCPzew9YHvgQuKCF72L5FNfGiUbD9/VW4mH+TZm9nkzm8siG8JKRMT9BWnf94FPgVWAG939EHcfaym9WE1ojHwkm/rKJtG0+TSbqrVytfo2wgJxH3AX4RO1KXAE8WZ7FrBYtu9ZwIfA/4AvpL7BVc+hyU3yqW9rqmyAHwAPAn8kUoKNI3LnLpztcxIwFriGsFLuXPW420E+kk09ZdO0+TS9VT4AtXo2YGD6dxDwW+DgdLF+nD63utEeS1gsXgOmq3oOTW6ST31bE2VDFvxDpJE6ID3UTwIWzbb9Os3nIGB2wpfyDWD+qufQVPlINrWWTaPm0w6t8gGo1bcBg7K/v5xutCfS+c02v9F+jkhYfhWwebaPVT2XJjbJp76tibLJxwL8H3AJsHiL+RwCfK74HYBF0t8D6zKvpslHsqm1bBo1n6a3ygeg1j9auoH+tHSjLd5sDwZGZf3D0r/T0PGmPLCvxtqOTfKpb2uqbIAZsr9/QVi2Os0n2z4XsC8poCj11eIh30T5SDb1lE3T5tPEVvkA1OrfyN50s77vA68DBxYXMp2tFbMSqYBOzC5uXdCST1u1JsqmrDARS/EPA+cDI7P+fFl/bSKw6Flgu6rn0FT5SDa1lk2j5tPUVqeIU1FTvFSTPZWz/ALwFHCup/Kinq7YxFgionZ+4C9mNq1HxR+dcz2M5FNfmiib0lgBnEgz9aS7v5ZvMLOFzGw0cJu770JYxX5pZl/rk8FOhKbJR7KptWwaNZ+moh9WTDLp4v4ImBp4t+g3s5FmtqGZ7QGs6O6/J6Jt/wNcbGaDvYsE56LnkHzqSxNl4+4vA6cAPzWzjQDMbBHgZOAR4DbgFjNb0N2PAa4FtqhqvBOiafKRbGotm0bNpzFUbapW65+N8I16FriBSAg/L7FENw54iUgJdHbadz3iBjxz1eNulyb51Lc1STZ0Xur9IbB9+vtIIsp+R2BL4A/AW8DMwB7Ekn4tU001RT6STX1l09T5NKFVPgC1/tdIvlFE2dBN09+7pwt5bWA6YP10QZ8DLEEkjF+k6rG3Q5N86tuaKBs6+6caETn/MvDrrH8WYnn+feBV4KSqx90O8pFsai2bRs2nKc2SIISYJMxskGc+Umb2OyLn4pfS56mI1DIPAMOBa919k0oG24ZIPvWl6bIxs5HEMvzdwPfc/cPUvwixXP8P4Mfu/pfqRtk1TZaPZFMvmjafJiCfYjFZeCloAHgBmNnMFkzbPyQq+bxJBAqc1rcjbG8kn/rSdNl4BHSdRyzP725m3zSzrYBbiCXhU+qqdEGz5SPZ1IumzacJyFIseoQUSftX4D3gUeADon77y8DB7n5hhcNreySf+tIk2ZiZeXqomNmewIzArsSy/XPAT939kvK++WczG+Du48rbq6Ip8pFs6k/T5tMfkVIspphiCShd0GsDmwObEalmDnf3S9N+5RvtAM+iaMtLSaJnkHzqSxNlk4/NzKYFHieCiX7l7hel/pZKV/p7Xnf/d9+PfHyaJh/JptayadR8+itynxBTTLqQB7r7J+5+DZEb82XgqAlcyJbdnDc1s2Hpe6ySSTQYyae+NFE2+QPa3d8Bvgec3k2lawfgaTNbKN/eV2Mv0zT5SDa1lk2j5tNv8RpE+6k1qxE+UGtkn8tVlvJUQVsS0bZ7kyKlUcUeyadNW5NkUx576hvQ1T7Atmk+/yVSgy0zoe+SfCSbJsqmqfPpL02WYtGjpDfdT939pvR5gKcrNH3OLQ/bEzkyHVgaOMg6KvYMrGL8TUfyqS9Nk00+9qwvX+bN57MzcC7wJHAxsBRwZZpny+/qa5okH8mmvrKB5s2nPyGfYtFntLjRnk5EPZ8LDAE2JZKXL+nu71c20DZF8qkvTZNNF/M5g/Bt/aeZTQccAawOrO7ur1Q32onTJPlINvWmafOpHVWbqtXao9E5ifzOxFLPqcD8Wf8qRCnL9bO+WizNNb1JPvVtTZYNkRqsmM88pW0HE8UkZiz112peTZWPZFP9+Js+nzq2QRNSmIXoKbwjGGAn4s32NOBId38m2204UWp0XLHs47EEpGjaXkbyqS9NlY2Z7QGcCJwMHJPPx8yWJR76APuZ2UfAQ+5+paenfF1oonwkm3rKpmnzqSPyKRZ9hpltSCzDnULpQjazFYFfpI/fBC4DLjez4R7RtDpXexnJp740VDaDiEISx5bmsxxwLFHS9jfA08Sz6iIz27iCcU6UBspHsqmpbJo2n7ohn2LRZ1jkX9wHuMLdn8r6lwOOAYYBZwNnAl8EDiVuvuu5+8d9PuA2Q/KpL02VjZlN5+5vZZ+L+UwF/NLdz0/9BlwPvAFsDXxaJ6tkE+Uj2dRTNk2bT+2o2n9DrT0aMKiL/uWAO4gqPluVth1AVFoaVfX4m956Uj7A7MC8Vc+pKa2J1w6tU4KtANzaxXxmA54Afl312HtTPnW4diSb+l47TZtPHZtM6aJP8Ba+TGa2MnAkETF7vLv/Ids2M5Fe5llAEbS9zBTK54PSoSsAT5nZqGRFElNAE68dT0/rAjP7HLEcPAOxZJ/PZxbgO8BQ4N4+HGa3aNq1MwWyuafV91V5D2jatdO0+dQRKcWiSrYC5gaOK13II4Hdibffa939zfKBUrb6hO7I5xp3fyP1z2Jmg939EmBVd3+1/IAVPUZ3ZHN1fu0U14zVMHepu48Bfk481C8s+tN8dgV2A65y93O7+o6a3RMac+10UzZXuPvvUv+0Zra0mS2Zjvd+KJv+9Nxp2nyqpWpTtVr7NqJiz5qlvpHAgcDrwKmlbSMAo6NiT8ulJLU+k8/JWf9chB/bQ5JLLWTzm6x/OmBt4AfANKmvNjKii3RRRAR9MZ8zs/61gP2AK4jUYGtO7LtqKJ9+ce10UzZnlLZ9Ic3nH8DWVc9hMmTTr547kzGfWYDpi3nUbT5Vt8oHoNaejRYlKNON9oAWD8FticTkrxNLdL8Hpu3qe9T6RD6nZ/2jgROAT4D9S8cM6M1xtmOblGsnbRsE7AXcBtwHDK+7bNKD+yBgTK50AfsCjxI+kjcTfpRvAjtWPeZJkE+/vnZKssnv0zMDU6W/Pwd8j1iy37LqMU+CbPrVc2cS5/MN4ELgw3QNXZ3dC2oxnzq0ygegpubuEP5qRxJJx0/L+vcFXkwPwP2ISNrbicCOYVWPu13aBORTPNTHAbtl/ZY9IGthvWtqK8kmVyCXAJYlBW4RAVG3AncDU1c97onMaZl0Tv0269szKcAvAgumvlmB76a5L171uLshn35/7WSyOTvr2wO4DngEOJ9UTAI4JN2vaxnk1bTnTmk++cvXD9LY7wP2BvYnXigfIa0eqaXfquoBqKm5f/YguBa4MOv7JvBSugF/P/UNAuYBHgN+VPW426Vl8rk468sf6t/J+gcRFpY76VhyrKXVqwktk81FWd/3gb8BH6dr5dTU/4X0IFyv6nF3Y16rZ3/PRpSyvRZ4Gfhjtm1x4AXCF7fycU9APo25dkqy2YGw3F9OFPz4E/BfYEFgjbRtlqrHPBHZNOK508V8tgQeJ1K0LZ7ttyDwb2CXqsddp6aKdqJyilruZra+p4o9idWIi/ki4Agze8EjkOAZM/sv8aAUvUwr+ZjZaMJCtxewh7ufnPoHEJaKbYnluTXM7A53/0AVlXqeLmSzLvBt4CbgMCJv6eFmdknqHwRMU9WYJ4aZDXD3ce5+c9ZdRNGvRSzb32Zm17r7eoRVbHYif26taNq1U5ZNGvOqhAV/L3d/0cxmJSyrfwF+S7iGTEvIKf8u86ShVUHTnjsTmM+GwHvACe7+90KGZvYiMJi4dkRC2SdE5aQLeUD20BhkZjMAXwUucfd9iKCB88xsg3TYgsDz5e9SxZ6ep4V8ZiUCnPYEdi891H9DJJa/A/gfsVR3r5lN61FRSdHOPUhZNolZCZ/Os939UqIy2QaEUnkbMAfhbwh0ZKOoS1aK0lwKPgXeBd73qOD1NWBRM7uFsLjeSZYSzMyGZ39XmRKsUddOC9kYEcj5MvBK2ucVwqf1cWI+p7r7vwHMbHUzWz/tV2l2jaY9d1rMZ2BKybYpcH5SiI0OvW8O4j7xajUjrieVC1II6HyzdfdPPFIV/QOYLbOgHABcaWb3Ef5Rn1mSzGy+lNJoXF0e7k2i9DAcQyzD/9DdT4HPHgonE0upewCbuvv2hD8ewPHphl2LNFNNooWiMoSw3L2drEefEnlK/wAsDPzc3W+Fz6yWfzCzadJ+deVd4rxb3cwGuvvfCEV/JLAO8AdP1dfMbD3ghPRvoSxUqRg39tpJ58zVwCrAzma2kJktRSj0SxNliI/KDlkKuNDMhtdNyW/Cc6c0n0+JoLqngSGFJTm9YM1O5J5+FPhzq++qw3wqobf9M9TUJrcBvyYq9Hw+69uXeLP9Sdb3ZeLCvoiaRQc3rdHh52h5H3GD/QDYBRhaOuY64MayTIil8CWqnlPTGpGi6QHC4rgP4SN5HrGMfRwpiCvtuxThO1l7OaS5vAtsTwrcIvyJv1Xab33CevwfYPuqx52Nq1HXTmke+xKBaDcD76Rz6jhgSGnuRwJ35H11a0167hCW/EuJALtFiUC8hYgVo8eBndJ+swA7EW4v+5OCCelI29Y2wXiVD0BNrdyKmy3h+/i39HBfL7vxfLG0/1TpIXED8M9sv9pFbjehlR6GRqT5eQfYucVDfUViufHI8kMwKTn/JvxbJauekc3A9O8g4NikcBVKylGFkpLtvy4wlggiGpD110YepfPt+4Tl6/fAcln/HISldU1gxtS3L7Gkv3pfjncS5tLvr53SOTMb8NN0rh1MluGEUP6nAu6nlDe3Ludc05472XwGEy/JzxFuLh8TObF3TNs3JjLSfJjOqZeAp+hI17YiEUS5SNVz6otW/GhC1Iq0RPqpmQ0CLgFmBC539+PT9lWIqFoDHnD3M5P/1MXEUvHO3to3UfQgaYnxJuKmebq7v59tWx74MVGcYF/vCM5ZHvibRwDRbO7+cgVDbyx5UJaZrUZHwN2R7v5ekpm7u5vZLwmL42rl41v4KldGyVdyfWBO4hy6x8x2Jnx0lwCeIR76X3f3R8zsNuAGdz+8qrF3RVOunTxgzsx2J9IA7ubuH5T2G0LM96+EK89qRCW2u7xzJbY53P3Fvhp/aYyNeu5k1/JAYBtgFBFPcKO7P2dmGwGHE4F42xNZXEYR1vLfE+4iBxIrSj9y999XMI0+RUqxqC35wxlYCfifu//TzLYgksfPRPhELUAsyX3DzPYBNnb3VasbeXuQRTEbYYH8MNtWPNQXBA4tbqZmtjbwK8IisX5SzCqNQm8y6eG+L3B0/huna8qInLJLEFkCViKC9IYAyxbyNLPFgMHu/mAfD78T5aCodG+YifBp/Zh46M8GfIcILlqAcKN43923q6OS3+Rrx8ymI2QwB+H3PQ2RWeNTwoJvREDhrcCB7v62mQ0j3EVucvcDKxp3o547haLfon9m4I/E9b5J8SKSzsnjgPmBjwgf+CM9lfFuOkrJJmpLyVp1e7ZpG+KGOjuxVLcKESF8LuEnNTrdXD9y94+h8wNV9AzFQz09lPOH+nLAj4iHxuHZQ3094mE/BPhp8TBPD/eWN24x+WQW46PS580In8IFiAj7scTSKETQ2keEj+TLRABbYa3bENjKzJYpW//6Ei8FRaU/5yOsWNu6+/Nm9hJxD3idGL8Ty8PU6fpv4rVTshivQVhPpyUU3weBRYi5bgO86O5/LR0/nbu/lRTM//bp4DOa9tyZwLmxPlGIZUmPVHqDgHHp3HydDved4wqFuA7z6W2kFItaU74AzWxGwvJwc3Gxm9ntxJLdU8Q5vY+7v5sdM8TdP6qzVaW/Uv49zWwRwpd1RuKhfm7qX4+I4p4L2NDdHzazwYQv6wv50rHoGTLFsbAMfxtYjkhddh1xraxAKMQ3uftH+fFm9nngJXf/uZmdXqVCPAHeJwLq5jSzqdMY3zCzZ4jAwmOAu4qdzexUovDHNZWMNqNp104+H3e/ycy2JfxT33X3p83sSMJV5zLLUpilv5cALjKzPdz9hj4ffIk2ee7MSWRDeSV9LhTilYkc2u8Cx7STQgxSikX/YyzxVr6KmS1Kxw1pU2AocDyxnAqAmf0WmMXMNvaa+Uk2lMcIC915E3moDyIsR5cC9wJbp31rYfVqEumhPM7M1iEKebyTLIxrE1bgp1soxIsTAUR3Ef65r/X5wLuBu//DzE4iArxmN7OPiPNsK+AM4Hfu/j8AM/sZkeHhAzN7wN3HpP66KC2NuHaKe2yLF48vk3Lilu7BixKyGkLKAVzIpEayaeJz523iNx9lZm9kCvFBxHwPz10m8vHX5VzrDeRTLPoN2Y1yEPGw/oB4mx0JfIm4sR7n7o+n/Y8g/CnvBn4O3Jbe3GtRHapp5H6S2TLqusSy72g6HuoDiQfhr4klutOI6Ojz3f3dJt9wqyL/TS3yqn5sZscS5Z4XLu27OFFqeCDwDXd/JNtWm2undJ59g3jIn0tkBTiXqOD1RNr+SyJ37n+AfxEBedd5TQKHmnztJB/VgYQf7i/d/TfZtsWB3xEZEjZ29ydT/zTu/l4V4y3T1OdOms+DhHvLPwmXlT0I16Nj3P2MtN9mhDvP54gXg7M9AnZrNZ+eQkqx6FdY52jaWYh67qsQgUK/yW5MvyKsQq8QEbQfEMuNq3gEdNTFAtEoSg/15YGz6QjkyB/q5xDLxL8nHjArAvMCS+dLkKL3MLOLiZzFX8v6ulJSZgPeSA/D2lw71jn4bm5C2T0Z+LW7P5r6TwQ2Iyx5PwdGEKWJDyMKZVxfxdjLNPXaSe4Ro4lUeuu4+59Tf36ubZK9wMwFPEK8sN2Z+qouCd2o5451zrKxFbHa8CVivAdkLhNnEWkOPwVeI1aaINIh1mY+PYrXIC+cmtqkNLIE6cBfiAj6hbO+E4mb0tF0JPn/fNr3YmqWYL2pjbAOXUY8rIvPSwAPE1ajxbJ9lyQsFLumz7XI9dnERocx5EbCmgoRQLR4JpsFsv2LwKKT6SjCUMtrCFiDLJ9qdi/4OTB71j+CUFp+UfWYu5hH464dQvlaaiLn2pyEa8I4YLfSfb3qPMaNeu6U5nN4Gvs2Wd9ZSQ7nZX0LEdlCrqjbfHqqyadY9Ds83nCLN9QvA3O4e+GL9ivCKnQW8Ct3L3zYnjaz5wnfPNHLZMu4m6bPg4gSw+eQWSGzJbixhDVi5uIrAG/qEl1NWIdQTgAWI1wOWlnt7kj7DAd2M7OzPSzGtVmqz/xYb8r6TiaCCM8muxck5iMyI7zVpwPtBk27djLZ/CF9XgK4gEijl59rcxNBoCOJfMZrAHua2bHufla631dG0547pfn8FLjb3a8GMLPNCQvxVcAWZnaZu18KPEm8TG9CFGSpxfXfkwyY+C5C1A939+xmW9yYziUeJJ1uTGnb/ET08BhSHfgqxt0utFCWRgPXExav4qE+MHtoH0A81L9kZr8ATjOz4d6RL1T0ENm186m7P2Rm8xCp2ArZ5ErKfcQS/q8IX9x1gYeTz2dtHog+fraAVYEdiSIRJ5TuBfMQ0fXTEg/4WtG0a6csG8IFZHpgi9K59lciHeDX3X0tYHeisMlRZrZAHw65S5r23EnzGZjuBVdnm+YjfIx3IPzaLzazbdK5+R5R4GPaOpxfPY0sxaLf4p2jYQcQOVj/yPg3ptFEedhlCT+1WqQwajOeAW4mIpqLh3oR+HUmsB3hn3cmUeVqZ+BmM1vJs8IGomcoKSrPE8rhz73Dh7hQUl4AfujuN6b+xYisFDsSwV61xN1vM7PVgWdb3At+BHyTCCL8axdfUScade24+3lmdrW7vwHjnWv7Anem/V4zs/uJwMkh+XdU6cvatOdOFy+3MxBFb8aa2THEC9k5ZjYv6WXT3cf24TD7jqr8NtTUeqrR4edowIylbaOB0wnfqG8U+03k+4ZVPacmNUq+Z3T2ZfstUTTi/4CZs/6NiRyny1Q9/ia3smxS32giddb9wMqlbbMSwTh7pM+Ff/KgqueSjXFAF/1zZ/eCbfLx17U17dop/96ERfINwm1ildK2aYArk7z2JF5kvlL1HMrnWROfO8SL1iPAUlnfDwl3ibuIgh+Vj7M3WuNM36L98CyVkaecpPCZBeLHhOXkm+5+QSsLQ76klZZdz7CoLCV6AC9ZIrzDynUGUSXq+8A57p5XsVqO8GF9JT+2ict1VVKWjZnNSpQRfpYoRnBHtm16orjEEGBZi1ysv6nTUj20rlyXLFyHANsT94Lz0r7le8HA0ue5zGy25Nfb5zTt2in/3sDqRJq8H7j7Z9XjzGxaYiViBeAWYj5fBq60SL9XOT383FkZOKtGz53ziBSHvzazxSwKkfwC+AaRVu9v5QPq5hoy2VStlaup9UYjfJ4uIYI5tp7Afpb9vQIR8T2OsMCMoOaWpP7aiCXscUTu2BlK23YmlLLjgWlS30BgSPq7pSVQrUfkMpBYhl+z1D8tYZl8igiSWoBQyq4jlr6nqnrsE5nXGul8224C+wxK/w4GfkAoY68AbwIXAXtXPY80vkZdO8BsXZxrjxNZQwYXYydyAt9NLO/X7t48mc+d5Yhy0uMIS+zKvT3OicxhYPp3MBFk+xfgyOJ86mIOxTFGBEquB3ydCBCduS/G3VNNPsWiqfyHuEHt7u7nt9ohf3tPb+hHEjeol4ALvak+U/XgaCJw60ZPvoUAZrYDsA+Rrul3HlkOBhNBLNOa2RZeqr4meobMV3XnUv+0RLqpFYjUUod5+Ec+YWbvEorK4kRQXi3xKDu8kKfArjJF4JqZTUUo+gsTkfY/JJS0hYHvp+/4Tp8NvDWNuHYyK+vLWd+0wC+J3MuXAYe4+8fwmWV2DuCjfN41Y1KfO8sS852HSK23OXCTma3h2SpNX+Ipf7FHgZ+vAN8lgu6mJ4Lsiv2KORQ5nIcQL2NrEano3iVyG79sZvt6//Dfl1Ismod1JPRfKX1uuXRVujGdRNzM9gLWJ6xElSeNbyLWkSrqolL/9sD+wHNE/tgH00P9l8AXCYvFYMKKL3oYbxFwY2bDiQfd8nQoKXnw1vLEw+/10nG1uW6y+8GTE9j+aVKI7yB8j38BnOqpqlpS1q4CfmdREvdHfTT88lgbc+20uCcPIizESxPL9z8rzrXk1rIV4at7pXVUZRxSF0V/Mp47XybOs7mA1dz9MTN7hiiQcYOZrevut/XtLIKk5BYvycd2tV/pZfJ24tq5ibASv04o+psSgZ9ru/vdvT/6KaRqU7WaWl83Oi/7LAvcS0Q+jySqRN2WbW9kgvK6NSL90sPAn4AVUt9g4DdEkYXjgOEtjqvdcnCTGuEbORb4GZmLBLFM+g2iPOwf6Fiqn5Fsubvq8U/CPAcSeYBfBr5VzJUsgDDtswvwN+CLVY85G1djrh0imO74Fufa5sDf0716ZOqflahKuHHV4+7m3MouE9cTLhObpL7CxWUosTLzJvD5qsc9sfkQMQZ3EWnndgOmLe03fZrPP4C5qh73ROdV9QDU1Kpq6cb0FyLSfj4i6ONfZNWisn23JKtepNajcjCi3Os9wLKpbzCRG/eJ9JAcXjpmBDX3Y21CS4rgt4GpS/LaPD3kHgBmSf2jiKX924oHfH9phI/080Rw17Bini32m494Sdix6jFnsuj3105JYRyQ92fn2t/pqBQ3G5Gf+UPgjKrHP4lzLZ47YwiL/1ukTCF0ZLQYShQDWbLq8XZjPj9Pz9A96HiZLGcZWZrIZrN21eOd6HyqHoCaWhWNsBAXN6aFUt+phM/kTOnzdERN+LOJN/r3pBj3uBwKa8OATA4GfIcI6joBmK50zEgiVdM9yJLfm7Jpla7N0sO6UFJmTf2zAT8hfCrP68tx9tBcdwE+KeaT9beyGP+MyDvbSYGrYMyNvnbSvLYEHiopxLMTFdj+B5yY71/1mLsxp+K58yowf+o7nXBBWiqfB2nFJf09f9Vj72I+Q4gX4z+RLMRlhTjb9wbgJ1WPeaJzqnoAamp93QifrzuJyPIFUt9qRMTwl4j8mGsDjwGXElHQzxE+iV1GFKtNtjwGlD4PTjfQv+YPhrRtFmKZ+y1SrtlsW60f8v29JYVrO8J94G9dKCm/yfYfkv1da4UF+B7wZD5WYnn+VGDp9LmIsJ+9rGxWOO7GXjtEsNbz6T78udQ3Cjiwxbk2tOrxdmM+XyFWUV7JFOIB6XlzMmF4WSL15y9hf0nHzNiX4+3mnJYE3gfWKObTYp/iulmRfrCCpEA70Y58k7hAF3D3p8xsBqJKz/lEUMqRxA35d0QC89kIn+PzvIuIYjH5+Ph5ZUcQCsm1niLPUw7MkcAxhKVyW3e/LG0bAXzgNa0Y1RTc3c1sTeBzxLLuGDMbSSjK3wUucPfdAcxsanf/IP+7TsF3LXgSGGZmC7v7o6nvTcJ95C4zW9bdHwZw95fgs9zHuxCK2zPufn1fD7rJ1467P21mPwRuSefaIGAjIqDwD6Vz7f30961EHt0rKhr2hFgbWBlY3aMyYRGY956Z7Ue4gixqZo95Ch40s+sIl50DCOWzbgxK7SNonSPcUwCvpyC7LFi0nlStlaup9XUjHnRzpr8HENG/jxE5It8mrF6LADMTEbX3Ajtnx9fa6tXfG2GRvJBYMl2KiMZemgjoeg/4erbvLEl2F9GxnFw7q1dTWrp2RmSfNyGUx1Ozvtz/+L50DdVaJsCiRNDgPoxvYX0ceJpkGU99XyHSVL1IuCo8AZxQg3k04tqhtT/3MCIQ8vwuzrWrCeVsJ0JRG9DVd1U0p0HAo0TJ7nlabB9IVtWOSA04lnjpnKavxjmJc5oxPR/3mtTnYl3OtXKrvMKNEH1JkWbG3V9IlqtxRJnRK4iAgaXc/VDirf0ywq/4FHc/Mx1fvN2LXiD9vg5sS+S4PIR42F0NbEhYuS5N+85CPOwXIKwoO5jZNB7ptQa2/A/EZJNdO2MtQfgR/83dv532yS3E1xEptM4Hps6/p+9HP2Hc/Z9E0ZKfAVumFGyY2YFEnuI7ieAnzGxq4DRC8VwMWJNwv/iGmR3Z96MPmnTtpHmUWQKYici7TErHVpxr1xOrfzsDl3pYIqeZwHf1KZl1dAlipeU8M5urvJu7v5v2v54IyPs/4BKPnNN1rBg3lnix2obItQzENV4er5mtbGabm9kG6T7xqVVUKXKCVK2Vq6lV2ciCGkhWB8JqdAfwIJmFuMWxtXzT7e+NjspiA4CFCCvdGOBr2T4jCQvF60TxiJ8TAUSPk1lb1HpVTjMQEeXHttiWW7mmS31zk3w/qYn1rjwWIgXdS4Srwfnp75OAebN9Pp/mdgwdFtZBwA6ExXijCufS2GuHWL37L6EoFvOcikhtNpYo4V2ca18gXGJWzo6v9Jyjc8XEfxKrKMWK5cBsv+sIQ832lNKb1WEe5XGkc/9h4sXxa8TLSH5NfZEIVv+Y8Gd/gQjOG16cq1XPpdO8qh6AmlrVjVhyLJTjhQmF+AFgl2yfVQhf5IOI/KzFw31QX4+3HVomDwNuBbbKto0kAon+TVRPKm7OiyQFYN/i2Krn0eSWZHMpkR1gYcKKNwq4sYWSMg+hYJ5CR+BNbV4q6ZwG7BtJWfwXkRlgntK+0xKBt38u9c+dFLTvVvmgb/K1Q1QYfB04ikif91D6XD7X/gK8Q/jizpMdXyfF+GGiMEbZZeIN4mWyyOYwDeGmMDyTbS2eO6X5XJGem08Am6X+LxLuIm8Q/uwLExbwW4nnbO2CcSsfgJpaXRqxFHor8Qaf+xD/mAim+Q/wDGGB+Ds1fdNtSmv1u9Jh5fo3sGpJmZk6PdgPTZ8/s2RUPZemNTpbie4mshrcna6T10pKynxJEfuEsEhuRUexj7oqxoUiUmQ9GFT6dw8iW8D+pe9YBpi5TnPJ+vrttVMa6ypp/OMI63bZQvxQ2nY3cE26n+9U9Ryy8Rfn0EBg+az/WjoU4mLVcnWiMMYzaR4XZM+dWlw7dF6dWIpw3ylSGZ5FxBx8sXTM5sRL5xdayHdklfORT7EQHXwHmBc4yTt8iH9M+Bk+QuQCnYfInfkacGkqNyof416g/LumLCF/JZSVHYA7S/scQTz45zOzXwDHmdm0nkqW9s2o2wN398xPciXi2lmEeADuD1zu7m+Z2RcIa/Jowh3hWeL6+buZDfMWpaWronQuve/u//PIerAKEbyFd0TN/5lQUnY3s8Wz77jP3f9rZgOq9AFt2rXj7uPMbED6+3bipWUcUSzmiuxcu5gImP6Ou69IVCe8HDjWzJaoZvSdSb/pAA///HsAzOxB4oVqb+Ayj2wt2xGlxV8mfNgvAuYE7qvTtZPNZ5y7P+Tuv3f3D9N1sw3wLY+y4wMLGRJW/HmImJ3Pzlcz+y5wvZnNV8FUAKQUC5GxF7Cbu58DYGarEmUrbwfWICLOIazElxBZK4blX2BmM5rZ3H014DbjbcKHdXfiof7ZQ8HMjiduwNcSS/T/JnwqbzezqeryAGkS+cMQ+ICwQp4M/LGkpHwC7OPu27n73oTSPJhwM6BK5bErPJmskkK4OvA9M5sn2/4kUTVuNOFjTPHAN7N5k4LgfT3uCdDvr51ciXf3a4nYj0vd/U0z+zwd59pR7n5q2u8lYt5TkwV7QrXnXT6XdN78h8i/fJm7v2NmqxNluocC17n7Ee5+NGFF/pBYqajNtdOFYWgGYqxPdOzm48xsZsKV4nFi3gCY2XeI1KhjSUGtVSClWAg6RdZfk3XPTkRm70r4pl1sZt9MD4kxRD7QGbLvmImo8f6wmc3YZ4NvAzL5bA7cVHqoHE/4gf6WeKm5w91PI/wOZyei0kUvUFjwkjzWAc5x9zeSknIRHUrKedlh7xD5dN9P31EooLWLRE/X+u3EPNYxs8HZ5ufSv4umfceZ2UrA07n1uGqadu1kFuPH07m2AOGW8wFxrl2U7TstsEX6uJKZbWRmX07He9VKZXbtrAeclRTiaYDDiCIsBwGnm9k30iH/Idwupod6ZNaYAB8S7iDD4bPrYyQhj62Ba9z9BfhMId6LcBP5rkc2mEqo3U1IiCrowhoyZ9r2JHB0emifbWbDiCXIm9z9WfhMIT6CSC6/l7v/ry/G3S7k8ik91I8lluPPIfLEvpIdthxhccj7RA+THnZFYY53zWwOQpF8gchMkSsp0xPpy6YFVjCz04mI9IPd/e1CgatgGl3i7jencR4FfGRRIGIMoUx+RKQ8IykzWxG+rP+tZrTj07Rrp4VV8kvEysMJpXNtOBGYtwlRlnwUkYJuczPbw93Pr1qpLF07H6buwUTQ6u/c/RQz+wQ438w+dfeL04vZOxCW4qrnMAFuJN0DzOxXxHX+deKc+6O77w9gZnsSKxhPAoe4+z9SfzX3Aq+Bo7aaWh0bsAGRbH2FrO9HdAR4fCX1jSR8vj4iliHzoIHhVc+jqY3wvxtHpJSatbRtU2LZ9A/UsDxqkxthyToT2L7UPxw4nPAr/jOwLJFX9moian3qvh5rN+aSp5b6PpHi7HHiYf8R8AM6MtHMBrwKHNbq+Dq1pl07pICtFufak0SwV5HxxIjVvL8Timft5EO4edwK/CB9HkAo958QqdxepYtgNOoTfFf83oPTtf5kOt/+Tljzi/32Ss/YK4gaARBp9oYUc+/rsRcRpkKIEsmf8DbiprQL8JS7f2xm2xJuE78mLF7HEb5e33L3c7Pjf0c8XE72VLZT9BxJPrsCV7n7i1n/RoTf6lTAnu7+1y6OVyGWHqYr605mtduCSMW0q6egNTNbh6jCtoG739mX4+0O+XliZl8h3CW+CPweuM/d307bfk8ESy3knS2AtWNKr5260Oo3Tufa9wirfadzLW3/HbAgsKKnUth1w8xOIc6lrd39idT3Q0JmP3T3i1Lg4BzA/MSL2i0eAXq1KKNcjCO5u4wmXlRec/eX0/a9iADdwkL8kEVhnEOJOW3p7h+2/vZepOo3CjW1OjY6v+neTaTFOYLM8ktEQJ9JLHttnfrmBP5GLB29D2xMTd7em9S6+k0J95U7iLRMK0/g+CLf5wBgT2CxqufU1Eb4Px5GPPx+SynNF1GMYQywZNVjncAcWloUi37C6v0PYD9Smeh0fxgNnEeFBT1ajHlKrx0r/93V71PB3AYTRorniBzTeVGMImjyCeB35fOwDi37PQclOdwBrAZMn/oXTv9+I83xPcKV4nkiG0rd0rW1tPQSgYJlC/FQokLmI0S+76FVjFmBdkK0wFMJSg9LwipE0MP/gJnhMx/iE4icjDu5+/np0A+J1FQrAru7+xVeMx/JJtDqN03BKPsQysje7n5Hq2MLy1+yYHyXcIm518zm78UhtzNbEorv7cRqSmEhtpThZVeiVOzj1Q1xwnh6aheksedWyi2Ie8P5HqtJixDZKW4hFLHFkxWscnrg2skzc8yRYi0Gt9q/r0n3678B9xBp2T6Fz7I0rEAozIOIuI9Pqg60K+PeKdXhMoRv+jHAD9L59qiZbU2kN7yTeNmaligsNRa4zcyGeufsIpWl1PMWK3Fmtj+R1ekJOizEQ4kVim2Il8tN3f39fOzWkc6t1wetpqbWRaOzpaGwLs4AnEpmIc72uZa4Of2D8D/sVNaWVAxArVdk9RvgRWClCeyTW4i/TygtYwh/t9WqnkMTG2Gh+zGZZY7w7VyJsIY9A8yUy6eL76mFNbLFuJYn/Dw3I4oXHE/4G99DBOcNooXlrk7z6c61k+07A5G67QFCaXsifV6i6nm0+m1L59qzwOypvyuLeeVyoXNBjNVIVe+IinD/JpTi+UtzXJbwRd4w61+KSHM2Q9VzysZ5HvGCvETqm4pwrXqcKE5SzDW/XxQ+xr0uG/kUCzEJmNl0hMvE+kQZ6POzbdcSN6Y9iET5FxKJ5Ff0sEyuQNy0Fnb3f/X12JtOsios4u5/72J7biHeh/ADfwOYhYiG/mGfDbZNaOVjnKxzXyYCnkYQS/UvTCjaPF136wN/d/dHenvc3cXMhhDpGvcklPvZiJfi04CLPGV0MLM5CUvytMAYj4w2tckeMLFrJ9tvJSJbxTREvMXFRPaH+Yk87hu5+129O9oJjq/T75nOtZWJcsojCaX/xa7Oy/KxVcqmVcyDma1NuBxs7O435Pulc3FrwiAzO3Fv+w1RaGpjd3+zD4c/HsXvme6/8xTPwOTHfgxh4d/B3d/N/aLN7Fw6fMB731e66jcHNbX+1Ig391uJwLvcGnEt8DoR3FGUuBxFPGigw0d5c2CWqufRtMZEfOjobCHel7BuXUpE2l8JrJ9tL/at3GLUtEZYilYnAlCfAebI5dPFMdMSD/sXCYv+IlXPozS+ndO4HiWWfxcsbd8VeDDtM5aoULZ91ePOxjexa6e4dy0HvERYwL9C8p1O2z5HuJO9Bixa9Zyycc1KWFZfmdC5ll3z8xAW/3OBy4iCGhtUPY9snLsB75JZktO/+bNoEOHuN47w45++6nGXf+dS32XAU5RKqae/f5Xmux8wbfmc7I2mPMVCdJPszf0rpf6rCP+vPQiL44cA7v4q8KqZ7Q1sb2YruPvFfTzstsAn4LfdwkK8C2FN2ZoI/JrB3a9O35NbZmYhXCtED+HublG+dhHg8+7+0oSygKTcvxsTqcNmI1yWapUxwN3PNLMngYfc/Z18m5l9DziEsO79mCi+sD7wGzN71d2v7+vxlpnQtVNst8jN/jvixWRHd38MOjIMeJTDPhKYF9jMzB7xpL1Uibu/YmaHETnlX+zC+lrcH1YGTiKCI58jXC02BHY0szXd/bt9O/qW/Jt4sVrVzG72UpaTZBlfEpibqGI4NalIzoRWYvqKFr/9DMSLy02erMDeYSE+iSjTvTdwQXFtJZ/p93vNkl/1m4OaWn9uxAP7I+DbJL+n0vadiQf5OcDMWX+f519sx0ZnC/F+RGTzlYRf5OrEQ2bRbP/pCGX5fMLSskzVc2haI3yMR+by6WK/aYhA1reI1ZkzgF+k42tx/ZTHQecYhI0Iq/CxxHJx0T+ccD04ofwd1DBXcxrXYcQL4rpMYAUF+CVwedXjTWOx0ucu/brTveA9ItPQjtn2mQjL8evAiTWY00AiE9IdRODg1Nk2I4wzDxP+ucsTFuOXSX66dWxEasNHiBflYUQGp9PSc/PbZJbu7P5d3Nd73GKs7BNCTBmfEsEmj3gpF7GZ7Uxc3KcQuSX/m/oHe4flUvQi6XceSGSZ2JFQgndx9zeI6koPufs/zWxOM1sLuJeogLV2+ood6pI1oAlYR8nh15KlZ0IW4k2Ja+dCwlq0FvBgOr4W1095/N7ZErcacW84y92fgc8sq28TL2lz5t9hZiOAU83siL4Y+ySyDKGQ3OJJG8nJsjicBFxfE9l46fN4VlJ3dzP7IvESfB/wI3c/Cz6T1euEm9XOwPpmtlXvj7w1maX3K6nrCGAbMxuQfu8vES+OUwHruvs9wE7A5YQLUvE9tfAQyM6ZHQj/5yOIsf6ZeBkuLMRvpv1vABYilOWtiiwbPT2fyk9cIfo51xBLbSea2axFZ6YQnwT8wjsCbo4ArpJi3KdsQSxdP0VYgf6TAoZ2I8p2b0w8TM4gHoD3EAFgZwJnuvsHlYy6gXjnksMtlz4zhfjXhK/+7sQD817Pyvim62dwft3VhfQitRbwsLv/M/UN8UgDNgeRFeDhbP8ZgIOJ1FovjveFFWFmAy2KYcwNXOvuH7ZSQjJZvgCclmRTq3RnrbAombwb4WpwpLvflvoHeMdyvhOW/UeJLBaV4J3ThH4FOJK4JsYRyuIZhLvEeu7+bDrmRXffA/jYzA4ws2Feg8IeMF76uZUJ98M5iLzmewEXZgrxXYQ1/0oi08vmwEO9MR89kIWYTLI391WIN90ZU/9OtFaI9yEu9ukJn7Ah6eFRizf3BnMRoeB+293/mzIZbE24R+xIpAH6C+HreT6xVHwZsbz9YDVDbk9KCvHt7r45sDixjH12tt+cFtlcbgYeNrMvVTDcCfEpYVkdmOaEu39kZjMCfySKLZwLn+U8P4xQ/ndy919XM+TxSVb5t4mAreWsi2pplnD3DwvLd1J66q4YDyP8hq9z92uKzsx6PyB9HktkFJq9yvt1eqkqVluuLV64CLeVEcAa7v5M/rub2VyE69hhdKyA1cJinOZT+Hm/SKysHEm44LwBnynEnycsypu6+15EefVhRG5jevI8q/xHEaK/UnpzXw0gBdUdzfgK8QFEkYj3iMCH7wCHpwCOt3staKDNyV5cfpA+DwA+SO1+wrK1uLs/mQJtrgduAg5y90fTMZJNH5CCuTYmUrXd6u4bpU0HE5a869N+2xPW/xkIv0qI4gbbu/v7fTjkLvEo4PFnIiPDbWb2bNp0MhEoeEhSXqYj5rcb8dJ2dvEdNTvv7iR+84WBv0OnFFudArjMbDZCXu8npacWZYe7YG7CT/d8+ExRdGAN4Dl3fyIb/+lEWrdKS8OX3UDMbDSRgu1Uj2DCPPBuDiK4eA/iRe3LZrYl4UL2DjWgWDFN/65K+D8XQXV3EgrxXkQQe/HbP0n4UA9N31HMd8qDCb0GjtZqak1o6SK9gwiqmzXrP5AIUjkNWCj1LUg8aP5IDcuNNr0RPnYjgWnS57UJRfmPdFHymZoEdzW1AesQCselWd/GhNvLUkRe6SuIl8q96bDwXwCsUPX4szHn6bF+QPixv0P4Qt4EfJVYpZ2asIh/TFiIi2NqUaK3NKdBhKX0biK7wXgBgUlWF6V5PgH8iZqVHW4x5nmJlF/bZH0DiHLD4+goQWylfxcCFgVG12AOlu5b95IVhyJcEX6V5nEk4c6zCRGodxU1C77Lr5v0+UkiSHALUprTbNtBaV6/I4JvDyWlbJvSc63yH0JNrQktu1kOIFXnSp8PSArxr4H5SsecSlhdpql6/O3UWjzg1sgU4oWy/b4AfDE9OAenPr3A9J5cBgFHZJ8HAz9L18/LRCnoE9LD/uvp86V1UoizsecZJRYFlga+kvVNQxRWKCvE0xEv1l+reg7ZmIo8xYOTQnUvkdVgbTqqj32PyMX8KJFG6wdExpB/1vn+RrwcX0VUIeykVBHpGt/J79vpN/hx6n8j/RaHVX2epXH9Czg9fc4V4t1Kx8xNuIzkmVJq9dJCWO+vIlwmygrxT4gX5fvSObgfcDXhoz/F2Vsqn7yaWlMa46dnOjI9zFspxLMD1xE+kdOBCkX0sawKhXg1wvJ4BR1W/OGE+8srwCfpwX4LNbd69edG6WUjPRQHpwf7w0QWiqIU9NcI6+sVwHJlmdaldTUeIvbgJMZXiIene8Lb1KhgRC6fJJddgeOIJW1LCtYDRJaQzxdzJ1wtnqNzirPaXTtEZok3iMIrxcrR1ESQ5xhCyS/uF9skRfP7hGV8t7TPr2ogm0FEbvXpCR/jcYRLTrFfUSp5JmAJYEtidWbq/HuqbnRW1geXtv2EWIU4l85pDlcg/JHXn+L/v+ofQE2tiS09FP5E1HkvK8SjiCWfT4FvdnG8lup7X0ZfSg+OS0m5iokXlDOTMrwTsCax5HgrYdUfmsu46jk0vRHBNPNnn3cAnias+i0txHW+dpJCcjLxIlZWiP9MRNavWxcFpTT2gaXPhZXyAiIvbqEQF6sqRvjsH1Q+nnjZ+VLF88ndXA5MivG+RIaJywlr5O6Z0jgknXdPkHLnEiuDXyOU/z3qIBsi+Gwc8K2sr5DJ5wkXntfSPi8Q5ZWLF/46XzsHpvPsXGCu0rbdkvymOK985RNVU2tao7MrxeylbaOI6lzjgB+Uti1MlBkt3uhr92BsUiMt1xOBdkXfr5NsxgFbZ3JcPCljm6a+/CE0vOq5NLGVXzqIl5QnCQvxl7P+mYkSxHNQczcXYNV0bn1WJp6wiN9AKMTrlccOzFb1uLuYy0CiyuC7wO7ZXIp5LZvmum3puOIa+z8qthzT2c3lO0lhfCnJYo9MIR6Q2iVEVcJR2XEzEisZfyAr0FThnAYBG+VySv/OR6x+3ZXmNpx4Abg5tfGKT9Wl0ZHL+KwWCvGqafy3kZXyZjILfFQ+WTW1JjZa13j/HHB4eiD8OOvfEbgx9b9K5Mkt3txljewd+YynNKWHxnOE1eEHSR7fSNs+nx6Gm5WO2SA9SBfv7TG3cyOC7J5ooRCfS/ixjiOsSDdRczcXYOHS50sJy91XWyjEmxIBbmtXPe4u5jKMcC/6CZ0trwsQVtUHSMFqqf8UYhVmH7LYi4rnkCvG8xE5czcj+bISFu9CsVyDCCQ8ovQdKwP71mAu5fOneEGZgViJuIvIcZzLaivCYvyFqsc/gXkNJCzF85T6VyJcjl4CVkp9g4n0mn8krexNynO08smqqbVDo8OP8BOiul3RfwSxLF/kyd2R8F+9gexNt64P+CY1YDHC6vXl9PmHSdnakLAUvwOsle2/DuFW8SZZqWi1HpfLeuk37hRUlx7ybxMvMF9MD/ebKLm51KXR+kV5I5IrBSVLHbEsfzPxMrBmaVstXpaTwngxEXC2RFJIliWKLDwN7JPtWyjE+wEzVj32if2eaW6rlvo+R2TY+ABYp9WxdbxXE9mOxgB7ZopyofSvTyj6y1U1vomMveXvmSnE/wFWS31DiCwbN6d790gyn+tu/X9VT1hNrR1ausHeRRapDGyfHuBn0DmVznJEoYJ16OzDuiA1jLRvSiOWgu8DtshkVijG44Dzsn3XTfJ8jpTCjVi2rN0Dsb834qXwCJIlKPX9jsj5/TGwQ+obTKRue54W2Rtqqih/nVgWXq7U/zXgduAhUtaK9MCfq6/HOIGx55kPipzfL6dr5f6SQnx6phB/rvQ9tbxmgBWJ4K3yC8lKRDzIMcX86zyPNLatklxmTJ+nyrb9AXiGjkwiA+iwjNfSx5iw2F9NrKzmCvGaxMrFm0TxnFOIbC7d9pmufHJqak1vlFKAZf03kOWWpMOXeE7irf51Im3bgKSg/Szd2Fbqq7G3U0u/8eXEcvUCWf++RIW7+dLn9Ygc088Bi2T7LU9Eqk9f9Vya0lopGkSw2gPEkv1B6ZrYJm0bnq6d7UrHfJF44Vmkt8c8ifNbKynFW9ORXWPr9CB/KHvgDyayJDxd9KX+Si3GdM5KsQ1htf8Gnf30z6BDIZ4l9U2X7m21e1HJxr1wuhf8qDzOpHjdQZYdAfg9JfequjTCZeKF/NxJ/YeTikmlz0OJvMB3Zc+j2inGRPaTcXR+YVyTcKF6ggiino1Yvbid8Dce3K3vrnpyamrt0JLClQfgLZgu6iJwa0i277Zp23XAJln/NMCxwEfAKlXPqUmNDqvXoKRw3ZrkUFhWin/XpMNCvHjq24bIl/kvIo1TrZLiN63RkTWksND/KH1eG1g9KWBfyvZfIsn0JVKGhDo1wlfyH4Tl6+p0ff+Nzgrx1klBG0esUuQvbVUrxl1aSOlsIS4U4uFEdcKxwOZV//4TmdtuhHFiWzr8U+cjLJHnZvutCvwP2LnqMXclI8JF727gm0Q6uQsJN5DjSIGDhKvYU8SL5WhqGvSd5vPF9HeuED9GyVcd+C6RwnHubn131ZNTU2vHRvimvUKWRzL1fzU9vO8m6tiXj5uajrf7SlMaNa3ROd/n2UkGW2bb10p9uUI8A2E1Hkfk0/0sOKfq+TS1EdkmHkgP92KZt1CM3ySKYkxFvHwuQbgoPUoWSEQNrF90DvBagEjT+DJRBCMPGtqGWN5+ifDfPSkpattVMe5JmN9vCAW/rBDfmBSYb1MqzFCXRudAtB8TfqvnEun0/pHk8cVsn18Qy/ULVj32FnPJi69cCzybnh//InJOFyuVG6fz6iXCCn4xYQCoVdA3nTP/DCJ8oh9J13ix2pIbmc4gAlmn79b3Vz1BNbV2aoS/cGEdvomw/qxEWB+2Sw/FW+gc0GWlB+hhSQF4H7lS9LR8igeI0Tk/7tp0uEwUCvFg4MuEFWIs4ctaKMWF5bmWD/3+3JJsLiOytOQy2pNwSSoCJRcnFOLH6awQL0qUiZ6uDnPJ/t4wnUu5hXjb9LB/EBiZ+oeke8Cr1KDMcBfz+l66R+2bjXtaQiF+gkh/Vg4sLFbSauGbW7rnbgWcn67/64FVsmt9ESJQLfehLud0rtqaX7zwDyDSfi5F53iVjYkXysvT9TE8/XsvUVmuVpbi0tzOJ5T5Vgrx/sBbwP7dlUPlE1JTa5dGvNX+ND3Qh6aH3pVE0NDr6SHyF8ZXiPMH5yFpv0OA04jlr5WrnluTGuNXJtyY8Et7lg6FeAjxMvME4fu5VFLIXqajKtaMhN+k5NPDsknX0oOEr+AudBRTmC39uyjh9/koSXEmLMsrEllELqYU8FWHRkee1cJl4jHCF3pEab+D0rZRpf66WPMGJUWyUIiHEjl+H6O1QrwCUVa5ViWh6awYD6SUk5xwaTuFeGEu56SflxarfXWYS9Y3hMiA8maSz3Sl7Yckmc3eF2OczHkNLMZNZ4V4X8IN5GQm4eWx8gmpqbVTA76QLtSz6fDj2p4oSnBDfhOdgEL8U2IJeSAdAQeLVT23prZ0U/2ADh/WIYSF+AnC97OQ45xJHjNkx55EKgKi1mPyKLu5XAVsnG1fvKwQp/4in/GdRN7p2li/6BxvYESKtn8RL8kzlPaZG7iG8GE9iyiCUZuy0K1+VyJ46zkiEG/q0ralCR/+cZSqf9aple7Fha/tKOBF4NBs22pEtpRPiGX9efpynJMyH6LU83+Jl8RCscxfBO4kXMZqc62U5tDJlSL7ex/C5eV0Mv/7bn1n1ZNSU2u3BixJ+BM/SyyXfky8pXfHQvxTkn9e6h9GWF6WqHpeTW3Ey8e82eeliCX5v9Gi2hiRG/MQFHDXqzJJ/xqxHFxY5wuXibJCvCjh4vIa4eZSKADdikivYH7fJF66RqTPhYV8ViJP7nuEH+tuhCvFO6RUgnVshGX7fyTLcda/NOG/+hopM0i63mrlSjGBef023QeGEG4UpyYl8xEinWNtUuh1Mf7riZy+nRRi4uXsV4Tv9LfLz6O0Ty1WJVrM6afEit1pk6oQu/tnJ54Qog8xs3kJH8LPEz7E97v782mbAXhx5zE7hEg/dTBwsru/Vvouc13IvYKZDXT3T0t9dwCzEP7cr5nZAHcfl7aNJoK+vkVY767p6zG3C/nvnj5/CTiTsCBv6u5Ppv7FiGvsUaLE+v5EgYlZ3f3ddL2dA5zq7nf18TS6xMyGuPtHxTzNbFbgBCIl4FbAde7+iZlNSyjK7xCrEp9UOOyWmNn3CeVqA3d/LPV9mVDsFyJSa/3TzIpVmB8QpYo/LMu5LpjZssT5dgcROL0oYaT4KfCgu/8t23ck8K67v5c+1+KebWYDCN/id7PzbABwPHGeXU1Ywcea2dSEW8/HwEdp30F1Ot/StXwVEW9zoLs/ManfMajHRyWEmCju/m/iAdeJSVWI831Fz9NCIZ6PsASfnhRiKynEBxD5ZLctK8R1e4D0d1ooSv8HzEEUuHkKxlOIv+vuD5rZ34DdCWsY7u5m9l9gLjO7u+rrqVCY3P2jNL5xZjYbkY5xPSIjyjVmNjBtf8fMpiMyblhlA58wxxE+0iea2b2Ej/HXCZ/cQiEeRFiOj0j/zm5mL7j7xzW9dqYirMOLEFkazgTOdvf/FDuY2UFE/vLFgUfN7HR3v6jqcww6vVSWFeLjiPPsejoU4jWJ4NTFCCvsM2a2Wzr3xjMcVEFx3ZjZ1wjf/Ocn63tqIBshRAvSDfWnxFJ8S4VY9C2ZJWJmwo/11dQ/mg6FeDt3/33qnwo4FDjE3d+ri4WoiSQlcTZ3fyF9/gKRoeIRIjPAA6m/eHgakd7wZ0k2dVS8MLNpCPeqNQnr6bX5eWRmewG/JHL+XpbPrw7nWqE0JaV3F8J9bFfCF3ddd380bVuWSHs2M7Fs/yjhp7++u79dl/nkmNlOREDt8cT7Vb5ycQ2hMD9G+IbPQ1j4t3f3CyoY7gRJ18NZhBxuAg5299fNbHOi6t01hGvS+0QatJmItHTvVjTk8eiRVYVJ9bdQU1Pr/UYoV+OIxP4jqx6PWid/u8FEENRJ6fNoIqBjHGEhLvafivDLG0ck9/8sc0LVc2laY/wUWNMSD/AHyfzt6Zy94twkm/WqHn835vcdUjAhnWMNdktzOI5S0YLS8VWnBMsDoo5IY14gk8WKRIzFPenvWQi3ivuIIka18i2mdSaHXC7nEvl+d6LDX3cocAyhIM9UtUxazYlQfs+lw5d9LaKk9Thg72zf0URe9gPrcH71ZJP7hBD15BziRnSNy0JcC7zDh+5jM5sfGG5mMxM+xDvR2UI8mAi62Y4IvNke2MHM9vawetViybEptPgtpyMe5le7+8PQYUVKVslzCJ/+bYgI+8+ok2wyV4qT88/p792IAhmnAkd7WPWGEvmaZyOWue9199vdq7Ucl37PnxDuR//OLMTnE0rkFu7+EoCZvUbkyV06/646yMdbWCMzuaxMZKA4EbjIw8VggLu/b2avEwrlJ7kszGxxYFV3P7FPJlAiuza2IYJW30nn0h50WIh/aWZj3P1CIrPDh8D00CwXvgFVD0AI0ZliGdfdz3L3MVWPR3TgEdg00N0/dfc3CEvdtwgLcaEQDyGUrs0IX/ANidzUcwG3mtlUVT/U24BXCKVwNTP7XOH7nZSwc4GvEYFfV6WXlKnN7DdmNtRjqb8WvrllZSNTvPYEfk2kCzwceNXM1ieUl/2BNYg5XmVmO7T6rirI7m3/Tu4uKxN52l8k3D9eynylnfBhnR9Yw8zWzeQzsLJJTJylCBeQ3yTlMjc+zkSkahtSdJjZwkRqwcPMbJa+HGhBujYGuPs4d38ndQ8AFgZuc/cDiMp+fzCzjTwCBmcmFGPqcr30BLIUC1EzvIZ+jaKDkkL7C+Cv7n4DfKYQn00owt8ilK530rYPCWV5VSIntegFMqvXxoSy9V13/1GmEG9I+LRelZSWIYRi9jXgDjO70GsYWQ+fKR+LEm45pwPHuvuLZrYCoRzPSaQB+4+ZfY5YoTjazP7hyae6SvLfMym3exJK4hbu/nJuBTazqwml+TpiGX8JYi7Le438WFvwCXHeDYGOOZvZhkRBiYM8Za0hXETOIQIOv+xZkF5f08L6/T6RW3rq9PkoUjVJM3uCcA87OR3b6YWrDtb8yUWWYiGEmAwyq1euEJ8JbEBYIS/LrC4QpbxnJqoXil4id3Mh3AgOTwrIWYRscoV4KqJwwdeJ5ftNgRPMbHixKlDRNFqSXCn+AawOHOmRxQYik81URGGgM9K+Y4iyve8T514d2YJQBssK8Z+IipHfJizI3yOW8ocSJaTrzL8In+ivmtnMFmxFvCxfDVyY9isU4umJFIKPVDHYrkhK8lPAJmY2f7IiH0G4i01NBOK9ZGaLmNnGZnakmX3XzIb1A2t+lyj7hBBC9ABmdjCRLWRr4HJ3/zDbtgZRZOF/wK7u/nLp2NpF1vd3rHP+6CK14dbAnzKF+HJgQSKjyIlE6qzdgBFEee4PKhl8F7SKrjezJYA/ApsDbwN/Bv7u7uub2ZxEYNe33f3qPh/wBCgpwbmsriJWU/6P8Ml9NzvmBeJl8/+qGHN3MbMDgP2IF63BhAvIhcBx7n6fmS0E/J7wfd/I3R+tbLAtKO5HaXXlPuBd4t52v7u/ZWZLuPvDKTPFfkRGkf8Qcx0LLFsyCPQbZCkWQoie4WfA2sAVJYX4K4R1a1qiQMTL2bapSw8g0UOUlMfDgHWIwLtcIZ6fSKd1gLs/7e6XAUcTltWV+nrME6NVgBdRgnwEMKO7P05YwxdLyuUOwHDCelkr8uX1TCH+C1G8Y2/GV4h/SFxD4xXEqYtVMq1I4O6HE7/9hUSe7K8RLyb3mdmCwHmUFGIzWym5/FROcT9Krh/LEKtbRwJ7pfvVw2b2DSIV4AvAV919LsLN5WVg4+K76iKb7qKbsBBCTCHZA+TGUv+qRLnXuQlfwqtS/17AcsDswLNmtpfXKBF+k8hk8+f0eWrCZeILJN/ckkV4HsJ3ss5+qzlvAf8EljCzW5LCsgFh/V4f+Ja7P1731YikUL5NWPQvLinE3yX8oy8D/pH6hhPlsM9y9/f7fMAtKAWsXQlcmW9PQXW/J15UcoV4eeLecY+Z3VgHK2tyHxqU/t0E+ArwaFKYFyReXP5C5Pl+OB3zTzO7kwiMXB/YqS6y6S5SioUQYgppFZBlZusQD445gZ+6+8Wp/1oiqf8Y4EkiJdU9ZraMR9qmWisv/Y0WsvkZ8YD/EXBGrhBblIrenggwerL8XXWUjbu/YmYXE4F2r5vZZe7+d4vKXosRhT9qkX2iK7LgyHWAqUoy2YfwLf4HEVj4skVp6/0IV5dRxNJ+LapGdmHNL6ornsH4CvEKwJ+Ighn71UEhLkgKceHacnO26fPAl4CfeEfKw6nSCtknRKrDF4hAyX4VVCylWAgheod1iBKvu7j75fBZlavFgYOAc939I4vKaycCc5rZ09lS8jBgWldavp7mB0SRiKtLyteSRIT9IsCK7t4pIDJ3c6la8SooxuTuv0yBngcAK5vZxe5+tUXWiSmr8NUHJIW4eOHIZfJ/RBaXvxNVIR9NFuL9iEwOnxDluX8I/NprmgPczGYkrNxDiGIxuYX4T8D9RNXFJ7NjavEC1sX5szzwurvfBB0KsZnNCuxFVPE7glD0P6OVT3zdkE+xEEL0Dt8DNskU4u8Q6bQOBn7v7h+l/f5FLNX/FngwKTcQytmjZrZan466wVhHjulL8mXdpBAfSxSKWCspX7Ob2ZZmdp2Z3Qica2bTFtaziqbQiaSkF36sRwH7ED6dr6S+WisgOWUF0My2IF4eC4X4kaQQf5+w5v+V8F29iwjMu8tSHuM+HXg3cPf/ESsUG2YK8TyEi8v9wF4lhXgI8BMzW7eK8ZYxizzE6feHKMM9xMyWAsgU4psJd55DCZ/wT81sHjNbOSnORa7w2qLsE0II0cO0slaZ2dlElPYyHlXxCsVqGHAtkYf1fGJJ8j8WFaWOJ6rirePud/TV+NsJM1uWUFiWIRTi+81sAcLqujrwPKEELE5Y+pZw9w8yZTS3clZC/v+b2ZC0AtHtMdXJ+l2QArQOBi51979lCvGWhC/rt4oXSzNbBrieyOxySUVDbkkrOZjZNIQP8VREJcxHsm2DiZebPYlSypsA46p+wTGz6Yi87BcSLyL3Ah8R83DCxeVt4rq5OF0Xg4mMLxsSbiNfr3vsRC3edoUQokm0UIgHAnMAT3nkz027+TjCL3JFomTvzz0l8E+WzL2BU4Abk/Imep5tgDWB1ZNCPCtRvWsp4CR3X9HdtyGC1h4F5k9K5DhCGYDwa62MktL1cYu+LjGzEcDtuVWysAxWRWbR/0lSiKcnKvVtQbi+fCsp/sWqygAiM8XQiobcJV3IYVoih/YNJYV4CDHPnQjF/5vpZWWqvhjrRBhM5FbeJY1paWJ1ZWbiehlDuCYVCvGM7v6xu59DZK4YCNxtKY9xNVOYOFKKhRCiFzGzRdND4DHgS2a2cFpCnM7Mvk88ME4ATnT359IxRanbj4jgnDeAW8zsy1XMoeHsByzo7g+lzysSqaXOd/dfwGeZEV4hCi38DrgkWWQ9yeQli0wjlZPG1C2lNimbJxM+orOa2XyT+h29gXfkLy7GsAXwXeBuQin7KL2YfJSupf8j8uNeX8V4J4OpgXFEYRWgk0K8HfAIYfV+w8xGAz+1yABRGcnHfm9gwxTYuRXh0/1VIpfxEUR+9nHpvvYXM7vWzH7s7vcRmUI+Iq632iL3CSGE6CVSIM1xxIPueeAO4EMi7+dsxJL9CaRyvemYvJDBPEQu0JXSsV8lLJp39u1MmknuNmAdGRCuAka5+zKpf0i2TH8nYfG/jygf/VJaCj+SsO6t5+63VzSXA4AP3P3Y9HmC7hNJIb6IeAF4nnhpm50oMHF274+4+6SXxB8BRyf/1SJVmBEvjd8AtnH3K/pDMBeAme1HuIecQJSFXpu4H9wL7Ozub5rZHMDORPDaDcD2xUpTVS47ZrY4UR1yDmAkkXbueOCOdP0cTliMzyZSG64OPOTumyYXsqHuvmXpO+cE3nD3t/tsIl3h7mpqampqvdCAuYDHifLPA1LfUoRF631CmZoj239A9ve8xAPnTeArqe9YQqles+q5NaWRjEPZ5+uAS1rI42jCuncaMHfpmGkJf8uPiGpeVczjO2l8e3Y1t6x/eiIzwBhgd0K5GU0oM+8By1ctl2ysA0ufBxVzS9fVe0ShjKmqHuuknm/ARoTrwROE1fV8YPq0bQ4iYO1jInjyaiIwb4cazGEmIvf6UqX+LxDZQw5Jn6cmXuhfSNfVzcARxbb077LE6st3CtlWOreqB6CmpqbW5EYE0L1BBKScQgSqjCPyys6d7VdWiP+YFOLVsv6hRJaKs6qeV9Ma8OX078nAU0kGxYP7mPRgPx2YM/VZasXLzvREoNG7REq3KuawE/Apkc2g6Csr/TPQ8bK1EWG5y8+7/wA7Vi2PiczTkizeTQrx0KrHNInjz6/1pYi82JcD06W+OZNC/Haa5wgi5/T/JaVz1arn0MW85kpz+b9S/8ZEVoongY1T31eBW9N9cQzpxb/qVuvUGEII0d/xqDC2Ih0ZDgYQEfTXekrUX3KZmJdYjvwKkeT/1mKp1KO4x7e9I1hP9AAWpbi/Y2YvEEvVDxCBj8NSBP2yhFXycHd/oZBH5nIxiCg//D8iGO9WM+tzNxd3/23yfz7VzHD3E0vjnJ5Y+t6AeLm6yTsXjJmJCIga25fjngx+SLgV7Eik/upXVdM8c+9w94fMbHfgbnd/K7lMfIs4D89w933SrmPTufgu4ebyGVVnP8l4EXga2NLM7iZeLucnZDUW+I27X5H2nYooADIVsHm6zw0BPvEK3V+kFAshRC/jkfd2a49UXsO8cwnbCSrE6XjPvqtSn8KG8hSRcu1wd/8mUTJ5AUKBnJVQiA/tQiEeTFi8RhNLwLcT/q+3m9lq7n5bX07E3c8wMwdOT1kcfpnGOR2x+rAA4T98T/FSluYzGNiMSDv3dPF9Jb/rupxzRxPpym7tbwpxTvaye3X6PDeR2uxbwDmFQmwd6RvnIdwpVjazBwmXkYeT/KpOC1hcD+sSKSZ/Qli35yKsxEe4+y/TvmsAu6b+V4ATzOxWd38zbf8cUbjoX30+j3qc30II0WwyZcqgZbGCeYmgvNXIFGLRN5jZEoRy+zDwIKE8LkEokke7+7MTUYh3B67xyEE9jPAxfoco8d3nln0z25VQzr9OFIi5DFgQ2JbIffsmcJhHBocRhBvCMYTrxW+SIrYEkY3jUXe/JX1v1cpX7XIq9xQpMPduwqK6Z2nbisB5hDHzbsLy+lXgQHc/t6/H2opSAOTyxAv+bMT1c2LaZ00iy8ZChAX5FsLf+HVg63RdbQb8HFjX3f/dp3OQUiyEENViZrMRmQAWBjaVQlwNZrYw4es9HxE09ysil+yr3bAQX+cdVQpJLhVTF9bYKrCoOPYYsAdhXV3P3a83s+0I3+mT065fIAosHOHuB2THfykduyY1Ur6ajJmt4qUMJma2AqEQjyHymv8DmIZwI9kJWM7dX+jrsbYiv06IMf/L3Q9M29YgFOJFiNR6NyQFeiQRrPoSUajkYzPb1N0v6/PxSykWQohqSQ+Q64jUbP0l12ojSX6N5Aputq1bCnEdMbPFgCe8I73cfkThkhkJS/IZ7v4Hi0qKSxJW7qeTz3GRPuwbxVK/6Fm6SiWXrMfnEwrx5p5SN6ZtOxHW/eXc/ak+G+xEyK6TvNLiwsBRhEvFt5JCnFeFHJm2f45YKSvcxPq0+p2UYiGEqJCuHoaiekr+3oUFbDCx5DsP/UAhLp9fuZKRfDffJVJhvWFmMxJpv+YnLOUvEDlzHzGzM4m0YXsAn9bEt7jRmNnKwAVEmfFvuPuLJfndDMzo7ktWOMyWlN1szGwLYi6buPsfk0Js7v5pOg9/DKwH3OnuO2XH9am7jCraCSFEhUghri/eOUuAJ5eIfxLBQ7VXiGH88yspIYVf+xh3fycpxFMRSsv0wPcI5fc5ojTvokRawdHu/okU4j5jPsL3e6sWCvEfiXRuP+7q4ELOVdDiHJkF+Dfhsw/jK8QbAFcXCrGZHZKCkvvUf1yWYiGEEKIbWFRWuxT4A1HSttYK8aRgZjMBfwWuzLIeTEcUjNkCeBb4C1EcpJGBbnWkyFZTygLyR2ANIm3b+R5V/j5HBFJ+CPynCFCrOjCyIF07DxJK/tfd/bWSQvxHd9837fs14Arga+7+p74cpyzFQgghxEQorHTuvjFwaZMU4sTUROGPtzNL8luE9XgwEQh1iBTiviGTwbtJsS0U4kuJ0sn/RwSyTZUyjTxOvLD9Cbg2BVO2stj2OZmFe2nCFWT65EP8I8ZXiAeT3CiAv/f1WJWnWAghhJgIebBPExVDd3/JzE4hKqlhZv8gAu62IPLJru3ur8gHvm/IldksWO0mYGUiA8XFhI/3d4ADiGIspxCZHFYFTjOzse5+VR8PfTySm8SgFDy3fXLVOQX4MnCFu++X7b4AsAtRFe95GM8PfiCRoaJXlH25TwghhBBtTClLwPZEbuIliQqMLxAK8RNSiKvFzL5DlLj+vUf1uzWISopGVMjcKO03nCg88zqRP7vXlMjJIQXZXQyMdfddUp8RZezPIRT7r7r7e9kxw4GP3f2DrG96TwU/egq5TwghhBBtTJZXFnc/B/gpoaD8B1hNCnG1ZK4UJwOnJYV4KmBforjMksAXzeyatN/bwAhgeHL5qZVCnM6jzTKFeEga42iipPpVhUJsZouZ2W5E5bsLU7ArFhUnrzCz5XpyfHKfEEIIIdqcUqaNMWZ2EpEe699SiKul5EpRuO58DMwAPO5RRn4D4Cozu5oo6DEzcE9fj3VipJzEearDgZl//kGEdfu4lB7wO8D2wDDgXMJC/kla2XjCzO4iUgf2GHKfEEIIIQTQMr+sFOKakaVaO4VY8d/L3T+wKFV+OWFxfQFYwt3fqGSQk0hyBfk1cCIwO7AOkU3jOOA2d78x7bc2sDWwe2ZNHkS8O0xxkQ+5TwghhBACGD9bgRTi+uEJ4A4iEHJLM5vG3R8GNiOyUOyY8k8PrHKsk8DCRNGYE4FNiPSAS7n7TzKFeB0iY8U2hOJcsBSwo5lNP6WDkKVYCCGEEKIfYmY/BvYhLKr3uPstKSjt/f6UJSVZe68EngR+AbxRCqpbGziQsIJv5O4Pmdk3gNeAk4k82pu6+ztTNA4pxUIIIYQQ/YeSX+7OwMaEy8Se/c26XxQmKWVByf9ekwj+nIdQiB9MVuGrgJWAx4Bl8mwVkz0WKcVCCCGEEP2LkmI8O/BqT/jVVkFJCc7zErdSiAcDXwJ+R6Rvewv4gkdlv4FT8hvIp1gIIYQQop+RMjkU6dpeSkUybGLHFZjZKma2Se+NsPuUMmwUCvEGRGGS0XRWiJchipV8AKxJVPN7NvlVT9JvUEZKsRBCCCFEP6SUKaRT5pAJYWbLEu4Hu+cBalOiUPYC6xHuEblCvDRwBlGSfF13v5+ogPcHYFiR83hy5yH3CSGEEEKIfoiZzQKs4O5/TJ8nqhib2ZeJinKjgDuB54lyy5f09ngnhZQ5Y153fyp9XgI4HxgHrOfuL5b2PwxYG1hxcl0oZCkWQgghhOifLAVcbmZ7wWfVCbu0kiaF+CqiQtz+wLHAM8BFdXGlgA6/4kIhTvwSmBpYq1CIrYMZCT/jB4HBk/v/qqKdEEIIIUQ/xN2vN7NvAackK/GvCsW4bDE2sxWIstAPAPu6+z9S/58Jt4TNU0W8j6suDV229JrZPIRl+1R3f7W0rydlfy3CpSJP5dZtlxKQpVgIIYQQot/i7mcSJZGPN7O9U18ni7GZLQ9cSBhDfwb8Mzv+PWA4MMTdP6paIe6CZ4Gnga8nlxHgs3nOSORq/r273wSRmaPYnj53q4iJlGIhhBBCiH6Mu58BfBs41sy+m/oKhXAl4BpgDuAv7n5brjSb2SLAbMDTrVwvqg6+S6nnHNgUmBE4zMwGFIov8DngC8Ct2WEjzGxtMzvPzIZ218dY7hNCCCGEEP0cdz8jKbCnmNnL7n5Rciu4HLgJ+DOwlpkNd/e3k2I8Ejic8NU9NvVNR6RBG+fu/+zKHaMP5zUuFfj42MwWBGYqFSj5BfCEu58DYGb7EK4UqxM+xmsAf+rO/6XsE0IIIYQQDcHMNgSGAA8RLgfXAnsRBTDOJtwnHgTmBvYAlgQ2d/cb0/HzE+4YqwFHuvsf+nYG45PlYy6s34OItGxfIoIFryMU+42BzwOXApe4+5WT8v/IUiyEEEII0RDc/aribzP7AfAnd/838G8zO42wrA4F3gT+B6zu7g9lFuQnzewYYAxwZrISn1/BVD4jU4bncfdn3P2T9HlloGj3ES8CGxHV/T7o6vu6QpZiIYQQQogGkZeAbrFtBWAEoRQ/5O7vmdl6wHeBGYBHgcPd/WkzOwhYB9jM3V/uk8F3gZlNA5wCvA88BSwM7AA8SVS4+wOhDH88oflP8P+QUiyEEEII0Wy6UhTNbC3gCiJQ7b/ATMDywHLASOBKYPlkba4UM1uSGM8bwIfA0cBf3f3ZbJ/J9n+W+4QQQgghRMMpK8SZkrw58BzwXXd/ysxmBw4m/I5/BbxITfRFd/9bqmw3jggEfLvYVsxnSgIClZJNCCGEEKLNyJTkQYTl9aXU/xJwIFHk4wDgPnd/sooxtsLdx7r7m+7+dp4ubnLcJcpIKRZCCCGEaF9uAxYnKtrNb2ajgF2AZYCL3P3bUH2+4lb0dJo4+RQLIYQQQrQxZvZ9YANgMJG6bRbgCnffNG2vLE9xXyKlWAghhBCiDcmD78xsPmAz4AjgbHffKfW3hUIMcp8QQgghhGhLUrW4ojDGv4DXgNvbUSEGWYqFEEIIIUSJdlOIQZZiIYQQQoi2pxxI124KMchSLIQQQgghhCzFQgghhBBCSCkWQgghhBBtj5RiIYQQQgjR9kgpFkIIIYQQbY+UYiGEEEII0fZIKRZCCCGEEG2PlGIhhBBCCNH2SCkWQgghhBBtz/8DRdzq25GoIC0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "importances = rf.feature_importances_\n",
    "included = X.columns.values\n",
    "indices = np.argsort(importances)[::-1]\n",
    "\n",
    "plt.figure(figsize=(10,5))\n",
    "plt.bar(x = included[indices][0:10], height = importances[indices][0:10],\n",
    "       color = 'g', edgecolor = 'k')\n",
    "plt.xticks(rotation = -45, fontsize = 15, ha='left')\n",
    "plt.yticks(fontsize = 15)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "9145b9a6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 4 2 1 3]\n",
      "[0 2 3]\n",
      "[3 1 2 4 0]\n"
     ]
    }
   ],
   "source": [
    "a = [1,5,3,6,2]\n",
    "b = np.argsort(a)[::1]\n",
    "c = np.argsort(a)[::2]\n",
    "d = np.argsort(a)[::-1]\n",
    "print(b)\n",
    "print(c)\n",
    "print(d)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "180130ef",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}